Citation
Administrative and technical feasibility of alluvial aquifer storage and recovery on the South Platte River / |c by Cibi Vishnu Chinnasamy

Material Information

Title:
Administrative and technical feasibility of alluvial aquifer storage and recovery on the South Platte River / |c by Cibi Vishnu Chinnasamy
Creator:
Chinnasamy, Cibi Vishnu ( author )
Publication Date:
Language:
English
Physical Description:
1 electronic file (55 pages) + : ;

Subjects

Subjects / Keywords:
Aquifer storage recovery ( lcsh )
South Platte River (Colo. and Neb.) ( lcsh )
Genre:
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )

Notes

Review:
Increasing population growth coupled with climate variability is causing water managers to reassess water storage. In Colorado, the era of large above-ground dams and reservoirs is probably over, due to environmental and endangered species concerns. In this context, an alluvial aquifer storage and recovery (ASR) system presents an alternative option for water storage. A recent study estimated approximately 12 km3 (10 million acre-feet) of storage may be possible in the South Platte River alluvium of northeast Colorado. To investigate this option, a case study was conducted to examine the availability of free river conditions on the South Platte River, and simulations of soil matrix stability, aquifer recharge and extraction rate under various levels of clogging, to test the feasibility of an alluvial ASR system for water storage purpose. An alluvial ASR site near U.S. Highway 7 at Brighton, Colorado with a storage capacity of 118,500 m3 (96 ac-ft) was considered for this study. Analyzing river flow data at this site confirms the availability of excess water, during the wet season, to fill the proposed alluvial ASR facility in compliance with Colorado's water laws. This suggests that alluvial ASR facilities could be a viable option to meet rising water demands of Colorado, prevent water loss due to evaporation, reduce the effect of climate stress on water resources, and avoid the need to purchase lands for above-ground water storage facilities.
Thesis:
Thesis (M.S.)--University of Colorado Denver, 2017.
Bibliography:
Includes bibliographical references (leaves 85-86).
System Details:
System requirements: Adobe Reader.

Record Information

Source Institution:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
on10047 ( NOTIS )
1004775798 ( OCLC )
on1004775798

Downloads

This item has the following downloads:


Full Text
ADMINISTRATIVE AND TECHNICAL FEASIBILITY OF ALLUVIAL AQUIFER
STORAGE AND RECOVERY ON THE SOUTH PLATTE RIVER
by
CIBI VISHNU CHINNASAMY B.E., Anna University, 2014
A thesis submitted to the Faculty of The Graduate School of the University of Colorado in partial fulfillment of the requirements for the degree of Master of Science Civil Engineering Program
2017


2017
CIBI VISHNU CHINNASAMY ALL RIGHTS RESERVED
11


This thesis for the Master of Science degree by Cibi Vishnu Chinnasamy has been approved for the Civil Engineering Program by
Caroline Clevenger, Chair David C. Mays Dharmarajan Ramaswami
Date: May 13, 2017
m


Chinnasamy, Cibi Vishnu (M.S., Civil Engineering)
Free River Analysis in South Platte River for Groundwater Storage Using Alluvial
Aquifer Storage and Recovery System
Thesis directed by Associate Professor David C. Mays.
ABSTRACT
Increasing population growth coupled with climate variability is causing water managers to reassess water storage. In Colorado, the era of large above-ground dams and reservoirs is probably over, due to environmental and endangered species concerns. In this context, an alluvial aquifer storage and recovery (ASR) system presents an alternative option for water storage. A recent study estimated approximately 12 km3 (10 million acre-feet) of storage may be possible in the South Platte River alluvium of northeast Colorado. To investigate this option, a case study was conducted to examine the availability of free river conditions on the South Platte River, and simulations of soil matrix stability, aquifer recharge and extraction rate under various levels of clogging, to test the feasibility of an alluvial ASR system for water storage purpose. An alluvial ASR site near U.S. Highway 7 at Brighton, Colorado with a storage capacity of 118,500 m3 (96 ac-ft) was considered for this study. Analyzing river flow data at this site confirms the availability of excess water, during the wet season, to fill the proposed alluvial ASR facility in compliance with Colorado's water laws. This suggests that alluvial ASR facilities could be a viable option to meet rising water demands of Colorado, prevent water loss due to evaporation, reduce the effect of climate stress on water resources, and avoid the need to purchase lands for above-ground water storage facilities.
The form and content of this abstract are approved. I recommend its publication.
Approved: David C. Mays
IV


ACKNOWLEDGEMENTS
I would like to thank Mr. Brent Schantz and Mr. Colin Watson at the State Engineers Office at Greeley, Colorado for hosting a field trip to show how water decrees and administrative calls affect the river flow conditions, and for providing data on reusable effluent water. Special thanks to Mr. Joe Frank, Chair of the South Platte Basin Roundtable Chair, for his valuable insights on free river analysis. Most importantly this work gained momentum under the guidance of Dr. William McIntyre, whose patent work and graduate thesis were the cornerstone for this thesis report. I would also like to acknowledge the University of Colorado Denver for providing valuable academic support.
v


TABLE OF CONTENTS
CHAPTER
I. INTRODUCTION....................................................1
II. ALLUVIAL AQUIFER STORAGE AND RECOVERY..........................6
III. CASE STUDY ON SOUTH PLATTE RIVER ALLUVIAL ASR.................9
Background information.......................................9
Facility design.............................................12
Technical feasibility.......................................16
Administrative feasibility..................................22
IV. DISCUSSION AND CONCLUSION.....................................33
REFERENCES...........................................................34
APPENDIX
A. Algorithm for recharge alluvial ASR...........................36
B. Soil matrix consolidation.....................................37
C. RStudio codes for free river analysis.........................39
vi


LIST OF TABLES
TABLE
1 Available storage capacity in South Platte River Basin alluvium........10
2 Yearly average of unused reusable effluent flow from Denver and Aurora.27
vii


LIST OF FIGURES
FIGURE
1 Map of South Platte River from Denver Metro to Liddle Ditch near Colorado-
Nebraska state line (CDM Smith 2013)......................................3
2 Selected alluvial site at Brighton, CO. 3959'07.25" N 10449'49.96" W (Google
Earth)....................................................................11
3 Gravity powdered recharging of alluvial ASR (McIntyre et al. 2016)........14
4 Confinement using Slurry walls/ sheet piling to setup a static alluvial ASR system ...........................................................................14
5 3D model of modeled alluvial ASR system showing wells at different locations in
layer 2........................................................................15
6 Perforated pipe assembly in layer 2, 9.5 m below surface level..................15
7 Time taken to saturate the alluvial ASR under varying levels of clogging........19
8 Time taken to pump out all usable water from the alluvial ASR under varying
levels of clogging.............................................................20
9 Total soil consolidation at the case study site.................................21
10 Daily summary of free river available from Water Year 2000 to 2013 .............29
11 Annual free river available from water year 2000 to 2013 .......................30
12 Monthly mean, minimum and maximum available free river from water year 2000
to 2013 .......................................................................31
13 Time series correlation analysis of Free River available from WY 2000 to WY
2013 ..........................................................................32
vm


LIST OF ABBREVIATIONS
ASR Aquifer Storage and Recovery
CWCB Colorado Water Conservation Board
SPBIP South Platte Basin Implementation Plan
SPDSS South Platte Decision Support System
SWSI Statewide Water Supply Initiative
WY Water Year (October 1st to September 30th)
IX


CHAPTERI
INTRODUCTION
Water, one of the five basic elements of nature (Ma et al. 2014), is responsible for sustaining life on Earth. From the dawn of civilization, a communitys growth and progress was determined by their ability to distribute, manage and sustainably use water resources. This practice of managing water bodies and structures has advanced from dawn of civilization to the present 21st century, where water drives our economy and commerce. Water has even turned scorching deserts into vibrant cities, like Dubai, with urban green spaces and urban forestry. With increasing global human population resulting in higher consumption of basic resources, the task to supply clean water is becoming more challenging, and water managers around the world must be prudent and innovative to meet the ever-growing water demand.
Colorado is blessed to be the home to seven major head waters in the USA. Rio Grande, North Platte, South Platte, Arkansas, Cache la Poudre, and the Republican Rivers flow East from the Continental Divide, and the Colorado River flows toward the Pacific Ocean. These head waters, which originate in the Rockies, have converted Colorado from a semi-arid to a bread-basket state in the USA. Colorado receives approximately on an average 41 cm (16 in) of annual rainfall combined with high evaporative losses (statewide average of approximately 81%) results in a water balance deficit over most of the state, with the exception of the higher mountainous regions (Topper et al. 2003). Based on State Demographers Office population projections, the South Platte and Metro Basins are projected to grow from approximately 3.5 million people in the year 2008 to about 6 million people by the year 2050 (SWSI2010). Population growth will significantly increase the basins future municipal and industrial
1


water needs. The water available to Colorado within its state boundaries is finite, and with climate variability, it is anticipated the amount of aggregate annual rainfall will decrease (Kennedy 2014). Therefore, new water storage methods must be developed that will help solve the storage issue, while also minimizing losses due to evaporation.
A recent study conducted by the Colorado Water Conservation Board (CWCB) estimated an abundance of alluvial aquifer area, useful for storage and recovery, of 6500 km2 (2500 mi2) along the South Platte River from Denver up to the Colorado-Nebraska Stateline as shown in Figure 1 (CDM Smith 2013). The science of using groundwater for domestic and recreational use is not new. Persians, in the early part of first millennium B.C., built elaborate tunnels systems called qanats to extract groundwater from dry mountain basins. These underground aquifers act like natural sponges to store infiltrated water from surface runoffs and precipitation, and can be pumped out for use anytime. Storing water in underground aquifers is advantageous compared to storing water in conventional dams and reservoirs that have a minimum of 10 % evaporation loss (Finley 2015). Hence, storing water underground into alluvial aquifers will potentially benefit assist water managers in maximizing the beneficial use of the states water resources, in the most efficient method practicable. Accordingly, this study focuses on alluvial aquifer storage and recovery (ASR), with attention to physical and administrative considerations.
2


Legend
03 Active Alluvial Model Extent
Approximately 2,507 square miles C3 Denver Basin Bedrock Aquifer Extent Contributes to bedrock and alluvial groundwater flux
[~* | County City
Stream
Figure 1: Map of South Platte River from Denver Metro to Liddle Ditch near Colorado-Nebraska state line (SPDSS Alluvial Groundwater Model Report CDM Smith 2013).
3


In terms of physical considerations, the alluvium must be able to store useful quantities of water for subsequent use, the system must be located adjacent to recharge water and near water consumers, and adequate hydraulic conductivity must be maintained. Fortunately, a study has been performed to evaluate the alluvium and an estimated 12 km3 (10 million ac-ft) of storable void space physically exists in the South Platte basin (HDR Engineering and West Sage Water Consultants 2015). Comparing it with existing 9.5 km3 (7.7 million ac-ft) of storage available in dams and reservoirs across Colorado from Division of Water Resources, alluvial ASRs could potentially reduce the need to use up vast open spaces to construct new dams.
In terms of legal considerations, alluvial ASR depends on availability of water timing, quantity, and qualitywithin Colorados doctrine of prior appropriation. Like most states in the Western U.S., Colorado water law is based upon the recognition that the right to use decreed water, either surface water or groundwater, is property of the owner and therefore protected by law from injury (McIntyre and Mays 2017). Injury occurs when a decreed water right cannot be diverted in accordance with its decree, due to the actions of another. Other water right holders have an opportunity to oppose a water rights application, if they feel injury may occur.
This paper deals with alluvial aquifer storage and recovery and how it was applied to a case study along the South Platte River, about 40 km (25 mi) northeast of Denver, the capital of Colorado, near the city of Brighton with a storage capacity of 148,000 m3 (120 ac-ft). The case study involved site selection, review of available soils data, modeling of water recharge rate and extraction rates under varying levels of soil matrix clogging, and determining the optimal configuration of storage inflow and outflow
4


piping. In addition, and critical to the operation of such a storage vessel, a detailed study of legally available surface water was performed. This was necessary since a junior water storage right, which this storage facility would be granted in the water court, would not be a dependable storage structure.
Chapter 2 looks at alternative water storage facilities, Chapter 3 will discuss about the technical and administrative feasibility of such system, and Chapter 4 will summarize the study and provide recommendations for further research.
5


CHAPTER II
ALLUVIAL AQUIFER STORAGE AND RECOVERY
Owing to various problems with above the ground surface-dams and reservoirs, subsurface dams are now widely used for water storage (Ishida et al. 2011). Topper et al. (2010) discuss artificially recharge methods to introduce water to, store water in, underlying aquifer system during wet season, when the river carries excess water. The stored water can be later pumped out for use, either drinking or irrigation, during dry season. Primary objectives of artificial recharge are; manage water supply, meet legal obligations (augment water to meet downstream water rights or facilitate compliance with interstate agreements), manage/ mitigate water quality, restore/ protect aquifers (restore groundwater levels, limit aquifer compaction, mitigate saltwater intrusion), and protect the environment (maintain wetland hydrology, control migration of groundwater contamination).
Artificial recharge of aquifers is achieved by natural/ enhanced recharge through surface flooding on infiltration basins and well injection. Alluvial aquifer storage and recovery is one way of artificially storing water, by the use of injection/pumping well systems to inject water directly into the receiving aquifer, for future recovery at the same location (Topper et al. 2010). Hanson and Nilsson (1986) classified underground dams into two types: sand storage dams, where the sand is behind the above-ground dam, and those that are constructed below ground to stop the flow of a natural aquifer. Alluvial ASR has been a central component of water storage internationally. Australia has a concentrated effort in maximizing storage of water underground due to reoccurring droughts (Simmons 2014). The literature provides evidence for a wide application in
6


