Citation
Spatial and temporal variability of fire regimes across the biodiverse Klamath-Siskiyou ecoregion, Northern California and Southern Oregon, USA

Material Information

Title:
Spatial and temporal variability of fire regimes across the biodiverse Klamath-Siskiyou ecoregion, Northern California and Southern Oregon, USA
Creator:
Morton, Shelley J.
Place of Publication:
Denver, CO
Publisher:
University of Colorado Denver
Publication Date:
Language:
English

Thesis/Dissertation Information

Degree:
Master's ( Master of science)
Degree Grantor:
University of Colorado Denver
Degree Divisions:
Department of Geography and Environmental Sciences, CU Denver
Degree Disciplines:
Environmental sciences
Committee Chair:
Briles, Christy E.
Committee Members:
Anthamatten, Peter
Moreno, Rafael

Notes

Abstract:
The Klamath-Siskiyou Ecoregion in northern California and southern Oregon is a biodiversity hotspot where future climate change and fire activity are significant concerns. The historical range of variability of fire has been difficult to examine in this region due to the generally low temporal resolution of historical documents and temporally restricted proxy records like those from tree rings. Paleoenvironmental records from lake sediments, based on proxies of pollen and macroscopic charcoal, spanning the last 5000 years were analyzed to temporally expand knowledge of the relationships between fire activity, climate change, and forests. Reconstruction of these histories enables quantification of fire frequency, biomass burned and fire severity along elevational, latitudinal and coastal-to-inland gradients. The results of the study indicate that fire activity was less frequent and more severe at wetter northern and coastal sites, while it was frequent and less severe at drier southern and inland locations. The Klamath-Siskiyou Ecoregion collectively experienced higher fire activity and severity during the Medieval Climate Anomaly when conditions were warmer and drier. These trends were exaggerated at southern and inland sites that received less precipitation than northern and coastal sites. During the Little Ice Age when conditions were cooler and wetter, fewer and less severe fires occurred, especially at northern and coastal sites that receive more precipitation. Current forest and fire conditions are a legacy of the LIA, and as conditions become warmer, and effectively drier, the region will likely experience higher fire activity and severity. Combining these local and regional data sets allows for the creation of a temporal and spatial depiction of how fire regimes, and more specifically fire severity, has changed. This information has important implications for ecological and forest management goals in northern California and the Northwest coast of the United States

Record Information

Source Institution:
University of Colorado Denver
Holding Location:
Auraria Library
Rights Management:
Copyright Shelley J. Morton. Permission granted to University of Colorado Denver to digitize and display this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.

Downloads

This item has the following downloads:


Full Text
SPATIAL AND TEMPORAL VARIABILITY OF FIRE REGIMES ACROSS THE BIODIVERSE
KLAMATH-SISKIYOU ECOREGION, NORTHERN CALIFORNIA AND SOUTHERN
OREGON, USA
by
SHELLEY J. MORTON
B.S., Metropolitan State University Denver, 2017
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 Environmental Sciences Program
2017


11
This thesis for the Master of Science degree by Shelley J. Morton has been approved for the
Environmental Sciences program
by
Christy E Briles, Chair Peter Anthamatten Rafael Moreno
Date: December 16, 2017


Morton, Shelley J. (M.S., Environmental Sciences Program)
Spatial and Temporal Variability of Fire Regimes across the Biodiverse Klamath-Siskiyou Ecoregion, Northern California and Southern Oregon.
Thesis directed by Assistant Professor Christy E Briles Ph.D.
ABSTRACT
The Klamath-Siskiyou Ecoregion in northern California and southern Oregon is a biodiversity hotspot where future climate change and fire activity are significant concerns. The historical range of variability of fire has been difficult to examine in this region due to the generally low temporal resolution of historical documents and temporally restricted proxy records like those from tree rings. Paleoenvironmental records from lake sediments, based on proxies of pollen and macroscopic charcoal, spanning the last 5000 years were analyzed to temporally expand knowledge of the relationships between fire activity, climate change, and forests. Reconstruction of these histories enables quantification of fire frequency, biomass burned and fire severity along elevational, latitudinal and coastal-to-inland gradients. The results of the study indicate that fire activity was less frequent and more severe at wetter northern and coastal sites, while it was frequent and less severe at drier southern and inland locations. The Klamath-Siskiyou Ecoregion collectively experienced higher fire activity and severity during the Medieval Climate Anomaly when conditions were warmer and drier. These trends were exaggerated at southern and inland sites that received less precipitation than northern and coastal sites. During the Little Ice Age when conditions were cooler and wetter, fewer and less severe fires occurred, especially at northern and coastal sites that receive more precipitation. Current forest and fire conditions are a legacy of the LIA, and as conditions become warmer, and effectively drier, the region will likely experience higher fire activity and severity. Combining these local and regional data sets allows for the creation of a temporal and spatial depiction of how fire regimes, and more specifically fire severity, has changed. This information has important


implications for ecological and forest management goals in northern California and the Northwest coast of the United States
IV
The form and content of this abstract are approved. I recommend its publication.
Approved: Christy Briles


V
DEDICATION
I would like to dedicate this thesis to my loving partner. His support and understanding were essential to my success and I am eternally grateful. I would also like to thank my family for instilling a deep love of science and respect for the environment at an early age, and my friends for their encouragement and support.


VI
ACKNOWLEDGEMENTS
I would like to thank my advisor, Christy Briles for her guidance, time, insight, and for introducing me to the world of paleoecology. Her dedication to scientific research has always astounded me and I hope to be as devoted as her in my future endeavors. In addition, I would like to thank my other committee members, Peter Anthamatten and Rafael Moreno for their encouragement and support throughout my entire scholastic career at UC Denver.


vii
TABLE OF CONTENTS
CHAPTER
I. INTRODUCTION......................................................................1
II. BACKGROUND INFORMATION...........................................................4
Physical Environment...........................................................4
Climate and Vegetation.........................................................7
Modem Climate...............................................................7
Modem Forests...............................................................8
Paleoclimate and historical vegetation......................................9
Abmpt climate fluctuations..................................................9
Disturbance...................................................................10
Fire Regimes...............................................................10
Increasing Fire Activity...................................................12
Regional Trends in Biomass Burned..........................................14
III. METHODS AND DATA ANALYSIS.....................................................15
Site Descriptions.............................................................15
Oak Woodland Zone..........................................................16
Mixed Conifer Zone.........................................................16
White Fir Zone.............................................................17
Red Fir Zone...............................................................17
Faboratory Methods............................................................19
Charcoal...................................................................20
Pollen.....................................................................20
Data Analysis.................................................................20
Chronology.................................................................20
Charcoal Analysis..........................................................21
Determining Fire Severity..................................................22
Regional Biomass Burned Reconstmction......................................23
IV. RESUFTS........................................................................27
Chronology....................................................................27
Vegetation Zone Fire History Reconstructions..................................28
Oak Woodlands Zone.........................................................28


viii
Mixed Conifer Zone..........................................................32
White Fir Zone..............................................................33
Red Fir Zone................................................................35
Biomass Burned Reconstruction..................................................37
5,000-year biomass burned trends in the Klamath-Siskiyou Ecoregion..........37
Biomass Burned Trends in the Klamath-Siskiyou Ecoregion and Northwestern US.39
V. DISCUSSION.......................................................................40
Influences on Fire Variability.................................................46
Oak Woodland Vegetation Zone................................................46
Mixed Conifer Vegetation Zone...............................................47
White and Red Fir Vegetation Zones..........................................48
Regional Biomass Burned Patterns and Controls..................................49
Spatial Variability of Fire in the Klamath-Siskiyou Ecoregion and Implications for Fire Management.....................................................................52
Future research................................................................54
VI. CONCLUSIONS......................................................................56
REFERENCES...........................................................................58
APPENDIX
A. ANALYSISES PERFORMED.............................................................65
B. CHRONOLOGY FOR ALL SITES.........................................................66
C. POLLEN PERCENTAGES AND CANOPY-TO- UNDERSTORY RATIOS
67


IX
LIST OF TABLES
TABLE
1. Summary of Site Descriptions............................................................18
2. Pollen types and associated fire sensitivity used in this study.........................23
3. Location, elevations, and citation for the sites used to create regional biomass burned.26
4. Radiocarbon dates used to create Kelly and Miller Lake's age models.....................28
5. Fire-related sample ratio comparisons to baseline means, and fire type determinations...35


X
LIST OF FIGURES
FIGURE
1. Klamath-Siskiyou Ecoregion...............................................................6
2. Site Map................................................................................16
3. Vegetation zones and site locations. Dashed lines represent plant species distributions.19
4a. CHAR, peak magnitude, fire events, fire frequency and pollen ratios of oak woodland and mixed
conifer vegetation zones.................................................................30
4b. CHAR, peak magnitude, fire events, fire frequency and pollen ratios for White and Red Fir
vegetation zones.........................................................................32
5a. Biomass burned curves for northern vs southern sites....................................38
5b. Biomass burned curves for coastal vs inland sites......................................38
5c. Klamath-Siskiyou Ecoregion and Northwest US coast trends in biomass burned..............39
Figure 6: Scatter plot comparing charcoal values for BCHAR and peak magnitude............41
Figure 7: Biomass burned, fire events, temperature, climatic fluctuations and inferred vegetation ....46


1
CHAPTER I INTRODUCTION
The Klamath-Siskiyou Ecoregion is a rugged series of mountain ranges, straddling the California - Oregon border. It is an ecologically and geographically unique region of almost 100 million acres. It is one of seven areas of global botanical significance in North America determined by the International Union of Conservation of Nature, is considered a biodiversity hotspot, and has been the focus of biogeographic research for many decades. The region’s complex topography, steep coastal-to inland precipitation gradients, and mixed-severity fire regimes encourage high levels of biodiversity and fire activity is considered essential in maintaining biodiversity (Martin and Sapsis, 1992; Bond and van Wilgen, 1996). With the region largely wildlands and forested areas, and fire being an integral disturbance element, the change in fire activity in response to future climate change and anthropogenic fire management is a significant concern (Briles et al. 2011; Crawford et al., 2015, Miller et al., 2009; Odion and Hanson, 2006).
The historical range of variability of fire has been difficult to examine in the region due to the generally low temporal resolution of historical documents, and short-term records like those from tree rings, satellite, and forestry records. The reconstruction of fire severity has yielded mixed results, leading to debates between fire researchers about whether fire severity is increasing or simply returning to historical conditions (Baker, 2015; Odion and Hansen, 2006; Safford et al., 2007; Westerling, et al., 2006). This past summer, the Klamath Mountains experienced several large fires (e.g., the Salmon-August Complex in the Marble Mountains, the Eclipse Complex in the Siskiyou Mountains, and the Orleans Complex on the western edge of the Marble Mountains between the other two fire complexes) contributing -80,000 hectares to the estimated 570,000 hectares burned in California, Oregon, Washington, and Montana. Questions around whether these and other fires in the past few decades are unprecedented, and whether they are the new normal under projected climate change are being raised within forest management, academic, and local western US communities.


2
Paleoenvironmental records of pollen and charcoal preserved in lake sediments allow for the examination of relationships between fire activity, climate change, and forests. Reconstruction of these histories enables quantification of fire frequency, biomass burned, fire severity and forest changes in response to past climate variability that span millennia. The Klamath Mountains have a sizable number of paleoecological records. The focus of past paleoenvironmental studies have been to examine how vegetation and fire activity have responded on different substrates to climate variations since the last glacial period (-15,000 years) (Briles et al., 2011), to access regional and local controls on postglacial forest development and fire regimes (Briles et al., 2008) and determine late-Holocene human land-use practices on forests and fire regimes at low elevations (Crawford et al., 2015). In addition, several individual site studies have documented the Holocene paleoenvironmental history of the Klamath and Siskiyou Mountains (Colombaroli and Gavin, 2010; Daniels et al, 2005; Briles et al., 2005; Mohr et al., 2000). These records present an opportunity to evaluate how local and regional fire regimes, and more specifically fire severity, have fluctuated during the last 5,000 years across the heterogenous region. A regional paleoenvironmental study that evaluates fire regime variability, and the controls on fire, has the potential to inform current United States Forest Service (USFS) adaptive management plans that recognize the importance of fire in protecting and enhancing “Old Growth” forest habitat and the biological diversity (Kaufmann et al. 2007; Olson et al., 2012)
The main objective of this research is to obtain a temporal and spatial understanding of fire regimes within the Klamath-Siskiyou Ecoregion and compare them with other published reconstructions in the western US. The study utilizes pollen and macroscopic charcoal proxy records from nine sites to explore how fire regimes respond to changes in vegetation and climate. A composite analysis utalizing the nine sites, and other sites in the Global Charcoal Database (an online database of charcoal data), are combined along elevational, latitudinal, and a costal-to-inland gradient, based on modem precipitation and vegetation characteristics, and examined for changes in fire activity, biomass burned, and fire severity for the last 5000 years.


3
The purpose of this study is to address the following questions:
(1) What is the spatial and temporal variability of fire in the Klamath-Siskiyou Ecoregion during the last 5000 years?
(2) How has fire severity changed during the last 5000 years in the Klamath-Siskiyou Ecoregion?
The thesis is composed of six chapters. Chapter two provides a background on the physical characteristics of the region, including climate, geology and fire regimes that maintain the current biodiversity of the Klamath-Siskiyou Ecoregion. Chapter two examines the disturbance regimes within the region, particularly that of fire, along with a discussion of regional debate on fire severity. Chapter three describes the nine sites used in this study, outlines the field and laboratory procedures of data collection and analyses performed on the lake sediment cores, and the modeling and statistical analyses performed on the radiocarbon, pollen and charcoal data. Chapter four presents the results of the age-depth modeling and charcoal analysis, including composite analyses to reconstruct biomass burned and fire severity estimates using charcoal and pollen data from individual sites. Chapter five discusses the spatial variability of fire in the Klamath-Siskiyou Ecoregion along environmental gradients and what we can glean about fire disturbance regimes in the region. It also addresses how results of the study can inform forest management strategies and lend to the ongoing fire debate in the region. Chapter six summarizes the findings of the study.


4
CHAPTER II
BACKGROUND INFORMATION Physical Environment
Today the Klamath-Siskiyou Mountain Ecoregion extends from the northern Sacramento Valley in northern California to the Umpqua Valley in southern Oregon and lies to the east of the California Coast Range (Figure 1). Within this ecoregion reside the Klamath Mountains and the many subranges that make up the mountainous complex, including the Siskiyou, Marble, Trinity, Scotts, Salmon, Russian, Trinity, and Yolla-Bolly mountains, which support a wide variety of vegetation, including 39 species and subspecies of conifers, seven of which are endemic (found nowhere else) to the region (DellaSala et al., 1999).
The topographic relief in the Klamath Mountains has a large effect on species distributions. Steep elevational gradients vary from sea level to 2900 meters over a small geographic area creating one of the steepest coastal-to-inland precipitation gradients in North America (Franklin and Dymess, 1988; Whittaker, 1960). The rugged topography also creates pronounced differences in precipitation, humidity, and temperature on small scales, usually less than one kilometer (Dobrowski et al., 2011). These complexities affect the distribution of species and encourage contrasting biotic communities to thrive near one another. Besides the west-to-east climate gradient, the region is a transition zone between the drier Mediterranean climate to the south and the wetter Pacific Northwest Temperate Rainforest climate to the north. This results in many plant species either originating or terminating their ecologic ranges in the Klamath Mountains.
Geology also influences plant diversity in the Klamath Mountains. Ultramafic soils, derived from serpentine and peridotite bedrock, have high concentrations of heavy metals and are low in nitrogen, phosphorus, and calcium. Consequently, they are inhabited by endemic xerophytic plants that have evolved ways of concentrating the metals and toxins in their plant tissues or restricting their uptake (Alexander, 2007). The forests on these ultramafic substrates are open due to the nutrient


5
restrictions and support a wide diversity of plants (Kruckeburg, 2002; Alexander, 2007). Other soil types in the region, derived from a range of metasedimentary and igneous rocks, do not have the same
nutrient restrictions and support different plants, but are of lower diversity than ultramafic substrates. These more fertile soils yield more closed forests than their ultramafic counterparts. The type of substrate greatly affects present plant communities, as well as plant response to abrupt climate changes, such as those seen in the Little Ice Age cooling (LIA, -500-100 cal yr B.P; Grove, 2001) and Medieval Climate Anomaly (MCA, -1400-900 cal yr BP; Graham et al, 2007). For example, lake records suggest little vegetation response to these climate events on ultramafic substrates, in contrast to rapid responses on other soil types. Fire responded in concert with climate change on all soil types, except when ultramafic substrates became too open during the MCA to support fire spread (Briles, 2017; Briles et al., 2011).
How vegetation will respond to future climatic variations is a point of debate. Predictions are difficult to make for such complex landscapes, but forecasting these responses are necessary to maintain the current biodiversity (Harrison et al., 2010). The lack of high-resolution temporal records in the Pacific Northwest of response to climatic change is noted in several studies (Harrison et al., 2014, Walsh, et al. 2010). For example, Harrison et al. (2010) note that the Klamath-Siskiyou region, with its steep precipitation and temperature gradients, combined with its unique and endemic flora, provides an ideal location in which to study the impacts that changing climates may have on future biodiverse temperate communities.


6
124°W
RoseBurg
[Medford
Klamath Falls
'Ashland
|l23°W
t ' â– 
r
1122°W
m


43 N
41°N
â–¡ KlamathEcoregion
â–¡ National Forest
0 25 50 75 kill
Figure 1: Klamath-Siskiyou Ecoregion


7
Climate and Vegetation
Modem Climate
The modem climate of the northwest US coast can be generally characterized by warm dry summers and cool wet winters, where fluctuations in the Pacific subtropical high- and the Aleutian low-pressure system, and the position of the jet stream (westerlies), influence precipitation on multiple temporal scales. In the summer, a strengthened and expanded high-pressure system develops off the coast that results in warm conditions with little to no precipitation. In winter a strong low-pressure system pushes the westerlies south and brings Pacific storms and snow to the mountains. However, due to steep environmental lapse rates in summer, caused by surface warming, variable atmospheric conditions often result in conventional storms that produce lightning that start wildfires. Ocean currents also influence the climate, especially along the coast. Today, the California Current is associated with strong upwelling that brings fog and cool moist conditions to the coastal area and mountains. The westernmost Klamath Mountains, including forests and fire regimes, have been influenced by the fog for millennia (Briles et al., 2008).
Large-scale interannual-to-decadal climatic oscillations have also helped to shape the current landscape in this region. The El Nino-Southern Oscillation (ENSO) causes fluctuations in ocean temperatures along the equator influencing the trajectory of storms to the Klamath region and has a cycle of 6 to 18 months. The El Nino phase of ENSO results in warm ocean conditions along the equatorial eastern Pacific Ocean which causes warmer and drier conditions in the Pacific Northwest and results in reduced snowpack in the winter and spring. The La Nina phase, or cool ocean conditions along the equatorial eastern Pacific Ocean, causes cooler and wetter condition. The pattern of variability in North Pacific sea surface temperatures, or the Pacific Decadal Oscillation (PDO) is another climatic phenomenon that can impact oceanic precipitation with phases that can last up to 30 years. During the positive phase of the PDO, sea surface temperatures (SSTs) increase along the west coast of North America and the Aleutian low pressure system intensifies resulting in warmer


8
and drier winter conditions in the northwestern North America. Cooler and wetter winters are associated with cooler SSTs and a reduced Aleutian low pressure system, and the negative phase of the PDO (Mantua & Hare, 2002; Newman et al., 2016). The combination the positive phase of the PDO and the El Nino phase of ENSO extended the fire season especially for inland sites within the Pacific Northwest (Gershunov et al., 1999; Hessl et al, 2004).
Modem Forests
Elevational changes in precipitation and temperature influence plant distributions in the Klamath Mountains (Figure 2 &3). At the highest elevations above 2300 m, the Mountain Hemlock Zone forests are dominated by Tsuga mertensiana (mountain hemlock), Abies magnified (red fir) and several species of Finns (pine) and conditions are cold and wet. Fires, although less frequent, are generally of higher intensity and severity than other zones, due to the wet conditions that allow significant fuel buildup, resulting intense and severe fires, with most species in the zone easily killed by fire (Taylor and Skinner, 1998, 2003). In the Red Fir Zone (2300 to 1900 m), Tsuga mertensiana dissipates with decreasing elevation and a more diverse assemblage is present in the forest. Abies magnified, Abies concolor (White fir), Pinus contorta (lodgepole pine) and Pinus monticola (western white pine) are the dominant canopy species. Snowpacks remain high and cool summers allow moisture to persist through much of the year (Franklin and Dymess, 1988). The White Fir Zone lies between 1900-1300 m elevation, and Abies concolor generally forms nearly pure stands, but Pseudotsuga menziesii (Douglas fir), Pinus lambertiana (sugar pine), Pinus ponderosa (ponderosa pine) and Pinus monticola are also present in more minor abundances within this zone. Significant snowpack still occurs in the higher elevations of this zone, and fires occur more regularly than at higher elevations (Agee, 1993; Franklin and Dymess, 1988). The Mixed Conifer Zone, from 1300-1100 m elevation, consists of P. monticola, Pseudotsuga menziesii, P. lambertiana, P. ponderosa, Calocedrus decurrens (California incense cedar), and Abies concolor. There is a gradual warming moving into the lower zones with lower snow packs and an earlier melt in the spring, resulting in


9
progressively drier conditions at lower elevations. The Mixed Evergreen Zone, between 1100-800 m elevations, has a mixture of Pseudotsuga menziesii, Pinus ponderosa, Pinus jeffreyi (Jeffrey pine), Calocedrus decurrens, and Lithocarpus densiflora (tanbark oak) and are more open. Fire is more frequent at lower elevations and are generally of low severity (Skinner and Taylor, 1998). The lowest elevations, 800m and lower, consist of oak woodland (Quercus spp.) and receive little if any snow, tend to be very dry in the summer months, and generally have low severity, high-frequency fire regimes (Taylor and Skinner 1998, Franklin and Dymess, 1988).
Paleoclimate and historical vegetation
The paleoclimate of the region was driven by long-term changes in incoming solar radiation and the strength and movement of the pressure systems and ocean currents. During the middle (8200 to 4500 cal yr B.P) and late Holocene (4500 cal yr B.P. to present), summer insolation declined throughout the western United States, which resulted in cooler summers with increased precipitation (Bartlein et al., 1998). Paleoecological data collected from the Klamath Mountains and throughout the Pacific Northwest, confirms that transition to a wetter and cooler climate began around 4500 cal yr BP, although vegetation and fire responses are not synchronous (Briles et al., 2005). Ocean upwelling increased between 5200 and 3500 cal yr BP creating wetter conditions and increased fog production, as reflected in the increase in cool water diatoms and the presence of redwood pollen in oceanic sediment cores (Barron et al., 2003). These cooler and wetter conditions lead to forest becoming more closed where Abies, Pseudotsuga, and Tsuga became more prominent (Briles et al., 2005). By 2000 cal yr B.P., the cooler and wetter conditions resulted in the establishment modem forest communities (Briles et al., 2011; Crawford et al, 2015).
Abrupt climate fluctuations
There have been two abrupt climate change periods identified within the late Holocene including the Fittle Ice Age (FIA, -500-100 cal yr B.P; Grove, 2001), where the climate was cooler


10
and wetter than today, and Medieval Climate Anomaly (MCA, -1400-900 cal yr BP; Graham et al, 2007), where the climate was warmer and drier. These events have been attributed to changes in sunspot activity, volcanism, and long-term changes in ocean-atmospheric dynamics that led to prolonged climate conditions that had noticeable impacts on ecosystems and fire regimes (Deng et al. 2007; Mann et al. 2009). Both short-term climate events had effects on the modem vegetation and fire regimes seen in this region. In the Klamath Mountains, the cool wet conditions during the LIA resulted in less charcoal influx into the lake system, increased biomass accumulation in the forests, and more shade-tolerant trees and shrubs. The warm conditions during the MCA led to increases in charcoal accumulations, fuels biomass in the forest decreased due to more frequent fires, and shade-intolerant species became more prominent (Briles et al., 2008 and 2017; White et al.,2015; Crawford et al. 2015).
Disturbance
Forest disturbance consists of an event that impacts the landscape or alters the physical structure of the vegetation. The effect of disturbances on the environment can last a few years to centuries. As climate changes, there are corresponding changes in the disturbance regime of an area (Westerling, 2006 and Minckley and Long, 2016). Disturbance within forest ecosystems can present in many forms (Baker, 1992). Pine beetles, large weather events, flooding, erosion, avalanches, landslides, and fire can change the current structure of the forest as well as how the forest responds to future disturbances (Fried et al, 2004 and Higuera et al. 2014).
Fire Regimes
A forest’s fire regime, or the pattern of fire occurrences through time, is the result of the composition of the surrounding vegetation, the amount of fuels present, time since the last disturbance, the type of anthropogenic management, and climate. Historical and current fire regimes can provide insights into how the vegetation within a forest will respond to climatic changes in the


11
future and the major controls driving those changes. Climate also influences the structure and composition of the vegetation, which is correlated with the severity, size, and frequency of fires. High-severity fire usually removes most aboveground vegetation and replaces entire stands. Mixed-severity regimes result in fires of varying severity and are the most common type (Sommers, 2011)
Lakes are repositories of paleoenvironmental data. During a fire event, and for some time after, ash and charred particles are deposited into the lake through airborne fallout. Streamflow, runoff, and other disturbance events (i.e. landslides) continue to introduce charcoal to the lake sediments for months to years after a fire event. This deposition of charcoal and other disturbance proxies, through time, allows for the reconstruction of the surrounding areas fire regime (Whitlock, 2004). Macroscopic charcoal (>125 microns) is a common proxy used to reconstruct fire activity and biomass burned. The mechanisms that introduce charcoal into the lake system can vary from one lake to another due to the size of the lake and surrounding topography.
Fire regimes are subject to many environmental drivers, but it is generally understood that the main drivers are fuel loads and climate. Fuel-driven regimes can be the result of the varying levels of biomass produced, but can also be the product of anthropogenic clearing of debris from forest floors and other suppression methods. Climate-driven regimes are more controlled by temperature and precipitation (Sommers, 2011). Northern California and southern Oregon forest have a mixture of fuels-driven and climate-driven fire regimes. In the forests closer to the ocean, especially at high elevations, climate limits fire activity even though there are sufficient fuels, due to high moisture content both in the atmosphere and in the fuels. At lower elevations and more inland open forests, fire regimes can be fuel-limited (Steel et al., 2015). In the western US, both rain and snowfall winter precipitation, are major contributor to annual precipitation (Mock, 1996). El Nino variability can have large impacts on winter precipitation and snowpack, especially in conjunction with the positive phase of the Pacific Decadal Oscillation (Taylor and Beaty, 2005).


12
Increasing Fire Activity
Due to anthropogenic climate change, the western US is experiencing increased fire activity and area burned per individual fire since the late 20th century (Baker 2015; Marlon et al., 2008; Westerling, 2006). Fire regime changes are attributed to lower than average winter precipitation coupled with snowmelt occurring earlier in the spring, and warmer and drier summers (Westerling, 2016). There is debate among fire researchers as to how drastically fire in the western US is changing in response to our current climate change. Westerling et al. (2006) reported that in the mid-eighties the wildfire season was extended and has continued to elongate since. Mid-elevation forests of the Northern Rockies were most impacted by the increased summer and spring temperatures and extended wildfire season. Areas that experience short snow-free seasons and high evapotranspiration are considered at the highest risk for increased wildfire activity and larger fires (Westerling et al.2006)
The status of high severity fires is under debate in the wester US. Odion and Hanson (2006) found that high severity fires in the Sierra Nevada Mountains have not been a common occurrence, and suggest introducing more fire into management strategies. In their 2014 article Hanson and Odion state that since 1984, modem high-severity fires have not increased in proportion, area, or patch size in western North America according to Burned Area Emergency Rehabilitation (BAER) fire-severity data. They suggest that climate change might not have as large of an impact on fire severity in the future as others have forecasted (Hanson and Odion, 2014). Baker (2015) suggests that, at least in dry forests, rates of high-severity fire are increasing over historical rates, but are not doing so at the exceptionally high rates that others have indicated (Baker, 2015). According to the majority of studies, high severity fires are increasing in frequency over historical rates (Colombaroli and Gavin, 2010; Higuera et al., 2014; Miller et al., 2009 and others). The increase in high severity fires has been partly attributed to anthropogenic fire suppression, which has led to the accumulation of fuels on the forest floor. The effects of fire suppression have changed the structure of many


13
landscapes since the mid-19th-century settlement of the western US (Nagel and Taylor, 2005). Historically, high-elevation forests in the Klamath-Siskiyou Ecoregion, that receive high amounts of precipitation, had infrequent but sever fires. These “wet” forests have not experienced an increase in severity, and are within their historical range of variability. High severity fires at drier low- and middle- elevation sites, that historically burned more frequently but at lower severity than wetter sites, are burning more severely compared to pre-Euro-American settlement rates (Miller et al. 2009). The rate at which high severity fires are increasing can change depending on forest composition. Stephens et al. (2013) found that if current climate models are accurate the planets forest ecosystems will experience large-scale changes in structure and composition in the next few decades (Stephens et al., 2013).
While most research indicates increasing severity, a limitation of most of the studies is that the data is based on current remotely sensed images, that only extend back to the 1980’s, forestry records that only extend back a century, or dendrochronology records that only extend back -500 years. These records do not examine prior to the LIA when climatic conditions were similar to conditions we are encountering today. A better understanding of the trends observed in fire severity might be gleaned from identifying high severity fire from paleoecological data. In addition, late-Holocene paleoclimate data are available for determining climate variability and can be used to compare with charcoal-based fire history reconstructions. Reconstruction severity from macroscopic charcoal records is in its infancy. Studies on relationships between charcoal peak height and area burned to suggest severity using a combination of dendrochronology and lake sediment data have been conducted (Whitlock et al., 2004; Higuera et al. 2011). Other studies have identified high severity fires using changes in pollen spectra following fire events and compared them to pollen spectra of fire-free periods (Minckley and Shriver 2011; Minckley and Long, 2016).


