Citation
Geographic information system applications for cultural resource management

Material Information

Title:
Geographic information system applications for cultural resource management
Creator:
Calamia, Mark Anthony
Language:
English
Physical Description:
xiii, 207 leaves : folded charts, maps (some color) ; 28 cm

Subjects

Subjects / Keywords:
Land use -- Planning -- Environmental aspects ( lcsh )
Environmental archaeology ( lcsh )
Environmental archaeology ( fast )
Land use -- Planning -- Environmental aspects ( fast )
Genre:
Academic theses. ( lcgft )
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )
Academic theses ( lcgft )

Notes

Bibliography:
Includes bibliographical references (leaves 200-207).
General Note:
Submitted in partial fulfillment of the requirements for the degree, Master of Planning and Community Development, College of Design and Planning.
Statement of Responsibility:
by Mark Anthony Calamia.

Record Information

Source Institution:
University of Colorado Denver
Holding Location:
Auraria Library
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
15344036 ( OCLC )
ocm15344036
Classification:
LD1190.A78 1986 .C325 ( lcc )

Downloads

This item has the following downloads:


Full Text
GEOGRAPHIC INFORMATION SYSTEM
//
APPLICATIONS FOR CULTURAL RESOURCE MANAGEMENT
by
Mark Anthony Cal ami a
/
auraria library
Thesi s
Submitted in partial fulfillment of the requirements for the Masters of Planning and Community Development Degree
University of Colorado at Denver, CO
August 5, 1986


For
Mom and Dad


ACKNOWLEDGEMENTS
This thesis might not have been completed if it had not been for many people who gave much of their time in assisting me with this project.
Several BLM employees and Technical Government Services (TGS) Inc. personnel gave me valuable advice and assistance throughout the project. Daniel Webster initially trained me to use the BLM's Geographic Information System (GIS) -- MOSS/MAPS. Daniel Martin provided excellent suggestions in applying a GIS to cultural resource management situations. Sol Katz, Wendy Telley, Paul Kimberling, John Foster, John Russell, Carl Zulick, and Robert Ader taught me techniques on the GIS to accomplish some tasks more efficiently. Jim Turner answered questions on geomorphology and A1 Amen assisted me with the soils data. Rose Maruska prepared and printed the flow diagrams. Randy McKinley formatted the digital elevation model data and Ed Chine produced the color Applicon plots for this study. Doug Sipes, Steve Russell, and Sharon Chomas assisted me in loading and unloading data and in solving problems encountered with the computer's operating system. Mike Carsella, Mike Fiebach, Dennis Colarelli answered my questions on command algorithms.
Outside of the BLM Service Center, several individuals were particularly helpful in gathering data and answering technical questions. Those individuals were John Roney (BLM Albuquerque District Office archaeologist), Robert Bewley (BLM Albuquerque District Office geographer), and Jeff Neibert (BLM geographic information system coordinator for the BLM New Mexico State Office). Professor Steven Wells, geomorphologist at the University of New Mexico allowed me to use several geomorphological map overlays. I am also thankful to Steve Fosberg (BLM New Mexico State Office archaeologist) for


his support and encouragement throughout the course of this project. Marsha Jackson at the New Mexico Laboratory of Anthropology answered my questions of the Archaeological Records Management System (ARMS).
Ken McGinty assisted in editing this report, and Gerry Stakes spent many hours formatting and typing the manuscript. I am indebted to them for their valuable criticisms and comments during preparation of the final product.
My thesis advisor, Dr. Thomas Clark, provided much encouragement and general guidance for this study.
Above all others, I would like to say a special thank you to my coworker Mike Garratt (BLM statistician) who assisted with the quantitative aspects of this study. In addition, he provided many comments related to the technical accuracy and content of the report. Without his candid criticisms and remarks, this manuscript would have suffered. For his constant encouragement, patience, and generosity, I am grateful.
Finally, it should be said that any shortcomings or faults which may occur in any portion of this thesis ultimately rest with the author himself. Thus, I alone claim full responsibility for the contents of this report.
Mark A. Cal ami a August 1986


TABLE OF CONTENTS
Page
ACKNOWLEDGEMENTS......................................................... 111
LIST OF TABLES........................................................ viii
LIST OF FIGURES........................................................... xi
ABSTRACT................................................................. xiv
Chapter
I INTRODUCTION ...................................................... 1
Project Goals and Concept.......................................... 1
The Cultural Resource Management Problem .......................... 2
BLM Resource Management Plans...................................... 3
Planning and Historic Preservation ................................ 4
Spatial Analysis and Support Program Needs ........................ 9
The New Mexico Data Synthesis Project............................. 10
Structure of Report............................................... 13
II GEOGRAPHIC INFORMATION SYSTEMS AMD
CULTURAL RESOURCE MANAGEMENT STUDIES ........................... 16
Geographic Information Systems ................................... 16
The MOSS/MAPS System ............................................. 22
Previous CRM Studies Using Geographic Information Systems. . 23
III EXISTING ENVIRONMENT ............................................. 29
Location and Physiographic Setting ............................... 29
Climate........................................................... 31
Hydrology......................................................... 31
Geology and Geomorphology......................................... 32
Local Lithic Raw Materials........................................ 36
Soils............................................................. 37
Flora............................................................. 37
Fauna............................................................. 42
IV THE ARCHAIC PERIOD................................................ 43
Overview of the Archaic........................................... 43
v


TABLE OF CONTENTS (CONT.)
Chapter Page
V ASSESSING ARCHAEOLOGICAL INVENTORY
AND ENVIRONMENTAL DATA BASES.................................... 52
Selecting the Study Area.......................................... 52
Planned Development in the Study Area............................. 54
BLM Inventory Levels ............................................. 54
Problems and Biases in Using Existing Site and
Inventory Data.................................................. 56
Data Themes....................................................... 60
Procedures for Assessing Inventory Data Usina
MOSS/MAPS....................................................... 65
Evaluating the,Inventory and Environmental Data................... 68
Results of Analysis............................................... 71
Evaluating the Inventory and Administrative Data ................. 90
Results of Analysis............................................... 94
Conclusions...................................................... 116
VI USE OF MOSS/MAPS FOR IDENTIFYING POTENTIAL CULTURAL
RESOURCE AND LAND USE CONFLICT AREAS........................... 118
Procedures for Using MOSS/MAPS to Prepare Data for Identification of Potential Cultural
Resource and Land Use Conflict Area............................ 119
Methodology Used for the Analysis of Potential
Cultural Resource and Land Use Conflict Areas.................. 127
Results of Analysis.............................................. 129
Conclusions...................................................... 145
VII USING A GIS IN CULTURAL RESOURCE MANAGEMENT SITUATIONS ... 149
Scenario 1....................................................... 149
Scenario 2....................................................... 155
Conclusions...................................................... 160
VIII EVALUATION OF MOSS/MAPS FOR MEETING CULTURAL RESOURCE
MANAGEMENT NEEDS .............................................. 161
Assessment of MOSS/MAPS.......................................... 161
Capabilities and Benefits of MOSS/MAPS for CRM................... 162
Problems of Using MOSS/MAPS for CRM.............................. 164
The Relational Data Base Management Approach .................... 170
Cost Estimates................................................... 171
Conclusions...................................................... 175
vi


TABLE OF CONTENTS (CONT.)
Chapter Page
IX THE USE OF A GIS IN THE BLM RESOURCE MANAGEMENT
PLANNING PROCESS AND FOR SUPPORT PROGRAM NEEDS ............ 176
Using MOSS/MAPS in the Planning Process.................... 177
GIS and Support Program Needs.............................. 187
Summary...................................................... 189
X CONCLUSION................................................... 191
The Need for a GIS in Site Location Modeling................. 191
GIS and the New Mexico Data Synthesis Project................... 197
GLOSSARY................................................................. 198
BIBLIOGRAPHY........................................................ 200
vi i


LIST OF TABLES
Table Page
3.1 Soil Types and Associated Range Site Data
for the Project Area....................................... 38
3.2 Major Vegetation Units and Associated Land Forms ................... O
5.1 Elevation Data Overlaid with Inventory data........................ 73
5.2 Slope Data Overlaid with Inventory Data............................ 75
5.3 Aspect Data Overlaid with Inventory Data........................... 78
5.4 Drainage Data Overlaid with Inventory Data......................... 80
5.5 Stream Data Overlaid with Inventory Data........................... 82
5.6 Soils Data Overlaid with Inventory Data............................ 85
5.7 Surface Geology Data Overlaid with Inventory Data.................. 88
5.8 Geologic Surficial Process Units Data Overlaid with
Inventory Data.................................................. 91
5.9 Land Status (Ownership) Data Overlaid with Inventory Data. . 95
5.10 Wilderness Study Areas (WSA) Data Overlaid with
Inventory data.................................................. 98
5.11 Linear Development in WSA Data Overlaid with
Inventory Data................................................. 100
5.12 Road Network Data Overlaid with Inventory Data.................... 103
5.13 Proposed Areas of Critical Environmental Concern (ACEC)
Data Overlaid with Inventory Data.............................. 105
5.14 Range Allotment Data Overlaid with Inventory Data................. 108
5.15 Oil and Gas Lease Data Overlaid with Inventory Data............... 110
5.16 Preference Right Lease Applications (PRLA) Data
Overlaid with Inventory Data................................... 112
5.17 Industry Expressions of Interest (IEOI) in Coal Leasing
Data Overlaid with Inventory Data.............................. 114
vi i i


LIST OF TABLES (CONT.)
Table Page
6.1 Area Table of Preference Right Lease Applications (PRLA)
Data Overlaid with Archaic Site Data for Two Leases. ... 142
6.2 Area Table of Oil and Gas Lease Data Overlaid with
Archaic Site Data for Four Leases......................... 148
8.1 Estimated time for Digitizing and Editing Selected Themes. . 173
8.2 Actual Time for Digitizing and Editing Selected Themes . . . 174
9.1 Acreage Table of Total Land Available for Uses Under
Various RMP Alternative Management Schemes .............. 182
IX


LIST OF FIGURES
Figure Page
2.1 Construction of a GIS.............................................. 20
2.2 Character!sties of Map Information in Vector-Based and
Cell-Based Geographic Information Systems ...................... 21
3.1 Location Map....................................................... 30
5.1 Data Themes Used for Analysis...................................... 61
5.2 Cartographic Model for Assessing the Archaeological
Inventory Data.................................................. 66
5.3 Map of Elevation Data Overlaid with Inventory Data................ 74
5.4 Map of Slope Data Overlaid with Inventory Data.................... 76
5.5 Map of Aspect Data Overlaid with Inventory Data................... 79
5.6 Map of Drainage Data Overlaid with Inventory Data................. 81
5.7 Map of Stream Data Overlaid with Iventory Data.................... 83
5.8 Map of Soils Data Overlaid with Inventory Data.................... 86
5.9 Map of Surface Geology Data Overlaid with Inventory Data. . 89
5.10 Map of Geologic Surficial Process Units Data Overlaid
with Inventory Data............................................. 92
5.11 Map of Land Status (Ownership) Data Overlaid with
Inventory Data.................................................. 96
5.12 Wilderness Study Areas (WSA) Data Overlaid with
Inventory Data.................................................. 99
5.13 Linear Development in WSA Data Overlaid with Inventory Data 101
5.14 Road Network Data Overlaid with Inventory Data.................... 104
5.15 Map of Proposed Areas of Critical Environmental Concern
(ACEC) Data Overlaid with Inventory Data....................... 106
x


LIST OF FIGURES (CONT.)
Figure Page
5.16 Range Allotment Data Overlaid with Inventory Data .......... 109
5.17 Oil and Gas Lease Data Overlaid with Inventory DAta .... Ill
5.18 Preference Right Lease Applications (PRLA) Data Overlaid
with Inventory Data...................................... 113
5.19 Industry Expressions of Interest (IEOI) in Coal Leasing
Data Overlaid with Inventory Data........................ 115
6.1 Cartographic Model for Preparing Archaeological Site Map. . 122
6.2 Cartographic Model for Identifying Cultural Resource and
Land Use Conflict Areas.................................. 123
6.3 Map of Range Data Overlaid with Archaic Site Data ............... 130
6.4 Map of Proposed ACEC Data Overlaid with Archaic Site Data . 132
6.5 Map of Fossil Forest Data Overlaid with Archaic Site Data . 133
6.6 Map of WSA Data Overlaid with Archaic Site Data............ 134
6.7 Map of Land Status Data Overlaid with Archaic Site Data . . 135
6.8 Map of Linear Development in WSA Data Overlaid with
Archaic Site Data........................................ 137
6.9 Map of Road Network Data Overlaid with Archaic Site Data. . 138
6.10 Map of Preference Right Lease Applications (PRLA) Data
Overlaid with Archaic site data.......................... 140
6.11 Enlarged Map of Preference Right Lease Applications (PRLA)
Data Overlaid with Archaic Site Data..................... 141
6.12 Map of Industry Expressions of Interest (IEOI) in Coal
Leasing Data Overlaid with Archaic Site Data............. 143
6.13 Enlarged Map of Industry Expressions of Interest (IEOI) in
Coal Leasing Data Overlaid with Archaic Site Data .... 144
xi


LIST OF FIGURES (CONT.)
Figure Page
6.14 Map of Oil and Gas Lease Data Overlaid with Archaic
Site Data................................................ 146
6.15 Enlarged Map of Oil and Gas Lease Data Overlaid with
Archaic Site Data........................................ 147
7.1 Map of Distribution of Archaic Site Data Relative to
Preference Right Lease Applications (PRLA) Data ......... 151
7.2 Enlarged Map of Preference Right Lease Applications (PRLA)
Data Overlaid with Inventory Data........................ 152
7.3 Enlarged Map of Preference Right Lease Applications (PRLA)
Data Plotted with Archaic Site Data...................... 153
7.4 Enlarged Map of Archaic Sites Plotted on Preference Right
Lease Applications (PRLA) Data as Overlaid with Geologic Surficial Process Units Data......................... 154
7.5 Map of Distribution of Archaic Site Data Relative to
Road Network Data........................................ 157
7.6 Enlarged Map of Distribution of Archaic Site Data
Relative to Road Segments................................ 158
XI 1


ABSTRACT
The purpose of this thesis is to identify and evaluate the applications of Geographic Information System (GIS) technology for federal government cultural resource management situations, especially those involving planning and compliance for fossil fuel development. Procedures are developed for performing spatial analysis involving administrative, environmental and cultural resource data. The demonstration of these procedures will aid BLM state, district, and resource area offices in meeting their resource management planning needs. Capabilities, limitations, and deficiencies as currently exist for performing archaeological spatial analysis will be identified for BLM's primary GIS --MOSS/MAPS.
XI 1 1


CHAPTER I
INTRODUCTION
Project Goals and Concept
The purpose of this study is to identify and evaluate the applications of Geographic Information System (GIS) technology to federal government cultural resource management situations, especially those involving planning and compliance for fossil fuel development. Using some of the technological aspects described in the Bureau of Land Management (BLM) technical volume, BLM Predictive Modeling Draft, DPP No. 8000.001, procedures are developed for performing spatial analysis involving cultural resources. The demonstration of these procedures will aid BLM state, district, and resource area offices in meeting their resource management planning needs. It is suggested that the reader of this report have some familiarity with GISs, and MOSS/MAPS in particular, to facilitate understanding of the material. Capabilities, limitations, and deficiencies as currently exist for performing archaeological spatial analysis within a planning context, will be identified for BLM's primary GIS.
The aim of this project is to show how a GIS can serve as a valuable tool for a BLM resource area in planning for energy development (coal, oil, gas) and cultural resource management (CRM). This study identifies applications to CRM in a multiple land use context and applies the Map Overlay and Statistical System (MOSS) and the Map Analysis and Processing System (MAPS) to functions that would be more effectively accomplished
1


through the analytical capabilities of a GIS than through conventional means. Evaluation of the utility of a GIS for compliance and planning includes time, accuracy, monetary cost, and overall effectiveness. The MOSS/MAPS is BLM's GIS
The Cultural Resource Management Problem
The project area consists of a small portion (154,187 acres or 10.3 percent) of the BLM Farmington Resource Area (FRA) in northwestern New Mexico. The area itself falls within the northern San Juan Basin. The FRA has a large amount of existing and proposed mineral development and leasing. Most of the proposed development is in the form of surface coal mining and oil and gas extraction. Federal law requires that all cultural resources on federal properties be identified and assessed by the significance criteria listed in the National Historic Preservation Act of 1966, as amended.
The FRA is faced with a problem regarding the distribution and location of cultural resources with respect to expected coal and oil and gas development. Specifically, the FRA requires (1) a qualitative and quantitative assessment of its existing cultural site, survey, and environmental data bases, and (2) the identification of potential conflict areas between cultural resources and specific land uses.
This study shows and documents the utility of a GIS for assessing data bases. Moreover, this study identifies these existing site locations with respect to existing coal leases and oil and gas leases so that potential conflict areas can be addressed for compliance and planning needs.
2


BLM Resource Management Plans
BLM receives its authority to balance development with protection of natural and cultural resources from the National Environmental Policy Act of 1969 (NEPA); the Federal Land Policy Management Act of 1976 (FLPMA); and other federal, state, and local laws. NEPA requires evaluation of the effects of major federal actions on environmental (including cultural) resources. Under NEPA, every federally funded or licensed entity must consider the total environment. FLPMA specifies several key directions for BLM, notably that goals and objectives be established as guidelines for public land use planning, and that public lands be managed on the basis of multiple-use and sustained yield, unless otherwise specified by law.
To achieve its land use management goals, BLM develops comprehensive land use plans called resource management plans (RMPs) to set guidelines for multiple- use decisionmaking. The plans are used by field-level managers in BLM resource areas. Individual plans are prepared for each area, which may vary in size from 100,000 acres to several million acres. Because of geographic and natural diversity, each plan is unique to the area it addresses. RMPs must be both comprehensive and issue oriented, must discuss all affected resources and concerns, and must focus on actual problems. The plans must allocate resources for specific uses.
Environmental impact statements (EISs) are fully integrated into the RMPs and discuss the effects of implementing each management alternative presented in the RMP. Once all alternatives have been thoroughly analyzed, BLM tentatively selects a preferred, multiple use-oriented alternative. BLM also follows a public involvement plan during the
3


typical 2 years of RMP preparation. Once completed, the RMP is implemented through the decisions made in the document and for on-going activities by such site-specific and resource activity plans as cultural resource management plans. The plans are monitored throughout implementation (about 10 years) to ensure that impacts are accurately assessed and goals are being met.
BLM uses a GIS to assist in thorough analysis and timely completion of projects that use large data bases. In some cases, these tasks would be impossible to complete manually while maintaining the high-quality output that a GIS can produce.
Land use planning is often the medium for implementing a GIS and building a comprehensive digital data base. Planning integrates all resource and mineral disciplines and lends a common purpose and consistency to goals for each resource represented in the planning effort.
Interdisciplinary teams prepare the RMPs. The planning team assesses funding and data needs and either guides resource specialists in preparing data for digitizing or procures existing digital data. With the assistance of management and resource specialists, planners often guide and shape the implementation of a GIS. This report describes the specific applications of a GIS to aid in the RMP process.
Planning and Historic Preservation
The National Historic Preservation Act (NHPA) as amended and as supplemented by Executive Order 11593 requires federal agencies to locate significant cultural resource properties and allows for regulatory agency comment before implementing actions that would effect such properties.
The quasi-regulatory agency created by the act is the Advisory Council on
4


Historic Preservation, which has supported the integration of cultural resource management considerations into federal agency planning and thus into land use planning.
Other environmentally influenced legislation is equally responsible for implementing cultural resource management activities. NEPA and FLPMA are good examples. As mentioned above, these acts mandate an emphasis on land use planning and designate a wide spectrum of natural and cultural resources (specifically including archaeological and historical properties) that must be considered in making specific land use decisions. These mandates constitute the first time in government decisionmaking when specified information about the locations of a whole range of resources has been needed.
The requirements of FLPMA and NEPA complement those of the National Historic Preservation Act. NEPA mandates that during federal decisionmaking all components of the environment be considered in an EIS. Although preparation of an EIS requires planning, FLPMA further formalizes the planning phase of decisionmaking and cites cultural resources as one of the types of resources to be included among the factors influencing decisions. The policy portion of FLPMA specifies protection of the quality of historical and archaeological values as a management responsibility of the federal government, specifically in this case, BLM. One of the central concerns of FLPMA is areas of historical and cultural value that require special management attention. FLPMA sets forth a management process wherein BLM must inventory such resources and other values encompassed by the other sections of FLPMA when they lie within BLM's areas of jurisdiction.
5


The planning process gives priority to designating and protecting these areas of historical and cultural value, but all resources must be identified and considered on the basis of their relative scarcity.
FLPMA, therefore, provides an orientation for federal land-managing agencies by explicitly naming cultural resources as one of the spectrum of resources that must be inventoried and included in the planning process and for which protection is a defined management option. The specific management prescriptions, however, are not defined, either in the act or in the 36 CFR 800 regulations. Instead, BLM has developed internal directives and manuals that set basic procedures and standards for cultural resource management. These procedures and standards define administrative and physical inventory methods, evaluation frameworks, and cultural resource protection measures.
The BLM inventory structure includes three classes of inventory (BLM 8100-CRM Manual, 1978). The Class I inventory is a review and compilation of existing data. The Class II inventory (survey) is defined as a sampling field inventory, but this definition does not address relationships between sampling and spatial analysis. The Class III inventory is a complete surface inventory of a specified area, i.e., an intensive field inventory. This report discusses the capabilities of a GIS to meet Class II and Class III inventory objectives.
The BLM inventory structure is intended to be a sequential tiering system. Some measure of Class I inventory is always conducted at first, whatever the eventual goal of the specific inventory situation. A complete Class I inventory treatment is conducted for every BLM administrative unit. As part of this inventory, a list of all previously recorded cultural resources is compiled, a narrative history and
6


prehistory of the area are written, and major research domains and data gaps are identified. A shorter version of a Class I inventory is performed for specific projects or programs to identify known cultural properties and inventory needs.
When the BLM CRM manual was written in the mid-1970s, much of the BLM administered land base had received little or no inventory. Class II inventory was envisioned for large-area survey applications, wherein a sampling approach would be used to estimate the density and distribution of cultural properties. Neither the manual nor associated BLM directives state that the objective of the Class II inventory should be to create a model to predict locations or environments having defined classes of cultural properties. Rather, the main objectives of Class II inventory are to (a) identify management opportunities, to protect cultural properties in their own right, (b) identify potential constraints on other land use decisions, and (c) identify general areas of sensitivity for cultural resource issues.
BLM has used the Class III inventory as the culmination of inventory for any project or undertaking that could affect cultural resources, or as the highest precision level of inventory for identifying resources within an area managed for its cultural resources. Class III is generally the level of inventory chosen when cultural properties are expected to occur because this is the most certain means of satisfying the legislative framework for cultural resource protection. Most of the contributing surveys in the project area were conducted at the Class III level as intensive survey units selected as part of Class II inventories in and near the project area.
7


This project has investigated the possibility of using a GIS for examining existing Class II and Class III inventory data to address Class II and Class III inventory objectives. More specifically, the survey data base has been analyzed to determine areas where future surveys need to be concentrated to provide more representative samples of cultural site data and environmental variables in the study area. In this manner, one can assess and measure the potential for biases associated with certain environmental characteristics resulting from nonrandom sample survey units.
In BLM, planning system products normally lead to a resource allocation decision. By considering the resources on BLM-administered lands, decisionmakers make trade-offs to achieve the most environmentally and publicly acceptable balance of resource uses.
To fit within the multiple-use concept, BLM has developed by internal directive a cultural resource evaluation system that assigns uses to individual cultural resources-. The system assigns value to the cultural resources categories commensurate with potential resource use. The seven uses allowable in the BLM system include the following: (1) current scientific use; (2) potential scientific use; (3) conservation for future use; (4) management use; (5) socio-cultural use; (6) public use; and (7) discharged use. See BLM 8100-Cultural Resource Management Manual (1978) for a discussion of these seven allowable uses.
This evaluation system provides a means of assessing the precedence of a cultural resource relative to other land uses. In a multiple-use management situation, the presence of a cultural property segregated from other resource use could prevent coal mining at a particular location.
On the other hand, the presence of a resource designated for excavation
8


could not prevent mining; once mitigation had been carried out, coal mining could occur.
The BLM's cultural resource evaluation system is not exempt from the NHPA requirements. Where a conflict exists between the proposed use of a cultural resource and the use of another resource such as coal mining, the site must be evaluated for NHPA eligibility. Moreover, potential effects on the property must be determined in consultation with the Advisory Council on Historic Preservation. Presumably, however, the BLM-determined use evaluation does influence the council's comments on effect.
Spatial Analysis and Support Program Needs
Cultural resource management programs maintained by federal, state, or other agencies are generally guided by a set of policies, procedures, and priorities. In virtually every case, some part of direction is to support other resource programs in meeting the requirements of historic preservation law, regulation, and agency policy. For example, cultural resource staff members generally provide support to coal program specialists when coal lease applications are being processed. Cultural resource specialists are involved throughout the project, during successive stages of alternative lease area selection--stipulation development, inventory, evaluation and mitigation, and monitoring.
Despite the importance of the support role filled by cultural resource staff members, the support program has not been formalized within any land-managing agency. Support has mostly been interpreted as compliance, which is only one aspect of support. A full support program should include pianning and implementation as they relate to an administration and compliance plan.
9


Two main issues need to be addressed in developing a cultural resource support program: (a) the types of information that are needed and (b) the strategies for tailoring a general support program to the needs of a specific project. The main components of a general support program--!'nventory and evaluation--fall in the domains of planning and implementation, and These components, in turn, should relate to separate administrative and compliance plans. Each of these components should be addressed independently to determine information needs and appropriate strategies. Requirements of each should also assist decisions made within the others. This report describes how information needs and strategies for tailoring a general support program for a project can be facilitated through the use of spatial analysis methodologies via a GIS.
Theoretical and methodological issues must ultimately provide a basis for determining why cultural resource information should be collected and analyzed, what types of information should receive priority, and how data should be collected and analyzed. To better realize this goal, some federal, state, and local agencies and institutions are recognizing the need to synthesize their growing data bases to reevaluate and reduce intensive survey requirements.
The New Mexico Data Synthesis Project
The Data Synthesis Project began in the Farmington and Carlsbad resource areas during fiscal year 1983 as part of the BLM New Mexico Cultural Resource Program. A statewide priority was granted to the project by the New Mexico BLM State Director, who saw as a main goal of the project the reevaluation and possible reduction of intensive inventory requirements in the two main oil and gas areas in New Mexico--Farmington and Carlsbad.
10


The rationale for this approach was that, though hundreds of inventories of well pads and access roads had been performed in these areas, no one had tried to synthesize the data from these "clearance" inventories and postulate where cultural resource properties would most likely occur. Because of the size and complexity of the information base, existing site and survey data had to be used through an automated data processing system. Once the needed hardware was purchased and located at the BLM Albuquerque District Office, the Data Synthesis Project was selected as a pilot project for exploring the applications of this equipment.
The Data Synthesis objectives are twofold: (1) to provide a means for the integrated evaluation and analysis of cultural resource site data, survey data, and appropriate environmental information and (2) using existing data, to provide the most information possible on the nature and distribution of cultural resources to inform and guide resource allocation decisions in a multiple-use context, in a timely and cost-effective manner.
The two basic project objectives reflect the dual emphasis of the project as it developed: (1) a primary push towards automating the large and complex data bases involved and (2) a secondary long-term effort to develop and refine potential applications of the data base. The development of spatial analysis methodologies per se was not considered to be one of the main objectives, though it was clearly recognized that model development would be facilitated in areas where the requisite data bases had been created.
When the Data Synthesis Project was conceived during 1982, the term GIS was not widely used within BLM. BLM offices lacked graphics hardware
11


and applications for planning and resource data display. In addition, data manipulation had not been established or demonstrated. In retrospect, however, the Data Synthesis Project relates to a GIS application project developed specifically for the display and manipulation of cultural resource and other relevant data. Additionally, the GIS allows for expanded multiple-use management planning capabilities involving cultural resources.
The Data Synthesis Project identified three types of project-specific data that had to be simultaneously displayed and manipulated in a graphics format. These included the site data, the surveyed area data, and the environmental data. Development of the data base was greatly facilitated by the presence of an existing automated statewide cultural site data base maintained by the Laboratory of Anthropology, a New Mexico State agency. Although certain backlogs in site entry are present, the scope of the file is statewide. With the exception of selected United States Forest Service sites, the file contains and receives information on all sites recorded within the State regardless of land ownership. The file is updated every 6 months.
The site file is maintained as part of the Archaeological Records Management System (ARMS). ARMS is the site data base that has been integrated with the MOSS graphic data base for use in this study.
BLM views the results of this GIS study as the next step in the New Mexico Data Synthesis Project towards achieving its two objectives, with emphasis on developing and refining potential applications for the system that could aid in CRM and support programs.
To evaluate the potential of a GIS for assessing data bases and providing information on distribution of cultural resources for guiding
12


resource allocation decisions, this study must present an overview of prehistoric settlement pattern behavior. For this study, the settlement and adaptive system of the Archaic period (5,500 B.C. to A.D. 400) has been explored for a portion of the upper San Juan Basin. A glossary is provided at the end of this report to aid readers who may be unfamiliar with some of the anthropology, archaeology, and geology terms.
Structure of Report
The remainder of this chapter describes the organization of this report. A brief description of the topics covered in each chapter is given below.
Chapter II briefly reviews what a GIS is and, in particular, the general capabilities of MOSS/MAPS. Chapter II also presents an overview of some of the major CRM projects involving spatial analysis using a GIS.
Chapter III briefly discusses the physiographic setting and natural environment of the project area.
Chapter IV presents an overview of the Archaic period, ethnographic data on hunting and gathering societies, and research on the Archaic period conducted in the northern San Juan Basin. In this study the GIS will be used to identify potential conflict areas between Archaic sites and specific land uses in the project area.
Chapter V describes and explains procedures and methods using a GIS for evaluating the archaeological inventory and environmental data for potential biases. The data sets used for analysis are discussed. Nine administrative themes (areas of planned and existing development or conservation) are also analyzed to determine the amount of inventory conducted for them. A cartographic model schematically documents a GIS user's procedures for combining layers of map data. The cartographic
13


model also documents how a planner or archaeologist can use a GIS, specifically MOSS/MAPS, to provide technical assistance in making a resource management decision. Results of the analysis are shown in graphic and tabular form with associated text.
Chapter VI includes a series of GIS-generated maps and tables that will aid in identifying cultural resource and land use conflict areas in a multiple-use setting. Specifically, the distribution of archaeological sites of a specific cultural period are analyzed in relation to their distribution with areas of proposed coal and oil and gas development. However, areas designated or proposed for conservation are also analyzed using the archaeological site data.
Chapter VII shows how MOSS/MAPS might be used in a typical CRM compliance situation. Analytical results from Chapters V and VI are combined to show how a resource area archaeologist can use output generated by MOSS/MAPS to meet BLM requirements. These requirements may include the preparation of environmental assessments (EAs), environmental impact statements (EISs), or resource management plans (RMPs).
Chapter VIII is an evaluation of the capabilities of the MOSS/MAPS software as integrated with ARMS for purposes of CRM and support program planning. The criteria used for this evaluation are monetary cost, time required, and level of accuracy compared to existing methods for accomplishing the same tasks. Additionally, some specific suggestions are given for enhancing MOSS/MAPS and ARMS for meeting CRM needs.
Chapter IX centers on GIS, specifically MOSS/MAPS, as a tool in the BLM resource management planning process involving cultural resources.
The phases in the RMP process where GIS could be used for the FRA are described. The remainder of the chapter focuses on the capabilities of
14


GIS in meeting federal support program needs for project planning.
Finally, Chapter X draws conclusions for using a GIS, as contrasted with manual methods, for performing site location modeling. Reference is also made to the use of the applications in this study for aiding the development of the New Mexico Data Synthesis Project.
15


CHAPTER II
GEOGRAPHIC INFORMATION SYSTEMS AND CULTURAL RESOURCE MANAGEMENT STUDIES
This chapter discusses what a GIS is and specifically the capabilities of MOSS/MAPS. Chapter II also presents an overview of some of the major CRM projects involving spatial analysis using a GIS.
Geographic Information Systems Geographic Information Systems are computer-based means for assembling, analyzing, and storing varied forms of data corresponding to specific geographic areas, with the spatial locations forming the basis of the system (Tomlinson and others 1976). The term GIS, as used here, is restricted to computer systems that can interrelate sets of data representing different geographical variables, as opposed to systems that merely manipulate or map individual files or geographical data (Rhind 1975).
Virtually any type of geographically distributed information from any source can be encoded in computer-compatible form. Computers can extract geographic information from digital geographic data bases, manipulate the data, derive new data, and analyze this information to propose solutions to problems. The analytical capabilities of such a system make it a powerful tool for spatial analysis. Thus, a GIS can go beyond the role of merely processing and displaying information. A GIS can also be incorporated into the analysis, interpretation, and problem solving aspects of research in geographically distributed phenomena and processes (Hasenstab 1983a).
16


Many types of geographically distributed data can serve as the primary information portion of a GIS: elevation data, river and stream locations, vegetation patterns and soil types (which might be derived from satellite remote sensing), known archaeological site locations, and regions of planned construction or development. In its most elementary use, a GIS can retrieve geographic information that is encoded in data bases for a specified coordinate point, such as the location of an archaeological site.
For archaeological research, additional information might include cultural data, such as known archaeological site locations, surveyed areas, access roads, and areas of planned development or impact. A GIS is also capable of using associated characteristic site and survey information from multiple attribute data derived from data base management systems. This attribute information can be used to select subsets of the data base beyond merely geographic characteristics.
Such a procedure, however, does not fully use the central capability of a GIS--the ability to derive new information beyond that originally encoded in the data base (Collins and Moon 1981). For example, from interrelationships between known points of elevation in the data base, it is possible to estimate at any locus the values of slope, aspect, and a variety of relief and terrain roughness measures (Monmonier 1982:76-29). Points of vantage, such as hilltops and ridges, can also be defined (Kvamme 1983a). From a digital hydrology net, distances to nearest seasonal or permanent streams can be computed, and from digitized vegetation data, distances to a specified plant community (Lee and others 1984). Listings of nearest neighbor sites and distances can be obtained, as well as the distance to a central place village from a data "layer" containing known archaeological site locations.
17


An important benefit for the data-generating capabilities of a GIS is that it can derive information that was previously impossible to obtain due to the sheer number of required calculations. Maximum view distances, measures of shelter and view quality, and least-effort travel distances are all potential information classes that illustrate this property.
One can obtain through the U.S. Geological Survey (USGS) and other government agencies or private companies many types of geographical data, particularly regional elevation data, already in digital form and on computer tape. For example, the USGS now produces highly accurate digital elevation models (DEMs) obtained through digitizing 1:24,000-scale topographic maps (Doyle 1978:1484). Most often, other sources of information, such as vegetation and soil data, do not exist in digital form; archaeological data usually is not available. Therefore, these data often need to be electronically digitized.
A common digitizing procedure employs a digitizing tablet and cursor (Monmonier 1982:7; Rogers and Dawson 1979). With these devices, such potential information as geomorphological mapping units or stream courses are manually traced and encoded in computer compatible form.
The primary data, which can be derived from many sources, are often digitized from 1:24,000-scale topographic maps, but other sources, such as remotely sensed digital satellite images, can be used (Shelton and Estes 1981). However they are acquired, the several primary surfaces of digital information that a GIS needs are encoded and stored in the initial data base. Computer programs are then able to use these primary data to generate secondarily derived information that is often more useful than the primary data (Collins and Moon 1981). For example, slope
18


estimates, aspect estimates, or distances to nearest drainages might be derived (from elevation and hydrology surfaces, respectively) and stored as new and distinct analytical surfaces.
The principle of GISs is that users input the system through digitization or other manual means with base information (primary layers) which can be used to derive additional or secondary data layers. Both primary and secondary surfaces can then be used for analysis or display. The ways in which these data are used, however, depend on the nature of the particular GIS (Figure 2.1).
There are two basic types of GIS designs: the vector-based and the cell-based GIS. The two designs are sometimes included in one GIS package.
The vector-based GIS stores data as a series of points, lines, and polygons that are used to identify discrete features (vector is another term for a line between two points). The computer storage requirements for this information are smaller than cell data because only the digitized points along lines or polygon boundaries need to be stored (Figure 2.2).
The cell-based (sometimes called raster based) GIS superimposes a regular grid of rows and columns of cells over the region and assigns a numeric value to each cell. Each cell corresponds to a fixed area in real space and each contains a value for that area. With a cell-based GIS dichotomous, discrete, and continuous data can be used.
19


ORIGINAL GROUND SURFACE AND THEMATIC MAPS
PRIMARY
SURFACES
SECONDARY
SURFACES
FIGURE 2.1 Construction of a GIS. From the original land surface (b), various thematic maps are
produced, such as elevation contours (c), hydrology (d), and forested areas (e). These maps are digitized and converted to primary layers in a GIS representing an elevation surface (f), a hydrology surface (g), and a forest location surface (h), which are all referenced to a reference grid, such as the UTM grid (a). From the elevation surface, secondary surfaces, such as slope (i), aspect (j), and local relief (k), might be obtained. The hydrology surface might provide a surface showing distance to nearest drainage (1), and the forest location surface might yield a surface showing distance to nearest forest (m).