Australia and North America. Australia is one of the pioneers to integrate alluvial ASR into their urban stormwater harvest program. Peak surface runoffs are diverted into wetlands and settling ponds for primary treatment and finally infiltrate into underground aquifers. In the rural areas, alluvial ASRs are used to combat drought scenarios. Florida, Oregon and Texas are some of the states in the U.S.A., which have adopted aquifer storage and recovery systems to tackle changing weather conditions and increasing water demand because of population growth. Brown et al. (2006) synthesized lessons learned from 50 ASR projects worldwide. Their key findings were as follows: First, well clogging, including air binding, is problematic. However, a potential solution is to incorporate regular back flushing programs. Second, water quality can diminish the usefulness of ASR, particularly when arsenic, iron, manganese or other metals can be released from the local geologic materials. Third, hydraulic analysis is important when evaluating multi-well clusters in order to avoid interference with other wells. And fourth, alluvial ASR systems need to incorporate monitoring equipment, such as sampling ports on recharge or discharge lines, to allow real-time monitoring of specific conductivity or turbidity.
To recapitulate, alluvial ASR is not a new concept, in either the United States or elsewhere. Depending on site location, soil conditions, and project design, there are various levels of operational success. Each potential project site requires a detailed analysis to include influent water quality, soil chemistry and physical properties, and intended use of the recovered water. In the era of climate change and increasing population, alluvial ASR offers a viable alternative to surface reservoirs. Alluvial ASR costs are less than above ground water storage (dams and reservoirs) and water loss due
7


to the evaporation is significantly decreased. Finally, in areas with limited resources to fund construction of traditional above ground water storage facilities or areas with limited technical expertise, alluvial ASR provides an alternative water storage vessel. With climate change anticipating shifting hot seasons and precipitation patterns in the western United States, Colorado is well suited, due to physical stream alluvial properties, its water law and administration, increased population predictions, to aggressively implement alluvial ASR as a water storage strategy.
This following chapter presents a case study for a theoretical ASR facility located adjacent to the South Platte River in northeastern Colorado, paying due attention to the relevant physical constraints (i.e., well clogging, water quality, hydraulic analysis, and monitoring) and administrative constraints (i.e., Colorados doctrine of prior appropriation).
8


CHAPTER III
CASE STUDY ON SOUTH PLATTE RIVER ALLUVIAL ASR
A case study was developed to evaluate the application of alluvial ASR in Colorado. Colorado has water right to only one-third of the total water available within its territories, while remaining two-third of the water sent downstream to neighboring states as part of various river compacts as dictated by Colorado Water Law. This has put a lot of stress on water administrators and commissioners to efficiently collect, distribute and manage this precious and life-sustaining element. Topics presented in this section include background information, facility design, technical feasibility, and administrative feasibility. Each of these topics will be presented in turn.
Background Information
A study by the Colorado Water Conservation Board (2007) assessed the volume of potential alluvial storage in the South Platte River basin (Table 1). The criteria for site selection were based on available alluvium and the distance between the site and the Denver metropolitan area (Denver Metro). Therefore, the area selected for detailed study was along the main stem of the South Platte River between the northeast portion of Denver Metro and Greeley, Colorado (SPMetro to Greeley). Along this reach, a candidate site was identified on the east bank of the South Platte River immediately upstream of the bridge crossing Colorado Highway 7 in Brighton, Colorado (Figure 2). As is typical for Colorados high plains, this site has high summer temperatures and high wind speeds, which would predict high evaporative rates for above ground water storage reservoirs. Brighton site takes advantage of evaporative loss savings, in addition to the relative proximity to Denver Metro.
9


Table 1: Available storage capacity in South Platte River Basin alluvium (CWCB, 2007)
Subregion Number Subregion Name Available Storage Capacity cubic meter (acre-foot)
1 SP- Denver Metro 435,000,000 (353,000)
2 SP Metro to Greeley 208,000,000 (169,000)
3 Cache la Poudre River 359,000,000 (291,000)
4 Upper Beebe/Box Elder 331,000,000 (268,000)
5 Lower Beebe/Box Elder 75,200,000 (61,000)
6 SP Greeley to Ft. Morgan 116,000,000 (94,000)
7 Upper Lost Creek 1,550,000,000 (1,260,000)
8 Lower Lost Creek 194,000,000 (157,000)
9 Upper Kiowa Creek 289,000,000 (234,000)
10 Lower Kiowa Creek 994,000,000 (806,000)
11 Upper Bijou Creek 575,000,000 (466,000)
12 Lower Bijou Creek 1,320,000,000 (1,067,000)
13 Badger/Beaver Ck. 384,000,000 (311,000)
14 SP Ft. Morgan Area 1,190,000,000 (968,000)
15 SP Balzak to State Line 1,100,000,000 (890,000)
16 SP South Park 1,110,000,000 (899,000)
Note: SP denotes areas along the main stem of South Platte River SP South Park data from Topper et al, 2004.
10


Figure 2: Selected alluvial site at Brighton, CO. 3959'07.25" N 10449'49.96" W (Google Earth)
11


Facility Design
Design of this alluvial ASR is based on a patent by McIntyre and Rens (2016). Water is diverted from the river and into the alluvial ASR by a reverse spillway (Figure C), to divert water from the river into smaller canals and networks powered by gravity. Since peak flows typically satisfy all decreed water rights, any water available above a certain stage will result in free river, the diversion will comply with Colorado water laws. Should a portion of stored water be considered out of priority, then this water can be measured and pumped back into the stream.
A key component is the stream diversion structure because contained within the structure, is a solids removal treatment system. The patent portrays alluvial ASR as viable, if, solids are removed prior to water injection in order to maintain water quality in the facility. A design for solids removal is not contained within the scope of this research effort, other to alert, if not properly designed and constructed, an efficiently operating ASR facility is problematic.
A field visit to the case study site in Brighton, CO. at the start of this research work assisted in designing the dimensions of an alluvial ASR facility. The facility, designed underground on alluvial land beside the river, will have an areal extent denoted by the box in Figure 2, and an assumed depth to store water. The Brighton alluvial site is 195 m (639.8 ft) wide, 380 m (1246.7 ft) long, and 10 m (32.8 ft) deep with an assumed porosity of 20%. These dimensions equate to approximately 148,000 m3 (120 ac.ft) of storage volume. The ASR vessel will be enclosed by an impermeable slurry wall, keyed into bedrock, required to contain and control the surface water inflow. This could be constructed, for example, from interlocking sheet piles or from a bentonite slurry placed
12


in an excavated trench, the latter of which is assumed to purposes of cost estimation. Figure 4 shows the underground view of alluvial ASR confined by slurry wall on four sides.
Soils information was provided by the U.S. Soil Conservation Service (1974) and the U.S. Natural Resources Conservation Service (2014). The three types of soils in and near the case study site are sandy alluvial land, loamy alluvial land with gravelly substratum, and wet alluvial land. All three soil types are indicative of gravelly soils which makes the selected site a good alluvial ASR case study. Pumping of water in and out of the alluvial ASR is achieved by a set of 8 wells that symmetrically recharge and extract water. These wells are designed to be 9.5 m deep (Figure 5), with perforated pipes connecting them to form a rectangular network as shown in Figure 6. Perforated pipes aid in pumping out water without reducing the head to a small amount in the bottom layer so that dry cells do not affect the performance of the pumps.
Recharging water into the alluvial ASR is achieved with by the gravitational forces which carry water from a river upstream at a higher elevation and into this alluvial ASR at a lower elevation. The network of wells and perforated pipes carry water into the soil matrix, which stores water in pore spaces between soil particles. While alluvial ASR recharge can be facilitated by gravity, working out against it calls for mechanical pumps that are powered by electricity. Hence, the major energy expense, in operating this alluvial ASR facility, comes from running those 8 pumps.
Additional use of instruments, to test water quality, soil matrix porosity, water turbidity, and water level, which assists in the automation of operating this alluvial ASR facility would result in extra energy and cost.
13


Inlet with shutoff valve
j
Water source (stream or river)
\
Figure 3: Gravity powdered recharging of alluvial ASR (McIntyre et al. 2016)
Figure 4: Confinement using Slurry walls/ sheet piling to setup a static alluvial ASR system.
14


Figure 5: 3D model of modeled alluvial ASR system showing wells at different locations in layer 2.
Figure 6: Perforated pipe assembly in layer 2, 9.5 m below surface level.
15


Technical Feasibility
The study by CDM Smith (2013) for Colorado Water Conservation Board analyzed samples from the South Platte River alluvium, reporting median values for hydraulic conductivity, K = 130 m/d (425.4 ft/d), transmissivity T= 1300 m2/d (14,000 ft2/d), and specific yield Sy = 0.2 These values give a rough estimate of the median saturated thickness of b = T/K = 10 m (32.8 ft). The filling constraint is the maximum infiltration rate permitted by the in-situ soils on site since the vessel is below ground and the intent is no evaporation component associated with any of the inflow quantity. This aspect of the case study is examined via a 3-dimensional groundwater model described below.
A groundwater simulation model, developed with MODFLOW (Harbaugh and Barlow 2006), was constructed to evaluate the technical feasibility of the alluvial ASR facility described above. The MODFLOW model assumes a two-layered alluvial aquifer with varying transmissivity and dimensions as discussed under Facility Design above. The first layer from top surface is 9.5m (31 ft) thick contains the wells and perforated pipe system. The bottom layer, below first layer, is 0.5m (1.6 ft) thick and is assumed to be saturated with water all the time to prevent running into dry cells at the time of simulation. During MODFLOW simulations for water recharge and extraction, unsaturated condition implies there is 1.5m depth of water, and saturation condition implies 9.5 m depth of water is totally available. Therefore, at any given time of operation, the useable level of water from this facility is 8 m, which translates to a useable storage volume of 118,500 m3 (~96 ac-ft). Recharging and pump-out simulations were implemented with MODFLOWs PCG2 solver, which uses Modified Incomplete
16


Cholesky and the Polynomial methods to check convergence conditions in each cell, thereby preventing dry cell conditions when water is being pumped out through the wells.
The model was used to answer the following questions: First, what is the operating protocol or algorithm to store and extract water without surface ponding or running into dry cells? Second, how many days are required to fill/empty the facility as a function of pumping rate? And third, to account for clogging, how do results depend on reduced hydraulic conductivity? What is the optimal design configuration for the manifold of injection-extraction wells?
Pumping algorithm for this alluvial ASR was designed with a conservative approach that there is at least a minimum flow of 2500 m3/day (1 cfs) in the river. This algorithm ensures that alluvial ASR is not over filled during recharge phase, and does not pump out too much water to cause dry cells in the soil matrix around extraction bores. Appendix A of this report contains the algorithm for operating this alluvial aquifer storage and recovery facility.
Under no clogging conditions (K= 130 m/day), this alluvial ASR facility could be filled up from 1.5 m to 9.5 m in about six and a half days, and this 8 m (26 ft) depth of water could be extracted in nine days using pumps. These results were obtained under an assumption that not more than 2500 m3 of water could be removed from the river on any single day.
The question of clogging is a critical design and operational consideration in any proposed alluvial ASR project. The alluvial ASR site at Brighton, Colorado has an assumed unclogged hydraulic conductivity of K= 130 m/d (426.5 ft/d). Several MODFLOW simulations were generated by varying the value of K to simulate clogging
17


by setting K = {K, Kl2, K/4, KJ8}. Figure 7 and Figure 8 respectively show the time taken to completely recharge and empty the alluvial ASR under various levels of clogging. The change in hydraulic results were minimal and did not change any of the assumed initial design parameters such as diversion rates, pipe diameter, or pump out requirements. A pretreatment structure before pumping water into the alluvial ASR will remove suspended solids and other chemical colloids that clog up the soil matrix.
Repeated filling and emptying of the alluvial ASR causes varying effective and pore pressure in the soil matrix, causing soil consolidation. Figure 9 is a graph of varying total stress in the soil matrix. Knowing the pore pressure and total stress for varying levels of water level saturation, overall soil consolidation in this ASR facility was theoretically calculated to be only 0.75 cm (0.025 ft). This negligible amount of consolidation does not have any impact on geological conditions on the land where the alluvial ASR facility is constructed. Appendix B shows the calculations to compute total soil matrix consolidation in the alluvial ASR case study site.
18


Head in the ASR
10 9 8 7 6
I 5
4 3 2 1 0
0 4 8 12 16 20 24
Time
Initial head
Final head
Figure 7: Time taken to saturate the alluvial ASR under varying levels of clogging.
19


Head in the ASR
10
12
Time
[d]
16
20
24
Initial head
Final head
Figure 8: Time taken to pump out all usable water from the alluvial ASR under varying levels of clogging.
20


Total stress @ Saturation Pore Pressure @ Saturation
....Total stress @ Unsaturation - -Pore Pressure @ Unsaturation
Figure 9: Total soil consolidation at the case study site.
21


Administrative Feasibility
Water law in Colorado is based on the doctrine of prior appropriation (Hobbs, 2004; McIntyre and Mays, 2017), which grants water rights in the order in which they were put to beneficial use, with accounting for expected return flows via surface- or groundwater. When flow is insufficient to meet demand, those holding senior water rights my place a call on the river to guarantee delivery of their appropriation. Accordingly, feasibility of the alluvial ASR design depends not only on technical feasibility, as discussed above, but also on administrative feasibility. For surface water diversions, Colorado water law establishes a date of priority decreed by the water court and administered by Colorado Division of Water Resources (CDWR). The dates of priority, with the most senior being first, are typically listed on a tabulation of water rights that the CDWR references when establishing who on a stretch of stream or river, is permitted to divert and by how much. The South Platte River is over-appropriated, meaning there is seldom sufficient water to satisfy all decreed surface water rights (HDR Engineering and West Sage Water Consultants 2015).
For a new water storage structure to be able to divert legally, must first secure a court decree with a priority date. The alluvial ASR facility at Brighton would be granted a junior priority date, meaning more senior water rights would take priority. Therefore, an analysis of the nature of the flow in the river at Brighton is required to access when and if the vessel can fill. Two administrative classifications of surface water are available to fill the alluvial ASR facility: (1) free river, and (2) fully reusable effluent water. Each of these classifications will be discussed in turn. Free river occurs when flow remains in the river even after all decreed water rights are diverting their full entitlements without
22


reduction. The excess water in the river is termed free river and is available to even the
most junior decreed priority.
To identify the specific time periods of free river, an analysis was performed to determine, based upon 14 years of flow data WY (2000 to WY 2013), which periods of flow were classified free river at the Brighton site and therefore available for diversion, even by the most junior water right priority.
Free River Analysis Results for Periods at Brighton between 2000 and 2013
The framework and reference for this study is based on the South Platte Implementation Plan (HDR Engineering and West Sage Water Consultants 2015). Appendix G of this report provides the basic information to perform a Point-Flow Analysis at the Henderson Gage of the South Platte River with an aim to calculate the quantitative availability of free river in a Water Year (Oct. 1st to Sept. 30th) from 2000 to 2013. Flow through the Henderson Gage is roughly the same as flow on South Platte River at the Brighton case study site, since no water diversion of augmentation is present between the two places.
Original stream flow data along with daily call chronology was downloaded from South Platte Basin Implementation Plan report compiled by HDR Engineering and West Sage Water Consultants (2015). Free-River analysis was carried out using an RStudio program. Appendix C contains the lines of code written in RStudio to calculate free river availability at Henderson Gage.
Call Chronology
Appendix G of South Platte Basin Implementation Plan by HDR Engineering and West Sage Water Consultants (2015) dictates that a call placed anywhere downstream
23


from the point of analysis implies absence of free river conditions in the river on that day. Hence a call anywhere from Henderson gage to Liddle Ditch implies a call on the river. 0 denotes no recorded call for a day, while 1 indicates presence of a call placed at one or many downstream gages/ditches from Henderson Gage.
Call Chronology is calculated from the Point Flow Spreadsheet 01/16/15.xlsx containing call records for each gage/ditch on the South Platte River. If a senior water right holder does not obtain allocated supply, then is a call is placed anywhere from Denver metro to Liddle Ditch. Water is diverted to this senior water right holder before supplying to other.
River Compact Deal at Liddle Ditch
South Platte River Compact requires a minimum flow of 120 cfs to be sent across the state border to Nebraska between April 1 and October 15. A Compact Call is placed on the river whenever physical flow at Liddle Ditch is below 120 cfs, to indicate absence of free river condition. However, the Water Availability Model from South Platte River Metro Roundtable have not applied Compact Deal condition while computing annual free-river availability.
Water Availability Analysis
Daily amount of Free-River available at Henderson Gage is analyzed using Point Flow method, which states that total amount of free-river available at Henderson gage is the minimum flow among the total flow across all the downstream ditches and gages from Henderson Gage to Liddle Ditch, minus the diversions.
24