14
Regional Trends in Biomass Burned
Several studies have compiled lake sediment charcoal records from around the world to examine global-to-regional fire patterns and their drivers. Power et al. (2008) examined biomass burned across the globe and on a continental scale since the last glacial maximum. Their findings indicate that global biomass burned increased through time and this increase corresponds to increases in global summer insolation. Marlon et al. (2012) examined biomass burned throughout the last 3,000 years and determined that biomass burned has generally declined but in periods of warmer and drier conditions like during the MCA and within the last 100 years large peaks in biomass burned occurred. In the Pacific Northwest, Walsh et al. (2015) have reconstructed biomass burned through the Holocene. The late Holocene began with increasing biomass burned contrary to the cooler wetter climate during this time until 600 cal yr BP and has decreased since (Walsh et al., 2015). Very few studies examine controls on regional biomass burned, especially in mountainous environments or along steep environmental gradients.


15
CHAPTER III
METHODS AND DATA ANALYSIS Site Descriptions
The sites used in this study were selected because field methods used to collect sediment cores and data were similar, in addition to the sites geographical locations within the Klamath-Siskiyou Ecoregion. Each lake is described by its physical attributes as well as the attributes of the surrounding forest. All these descriptions can be referenced in Table 1. Several other sites had appropriate charcoal records, but were excluded from the study due to their ultramafic soils, size, or lack of available pollen data. However, charcoal records from ultramafic sites from the Global Charcoal Database (Cedar, Bluff, and Crater) were included in the composite analyses as Briles et al. (2012) found that fire on ultramafic and other soils showed similar responses to climate. Squaw Lake (Colombaroli and Gavin, 2010) was left out due to its large size, inflowing stream, and large charcoal catchment area. All sites discussed here were relatively small is size (<17 ha) and have minimal or
lack stream inputs.


16
Figure 2: Site Map. Location of lakes (blue circles) used in the Klamath-Siskiyou Ecoregion with average annual precipitation from 1981-2010 (PRISM Climate Group, 2014). Cities are identified with black hexagons.
Oak Woodland Zone
Fish Lake is the lowest and westernmost site in the study area and lies within the Six Rivers National Forest. Ogaromtoc Lake is within the Klamath National Forest. The climate at both lakes consists of warm summers and mild winters and precipitation comes mainly in the form of fall and winter rain. Alnus, Cupressaceae Pseudotsuga menziesii, Lithocarpus, and Pinas pondersoa are the dominant trees making up the open forest around the lakes (Crawford et al., 2015)
Mixed Conifer Zone
Kelly Lake is the more coastal of the sites in the mixed conifer zone and lies in the Klamath
National Forest. The climate consists of warm summers and mild winters. Precipitation comes in


17
both rain and snow. The closed forests around Kelly Lake consist mainly of Abies concolor, Pinus lambertiana, and Pseudotsuga menziesii.
Hobart Lake is the easternmost site in this study and lies in the Cascade-Siskiyou National Monument. The climate consists of mild summers and cool winters. Precipitation comes in both rain and snow, more than half of which falls in winter. The forests around Hobart Lake are more open than sites at the same or higher elevations and Pinus ponderosa, Abies concolor, and Pseudotsuga menziesii are the dominant forest species. (White et al. 2015).
White Fir Zone
Sanger Lake is in the Six Rivers National Forest, Miller Lake lies within the Rogue River National Forest, and Bolan Lake is just north of the Califomia-Oregon border within the Siskiyou National Forest. Campbell Lake is the southernmost site in the zone in the Klamath National Forest and is in the Marble Mountain Wilderness.
The summers at these sites are generally mild and winters are cool and wet, with more than half of the annual precipitation in the form of winter snows. The forests are around these lakes closes as elevation increases, and the dominant trees consist of Abies concolor, Abies magnifica, Picea breweriana (Brewer spruce), Pinus monticola, and Pseudotsuga menziesii (Briles et al., 2008; Briles etal., 2011).
Red Fir Zone
Taylor Lake is the highest in elevation of all the study sites and lies within the Klamath National Forest and Russian Wilderness Area. The climate consists of cool summers and cold winters with more than half the annual precipitation in the form of snow. The closed forest around Taylor Lake consists of Abies concolor, Abies magnifica, Pinus monticola, and Tsuga mertensiana (Briles et
al, 2008).


18
Table 1: Summary of Site Descriptions
Name County, State Lat, Long Elev. (m) Size(ha)/ Depth (m) Precip.* (mm) Temp* (°C; min, max) Reference
Fish Lake Humbolt, CA 41° 15.813' N, 123° 41.014'W 552 9.8/13 1541 6,21 Crawford et al. (2015)
Ogaromtoc Lake Siskiyou, CA 41° 29.167' N, 123°32.474'W 594 1.5/6.3 1529 5,21 Crawford et al. (2015)
Kelly Lake Siskiyou, CA 41°54.795'N, 123°31.040'W 1328 4/5.17 1813 5, 15 This study-
Hobart Lake Jackson, OR 42° 5.866'N, 122° 28.848'W 1463 5.3/3.61 765 4, 14 White et al. (2015)
Sanger Lake Del Norte, CA 41° 54.106' N, 123°38.840'W 1558 2.5/7.19 2647 3, 14 Briles et al. (2008, 2011)
Miller Lake Josephine, OR 42° 3.848'N, 123° 18.183' W 1585 2.5/7.70 1590 2, 12 This study-
Bolan Lake Josephine, OR 42° 1.331' N, 123° 27.594'W 1640 5/11.25 1697 2, 12 Briles et al. (2008, 2011)
Campbell Lake Siskiyou, CA 41° 31.998' N, 123° 6.333'W 1756 17/7.12 1509 2,14 Briles et al. (2011)
Taylor Lake Siskiyou, CA 41° 21.674' N, 122°58.102'W 1981 6.5/9.45 1351 2, 11 Briles et al. (2011)
* 30 yr average precipitation and temperature from 1981-2010 (PRISM Climate Group, 2014)


19
Figure 3: Vegetation zones and site locations. Dashed lines represent plant species distributions. Solid vertical lines indicate species dominance and dashed lines indicate species is present (based on Agee (1993), Briles et al. (2005) and Franklin and Dymess (1988)).
Laboratory Methods
Cores were sectioned longitudinally in 1-cm increments and stored and refrigerated in 2-oz Whirl-Pak bags. While sectioning cores, macrofossils were collected to perform AMS-radiocarbon dating. These macrofossils were processed and sent to the Livermore National Laboratory, Center for Accelerator Mass Spectrometry. Five wood samples from Kelly Lake and four wood samples from
Miller Lake were sent for dating.


20
Charcoal
Macroscopic charcoal was used to reconstruct local fire histories and to estimate regional biomass burned. Two-cm3 samples for Miller and Kelly lakes were processed every cm. Samples were disaggregated and soaked in household bleach for a minimum of 24 hours as outlined in Whitlock and Larsen (2001), washed through a 125- micron metal-mesh sieve, and transferred to a gridded petri-dish and all charcoal pieces counted using a stereomicroscope.
Pollen
Analysis of the presence and quantity of pollen provides information on local vegetation history and forest structure. The temporal resolution at which pollen was sampled varies between sites, with an average of every 125 years. At Kelly and Miller lakes pollen was sampled on average every 100 years. A total of 42 samples were processed for Kelly Lake and 43 samples for Miller Lake. Removal of organics with potassium hydroxide (KOH), silicates with hydrofluoric acid (HF), and insoluble organics using acetolysis (a 6:lmixture of acetic anhydride to sulfuric acid) was conducted following the methods outlined by the LacCore Pollen Preparation Procedure (University of Minnesota; http://lrc.geo.umn.edu/laccore). Lycopodium tracers were added to aid in calculating pollen concentrations (grains cm'3). Grains were counted at 500 to 125Ox magnification and identified to the lowest taxonomic level possible with the use of atlases and reference collections (Jones et al., 1995; Kapp et al., 2000). Modem phytogeography was used in determining possible genera or species occurring at different elevations. Pinus grains were separated into Haploxylon (.moticola-type) and Diploxylon (ponderosa-type), and all others were classified as undifferentiated.
Data Analysis
Chronology
Age-depth models have evolved in the last decade from linear/polynomial regressions to more iterative models that allow each date and its associated error to influence the reconstruction of


21
210
sediment accumulation in lakes through time. Radiocarbon and Pb age dates, as well as tephrochronology (using the Mazama and Little Glass Mountain eruptions, which occurred at 7627 ± 150 and 986 ± 200 cal. yr BP respectively), were compiled from previously conducted field collections (Briles et al., 2005, 2008; Crawford et al., 2015; White et al., 2015) and new dates for Kelly and Miller lakes. Age-depth models were created using a Bayesian accumulation model called BACON, an R-based statistical package (Blaauw and Christen, 2011), and using the IntCall3 radiocarbon calibration curve (Reimer et al., 2013). BACON assumes a monotonic rate of accumulation and uses a gamma autoregressive process to model sedimentation rates as a function of depth. A ‘shift outlier’ method, is used to identify outliers. A Markov Chain Monte Carlo “t-walk” algorithm is used in producing the posterior distributions (Christen, 1994; Blaauw and Christen 2001). The models were run accepting recommended parameters for memory strength (4) and memory mean (0.7). Accumulation rate means, and the number of sections analyzed were determined by the program after making initial estimates.
Charcoal Analysis
To determine fire history, charcoal accumulation rates (CHAR) were determined using the methods described in Higuera et al. (2009) and the program CharAnalysis
(http://phiguera.github.io/CharAnalysis). Counts, volume, and depths were interpolated to the median sample resolution (yr sample"1-1 for each site. CHAR (particles cm"2 yr"1) are the product of the interpolated charcoal concentrations (particles cm'3) and the interpolated sedimentation rate (cm yr"1). Only the last 5,000 years of data was analyzed to provide baseline conditions during the late Holocene.
CharAnalysis also calculates background charcoal using a tricubic locally-weighted regression and all sites smoothed using a 700-year window. Background charcoal (BCHAR), which represents levels of fuel biomass (Higuera et al. 2007), is useful in identifying charcoal peaks that are


22
then interpreted as local fire episodes. When BCHAR is removed, two components of CHAR remain: (1) the residuals or peaks that represent local fire episodes within the watershed of the lake, and (2) the variability not related to fire, or natural noise, typically from charcoal stored in the watershed or from regional fires. A signal-to-noise index is applied to equally spaced intervals and indicates where CHAR surpasses random noise or where fire events are detected. The noise distribution was determined with a Gaussian mixture model with a locally-defined threshold to determine peak charcoal events and represent fire events. Peaks were then screened using a Poisson-distribution, which identifies single fire events within multiple peaks. Peak magnitude (particles cm"2 peak'1) is the sum of positive CHAR threshold values and is thought to be related to fire severity and/or charcoal transport and delivery into the lake system (Whitlock et al., 2004; Higuera et al., 2007).
Determining Fire Severity
The identification of high severity fires in lake sediment charcoal records is not well established. Examining plant community-level changes using pollen abundances post-fire can provide a more robust determination of fire severity, as high severity fires consume the larger trees and biomass in the forest. To gauge community-level responses to disturbances in historical forests, canopy-to-understory ratios of pollen percentages were used. The ratio for canopy pollen species, a, to understory, b, was calculated as (a-b)/(a+b) in order to normalize values (Jimenez-Moreno et al., 2010). Canopy-to-understory ratios were grouped based on their proximity to an identified fire event within the lake sediment core. Fire pollen ratios came from the same or the preceding pollen sample depth that a fire event that was identified in the charcoal record. All other samples were considered non-fire pollen ratios, were summed, averaged, and used to identify a non-fire pollen ratio mean to characterize vegetation composition during intervals where no fires events occurred (as outlined in Minckley and Shriver (2011)). This study added metric ratios for fire-sensitive and fire-adapted species (Table 2; USDA Fire Effects Information System (FEIS) www.feis-crs.org/feis/), in addition


23
to the original ratio for all pollen types, to further validate or invalidate the characterizations of fire types done with all pollen types.
Only tree or shrub species were used, as herbs are limited in the pollen record and riparian/aquatic species typically are spared from fires due to the proximity to the lake. Fire pollen ratios were then compared to the previously identified non-fire pollen ratio mean. If a fire pollen ratio for all pollen types was below the non-fire pollen mean then the corresponding fire event was characterized as a canopy fire. This assumes that understory taxa will re-establish at a higher rate post-fire relative to canopy taxa, resulting in a lower canopy-to-understory ratio after a fire event (Minckley and Shriver, 2011). These are not the only canopy fires that occurred, but only the fires that this study was able to identify, due to the coarsely sampled pollen records. However, when all records are analyzed together, given the even interval pollen sampling methods, a regional bum severity record can be reconstructed. Microsoft Excel was used to calculate ratios and C2 software version 1.7.7 (Steven Juggins, U of Newcastle, UK) was used to graphically display the data.
Table 2: Pollen types and associated fire sensitivity used in this study
Fire-Sensitive Canopy Pollen Taxa Fire-Adapted Canopy Pollen Taxa Fire-Sensitive Understory Pollen Taxa Fire-Adapted Understory Pollen Taxa
Abies Ainas rubra-type Cercocarpus/Purshia Amelanchier
Acer macrophyllum Cupressaceae Rosaceae Ceanothus
Be tula Pinas monticola-type Salix Chyrsolepis /Lithocarpus
Corylus Pinas Ponderosct-type Qaercas vaccinifolia type
Picect Pseudotsuga Scircobcitas
Tsuga heterophylla Qaercas Large Spiraea
Tsuga mertensicma
Regional Biomass Burned Reconstruction
To observe changes in fire activity over an entire region a synthesis of charcoal data was conducted. This synthesis provides information on how biomass burned has changed through time.


24
Two levels of biomass burned reconstructions were conducted. 1) Within the Klamath Siskiyou Ecoregion, north-to-south and coastal-to-inland gradients were compared. These comparisons were conducted with the nine sites discussed in detail in the study, in addition to four other sites from the Global Charcoal Database, Bluff and Crater Lake CA (Mohr et al., 2000), Mumbo (Daniels et al., 2005), and Cedar (Briles et al. 2008), that are located in the southern Klamath Mountains. The sites were broken into northern (n=5) and southern (n=8) regions, with the northern region north of 41.5° Latitude. Coastal (n=5) and inland (n=8) sites with were located west of 123.2° Longitude. 2) The nine sites used in this study were combined with 32 (Table 3) other sites from the Global Charcoal Database to create biomass burned curves for the northwest US coast to examine how trends observed within the Klamath-Siskiyou Ecoregion compared with trends at a larger spatial scale.
Both sets of biomass burned reconstructions used the same methods and utilized the paleofire R-software package. All sites were standardized using a protocol designed by Blarquez et al. (2014), using four steps. (1) Raw charcoal counts was rescaled to values from 0 to 1 with a minimax transformation. (2) Variances were homogenized with a Box-Cox transformation due to the long upper tail skewness that is inherently present in charcoal data. (3) Z-scores were then determined so the mean and variance were equivalent across all sites (Power et al., 2008). An additional minimax rescaling was performed after the Box-Cox step to ensure that all series had equal values (Marlon et al., 2008). To create the regional mean charcoal composite or the smoothed mean value for charcoal across all sites for a given timeframe (i.e. biomass burned), a two-step data-binning sequence was used. Lirst, the charcoal series was pre-binned with 10-year non-overlapping bins, so that all records have the same influence on the composite. Second, the pre-binned series was smoothed with a “LOWESS” locally weighted smoother and a half width 500-year window. The pre-binned series is then resampled using a bootstrap approach and a LOWESS curve fitting is applied to calculate the confidence intervals (Blarquez et al., 2014).


25
The northwestern US coast regional biomass burned reconstructions was determined using a rectangular area from 40° to 50° North and -118° to -125°S and the 32 Global Charcoal Database sites within were selected space (Table 3). Records for Bolan, Campbell Lake CA, Taylor lake CA, and Sanger Lake CA are in the Global Charcoal Database, but were excluded as the lakes received updated age-depth chronologies. Instead the four sites were added as user-defined data with their updated age-depth models.


26
Table 3: Location, elevations, and citation for the sites used to create regional biomass burned for the northwest US coast.
Site Lat /long Elevation Reference
Bluff 41.340, -122.550 1921 Mohr et al.(2000)
Barrett Lake 37.595, -119.007 2816 Hallett et al.(2002)
Battle Ground Lake 45.800, -122.492 154 Walsh et al.(2002)
Beaver2 44.918, -123.305 69 Walsh et al.(2002)
Cedar 41.200, -122.490 1740 Briles et al. (2011)
Coast Trail Pond 37.985, -122.800 230 Walsh et al.(2002)
Crater Lake CA 41.384, -122.578 2288 Mohr et al.(2000)
Dead Horse Lake 42.561, -120.778 2248 Minckley et al.(2007)
East Lake 37.178, -119.028 2863 Power et al.(2002)
East Sooke Fen 48.352, -123.682 155 Brown et al.(2002)
Five Lakes 48.082, -118.929 780 Scharf et al.(2002)
Glenmire 37.993, -122.777 399 Anderson et al.(2002)
Lake Oswego 45.411, -122.667 30 Walsh et al.(2002)
Lily Lake 41.976, -120.210 2042 Minckley et al.(2007)
Little Lake 44.168, -123.584 703 Long et al.(2002)
Lost 45.824, -123.579 449 Long et al.(2007)
Lower Gaylor Lake 37.909, -119.286 3062 Hallett et al.(2002)
Martins 47.714, -123.540 1415 Gavin et al.(2001)
Moose 47.883, -123.350 1508 Gavin et al.(2001)
Mumbo 41.191, -122.509 1860 Daniels et al.(2005)
Patterson Lake 41.388, -120.224 2743 Minckley et al.(2007)
Porphyry 48.906, -123.883 1100 Brown et al.(2002)
Porter Lake 44.448, -123.243 73 Walsh et al.(2002)
Siesta Lake 37.850, -119.667 2430 Brunelle et al.(2002)
Swamp Lake 37.950, -119.817 1554 Smith et al.(1992)
Taylor 46.101, -123.907 6 Long et al.(2002)
Three Creeks 44.099, -121.627 1996 Long (unpublished)
Todd Lake 44.028, -121.685 1875 Long (unpublished)
Tumalo Lake 44.022, -121.544 1536 Long et al.(2002)
Upper Squaw Lake 42.033, -123.015 930 Colombaroli et al.(2002)
Warner Lake 44.246, -122.958 590 Walsh et al.(2002)
Wildcat Lake 37.968, -122.785 67 Anderson et al.(2002)


27
CHAPTER VI
RESULTS
Charcoal and pollen results for Kelly and Miller lakes have yet to be published and are discussed below in detail. Data for Bolan, Campbell, Fish, Hobart, Ogaromtoc, Sanger, and Taylor Lakes can be found in the publications listed in Table 1. In this chapter, results from chronology, pollen and fire history reconstructions of all sites are discussed.
Chronology
Age determinations based on 14C accelerated mass spectrometry dates and results for Kelly and Miller lakes are shown in Table 2. Dates from previously published sites were used to create new age models (Appendix A) using BACON (Bayesian age-depth modeling), as the previous models were determined using older less robust methods (Ramsey, 2007). Age differences between the new chronologies and previously determined age models averaged around 60 years, in most cases within the error of individual radiocarbon dates, and do not change previously published interpretations of
the data.


28
Table 4: Radiocarbon dates used to create Kelly and Miller Lake's age models. Single "best" Model, min and max are outputs of Bacon and are reported in cal yr BP.
Depth (m) Uncalibrated 14C Age (14C yr B.P.) Single "best" Model (trmean) Min Max Material dated Lab. Reference*
Kelly Lake
62 800 ±30 721 667 810 PlantAVood KL11B 523- 523.5
164 2030 ±35 2000 1881 2128 PlantAVood KL11B 625- 625.5
229.5 2820 ± 25 2922 2802 3064 PlantAVood KL11B 690.5-693
259 3300 ±25 3502 3362 3631 PlantAVood KL11B 720- 720.5
363.5 5960 ±45 6709 6425 6920 PlantAVood KL11B 824.5-826
Miller Lake
34 510 ± 30 352 88 552 PlantAVood ML10 128.5-130
128.5 1460 ±20 977 627 1375 PlantAVood ML11B 1106-1107.5
134.5 770 ± 20 1017 667 1423 PlantAVood ML11B 911.5
227 2110 ± 30 2123 1964 2311 PlantAVood ML11B 1004-1009
329 3590 ±35 3797 3545 3972 PlantAVood ML11B 811-811.5
*C14 dates were processed and run at the Lawrence Livermore National Laboratory in July of 2011
Vegetation Zone Fire History Reconstructions
Oak Woodlands Zone
Fish Lake, has the highest average charcoal accumulation rates (CHAR) (6.84 particles cm"2 yr'1) of all sites (Figure 4a). Background charcoal influx (BCHAR), is initially 2.2 particles cm"2 yr"1, reaches a maximum of 9.8 particles cm"2 yr"1 around 1400 cal yr BP, and decreases to 3.8 particles cm" 2 yr"1 thereafter. A smaller peak of 4.8 particles cm"2 yr"1 occurs around 750 cal yr BP. Between 3000 and 500 cal yr BP peak magnitude averages ~80 particles cm peak " and increases to -1000 cm peak _1 at present. Fire frequency is highest at the beginning of the record (3000 cal yr BP) at -17


29
fires per 1000 years, decreases from 3000-450 cal yr BP to ~ 6 fires per 1000 years, and increases slightly to ~ 7 fires per 1000 years in the last 450 years.
At Ogaromtoc Lake BCHAR slowly increases from 0.19 particles cm'2 yr"1 at 3700 cal yr BP to a maximum of 1.5 particles cm'2 yr"1 at 1200 cal yr BP and decreases to 0.6 particles cm'2 yr"1 at present. Between 3600 and 2000 cal yr BP peak magnitude averages ~10 particles cm"2 peak _1, decreases from 2000 to 500 cal yr BP averaging ~4 cm"2 peak _1, and increases to ~10 cm"2 peak _1 at present. Fire frequency at Ogaromtoc Lake increases from ~1 fire per 1000 years at 3700 cal yr BP to ~7 fires per 1000 years around 2200 cal yr BP. Frequency remains stable, averaging ~6 fires per 1000 years, from 2200 cal yr BP until 400 cal yr BP, then increases to ~7 fires per 1000 years at present.
At Fish and Ogaromtoc Lakes the dominant canopy pollen types that contributed most to the understory-to-canopy ratios were from fire-adapted Alnus rubra-type and Quercus with secondary pollen contributions of fire-adapted Pinus ponderosa-type and Pseudotsuga menziesii. Understory types were based on pollen from fire-adapted Chyrsolepis /Lithocarpus. At Fish Lake 35 pollen samples were processed. Three samples (126 cm, 904 cal yr BP; 151cm, 1139 cal yr BP; 215 cm, 1599 cal yr BP) have fire pollen ratios lower than the non-fire pollen ratio mean in all and fire-adapted pollen types, identify them as canopy fires (Table 5). The non-fire ratio mean at Fish Lake (0.66) is lower than most sites. Of the 33 pollen samples processed at Ogaromtoc Lake, one sample (367 cm, 2549 cal yr) has a fire pollen ratio lower than the non-fire pollen ratio mean in all and fire-adapted pollen types, identifying it as a canopy fire. At both sites, there are more negative fire-sensitive non-fire and fire pollen ratios than at the other sites. The non-fire ratio mean at Ogaromtoc Lake (0.71) is average when compaired to other sites.


30
particles particles fires
cm-2 yr1____________cm-2 peak-1________1000 yr1
Canopy : Understory Ratio
Figure 4a: CHAR, peak magnitude, fire events, fire frequency and pollen ratios of oak woodland and mixed conifer vegetation zones. Canopy-to-understory non-fire pollen ratios (grey) with non-fire-pollen ratio mean (black vertical line). Fire-pollen ratios that correspond with a canopy fire are represented by red bars and those that correspond with understory fires are represented by yellow bars. Dotted lines connect fire events to corresponding fire-pollen ratio.


31
50-1,------,------,--- ,---,----,--- ,----,---r
0 2 4 0 200 600 0 5 10 -1 0 1 -1 0 1 -1 0 1
U1
O “
CQ
UJ
j!
dl
U £
ops
Oh
<
u
Red Fir Zone
Understory Fire Pollen Ratio
Canopy Fire Pollen Ratio
m Non-Fire Pollen Ratio — Non-Fire Pollen Ratio Mean


32
Figure 4b: CHAR, peak magnitude, fire events, fire frequency and pollen ratios for White and Red Fir vegetation zones. Canopy-to-understory non-fire pollen ratios (grey) with non-fire-pollen ratio mean (black vertical line). Fire-pollen ratios that correspond with a canopy fire are represented by red bars and those that correspond with understory fires are represented by yellow bars. Dotted lines connect fire events to corresponding fire-pollen ratio.
Mixed Conifer Zone
BCHAR at Kelly Lake is ~0.1 particles cm'2 yr"1 from 5000 cal yr BP to 3500 cal yr BP, increases to 0.3 particles cm"2 yr"1 by 2600 cal yr BP and remains stable -0.36 particles cm"2 yr"1 thereafter. Between 5000 and 600 cal yr BP peak magnitude averages -2 particles cm"2 peak _1 and increases to -40 particles cm"2 peak_1 at present. Fire frequency at Kelly Lake was highest, -17 fires per 1000 years at 3000 cal yr BP, decreases until -2500 cal yr BP and has remains stable until present at - 9 fires per 1000 years.
BCHAR at Hobart Lake declines during the last 5,000 years from a maximum of 2.9 particles cm"2 yr"1 at 4400 cal yr BP to 0.4 particles cm"2 yr"1 at present. Between 5000 and 4000 cal yr BP
peak magnitude averages -35 particles cm"2 peak_1, decreases from 4000 to 1000 cal yr BP to -11
cm"2 peak_1, and increases to -28 particles cm"2 peak _1 at present. Fire frequency at Hobart is high at 5000 years with -12 fires per 1000 year then declines until 3000 cal yr BP. Frequency remains stable at -6 fires per 1000 years from 3000 to 1200 cal year BP then increases to - 9 fires per 1000 years from the MCA to present.
At Kelly and Hobart lakes, the dominant canopy pollen types that contributed most to the understory-to-canopy ratios were from fire-adapted Pinus monticola, Pseudotsuga menziesii, and fire-sensitive Abies. Secondary contributions of fire-adapted Pinus ponderosa, Alnus rubra-type and Quercus were observed. Pollen from fire-adapted Quercus vaccinifolia-type was the dominant understory taxa at Kelly Lake and fire-sensitive Salix was dominant at Hobart Lake. Of the 42 processed pollen samples at Kelly Lake, five samples (207 cm, 2609 cal yr BP; 239 cm, 3097 cal yr
BP; 247 cm, 3257 cal yr BP; 255 cm, 3427 cal yr BP; 279 cm, 4063 cal yr BP) have fire pollen ratios


33
lower than the non-fire pollen ratio mean in all and fire-adapted pollen types identifying them as canopy fires. The non-fire ratio mean at Kelly Lake (0.78) is average when compared to other sites. At Hobart Lake, of the 42 processed pollen samples, four samples (207 cm, 2785 cal yr BP; 239 cm, 3907 cal yr BP; 247 cm, 4454 cal yr BP; 255 cm, 4657 cal yr BP) were identified as canopy fires.
The non-fire ratio mean at Hobart Lake (0.91) is the highest of all sites.
White Fir Zone
BCHAR at Sanger Lake averages 0.69 particles cm"2 yr'1 and was stable through time reaching a maximum value of 0.96 particles cm"2 yr"1 around 3500 cal yr BP (Figure 4b). Between 5000 and 3000 cal yr BP peak magnitude averages -100 particles cm"2 peak_1, drops to -50 particles cm"2 peaks'1 until 1500 cal yr BP, increases to -220 cm"2 peaks'1 during the MCA and decreases to -50 particles cm"2 peaks'1 at present. From 5000 to 3200 cal yr BP fire frequency at Sanger Lake increases from -2 to -5 fires per 1000 years. Frequency then generally declines to -2 fires per 1000 years at the beginning of the LIA -500 cal yr BP. From 500 cal yr BP to present frequency increases to -4 fires per 1000.
BCHAR at Miller Lake averaged 0.47 particles cm"2 yr"1 and is variable through time (ranging between 0.12 and 0.73 particles cm"2 yr"1) reaching a maximum of 0.73 particles cm"2 yr"1 right before the onset of the LIA around 550 cal yr BP. Between 5000 and 2300 cal yr BP peak magnitude
increases from -40 particles cm"2 peak _1 to -600 particles cm"2 peak _1 and decreases to -10 cm"2 peak _1 at present. Miller Lake fire frequency is relatively stable and reaches a maximum frequency at 1000 cal yr BP of -10 fires per 1000 years and declines to -1 fire per 1000 years at present.
BCHAR at Bolan Lake averages 3.1 particles cm"2 yr"1 and is relatively stable through time reaching a maximum of 4.8 particles cm"2 yr"1 around 3000 cal yr BP. Between 5000 and 1500 cal yr BP peak magnitude increases from -15 particles cm"2 peak_1 to -90 particles cm"2 peak_1 and decreases to -70 cm"2 peak _1 at present. Maximum fire frequency at Bolan Lake occurs at the