BASE MAP
VECTOR /
\
II 1
CELL
I I I 1112
n
oio
Wo
O'O'O
MO
LjflSB
LUuQr
tTiTiTi
2J2I2I2I2)
I [11212 2
r“
1
IHhFJB
â– kumrf-jnhrp-i HMHRHHHnRinfl Kn^^puiniui
1 1

l 2 l ill
LI
tun
Li
lTIW
lllllll
I I
ITT
FIGURE 2.2 Characteristics of map information in vector-based and cell-based geographic information systems (after Lee and others. 1985)


The MOSS/MAPS System
MOSS (Map Overlay and Statistical System) is a GIS originally developed by the Western Energy and Land Use Team, U.S. Fish and Wildlife Service. MOSS has been in continuous development over the last few years with cooperation from the Bureau of Indian Affairs, the BLM, the Forest Service, the Geological Survey, and the Soil Conservation Service (Lee and others 1985). Thus, unlike most geographic systems, MOSS is in the public domain, although a superset of MOSS is marketed by Autometric of Fort Collins, Colorado. Most storage and processing in MOSS is in vector or polygon format, although some raster capabilities are available.
MOSS, as used by BLM, has both MOSS and MAPS. The version of MOSS/MAPS used for this project has not been released as of this writing.
Additional raster capabilities, designed in part to allow the incorporation of data derived from cell images, may be obtained through the Map Analysis Package Subsystem (MAPS), originally developed at Yale University. Most of the analytical manipulation in this study was performed using MAPS. To some extent, MAPS and MOSS can pass files back and forth. Input to MOSS is through the Analytical Digitizing System (ADS) or the Analytical Mapping System (AMS). Enhanced cartographic plotting beyond the normal capabilities of MOSS or MAPS, is provided by the Cartographic Output System (COS). For this report two maps were processed on the IDIMS system (see below) and plotted in color using the Applicon plotter at the BLM Service Center in Denver. The remainder of the maps used in this report were output on an Anadex printer.
Another system, the Interactive Digital Image Manipulation System (IDIMS) is mainly an image-processing system. For this reason IDIMS is organized in a raster format and includes many functions that address
22


problems specific to the processing of digital images, such as image-enhancement routines for remotely sensed data. The DEMs used for this project were processed on I DIMS into a MOSS/MAPS readable format at the BLM Service Center.
The MOSS/MAPS package provides flexible routines for overlay and neighborhood analysis, map description, and data management. A main advantage of this package is that it is used and supported by several federal agencies. At present, MOSS/MAPS has limited capabilities for internal inferential statistical analysis. However, data may be transported to a separate statistical package, e.g., SPSS, for analysis. Afterwards, the statistical results can be reintroduced into the system.
The version of MOSS/MAPS used in this project operates on Data General minicomputers and microcomputers, using the Advanced Operating System (AOS). MOSS/MAPS is operational on VAX and Hewlett Packard hardware. The advantages and limitations of using MOSS/MAPS for archaeological research and CRM planning will be described later in this report.
Previous CRM Studies Using Geographic Information Systems
Although many types of spatial analysis of archaeological and historic values have been used in cultural resource planning in a multiple land-use context, few have involved the use of a GIS. Only since the mid-1970s have GISs been used by public land holding agencies for management needs. What follows is a brief review of some of the major CRM locational studies using GISs.
23


The Wasson Field-Denver Unit COg Project
The Wasson Field-Denver Unit CC^ Project was funded by a private oil company in preparing an environmental impact statement (EIS) for a carbon dioxide well-field project in southwestern Colorado. The cultural resource portion of the EIS was needed in part because BLM required a right-of-way permit. Woodward-Clyde Consultants prepared a planning study that would improve well-field layout by reducing impacts to significant archaeological sites. This summary is based on two draft documents (James and others 1983; Woodward-Clyde Consultants 1978).
The project area contained 263,158 hectares (ha), which included plateaus, canyons, farmland, rangeland, and forests. Environmental and cultural data were entered, compiled, analyzed, and displayed with a GIS. Map-based information was coded and digitized for 175,000 cells (each about 1.5 ha) for land use and soil association, prehistoric farming areas, topography, roads, archaeological sites (e.g., period, size, type, and condition), biological communities, and geologic materials. Site significance was identified as the dependent variable and defined in part on the basis of age, type, size, and number of components for hundreds of known Basketmaker, Anasazi, and post-Anasazi sites. Fundamental to the definition of significance were the "subjective" attitudes of professional archaeologists. The archaeologists developed a seven point scale believed to conform to prevailing opinions of the professional archaeological community. Ultimately, three independent environmental variables--soi1, drainage rank, and slope--were used in a step-wise multiple regression, with the computed site significance values being the dependent variable. Sets of surveyed cells without sites were also included in the analysis. The
24


analysis yielded significance values for each cell, and scaled values were then color-coded and plotted on 1:24,000-scale maps. A total of 140 cells were randomly visited in the field as means of verification. The model was supported to the extent that the standard error of projected-to- observed value was identical to the standard error of the model (James and others 1983:23).
The Passaic River Project
The Passaic River Project was funded by the New York District Corps of Engineers. Robert Hasenstab (University of Massachusetts, Amherst) implemented
the project through a subcontract with Soils Systems Inc., an environmental consulting firm based in Marietta, Georgia. The project's objectives were to estimate the amounts of cultural materials likely to be affected by post- flood-control facilities and to define areas with a high probability of site occurrence (Hasenstab 1983b).
The 1,619-ha project area extends 160 linear km along the Passaic River, crosscutting ridge and valley, piedmont, coastal plain, and tidal-estuarine areas. Urban and commercial developments occupy most of the impact zone, but 42 percent is either agricultural, forested, or classed as wetlands.
The project area was subdivided into a high-resolution grid of 0.47 ha units (pixels) for which environmental variables were coded. All manipulation and mapping was performed on a GIS. Univariate statistical tests were employed to determine which environmental variables were most useful for their power to "retrodict" known site locations. Significant
25


variables were found to be soil drainage, distance to nearest river, distance to minor tributary/river confluence.
Grid cells were assigned a sensitivity rating by summarizing the component-variable ratings. The sensitivity models were then tested and revised with data derived from a survey of 300 pixels (ca. 140 ha) representing a stratified random sample (with some modifications). Overall, the sample fraction was about 6.5 percent of the impact zone.
The survey techniques included limited but systematic subsurface testing judgmentally selected pixels. Totals of 28 historical sites and 16 prehistoric sites were recorded. A series of computer-generated maps illustrated the final model on a pixel-by-pixel basis for prehistoric archaeological sensitivity (high, medium, or low, based on the component variable ratings) and a combination of historical and prehistoric sensitivity. Hasenstab (1983b:13) concluded that the GIS approach "has greatly enhanced the capabilities for archaeological prediction and land use management . . but it cannot be taken as a final solution to all cultural resource management problems."
Grand Junction Area Project
BLM funded the Grand Junction Area Project as a overview of statistical classification procedures for predicting archaeological site locations. This summary emphasizes aspects of the project related to the development and testing of models in the Grand Junction Resource Area.
For that area, the objectives were to develop quantitative models that could be used to predict likely locations of prehistoric sites (Kvamme 1983b).
26


The project encompasses some 438,966 ha of western Colorado uplands. Vegetation types characteristic of the area include desert grasslands, pinyon-juniper woodlands, and spruce-fir forests. The subareas of the district were stratified into five major biotic communities considered to occur in significant proportions across the landscape. A stratified proportional random sample of 65 ha quadrants (quarter sections) were selected from the physiographically defined subareas. One hundred quadrants were selected for survey, specifically to provide the data base for generating the models. The surveyed area amounted to about 1.5 percent of the project area. Environmental data were coded for site and non-site locations.
Through a series of statistical analyses, the following variables were found to be important in distinguishing between site and non-site locations: biotic zone, vertical distance to permanent water, vantage point distance, slope, view, aspect, shelter within 100 m, and shelter within 250 m. The models were developed through a pattern-recognition approach using multivariate analyses as classification tools. The most successful analysis was logistic regression.
Depending upon the particular approach, GIS-based probability surface maps were generated to illustrate predictions for sites and siteless locations in unsurveyed areas covering from 0.6 to 1 and 1 to 25 ha. The accuracy of the models was tested independently using site-file and non-site data, as well as split sampling techniques. Kvamme's approach to predictive locational modeling is statistically and computationally one of the more sophisticated attempts at spatial analysis using a GIS.
27


The Pinyon Canyon Project
The Pinyon Canyon Project in southeastern Colorado focused on the investigation of prehistoric settlement patterns within this high plains setting. The project was funded by the U.S. Army. The goal was to develop models of prehistoric site location (Kvamme 1984). For the study, a GIS was established for the entire region, which encompasses nearly 1,000 sq.km.
The computer data base used a cell-based GIS of more than 400,000 cells, each 50 m on a side, and more than 20 analytical and management surfaces or themes. The system includes environmental surfaces representing elevation, slope, aspect, measures of local relief, a measure of relative view quality, vantage locations and hydrology network, and horizontal and vertical distances to streams of Strahler order ranks. Management surfaces include the locations of nearly 1,200 archaeological sites, with information on site number, site type, temporal period association, and surfaces depicting field inspected regions, including date of inspection and several management boundaries.
The above GIS could retrieve sources and combinations of management and environmental data, such as archaeological information about a site and its environmental properties or scaled maps of any surface or combination of surfaces. One of the chief uses of the geographic data bases in all of the above studies is to examine and test environmental hypotheses about archaeological site locations and to develop settlement pattern models, including models used for site density projection.
28


CHAPTER III
EXISTING ENVIRONMENT
Chapter III describes the physiographic and general environmental setting of the project area. Many of the environmental factors discussed in this chapter serve as the primary surfaces for spatial manipulation in the GIS.
r
Location and Physiographic Setting
The four-quadrangle study area (Pretty Rock, Tanner Lake, Alamo Mesa East, Alamo Mesa West) lies in the northern part of the San Juan Basin portion of the Colorado Plateau (Figure 3.1). The central part of the basin is elliptical, 160 km north to south and 145 km east to west, and contains 19,425 sq. km (Fassett and Hinds 1971). The project area lies on the west side of the central basin on the Chaco Plateau. The Chaco Plateau is the area north and east of the Chaco River, south of the San Juan River, and west of Canon Largo (Warren 1967a).
This portion of the Colorado Plateau consists of "high plateaus and mesas interspersed with broad basins and intervening valleys and arroyos containing ephemeral streams" (Metric Corp. 1982). The area has a gently sloping upland erosional surface, partially covered by sheet sand and dunes, with badlands exposed where drainages have cut through the upland surface. The elevation ranges from 1,730 m to 2,010 m. Ephemeral streams within the area include Coal Creek, which crosscuts the northeastern and north-central sections and runs into De-na-zin Wash. De-na-zin Wash and Alamo Wash drain the north and northwestern portions
29


FIGURE 3.1 Location Map
30


of the study area, respectively, and join 1.7 km west of the project area. De-na-zin Wash then continues about 30 km to the west. To the northeast lies Hunter Wash, which drains the northwest portion of the project area and is a tributary to the Chaco River to the west. The southern portion of the area is drained by Tsaya Arroyo and Ah-shi-sie-pah Wash, which are tributary to the southern portion of the Chaco Wash. Other features within the area are exposed badlands in its east-central and north-western sections; red dog (burned shale) hills in the southeast and west; Tanner Lake in the west; and the end of Alamo Mesa, which projects into the central part of the study area.
Climate
The project area has a semiarid climate (Fern'll 1978). The mean annual temperature is 10°C, with winter low temperatures below 6°C and summer highs above 32°C. Mean annual precipitation is 25 cm with 50 percent of the precipitation occurring in thunderstorms between July and October (Berger and Lucas, n.d.). Rainfall is variable within the region. Chaco Canyon records show from 1933 to 1979 a range from 8.2 to 4.8 cm (Gutierrez 1980). June and November are the driest months, with the highest precipitation occurring in the summer, less in the fall, and the lowest in the winter. The project area has from 160 to 180 frost-free days (Ferrill 1978).
Hydro!ogy
The project area is in the San Juan River watershed (Dingham 1978) and contains six ephemeral streams: Coal Creek, De-na-zin Wash, Alamo Wash, Hunter Wash, Ah-shi-sie-pah Wash, and Tsaya Arroyo. The only major drainage basin is De-na-zin Wash. Black Lake, located to the southeast,
31


is a dry Holocene lake. Tanner Lake, another dry lake, may have been modified during historical times.
Geology and Geomorphology
The San Juan is an asymmetrical structural basin containing 4,572 m of sedimentary deposits ranging in age from Cambrian to Holocene. The Cretaceous (120 to 65 million years ago) geologic units within the study area are, from oldest to youngest, Cliff House Sandstone, Lewis Shale, Pictured Cliffs Sandstone, and the Fruitland and Kirtland Formations. Tertiary (65 to 3 million years ago) formations have been eroded from the project area.
The Cliff House is a thick (495 m to 990 m), transgressive strand-line sandstone. The units represent long stands of the beach line between the marine Lewis Shale and the continental Menefee Formation. In many places the Cliff House Sandstone conformably contacts both of these units.
The Lewis Shale unit is the highest marine shale in the basin and has marine shell fragments including gastropods and ammonites (Marshall and Breed 1974). The dark shale contains scattered interbeds of bentonite, marine sandstone, and calcareous silty concretions with veins of anhedral barite. The Lewis Shale is easily eroded into badlands that are often covered with gypsum crystals and crusts of soluble salts. Along the Chaco River, the Lewis Shale has been redeposited as slopes or terraces with low gradients.
The Pictured Cliffs Sandstone was defined from outcrops north of the San Juan River and west of Fruitland. This sandstone forms cliffs in some parts of the study area. Lithologically, the Pictured Cliffs can be
32


divided into two units: an upper massive sandstone and a lower shaley unit that grades into the Lewis Shale. The sandstone beds are medium to fine grained, well sorted, and composed of 8 percent quartzes, 13 percent feldspars, and 4 percent coal fragments (Burgener 1953). Fossils included in this formation include shark teeth and fragments of teleosts, turtles, and dinosaurs.
The Fruitland and Kirtland formations have been defined by Bauer (1916). The Fruitland Formation is the coal-bearing unit in this portion of the San Juan Basin, and the Fruitland/Kirtland contact is gradational and arbitrary. The contact is commonly defined by the limit of the uppermost Fruitland coal deposits. The two formations are similar in flora and fauna. Further work is needed to accurately distinguish the conformable contact in surface expression (Marshall and Breed 1974).
The Fruitland is lithologically more complex than the Kirtland and includes 990 to 1159 m of interbedded sandstone, siltstone, shale, clays, carbonaceous shale, carbonaceous siltstone, coal, and thin limestone beds composed entirely of brackish-water pelecypod shells. All beds within the Fruitland are discontinuous and pinch out laterally (Baltz 1967).
The Fruitland commonly has a sequence of thin coal seams overlain by fine laminated clays and sandstone and weathers to a badland condition.
In some areas, the Fruitland is marked by low hills of baked shale which has been fired to a porcelain-like material by spontaneous oxidation of underlying coals. In a few areas, this porcelain material is of suitable quality to have been used prehistorically for chipped stone tools.
The Kirtland Shale is divided into two or three units and varies greatly in thickness throughout the San Juan Basin. The lower and upper shales are often divided by a sandstone layer that is medium to fine
33


grained. The upper unit of the Kirtland is thin or missing in some areas but is conformable with the overlying Ojo Alamo where contact can be traced (O'Sullivan and others 1972).
The fossil inventory from the Fruitland and Kirtland formations is extensive, including fossil wood, leaf impressions, fresh water mollusks (11 species), and many microvertebrate and macrovertebrate remains. Turtles (nine species) and the hadrosaurian dinosaur (Kirtosaurus navajovnus), the most abundant large vertebrates, are typical of late Cretaceous marginal marine assemblages.
The Quaternary (3 million years ago) deposits include aeolian soils. The following discussion is based on Schultz's (1980) work on the Chaco dune field. This dune field is bound by Brumhall Wash to the north, the Chaco River to the west and south, and De-na-zin and Alamo washes to the south and east. The northwest part of the study area (north of Alamo Wash) lies within the Chaco dune field, and the remainder of the area is covered by Quaternary alluvial deposits associated with the De-na-zin and Alamo wash systems.
Any of five aeolian landforms may occur. Sand sheets, which are "usually thin, well-laminated, gently undulating sand bodies that form the flat topography" (Love and Schultz 1980:30), constitute the major landform within the project area. Barchan dunes, which have a "crescentic form with the horns of the crescent extending downwind" (Schultz 1980:30), are active dunes that are restricted to a 5 km zone downwind from the Chaco River (Schultz 1980:30). No examples of this landform exist in the study area. Parabolic dunes, which have a crescent shape similar to that of barchan dunes but with the horns pointing toward the dominant wind direction (Love and Schultz 1980:30), occur in the
34


northwest and east sections of the project area. The south landform, linear dunes, consists of "aeolian deposits that are elongated parallel or nearly parallel to the dominant wind direction" (Schultz 1980:39). Although there are dunes within portions of the surveyed areas with a linear form, these are actually the detached arms of parabolic dunes. Finally, ridge dunes are "accumulations of wind-blown sand that form large irregular mounds along the dissected edges of upland surfaces" (Schultz 1980:46-48), for example along badlands.
In the central portion of the four quadrangle study area, the dunes are generally oriented southwest to northeast, parallel to the dominant wind direction. This pattern has been observed by Hack (1941) and by Cooley and others (1969) across the Navajo Reservation, as well as by Schultz (1980). Schultz also found that aeolian landforms within the Chaco dune field represented 88 percent sand sheet, 8 percent parabolic dunes, and 4 percent other forms. Coppice dunes, "mounds of wind-blown sand that form around clumps of vegetation" (Schultz 1980:50), occur on sand sheets and along the crests of linear and parabolic dunes, creating a humocky appearance in sand deposits observed in isolated locations in the study area.
The aeolian deposits are generally considered to be of late Pleistocene and Holocene age (Cooley and others 1969; Hack 1941; Hall 1979). Schultz (1980) and Wells (1982) have defined three major periods of aeolian deposition: Aeolian Unit 1 dates from the late Pleistocene (7,000 years ago); Aeolian Unit 2 dates from 2,800 to 7,000 years ago; and Aeolian Unit 3 dates from 1,500 years ago to the present. Wells (1982:136-138) notes that Archaic sites do not occur in Unit 1 but are present within Unit 2. The parabolic dunes are the oldest landform,
35


dating from late Pleistocene to recent Holocene, while linear and barchan dunes originate in the recent Holocene. Studies that correlate aeolian depositional units with cultural-temporal units include Hall (1979) and Wells (1982). The sources of the sand for the Chaco dune field are Upper Cretaceous sandstone outcrops (the dominant source since the Pleistocene) and sediments of the Chaco River (Schultz 1980). The sandstone is a more important source within the survey area, but some local sand from washes and sand sheets is also reworked and redeposited (Love and Schultz 1980).
Local Lithic Raw Materials
The Chaco Plateau contains terrace and stream gravel, which is a major source of such lithic materials as quartzites, cherts, sandstone, and igneous cobbles (Warren 1967b:118). Chapman (1977:429), who used Warren's lithic type codes in a later analysis, describes the lag gravels in the lower Chaco River area as "waterworn nodules and cobbles of silicified woods, cherts, chalcedonies, and quartzites."
Surface gravel deposits have been found at several locations in the study area: along eroded drainages; along the eroded edges of dunes at dune-badland contacts, where dunes were cut by drainages; in interdunal areas on eroded/deflated hardpans and playas; and on exposed badlands.
The first four locations contain surface gravel/scatters of varying sizes. Gravels on exposed badlands occur in isolated pockets. Overall, gravel deposits appear to lie above the badland formations and below the aeolian deposits. The original source of these surface lag gravels is probably the Ojo Alamo Formation, where they were reworked and deposited by alluvial action during the Pleistocene (Vierra and others 1986).
Lithic raw materials within the project area include chert, siliceous and
36


nonsiliceous petrified woods, chalcedonies, quartzites, sandstone, igneous rocks, and claystone (Vierra and others 1986).
Soi 1 s
Most of this discussion on soils is based on the soils inventory performed by the Soil Conservation Service (SCS) and documented in Soil Survey of San Juan County New Mexico, Eastern Part (1980). Soils are natural bodies at the surface of the earth that support or are capable of supporting plants (Daugherty and Buchanan 1981:3). Soil formation is a function of climate and biota, parent material, topography, and time.
Most soils within the study area formed from alluvium and aeolian sediments derived from shale and related sandstones. The soils are very young, having formed in the late Pleistocene and Holocene eras. The soil types and associated map symbols are shown in Table 3.1. These soil mapping units are the result of the SCS Order 3 range survey completed at the series level of mapping precision.
FI ora
Much of the project area lies in a heterogenous region of the Great Basin sagebrush shrublands in the grama/galleta (Bouteloua-Hilaria) steep association (Morain and others 1977). The main vegetation types in the study area and their associated landforms are shown in Table 3.2. In of this study, all the vegetation units are described by characteristic vegetation units associated with their respective SCS soil mapping unit names (Table 3.1).
37


TABLE 3.1
SOIL TYPES AND ASSOCIATED RANGE SITE DATA FOR THE PROJECT AREA
Soil Name/Symbol Range-Site Characteristic Vegetation Physiographic Setting X Slope Elevation (m)
Badland/BA — barren shale uplands with deep drainages and gulleys 5-80 1,455-2,182
Blancot Notal Assoc. Gently Sloping/BT Loamy ND-1 Big sagebrush Galleta, Indian ricegrass, Fourwing saltbush Western wheatgrass (Blancot) fans and upland (Notal) fans and valley bottoms 0-5 1,697-1,939
Doak Loam/DC Loamy ND-1 Blue grama, Indian ricegrass, Big Sagebrush, Needleand-thread, Western wheatgrass Galleta deep and well-drained on mesas, plateaus, and terraces 3-5 1,697-1,939
Doak Avalon/DN Loamy ND-1 Big sagebrush, Needleand-thread, Blue grama, Western wheatgrass Galleta Indian ricegrass mesas, plateaus, and terraces 0-5 1,697-1,939
Doak-Sheppard Shiprock Assoc., Rolling/DS Loamy ND-1 Big sagebrush, Needleand-thread, Blue grama Western wheatgrass Galleta, Indian ricegrass mesas, plateaus, and terraces 0-15 1,697-1,939


TABLE 3.1 (continued)
SOIL TYPES AND ASSOCIATED RANGE SITE DATA FOR THE PROJECT AREA
Soil Name/Symbol Range-Site Characteristic Vegetation Physiographic Setting Z Slope Elevation (m)
Duneland/DZ Scattered vegetation mainly In lnterdunal areas mesas, plateaus, major dralnageways, deep excessively drained active dunes 5-25 1,455-2,182
Frultland-Persayo-Sheppard Complex Hllly/FX Sandy ND-1 Indian rlcegrass Blue grama, Big sagebrush Fourwing saltbush Giant dropseed hills, mesas, plateaus, fans, and breaks 5-30 1,455-2,182
Huerfano-Muff-Uffens Complex/HU Sodic Slope Alkali sacaton, Fourwing saltbush, Galleta, Polack greasewood mesas and valleys 0-8 1,697-1,939
Rlverwash/RA Little or no vegetation due to frequent flooding and reworking by fluvial action unstablized sandy, silty, clayey, or gravelly sediments on flood plains, streambeds, riverbeds, and arroyos 0-3 1,455-2,182
Rock Outcrop/RO No vegetation exposures of barren sandstone on cliffs, breaks, bluffs, and ridges 5-100 1,455-2,182


TABLE 3.1 (continued)
SOIL TYPES AND ASSOCIATED RANGE SITE DATA FOR THE PROJECT AREA
Soil Name/Symbol Range-Site Characteristic Vegetation Physiographic Setting X Slope Elevation (m)
Sheppard Huerfano Notal Complex Gently Sloping/SC Deep Sand ND-1 Indian ricegrass, Giant dropseed, Alkali sakaton, Needleandthread valley bottoms, fans, mesas, plateaus 0-8 1,455-2,182
Shlprock Fine Sandy Loam/SO Sandy ND-1 Indian ricegrass, Blue grama, Big sagebrush, Giant dropseed, Fourwlng saltbush mesas and plateaus 2-5 1,697-1,939
Sheppard Mayqueen Shlprock Complex/SD Deep Sand ND-1 Indian ricegrass, Giant dropseed, Alkali sakaton, Mormon tea, Sand dropseed mesas and plateaus in loamy fine sand 0-8 1,697-1,939
Source: Soil Survey of San Juan County,
New Mexico: Eastern Part 1980~ SCS (1980)


TABLE 3.2
MAJOR VEGETATION UNITS AND ASSOCIATED LAND FORMS
Landform Rice grass/ Gall eta grass Rice grass/ Big sagebrush Big sagebrush/ Blue grama Shadscale/mixed saltbrush-alakali sakaton
Plains X X X
Badlands X X
Erosional Areas X X X
Hill X X X
Landforms:
Plains - nearly level surfaces with some undulations or dissections
Badlands - high density of gullies, ravines and sharp-backed ridges; usually clay or shale
Erosional - nearly devoid of vegetation with
Areas alkaline or saline soils
Hill - moderately steep to steep slopes
and somewhat rounded relief of less than 151 meters
Source: Rangeland Resources International, Inc. (1978)
41


Fauna
Modern animal populations appear to be small. The birds present include horned larks, Says phoebes, loggerhead, shrikes, scaled quail, common raven, mountain bluebird, and various hawks. Reptiles include various lizards and rattlesnakes. The mammals present include the prairie dog, cottontail rabbit, jackrabbit, kit fox, and coyote. Parts of the study area are used for grazing cattle and sheep (Berger and Lucas 1972).
42


CHAPTER IV
THE ARCHAIC PERIOD
The Archaic period is unique within the culture history of the San Juan Basin. The hunter-gatherer mode of subsistence of this period persisted for about 6,000 years--the longest adaptation in the Basin's history. This chapter focuses on previous research of the Archaic in the Basin and ethnographic examples of hunter-gatherer systems with organizational strategies similar to those postulated for the Archaic. Since the Archaic is the best represented cultural period in the study area, the Archaic site data base will be used for identifying land use and cultural resource conflict areas with GIS technology.
The prehistory of the study area can be divided into five basic cultural/temporal periods: the Paleo Indian (9,500 BC to 5,000 BC) the Archaic (5,500 BC to AD 400) the Anasazi (AD 1,000 to AD 1,300) the Historic Navajo (AD 1,500 to 1930) and the Anglo (1846 to present). Only the Archaic period is discussed here.
Overview of the Archaic
The following summary of the Archaic period is based on the information from a recent study of a portion of the study area by Vierra and others (1986).
The Archaic culture has been defined as a "broadly based mixed gathering and hunting economy" (Irwin-Williams, 1973:4) and as groups that manipulated higher life zones and conducted intensified food production (Stuart and Gauthier 1981:33). It was a post-Pleistocene
43


adaption. Lee and Devore (1968) describe general hunter-gatherer systems as small flexible groups of 25 to 30 individuals and a population density of 1 to 25 people per 260 sq. km. Each of these small groups is associated with a geographical region but interacts with other groups, creating a large breeding and linguistic community. The economic system includes such features as a home base, a sexual division of labor (with men hunting and women gathering), and a pattern of sharing of food and other goods.
The term "Archaic" is sometimes used to denote a specific cultural/temporal unit; at other times it describes a general cultural adaptation. This discussion deals with the Archaic as a cultural/temporal unit. It touches on the transition from Paleo Indian to Archaic, reviews the Oshara chronology from a regional perspective, and considers Archaic settlement-subsistence system and site organization studies, with an emphasis on site reoccupation. Finally, the discussion considers the transition from mobile hunter-gatherer to sedentary agriculturalist and the coexistence of hunter-gatherer and agricultural strategies.
In northwest New Mexico the Archaic has been defined as a hunter-gatherer adaptation, the Oshara tradition, which lasted from about 550 BC to AD 400 (Irwin-William 1973). Irwin-William (1973) considers the Archaic hunter-gatherers to be predecessors of the Anasazi. Although her work centered in the Arroyo Cuervo area, southwest of San Ysidro, New Mexico, projectile points similar to those in the Arroyo Cuervo region occur throughout the San Juan Basin. Irwin-Wil1iams states that Archaic
44


assemblages "differ so greatly in technology, typology, and functional classes from those of the preceding Cody and other Paleo Indian phases, that there is evidently no generic connection between them" (1973:4-5). Because of similarities between Lake Mojave Complex of southern California and the Jay Phase, she suggests that the beginnings of the Oshara tradition represent the eastward movement of western-based Archaic groups into northwest New Mexico at about 6,000 to 5,000 BC and that these groups occupied the area abandoned by the plains-based Paleo Indians. This interpretation has recently been questioned by Stuart and Gauthier (1981:33), who argue that the Archaic may be an indigeneous development from the Paleo Indian.
The Oshara tradition provides the Archaic period with a chronological framework, in part based on projectile point typology. This tradition has been divided into five phases: Jay (5,500 to 4,800 BC), Bajada (4,800 to 3,200 BC), San Jose (3,200 to 1,800 BC), Armijo (1,800 to 800 BC) and En Medio/ Basketmaker II (800 BC - AD 400). Irwin-Williams1 (1973) generalizations concerning the Arroyo Cuervo area, however, can be considered representative of only a portion of the regional Archaic settlement-subsistence system.
The notion of a regional approach to the study of past cultural systems was advocated by Lewis Binford in 1964. But few investigators have yet applied this approach to settlement-subsistence systems in the San Juan Basin. Several classes of data could be used in developing such a regional perspective, including projectile point distributions, intrusive material types, ethnographic information, and subsistence data.
Hunter-gatherer groups tend to interact across a broad geographic region, allowing individual groups to collect and exchange information on
45


the location and seasonal availability of resources. This interaction could be indicated by the distribution of projectile point styles.
Oshara tradition points are distributed throughout the San Juan Basin and in the adjacent areas of southeast Utah and southwest Colorado, the Jicarilla Reservation, the Jemez Mountains, Cebolleta Mesa, and in the Defiance Plateau and Chuska Mountains.
Binford (1979:259) stated that "raw materials used in the manufacture of implements are normally obtained incidentally to the execution of basic subsistance tasks. In other words, procurement of raw materials is embedded in basic subsistence schedules." Given this perspective, the intrusive materials in an area can provide clues to a group's seasonal movements and to their settlement-subsistence region (Clark 1976:43;
Issac 1977:94-95). Intrusive materials on sites near the project area include Jemez and Polvadera obsidian (Chapman 1977; R. Moore 1982;
Gomolak 1980) and Cerro Pedernal chert (Acklen and others 1982; P. Moore 1982).
The ethnographic record is replete with documentation of the movement of hunter-gatherer groups between higher and lower elevations to increase seasonal availability of resources (Krober 1925; Powers 1877; Steward 1938). Steward (1938:52) noted that the Shoshonean territory "embraced the various life zones, thus providing all possible local varieties of essential foods."
It is not surprising, therefore, that the intrusive lithic materials found in the project area appear to have been procured when local groups were exploiting resources at higher elevations. These uplands were the focus of a portion of a regional settlement-subsistence system. Lowlands
46


such as the project area were the focus of activities for other portions of the system.
Other researchers have discussed the recent use of both higher and lower elevation areas. Irwin-Williams (1973) reported that "two kinds of special activity sites are known outside the Arroyo Cuervo region: isolated hunting camps in the Jemez Mountains and repeated quarry workshop camps." More recently, Stewart and Gauthier (1981:407) suggested that "Archaic sites...in higher elevation(s) contain more diagnostic points than do lithic scatters in many of the extensive dune settings. It may be that two kinds of site locations reflect the collecting versus the hunting strategy, and are seasonably distinct assemblage sets of the same population."
The limited evidence of Archaic use of plants in the northern San Juan Basin has been summarized by Toll and Cully (1983). Their information, which was derived from the Navajo Mine Archaeological Program (NMAP) and the Navaho Indian Investigation Project (NIIP), suggests the exploitation of Indian ricegrass and mustard in the late spring/early summary and the exploitation of dropseed, goosefoot, and pigweed in the mid-summer/early fall. They have developed a settlement-subsistence model of Archaic seasonal movement involving summer occupation of lower elevation dunes (based on archaeological evidence), with movement into high elevations during the fall for pinyon collecting and possible winter residence (Steward 1938).
The Archaic settlement-subsistence system has been viewed in various ways. Winters (1969:110) defined a settlement pattern as "the geographic and physiographic relationship of a contemporaneous group of sites within a single culture" and a settlement system as "the functional relationship
47


among the sites contained within the settlement pattern." In other words, settlement patterns are studied on the intersite level, and settlement systems are studied on the intrasite level.
Studies of settlement patterns test ecological and locational hypotheses about the causal variables affecting the selection of site locations. Studies of Archaic site locations have stressed three environmental variables: (1) vegetation diversity (Elyea and others 1979; Sessions 1979; Reher and Witter 1977); (2) vegetation diversity and proximity to water (Elyea and others 1979; Ganas 1980; J. Moore 1980; Powers 1979); and (3) aeolian soils (Ganas 1980; Huse and others 1978; Powers 1979; Reher and Witter 1977; Sessions 1979). Kemrer (1982:103) summarized these three factors when he observed that locations of lithic sites seem to correlate with soils that largely support grasses and with areas where seed grasses are differentially distributed. He noted that lithic sites are most likely to occur in areas with a mixture of alluvial and aeolian soils.
Reher's (1977) study of the lower Chaco River presents an early view of the Archaic settlement system in the northern San Juan Basin. Reher suggested that this system was composed of "two basic types: campsites (i.e., habitation), and several kinds of limited activity sites where various plant and animal resources were procured and sometimes processed in areas surrounding habitation sites" (1977:96). He further observed that the lithic assemblages form a continuum from large habitational sites (which show a high degree of functional variability and frequent presence of hearths) to small special use sites (with little functional variability and few or no hearths).
48


In a later study, Vierra (1980a) argued that the "continuum" actually represents only one part of the settlement system, the limited basecamp, and that the difference in size and content is mainly due to variations in the length and number of occupations of the site area" (1980:356). He went on to say that the only remains suggesting task-specific sites are isolated artifacts.
Although Kemrer (1982:19) suggested that large camps represent a seasonal macroband (i.e., a population composed of several microbands) phase, large sites are more likely to represent locales reoccupied by microbands (i.e., the small primary foraging units) and are not single occupational episodes representing microband encampments (i.e., large aggregations of microbands or central basecamps). This view of microband reoccupation has been supported by Eschman (1983) for the NMAP project area, which lies north of the present survey area.
Other researchers, however, have taken a different approach, defining multiple archeological site types on the basis of assemblage content and specific site function. Suggested types include (a) large/main camps,
(b) small camps (c) plant processing sites, (d) plant procurement sites, (e) hunting camps, and (f) lithic procurement sites (Kemrer 1982:18-23). Huse and others (1978), Powers (1979), Sessions (1979) and Ganas (1980) also provide examples of multiple site type definitions.
Little research has been conducted toward an understanding of Archaic group size, structure, and organization. Site structure studies can provide information relevant to these questions, but few such studies exist.
Vierra (1980b:385) described a limited pattern recognition study that defined four possible patterns in the organization and use of space on
49


Archaic sites: (a) an association among ground stone, fire-cracked cobbles, and hearths; (b) areas with the highest density of nonutilized chipped stone debris (debitage) also being the areas with the highest density of utilized debitage (which shows that debitage was produced, used, and discarded in the same location); (c) nonpatterned scatters representing refuse areas or areas affected by postdepositional processes; and (d) activity/refuse areas consisting of large patterned locales that included evidence of multiple activities. This fourth pattern is probably a result of reoccupation and reuse of site locations.
Vierra (1981) proposed a model for the organization of an Archaic camp. From his structural study of a Basketmaker II site, this model suggests that an Archaic camp should consist of (a) a hearth area that was the focus of camp activities and that exhibits the greatest amount of functional variability as indicated by lithic reduction debris and (b) a separate lithic reduction locus with some expedient tool use. See Chapman (1980) and Camilli (1979) for other examples of site structural studies.
Preventing an understanding of Archaic sites is the problem of identifying and interpreting reoccupied site locations. The multiple activity loci visible on many sites may reflect (a) separate but contemporaneous residential units, (b) contemporaneous but functionally differentiated activity locales, (c) multiple occupational episodes possibly involving the previous two patterns, or (d) differential surface exposure of all of the above patterns. Reoccupied sites are often identified by absolute or relative dating techniques, the existence of multiple stratigraphic levels, and the extensive site size with multiple activity locales. A suggested avenue of research would be to determine
50


the organization of the residential unit by trying to isolate redundant patterning in the archaeological record (Vierra 1980b, 1981). Once a redundant unit has been identified, this information can be used to interpret the structure of sites with multiple activity locales.
A major interest in modern anthropological theory involves the question of what causal variables led to the transition from mobile hunting and gathering to sedentary agriculturalist adaptation (Reed 1977). Some of the suggested causes include environmental change, climatic variability, and population pressure. The Archaic exhibits an intensification of the economy through time, which appears to be characteristic of the transition to agriculture (Cohen 1977 and Klein 1977). But the causal variables that led to this intensification have yet to be identified.
Recent work by Eschman (1982) in the NMAP project area suggests that a post-Archaic hunter-gatherer strategy coexisted with the Anasazi adaptation. This hypothesis is based on the association of four late radiocarbon dates (between AD 1280 and 1610) with assemblages that would otherwise be considered Archaic. Eschman proposed that initial Anasazi populations displaced Archaic groups in the NMAP area but that once the Anasazi abandoned this territory, the hunter-gatherer groups returned (Stuart and Gauthier 1981).
51


CHAPTER V
ASSESSING ARCHAEOLOGICAL INVENTORY AND ENVIRONMENTAL DATA BASES
This chapter focuses on the GIS-aided methodology used in evaluating the archaeological inventory and environmental data for potential biases. Administrative themes (areas of planned and existing development or conservation) are also analyzed according to the amount of inventory conducted for them.
Chapter V provides the rationale for selecting the four-quadrangle study area in the northern San Juan Basin. It also briefly describes the planned and existing development in the area, and discusses BLM inventory classes as they relate to the project area and BLM inventory criteria for the FRA. Next, Chapter V describes the problems of and potential biases involved in using existing site and inventory data. A description is then given of the specific environmental and administrative themes used for analysis using the GIS. The next section focuses on the procedures used to evaluate the project inventory data with the environmental and administrative data using MOSS/MAPS. A cartographic model is included to aid the reader in following the steps used for each analytical procedure. Finally, the results of the analysis are interpreted with the aid of tables and maps.
Selecting the Study Area
The four-quadrangle study area lies in the northern San Juan Basin of New Mexico, within the Farmington Resource Area (FRA). Since the BLM in New Mexico has a large digitized data base and the Data Synthesis
52


Project (Chapter I) has begun, the state of New Mexico was a logical choice for a cultural resource and land use study involving a GIS.
The reasons for choosing this particular study area are threefold. First, it lies in an area where multiple land use is planned or occurring. The extraction of coal and oil and gas, in particular, presents a source of possible conflict with cultural resources occurring nearby. One of the aims of this project is to evaluate the potential of using a GIS in a multiple-use setting where cultural resources are involved.
Second, existing contributory surveys (see below) provide an adequate sample for statistical analysis. Within the four-quadrangle project area, 49,308.7 acres (29.23%) had been inventoried. An inventory level of at least 20 percent was determined to be desirable for this study. In this way, the nature of potential inventory bias in existing data could be described. Even though most resource areas have inventory levels closer to 5 percent, the same or similar biases would be expected.
Third, most of environmental and administrative themes for the study area had previously been digitized at the BLM New Mexico State Office and the Albuquerque District Office. Because none of the one administrative themes and only three environmental themes had to be digitized or procured from an outside source, the cost of conducting the study was significantly reduced.
The use of existing data bases for archaeological analysis via a GIS is of major interest to cultural resource managers and land use planners, particularly because limited budgets often make it difficult to procure digitized maps.
53


Planned Development in the Study Area
The four-quadrangle project area of Pretty Rock and Tanner Lake to the south and Alamo Mesa East and Alamo Mesa West to the north has an extensive amount of proposed mineral development. The study area comprises lands of three different statuses relative to coal leasing in the northern San Juan Basin: (a) preference right lease applications; (b) industrial expressions of interest (proposed competitive tract leases); and (c) areas of existing but inactive leases.
There are also other types of proposed and existing development under lease. Oil and gas leasing tends to be restricted to the northern half of the study area. Range allotments are found in smaller portions of the area. A few unimproved dirt roads can be found as well as pipe and transmission lines. The study area also contains two wilderness areas and two proposed areas of critical environmental concern (ACEC).
BLM Inventory Levels
The BLM inventory levels are described in Chapter I. Most of the surveys in the project area were conducted at the Class III level as intensive survey units selected as part of Class II surveys in and near the project area. This chapter presents the results of the analysis of Class II and Class III inventory data using a CIS to address Class II inventory objectives. These objectives are given in the BLM cultural resource manual; (BLM 8100-CRM Manual, 1978).
1. Discovery, recognition, or elaboration of patterns of past human use and occupation of given regions.
2. Determination of the cultural resource potential of an inventory area.
54


3. Prediction of zones of greater or lesser activity by past human populations.
4. Identification and assessment of the environmental and/or cultural variables, or combination of variables, that form the most accurate predictors of cultural resources.
5. Development of projections of expected density distribution and diversity of cultural resources.
6. Discovery of the range of cultural resource variability within an inventory area.
7. Provision of an objective means of assessing the existing cultural resource inventory.
8. Development of a systematic base for planning decisions concerning cultural resources.
The analysis of the existing inventory data should eventually aid in meeting most of these Class II objectives. Specifically, the inventory base has been analyzed to determine areas where future surveys need to be concentrated to provide more representative samples of cultural site and environmental data in the project area. In this manner, one can assess and measure the biases associated with particular environmental character!'sties resulting from non-random survey units.
Many intensive surveys have been conducted in and near the study area, especially for "clearing" oil and gas drill pads, pipelines, and access roads. The clearance of an area slated for development is required to meet federal compliance requirements. In addition, portions of coal tracts and areas for coal lease applications and expressions of interest have also been inventoried. The major surveys and portions of major surveys occurring within the study area include projects by Huse and others (1978), Chapman and Biella (1980), Vogler and others (1982), Harlan (1982), Vierra and others (1986), and various BLM intensive surveys.
55


The FRA and BLM Albuquerque District Office have established six criteria for determining whether archaeological inventory reports are contributing or not contributing valid survey information. This study used only digitized contributing inventory information. The following types of reports are considered non-contributing in the FRA:
1. Known field operations of a cultural contractor did not yield contributing information. For example, a pre-1982 operation of San Juan College consisted of moving planned locations of gas well/ pipeline to avoid cultural resources. This procedure resulted in a lower density of cultural evidence in the records than actually exists.
2. The description in reports is inadequate for field methods used. Field methods not acceptable to BLM include survey by vehicle, aerial surveying, and transects too far apart for terrain and site density.
3. Inventories were conducted by unauthorized individuals or groups, e.g., geological survey or contractors with expired permits.
4 Excessive actions were inventoried during each day.
5. Inadequate observation fails to identify cultural resources present as verified by a BLM field check.
6. Reports mention the presence of artifact assemblages but do not record these as sites.
Problems and Biases in Using Existing Site and Inventory Data When examining existing archaeological data bases within a region, data often greatly varies. This variation stems from several factors, ranging from differing standards of quality or practice between different archaeologists, to changes through time in accepted field practices, to variability in the goals and research plans of individual survey projects. The following is a general discussion of potential sources of
56


bias and variations in data quality that has been summarized from Kvamme (1985).
Variation, bias, and inconsistency in existing site data bases result mainly from the ways in which different field projects, archaeologists, and field crews perform fieldwork and define, identify, and record archaeological sites. Operational problems can result in defining sites from diffuse scatters of artifacts. For example, lithic sites defined by one project may not constitute sites according to another project's definition.
Inconsistencies in regional data bases result not only from the lack of standard archaeological procedures, such as field methods and operational definitions of sites, but from differences in research goals from project to project. For example, one project might focus its inventory methodology on addressing research goals dealing with hunting and gathering adaptations, whereas another project (in the same area) might be more concerned in inventorying cultural sites that pertain to sedentary agriculturalists.
Even within a single project, the potential for introducing bias and inconsistencies exists. Sites might, in practice, be defined differently owing to differences in the quality of individual field personnel and crews or because of other factors such as adverse weather, rough terrain, or a fatigued survey crew.
Budgetary constraints can also influence the quality of data collected. For example, when a contractor with a fixed price contract finds site densities to be greater than expected, the survey may be rushed. Schiffer and Wells (1982:346) note that "this is accomplished by
57


increasing crew spacing or reducing the recording time". These practices lower the quality of the resulting data.
Schiffer and Wells (1982) summarized several factors influencing archaeological inventory data quality. A main factor is survey intensity or crew spacing. Crew spacing not only affects site discovery rates but also the sizes of discovered sites (Plog and others 1978). Crews tend to miss small sites and cultural features when crew spacing is large (Wandsrider and others 1984) and small sites are not necessarily less important than larger ones. Narrow spacing dramatically increases survey time and effort, and also increases costs. It should be remembered that small sites are not necessarily less important than larger ones.
Given a specified level of survey intensity (Schiffer and others 1978), the nature or obtrusiveness of the archaeological evidence determines the likelihood that a particular archaeological feature, such as a site or an artifact, will be discovered. An architectural feature, for example, has a higher chance of discovery than a single, isolated flake. Low intensity surveys (those with wide spacing) tend to bias resulting archaeological samples in favor of more obtrusive remains (Schiffer and Wells 1982).
Another source of bias that affects the obtrusiveness of cultural remains involves geologic processes. Although the pattern of major landform and drainages in an area may not have changed since prehistoric times, alluvial material may have been deposited. Deeply buried sites in the areas of deposition will bias the sample away from locations in valley floors. Erosion, on the other hand, may destroy sites on steep slopes or along meandering streams, thus biasing the sample away from steep slopes and drainage locations.
58