Reusable Effluent Water
The second administrative classification of surface water available to fill the alluvial ASR facility is fully reusable effluent water, which is water supply from nontributary sources. Non-tributary sources are those that would not be present in the river in the absence of engineering intervention, such as tunnels under the continental divide that provide water from the western slope of the Rocky Mountains, or wells extracting nontributary groundwater. At the Brighton case study ASR site, the following water providers have fully reusable effluent water that passes the site: Arvada, Aurora,
Brighton, Denver, Ft. Lupton, Thornton, and Westminster. Broomfield and Northglenn discharge their reusable effluent into Big Dry Creek, which enters the South Platte River downstream of Brighton; however, this water could be exchanged (move point of diversion upstream) up to the Brighton ASR site to fill the vessel.
Under Colorado water law, municipal water providers are entitled to capture their fully reusable effluent water for additional use. The Colorado Division of Water Resources maintains an accounting of reusable effluent flows as submitted by the individual municipal water providers. The yearly average of available fully reusable effluent water, assuming permission from the municipality is granted, to store in the alluvial ASR facility is found in Table 2. Table 2 shows that there is sufficient fully reusable effluent water at the alluvial ASR site to fill and refill the vessel multiple times each month of the year. However, permission must first be granted by the municipal entity to store their water in the vessel.
Water effluent from Denver and Aurora's wastewater treatment facilities belong to the respective cities under Water Rights Law. Some of the reusable effluent water are
25


sold to utilities by the City of Denver and Aurora. Remaining reusable effluent water, which is a portion of physical flow in the South Platte, can be reclaimed for use in the future, and hence it should not be accounted for in free- river analysis. South Platte Basin Implementation Plans Water Availability Model contains daily unused reusable effluent water from Denver and Aurora in the Unused Reusable Flow.
26


Table 2: Yearly average of unused reusable effluent flow from Denver and Aurora
Water Year Unused reusable flow |m3/sl Unused reusable flow Icfsl
2000 18.6 658.0
2001 42.5 1502.3
2002 50.6 1789.3
2003 31.4 1109.3
2004 18.7 661.1
2005 21.1 745.2
2006 19.4 683.8
2007 0.0 0.0
2008 16.6 586.2
2009 0.0 0.0
2010 21.1 745.4
2011 28.1 991.1
2012 26.7 943.7
2013 42.9 1517.5
27


Total amount of Free River available at Henderson Gage
Actual water available minus the unused effluent flow gives total free-river at Henderson Gage. Total amount of Free-River available m3 (ac-ft) is calculated by summing daily amount of free-river for each Water Year (WY) from 2000 to 2013. Figure 10 is the graph free river available daily, and Figure 11 is the graph of annual free river available each year. Water years 2002 and 2003 were completely dry with no free river available, while water years 2000 and 2009 had lot of free river and call diversions were minimum. Figure 12 highlights log of monthly mean, minimum and maximum amount of available free river over those 14 water years. April, May and June have more number of days with free river since spring melt adds significantly high amount of flow to the river. By end of summer free river conditions are fewer and more calls are placed on the river. January and February have least amount of free river availability. Time series correlation analysis showed a weak positive relationship between the water years. Climate variability and stream flow modifications can be hypothesized to result in a poor R2 value (Figure 13).
Results obtained by the free river analysis using RStudio was validated by comparing it with the water availability model for Henderson gage from South Platte Basin Implementation Basin Surface Water Availability Model (HDR Engineering and West Sage Water Consultants 2015). Except for water year 2013, total free river availability calculated from water year 2000-2012 closely matched with the validation data. In the water year 2013, September flood was believed to be neglected in the free analysis by HDR engineering and West Sage Water Consultants during the, while this research work considered all the natural events contributing to river flow variability.
28


[09s/£lu] 9|qe|!eAe J9AU 99jj
Figure 10: Daily summary of free river available from Water Year (WY) 2000 WY 2013.
29
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014
Water Year [10/1 to 09/30]


[y-oe £v0 !]
Apjenb JS}e/v\
Figure 11: Annual free river available from WY 2000 WY 2013.
3000 -r
----1----------1----------1----------1---------1----------1----------1----------1----------1----------1----------1----------1----------1----------1--
WY2000 WY2001 WY2002 WY2003 WY2004 WY2005 WY2006 WY2007 WY2008 WY2009 WY2010 WY2011 WY2012 WY2013
Water Year [10/1 to 09/30]


[09S/£UJ]
Ajjjuenb J8}e/v\
Figure 12: Monthly mean, minimum and maximum available free river from water year 2000 to 2013.
31
January February March April May June July August September October November December
Months


Free River available [m3]
250
200
y = 2.35x-4,662.14 O R2 = 0.03
O water available Linear fitted line
Figure 13: Time series correlation analysis of Free River available from WY 2000 to WY 2013
32


CHAPTER 4
DISCUSSION AND CONCLUSION
Based on the literature review of existing alluvial ASR projects in the US and worldwide, the MODFLOW simulations, and the legal availability of surface water in the South Platte River at the Brighton, Colorado ASR case study site, alluvial ASR is a viable water storage option in the South Platte River basin in Colorado. However, pretreatment of the diverted South Platte River surface water is required to ensure water quality and prevent clogging of soil matrix by silt and other particles carried down by the river. Free river analysis from water year 2000 to 2013 provide proof that free river conditions exist for an average of 103 days per year, with an average flow of 2.15 m3/s (76 cfs) and satisfies the recharging rate needed to fill the alluvial ASR facility. However, to setup the perforated pipe system, for the purpose of faster water extraction from the alluvial ASR, the ground would first have to be dug out, install the pipes and then repack the facility with soil. Construction cost of an alluvial ASR site was beyond the scope of this research, hence future work should compare the cost analysis of building an alluvial ASR versus an above-ground dam or reservoir.
In Colorado, the property nature of water would constrain a junior priority ASR water storage vessel, but not for free river and fully reusable effluent. Because of these two classes of water, alluvial ASR is a viable water storage methodology at the case study site near Brighton, Colorado. However, the potential of clogging could temper wide use of ASR elsewhere in Colorado. Because of this, site selection including detailed evaluation of in situ materials, is a critical consideration. In addition, the existing water rights in a given location must also be evaluated.
33


REFERENCES
Barlow, M. P., and Harbaugh, W. A. (2006). USGS Directions in MODFLOW Development. Ground Water, Vol. 44 (6), 771-774.
Brown, C. J. et al. (2006). Lessons Learned from a Review of 50 ASR projects from the United States, England, Australia, India, and Africa. Conference Proceedings Southern Illinois University, Carbondale, IL, (Mar. 5, 2017).
CDM Smith (2013). South Platte Decision Support System Alluvial Groundwater Model Report. Colorado Water Conservation Board, April 2013.
Finley, B. (2015). Wests water reservoir managers wrestle with evaporation. The Denver Post, (Mar. 6, 2017).
Hanson, G., and A. Nilsson (1986). Groundwater dams for rural-water supplies in developing-countries. Ground Water, 24(4), 497-506.
HDR Engineering and West Sage Water Consultants (2015). South Basin Implementation Plan. South Platte Basin Roundtable, April 17, 2015.
HDR Engineering and West Sage Water Consultants (2015). Appendix G South Platte Basin Surface Water Availability Analysis. South Platte Basin Roundtable, March 16, 2015.
Ishida, S., Tsuchihard, T., Yoshimoto, S., and Imaizumi, M. (2011) Sustainable Use of Groundwater with Underground Dams. Japan Agricultural Research Quarterly, Vol. 45 (1), 51-61 (2011).
Kennedy, C. (2014). Future Temperature and Precipitation Change in Colorado.
NOAA, (Mar. 5, 2017).
Ma, Z., Jia, C., Guo, J., Gu, H., and Miao, Y. (2014). Features analysis of five-element theory and its basal effects on construction of visceral manifestation theory. J. Tradit. Chin. Med, 34(1), 115-121.
McIntyre, W. C. and Mays, D. C. (2017). Roles of the Water Court and the State Engineer for Water Administration in Colorado. Water Policy, in press.
McIntyre, W. C. and Rens, K. L. (2016). "System And Method Of Using Differential Elevation Induced Energy For The Purpose Of Storing Water Underground, To The Regents of University of Colorado, a body corporate, Denver, CO (USA), U.S. Patent 9,278,808 Bl, March 28, 2016.
34


Topper, R., Barkmann, P. E., Bird, D. A., and Sares, M. A. (2004). Artificial Recharge of Ground Water in Colorado -A Statewide Assessment. Environment Geology 13, Colorado Geological Survey, Denver, CO, 2004.
Simmons, G. (2014). Underground dams the solution to Australias drought problems. The Sydney Morning Herald, (Mar. 5, 2017).
Statewide Water Supply Initiative (2010). Colorados Water Supply Future. Colorado Water Conservation Board, January 2010.
35


APPENDIX A
ALGORITHM TO RECHARGE ALLUVIAL ASR
36


APPENDIX B
SOIL MATRIX CONSOLIDATION
FULLY WET CONDITION
Layer Layer thickness [m] Elevat ion,z [m] Head [m] Stress[N/m2] Total stress [N/m2] @ h=10m Pore Pressure [N/m2]@ h=10m Eff. Stress [N/m2] @ h=10m
0 0 10 10 0 0 0 0
1 0.5 9.5 10 12262.5 12262.5 4905 7357.5
2 0.5 9 10 12262.5 24525 9810 14715
3 0.5 8.5 10 12262.5 36787.5 14715 22072.5
4 0.5 8 10 12262.5 49050 19620 29430
5 0.5 7.5 10 12262.5 61312.5 24525 36787.5
6 0.5 7 10 12262.5 73575 29430 44145
7 0.5 6.5 10 12262.5 85837.5 34335 51502.5
8 0.5 6 10 12262.5 98100 39240 58860
9 0.5 5.5 10 12262.5 110362. 5 44145 66217.5
10 0.5 5 10 12262.5 122625 49050 73575
11 0.5 4.5 10 12262.5 134887. 5 53955 80932.5
12 0.5 4 10 12262.5 147150 58860 88290
13 0.5 3.5 10 12262.5 159412. 5 63765 95647.5
14 0.5 3 10 12262.5 171675 68670 103005
15 0.5 2.5 10 12262.5 183937. 5 73575 110362.5
16 0.5 2 10 12262.5 196200 78480 117720
17 0.5 1.5 10 12262.5 208462. 5 83385 125077.5
18 0.5 1 10 12262.5 220725 88290 132435
19 0.5 0.5 10 12262.5 232987. 5 93195 139792.5
20 0.5 0 10 12262.5 245250 98100 147150
37


DRY CONDITION
Head Total stress Pore Pressure Eff. Stress [N/m2]
[m] Stress[N/m2] [N/m2] @ h=0.5m [N/m2] @ h=0.5m @ h=0.5m
10 0 0 0 0
9.5 11281.5 11281.5 0 11281.5
9 11281.5 22563 0 22563
8.5 11281.5 33844.5 0 33844.5
8 11281.5 45126 0 45126 |
7.5 11281.5 56407.5 0 56407.5
7 11281.5 67689 0 67689
6.5 11281.5 78970.5 0 78970.5
6 11281.5 90252 0 90252
5.5 11281.5 101533.5 0 101533.5
5 11281.5 112815 0 112815
4.5 11281.5 124096.5 0 124096.5
4 11281.5 135378 0 135378
3.5 11281.5 146659.5 0 146659.5
3 11281.5 157941 0 157941
2.5 11281.5 169222.5 0 169222.5
2 11281.5 180504 0 180504
1.5 11281.5 191785.5 0 191785.5
1 11281.5 203067 0 203067
0.5 11281.5 214348.5 0 214348.5
0.5 12262.5 226611 4905 221706
Dry Bulk Density = 2300 kg/m3
Wet bulk density = Dry bulk density + (Soil porosity Density of water) Wet Bulk Density = 2500 kg/m3
Soil compressibility, a = IE-08 m2/N
Change in effective stress = dry condition wet condition 221706 147150 [N/m2]
Aove= 74556 N/m2
Ab/ b0 = -aAove
Ab = b0aAove
Ab = -0.0075 m
Ab = -0.75 cm
Soil matrix compresses by: 0.75 cm
38


APPENDIX C
RSTUDIO CODE FOR FREE RIVER ANALYSIS
Download all spreadsheets from supplemental materials:
## codes to process the raw data from south platte metro roundtable ##
## set directory where the spreadsheet and RStudio code is present ##
setwd("C:/Users/........ ")
install.packages("data.table")
install.packages("ggplot2") # for creating graphs
install.packages("psych")
install.packages("car")
install. packages(" sm")
install.packages("lsr")
install.packages("reshape")
install. packages(" scales")
library(scales) # to access breaks/formatting functions
library(data.table)
library(psych)
library(car)
library(lsr)
library(ggplot2)
library(reshape)
library(sm)
# River data with flow from "At the structure" # rawdata<-read.csv("Point Flow.csv",header = TRUE)
river<- cbind(rawdata$X,rawdata[,seq(2,ncol(rawdata),2)]) # from column 2,
alternatively choose columns to be used
setnames(river, "rawdata$X", "Date")
river[,l] <- as.Date(river[,l], format = "%m-%d-%Y")
r=nrow(river)
# compact at liddle ditch for (i in l:r){
if (river[i,62] <= 120) { river[i,62] <- 0
}
else {
river[i,62] <- river[i,62]
}
}
View(river)
39