34
beginning of the record at 5000 cal yr BP (~10 fires per 1000 years) and declines to ~3 fires per 1000 years at present.
BCHAR at Campbell Lake averaged 0.16 particles cm"2 yr'1 and was variable through time (ranging between 0.06 and 0.3 particles cm"2 yr"1) reaching a maximum value of 0.3 particles cm"2 yr"1 right before the onset of the MCA around 1500 cal yr BP and decrease to 0.9 particles cm"2 yr"1 at present. Between 5000 and 3500 cal yr BP, peak magnitude increases from ~4 particles cm"2 peak _1
to ~40 particles cm"2 peak _1 and decreases to ~20 cm"2 peak_1 at 2700 cal yr BP. There is a gap in fire events and peak magnitude peaks from 2700 to 1700 cal yr BP. From 1700 cal yr BP to present
peak magnitude remains low and relatively stable, averaging ~5 cm"2 peak "\ Maximum fire frequency at Campbell Lake occurs at 5000 cal yr BP (~4 fires per 1000 years) and declines <1 fires per 1000 years at -2500 cal yr BP. Frequency increases from 2500 to around 1000 cal yr BP to -2 fires per 1000 years and declines to <1 fire per 1000 years at present.
At Sanger, Miller, Bolan, and Campbell sites the dominant canopy pollen types that contributed most to the understory-to-canopy ratios were from fire-adapted Pseudotsuga menziesii and fire-sensitive Abies. Secondary pollen abundances of fire-adapted Pinus ponderosa, Alnus rubra-type and Quercus were present. Pollen from fire-adapted Quercus vaccinifolia-sm-type suggests it as the dominant understory taxa. Of the 45 pollen samples processed at Sanger Lake, five samples (21 cm, 144 cal yr BP; 73 cm, 960 cal yr BP; 205 cm, 2877 cal yr BP; 245 cm, 3480 cal yr BP; 281 cm, 4350 cal yr BP) have fire pollen ratios lower than the non-fire pollen ratio mean in all and fire-adapted pollen types and identify canopy fires. The non-fire ratio mean at Sanger Lake (0.64) is the lowest of all sites. At Miller Lake, of the 47 processed pollen samples, one sample (188 cm, 1654 cal yr BP) identifies a canopy fire. The non-fire ratio mean at Miller Lake (0.77) is average when compared to other sites. At Bolan Lake, of the 19 processed pollen samples, one sample (205 cm, 2719 cal yr BP) identifies a canopy fire. The non-fire ratio mean at Bolan Lake (0.73) is average when compared to other sites. At Campbell Lake, of the 42 processed pollen samples, three samples


35
(34cm, 612 cal yr BP; 150 cm, 2740 cal yr BP; 230 cm, 4454 cal yr BP) identify canopy fires. The non-fire ratio mean at Campbell Lake (0.81) is high when compared to other sites.
Red Fir Zone
BCHAR at Taylor Lake averaged 0.62 particles cm"2 yr"1, reaching a maximum of 0.8 particles cm"2 yr"1 early in the record at 4000 cal yr BP, then decreasing to 0.5 particles cm"2 yr"1 between 4000 and 1500 cal yr BP, increasing to 0.73 particles cm"2 yr"1 in the MCA and then declines
to 0.3 particles cm"2 yr"1 toward present. Peak magnitude averages ~2 particles cm"2 peak _1 from 4300 to 1300 cal yr BP and ~18 particles cm"2 peak _1 between 1300 cal yr BP and present. Maximum fire frequency at Taylor Lake occurs at 5000 cal yr BP ~8 fires per 1000 years and declines until -2500 cal yr BP to -3 fires per 1000 years, increases and peaks around 1000 cal yr BP at -5 fires per 1000 years, and then decreases to -3 fires per 1000 years at present.
At Taylor Lake, the dominant canopy pollen types that contributed most to the understory-to-canopy ratios were from fire-adapted Firms monticolct with secondary contributions of fire-sensitive Abies. Pollen from fire-sensitive Rosacecte is the dominant understory. Of the 52 pollen samples processed, one sample (75cm, 1098 cal yr BP) has a fire ratio lower than the non-fire ratio mean in all and fire-adapted pollen types and identifies a canopy fires. The non-fire ratio mean at Miller Lake (0.71) is average when compared to other sites.
Table 5: Fire-related sample ratio comparisons to baseline means, and fire type determinations. Highlighted cells represent ratios that characterized their corresponding fire events as canopy fires
Lake Fire Event age (cal yr BP) All Comparison (fire ratio; non-fire mean) Fire-Sensitive Comparison (fire ratio; non-fire mean) Fire-Adapted Comparison (fire ratio; non-fire mean)
Fish 904 0.54; 0.66 -0.58; -0.25 0.60; 0.70
1048 0.68; 0.66 -0.68; -0.25 0.76; 0.70
1139 0.61; 0.66 0.33; -0.25 0.65; 0.70
1599 0.62; 0.66 -0.13; -0.25 0.65; 0.70
Ogaromtoc 296 0.83; 0.77 -0.5;-0.32 0.89; 0.81
2549 0.63; 0.77 -0.09;-0.32 0.65; 0.81
Kelly 331 0.85; 0.78 0.98; 0.92 0.79; 0.74


36
796 0.81; 0.78 0.98; 0.92 0.79; 0.74
1437 0.83; 0.78 0.83; 0.92 0.84; 0.74
2609 0.65; 0.78 0.94; 0.92 0.60; 0.74
2948 0.79; 0.78 0.93; 0.92 0.71; 0.74
3097 0.58; 0.78 0.87; 0.92 0.48; 0.74
3257 0.73; 0.78 0.84; 0.92 0.71 0.74
3427 0.72; 0.78 0.96; 0.92 0.65; 0.74
4063 0.72; 0.78 0.92; 0.92 0.66; 0.74
Hobart 2785 0.91 ; 0.91 0.59; 0.68 0.97; 0.96
3781 0.97 ; 0.91 0.87; 0.68 0.99; 0.96
3907 0.88 ; 0.91 0.63; 0.68 0.93; 0.96
4454 0.87 ; 0.91 0.35; 0.68 0.96; 0.96
4657 0.89 ; 0.91 0.45; 0.68 0.97; 0.96
Sanger -46 0.72; 0.64 0.78; 0.88, 0.71; 0.59
144 0.53; 0.64 0.93; 0.88, 0.47; 0.59
960 0.51; 0.64 0.80; 0.88, 0.45; 0.59
2877 0.36; 0.64 0.65; 0.88, 0.32; 0.59
3480 0.61; 0.64 0.95; 0.88, 0.55; 0.59
4350 0.49; 0.64 0.92; 0.88, 0.42; 0.59
Miller 676 0.77; 0.77 0.94; 0.89 0.58; 0.68
1654 0.76; 0.77 0.96; 0.89 0.68; 0.68
1847 0.91; 0.77 0.95; 0.89 0.88; 0.68
3055 0.86; 0.77 0.96; 0.89 0.83; 0.68
3454 0.87; 0.77 1; 0.89 0.82; 0.68
3586 0.83; 0.77 0.96; 0.89 0.76; 0.68
3857 0.87; 0.77 0.85; 0.89 0.89; 0.68
Bolan 2719 0.65; 0.73 0.76; 0.91 0.62; 0.64
Campbell 612 0.72; 0.81 0.39; 0.55 0.86; 0.90
1768 0.88; 0.81 0.71; 0.55 0.95; 0.90
2740 0.70; 0.81 0.27; 0.55 0.87; 0.90
4538 0.73; 0.81 0.26; 0.55 0.85; 0.90
Taylor 139 0.89; 0.71 0.80; 0.32 0.94; 0.91
371 0.77; 0.71 0.56; 0.32 0.96; 0.91
1098 0.69; 0.71 0.46; 0.32 0.83; 0.91
1639 0.77; 0.71 0.41; 0.32 0.90; 0.91
2291 0.74; 0.71 0.36; 0.32 0.91; 0.91
4200 0.80; 0.71 0.43; 0.32 0.95; 0.91


37
Biomass Burned Reconstruction
5.000-vear biomass burned trends in the Klamath-Siskivou Ecoregion
Biomass burned has been slowly declining over the last 5,000 years at northern sites in the Klamath-Siskiyou Ecoregion (Figure 4a). Biomass burned was at a maximum z-score of 0.7 ~ 4500 cal yr BP and declines to a historical low of -0.5 at present. A slight increase in biomass burned occurs during the MCA, and in the middle (-0.2) of the climate event, and then continues to decline through the LIA. The southern sites show opposite trends in the initial portion of the record when compared to northern sites. Biomass burned increases from 5000 cal yr BP to 4500 cal yr BP (-0.8 to -0.2), remains relatively stable from 4500 cal yr BP to 2200 cal yr BP (-0.4), then the increases rapidly until peaking (0.7) during the early portion of the MCA -1300 cal yr BP. Biomass burned decreases during the last 1,300 years, but present-day values (-0.4) are within the 5000-year historical range.
Biomass burned at coastal sites (Figure 4b) are high (0.5) at the beginning of the record and remain relatively stable until 3700 cal yr BP, dropping from 0.4 to -0.5 over the next 500 years, and remains low until 2200 cal yr BP. After 2200 cal yr BP, biomass burned increases and peaks at 0.5 around 1300 cal yr BP in the MCA. Biomass burned decreases at coastal sites toward present, but present-day values (0.0) are within the 5000-year historical range. Biomass burned at inland sites was much less variable over time than coastal sites, remaining stable at 0.0 from about 4000 cal yr BP until 1000 cal yr BP and then decreasing toward present to -1.0, a significant 5000-year historic low.


38
o
©
o

5000 4500 4000 3500 3000 2500 2000 1500 1000 500 0
Age (cal yr BP)
Figure 5a: Biomass burned curves for northern vs southern sites, including the number of sites making up the composite. The solid and dashed lines represent biomass burned curves calculated with 500 (black) and 200(orange) year window half width smoothers (Blarquez et al. 2014). The grey band is the 95% confidence interval. Vertical bars represent the time periods of the MCA (green) and LIA (blue).
o
-a

Figure 5b: Biomass burned curves for coastal vs inland sites. The number of records shows the number of sites contributing to the reconstruction and each point. The center lines represent biomass burned curves calculated with 500 (black) and 200(orange) year window half width smoothers (Blarquez et al. 2014). The grey is the 95% confidence interval. Vertical Bars represent the time periods of the MCA (green) and LIA (blue).


39
Biomass Burned Trends in the Klamath-Siskivou Ecoregion and Northwestern US
Biomass burned in the Klamath-Siskiyou Ecoregion (Figure 4c) is highest around 4500 cal yr BP (0.5) and declines until 2400 cal yr BP. After 2400 cal yr BP, it peakes (0.1) at the beginning of the MCA and then declines toward present. Present day biomass burned for the region is at a 5000-year historic low (-1.0),
For the Northwestern US coast biomass burned was stable from 5000 cal yr BP until around 2500 cal yr BP, and then slowly began to increase, peaking (0.1) in the middle of the MCA and has been on a steep downward trend since. Today values (-0.4) are the lowest seen in the past 5,000 years.
N o _
Klamath-Siskiyou Ecoregion k'l'~A 1 IA

Northwest US Coast
l
5000
4500
4000
3500
3000 2500
Age (cal yr BP)
2000
No. of Sites
r 20
No. of Sites
Figure 5c: Klamath-Siskiyou Ecoregion and Northwest US coast trends in biomass burned for the last 5,000 years. The number of records shows the number of sites contributing to the reconstruction and each point. The center lines represent biomass burned curves calculated with 500 (black) and 200(orange) year window half width smoothers (Blarquez et al. 2014). The grey is the 95% confidence interval. Vertical Bars represent the time periods of the MCA (green) and FIA (blue).


40
CHAPTER V
DISCUSSION
Fire is an integral disturbance mechanism in the Klamath-Siskiyou Ecoregion. The complex fire regime in the region has created forests with varying stand compositions and ages. Fire regime designations (i.e. replacement, mixed, surface or low) have large impacts on forest and fire management strategies (Agee, 1993, Barrett et al., 2010; Stephens, 2009). In this study, the spatial variability of fire was determined using fire histories from nine lake sediment cores. These local fire histories were compared along an elevational gradient and grouped within vegetation zones to visualize how changes climate and forest composition influence fire regimes (Figure 7).
Fire severity impacts were determined by examining changes in pollen slightly after a fire event to distinguish between understory (low severity) and canopy (medium to high severity) fires. While complete fire severity records are unable to be reconstructed for individual sites due to the low pollen sample resolution, combining the records from nine sites helps approximate how fire severity has changed in the region during the late Holocene. The identification of high severity fires in lake sediments using proxies, such as pollen and charcoal, is not well established. Some studies have examined the ratio of charcoal accumulation rates to fire frequency as a proxy of fire severity (Ali et al. 2012; Kelly et al 2013). Other studies have identified relationships between charcoal peak height and area burned using a combination of dendrochronology and lake sediment data (Whitlock et al., 2004; Higuera et al. 2011). Minckley and Shriver (2011) conducted a study using changes in pollen spectra following fire events and compared them to pollen spectra of fire-free periods at a single site. This study expands on the Minckley and Shriver (2011) analysis using multiple sites to characterize fire severity using not considering canopy and understory plant impacts inferred, but also fire-adapted and fire-sensitive plant impacts using pollen.
An original goal of this study was to determine if it was possible to identify high severity fires with charcoal peak magnitude heights relative to background charcoal (BCHAR) and to validate


41
those findings by also identifying the same high severity fires with canopy-to-understory ratios. However, based on this comparison, fire severity is not related to peak magnitude (Figure 6) as there is no clustering of high peak magnitude values relative to BCHAR or of high severity fires identified with the canopy-to-understory ratios. This was also demonstrated in the Rocky Mountains of Wyoming (Minckley and Shriver, 2011). As peak size is a result of several factors (e.g., fire size and distance from the repository, wind direction and intensity), these findings suggest that peak magnitude is not a reliable proxy for fire severity.
Figure 6: Scatter plot comparing charcoal values for BCHAR and peak magnitude for all fires at Kelly Lake since 5000 cal yr BP. Circled events are fire events with associated pollen samples. Shaded circles are fire events that were interpreted as canopy fires. All sites were evaluated, and none show a clear trend. Kelly Lake had the greatest variety of fires and was used here as an example.
Biomass burned reconstructions within Klamath-Siskiyou Ecoregion were compared along latitudinal and coastal-to-inland gradients, and regional scale comparisons between the Klamath-Siskiyou Ecoregion and the Northwest US coast. The trends in the data were then compared with independent paleoclimate proxy data (modeled ENSO events, summer solar radiation, and alkenone-


42
derived sea surface temperatures) to infer trends seen in fire activity, biomass burned, and fire severity during the last 5000 years (Figure 7).
A total of 201 fires were identified between the nine sites used in this study, 56 of which are visible across multiple records (Figures 3a and 3b), suggesting that fires in this region have been historically widespread. For example, a fire event was identified -900 cal yr BP in all northern sites (Kelly, Hobart, Sanger, Miller, and Bolan lakes), suggesting a widespread high-severity fire throughout this region. In 1987, the Silver Fire Complex burned nearly 150,000 acres, which was also identified at two sites (Sanger and Bolan lakes).
During the last 5,000 years, summer insolation (Figure 7) declined and resulted in cooler summers than before and a weakened northeastern Pacific subtropical high-pressure system that increased effective moisture (Bartlein et al., 1998). From 5000 to 3300 cal yr BP, alkenone-derived SSTs indicate cool oceans (1-2°C below modem). SST’s rapidly increased to 1-2 °C above modem SSTs within a -200-year period between 3500 and 2200 cal yr BP. After 2200 cal yr BP, SST’s declined to slightly below modem temperatures through present, with near modem SSTs during the early part of the MCA and lower SST’s during the LIA (Barron et ah, 2003). The Palmer Drought Severity Index which uses precipitation and temperature variations to create regional composites of climate conditions across North America, highlights periods of intense drought during the MCA along the Oregon and northern California coast, with the most intense drought occurring south of the CA-OR border (Cook and Kmsic, 2004). On interannual time scales, current El Nino conditions create warmer and drier winters in the Pacific Northwest (Mcafee et al., 2016). El Nino conditions result in early drying of summer fuels and lightning ignitions, which increase the probabilities of large fires (Westerling et al. 2006). ENSO events per 100 years have increased during the last 5,000 years and with the greatest variation prior to and during warmer and drier periods like the MCA. In the last 200 years, ENSO events have declined and are currently near historic lows (Moy et al., 2002). Like El Nino, the Pacific Decadal Oscillation (PDO) also impacts the climate of the Klamath-


43
Siskiyou Ecoregion on decadal time scales. The positive PDO results in warmer SSTs along the western coast of the northern US and lower sea level pressure over the North Pacific Ocean. These anomalies create warmer and drier conditions in the region. When El Nino phases occur in conjunction with the positive PDO phase, the combined effects enhance drying and results in prolonged drought (Mantua et al. 2001; Hessl, 2010; Mcafee, 2016; Newman et al., 2016). Brownand Comrie (2004) found that the connections between ENSO and PDO resulted in a ‘dipole’ between southern and northern locations across the western US, and for the Klamath-Siskiyou’s at the OR-CA border. Strong ENSO signals, in the form of increased ENSO events, only occur during the negative phase of the PDO in the north and during the positive phase in the south (Moy et al.,2002). Spatial variations in fire activity that cannot be explained by large-scale climatic variability, may be the result of the interplay of these interannual- and decadal- climate phenomena.
The interpretation of the results to follow are limited by the number of available records in the region, but also by the proxies that the histories are based upon. Several site factors such as lake size and inflowing streams can have a significant impact on the transport and deposition of pollen and charcoal. Therefore, sites were chosen that were smaller and had minimal, or no inflowing streams. Briles et al. (2012) also suggested that the vegetation history was significantly influenced by soils in the region; therefore, sites on ultramafic substrates were left out of the individual site analyses, but were included in the composite analyses as fire activity was shown to be similar on all substrates. Proxy data inherently have assumptions tied to them. For example, the pollen identified for the sites primarily came from wind-pollinated species, which captures many of the trees and shrubs in the forests, but there are other insect and bird/animal pollinated species that occur in minor abundances that are not represented in the pollen spectra (e.g., all species in the Ericaceae family). Charcoal is also transported and deposited in several ways and, while we can account for these, it is assumed site characteristics do not change through time (e.g., landscape geomorphology). Finally, dating techniques have error associated with them and are costly. Some sites are better dated than others,


44
which influences the sedimentation rate and CHAR calculations and levels. To account for this, the study only evaluates long-term trends in the data at individual sites and standardizes those levels in the composite analyses.


Z-score for transformed charcoal influx Aleknone SST’s Wnr2 Events per lOOyr
-1.0 0.0 1.0 -1.0 0.0 1.0 -1.0 0.0 1.0 -1.0 0.0 1.0 -1.0 0.0 1.0 -1.0 0.0 1.0 10 (°C) 12 20 40 10 30 .
45
Dominant Forest
More Open Pinus monticola /Abies / Quercus Forest
More Closed Finns monticoIa/Abies/ Pseudotsuga Forest
Closed
Abies/Tsuga/ Pseudotsuga Forest
ENSO Event Time Series


Biomass Burning and Fire Events
Northern Sites
Inland Sites
Age (cal yr BP)


46
Figure 7: Biomass burned and fire events, temperature, climatic fluctuations and inferred vegetation history for Abies-dominated forests to illustrate timing of major vegetation shifts (Briles et al., 2012). All identified fire events (black dots), identified canopy fires (red ovals) from the nine sites used in this study. Biomass burned curves are calculated with 500 (black) and 200(orange) year window half width smoothers and associated 95% confidence interval (Blarquez et al. 2014). MCA (green) and the LIA (blue) are highlighted. SST’s (dark blue) derived from phytoplankton alkenones (Barron et al. 2003), July 45° insolation (Bartlein et al., 1998), ENSO event time series (Moy et al., 2002) are used to infer historical change changes and mechanisms in the region.
Influences on Fire Variability
Oak Woodland Vegetation Zone
Pollen records at Ogaromtoc Lake suggest more closed forest during cool wet periods (e.g., LIA and 2300- 2000 cal yr BP), and more open forests during warmer and drier periods (e.g. MCA, and the last 100 years) (Crawford et al., 2015). Increasing then stable fire activity and increasing biomass burned from -3500 cal yr BP until the middle of the MCA, suggest cooler and wetter conditions due to decreasing summer insolation. The decreased biomass burned in the last 1,000 years suggests that warm dry conditions of the MCA likely resulted in larger fires that consumed the fuels that accumulated during the previous cool wet period. Increased fire activity in the last 200 years, and the continued decline of biomass burned, reflects the frequent, low severity fire regime currently observed at lower elevations in the region.
Fish Lake is influenced by more coastal conditions than Ogaromtoc Lake, which likely accounts for some of the contrasting trends in fire at the two sites (Crawford et al., 2015). Fire frequency at Fish Lake decreased from 3000 cal yr BP until the beginning of the LIA (-400 cal yr BP), suggesting that the higher annual precipitation than at Ogaromtoc Lake significantly effects fire activity. At Fish Lake, despite fire activity decreasing throughout most of the record, biomass burned exceeded all other sites in the study. Crawford et al. (2015) indicate that after 1500 cal yr BP at Fish Lake, and to a lesser degree Ogaromtoc Lake, Native American burning altered the landscape with fire, encouraging a more open forest that was more suitable for foraging and hunting. BCHAR at both lakes decreased significantly since -500 cal yr BP, suggesting that European settlers and fire


47
suppression may have changed the levels of fuels burned in the forests (Crawford et al., 2015). However, fire frequency at both sites increased at the beginning of the LIA, suggesting that while biomass burned was less than before due to fire suppression efforts, there was heightened fire activity possibly due to greater human accessibility and ignitions.
At Fish and Ogaromtoc lakes, the lowest elevation sites, negative fire-sensitive non-fire and fire pollen ratios indicate that fire-sensitive understory taxa were more prevalent than fire-sensitive canopy taxa. This is likely because the forest surrounding these lakes are more open than lakes at higher elevations, due to the low severity, high frequency fire regime.
Mixed Conifer Vegetation Zone
Fire frequency during the past 5000 cal yr BP at Hobart and Kelly lakes have both remained relatively high in comparison with sites in other vegetation zones. They also have the most identified fire events. Biomass burned increased at Kelly Lake and decreased at Hobart Lake, indicating trends are inverse of one another during the past 5000 years. The contrast is likely the result of Kelly Lake being in closer proximity to the ocean (~80 km) than Hobart Lake, which lies 110 km east of Kelly. Biomass burned nearly doubled at Kelly Lake at the same time there was a significant increase in SSTs (1-2°C warmer than modem) likely creating warmer conditions there than at Hobart Lake. Alternatively, at Hobart Lake, declining biomass burned after 5000 cal yr BP, is likely due to longterm climatic changes of progressively cooler and wetter conditions than before, due to the gradual decrease in summer insolation during the late Holocene (White et al, 2015). During the MCA, fire activity at Kelly Lake remained stable where at Hobart Lake it increased, suggesting that the increased influence from coastal conditions that resulted in increased moisture, increased biomass, and less frequent fire at Kelly Lake than at Hobart Lake. At Hobart Lake, lightning is generally the source of fire ignition, but there is also evidence of anthropogenic ignitions. Logging in the last 100 years has also changed the forest structure from closed to more open (Agee, 1993; White et al., 2015).


48
The contrasting records suggest that the coastal-to-inland climate gradient is a strong driver of fire regimes at these sites (discussed further below).
White and Red Fir Vegetation Zones
Coastal sites (Bolan and Sanger) within the White Fir Zone have less variation in biomass burned when compared to inland sites in the same zone, especially during the last 2500 years. Fire frequency at both sites was highest prior to 3000 cal yr BP after which frequency at Bolan and Sanger declines, consistent with large-scale climatic trends. Fire frequency at Sanger Lake is lower than Bolan throughout the record and shows more variation in the last 1000 years, suggesting that the increased coastal influence Sanger receives as a result of being 20 km close to the Pacific Ocean than Bolan has a marked impact on precipitation and fire frequency. Since 2500 cal yr BP, fire events were more sporadic resulting in longer, fire-free periods at the coastal sites. This suggests that the influence of coastal weather patterns that create cooler and wetter conditions have shifted the fire regime from a more frequent (5000 - 2500 cal yr BP) to a less frequent fire regime for sites closer to the coast (2500 cal yr BP - present). Briles et al. (2017) suggests a shift in forest composition at Bolan and Sanger Lakes around 4500 cal yr BP from a more open, pine-shrub oak forest, to a closed forest consisting of higher abundances of Abies, as a result of wetter and cooler conditions than before. Interestingly, the timing of the 4500 cal yr BP vegetation shift was different between sites, suggesting the effects of coastal weather patterns (fog production) were stronger at Sanger Lake than the more inland site Bolan Lake (Briles et al., 2008). An 800-year gap in fire events from the middle of the MCA to the end of the LIA occurred at Sanger Lake, which likely allowed for a large buildup of fuels biomass. Fires toward during the LIA at Sanger Lake have burned more fuels (the upper fire likely being the 1987 Silver Fire) likely the result of increased fuels during the previous fire-free period and the warmer and drier conditions of the current climate.
Biomass burned at both inland sites (Miller and Campbell) were more variable than the two coastal sites and Miller Lake had higher biomass burned than Campbell Lake. Biomass burned at


49
Miller and Campbell lakes was initially low until -2400 cal yr BP, then increased at Campbell while remaining low at Miller until the MCA, then increased at both sites during the MCA, and dropped drastically during the LIA. The contrasts and variability in biomass burned between these sites prior to the MCA may be due to their more central location and at the transition between coastal wet forests and the drier inland forests. While the trends in fire frequency are similar at these sites (initially high at 5000 years ago and decreasing through present), Miller Lake had a higher fire frequency and is more variable. The two sites showed the greatest increase in fire frequency during the MCA than any of the other sites. This suggests that the more northern position of Miller Lake results in more biomass accumulation, which during dry summers, is more likely to ignite resulting in more frequent fires.
Fire frequency at Taylor Lake in the Red Fir Zone displayed similar overall trends in fire frequency as the two inland sites (Campbell and Miller). Biomass burned trends are comparable to Campbell Lake, except during and just after the MCA, when it significantly increases, suggesting that the drier conditions resulted biomass burned levels comparable to 5000 years ago. Briles et al. (2011) suggests that increased abundances of Tsuga mertensiana and Abies in the last 2000 years indicate that the forests at higher elevations became more closed likely due to increased precipitation, which would have resulted in more fuels to bum in the drier MCA.
Regional Biomass Burned Patterns and Controls
To characterize changes in fire activity, biomass burned and bum severity across the Klamath-Siskiyou Ecoregion, composite analyses were determined for northem/southem and coastal/ inland locations. The composites were compared with climate and vegetation reconstmctions (Figure 7). The comparisons enable an understanding of how fire operates in a region that is not only at a biogeographic transition zone between drier, more open forests to the south and wetter, more


50
closed forests to the north, but also how the steep coastal-to-inland precipitation gradient influences bum patterns.
Biomass burned at northern sites decreased during the last 5000 years, which coincides with a shift in forest composition from more open forests of Pinus monticola, Abies, and Quercus to more closed forests with less Quercus and more Pseudotsuga menziesii in the forest. The decreasing trend of biomass burned over the last 5000 years at northern sites corresponds to decreasing summer insolation over the same period. A slight increase fire activity occurs from 1200-500 cal yr BP which corresponds with increased ENSO variability and the warmer and drier conditions of the MCA. Biomass burned at northern sites reached a historical low during the LIA, and continues to decrease toward present, but a slight increase in fire activity is likely a response to today’s warmer and drier conditions and/or increased human ignitions.
At southern sites, biomass burned gradually increased from 5000 cal yr BP until ~ 2400 cal yr BP, then increased more rapidly and peaked at -1500 cal yr BP, and then declined through present. Present biomass burned is lower than it has been in the last 2000 years at southern sites, but near levels prior to that time. Variability in biomass burned at southern sites coincides with a rise in ENSO variability and the warmer, drier conditions during the MCA. The 2000-year Palmer Drought Severity Index (PSDI) for grid points 33-36 also indicate that there was less precipitation during the MCA and that drought conditions were intensified at more southern latitudes (Cook et al., 2004).
Fire frequency at northern sites was more frequent, and of higher severity than southern sites, likely due to the higher accumulation of biomass at the wetter northern sites that increase the chances of fire spread and increased fuels burned. A notable gap in high severity fires from -2500 cal yr BP to -1600 cal yr BP at all locations corresponds to a drop in SSTs through present day and suggests that the climate during this time was cool and wet enough to limit high severity fires. All sites record more closed forests than before, so fuel availability was likely not a factor influencing fire severity. The discrepancies observed between biomass burned and fire events at different latitudes highlights


51
the significant differences in precipitation and climatic impacts over this relatively small geographic area.
Trends in biomass burned observed at coastal sites follows SSTs and ENSO variability throughout the record. From 5000 cal yr BP to ~ 3700 cal yr BP, when SSTs were cooler than present day, and ENSO variability was higher, biomass burned was high and fire activity more variable. Fire activity remained low from 3700 to 2500 cal yr BP when SSTs were warm and ENSO events were less variable. Biomass burned increased and peaked after the beginning of the MCA -1300 cal yr BP when SSTs became more stable and ENSO variability increased drastically.
Biomass burned at coastal sites is currently near 5000 year historical lows, and like northern sites, more fire events and high severity fires were identified at coastal sites.
Unlike coastal sites, trends in biomass burned at inland sites remained relatively stable throughout the record and is currently at a 5000 year historical low. During the middle of the MCA (-1100 cal yr BP) biomass burned and fire activity decreased dramatically at inland sites where at coastal sites it declined only slightly. Fire events and high severity fires are more prevalent at coastal sites due to increased precipitation and increased fuels availability in comparison to the drier inland sites. Inland sites do not appear to follow trends in SSTs or ENSO events to the extent that coastal sites do likely because of the lack of coastal influences on precipitation and moisture availability. The notable gap in high severity fires from -2500 cal yr BP until -1600 cal yr BP is also evident in this comparison of coastal-to-inland sites.
The coastal-to-inland precipitation gradient present in the Klamath-Siskiyou Ecoregion and along the western coast of the US is well documented (Briles et al, 2008; Walsh et al, 2015); however, the Klamath-Siskiyou Mountains are at a transition (~41°N) between climate regimes to the north and south, which result in contrasting fire regimes over a short distance. Northern sites are cool and wet, resulting in more fuels and more frequent high severity fires than drier southern sites.