Modern land use can also influence site visibility. Cultural materials are often exposed in areas of cultivation or surface disturbance. In nondeveloped areas such as remote arid regions, good site visibility is sometimes lacking because of natural processes that tend to obscure sites.
Difficulty of access, a common problem in many regions of the country, might mean that samples are biased against difficult-to-reach regions. Surveys are commonly located within a specified distance from access roads. Even when hard-to-reach places can be reached, the small amount of time left in the day after travel might lower the quality of surveys in those regions.
Private land ownership presents similar difficulties when landowners refuse access (Schiffer and Gumerman 1977:187). Where most archaeological survey work tends to be conducted on federal or state lands, the lack of comparable site data from private properties presents a severe source of bias to regional archaeological data bases. Private property often includes some of the best farmlands and the best areas for hunting and plant collecting.
Variable archaeological visibility, due mainly to plant cover, introduces another major source of potential bias. Land cover, however, is not a major problem in most arid regions of the country.
One of the main weaknesses of using existing data is that the sum total of previous work in a given region constitutes an unplanned effort. In other words, strong locational biases typically exist in the areas that have been field inspected within a region. For example, early work was often conducted only at the most accessible and visible sites, whereas much contemporary survey is conducted in areas of planned
59


development. Thus, existing site data may be strongly biased toward certain types of settings and may not constitute a representative sample of sites within a region.
Sites also might be inaccurately located on maps. For modeling approaches that focus on the specific locations of sites, accurate placement of sites on maps is of critical importance. Characteristics of the actual locations, such as environmental properties, are often used as a basis for modeling. It is often hard to locate oneself precisely in the field, particularly in areas with few nearby landmarks. Field crews sometimes get lost or misread maps.
Moreover, early archaeological surveys often did not have access to good maps and offered only verbal descriptions, directions, and rough locational sketch maps. This problem is further compounded as site locations are transferred from map to map.
Data Themes
For this report, a theme is defined as the name of a specific environmental or administrative map layer used in analysis. On the basis of its characteristics, each feature or cell in a map is assigned an identifier called a subject. For example, a soil thematic map could consist of several soil types. Each soil type would be considered a subject for the soils theme (see Figure 5.1).
Figure 5.1 also shows the environmental and administrative themes used in the assessment of the archaeological inventory data. Each theme is briefly discussed below. See Chapter VIII for a discussion of problems of acquiring and preparing some of the data sets for analysis.
The elevation map was derived from 1:24,000-scale, 7.5 minute digital elevation models (DEMs) at 30 m X 30 m cell size. The DEMs were
60


DATA THEMES
A THEME IS THE NAME OF A SPECIFIC ENVIRONMENTAL OR ADMINISTRATIVE MAP LAYER USED IN ANALYSIS. ON THE BASIS OF ITS CHARACTERISTICS, EACH FEATURE OR CELL IN A MAP IS ASSIGNED AN IDENTIFIER CALLED A SUBJECT. A SOIL THEMATIC MAP COULD CONSIST OF SEVERAL SOIL TYPES OR ASSOCIATIONS; EACH SOIL TYPE WOULD BE A SUBJECT.
SOME THEMES DO NOT HAVE DIGITIZED DATA FOR ALL FOUR QUADS IN THE STUDY AREA.
ENVIRONMENTAL AND ADMINISTRATIVE THEMES USED TO ASSESS THE ARCHEOLOGICAL INVENTORY DATA:
ENVIRONMENTAL
DRAINAGES
STREAMS
GEOLOGIC SURFICIAL PROCESS UNITS SURFACE GEOLOGY SOILS
* ELEVATION
‘slope
‘aspect
ADMINISTRATIVE WILDERNESS STUDY AREAS (WSA)
LINEAR DEVELOPMENT IN WSA ROAD NETWORKS RANGE ALLOTMENTS OIL AND GAS LEASES PREFERENCE RIGHT LEASE APPLICATIONS FOR COAL INDUSTRY EXPRESSIONS
OF INTEREST IN COAL LEASING PROPOSED AREAS OF CRITICAL ENVIRONMENTAL CONCERN LAND STATUS (OWNERSHIP)
* ELEVATION WAS DERIVED FROM DIGITAL ELEVATION MODELS (DEMS). SLOPE AND ASPECT WERE DERIVED FROM ELEVATION MAPS USING SLOPE AND ASPECT COMMANDS IN MAPS, RESPECTIVELY.
FIGURE 5.1 Data Themes used for analysis
61


procured from the United States Geological Survey (USGS).
The DEM data was processed through the Interactive Digital Image Processing System (IDIMS) at the BLM Service Center in Denver. See Chapter II for a discussion of IDIMS.
Since the DEM data is inherently raster in format, the MOSS cell processing subsystem, MAPS, is the main tool for the manipulating digital elevation data.
The SLOPE command in MAPS was used to access elevation data, i.e., a
7.5 minute DEM cell map, for computing slope (in percent rise over run) to create a new slope map.
The ASPECT command in MAPS can also access elevation data and compute azimuthal aspect or direction of surface slope to create a new aspect map.
The geologic surficial process units and surface geology themes were digitized from mylar maps created by graduate students of Dr. Steven Wells, a geomorphologist at the University of New Mexico. The geologic surficial process units theme shows polygon geologic mapping units that represent geologic processes affecting landscape morphology. The surface geology theme also has polygon mapping units, but these units show where stratigraphic units occur within portions of the study area.
The soils data was derived from a soils inventory performed for BLM by the Soil Conservation Service (SCS) and documented in Soil Survey of San Juan County, New Mexico, Eastern Part (1980). The soil types and associated map symbols are shown in Table 3.1. The soil mapping was performed at the SCS Order 3 range survey completed using the series level of mapping precision.
Four primary drainage basins were digitized as polygons for the
62


drainage theme. These drainages were delineated by major arroyos and tributaries that feed into common basins.
The stream data was digitized as line data. Several major streams are ranked according to Strahler's rank-order systems. Six orders of streams occur within the project area.
The soils, geologic surficial process units, surface geology, drainage, and stream data are on file in digital form at the BLM New Mexico State Office in Santa Fe.
The administrative themes used in assessing the inventory data include wilderness study areas (WSA), linear development in proposed WSAs, road networking, range allotments, oil and gas leases, preference right lease applications for coal, industry expressions of interest in coal leasing, proposed areas of critical environmental concern (ACEC), and land status (ownership) (see Figure 5.1). Data for all of these digitized themes are on file at the BLM Mew Mexico State Office.
The WSAs and the proposed ACECs were presented in the 1981 BLM Management Framework Plan (MFP) for the FRA. The plan had two WSAs, Bisti and De-na-zin, which fall partly in the badlands area. In addition, the plan contained data on two proposed ACECs, Bisti/De-na-zin and Fossil Forest.
In 1984 the two WSAs were designated by Congress as wilderness areas. The two proposed ACECs are still pending designation as ACECs by BLM. This GIS study used the digitized boundaries of the WSAs and proposed ACECs even though the WSA boundaries changed slightly after receiving wilderness designation (October 30, 1984). The two proposed ACEC boundaries are the same as included in to the MFP decisions of 1981. The proposed Fossil Forest ACEC is now encompassed within the
63


Fossil Forest Natural Research Area, which was also designated on October 30, 1984.
Preference right lease applications (PRLA) for coal show where applications have been made for coal leasing. This information will be used for developing action plans for coal development and for site-specific impact analysis. Industry expressions of interest in coal leasing contain data on tract delineations received from USGS. Most of the tracts are located in the "Chaco Stripable Coal Belt" near the Bisti Badlands. The coal tracts could be considered for competitive lease sales under appropriate market conditions. Finally, land status (ownership) contains five types of ownership for the study area: Indian, Indian withdrawn, state, BLM, and private.
Not all environmental and administrative themes have digital data for the entire four-quadrangle area. In some cases, data for a theme is missing either one or two of the quadrangles. In other cases, only portions of the four-quadrangles have been digitized.
The digitized data was mostly analyzed using the MAPS subpackage of MOSS, which is designed exclusively for raster (cell) format. MAPS was used for two reasons: (1) the DEMs already come in raster format, and other themes had to be in cell format for overlaying; (2) statistical analysis is performed more easily with cell data using statistical packages outside of MOSS/MAPS.
Procedures for Assessing Inventory Data Using MOSS/MAPS
The inventory data base was assessed to measure the biases associated with particular environmental features resulting from nonrandom survey units. In this way it is possible to determine where future surveys need to be concentrated to provide more representative
64


samples of cultural site and environmental data in the project area.
The preparation of the data base for analysis included the merging of the four 7.5 minute maps for each environmental and administrative theme. Each resulting map was converted to cell-format. The archaeological survey data was prepared in a similar fashion except that line and point data had to be converted to polygon data. After converting this inventory map to cell-format, all the map cells values showing inventoried areas were combined into one classification and assigned a new single value. Once this step was performed the new inventory map was overlaid with each of the environmental and administrative themes listed in Figure 5.1. Finally, area information was calculated for each thematic subject intersected with areas surveyed.
The remainder of this section shows the technical steps used in analyzing the inventory data base. This portion of the report will be of particular interest to those who are familiar with MOSS/MAPS. (All commands are capitalized. MOSS commands are underscored, whereas MAPS commands are not.) Figure 5.2 is a cartographic model showing the steps followed and their sequence. The circled numbers correspond to the respective numbered steps listed below. However, the numbering does not necessarily imply consecutiveness, since one step may not have to precede another. Note that the continuous data is data that are uninterrupted and has real number values. The elevation, slope, and aspect data sets are examples of continuous type data. Discrete data, on the other hand, has distinct units or classes represented by integer values, (whole numbers).
The SLOPE and ASPECT commands in MAPS were used to access continuous elevation data, i.e., 7.5 minute DEM cell maps, and the resulting maps
65


were continuous slope and aspect maps, respectively.
The EXTRACT command in MAPS is a data reclassification command that creates a new discrete map by assigning new values to the cell values of an existing discrete or continuous map. EXTRACT was used to convert the continuous slope and aspect maps to discrete maps.
1. MERGE four 7.5 minute quadrangles for each environmental or administrative data theme.
2. RASTERIZE to convert to cell format with background and nonbackground values. A 30 m X 30 m cell size was used to obtain the highest resolution possible. In this way more accurate area measurements could be derived.
3. SLOPE or ASPECT continuous elevation map to produce continuous slope and aspect maps.
4. EXTRACT only the continuous elevation, slope, and aspect maps to assign new values and convert to discrete data classes.
5. MERGE the archaeological inventory data. The BLM Albuquerque District Office has distributed all their inventory data from 1:100,000-scale maps. There are 32
7.5 minute quandrangles in a 1:100,000-scale map. Thus 32 maps with inventory data had to be merged by the District Office.
6. BUFFER archaeological point and line inventory data at 1:000,000-scale. All of the point and line inventory data shown on the 1:100,000-scale maps mentioned in step 5 was BUFFERED at the District Office. In this manner all survey data was converted to polygon data so that area measurements could be made.
7. GENERATE a WINDOW border using UTM coordinate pair data for the four-quadrangle study area. The GENERATE command allows for the creation of new maps or rectangular features such as a frame. The WINDOW was expanded beyond the edge of the map to create even multiples of the 30 m X 30 m cell size. In this way, each cell map had the exact number of cells, with no rounding error. By enlarging the window, the edge of the map was not obscured by the window frame when viewed on the terminal screen.
67


8. OVERLAY 1:100,000-scale BUFFERED archaeological inventory maps with WINDOW border using the "intersect" option of the OVERLAY command. The resulting map included only the inventory data for the four-quadrangle study area.
9. RASTERIZE BUFFERED archaeological inventory data to convert to cell format with background and non-background values. A 30 m X 30 m cell size was used to obtain the highest resolution possible. (Cell size must be the same as in Step 2.)
10. EXTRACT to combine all map values into one classification and assign a new single value to all cells having inventory data.
11. CROSS environmental or administrative data theme with archaeological inventory data to create a new discrete map showing logical intersection of inventoried and noninventoried areas by subject. The logical operations "AND" and "OR" in the CROSS command were used to assign new values to cells.
12. AREA the map from step 11 to obtain the number and percentage of acres inventoried for each subject of a theme.
Evaluating the Inventory and Environmental Data Methodology for Analysis
By analyzing nonrandom survey units, we can quantify the potential biases associated with the under or over sampling of particular environmental features. In this way, it is possible to determine where additional areas need to be surveyed that will provide more representative samples of sites and environmental features.
To accomplish this objective the percentage change was calculated between the cell frequency of each environmental subject that has been archaeologically surveyed and the cell frequency of the same subject throughout the total study area. These frequencies were converted to percentages and then weighed to better show the relative biases associated with the distribution of the surveyed environmental features.
68


The remainder of this section is a technical discussion of the methodology used to measure the potential biases existing with particular environmental features that have been oversurveyed or undersurveyed.
Some readers may wish to skip to the section that discusses the results of this analysis.
The percentage change between the frequency of distribution of each environmental subject inventoried and the frequency of the same subject within the total study area was calculated using cell frequencies that were converted to observed and expected values. The observed values indicate the area of each subject's intersection with inventoried areas. The expected values show the area of each subject in the entire four-quadrangle study area. By comparing the observed values to the expected values, one can measure the degree of bias associated with each subject.
The observed values were obtained using the AREA command in MAPS (Figure 5.2) for each of the maps derived by overlaying a particular environmental theme with the area inventoried. Both the intersected and the nonintersected areas were revealed by assigning unique values using the CROSS command in MAPS (Figure 5.2). The analysis used only the intersected values — where environmental subjects coincided with surveyed areas. The observed values were converted to a percentage.
The expected percentages of each subject were taken from the area table for each environmental theme. These expected values for the four-quadrangle study area represent the underlying or background occurrence of these mapped features (subjects). Because portions of the four-quadrangle study area did not have digitized data, the expected percentage values pertain only to the area of the subjects that was
69


actually mapped, and not necessarily to the distribution in the entire four-quadrangle study area.
Inventory bias can be quantified by showing the percentage of areas overinventoried and underinventoried. The percentages reflecting the degree of inventory bias for each subject were derived in the following manner. For each theme, all observed subject frequency values that intersected with the inventory data were summed. Each total was then multiplied by the percent expected of each subject for the respective theme.
The percentage or difference between the subjects that had been inventoried (observed) and expected distribution of the subjects in the project area was obtained using the following formula:
100% x Observed Frequency - Expected Frequency _ 0. „ .
0 Expected Frequency
The resulting figures represent the percentage of positive or negative bias, associated with each subject inventoried. The positive values reveal the percentage of oversurveyed area associated with a particular subject. The negative values reflect the undersurveyed percentage.
The percent bias, however, does not offer as clear an understanding of the relative amount of inventory bias of each subject in the project area as does a weighted bias. This is because large survey units may have been coincidentally located in areas where the distribution of a particular environmental feature (subject) was concentrated. The converse may have also occurred, i.e., inventories were not
70


coincidentally located where certain environmental features were concentrated. Therefore, the percent bias figures may be somewhat misleading by suggesting that a given environmental feature was highly surveyed or had received little inventory. This problem could be avoided if smaller sample survey units are selected at random rather than using larger block surveys that focus only on areas slated for development.
The weighted percent bias was used to obtain a measure of the amount of bias associated with each of the subjects (of a theme) relative to all other subjects in the study area. The weighted percent bias figures were obtained by multiplying the percent expected by the percent bias for each subject.
The foregoing methods were used to assess the quality of inventory and environmental data. These methods measure the potential for bias that may have been introduced from the lack of any overall random sampling plan for the archaeological surveys.
I
Results of Analysis
The eight environmental themes overlaid with the inventory data were elevation, slope, aspect, drainages, streams, soils, surface geology, and geologic surficial process units. The results of the analysis of each theme are shown along with an interpretation of the larger biases. A table is included for each theme. A program written in FORTRAN was used to calculate the biases for each subject. Each table contains the percent observed, the percent expected, percent bias, and weighted percent bias for each thematic subject. Following each table is a map, generated in MAPS, showing only the subjects intersected with inventory
71


data. Each subject, where appropriate, is shaded with a different pattern. Noninventoried subjects were left unshaded, or in some cases not shown. Most of the maps were derived from an Anadex printer. The Anadex printer maps are not to scale and most do not have a legend, since they are used only for analysis. Exact location, subject, and cell value information can be obtained with the MAPS QUERY command (using the crosshair input mark) to interactively select a desired point or area.
In addition, a map containing only landline data can be plotted over any map to determine sectional boundaries. Two example maps were processed on the IDIMS system and plotted on the Applicon. The Applicon maps are in color. All map projections are based on the UTM coordinate system; this explains why the printed maps are canted.
The result of overlaying the elevation data with the areas inventoried is shown in Table 5.1 and Figure 5.3. A lack of strong weighted biases larger than an absolute value of approximately 10 percent suggests that no strong factors influenced where inventories were conducted with respect to elevation classes. The highest weighted bias, 5.55 percent, is for elevation class 1,791 m to 1,820 m.
Table 5.2 and Figure 5.4 show the result of overlaying the slope data with the inventory data. Here again, no strong biases are associated with overinventoried and underinventoried slope classes. The 0 to 5 percent slope class, however, shows a 10.83 percent bias and a 6.30 percent weighted bias. This bias could be accounted for, in part, by the inventory performed in areas of little or no slope where oil and gas and coal would be easiest to develope. In addition, portions of Blocks X and XI of the Navajo Indian Irrigation Project (NIIP) lies in the northwest corner of the study area, and includes about 11 percent of
72


TABLE 5.1
ELEVATION DATA OVERLAID WITH INVENTORY DATA Subject % Observed
(Elevation Class) (Intersected) % Expected % Bias Weighted % Bias
1730 m - 1760 m 0 0.15 -100.00 -0.15
1761 m - 1790 m 1.19 5.14 -76.85 -3.95
1791 m - 1820 m 19.25 13.70 40.51 5.55
1821 m - 1850 m 20.69 18.34 12.81 2.35
1851 m - 1880 m 22.53 27.52 -18.13 -4.99
1881 m - 1910 m 20.79 19.10 8.86 1.69
1911 m - 1940 m 13.12 10.50 24.93 2.62
1941 m - 1970 m 2.42 4.51 -46.27 -2.09
1971 m - 2000 m 0.01 0.99 -99.33 -0.98
2001 m - 2030 m 0 0.05 -100.00 -0.05
73


FIGURE 5.3 Map of elevation data overlaid with inventory data
74


TABLE 5.2
SLOPE DATA OVERLAID WITH INVENTORY DATA Subject % Observed
(% Slope Class) (Intersected) % Expected % Bias Weighted % Bias
0 - 5 64.44 58.14 10.83 6.30
6 - 10 25.14 25.84 -2.71 -0.70
11 - 15 7.35 9.60 -23.48 -2.25
16 - 30 2.98 5.87 -49.18 -2.89
31 - 50 0.09 0.55 -83.10 -0.46
75


FIGURE 5.4 Map of slope data overlaid with inventory data
76


Table 5.3 and Figure 5.5 show the result of overlaying the aspect data with areas inventoried. Table 5.3 shows that no strong biases are associated with any of the aspect classes shown.
Table 5.4 and Figure 5.6 show the result of overlaying the drainage data with the areas inventoried. The Hunter/Brimhall drainage has the highest weighted bias of 8.58 percent and a percent bias of 34.14 percent. This bias, indicating overinventory, may have resulted from the presence of the Hunter/Brimhall drainage in NIIP, Block XI. Since the entire NIIP Block XI was inventoried and makes up about 11 percent of the total inventory in the project area, more of a sampling error (bias) is associated with this one large inventory sample than with other areas.
On the other hand, if many smaller size sampling units, for example, 100 10-acre inventory blocks were randomly distributed in the study area, a smaller sampling error or bias would be associated with the NIIP area and therefore the Hunter/Brimhall drainage.
Table 5.5 and Figure 5.7 represent the results of overlaying the stream data with the area inventoried. Although the biases are relatively small for all stream orders, there appears to be some positive bias towards the lowest order streams - first and second orders. The fourth order stream class shows a relatively high negative bias, the study area. This area contains relatively flat land, favorable for irrigation and agriculture, and therefore could also account for some of the overinventory observed for slope class 0 to 5 percent.
First and second order streams show a weighted bias of 7.79 percent and 6.91 percent respectively. This result can be accounted for by the
77


TABLE 5.3
ASPECT DATA OVERLAID WITH INVENTORY DATA
Subject (Aspect Class Degrees) % Observed (Intersected) % Expected % Bias Weighted % Bias
0-45 (NNE) 10.14 9.71 4.39 0.43
46 - 90 (ENE) 6.56 6.69 -1.92 -0.13
91 - 135 (ESE) 8.26 9.41 12.27 -1.15
136 - 180 (SSE) 18.92 18.45 2.57 0.47
181 - 225 (SSW) 18.35 16.69 9.96 1.66
226 - 270 (WSW) 11.86 11.84 0.15 0.02
271 - 315 (UNU) 12.17 13.23 -7.99 -1.06
316 - 359 (NNW) 13.74 13.98 -1.73 -0.24
78


FIGURE 5.5 Map of aspect data overlaid with inventory data
79


TABLE 5.4
DRAINAGE DATA OVERLAID WITH INVENTORY DATA
Subject Drainage % Observed (Intersected) % Expected % Bias Weighted % Bias
De-na-zi n 66.17 68.83 -3.87 -2.66
Escavada 0.12 4.19 -97.13 -4.07
Hunter/Brimhal1 33.70 25.12 34.14 8.58
Kim-me-ni-oli 0.02 1.86 -99.01 -1.84
80


FIGURE 5.6 Map of drainage data overlaid with inventory data
81


TABLE 5.5
STREAM DATA OVERLAID WITH INVENTORY DATA
Subject Stream Order % Observed (Intersected) % Expected % Bias Weighted %
First 39.85 32.06 24.29 7.79
Second 17.78 10.87 63.56 6.91
Third 36.78 39.76 -7.50 -2.98
Fourth 4.91 12.68 -61.25 -7.77
F1 f th 0.62 2.21 -71.73 -1.59
Sixth 0.06 2.42 -97.65 -2.36
Bias
82


J
r
f
FIGURE 5.7 Map of stream data overlaid with inventory data
83


two largest surveys in the study area - NIIP (Vogler and others 1982) -being conducted in upland topography where first and second order streams tend to occur. As mentioned above, the NIIP (BLock X and XI) encompasses about 11 percent of the study area and is situated on a plateau drained mainly by first and second order ephemeral streams.
The Joint Venture Survey, (Vierra and others 1986) in the central part of the study area, concentrated on the areas overlying major coal seams. Many of the coal outcrops lie near the tops of hills and ridges. Because of less overburden, coal at or near the outcrop is generally easier and less expensive to extract. Conversely, higher order streams, which have dissected the upland topography may have eroded the coal. The discontinuity of the seam along the outcrop may make the deposit more difficult or expensive to remove. Hence, the larger order streams, such as the fourth order, might have been undersurveyed, whereas first and second order streams occurring near the tops of hills and ridges could have been overinventoried (John Roney, personal communication, 1986).
The result of overlaying the soils data with inventory data is shown in Table 5.6 and Figure 5.8. Figure 5.8 is a color map produced on the Applicon plotter. Two soil types show marked biases with respect to the degree of inventory conducted on them. The Badland (BA) soil type has been undersurveyed by -42.23 percent and has a weighted bias of -15.06 percent. The Sheppard Mayqueen Shiprock Complex (SD) has been oversurveyed by 194.73 percent and has a weighted bias of 21.44 percent.
Much of the Badlands soil (18.46 percent) tends to co-occur in the Bisti and De-na-zin WSA. Because no oil and gas or coal development is allowed in WSAs, the Bisti and De-na-zin WSAs have probably been underinventoried. Another contributor to this negative bias may have been deliberate avoidance of badlands for oil and gas development because
84


TABLE 5.6
SOILS DATA OVERLAID WITH INVENTORY DATA Subject % Observed
(Soil Type) (Intersected) % Expected % Bias Weighted % Bias
Badland (BA) 20.61 35.67 -42.23 -15.06
Blancot Notal Assoc. Gently Sloping (BT) 5.42 8.87 -38.93 -3.45
Doak Loam (DC) 0.01 0.37 -98.46 -0.36
Doak Avalon (DN) 1.75 4.45 -60.74 -2.70
Doak-Sheppard Shiprock Assoc. Rolling (DS) 1.77 5.44 -67.44 -3.67
Dune Land (DZ) 0 0.03 -100.00 -0.03
Fruitland-Persayo-Sheppard Complex Hilly (FX) 0.01 0.01 -33.66 -0.00
Huerfano-Muff Uffens Complex (HU) 11.57 6.32 83.04 5.25
Riverwash (RA) 0 0.48 -99.31 0.48
Rock Outcrop (RO) 0.16 4.16 -96.15 -4.00
Sheppard Huerfano Notal Complex Gently Sloping (SC) 26.27 23.10 13.71 3.17
Sheppard Mayqueen Shiprock Complex (SD) 32.45 11.01 194.73 21.44
Shiprock Fine Sandy Loam (SO) 0 0.08 -100.00 -0.08
85


BR/SURUEY
BT/SURUEY
llVSUHUEY
. .
FX/SURUEY
HLi/SURUEY
:
RFVSURVEY
IsiySUHUEY
Wm | Ilp||||i|
SD/SURUEY
1:197,000
FIGURE 5.8 Map of soils data overlaid with inventory data (See Soil
Conservation Service nomenclature for soil types, Chapter III)
86


of their rough terrain, high scenic quality, and potential for paleontological resources.
In contrast, the Sheppard Mayqueen Shiprock Complex (SD) has been more thoroughly inventoried, because the SD soil is restricted to the NIIP area, which is found on a plateau. This soil type often occurs in relatively flat areas (0 to 5 percent slope). Most areas of this soil type are used for irrigated crops, mainly alfalfa and pasture (Soil Conservation Service 1980).
Overlaying the surface geology data with the areas inventoried, revealed minor biases,as shown in Table 5.7 and Figure 5.9. The weighted percentage bias column in Table 5.7 shows no large biases, but the Kkf unit (Kirtland/Fruitland - shale, sandstone, and coal) and Qa3 unit (Quaternary alluvial unit of late Pleistocene to early Holocene age) show a greater negative and higher positive bias, respectively than the other units.
The Kkf unit often occurs as steep cliffs next to badland areas, which are generally avoided for oil and gas development because of scenic quality, potential for significant paleontological resources, and the two WSAs within portions of these unique areas. In addition, the steep cliffs of this formation may dramatically increase overburden thickness, therefore making it unprofitable to mine coal (Jim Turner, personal communication 1986). These factors would account for the -6.20 percent weighted bias for Kkf.
In contrast, the Qa3 unit has a bias toward overinventory with a 7.48 percent weighted bias. This geologic unit occurs on topographically higher geomorphic surfaces (strath terraces and pediments), is coarse grained, and contains more well developed soils than Holocene units (Wells and others 1983). Because of its stable soils, Qa3 lends itself to locating equipment for oil and gas development. It also tends to occur in areas of low slope, which are ideal for development. Therefore, clearance inventories are
87


TABLE 5.7
SURFACE GEOLOGY DATA OVERLAID WITH INVENTORY DATA
Subject (Surface Geology Units) % Observed (Intersected) % Expected % Bias Weighted % Bias
K1 2.26 5.89 -61.67 -3.63
Kch 0.06 1.96 -97.16 -1.90
Kkf 14.98 21.18 -29.29 -6.20
Kmf 0 0.06 -100.00 -0.06
Kmft 0.02 0.36 -93.92 -0.34
Kpc 4.69 3.54 32.58 1.15
Qal 0 0.18 -100.00 -0.18
Qa2 0.46 4.32 -89.35 -3.86
Qa3 28.63 21.15 35.36 7.48
Qa4 19.67 21.30 -7.63 -1.63
Qa4A 0 0.02 -100.00 -0.02
Qa4B 15.13 9.98 51.63 5.15
Qel-3 6.22 5.02 23.94 1.20
Qe2-3 0.10 0.37 -71.67 -0.27
Qe3 7.77 4.67 66.42 3.10
88


Full Text

PAGE 1

A+P LD 1190 A78 1986 C325 c.l GEOGRAPHIC INFORMATION SYSTEM APPLICATIONS FOR CULTURAL RESOURCE MANAGEMENT by Mark Anthony Calamia / Al:JRARIA LIBRARY Thesis Submitted in partial fulfillment of the requirements for the Masters of Planning and Community Development Degree University of Colorado at Denver, CO August 5, 1986

PAGE 2

F _or Mom and Dad Date Due 1 --! I ---... -I I . ---• I l ...._..... -.. i i

PAGE 3

ACKNOWLEDGEMENTS This thesis might not have been completed if it had not been for many people who gave much of their time in assisting me with this project. Several BLM employees and Technical Government Services (TGS) Inc. personnel gave me valuable advice and assistance throughout the project. Daniel Webster initially trained me to use the BLM•s Geographic Information System (GIS) -MOSS/MAPS. Daniel Martin provided excellent suggestions in applying a GIS to cultural resource management situations. Sol Katz, Wendy Telley, Paul Kimberling, John Foster, John Russell, Carl Zulick, and Robert Ader taught me techniques on the GIS to accomplish some tasks more efficiently. Jim Turner answered questions on geomorphology and Al Amen assisted me with the soils data. Rose Maruska prepared and printed the flow diagrams. Randy formatted the digital elevation model data and Ed Chine produced the color Applicon plots for this study. Doug Sipes, Steve Russell, and Sharon Chomas assisted me in loading and unloading data and i n solving problems encountered with the computer• s operating system. Carsella, Mike Fiebach, Dennis Colarelli answered my questions on command a 1 gori thms. Outside of the BLM Service Center, several individuals were particularly helpful in gathering data and answering technical questions. Those individuals were John Roney (BLM Albuquerque District Office archaeologist ) , Robert Bewley (BLM Albuquerque District Office geographer), and Jeff Neibert (BLM geographic information system coordinator for the BLM New M exico State Office). Professor Steven Wells, geomorphologist at the University of New t 1exico allowed me to use several geomorphological map overlays. I am also thankful to Steve Fosberg (BLM New M exico State Office archaeologist) for iii

PAGE 4

his support and encouragement throughout the course of this project. Marsha Jackson at the New Mexico Laboratory of Anthropology answered my questions of the Archaeological Records Management System (ARMS). Ken McGinty assisted in editing this report, and Gerry Stakes spent many hours formatting and typing the manuscript. I am indebted to them for their valuable criticisms and comments during preparation of the final product. My thesis advisor, Dr. Thomas Clark, provided much encouragement and general guidance for this study. Above all others, I would like to say a special thank you to my coworker Mike Garratt (BLM statistician) who assisted with the quantitative aspects of this study. In addition, he provided many comments related to the technical accuracy and content of the report. Without his candid criticisms and remarks, this manuscript would have suffered. For his constant encouragement, patience, and generosity, I am grateful. Finally, it should be said that any shortcomings or faults which may occur in any portion of this thesis ultimately rest with the author himself. Thus, I alone claim full responsibility for the contents of this report. iv Mark A. Calamia August 1986

PAGE 5

TABLE OF CONTENTS ACKNOWLEDGEMENTS. LIST OF TABLES. LIST OF FIGURES . ABSTRACT. Chapter I INTRODUCTION Project Goals and Concept ........... . The Cultural Resource Management Problem BLM Resource Management Plans ..... . Planning and Historic Preservation ........ . Spatial Analysis and Support Program Needs The New Mexico Data Synthesis Project .. Structure of Report. . . . . . . . . . II GEOGRAPHIC INFORMATION SYSTEMS AND CULTURAL RESOURCE MANAGEMENT STUDIES Page iii viii xi xiv 1 2 3 4 9 10 13 16 Geographic Information Systems . . . . 16 III The MOSS/MAPS System . . . . . . . . . . . . . . . . . . . 22 Previous CRM Studies Using Geographic Information Systems. 23 EXISTING ENVIRONMENT . . . Location and Physiographic Setting Climate ......... . Hydrology ............. . Geology and Geomorphology .. Local Lithic Raw Materials Soi 1 s. . Flora. . . . . . . . Fauna ....... . 29 29 31 31 32 36 37 37 42 IV THE ARCHAIC PERIOD 43 43 Overview of the Archaic. v

PAGE 6

TABLE OF CONTENTS (CONT.) Chapter V ASSESSING ARCHAEOLOGICAL INVENTORY AND ENVIRONMENTAL DATA BASES . Selecting the Study Area . . . . . Planned Development in the Study Area. BLM Inventory Levels . . ........ . Problems and Biases in Using Existing Site and Inventory Data . . . . . . . . . . . . . . . Data Themes ................. . Procedures for Assessing Inventory Data Using MOSS/MAPS. . . . . . . . . . . . . . . . . . Evaluating the.Inventory and Environmental Data. Results of Analysis .............. . Evaluating the Inventory and Administrative Data . Results of Analysis. . .......... . Conclusions .................. . VI USE OF MOSS/MAPS FOR IDENTIFYING POTENTIAL CULTURAL RESOURCE AND LAND USE CONFLICT AREAS . . . . . Procedures for Using MOSS/MAPS to Prepare Data for Identification of Potential Cultural Resource and Land Use Conflict Area ..... . Methodology Used for the Analysis of Potential Cultural Resource and Land Use Conflict Areas. Results of Analysis .............. . Conclusions .................. . . . . . VII USING A GIS IN CULTURAL RESOURCE MANAGEMENT SITUATIONS Scenario 1 . Scenario 2 ................ . Conclusions ................ . VIII EVALUATION OF MOSS/MAPS FOR MEETING CULTURAL RESOURCE MANAGEMENT NEEDS . . . . . . . . . . . . . . Assessment of MOSS/MAPS ............ . Capabilities and Benefits of MOSS/MAPS for CRM Problems of Using MOSS/MAPS for CRM ..... The Relational Data Base Management Approach Cost Estimates . . . . . . . . . . . . . . . Conclusions ................ . vi Page 52 52 54 54 56 60 65 68 71 90 94 116 118 119 127 129 145 149 149 155 160 161 161 162 164 170 171 175

PAGE 7

TABLE OF CONTENTS (CONT.) Chapter IX THE USE OF A GIS IN THE BLM RESOURCE MANAGEMENT PLANNING PROCESS AND FOR SUPPORT PROGRAM NEEDS X Using MOSS/MAPS in the Planning Process .. GIS and Support Program Needs .. Summary. . . .... CONCLUSION . . . . . . . . . . . The Need for a GIS in Site Location Modeling . GIS and the New Mexico Data Synthesis Project .. GLOSSARY. . . BIBLIOGRAPHY. vii Page 176 177 187 189 1 91 . . . . 191 . . . . 197 198 200

PAGE 8

Table 3.1 3.2 5. 1 5.2 5.3 5.4 5.5 5.6 5.7 5.8 5.9 5.10 LIST OF TABLES Soil Types and Assoiciated Range Site Data for the Project Area ......... . Major Vegetation Units and Associated Land Forms Elevation Data Overlaid with Inventory data. Slope Data Overlaid with Inventory Data .. Aspect Data Overlaid with Inventory Data Drainage Data Overlaid with Inventory Data Stream Data Overlaid with Inventory Data Soils Data Overlaid with Inventory Data .. Surface Geology Data Overlaid with Inventory Data. Geologic Surficial Process Units Data Overlaid with Inventory Data ................. . Land Status (Ownership) Data Overlaid with Inventory Data. Wilderness Study Areas (WSA) Data Overlaid with Inventory data . . . . . . . . . . . . . . 5.11 Linear Development in WSA Data Overlaid with Inventory Data . . . . . . . . . . . . . . 5.12 Road Network Data Overlaid with Inventory Data . 5.13 Proposed Areas of Critical Environmental Concern (ACEC) Data Overlaid with Inventory Data ..... 5.14 Range Allotment Data Overlaid with Inventory Data .. 5.15 Oil and Gas Lease Data Overlaid with Inventory Data. 5.16 Preference Right Lease Applications (PRLA) Data Overlaid with Inventory Data ..... . 5.17 Industry Expressions of Interest (IEOI) in Coal Leasing Page 38 73 75 78 80 82 85 88 91 95 98 100 103 1 0 5 108 110 112 Data Overlaid with Inventory Data. . . . . . . . . . . 114 viii

PAGE 9

Table 6.1 6.2 8.1 8.2 9.1 LIST OF TABLES (CONT.) Area Table of Preference Right Lease Applications (PRLA) Data Overlaid with Archaic Site Data for Two Leases. Area Table of Oil and Gas Lease Data Overlaid with Archaic Site Data for Four Leases ......... . Estimated time for Digitizing and Editing Selected Themes .. Actual Time for Digitizing and Editing Selected Themes Acreage Table of Total Land Available for Uses Under Various RMP Alternative Management Schemes ..... ix Page 142 148 173 174 182

PAGE 10

Figure 2.1 2.2 3.1 5.1 5.2 5.3 5.4 LIST OF FIGURES Construction of a GIS ................ . Characteristics of Map Information in Vector-Based and Cell-Based Geographic Information Systems . Location Map ......... . Data Themes Used for Analysis Cartographic Model for Assessing the Archaeological Inventory Data. . . . . . . . . . . . . . . . . . . Map of Elevation Data Overlaid with Inventory Data. Map of Slope Data Overlaid with Inventory Data. Page 20 21 30 61 66 74 76 5.5 Map of Aspect Data Overlaid with Inventory Data . 79 5.6 of Drainage Data Overlaid with Inventory Data 81 5.7 Map of Stream Data Overlaid with Iventory Data. 83 5.8 Map of Soils Data Overlaid with Inventory Data. 86 5.9 of Surface Geology Data Overlaid with Inventory Data. . 89 5.10 Map of Geologic Surficial Process Units Data Overlaid with Inventory Data . . . . . . . . . . . . . . 92 5.11 5.12 Map of Land Status (Ownership) Data Overlaid with Inventory Data ................ . Wilderness Study Areas (WSA) Data Overlaid with Inventory Data. . . . . . . . . . . . . . . . 96 99 5.13 Linear Development in WSA Data Overlaid with Inventory Data 101 5.14 Road Network Data Overlaid with Inventory Data. 104 5.15 Map of Proposed Areas of Critical Environmental Concern (ACEC) Data Overlaid with Inventory Data ...... . 106 X

PAGE 11

LIST OF FIGURES (CONT.) Figure Page 5.16 Range Allotment Data Overlaid with Inventory Data . 109 5.17 Oil and Gas Lease Data Overlaid with Inventory DAta 111 5.18 Preference Right Lease Applications (PRLA) Data Overlaid with Inventory Data . . . . . . . . . . . . . . . . . 113 5.19 Industry Expressions of Interest (IEOI) in Coal Leasing Data Overlaid with Inventory Data . . . . . . . . . . 115 6.1 Cartographic Model for Preparing Archaeological Site Map. . 122 6.2 Cartographic Model for Identifying Cultural Resource and Land Use Conflict Areas . . . . . . . . . . . . 123 6.3 Map of Range Data Overlaid with Archaic Site Data . . . . 130 6.4 Map of Proposed ACEC Data Overlaid with Archaic Site Data 132 6.5 Map of Fossil Forest Data Overlaid with Archaic Site Data . 133 6.6 Map of WSA Data Overlaid with Archaic Site Data . . . . 134 6.7 Map of Land Status Data Overlaid with Archaic Site Data 135 6.8 Map of Linear Development in WSA Data Overlaid with Archaic Site Data ............... . 137 6.9 Map of Road Network Data Overlaid with Archaic Site Data. . 138 6.10 Map of Preference Right Lease Applications (PRLA) Data Overlaid with Archaic site data . . . . . . . . . 140 6.11 Enlarged Map of Preference Right Lease Applications (PRLA) Data Overlaid with Archaic Site Data. . . . . . 141 6.12 Map of Industry Expressions of Interest (IEOI) in Coal Leasing Data Overlaid with Archaic Site Data. . . . . 143 6.13 Enlarged Map of Industry Expressions of Interest (IEOI) in Coal Leasing Data Overlaid with Archaic Site Data . . 144 xi

PAGE 12

LIST OF FIGURES (CONT.) Figure Page 6.14 Map of Oil and Gas Lease Data Overlaid with Archaic Site Data . . . . . . . . . . . . . 146 6.15 Enlarged Map of Oil and Gas Lease Data Overlaid with Archaic Site Data . . . . . . . . . . . . . . . 147 7.1 Map of Distribution of Archaic Site Data Relative to Preference Right Lease Applications (PRLA) Data . . 151 7.2 Enlarged Map of Preference Right Lease Applications (PRLA) Data Overlaid with Inventory Data . . . . . . . . . . . . 152 7.3 Enlarged Map of Preference Right Lease Applications (PRLA) Data Plotted with Archaic Site Data . . . . . . . . . . . 153 7.4 Enlarged Map of Archaic Sites Plotted on Preference Right Lease Applications (PRLA) Data as Overlaid with Geologic Surficial Process Units Data. . . . . . . . . . 154 7.5 Map of Distribution of Archaic Site Data Relative to Road Network Data . . . . . . . . . . . . . 157 7.6 Enlarged Map of Distribution of Archaic Site Data Relative to Road Segments .......... . 158 xii

PAGE 13

ABSTRACT The purpose of this thesis is to identify and evaluate the applications of Geographic Information System (GIS) technology for federal government cultural resource management situations, especially those involving planning and compliance for fossil fuel development. Procedures are developed for performing spatial analysis involving administrative, environmental and cultural resource data. The demonstration of these procedures will aid BLM state, district, and resource area offices in meeting their resource management planning needs. Capabilities, limitations, and deficiencies as currently exist for performing archaeological spatial analysis will be identified for BLM's primary GIS MOSS/MAPS. xiii