# call data
rawcall<-read.csv("Call.csv",header = TRUE) call <- rawcall[seq(10,nrow(rawcall),l),] row.names(call) <- l:nrow(call) # reset serial no to 1,2,3,.. setnames(call,"X",''Date")
call[is.na(call)] <- 0 #set all n/a to 0, representing no call for (i in l:r) {
river[i,64]<- max(call[i,l 1:63]) #call is placed on the river if any structure below Henderson raise a call
}
setnames(river,"V64",''Call")
# Unused Reusable Flow
rawunused<- read.csv("Unused Reusable Flow.csv", header=TRUE) river[,65] <- rawunused[277:5390,12] setnames(river," V65", "Unused") river[,65]<-as.numeric(as.character(river[,65]))
# Gross free river: min flow from Henderson to Liddle Ditch and sum of channels, corresponding to no call days
for(i in l:r){ if (river[i,64]==l){
river[i,66] <- 0 #if call=l, no free river exists } else {
river[i,66] <- min(river[i,l 1:63]) #else if call=0, potential free river = min flow in SP
}
}
setnames(river,"V66","Gross free river")
#Final free river available everyday for (i in l:r){
river[i,67]= river[i,66]-river[i,65] if (river[i,67]<0) { river[i,67] <- 0 } else {
river[i,67] <- river[i,67]
}
}
setnames(river," V67", "Freeriver")
river[,68] <- river[,67]*0.0283 setnames(river,"V68","Freeriver [m3/sec]")
river[,69] <- river[,67]* 1.9835 setnames(river,"V69", "Freeriver [ac.ft/day]")
40


## read climate data ##
climate <- read.csv("monthlytempnprecip.csv", header = TRUE)
climate [1,6] = sum(river$Freeriver[2:32])
climate [2,6] <- sum(river$Freeriver[33:62])
climate [3,6] <- sum(river$Freeriver[63:93])
climate [4,6] <- sum(river$Freeriver[94:124])
climate [5,6] <- sum(river$Freeriver[125:153])
climate [6,6] <- sum(river$Freeriver[154:184])
climate [7,6] <- sum(river$Freeriver[185:214])
climate [8,6] <- sum(river$Freeriver[215:245])
climate [9,6] <- sum(river$Freeriver[246:275])
climate [10,6] <- sum(river$Freeriver[276:306])
climate [11,6] <- sum(river$Freeriver[307:337])
climate [12,6] <- sum(river$Freeriver[338:367])
climate [13,6] <- sum(river$Freeriver[368:398])
climate [14,6] <- sum(river$Freeriver[399:428])
climate [15,6] <- sum(river$Freeriver[429:459])
climate [16,6] <- sum(river$Freeriver[460:490])
climate [17,6] <- sum(river$Freeriver[491:518])
climate [18,6] <- sum(river$Freeriver[519:549])
climate [19,6] <- sum(river$Freeriver[550:579])
climate [20,6] <- sum(river$Freeriver[580:610])
climate [21,6] <- sum(river$Freeriver[611:640])
climate [22,6] <- sum(river$Freeriver[641:671])
climate [23,6] <- sum(river$Freeriver[672:702])
climate [24,6] <- sum(river$Freeriver[703:732])
climate [25,6] <- sum(river$Freeriver[733:763])
climate [26,6] <- sum(river$Freeriver[764:793])
climate [27,6] <- sum(river$Freeriver[794:824])
climate [28,6] <- sum(river$Freeriver[825:855])
climate [29,6] <- sum(river$Freeriver[856:883])
climate[30,6] <- sum(river$Freeriver[884:914])
climate[31,6] <- sum(river$Freeriver[915:944])
climate[32,6] <- sum(river$Freeriver[945:975])
climate[33,6] <- sum(river$Freeriver[976:1005])
climate[34,6] <- sum(river$Freeriver[1006:1036])
climate[35,6] <- sum(river$Freeriver[1037:1067])
climate[36,6] <- sum(river$Freeriver[ 1068:1097])
climate[37,6] <- sum(river$Freeriver[1098:1128])
climate[38,6] <- sum(river$Freeriver[l 129:1158])
climate[39,6] <- sum(river$Freeriver[l 159:1189])
climate[40,6] <- sum(river$Freeriver[l 190:1220])
climate[41,6] <- sum(river$Freeriver[1221:1248])
climate[42,6] <- sum(river$Freeriver[1249:1279])
climate[43,6] <- sum(river$Freeriver[1280:1309])
41


climate[44,6] <- sum(river$Freeriver[1310:1340]) climate[45,6] <- sum(river$Freeriver[1341:1370]) climate[46,6] <- sum(river$Freeriver[1371:1401]) climate[47,6] <- sum(river$Freeriver[1402:1432]) climate[48,6] <- sum(river$Freeriver[1433:1462]) climate[49,6] <- sum(river$Freeriver[1463:1493]) climate[50,6] <- sum(river$Freeriver[1494:1523]) climate[51,6] <- sum(river$Freeriver[1524:1554]) climate[52,6] <- sum(river$Freeriver[1555:1585]) climate[53,6] <- sum(river$Freeriver[1586:1614]) climate[54,6] <- sum(river$Freeriver[1615:1645]) climate[55,6] <- sum(river$Freeriver[1646:1675]) climate[56,6] <- sum(river$Freeriver[1676:1706]) climate[57,6] <- sum(river$Freeriver[1707:1736]) climate[58,6] <- sum(river$Freeriver[1737:1767]) climate[59,6] <- sum(river$Freeriver[ 1768:1798]) climate[60,6] <- sum(river$Freeriver[1799:1828]) climate[61,6] <- sum(river$Freeriver[1829:1859]) climate[62,6] <- sum(river$Freeriver[1860:1889]) climate[63,6] <- sum(river$Freeriver[1890:1920]) climate[64,6] <- sum(river$Freeriver[1921:1951]) climate[65,6] <- sum(river$Freeriver[1952:1979]) climate[66,6] <- sum(river$Freeriver[1980:2010]) climate[67,6] <- sum(river$Freeriver[2011:2040]) climate[68,6] <- sum(river$Freeriver[2041:2071]) climate[69,6] <- sum(river$Freeriver[2072:2101]) climate[70,6] <- sum(river$Freeriver[2102:2132]) climate[71,6] <- sum(river$Freeriver[2133:2163]) climate[72,6] <- sum(river$Freeriver[2164:2193]) climate[73,6] <- sum(river$Freeriver[2194:2224]) climate[74,6] <- sum(river$Freeriver[2225:2254]) climate[75,6] <- sum(river$Freeriver[2255:2285]) climate[76,6] <- sum(river$Freeriver[2286:2316]) climate[77,6] <- sum(river$Freeriver[2317:2344]) climate[78,6] <- sum(river$Freeriver[2345:2375]) climate[79,6] <- sum(river$Freeriver[2376:2405]) climate[80,6] <- sum(river$Freeriver[2406:2436]) climate[81,6] <- sum(river$Freeriver[2437:2466]) climate[82,6] <- sum(river$Freeriver[2467:2497]) climate[83,6] <- sum(river$Freeriver[2498:2528]) climate[84,6] <- sum(river$Freeriver[2529:2558]) climate[85,6] <- sum(river$Freeriver[2559:2589]) climate[86,6] <- sum(river$Freeriver[2590:2619]) climate[87,6] <- sum(river$Freeriver[2620:2650]) climate[88,6] <- sum(river$Freeriver[2651:2681]) climate[89,6] <- sum(river$Freeriver[2682:2709])
42


climate[90,6] <- sum(river$Freeriver[2710:2740]) climate[91,6] <- sum(river$Freeriver[2741:2770]) climate[92,6] <- sum(river$Freeriver[2771:2801]) climate[93,6] <- sum(river$Freeriver[2802:2831]) climate[94,6] <- sum(river$Freeriver[2832:2862]) climate[95,6] <- sum(river$Freeriver[2863:2893]) climate[96,6] <- sum(river$Freeriver[2894:2923]) climate[97,6] <- sum(river$Freeriver[2924:2954]) climate[98,6] <- sum(river$Freeriver[2955:2984]) climate[99,6] <- sum(river$Freeriver[2985:3015]) climate[ 100,6] <- sum(river$Freeriver[3016:3046]) climate[101,6] <- sum(river$Freeriver[3047:3075]) climate[ 102,6] <- sum(river$Freeriver[3076:3106]) climate[ 103,6] <- sum(river$Freeriver[3107:3136]) climate[ 104,6] <- sum(river$Freeriver[3137:3167]) climate[ 105,6] <- sum(river$Freeriver[3168:3197]) climate[ 106,6] <- sum(river$Freeriver[3198:3228]) climate[ 107,6] <- sum(river$Freeriver[3229:3259]) climate[ 108,6] <- sum(river$Freeriver[3260:3289]) climate[ 109,6] <- sum(river$Freeriver[3290:3320]) climate[l 10,6] <- sum(river$Freeriver[3321:3350]) climate[l 11,6] <- sum(river$Freeriver[3351:3381]) climate[l 12,6] <- sum(river$Freeriver[3382:3412]) climate[l 13,6] <- sum(river$Freeriver[3413:3440]) climate[l 14,6] <- sum(river$Freeriver[3441:3471]) climate[l 15,6] <- sum(river$Freeriver[3472:3501]) climate[l 16,6] <- sum(river$Freeriver[3502:3532]) climate[l 17,6] <- sum(river$Freeriver[3533:3562]) climate[l 18,6] <- sum(river$Freeriver[3563:3593]) climate[l 19,6] <- sum(river$Freeriver[3594:3624]) climate[ 120,6] <- sum(river$Freeriver[3625:3654]) climate[121,6] <- sum(river$Freeriver[3655:3685]) climate[ 122,6] <- sum(river$Freeriver[3686:3715]) climate[123,6] <- sum(river$Freeriver[3716:3746]) climate[ 124,6] <- sum(river$Freeriver[3747:3777]) climate[125,6] <- sum(river$Freeriver[3778:3805]) climate[ 126,6] <- sum(river$Freeriver[3806:3836]) climate[ 127,6] <- sum(river$Freeriver[3837:3866]) climate[128,6] <- sum(river$Freeriver[3867:3897]) climate[ 129,6] <- sum(river$Freeriver[3898:3927]) climate[130,6] <- sum(river$Freeriver[3928:3958]) climate[131,6] <- sum(river$Freeriver[3959:3989]) climate[ 132,6] <- sum(river$Freeriver[3990:4019]) climate[133,6] <- sum(river$Freeriver[4020:4050]) climate[ 134,6] <- sum(river$Freeriver[4051:4080]) climate[135,6] <- sum(river$Freeriver[4081:4111])
43


climate[136,6] <- sum(river$Freeriver[4112:4142]) climate[137,6] <- sum(river$Freeriver[4143:4170]) climate[138,6] <- sum(river$Freeriver[4171:4201]) climate[139,6] <- sum(river$Freeriver[4202:4231]) climate[ 140,6] <- sum(river$Freeriver[4232:4262]) climate[141,6] <- sum(river$Freeriver[4263:4292]) climate[ 142,6] <- sum(river$Freeriver[4293:4323]) climate[143,6] <- sum(river$Freeriver[4324:4354]) climate[ 144,6] <- sum(river$Freeriver[4355:4384]) climate[145,6] <- sum(river$Freeriver[4385:4415]) climate[146,6] <- sum(river$Freeriver[4416:4445]) climate[ 147,6] <- sum(river$Freeriver[4446:4476]) climate[148,6] <- sum(river$Freeriver[4477:4507]) climate[ 149,6] <- sum(river$Freeriver[4508:4536]) climate[ 150,6] <- sum(river$Freeriver[4537:4567]) climate[151,6] <- sum(river$Freeriver[4568:4597]) climate[ 152,6] <- sum(river$Freeriver[4598:4628]) climate[153,6] <- sum(river$Freeriver[4629:4658]) climate[ 154,6] <- sum(river$Freeriver[4659:4689]) climate[155,6] <- sum(river$Freeriver[4690:4720]) climate[ 156,6] <- sum(river$Freeriver[4721:4750]) climate[ 157,6] <- sum(river$Freeriver[4751:4781]) climate[158,6] <- sum(river$Freeriver[4782:4811]) climate[ 159,6] <- sum(river$Freeriver[4812:4842]) climate[ 160,6] <- sum(river$Freeriver[4843:4873]) climate[161,6] <- sum(river$Freeriver[4874:4901]) climate[ 162,6] <- sum(river$Freeriver[4902:4932]) climate[ 163,6] <- sum(river$Freeriver[4933:4962]) climate[ 164,6] <- sum(river$Freeriver[4963:4993]) climate[ 165,6] <- sum(river$Freeriver[4994:5023]) climate[ 166,6] <- sum(river$Freeriver[5024:5054]) climate[ 167,6] <- sum(river$Freeriver[5055:5085]) climate[ 168,6] <- sum(river$Freeriver[5086:5114])
# rename the columns
setnames(climate,"V6",''Total_free_river_cfs") climate[,7]=climate[, 6] *0.0283 setnames(climate,"V7",''Total_free_river_m3/sec") climate[,8]=climate[, 6]* 1.9835/1000 setnames(climate,"V8",''Total_free_river_10A3 acft") View(climate)
## Annual free river availability dataframe ## water_available = data.frame(matrix(NA,nrow=14,ncol=2))
44


water_available[,l]=c("WY2000","WY2001","WY2002","WY2003","WY2004","WY20
05","WY2006","WY2007","WY2008","WY2009","WY2010","WY2011","WY2012","
WY2013")
water_available[l,2]=sum(climate[l: 12,6]) water_available[2,2]=sum(climate[ 13:24,6]) water_available[3,2]=sum(climate[25:36,6]) water_available[4,2]=sum(climate[37:48,6]) water_available[5,2]=sum(climate[49:60,6]) water_available[6,2]=sum(climate[61:72,6]) water_available[7,2]=sum(climate[73:84,6]) water_available[8,2]=sum(climate[85:96,6]) water_available[9,2]=sum(climate[97:108,6]) water_available[ 10,2]=sum(climate[ 109:120,6]) water_available[ 11,2]=sum(climate[ 121:132,6]) water_available[12,2]=sum(climate[133:144,6]) water_available[ 13,2]=sum(climate[ 145:156,6]) water_available[ 14,2]=sum(climate[ 157:168,6])
setnames(water_available,"Xl",''Water_Year") setnames(water_available, "X2"," Annual_free_river_cfs") water_available[,3]= round((water_available[,2]*0.0283)) water_available[,4]=round((water_available[,2]* 1.9835/1000), digits=l) setnames(water_available,V3"," Annual_free_river_m3") setnames(water_available,V4","Annual_free_river_l0A3 acft")
## Use free river availability data from South Platte Implementation Plan report for validation ##
water_available[,5] = c(141.1,45.7,3.4,0,0,29.2,.9,46.6,4.7,104.4,194.8,84.7,39,0)
setnames(water_available,"V5",''validation_data_10A3 acft")
View(water_available)
#Monthly free river availability (box plot)
ggplot(climate, aes(x=climate[,l], y=climate[,7]))+ geom_boxplot() + scale_y_logl0() + xlab("Months") + ylab("Water quantity \n [m3/sec]") +
scale_x_discrete(limits=c(" January", "February","March", "April","May","June", "July", "A ugust","September", "October","November", "December")) + theme_bw() +
theme(axis.line = element_line(colour = "black"), panel.grid.major = element_blank(), panel.grid.minor = element_blank(), panel.background = element_blank())
# Daily summary of free river availability
ggplot(river, aes(x=river[,l], y=river[,68]))+ geom_line() + xlab("Water Year \n [10/1 to 09/30]") + ylab("Free river available [m3/sec]") + scale_x_date(breaks=date_breaks("l years"),labels=date_format("%Y")) +
45


theme_bw() +
theme(axis.line = element_line(colour = "black"), panel.grid.major = element_blank(), panel.grid.minor = element_blank(), panel.background = element_blank())
# Annual free river availability (box plot)
ggplot(climate, aes(x=climate[,2], y=climate[,7]))+ geom_boxplot() + xlab("Water Year \n [10/1 to 09/30]") + ylab("Water quantity \n [cubic meters]") + theme_bw() +
theme(axis.line = element_line(colour = "black"), panel.grid.major = element_blank(), panel.grid.minor = element_blank(), panel.background = element_blank())
# Annual free river availability (bar plot)
ggplot(water_available, aes(x=water_available[,l], y=water_available[,3])) + geom_bar(stat = "identity") + xlab("Water Year \n [10/1 to 09/30]") + ylab("Water quantity \n [cubic meters]") + theme_bw() +
theme(axis.line = element_line(colour = "black"), panel.grid.major = element_blank(), panel.grid.minor = element_blank(), panel.background = element_blank())
# Correlation test to prove the accuracy of free river analysis with RStudio ## cor.test(water_available[,4],water_available[,5])
46