52
In western the United States, despite over a century of fire suppression, frequency, area burned, and wildfire size have all increased significantly since the mid-1980’s (Stephens, 2005; van Mantgem et al, 2013; Westerling et al. 2006). Biomass burned in the western US has been relatively stable during the previous 5,000 years, compared to the prior 8,000 years of increasing biomass burned (Marlon et al. 2012). Biomass burned reconstructions for the Pacific Northwest from northern Oregon to southern British Columbia differ substantially from those for the western US. For example, biomass burned increased from 5000 cal yr BP until the end of the MCA and substantially decreased during the last 1000 years with only a slight increase around 350 cal yr BP in the Pacific Northwest (Walsh et al. 2015). The northwest US coast biomass burned reconstruction includes sites in Washington, Oregon, and Northern California that span the north/south precipitation gradient. Biomass burned in the northwest coast was stable from the beginning of the record until the onset of the MCA when it increased slightly for about 500 years then decreased dramatically. These trends highlight increased fires during the MCA. The scaling of composite records helps identify key locations within the western US where there are transitions in fire regimes, useful for developing fire management plans for diverse forests.
Spatial Variability of Fire in the Klamath-Siskiyou Ecoregion and Implications for Fire
Management
The identification of the historical range of variability of diverse forests requires long-term historical data sets (Gavin et al. 2007; Kaufmann et al., 2007). Paleoenvironmental records have the potential to greatly influence future ecosystem and fire management strategies, which is invaluable particularly given increasing wildfire activity, coupled with the increasing urban-wildland interface across the western United States. It is important to note that the records and results here should not be used as predictive tools, but rather for informing about controls that have resulted in changing fire conditions and the spatial tends across the region. Many current forest management plans (e.g., Healthy Forest Initiative, US Forest Service fire management plans, and CAU FIRE) rarely consider


53
the changing role of fire across diverse landscapes and forests, but rather focus on how to reduce hazards and risks associated with wildland fire, and improve environmental resilience to fire. For example, the 2010 Strategic Fire Plan for California suggests that emerging research be used in management strategies, but does not specifically mention the types of research useful for developing better management plans (Dixon et al, 2010). The Klamath National Forest has created watershed-based fire management plans within the forest, and while they do take into consideration vegetation types within the watershed, little to no consideration is given to precipitation gradients in the region and how fire will respond to the warmer and drier climate predicted in the future.
Historically, northern forests in the Klamath-Siskiyou Mountain Ecoregion, that receive higher amounts of precipitation than southern forests, had fire regimes that burned frequently and at high severity. Current Klamath-Siskiyou forests are a product of LIA conditions and fire suppression at lower elevations where ample biomass exists. As conditions become warmer and drier in the future, these forests will likely bum more frequently and at higher severity than they have historically.
Lower, drier sites that historically burned more often than wetter sites, but at lower severity, are experiencing fire activity already beyond their historical range of variability. The spatial variability of fire is higher in the Klamath-Siskiyou Ecoregion than elsewhere along the northwest US coast, as documented through the gradients analyzed in this study, and forest managers and management plans need to consider how fire may change along these gradients in the future.
In addition to informing future ecological and fire management strategies, the information in the thesis addresses the debate on fire severity in the western US. Some studies have shown that high-severity fires have increased in comparison to historical trends (Colombaroli and Gavin, 2010; Higuera et al., 2014; Miller et al., 2009). On the other side of the debate, Hanson and Odion (2014) suggest that modem high-severity fires have not increased in proportion, area, or patch size in western North America and conclude that climate change might not have as large of an impact on fire severity in the future as the others have projected (Hanson and Odion, 2014). This study indicates a decrease


54
in high severity fires during cool and wet periods, like those -2500 to 1600 cal yr BP and during the LIA, and increased high severity fires during warm dry periods, such as the MCA and prior to 2500 cal yr BP. Within the limitations of this study it does not appear as though high severity fires have increased in frequency in the recent past in the Klamath-Siskiyou Ecoregion; however, prior to the LIA, during warm and dry periods, there were periods of more severe and widespread fire and these can be expected if conditions continue to warm and become drier than before.
Future research
A goal of this study was to determine a way of differentiating high severity fire from moderate-to-low severity fires. The study showed that it is possible to identify high severity, canopy replacing fires by using high-resolution charcoal records in conjunction with changes in pollen ratios during fire events in comparison with ratios during non-fire periods. However, a clear method of identifying all high severity fires with the limited number pollen samples available was not determined. To get a complete history of fire severity for an individual site, pollen samples around each fire event need to be analyzed. Future research on fire severity should include additional pollen processing at the same depth and immediately following fire events to determine fire severity for all identified fire events and vegetation response post-fire. Additional sampling of non-fire time periods will enable a more robust non-fire to fire pollen ratio, and a moving average record could be more representative of normal variation through time than a single average. This would allow for the spatial variability of high severity fires to be determined along elevational gradients in addition to coastal-to-inland and latitudinal gradients. It would also be advantageous to determine pollen zones at each site so that fuel types and connectivity can be examined for their effects on fire severity similar to the study conducted in the Cascades of northern Oregon (Minckley and Long, 2016).
Examining fire pollen ratios after known high severity fires would increase the validity of the understory-to-pollen ratio method used in this study and described by Minckley and Shriver (2011).


55
Modem calibration studies from recent fires would also help develop more accurate reconstructions of fire severity.
Additionally, there is a lack of data in the southern portion of this study area. To better understand the latitudinal gradient proposed by this study, pollen and charcoal data from new sites would be useful in increasing our understanding of how differences in precipitation along these gradients have influenced vegetation and fire regimes in the Klamath-Siskiyou Ecoregion.
To increase knowledge of the main climatic factors that influence the spatial variability of fire in the Klamath-Siskiyou Ecoregion additional information regarding climate-fire relationship is needed. How fire activity interacts with interannual and decadal variability in climate, like the effects of El Nino/ Southern Oscillations and Pacific Decadal Oscillations especially along the gradients discussed in this study will enable better predictions of wildfire severity, occurrence, and extent (Hessl et al. 2010).


56
CHAPTER VI
CONCLUSIONS
This study reports on the pyrogeography of the Klamath-Siskiyou Ecoregion. Charcoal and pollen records from nine lakes were examined along elevational, coastal-to-inland, and latitudinal gradients to determine the variability in fire regimes and controls. In addition, fire severity was explored to determine how it has changed along these gradients. The research addressed the following questions:
1) What is the spatial and temporal variability of fire during the last 5000 years in the Klamath-Siskiyou Ecoregion?
Fire activity not only varied temporally, but also along elevational, coastal-to-inland, and latitudinal gradients in the Klamath-Siskiyou Ecoregion. While long-term fire activity is largely governed by changes in solar radiation influencing ocean temperatures, pressure systems and the position of the jet stream and storm tracks, there is evidence that native people influenced fire regimes at very small local scales and at low elevations. As vegetation moved along elevational gradients and fuel loads and moisture levels changed, so did fire activity and fuel loads. Sites more sensitive to moisture fluctuations such as lower and more southem/inland sites burned more frequently and were driven more by fuel availability than climate. Higher, more coastal, and northern sites burned less frequently and had more climate driven fire regimes. Interestingly, fire activity and biomass burned increased at most sites during the Medieval Climate Anomaly (MCA) and decreased during the Little Ice Age (LIA), suggesting regional coherency to drought conditions and cool wet conditions of those events, respectively. There is also evidence of a widespread fire across the northern Klamath-Siskiyou Ecoregion during the MCA (-900 cal yr BP). At no other time in the last 5000-years did this level of coherency exist across all sites and all composite records, suggesting that drivers of fire were operating across a much boarder region than they had in the past. Biomass burned is at historical lows


57
in the ecoregion, when compared to the last 5000 years, and primarily at northern and inland locations, due to LIA conditions and fire suppression. Fire managers should consider the complexity at which fires have burned in the region historically when developing future management plans.
2) How has fire severity changed during the last 5000 years in the Klamath-Siskiyou Ecoregion?
The resolution of the pollen records used in this study are not high enough to distinguish fire severity trends at individual sites, but when they are combined, and examined along latitudinal and coastal-to-inland gradients trends can be inferred. Southern sites had less frequent high severity fires than northern sites, due to increased precipitation and increased fuel availability. Coastal sites were like northern sites and recorded more high severity fires during the MCA than inland sites. Southern and inland locations, receiving lower precipitation than the other locations, recorded less frequent higher severity fires. The highest concentration of high severity fires occurred prior to 2500 cal yr BP. No high severity fires were identified from 2500 cal yr BP to -1600 cal yr BP; which corresponds with a drop in SSTs through present day and suggests that the climate during this time was cool and wet enough to limit high severity fires. However, they increased through the MCA and became less severe during the LIA. Fire severity in the Klamath-Siskiyou Ecoregion has not increased in the last 500 years. Higher resolution pollen records are needed to reconstruct individual site fire severity and to enhance the composite analyses through the region.
Regional historical fire severity reconstructions helps to inform the debate on changing patterns of fire severity in the western US. Trends can be observed over much longer timelines with paleoeclogical data than is possible with other methods of measuring fire severity (i.e dendrochronology, and remotely sensed data). These longer timelines increase knowledge of how fire severity responded to warmer drier conditions in the past, like during the MCA, and have the potential to contribute greatly to ecological and fire management strategies to better protect the forest in the Klamath-Siskiyou Ecoregion in the face of future warmer and drier conditions.


58
REFERENCES
Ali A. A., Carcaillet C., Bergeron Y. (2009). Long-term fire frequency variability in the eastern Canadian boreal forest: the influences of climate vs. local factors. Global Change Biology 15, 1230-1241
Agee, J. K. (1993). Fire ecology of Pacific Northwest forest. Washington DC: Island Press.
Alexander, E.B., Coleman, R.G., Keeler-Wolfe, T., Harrison, S.P., 2007. Serpentine geoecology of western North America. Oxford University Press.
Baker, W. L. (2015). Are high-severity fires burning at much higher rates recently than historically in dry-forest landscapes of the western USA? PLoS ONE, 10(9), 1-26.
Baker, W. L. (1992). The landscape ecology of large disturbances in the design and management of nature reserves. Landscape Ecology, 7(3), 181-194.
Blaauw, M., & Christen, J. A. (2011). Flexible paleoclimate age-depth models using an autoregressive gamma process. Bayesian Analysis, 6(3), 457-474.
Blarquez, O., Vanniere, B., Marlon, J. R., Daniau, A. L., Power, M. J., Brewer, S., & Bartlein, P. J. (2014). Paleofire: An R package to analyze sedimentary charcoal records from the Global Charcoal Database to reconstruct past biomass burning. Computers and Geosciences, 72(September), 255-261.
Bartlein, P. J., Anderson, K. H., Edwards, M. E., Mock, C. J., Thompson, R. S., Webb, T. I., & Whitlock, C. (1998). Paleoclimate simulations for North America over the past 21,000 years: Features of the Simulated Climate and Comparisons with Paleoenvironmental Data. Quaternary Science Reviews, 17, 549-585.
Barrett, S.; Havlina, D.; Jones, J.; Hann, W.; Frame, C.; Hamilton, D.; Schon, K.; Demeo, T.; Hutter, L.; Menakis, J. (2010). Interagency Fire Regime Condition Class Guidebook. Version 3.0, https://www.frcc.gov/.
Barron, J. A., Heusser, L., Herbert, T., & Lyle, M. (2003). High-resolution climatic evolution of coastal northern California during the past 16,000 years. Paleoceanography, 18(1).
Bond, W.J., & van Wilgen, B.W. (1996) Prescribed burning in California wildlands vegetation management. University if California Press, Berkley, California, USA.
Briles, C. E., Whitlock, C., & Bartlein, P. J. (2005). Postglacial vegetation, fire, and climate history of the Siskiyou Mountains, Oregon, USA. Quaternary Research, 64(1), 44-56.
Briles, C. (2008) Holocene Vegetation and fire history of the floristically Diverse Klamath Mountains, Northern California, USA. PhD Dissertation
Briles, C. E., Whitlock, C., Bartlein, P. J., & Higuera, P. (2008). Regional and local controls on postglacial vegetation and fire in the Siskiyou Mountains, northern California, USA.
Palaeogeography, Palaeoclimatology, Palaeoecology, 265(1-2), 159-169.


59
Briles, C. E., Whitlock, C., Skinner, C. N., & Mohr, J. (2011). Holocene forest development and maintenance on different substrates in the Klamath Mountains, northern California, USA. Ecology, 92(3), 590-601.
Briles, C. E. (2017). Controls on Mountain Plant Diversity in Northern California: A 14,000-Year Overview. Annals of the American Association of Geographers, 107(2), 238-249.
Brown, D. P., & Comrie, A. C. (2004). A winter precipitation “dipole” in the western United States associated with multidecadal ENSO variability. Geophysical Research Letters, 31(9), 1-4.
Christen, J.A., 1994. Summarizing a set of radiocarbon determinations - a robust approach. Applied Statistics-Journal of The Royal Statistical Society Series C, 43 (3), 489 - 503.
Christen, J.A. & Fox, C. (2010). A general purpose sampling algorithm for continuous distributions (the t-walk). Bayesian Analysis, 4(2), 263-282
Cook, E.R. & Krusic, P.J., (2004). The North American Drought Atlas.Lamont-Doherty Earth Observatory and the National Science Foundation.
Cook, E.R., Woodhouse, C. A., Eakin, C. M., Meko, D. M., & Stahle, D. W. (2004). Long-term aridity changes in the western United States. Science, 306(5698), 1015-1018.
Colombaroli, D., & Gavin, D. G. (2010). Highly episodic fire and erosion regime over the past 2,000 y in the Siskiyou Mountains, Oregon. Proceedings of the National Academy of Sciences, 107(44), 18909.
Crawford, J. N., Mensing, S. A., Lake, F. K., & Zimmerman, S. R. H. (2015). Late Holocene fire and vegetation reconstruction from the western Klamath Mountains, California, USA: A multidisciplinary approach for examining potential human land-use impacts. Holocene, 25(8), 1341— 1357.
Daniels, M. L., Anderson, R. S., & Whitlock, C. (2005). Vegetation and fire history since the Late Pleistocene from the Trinity Mountains, northwestern California, USA. The Holocene, 75(7), 1062-1071.
DellaSala, D. A., S. R. Reid, T. J. Frest, J. R. Strittholt, & D. M. Olson. (1999). A global perspective on the biodiversity of the Klamath-Siskiyou ecoregion. Natural Areas Journal 19:300-319.
Deng, W., Liu, X., Chen, X., Wei, G., Zeng, T., Xie, L., & Zhao, J. (2017). A comparison of the climates of the Medieval Climate Anomaly, Little Ice Age, and Current Warm Period reconstructed using coral records from the Northern China Sea. Journal of Geophysical Research: Oceans, 122, 264-275.
Dixon, S. L., Giocomini, P., Andre, M., Bradshaw, LI., Nakamura, G., et al. (2010). 2010 Strategic Fire Plan For California. State Board of Forestry and Fire Protection, 47. Retrieved from http://cdfdata.fire.ca.gov/pub/fireplan/fpupload/fpppdf668.pdf
Dobrowski, S. Z. (2011). A climatic basis for microrefugia: The influence of terrain on climate.
Global Change Biology, 17(2), 1022-1035.


60
Franklin, J. F., & C. T. Dymess (1988), Natural Vegetation of Oregon and Washington Rep., Oregon State University Press, Oregon State University, Corvallis, Oregon.
Fried, J.S., Tom, M. S., & Mills, E. (2004). The Impact of Climate Change on Wildfire Severity: A Regional Forecast for Northern California. Climatic Change, 64, 169-191.
Gavin, D. G., Hallett, D. J., Hu, F. S., Uertzman, K. P., Prichard, S. J., Brown, K. J., Uynch, J.A., Bartlein, P., & Peterson, D. U. (2007). Forest fire and climate change in western North America: insights from sediment charcoal records. Frontiers in Ecology and the Environment, 5(9), 499-506.
Gershunov, A., T. P Barnett, & D. R. Cayan. (1999). North Pacific interdecadal oscillation seen as factor in ENSO- related North American climate anomalies. EOS Transactions 80 (3): 25-30.
Graham, N. E., M. K. Hughes, C. M. Ammann, K. M. Cobb, M. P. Hoerling, D. J. Kennett, J. P. Kennett, B. Rein, L. Stott, P. E. Wigand, and T. Xu. (2007). Tropical Pacific-mid- latitude teleconnections in medieval times. Climate Change 83:241-285.
Grove, A. T. (2001). The “Little Ice Age” and its geomorphological consequences in Mediterranean Europe. Climatic Change, 48(1), 121-136.
Hanson, C. T., & Odion, D. C. (2014). Is fire severity increasing in the Sierra Nevada, California, USA? International Journal of Wildland Fire, 23(2009), 1-8.
Harrison, S., Damschen, E. I., & Grace, J. B. (2010). Ecological contingency in the effects of climatic warming on forest herb communities. Proceedings of the National Academy of Sciences, 107(45), 19362-19367.
Harrison, S., Damschen, E., Fernandez-Going, B., Eskelinen, A., & Copeland, S. (2014). Plant communities on infertile soils are less sensitive to climate change. Annals of Botany, 116(6), 1017-1022.
Hessl, A. E., Mckenzie, D., & Schellhaas, R. (2010). Drought and Pacific Decadal Oscillation Linked to Fire Occurrence in the Inland Pacific Northwest. Ecological Applications, 14(2), 425-442.
Higuera PE, Peters ME, Brubaker LB and Gavin DG (2007) Understanding the origin and analysis of sediment-charcoal records with a simulation model. Quaternary Science Reviews 26: 1790-1809
Higuera, P.E., L.B. Brubaker, P.M. Anderson, F.S. Hu, T.A. Brown. (2009). Vegetation mediated the impacts of postglacial climatic change on fire regimes in the south-central Brooks Range, Alaska. Ecological Monographs, 79: 201-219.
Higuera, P. E., Whitlock, C., & Gage, J. a. (2011). Linking tree-ring and sediment-charcoal records to reconstruct fire occurrence and area burned in subalpine forests of Yellowstone National Park, USA. The Holocene, 21(2), 327-341.
Higuera, P. E., Briles, C. E., & Whitlock, C. (2014). Fire-regime complacency and sensitivity to centennial through millennial-scale climate change in Rocky Mountain subalpine forests, Colorado, USA. Journal of Ecology, 102(6), 1429-1441.


61
Jimenez-Moreno, G., Anderson, R. S., Desprate, S., Grigg, L. D., Grimm, E. C., & Al., E. (2010). Millennial-scale variability during the last glacial in vegetation records from Europe. Quaternary Science Reviews, 29(21-22), 2839-2864.
Jones, G.D., Bryant, V.M., Lieux, M.H., Jones, S.D., and Lingren, P.D. (1995). “Pollen of the southeastern United States: with emphasis on melissopalynology and entomopalynology.” American Association of Stratigraphic Palynologists Foundation, College Station, TX.
Kaufmann, M.R., D. Binkley, P.Z. Fule, M. Johnson, S.L. Stephens, & T.W. Swetnam. (2007). Defining old growth for fire-adapted forests of the western United States. Ecology and Society 12(2): 15.
Kapp, R.O., Davis, O.K., and King, J.E. (2000). “Ronald O. Kapp’s pollen and spores.” American Association of Stratigraphic Palynologists Foundation, College Station, TX.
Kelly, R., Chipman, M. L., Higuera, P. E., Stefanova, I., Brubaker, L. B., & Hu, F. S. (2013). Recent burning of boreal forests exceeds fire regime limits of the past 10,000 years. Proceedings of the National Academy of Sciences, 110(32), 13055-13060.
Kruckeberg, A.R., (2002). Geology and plant life: the effects of landforms and rock types on plants. University of Washington Press. Seattle, WA. p. 362.
Mann, M. E., Zhang, Z., Rutherford, S., Bradley, R. S., Hughes, M. K., Shindell, D., Ammann, C., Faluvegi, G., & Ni, F. (2009). Global signatures and dynamical origins of the Fittle Ice Age and Medieval Climate Anomaly. Science, 326(5957), 1256-60.
Mantua, N. J., & Hare, S. R. (2002). The Pacific Decadal Oscillation. Journal of Oceanography.
Marlon, J. R., Bartlein, P. J., Carcaillet, C., Gavin, D. G., Harrison, S. P., Higuera, P. E., Joos, F., Power, M.J., & Prentice, I. C. (2008). Climate and human influences on global biomass burning over the past two millennia. Nature Geoscience, 1, 697-702.
Marlon, J. R., Bartlein, P. J., Gavin, D. G., Fong, C. J., Anderson, R. S., Briles, C. E., Walsh, M. K. et al. (2012). Fong-term perspective on wildfires in the western USA. Proceedings of the National Academy of Sciences, 109(9), 535-543.
Martin, R.E., & Sapsis, D.B., (1992). Fires as agents of Biodiversity: Pyrodiversity promoted
biodiversity. In Harris, R.R., Erman, & Kemer, H.M (Editors), Proceedings of the Symposium on Biodiversity of Northwest California. University of California, Berkley, CA, Wildland Resources Center Repot No. 29, pp. 150-157.
Mcafee, S. A., & Wise, E. K. (2016). Intra-seasonal and inter-decadal variability in ENSO impacts on the Pacific Northwest. International Journal of Climatology, 36(1), 508-516.
Miller, J. D., Safford, H. D., Crimmins, M., & Thode, A. E. (2009). Quantitative evidence for increasing forest fire severity in the the Sierra Nevada and the southern Cascade Mountains, California, and Nevada, USA. Ecosystems, 12(1), 16-32.


62
Minckley, T. A., & Shriver, R. K. (2011). Vegetation responses to changing fire regimes in a rocky mountain forest. Fire Ecology, 7(2), 66-80.
Minckley, T. A., & Long, C. J. (2016). Paleofire severity and vegetation change in the Cascade Range, Oregon, USA. Quaternary Research, 85(2), 211-217.
Mock, C. J. (1996). Climatic controls and spatial variations of precipitation in the western United States. Journal of Climate.
Mohr, J. A., C. Whitlock, and C. N. Skinner. (2000). Postglacial vegetation and fire history, Eastern Klamath Mountains, California. Holocene 10:587-601.
Moy, C. M., Seltzer, G. O., Rodbell, D. T., & Anderson, D. M. (2002). Variability of El
Nino/Southem Oscillation activity at millennial timescales during the Holocene epoch. Nature, 420(November).
Nagel TA, Taylor AH (2005) Fire and persistence of montane chaparral in mixed conifer forest
landscapes in the northern Sierra Nevada, Lake Tahoe Basin, California, USA. The Journal of the Torrey Botanical Society, 132, 442-457.
Newman, M., Alexander, M. A., Ault, T. R., Cobb, K. M., Deser, C., Di Lorenzo, et al., (2016). The Pacific Decadal Oscillation, Revisited. Journal of Climate, 29(12), 4399-4427.
Odion, D. C., & Hanson, C. T. (2006). Fire severity in conifer forests of the Sierra Nevada,
California. Ecosystems, 9(7), 1177-1189.
Olson, D., DellaSala, D. a., Noss, R. F., Strittholt, J. R., Kass, J., Koopman, M. E., & Allnutt, T. F. (2012). Climate Change Refugia for Biodiversity in the Klamath-Siskiyou Ecoregion. Natural Areas Journal, 32(1), 65-74.
Power, M. J., Marlon, J., Oritz, P. J., Bartlein, P. J., Harrison, S. P., Mayle, F. E., Ballouche, A., et al. (2008). Changes in fire regimes since the Last Glacial Maximum: an assessment based on a global synthesis and analysis of charcoal data. Climate Dynamics, 30, 887-907.
PRISM Climate Group, Oregon State University, http://prism.oregonstate.edu, created 4 Feb 2004.
Ramsey, C. (2007). Deposition Models for Chronological Records, Quaternary Science Reviews, 1-42.
Reimer, P.J., Bard, E., Bayliss, A., Beck, J.W., Blackwell, P.G., Bronk, R.C., Buck, C.E., Cheng, H., Edwards, R.L., et al. (2013). IntCall3 and Marinel3 radiocarbon age calibration curves 0-50000 years cal BP. Radiocarbon 55, 1869-1887.
Safford, H. D., Miller, J., Schmidt, D., Roath, B., & Parsons, A. (2007). BAER soil bum severity maps do not measure fire effects to vegetation: A comment on Odion and Hanson (2006).
Ecosystems, 11(1), 1-11.


63
Sommers, W. T., Coloff, S. G., & Conard, S. G. (2011). Synthesis of Knowledge: Fire History and Climate Change. JFSP Synthesis Reports. Paper 19, 190p. Retrieved from http://www.firescience.gov/JFSP_fire_history.cfm
Steel, Z. L., Safford, H. D., & Viers, J. H. (2015). The fire frequency-severity relationship and the legacy of fire suppression in California forests. Ecosphere, 6(1), 8.
Stephens SL. (2005). Forest fire causes and extent on United States forest service lands. International Journal of Wildland Fire, 14:213-22.
Stephens, S. L., Moghaddas, J. J., Edminster, C., Fiedler, C. E., Haase, S., Harrington, M.,
Youngblood, A. et al. (2009). Fire Treatment Effects on Vegetation Structure, Fuels , and Potential Fire Severity in Western U.S. Forests. Ecological Applications, 19(2), 305-320.
Stephens, S. L., Agee, J. K., Fule, P. Z., North, M. P., Romme, W. H., Swetnam, T. W., & Turner, M. G. (2013). Managing Forests and Fire in Changing Climates. Science, 342(6154), 41-2.
Taylor, A. H., & Beaty, R. M. (2005). Climatic influences on fire regimes in the northern Sierra Nevada mountains, Lake Tahoe Basin, Nevada, USA. Journal of Biogeography, 32(3), 425-438.
Taylor, A. H., & Skinner, C. N. (1998). Fire history and landscape dynamics in a late-successional reserve, Klamath Mountains, California, USA. Forest Ecology and Management, 111, 285-301.
Taylor, A. H., & Skinner, C. N. (2003). Spatial Patterns and Controls on Historical Fire Regimes and Forest Structure in the Klamath Mountains. Ecological Applications, 13(3), 704-719.
van Mantgem, P. J., Nesmith, J. C. B., Keifer, M., Knapp, E. E., Flint, A., & Flint, L. (2013).
Climatic stress increases forest fire severity across the western United States. Ecology Letters, 16(9), 1151-1156.
Walsh, M. K., Whitlock, C., & Bartlein, P. J. (2010). 1200 years of fire and vegetation history in the Willamette Valley, Oregon and Washington, reconstructed using high-resolution macroscopic charcoal and pollen analysis. Palaeogeography, Palaeoclimatology, Palaeoecology, 297(2), 273-289.
Walsh, M. K., Marlon, J. R., Goring, S. J., Brown, K. J., & Gavin, D. G. (2015). A Regional
Perspective on Holocene Fire-Climate-Human Interactions in the Pacific Northwest of North America. Annals of the Association of American Geographers, 105(6), 1135-1157.
Westerling, A. L., Hidalgo, H. G., Cayan, D. R., & Swetnam, T. W. (2006). Warming and earlier spring increase western US forest wildfire activity. Science, 313(5789), 940-943.
Westerling, A. L. (2016). Increasing western US forest wildfire activity: sensitivity to changes in the timing of spring. Philosophical Transactions of the Royal Society of London. Series B, Biological Sciences, 371(1696), 717-728.
White, A., Briles, C., & Whitlock, C. (2015). Postglacial vegetation and fire history of the southern Cascade Range, Oregon. Quaternary Research, 84(3), 348-357.


64
Whitlock, C., & Larsen C. P. S. (2001). Charcoal as a fire proxy. Pages 75-97 in J. P. Smol, H. J. B. Birks, and W. M. Last, editors. Tracking environmental change using lake sediments. Volume 3. Terrestrial, algal, and siliceous indicators. Kluwer Academic, Dordrecht, The Netherlands.
Whitlock, C., Skinner, C. N., Bartlein, P. J., Minckley, T., & Mohr, J. A. (2004). Comparison of charcoal and tree-ring records of recent fires in the eastern Klamath Mountains, California, USA.
Canadian Journal of Forest Research, 34(10), 2110-2121.
Whittaker, R. H. (1960). Vegetation of the Siskiyou Mountains, Oregon and California. Ecological Monographs, 30(3), 279-338.