PAGE 14

CHAPTER I INTRODUCTION Project Goals and Concept The purpose of this study is to identify and evaluate the applications of Geographic Information System (GIS) technology to federal government cultural resource management situations, especially those involving planning and compliance for fossil fuel development. Using some of the technological aspects described in the Bureau of Land Management (BLM) technical volume, BLM Predictive Modeling Draft, DPP No. 8000.001, procedures are developed for performing spatial analysis involving cultural resources. The demonstration of these procedures will aid BLM state, district, and resource area offices in meeting their resource management planning needs. It is suggested that the reader of this report have some familiarity with GISs, and MOSS/MAPS in particular, to facilitate understanding of the material. Capabilities, limitations, and deficiencies as currently exist for performing archaeological spatial analysis within a planning context, will be identified for BLM1s primary GIS. The aim of this project is to show how a GIS can serve as a valuable tool for a BLM resource area in planning for energy development (coal, oil, gas) and cultural resource management (CRM). This study identifies applications to CRM in a multiple land use context and applies the Map Overlay and Statistical System (MOSS) and the Map Analysis and Processing System (MAPS) to functions that would be more effectively accomplished

PAGE 15

through the analytical capabilities of a GIS than through conventional means. Evaluation of the utility of a GIS for compliance and planning includes time, accuracy, monetary cost, and overall effectiveness. The MOSS/MAPS is BLM's GIS The Cultural Resource Management Problem The project area consists of a small portion (154,187 acres or 10.3 percent) of the BLM Farmington Resource Area (FRA) in northwestern New Mexico. The area itself falls within the northern San Juan Basin. The FRA has a large amount of existing and proposed mineral development and leasing. Most of the proposed development is in the form of surface coal mining and oil and gas extraction. Federal law requires that all cultural resources on federal properties be identified and assessed by the significance criteria listed in the National Historic Preservation Act of 1966, as amended. The FRA is faced with a problem regarding the distribution and location of cultural resources with respect to expected coal and oil and gas development. Specifically, the FRA requires (1) a qualitative and quantitative assessment of its existing cultural site, survey, and environmental data bases, and (2) the identification of potential conflict areas between cultural resources and specific land uses. This study shows and documents the utility of a GIS for assessing data bases. Moreover, this study identifies these existing site locations with respect to existing coal leases and oil and gas leases so that potential conflict areas can be addressed for compliance and planning needs. 2

PAGE 16

BLM Resource Management Plans BLM receives its authority to balance development with protection of natural and cultural resources from the National Environmental Policy Act of 1969 (NEPA); the Federal Land Policy Management Act of 1976 (FLPMA); and other federal, state, and local laws. NEPA requires evaluation of the effects of major federal actions on environmental (including cultural) resources. Under NEPA, every federally funded or licensed entity must consider the total environment. FLPMA specifies several key directions for BLM, notably that goals and objectives be established as guidelines for public land use planning, and that public lands be managed on the basis of multiple-use and sustained yield, unless otherwise specified by law. To achieve its land use management goals, BLM develops comprehensive land use plans called resource management plans (RMPs) to set guidelines for multiple-use decisionmaking. The plans are used by field-level managers in BLM resource areas. Individual plans are prepared for each area, which may vary in size from 100,000 acres to several million acres. Because of geographic and natural diversity, each plan is unique to the area it addresses. RMPs must be both comprehensive and issue oriented, must discuss all affected resources and concerns, and must focus on actual problems. The plans must allocate resources for specific uses. Environmental impact statements (EISs) are fully integrated into the RMPs and discuss the effects of implementing each management alternative presented in the RMP. Once all alternatives have been thoroughly analyzed, BLM tentatively selects a preferre d, multiple use-oriented alternative. BLM also follows a public involvement plan during the 3

PAGE 17

typical 2 years of RMP preparation. Once completed, the RMP is implemented through the decisions made in the document and for on-going activities by such site-specific and resource activity plans as cultural resource management plans. The plans are monitored throughout implementation (about 10 years) to ensure that impacts are accurately assessed and goals are being met. BLM uses a GIS to assist in thorough analysis and timely completion of projects that use large data bases. In some cases, these tasks would be impossible to complete manually while maintaining the high-quality output that a GIS can produce. Land use planning is often the medium for implementing a GIS and building a comprehensive digital data base. Planning integrates all resource and mineral disciplines and lends a common purpose and consistency to goals for each resource represented in the planning effort. Interdisciplinary teams prepare the RMPs. The planning team assesses funding and data needs and either guides resource specialists in preparing data for digitizing or procures existing digital data. With the assistance of management and resource specialists, planners often guide and shape the implementation of a GIS. This report describes the specific applications of a GIS to aid in the RMP process. Planning and Historic Preservation The National Historic Preservation Act (NHPA) as amended and as supplemented by Executive Order 11593 requires federal agencies to locate significant cultural resource properties and allows for regulatory agency comment before implementing actions that would effect such properties. The quasi-regulatory agency created by the act is the Advisory Council on 4

PAGE 18

Historic Preservation, which has supported the integration of cultural resource management considerations into federal agency planning and thus into land use planning. Other environmentally influenced legislation is equally responsible for implementing cultural resource management activities. NEPA and FLPMA are good examples. As mentioned above, these acts mandate an emphasis on land use planning and designate a wide spectrum of natural and cultural resources (specifically including archaeological and historical properties) that must be considered in making specific land use decisions. These mandates constitute the first time in government decisionmaking when specified information about the locations of a whole range of resources has been needed. The requirements of FLPMA and NEPA complement those of the National Historic Preservation Act. NEPA mandates that during federal decisionmaking all components of the environment be considered in an EIS. Although preparation of an EIS requires planning, FLPMA further formalizes the planning phase of decisionmaking and cites cultural resources as one of the types of resources to be included among the factors influencing decisions. The policy portion of FLPMA specifies protection of the quality of historical and archaeological values as a management responsibility of the federal government, specifically in this case, BLM. One of the central concerns of FLPMA is areas of historical and cultural value that require special management attention. FLPMA sets forth a management process wherein BLM must inventory such resources and other values encompassed by the other sections of FLPMA when they lie within BLM's areas of jurisdiction. 5

PAGE 19

The planning process gives priority to designating and protecting these areas of historical and cultural value, but all resources must be identified and considered on the basis of their relative scarcity. FLPMA, therefore, provides an orientation for federal land-managing agencies by explicitly naming cultural resources as one of the spectrum of resources that must be inventoried and included in the planning process and for which protection is a defined management option. The specific management prescriptions, however, are not defined, either in the act or in the 36 CFR 800 regulations. Instead, BLM has developed internal directives and manuals that set basic procedures and standards for cultural resource management. These procedures and standards define administrative and physical inventory methods, evaluation frameworks, and cultural resource protection measures. The BLM inventory structure includes three classes of inventory (BLM 8100-CRM Manual, 1978). The Class I inventory is a review and compilation of existing data. The Class II inventory (survey) is defined as a sampling field inventory, but this definition does not address relationships between sampling and spatial analysis. The Class III inventory is a complete surface inventory of a specified area, i.e., an intensive field inventory. This report discusses the capabilities of a GIS to meet Class II and Class III inventory objectives. The BLM inventory structure is intended to be a sequential tiering system. Some measure of Class I inventory is always conducted at first, whatever the eventual goal of the specific inventory situation. A complete Class I inventory treatment is conducted for every BLM administrative unit. As part of this inventory, a list of all previously recorded cultural resources is compiled, a narrative history and 6

PAGE 20

prehistory of the area are written, and major research domains and data gaps are identified. A shorter version of a Class I inventory is performed for specific projects or programs to identify known cultural properties and inventory needs. When the BLM CRM manual was written in the mid-1970s, much of the BLM administered land base had received little or no inventory. Class II inventory was envisioned for large-area survey applications, wherein a sampling approach would be used to estimate the density and distribution of cultural properties. Neither the manual nor associated BLM di-rectives state that the objective of the Class II inventory should be to create a model to predict locations or environments having defined classes of cultural properties. Rather, the main objectives of Class II inventory are to (a) identify management opportunities, to protect cultural properties in their own right, (b) identify potential constraints on other land use decisions, and (c) identify general areas of sensitivity for cultural resource issues. BLM has used the Class III inventory as the culmination of inventory for any project or undertaking that could affect cultural resources, or as the highest precision level of inventory for identifying resources within an area managed for its cultural resources. Class III is generally the level of inventory chosen when cultural properties are expected to occur because this is the most certain means of satisfying the legislative framework for cultural resource protection. Most of the contributing surveys in the project area were conducted at the Class III level as intensive survey units selected as part of Class II inventories in and near the project area. 7

PAGE 21

This project has investigated the possibility of using a GIS for examining existing Class II and Class III inventory data to address Class II and Class III inventory objectives. More specifically, the survey data base has been analyzed to determine areas where future surveys need to be concentrated to provide more representative samples of cultural site data and environmental variables in the study area. In this manner, one can assess and measure the potential for biases associated with certain environmental characteristics resulting from nonrandom sample survey units. In BLM, planning system products normally lead to a resource allocation decision. By considering the resources on BLM-administered lands, decisionmakers make trade-offs to achieve the most environmentally and publicly acceptable balance of resource uses. To fit within the multiple-use concept, BLM has developed by internal directive a cultural resource evaluation system that assigns uses to individual cultural The system assigns value to the cultural resources categories commensurate with potential resource use. The seven uses allowable in the BLM system include the following: (1) current scientific use; (2) potential scientific use; (3) conservation for future use; (4) management use; (5) socio-cultural use; (6) public use; and (7) discharged use. See BLM 8100-Cultural Resource Management Manual (1978) for a discussion of these seven allowable uses. This evaluation system provides a means of assessing the precedence of a cultural resource relative to other land uses. In a multiple-use management situation, the presence of a cultural property segregated from other resource use could prevent coal mining at a particular location. On the other hand, the presence of a resource designated for excavation 8

PAGE 22

could not prevent mining; once mitigation had been carried out, coal mining could occur. The BLM's cultural resource evaluation system is not exempt from the NHPA requirements. Where a conflict exists between the proposed use of a cultural resource and the use of another resource such as coal mining, the site must be evaluated for NHPA eligibility. Moreover, potential effects on the property must be determined in consultation with the Advisory Council on Historic Preservation. Presumably, however, the BLM-determined use evaluation does influence the council's comments on effect. Spatial Analysis and Support Program Needs Cultural resource management programs maintained by federal, state, or other agencies are generally guided by a set of policies, procedures, and priorities. In virtually every case, some part of direction is to support other resource programs in meeting the requirements of historic preservation law, regulation, and agency policy. For example, cultural resource staff members generally provide support to coal program specialists when coal lease applications are being processed. Cultural resource specialists are involved throughout the project, during successive stages of alternative lease area selection--stipulation development, inventory, evaluation and mitigation, and monitoring. Despite the importance of the support role filled by cultural resource staff members, the support program has not been formalized within any land-managing agency. Support has mostly been interpreted as compliance, which is only one aspect of support. A full support program should include planning and implementation as they relate to an administration and compliance plan. 9

PAGE 23

Two main issues need to be addressed in developing a cultural resource support program: (a) the types of information that are needed and (b) the strategies for tailoring a general support program to the needs of a specific project. The main components of a general support program--inventory and evaluation--fall in the domains of planning and implementation, and These components, in turn, should relate to separate administrative and compliance plans. Each of these components should be addressed independently to determine information needs and appropriate strategies. Requirements of each should also assist decisions made within the others. This report describes how information needs and strategies for tailoring a general support program for a project can be facilitated through the use of spatial analysis methodologies via a GIS. Theoretical and methodological issues must ultimately provide a basis for determining why cultural resource information should be collected and analyzed, what types of information should receive priority, and how data should be collected and analyzed. To better realize this goal, some federal, state, and local agencies and institutions are recognizing the need to synthesize their growing data bases to reevaluate and reduce intensive survey requirements. The New Mexico Data Synthesis Project The Data Synthesis Project began in the Farmington and Carlsbad resource areas during fiscal year 1983 as part of the BLM New Mexico Cultural Resource Program. A statewide priority was granted to the project by the New Mexico BLM State Director, who saw as a main goal of the project the reevaluation and possible reduction of intensive inventory requirements in the two main oil and gas areas in New Mexico--Farmington and Carlsbad. 10

PAGE 24

The rationale for this approach was that, though hundreds of inventories of well pads and access roads had been performed in these areas, no one had tried to synthesize the data from these .. clearance .. inventories and postulate where cultural resource properties would most likely occur. Because of the size and complexity of the information base, existing site and survey data had to be used through an automated data processing system. Once the needed hardware was purchased and located at the BLM Albuquerque District Office, the Data Synthesis Project was selected as a pilot project for exploring the applications of this equipment. The Data Synthesis objectives are twofold: (1) to provide a means for the integrated evaluation and analysis of cultural resource site data, survey data, and appropriate environmental information and (2) using existing data, to provide the most information possible on the nature and distribution of cultural resources to inform and guide resource allocation decisions in a multiple-use context, in a timely and cost-effective manner. The two basic project objectives reflect the dual emphasis of the project as it developed: (1) a primary push towards automating the large and complex data bases involved and (2) a secondary long-term effort to develop and refine potential applications of the data base. The development of spatial analysis methodologies per se was not considered to be one of the main objectives, though it was clearly recognized that model development would be facilitated in areas where the requisite data bases had been created. When the Data Synthesis Project was conceived during 1982, the term GIS was not widely used within BLM. BLM offices lacked graphics hardware 11

PAGE 25

and applications for planning and resource data display. In addition, data manipulation had not been established or demonstrated. In retrospect, however, the Data Synthesis Project relates to a GIS application project developed specifically for the display and manipulation of cultural resource and other relevant data. Additionally, the GIS allows for expanded multiple-use management planning capabilities involving cultural resources. The Data Synthesis Project identified three types of project-specific data that had to be simultaneously displayed and manipulated in a graphics format. These included the site data, the surveyed area data, and the environmental data. Development of the data base was greatly facilitated by the presence of an existing automated statewide cultural site data base maintained by the Laboratory of Anthropology, a New Mexico State agency. Although certain backlogs in site entry are present, the scope of the file is statewide. With the exception of selected United States Forest Service sites, the file contains and receives information on all sites recorded within the State regardless of land ownership. The file is updated every 6 months. The site file is maintained as part of the Archaeological Records Management System (ARMS). ARMS is the site data base that has been integrated with the MOSS graphic data base for use in this study. BLM views the results of this GIS study as the next step in the New Mexico Data Synthesis Project towards achieving its two objectives, with emphasis on developing and refining potential applications for the syste m that could aid in CRM and support programs. To evaluate the potential of a GIS for assessing data bases and providing information on distribution of cultural resources for guiding 12

PAGE 26

resource allocation decisions, this study must present an overview of prehistoric settlement pattern behavior. For this study, the settlement and adaptive system of the Archaic period (5,500 B.C. to A.D. 400) has been explored for a portion of the upper San Juan Basin. A glossary is provided at the end of this report to aid readers who may be unfamiliar with some of the anthropology, archaeology, and geology terms. Structure of Report The remainder of this chapter describes the organization of this report. A brief description of the topics covered in each chapter is given below. Chapter II briefly reviews what a GIS is and, in particular, the general capabilities of MOSS/MAPS. Chapter II also presents an overview of some of the major CRM projects involving spatial analysis using a GIS. Chapter III briefly discusses the physiographic setting and natural environment of the project area. Chapter IV presents an overview of the Archaic period, ethnographic data on hunting and gathering societies, and research on the Archaic period conducted in the northern San Juan Basin. In this study the GIS will be used to identify potential conflict areas between Archaic sites and specific land uses in the project area. Chapter V describes and explains procedures and methods using a GIS for evaluating the archaeological inventory and environmental data for potential biases. The data sets used for analysis are discussed. Nine administrative themes (areas of planned and existing development or conservation) are also analyzed to determine the amount of inventory conducted for them. A cartographic model schematically documents a GIS user ' s procedures for combining layers of map data. The cartographic 13

PAGE 27

model also documents how a planner or archaeologist can use a GIS, specifically MOSS/MAPS, to provide technical assistance in making a resource management decision. Results of the analysis are shown in graphic and tabular form with associated text. Chapter VI includes a series of GIS-generated maps and tables that will aid in identifying cultural resource and land use conflict areas in a multiple-use setting. Specifically, the distribution of archaeological sites of a specific cultural period are analyzed in relation to their distribution with areas of proposed coal and oil and gas development. However, areas designated or proposed for conservation are also analyzed using the archaeological site data. Chapter VII shows how MOSS/MAPS might be used in a typical CRM compliance situation. Analytical results from Chapters V and VI are combined to show how a resource area archaeologist can use output generated by MOSS/MAPS to meet BLM requirements. These requirements may include the preparation of environmental assessments (EAs), environmental impact statements (EISs), or resource management plans (RMPs). Chapter VIII is an evaluation of the capabilities of the MOSS/MAPS software as integrated with ARMS for purposes of CRM and support program planning. The criteria used for this evaluation are monetary cost, time required, and level of accuracy compared to existing methods for accomplishing the same tasks. Additionally, some specific suggestions are given for enhancing MOSS/MAPS and ARMS for meeting CRM needs. Chapter IX centers on GIS, specifically MOSS/MAPS, as a tool in the BLM resource management planning process involving cultural resources. The phases in the RMP process where GIS could be used for the FRA are described. The remainder of the chapter focuses on the capabilities of 14

PAGE 28

GIS in meeting federal support program needs for project planning. Finally, Chapter X draws conclusions for using a GIS, as contrasted with manual methods, for performing site location modeling. Reference is also made to the use of the applications in this study for aiding the development of the New Mexico Data Synthesis Project. 15

PAGE 29

CHAPTER II GEOGRAPHIC INFORMATION SYSTEMS AND CULTURAL RESOURCE MANAGEMENT STUDIES This chapter discusses what a GIS is and specifically the capabilities of MOSS/MAPS. Chapter II also presents an overview of some of the major CRM projects involving spatial analysis using a GIS. Geographic Information Systems Geographic Information Systems are computer-based means for assembling, analyzing, and storing varied forms of data corresponding to specific geographic areas, with the spatial locations forming the basis of the system (Tomlinson and others 1976). The term GIS, as used here, is restricted to computer systems that can interrelate sets of data representing different geographical variables, as opposed to systems that merely manipulate or map individual files or geographical data (Rhind 1975). Virtually any type of geographically distributed information from any source can be encoded in computer-compatible form. Computers can extract geographic information from digital geographic data bases, manipulate the data, derive new data, and analyze this information to propose solutions to problems. The analytical capabilities of such a system make it a powerful tool for spatial analysis. Thus, a GIS can go beyond the role of merely processing and displaying information. A GIS can also be incorporated into the analysis, interpretation, and problem solving aspects of research in geographically distributed phenomena and processes (Hasenstab 1983a). 16

PAGE 30

Many types of geographically distributed data can serve as the primary information portion of a GIS: elevation data, river and stream locations, vegetation patterns and soil types {which might be derived from satellite remote sensing), known archaeological site locations, and regions of planned construction or development. In its most elementary use, a GIS can retrieve geographic information that is encoded in data bases for a specified coordinate point, such as the location of an archaeological site. For archaeological research, additional information might include cultural data, such as known archaeological site locations, surveyed areas, access roads, and areas of planned development or impact. A GIS is also capable of using associated characteristic site and survey information from multiple attribute data derived from data base management systems. This attribute information can be used to select subsets of the data base beyond merely geographic characteristics. Such a procedure, however, does not fully use the central capability of a GIS--the ability to derive new information beyond that originally encoded in the data base {Collins and Moon 1981). For example, from interrelationships between known points of elevation in the data base, it is possible to estimate at any locus the values of slope, aspect, and a variety of relief and terrain roughness measures {Monmonier 1982:76-29). Points of vantage, such as hilltops and ridges, can also be defined (Kvamme 1983a). From a digital hydrology net, distances to nearest seasonal or permanent streams can be computed, and from digitized vegetation data, distances to a specified plant community (Lee and others 1984). Listings of nearest neighbor sites and distances can be obtained, as we 11 as the distance to a centra 1 p 1 ace vi 11 age from a data 111 ayer .. containing known archaeological site locations. 17

PAGE 31

An important benefit for the data-generating capabilities of a GIS is that it can derive information that was previously impossible to obtain due to the sheer number of required calculations. Maximum view distances, measures of shelter and view quality, and least-effort travel distances are all potential information classes that illustrate this property. One can obtain through the U.S. Geological Survey (USGS) and other government agencies or private companies many types of geographical data, particularly regional . elevation data, already in digital form and on computer tape. For example, the USGS now produces highly accurate digital elevation models (DEMs) obtained through digitizing 1 :24,000-scale topographic maps (Doyle 1978:1484). Most often, other sources of information, such as vegetation and soil data, do not exist in digital form; archaeological data usually is not available. Therefore, these data often need to be electronically digitized. A common digitizing procedure employs a digitizing tablet and cursor (Monmonier 1982:7; Rogers and Dawson 1979). With these devices, such potential information as geomorphological mapping units or stream courses are manually traced and encoded in computer compatible form. The primary data, which can be derived from many sources, are often digitized from 1 :24,000-scale topographic maps, but other sources, such as remotely sensed digital satellite images, can be used (Shelton and Estes 1981). However they are acquired, the several primary surfaces of digital information that a GIS needs are encoded and stored in the initial data base. Computer programs are then able to use these primary data to generate secondarily derived information that is often more useful than the primary data (Collins and Moon 1981). For example, slope 18

PAGE 32

estimates, aspect estimates, or distances to nearest drainages might be derived (from elevation and hydrology surfaces, respectively) and stored as new and distinct analytical surfaces. The principle of GISs is that users input the system through digitization or other manual means with base information (primary layers) which can be used to derive additional or secondary data layers. Both primary and secondary surfaces can then be used for analysis or display. The ways in which these data are used, however, depend on the nature of the particular GIS (Figure 2.1). There are two basic types of GIS designs: the vector-based and the cell-based GIS. The two designs are sometimes included in one GIS package. The vector-based GIS stores data as a series of points, lines, and polygons that are used to identify discrete features (vector is another term for a line between two points). The computer storage requirements for this information are smaller than cell data because only the digitized points along lines or polygon boundaries need to be stored (Figure 2.2). The cell-based (sometimes called raster based) GIS superimposes a regular grid of rows and columns of cells over the region and assigns a numeric value to each cell. Each cell corresponds to a fixed area in real space and each contains a value for that area. With a cell-based GIS dichotomous, discrete, and continuous data can be used. 19

PAGE 33

N C> ORIGINAL GROUND SURFACE AND THEMATIC MAPS PRIMARY SURFACES SECONDARY SURFACES FIGURE 2.1 Construction of a GIS. From the original land surface (b), various thematic maps are produced, such as elevation contours (c), hydrology (d), and forested areas (e). These maps are digitized and converted to primary layers in a GIS representing an elevation surface (f), a hydrology surface (g), and a forest location surface (h), which are all referenced to a reference grid, such as the UTM grid (a). From the elevation surface, secondary surfaces, such as slope (i), a s pect (j), and local relief (k), might be obtained. The hydrology surface might provide a surface showing distance to nearest drainage (1), and the forest location surface might yield a surface showing distance to nearest forest (m).

PAGE 34

N ..... BASE MAP VECTOR / " CELL I 1!1 I I I I I I I 12 22 222 Ill !I I I Jl I I I 12 2 212 212 I ll I II II I I I [ 12 r'r2 I Ill I I I I f21Z rz 1"2" I 1100 I I 1 1 0 I I 22 121Z [! rz I )'00 1001( 1212 I"! 12 Ill :o:c OOIOIC 12, _l[l 0010 I I I ll i ! Q I I I IC IC 12 I I I Ill I I I ll I I I rr IT IT tiT II rr TIT I I I II II I I I I I I I I I I Ill FIGURE 2.2 Characteristics of map information in vector-based and cell-based geographic information systems (after Lee and others. 1985)

PAGE 35

The MOSS/MAPS System MOSS (Map Overlay and Statistical System) is a GIS originally developed by the Western Energy and Land Use Team, U.S. Fish and Wildlife Service. MOSS has been in continuous development over the last few years with cooperation from the Bureau of Indian Affairs, the BLM, the Forest Service, the Geological Survey, and the Soil Conservation Service (Lee and others 1985). Thus, unlike most geographic systems, MOSS is in the public domain, although a superset of MOSS is marketed by Autometric of Fort Collins, Colorado. Most storage and processing in MOSS is in vector or polygon format, although some raster capabilities are available. MOSS, as used by BLM, has both MOSS and MAPS. The version of MOSS/MAPS used for this project has not been released as of this writing. Additional raster capabilities, designed in part to allow the incorporation of data derived from cell images, may be obtained through the Map Analysis Package Subsystem (MAPS), originally developed at Yale University. Most of the analytical manipulation in this study was performed using MAPS. To some extent, MAPS and MOSS can pass files back and forth. Input to MOSS is through the Analytical Digitizing System (ADS) or the Analytical Mapping System (AMS). Enhanced cartographic plotting beyond the normal capabilities of MOSS or MAPS, is provided by the Cartographic Output System (COS). For this report two maps were processed on the !DIMS system (see below) and plotted in color using the Applicon plotter at the BLM Service Center in Denver. The remainder of the maps used in this report were output on an Anadex printer. Another system, the Interactive Digital Image t1anipulation System (!DIMS) is mainly an image-processing system. For this reason !DIMS is organized in a raster format and includes many functions that address 22

PAGE 36

problems specific to the processing of digital images, such as image-enhancement routines for remotely sensed data. The DEMs used for this project were processed on !DIMS into a MOSS/MAPS readable format at the BLM Service Center. The MOSS/MAPS package provides flexible routines for overlay and neighborhood analysis, map description, and data management. A main advantage of this package is that it is used and supported by several federal agencies. At present, MOSS/MAPS has limited capabilities for internal inferential statistical analysis. However, data may be transported to a separate statistical package, e.g., SPSS, for analysis. Afterwards, the statistical results can be reintroduced into the system. The version of MOSS/MAPS used in this project operates on Data General minicomputers and microcomputers, using the Advanced Operating System (AOS). MOSS/MAPS is operational on VAX and Hewlett Packard hardware. The advantages and limitations of using MOSS/MAPS for archaeological research and CRM planning will be described later in this report. Previous CRM Studies Using Geographic Information Systems Although many types of spatial analysis of archaeological and historic values have been used in cultural resource planning in a multiple land-use context, few have involved the use of a GIS. Only since the mid-1970s have GISs been used by public land holding agencies for management needs. What follows is a brief review of some of the major CRM locational studies using GISs. 23

PAGE 37

The Wasson Field-Denver Unit co2 Project The Wasson Field-Denver Unit co2 Project was funded by a private oil company in preparing an environmental impact statement (EIS) for a carbon dioxide well-field project in southwestern Colorado. The cultural resource portion of the EIS was needed in part because BLM required a right-of-way permit. Woodward-Clyde Consultants prepared a planning study that would improve well-field layout by reducing impacts to significant archaeological sites. This summary is based on two draft documents (James and others 1983; Woodward-Clyde Consultants 1978). The project area contained 263,158 hectares (ha), which included plateaus, canyons, farmland, rangeland, and forests. Environmental and cultural data were entered, compiled, analyzed, and displayed with a GIS. Map-based information was coded and digitized for 175,000 cells (each about 1.5 ha) for land use and soil association, prehistoric farming areas, topography, roads, archaeological sites (e.g., period, size, type, and condition), biological communities, and geologic materials. Site significance was identified as the dependent variable and defined in part on the basis of age, type, size, and number of components for hundreds of known Basketmaker, Anasazi, and post-Anasazi sites. Fundamental to the definition of significance were the .. subjective .. attitudes of professional archaeologists. The archaeologists developed a seven point scale believed to conform to prevailing opinions of the professional archaeological community. Ultimately, three independent environmental variables--soil, drainage rank, and slope--were used in a step-wise multiple regression, with the computed site significance values being the dependent variable. Sets of surveyed cells without sites were also included in the analysis. The 24

PAGE 38

analysis yielded significance values for each cell, and scaled values were then color-coded and plotted on 1 :24,000-scale maps. A total of 140 cells were randomly visited in the field as means of verification. The model was supported to the extent that the standard error of projected-to-observed value was identical to the standard error of the model (James and others 1983:23). The Passaic River Project The Passaic River Project was funded by the New York District Corps of Engineers. Robert Hasenstab (University of Massachusetts, Amherst) implemented the project through a subcontract with Soils Systems Inc., an environmental c o nsulting firm based in Marietta, Georgia. The project's objectives were to estimate the amounts of cultural materials likely to be affected by postflood-control facilities and to define areas with a high probability of site occurrence (Hasenstab 1983b). The 1,619-ha project area extends 160 linear km along the Passaic River, crosscutting ridge and valley, piedmont, coastal plain, and tidal-estuarine areas. Urban and commercial developments occupy most of the impact zone, but 42 percent is either agricultural, forested, or classed as wetlands. The project area was subdivided into a high-resolution grid of 0.47 ha units (pixels} for which environmental variables were coded. All. manipulation and mapping was performed on a GIS. Univariate statistical tests were employed to determine which environmental variables were most useful for their power to "retrodict" known site locations. Significant 25

PAGE 39

variables were found to be soil drainage, distance to nearest river, distance to minor tributary/river confluence. Grid cells were assigned a sensitivity rating by summarizing the component-variable ratings. The sensitivity models were then tested and revised with data derived from a survey of 300 pixels (ca. 140 ha) representing a stratified random sample (with some modifications). Overall, the sample fraction was about 6.5 percent of the impact zone. The survey techniques included limited but systematic subsurface testing judgmentally selected pixels. Totals of 28 historical sites and 16 prehistoric sites were recorded. A series of computer-generated maps illustrated the final model on a pixel-by-pixel basis for prehistoric archaeological sensitivity (high, medium, or low, based on the component variable ratings) and a combination of historical and prehistoric sensi ti vi ty. Hasen stab ( 1983b: 13) concluded that the GIS approach "has greatly enhanced the capabilities for archaeological prediction and land use management .. but it cannot be taken as a final solution to all cultural resource management problems." Grand Junction Area Project BLM funded the Grand Junction Area Project as a overview of statistical classification procedures for predicting archaeological site locations. This summary emphasizes aspects of the project related to the development and testing of models in the Grand Junction Resource Area. For that area, the objectives were to develop quantitative models that could be used to predict likely locations of prehistoric sites (Kvamme 1983b). 26

PAGE 40

The project encompasses some 438,966 ha of western Colorado uplands. Vegetation types characteristic of the area include desert grasslands, pinyon-juniper woodlands, and spruce-fir forests. The subareas of the district were stratified into five major biotic communities considered to occur in significant proportions across the landscape. A stratified proportional random sample of 65 ha quadrants (quarter sections) were selected from the physiographically defined subareas. One hundred quadrants were selected for survey, specifically to provide the data base for generating the models. The surveyed area amounted to about 1.5 percent of the project area. Environmental data coded for site and non-site locations. Through a series of statistical analyses, the following variables were found to be important in distinguishing between site and non-site locations: biotic zone, vertical distance to permanent water, vantage point distance, slope, view, aspect, shelter within 100m, and shelter within 250m. The models were developed through a pattern-recognition approach using multivariate analyses as classification tools. The most successful analysis was logistic regression. Depending upon the particular approach, GIS-based probability surface maps were generated to illustrate predictions for sites and siteless locations in unsurveyed areas covering from 0.6 to 1 and 1 to 25 ha. The accuracy of the models was tested independently using site-file and non-site data, as well as split sampling techniques. Kvamme•s approach to predictive locational modeling is statistically and computationally one of the more sophisticated attempts at spatial analysis using a GIS. 27

PAGE 41

The Pinyon Canyon Project The Pinyon Canyon Project in southeastern Colorado focused on the investigation of prehistoric settlement patterns within this high plains setting. The project was funded by the U.S. Army. The goal was to develop models of prehistoric site location (Kvamme 1984). For the study, a GIS was established for the entire region, which encompasses nearly 1,000 sq.km. The computer data base used a cell-based GIS of more than 400,000 cells, each 50 m on a side, and more than 20 analytical and management surfaces or themes. The system includes environmental surfaces representing elevation, slope, aspect, measures of local relief, a measure of relative view quality, vantage locations and hydrology network, and horizontal and vertical distances to streams of Strahler order ranks. Management surfaces include the locations of nearly 1,200 archaeological sites, with information on site number, site type, temporal period association, and surfaces depicting field inspected regions, including date of inspection and several management boundaries. The above GIS could retrieve sources and combinations of management and environmental data, such as archaeological information about a site and its environmental properties or scaled maps of any surface or combination of surfaces. One of the chief uses of the geographic data bases in all of the above studies is to examine and test environmental hypotheses about archaeological site locations and to develop settlement pattern models, including models used for site density projection. 28

PAGE 42

CHAPTER III EXISTING ENVIRONMENT Chapter III describes the physiographic and general environmental setting of the project area. Many of the environmental factors discussed in this chapter serve as the primary surfaces for spatial manipulation in the GIS. Location and Physiographic Setting The four-quadrangle study area (Pretty Rock, Tanner Lake, Alamo Mesa East, Alamo Mesa West) lies in the northern part of the San Juan Basin portion of the Colorado Plateau (Figure 3.1). The central part of the basin is elliptical, 160 km north to south and 145 km east to west, and contains 19,425 sq. km (Fassett and Hinds 1971). The project area lies on the west side of the central basin on the Chaco Plateau. The Chaco Plateau is the area north and east of the Chaco River, south of the San Juan River, and west of Canon Largo (Warren l967a). This portion of the Colorado Plateau consists of 11high plateaus and mesas interspersed with broad basins and intervening valleys and arroyos containing ephemeral streams11 (Metric Corp. 1982). The area has a gently sloping upland erosional surface, partially covered by sheet sand and dunes, with badlands exposed where drainages have cut through the upland surface. The elevation ranges from 1,730 m to 2,010 m. Ephemeral streams within the area include Coal Creek, which crosscuts the northeastern and north-central sections and runs into De-na-zin Wash. De-na-zin Wash and Alamo Wash drain the north and northwestern portions 29

PAGE 43

....----' • Carson Trading Post Burnham ., .. .. .. wJ-e::: Traeling-hst ... . . .. ;;,:Cii"Q .;::v .p . ... ' FIGURE 3.1 Location Map 30

PAGE 44

of the study area, respectively, and join 1.7 km west of the project area. De-na-zin Wash then continues about 30 km to the west. To the northeast lies Hunter Wash, which drains the northwest portion of the project area and is a tributary to the Chaco River to the west. The southern portion of the area is drained by Tsaya Arroyo and Ah-shi-sie-pah Wash, which are tributary to the southern portion of the Chaco Wash. Other features within the area are exposed badlands in its east-central and north-western sections; red dog (burned shale} hills in the southeast and west; Tanner Lake in the west; and the end of Alamo Mesa, which projects into the central part of the study area. Climate The project area has a semiarid climate (Ferrill 1978}. The mean annual temperature is 10C, with winter low temperatures below 6C and summer highs above 32C. Mean annual precipitation is 25 em with 50 percent of the precipitation occurring in thunderstorms between July and October (Berger and Lucas, n.d.}. Rainfall is variable within the region. Chaco Canyon records show from 1933 to 1979 a range from 8.2 to 4.8 em (Gutierrez 1980}. June and November are the driest months, with the highest precipitation occurring in the summer, less in the fall, and the lowest in the winter. The project area has from 160 to 180 frost-free days (Ferrill 1978}. Hydrology The project area is in the San Juan River watershed (Dingham 1978} and contains six ephemeral streams: Coal Creek, De-na-zin Wash, Alamo Wash, Hunter Wash, Ah-shi-sie-pah Wash, and Tsaya Arroyo. The only major drainage basin is De-na-zin Wash. Black Lake, located to the southeast, 31

PAGE 45

is a dry Holocene lake. Tanner Lake, another dry lake, may have been modified during histori cal times. Geology and Geomorphology The San Juan is an asymmetrical structural basin containing 4,572 m of sedimentary deposits ranging in age from Cambrian to Holocene. The Cretaceous (120 to 65 million years ago) geologic units within the study area are, from oldest to youngest, Cliff House Sandstone, Lewis Shale, Pictured Cliffs Sandstone, and the Fruitland and Kirtland Formations. Tertiary (65 to 3 million years ago) formations have been eroded from the project area. The Cliff House is a thick (495 m to 990 m), transgressive strand-line sandstone. The units represent long stands of the beach line between the marine Lewis Shale and the continental Menefee Formation. In many places the Cliff House Sandstone conformably contacts both of these units. The Lewis Shale unit is the highest marine shale in the basin and has marine shell fragments including gastropods and ammonites (Marshall and Breed 1974). The dark shale contains scattered interbeds of bentonite, marine sandstone, and calcareous silty concretions with veins of anhedral barite. The Lewis Shale is easily eroded into badlands that are often covered with gypsum crystals and crusts of soluble salts. Along the Chaco River, the Lewis Shale has been redeposited as slopes or terraces with low gradients. The Pictured Cliffs Sandstone was defined from outcrops north of the San Juan River and west of Fruitland. This sandstone forms cliffs in some parts of the study area. Lithologically, the Pictured Cliffs can be 32

PAGE 46

divided into two units: an upper massive sandstone and a lower shaley unit that grades into the Lewis Shale. The sandstone beds are medium to fine grained, well sorted, and composed of 8 percent quartzes, 13 percent feldspars, and 4 percent coal fragments (Burgener 1953). Fossils included in this formation include shark teeth and fragments of teleosts, turtles, and dinosaurs. The Fruitland and Kirtland formations have been defined by Bauer (1916). The Fruitland Formation is the coal-bearing unit in this portion of the San Juan Basin, and the Fruitland/Kirtland contact is gradational and arbitrary. The contact is commonly defined by the limit of the uppermost Fruitland coal deposits. The two formations are similar in flora and fauna. Further work is needed to accurately distinguish the conformable contact in surface expression (Marshall and Breed 1974). The Fruitland is lithologically more complex than the Kirtland and includes 990 to 1159 m of interbedded sandstone, siltstone, shale, clays, carbonaceous shale, carbonaceous siltstone, coal, and thin limestone beds composed entirely of brackish-water pelecypod shells. All beds within the Fruitland are discontinuous and pinch out laterally (Baltz 1967). The Fruitland commonly has a sequence of thin coal seams overlain by fine laminated clays and sandstone and weathers to a badland condition. In some areas, the Fruitland is marked by low hills of baked shale which has been fired to a porcelain-like material by spontaneous oxidation of underlying coals. In a few areas, this porcelain material is of suitable quality to have been used prehistorically for chipped stone tools. The Kirtland Shale is divided into two or three units and varies greatly in thickness throughout the San Juan Basin. The lower and upper shales are often divided by a sandstone layer that is medium to fine 33

PAGE 47

grained. The upper unit of the Kirtland is thin or missing in some areas but is conformable with the overlying Ojo Alamo where contact can be traced (O'Sullivan and others 1972). The fossil inventory from the Fruitland and Kirtland formations is extensive, including fossil wood, leaf impressions, fresh water mollusks (11 species), and many microvertebrate and macrovertebrate remains. Turtles (nine species) and the hadrosaurian dinosaur (Kirtosaurus navajovnus), the most abundant large vertebrates, are typical of late Cretaceous marginal marine assemblages. The Quaternary (3 million years ago) deposits include aeolian soils. The following discussion is based on Schultz's (1980) work on the Chaco dune field. This dune field is bound by Brumhall Wash to the north, the Chaco River to the west and south, and De-na-zin and Alamo washes to the south and east. The northwest part of the study area (north of Alamo Wash) lies within the Chaco dune field, and the remainder of the area is covered by Quaternary alluvial deposits associated with the De-na-zin and Alamo wash systems. Any of five aeolian landforms may occur. Sand sheets, which are "usually thin, well-1 aminated, gently undulating sand bodies that form the flat topography" (Love and Schultz 1980:30), constitute the major landform within the project area. Barchan dunes, which have a "crescentic form with the horns of the crescent extending downwind" (Schultz 1980:30), are active dunes that are restricted to a 5 km zone downwind from the Chaco River (Schultz 1980:30}. No examples of this landform exist in the study area. Parabolic dunes, which have a crescent shape similar to that of barchan dunes but with the horns point ing toward the dominant wind direction (Love and Schultz 1980:30}, occur in the 34

PAGE 48

northwest and east sections of the project area. The south landform, linear dunes, consists of 11aeolian deposits that are elongated parallel or nearly parallel to the dominant wind direction•• (Schultz 1980:39). Although there are dunes within portions of the surveyed areas with a linear form, these are actually the detached arms of parabolic dunes. Finally, ridge dunes are 11accumulations of wind-blown sand that form large irregular mounds along the dissected edges of upland surfaces11 (Schultz 1980:46-48), for example along badlands. In the central portion of the four quadrangle study area, the dunes are generally oriented southwest to northeast, parallel to the dominant wind direction. This pattern has been observed by Hack (1941) and by Cooley and others (1969) across the Navajo Reservation, as well as by Schultz (1980). Schultz also found that aeolian landforms within the Chaco dune field represented 88 percent sand sheet, 8 percent parabolic dunes, and 4 percent other forms. Coppice dunes, 11mounds of wind-blown sand that form around clumps of vegetation11 (Schultz 1980:50), occur on sand sheets and along the crests of linear and parabolic dunes, creating a humocky appearance in sand deposits observed in isolated locations in the study area. The aeolian deposits are generally considered to be of late Pleistocene and Holocene age (Cooley and others 1969; Hack 1941; Hall 1979). Schultz (1980) and Wells (1982) have defined three major periods of aeolian deposition: Aeolian Unit 1 dates from the late Pleistocene (7,000 years ago); Aeolian Unit 2 dates from 2,800 to 7,000 years ago; and Aeolian Unit 3 dates from 1,500 years ago to the present. Wells (1982:136-138) notes that Archaic sites do not occur in Unit 1 but are present within Unit 2. The parabolic dunes are the oldest landform, 35