Full Text

PAGE 1

ADMINISTRATIVE AND TECHNICAL FEASIBILITY OF ALLUVIAL AQUIFER STORAGE AND RECOVERY ON THE SOUTH PLATTE RIVER by CIBI VISHNU CHINNASAMY B.E., Anna University 20 14 A thesis submitted to the Faculty of The Graduate School of the University of Colora do in partial fulfillment of the requirements for the degree of Master of Science Civil Engineering Program 2017

PAGE 2

ii 2017 CIBI VISHNU CHINNASAMY ALL RIGHTS RESERVE D

PAGE 3

iii This thesis for the Master of Science degree by Cibi Vishnu Chinnasamy has been approved for the Civil Engineering Program by Carolin e Clevenger, Chair David C. Mays Dharmarajan Ramaswami Date : May 13, 2017

PAGE 4

iv Chinnasamy, Cibi Vishnu (M.S., Civil Engineering) Free River Analysis in South Platte River for Ground water Storage Using Alluvial Aquifer Storage and Recovery System Thesis directed by Associate Professor David C. Mays. ABSTRACT Increasing population growth coupled with climate variability is causing water managers to reassess water storage. In Colorado, the era of large above ground dams and reservoirs is probably over, due to environmental and endangered species concerns. In this context, an alluvial aquifer storage and recovery (ASR) system presents an alternative option for water storage. A recent stu dy estimated approximately 12 km 3 (10 million acre feet) of storage may be possible in the South Platte River alluvium of northeast Colorado. To investigate this option a case study was conducted to examine the availability of free river conditions on the South Platte River, and simulations of soil matrix stability, aquifer recharge and extraction rate under various levels of clogging, to test the feasibility of an alluvial ASR system for water storage purpose An alluvial ASR site near U.S. Highway 7 at B righton, Colorado with a storage capa city of 118,500 m 3 ( 96 ac ft) was considered for this study. Analyzing river flow data at this site confirms the availability of excess water, during the wet season, to fill the proposed alluvial ASR facility in complia nce with Colorado's water laws. This suggests that alluvial ASR facilities could be a viable option to meet rising water demands of Colorado, prevent wa ter loss due to evaporation, reduce the effect of climate stress on water resources, and avoid the need to purchase lands for above ground water storage facilities. The form and content of this abstract are approved. I recommend its publication. Approved: David C. Mays

PAGE 5

v ACKNOWLEDGEMENTS I would like to thank Mr. Brent Schantz and Mr. Colin Watson at the Sta te at Greeley, Colorado for hosting a field trip to show how water decrees and administrative call s affect the river flow conditions, and for providing data on reusable effluent water. Special thanks to Mr. Joe Frank, Chair of the South P latte Basin Roundtable Chair, for his valuable insights on free river analysis. Most importantly this work gained momentum under the guidance of Dr. William McIntyre, whose patent work and graduate thesis were the cornerstone for this thesis report. I woul d also like to acknowledge the University of Colorado Denver for providing valuable academic support.

PAGE 6

vi TABLE OF CONTENTS CHAPTER I. INTRODU CTION 1 II. ALLUVIAL AQUIFER STORAGE AND RECOVERY .6 III. CASE STUDY ON SOUTH PLATTE RIVER ALLUVIAL ASR 9 Background info rmation .. .... .9 Facility design .. 12 Technical feasibility .. 16 Administrative feasibility .. ..... 22 IV. DISCUSSION AND CONCLUSION 33 R EFERENCES APPENDIX A. Algorithm for recharge alluvial ASR B. Soil matrix consolidation C. RS tudio codes for free river analysis ... ... ...... 39

PAGE 7

vii LI ST OF TABLES TABLE 1 Available storage capacity in South Platte R 10 2 Yearly average of unused reusable effluent flow from Denver and Aurora 27

PAGE 8

viii LIST OF FIGURES FIGURE 1 Map of South Platte River from Denver Metro to Liddle Ditch near Colorado Nebraska state line (CDM Smith 2013) .. 3 2 Selected alluvial site at Brighton, CO. 3959'07.25" N 10449'49.96" W (Google Earth) 11 3 Gravity powdered recharging of alluvial ASR (McIn tyre et al. 2016) ... 14 4 Confinement using Slurry walls/ sheet piling to setup a static alluvial ASR system .. .14 5 3D model of modeled alluvial ASR system showing wells a t different locations in layer 2 .... ..15 6 Perforated pipe assembly in layer 2, 9.5 m below surf ...... 7 Time taken to saturate the alluvial ASR under .19 8 Time taken to pump out all usable water from the alluvial ASR und er varying levels of c logging 20 9 Total soil consolidation at the case study site .21 10 Daily summary of free river availabl e from Water Year 2000 to 2013 .29 11 Annual free r iver available from water year 2 000 to 2013 .30 12 Monthly mean, minimum and max imum ava ilable free river from water year 2000 to 2013 13 Time series correlation analysis of Free River available from WY 2000 to WY 2013 .32

PAGE 9

ix LIST OF ABBREVIATIONS ASR Aquifer Storage and Recovery CWCB Colorado Water Conservation Board SPBIP South Platte Basin Implementation Plan SPDSS South Platte Decision Support System SWSI Statewide Water Supply Initiative WY Water Year (October 1 st to September 30 th )

PAGE 10

1 CHAPTER I INTRODUCTION Water, one of the five basic elements of nature (Ma et al. 2014) is responsible for s growth and progress was determined by thei r ability to distribute, manage and sustainably use water resources. This practice of managing water bodies and structures has advanced from dawn of civilization to the present 21 st century where water drives our economy and commerce. Water has even turne d scorching deserts into vibrant cities, like Dubai, with urban green spaces and urban forestry. With i ncreasing global human population resulting in higher consumption of basic resources, t he task to supply clean water is becoming more challenging and wa ter managers around the world must be prudent and innovative to meet the ever growing water demand. Colorado is blessed to be the home to seven major head waters in the USA. Rio Grande, North Platte, South Platte, Arkansas, Cache la Poudre, and the Republi can Rivers flow East from the Continental Divide, and the Colorado River flows toward the Pacific Ocean These head waters, which originate in the Rockies, have converted Colorado from a semi arid to a bread basket state in the USA. Colorado receives appro ximately on an average 41 cm (16 in) of annual rainfall combined with high evaporative losses (statewide average of approximately 81 % ) result s in a water balance deficit over most of the state, with the exception of the higher mountainous regions (Topper e t al 2003). Based on S tate Demographer s Office population projections, the South Platte and Metro Basins are projected to grow from approximately 3.5 million people in the year 2008 to about 6 million people by the year 2050 ( SWSI 2010 ) Population grow industrial

PAGE 11

2 water needs. The water available to Colorado within its state boundaries is finite, and with climate variability it is anticipated the amount of aggregate annual rainfall will decr ease ( Kennedy 2014 ) Therefore, new water storage methods must be developed that will help solve the storage issue, while also minimizing losses due to evaporation. A recent study conducted by the Colorado Water Conservation Board (CWCB) estimated an abu ndance of alluvial aquifer area useful for storage and recovery, of 6500 km 2 (2500 mi 2 ) along the South Platte River from Denver up to the Colorado Nebraska Stateline as shown in Figure 1 (CDM Smith 2013) The science of using groundwater for domestic and recreational use is not new. Persians, in the early part of first millennium B.C., built elaborate tunnels systems called qanats to extract groundwater from dry mountain basins. These underground aquifers act like natural sponges to store infiltrated wate r from surface runoffs and precipitation and can be pumped out for use anytime. S toring water in underground aquifers is advantageous compared to storing water in conventional dams and reservoirs that have a minimum of 1 0 % evaporation loss ( Finley 2015 ) Hence, storing water underground into alluvial aquifers will potentially benefit the most efficient method practicable. Accordingly, this study focuses on alluvial a quifer s torage and r ecovery (ASR) with attention to physical and administrative considerations.

PAGE 12

3 Figure 1 : Map of South Platte River from Denver Metro to Liddle Ditch near Colorado Nebraska state line (SPDSS Alluvial Groundwater Model Report CDM Smith 2 013).

PAGE 13

4 In terms of physical considerations, the alluvium must be able to store useful quantities of water for subsequent use, the system must be located adjacent to recharge water and near water consumers, and adequate hydraulic conductivity must be maint ained. Fortunately, a study has been performed to evaluate the alluvium and an estimated 12 k m 3 ( 10 million ac ft ) of stora ble void space physically exist s in the South Platte basin ( HDR Engineering and West Sage Water Consultants 2015) Comparing it with existing 9.5 km 3 (7.7 million ac ft) of storage available in dams and reservoirs across Colorado from Division of Water Resources alluvial ASRs could potentially reduce the need to use up vast open spaces to construct new dams. In terms of legal consider ations, alluvial ASR depends on availability of water timing, quantity, and quality most states in the Western U.S., Colorado water law is based upon the recognition that the right to use decreed wate r, either surface water or groundwater, is property of the owner and therefore protected by law from injury (McIntyre and Mays 2017) Injury occurs when a decreed water right cannot be diverted in accordance with its decree, d ue to the actions of another. Other water right holders have an opportunity to oppose a water rights application, if they feel injury may occur. This paper deals with alluvial aquifer storage and recovery and how it was applied to a case study along the South Platte River, about 40 km ( 2 5 mi ) north east of D enver, the capital of Colorado, near the city of Brighton with a storage capacity of 148,000 m 3 (120 ac ft) The case study involved site selection, review of available soils data, modeling of water recharge rate and extraction rat es under varying levels of soil matrix clogging and determining the optimal configuration of storage inflow and outflow

PAGE 14

5 piping. In addition, and critical to the operation of such a storage vessel, a detailed study of legally available surface water was pe rformed. This was necessary since a junior water storage right, which this storage facility would be granted in the water court, would not be a dependable storage structure. Chapter 2 looks at alternative water storage facilities, Chapter 3 will discuss about the technical and administrative feasibility of such system, and Chapter 4 will summarize the study and provide recommendations for further research.

PAGE 15

6 CHAPTER II ALLUVIAL AQUIFER STORAGE AND RECOVERY Owing to various problems with above the ground surface dams and reservoirs, s ubsur face dams are now widely used for water storage ( Ishida et al. 2011 ). Topper et al ( 2010 ) discuss artificially recharge methods to introduce water to, store water in, underlying aquifer system during wet season, when th e river carries excess water The stored water can be later pumped out for use, either drinking or irrigation, during dry season. Primary objectives of artificial recharge are; manage water supply, meet legal obligations (augment water to meet downstream w ater rights or facilitate compliance with interstate agreements), manage/ mitigate water quality, restore/ protect aquifers (restore groundwater levels, limit aquifer compaction, mitigate saltwater intrusion), and protect the environment (maintain wetland hydrology, control migration of groundwater contamination). Artificial rec harge of aquifers is achieved b y natural/ enhanced recharge through surface flooding on infiltration basins and well injection. Alluvial aquifer storage and recovery is one way of a rtificially storing water, by the use of injection/pumping well systems to inject water directly into the receiving aquifer, for future recovery at the same location (Topper et al. 2010). Hanson and Nilsson (1986) classified u nderground dams into two type s: sand storage dams, where the sand is behind the above ground dam, and those that are constructed below ground to stop the flow of a natural aquifer. Alluvial ASR has been a central component of water storage internationally. Australia has a concentrated effort in maximizing storage of water underground due to reoccurring droughts (Simmons 2014). The literature provides evidence for a wide application in

PAGE 16

7 Australia and North America Australia is one of the pioneers to integrate alluvial ASR into their urb an stormwater harvest program. Peak surface runoffs are diverted into wetlands and settling ponds for primary treatment and finally infiltrate into underground aquifers. In the rural areas, alluvial ASRs are used to combat drought scenarios. Florida, Orego n and Texas are some of the states in the U.S.A., which have adopted aquifer storage and recovery systems to tackle changing weather conditions and increasing water demand because of population growth. Brown et al. (2006) synthesized lessons learned from 5 0 ASR projects worldwide. Their key findings were as follows: First, well clogging, including air binding, is problematic. However, a potential solution is to incorporate regular back flushing programs. Second, water quality can diminish the usefulness of ASR, particularly when arsenic, iron, manganese or other metals can be released from the local geologic materials. Third, hydraulic analysis is important when evaluating multi well clusters in order to avoid interference with other wells. And fourth, alluv ial ASR systems need to incorporate monitoring equipment, such as sampling ports on recharge or discharge lines, to allow real time monitoring of specific conductivity or turbidity. To recapitulate, alluvial ASR is not a new concept, in either the United States or elsewhere. Depending on site location, soil conditions, and project design, there are various levels of operational success. Each potential project site requires a detailed analysis to include influent water quality, soil chemistry and physical p roperties, and intended use of the recovered water. In the era of climate change and increasing population, alluvial ASR offers a viable alternative to surface reservoirs. Alluvial ASR costs are less than above ground water storage (dams and reservoirs) an d water loss due

PAGE 17

8 to the evaporation is significantly decreased. Finally, in areas with limited resources to fund construction of traditional above ground water storage facilities or areas with limited technical expertise, alluvial ASR provides an alternati ve water storage vessel. With climate change anticipating shifting hot seasons and precipitation patterns in the western United States, Colorado is well suited, due to physical stream alluvial properties, its water law and administration, increased populat ion predictions, to aggressively implement alluvial ASR as a water storage strategy. This following chapter presents a case study for a theoretical ASR facility located adjacent to the South Platte River in northeastern Colorado, paying due attention to the relevant physical constraints ( i.e. well clogging, water quality, hydraulic analysis, and monitoring) and administrative constraints ( i.e. appropriation).