65
APPENDIX A: ANALYSISES PERFORMED
Analysis performed Program the file was used in Puropose of the analysis File name of data set or file folder Reference/ Online Location
Chronology R: Bacon package Age-depth models All_Lakes_Bacon.xls http: //www. chrono. qub .ac .uk/blaauw/b acon.html
Peak detection in sediment charcoal Char Analysis Fire history reconstruction CharAnalysis File Folder http s: //github .com/ phiguera/ CharAnal ysis
Z-score charcoal composites R: Paleofire package Biomass burning/ fire frequency reconstructions Paleofire File Folder https://cran.r-proj ect.org/web/pac kages/paleofire/ind ex.html
Pollen percent calculations Microsoft Excel Vegetation history reconstructions Pollen File Folder
Canopy: understory ratios Microsoft Excel Fire severity type characterization All Fakes_Fast5k_Ratios .xls Minckley and Shriver 2011 & 2016


66
APPENDIX B: CHRONOLOGY FOR ALL SITES
_ accshape 1.5
to - ism.stength: 4 “ iin Ann 0.7 _£0
0.4 0.8 Memory
2000 4000 teratioo
Depth
II I II 2000 5000 Iteraton
Acc. rale (yr/cm)
Werrory
Acc. rate (yr/cm)
Memory
Depth
Depth
Depth
2000
ltera:on
0 100 200 Acc. rate (yr/cm)
Depth Depth Depth


67
APPENDIX C: POLLEN PERCENTAGES AND CANOPY-TO- UNDERSTORY RATIOS
Bolan Lake Fire Ratio
Depth Age Fire Fire Adapted Sensitive Canopy Canopy Fire Adapted Shrubs Fire Sensitive Shrub All Canopy: Understory Sensitive Canopy: Understory Adapted Canopy: Understory
1 -47 29.40 20.72 5.30 0.48 0.79 0.95 0.69
8 -4 33.24 26.24 6.41 0.29 0.80 0.98 0.68
25 132 16.83 9.06 5.50 0.32 0.63 0.93 0.51
45 309 22.12 13.27 4.13 0.00 0.79 1.00 0.69
65 522 19.48 19.58 5.19 0.00 0.77 1.00 0.58
85 843 30.93 13.40 6.19 0.52 0.74 0.93 0.67
105 1154 23.12 12.01 6.31 1.50 0.64 0.78 0.57
125 1474 26.69 8.90 7.98 0.31 0.62 0.93 0.54
145 1787 24.29 9.89 6.50 0.56 0.66 0.89 0.58
165 2077 20.00 9.65 5.41 0.24 0.68 0.95 0.57
185 2406 28.49 15.62 4.93 1.37 0.75 0.84 0.70
205 2719 23.75 9.06 5.63 1.25 0.65 0.76 0.62
225 3056 23.62 12.88 5.21 0.00 0.75 1.00 0.64
245 3377 29.51 23.77 5.46 0.55 0.80 0.96 0.69
265 3690 24.65 15.30 5.67 0.85 0.72 0.89 0.63
285 4076 22.30 16.39 2.30 0.66 0.86 0.92 0.81
305 4481 26.27 16.72 5.07 0.60 0.77 0.93 0.68
325 4843 31.58 7.24 7.24 0.66 0.66 0.83 0.63
345 5186 34.21 10.20 7.89 1.64 0.65 0.72 0.63
Campbell Lake Fire Ratio
Depth Age Fire Fire Adapted Sensitive Canopy Canopy Fire Adapted Shrubs Fire Sensitive Shrub All Canopy: Understory Sensitive Canopy: Understory Adapted Canopy: Understory
0 -52 45.86 28.99 0.89 3.25 0.90 0.80 0.96
4 -17 40.23 23.51 2.55 3.12 0.84 0.77 0.88
8 23 49.86 13.97 2.47 3.29 0.83 0.62 0.91
14 228 46.69 18.37 3.01 2.41 0.85 0.77 0.88
18 344 42.86 18.96 2.60 2.86 0.84 0.74 0.89
22 437 48.44 19.56 0.67 1.11 0.95 0.89 0.97
26 518 47.90 8.24 1.34 4.54 0.81 0.29 0.95
30 578 45.34 14.51 2.59 2.33 0.85 0.72 0.89
34 612 32.58 10.66 2.46 4.71 0.72 0.39 0.86


68
38 647 44.05 18.72 1.32 2.20 0.89 0.79 0.94
42 680 44.72 9.52 1.04 4.14 0.83 0.39 0.95
46 713 43.44 14.48 1.17 2.74 0.87 0.68 0.95
50 760 37.44 10.76 2.91 4.71 0.73 0.39 0.86
54 848 44.96 20.16 0.82 4.63 0.85 0.63 0.96
58 925 41.52 13.45 1.17 3.31 0.85 0.60 0.95
62 1019 40.98 13.66 3.09 1.03 0.86 0.86 0.86
66 1105 44.29 13.67 1.63 5.51 0.78 0.43 0.93
70 1194 49.72 12.43 2.82 3.77 0.81 0.53 0.89
74 1269 39.77 14.19 3.26 4.65 0.74 0.51 0.85
78 1342 45.64 14.22 2.06 5.50 0.78 0.44 0.91
82 1419 39.44 9.05 1.61 4.83 0.77 0.30 0.92
86 1489 32.06 8.18 1.45 2.24 0.83 0.57 0.91
90 1553 35.12 10.12 3.51 4.96 0.68 0.34 0.82
94 1632 43.63 12.74 2.17 4.34 0.79 0.49 0.91
98 1699 44.83 9.92 1.03 7.02 0.74 0.17 0.95
102 1768 42.59 16.08 1.04 2.71 0.88 0.71 0.95
106 1837 48.88 9.13 1.42 3.25 0.85 0.48 0.94
110 1902 49.10 14.03 2.61 2.40 0.85 0.71 0.90
118 2059 33.77 13.29 1.96 3.70 0.79 0.56 0.89
126 2228 39.56 13.56 2.89 3.78 0.78 0.56 0.86
134 2393 27.41 28.70 3.52 1.85 0.83 0.88 0.77
142 2568 40.49 10.49 2.68 3.90 0.77 0.46 0.88
150 2740 41.25 10.94 2.81 6.25 0.70 0.27 0.87
158 2910 52.45 11.76 2.70 2.70 0.85 0.63 0.90
166 3086 57.28 12.58 1.49 4.97 0.83 0.43 0.95
174 3267 49.91 7.10 4.09 3.37 0.77 0.36 0.85
182 3464 43.55 16.94 2.96 2.69 0.83 0.73 0.87
190 3666 38.38 11.18 5.48 1.97 0.74 0.70 0.75
198 3871 40.19 11.92 5.61 3.50 0.70 0.55 0.76
206 4058 42.40 11.66 4.42 3.53 0.74 0.53 0.81
214 4225 43.34 10.32 2.63 3.38 0.80 0.51 0.89
222 4380 46.49 10.46 1.49 3.74 0.83 0.47 0.94
230 4538 42.90 7.50 3.43 4.40 0.73 0.26 0.85
234 4627 49.17 10.47 1.16 4.82 0.82 0.37 0.95
242 4782 44.06 12.47 3.42 4.02 0.77 0.51 0.86
250 4950 39.85 8.23 0.91 4.75 0.79 0.27 0.96
258 5111 34.08 11.80 0.67 6.68 0.72 0.28 0.96
Fire Ratio
Fish Lake


69
Depth Age Fire Fire Adapted Sensitive Canopy Canopy Fire Adapted Shrubs Fire Sensitive Shrub All Canopy: Understory Sensitive Canopy: Understory Adapted Canopy: Understory
6 -47 78.00 2.75 14.25 1.75 0.67 0.22 0.69
11 -28 72.87 2.58 16.02 2.58 0.60 0.00 0.64
20 9 81.10 0.79 11.29 2.89 0.70 -0.57 0.76
27 47 77.04 4.23 7.25 3.63 0.76 0.08 0.83
36 143 85.35 1.41 6.20 1.69 0.83 -0.09 0.86
39 195 79.52 2.66 9.04 3.19 0.74 -0.09 0.80
46 314 80.38 2.99 7.36 4.90 0.74 -0.24 0.83
56 368 78.03 2.54 7.61 5.07 0.73 -0.33 0.82
66 416 72.32 0.78 14.36 6.27 0.56 -0.78 0.67
76 467 79.13 1.63 10.84 2.44 0.72 -0.20 0.76
81 496 71.20 0.52 19.63 4.71 0.49 -0.80 0.57
86 531 79.95 1.03 10.54 2.57 0.72 -0.43 0.77
101 662 76.47 1.07 9.09 4.81 0.70 -0.64 0.79
111 758 77.11 1.32 9.47 6.32 0.66 -0.66 0.78
126 904 70.99 1.10 17.68 4.14 0.54 -0.58 0.60
141 1048 76.88 0.84 10.31 4.46 0.68 -0.68 0.76
151 1139 75.59 1.31 16.27 2.62 0.61 -0.33 0.65
166 1280 75.86 1.06 14.59 1.86 0.65 -0.27 0.68
176 1371 77.84 0.53 11.35 3.43 0.68 -0.73 0.75
187 1448 75.13 1.05 13.09 3.40 0.64 -0.53 0.70
201 1538 73.09 1.00 16.36 2.90 0.59 -0.49 0.63
215 1599 71.85 1.87 15.01 2.41 0.62 -0.13 0.65
226 1642 76.52 0.79 10.29 2.11 0.72 -0.45 0.76
241 1692 76.08 2.14 12.90 1.34 0.69 0.23 0.71
251 1725 70.69 1.28 17.99 2.06 0.56 -0.23 0.59
262 1762 74.27 0.53 18.83 1.59 0.57 -0.50 0.60
276 1808 70.90 2.38 16.93 1.06 0.61 0.38 0.61
292 1877 71.86 0.82 21.31 0.55 0.54 0.20 0.54
301 1920 71.05 1.05 17.89 1.58 0.57 -0.20 0.60
309 1956 73.85 1.08 17.79 1.08 0.60 0.00 0.61
326 2052 78.97 0.77 13.85 1.28 0.68 -0.25 0.70
346 2213 82.71 1.06 5.32 1.06 0.86 0.00 0.88
391 2552 72.51 0.79 18.59 0.79 0.58 0.00 0.59
411 2677 72.41 1.33 18.04 2.12 0.57 -0.23 0.60
446 2865 75.82 0.82 13.74 1.37 0.67 -0.25 0.69
Hobart Lake Fire Ratio


70
Depth Age Fire Fire Adapted Sensitive Canopy Canopy Fire Adapted Shrubs Fire Sensitive Shrub All Canopy: Understory Sensitive Canopy: Understory Adapted Canopy: Understory
1 -59 59.62 11.12 1.69 0.84 0.93 0.86 0.94
16 -6 64.07 8.29 1.54 2.15 0.90 0.59 0.95
32 57 49.58 6.48 1.67 3.35 0.84 0.32 0.93
48 126 53.75 3.33 0.42 2.92 0.89 0.07 0.98
64 187 65.19 8.92 1.70 1.70 0.91 0.68 0.95
80 250 59.13 3.04 3.04 9.57 0.66 -0.52 0.90
96 314 60.51 11.02 0.46 2.30 0.93 0.66 0.98
119 402 53.26 9.44 1.32 4.83 0.82 0.32 0.95
135 459 60.50 6.73 0.69 1.72 0.93 0.59 0.98
151 520 71.03 9.78 0.38 1.92 0.94 0.67 0.99
167 555 41.49 21.72 0.39 0.39 0.98 0.96 0.98
183 611 63.72 10.08 0.00 2.42 0.94 0.61 1.00
199 686 52.01 16.16 1.49 4.48 0.84 0.57 0.94
202 706 65.04 7.82 0.41 4.11 0.88 0.31 0.99
217 836 65.56 7.92 1.28 2.57 0.90 0.51 0.96
218 840 66.66 5.52 2.29 3.05 0.86 0.29 0.93
233 916 60.03 18.70 0.73 0.73 0.96 0.92 0.98
234 917 56.33 16.35 1.54 1.54 0.92 0.83 0.95
249 978 68.12 8.77 1.99 0.80 0.93 0.83 0.94
250 982 61.12 11.56 0.90 2.40 0.91 0.66 0.97
266 1076 46.31 18.50 0.90 0.30 0.96 0.97 0.96
282 1173 57.70 11.51 1.24 1.24 0.93 0.80 0.96
298 1269 67.84 11.31 1.26 0.50 0.96 0.91 0.96
314 1362 67.30 10.28 1.23 1.23 0.94 0.79 0.96
330 1463 62.74 10.60 1.21 2.12 0.91 0.67 0.96
346 1558 58.36 13.77 2.32 1.99 0.89 0.75 0.92
362 1654 47.02 19.71 1.08 3.25 0.88 0.72 0.95
378 1749 56.50 15.55 1.18 1.97 0.92 0.78 0.96
394 1852 52.79 11.54 0.84 1.26 0.94 0.80 0.97
410 1958 55.36 15.41 1.19 0.79 0.95 0.90 0.96
426 2068 57.16 9.52 1.98 1.19 0.91 0.78 0.93
444 2188 52.70 9.70 2.43 2.43 0.86 0.60 0.91
460 2294 51.94 11.12 2.52 1.26 0.89 0.80 0.91
476 2405 56.07 12.52 1.99 3.18 0.86 0.59 0.93
492 2530 52.85 13.58 0.38 1.13 0.96 0.85 0.99
508 2658 58.91 12.63 0.71 0.36 0.97 0.95 0.98
524 2785 66.59 9.27 1.18 2.36 0.91 0.59 0.97


71
540 2908 63.90 8.09 0.62 1.56 0.94 0.68 0.98
556 3031 41.63 19.46 1.59 1.59 0.90 0.85 0.93
572 3153 42.21 18.88 1.53 1.14 0.92 0.89 0.93
588 3279 74.08 7.99 0.65 1.30 0.95 0.72 0.98
604 3399 51.40 14.19 0.33 1.67 0.94 0.79 0.99
620 3528 51.57 10.37 0.68 0.68 0.96 0.88 0.97
636 3654 50.13 11.63 0.00 0.00 1.00 1.00 1.00
652 3781 60.21 9.26 0.32 0.65 0.97 0.87 0.99
668 3907 55.51 10.48 1.98 2.37 0.88 0.63 0.93
684 4034 60.45 6.83 0.95 3.18 0.88 0.37 0.97
706 4208 61.91 6.65 0.40 0.81 0.97 0.78 0.99
722 4333 58.08 7.58 1.35 0.67 0.94 0.84 0.95
738 4454 56.36 6.46 1.17 3.13 0.87 0.35 0.96
754 4554 56.52 6.50 1.33 0.33 0.95 0.90 0.95
770 4657 64.19 7.83 1.12 2.98 0.89 0.45 0.97
786 4764 72.14 5.79 1.35 1.62 0.93 0.56 0.96
808 4907 67.13 7.68 1.54 0.61 0.94 0.85 0.96
824 5014 64.56 8.86 2.04 1.36 0.91 0.73 0.94
Kelly Lake Fire Ratio
Depth Age Fire Fire Adapted Sensitive Canopy Canopy Fire Fire Adapted Sensitive Shrubs Shrub All Canopy: Understory Sensitive Canopy: Understory Adapted Canopy: Understory
11 67 42.86 15.82 4.59 0.51 0.84 0.94 0.81
15 123 31.97 11.15 7.43 0.37 0.69 0.94 0.62
23 233 30.18 17.68 6.10 0.00 0.77 1.00 0.66
27 283 35.14 23.55 0.39 1.16 0.95 0.91 0.98
31 331 35.56 22.46 4.28 0.27 0.85 0.98 0.79
37 408 26.98 13.23 5.82 0.00 0.75 1.00 0.65
39 431 36.59 14.29 8.01 2.09 0.67 0.74 0.64
47 534 44.33 8.59 8.25 0.69 0.71 0.85 0.69
51 590 39.11 8.12 5.90 0.00 0.78 1.00 0.74
55 639 41.20 4.87 7.49 0.75 0.70 0.73 0.69
63 728 47.89 9.96 9.96 0.38 0.70 0.93 0.66
65 751 41.83 9.13 5.77 0.48 0.78 0.90 0.76
69 796 56.62 10.14 6.76 0.00 0.82 1.00 0.79
73 848 46.43 19.94 2.38 0.00 0.93 1.00 0.90
77 897 45.51 18.84 2.61 0.58 0.91 0.94 0.89
85 996 41.95 9.73 4.36 0.00 0.84 1.00 0.81
91 1072 25.43 12.50 5.17 0.00 0.76 1.00 0.66
95 1125 45.07 9.87 3.29 0.00 0.89 1.00 0.86


72
103 1226 51.09 9.35 5.61 0.31 0.82 0.94 0.80
107 1278 32.46 12.72 5.92 0.00 0.77 1.00 0.69
111 1332 41.97 6.48 5.35 0.00 0.80 1.00 0.77
119 1437 42.80 8.33 3.79 0.76 0.84 0.83 0.84
127 1538 48.38 9.42 3.25 0.32 0.88 0.93 0.87
137 1663 37.30 12.85 5.64 0.00 0.80 1.00 0.74
143 1740 38.41 8.28 4.30 0.99 0.80 0.79 0.80
151 1839 36.84 4.86 6.07 0.40 0.73 0.85 0.72
159 1935 38.91 10.29 2.57 0.32 0.89 0.94 0.88
167 2034 42.99 14.01 4.78 0.32 0.84 0.96 0.80
175 2148 38.00 17.43 5.71 0.00 0.81 1.00 0.74
183 2258 36.05 9.09 7.84 0.00 0.70 1.00 0.64
199 2494 40.82 12.60 4.93 0.27 0.82 0.96 0.78
207 2609 38.90 8.73 9.73 0.25 0.65 0.94 0.60
215 2726 41.18 14.82 11.53 0.24 0.65 0.97 0.56
223 2841 31.91 8.21 6.99 0.00 0.70 1.00 0.64
231 2948 36.96 23.91 6.21 0.93 0.79 0.93 0.71
239 3097 29.47 13.17 10.34 0.94 0.58 0.87 0.48
247 3257 34.14 10.27 5.74 0.91 0.74 0.84 0.71
255 3427 33.55 14.47 7.24 0.33 0.73 0.96 0.65
263 3593 33.63 12.39 7.37 1.47 0.68 0.79 0.64
271 3811 35.92 12.64 8.33 0.00 0.71 1.00 0.62
279 4063 36.03 12.27 7.31 0.52 0.72 0.92 0.66
287 4319 42.57 5.28 6.27 0.66 0.75 0.78 0.74
Miller Lake Fire Ratio
Depth Age Fire Fire Adapted Sensitive Canopy Canopy Fire Fire Adapted Sensitive Shrubs Shrub All Canopy: Understory Sensitive Canopy: Understory Adapted Canopy: Understory
1 -57 18.47 24.44 2.61 0.00 0.89 1.00 0.75
8 12 22.12 20.06 4.05 0.67 0.80 0.93 0.69
12 67 20.77 10.32 4.13 0.26 0.75 0.95 0.67
16 110 20.30 21.99 2.46 0.35 0.88 0.97 0.78
24 223 14.15 13.87 0.76 0.25 0.93 0.96 0.90
28 281 17.25 14.59 3.50 1.17 0.74 0.85 0.66
32 324 24.59 16.67 2.01 0.00 0.91 1.00 0.85
42 399 22.30 18.85 2.82 0.70 0.84 0.93 0.78
44 412 16.36 16.10 2.81 0.77 0.80 0.91 0.71
52 468 18.25 19.34 2.81 0.62 0.83 0.94 0.73
60 518 17.86 17.62 6.02 0.60 0.69 0.93 0.50
62 532 18.08 17.37 3.16 0.00 0.84 1.00 0.70


73
68 573 17.98 14.36 4.33 0.27 0.75 0.96 0.61
76 625 25.61 5.73 2.80 0.51 0.81 0.84 0.80
84 676 11.49 16.04 3.03 0.51 0.77 0.94 0.58
92 730 24.40 11.84 5.00 0.87 0.72 0.86 0.66
100 780 12.05 17.96 4.92 0.58 0.69 0.94 0.42
108 833 23.19 15.12 2.59 0.78 0.84 0.90 0.80
116 887 21.13 17.05 3.44 0.86 0.80 0.90 0.72
124 946 18.67 12.27 6.22 1.21 0.61 0.82 0.50
132 1006 18.99 12.57 6.86 1.43 0.58 0.80 0.47
140 1076 19.68 10.80 2.91 0.42 0.80 0.93 0.74
156 1272 18.88 19.80 5.22 1.53 0.70 0.86 0.57
172 1464 12.92 17.24 4.46 0.50 0.72 0.94 0.49
188 1654 19.01 8.82 3.67 0.18 0.76 0.96 0.68
204 1847 27.86 21.32 1.78 0.59 0.91 0.95 0.88
220 2030 26.35 12.28 3.29 2.40 0.74 0.67 0.78
236 2233 23.94 14.02 4.08 0.25 0.80 0.96 0.71
252 2513 23.94 12.27 4.04 0.00 0.80 1.00 0.71
260 2649 10.34 11.76 1.67 0.83 0.80 0.87 0.72
268 2782 18.43 3.90 2.14 1.33 0.73 0.49 0.79
276 2918 6.86 4.15 3.79 0.72 0.42 0.70 0.29
284 3055 27.86 11.19 2.62 0.24 0.86 0.96 0.83
292 3185 12.44 10.60 2.07 0.26 0.82 0.95 0.71
300 3320 24.61 14.80 3.48 0.58 0.81 0.92 0.75
308 3454 17.94 9.30 1.83 0.00 0.87 1.00 0.82
316 3586 29.02 14.89 3.91 0.30 0.83 0.96 0.76
324 3723 13.95 12.34 3.49 0.54 0.73 0.92 0.60
332 3857 34.75 14.96 2.12 1.24 0.87 0.85 0.89
340 3944 22.85 10.92 2.61 0.95 0.81 0.84 0.79
348 4020 21.31 13.25 4.21 0.90 0.74 0.87 0.67
364 4174 24.76 17.07 1.88 0.00 0.91 1.00 0.86
372 4256 18.71 21.79 5.41 0.32 0.75 0.97 0.55
Ogaromtoc Lake Fire Ratio
Depth Age Fire Fire Adapted Sensitive Canopy Canopy Fire Fire Adapted Sensitive Shrubs Shrub All Canopy: Understory Sensitive Canopy: Understory Adapted Canopy: Understory
1 -59 78.61 1.03 12.37 0.52 0.72 0.33 0.73
10 -46 70.00 1.28 18.72 2.56 0.54 -0.33 0.58
20 -26 75.42 0.96 14.46 1.93 0.65 -0.33 0.68
40 72 88.43 0.77 4.88 0.77 0.88 0.00 0.90
60 220 90.89 0.52 2.60 1.04 0.92 -0.33 0.94


74
70 296 86.17 0.99 4.94 2.96 0.83 -0.50 0.89
80 383 81.39 1.74 7.44 3.23 0.77 -0.30 0.83
100 562 89.09 0.78 4.42 2.08 0.87 -0.45 0.91
120 775 81.98 2.28 6.09 4.06 0.78 -0.28 0.86
135 865 83.25 0.25 8.12 4.82 0.73 -0.90 0.82
150 953 80.10 2.07 7.49 2.84 0.78 -0.16 0.83
166 1046 83.10 1.19 10.00 2.62 0.74 -0.38 0.79
180 1126 87.19 1.01 3.77 3.02 0.86 -0.50 0.92
190 1185 81.44 0.00 7.22 3.09 0.78 -1.00 0.84
200 1242 84.12 0.50 7.20 2.73 0.79 -0.69 0.84
208 1287 81.25 1.50 8.25 3.75 0.75 -0.43 0.82
220 1361 75.73 1.58 9.50 1.85 0.74 -0.08 0.78
231 1428 81.09 1.49 8.21 3.48 0.75 -0.40 0.82
250 1553 81.65 1.81 7.24 2.84 0.78 -0.22 0.84
259 1601 80.90 1.26 7.79 2.76 0.77 -0.38 0.82
270 1662 82.37 1.76 10.83 1.51 0.74 0.08 0.77
280 1720 84.29 1.05 7.59 1.31 0.81 -0.11 0.83
290 1781 84.18 0.97 10.22 1.95 0.75 -0.33 0.78
306 1907 83.91 1.73 8.66 2.48 0.77 -0.18 0.81
310 1940 83.42 0.75 8.29 2.26 0.78 -0.50 0.82
320 2027 81.30 0.52 9.61 2.34 0.75 -0.64 0.79
332 2133 80.15 1.96 10.78 3.19 0.71 -0.24 0.76
350 2293 88.43 1.54 4.37 1.03 0.89 0.20 0.91
367 2549 76.50 1.25 16.25 1.50 0.63 -0.09 0.65
383 2673 72.10 2.22 9.88 8.15 0.61 -0.57 0.76
400 2856 81.23 0.77 6.43 1.54 0.82 -0.33 0.85
410 3037 78.88 1.46 10.92 3.16 0.70 -0.37 0.76
428 3363 88.40 1.39 7.19 1.62 0.82 -0.08 0.85
Sanger Lake Fire Ratio
Depth Age Fire Fire Adapted Sensitive Canopy Canopy Fire Adapted Shrubs Fire Sensitive Shrub All Canopy: Understory Sensitive Canopy: Understory Adapted Canopy: Understory
1 -46 45.99 6.94 7.81 0.87 0.72 0.78 0.71
5 -14 45.58 9.12 5.63 0.54 0.80 0.89 0.78
9 19 40.85 14.08 4.23 0.56 0.84 0.92 0.81
13 56 36.11 10.90 4.06 1.28 0.80 0.79 0.80
17 95 26.14 5.61 5.44 0.35 0.69 0.88 0.66
21 144 24.83 4.70 8.89 0.17 0.53 0.93 0.47
25 202 29.14 7.58 7.19 0.40 0.66 0.90 0.60
29 252 42.29 11.97 6.12 0.27 0.79 0.96 0.75


75
35 323 44.83 29.47 3.76 0.00 0.90 1.00 0.85
37 350 37.05 16.71 6.13 1.11 0.76 0.88 0.72
41 404 48.18 11.48 7.00 1.12 0.76 0.82 0.75
45 458 34.26 7.61 11.93 0.51 0.54 0.88 0.48
49 507 37.50 16.19 6.25 0.00 0.79 1.00 0.71
53 553 40.00 7.86 9.29 0.95 0.65 0.78 0.62
57 612 38.12 11.49 6.53 0.78 0.74 0.87 0.71
61 683 36.97 4.70 11.32 0.64 0.55 0.76 0.53
65 772 34.74 6.70 7.69 0.25 0.68 0.93 0.64
69 867 28.24 4.50 12.23 0.54 0.44 0.79 0.40
73 960 32.21 8.11 12.16 0.90 0.51 0.80 0.45
77 1051 43.32 10.08 7.08 0.54 0.75 0.90 0.72
81 1124 42.65 7.60 13.73 0.25 0.56 0.94 0.51
85 1176 47.67 8.81 6.74 0.26 0.78 0.94 0.75
95 1315 33.42 8.29 9.84 0.52 0.60 0.88 0.54
103 1429 38.19 8.04 8.04 0.25 0.70 0.94 0.65
111 1533 38.50 11.74 9.62 0.23 0.67 0.96 0.60
119 1651 47.07 7.56 12.68 0.24 0.62 0.94 0.58
127 1759 34.72 9.07 9.07 1.30 0.62 0.75 0.59
135 1858 35.71 5.56 10.05 0.53 0.59 0.83 0.56
139 1907 31.70 6.70 14.96 0.00 0.44 1.00 0.36
149 2039 38.91 6.11 11.99 0.45 0.57 0.86 0.53
161 2202 29.20 4.42 15.34 1.47 0.33 0.50 0.31
173 2362 35.07 7.58 12.32 0.24 0.55 0.94 0.48
188 2601 36.83 8.57 10.71 0.00 0.62 1.00 0.55
197 2740 35.25 8.53 14.52 0.46 0.49 0.90 0.42
205 2877 27.22 3.52 13.89 0.74 0.36 0.65 0.32
213 3001 38.83 6.91 8.78 0.80 0.65 0.79 0.63
225 3178 41.37 6.35 10.66 0.25 0.63 0.92 0.59
233 3295 37.87 10.40 6.13 0.00 0.77 1.00 0.72
245 3480 39.10 9.02 11.28 0.25 0.61 0.95 0.55
257 3721 23.68 8.70 14.37 0.20 0.38 0.95 0.24
265 3889 35.60 10.99 8.64 0.00 0.69 1.00 0.61
277 4241 40.34 5.80 11.35 0.72 0.59 0.78 0.56
281 4350 25.82 5.82 10.63 0.25 0.49 0.92 0.42
297 4886 32.65 7.14 9.18 0.77 0.60 0.81 0.56
305 5138 28.04 6.30 18.70 0.43 0.28 0.87 0.20
Taylor Lake
Fire Ratio


76
Depth Age Fire Fire Adapted Sensitive Canopy Canopy Fire Adapted Shrubs Fire Sensitive Shrub All Canopy: Understory Sensitive Canopy: Understory Adapted Canopy: Understory
1 -48 34.29 14.40 1.57 4.19 0.79 0.55 0.91
4 -30 38.73 8.35 3.80 9.37 0.56 -0.06 0.82
8 7 28.68 12.22 2.99 7.48 0.59 0.24 0.81
12 58 44.28 14.43 1.49 7.46 0.74 0.32 0.93
16 119 37.56 14.93 0.00 3.73 0.87 0.60 1.00
17 139 45.56 20.60 1.34 2.35 0.89 0.80 0.94
21 206 44.29 21.45 0.93 3.03 0.89 0.75 0.96
24 265 36.03 13.58 0.78 5.74 0.77 0.41 0.96
25 284 34.73 18.72 1.97 6.90 0.72 0.46 0.89
29 371 29.82 20.18 0.66 5.70 0.77 0.56 0.96
32 439 39.23 11.21 3.24 4.42 0.74 0.43 0.85
33 459 29.10 14.44 1.31 5.91 0.72 0.42 0.91
36 515 44.07 17.93 0.30 3.04 0.90 0.71 0.99
37 537 30.47 16.25 1.35 7.00 0.70 0.40 0.91
43 705 32.70 19.53 1.49 4.03 0.81 0.66 0.91
49 840 34.35 14.49 1.87 8.41 0.65 0.27 0.90
53 887 30.59 12.56 3.42 4.11 0.70 0.51 0.80
57 926 31.15 18.06 1.13 4.29 0.80 0.62 0.93
59 947 33.57 12.82 2.80 6.99 0.65 0.29 0.85
71 1062 39.57 14.04 1.28 6.17 0.76 0.39 0.94
75 1098 37.17 17.38 3.48 6.42 0.69 0.46 0.83
79 1147 36.32 20.35 0.44 3.94 0.86 0.68 0.98
83 1212 36.24 10.82 4.00 6.12 0.65 0.28 0.80
87 1292 41.79 11.59 1.45 4.11 0.81 0.48 0.93
91 1376 38.96 17.47 2.61 8.23 0.68 0.36 0.87
95 1456 30.22 13.78 2.00 11.78 0.52 0.08 0.88
99 1552 34.29 12.57 1.71 6.86 0.69 0.29 0.90
103 1639 44.35 11.75 2.44 4.88 0.77 0.41 0.90
107 1726 44.70 16.23 2.98 8.28 0.69 0.32 0.88
111 1817 43.36 15.62 0.47 7.69 0.76 0.34 0.98
115 1915 34.20 9.43 0.94 8.73 0.64 0.04 0.95
119 2004 38.44 18.84 2.01 5.78 0.76 0.53 0.90
127 2157 41.81 9.89 0.56 8.19 0.71 0.09 0.97
135 2291 41.01 13.15 1.93 6.19 0.74 0.36 0.91
143 2433 48.99 11.69 0.90 13.26 0.62 -0.06 0.96
151 2586 48.31 12.80 2.17 9.42 0.68 0.15 0.91
159 2734 40.85 17.96 0.00 12.68 0.65 0.17 1.00