PAGE 49

dating from late Pleistocene to recent Holocene, while linear and barchan dunes originate in the recent Holocene. Studies that correlate aeolian depositional units with cultural-temporal units include Hall (1979) and Wells (1982). The sources of the sand for the Chaco dune field are Upper Cretaceous sandstone outcrops (the dominant source since the Pleistocene) and sediments of the Chaco River (Schultz 1980). The sandstone is a more important source within the survey area, but some local sand from washes and sand sheets is also reworked and redeposited (Love and Schultz 1980). Local Lithic Raw Materials The Chaco Plateau contains terrace and stream gravel, which is a major source of such lithic materials as quartzites, cherts, sandstone, and igneous cobbles (Warren 1967b:ll8). Chapman (1977:429), who used Warren•s lithic type codes in a later analysis, describes the lag gravels in the lower Chaco River area as ••waterworn nodules and cobbles of silicified woods, cherts, chalcedonies, and quartzites ... Surface gravel deposits have been found at several locations in the study area: along eroded drainages; along the eroded edges of dunes at dune-badland contacts, where dunes were cut by drainages; in interdunal areas on eroded/deflated hardpans and playas; and on exposed badlands. The first four locations contain surface gravel/scatters of varying sizes. Gravels on exposed badlands occur in isolated pockets. Overall, gravel deposits appear to lie above the badland formations and below the aeolian deposits. The original source of these surface lag gravels is probably the Ojo Alamo Formation, where they were reworked and deposited by alluvial action during the Pleistocene (Vierra and others 1986). Lithic raw materials within the project area include chert, siliceous and 36

PAGE 50

nonsiliceous petrified woods, chalcedonies, quartzites, sandstone, igneous rocks, and claystone (Vierra and others 1986). Soils t1ost of this discussion on soils is based on the soils inventory performed by the Soil Conservation Service (SCS) and documented in Soil Survey of San Juan County New Mexico, Eastern Part (1980). Soils are natural bodies at the surface of the earth that support or are capable of supporting plants (Daugherty and Buchanan 1981 :3). Soil formation is a function of climate and biota, parent material, topography, and time. Most soils within the study area formed from alluvium and aeolian sediments derived from shale and related sandstones. The soils are very young, having formed in the late Pleistocene and Holocene eras. The soil types and associated map symbols are shown in Table 3.1. These soil mapping units are the result of the SCS Order 3 range survey completed at the series level of mapping precision. Flora Much of the project area lies in a heterogenous region of the Great Basin sagebrush shrublands in the grama/galleta (Bouteloua-Hilaria) steep association (Morain and others 1977). The main vegetation types in the study area and their associated landforms are shown in Table 3.2. In of this study, all the vegetation units are described by characteristic vegetation units associated with their respective SCS soil mapping unit names (Table 3.1). 37

PAGE 51

TABLE 3.1 SOIL TYPES AND ASSOCIATED RANGE SITE DATA FOR THE PROJECT AREA Soil Name/Symbol Range-Site Characteristic Vegetation Phisiogra2hic % SloEe Elevation (m) Badland/BA barren shale uplands 5-80 1,455-2,182 with deep drainages and gulleys Blancot Loamy ND-1 Big sagebrush (Blancot) fans and upland 0-5 1,697-1,939 Notal Assoc. Galleta, Indian ricegrass, (Notal) fans and Gently Sloping/BT Fourwing saltbush valley bottoms Western wheatgrass Doak Loam/DC Loamy ND-1 Blue grama, Indian ricegrass, deep and well-drained 3-5 1,697-1,939 Big Sagebrush, Needleand-on mesas, plateaus, and thread, Western wheatgrass terraces Galleta w co Doak Avalon/ON Loamy ND-1 Big sagebrush, Needleand-mesas, plateaus, 0-5 1,697-1,939 thread, Blue grama, and terraces Western wheatgrass Galleta Indian ricegrass Doak-Sheppard Loamy ND-1 Big sagebrush, Needleand-mesas, plateaus, 0-15 1,697-1,939 Shiprock Assoc. , thread, Blue grama and terraces Rolling/ OS Western wheatgrass Galleta, Indian ricegrass

PAGE 52

w 1..0 Soil Name/Symbol Duneland/DZ Fruitland Persayo-Sheppard Complex Hilly/FX Huerfano Muff-Uffens Complex/HU Riverwash/RA Rock Outcrop/RO TABLE 3.1 (continued) SOI L TYPES AND ASSOCIATED RANGE SITE DATA FOR THE PROJECT AREA Range-Site Sandy ND-1 Sodic Slope Characteristic Vegetation Scattered vegetation mainly in interdunal areas Indian ricegrass Blue grama, Big sagebrush Fourwing saltbush Giant dropseed Alkali sacaton, Fourwing saltbush, Galleta, Polack greasewood Little or no vegetation due to frequent flooding and reworking by fluvial action No vegetation Physiographic Setting mesas, plateaus, major drainageways, deep excessively drained active dunes hills, mesas, plateaus, fans, and breaks mesas and valleys unstablized sandy, silty, clayey, or gravelly sediments on flood plains, streambeds, riverbeds, and arroyos exposures of barren sandstone on cliffs; breaks, bluffs, and ridges % Slope Elevation (m) 5-25 1,455-2,182 5-30 1,455-2,182 0-8 1,697-1,939 0-3 1,455-2,182 5-100 1,455-2,182

PAGE 53

TABLE 3.1 (continued) SOIL TYPES AND ASSOCIATED RANGE SITE DATA FOR THE PROJECT AREA Soil Range-Site Sheppard Huerfano Deep Sand ND-1 Notal Complex Gently Sloping/SC Shiprock Fine Sandy ND-1 Sandy Loam/SO Sheppard Mayqueen Deep Sand ND-1 Shiprock Complex/SO Characteristic Indian ricegrass, Giant dropseed, Alkali sakaton, Needleandthread Indian ricegrass, Blue grama, Big sagebrush, Giant dropseed, Fourwing saltbush Indian ricegrass, Giant dropseed, Alkali sakaton, Mormon tea, Sand dropseed Source: Soil Survey of San Juan County, New Mexico: Eastern Part 1980 SCS (1980) valley bottoms, fans, mesas, plateaus mesas and plateaus mesas and plateaus in loamy fine sand % Sloee Elevation (m) 0-8 1,455-2,182 2-5 1,697-1,939 0-8 1,697-1,939

PAGE 54

TABLE 3.2 MAJOR VEGETATION UNITS AND ASSOCIATED LAND FORMS Rice grass/ R1 ce grass/ Big sagebrush/ Landform Galleta grass Big sa9ebrush Blue grama Plains Badlands Erosional Areas Hi 11 Landforms: Plains X X X X X nearly level surfaces with some undulations or dissections Badlands -high density of gullies, ravines and sharp-backed . ridges; usually clay or shale Erosional nearly devoid of vegetation with Areas alkaline or saline soils Hill moderately steep to steep slopes X X X X and somewhat rounded relief of less than 151 meters Source: Rangeland Resources International, Inc. (1978) 41 Shadscale/mixed saltbrush-alakali X X sakaton

PAGE 55

Fauna Modern animal populations appear to be small. The birds present include horned larks, Says phoebes, loggerhead, shrikes, scaled quail, common raven, mountain bluebird, and various hawks. Reptiles include various lizards and rattlesnakes. The mammals present include the prairie dog, cottontail rabbit, jackrabbit, kit fox, and coyote. Parts of the study area are used for grazing cattle and sheep (Berger and Lucas 1972). 42

PAGE 56

CHAPTER IV THE ARCHAIC PERIOD The Archaic period is unique within the culture history of the San Juan Basin. The hunter-gatherer mode of subsistence of this period persisted for about 6,000 years--the longest adaptation in the Basin•s history. This chapter focuses on previous research of the Archaic in the Basin and ethnographic examples of hunter-gatherer systems with organizational strategies similar to those postulated for the Archaic. Since the Archaic is the best represented cultural period in the study area, the Archaic site data base will be used for identifying land use and cultural resource conflict areas with GIS technology. The prehistory of the study area can be divided into five basic cultural/temporal periods: . the Paleo Indian (9,500 BC to 5,000 BC) the Archaic (5,500 BC to AD 400) the Anasazi (AD 1,000 to AD 1,300) the Historic Navajo (AD 1,500 to 1930) and the Anglo (1846 to present). Only the Archaic period is discussed here. Overview of the Archaic The following summary of the Archaic period is based on the information from a recent study of a portion of the study area by Vierra and others (1986). The Archaic culture has been defined as a 11broadly based mixed gathering and hunting economy11 (Irwin-Williams, 1973:4) and as groups that manipulated higher life zones and conducted intensified food production (Stuart and Gauthier 1981 :33). It was a post-Pleistocene 43

PAGE 57

adaption. Lee and Devore (1968) describe general hunter-gatherer systems as small flexible groups of 25 to 30 individuals and a population density of 1 to 25 people per 260 sq. km. Each of these small groups is associated with a geographical region but interacts with other groups, creating a large breeding and linguistic community. The economic system includes such features as a home base, a sexual division of labor (with men hunting and women gathering), and a pattern of sharing of food and other goods. The term "Archaic'' is sometimes used to denote a specific cultural/temporal unit; at other times it describes a general cultural adaptation. This discussion deals with the Archaic as a cultural/temporal unit. It touches on the transition from Paleo Indian to Archaic, reviews the Oshara chronology from a regional perspective, and considers Archaic settlement-subsistence system and site organization studies, with an emphasis on site reoccupation. Finally, the discussion considers the transition from mobile hunter-gatherer to sedentary agriculturalist and the coexistence of hunter-gatherer and agricultural strategies. In northwest New Mexico the Archaic has been defined as a hunter-gatherer adaptation, the Oshara tradition, which lasted from about 550 BC to AD 400 (Irwin-William 1973). Irwin-William (1973) considers the Archaic hunter-gatherers to be predecessors of the Anasazi. Although her work centered in the Arroyo Cuervo area, southwest of San Ysidro, New Mexico, projectile points similar to those in the Arroyo Cuervo region occur throughout the San Juan Basin. Irwin-Williams states that Archaic 44

PAGE 58

assemblages 11differ so greatly in technology, typology, and functional classes from those of the preceding Cody and other Paleo Indian phases, that there is evidently no generic connection between them11 (1973:4-5). Because of similarities between Lake Mojave Complex of southern California and the Jay Phase, she suggests that the beginnings of the Oshara tradition represent the eastward movement of western-based Archaic groups into northwest New Mexico at about 6,000 to 5,000 BC and that these groups occupied the area abandoned by the plains-based Paleo Indians. This interpretation has recently been questioned by Stuart and Gauthier (1981:33), who argue that the Archaic may be an indigeneous development from the Paleo Indian. The Oshara tradition provides the Archaic period with a chronological framework, in part based on projectile point typology. This tradition has been divided into five phases: Jay (5,500 to 4,800 BC), Bajada (4,800 to 3,200 BC), San Jose (3,200 to 1,800 BC), Armijo (1 ,800 to 800 BC) and En Media/ Basketmaker II (800 BC -AD 400). Irwin-Williams• (1973) generalizations concerning the Arroyo Cuervo area, however, can be considered representative of only a portion of the regional Archaic settlement-subsistence system. The notion of a regional approach to the study of past cultural systems was advocated by Lewis Binford in 1964. But few invest i gators have yet applied this approach to settlement-subsistence systems in the San Juan Basin. Several classes of data could be used in developing such a regional perspective, including projectile point distributions, intrusive material types, ethnographic information, and subsistence data. Hunter-gatherer groups tend to interact across a broad geographic region, allowing individual groups to collect and exchange information on 45

PAGE 59

the location and seasonal availability of resources. This interaction could be indicated by the distribution of projectile point styles. Oshara tradition points are distributed throughout the San Juan Basin and in the adjacent areas of southeast Utah and southwest Colorado, the Jicarilla Reservation, the Jemez Mountains, Cebolleta Mesa, and in the Defiance Plateau and Chuska Mountains. Binford (1979:259) stated that "raw materials used in the manufacture of implements are normally obtained incidentally to the execution of basic subsistance tasks. In other words, procurement of raw materials is embedded in basic subsistence schedules." Given this perspective, the intrusive materials in an area can provide clues to a group•s seasonal movements and to their settlement-subsistence region (Clark 1976:43; Issac 1977:94-95). Intrusive materials on sites near the project area include Jemez and Polvadera obsidian (Chapman 1977; R. Moore 1982; Gomolak 1980) and Cerro Pedernal chert (Acklen and others 1982; P. Moore 1982) . The ethnographic record is replete with documentation of the movement of hunter-gatherer groups between higher and lower elevations to increase seasonal availability of resources (Krober 1925; Powers 1877; Steward 1938). Steward (1938:52) noted that the Shoshonean territory "embraced the various life zones, thus providing all possible local varieties of essential foods." It is not surprising, therefore, that the intrusive lithic materials found in the project area appear to have been procured when local groups were exploiting resources at higher elevations. These uplands were the focus of a portion of a regional settlement-subsistence system. Lowlands 46

PAGE 60

such as the project area were the focus of activities for other portions of the system. Other researchers have discussed the recent use of both higher and lower elevation areas. Irwin-Williams (1973) reported that 11two kinds of special activity sites are known outside the Arroyo Cuervo region: isolated hunting camps in the Jemez Mountains and repeated quarry workshop camps... More recently, Stewart and Gauthier (1981 :407) suggested that 11Archaic sites ••. in higher elevation(s) contain more diagnostic points than do lithic scatters in many of the extensive dune settings. It may be that two kinds of site locations reflect the collecting versus the hunting strategy, and are seasonably distinct assemblage sets of the same population ... The limited evidence of Archaic use of plants in the northern San Juan Basin has been summarized by Toll and Cully (1983). Their information, which was derived from the Navajo Mine Archaeological Program (NMAP) and the Navaho Indian Investigation Project (NIIP), suggests the exploitation of Indian ricegrass and mustard in the late spring/early summary and the exploitation of dropseed, goosefoot, and pigweed in the mid-summer/early fall. They have developed a settlement-subsistence model of Archaic seasonal movement involving summer occupation of lower elevation dunes (based on archaeological evidence), with movement into high elevations during the fall for pinyon collecting and possible winter residence (Steward 1938). The Archaic settlement-subsistence system has been viewed in various ways. Winters (1969:110) defined a settlement pattern as 11the geographic and physiographic relationship of a contemporaneous group of sites within a single culture11 and a settlement system as 11the functional relationship 47

PAGE 61

among the sites contained within the settlement pattern." In other words, settlement patterns are studied on the intersite level, and settlement systems are studied on the intrasite level. Studies of settlement patterns test ecological and locational hypotheses about the causal variables affecting the selection of site locations. Studies of Archaic site locations have stressed three environmental variables: (l) vegetation diversity (Elyea and others 1979; Sessions 1979; Reher and Witter 1977); (2) vegetation diversity and proximity to water (Elyea and others 1979; Ganas 1980; J. Moore 1980; Powers 1979); and (3) aeolian soils (Ganas 1980; Huse and others 1978; Powers 1979; Reher and Witter 1977; Sessions 1979). Kemrer (1982:103) summarized these three factors when he observed that locations of lithic sites seem to correlate with soils that largely support grasses and with areas where seed grasses are differentially distributed. He noted that lithic sites are most likely to occur in areas with a mixture of alluvial and aeolian soils. Reher•s (1977) study of the lower Chaco River presents an early view of the Archaic settlement system in the northern San Juan Basin. Reher suggested that this system was composed of "two basic types: campsites (i.e., habitation), and several kinds of limited activity sites where various plant and animal resources were procured and sometimes processed in areas surrounding habitation sites" (1977:96). He further observed that the lithic assemblages form a continuum from large habitational sites (which show a high degree of functional variability and frequent presence of hearths) to small special use sites (with little functional variability and few or no hearths). 48

PAGE 62

In a later study, Vierra (1980a) argued that the "continuum" actually represents only one part of the settlement system, the limited basecamp, and that the difference in size and content is mainly due to variations in the length and number of occupations of the site area" (1980:356). He went on to say that the only remains suggesting task-specific sites are isolated artifacts. Although Kemrer (1982:19) suggested that large camps represent a seasonal macroband (i.e., a population composed of several microbands) phase, large sites are more likely to represent locales reoccupied by microbands (i.e., the small primary foraging units) and are not single occupational episodes representing microband encampments (i.e., large aggregations of microbands or central basecamps). This view of microband reoccupation has been supported by Eschman (1983) for the NMAP project area, which lies north of the present survey area. Other researchers, however, have taken a different approach, defining multiple archeological site types on the basis of assemblage content and specific site function. Suggested types include (a) large/main camps, (b) small camps (c) plant processing sites, (d) plant procurement sites, (e) hunting camps, and (f) lithic procurement sites (Kemrer 1982:18-23). Huse and others (1978), Powers (1979), Sessions (1979) and Ganas (1980) also provide examples of multiple site type definitions. Little research has been conducted toward an understanding of Archaic group size, structure, and organization. Site structure studies can provide information relevant to these questions, but few such studies exist. Vierra (1980b:385) described a limited pattern recognition study that defined four possible patterns in the organization and use of space on 49

PAGE 63

Archaic sites: (a) an association among ground stone, fire-cracked cobbles, and hearths; (b) areas with the highest density of nonutilized chipped stone debris (debitage) also being the areas with the highest density of utilized debitage (which shows that debitage was produced, used, and discarded in the same location); (c) nonpatterned scatters representing refuse areas or areas affected by postdepositional processes; and (d) activity/refuse areas consisting of large patterned locales that included evidence of multiple activities. This fourth pattern is probably a result of reoccupation and reuse of site locations. Vierra (1981) proposed a model for the organization of an Archaic camp. From his structural study of a Basketmaker II site, this model suggests that an Archaic camp should consist of (a) a hearth area that was the focus of camp activities and that exhibits the greatest amount of functional variability as indicated by lithic reduction debris and (b) a separate lithic reduction locus with some expedient tool use. See Chapman (1980) and Camilli (1979) for other examples of site structural studies. Preventing an understanding of Archaic sites is the problem of identifying and interpreting reoccupied site locations. The multiple activity loci visible on many sites may reflect (a) separate but contemporaneous residential units, (b) contemporaneous but functionally differentiated activity locales, (c) multiple occupational episodes possibly involving the previous two patterns, or (d) differential surface exposure of all of the above patterns. Reoccupied sites are often identified by absolute or relative dating techniques, the existence of multiple stratigraphic levels, and the extensive site size with multiple activity locales. A suggested avenue of research would be to determine 50

PAGE 64

the organization of the residential unit by trying to isolate redundant patterning in the archaeological record (Vierra 1980b, 1981). Once a redundant unit has been identified, this information can be used to interpret the structure of sites with multiple activity locales. A major interest in modern anthropological theory involves the question of what causal variables led to the transition from mobile hunting and gathering to sedentary agriculturalist adaptation (Reed 1977). Some of the suggested causes include environmental change, climatic variability, and population pressure. The Archaic exhibits an intensification of the economy through time, which appears to be characteristic of the transition to agriculture (Cohen 1977 and Klein 1977). But the causal variables that led to this intensification have yet to be identified. Recent work by Eschman (1982) in the NMAP project area suggests that a post-Archaic hunter-gatherer strategy coexisted with the Anasazi adaptation. This hypothesis is based on the association of four late radiocarbon dates (between AD 1280 and 1610) with assemblages that would otherwise be considered Archaic. Eschman proposed that initial Anasazi populations displaced Archaic groups in the NMAP area but that once the Anasazi abandoned this territory, the hunter-gatherer groups returned (Stuart and Gauthier 1981). 51

PAGE 65

CHAPTER V ASSESSING ARCHAEOLOGICAL INVENTORY AND ENVIRONMENTAL DATA BASES This chapter focuses on the GIS-aided methodology used in evaluating the archaeological inventory and environmental data for potential biases. Administrative themes (areas of planned and existing development or conservation) are also analyzed according to the amount of inventory conducted for them. Chapter V provides the rationale for selecting the four-quadrangle study area in the northern San Juan Basin. It also briefly describes the planned and existing development in the area, and discusses BLM inventory classes as they relate to the project area and BLM inventory criteria for the FRA. Next, Chapter V describes the problems of and potential biases involved in using existing site and inventory data. A description is then given of the specific environmental and administrative themes used for analysis using the GIS. The next section focuses on the procedures used to evaluate the project inventory data with the environmental and administrative data using MOSS/MAPS. A cartographic model is included to aid the reader in following the steps used for each analytical procedure. Finally, the results of the analysis are interpreted with the aid of tables and maps. Selecting the Study Area The four-quadrangle study area lies in the northern San Juan Basin of New Mexico, within the Farmington Resource Area (FRA). Since the BLM in New Mexico has a large digitized data base and the Data Synthesis 52

PAGE 66

Project (Chapter I) has begun, the state of New Mexico was a logical choice for a cultural resource and land use study involving a GIS. The reasons for choosing this particular study area are threefold. First, it lies in an area where multiple land use is planned or occurring. The extraction of coal and oil and gas, in particular, presents a source of possible conflict with cultural resources occurring nearby. One of the aims of this project is to evaluate the potential of using a GIS in a multiple-use setting where cultural resources are involved. Second, existing contributory surveys (see below) provide an adequate sample for statistical analysis. Within the four-quadrangle project area, 49,308.7 acres (29.23 % ) had been inventoried. An inventory level of at least 20 percent was determined to be desirable for this study. In this way, the nature of potential inventory bias in existing data could be described. Even though most resource areas have inventory levels closer to 5 percent, the same or similar biases would be expected. Third, most of environmental and administrative themes for the study area had previously been digitized at the BLM New Mexico State Office and the Albuquerque District Office. Because none of the one administrative themes and only three environmental themes had to be digitized or procured from an outside source, the cost of conducting the study was significantly reduced. The use of existing data bases for archaeological analysis via a GIS is of major interest to cultural resource managers and land use planners, particularly because limited budgets often make it difficult to procure digitized maps. 53

PAGE 67

Planned Development in the Study Area The four-quadrangle project area of Pretty Rock and Tanner Lake to the south and Alamo Mesa East and Alamo Mesa West to the north has an extensive amount of proposed mineral development. The study area comprises lands of three different statuses relative to coal leasing in the northern San Juan Basin: (a) preference right lease applications; (b) industrial expressions of interest (proposed competitive tract leases); and (c) areas of existing but inactive leases. There are also other types of proposed and existing development under lease. Oil and gas leasing tends to be restricted to the northern half of the study area. Range allotments are found in smaller portions of the area. A few unimproved dirt roads can be found as well as pipe and transmission lines. The study area also contains two wilderness areas and two proposed areas of critical environmental concern (ACEC). BLM Inventory Levels The BLM inventory levels are described in Chapter I. Most of the surveys in the project area were conducted at the Class III level as intensive survey units selected as part of Class II surveys in and near the project area. This chapter presents the results of the analysis of Class II and Class III inventory data using a GIS to address Class II inventory objectives. These objectives are given in the BLM cultural resource manual; (BLM 8100-CRM Manual, 1978). 1. Discovery, recognition, or elaboration of patterns of past human use and occupation of given regions. 2. Determination of the cultural resource potential of an inventory area. 54

PAGE 68

3. Prediction of zones of greater or lesser activity by past human populations. 4. Identification and assessment of the environmental and/or cultural variables, or combination of variables, that form the most accurate predictors of cultural resources. 5. Development of projections of expected density distribution and diversity of cultural resources. 6. Discovery of the range of cultural resource variability within an inventory area. 7. Provision of an objective means of assessing the existing cultural resource inventory. 8. Development of a syste ' matic base for planning decisions concerning cultural resources. The analysis of the existing inventory data should eventually aid in meeting most of these Class II objectives. Specifically, the inventory base has been analyzed to determine areas where future surveys need to be concentrated to provide more representative samples of cultural site and environmental data in the project area. In this manner, one can assess and measure the biases associated with particular environmental characteristics resulting from non-random survey units. Many intensive surveys have been conducted in and near the study area, especially for 11Clearing11 oil and gas drill pads, pipelines, and access roads. The clearance of an area slated for development is required to meet federal compliance requirements. In addition, portions of coal tracts and areas for coal lease applications and expressions of interest have also been inventoried. The major surveys and portions of major surveys occurring within the study area include projects by Huse and others (1978), Chapman and Biella (1980), Vogler and others (1982), Harlan (1982), Vierra and others (1986), and various BLM intensive surveys. 55

PAGE 69

The FRA and BLM Albuquerque District Office have established six criteria for determining whether archaeological inventory reports are contributing or not contributing valid survey information. This study used only digitized contributing inventory information. The following types of reports are considered non-contributing in the FRA: 1. Known field operations of a cultural contractor did not yield contributing information. For example, a pre-1982 operation of San Juan College consisted of moving planned locations of gas well/ pipeline to avoid cultural resources. This procedure resulted in a lower density of cultural evidence . in the records than actually exists. 2. The description in reports is inadequate for field methods used. Field methods not acceptable to BLM include survey by vehicle, aerial surveying, and transects too far apart for terrain and site density. 3. Inventories were conducted by unauthorized individuals or groups, e.g., geological survey or contractors with expired permits. 4 Excessive actions were inventoried during each day. 5. Inadequate observation fails to identify cultural resources present as verified by a BLM field check. 6. Reports mention the presence of artifact assemblages but do not record these as sites. Problems and Biases in Using Existing Site and Inventory Data When examining existing archaeological data bases within a region, data often greatly varies. This variation stems from several factors, ranging from differing standards of quality or practice between different archaeologists, to changes through time in accepted field practices, to variability in the goals and research plans of individual survey projects. The following is a general discussion of potential sources qf 56

PAGE 70

bias and variations in data quality that has been summarized from (1985). Variation, bias, and inconsistency in existing site data bases result mainly from the ways in which different field projects, archaeologists, and field crews perform fieldwork and define, identify, and record archaeological sites. Operational problems can result in defining sites from diffuse scatters of artifacts. For example, lithic sites defined by one project may not constitute sites according to another project's definition. Inconsistencies in regional data bases result not only from the lack of standard archaeological procedures, such as field methods and operational definitions of sites, but from differences in research goals from project to project. For example, one project might focus its inventory methodology on addressing research goals dealing with hunting and gathering adaptations, whereas another project (in the same area) might be more concerned in inventorying cultural sites that pertain to sedentary agriculturalists. Even within a single project, the potential for introducing bias and inconsistencies exists. Sites might, in practice, be defined differently owing to differences in the quality of individual field personnel and crews or because of other factors such as adverse weather, rough terrain, or a fatigued survey crew. Budgetary constraints can also influence the quality of data collected. For example, when a contractor with a fixed price contract finds site densities to be greater than expected, the survey may be rushed. Schiffer and Wells (1982:346) note that "this is accomplished by 57

PAGE 71

increasing crew spacing or reducing the recording time11• These practices lower the quality of the resulting data. Schiffer and Wells (1982) summarized several factors influencing archaeological inventory data quality . A main factor is survey intensity or crew spacing. Crew spacing not only affects site discovery rates but also the sizes of discovered sites (Plog and others 1978). Crews tend to miss small sites and cultural features when crew spacing is large (Wandsrider and others 1984) and small sites are not necessarily less important than larger ones. Narrow spacing dramatically increases survey time and effort, and also increases costs. It should be remembered that small sites are not necessarily less important than larger ones. Given a specified level of survey intensity (Schiffer and others 1978), the nature or obtrusiveness of the archaeological evidence determines the likelihood that a particular archaeological feature, such as a site or an artifact, will be discovered. An architectural feature, for example, has a higher chance of discovery than a single, isolated flake. Low intensity surveys (those with wide spacing) tend to bias resulting archaeological samples in favor of more obtrusive remains (Schiffer and Wells 1982). Another source of bias that affects the obtrusiveness of cultural remains involves geologic processes . Although the pattern of landform and drainages in an area may not have changed since prehistoric times, alluvial material may have been deposited. Deeply buried _sites i n the areas of deposition will bias the sample away from locations in valley floors. Erosion, on the other hand, may destroy sites on steep slopes or along meandering streams, thus biasing the away from steep slopes and drainage locations. 58

PAGE 72

Modern land use can also influence site visibility. Cultural materials are often exposed in areas of cultivation or surface disturbance. In nondeveloped areas such as remote arid regions, good site visibility is sometimes lacking because of natural processes that tend to obscure sites. Difficulty of access, a common problem in many regions of the country, might mean that samples are biased against difficult-to-reach regions. Surveys are commonly located within a specified distance from access roads. Even when hard-to-reach places can be reached, the small amount of time left in the day after travel might lower the quality of surveys in those regions. Private land ownership presents similar difficulties when landowners refuse access (Schiffer and Gumerman 1977:187). Where most archaeological survey work tends to be conducted on federal or state lands, the lack of comparable site data from private properties presents a severe source of bias to regional archaeological data bases. Private property often includes some of the best farmlands and the best areas for hunting and plant collecting. Variable archaeological visibility, due mainly to plant cover, introduces another major source of potential bias. Land cover, however, is not a major problem in most arid regions of the country. One of the main weaknesses of using existing data is that the sum total of previous work in a given region constitutes an unplanned effort. In other words, strong locational biases typically exist in the areas that have been field inspected within a region. For example, early work was often conducted only at the most accessible and visible sites, whereas much contemporary survey is conducted in areas of planned 59

PAGE 73

development. Thus, existing site data may be strongly biased toward certain types of settings and may not constitute a representative sample of sites within a region. Sites also might be inaccurately located on maps. For modeling approaches that focus on the specific locations of sites, accurate placement of sites on maps is of critical importance. Characteristics of the actual locations, such as environmental properties, are often used as a basis for modeling. It is often hard to locate oneself precisely in the field, particularly in areas with few nearby landmarks. Field crews sometimes get lost or misread maps. Moreover, early archaeological surveys often did not have access to good maps and offered only verbal descriptions, directions, and rough locational sketch maps. This problem is further compounded as site locations are transferred from map to map. Data Themes For this report, a theme is defined as the name of a specific environmental or administrative map layer used in analysis. On the basis of its characteristics, each feature or cell in a map is assigned an identifier called a subject. For example, a soil thematic map could consist of several soil types. Each soil type would be considered a subject for the soils theme (see Figure 5.1). Figure 5.1 also shows the environmental and administrative themes used in the assessment of the archaeological inventory data. Each theme is briefly discussed below. See Chapter VIII for a discussion of problems of acquiring and preparing some of the data sets for analysis. The elevation was derived from 1 :24,000-scale, 7.5 minute digital elevation models (DEMs) at 30m X 30m cell size. The DEM s were 60

PAGE 74

DATA THEMES A THEME IS THE NAME OF A SPECIFIC ENVIRONMENTAL OR ADMINISTRATIVE MAP LAYER USED IN ANALYSIS . ON THE BASIS OF ITS CHARACTERISTICS, EACH FEATURE OR CELL IN A MAP IS ASSIGNED AN IDENTIFIER CALLED A SUBJECT. A SOIL THEMATIC MAP COULD CONSIST OF SEVERAL SOIL TYPES OR ASSOCIATIONS; EACH SOIL TYPE WOULD BE A SUBJECT. SOME THEMES DO NOT HAVE DIGITIZED DATA FOR ALL FOUR QUADS IN THE STUDY AREA. ENVIRONMENTAL AND ADMINISTRATIVE THEMES USED TO ASSESS THE ARCHEOLOGICAL INVENTORY DATA: ENVIRONMENTAL DRAINAGES STREAMS ADMINISTRATIVE WILDERNESS STUDY AREAS (WSA) GEOLOGIC SURFICIAL PROCESS UNITS SURFACE GEOLOGY SOILS *ELEVATION *sLOPE *ASPECT LINEAR DEVELOPMENT IN WSA ROAD NETWORKS RANGE ALLOTMENTS OIL AND GAS LEASES PREFERENCE RIGHT LEASE APPLICATIONS FOR COAL INDUSTRY EXPRESSIONS OF INTEREST IN COAL LEASING PROPOSED AREAS OF CRITICAL ENVIRONMENTAL CONCERN LAND STATUS (OWNERSHIP) * ELEVATION WAS DERIVED FROM DIGITAL ELEVATION MODE L S (OEMS). SLOPE AND ASPECT WERE DERIVED FROM ELEVATION MAPS USING SLOPE AND ASPECT COMMANDS IN MAPS, RESPECTIVELY. FIGURE 5 . 1 D ata Themes used for analysis 61

PAGE 75

procured from the United States Geological Survey (USGS). The OEM data was processed through the Interactive Digital Image Processing System (!DIMS) at the BLM Service Center in Denver. See Chapter II for a discussion of !DIMS. Since the OEM data is inherently raster in format, the MOSS cell processing subsystem, MAPS, is the main tool for the manipulating digital elevation data. The SLOPE command in MAPS was used to access elevation data, i.e., a 7.5 minute OEM cell map, for computing slope (in percent rise over run) to create a new slope map. The ASPECT command in MAPS can also access elevation data and compute azimuthal aspect or direction of surface slope to create a new aspect map. The geologic surficial process units and surface geology themes were digitized from mylar maps created by graduate students of Dr. Steven Wells, a geomorphologist at the University of New Mexico. The geologic surficial process units theme shows polygon geologic mapping units that represent geologic processes affecting landscape morphology. The surface geology theme also has polygon mapping units, but these units show where stratigraphic units occur within portions of the study area. The soils data was derived from a soils inventory performed for BLM by the Soil Conservation Service (SCS) and documented in Soil Survey of San Juan County, New Mexico, Eastern Part (1980). The soil types and associated map symbols are shown in Table 3.1. The soil mapping was performed at the SCS Order 3 range survey completed using the series level of mapping precision. Four primary drainage basins were digitized as polygons for the 62

PAGE 76

drainage theme. These drainages were delineated by major arroyos and tributaries that feed into common basins. The stream data was digitized as line data. Several major streams are ranked according to Strahler•s rank-order systems. Six orders of streams occur within the project area. The soils, geologic surficial process units, surface geology, drainage, and stream data are on file in digital form at the BLM New Mexico State Office in Santa Fe. The administrative themes used in assessing the inventory data include wilderness study areas (WSA), linear development in proposed WSAs, road networking, range allotments, oil and gas leases, preference right lease applications for coal, industry expressions of interest in coal leasing, proposed areas of critical environmental concern (ACEC), and land status (ownership) (see Figure 5.1). Data for all of these digitized themes are on file at the BLM New Mexico State Office. The WSAs and the proposed ACECs were presented in the 1981 BLM Management Framework Plan (MFP) for the FRA. The plan had two WSAs, Bisti and De-na-zin, which fall partly in the badlands area. In addition, the plan contained data on two proposed ACECs, Bisti/De-na-zin and Fossil Forest. In 1984 the two WSAs were designated by Congress as wilderness areas. The two proposed ACECs are still pending designation as ACECs by BLM. This GIS study used the digitized boundaries of the WSAs and proposed ACECs even though the WSA boundaries changed slightly after receiving wilderness designation (October 30, 1984). The two proposed ACEC boundaries are the same as included in to the MFP decisions of 1981. The proposed Fossil Forest ACEC is now encompassed within the 63

PAGE 77

Fossil Forest Natural Research Area, which was also designated on October 30, 1984. Preference right lease applications {PRLA) for coal show where applications have been made for coal leasing. Thi s information will be used for developing action plans for coal development and for site-specific impact analysis. Industry expressions of interest in coal leasing contain data on tract delineations received from USGS. Most of the tracts are located in the 11Chaco Stripable Coal Belt11 near the Bisti Badlands. The coal tracts could be considered for competitive lease sales under appropriate market conditions. Finally, land status {ownership) contains five types of ownership for the study area: Indian, Indian withdrawn, state, BLM, and private. Not all environmental and administrative themes have digital data for the entire four-quadrangle area. In some cases, data for a theme is missing either one or two of the quadrangles. In other cases, only portions of the four-quadrangles have been digitized. The digitized data was mostly analyzed using the MAPS subpackage of MOSS, which is designed exclusively for raster (cell) format. MAPS was used for two reasons: {1) the DEMs already come in raster format, and other themes had to be in cell format for overlaying; (2) statistical analysis is performed more easily with cell data using statistical packages outside of MOSS/MAPS. Procedures for Assessing Inventory Data Using MOSS/MAPS The inventory data base was assessed to measure the biases associated with particular environmental features resulting from nonrandom survey units. In this way it is possible to determine where future surveys need to be concentrated to provide more representative 64

PAGE 78

samples of cultural site and environmental data in the project area. The preparation of the data base for analysis included the merging of the four 7.5 minute maps for each environmental and administrative theme. Each resulting map was converted to cell-format. The archaeological survey data was prepared in a similar fashion except that line and point data had to be converted to polygon data. After converting this inventory map to cell-format, all the map cells values showing inventoried areas were combined into one classification and assigned a new !ingle value. Once this step was performed the new inventory map was overlaid with each of the environmental and administrative themes listed in Figure 5. 1. Finally, area information was calculated for each thematic subject intersected with areas surveyed. The remainder of this section shows the technical steps used in analyzing the inventory data base. This portion of the report will be of particular interest to those who are familiar with MOSS/MAPS. (All commands are capitalized. MOSS commands are underscored, whereas MAPS commands are not.) Figure 5.2 is a cartographic model showing the steps followed and their sequence. The circled numbers correspond to the respective numbered steps listed below. However, the numbering does not necessarily imply consecutiveness, since one step may not have to precede another. Note that the continuous data is data that are uninterrupted and has real number values. The elevation, slope, and aspect data sets are examples of continuous type data. Discrete data, on the other hand, has distinct units or classes represented by integer values, (whole numbers). The SLOPE and ASPECT commands in were used to access continuous elevation data, i.e., 7.5 minute OEM cell maps, and the resulting maps 65

PAGE 79

were continuous slope and aspect maps, respectively. The EXTRACT command in MAPS is a data reclassification command that creates a new discrete map by assigning new values to the cell values of an existing discrete or continuous map. EXTRACT was used to convert the continuous slope and aspect maps to discrete maps. 1. MERGE four 7.5 minute quadrangles for each environmental or administrative data theme. 2. RASTERIZE to convert to cell format with background and nonbackground values. A 30m X 30m cell size was used to obtain the highest resolution possible. In this way more accurate area measurements could be derived. 3. SLOPE or ASPECT continuous elevation map to produce continuous slope and aspect maps. 4. EXTRACT only the continuous elevation, slope, and aspect maps to assign new values and convert to discrete data classes. 5. MERGE the archaeological inventory data. The BLM Albuquerque District Office has distributed all their inventory data from 1 :100,000-scale maps. There are 32 7.5 minute quandrangles in a 1 :100,000-scale map. Thus 32 maps with inventory data had to be merged by the District Office. 6. BUFFER archaeological point and line inventory data at 1 :000,000-scale. All of the point and line inventory data shown on the 1 :100,000-scale maps mentioned in step 5 was BUFFERED at the District Office. In this manner all survey data was converted to polygon data so that area measurements could be made. 7. GENERATE a WINDOW border using UTM coordinate pair data for the four-quadrangle study area. The GENERATE command allows for the creation of new maps or rectangular features such as a frame. The WINDOW was expanded beyond the edge of the map to create even multiples of the 30m X 30 m cell size. In this way, each cell map had the exact number of cells, with no rounding error. By enlarging the window, the edge of the map was not obscured by the window frame when viewed on the terminal screen. 67

PAGE 80

8. OVERLAY 1 :100,000-scale BUFFERED archaeological inventory maps with WINDOW border using the 11intersect11 option of the OVERLAY command. The resulting map included only the inventory data for the four-quadrangle study area. 9. RASTERIZE BUFFERED archaeological inventory data to convert to cell format with background and non-background values. A 30m X 30m cell size was used to obtain the highest resolution possible. (Cell size must be the same as in Step 2.) 10. EXTRACT to combine all map values into one classification and assign a new single value to all cells having inventory data. 11. CROSS environmental or administrative data theme with archaeological inventory data to create a new discrete map showing logical intersection of inventoried and noninventoried areas by subject. The logical operations 11AND11 and 110R11 in the CROSS command were used to assign new values to cells. 12. AREA the map from step 11 to obtain the number and percentage of acres inventoried for each subject of a theme. Evaluating the Inventory and Environmental Data Methodology for Analysis By analyzing nonrandom survey units, we can quantify the potential biases associated with the under or over sampling of particular environmental features. In this way, it is possible to determine where additional areas need to be surveyed that will provide more representative samples of sites and environmental features. To accomplish this objective the percentage change was calculated between the cell frequency of each environmental subject that has been archaeologically surveyed and the cell frequency of the same subject throughout the total study area. These frequencies were converted to percentages and then weighed to better show the relative biases associated with the distribution of the surveyed environmental features. 68

PAGE 81

The remainder of this section is a technical discussion of the methodology used to measure the potential biases existing with particular environmental features that have been oversurveyed or undersurveyed. Some readers may wish to skip to the section that discusses the results of this analysis. The percentage change between the frequency of distribution of each environmental subject inventoried and the frequency of the same subject within the total study area was calculated using cell frequencies that were converted to observed and expected values. The observed values indicate the area of each subject's intersection with inventoried areas. The expected values show the area of each subject in the entire four-quadrangle study area. By comparing the observed values to the expected values, one can measure the degree of bias associated with each subject. The observed values were obtained using the AREA command in MAPS (Figure 5.2) for each of the maps derived by overlaying a particular environmental theme with the area inventoried. Both the intersected and the nonintersected areas were revealed by assigning unique values using the CROSS command in MAPS (Figure 5.2). The analysis used only the intersected values --where environmental subjects coincided with surveyed areas. The observed values were converted to a percentage. The expected percentages of each subject were taken from the area table for each environmental theme. These expected values for the four-quadrangle study area represent the underlying or background occurrence of these mapped features (subjects). Because portions of the four-quadrangle study area did not have digitized data, the expected percentage values pertain only to the area of the subjects that was 69

PAGE 82

actually mapped, and not necessarily to the distribution in the entire four-quadrangle study area . Inventory bias can be quantified by showing the percentage of areas overinventoried and underinventoried. The percentages reflecting the degree of inventory bias for each subject were derived in the following manner. For each theme, all observed subject frequency values that intersected with the inventory data were summed. Each total was then multiplied by the percent expected of each subject for the respective theme. The percentage or difference between the subjects that had been inventoried (observed) and expected distribution of the subjects in the project area was obtained using the following formula: lOO% X Observed Frequency Expected Frequency = % Bias Expected Frequency The resulting figures represent the percentage of positive or negative bias, associated with each subject inventoried. The positive values reveal the percentage of oversurveyed area associated with a particular subject. The negative values reflect the undersurveyed percentage. The percent bias, however, does not offer as clear an understanding of the relative amount of inventory bias of each subject in the project area as does a weighted bias. This is because large survey units may have been coincidentally located in areas where the distribution of a particular environmental feature (subject) was concentrated. The converse may have also occurred, i.e., inventories were not 70

PAGE 83

coincidentally located where certain environmental features were concentrated. Therefore, the percent bias figures may be somewhat misleading by suggesting that a given environmental feature was highly surveyed or had received little inventory. This problem could be avoided if smaller sample survey units are selected at random rather than using larger block surveys that focus only on areas slated for development. The weighted percent bias was used to obtain a measure of the amount of bias associated with each of the subjects {of a theme) relative to all other subjects in the study area. The weighted percent bias figures were obtained by multiplying the percent expected by the percent bias for each subject. The foregoing methods were used to assess the quality of inventory and environmental data. These methods measure the potential for bias that may have been introduced from the lack of any overall random sampling plan for the archaeological surveys. Results Analysis The eight environmental themes overlaid with the inventory data were elevation, slope, aspect, drainages, streams, soils, surface geology, and geologic surficial process units. The results of the analysis of each theme are shown along with an interpretation of the larger biases. A table is included for each theme. A program written in FORTRAN was used to calculate the biases for each subject. Each table contains the percent observed, the percent expected, percent bias, and weighted percent bias for each thematic subject. Following each table is a map, generated in MAPS, showing only the subjects intersected with inventory 71

PAGE 84

data. Each subject, where appropriate, is shaded with a different pattern. Noninventoried subjects were left unshaded, or in some cases not shown. Most of the maps were derived from an Anadex printer. The Anadex printer maps are not to scale and most do not have a legend, since they are used only for analysis. Exact location, subject, and cell value information can be obtained with the MAPS QUERY command (using the crosshair input mark) to interactively select a desired point or area. In addition, a map containing only landline data can be plotted over any map to determine sectional boundaries. Two example maps were processed on the !DIMS system and plotted on the Applicon. The Applicon maps are in color. All map projections are based on the UTM coordinate system; this explains why the printed maps are canted. The result of overlaying the elevation data with the areas inventoried is shown in Table 5.1 and Figure 5.3. A lack of strong weighted biases larger than an absolute value of approximately 10 percent suggests that no strong factors influenced where inventories were conducted with respect to elevation classes. The highest weighted bias, 5.55 percent, is for elevation class 1,791 m to 1,820 m. Table 5.2 and Figure 5.4 show the result of overlaying the slope data with the inventory data. Here again, no strong biases are associated with overinventoried and underinventoried slope classes. The 0 to 5 percent slope class, however, shows a 10.83 percent bias and a 6.30 percent weighted bias. This bias could be accounted for, in part, by the inventory performed in areas of little or no slope where oil and gas and coal would be easiest to develope. In addition, portions of Blocks X and XI of the Navajo Indian Irrigation Project (NIIP) lies in the northwest corner of the study area, and includes about 11 percent of 72

PAGE 85

. TABLE 5.1 ELEVATION DATA OVERLAID WITH INVENTORY DATA Subject % Observed (Elevation Class) (Intersected) % Expected % Bias Weighted% Bias 1730 m 1760 m 0 0.15 100.00 -0.15 1761 m1790 m 1.19 5.14 -76.85 -3.95 1791 m 1820 m 19.25 13.70 40.51 5.55 1821 m 1850 m 20.69 18.34 12.81 2.35 1851 m 1880 m 22.53 27.52 -18.13 -4.99 1881 m 1910 m 20.79 19.10 8.86 1.69 1911 m 1940 m 13.12 10.50 24.93 2.62 1941 m 1970 m 2.42 4. 51 -46.27 -2.09 1971 m 2000 m 0. 01 0.99 -99.33 -0.98 2001 m 2030 m 0 0.05 -100.00 -0.05 73

PAGE 86

FIGURE 5.3 Map of elevation data overlaid with inventory data 74

PAGE 87

TABLE 5.2 SLOPE DATA OVERLAID WITH INVENTORY DATA Subject % Observed ( % Slope Class) (Intersected) % Expected % Bias Weighted % Bias 0 -5 64.44 58.14 10.83 6.30 6 10 25.14 25.84 -2.71 -0.70 11 -15 7.35 9.60 -23.48 -2.25 16 30 2.98 5.87 -49.18 -2.89 31 50 0.09 0.55 -83. 1 0 -0.46 75

PAGE 88

-•• ( ' , . . ... \ FIGURE 5.4 Map of slope data overlaid with inventory data 76 • . . . . ' ..