PAGE 18

9 CHAPTER III CASE STUDY ON SOUTH PLATTE RIVER ALLUVIAL ASR A cas e study was developed to evaluate the application of alluvial ASR in Colorado. Colorado has water right to only one third of the total water available within its territories, while remaining two third of the water sent downstream to neighboring states as p art of various river compacts as dictated by Colorado Water Law This has put a lot of stress on water administrators and commissioners to efficiently collect, distribute and manage this precious and life sustaining element. Topics presented in this sectio n include background information, facility design, technical feasibility, and administrative feasibility Each of these topics will be presented in turn. Background Information A study by the Col orado Water Conservation Board ( 2007 ) assessed the volume of potential alluvial storage in the South Platte River basin (Table 1 ). The criteria for site selection were based on available alluvium and the distance between the site and the Denver metropolitan area (Denver Metro). Therefore, the area selected for deta iled study was along the main stem of the South Platte River between the nor theast portion of Denver Metro and Greeley, Colorado (SP Metro to Greeley). Along this reach, a candidate site was identified on the east bank of the South Platte River immediately upstream of the bridge crossing Colorado Highway 7 in Brighton, Colorado (Figure 2 ). wind speeds, which would predict high evaporative rates for above ground water s torage reservoirs. Brighton site takes advantage of evaporative loss savings in addition to the relative proximity to Denver Metro.

PAGE 19

10 Table 1 : Available storage capacity in South Platte River Basin alluvium (CWCB, 2007) Subregion Subregion Name Available S torage Capacity cubic meter (acre foot) Number 1 SP Denver Metro 435,000,000 (353,000) 2 SP Metro to Greeley 208,000,000 (169,000) 3 Cache la Poudre River 359,000,000 (291,000) 4 Upper Beebe/Box Elder 331,000,000 (268,000) 5 Lower Beebe/Box Elder 75,200,000 (61,000) 6 SP Greeley to Ft. Morgan 116,000,000 (94,000) 7 Upper Lost Creek 1,550,000,000 (1,260,000) 8 Lower Lost Creek 194,000,000 (157,000) 9 Upper Kiowa Creek 289,000,000 (234,000) 10 Lower Kiowa Creek 994,000,000 (806,00 0) 11 Upper Bijou Creek 575,000,000 (466,000) 12 Lower Bijou Creek 1,320,000,000 (1,067,000) 13 Badger/Beaver Ck. 384,000,000 (311,000) 14 SP Ft. Morgan Area 1,190,000,000 (968,000) 15 SP Balzak to State Line 1,100,000,000 (890,000) 16 SP South Park 1,110,000,000 (899,000) SP South Park data from Topper et al, 2004.

PAGE 20

11 Figure 2 : Selected alluvial site at Brighton, CO. 3959'07.25" N 10449'49.96" W (Google Earth)

PAGE 21

12 Facility Design Design of this alluvial ASR is based on a patent by McIntyre and Rens (2016). Water is diverted from the river and into the alluvial ASR by a reverse spillway (Figure C), to divert water from the river into smaller canals and networks powe red by gravity. Since peak flows typically satisfy all decreed water rights, any water available above a certain stage will result in free river, the diversion will comply with Colorado water laws. Should a portion of stored water be considered out of prio rity, then this water can be measured and pumped back into the stream. A key component is the stream diversion structure because contained within the structure, is a solids removal treatment system. The patent portrays alluvial ASR as viable, if, solids ar e removed prior to water injection in order to maintain water quality in the facility. A design for solids removal is not contained within the scope of this research effort, other to alert, if not properly designed and constructed, an efficiently operating ASR facility is problematic. A field visit to the case study site in Brighton, CO. at the start of this research work assisted in designing the dimensions of an alluvial ASR facility. The facility, designed underground on alluvial land beside the river, will have an areal extent denoted by the box in Figure 2 and an assumed depth to store water. T he Brighton alluvial site is 195 m (639.8 ft) wide, 380 m (1246.7 ft) long, and 10 m (32.8 ft) deep with an assumed porosity of 20%. These dimensions equate to approximately 148, 0 00 m 3 (120 ac.ft) of storage volume. The ASR vessel will be enclosed by an impermeable slurry wall, keyed into bedrock, required to contain and control the surface water inflow. This could be constructed, for example, from interlocking s heet piles or from a bentonite slurry placed

PAGE 22

13 in an excavated trench, the latter of which is assumed to purposes of cost estimation Figure 4 shows the underground view of alluvial ASR confined by slurry wall on four sides. Soils information was provided by the U.S. Soil Conservation Service (1974) and the U.S. Natural Resources Conservation Service (2014). The three types of soils in and near the case study site are sandy alluvial land, loamy alluvial land with gravelly substratum, and wet alluvial land. Al l three soil types are indicative of gravelly soils which makes the selected site a good alluvial ASR case study. Pumping of water in and out of the alluvial ASR is achieved by a set of 8 wells that symmetrically recharge and extract water. These wells ar e designed to be 9.5 m deep (Figure 5 ), with perforated pipes connecting them to form a rectangular network as shown in Figure 6 Perforated pipes aid in pumping out water without reducing the head to a small amount in the bottom layer so that dry cells do not affect the performance of the pumps. Recharging water into the alluvial ASR is achieved with by the gravitational forces which carry water from a river upstream at a higher elevation and into this alluvial ASR at a lower elevation. The network of wel ls and perforated pipes carry water into the soil matrix, which stores water in pore spaces between soil particles. While alluvial ASR recharge can be facilitated by gravity, working out against it calls for mechanical pumps that are powered by electricity Hence, the major energy expense, in operating this alluvial ASR facility, comes from running those 8 pumps. Additional use of instruments, to test water quality, soil matrix porosity, water turbidity, and water level, which assists in the automation of operating this alluvial ASR facility would result in extra energy and cost.

PAGE 23

14 Figure 3 : Gravity powdered recharging of alluvial ASR (McIntyre et al. 2016) Figure 4 : Confinement using Slurry walls/ sheet piling to setup a static alluvial ASR syst em.

PAGE 24

15 Figure 5 : 3D model of modeled alluvial ASR system showing wells at different locations in layer 2. Figure 6 : Perforated pipe assembly in layer 2, 9.5 m below surface level.

PAGE 25

16 Technical Feasibility The study by CDM Smith (2013) for Colo rado Water Conservation Board analyzed samples from the South Platte River alluvium, reporting median values for hydraulic conductivity, K = 130 m /d (425.4 ft/d), transmissivity T = 1300 m 2 /d ( 14,000 ft 2 /d), and specific yield S y = 0.2 These values giv e a rough estimate of the median saturated thickness of b = T / K = 10 m ( 32.8 ft). The filling constraint is the maximum infiltration rate permitted by the in situ soils on site since the vessel is below ground and the intent is no evaporation component ass ociated with any of the inflow quantity. This aspect of the case study is examined via a 3 dimensional groundwater model described below. A groundwater simulation model, developed with MODFLOW (Harbaugh and Barlow 200 6 ), was constructed to evaluate the t echnical feasibility of the alluvial ASR facility described above. The MODFLOW model assumes a two layered alluvial aquifer with varying transmissivity and dimensions as discussed under Facility Design above The first layer from top surface is 9.5m (31 ft ) thick contains the wells and perforated pipe system. The bottom layer, below first layer, is 0.5m (1.6 ft) thick and is assumed to be saturated with water all the time to prevent running into dry cells at the time of simulation During MODFLOW simulation s for water recharge and extraction unsaturated condition implies there is 1 .5m depth of water, and saturation condition implies 9.5 m depth of water is totally available Therefore, at any given time of operation, the useable level of water from this fac ility is 8 m, which translates to a useable storage volume of 118,500 m 3 (~96 ac ft) Recharging and pump out simulations which uses Modified Incomplete

PAGE 26

1 7 Cholesky and the Polynomial methods to check convergence c onditions in each cell thereby preventing dry cell conditions when water is being pumped out through the wells. The model was used to answer the following questions: First, what is the operating protocol or algorithm to store and extract water without s urface ponding or running into dry cells? Second, how many days are required to fill/empty the facility as a function of pumping rate? And third, to account for clogging, how do results depend on reduced hydraulic conductivity? What is the optimal design c onfiguration for the manifold of injection extraction wells ? P umping algorithm for this alluvial ASR w as designed with a conservative a pproach that there is at least a minimum flow of 2500 m 3 /day (1 cfs) in the river. This algorithm ensures that alluvial ASR is not over filled during recharge phase, and does not pump out too much water to cause dry cells in the soil matrix around extraction bores. Appendix A of this report contains the algorithm for operating this alluvial aquifer storage and recovery faci lity. Under no clogging conditions (K= 130 m/day), this alluvial ASR facility could be filled up from 1.5 m to 9.5 m in abou t six and a half days, and this 8 m (26 ft) depth of water could be extracted in nine days using pumps These results were obtain ed under an assumption that not more than 2500 m 3 of water could be removed from the river on any single day. The question of clogging is a critical design and operational consideration in any proposed alluvial ASR project. The alluvial ASR site at Brig hton, Colorado has an assumed unclogged hydraulic conductivity of K = 130 m/d (42 6 .5 ft/d) Several MODFLOW simulations were generated by varying the value of K to simulate clogging

PAGE 27

18 by setting K = { K K /2, K /4, K /8} Figure 7 and Figure 8 respectively show the time taken to completely recharge and empty the alluvial ASR under various levels of clogging. The change in hydraulic results were minimal and did not change any of the assumed initial design parameters such as diversion rates, pipe diameter, or pum p out requirements. A pretreatment structure before pumping water into the alluvial ASR will remove suspended solids and other chemical colloids that clog up the soil matrix. Repeated filling and emptying of the alluvial ASR causes varying effective and po re pressure in the soil matrix, causing soil consolidation. Figure 9 is a graph of varying total stress in the soil matrix. Knowing the pore pressure and total stress for varying levels of water level saturation, overall soil consolidation in this ASR faci lity was theoretically calculated to be only 0.75 cm (0.025 ft). This negligible amount of consolidation does not have any impact on geological conditions on the land where the alluvial ASR facility is constructed. Appendix B shows the calculations to comp ute total soil matrix consolidation in the alluvial ASR case study site.

PAGE 28

19 Figure 7 : Time taken to saturate the alluvial ASR under varying levels of clogging. 0 1 2 3 4 5 6 7 8 9 10 0 4 8 12 16 20 24 Head in the ASR [m] Time [d] K K/2 K/4 K/8 Initial head Final head

PAGE 29

20 Figure 8 : Time taken to pump out all usable water from the alluvial ASR under varying levels of clogging. 0 1 2 3 4 5 6 7 8 9 10 0 4 8 12 16 20 24 Head in the ASR [m] Time [d] K K/2 K/4 K/8 Initial head Final head

PAGE 30

21 Figure 9 : Total soil consolidation at the case study site.

PAGE 31

22 Administrative Feasibility Water law in Colorado is based on the doctrine of prior appropriation ( Hobbs, 2004 ; McIntyre and Mays, 2017 ) which grants water rights in the ord er in which they were put to beneficial use, with accounting for expected return flows via surface or groundwater. When flow is insufficient to meet demand, those holding senior water rights my place a call on the river to guarantee delivery of their appr opriation. Accordingly, feasibility of the alluvial ASR design depends not only on technical feasibility, as discussed above, but also on administrative feasibility. For surface water diversions, Colorado water law establishes a date of priority decreed by the water court and administered by Colorado Division of Water Resources (CDWR). The dates of priority, with the most senior being first, are typically listed on a tabulation of water rights that the CDWR references when establishing who on a stretch of s tream or river, is permitted to divert and by how much. The South Platte River is over appropriated, meaning there is seldom sufficient water to satisfy all decreed surface water rights ( HDR Engineering and West Sage Water Consultants 2015 ) For a new wa ter storage structure to be able to divert legally, must first secure a court decree with a priority date. The alluvial ASR facility at Brighton would be granted a junior priority date, meaning more senior water rights would take priority. Therefore, an an alysis of the nature of the flow in the river at Brighton is required to access when and if the vessel can fill. Two administrative classifications of surface water are available to fill the alluvial ASR facility: (1) free river, and (2) fully reusable eff luent water. Each of these classifica tions will be discussed in turn. Free river occurs when flow remains in the river even after all decreed water rights are diverting their full entitlements without

PAGE 32

23 reduction. The excess water in the river is termed free river and is available to even the most junior decreed priority. To identify the specific time periods of free river, an analysis was performed to determine, based upon 1 4 years of flow data WY (200 0 to WY 2013 ), which periods of flow were classified fr ee river at the Brighton site and therefore available for diversion, even by the most junior water right priority. Free River Analysis Results for Periods at Brighton between 20 00 and 201 3 The framework and reference for this study is based on the South Platte Implementation Plan ( HDR Engineering and West Sage Water Consultants 2015 ) Appendix G of this report provides the basic information to perform a Point Flow Analysis at the Henderson Gage of the South Platte River with an aim to calculate the quanti tative availability of free river in a Water Year (Oct. 1 st to Sept. 30 th ) from 2000 to 2013. Flow through the Henderson Gage is roughly the same as flow on South Platte River at the Brighton case study site, since no water diversion of augmentation is pre sent between the two places. Original stream flow data along with daily call chronology was downloaded from South Platte Basin Implementation Plan report compiled by HDR Engineering and West Sage Water Consultants (2015). Free River analysis was carried o ut using an RStudio program. Appendix C contains the lines of code written in RS tudio to calculate free r iver availability at Henderson Gage. Call Chronolog y Appendix G of South Platte Basin Implementation Plan by HDR Engineering and West Sage Water Consul tants ( 2015 ) dictates that a call placed anywhere downstream