77
167 2881 40.39 9.73 1.70 6.81 0.71 0.18 0.92
175 3029 40.47 13.26 2.33 9.53 0.64 0.16 0.89
183 3179 48.39 13.90 1.99 8.68 0.71 0.23 0.92
187 3260 42.94 20.72 0.60 6.01 0.81 0.55 0.97
195 3432 35.97 8.35 1.71 7.07 0.67 0.08 0.91
203 3596 40.33 11.69 0.95 9.07 0.68 0.13 0.95
211 3746 28.47 12.68 1.91 5.98 0.68 0.36 0.87
219 3897 45.54 13.61 1.98 4.95 0.79 0.47 0.92
227 4048 36.22 10.21 1.73 5.59 0.73 0.29 0.91
235 4200 39.38 11.93 0.95 4.77 0.80 0.43 0.95
243 4363 35.39 9.55 2.25 9.55 0.58 0.00 0.88
251 4506 35.04 6.15 1.54 7.86 0.63 -0.12 0.92
259 4645 42.22 10.87 0.21 7.25 0.75 0.20 0.99
269 4905 42.50 7.60 1.56 7.41 0.70 0.01 0.93
277 5121 49.29 8.53 2.84 11.61 0.60 -0.15 0.89


Full Text

PAGE 1

SPATIAL AND TEMPORAL VARIABILIT Y OF FIRE REGIMES ACROSS THE BIODIVERSE KLAMATHSISKIYOU ECO REGION, N ORTHERN CALIFORNIA AND SOUTHERN OREGON, USA b y SHELLEY J . MORTON B.S., Metropolitan State University Denver, 2017 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 Environmental Sciences Program 2017

PAGE 2

ii This thesis for the Master of Science degree by Shelley J. Morton has been approved for the Environmental Sciences program by Christy E Briles, Chair Peter Anthamatten Rafael Moreno D ate: December 16, 2017

PAGE 3

iii Morton, Shelley J. (M.S., Environmental Sciences Program ) Spatial and Temporal Variability of Fire Regimes across the Biodiverse Klamath -Siskiyou Ecoregion, Northern California and Southern Oregon. Thesis directed by Assistant Professor Christy E Briles Ph .D. ABSTRACT The Klamath -Siskiyou Ecoregion in northern California and southern Oregon is a biodiversity hotspot where future climate change and fire activity are significant concerns. The historical range of variability of fire has been difficult to examine in this region due to the generally low temporal resolution of historical documents and t emporally restricted proxy records like those from tree rings. Paleoenvironmental records from lake sediments, based on proxies of pollen and macroscopic char coal, spanning the last 5000 years were analyzed to temporally expand knowledge of the relationships between fire activity, climate change, and forests. Reconstruction of these histories enables quantification of fire frequency , biomass burned and fire se verity along elevational, latitudinal and coastal -to -inland gradients. The results of the study indicate that fire activity was less frequent and more severe at wetter northern and coastal sites, while it was frequent and less severe at drier southern and inland locations . The Klamath -Siskiyou Ecoregion collectively experienced higher fire activity and severity during the Medieval Climate Anomaly when conditions were warmer and drier. These trends were exaggerated at southern and inland sites that receiv ed less precipitation tha n northern and coastal sites. During the Little Ice Age when conditions were cooler and wetter, fewer and less severe fires occurred, especially at northern and coastal sites that receive more precipitation. Current forest and fi re conditions are a legacy of the LIA, and as conditions become warmer, and effectively drier, the region will likely experience higher fire activity and severity. Combining these local and regional data sets allows for the creation of a temporal and spatial depiction of how fire regimes, and more specifically fire severity, has changed. This information has important

PAGE 4

iv implications for ecological and forest management goals in northern California and the Northwest coast of the United States The form and content of this abstract are approved. I recommend its publication. Approved: Christy Briles

PAGE 5

v DEDICATION I would like to dedicate this thesis to my loving partner . His support and understanding w ere essential to my success and I am eternally grateful . I would also like to thank my family for instilling a deep love of science and respect for the environment at an early age, and my friends for their encouragement and support .

PAGE 6

vi ACKNOWLEDGEMENTS I would like to thank my advisor, Christy Briles for her guidance, time, insight, and for introducing me to the world of paleoecology. Her dedication to scientific research has always astounded me and I hope to be as devoted as her in my future endeavors. In addition, I would like to thank my other committee members, Peter Anthamatten and Rafael Moreno for their encouragement and support throughout my entire scholastic career a t UC Denver.

PAGE 7

vii TABLE OF CONTENTS CHAPTER I. INTRODUCTION ............................................................................................................................ 1 II. BACKGROUND INFORMATION ................................................................................................. 4 Physical Environment ................................................................................................................. 4 Climate and Vegetation ............................................................................................................... 7 Modern Climate ..................................................................................................................... 7 Modern Forests ...................................................................................................................... 8 Paleoclimate and historical vegetation ................................................................................... 9 Abrupt climate fluctuations .................................................................................................... 9 Disturbance ............................................................................................................................... 10 Fire Regimes ........................................................................................................................ 10 Increasing Fire Activity ........................................................................................................ 12 Regional Trends in Biom ass Burned .................................................................................... 14 III. METHODS AND DATA ANALYSIS ......................................................................................... 15 Site Descriptions ....................................................................................................................... 15 Oak Woodland Zone ............................................................................................................ 16 Mixed Conifer Zone ............................................................................................................. 16 White Fir Zone ..................................................................................................................... 17 Red Fir Zone ........................................................................................................................ 17 Laboratory Methods .................................................................................................................. 19 Charcoal ............................................................................................................................... 20 Pollen ................................................................................................................................... 20 Data Analysis ............................................................................................................................ 20 Chronology ........................................................................................................................... 20 Charcoal Analysis ................................................................................................................ 21 Determining Fire Severity .................................................................................................... 22 Regional Biomass Burned Reconstruction ........................................................................... 23 IV. RESULTS ..................................................................................................................................... 27 Chronology ............................................................................................................................... 27 Vegetation Zone Fire History Reconstructions ......................................................................... 28 Oak Woodlands Zone ........................................................................................................... 28

PAGE 8

viii Mixed Conifer Zone ............................................................................................................. 32 White Fir Zone ..................................................................................................................... 33 Red Fir Zone ........................................................................................................................ 35 Biomass Burned Reconstruction ............................................................................................... 37 5,000year biomass burned trends in the Klamath -Siskiyou Ecoregion .............................. 37 Biomass Burned Trends in the Klamath Siskiyou Ecoregion and Northwestern US .......... 39 V. DISCUSSION ................................................................................................................................ 40 Influences on Fire Variability ................................................................................................... 46 Oak Woodland Vegetation Zone .......................................................................................... 46 Mixed Conifer Vegetation Zone .......................................................................................... 47 White and Red Fir Vegetation Zones ................................................................................... 48 Regional Biomass Burned Patterns and Controls ..................................................................... 49 Spatial Variability of Fire in the Klamath -Siskiyou Ecoregion and Implications for Fire Management .............................................................................................................................. 52 Future research .......................................................................................................................... 54 VI. CONCLUSIONS ........................................................................................................................... 56 REFERENCES ..................................................................................................................................... 58 APPENDIX A . ANALYSISES PERFORMED .................................................................................................... 65 B. CHRONOLOGY FOR ALL SITES ............................................................................................ 66 C. POLLEN PERCENTAGES AND CANOPY-TO UNDERSTORY RATIOS ........................... 67

PAGE 9

ix LIST OF TABLES TABLE 1 . Summary of Site Descriptions ...................................................................................................... 18 2 . Pollen types and associated fire sensitivity used in this study ...................................................... 23 3 . Location, elevations, and citation for the sites used to create regional biomass burned. ............. 26 4 . Radiocarbon dates used to create Kelly and Miller Lake’s age models ....................................... 28 5 . Fire -related sample ratio comparisons to baseline means, and fire type determinations. ............ 35

PAGE 10

x LIST OF FIGURES FIGURE 1. Klamath -Siskiyou Ecoregion .......................................................................................................... 6 2. Site Map ....................................................................................................................................... 16 3 . Vegetation zones and site locations. Dashed lines represent plant species distributions. ........... 19 4a . CHAR, peak magnitude, fire events, fire frequency and pollen ratios of oak woodland and mixed conifer vegetation zones. ...................................................................................................................... 30 4b. CHAR, peak magnitude, fire events, fire frequency and pollen ratios for White and Red Fir vegetation zones. .................................................................................................................................. 32 5a . Biomass burned curves for northern vs southern sites ................................................................. 38 5b. Biomass burned curves for coastal vs inland sites ....................................................................... 38 5c . Klamath -Siskiyou Ecoregion and Northwest US coast trends in biomass burned. ...................... 39 Figure 6: Scatter plot comparing charcoal values for BCHAR and peak magnitude ........................... 41 Figure 7: Biomass burned, fire events, temperature, climatic fluctuations and inferred vegetation .... 46

PAGE 11

1 CH APTER I INTRODUCTION The Klamath -Siskiyou Ecoregion is a rugged series of mountain ranges, straddling the California – Oregon border . It is an ecologically and geographically unique region of almost 100 million acres . It is one of seven a reas of g lobal b otanical s ignificance in North America determined by the International Union of Conservation of Nature , is considered a biodiversity hotspot , and has been the focus of biogeographic research fo r many decades. The region’s complex topography, steep coastal -to inland precipitation gradients, and mixed-severity fire regimes encourage high levels of biodiversity and fire activity is considered essential in maintaining biodiversity (Martin and Sapsis, 1992; Bond and van Wilgen, 1996) . With the region largely wildlands and forested areas, and fire being an integral disturbance element , the change in fire activity in response to future climate change and anthropogenic fi re management is a significant concern (Briles et al. 2011 ; Crawford et al., 2015 , Miller et al., 2009; Odion and Hanson, 2006). The h istorical range of variability of fire has been difficult to examine in the region due to the generally low temporal resolution of historical documents , and short -term records like those from tree rings , satellite, and forestry records. The reconstruction of fire severity has yielded mixed results, leading to debates between fire researchers about whether fire severity is increasing or simply returning to historical conditions (Baker, 2015; Odion and Hansen , 2006; Safford et al., 200 7 ; Westerling, et al., 2006 ). This past summer, the Klamath Mountains experienced several large fires (e.g., the Salmon -August Complex in the Marble Mountains , the Eclipse Complex in the Sisk iyou Mountains , and the Orleans Complex on the western edge of the Marble Mountains between the other two fire complexes) contributing ~8 0,000 hectares to the estimated 570,000 hectares burned in California, Oregon, Washington , and Montana . Questions around whether these and other fires in the past few decades are unprecedented , and whether they are the new normal under projected climate change are being raised within forest man agement , academic, and local western US communities.

PAGE 12

2 Paleoenvironmental records of pollen and charcoal preserved in lake sediments allow for the examination of relationships between fire activity, climate change, and forests. Reconstruction of these histories enables quantification of fire frequency, biomass burned, fire severity and forest changes in response to past climate variability that span millennia . The Klamath Mountains have a sizable number of paleoecological reco rds . The focus of past paleoenvironmental studies have been to examine how vegetation a nd fire activity h ave responded on different substrates to climate variations since the last glacial period (~15,000 years) (Briles et al., 201 1 ), to a ccess regional and local control s on postglacial forest development and fire regimes (Briles et al., 2008 ) and determine late -Holocene human land -use practices on forests and fire regimes at low elevations (Crawford et al., 2015) . In addition, several individua l site studies have document ed the Holocene paleoenvironmental history of the Klamath and Siskiyou Mountains ( Colombaroli and Gavin, 2010; Daniels et al, 2005; Briles et al., 2005; M ohr et al., 2000). The se records present an opportunity to evaluate how local and regional fire regimes, and more specifically fire severity, have fluctuated during the last 5 , 000 years across the heterogenous region. A regional paleoenvironmental study that evaluates fire regime variability, and the controls on fire, has the potential to inform current United S tates F orest Service (USFS) adaptive management plans that recognize the importance of fire in protect ing and enhancing “Old Growth ” forest habitat and the biological diversity (K aufmann et al. 2007; Olson et al., 2012) T he main objective of this research is to obtain a temporal and spatial understanding of fire regimes within the Klamath -Siskiyou Eco region and compare them with other published reconstructions in the western US . The study u tilizes pollen and macroscopic charcoal proxy records from nine sites to explore how fire regimes respond to changes in vegetation and climate . A composite analysis utalizing the nine sites , and other site s in the Global Charcoal Database (an online database of charcoal data) , are com bined along elevational, latitudinal, and a costal -to -inland gradient, based on modern precipitation and vegetation characteristics, and examined for changes in fire activity, biomass burned, and fire severity for the last 5000 years.

PAGE 13

3 The purpose of this s tudy is to address the following questions: (1) What is the spatial a nd temporal variability of fire in the Klamath Siskiyou Ecoregion during the last 5000 years? (2) How has fire severity changed during the last 5000 years in the Klamath -Siskiyou Ecoregion ? The thesis is comp osed of s ix chapters. Chapter t wo provides a background on the physical characteristics o f the region, including climate, geology and fire regimes that maintain the current biodiversity of the Klamath -Siskiyou Ecoregion. Chapter two examines the disturbance regimes within the region , particularly that of fire, along with a discussion of regional debate on fire severity . Chapter three describes the nine sites used in this study , outline s the field and laboratory procedures of data col lection and analyses performed on the lake sediment cores, and the modeling and statistical analyses performed on the radiocarbon, pollen and charcoal data. Chapter four presents the results of the age-depth modeling and charcoal analysis, including composite analyses to reconstruct biomass burned and fire severity estimates using charcoal and pollen data from individual sites . Chapter five discusses the spatial variability of fire in the Klamath Siskiyou Ecoregion along environmental gradients and w hat we can glean about fire disturbance regimes in the region . It also addresses how results of the study can inform forest management strategies and lend to the on going fire debate in the region. Chapter six summarizes the findings of the study.

PAGE 14

4 CHAPTER II BACKGROUND INFORMATION Physical Environment Today the Klamath -Siskiyou Mountain E coregion extends from the northern Sacramento Valley in northern California to the Umpqua Valley in southern Oregon and lies to the east of the California Coast Range (Figure 1) . Within this ecoregion reside the Klamath Mountains and the many subranges that make up the mountainous complex, including the Siskiyou , Marb le, Trinity, Scotts , Salmon, Russian, Trinity, and Yolla -Bolly mountains , which support a wide variety of vegetation, including 39 species and subspecies of con i fers , seven of which are endemic (found nowhere else) to the region ( DellaSala et al., 1999 ). The topographic relief in the Klamath Mountains has a large effect on species distributions. Steep elevational gradients vary from sea level to 2900 meters over a small geographic area creating one of the steepest coastal to -inland precipitation gradients in North America (Franklin and Dyrness, 1988; Whittaker, 1960) . The rugged topography also create s pronounced differences in precipitation, humidity, and temperature on small scales , usually less than one kilometer (Dobrowski et al., 2011). These complexities affect the distribution of species and encourage contrasting biotic communities to thrive near one another. Besides the west -to -east climate gradient, the region is a transition zone between the drier Mediterranean climate to the south and the wetter Pacific Northwest T emperate Rainforest c limate to the north. This result s in many plant species either originating or terminating their ecologic ranges in the Klamath Mountains. Geology also influences plant diversity in the Klamath Mountains . Ultramafic soils , derived from serpentine and peridotite bedrock, have high concentrations of heavy metals and are low in nitrogen, phosphorus, and calcium . Consequently, they are inhabited by endemic xerophytic plants that have evolved ways of concentrating the metals and toxins in their plant tissues or restricting their uptake ( Alexander, 2007). The forests on these ultramafic substrates are open due to the nutrient

PAGE 15

5 restrictions and support a wide diversity of plants ( Kruckeburg, 2002; Alexander, 2007). Other soil types in the region, derived from a range of metasedimenta ry and igneous rocks , do not have the same nutrient restrictions and support different plants , but are of lower diversity than ultramafic substrates. T hese more fertile soils yield more closed forests than the ir ultramafic counterparts. T he type of substrate greatly affect s present plant communities , as well as plant response to abrupt climate changes, such as those seen in the Little Ice Age cooling (LIA, ~500 -100 cal yr B.P ; Grove, 2001) and Medieval Climate Anomaly (MCA, ~1 4 00-9 00 cal yr BP ; Graham et al, 2007). For example, lake records suggest little vegetation response to these climate events on ultramafic substrates, in contrast to rapid responses on other soil types . Fire responded in concert with climate change on all soil types, except when ult ramafic substrates became too open during the MCA to support fire spread (Briles , 2017; Briles et al., 201 1 ). How vegetation will respond to future climatic variations is a point of debate . P redictions are difficult to make for such complex landscapes, but forecasting these responses are necessary to maintain the current biodiversity (Harrison et al., 2010). The lack of high-resolution temporal records in the Pacific Northwest of response to climatic change is noted in several studies (Harrison et al., 2014, Walsh, et al. 2010). For example, Har ris on et al. (2010) note that the Klamath-Siskiyou region, with its steep precipitation and temperature gradients, combined with its unique and endemi c flora, provides an ideal location in which to study the impacts that changing climates may have on future biodiverse temperate communities.

PAGE 16

6 Figure 1: Klamath -Siskiyou Ecoregion

PAGE 17

7 Climate and Vegetation Modern Climate The modern climate of the n orthwest US coast can be generally characterized by warm dry summers and cool wet winters, where fluctuations in the Pacific subtropical high and the Aleutian low -pressure system, and the position of the jet stream (westerlies) , influence precipitation on multiple temporal scales. In the summer, a strengthened and expanded high -pressure system develops off the coast that results in warm conditions with little to no precipitation . In winter a strong lowpressure system pushes the westerlies south and brin gs P acific storms and snow to the mountains. However, due to steep environmental lapse rates in summer, caused by surface warming, variable atmospheric conditions often result in conventional storms that produce light n ing that start wildfires. Ocean curr ents also influence the climate, especially along the coast . Today, the California Current is associated with strong upwelling that brings fog and cool moist conditions to the coastal area and mountains. The westernmost Klamath Mountains, including forests and fire regimes, have been influenced by the fog for millennia ( Briles et al., 2008 ). Large-scale interannual -to -decadal climatic oscillations have also helped to shape the current landscape in this region. The El N io -Southern Oscillation (ENSO) causes fluctuation s in ocean temperatures along the equator influencing the trajectory of storms to the Klamath region and has a cycle of 6 to 18 months . The El Nino phase of ENSO result s in warm ocean conditions along the equatorial eastern Pacific Ocean which causes warmer and drier conditions in the Pacific Northwest and results in reduced snowpack in the winter and spring. The La Nia phase, or cool ocean conditions along the equatorial eastern Pacific Ocean, causes cooler and wetter condition . The pattern of variability in North Pacific s ea surface temperatures, or t he Pacific Decadal Oscillation (PDO) is another climatic phenomenon that can impact oceanic precipitation with phases that can last up to 30 yea rs . During the positive phase of the PDO, s ea surface temperatures (SSTs) increase along the west coast of North America and the Aleutian low pressure system intensifies resulting in warmer

PAGE 18

8 and drier winter conditions in the northwestern North America . C ooler and wetter winters are associated with cooler SSTs and a reduced Aleutian low pressure system, and the negative phase of the PDO (Mantua & Hare, 2002; Newman et al., 2016) . The combination the positive phase of the PDO and the El Nio phase of ENSO extend ed the fire season especially for inland sites within the P acific N orthwest (Gershunov et al., 1999; Hessl et al, 2004). Modern Forests Elevational changes in precipitation and temperature influence plant distributions in the Klamath Mountains (Figure 2 & 3 ). At the highest elevations above 2300 m, the Mountain Hemlock Zone forests are dominated by Tsuga mertensiana (mountain hemlock), Abies magnifica (red fir) and several species of Pinus (pine) and conditions are cold and wet . Fires, although less frequent, are generally of higher intensity and severity than other zones, due to the wet conditions that allow significant fuel buildup , resulting intense and severe fires, with most species in the zone easily killed by fire (Taylor and Skinner, 1998, 2003). In t he Red Fir Zone (2300 to 1900 m ), Tsuga mertensiana dissipates with decreasing elevation and a more diverse assemblage is present in the forest. Abies magnifica, Abies concolor (W hite fir), Pinus contorta (lodgepole pine) and Pinus monticola (western white pine) are the dominant canopy species. Snowpacks remain high and cool summers allow moisture to persist through much of the year ( Franklin and Dyrness, 1988). The White Fir Zone lies between 1900 -1300 m elevation, and Abies concolor generally forms nearly pure stands, but Ps eu dotsuga menziesii (Douglas fir) , Pinus lambertiana (sugar pine) , Pinus ponderosa (ponderosa pine) and Pinus monticola are also present in more minor abundances within this zone. Si gnificant snowpack still occurs in the higher elevations of this zone, and fires occur more regularly than at higher elevations (Agee, 1993 ; Franklin and Dyrness, 1988). The Mixed Conifer Zone, from 13001100 m elevation, consists of P.monticola , Ps eu dotsuga menziesii , P. lambertiana , P. ponderosa, Calocedrus decurrens (California incense cedar), and Abies concolor. There is a gradual warming moving into the lower zones with lower snow packs and an earlier melt in the spring, resulting in

PAGE 19

9 progressivel y drier conditions at lower elevations. The Mixed Evergreen Zone, between 1100-800 m elevations, has a mixture of Ps eu dotsuga menziesii , Pinus ponderosa, Pinus jeffreyi (Jeffrey pine), Calocedrus decurrens, and Lithocarpus densiflora (tanbark oak) and are more open. Fire is more frequent at lower elevations and are generally of low severity ( Skinner and Taylor, 1998). The lowest elevations, 800m and lower, consist of oak woodland ( Quercus spp. ) and receive little if any snow, tend to be very dry in the s ummer months , and generally have low severity, high -frequency fire regimes (Taylor and Skinner 1998, Franklin and Dyrness, 1988). Paleoclimate and historical vegetation The paleoclimate of the region was driven by long -term changes in incoming solar radiation and the strength and movement of the pressure systems and ocean currents. During the m iddle (8 200 to 4 500 cal yr B.P) and l ate Holocene (4 500 cal yr B.P. to present) , summer insolation declined throughout the western United States, which resulted in cooler summers with increased precipitation (Bartlein et al. , 1998). Paleoecological data collected from the Klamath Mountains and throughout the Pacific Northwest, confirms that transition to a wetter and cooler climate began around 4 500 cal yr BP , although vegetation and fire responses are not synchronous ( Briles et al., 2005 ). Ocean u pwelling increased between 5200 and 3500 cal yr BP creating wetter conditions and increased fog production, as reflected in the increase in cool water diatoms and the presence of redwood pollen in oceanic sediment cores ( Barron et al., 2003). These cooler and wetter conditions lead to forest becoming more closed where Abies, Pseudotsuga, and Tsuga became more prominent ( Briles et al. , 2005). By 2000 ca l yr B.P. , the cooler and wetter conditions resulted in the establishment modern forest communities (Briles et al., 2011 ; Crawford et al, 2015 ). Abrupt climate fluctuation s There have been two abrupt climate change periods identified within the late Holocene including the Little Ice Age (LIA, ~500100 cal yr B.P ; Grove, 2001), where the climate was cooler

PAGE 20

10 and wetter than today, and Medieval Climate Anomaly (MCA, ~1400-900 cal yr BP ; Graham et al, 2007), where the climate was warmer and drier . These events have been attributed to changes in sunspot activity, volcanism, and long-term changes in ocean -atmospheric dynamics that led to prolonged climate conditions that had noticea ble impacts on ecosystems and fire regimes ( Deng et al. 2007; Mann et al. 2009). Both short -term climate events had effects on the modern vegetation and fire regimes seen in this region. In the Klamath Mountains , the cool wet conditions during the LIA resulted in less charcoal influx into the lake system , increased biomass accumulation in the forests, and more shade -tolerant trees and shrubs. The warm conditions during the MCA led to increases in charcoal accumulations, fuels biomass in the forest decr eased due to more frequent fires, an d shadeintolerant species became more prominent (Briles et al., 2008 and 2017; White et al.,2015; Crawford et al. 2015 ). Disturbance Forest d isturbance consists of an event that impacts the landscape or alters the ph ysical structure of the vegetation . The effect of d isturbances on the environment can last a few years to centuries. A s climate changes, there are corresponding changes in the disturbance regime of an area (Westerling, 2006 and Minckley and Long, 201 6 ). Disturbance within forest ecosystems can present in many forms ( Baker, 1992). Pine beetles, large weather events, flooding, erosion, avalanches, landslides, and fire can change the current structure of the forest as well as how the forest respond s to fut ure disturbances ( Fried et al, 2004 and Higuera et al. 2014). Fire Regimes A forest ’ s fire regime , or the pattern of fire occurrences through time , is the result of the composition of the surrounding vegetation, the amount of fuels present, time since t he last disturbance, the type of anthropogenic management, and climate. Historical and current fire regimes can provide insights into how the vegetation within a forest will respond to climatic changes in the

PAGE 21

11 future and the major controls driving those changes . Climate also influences the structure and composition of the vegetation, which is correlated with the severity, size, and frequency of fires . Hi gh-severity fire usually remove s most aboveground vegetation an d replaces entire stands. Mixed severity regimes result in fires of varying severity and are the most common type (Sommers, 2011) Lakes are repositories of paleoenvironmental data. During a fire event, and for some time after, ash and charred particles are deposited into the lake through airborne fallout. Streamflow , runoff , and other disturbance events (i.e. landslides) continue to introduce charcoal to t he lake sediments for months to years after a fire event. This deposition of charcoal and other disturbance proxies, through time , allows for the reconstruction of the surrounding areas fire regime (Whitlock, 2004) . Macroscopic charcoal (>125 microns) is a common proxy used to reconstruct fire activity and biomass burned. T he mechanisms that introduce charcoal into the lake system can vary from one lake to another due to the size of the lake and surrounding topography . Fire regimes are subject to man y environmental drivers , but it is generally understood that the main drivers are fuel loads and climate. F uel -driven regimes can be the result of the varying levels of biomass produced, but can also be the product of anthropogenic clearing of debris from forest floors and other suppression methods . Climate -driven regimes are more controlled by temperature and precipitation ( Sommers, 2011). Northern California and s outhern Oregon forest have a mixture of fuels-driven and climate -driven fire regimes. In the forests closer to the ocean, especially at high elevations, climate limits fire activity even though there are sufficient fuels, due to high moisture content both in the atmosphere and in the fuels. At lower elevations and more inland open forests , fire regimes can be fuel -limited (Steel et al., 2015 ). In the western US, both rain and snowfall winter precipitation, are major contributor to annual precipitation (Mock, 1996). El Nio variability can have large impacts on winter precipitation and snowpack, especially in conjunction with the positive phase of the Pacific Decadal Oscillation (Taylor and Beaty, 2005).

PAGE 22

12 Increasing Fire Activity Due to anthropogenic climate change , the western US is experiencing increased fire activity and area burned per individual fire since the late 20th century (Baker 2015; Marlon et al., 2008; Westerling, 2006). F ire regime changes are attributed to lower than aver age winter precipitation coupled with snowmelt occurring earlier in the spring, and warmer and drier summers ( Westerling, 2016). There is debate among fire researchers as to how drastically fire in the western US is changing in response to our current cli mate change . Westerling et al. (2006) reported that in the mid-eighties the wildfire season was extended and has continued to elongate since . M id -elevation forest s of the Northern Rockies were most impact ed by the increased summer and spring temperatures and extended wildfire season . Areas that experience short snow -free seasons and high evapotranspiration are considered at the highest risk for increased wildfire activity and larger fires (Westerling et al.2006) The status of high severity fires is under debate in the wester US. Odion and Hanson (2006) found that high severity fires in the Sierra Nevada Mountains have not been a common occurrence, and suggest introducing more fire into management strategies. In their 2014 article Hanson and Odion state that since 1984, modern high-severity fires have not increased in proportion, area, or patch size in western North America according to Burned Area Emergency Rehabilitation (BAER) fire -severity data. They suggest that climate change might not have as larg e of an impact on fire severity in the future as others have forecasted (Hanson and Odion, 2014). Baker (2015) suggests that, at least in dry forests, rates of high severity fire are increasing over historical rates, but are not doing so at the exceptiona lly high rates that others have indicated (Baker, 2015) . According to the majority of studies, high severity fires are increasing in frequency over historical rates (Colombaroli and Gavin, 2010; Higuera et al., 2014 ; Miller et al., 2009 and others ). Th e increase in high severity fires has been partly attributed to anthropogenic fire suppression , which has led to the accumulation of fuels on the forest floor. The effects of fire suppression have changed the structure of many

PAGE 23

13 landscapes since the mid -19th-century settlement of the western US (Nagel and Taylor, 2005). Historically, high-elevation forest s in the K la math -Siskiyou E coregion, that receive high amounts of precipitation , had infrequent but sever fires . T hese “wet” forest s have not experienced an increase in severity , and are within their historical range of variability . High severity fires at drier l ow and middle elevation sites, that historically burned more frequently but at lower severity than wetter sites, are burning more severely compared to pre -Euro -American settlement rates (Miller et al. 2009). The rate at which high severity fires are increasing can change depending on forest composition. Stephens et al. (2013) found that if current climate models are accurate the planets for est ecosystems will experience large-scale changes in structure and composition in the next few decades (Stephens et al., 2013) . While most research indicate s increasing severity , a limitation of most of the studies is that the d ata is based on current remotely sensed images, that only extend back to the 1980’s , forestry records that only extend back a century , or dendrochronolo g y records that only extend back ~ 500 years. T hese records do not examine prior to the LIA wh e n climatic conditions were simil ar to conditions we are encountering today . A better understanding of the trends observed in fire severity might be gleaned from identifying high severity fire from paleoecological data . In addition, late Holocene paleoclimate data are available for determining climate variability and can be used to compare with charcoal -based fire history reconstructions . Reconstruction severity from macroscopic charcoal records is in its infancy . Studies on r elationships between charcoal peak height and area burned to suggest severity using a combination of dendrochronology and lake sediment data have been conducted (Whitlock et al., 2004; Higuera et al. 2011). Other studies have identified high severity fires using changes in pollen spectra following fire events and compared them to pollen spectra of fire -free periods (Minckley and Shriver 2011; Minckley and Long, 2016) .