PAGE 89

Table 5.3 and Figure 5.5 show the result of overlaying the aspect data with areas inventoried . Table 5.3 shows that no strong biases are associated with any of the aspect classes shown. Table 5.4 and Figure 5.6 show the result of overlaying the drainage data with the areas inventoried. The Hunter/Brimhall drainage has the highest weighted bias of 8.58 percent and a percent bias of 34.14 percent. This bias, indicating overinventory, may have resulted from the presence of the Hunter/Brimhall drainage in NIIP, Block XI. Since the entire NIIP Block XI was inventoried and makes up about 11 percent of the total inventory in the project area, more of a sampling error (bias) is associated with this one large inventory sample than with other areas. On the other hand, if many smaller size sampling units, for example, 100 10-acre inventory blocks were randomly distributed in the study area, a smaller sampling error or bias would be associated with the NIIP area and therefore the Hunter/Brimhall drainage. Table 5.5 and Figure 5.7 represent the results of overlaying the stream data with the area inventoried. Although the biases are relatively small for all stream orders, there appears to be some positive bias towards the lowest order streams -first and second orders. The fourth order stream class shows a relatively high negative bias. the study area. This area contains relatively flat land, favorable for irrigation and agriculture, and therefore could also account for some of the overinventory observed for slope class 0 to 5 percent. First and second order streams show a weighted bias of 7.7 9 percent and 6.91 percent respectively. This result can be accounted for by the 77

PAGE 90

TABLE 5.3 ASPECT DATA OVERLAID WITH INVENTORY DATA Subject % Observed (Aspect Class Degrees) (Intersected) % Expected % Bias Weighted % Bias 0 -45 (NNE) 10.14 9 . 71 4.39 0.43 46 90 (ENE) 6.56 6.69 -1.92 -0.13 91 -135 (ESE) 8 :26 9.41 -12.27 1.15 136 180 (SSE) 18.92 18.45 2.57 0.47 181 -225 (SSW) 18.35 16.69 9.96 1.66 226 270 (WSW) 11 .86 1 1 .84 0.15 0.02 271 -315 (WNW) 12.17 1 3 .23 -7.99 -1 .06 316 -359 (NNW) 13.74 13.98 -1.73 -0.24 78

PAGE 91

.• ... \ r FIGURE 5.5 Map of aspect data overlaid with inventory data 79 •

PAGE 92

TABLE 5.4 DRAINAGE DATA OVERLAID WITH INVENTORY DATA Subject % Observed Drainage (Intersected) % Expected % Bias Weighted % Bias De-na-zin 66.17 68.83 -3.87 -2.66 Escavada 0.12 4.19 -97.13 -4.07 Hunter/ B rimhall 33.70 25.12 34.14 8.58 Kim-me-ni-oli 0.02 1.86 -99.01 -1.84 80

PAGE 93

. . . .. , . I jilllll ) { '\ FIGURE 5.6 Map of drainage data overlaid with inventory data 81

PAGE 94

TABLE 5.5 STREAM DATA OVERLAID WITH INVENTORY DATA Subject % Observed Stream Order (Intersected) % Expected % Bias Weighted % Bias First 39.85 32.06 24.29 7.79 Second 17.78 10.87 63.56 6. 91 Third 36.78 39.76 -7.50 -2.98 Fourth 4.91 12.68 -61.25 -7.77 Fifth 0.62 2.21 -71 . 73 -1.59 Sixth 0.06 2.42 -97.65 -2.36 82

PAGE 95

, j { / lb .... , , .... J -.' -. FIGURE 5.7 Map of stream data overlaid with inventory data 8 3 /

PAGE 96

two largest surveys in the study area -NIIP (Vogler and others 1982) being conducted in upland topography where first and second order streams tend to occur. As mentioned above, the NIIP (Block X and XI) encompasses about 11 percent of the study area and is situated on a plateau drained mainly by first and second order ephemeral streams. The Joint Venture Survey, (Vierra and others 1986) in the central part of the study area, concentrated on the areas overlying major coal seams. Many of the coal outcrops lie near the tops of hills and ridges. Because of less overburden, coal at or near the outcrop is generally easier and less expensive to extract. Conversely, higher order streams, which have dissected the upland topography may have eroded the coal. The discontinuity of the seam along the outcrop may make the deposit more difficult or expensive to remove. Hence, the larger order streams, such as the fourth order, might have been undersurveyed, whereas first and second order streams occurring near the tops of hills and ridges could have been overinventoried (John Roney, personal communication, 1986). The result of overlaying the soils data with inventory data is shown in Table 5.6 and Figure 5.8. Figure 5.8 is a color map produced on the Applicon plotter. Two soil types show marked biases with respect to the degree of inventory conducted on them. The Badland (BA) soil type has been undersurveyed by -42.23 percent and has a weighted bias of -15.06 percent. The Sheppard Mayqueen Shiprock Complex (SO) has been oversurveyed by 194.73 percent and has a weighted bias of 21.44 percent. Much of the Badlands soil (18.46 percent) tends to co-occur in the Bisti and De-na-zin WSA. Because no oil and gas or coal development is allowed in WSAs, the Bisti and De-na-zin WSAs have probably been underinventoried. Another contributor to this negative bias may have been deliberate avoidance of badlands for oil and gas development because 84

PAGE 97

TABLE 5.6 SOILS DATA OVERLAID WITH INVENTORY DATA Subject % Observed (Soil Type) (Intersected) % ExEected % Bias Weighted % Bias Badland (BA) 20.61 35.67 -42.23 -15.06 Blancot Notal Assoc. Gently Sloping 5.42 8.87 -38.93 -3.45 (BT) Doak Loam (DC) 0.01 0.37 -98.46 -0.36 Doak Avalon (ON) 1. 75 4.45 -60.74 -2.70 Doak-Sheppard Shiprock Assoc. Rolling (OS) 1.77 5.44 -67.44 -3.67 Dune Land (DZ) 0 0.03 -100.00 -0.03 Fruitland-PersayoSheppard Complex Hilly (FX) 0.01 0.01 -33.66 -0.00 Huerfano-Muff Uffens Complex (HU) 11.57 6.32 83.04 5.25 Riverwash (RA) 0 0.48 -99.31 0.48 Rock Outcrop (RO) 0.16 4.16 -96.15 -4.00 Sheppard Huerfano Notal Com)lex Gently Sloping (SC 26.27 23.10 13.71 3.17 Sheppard Mayqueen Shiprock Complex (SO) 32.45 11 . 01 194.73 21 .44 Shiprock Fine Sandy Loam (SO) 0 0.08 -100.00 -0.08 85

PAGE 98

FIGURE 5.8 Map of soils data overlaid with inventory data (See Soil Conservation Service nomenclature for soil types, Chapter III) 86

PAGE 99

of their rough terrain, high scenic quality, and potential for paleontological resources. In contrast, the Sheppard Mayqueen Shiprock Complex (SO) has been more thoroughly inventoried, because the SO soil is restricted to the NIIP area, which is found on a plateau. This soil type often occurs in relatively flat areas (0 to 5 percent slope). Most areas of this soil type are used for irrigated crops, mainly alfalfa and pasture Conservation Service 1980). Overlaying the surface geology data with the areas inventoried, revealed minor biases,as shown in Table 5.7 and Figure 5.9. The weighted percentage bias column in Table 5.7 shows no large biases, but the Kkf unit (Kirtland/Fruitland-shale, sandstone, and coal) and Qa3 unit (Quaternary alluvial unit of late Pleistocene to early Holocene age) show a greater negative and higher positive bias, respectively than the other units. The Kkf unit often occurs as steep cliffs next to badland areas, which are generally avoided for oil and gas development because of scenic quality, potential for significant paleontological resources, and the two WSAs within portions of these unique areas. In addition, the steep cliffs of this formation may dramatically increase overburden thickness, therefore making it unprofitable to mine coal (Jim Turner, personal communication 1986). These factors would account for the -6.20 percent weighted bias for Kkf. In contrast, the Qa3 unit has a bias toward overinventory with a 7.48 percent weighted bias. This geologic unit occurs on topographically higher geomorphic surfaces (strath terraces and pediments), is coarse grained, and contains more well developed soils than Holocene units (Wells and others 1983). Because of its stable soils, Qa3 lends itself to locating equipment for oil and gas development. It also tends to occur in areas of low slope, which are ideal for development. Therefore, clearance inventories are 87

PAGE 100

TABLE 5.7 SURFACE GEOLOGY DATA OVERLAID WITH INVENTORY DATA Subject (Surface % Observed Geology Units) (Intersected) % Expected % Bias Weighted % Bias Kl 2.26 5.89 -61.67 -3.63 Kch 0.06 1.96 -97. 16 -1.90 Kkf 14.98 21 . 18 -29.29 -6.20 Kmf 0 0.06 -100.00 -0.06 Kmft 0.02 0.36 -93.92 -0.34 Kpc 4.69 3.54 32.58 1.15 Qal 0 0.18 -100.00 -0.18 Qa2 0.46 4.32 -89.35 -3.86 Qa3 28.63 21 . 15 35.36 7.48 Qa4 19.67 21 .30 -7.63 -1.63 Qa4A 0 0.02 -100.00 -0.02 Qa4B 15.13 9.98 51.63 5.15 Qel-3 6.22 5.02 23.94 1.20 Qe2-3 0.10 0.37 -71 . 67 -0.27 Qe3 7. 77 4.67 66.42 3.10 88

PAGE 101

FIGURE 5.9 Map of surface geology data overlaid with inventory data 89

PAGE 102

probably focusing on those areas where drilling is situated on Qa3 units. (Jim Turner, personal communication, June 1986.) Table 5.8 and Figure 5.10 show the result of overlaying the geologic surficial process units with the inventory data. Process units KTbb and QTps show strong negative and positive weighted biases, respectively. KTbb tends to occur in the badland landscape. Once again, the badland areas tend to be avoided for development due to their environmental fragility, WSA status, and fossil remains. Thus, a negative bias of -64.61 percent or -16.13 percent weighted bias is understandable. Conversely, the QTps process unit has a 31.27 percent bias or 12.82 percent positive weighted bias. The QTps unit occurs in pediment and terrace remnants that contain clay, silt, sand, and gravel and have erosional stability and aeolian deposition (Wells and others 1983). The QTps unit can support heavy oil and gas drilling equipment due to its stability and low slopes. If such areas were used for drilling, the inventories would, in turn, be biased toward this process unit. Evaluating the Inventory and Administrative Data Methodology for Analysis This section assesses the amount of archaeological inventory that has been conducted for the administrative themes listed in Figure 5.1. This assessment involves the use of MOSS/MAPS to overlay the inventory data with each of the administrative themes to identify subject areas that have been inventoried and those that have not. This analysis uses procedures presented in Figure 5.2 but uses administrative rather than environmental themes. 90

PAGE 103

TABLE 5.8 GEOLOGIC SURFICIAL PROCESS UNITS DATA OVERLAID WITH INVENTORY DATA Subject (Geologic % Observed Surficial Process Unit) (Intersected) % Expected % Bias Weighted % Bias Ktbb 8.83 24.96 -64.61 -16.13 QA2 0.63 1.94 -67.76 -1. 3 . 1 QTps 53.81 40.99 31.27 12.82 QTpt 23.65 18.74 26.19 4. 91 Qall 0.90 l. 75 -48.66 -0.85 Qa 12 6.88 8.75 -21.38 -1 .87 Qal3 5. 12 2.75 86.06 2.37 Qds 0.11 0.08 42.58 0.03 Qtpt 0.08 0.03 152.28 0.05 91

PAGE 104

FIGURE 5.10 Map of geologic surficial pr ocess units data overlaid with inventory data 92

PAGE 105

The land status (ownership) area summary table was computed differently from the area tables for other administrative themes (see below). The objective of overlaying the land status data with the inventory data was to determine possible inventory biases due to ownership designation. Therefore, percent bias and weighted percent biases has to be calculated in the same way as for environmental themes. A color map plotted on the Applicon plotter is included to show the ownership types intersected with the inventory data. Intersections are shown with different colors for each ownership type intersected with inventory data. Most of the administrative themes have a null class. The null class represents areas for a particular theme that were not assigned to any subject. For example, the proposed WSA theme has three subjects: null, De-na-zin, and Bisti. The null subject class was not considered in the interpretation of the analysis results. Area summary tables were developed for all the themes using the output from "AREA" command in Maps (Figure 5.2). The summary tables show the number of acres inventoried , the percent inventoried, the number of acres not inventoried, and the percent not inventoried for each subject (administrative feature). This data will enable the user to determine the amount of inventory conducted for each subject of a particular theme. Following each table is a map generated showing only the subjects intersected with the inventory data. Noninventoried areas are not shown. Some subjects are shaded to enhance interpretation while others are simply plotted. All plotting was performed on an Anadex printer with the exception of the land status map. The Anadex printer maps are not to scale and do not have a legend, since the maps are used strictly for analysis. 93

PAGE 106

Results of Analysis Table 5.9 and Figure 5.11 show the result of overlaying the land status (ownership) data with the inventory data. The BLM land has the highest weighted bias (10.45 percent) as might be expected because of legal requirements to systematically survey cultural resources on any federal land planned for development or surface disturbance. The Indian withdrawn lands have weighted biases of -1.35 percent and -2.92 percent, respectively. Indian withdrawn lands are parcels that have been acquired through purchase or exchange agreements with the BLM and are administered by the Bureau of Indian Affairs (BIA). Title, however still rests with the U.S. Government. Indian lands are associated with reservations, Pueblos, and special treaties. The Indian lands can be held in trust by the U.S. Government and administered by the BIA, or held in fee status and administered by tribes. The Indian and Indian withdrawn lands in the study area are used by the Navajos mainly for grazing cattle and are not subject to oi l and gas or coal development. As seen in Table 5.9, the private and state lands show negative weighted biases of -3.07 percent and -3.10 percent, respectively. These lands are often the least inventoried due to the lack of strong legislation that requires cultural resource inventories be conducted on state and private lands planned for development. Most archaeological survey work tends to be concentrated on federal or state lands, while comparable site data is often lacking fro m properties under private ownership. This situation can introduce a severe source of bias in a project area because private properties often include some of the best areas for hunting and plant collecting. 94

PAGE 107

TABLE 5.9 LAND STATUS (OWNERSHIP) DATA OVERLAID WITH INVENTORY DATA Subject % Observed (Land Ownership) (Intersected) % Expected % Bias Weighted % Bias Indian 35.76 37.11 -3.63 -1.35 Private 2.11 5.18 -59.32 -3.07 State 10.65 13.75 -22.56 -3.10 BLM 49.43 38.98 26.81 10.45 Indian withdrawn 2.05 4.97 -58.74 -2.92 95

PAGE 108

FIGURE 5.11 Map of land status (ownership) data overlaid with inventory data 96

PAGE 109

Table 5.10 and Figure 5.12 show the result of overlaying wilderness study areas (WSA) data with inventory data. The Bisti WSA has the least amount of inventory (.04 percent) while the De-na-zin WSA has an inventory percentage of 16.85 percent. Table 5.10 shows that the majority of these two WSAs are lacking inventory. The result of overlaying the linear development in WSAs (road and transmission lines) with inventory . data is shown in Table 5.11 and Figure 5.13. As seen in Table 5.11, none of the transmission line corridor area has been surveyed, and only 29.57 percent of the road has been inventoried. The transmission lines and roads in the WSA pose an interesting area measurement problem for a GIS. Both the road and the transmission line represent linear features on the landscape and must be digitized as lines. The transmission line corridor that would be inventoried would normally be about 30 m wide, while the road might only require about a 3 m wide right-of-way because most of the dirt roads are made by narrow two-track vehicles. Linear features are inventoried to include a buffer zone so that the right-of-way will contain areas of direct impact as well as potential or indirect impacts. An example of an indirect impact is a vehicle that turns off a two-track road and heads cross-country. The transmission line was not buffered because the 30 m cell size creates a 15m buffer on either side of the line representing the transmission line. Similarly, the two-track road didn't need to be buffered because the 30m cell size creates a buffer of 15 m was created on either side of the line representing the 3m wide road. 97

PAGE 110

TABLE 5.10 WILDERNESS STUDY AREAS (WSA) DATA OVERLAID WITH INVENTORY DATA Subject (WSA) Bisti De-na-zin Inventoried Non-Inventoried % Non-Acres % Inventoried Acres Inventoried 1.33 0.04 3,795.84 99.96 3,563.43 16.85 17,579.20 83.15 98

PAGE 111

• _ ,. n J FIGURE 5.12 Wilderness study areas (WSA) data overlaid with inventory data 99

PAGE 112

TABLE 5.11 LINEAR DEVELOPMENT IN WSA DATA OVERLAID WITH INVENTORY DATA Subject (Linear Inventoried Non-Inventoried % Non-Development Boundary) Acres % Inventoried Acres Inventoried Road 25.58 29.57 60.94 70.43 Transmission Line 0 0 25.35 100.00 100

PAGE 113

/ ' . ...... FIGURE 5.13 Linear development in WSA data overlaid with inventory data 101

PAGE 114

If the potential impact zone of a linear feature is expected to be large for certain types of development projects, then the ZONE command in MAPS should be used to buffer accordingly. Another reason for buffering linear features is that the feature itself may potentially be inaccurately located on a map. Buffering compensates for any discrepancy resulting from inaccurate mapping. Finally, one cannot always assume that a linear feature will pass through the centroid of any given cell, and therefore be equidistant from each side of the cell. Often a line will pass through only a small portion of the cell, thus activating it. If this occurs, buffering is a conservative way to solve this problem. This technical discussion on using cell data with unbuffered linear features highlights the fact that the area figures provided in Table 5.11 are only estimates of the impacted area of linear development features that coincides with inventoried areas. Table 5.12 and Figure 5.14 show the result of overlaying the road network data with the inventory data. Once again, the linear features or subjects, a jeep trail and a secondary unpaved road (two-track dirt road) would have been automatically buffered to 15m on either side of the line representing the road. The rough area estimates in Table 5.12 reveal that only 2.47 percent of jeep trail has been inventoried, and a little under half (45.47 percent) of the secondary unpaved road has been surveyed. Table 5.13 and Figure 5.15 show the result of overlaying the proposed areas of critical environmental concern (ACEC) with areas inventoried. Table 5.13 show that the Fossil Forest has received the most inventory (75.10 percent), while the Bisti/De-na-zin has received only about 10 percent inventory. 102

PAGE 115

TABLE 5.12 ROAD NETWORK DATA OVERLAID WITH INVENTORY DATA Subject Inventoried Non-Inventoried % Non-(Road/Trail) Acres % Inventoried Acres Inventoried Jeep Trail .44 2.47 17.35 97.53 Secondary Unpaved Road 297.79 45.47 357.17 54.53 103

PAGE 116

FIGURE 5.14 Road network data overlaid with inventory data 104 /

PAGE 117

TABLE 5.13 PROPOSED AREAS OF CRITICAL ENVIRONMENTAL CONCERN (ACEC) DATA OVERLAID WITH INVENTOPY DATA Subject Inventoried Non-Inventoried % Non(Proposed ACEC) Acres % Inventoried Acres Inventoried Bisti/De-na-zin 3,565.43 10.10 31,735.74 89.90 Fossil Forest 2,129.21 75.10 705.88 24.90 105

PAGE 118

,,. FIGURE 5.15 Map of proposed areas of critical environmental concern (ACEC) data overlaid with inventory data 106 •

PAGE 119

Table 5.14 and Figure 5.16 show the result of overlaying the range allotment data with the inventory data. The only range allotment, Paragon, in the study area has had 14.47 percent of its area inventoried. Table 5.15 and Figure 5.17 show the result of overlaying oil and gas data with inventory data. None of oil and gas leases NM-02311, NM-16054, NM-19570, NM-20311, NM-22595, NM-23441, NM-27437, and NM-36584 have been inventoried. In contrast, leases NM-23471, NM-26660, NM-29344, and NM-32471 have been completely surveyed (100 percent). Table 5.16 and Figure 5.18 show the result of overlaying preference right lease applications (PRLA) data with inventory data. Table 5.16 reveals that PRLA NM-3752, NM-3753, NM-3754 have the most area inventoried with 88.99 percent, 97.96 percent, and 76.75 percent, respectively. Conversely, PRLA NM-11916 and NM-6802 do not have any area inventoried for them. PRLA NM-3834, NM-3835m NM-3838, and NM-7235 have been only slightly inventoried. Table 5.17 and Figure 5.19 show the result of overlaying industry expressions of interest (IEOI) in coal leasing data with inventory data. The IEOI NM-0186613 and NM-0186615 each have about 99 percent of their areas surveyed. On the other hand, Sisti 6 Tract and NM-0186613 only have 1.04 percent and 0.11 percent, respectively, of their areas inventoried. 107

PAGE 120

TABLE 5.14 RANGE ALLOTMENT DATA OVERLAID WITH INVENTORY DATA Subject Inventoried Non-Inventoried % Non-(Range Allotment) Acres % Inventoried Acres Inventoried Paragon 3,697.09 14.47 21,850.96 85.53 108

PAGE 121

• l ....... I l • ' •. ; . . , _ .. FIGURE 5.16 Range allotment data overlaid with inventory data 109 .. ' l .

PAGE 122

TABLE 5.15 OIL AND GAS LEASE DATA OVERLAID WITH INVENTORY DATA Subject Inventoried Non-Inventoried % Non-(Oil/Gas Lease) Acres % Inventoried Acres Inventoried NM02311 0 0 273.32 100.00 NM13612 2 0.31 651.84 99.69 NM14443-A 153.45 99.57 .67 0.43 NM16054 0 0 239.07 100.00 NM19570 0 0 685.64 100.00 NM20308 2.89 0.74 386.97 99.26 NM20311 0 0 210.16 100.00 NM22595 0 0 224.62 100.00 NM23239 3.11 1.87 163.46 98.13 NM23441 0 0 76.28 100.00 NM23471 45.37 100.00 0 0 NM25847 0.89 0.94 93.85 99.06 NM26660 67.16 100.00 0 0 NM27023 0.67 0.93 71.17 99.07 NM27437 0 0 169. 91 100.00 NM29344 75.17 100.00 0 0 NM32471 82.51 100.00 0 0 NM36584 0 0 87.18 100.00 110

PAGE 123

• r l L . "' FIGURE 5.17 Oil and gas lease data overlaid with inventory data 111

PAGE 124

TABLE 5.16 PREFERENCE RIGHT LEASE APPLICATIONS (PRLA) DATA OVERLAID WITH INVENTORY DATA Subject Inventoried Non-Inventoried % Non-(PRLA) Acres % Inventoried Acres Inventoried NM-11916 0 0 2889.53 100.00 NM-3752 3394.64 88.99 420.10 11.01 NM-3753 2988.32 97.96 62.27 2.04 NM-3754 2411.21 76.75 730.57 23.25 NM-3755 17.12 1.03 1,651.95 98.97 NM-3834 4.89 0. 11 4' 615.51 99.89 NM-3835 54.71 1.21 4,484.81 98.79 NM-3836 384.97 12.56 2,680.24 87.44 NM-3337 424.11 28.53 1,062.60 71 .47 NM-3838 6.89 0.14 4,799.28 99.86 NM-6801 772.33 21 . 19 2,872.90 78.81 NM-6802 0 0 345.16 100.00 NM7235 0.89 0.55 161 . 90 99.45 112

PAGE 125

,.. I "" • . . . . . ... \ ht lease app Preference rlgwith inventory data data overlaid I FIGURE 5.18 113

PAGE 126

TABLE 5.17 INDUSTRY EXPRESSIONS OF INTEREST (IEOI) IN COAL LEASING DATA OVERLAID WITH INVENTORY DATA Subject Inventoried Non-Inventoried % Non-(IEOI) Acres % Inventoried Acres Inventoried Bi sti 1 Tract 2,742.57 67.72 1 ,307. 01 32.28 Bi sti 2 Tract 81.62 22.38 283.11 77.62 Bi sti 4 Tract 12.23 40.14 18.24 59.86 Bisti 4 (a & b) Tract 380.29 14.12 2,315.13 85.88 Bi sti 6 Tract 2.89 1.04 275.32 98.96 NM-0186612 184.37 34.60 348.49 65.40 NM-0186613 1.33 0.11 1,193.37 99.89 NM-0186615 2,006.45 99.64 7.34 0.36 NM-1 0931 1,955.74 99.89 2.22 0.11 114

PAGE 127

.-FIGURE 5.19 Industry expressions of interst (IEOI) in coal leasing data overlaid with inventory data 115

PAGE 128

Conclusions Chapter V has shown the utility of MOSS/MAPS for assessing archaeological inventory and environmental data bases. This chapter presented specific MOSS/MAP commands and procedures that can aid federal archaeologists or land use planners in assessing data biases. Methodologies for interpreting the results along with the methodology used in the analysis were also presented. On the whole, no particular elevation, slope, or aspect class tends to be strongly biased in the degree of inventory conducted. The drainage data analysis, on the other hand, reveals that the Hunter/Brimhall drainage has a high overinventory bias, which may be largely due to the occurrence of the Hunter/Brimhall Wash in the NIIP, which occurs within 11 percent of the study area. The NIIP area has been completely inventoried. The analysis of the stream data reveals a strong inventory bias toward first and second order streams and away from higher order streams. First and second order streams tend to occur more in upland topography where the NIIP and coal seams are located. Thus, the survey would have been concentrated in these areas. The soils data analysis shows a strong bias away from badland soils and a relatively high bias toward the Mayqueen Shiprock Complex, which occurs only in the NIIP area. The surface geology analysis shows no strong biases for overinventoried or underinventoried surface geologic units. The results of the analysis of the geologic surficial process data show a strong negative bias toward the badlands area (KTbb) and a strong bias toward the QTps units. Avoidance of badland areas and the need for stable soils for equipment in oil and gas development, respectively were suggested as possible explanations for these patterns. 116

PAGE 129

The land status theme was assessed for inventory bias, and eight other administrative themes were also analyzed to determine how much inventory has been conducted for the different subjects for each theme. However, for land status inventory biases were calculated. The land status analysis reveals a strong positive bias for SLM land and negative biases for private, state, Indian, and Indian Withdrawn lands. This result is not surprising in light of the amount of mineral development conducted on SLM lands. The WSA data analysis shows that the De-na-zin WSA has the highest inventory. The linear development for the WSA analysis makes clear that none of the transmission line corridors have been inventoried in the Sisti WSA. The road network analysis suggests that a little under half of the secondary unpaved roads in the study area have been surveyed. The results of the proposed ACEC show that Fossil Forest has received the most inventory. Oil and gas data analysis results reveal that NM23471, NM26660, NM29344, and NM32471 have been completely inventoried. The results of the PRLA data analysis show that PRLAs NM-3752, NM-3753, and NM-3754 have received the most inventory. Finally, the IEOI in coal analysis reveals that Sisti 6 Tract and NM-0186613 have almost none of their areas surveyed. 117

PAGE 130

CHAPTER VI USE OF MOSS/MAPS FOR IDENTIFYING POTENTIAL CULTURAL RESOURCE AND LAND USE CONFLICT AREAS This chapter presents MOSS/MAPS spatial analysis procedures. These procedures are useful for preparing maps for identifying potential cultural resource and land use conflict areas in a multiple-use setting. More specifically, this chapter focuses on the distribution of Archaic sites in relation to areas of proposed coal and oil and gas development. Areas designated or proposed for nondevelopment in relation to site distribution are also examined. Land use compatibilities, however must, be understood before spatial analysis can be conducted with a GIS. The GIS is only a tool for conducting spatial analysis of different data themes. The archaeologist or planner must identify cultural resource and land use conflicts. Only those familiar with land use conflict and compatibility situations can interpret whether the GIS spatial analysis represents an potential or existing conflict. The procedures followed to prepare the archaeological site data for overlaying with the administrative themes are presented along with a cartographic model that will schematically show the steps involved in preparing the data for analysis. The archaeological site data in this analysis represents sites of the Archaic period (5,500 BC to AD 400). The Archaic period was chosen for this analysis, because it is one of the cultural periods best represented in the study area and can be used for site location analysis. Chapter IV presents an overview of the Archaic period for the northern San Juan Basin. The Archaic period is 118

PAGE 131

characterized as having had a hunting and gathering mode of adaptation where particular resources were being exploited. By understanding the environmental factors that influenced decisionmaking in the location of sites, one can project where other areas might exist that contain similar site densities. The concept of site projection is beyond the scope of this report, but by using the archaeological inventory assessment information (Chapter V) and the results of the spatial analysis for identifying potential land use conflict areas (Chapter VI), one can plan for the avoidance or mitigation of impacts to cultural sites. This concept is discussed in Chapter VII. The data themes used for analysis are the same administrative themes described in Chapter V. Procedures for Using MOSS/MAPS to Prepare Data for Identification of Potential Cultural Resource and Land Use Conflict Areas The federal archaeologist analyzes cultural site data to meet the planning and compliance requirements that include preparing resource management plans, environmental assessments, and environmental impact statements. To accomplish these work requirements, the archaeological inventory data must be assessed for potential biases (Chapter V) and the potential for cultural resource and land use conflict areas must be identified (Chapter VI). This section presents the steps followed in preparing data sets for spatial analysis using MOSS/MAPS. The results of this analysis can then be interpreted by identifying potential conflict areas between cultural resources and specific land uses. 119

PAGE 132

The aim of this study was to select all Archaic period sites in the study area that are encoded in the Archaeological Records Management System (ARMS) and overly that data with each administrative theme. See Chapter I for a discussion of ARMS. Several major procedures were required on the GIS to prepare the Archaic site data for overlaying with each of the administrative themes. The administrative themes were prepared to be in the same format as the inventory data (Chapter V). After the archaeological site data and associated attribute information was received from the New Mexico Laboratory of Anthropology it was converted to MOSS format using a conversion program that was developed at the BLM Service Center in cooperation with the BLM New Mexico State Office in Santa Fe. This information was sent to the Service Center on tape. The area on the site map corresponding to the study area was extracted along with all the site and multiple attribute data contained therein. Following this step the ARMS attribute information was searched to extract only those sites that had an Archaic period component. This resulted in a map containing 137 locations representing only the Archaic sites. This map was then converted to a polygon format by buffering the 137 sites using their respective sizes as contained in the ARMS data base. This step was performed as an aid to plotting the sites and to obtain area measurements for the sites. An optional step is to extract only those Archaic sites that have a specific National or State register status. This information is also provided in the ARMS data base. The map containing the 137 polygons representing the sites was then converted to cell-format in order to overlay it with each of the 120

PAGE 133

administrative themes. Once overlaid, the sites that coincided or intersected with particular administrative theme subjects were then shaded, and area figures were calculated for those intersected sites. One limitation of working with the site data in cell format was that all attribute information was lost. The remainder of this section lists the technical procedures used for preparing the archaeological site map for analysis with the administrative data themes. This technical section is intended for those readers who have had some experience with MOSS/MAPS or another GIS. Figures 6.1 and 6.2 are cartographic models that schematically show the steps followed in preparing the data for analysis. The circled numbers in Figure 6.1 and the circled letters in Figure 6.2 correspond to the steps listed below. Figure 6.1 shows the steps (1 through 13) followed in preparing the archaeological site data for analysis. Figure 6.2 shows the steps (A through J) needed to prepare the site data and administrative data themes for spatial analysis. Using point data to represent archaeological sites requires two fundamental assumptions. First, all sites are assumed to be circular when buffered in MOSS and the digitized point is assumed to be located at the actual center of the site. The circular shape is an abstraction of reality needed for the GIS algorithm to perform its function. In point of fact, few sites are perfectly circular. Digitizing sites as polygons to their actual configuration would preclude the need for this abstraction. The shape of the buffered sites does change slightly, however, once rasterized. The shape of the site assumed in cell format largely depends on its size in vector format as a circular polygon. 121

PAGE 134

The second assumption is that a cell is assigned the class of the first location that is inside the cell. Some of the points representing sites can be 11lost11 if more than one site location occurs in a cell, since the cell can be activated to indicate the presence of sites but not the number of sites. By using small cell size, such as 30m X 30m, the potential of having more than one site per cell is small. The fo 11 owing is the 1 i st of the processing steps in necessary to prepare the archaeological site data for analysis. All commands are capitalized. MOSS commands are underscored, whereas MAPS commands are not. 1. The archaeological site and associated attribute information is kept on the Archaeological Records Management System (ARMS), a data base containing only alphanumeric data. The data base is maintained and updated regularly by the Laboratory of Anthropology, an agency of the New Mexico State Government. This data base management system is used by the state for data retrieval using an IBM mainframe computer . 2. To prepare this data base, the Laboratory transferred UTM site location information from field reports to a base map. A digital planimeter was then used to verify the location. This site data was then encoded into ARMS. 3. A special conversion program was developed at the BLM Service Center to convert the data from the IBM format to format. At the same time, any site that did not have good or excellent location accuracy, as coded on the data file, was omitted by the program. Specific criteria for determining good and excellent locational information were developed by the Laboratory. The MOSS graphic data base for this project was stored on 1 :100,000-scale maps subdivided into four columns, each being two 7.5-minute quadrangles wide. The 1 :100,000-scale was divided because of the limitations of the MOSS ADD command, which accepts only 2,750 unique subjects (sitesr-in any one run. 4. The file containing the point data for the sites was then linked to the associated attribute files from the ARMS data base. Steps 3 and 4 were performed outside of MOSS/MAPS at the BLM New Mexico State Office. 5. The site and attribute data was copied onto tape and sent to the BLM Service Center for manipulation on MOSS/MAPS. 124

PAGE 135

6. GENERATE a WINDOW border using UTM coordinate pair data for the fourquadrangle study area. The WINDOW was expanded beyond the edge of the map to create even multiples of the 30 m X 30 m cell size. In this way, each cell map had the exact number of cells, with no rounding error. By enlarging the WINDOW, the edge of the map did not obscure the border when viewed on the terminal screen. 7. Overlay using LPOVER the portion of the 1:100,000-scale map containing the s1te data with the WINDOW border. The resulting map included only the archaeological site data for the four-quadrangle study area (603 sites). 8. BSEARCH the map•s 603 sites to select all sites and occupations (components) dating to the Archaic period with their multiple attributes. This point data map had a total of 137 Archaic sites with multiple attributes. 9. Use REPORT to generate a tabular listing of the multiple attr1bute data base for the map with the 137 Archaic sites represented as points. 10. PLOT the map of 137 Archaic sites as points on the terminal screen to view the distribution of the sites. 11. NUMBER the map of 137 Archaic sites to display the ID number of the sites on the map. 12. EDITATT using the cursor option to interactively edit the contents of the map•s multiple attribute file. The objective is to place all of the Archaic period attribute information into the first of the three nested temporal matrices (components) of the ARMS file. This procedure was performed in order to retrieve attribute information from only one set of nested attribute matrices, rather than from all three. Most of the Archaic sites had been encoded in the first nested matrix for the first component. This step was performed because ARMS is not a relational data base. 13. The remaining attribute 11keys11 were then deleted with UTILITY using the 11attribute11 and 11delete11 options. This step was performed to reduce the processing time when using BSEARCH to search the multiple attribute list. A. BSEARCH the 137 Archaic site point map nine times to obtain a separate map for each of the nine site size ranges encoded in the ARMS file. B. BUFFER each of the nine site size maps containing points using the radius calculated for the value at the upper end of the size range (Size ranges and not actual site sizes are provided in ARMS file). This procedure will give a maximum estimate of 125

PAGE 136

the area to be BUFFERED for the sites occurring in that size category. The "resolve overlap11 option for the BUFFER command was not used, because inaccurate area figures were being calculated. This is because the portion of the algorithm responsible for selecting new polygons out of the area of overlap in intersecting polygonswas not working correctly. As a result, each site retained its own BUFFERED area. Thus, all point data is converted to polygon data. Note: Sixteen sites were 11lost11 after RASTERIZING the BUFFERED data because more than one of the polygons of the smaller size ranges fell within a single 30m X 30m cell. This resulted in the loss of one or more cell frequencies for that particular site size map. This problem was experienced for some or all of the sites having 25 sq. m, and 500 sq. m areas. A larger cell size would have resulted in greater site .. losses ... The problem was resolved by SELECTING by item number only those sites that did not receive a cell frequency and rebuffering them using the upper end of the range of ARMS site size category 4, which was 1,000 sq. m. No sites were lost at this area size when the BUFFERED sites were later RASTERIZED. Although the BUFFERED Archaic sites may not have had highly accurate areas calculated, the sites could easily be seen on each graphic display. C. MERGE all BUFFERED site maps to produce a map containing the BUFFERED 137 Archaic sites that represent area approximations for each of the sites. D. RASTERIZE the BUFFERED 137 Archaic site map to convert to cell format with background and non-background values. A 30 m X 30m cell size was used to obtain a high resolution and to minimize the loss of sites in the event more than one polygon fell within a cell. After buffering, only one polygon (site) was lost. Note: Administrative themes were previously RASTERIZED to be in the same format as the inventory data (Chapter V). E. INTERSECT each of the RASTERIZED administrative data themes with the RASTERIZED BUFFERED 137 Archaic site map to create a new discrete map show1ng logical intersection of sites with the subjects of each administrative theme. F. PLOT and SHADE the INTERSECTED map on the terminal screen to visually inspect the areas of each administrative theme where proposed or existing development is conflicting with existing Archaic sites in the area. G. AREA the INTERSECTED map to obtain the acreage of Archaic sites that coincide with areas of planned development or conservation. 126