PAGE 33

24 from the point of analysis implies absence of free river conditions in the river on that day. Hence a call anywhere from Henderson gage to Liddle Ditch implies a call on the river. 0 denotes no r ecorded call for a day, while 1 indicates presence of a call placed at one or many downstream gages/ditches from Henderson Gage. Call Chronology is calculated from the Point Flow Spreadsheet 01/16/15.xlsx containing call records for each gage/ditch on the South Platte River. If a senior water right holder does not obtain allocated supply, then is a call is placed anywhere from Denver metro to Liddle Ditch. Water is diverted to this senior water right holder before supplying to other. River Compact Deal at Liddle Ditch South Platte River Compact requires a minimum flow of 120 cfs to be sent across the state border to Nebraska between April 1 and October 15. A Compact Call is placed on the river whenever physical flow at Liddle Ditch is below 120 cfs, to indi cat e absence of free river condition. However, the Water Availability Model from South Platte River Metro Roundtable ha ve not applied Compact Deal condition while computing annual free river availability. Water Availability Analysis Daily amount of Free R iver available at Henderson Gage is a nalyzed using Point Flow method which states that t otal amount of free river available at Henderson gage is the minimum flow among the total flow across all the downstream ditches and gages from Henderson Gage to Liddl e Ditch, minus the diversions

PAGE 34

25 Reusable Effluent Water The second administrative classification of surface water available to fill the alluvial ASR facility is fully reusable effluent water which is water supply from non tributary sources. Non tributa ry sources are those that would not be present in the river in the absence of engineering intervention, such as tunnels under the continental divide that provide water from the western slope of the Rocky Mountains, or wells extracting non tributary groundw ater. At the Brighton case study ASR site, the following water providers have fully reusable effluent water that passes the site: Arvada, Aurora, Brighton, Denver, Ft. Lupton, Thornton, and Westminster. Broomfield and Northglenn discharge their reusable ef fluent into Big Dry Creek, which enters the South Platte River downstream of Brighton; however, this water could be exchanged (move point of diversion upstream) up to the Brighton ASR site to fill the vessel. Under Colorado water law, municipal water pro viders are entitled to capture their fully reusable effluent water for additional use. The Colorado Division of Water Resources maintains an accounting of reusable effluent flows as submitted by the individual municipal water providers. The yearly average of available fully reusable effluent water, assuming permission from the municipality is granted, to store in the alluvial ASR facility is found in Table 2 Table 2 shows t hat there is sufficient fully reusable effluent water at the alluvial ASR site to fi ll and refill the vessel multiple times each month of the year. However, permission must first be granted by the municipal entity to store their water in the vessel. Water effluent from Denver and Aurora's wastewater treatment facilities belong to the resp ective cities under Water Rights Law. Some of the reusable effluent water are

PAGE 35

26 sold to utilities by the City of Denver and Aurora. Remaining reusable effluent water, which is a portion of physical flow in the South Platte, can be reclaimed for use in the fu ture, and hence it should not be accounted for in free river analysis. S outh P latte B asin I mplementation P lan Water Availability Model contains daily unused reusable effluent water from Denver and Aurora in the Unused Reusable Flow

PAGE 36

27 Table 2 : Yearly a verage of unused reusable effluent flow from Denver and Aurora Water Year Unused reusable flow [m 3 /s ] Unused reusable flow [cfs] 2000 18.6 658.0 2001 42.5 1502.3 2002 50.6 1789.3 2003 31.4 1109.3 2004 18.7 661.1 2005 21.1 745.2 2006 19.4 683.8 2007 0.0 0.0 2008 16.6 586.2 2009 0.0 0.0 2010 21.1 745.4 2011 28.1 991.1 2012 26.7 943.7 2013 42.9 1517.5

PAGE 37

28 Total amount of Free River available at Henderson Gage Actual water available minus the unused effluent flow gives total free river at Henderson Gage. Total amount of Free River available m 3 ( ac ft ) is calculated by summing daily amount of free river for each Water Year (WY) f rom 2000 to 2013 Figure 10 is the graph free river available daily, and Figure 11 is the graph of a nnual free river available each year. Water years 2002 and 2003 were completely dr y with no free river available, while water years 2000 and 2009 had lot of free river and call diversions were minimum Figure 12 highligh ts log of monthly mean, minimum and ma ximum amount of available free river over those 14 water years. April, May and June have more number of days with free river since spring melt adds signif icantly high amount of flow to the river. By end of summer free river conditions are fewer and more calls are placed on the river. January and February have least amount of free river availability. Time series correlation analysis showed a weak positive re lationship betw een the water years. Climate variability and stream flow modifications can be hypothesized to result in a poor R 2 value (Figure 13 ). Results obtained by the free river analysis using R Studio was validated by comparing it with the water avai lability model for Henderson gage from South Platte Basin Implementation Basin Surface Water Availability Model (HDR Engineering and West Sage Water Consultants 2015). Except for water year 2013, total free river availability calculated from water year 200 0 2012 closely matched with the validation data. In the water year 2013 September flood was believed to be neglected in the free analysis by HDR engineering and West Sage Water Consultants during the while this research work considered all the natural ev ents contributing to river flow variability.

PAGE 38

29 Figure 10 : Daily summary of free river available from Water Year (WY) 2000 WY 2013.

PAGE 39

30 Figure 11 : Annual free river available from WY 2000 WY 2013.

PAGE 40

31 Figure 12 : Monthly mean, minimum and maximum available free river from water year 2000 to 2013

PAGE 41

32 Figure 13 : Time series correlation analysis of Free River available from WY 2000 to WY 2013 y = 2.35x 4,662.14 R = 0.03 0 50 100 150 200 250 2000 2002 2004 2006 2008 2010 2012 2014 Free River available [m 3 ] WY 2000 to WY 2013 water available Linear fitted line

PAGE 42

33 CHAPTER 4 DISCUSSION AND CONCLUSION Based on the literature review of existing alluvial ASR projects in the US and wo rldwide, the MODFLOW simulations, and the legal availability of surface water in the South Platte River at the Brighton, Colorado ASR case study site, alluvial A SR is a viable water storage option in the South Platte River basin in Colorado. However, pretr eatment of the diverted South Platte River surface wat er is required to ensure water quality and prevent clogging of soil matrix by silt and other particles carried down by the river. Free river analysis from water year 2000 to 2013 provide proof that free river conditions exist for an average of 10 3 days per year, with an average flow of 2.15 m 3 /s (76 cfs) and satisfies the recharging rate need ed to fill the alluvial ASR facility. However, to setup the perforated pipe system, for the purpose of faster wate r extraction from the alluvial ASR, the ground would first have to be dug out, install the pipes and then repack the facility with soil. Construction cost of an alluvial ASR site was beyond the scope of this research, hence future work should compare the c ost analysis of building an alluvial ASR versus an above ground dam or reservoir. In Colorado, the property nature of water would constrain a junior priority ASR these two classes of water, alluvial ASR is a viable water storage methodology at the case study site near Brighton, Colorado. However, the potential of clogging could temper wide use of ASR elsewhere in Colorado. Because of this, site selection including detailed evaluation of in situ materials, is a critical consideration. In addition, the existing water rights in a given location must also be evaluated.

PAGE 43

34 REFERENCES Ground Water Vol. 44 (6), 771 774. Conference Proceedings Southern Illinois University, Carbondale, IL (Mar. 5, 2017). Report April 2013. The Denver Post (Mar. 6, 2017). water supplies in developing c Ground Water 24(4), 497 506. South Platte Basin Roundtable, April 17, 2015. HDR Engineering and West Sage Water Consultants (2015). Appendix G South Platte Basin Surface Water Availability Analysis South Platte Basin Roundtable, March 16 2015. Japan Agricultural Research Quart erly Vol. 45 (1), 51 61 (2011). NOAA (Mar. 5, 2017). Ma, Z., Jia element J. Tradit. Chin. Med. 34(1), 115 121. er Court and the State Water Policy, in press. McIntyre, W. C. and Rens, K. L. (2016). "System And Method Of Using Differential To The Regents of University of Colorado, a body corporate, Denver, CO (USA) U.S. Patent 9,278,808 B1, March 28, 2016.

PAGE 44

35 of Ground Water in Colorado Environment Geology 13, Colorado Geological Survey, Denver, CO, 2004. The Sydney Morning Herald (Mar. 5, 2017). Colorado Water Conservation Board, January 2010.

PAGE 45

36 APPENDIX A ALGORITHM TO RECHARGE ALLUVIAL ASR

PAGE 46

37 APPENDIX B SOIL MATRIX CONSOLIDATION FULLY WET CONDITION Layer Layer thickness [m] Elevat ion,z [m] Head [m] Stress[N/m 2 ] Total stress [N/m 2 ] @ h =10m Pore Pressure [N/m 2 ]@ h=10m Eff. Stress [N/m 2 ] @ h=10m 0 0 10 10 0 0 0 0 1 0.5 9.5 10 12262.5 12262.5 4905 7357.5 2 0.5 9 10 12262.5 24525 9810 14715 3 0.5 8.5 10 12262.5 36787.5 14715 22072.5 4 0.5 8 10 12262.5 49050 19620 29430 5 0.5 7.5 10 12262.5 61312.5 24525 36787.5 6 0.5 7 10 12262.5 73575 29430 44145 7 0.5 6.5 10 12262.5 85837.5 34335 51502.5 8 0.5 6 10 12262.5 98100 39240 58860 9 0.5 5.5 10 12262.5 110362. 5 44145 66217.5 10 0.5 5 10 12262.5 122625 49050 73575 11 0.5 4.5 10 12262.5 134887. 5 53955 80932.5 12 0.5 4 10 12262.5 147150 58860 88290 13 0.5 3.5 10 12262.5 159412. 5 63765 95647.5 14 0.5 3 10 12262.5 171675 68670 103005 15 0.5 2.5 10 12262.5 183937. 5 73575 110362.5 16 0.5 2 10 12262.5 196200 78480 117720 17 0.5 1.5 10 12262.5 208462. 5 83385 125077.5 18 0.5 1 10 12262.5 220725 88290 132435 19 0.5 0.5 10 12262.5 232987. 5 93195 139792.5 20 0.5 0 10 12262.5 245250 98100 147150

PAGE 47

38 DRY CONDITION Head [m] Stress[N/m 2 ] Total stress [N/m 2 ] @ h=0.5m Pore Pressure [N/m 2 ] @ h=0.5m Eff. Stress [N/m 2 ] @ h=0.5m 10 0 0 0 0 9.5 11281.5 11281.5 0 11281.5 9 11281.5 22563 0 22563 8.5 11281.5 33844.5 0 33844.5 8 11281.5 45126 0 45126 7.5 11281.5 56407.5 0 56407.5 7 11281.5 67689 0 67689 6.5 11281.5 78970.5 0 78970.5 6 11281.5 90252 0 90252 5.5 11281.5 101533.5 0 101533.5 5 11281.5 112815 0 112815 4.5 11281.5 124096.5 0 124096.5 4 11281.5 135378 0 135378 3.5 11281.5 146659.5 0 146659.5 3 11281.5 157941 0 157941 2.5 11281.5 169222.5 0 169222.5 2 11281.5 180504 0 180504 1.5 11281.5 191785.5 0 191785.5 1 11281.5 203067 0 203067 0.5 11281.5 214348.5 0 214348.5 0.5 12262.5 226611 4905 221706 Dry Bul k Density = 2300 kg/m 3 Wet bulk density = Dry bulk density + (Soil porosity Density of water) Wet Bulk Density = 2500 kg/m 3 Soil compressibility, 08 m 2 /N Change in effective stress = dry condition wet condition = 221706 147150 [N/m 2 ] ve = 74556 N/m 2 o = ve = b o ve 0.0075 m 0.75 cm Soil matrix compr esses by: 0.75 cm

PAGE 48

39 APPENDIX C RSTUDIO CODE FOR FREE RIVER ANALYSIS D ownload all spreadsheets from supplemental materials : ## codes to process the raw data from south platte metro roundtable ## ## set directory where the spreadsheet and RStudio code is present ## setwd("C:/Users/ ") install.packages("data.table") install.packages("ggplot2") # for creating graphs install.packages("psych") install.packages("car") install.packages("sm") install.packages("lsr") install.packages("reshape") install.p ackages("scales") library(scales) # to access breaks/formatting functions library(data.table) library(psych) library(car) library(lsr) library(ggplot2) library(reshape) library(sm) # River data with flow from "At the structure" # rawdata< read.csv("Poin t Flow.csv",header = TRUE) river< cbind(rawdata$X,rawdata[,seq(2,ncol(rawdata),2)]) # from column 2, alternatively choose columns to be used setnames(river,"rawdata$X","Date") river[,1] < as.Date(river[,1], format = "%m %d %Y") r=nrow(river) # compact at liddle ditch for (i in 1:r){ if (river[i,62] <= 120) { river[i,62] < 0 } else { river[i,62] < river[i,62] } } View(river)

PAGE 49

40 # call data rawcall< read.csv("Call.csv",header = TRUE) call < rawcall[seq(10,nrow(rawcall),1),] ro w.names(call) < 1:nrow(call) # reset serial no to 1,2,3,.. setnames(call,"X","Date") call[is.na(call)] < 0 #set all n/a to 0, representing no call for (i in 1:r) { river[i,64]< max(call[i,11:63]) #call is placed on the river if any structure below Henderson raise a call } setnames(river,"V64","Call") # Un u sed Reusable Flow rawunused< read.csv("Unused Reusable Flow.csv", header=TRUE) river[,65] < rawunused[277:5390,12] setnames(river,"V65","Unused") river[,65]< as.numeric(as.character(river[,65]) ) # Gross free river: min flow from Henderson to Liddle Ditch and sum of channels, corresponding to no call days for(i in 1:r){ if (river[i,64]==1){ river[i,66] < 0 #if call=1, no free river exists } else { river[i,66] < min(riv er[i,11:63]) #else if call=0, potential free river = min flow in SP } } setnames(river,"V66","Gross free river") #Final free river available everyday for (i in 1:r){ river[i,67]= river[i,66] river[i,65] if (river[i,67]<0) { river[i,67] < 0 } else { river[i,67] < river[i,67] } } setnames(river,"V67","Freeriver") river[,68] < river[,67]*0.0283 setnames(river,"V68","Freeriver [m3/sec]") river[,69] < river[,67]*1.9835 setnames(river,"V69", "Freeriver [ac.ft/day]")