PAGE 24

14 Regional Trends in Biomass Burned Several studies have compiled lake sediment charcoal records from around the world to examine global -to -regional fire patterns and their drivers. Power et al. (2008) examined biomass burned across the globe and on a continental scale since the last glacial maximum. Their findings indicate that global biomass burned increased through time and this increase corresponds to increases in global summer insolation . Marlon et al. (2012) examined biomass burned throughout the last 3,00 0 years and determined that biomass burned has generally declined but in periods of warmer and drier conditions like during the MCA and within the last 100 years large peaks in biomass burned occurred. In the P acific N orthwest , Walsh et al. (2015) have reconstructed biomass burned t hrough the Holocene. The late Holocene began with increasing biomass burned contrary to the cooler wetter climate during this time until 600 cal yr BP and has decreased since (Walsh et al., 2015). V ery few studies examine controls on regional biomass burned, especially in mountainous environments or along steep environmental gradients .

PAGE 25

15 CHAPTER III METHODS AND DATA ANALYSIS Site Descriptions The sites used in this study were selected because field methods used to collect sediment cores and data were similar, in addition to the sites geographical locations within the Klamath Siskiyou Ecoregion. Each lake is described by its physical attributes as well as the attributes of the surrounding forest. All these descriptions can be refer enced in Table 1. Several other sites had appropriate charcoal records, but were excluded from the study due to their ultramafic soils, size, or lack of available pollen data. However, charcoal records from ultramafic site s from the Global Charcoal Datab ase ( Cedar, Bluff, and Crater ) were included in the composite analyses as Briles et al. (2012) found that fire on ultramafic and other soils showed similar responses to climate. Squaw Lake (Colombaroli and Gavin, 2010) was left out due to its large size, i nflowing stream, and large charcoal catchment area. All sites discussed here were relatively small is size (<17 ha) and have minimal or lack stream inputs .

PAGE 26

16 Figure 2 : Site Map . Location of lakes (blue circles) used in the Klamath -Siskiyou Ecoregion with average annual precipitation from 19812010 (PRISM Climate Group , 2014). Cities are identified with black hexagons. Oak Woodland Zone Fish Lake is the lowest and westernmost site in the study area and lies within the Six Rivers National Forest . Ogaromtoc Lake is within the Klamath National Forest . The climate at both lakes consists of warm summers and mild winters and precipitation comes mainly in the form of fall and winter rain . Alnus, Cupressaceae Pseudotsuga menziesii, Lithocarpus , and Pinus ponderso a a re the dominant trees making up the open forest around the lake s (Crawford et al., 2015) Mixed Conifer Zone Kelly Lake is the more coastal of the sites in the mixed conifer zone and lies in the Klamath National Forest. The climate consists of warm summers and mild winters. Precipitation comes in

PAGE 27

17 both rain and snow. The closed forests around Kelly Lake consist mainly of Abies concolor, Pinus lambertiana, and Ps eu dotsuga menziesii . Hobart Lake is the easternmost site in this study and lies in the Cascade-Siskiyou National Monument . The climate consists of mild summers and cool winters. Precipitation comes in both rain and snow , more than half of which falls in winter . The forests around Hobart Lake are more open than sites at the same or higher elevations and Pinus ponderosa, Abies concolor, and Pseudotsuga menziesii are the dominant forest species. (White et al. 2015). White Fir Zone Sanger Lak e is in the Six Rivers National Forest, Miller Lake lies within the Rogue River National Forest , and Bolan Lake is just north of the California -Oregon border within the Siskiyou National Forest. Campbell Lake is the southernmost site in the zone in the Klamath National Forest and is in the Marble Mountain Wilderness. The summers at these sites are generally mild and winters are cool and wet, with more than half of the annual precipitation in the form of winter snows. The forests are around these lakes closes as elevation increases, and the dominant trees consist of Ab ies concolor, Abies magnifica , Picea breweriana (Brewer spruce) , Pinus monticola , and Pseudotsuga menziesii (Briles et al., 2008 ; Briles et al., 2011 ). Red Fir Zone Taylor Lake is the highest in elevation of all the study sites and lies within the Klamath National Forest and Russian Wilderness Area. The climate consists of cool summers and co ld winters with m ore than half the annual precipitation in the form of snow . The closed forest around Taylor Lake consists of Abies concolor, Abies magnifica, Pinus monticola, and Tsuga mertensiana (Briles et al, 2008).

PAGE 28

18 Table 1: Summary of Site Descriptions Name County, State Lat , Long Elev. (m) Size(ha)/ Depth (m) Precip.* (mm) Temp* (C; min, max) Reference Fish Lake Humbolt, CA 41 15.813' N, 123 41.014' W 552 9.8 / 13 1541 6, 21 Crawford et al. (2015) Ogaromtoc Lake Siskiyou, CA 4 1 29.167' N, 12332.474' W 594 1.5 / 6.3 1529 5, 21 Crawford et al. (2015) Kelly Lake Siskiyou, CA 4154.795' N, 12331.040' W 1328 4 / 5.17 1813 5, 15 This study Hobart Lake Jackson, OR 42 5. 866' N, 122 28.848' W 1463 5.3 / 3.61 765 4, 14 White et al. (2015) Sanger Lake Del Norte, CA 41 54.106' N, 12338.840' W 1558 2.5 / 7.19 2647 3, 14 Briles et al. (2008, 2011) Miller Lake Josephine, OR 42 3.848' N, 12318.183' W 1585 2.5 / 7.70 1590 2, 12 This study Bolan Lake Josephine, OR 42 1.331' N, 123 27.594' W 1640 5 / 11.25 1697 2, 12 Briles et al. (2008, 2011) Campbell Lake Siskiyou, CA 41 31.998' N, 123 6.333' W 1756 17 / 7.12 1509 2,14 Briles et al. (2011) Taylor Lake Siskiyou, CA 41 21.674' N, 12258.102' W 1981 6.5 / 9.45 1351 2, 11 Briles et al. (2011) * 30 yr average p recipitation and temperature from 1981-2010 (PRISM Cl imate Group, 2014)

PAGE 29

19 Figure 3 : Vegetation zones and site locations. Dashed lines represent plant species distributions. Solid vertical lines indicate species dominance and dashed lines indicate species is present (based on Agee (1993), Briles et al. (2005) and Franklin and Dyrness (1988)). Laboratory M ethods Cores were sectioned longitudinally in 1 -cm increments and stored and refrigerated in 2 -oz Whirl -Pak bags. While sectioning cores, macrofossils were collected to perform AMS -radiocarbon dating. These macrofossils were processed and sent to the Livermore National Laboratory, Center for Accelerator Mass Spectrometry. Five wood samples from Kelly Lake and four wood samples from Miller Lake were sent for dating .

PAGE 30

20 Charcoal Macroscopic charcoal was used to reconstruct local fire histories and to estimate regional biomass burned. Two -cm3 samples for Miller and Kelly l akes were processed every cm. Samples were disaggregated and soaked in house hold bleach for a minimum of 24 hours as outlined in Whitlock and Larsen (2001), washed through a 125micron metal mesh sieve, and transferred to a gridded petri -dish and all charcoal pieces counted using a stereomicroscope. Pollen Analysis of the presence and quantity of pollen provides information on local vegetation history and forest structure. The temporal resolution at which pollen was sampled varies between sites, with an average of every 125 years. At Kelly and Miller lakes pollen was sampled on average every 100 years. A total of 42 samples were processed for Kelly Lake and 43 samples for Miller Lake. Removal of organics with potassium hydroxide (KOH), silicates with hydrofluoric acid (HF), and insoluble organics using acetolysis (a 6:1mixture of acetic anhydride to sulfuric acid) was conducted following the methods outlined by the LacCore Pollen Preparation Procedure (University of Minnesota ; http://lrc.geo.umn.edu/laccore ). Lycopodium tracer s were added to aid in calcul ating pollen concentrations (grains cm-3). Grains were counted at 500 to 1250x magnification and identified to the lowest taxonomic level possible with the use of atlases and reference collections (Jones et al., 1995; Kapp et al., 2000). Modern phytogeography was used in determining possible genera or species occurring at different elevations. Pinus grains were separated into Haploxylon (moticola -type) and Diploxylon ( ponderosa-type), and all other s were classified as undifferentiated. Data A nalysis Chronology Age-depth models have evolved in the last decade from linear/polynomial regressions to more iterative models that allow each date and its associated error to influence the reconstruction of

PAGE 31

21 sediment accumulation in lakes through time . Radiocarbon and 210Pb age dates, as well as tephrochronology ( using the Mazama and Little Glass Mountain eruptions, which occurred at 7627 150 and 986 200 cal. yr BP respectively) , were compiled from previously conducted field collections (Briles et al., 2005, 2008; Crawford et al., 2015 ; White et al., 2015 ) and new dates for Kelly and Miller lakes. Age -depth models were created using a Bayesian accumulation model called BACON , an R-based statistical package (Blaauw and Christen, 2011) , and using the IntC al13 radiocarbon calibration curve (Reimer et al., 2013) . BACON assumes a monotonic rate of accumulation and uses a gamma autoregressive process to model sedimentation rates as a function of depth. A ‘shift outlier ’ method, is used to identify outliers. A Markov Chain Monte Carlo “t -walk” algorithm is used in producing the posterior distributions (Christen, 1994; Blaauw and Christen 2001 ). The models were run accepting recommended parameters for memory strength (4) and mem ory mean (0.7) . A ccumulation rate mean s, and the number of sections analyzed were determined by the program after making initial estimates. Charcoal Analysis To determine fire history , c harcoal accumulation rates (CHAR) were determined using the methods described in Higuera et al. (2009) and the program CharAnalysis (http://phiguera.github.io/CharAnalysis). Counts, volume , and depths were interpolated to the median sample resolution (yr sample-1 ) for each site. CHAR (particles cm-2 yr-1) are the product of the interpolated charcoal concentrations (particles cm-3) and the interpolated sedimentation rate (cm yr-1). Only the last 5,000 years of data w as analyzed to provide baseline conditions duri ng the l ate Holocene. CharAnalysis also calculates background charcoal using a tricubic locally -weighted regression and all sites smoothed using a 700year window. Background charcoal ( BCHAR), which represents levels of fuel biomass (Higuera et al. 2007), is useful in identifying charcoal peaks that are

PAGE 32

22 then interpreted as local fire episodes. When BCHAR is removed, two components of CHAR remain: (1) the residuals or peaks that represent local fire episodes within the watershed of the lake, and (2) the variability not related to fire, or natural noise, typically from charcoal stored in the watershed or from regional fires. A signal -to -noise index is applied to equally spaced intervals and indicates where CH AR surpasses random noise or where fire events are detected. The noise distribution was determined with a Gaussian mixture model with a locally -defined threshold to determine peak charcoal event s and rep resent fire events. Peak s were then screened using a Poisson -distribution, which identifies single fire events within multiple peaks . Peak magnitude (particles cm-2 peak-1) is the sum of positive CHAR threshold values and is thought to be related to fire severity and/or charcoal transport and delivery in t o the lake system (Whitlock et al., 2004; Higuera et al., 2007) . Determining Fire Severity The identification of high severity fires in lake sediment charcoal records is not well established. Examining plant community -level changes using pollen abundances post -fire can provide a more robust determination of fire severity , as high severity fires consume the larger trees and biomass in the forest . To gauge community -level responses to disturbances in h istorical forest s, canopy -to -understor y ratios of pollen percentages were used . The ratio for canopy pollen species, a , to understory, b, was calculated as ( a -b )/(a+b ) in order to normalize values (Jimenez -Moreno et al., 2010). Canopy -to -understory ratios were grouped based on their proximity to an identified fire event within the lake sediment core. Fire pollen ratios came from the same or the preceding pollen sample depth that a fire event that was identified in the charcoal record. All other samples were considered non-fire polle n ratios, were summed, averaged, and used to identify a non-fire pollen ratio mean to characterize vegetation composition during intervals where no fires events occurred (as outlined in Min c kley and Shriver (2011) ). This study added metric ratios for fire -sensitive and fire -adapted species (Table 2 ; USDA Fire Effects Information System (FEIS) www.feis crs.org/feis/) , in addition

PAGE 33

23 to the original ratio for all pollen types, to further validate or invalidat e the characterizations of fire types done with all pollen types. Only tree or shrub species were used , as herbs are limited in the pollen record and riparian / aquatic species typically are spared from fires due to the proximity to the lake. Fire pollen ratios were then compared to the previously identified non-fire pollen ratio mean. If a fire pollen ratio for all pollen types was below the non-fire pollen mean then the corresponding fire event was characterized a s a canopy fire . This assumes that understory taxa will re -establish at a higher rate post -fire relative to canopy taxa, resulting in a lower canopy -to understory ratio after a fire event (Minckley and Shriver, 2011). These are not the only canopy fires that occurred, but only the fires that this stu dy was able to identify, due to the coarsely sampled pollen records. However, when all records are analyzed together, given the even interval pollen sampling methods, a regional burn severity record can be reconstructed . Microsoft Excel was used to calcul ate ratios and C2 software version 1.7.7 (Steven Juggins, U of Newcastle, UK) was used to graphically display the data. Table 2 : Pollen types and associated fire sensitivity used in this study Regional Biomass Burned R econstruction To observe changes in fire activity over an entire region a synthesis of charcoal data was conducted. This synthesis provides information on how biomass burned has changed through time . Fire Sensitive Canopy Pollen Taxa Fire -Adapte d Canopy Pollen Taxa Fire -Sensitive Understory Pollen Taxa Fire -Adapted Understory Pollen Taxa Abies Alnus rubra-type Cercocarpus/Purshia Amelanchier Acer macrophyllum Cupressaceae Rosaceae Ceanothus Betula Pinus monticola -type Salix Chyrsolepis /Lithocarpus Corylus Pinus Ponderosa-type Quercus vaccinifolia type Picea Pseudotsuga Sarcobatus Tsuga heterophylla Quercus Large Spiraea Tsuga mertensiana

PAGE 34

24 Two levels of biomass burned reconstructions were conducted . 1) W ithin the Klamath Siskiyou E coregion, north -to -south and coastal -to -inland gradients were compared. These comparisons were conducted with the nine sites discussed in detail in the study , in addition to four other sites from the Global Charcoal Database, Bluff and Crater Lake CA (Mohr et al., 2000), Mumbo ( Daniels et al., 2005) , and Cedar (Briles et al. 2008), that are located in the southern Klamath Mountains . T he sites were broken into northern (n=5 ) and southern (n=8 ) regions , with the northern region north of 41.5 L atitude . C oastal (n= 5 ) and inland ( n=8 ) sites with were located west of 123.2 L ongitude . 2) The nine sites used in this study were combined with 3 2 ( Table 3 ) other sites from th e Global Charcoal Database to create biomass burned curves for the northwest US coast to examine how trends observed within the Klamath -Siskiyou Ecoregion compared with trends at a larger spatial scale. Both sets of biomass burned reconstruction s used the same methods and utilized the paleofire R-software package. All sites were standardized using a protocol designed by Blarquez et al. (2014), using four steps. (1) Raw charcoal counts was rescaled to values from 0 to 1 with a minimax transformation. (2) Variances were homogenized with a Bo x -Cox transformation due to the long upper tail skewness that is inherently present in charcoal data. (3) Z -scores were then determined so the mean and variance were equivalent across all sites (Power et al., 2008) . An additional minimax rescaling was performed after the Box -Cox step to e nsure that all series had equal values (Marlon et al., 2008). To create the regional mean charcoal composite or the smoothed mean value for charcoal across all site s for a given timeframe ( i.e. b iomass b urned), a two -step data-binning sequence was used . First, the charcoal series was prebinned with 10 year non-overlapping bins , so that all records have the same influence on the composite. Second , the pre -binned series was smoothed with a “LOWESS” locally weighted smoother and a half width 500 year window. The pre -binned series is then resampled using a bootstrap approach and a LOWESS curve fitting is applied to calculate the confidence intervals (Blarquez et al., 2014).

PAGE 35

25 The n orthwestern US coast regional biomass burned reconstruction s was det ermined using a rectangular area from 40 to 5 0 North and -1 18 to 125S and the 32 Global Charcoal Database sites within were selected space (Table 3) . Records for Bolan, Campbell Lake CA, Taylor lake CA, and Sanger Lake CA are in the G lobal Charcoal Database, but were excluded as the lakes received updated age-depth chronolog ies. Instead the four sites were added as user -defined data with their updated age -depth models .

PAGE 36

26 Table 3: Location, elevations, and citation for the s ites used to create regio nal biomass burned for the northwest US coast. Site Lat /long Elevation Reference Bluff 41.340, 122.550 1921 Mohr et al.(2000) Barrett Lake 37.595, 119.007 2816 Hallett et al.(2002) Battle Ground Lake 45.800, 122.492 154 Walsh et al.(2002) Beaver2 44.918, 123.305 69 Walsh et al.(2002) Cedar 41.2 00, 122.490 1740 Briles et al. ( 2011 ) Coast Trail Pond 37.985, 122.800 230 Walsh et al.(2002) Crater Lake CA 41.384, 122.578 2288 Mohr et al.(2000) Dead Horse Lake 42.561, 120.778 2248 Minckley et al.(2007) East Lake 37.178, 119.028 2863 Power et al.(2002) East Sooke Fen 48.352, 123.682 155 Brown et al.(2002) Five Lakes 48.082, 118.929 780 Scharf et al.(2002) Glenmire 37.993, 122.777 399 Anderson et al.(2002) Lake Oswego 45.411, 122.667 30 Walsh et al.(2002) Lily Lake 41.976, 120.210 2042 Minckley et al.(2007) Little Lake 44.168, 123.584 703 Long et al.(2002) Lost 45.824, 123.579 449 Long et al.(2007) Lower Gaylor Lake 37.909, 119.286 3062 Hallett et al.(2002) Martins 47.714, 123.540 1415 Gavin et al.(2001) Moose 47.883, 123.350 1508 Gavin et al.(2001) Mumbo 41.191, 122.509 1860 Daniels et al.(2005) Patterson Lake 41.388, 120.224 2743 Minckley et al.(2007) Porphyry 48.906, 123.883 1100 Brown et al.(2002) Porter Lake 44.448, 123.243 73 Walsh et al.(2002) Siesta Lake 37.850, 119.667 2430 Brunelle et al.(2002) Swamp Lake 37.950, 119.817 1554 Smith et al.(1992) Taylor 46.101, 123.907 6 Long et al.(2002) Three Creeks 44.099, 121.627 1996 Long (unpublished) Todd Lake 44.028, 121.685 1875 Long (unpublished) Tumalo Lake 44.022, 121.544 1536 Long et al.(2002) Upper Squaw Lake 42.033, -123.015 930 Colombaroli et al.(2002) Warner Lake 44.246, 122.958 590 Walsh et al.(2002) Wildcat Lake 37.968, 122.785 67 Anderson et al.(2002)

PAGE 37

27 CHAPTER VI R ESULTS Charcoal and pollen results for Kelly and Miller lakes have yet to be published and are discussed below in detail. Data for Bolan, Campbell, Fish, Hobart, Ogaromtoc, Sanger, and Taylor Lakes can be found in the publications listed in Table 1. In this chapter , results from chronology, pollen and fire history reconstructions of all sites are discussed. Chronology Age determinations based on 14C accelerated mass spectrometry dates and results for Kelly and Miller lakes are shown in Table 2 . D ates from previous ly published sites were used to create new age models (Appendix A ) using BACON (Bayesian age-depth modeling) , as the previous m odels were determined using older less robust methods (Ramsey, 2007). Age differences between the new chronologies and previously determined age models averaged around 60 years , in most cases within the error of individual radiocarbon dates, and do not change previously published int erpretation s of the data.

PAGE 38

28 Table 4 : Radiocarbon dates used to create Kelly and Miller Lake’ s age models. Single “best” Model, m in and m ax are outputs of Bacon and are reported in cal yr BP. *C14 dates were processed and run at the Lawrence Livermore National Laboratory in July of 2011 Vegetation Zone Fire History Recon structions Oak Woodlands Zone Fish Lake, has the highest average charcoal accumulation rates ( CHAR ) (6 .84 particles cm-2 yr1) of all sites (Figure 4 a) . Background charcoal influx (BCHAR), is initially 2.2 partic les cm-2 yr-1 , reaches a maximum of 9.8 particles cm2 yr1 around 1 400 cal yr BP , and decreases to 3.8 particles cm2 yr1 thereafter . A smaller peak of 4.8 particles cm-2 yr1 occurs around 750 cal yr BP . Between 3000 and 500 cal yr BP peak magnitude averages ~80 particles cm2 peak -1 and increases to ~1000 cm2 peak -1 at present . Fire frequency is highest at the beginning of the record (3000 cal yr BP) at ~17 Depth (m) Uncalibrated 14 C Age (14C yr B.P.) Single "best" Model ( w mean ) Min Max Material dated Lab. Reference* Kelly Lake 62 800 30 721 667 810 Plant/Wood KL11B 523 523.5 164 2030 35 2000 1881 2128 Plant/Wood KL11B 625 625.5 229.5 2820 25 2922 2802 3064 Plant/Wood KL11B 690.5 693 259 3300 25 3502 3362 3631 Plant/Wood KL11B 720 720.5 363.5 5960 45 6709 6425 6920 Plant/Wood KL11B 824.5 826 Miller Lake 34 510 30 352 88 552 Plant/Wood ML10 128.5 130 128.5 1460 20 977 627 1375 Plant/Wood ML11B 1106 1107.5 134.5 770 20 1017 667 1423 Plant/Wood ML11B 911.5 227 2110 30 2123 1964 2311 Plant/Wood ML11B 1004 1009 329 3590 35 3797 3545 3972 Plant/Wood ML11B 811 811.5

PAGE 39

29 fires per 1000 years, decreases from 3000-450 cal yr BP to ~ 6 fires per 1000 years , and increases slightly to ~ 7 fires per 1000 years in the last 450 years. At Ogaromtoc Lake BCHAR slowly increases from 0.19 particles cm-2 yr-1 at 3 700 cal yr BP to a maximum of 1.5 particles cm-2 yr-1 at 1200 cal yr BP and decreases to 0.6 particles cm-2 yr-1 at present . Between 3600 and 2000 cal yr BP peak magnitude averages ~10 particles cm2 peak 1, decreases from 2000 to 500 cal yr BP averaging ~ 4 cm2 peak -1, and increases to ~ 10 cm2 peak -1 at present . Fire frequency at Ogaromtoc Lake increas e s from ~1 fire per 1000 years at 3700 cal yr BP to ~7 fires per 1000 years around 2200 cal yr BP. Frequency remains stable, averaging ~6 fires per 1000 years, from 2200 cal yr BP until 400 cal yr BP , then increases to ~7 fires per 1000 years at present . At Fish and Ogaromtoc Lakes the dominant canopy pollen types that contributed most to the understory -to -canopy ratios were from fire-adapted Alnus rubra-type and Quercus with secondary pollen contributions of fire adapted Pinus ponderosa-type and Ps eu dotsuga menziesii . Understory types were based on pollen from fire -adapted Chyrsolepis /Lithocarpus . At Fish Lake 35 pollen samples were processed . T hree samples (126 cm, 904 cal yr BP; 151cm, 1139 cal yr BP; 215 cm, 1599 cal yr BP) have fire pollen ratios lower than the non -fire pollen ratio mean in all and fire adapted pollen types , identify them as canopy fires (Table 5). The non-fire ratio mean at Fish Lake (0.66) is lower than most sites. Of the 33 poll en samples processed at Ogaromtoc Lake, one sample (367 cm, 2549 cal yr) has a fire pollen ratio lower than the non-fire pollen ratio mean in all and fire adapted pollen types , identif ying it a s a canopy fire . At both sites, there are more negative fire s en sitive non fire and fire pollen ratios than at the other sites. The non-fire ratio mean at Ogaromtoc Lake (0.71) is average when compaired to other sites.

PAGE 40

30 Figure 4 a: CHAR, peak magnitude, fire events, fire frequency and pollen ratios of oak woodland and mixed conifer vegetation zones . Canopy-to -understory non-fire pollen ratios (grey) with non-fire pollen ratio mean (black vertical line) . F ire -pollen ratios that correspond with a canopy fire are represented by red bars and those that correspond with understory fires are represented by yellow bars. Dotted lines connect fire event s to corresponding fire -pollen ratio.