PAGE 137

Steps H, I, and J are optional. These steps would be followed if information on the National Historic Register status of the sites was required. H. REPORT to generate a tabular listing of the multiple attr1bute data base for the 137 Archaic site map in point data format. I. BSEARCH the 137 Archaic site map to select all sites with the desired national or state register status category as encoded in ARMS. The status categories include eligible, not eligible, on state register or in state register district, on national register or in national register district, unknown or not yet determined, and informal opinion that site may be eligible. J. BUFFER the 137 Archaic site map as described in step B above. Once the Archaic site map with the selected register status attributes has been BUFFERED, it can be converted to cell format and processed as described above for steps C through G. Note: Attributes obtained using the BSEARCH command from the ARMS data base are 11lost11 when the map is RASTERIZED, but the graphic does represent those sites that contained the desired attributes before conversion to cell format . Methodology Used for the Analysis of Potential Cultural Resource and Land Use Conflict Areas The purpose of the analysis is to prepare maps for the identification of potential conflict areas between the Archaic sites and land uses in the study area. In addition, particular areas that do not pose a threat to the cultural resources can be identified. Because the identifying of potential land use conflict areas occurring with cultural sites involves an assessment of spatial relationships, the GIS is a tool appropriate for meeting this need. The maps used for analysis are the result of overlaying each of the 127

PAGE 138

administrative themes with the cultural site data (see preceding section). The resultant maps show only the Archaic sites that were 11intersected11 with subjects of the administrative themes. For example, the map resulting from overlaying the PRLA map with the Archaic site map is a graphic of Archaic sites that .. intersected .. with particular lease applications for coal. By intersecting these maps, area figures can be obtained for each intersected site and a means can be provided to calculate the total acres of sites occurring on a particular lease. The PRLA map is then plotted with the map showing the ''intersected .. sites, thus depicting the spatial relation of intersected sites to their subjects. For estimating the time needed to excavate sites to recover data, the number of sites occurring in an area of planned development may be more useful than area information. Summary tables showing the number of acres of sites intersected with administrative subjects are given for two themes as examples. Four of the maps have been enlarged for portions of four of the administrative themes using the ZOOM command in MAPS. The maps were annotated using the TEXT command in MOSS. This textual information was plotted on the maps interactively at the computer terminal. All plots were made using an Anadex printer. Because the maps are used only as graphic outputs for potential conflict analysis, no scales or legends are shown. As mentioned in Chapter V, exact location information can be obtained using the QUERY command in MAPS, and a landline map can be plotted over any of the maps shown. The linear features --dirt roads and transmission line --were not buffered for the same reasons given in Chapter V. For some types of development projects, however, linear features may have to be buffered to 128

PAGE 139

ensure that a specified width will be inventoried for cultural resources. Rural roads may also have to be buffered to ensure that all sites visible within a specified distance from the road are also inventoried. This buffer would help protect the sites against possible vandalism. Results of Analysis The maps and tables shown in this section are grouped according to administrative themes that reflect nondevelopment and those that show where there is proposed or existing development. Most of the interpretation of the analytical results is based on visual inspection of the spatial data rather than measurement. Ultimately, all land use planning requires an assessment of the spatial context within which development is planned. An important point to keep in mind is that in each of the subjects for the administrative themes, e.g., a particular coal lease application or WSA, the possibility exists that more Archaic period sites or sites of other cultural periods exist. If the entire subject has not been inventoried, there is a potential for more sites to occur in these noninventoried areas. Each overlay represents only the Archaic sites found to date. The amount of area inventoried has been calculated previously in the section of Chapter V involving the results of the analysis. See this section, particularly Tables 5.9 through 5.17. Nondevelopment Themes Figure 6.3 shows the result of overlaying the range data with the Archaic site data. The single range allotment, Paragon, does have several sites in its east-central portion. Cattle grazing does not 129

PAGE 140

• • . • ., • ALLOT II. tNT •• . • . r • FIGURE 6.3 Map of range data overlaid with Archaic site data 130

PAGE 141

usually a conflict with cultural resources, because cattle do not significantly impact the resources. But range improvements, such as fences, gates, windmills, cattle guards, troughs, catchments, pipelines, pump jacks, chaining, and seeding can all potentially disturb surface manifestations of cultural sites. Figure 6.4 shows the result of overlaying the proposed ACEC data with the Archaic site data. The map shows the distribution of sites with respect to the Bisti/De-na-zin and Fossil Forest, which are the two proposed ACECs. Three large sites occur in the eastern portion of the Fossil Forest. These three sites and their site numbers are shaded in the enlarged map (Figure 6.5) of the Fossil Forest. The Bisti/De-na-zin proposed ACEC also contains sites that tend to concentrate in the southeast part of the area. A few sites also appear in the northern and far southwest portion of this proposed ACEC. Because no development is allowed in these two proposed ACECs, the Archaic sites occurring there are not in jeopardy of destruction. Figure 6.6 shows the result of overlaying the WSA data with the Archaic site data. Parts of the Bisti and De-na-zin WSAs contain a few sites, with most of them clustering near the southwest part of the De-na-zin WSA. Once again, no development or uses that create surface disturbances are permitted; therefore, the sites occurring in these areas are not in conflict with this use. Figure 6.7 shows the result of overlaying the land status (ownership) data with the Archaic site data. Figure 6.7 has been shaded for the different ownership classes. These classes include Indian, Indian withdrawn, state, BLM and private. See Chapter V for a discussion of Indian and Indian withdrawn lands. On this map, some of the Archaic sites are obscured by the shading. Although this graphic does not 131

PAGE 142

B'STI/OE-HA-11H . fOSSIL . . . . •'' FOREST •.c 0 .. .a . . .'t D, Oe 0 • • ... , . c l! I .o • . , . • . . c • FIGURE 6.4 Map of proposed ACEC data over l aid with Archaic site data l 3 2 0

PAGE 143

) r J ...__, . I rOSSIL , 4213@ roREsr i 42';5@ I -I ! j r FIGURE 6.5 Mao of Fossil Forest data overlaid with Archaic site data 133 J

PAGE 144

0 ;, .. . ... c. t • • 0 •.a Q ,a . ,Iii . . . ' FIGURE 6.6 of WSA data overlaid with Archaic site data 134

PAGE 145

FIGURE 6.7 Map of land status data overlaid with Archaic site data -Indian .,.,..,.., ;.'// Private 1# State 111111 BLM l \ J Indian withdrawn 135

PAGE 146

pertain to any particular land use, it does reveal who will administer the parcels which, in turn, may influence the manner in which cultural resources are managed. Development Themes For some of the subjects and sites described below, the QUERY command in MAPS was used to obtain exact location information. Figure 6.8 shows the result of overlaying the WSA linear development data--dirt roads and transmission (TM) line--with the Archaic site data. The unimproved dirt or two-track roads and transmission line occur only along the southern edge of the two WSAs. The roads and transmission line are represented on Figure 6.8 by the jagged lines running next to the subject labels. Most of the sites tend to cluster near the southwest portion of the De-na-zin WSA near one of the road segments. By enlarging this portion of the map, one could examine which sites are in easy viewing distance from the road and therefore need to be monitored more closely. Figure 6.9 shows the result of overlaying the road network data and the Archaic site data. Many sites occur close to several secondary unpaved roads, reflecting surveys for road clearances. A portion of the map could be enlarged for large-scale viewing. These roads could be buffered to a desired width to show which sites could be affected if the road corridor was widened to a specified distance. Buffering can aid in telling which sites may be seen from the road and, are thus susceptible to vandalism. 136

PAGE 147

0 .. .. . . . -•.a ... .o . 0. 0 ' . . 0 . .1:1 .. . . Cl DE:-NA.-ZIN WS.t.. Cl p D . D, ... ! I • c .?' • • .. 00 " FIGURE 6.8 Map of linear development in WSA data overlaid with Archaic site data 137

PAGE 148

SECO NOA.,RY ROA.D UNPAVED . -" 0 ,Q • • {> " . 0 . .. SECONDARY ROAD / • I) •J.I I .. DO D SECONDARY ROA.D UNP.4VED 0 FIGURE 6.9 Map of road network data overlaid with Archaic site data 138

PAGE 149

Figures 6.10 and 6.11 show the result of overlaying the PRLA data with the Archaic site data. PRLA NM-3752 and MN-3754 tend to have the highest site densities. Figure 6.11 is an enlargement of these two PRLAs. Fifteen sites lie within these PRLAs. Three of the larger sites have been shaded and their site numbers are given. These specific site numbers and cell values were identified using the QUERY command in MAPS, and the map was enlarged using the MAPS ZOOM command. Table 6.1 summarizes the acreage figures of the 15 sites occurring in each of the PRLAs. The MAPS AREA command was used to obtain the acreage figures. Archaic site 20078 is of particular interest because it lies on the boundary of the two leases. Thus, area figures can be obtained for the northern portion of the site that lies in PRLA NM-375 and in the southern portion of PRLA or 3754. All sites occurring within these two PRLAs would have to be assessed for their significance. Once significance has been established, particular sites can be avoided or excavated to recover the archaeological information prior to destruction from surface mining. Figures 6.12 and 6.13 show the result of overlaying the data for the Expressions of interest in coal leasing with the Archaic site data. The tract NM-1031 has the highest site density. As in the PRLA example, these sites can be enlarged and the exact locational information obtained using the MAPS ZOOM and QUERY commands, respectively. Figure 6.14 is an enlargement of NM-1093 plotted with the intersected site data. The AREA command in MAPS can then be used to obtain acreage figures for each site within NM-1031. As with the PRLAs, if the tracts are eventually developed, cultural resources may have to be avoided or impacts mitigated through excavation. 139

PAGE 150

0 . . . . . . .. PRLA • .'. •'' . r . • PRLA NM-3754 D 0 .. . . ,; "Lfp 0 FIGURE 6.10 Map of preference right lease applications (PRLA) data overlaid with Archaic site data 140

PAGE 151

D PRLA NM-3752 {W20099 • PRLA NM-3754 • • • } • • • Q . II FIGURE 6.11 Enlarged map of preference right lease applications (PRLA) data overlaid with Archaic site data 141

PAGE 152

TABLE 6.1 AREA TABLE OF PREFERENCE RIGHT LEASE APPLICATIONS (PRLA) DATA OVERLAID WITH ARCHAIC SITE DATA FOR TWO LEASES PRLA NM-3752 NM-3752 NM-3752 NM-3752 NM-3752 NM-3752 NM-3752 NM-3752 NM-3752 NM-3754 NM-3754 NM-3754 NM-3754 NM-3754 NM-3754 Site Number 34685 34684 38959 34683 38954 34680 34687 34678 20078 20078 20079 20099 20084 20091 20077 14927 142 Acres .44 1.33 1.33 1.33 1.11 12.23 .22 1.33 10.23 2.22 .22 12.45 .22 .44 .22 .22

PAGE 153

0 .. . •'' . ... , , . . 0 0 l.o ,0 • 0 .. . . FIGURE 6.12 Map of industry expressions of interest (IEOI) in coal leasing data overlaid with Archaic site data

PAGE 154

0 0 0 r D 0 • (]0 0 {;) Q D NM-1 \ II j D l ( 0 D 0 0 0 FIGURE 6.13 Enlarged map of industry expressions of interest (IEOI) in coal leasing data overlaid with Archaic site data 144

PAGE 155

Figures 6.14 and 6.15 show the result of overlaying the oil and gas lease data with the Archaic site data. In Figure 6.14, oil and gas leases with the highest site density area are shaded. Figure 6.15 is an enlargement of that shaded area. Four large sites (28800, 28802, 28805 and 28803) occur in this lease area and are shaded. Here again portions of some sites occupy the space of more than one lease. Table 6.2 shows the areas associated with the sites or portions of sites that fall within particular leases. Only site 28803 lies within a single lease. If portions of these oil and gas leases are developed, these sites or portions of them may be endangered. Conclusions Chapter VI has focused on the commands and procedures used in MOSS/MAPS to conduct spatial analysis that will aid in interpreting conflict areas between cultural resources and specific land uses. The ascribing of conflicts between cultural resources and specific land uses must be based on experience. A GIS can assist in conducting spatial analysis as an aid to identifying potential conflict areas between cultural values and specific land uses. But a GIS cannot identify conflicts, as only an experienced planner or archaeologist can interpret the results of the GIS-aided spatial analysis to identify cultural resource and land use conflicts. 145

PAGE 156

OIL/GAS L.E:AS(S FIGURE 6.14 Map of oil and gas lease data overlaid with Archaic site data 146

PAGE 157

0 OGL NM23471 OGLNII426660 2!SO OCI.NM25tl47 FIGURE 6.15 Enlarged map of oil and gas lease data overlaid with Archaic site data 147

PAGE 158

TABLE 6.2 AREA TABLE OF OIL AND GAS LEASE DATA OVERLAID WITH ARCHAIC SITE DATA FOR FOUR LEASES Oil and Gas Lease OGL-NM 23471 OGL-NM 29344 OGL-NM (Null) OGL-NM 26660 OGL-NM 26660 OGL-NM 25847 OGL-NM (Null) OGL-NM 2660 Site Number 2880 2880 28802 28802 28805 28805 28805 28803 148 Acres 4.52 8 .45 .22 2.22 2.67 2.67 7.12 1.33

PAGE 159

CHAPTER VII USING A GIS IN CULTURAL RESOURCE MANAGEMENT SITUATIONS A GIS can serve as a to meet cultural resource management requirements. Some of the results of the analysis in Chapters V and VI are used to address the questions posed in the two scenarios described below. The management questions and problems to be solved lend themselves to solution through spatial analysis. They are intended to represent questions or problems that a BLM district or resource area staff archaeologist or planner would have to answer to meet management and compliance needs. BLM's main problems for cultural resource management planning and compliance fall into two categories: (1) the location of inventory area and distribution of sites and (2) areas of potential conflict between cultural resources and other uses land that would disturb cultural resources. The two scenarios that follow are based on meeting compliance requirements. Scenario 1 To develop an environmental assessment, an archaeologist needs to know which preference right lease applications (PRLAs) for coal have the highest Archaic site density and what portion of each PRLA has been inventoried. The archaeologist also needs to know the geologic surficial process units that have higher site densities. We wish to determine if an association exists between the Archaic site locations and geologic surficial process units. This information will prove useful in 149

PAGE 160

estimating Archaic site densities in noninventoried areas that contain the same surficial process units. A description of the MOSS/MAPS processing steps necessary to answer these questions follows. The map containing the PRLA data overlaid with the buffered Archaic site data is plotted. This map shows only the distribution of intersected sites. To see these sites relative to the PRLAs, the PRLA map must also be plotted. When the two maps are plotted simultaneously (Figure 7.1) we see that PRLAs NM-3752 and NM-3754 have higher Archaic site densities than other PRLAs in the study area. Using the results of the inventory assessment in Chapter V, Table 5. 16, we can see that PRLA NM-3752 has 88.99 percent of its area inventoried, whereas PRLA NM-3754 has 76.75 percent of its area inventoried. Figure 7.2 shows a portion of the PRLAs and their inventoried areas; the shaded areas show where inventory has been conducted. This map was generated using the ZOOM, QUERY, and SHADE commands in MAPS. The map was annotated with the TEXT command in MOSS. Figure 7.3 shows an enlargement of the two PRLAs with plotted sites. The three largest sites are shaded and labeled with their respective numbers. Figure 7.4 is a map showing the results of intersecting the PRLA map with the geologic surficial process unit map. The two PRLAs studied here have been enlarged using the ZOOM command in MAPS. The Archaic site map has also been plotted over the intersected map and the three largest sites have been shaded. The smaller sites were left unshaded. Through the use of the QUERY command one can determine that the geologic process unit most common for these PRLAs is Qtps, which is a Quaternary unit that occurs in stable upland surfaces such as pediments and terrace 150

PAGE 161

• • . . . . . . . .. . . . . PRLA •• . .. , . l.a 0. , . •' PRLA NM-3754 • • 0 FIGURE 7.1 Map of distribution of Archaic site data relative to preference right lease applications (PRLA) data 1 5 1

PAGE 162

FIGURE 7.2 Enlarged map of preference right lease applications (PRLA) data overlaid with inventory data 152

PAGE 163

D PRLA NM-3752 PRLA NM-3754 a • • • • Cl • • • • • • FIGURE 7.3 Enlarged map of preference right lease applications (PRLA) data plotted with Archaic site data 153

PAGE 164

Otps D • • • D • • , • FIGURE 7.4 Enlarged map of Archaic sites plotted on preference right lease applications (PRLA) data as overlaid with geologic surficial process units data 154

PAGE 165

remnants and has erosional stability and aeolian (wind) deposition. The surficial material consists of fluvial clay, silt, and gravel and is usually indurated. By performing statistical tests for association --outside of MOSS/MAPS--the archaeologist can determine if a relationship exists between the occurrence of Qtps units and the occurrence of Archaic sites. The archaeologist would expect similar site densities in noninventoried Qtps units. Scenario 1 has illustrated the use of MOSS/MAPS in a management situation where compliance requirements would have to be met. The use of the Archaic site data overlaid with the PRLA data shows the distribution of sites and make it possible to identify where the highest densities are relative to the PRLAs. The percentage of area inventoried in the PRLAs was determined using results of overlaying the inventory data with the PRLA data (Chapter V). By assessing the geologic surficial process units data overlaid with the PRLA information and plotting the Archaic sites on the resultant map, the archaeologist can begin to identify which surficial process units are associated with Archaic sites. In this way site density projections can be made to other nonsurveyed PRLAs that contain the same surficial geologic process units. Scenario 2 A BLM resource area manager has proposed a plan to increase the width of two segments of an existing dirt road to facilitate the extension of a transmission line that would parallel the two road segments. The road segments are now about 3 m wide, and the corridor needs to be widened an 155

PAGE 166

additional 26 m, 13 m on either side of the road segments. The total width of the new right-of-way would be 29 m. The resource area archaeologist has been asked to provide the following information to assist in determining impact assessment for an environmental impact statement. 1. How much archaeological inventory has been conducted for all secondary unpaved roads in the study area? 2. the two road segments to be widened been inventoried? 3. What is the distribution of Archaic period sites with respect to the two road segments? 4. Will the additional 26m widening pose a danger to Archaic sites near the two road segments? A description of the MOSS/MAPS processing steps necessary to answer these questions follows. Figure 7.5 is a plot of the distribution of Archaic sites relative to the existing road network in the study area. The vector map was created by plotting unbuffered Archaic site data over the map containing the road network data. The map was annotated using the TEXT command in MOSS. The small rectangle shows the portion of the road network the two road segment rights-of-way are to be widened. The ZOOM command in MOSS was used to identify the location of this rectangle. Figure 7.6 presents the result of using the ZOOM command, showing the road segments with the three nearest Archaic sites. Each of the three sites has been labeled with its site number. The site numbers were obtained using the QUERY command in MOSS. The short lines shown between the road segments and the three nearest sites represent the distance measurements taken using the DISTANCE command in MOSS. Having a higher accuracy level, point and line data is better than cell data for obtaining distance measurements. But as great as a 12m 156

PAGE 167

.. • ..______ \.. • t. ....... . • • • •• .. . , . • •• •' • • • SECONDARY • • . • UNPAVED ROAD • • ,/ • • SECONDAR'r ROAD UNPAVED • • SECONDARY ROAD UNPAVED • t • -• Figure 7.5 Map of distribution of Archaic site data relative to road network data 157

PAGE 168

\ ' I \ •. .. .. + )21103 ' . . \ \ t2S288 + FIGURE 7.6 Enlarged map of distribution of Archaic site data relative to road segments 158 -

PAGE 169

error may occur in any direction for each distance because of the accuracy limitation of plotting point and line data on a map scale of 1:24,000. To compensate for the accuracy level all sites within 24m on either side of the right-of-way had to be examined for potential impacts. The distances to sites 26288, 21108, and 21103 are 115.2 m, 192.2 m, and 139.4 m respectively. Since all these distances are well beyond the 29 m potential impact zone, the increase in the right-of-way width should not affect any of the three sites. Table 5.12 and Figure 5.14 (Chapter V) present the result of overlaying the road network data with the archaeological inventory data in cell format. The data shows that about 45.47 acres or 2.47 percent of all secondary unpaved roads have been inventoried in the study area. After plotting the map and using the QUERY command in MAPS with the crosshair, the user can determine from the cell values whether the individual road segments have been inventoried. Other yet undiscovered sites could potentially still be located in the noninventoried areas . Scenario 2 has demonstrated the use of MOSS/MAPS for meeting another type of management situation. After plotting the road network and Archaic site data together in vector format, the archaeologist was able to focus on the two road segments of interest in determining the distribution of Archaic sites with respect to these segments. Once this was done, the distance between the road segments and the three nearest sites could be measured. Thus, the archaeologist can determine if these sites would be impacted by the widening of the right-of-ways. In addition it was possible to determine if the road segments had been inventoried. 159

PAGE 170

Conclusions Chapter VII has presented two examples using MOSS/MAPS to address the analytical requirements for cultural resource management. The MOSS/MAPS commands used to solve the problems were presented along with options to be considered when using rasterized line data for analysis. These two limited examples are aimed at demonstrating how a GIS can assist the archaeologist or land use planner in preparing environmental assessments and environmental impact statements. 160

PAGE 171

CHAPTER VIII EVALUATION OF MOSS/MAPS FOR MEETING CULTURAL RESOURCE MANAGEMENT NEEDS This chapter assesses the capabilities of the MOSS/MAPS software integrated with the Archaeological Resource Management System data base to meet cultural resource management (CRM) needs. Thi s chapter also discusses some of the benefits and problems of using this specific GIS for CRM tasks. A general discussion of the level of accuracy, time required for data capture and analysis, and monetary costs is also provided. Assessment of MOSS/MAPS The BLM Albuquerque District Office is establishing an automated graphics library of resource information for use by managers, planners, and resource specialists. The automated capabilities offered by the ADS/MOSS/COS system (Chapter II) can enhance each user•s ability to capture, store, analyze, and retrieve spatial data derived from maps, air photos, satellite imagery, and other forms of geographic information . The potential benefits of establishing such a system seem to outweigh the costs. Besides the direct cost-savings in time to produce actual graphics and statistics of resource a substantial cost-savings can potentially result in such things as: increased organizational effectiveness standardization ease of user access information update 161

PAGE 172

facilitated interagency data exchange increased ability to respond to new technologies and data needs. The acceptance of GIS throughout BLM signals its recognition as a prime tool in spatial analysis. In addition, most field managers view GIS as a tool to create maps, expedite the analysis of management issues, and resolve conflicts in planning and environmental documents. Capabilities and Benefits of MOSS/MAPS for CRM The MOSS/MAPS package provides flexible routines for overlay and neighborhood analysis, map description, and data management. One major strength of this package is that it is supported by several federal agencies. Versions exist for 16-and 32-bit microcomputers and minicomputers. Although MOSS/MAPS now has only limited capabilities for inferential statistical analysis, Kohler and Kvammel (1985) have noted some measurement capabilities of MOSS/MAPS for spatial analysis for locational modeling of archaeological sites. These capabilities include routines that: 1. collect a random sample of points or polygons for further analysis or for input to statistical procedures, 2. measure the distance between any two points along a path (which need not be straight) in a line map or the total distance around each subject in a polygon map, 3. identify locations with a specifiable distance of a point, line, or polygon subject map, 4. produce a three-dimensional display of an integer-valued continuous map, 5. create a map of azimuthal aspect or a slope map from a continuous-elevation map, 162

PAGE 173

6. create a map showing the visibility of locations from a specifiable observation point or points, 7. create a cross-sectional image between any two points (usually used for elevation data but suitable for any continuous map), 8. create a map showing the least effort path to a target cell so that the analyst can assign weights to features acting as partial barriers in the pathfinding process (an example of a common GIS capability for corridor analysis), and 9. create a map showing the steepest downhill path through varying terrain from a target area (essentially the path along which water would flow). The BLM is in the process of determining if the savings outweigh the high initial costs involved in gathering, compiling, digitizing, and editing of data to create accurate digital data bases. Unfortunately, no studies have been conducted to determine these upfront costs. For this demonstration project, most of the data bases had already been digitized, thus considerably reducing these upfront costs. Major costs for CRM will arise from developing a digital data base that includes archaeological inventory and cultural site data that must first be put on a data base management system. The automation of site and inventory records can be a costly venture when one considers the time required to transfer large amounts of site and inventory information to a data base management system. However, the costs of using environmental data bases can be substantially mitigated, since most of the data will be used by other resource specialists. After the initial costs have been realized, MOSS/MAPS has the potential to effectively evaluate archaeological inventory data, perform site location analyses, and examine potential cultural resource and land use conflict areas. Hopefully, much time used for preparing data as well 163

PAGE 174

as for analysis can be saved. The time saved would become apparent once the data bases and subjects are used for different projects and applications. This demonstration project presented some analysis procedures directed toward meeting CRM needs. The same analysis using conventional methods would have required much more time. Mylar overlays would have to be hand-drawn for 7.5 minute quadrangle sheets for each of the environmental and administrative themes. In addition, separate overlays would have had to be prepared for the inventory and site data sets, and resource data would have to be manually transferred from base maps for each resource to overlays. The analysis would have required that specific inventory locations, site location data, and site sensitivity and evaluation data be obtained from file drawers or from maps of different scales. And as overlays would have become more numerous, identifying and accurately measuring inventoried and potential conflict areas would have become more difficult. After the completion of the most tedious and expensive work--digitizing maps for the data base, correcting digitizing errors, correcting and merging digital elevation models--the GIS analysis was performed using the BLM Service Center's Data General 330 microcomputer. When the computer was being lightly used, the electronic overlaying took 5 minutes to an hour, depending upon the complexity of the maps. In contrast, applying all these procedures by hand would have required several weeks of intensive work. Problems of Using MOSS/MAPS for CRM Some technical difficulties arose during data preparation and analysis. Some of these problems have already been mentioned in Chapters 164

PAGE 175

V and VI. Although the difficulties found in using MOSS/MAPS for performing analytical tasks may seem numerous, all the problems can be resolved through software modification. This section discusses some of the problems found in using the ARMS data base as integrated with MOSS/MAPS and offers recommendations for enhancing MOSS/MAPS and ARMS for CRM work. Data Preparation Problems Although the Digital Elevation Models (DEMs) used in this study were formatted on the Interactive Digital -Image Processing System (!DIMS) at the BLM Service Center, 7.5 minute DEMs standard-format tapes can be read directly into MOSS/MAPS. !DIMS was used for formatting the DEMs because the system is faster than MOSS/MAPS. The DEMs tape file is read using the Data General Advanced Operating System (AOS) COPY command. The MAPS IMPORT command (with the Format = DEMs option) then reformats the OEMS disk file to create a cell file. The DEMs used in this study were formatted in !DIMS, but several problems and limitations were noted in the manipulation of the models (McKinley, In press). 1. DEMs files are generated using various processes, i.e., auto-correlation and manual correlation. Data generated by manual correlation tend to have less extraneous information than auto-correlated data. This phenomenon may not be of significance unless two extremely varying quadrangles are merged and displayed. To make the data values less variable, the auto-correlated data can be slightly smoothed. A moving window (3 x 3 cells) smoothing alogrithm does not now exist in MOSS/MAPS, but data files may be smoothed in !DIMS and transferred to MOSS/MAPS as an alternative. 2. When merging multiple DEMs files into one map, zero or fill values are likely to be generated along the quadrangle abutment lines. To avoid later problems with 165

PAGE 176

the data set, zeros within the merged map should be located and modified to an adjacent cell value. This process of zero replacement can be conducted in MOSS/MAPS through a sequence of commands that requires significant analyst effort and system disc space, depending on the number of quadrangles within the merged data set. For merged data sets with 10 or more quadrangles and substantial zero value problems, BLM data analysts should consider using !DIMS for fixing the zero problems and building the merged data set. For this project, the zero problems were corrected to some degree on the !DIMS system, but some zero values remained along the quadrangle edges. 3. Examination of multiple DEMs files has shown the presence of extraneous data or elevation data errors in most DEMs files. This problem can most easily be observed in areas of known flat terrain (0-2 percent slope), but the errors are not limited to those slopes. These errors are difficult if not impossible to eliminate and are a problem an analyst must accept. These errors can be reduced by avoiding excessively detailed elevation class, slope class, and aspect class output products. For example, slope class maps with a breakdown in slope of 0-3 percent, 4-6 percent, 7-10 percent, etc. will appear to exhibit significantly more erroneous data than will a slope class map with a more inclusive slope class breakdown (i.e., 0-10 percent, 11-20 percent, etc.), especially in areas of flat terrain and water bodies. This study used slope classes of 0-5 percent, 6-10 percent, 11-15 percent, 16-30 percent, and 31-50 percent. 4. The MAPS SLOPE command has been found to generate erroneous slope values for cells near the edge of a given elevation map. This error is caused by including nonmap data (fill data outside the map but within the data minimum bounding rectangle (MBR), in the slope calculations for edge cells. The slope algorithm uses a 3 x 3 cell window to calculate slope values for given target cells. For example, edge cells with elevations of 1,000 mare compared to adjacent fill cells with values of zero, and a large slope value is calculated. A modification of the SLOPE command has been suggested to eliminate fill cells (or any cell values specified by the user) when performing slope and aspect calculations. Planned modifications to existing software and the potential for new command development will enhance MOSS/MAPS capabilities for reducing the problems associated with using DEMs. In the interim, processing 166

PAGE 177

alternatives available through the BLM Service Center technical and system support should be explored . The soils data used for this project were digitized by Technicolor Government Services (TGS), a government contractor in Fort Collins, Colorado. When TGS reformatted the soils maps, which were originally prepared in the Farmington Resource Area Office, the Automated Digitizing System (ADS) calculated the projection of the maps to be in UTM zone 12. But the default UTM zone for the entire state of New Mexico (as defined by the BLM Albuquerque District Office) is zone 13. Therefore, the MOSS vector data projected by TGS in zone 12 had to be reprojected to match the projection of all other maps' in the District Office. This task was done using the PROJECTION command in MOSS. In future digitizing assignments, the user should specify in which UTM zone the maps are to be digitized. To ensure that all base maps in this project had the same projection parameters, the option of the MOSS UTILITY was used. TGS also digitized the geologic surficial process maps that were loaned to the BLM by Dr. Steven Wells of the University of New M exico. However two adjoining maps had previously been digitized by the BLM Albuquerque District Office. Several problems resulted from having these maps digitized at different locations. First and most important, the line and polygon data did not always match with lines and polygons of the same features that extend onto adjacent map edges. This lack of edge matching, which resulted from some of the features not having been initially mapped, required that the edges of the plots of the original base maps be redrafted. For the redrafting, the cartographer responsible for mapping the original surficial process unit maps was consulted, and 167

PAGE 178

both the BLM and TGS plots were changed. TGS performed the final editing on both the BLM and TGS plots. This problem was the result of the initial mapping process, not the digitizing process. Other problems with the geologic surficial process maps included not having all attribute labels for polygons and having a few adjacent polygons with different attribute labels. TGS made most of these corrections using the Automated Digitizing System (ADS) prior to conversion to MOSS. TGS also reprojected the maps to UTM zone 13 in ADS. Before all these problems were resolved, almost 4 months had passed, and much time was lost in the process. We recommend that when additional digitizing is expected for existing data bases, that all work be performed at the office where the original data sets were first digitized. In this way, standard data compilation, digitizing, projection, and editing procedures can be adhered to and similar digitizing menus can be developed. Perhaps the most important recommendation regarding the digitizing of data themes is that all those who compile data exercise careful planning and pre-digitizing quality control. By following this recommendation, the user will greatly reduce the possibility of delays due to data-compilation errors. The compiler should know the data quality requirements before submitting any maps for digitizing. These requirements can be learned in a digitizing course or through consultation with a digitizing team coordinator. As of this writing documentation does not exist for standard data compilation procedures. 168

PAGE 179

Problems Using Point and Line Data The use of point data to represent archaeological sites requires three fundamental assumptions. First, points do not occupy space and therefore have no area. For this reason, sites are buffered in a GIS to create polygons which allow for area measurements. Instead of initially digitizing sites as point data, sites should be digitized as polygons that approximate their actual shape. But if small-scale maps are used, it might be possible to map the sites as polygons. As discussed in Chapter V, buffered archaeological site data may not give accurate area results, depending on the limitations of the software to accurately buffer point data to a specified radius. In addition, it is difficult to buffer sites in raster format because what was a point before is now a 30 m X 30 m cell. Problems can result from attempts to buffer small or even medium-sized sites. Chapter VI contains more assumptions and potential problems to consider when working with point data to represent cultural sites. Step 8 of the section dealing with the procedures used to prepare site data for conflict analysis discusses the problems of trying to buffer sites using size information from the ARMS data base. As that section points out, some of the buffered sites were 1 ost after being rasterized because more than one site with a small area fell within a single 30m x 30 m cell. When the same data were rasterized before being buffered, only one site was lost. If a larger cell size had been used, such as 50 m x 50 m, the probability of more than one site occurring within a single cell would increase. 169

PAGE 180

Linear data also present some special considerations for CRM. Some of these considerations were mentioned in Chapters V and VI. Linear features such as roads and transmission-line corridors may have to be buffered for certain types of development projects to ensure that a specific corridor width is surveyed to identify cultural resources. Once cultural resources are identified, a plan for avoidance or mitigation can be developed. As of this writing, MAPS has no distance command to perform the same function as the DISTANCE command in MOSS. The MOSS DISTANCE command is designed to measure distances in vector format. By definition, line data do not have area; therefore, vector rather than rasterized line data should be used for distance measurements. Because of a possible error of plus or minus l/2 the cell width, inaccuracies will result if the user tries to measure road widths in cell format. Thus, vector data is best used for measuring distances from linear development features to specific cultural sites. The Relational Data Base Management Approach The BLM Service Center is preparing a prototype version of MOSS that can interface with a relational data base management system (DBMS). This prototype, however, does not apply to MAPS. Because MOSS does not now have a built-in DBMS, the prototype will be more powerful as well as flexible than MOSS as it exists now. The advantages of such a design are that it will (1) handle data more efficiently, (2) permit the use of elaborate report-generator facilities amenable to most DBMS, and (3) enable data bases to be accessed from the DBMS directly without having to go through MOSS. 170

PAGE 181

The ARMS data base is now kept on an IBM mainframe using DBMS. Once the prototype is completed, flat files containing multiple attribute information, such as ARMS, can be entered and reformatted using the relational DBMS. We suggest that the ARMS data base be entered into a relational DBMS in MOSS. With the multiple attribute information in ARMS accessible through the MOSS DBMS, the difficulties with nested matrices as noted in this project will be alleviated. However, prototype will not handle cell data, the use of multiple attribute information would still be limited. By importing ARMS data into a relational DBMS, the user can take advantage of sophisticated data base operations rather than relying on the current MO"SS operations, which have limited capabilities. This version of MOSS is only a prototype, and a fully operational production version may not exist for some time. Cost Estimates This project was funded largely from BLM's coal management program (subactivity 4121). The costs incurred by the project reflect the items that had to be procured, including digitizing services, training, and incidental costs, such as research materials. Other than the principal investigator's work months, the greatest cost was for DEMs and digitizing work. This section gives cost estimates for the time required to digitize the data themes used in this project. These estimates were provided by Robert Bewley, geographer at the BLM Albuquerque District Office. For the soils, surface geology, and the geologic surficial process unit maps, actual digitizing time is provided for each quadrangle used in the study. TGS the digitizing contractor, supplied the digitizing times. 171

PAGE 182

The BLM Service Center procured the DEMs from the USGS at the scale of 1:24,000. The cost for each 7.5 minute quadrangle was $100 plus $25 for the tape on which they were sent. The total cost for the four DEMs used in the study was $425. Table 8.1 presents an estimate of the average time required for digitizing and editing the data bases used in this study. These data bases were previously digitized at the BLM Albuquerque District Office, where actual digitizing time records were not kept. The estimates in Table 8.1 were based on the time required to digitize themes for a 7 . 5 minute quadrangle at a scale of 1:24,000. As of this writing, TGs•s digitizing rate is $17.83 per hour. As mentioned before, the soils, surface geology, and surficial geologic process units were digitized by TGS. Table 8.2 shows the actual time for digitizing all or portions of the 7.5 minute quadrangles that existed for these themes. However, the actual digitizing times do not include supervision time for quality control of plots. Other major costs that must be considered in a CRM project involving a GIS are the costs for training employees. But introductory training in ADS and MOSS/MAPS is only the first step for GIS familiarity. Based on the experience of this project, about 3 to 6 months of regular hands-on experience is required before the user is capable to begin a project similar to the one presented here. 172

PAGE 183

TABLE 8.1 ESTIMATED TIME FOR DIGITIZING AND EDITING SELECTED THEMES DATA THEME AVERAGE HOURS Stream Data 1.5 Wilderness Study Areas (WSA) 2.0 Linear development in WSA 1.0 Road networks 1 .0 ACEC 2.0 Range allotments 3.0 Oil and gas leases 2.5 PRLA 2.0 Industry expressions of interest in Coal leasing 2.0 Land status (ownership) 7.0 NOTE: All estimates are based on one 7.5 minute quadrangle at 1 :24,000-scale and of average complexity. Digitizing time and editing will vary with the number of points, lines, and polygons to be digitized. SOURCE: Robert Bewley (geographer) BLM Albuquerque District Office 173

PAGE 184

TABLE 8.2 ACTUAL TIME FOR DIGITIZING AND EDITING SELECTED THEMES SOILS Quadrangle Name Pretty Rock Tanner Lake Alamo Mesa East Alamo Mesa West GEOLOGIC SURFICIAL PROCESS UNITS Pretty Rock Tanner Lake Alamo Mesa East SURFACE GEOLOGY Pretty Rock Tanner Lake Alamo Mesa East Alamo Mesa West SOURCE: Technicolor Government Services (TGS) Inc. 174 Actual Hours 9.50 13. 15 28.00 2.25 12.50 6.25 14.25 41 .33 27.67 16.50 7.67

PAGE 185

Conclusions This chapter evaluates the capabilities of MOSS/MAPS for performing CRM work and presents some of the benefits and problems encountered in this project in using MOSS/MAPS with ARMS for CRM work. This chapter also presents a general discussion of the accuracy, time required for data capture and analysis, and monetary .costs. Beyond its role in helping to organize, overlay, and display data, a GIS can potentially help agencies make the results of CRM survey and site locational efforts, and land use conflict analysis more comparable from project to project. Once data compilation, digitizing, and editing have been performed, the ease and potential time saved in using a GIS for CRM needs should outweigh the initial cost. The GIS can also facilitate the development of site location models. MOSS/MAPS can aid in the evaluation of the cultural resource data base by providing a means for simultaneous visual display and manipulation of inventoried areas, site data, and environmental variability. These three data sources are the basis from which site location models can be formed. Together they form a body of data that is amenable to generalization of data in an automated format. Field guidance concerning GIS hardware, software, and environmental and administrative data needs will be increasingly important in the future. As an aid to preparing guidance, developmental work should be completed to tailor GISs to the requirements of the BLM cultural resource program. 175

PAGE 186

CHAPTER IX THE USE OF A GIS IN THE BLM RESOURCE MANAGEMENT PLANNING PROCESS AND FOR SUPPORT PROGRAM NEEDS This Chapter examines the advantages of using a GIS, specifically MOSS/MAPS, as a tool in the BLM resource management planning process and gives examples using cultural resources where appropriate. The Farmington Resource Area (FRA), is preparing a resource management plan (RMP) for which portions of this chapter could be useful for cultural resource management. This chapter concludes with a discussion of a GIS's value for performing spatial analysis to meet BLM support program needs. Chapter I discusses planning and historic preservation. It emphasizes that, although the BLM planning process gives priority to designating and protecting areas of historical and cultural value, all resources must be identified and considered on the basis of their relative scarcity. Guided by NEPA, FLPMA, and other federal, state, and local laws, BLM balances development with protection of its natural and cultural resources to ensure their continued existence. BLM develops comprehensive land use plans called resource management plans RMPs to set guidelines for multiple-use decisionmaking. Chapter I discusses the authorizing legislation for RMPs as well as the role RMPs play in BLM. 176

PAGE 187

Using MOSS/MAPS in the Planning Process The following list shows the steps in BLM's planning process, and the discussion summarizes the major parts of the process as described by Zulick ( 1986). 1. Identify issues 2. Develop planning criteria 3. Inventory and data call ecti on 4. Analyze management situation 5. Form alternatives 6. Estimate effects of alternatives 7. Select preferred alternative 8. Publish draft RMP/EIS 9. Publish proposed RMP/Final EIS 1 0. Se 1 ect RMP 11. Implement, monitor, and evaluate actions Collecting Inventory Data and Information Data needs are identified in the planning criteria and are limited to those that address key planning issues. If information does not exist, inventories or studies must be conducted. Since most natural and cultural resources are geographically referenced, they are mappable. If stored in a relational data base management system, they may be in text format with coordinate references . The data layers not already in a geographic information system are mapped and then digitized from a preferred 1:24,000 (7.5 minute) scale maps where available. Once entered into MOSS, the digitized layers are then referred to as resource data or resource data themes. 177

PAGE 188

Planning requirements vary with the complexity of resource issues, but an average of 50 themes are needed for a comprehensive RMP. Often the overlaying of data themes provides more valuable information than the original data layers. Analyzing the Current Management Situation Before management or the public can analyze or suggest methods of managing an area's resources, the physical, biological, and cultural characteristics of the area must be examined, and trends over time identified. This analysis, documented in map and narrative formats, provides the baseline from which alternatives may be examined to document proposed management changes. The data themes entered into the system are assessed through a series of intersection analyses, showing the extent of resources, existing and potential conflicts, and resource uses; providing acreage tables; and producing map graphics that illustrate the analysis . Because this project has demonstrated some of the analytical capabilities of MOSS/MAPS for cultural resource management, it has focused on assessing inventory data for bias identification and the identification of potential cultural resource and land use conflict areas. The study did not try to formulate alternative methods of resource management for this pilot project. The methods used in analyzing the current management situation are straight forward and logical. The first step is to merge all adjacent 7.5 minute quadrangles into one data set or new map. The planning area boundary is then overlaid on the new map. An acreage summary shows the extent of each resource within the boundary. 178