PAGE 50

41 ## read climate data ## climate < read.csv("monthlytempnprecip.csv", header = TRUE) climate [1,6] = sum(river$Freeriver[2:32]) climate [2,6] < sum(river$Freeriver[33:62]) climate [3,6] < sum(river$Freeriver[63:93]) climate [4,6] < sum(river$Freeriver[94:124]) climate [5,6 ] < sum(river$Freeriver[125:153]) climate [6,6] < sum(river$Freeriver[154:184]) climate [7,6] < sum(river$Freeriver[185:214]) climate [8,6] < sum(river$Freeriver[215:245]) climate [9,6] < sum(river$Freeriver[246:275]) climate [10,6] < sum(river$Freer iver[276:306]) climate [11,6] < sum(river$Freeriver[307:337]) climate [12,6] < sum(river$Freeriver[338:367]) climate [13,6] < sum(river$Freeriver[368:398]) climate [14,6] < sum(river$Freeriver[399:428]) climate [15,6] < sum(river$Freeriver[429:459]) c limate [16,6] < sum(river$Freeriver[460:490]) climate [17,6] < sum(river$Freeriver[491:518]) climate [18,6] < sum(river$Freeriver[519:549]) climate [19,6] < sum(river$Freeriver[550:579]) climate [20,6] < sum(river$Freeriver[580:610]) climate [21,6] < sum(river$Freeriver[611:640]) climate [22,6] < sum(river$Freeriver[641:671]) climate [23,6] < sum(river$Freeriver[672:702]) climate [24,6] < sum(river$Freeriver[703:732]) climate [25,6] < sum(river$Freeriver[733:763]) climate [26,6] < sum(river$Freer iver[764:793]) climate [27,6] < sum(river$Freeriver[794:824]) climate [28,6] < sum(river$Freeriver[825:855]) climate [29,6] < sum(river$Freeriver[856:883]) climate[30,6] < sum(river$Freeriver[884:914]) climate[31,6] < sum(river$Freeriver[915:944]) cli mate[32,6] < sum(river$Freeriver[945:975]) climate[33,6] < sum(river$Freeriver[976:1005]) climate[34,6] < sum(river$Freeriver[1006:1036]) climate[35,6] < sum(river$Freeriver[1037:1067]) climate[36,6] < sum(river$Freeriver[1068:1097]) climate[37,6] < sum(river$Freeriver[1098:1128]) climate[38,6] < sum(river$Freeriver[1129:1158]) climate[39,6] < sum(river$Freeriver[1159:1189]) climate[40,6] < sum(river$Freeriver[1190:1220]) climate[41,6] < sum(river$Freeriver[1221:1248]) climate[42,6] < sum(river$F reeriver[1249:1279]) climate[43,6] < sum(river$Freeriver[1280:1309])

PAGE 51

42 climate[44,6] < sum(river$Freeriver[1310:1340]) climate[45,6] < sum(river$Freeriver[1341:1370]) climate[46,6] < sum(river$Freeriver[1371:1401]) climate[47,6] < sum(river$Freeriver[14 02:1432]) climate[48,6] < sum(river$Freeriver[1433:1462]) climate[49,6] < sum(river$Freeriver[1463:1493]) climate[50,6] < sum(river$Freeriver[1494:1523]) climate[51,6] < sum(river$Freeriver[1524:1554]) climate[52,6] < sum(river$Freeriver[1555:1585]) c limate[53,6] < sum(river$Freeriver[1586:1614]) climate[54,6] < sum(river$Freeriver[1615:1645]) climate[55,6] < sum(river$Freeriver[1646:1675]) climate[56,6] < sum(river$Freeriver[1676:1706]) climate[57,6] < sum(river$Freeriver[1707:1736]) climate[58,6 ] < sum(river$Freeriver[1737:1767]) climate[59,6] < sum(river$Freeriver[1768:1798]) climate[60,6] < sum(river$Freeriver[1799:1828]) climate[61,6] < sum(river$Freeriver[1829:1859]) climate[62,6] < sum(river$Freeriver[1860:1889]) climate[63,6] < sum(ri ver$Freeriver[1890:1920]) climate[64,6] < sum(river$Freeriver[1921:1951]) climate[65,6] < sum(river$Freeriver[1952:1979]) climate[66,6] < sum(river$Freeriver[1980:2010]) climate[67,6] < sum(river$Freeriver[2011:2040]) climate[68,6] < sum(river$Freeriv er[2041:2071]) climate[69,6] < sum(river$Freeriver[2072:2101]) climate[70,6] < sum(river$Freeriver[2102:2132]) climate[71,6] < sum(river$Freeriver[2133:2163]) climate[72,6] < sum(river$Freeriver[2164:2193]) climate[73,6] < sum(river$Freeriver[2194:222 4]) climate[74,6] < sum(river$Freeriver[2225:2254]) climate[75,6] < sum(river$Freeriver[2255:2285]) climate[76,6] < sum(river$Freeriver[2286:2316]) climate[77,6] < sum(river$Freeriver[2317:2344]) climate[78,6] < sum(river$Freeriver[2345:2375]) climate [79,6] < sum(river$Freeriver[2376:2405]) climate[80,6] < sum(river$Freeriver[2406:2436]) climate[81,6] < sum(river$Freeriver[2437:2466]) climate[82,6] < sum(river$Freeriver[2467:2497]) climate[83,6] < sum(river$Freeriver[2498:2528]) climate[84,6] < s um(river$Freeriver[2529:2558]) climate[85,6] < sum(river$Freeriver[2559:2589]) climate[86,6] < sum(river$Freeriver[2590:2619]) climate[87,6] < sum(river$Freeriver[2620:2650]) climate[88,6] < sum(river$Freeriver[2651:2681]) climate[89,6] < sum(river$Fr eeriver[2682:2709])

PAGE 52

43 climate[90,6] < sum(river$Freeriver[2710:2740]) climate[91,6] < sum(river$Freeriver[2741:2770]) climate[92,6] < sum(river$Freeriver[2771:2801]) climate[93,6] < sum(river$Freeriver[2802:2831]) climate[94,6] < sum(river$Freeriver[283 2:2862]) climate[95,6] < sum(river$Freeriver[2863:2893]) climate[96,6] < sum(river$Freeriver[2894:2923]) climate[97,6] < sum(river$Freeriver[2924:2954]) climate[98,6] < sum(river$Freeriver[2955:2984]) climate[99,6] < sum(river$Freeriver[2985:3015]) cl imate[100,6] < sum(river$Freeriver[3016:3046]) climate[101,6] < sum(river$Freeriver[3047:3075]) climate[102,6] < sum(river$Freeriver[3076:3106]) climate[103,6] < sum(river$Freeriver[3107:3136]) climate[104,6] < sum(river$Freeriver[3137:3167]) climate[ 105,6] < sum(river$Freeriver[3168:3197]) climate[106,6] < sum(river$Freeriver[3198:3228]) climate[107,6] < sum(river$Freeriver[3229:3259]) climate[108,6] < sum(river$Freeriver[3260:3289]) climate[109,6] < sum(river$Freeriver[3290:3320]) climate[110,6] < sum(river$Freeriver[3321:3350]) climate[111,6] < sum(river$Freeriver[3351:3381]) climate[112,6] < sum(river$Freeriver[3382:3412]) climate[113,6] < sum(river$Freeriver[3413:3440]) climate[114,6] < sum(river$Freeriver[3441:3471]) climate[115,6] < su m(river$Freeriver[3472:3501]) climate[116,6] < sum(river$Freeriver[3502:3532]) climate[117,6] < sum(river$Freeriver[3533:3562]) climate[118,6] < sum(river$Freeriver[3563:3593]) climate[119,6] < sum(river$Freeriver[3594:3624]) climate[120,6] < sum(rive r$Freeriver[3625:3654]) climate[121,6] < sum(river$Freeriver[3655:3685]) climate[122,6] < sum(river$Freeriver[3686:3715]) climate[123,6] < sum(river$Freeriver[3716:3746]) climate[124,6] < sum(river$Freeriver[3747:3777]) climate[125,6] < sum(river$Free river[3778:3805]) climate[126,6] < sum(river$Freeriver[3806:3836]) climate[127,6] < sum(river$Freeriver[3837:3866]) climate[128,6] < sum(river$Freeriver[3867:3897]) climate[129,6] < sum(river$Freeriver[3898:3927]) climate[130,6] < sum(river$Freeriver[ 3928:3958]) climate[131,6] < sum(river$Freeriver[3959:3989]) climate[132,6] < sum(river$Freeriver[3990:4019]) climate[133,6] < sum(river$Freeriver[4020:4050]) climate[134,6] < sum(river$Freeriver[4051:4080]) climate[135,6] < sum(river$Freeriver[4081:4 111])

PAGE 53

44 climate[136,6] < sum(river$Freeriver[4112:4142]) climate[137,6] < sum(river$Freeriver[4143:4170]) climate[138,6] < sum(river$Freeriver[4171:4201]) climate[139,6] < sum(river$Freeriver[4202:4231]) climate[140,6] < sum(river$Freeriver[4232:4262]) climate[141,6] < sum(river$Freeriver[4263:4292]) climate[142,6] < sum(river$Freeriver[4293:4323]) climate[143,6] < sum(river$Freeriver[4324:4354]) climate[144,6] < sum(river$Freeriver[4355:4384]) climate[145,6] < sum(river$Freeriver[4385:4415]) climat e[146,6] < sum(river$Freeriver[4416:4445]) climate[147,6] < sum(river$Freeriver[4446:4476]) climate[148,6] < sum(river$Freeriver[4477:4507]) climate[149,6] < sum(river$Freeriver[4508:4536]) climate[150,6] < sum(river$Freeriver[4537:4567]) climate[151, 6] < sum(river$Freeriver[4568:4597]) climate[152,6] < sum(river$Freeriver[4598:4628]) climate[153,6] < sum(river$Freeriver[4629:4658]) climate[154,6] < sum(river$Freeriver[4659:4689]) climate[155,6] < sum(river$Freeriver[4690:4720]) climate[156,6] < sum(river$Freeriver[4721:4750]) climate[157,6] < sum(river$Freeriver[4751:4781]) climate[158,6] < sum(river$Freeriver[4782:4811]) climate[159,6] < sum(river$Freeriver[4812:4842]) climate[160,6] < sum(river$Freeriver[4843:4873]) climate[161,6] < sum(ri ver$Freeriver[4874:4901]) climate[162,6] < sum(river$Freeriver[4902:4932]) climate[163,6] < sum(river$Freeriver[4933:4962]) climate[164,6] < sum(river$Freeriver[4963:4993]) climate[165,6] < sum(river$Freeriver[4994:5023]) climate[166,6] < sum(river$Fr eeriver[5024:5054]) climate[167,6] < sum(river$Freeriver[5055:5085]) climate[168,6] < sum(river$Freeriver[5086:5114]) # rename the columns setnames(climate,"V6","Total_free_river_cfs") climate[,7]=climate[,6]*0.0283 setnames(climate,"V7","Total_free_r iver_m3/sec") climate[,8]=climate[,6]*1.9835/1000 setnames(climate,"V8","Total_free_river_10^3 acft") View(climate) ## Annual free river availability dataframe ## water_available = data.frame(matrix(NA,nrow=14,ncol=2))

PAGE 54

45 water_available[,1]=c("WY2000","WY2 001","WY2002","WY2003","WY2004","WY20 05","WY2006","WY2007","WY2008","WY2009","WY2010","WY2011","WY2012"," WY2013") water_available[1,2]=sum(climate[1:12,6]) water_available[2,2]=sum(climate[13:24,6]) water_available[3,2]=sum(climate[25:36,6]) water_availabl e[4,2]=sum(climate[37:48,6]) water_available[5,2]=sum(climate[49:60,6]) water_available[6,2]=sum(climate[61:72,6]) water_available[7,2]=sum(climate[73:84,6]) water_available[8,2]=sum(climate[85:96,6]) water_available[9,2]=sum(climate[97:108,6]) water_avail able[10,2]=sum(climate[109:120,6]) water_available[11,2]=sum(climate[121:132,6]) water_available[12,2]=sum(climate[133:144,6]) water_available[13,2]=sum(climate[145:156,6]) water_available[14,2]=sum(climate[157:168,6]) setnames(water_available,"X1","Water _Year") setnames(water_available,"X2","Annual_free_river_cfs") water_available[,3]= round((water_available[,2]*0.0283)) water_available[,4]=round((water_available[,2]*1.9835/1000), digits=1) setnames(water_available,"V3","Annual_free_river_m3") setnames(wa ter_available,"V4","Annual_free_river_10^3 acft") ## Use free river availability data from South Platte Implementation Plan report for validation ## water_available[,5] = c(141.1,45.7,3.4,0,0,29.2,.9,46.6,4.7,104.4,194.8,84.7,39,0) setnames(water_availabl e,"V5","validation_data_10^3 acft") View(water_available) #Monthly free river availability (box plot) ggplot(climate, aes(x=climate[,1], y=climate[,7]))+ geom_boxplot() + scale_y_log10() + xlab("Months") + ylab("Water quantity \ n [m3/sec]") + scale_x_disc rete(limits=c("January","February","March","April","May","June","July","A ugust","September","October","November","December")) + theme_bw() + theme(axis.line = element_line(colour = "black"), panel.grid.major = element_blank(), panel.gr id.minor = element_blank(), panel.background = element_blank()) # Daily summary of free river availability ggplot(river, aes(x=river[,1], y=river[,68]))+ geom_line() + xlab("Water Year \ n [10/1 to 09/30]") + ylab("Free river available [m3/sec]") + scale_x_date(breaks=date_breaks("1 years"),labels=date_format("%Y")) +

PAGE 55

46 theme_bw() + theme(axis.line = element_line(colour = "black"), panel.grid.major = element_blank(), panel.grid.minor = element_blank(), panel.background = element_blank()) #Annual free river availability (box plot) ggplot(climate, aes(x=climate[,2], y=climate[,7]))+ geom_boxplot() + xlab("Water Year \ n [10/1 to 09/30]") + ylab("Water quantity \ n [cubic meters]") + theme_bw() + theme(axis.line = eleme nt_line(colour = "black"), panel.grid.major = element_blank(), panel.grid.minor = element_blank(), panel.background = element_blank()) #Annual free river availability (bar plot) ggplot(water_available, aes(x=water_available[,1], y =water_available[,3])) + geom_bar(stat = "identity") + xlab("Water Year \ n [10/1 to 09/30]") + ylab("Water quantity \ n [cubic meters]") + theme_bw() + theme(axis.line = element_line(colour = "black"), panel.grid.major = element_blank(), panel.grid.minor = element_blank(), panel.background = element_blank()) # Correlation test to prove the accurac y of free river analysis with RS tudio ## cor.test(water_available[,4],water_available[,5])