PAGE 189

The next step is to intersect the land ownership theme (also merged) with each thematic layer. This step allows the resource specialist and managers to study the significance of resources on public lands with respect to all other ownership types. This can be a critical analytical step because certain resources may be located almost exclusively on private lands. If this is the case, the small amounts of these resources in public . ownership become important because there are so few means of protecting the resources in private ownership. Acreage tables and plotted maps are used to show these concepts and information. The use of the ownership data theme for assessing inventory biases and potential conflict areas with cultural resources is discussed in Chapters V and VI, respectively. Formulating Alternatives for Resource Management As in any other decisionmaking process, resource planning requires proposing an array of alternatives. These alternatives must be realistic, distinct, and implementable management alternatives. They must respond to the issues identified at the beginning of the planning process. If the analytical procedures presented in this study were developed further, they would show how to generate alternatives for protecting the cultural resources where potential land use conflicts have been identified (Chapter VI). Usually, at least four alternatives are proposed. Although the themes of the alternatives reflect management techniques ranging from resource protection to full-scale resource development, they are usually referred to thematically, such as cultural resources, forestry, or minerals. If all resources in the planning area are studied, one can 179

PAGE 190

determine that some resource uses are compatible with one another, some are compatible given stipulations, and some are incompatible, e.g, cultural sites and open pit mining. Other examples might include endangered plants/bald eagles (compatible), cattle grazing/oil and gas development (compatible), and scenic areas/surface coal mining ( i ncompa ti b 1 e). Each resource theme considered in the management scheme must be ranked in some order of preference. Each resource is compared to all other resources to determine the dominant or preferred use for given areas. If two resources exist in the same location, such as coal and cultural sites, the best use of the land would have to be determined. If the allocation involves compatible resource uses, the decision would be relatively easy. In developing alternatives, MOSS/MAPS can be used to sort areas by compatibility using logical (boolean) operators: intersection (A and B), exclusion (not B), intersection-complement (A and not B), and union (A or B). The theme of each alternative determines the ranking order of its component map themes. Components most compatible with the main themes of the alternative become the highest priority, whereas those least compatible become the lowest priority in use allocation. Graphics are used to show the highest priorities for any area by .. covering .. the lowest priority with the next higher priority and the next higher priority and the next, through the highest. The highest priority theme is shown on the graphic while the lower ones do not appear unless no higher priority covers them. The resultant maps are easy to interpret because intersecting areas, which would be complex and confusing, are not shown. 180

PAGE 191

Acreage summaries for each alternative show the relation of resource allocations from one alternative to the next. This method, however, does not consider dual uses when resource uses are compatible or compatible with limits, but reflects a kind of 11Ultimate priodty use .. concept. To generate analyses that reflect compatibility of uses, a series of exclusion and intersection type boolean operations must be performed on the map data. The resultant map illustrates the total land available for any use in any geographic area. Acreage calculations reflect total lands that can be used for each designated use. Comparing available acreages under differing alternatives allows RMP reviewers to easily target the areal extant of different management alternatives for any resource of interest, as shown in Table 9.1. These map products produce a focal point for interpretation of the RMP that can be invaluable for management use and public meetings. Environmental Consequences of Implementing Alternatives In this stage, potential physical, biological, cultural, and socioeconomic impacts from each management alternative are analyzed using standard environmental impact statement (EIS) procedures. The procedures involve the scientific comparison for each alternative of the relative effects of implementing differing management techniques and resource allocations. Recommendations are made for mitigating adverse impacts, and unavoidable and cumulative impacts are identified. Analysis relies largely on the ability of resource specialists and other professionals to interpret trends and predict the outcome of actions. In most cases, the specialist compares case histories of similar actions and makes judgments based on their outcome. A GIS cannot 181

PAGE 192

TABLE 9.1 ACREAGE TABLE OF TOTAL LAND AVAILABLE FOR USES UNDER VARIOUS RMP ALTERNATIVE MANAGEMENT SCHEMES MANAGEMENT PAIOR IT'f AREAS AL c--........-•-W"dttfe .,,.,.,..,. 0.1 end Gts A h etn•IIY! ():1 Shlll e AhHnAIIYf!' ,.. __ nl !M )I c 0 z 0 c: 3:

PAGE 193

replace the specialist's ability to measure impacts--such as the impact of a transmission power line on cultural site integrity--but it can help the specialist determine conflict areas as well as measure the potential disturbed acres. Chapter VI gives some examples using GIS to analyze cultural resource and mineral development conflicts. The area disturbed by potential coal and oil and gas development was calculated using acreage summaries derived from intersecting the archaeological site map with the development themes. The specialist, however, must consider other factors that may disturb adjacent cultural resource properties--new road development and the potential impact of vandals--and calculate the probable effect on the disturbed areas as well as the cumulative impact on cultural values proximate to the disturbed areas. Conflict areas in each alternative do not necessarily have to be pictured in graphic format, but they can be if a graphic format would be beneficial. Selecting the Preferred Alternative After potential impacts have been identified, a preferred alternative is selected. This alternative is intended to resolve planning issues and promote balanced multiple use principles with acceptable levels of impact. The preferred alternative can be a combination of aspects of several alternatives or a duplicate of one of the original alternatives. If new combinations of management techniques and allocations are proposed, the environmental consequences must be analyzed as described for the preceding stage. 183

PAGE 194

A GIS can be used to help manipulate alternatives by combining aspects of each. Alternatives may also be manipulated by extracting parts from alternatives and reforming new alternatives with a combination of cover, intersection, union, or complement operations. Unlike alternative formulation, this process may disregard compatibility. in the ranking order of the resources considered in the plan. In this case, simple mathematical data manipulation may be used to remove areas and replace them with different values. The process is fast enough to allow changes in the preferred alternative and make assessments of the effects within a reasonable timeframe. Furthermore, what is often perceived as a final product will probably be altered. A GIS handles changes well. Even 11final11 plots can be modified. Selecting and Publishing Draft and Final RMP This step involves preparing a draft RMP/EIS, including final graphics for all alternatives. The RMP/EIS is submitted to the Environmental Protection Agency (EPA) and undergoes a lengthy public review. After public comments are received, management evaluates the opinions, suggestions, and any new information within them. The draft may be reassessed and changed or may occasionally not change at all. Once a final plan is selected, it is published as a final RMP/EIS and filed with the EPA. If no official protests are made within 30 days, the final plan is approved. Protests may require revising the draft. Changes are handled as discussed in the section dealing with 11Selecting the Preferred Alternative11• GIS capabilities make these changes less troublesome than if manual methods were employed. Final graphics are polished. Black and white page-size plates can be made by 184

PAGE 195

producing plots on pen plotters and photographically reducing them. These plates can then be reproduced by offset printing at a low cost because they are camera ready. Full color plots can be generated by first plotting each color separately using a black pen. A photographic reversal serves as a color separation negative for a lithographic print. Map preparation costs can be reduced by as much as 80 percent because maps can then be economically reproduced in large numbers at a useful scale, usually l/2 inch to a mile or 1:100,000. A pen plotter can also plot on vellum paper or mylar, and these plots can then be reproduced in monotone on a blueprint machine at an even lower cost. GIS users contend that changes are less likely in RMPs and EISs that effectively use GISs because a GIS provides more thorough data analysis and more useful information. Advocates contend that a GIS allows impacts to be measured more effectively, produces a higher quality product that can be easily understood by the public, and reduces the likelihood of adverse comments or protests. Implementing, Monitoring, and Evaluating the RMP Once the RMP is approved, portions of the plan will be implemented by resource program managers. Implementation often requires completing site-specific plans that contain more detail than can be addressed in a comprehensive master plan. The plans may include grazing allotment management plans, recreation area management plans, off-road vehicle plans, mine plans, wild horse management plans, wildlife habitat management plans, forest management plans,or cultural resource site management plans. These plans provide the means to transform management decisions in the RMP into on-the-ground activities. 185

PAGE 196

RMP implementation is regularly monitored and evaluated to ensure that activities comply with the plan's provisions. Monitoring also determines the accuracy of judgments made by resource specialists in the EIS. Continued monitoring produces new and useful information that can be added to the data base, helping to make better land use decisions in the future. This stage of the planning process may last several years. The planning process is cyclic, and some stage is always active. The data prepared, entered, and processed during RMP preparation is reused during implementation and monitoring. Parts of the data and all of the computer hardware and technical expertise gained by the staff are used in every activity plan. Furthermore, resource management benefits from a GIS on a daily basis. A GIS can facilitate the production of site analyses, facility sitings, environmental assessments, surface reclamation plans, research projects, predictive models, conflict identification, suitability analyses, weighted composite analyses, acreage calculations, and graphic products. Most of the complex analytical procedures described for the RMP process can be performed in the MAPS subsystem. Matrix, math, and a wide variety of boolean operations--including union, intersection,. and exclusion--generally operate effectively in MAPS but are sometimes less effective in the vector structure of MOSS. This is because complex vector algorithms are prone to error. Batch processing is helpful on large projects because it allows updating input files, transferring analytic techniques to other offices, reducing telecommunication costs, performing analysis without an operator, and documenting analytical procedures. 186

PAGE 197

GIS and Support Program Needs Chapter I discusses the role of CRM support programs in assisting other resource programs to meet the requirements of historic preservation law, regulation, and agency policy. Federal archaeologists often provide support to the mineral staff when development plans are prepared and implemented. Archaeologists can be involved throughout the project, during successive stages of mine plan and stipulation development, inventory, evaluation and mitigation, and monitoring. Chapter I also discusses the main issues to be addressed in developing a cultural resource support program. The issues include the types of information needed and the strategies for tailoring a general support program to the needs of a project. The main parts of a general support program are inventory and evaluation. These two parts exist in the domains of planning and implementation, and in turn, should be part of separate administrative and compliance plans. Each of these parts should be addressed independently to determine information needs and strategies. Requirements of each should also assist decisions made within the others. A GIS can be useful in helping federal archaeologists conduct spatial analysis for project planning and implementation. The types of information used to make an informed choice among several options (and the strategies used to secure this information) may differ from the information needed and the strategies used to secure it when the choice has been made and the mitigation of project impacts has begun. Only GIS applications for project planning, however, are discussed here. Three types of information crucial to the planning phase of a project can be extracted from the inventory data: (1) potential cost and time 187

PAGE 198

expenditures for locating and protecting cultural resources, costs that may affect overall project feasibility; (2) potential for adverse effects (as defined in 36CFR 800) to significant cultural resources; and (3) potential long-term impacts on the resource base. Types (2) and (3) must be defined for each resource and assessed as specified in federal regulations and procedures. But ultimately they will be evaluated by the decisionmaker with respect to the BLM1s mission and policy. As shown in Chapter VI, a GIS can aid in identifying areas of potential adverse effects to significant cultural resources as well as areas that have the potential for long-term impacts on the resource base. By identifying potential biases in the inventory data (Chapter V), we can more easily plan to inventory areas that may have cultural resources but are severely undersurveyed. During project planning, evaluation activities should be directed toward weighing the information previously collected concerning the extent of needed mitigation the potential for adverse effects on properties, and the cumulative impact on the resource base. The cultural sensitivity of an area should be considered with respect to the potential impact to the resource base. Ultimately, evaluating an area•s sensitivity provides the basic information used to develop final recommendations to management concerning project feasibility. Using a GIS, we can identify potentially sensitive areas and delineate culturally sensitive zones by using environmental factors that correlate with cultural site locations. Thus, by analyzing different environmental factors together with the distribution of known cultural site data, we can identify site patterning on the landscape. 188

PAGE 199

We normally formulate measures to mitigate site-specific impacts through the inventory and mitigation standards carried out as part of project implementation. The entire planning phase is directed toward identifying nonmitigatable impacts, either site specific or regional. Although the loss of the resource base cannot be mitigated, identifying this impact should encourage agencies to begin addressing this problem directly and to establish ways to designate and permanently protect a representative array of sites. A GIS is an excellent tool for identifying both site specific and regional impact areas. In addition, areas that are likely to have a representative type of site can be depicted using a series of overlays that show the environmental conditions conducive for locating representative site types. Thus, the uses of a GIS for project planning lie mainly in identifying conflict areas where (1) significant cultural resources could be destroyed and (2) the potential for long-term impacts is great. Summary Using cultural resource examples, Chapter IX has examined the advantages of using a GIS, specifically MOSS/MAPS for aiding the BLM resource management planning process. The major phases of the RMP process where a GIS is useful include: (1) collecting inventory data and information (2) analyzing the current management situation (3) formulating alternative methods of resource management (4) examining the environmental effects of implementing alternatives 189

PAGE 200

(5) selecting the preferred alternative (6) selecting and publishing draft and final RMPs (7) implementing, monitoring, and evaluating the RMP. The remainder of this chapter focused on the use of a GIS as a tool for aiding other resource programs, such as coal development, in meeting the requirements of historic preservation law, regulations, and agency policy. Two of the main parts of a general support program are inventory and evaluation. These parts are subsumed by the domains of planning and implementation. This chapter has discussed possible applications for a GIS in project planning. 190

PAGE 201

CHAPTER X CONCLUSION The Need for GISs in Site Location Modeling Although this report only alludes to the concept of site location modeling, it is fitting to discuss the advantages of using a GIS to perform site location analysis. A GIS such as MOSS/MAPS can be valuable for site location modeling and testing in developing resource management plans, aiding in program support work, and conducting local and regional multiple-use planning. Environmental characteristics can be examined to determine if a relationship exists between the occurrence of specific environmental features and cultu:al site densities or types in inventoried areas. Once these relationships are established projections of similar site densities or types can be made for noninventoried areas. In this way, decisionmakers can be better informed about the cultural sensitivity of specific areas within the management area. Much of the following information is summarized from Kvamme•s discussion of GIS use for site location analysis (Kohleer and Kvamme 1985). Despite the growing literature on GISs and their increasing accessibility, they remain unfamiliar to most archaeologists and land use planners. The need for site locational techniques requiring GIS technology should be a concern of both land managers and archaeologists. Site location models can be both quantitative and nonquantitative (conceptual). However, GISs lend themselves to performing analyses for quantitative model building and testing. A variety of quantitative 191

PAGE 202

methods and models for classifying a location or region as site-likely (or site-type-likely) have been developed (see BLM Predicting Draft, DPP. No. 8000.001). At least during some of the modeling process, all of the procedures are based on measured data. Many of the procedures need numerous calculations. These quantitative approaches require measurements of environmental phenomena at each site location and at non-sites if the model development uses control groups. Non-site data is important because it enables one to contrast areas with sites to those without sites. This is useful for model building. For the simplest application of environmentally based models (without a GIS), projection variables, such as slope, aspect, and distance to water can be measured by hand at a specific location on a map. A simple site location model could then be applied to the measurements (usually requiring a few calculations) to assess the "site likelihood" or "site favorability" of the location. The approach can be useful to land use planners and managers in assessing archaeological sensitivity at, for example, the site of a proposed oil well pad. As the size of the area to be assessed increases, the number of well pads and access roads to the pads also increases. Thus, the labor-intensive measurement and calculation requirements rapidly become impractical. Many projects on federal lands involve large areas. In such cases, the logical approach would be to systematically replicate the above procedure across the area under consideration, performing the measurements and calculations every 50 m east-west and north-south, for example. The outcome would be a wide-area "site sensitivity surface11 depicting favorable or likely locations for cultural resource sites. Needless to say, performing measurements of multiple variables at some 192

PAGE 203

point on a map is tedious. Replicating this process every 50 m, even over a small area, is time-consuming and costly. In addition, once these data have been collected, one must consider the time and expense for all of the calculations required to apply most models. Manual techniques also pose several problems in model developing and testing. Perhaps most apparent is the effective limitation of sample sizes because of the labor requirements of measurement. For example, a region might have several hundred known sites, but it might be impossible to use all sites for model development or testing because of the difficulties of measurement. This problem is even more likely for non-sites if the control-group approach is used because potential sample sizes of many thousands of non-sites can be obtained. Perhaps more serious is the fact that the error in hand measurement introduces a large amount of variation into any analysis. Significant differences can be observed between measurements taken by different people or in measurements taken by the same person at different times. This factor can introduce major variations into the outcome of a model and can also affect how the model is applied. A less apparent effect is that measurement error violates one assumption common to many statistical procedures used for archaeological modeling, i.e., that the variables are measured without error (Poole and O'Farrell 1971). Measurement error not be confused with sampling error, which results when different areas are chosen at random. A major disadvantage of manual measurement has become apparent through implementing computer-based GIS technology in archaeological site location studies. Mainly because it is slow and time-consuming, manual measurement severely limits the kinds of phenomena that might be 193

PAGE 204

examined, or even conceived, in site location research. For example, for a given location on a map (such as a site location), or even for several locations, one might be able to estimate a least-effort travel distance (as opposed to a linear distance) or calculate a relative measure of view quality--the percentage of terrain visible within a given area. These values cannot be manually calculated for many hundreds or thousands of locations or, for example, for every 50 m across a map area. In fact, since many archaeologists think in a manual mode, such variables are rarely even considered. Thinking in a manual mode constrains all archaeological locational research. The nature of cultural resource management dictates that large amounts of information be collected. Data are gathered indicating where sites are found and where individual artifacts are located. Other information is collected describing regions surveyed, the intensity of the survey, where the region was surveyed, and who surveyed it. Data are collected on site content, the locations of features and artifacts within a site, cultural affiliation, site components, and the amount and kinds of work performed. Ecological data, such as environmental associations, might be recorded, as well as modern features, such as existing roads, trails, dwellings, and towns. Much, perhaps most, of the data are geographically distributed and have a mappable component. A major problem is that large bodies of regional data are often difficult to manage when retrieving particular information. Part of the data might exist on maps, whereas other information might exist on site forms, in project reports, in published articles, or even in museum collections. The usefulness of the vast amounts of data collected in such states is thus severely compromised. 194

PAGE 205

It is very expensive to initially automate data from different sources. For example, site data must be transferred to coding forms and key entered, whereas spatial data must be digitized. These are both very labor-intensive processes. For several years, archaeologists have been working with unmanageably large, geographically distributed computer data bases such as digital representations of remotely sensed images or digital terrain models. These information sources are often difficult to analyze, explore, manipulate, and use in drawing conclusions. Data sources might exist at different scales or on different map projections or might be geometrically distorted by the tilted angle of a remote sensing platform. One or more of these factors may make it difficult not only to register one source of data to another (lining up a particular point in space with the same point in all of the data sources over the entire region of study), but to locate even a single point in space in all data sources. Such problems are major limiting factors in the practical use of these data bases in regional archaeological investigations. Even automated data bases can be unwieldy if they are very large and computer resources are insufficient. GIS technology can help solve many of the problem areas and limitations discussed above. However, an important caveat must be made at this point. Although many of the technical difficulties encountered in site location modeling can be overcome with a GIS, an automated computer system may not always be cheaper than manual methods, only more accurate and efficient. For example, the high initial cost in building a data base, may not be justified if the project is of small scale. On the other hand, a medium-sized or large-scale project may stand to save money 195

PAGE 206

and time in the long term by automating the data bases and performing site location analyses on a GIS. A GIS can aid in overcoming many of the problems and limitations associated with manual methods for performing site location modeling: In seconds, computers can perform many thousands of measurements on all potential variables typically used in site location studies and permanently store those measurements for later use. This capability eliminates sample-size problems for known site locations and also permits us to obtain extremely large samples of background environment {or non-sites) for comparative analyses. Such complex calculations as probability estimates can be performed quickly and in large numbers. Variation in measurement is eliminated as the computer produces the same result every time. Depending on the ingenuity of the user, the existing software, and the software developer, there is a large potential for creating and exploring new types of information of relevance to archaeological research, cultural resource management, planning, and problem solving in site location studies. A GIS can provide a comprehensive tool for managing large, diverse, and unwieldly geographic data sets obtained from virtually any source, such as site files, aerial photographs, remotely sensed imagery, or conventional topographic maps. Despite their original disparity in projections and scales, all types of information are referenced to a common geographic coordinate base {such as longitude and latitude or Universal Transverse Mercator grid), providing a logical means for data storage, retrieval, manipulation, and interpretation. Only through GIS capabilities can we use much qf the data and many of the approaches to understand prehistoric site distributions. Using GISs, we can also develop strong support programs for project planning and implementation. 196

PAGE 207

GIS and the New Mexico Data Synthesis Project This study has attempted to contribute to the objectives of the New Mexico Data Synthesis Project described in Chapter I. By analyzing the Farmington Resource Area (FRA) inventory data and environmental information with MOSS/MAPS, we have been able to show where survey biases exist with respect to different environmental features. In addition , we have shown the percentage of areas inventoried and noninventoried for various administrative themes. We used existing data from the FRA to provide information on the distribution of cultural resources relative to planned development. This kind of information is useful in forming and guiding multiple-use resource allocation decisions in a timely and cost-effective manner. The methods shown here have focused on the use of site, survey, and environmental data for meeting Data Synthesis Project needs. We have not tried, however, to perform site location analysis. Rather, our aim was to show and describe the procedures involved in assessing the existing inventory and environmental data bases for potential biases and identifying potential conflict areas between cultural resources and particular land uses. By developing and refining such applications of MOSS/MAPS for contributing to the New Mexico Data Synthesis Project, more potential exists for meeting support program needs, such as for coal and oil and gas development. 197

PAGE 208

Anasazi Assemblage Basketmaker II Chipped Stone Component Fire-cracked cobble Ground stone Intrusive material Lithic assemblages Lithic reduction Lithic scatter Locus GLOSSARY Archaeology Terms One of the major prehistoric cultural traditions in ther American Southwest, which engaged in horticulture from about AD 500 to AD 1300. A set of objects found in association with each other and therefore assumed to belong to one temporal phase and one group of people, e.g., stone tools. A cultural period represented by the earliest stage of the Anasazi tradition in the American Southwest from about AD 100 to AD 400. Knives, scrapers, projectile points, and other stone artifacts produced by removing flakes. A period of occupation, i.e., Archaic, Basketmaker, or Pueblo, etc. at a site. A site may contain more than one component. Portions of cobble stones which have been broken by the extreme heat caused by fire. The cobbles were used in roasting pits or for building fire hearths. A stone that is finely pecked and ground to form a tool. The tool can be a mano or metate for processing plant material or a 5tone ax for cutting wood. Prehistoric objects that are not normally associated with a particular component, and are assumed to be introduced by cultural or natural processes, e.g., pottery from trade, displacement of projectile points due to erosional processes. A set of stone tools found in association with each other. The removal of flakes from stone material that is to be fashioned into a tool. A site characterized by a number of flakes and/or tools but having an ambiguous function. A specific location where specific human activities are evidenced by archaeological remains. 198

PAGE 209

Macroband . Microband Post-depositional processes Projectile point Blowout Playa A population composed of several foraging groups of hunters and gatherers . A small primary foraging group of hunter and gatherers. Natural processes such as alluviation or rodent activity which can affect cultural deposits on or in the soil. Normally a bifacially-flaked implement with a pointed distal end designed for penetrating an animal •s hide and a blunted proximal end designed for attachment to a shaft. The point can be a spear point, dart point, or arrowhead. Geology Terms A general term for various saucer-or trough-shaped depressions formed by wind erosion on a dune or other sand deposits. A dry, barren area in the lowest part of an undrained desert basin, underlain by clay silt, or sand, and commonly by soluble salts. It may be marked by an ephemeral lake. 199

PAGE 210

BIBLIOGRAPHY Acklen, J.C., M.E. Harlan, S.C. Hunt, and J.L. Moore. An Archaeological Survey of Approximately 4,000 Acres Located Near the Bisti Badlands South of Farmington, New Mexico. Public Service Company of New Mexico: Albuquerque, 1972. Baltz, E.H ... Stratigraphy and Regional Tectonic Implications of Part of Upper Cretaceous and Tertiary Rocks, East-Central San Juan Basin, New Mexico ... U.S. Geological Survey Professional Pap r 552. U.S. Government Printing Office, Washington, D.C., 1967/, Bauer, C.M. Bewley, R. (BLM Albuquerque District Geographer) Personal communication. 1986. Berger, S.P. and M. B. Lucas. Existing Environment In An Archaeological Survey of Approximately 4,000 Acres Located Near the Bisti Badlands South of Farmington, New Mexico. Public Service Company of New Mexico: Albuquerque, 1972. Binford, L.R. 11A Consideration of Archaeological Research Design ... American Antiquity. Vol. 29(4) (1964), 425-441 . .. Organization and Formation Processes: Looking at Curated Technologies... Journal of Anthropological Research. Vol. 35(3) ( 1979) ' 225-273. Bureau of Land Management. BLM 8100-Cultural Resource Management Manual. U.S. Department of Interior: 1978. BLM Predictive Modeling Draft. (DPP No. 8000.001 ). 1985. Burgener, J .A. 11The Stratigraphy and Sedimentation of Pictured Cliffs and Fruitland Formation. Upper Cretaceous of the San Juan Basin ... Unpublished Master•s Thesis. Illinois University. 1953. Cami 11 i, E. 11A Suggested Method for Recognizing Patterning in Lithic Artifact Distribution... In Archaeological Investigations in Cochiti Reservoir, Vol. 4., 339-354. J . Biella and R. Chapman, eds. Office of Contract Archaeology, University of New Mexico, Albuquerque: 1979. Chapman, R.C. 11Analysis of Lithic Assemblages ... In Settlement and Subsistence along the Lower Chaco River: The CGP Survey. C.A. Reher, ed., University of New Mexico Press, Albuquerque: 1977. 372-454. 200

PAGE 211

Chapman, R.C. 11The Archaic Period in the American Southwest: Factor Fantasy ... Unpublished Ph.D. Thesis. University of new Mexico Press, Albuquerque: 1980. Chapman, R.C. and T.V. Biella. An Archaeological Surgey on Four Sections of Land Near Black Lake, San Juan County, New Mexico. School of American Research, Santa Fe: 1980. Clark, J.D. 11The African Origins of Man the Toolmaker ... Origins, G. Issac and E. McGown, eds. Menlo Park: 1976. 1-54. In Human W.A. Benjamin, Cohen, N.M. The Food Crisis in Prehistory. Yale University Press: New Haven, 1977. Collins, S.H. and G.C. Moon. 11Algorithms for Dense Digital Terrain Models ... Photogrammetric Engineering and Remote Sensing, 44 ( 1981 ) 1481 -148 . Cooley, M.E., J.W. Hershbarger, J.P. Akers, and W.F. Hardt. 11Regional Hydrogeology of the Navajo and Hopi Indian Reservation, Arizona, New Mexico, and Utah.11 U.S. Geological Survey Professional Papers 521A. Washington, D.C. 1969. Daugherty, L.A. and B.A. Buchanan. Soils of the Sunbelt Mining Company, Alamo Mesa Project. Daugherty and Buchanan Associates, Las Cruces. 1981. Dingham, G.L. 11Surface Water. 11 In Western Area Survey. Public Service Company of New Mexico, Albuquerque, 1978. 149-162. Doyle, F.J. 11Digital Terrain Models:An Overview.11 Photorammetric Engineering and Remote Sensing, Vol. 44 (1978), 14 1-1485. Elyea, J.M. and E.K. Abbink, and P.N. Eschman. Cultural Resources of the N.I.I.P. Blocks IV and V Survey. Navajo Tribal Cultural Resource Management Program, Window Rock. 1979. Eschman, P.N. 11Archaic Site Typology and In Economy and Interaction Along The Lower Chaco River:The Navajo Mine Archaeo1oaica1 Program, Mining Area III. P. Rogan and J.C. Winter, e s. Office of Contract Archaeology and Maxwell Museum of Anthropology, University of New Mexico: Albuquerque, 1983, 375-384. Fassett, J.E. and J.S. Hinds. 11Geology and Fuel Resource of the Fruitland Formation and Kirtland Shale of the San Juan Basin, New Mexico and U.S. Geological Survey Professional Paper 676. Washington, D.C., 1971. 201

PAGE 212

Ferrill, J.T. "Climate". In Western Area Survey. Public Service Company of New Mexico, Albuquerque, 1978. 93-126. Ganas, M. "Local Variability in a Marginal Environment". The Archaeology of the Western Coal Comeany Bisti Lease contributors to Anthropology Series, No. 212. D1vision of Conservation Archaeology: Farmington. 1980. Gomolak, A.R. "Dating the Sites: The Development of a Knapped Stone Artifact Chronology." Block III Excavation Draft Manuscript. Navajo Cultural Resource Management Program, Windsor Rock. 1980. Gutierrez, A.A. "Dating the Site: The Development of Knapped Stone Artifact Chronology". N.I.I.P. Block III Excavation. Draft manuscript. Navajo Cultural Resource Management Program, Windsor Rock. 1980. Hack, J.T. "Dunes of the Navajo Country." Geological Review, Vol. 31 ( 1941). Hall, S.A. "Geology of Archaeologic Sites and Associated Sand Dunes, San Juan County, New Mexico." Manuscript in author's posession. 1979. Hasenstab, R.J. "The Application of Geographic Information Systems to the Analysis of Archaeological Site Distributions." Paper presented at the 48th Annual Meeting of the Society for American Archaeology, Pittsburgh. 1983a. A Preliminary Cultural Resource Sensitivity Analysis for the Proposed Flood Control Fac1l1t1es Construction in the Passaic River Basin of New Jersey. Submitted to Soil System, Inc., and the Passaic River Basin Special Studies Branch, Department of the Army. Copies available from Soil Systems, Inc., Marietta, Georgi a. 1983b. Huse, H., B.A. Noisat, and J.A. Halasi. The Bisti-Star Lake Project: A Sample Survey of Cultural Resources in Northwestern New Mexico. Bureau. of Land Management, Albuquerque D1str1ct, Number 1. 1978. Irwin-Williams, C. The Oshara Tradition: of Anasazi Culture. Eastern New Mexico University Contribut1ons in Archaeology, Vol. 5(1). Portales. 1973. Issac, G. "Olorgesailie. Archaeological Studies of a Middle Pleistocene Lake Basin in Kenya." Prehistoric Archaeology and Ecology Series. Chicago: University of Chicago Press. 1977. James, S.E., R. Knudson, A. Kane, and D. Breternitz. "Predicting Site Significance: A Management Application of High-Resolution Modeling." Paper presented at the 48th Annual Meeting of the Society for American Archaeology, Pittsburgh. 1983. 202

PAGE 213

Kemrer, M.F. Archaeological Variability Within the Bisti-Star Lake Region, Northwestern New Mexico. Bureau of Land Management, Albuquerque D1str1ct, Number 3. 1982. Klein, R.G. 11The Ecology of Early Man in Southern Africa.11 Science. Vol. 197 (1977) 115-126. Kohler, T.A. and K.L. Kvamme. Geographic Information Systems: Technical Aids for Data Collection, Analysis, and Display (Chapter 10). In BLM Predictive Modeling Draft (DPP No. 8000.001). 1985. Krober, A.L. Handbook of the Indians of California, Smithsonian Institution, Bureau of American Ethnology Bulletin 78. Washington, D.C. 1925. Kvamme, K.L. 11Computer Processing Techniques for Regional Modeling or Archaeological Site Locations.11 Advances in Computer Archaeology, Vol. 1 (1983a) 26-52. A Manual for Predictive Site Location Models: Examples from the Grand Junct1on D1str1ct, Colorado. Draft report subm1tted to the Bureau of Land Management, Grand Junction District, Colorado. l983b. 11Models of Prehistoric Site Location Near Pinyon Canyon, Colorado.11 In Papers of the Philmont Conference on the Archaeology of Northwestern Ne\'1 Mex1co; New Mex1co Archaeolo ical ounc1 roceed1ngs. Vo . 16, No. 1. 1984 349-370. 11Usi ng Existing Data for Model Bui 1 ding (Chapter 7). In BLM Predictive Modeling Draft (DPP No. 8000.001 ). 1985. Lee, J.E., C. N. Reed, R.L. Frash, and L. Sullivan. M oss Users M anual Version .8509 Western Energy and Land Use Team, Division of Biological Services. U. S. Fish and Wildlife Service, Washington, D.C. 1985. Lee, R.B. and I. DeVore. 11Problems in the Study of Hunters and Gatherers.11 In the Hunter. R. B. Lee and I. DeVore, eds. Aldine (1968) 3-12. Love, D.W. and J. Schultz. 11Surficial Geology an Geomorphology Related to Mitigbation of Archaeological Sites on the UII Lease.11 In Human Adaptations in a Marginal Environment: The UII Mitigation ProJect. J.L. Moore Off1ce of Contract Archaeology, Unv1ers1ty of New Mexico: Albuquerque, 1980. 29-38. Marshall, L.G. and W.J. Breed. Phase I Report of Paleontological Investigations on Navaho Tribal Lands leased by WESCO near Burnham, San Juan County, New Mexico. Ms. on file, Museum of Northern Arizona. 1974. 203

PAGE 214

McKinley, R.A. 11Map Overlay and Statistical System (Moss) Terrain Data Handling Capability.11 In Proceedings for the Third-National MOSS Users Workshop. In Press. Metric Corperation. 1982 Vegetation Survey. Arch-Mineral-Sunbelt Joint Venture Area San Juan County, New Mexico. Draft Manuscript. Metric Corp., Albuquerque. Monmonier, M.S. Computer Assisted Cartography: Principles and Prospects. Prentice-Hall, Englewood Cliffs, New Jersey. 1982. Moore, J .L. 11Archaic Settlement and Subsistence.11 In Human Adaptations in a Marginal Environment: The UII Mitigation Project. J. Moore and J. Winter, eds. Office of Contract Archaeology, University of New Mexico: Albaquerque, 1980. 358-381. Moore, R.A. 11Lithic Analyses.11 In Excavations in Blocks VI and VII, N.I.I.P., San Juan County ew Mexico. Vo1.3. T. Del Bene and D. Ford, eds. Navajo Nation Papers in Anthropology, Number 13. Navajo Nation C. Cultural Resource Management Program. Window Rock, 1982. 859-920. Morain, S.A., T.K., Bridge, and M.E. White. 11Vegetation and Land Use in New Mexico.11 New Mexico Bureau of Mines and Mineral Resources. Resource Map 8. Socorro, 1977. O'Sullivan, R.B. and Others. 11Stratigraphy of the Cretaceous Rocks and the Tertiary Ojo Alamo Sandstone Navajo and Hopi Indian Reservations, Arizona, New Mexico, and Utah.11 U.S. Geological Survey Professional Paper 521-E. U.S. Government Printing Office, Washington, D.C. 1972. Plog S., F. Plog, and W. Wait. 11Decision Making in Modern Surveys.11 Advances in Archaeological Method and Theory. Vol. 1. M.B. Schiffer, ed. Academic Press: New York, 1978. 383-421. Poole, M.A. and P.N. O'Farrell. 11The Assumptions of the Linear Regression Model .11 Transaction of the Institute of British Geographers. Vo 1 . 52 ( 1971 } 145-158. Powers, M.A. An Inventory and Analysis of Archaeological Resources on 38 Sections of Land Near the B1st1 Badlands, Northwestern New Mexico. Contributions to Anthropology Series, No. 119. Division of Conservation Archaeology: Farmington. 1979. Powers, S. Tribes of California. Contributions to North American Ethnology, Vol. 3. Department of the Interior, U.S. Geographical and Geological Survey of the Rocky Mountain Region. Washington, D.C. 1877. 204

PAGE 215

Rangeland Reso!Jrces International, Inc. 11Soils and Range Inventory ... Pater No. 9, San Juan Basin Regional Uranium Study. Bureau of nd1an Affairs, Albuquerque. 1978. Reed, C. (editor). Agricultural Origins. The Hague: Mouton. 1977. Reher, C.A. and D.C. Witter. Archaic Settlement and vegetation Diversity. In Settlement and Subsistence Along the Lower Chaco River: The CGP Survey. C.A. Reher, ed. University of New Mexico Press, Albuquerque, 1977. 113-126. Rehev, C.A. (editor) Settlement and Subsistence Along the Lower Chaco River: The CGP Survey. University of New Mexico Press, Albuquerque, 1977. Rhird, D. 11A Skeletal Overview at Spatial Interpolation Techniques ... • Computer Applications. Vol. 2 (1975) 293-309. Rogers, A. and J.A. Dawson. 11Which Digitizer? .. Area. Vol. 11 (1979) 67-73. Roney J. (BLM Albuquerque District Archaeologist) Personal communication. 1986. Schiffer, M.B. and G.J. Gumerman. Acquisition of Survey Data. In Conservation Archaeology. M.B. Schiffer. and G.J. Gumerman eds. Academic Press: New york, 1977. 183-190. Schiffer, M.B. and S.J. Wells ... Archaeological Surveys: Past and Future ... In Hohokam and Patayan Prehistory of Southwestern Arizona. R.H. McGuire and M.B. Schiffer eds. Academic Press, New York, 1982. 345-384. Schultz, J.D. ..Geomorphology, Sedimentology, and Quaternary History of the Aeolian Deposits, West-Central San Juan Basin, Northwest New Mexico ... Masters Thesis, Univeristy of New Mexico, Albuquerque. 1980. Sessions, S.E. (editor) The Archaeology of Southwest Gallegos Mesa: The EPCC Survey Project. Cultural Resource Management Program, Navajo Nation Anthropology, No. 1. Window Rock. 1979. Shelton, R.L. and J.E. Estes. 11Remote Sensing and Geographic Information Systems: An Unrealized Potential ... Geo-Processing. Vol. 1 (1981) 395-420. Soil Conservation Service. Soil Survey of San Juan County New Mexico, Eastern Part. U.S. Department of Agriculture in cooperation with the U.S. Department of Interior (Bureau of Indian Affairs and Bureau of Reclamation) and the New Mexico Agricultural Experiment Station. 1980. 205

PAGE 216

Steward, J.H. Basin-Plateau Aboriginal Socio-eolicitical Groups. Smithsonian Inst1tution, Bureau of Amer1can Ethnology Bulletin 120. Washington, D.C. 1938. Stuart, D.E., and R.P., Gauthier. Prehistoric New Mexico: Background for Survey. New Mexico Historic Preservation Bureau: Santa Fe. 1981. Toll, M.S. and A.C. Cully. 11Archaic Subsistence in the Four Corners area: Evidence for a Hypothetical Seasonal Round.11 In Economy and Interaction Along the Lower Chaco River: The Navajo Mine Archaeoloaical Program, Mining Area III. P. Hogan and J.c. Winter, e s. Office of Contract Archaeology and Maxwell Museum of Anthropology, . University of New Mexico: Albuquerque, 1983. 385-391. Tomlinson, R.F., H.W. Calkins, and D.F. Marble. Computer Handling of Geographical Data: An Examination of Selected Information Systems. Natural Resources Research!! . UNESCO, Paris. 1976. Turner, J. (BLM Service Center Geologist) Personal communication. 1986. Vierra, B.J. A Preliminary Ethnographic Model of the Southwestern Archaic Settlement System. In Human Adaptations in a Marginal Environment: The UII Mitigation Project. J. Moore and J. Winter eds. Office of Contract Archaeology, University of New Mexico, Albuquerque, 1980a. 351-357. A Summary Comparison of the Excavated Archaic and Anasazi Sites. In Human Adaptations in a Marginal Environment: the UII Mitigation Project. J. Moore and J. Winter, eds. Office of Contract Archaeology, University of New Mexico, Albuquerque, 1980b. 382-389. A Study in Site Structure at LA 18745. Unpublished manuscript in posession of author. 1981. Vierra, B., P. Prince, and M. Powers. Archaeolofical Inventory and Evaluation of Cultural Resources near Coa Creek and De-na-zin Wash, San Juan County New Mexico. (Draft report) Division of Conservation Archaeology, San Juan County Museum Association. Farmington, NM. 1986. Vogler, L.E., D. Gilpin and J.K. Anderson. 11Gallegos Mesa Settlement and Subsistence: A Set of Explanatory Models for Cultural Resources on Blocks VIII, IX, X, and XI, Navaho Indian Irrigation Project (Vol. 1).11 Navaho Nation Papers in Anthropology 12. Window Rock, Arizona. 1982. 206

PAGE 217

Wandsnider, L., and J.I. Ebert. "Accuracy in Archaeological Surface Survey in the Seedskadee Area, Southwestern Wyoming." Haliksa•i: Contributions to Anthropology. Vol. 3 (1984) 9-21. Warren, A.H. "The Land in the Chuska." In An Archaeological Survey of the Chuska Valley and the Chaco Plateau, New Mexico. Part I. Museum of New Mexico Research Records No. 4. Santa Fe, l967a. l-13. "Petrographic Analysis of Pottery and Lithics." In An Archaeological Survey of the Chuska Valley and the Cnaco Plateau, New Mexico, Part I. Museum of New Mexico Research Records No. 4. Santa Fe, l967b. 104-134. Wells, S.G. Geomorphology and Surface Hydrology in the Strippable Coal Belts of Northwestern New Mexico. New Mexico Energy Research and Development Institute, Santa Fe. 1982. Wells, S.G., D.W. Love and T.W. Gurder (editors). Chaco Canyon County. American Geomorphological Field Group Field Trip Guidebook, 1983 Conference Northwestern New Mexico. 1983. Winters, H.D. The Riverton Culture. Reports of Investigation No. 13. Springfield, Illinois State Mueseum. 1964. Woodward-Clyde Consultants. Appendix E. In Well Field Development Plan for the Wasson Field-Denver Unit C02 Project Environmental Impact Report. Draft report submitted to Shell Oil Corporation. Copies available from Woodward-Clyde Consultants, San Francisco. 1978. Zulick, C.A. Applications of a Geographic Information System to the Bureau of Land Managements Resource Management Planning Process. In Geographic Information Systems in Government Vol. l. B.K. Opitz ed., A. Decpak, Publishing. Hampton, VA: 1986. 309-327. 207