PAGE 41

31

PAGE 42

32 Figure 4b: CHAR, peak magnitude, fire events, fire frequency and pollen ratios for White and Red Fir vegetation zones. Canopy -to -understory non-fire pollen ratios (grey) with non -fire -pollen ratio mean (black vertical line) . Fire -pollen ratios that correspond with a canopy fire are represented by red bars and those that correspond with understory fires are represented by yellow bars. Dotted lines connect fire events to corresponding fire -pollen ratio. Mixed Conifer Zone BCHAR at Kelly Lake is ~ 0.1 particles cm-2 yr-1 from 5 000 cal yr BP to 3 5 00 cal yr BP, increase s to 0.3 particles cm-2 yr-1 by 2 6 00 cal yr BP and remains stable ~0.36 particles cm-2 yr1 thereafter . Between 5000 and 600 cal yr BP peak magnitude averages ~2 particles cm2 peak -1 and increases to ~40 particles cm2 peak -1 at present. Fire f requency at Kelly Lake was highest, ~17 fires per 1000 year s at 3 000 cal yr BP , decreases until ~2 5 00 cal yr BP and has remains stable until present at ~ 9 fires per 1000 years. BCH AR at Hobart Lake declines during the last 5,000 years from a maximum of 2.9 particles cm-2 yr1 at 4 400 cal yr BP to 0.4 particles cm-2 yr1 at present . Between 5000 and 4000 cal yr BP peak magnitude averages ~35 particles cm2 peak -1, decreases from 4000 to 1000 cal yr BP to ~11 cm2 peak -1, and increases to ~28 particles cm2 peak -1 at present . Fire frequen cy at Hobart is high at 5000 years with ~12 fires per 1000 year then declines until 3 000 cal yr BP . Frequency r emains stable at ~ 6 fires per 1000 years from 3000 to 1200 cal year BP then increases to ~ 9 fires per 1000 years from the MCA to present. At Kelly and Hobart lakes, the dominant canopy pollen types that contributed most to the understory -to -canopy ratios were from f ire -adapted Pinus monticola, Ps eu dotsuga menziesii , and fire sensitive Abies . Secondary contributions of fire -adapted Pinus ponder osa, Alnus rubra-type and Quercus were observed. Pollen from f ire -adapted Quercus vaccinifolia -type was the dominant understory taxa at Kelly Lake and fire-sensitive Salix was dominant at Hobart Lake. Of the 42 processed pollen samples at Kelly Lake, five samples (207 cm, 2609 cal yr BP; 239 cm, 3097 cal yr BP; 247 cm, 3257 cal yr BP; 255 cm, 3427 cal yr BP; 279 cm, 4063 cal yr BP) have fire pollen ratios

PAGE 43

33 lower than the non-fire pollen ratio mean in all and fire -adapted pollen types identifying them as canopy fires. The non-fire ratio mean at Kelly Lake (0.78) is average when compared to other sites. At Hobart Lake, of the 42 processed pollen samples , four samples (207 cm, 2785 cal yr BP; 239 cm, 3907 cal yr BP; 247 cm, 4454 cal yr BP; 255 cm, 4657 cal yr BP ) were identif ied as canopy fires. The non-fire ratio mean at Hobart Lake (0.91) is the highest of all sites. White Fir Zone BCHAR at Sanger Lake averages 0.69 particles cm-2 yr-1 and was stable through time reaching a maximum value of 0.96 particles cm-2 yr-1 around 3500 cal yr BP (Figure 4b) . Between 5000 and 3000 cal yr BP peak magnitude averages ~100 particles cm2 peak -1, d ro p s to ~50 particles cm2 peaks-1 until 1 50 0 cal yr BP, increases to ~220 cm2 peaks-1 during the MCA and decreases to ~50 particles cm2 peaks-1 at present . From 5000 to 3200 cal yr BP fire frequency at Sanger Lake increases from ~2 to ~5 fires per 1000 years . Frequency then generally declines to ~2 fires per 1000 years at the beginning of the LIA ~500 cal yr BP . From 500 cal yr BP to present frequency increases to ~4 fires per 1000 . BCHAR at Miller Lake averaged 0.47 particles cm-2 yr-1 and is variable through time (ranging between 0.12 and 0.73 particles cm-2 yr-1) reaching a maximum of 0.73 particles cm-2 yr-1 right before the onset of the LIA around 550 cal yr BP. Between 5000 and 2300 cal yr BP peak magnitude increases from ~40 particles cm2 peak 1 to ~600 particles cm2 peak -1 and decreases to ~10 cm2 peak -1 at present. Miller Lake fire frequency is relatively stable and reaches a maximum frequency at 1 000 cal yr BP of ~10 fires per 1000 years and declines to ~1 fire per 1000 years at present. BCHAR at Bolan Lake averages 3.1 particles cm-2 yr1 and is relatively stable through time reaching a maximum of 4.8 particles cm-2 yr-1 around 3 000 cal yr BP . Between 5000 and 1500 cal yr BP peak magnitude increases from ~15 particles cm2 peak -1 to ~90 particles cm2 peak -1 and decreases to ~70 cm2 peak -1 at present. Maximum fire frequency at Bolan Lake occurs at the

PAGE 44

34 beginning of the record at 5 000 cal yr BP (~10 fires per 1000 years) and declines to ~3 fires per 1000 years at present . BCHAR at Campbell Lake averaged 0.16 particles cm-2 yr-1 and was variable through time (ranging between 0. 06 and 0. 3 particles cm-2 yr-1) reaching a maximum value of 0.3 particles cm-2 yr-1 right before the onset of the MCA around 1500 cal yr BP and decrease to 0.9 particles cm-2 yr1 at present . Between 5000 and 3500 cal yr BP , peak magnitude increases from ~4 particles cm2 peak -1 to ~40 particles cm2 peak 1 and decreases to ~ 2 0 cm2 peak -1 at 2700 cal yr BP. There is a gap in fire events and peak magnitude peaks from 2700 to 1700 cal yr BP. From 1700 cal yr BP to present peak magnitude remains low and relatively stable, averaging ~5 cm2 peak -1. Maximum fire frequency at Campbell Lake occurs at 5 000 cal yr BP (~4 fires per 1000 years ) and declines <1 fires per 1000 years at ~2 500 cal yr BP . F requency increases from 2500 to around 1000 cal yr BP to ~2 fires per 1000 years and declines to <1 fire per 1000 years at present . At Sanger, Miller, Bolan, and Campbell sites the dominant canopy pollen types that contributed most to the understory -to -canopy ratios were from fireadapted Pseudotsuga menziesii and fire -sensitive Abies . Secondary pollen abundances of fire -adapted Pinus ponderosa, Alnus rubratype and Quercus were present . Pollen from f ire -adapted Quercus vaccinifolia -sm -type suggests it as the dominant understory taxa . Of the 45 pollen samples processed at Sanger Lake, five samples ( 21 cm, 144 cal yr BP; 73 cm, 960 cal yr BP; 2 05 cm, 2877 cal yr BP; 2 4 5 cm, 3480 cal yr BP; 2 81 cm, 4 350 cal yr BP) have fire pollen ratios lower than the non -fire pollen ratio mean in all and fire adapted pollen types and identify canopy fires. The non-fire ratio mean at Sanger Lake (0.64) is the lowest of all sites. At Miller Lake, of the 4 7 processed pollen samples, one sample (188 cm, 1654 cal yr BP ) identifie s a canopy fire. The nonfire ratio mean at Miller Lake (0.77) is average when compared to other sites. At Bolan Lake, of the 19 processed pollen samples , one sample (20 5 c m, 2 719 cal yr BP) identifies a canopy fire. The non-fire ratio mean at Bolan Lake (0.73) is average when compared to other sites. At Campbell Lake, of the 42 processed pollen sample s, three samples

PAGE 45

35 (34cm , 612 cal yr BP; 150 cm, 2740 cal yr BP; 230 cm, 4 454 cal yr BP ) identif y canopy fires. The non-fire ratio mean at Campbell Lake (0.81) is high when compared to other sites. Red Fir Zone BCHAR at Taylor Lake averaged 0.62 particles cm-2 yr1, reaching a maximum of 0.8 particles cm-2 yr-1 early in the record at 4 000 cal yr BP , then decreasing to 0.5 particles cm-2 yr1 between 4000 and 1500 cal yr BP , increasing to 0.73 particles cm-2 yr1 in the MCA and then declines to 0. 3 particles cm-2 yr1 toward present . Peak magnitude averages ~ 2 particles cm2 peak -1 from 4300 to 1300 cal yr BP and ~18 particles cm2 peak -1 between 1300 cal yr BP and present. Maximum fire frequency at Taylor Lake occurs at 5000 cal yr BP ~8 fires per 1000 years and declines until ~2 5 00 cal yr BP to ~3 fires per 1000 years, increase s and peaks around 1000 cal yr BP at ~5 fires per 1000 years, and then decreases to ~3 fires per 1000 years at present . At Taylor Lake, the dominant canopy pollen types that contributed most to the understory -to canopy ratios were from fire -adapted Pinus monticola with secondary contributions of fire -sensitive Abies . Pollen from fire sensitive Rosaceae is the dominant understory . O f the 52 pollen samples processed, one sample (75cm, 1098 cal yr BP) has a fire ratio lower than the nonfire ratio mean in all and fire -adapted pollen types and identif ies a canopy fires. The non -fire ratio mean at Miller Lake (0.71) is average when comp ared to other sites. Table 5 : Fire -related sample ratio comparisons to baseline means, and fire type determinations. Highlighted cells represent ratios that characterized their corresponding fire events as canopy fires Lake Fire Event age (cal yr BP ) All Comparison (fire ratio; non fire mean) Fire -Sensitive Comparison (fire ratio; non fire mean) Fire -Adapted Comparison (fire ratio; non fire mean) Fish 904 0.54; 0.66 0.58; 0.25 0.60; 0.70 1048 0.68; 0.66 0.68; 0.25 0.76; 0.70 1139 0.61; 0.66 0.33; 0.25 0.65; 0.70 1599 0.62; 0.66 0.13; 0.25 0.65; 0.70 Ogaromtoc 296 0.83; 0.77 0.5; 0.32 0.89; 0.81 2549 0.63; 0.77 0.09; 0.32 0.65; 0.81 Kelly 331 0.85; 0.78 0.98; 0.92 0.79; 0.74

PAGE 46

36 796 0.81; 0.78 0.98; 0.92 0.79; 0.74 1437 0.83; 0.78 0.83; 0.92 0.84; 0.74 2609 0.65; 0.78 0.94; 0.92 0.60; 0.74 2948 0.79; 0.78 0.93; 0.92 0.71; 0.74 3097 0.58; 0.78 0.87; 0.92 0.48; 0.74 3257 0.73; 0.78 0.84; 0.92 0.71 0.74 3427 0.72; 0.78 0.96; 0.92 0.65; 0.74 4063 0.72; 0.78 0.92; 0.92 0.66; 0.74 Hobart 2785 0.91 ; 0.91 0.59; 0.68 0.97; 0.96 3781 0.97 ; 0.91 0.87; 0.68 0.99; 0.96 3907 0.88 ; 0.91 0.63; 0.68 0.93; 0.96 4454 0.87 ; 0.91 0.35; 0.68 0.96; 0.96 4657 0.89 ; 0.91 0.45; 0.68 0.97; 0.96 Sanger 46 0.72; 0.64 0.78; 0.88, 0.71; 0.59 144 0.53; 0.64 0.93; 0.88, 0.47; 0.59 960 0.51; 0.64 0.80; 0.88, 0.45; 0.59 2877 0.36; 0.64 0.65; 0.88, 0.32; 0.59 3480 0.61; 0.64 0.95; 0.88, 0.55; 0.59 4350 0.49; 0.64 0.92; 0.88, 0.42; 0.59 Miller 676 0.77; 0.77 0.94; 0.89 0.58; 0.68 1654 0.76; 0.77 0.96; 0.89 0.68; 0.68 1847 0.91; 0.77 0.95; 0.89 0.88; 0.68 3055 0.86; 0.77 0.96; 0.89 0.83; 0.68 3454 0.87; 0.77 1; 0.89 0.82; 0.68 3586 0.83; 0.77 0.96; 0.89 0.76; 0.68 3857 0.87; 0.77 0.85; 0.89 0.89; 0.68 Bolan 2719 0.65; 0.73 0.76; 0.91 0.62; 0.64 Campbell 612 0.72; 0.81 0.39; 0.55 0.86; 0.90 1768 0.88; 0.81 0.71; 0.55 0.95; 0.90 2740 0.70; 0.81 0.27; 0.55 0.87; 0.90 4538 0.73; 0.81 0.26; 0.55 0.85; 0.90 Taylor 139 0.89; 0.71 0.80; 0.32 0.94; 0.91 371 0.77; 0.71 0.56; 0.32 0.96; 0.91 1098 0.69; 0.71 0.46; 0.32 0.83; 0.91 1639 0.7 7 ; 0.71 0. 41 ; 0.32 0.90; 0.91 2291 0.74; 0.71 0.36; 0.32 0.91; 0.91 4200 0. 80 ; 0.71 0. 43 ; 0.32 0.95; 0.91

PAGE 47

37 Biomass Burned Reconstruction 5,000year biomass burned trends in the Klamath -Siskiyou Ecoregion Biomass burned has been slowly declining over the last 5 , 000 years at northern s it es in the Klamath -Siskiyou Ecoregion (Figure 4 a ). Biomass burned was at a m aximum z -score of 0.7 ~ 4 500 cal yr BP and declines to a historical low of -0. 5 at present . A slight increase in biomass burned occurs during t he MCA, and in the middle (-0.2) of the climate event , and then continues to decline through the LIA . The southern sites show opposite trend s in the initial portion of the record when compared to northern sites . Biomass burned increases from 5 000 cal yr BP to 4 5 00 cal yr BP ( -0.8 to -0.2), remains relatively stable from 45 00 cal yr BP to 2 2 00 cal yr BP (-0.4), then the increases rapidly until peaking (0.7) during the early portion of the MCA ~ 1 3 00 cal yr BP. Biomass burned decreases during the last 1,300 years, but present -day values ( -0.4) are within the 5000year historical range . Biomass burned at coastal sites (Figure 4b) are hig h (0.5 ) at the beginning of the record and remain relatively stable until 37 00 cal yr BP , dropping from 0.4 to 0.5 over the next 500 years , and remain s low until 22 00 cal yr BP . After 2 2 00 cal yr BP , biomass burned increases and pea ks at 0.5 around 1300 cal yr BP in the MCA. Biomass burned decreases at coastal sites toward present , but present -day values (0.0) are within the 5000year historical range . Biomass burned at inland sites was much less variable over time than coastal sites, remaining stable a t 0.0 from about 4000 cal yr BP until 1 000 cal yr BP and then decreasing toward present to -1.0 , a significant 5000 year historic low.

PAGE 48

38 Figure 5a: Biomass burned curves for northern vs southern sites , including the number of sites making up the composite. The solid and dashed lines represent biomass burned curves calculated with 500 (black) and 200( orange ) year window half width smoother s (Blarquez et al. 2014) . The grey band is the 9 5 % confidence interval . Vertical bars represent the time periods of th e MCA (green) and LIA (blue) . Figure 5b : Biomass burned curves for coastal vs inland sites. The number of records show s the number of sites contributing to the reconstruction and each point. The center lines represent biomass burned curves calculated with 500 (black) and 200( orange ) year window half width smoother s (Blarquez et al. 2014) . The grey is the 9 5 % confidence interval . Vertical Bars represent the time periods of the MCA (green) and LIA (blue).

PAGE 49

39 Biomass Burned Trends in the Klamath Siskiyou Ecoregion and Northwestern US Biomass burned in the Klamath -S iskiyou Ecore gion (Figure 4c) is highest around 4500 cal yr BP (0.5) and decline s until 2 400 cal yr BP. After 2400 cal yr BP, it peakes (0.1) at the beginning of the MCA and then declines toward present . Present day biomass burned for the region is at a 5000 year historic low ( -1.0) , For the Northwestern US coast biomass burned was stable from 5000 cal yr BP until a round 2 500 cal yr BP , and then slowly began to increase, peaking (0.1) in the middle of the MCA and has been on a steep downward trend since . Today values ( -0. 4 ) are the lowest seen in the past 5,000 years. Figure 5c : Klamath -Siskiyou Ecoregion and Northwest US coast trends in biomass burned for the last 5 ,000 years . The number of records show s the number of sites contributing to the reconstruction and each point. The center lines represent biomass burned curves calculated with 500 (black) and 200( orange ) year window half width smoother s (Blarquez et al. 2014) . The grey is the 9 5 % confidence interval . Vertical Bars represent the time periods of the MCA (green) and LIA (blue).

PAGE 50

40 CHAPTER V DISCUSSIO N Fire is an integral disturbance mechanism in the Klamath -Siskiyou Ecoregion. The com plex fire regime in th e region has created forests with varying stand compositions and ages. Fire regime designations (i.e. replacement, mixed, surface or low) have large impacts on forest and fire management strategies (Agee, 1993, Barrett et al., 2010; Stephens, 2009) . In this study, the s patial variability of fire was determined using fire histories from nine lake sediment cores. These l ocal fire histories were compared along an elevational gradient and grouped within vegetation zones to visualize how changes climate and forest composition influence fire regimes (Figure 7). Fire severity impacts were determined by examining changes in pollen slightly after a fire event to distinguish between understory (low severity) and canopy (med ium to high severity) fires. While complete fire severity records are unable to be reconstructed for individual sites due to the low pollen sample resolution, combining the records from nine sites helps approximate how fire severity has changed in the regi on during the late Holocene. The identification of high severity fires in lake sediments using proxies, such as pollen and charcoal, is not well established. Some studies have examined the ratio of charcoal accumulation rates to fire frequency as a proxy of fire severity (Ali et al. 2012; Kelly et al 2013). Other studies have identified relationships between charcoal peak height and area burned using a combination of dendrochronology and lake sediment data (Whitlock et al., 2004; Higuera et al. 2011). Mi nckley and Shriver (2011) conducted a study using changes in pollen spectra following fire events and compared them to pollen spectra of fire -free periods at a single site . This study expands on the Minckley and Shriver (2011) analysis using multiple sites to characterize fire severity using not considering canopy and understory plant impacts inferred, but also fire -adapted and fire -sensitive plant impacts using pollen. An original goal of this study was to determin e if it was possible to identify high severity fires with charcoal peak magnitude heights relative to background charcoal (BCHAR) and to validate

PAGE 51

41 th ose findings by also identifying the same high severity fires with canopy -to -understory ratios . However, b ased on this comparison, fire severity is not related to peak magnitude (Figure 6) as there is no clustering of high peak magnitude values relative to BCHAR or of high severity fires identified with the canopy -to -understory ratios . This was also demonstrated in the Rocky Moun tains of Wyoming (Minckley and Shriver, 2011) . As p eak size is a result of several factors (e.g., fire size and distance from the repository, wind direction and intensity), these findings suggest that peak magnitude is not a reliable proxy for fire sever ity . Figure 6: Scatter plot comparing charcoal values for BCHAR and peak magnitude for all fires at Kelly Lake since 5000 cal yr BP. Circled events are fire events with associated pollen samples. Shaded circles are fire events that were interpreted as canopy fires. All sites were evaluated, and none show a clear trend . Kelly Lake had the greatest variety of fires and was used here as an example. Biomass burned reconstructions within Klamath -Sis kiyou Ecoregion were compared along latitudinal and coastal -to -inland gradients , and regional scale comparisons between the Klamath Siskiyou Ecoregion and the Northwest US c oast . The trends in the data were then compared with independent paleoclimate prox y data ( modeled ENSO events, summer solar radiation, and alkenone 0 1 2 3 4 5 6 7 8 9 10 0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.40Charcoal Peak Magnitude (patricles cm2peak1)Charcoal background (paticles cm 2 yr 1)

PAGE 52

42 derived sea surface temperatures) to infer trends seen in fire activity, biomass burned, and fir e severity during the last 5000 years ( Figure 7 ). A total of 201 fires were identified between the nine sites used in this study, 56 of which are visible across multiple records (Figures 3a and 3b), suggesting that fires in this region have be en historically widespread. For example, a fire event was iden tified ~ 900 cal yr BP in all northern sites (Kelly, Hobart, Sanger, Miller, and Bolan lakes), suggesting a widespread high -severity fire throughout this region. In 1987 , the Silver Fire Complex burned nearly 150,000 acres, which was also identified at two sites (Sanger and Bolan lakes). During the last 5,000 years, summer insolation (Figure 7 ) declined and resulted in cooler summers than before and a weakened northeastern Pacific subtropical highpressure system that increased effective moisture (Bartlei n et al., 1998). From 5 000 to 3 300 cal yr BP, alkenone -derived SSTs indicate cool oceans (1 -2C below modern). SST’s rapidly increased to 1 -2 C above modern SSTs within a ~200 year period between 3 500 and 2 200 cal yr BP. After 22 00 cal yr BP, SST’s declined to slightly below modern temperature s through present , with near modern SSTs during the early part of the MCA and lower SST’s during the LIA (Barron et al., 2003) . The Palmer Drought Severity Index which uses precipitation and temperat ure variations to create regional composites of climate conditions across North America, highlights periods of intense drought during the MCA along the Oregon and northern California coast , with the most intense drought occurring south of the CA -OR border (Cook and Krusic, 2004) . On inter annual time scales, current E l N io conditions create warmer and drier winters in the P acific Northwest ( M cafee et al., 2016). El Nio conditions result in early drying of summer fuels and lightning ignitions, which increase the probabilities of large fires (Westerling et al. 2006). ENSO events per 100 years have increased during the last 5,000 years and with the greatest variation prior to and during warmer and drier periods like the MCA . In the last 200 years, ENSO events have declined and are currently near historic lows (Moy et al., 2002) . Like El Nino, the Pacific Decadal Oscillation (PDO) also impacts the climate of the Klamath -

PAGE 53

43 Siskiyou Ecoregion on decadal time scales. The positive PDO results in warmer SSTs along the western coast of the northern US and lower sea l e vel pressure over the North Pacific Ocean. These anomalies create warmer and drier condition s in the region. When El Nio phases occur in conjunction with the positive PDO phase, the combine d effects enhance drying and results in prolonged drought (Mantua et al. 2001; Hessl, 2010; Mcafee, 2016; Newman et al., 2016) . Brown and Comrie (2004) found that the connections between ENSO an d PDO resulted in a ‘dipole’ between southern and norther n locations across the western US, and for the Klamath -Siskiyou’s at the OR -CA border . Strong ENSO signals, in the form of increased ENSO events, only occur during the negative phase of the PDO in the north and during the positive phase in the south (Moy e t al.,2002 ). Spatial v ariations in fire activity that cannot be explained by large -scale climatic variability , may be the result of the interplay of these interannual and decadal climate phenomena. The interpretation of the results to follow are limited by the number of available records in the region, but also by the proxies that the histories are based upon. Several site factors such as lake size and inflowing stream s can have a significant impact on the transport and deposition of pollen and charcoal. Therefore, sites were chosen that were smaller and had minimal, or no inflowing streams. Briles et al. ( 2012) also suggested that the vegetation history was significantly influenced by soils in the region; therefore , sites on ultr amafic substrates were left out of the individual site analyses, but were included in the composite analyses as fire activity was shown to be similar on all substrates. Proxy data inherently have assumptions tied to them . For example, the pollen identifi ed for the sites primarily came from wind -pollinated species, which captures many of the trees and shrubs in the forests, but there are other insect and bird/animal pollinated species that occur in minor abundances that are not represented in the pollen sp ectra (e.g., all species in the Ericaceae family). Charcoal is also transported and deposited in several ways and, while we can account for these, it is assumed site characteristics do not change through time (e.g., landscape geomorphology ). Finally, dating techniques have error associated with them and are costly. Some sites are better dated than others,

PAGE 54

44 which influences the sedimentation rate and CHAR calculations and levels. To account for this, the study only evaluates long -term trends in the data at individual sites and standardizes those levels in the composite analyses.

PAGE 55

45

PAGE 56

46 Figure 7: Biomass burned and fire events, temperature, climatic fluctuations and inferred vegetation history for Abies -dominated forests to illustrate timing of m ajor vegetation shifts (Briles et al., 2012) . All identified fire events (black dots), identified canopy fires (red ovals) from the nine sites used in this study. Biomass burned curves are calculated with 500 (black) and 200( orange) year window half width smo others and associated 9 5 % confidence interval (Blarquez et al. 2014) . MCA (green) and the LIA (blue) are highlighted . SST’s (dark blue) derived from phytoplankton alkenones (Barron et al. 2003) , July 45 insolation (Bartlein et al., 1998 ), ENSO event time series (Moy et al., 2002) are used to infer historical change changes and mechanisms in the region. Influence s on Fire V ariability Oak Woodland Vegetation Zone Pollen records at Ogaromtoc Lake suggest more closed forest during cool wet pe riods (e.g., LIA and 2300 2 000 cal yr BP ), and more open forests during warmer and drier periods (e.g. MCA , and the last 100 years) (Crawford et al., 2015) . Increasing then stable fire activity and increasing biomass burned from ~3500 cal yr BP until the middle of the MCA , suggest cooler and wetter conditions due to decreasing summer insolation . The d ecrea sed biomass burned in the last 1,000 years suggests that warm dry conditions of the MCA likely resulted in larger fires that consumed the fuels that accumulated during the previous cool wet period. Increased fire activity in the last 200 years, and the continued decline of biomass burned, reflects the frequent, low severity fire regime currently observed at lower elevation s in the region . Fish Lake is influence d by more coastal conditions than Ogaromtoc Lake, which likely accounts for some of the contrasting trends in fire at the two sites ( Crawford et al., 2015). Fire frequenc y at F ish Lake decreased from 3 000 cal yr BP until the beginning of the LIA (~400 cal yr BP ), suggesting that the higher annual precipitation than at Ogaromtoc Lake significant ly effect s fire activity. At Fish Lake, despite fire activity decreasing throughout most of the record, biomass burned exceeded all other sites in the stud y . Crawford et al. (2015) indicate that after 1500 cal yr BP at Fish Lake, and to a lesser degree Ogaromtoc Lake, Native American burning altered the landscape with fire, encouraging a more open forest that was more suitable for foraging and hunting. BCHAR at both lakes decreased significantly since ~ 500 cal yr BP , suggesting that European settlers and fire

PAGE 57

47 suppression may have changed the levels of fuels burned in the forest s (Crawford et al., 2015) . However, fire frequency at both sites increased at the beginning of the LIA , suggesting that while biomass burned was less than before due to fire suppression efforts, there was heightened fire activity possibly due to greater human accessibility and ignitions . At Fish and Ogaromtoc lakes, the lowest elevation sites, negative fire-sensitive non -fire and fire pollen ratios indicate that fire -sensitive understory taxa were more prevalent tha n fire -sensitive canopy taxa. This is likely because the forest surrounding these lakes are more open than lakes at higher elevations , due to the low severity, high frequency fire regime. Mixed Conifer Vegetation Zone Fire frequency during the past 5000 cal yr BP at Hobart and Kelly lakes have both remained relatively high in comparison with sites in other vegetati on zones. They also have the most identified fire events. Biomass burned increased at Kelly Lake and decreased at Hobart Lake, indicating trends are inverse of one another during the past 5000 years . The contrast is likely the result of Kelly Lake being in closer proximity to the ocean (~80 km) than Hobart Lake, which lies 110 km east of Kelly . Biomass burn ed nearly doubled at Kelly Lake at the same time there was a significant increase in SSTs (1 -2C warmer than modern) likely creating warmer conditions there than at Hobart Lake. Alternatively , at Hobart Lake, declining biomass burned after 5000 cal yr BP, is likely due to long term climatic changes of progressively cooler and wetter conditions than before , due to the gradual d ecrease in summer insolation during the late Holocene (White et al, 2015) . During the MCA , fire activity at Kelly Lake remained stable where at Hobart Lake it increased, suggesting that the increased influence from coastal conditions that result ed in increased moisture, increased biomass, and less frequent fire at Kelly Lake than at Hobart Lake. At Hobart Lake, lightning is generally the source of fire ignition , but there is also evidence of anthropogenic ignitions . Logging in the last 100 years has also changed the forest structure from closed to more open ( Agee, 1993; White et al., 2015) .

PAGE 58

48 The contrasting records suggest that the coastal -to -inland climate gradient is a strong driver of fire regimes at these sites (discussed further below) . White and Red Fir Vegetation Zone s Coastal sites (Bolan and Sanger) within the White Fir Zone have less variation in biomass burned when compared to inland sites in the same zone, especially during the last 25 00 years. Fire frequency at both sites was highest prior to 3000 cal yr BP after which frequency at Bolan and Sanger declines, consistent with large-scale climatic trends. Fire frequency at Sanger Lake is lower than Bolan throughout the record and shows more variation in the last 1000 years , sugge sting that the increased coastal influence Sanger receives as a result of being 20 km close to the Pacific Ocean than Bolan has a marked impact on precipitation and fire frequency . Since 2 5 00 cal yr BP , fire event s were more sporadic resulting in longer , fire -free periods at t he co a st al sites. This suggests that the influence of coastal weather patterns that create cooler and wetter conditions have shifted the fire regime from a more frequent (5 000 2 5 00 cal yr BP ) to a less frequent fire regime for sites closer to the coast (2 5 00 cal yr BP present ). Briles et al. (2017) suggest s a shift in forest composition at Bolan and Sanger Lakes around 45 00 cal yr BP from a more open, pine -shrub oak forest , to a closed forest consisting of higher abundances of Abies , as a result of wetter and cooler conditions than before . Interestingly, the timing of the 4500 cal yr BP vegetation shift was different between sites, suggest ing the effects of coastal weather patterns (fog production) were stronger at Sanger Lake than the more inland site Bolan Lake (Briles et al., 20 08) . An 800year gap in fire events from the middle of the MCA to the end of the LIA occurred at Sanger Lake , which likely allowed for a large buildup of fuels biomass. Fires toward during the LIA at Sanger Lake have burned more fuels (the upper fire likely being the 1987 Silver Fire) likely the result of increased fuels during the previous fire -free period and the warmer and drier conditions of the current climate. Biomass burn ed at both inland s ites (Miller and Campbell) were more variable than t he two coastal sites and Miller Lake had higher biomass burned than Campbell Lake . Biomass burned at

PAGE 59

49 Miller and Campbell lakes was initially low until ~2400 cal yr BP, then increased at Campbell while re maining low at Miller until the MCA , then increased at both sites during the MCA, and dropped drastically during the LIA . The contrasts and variability in biomass burned between these sites prior to the MCA may be due to their more central location and at the transition between coastal wet forests and the drier inland forests . While the trends in fire frequency are similar at these sites (initially high at 5000 years ago and decreasing through present) , Miller Lake ha d a higher fire frequency and is more variable . The two sites showed the greatest increase in fire frequency during the MCA than any of the other sites. This suggest s that the more northern position of Miller Lake results in more biomass accumulation , which during dry summers , is more likely to ignite resulting in more frequent fires . Fire frequency at Taylor Lake in the Red Fir Zone displayed similar overall trends in fire frequency as the two inland sites (Campbell and Miller) . Biomass burned trends are comparable to Campbell Lake , e xcept during and just after the MCA , when it significantly increases, suggesting that the drier conditions resulted biomass burned level s comparable to 5000 years ago . Bril es et al. (2011) suggest s that increased abundances of Tsuga mertensiana and Abies in the last 2000 years indicate that the forests at higher elevations became more closed likely due to increased precipitation , which would have resulted in more fuels to burn in the drier MCA. Regional Biomass Burned Patterns and Controls T o characterize changes in fire activity, biomass burned and burn severity across the Klamath -Siskiyou Ecoregion, composite analyses were determined for northern / southern and coastal / inland locations . The composites were compared with climate and vegetation reconstructions (Figure 7 ). The comparisons enable an understanding of how fire operates in a region that is not only at a biogeographic transition zone between drier , more open forest s to the south and wetter , more

PAGE 60

50 closed forests to the north , but also how the steep coastal -to -inland precipita tion gradient influence s bur n patterns. Biomass burned at northern sites decreased during the last 5000 years , which coincides with a shift in forest composition from more open forest s of Pinus monticola, Abies , and Quercus to more closed forest s with less Quercus and more Pseudotsuga menziesi i in the forest . The decreasing trend of biomass burned over the last 5000 years at northern sites corresponds to decreasing summer insolation over the same period . A slight increase fire activity occurs from 1200-500 cal yr BP which corresponds with increased ENSO variability and the warmer and drier conditions of the MCA. Biomass burned at northern sites reached a historical low during the LIA, and continue s to decrease toward present , but a slight increase in fire activity is likely a response to today’s warmer and drier conditions and/or increased human ignitions . At southern sites , biomass burned gradually increased from 5 000 cal yr BP until ~ 2400 cal yr BP , then increased more rapidly and peak ed at ~15 00 cal yr BP, and then declined through present . Present biomass burn ed is lower than it has been in the last 2000 years at southern sites, but near levels prior to that time. V ariability in biomass burned at southern sites coincid es with a rise in ENSO variability and the warmer, drier conditions during the MCA . The 2000 -year P almer D rought S everity Index (PSDI) for grid points 33-36 also indicate that there was less precipitation during the MCA and that drought conditions were intensified at more southern latitudes (Cook et al., 2004) . Fire frequency at northern sites was more frequent, and of higher severity than southern sites, likely due to the higher accumulation of biomass at the wetter northern sites that increase the chances of fire spread and increased fuels burned. A notable gap in high severity fires from ~2500 cal yr BP to ~16 00 cal yr BP at all lo cations corresponds to a drop in SSTs through present day and suggest s that the climate during this time was cool and wet enough to limit high severity fires. All sites record more closed forests than before, so fuel availability was likely not a factor i nfluencing fire severity. The discrepancies observed between biomass burned and fire events at different latitudes highlights

PAGE 61

51 the significant differences in precipitation and climatic impacts over this relatively small geographic area. Trends in biomass b urned observed at coastal sites follows SSTs and ENSO variability throughout the record. F rom 5 000 cal yr BP to ~ 37 00 cal yr BP , when SSTs were cooler than present day , and ENSO variability was higher, b iomass burned was high and fire activity more variable. Fire activity remained low from 3 700 to 2 5 00 cal yr BP when SSTs were warm and ENSO event s were less variable. Biomass burned increased and peaked after the beginning of the MCA ~1 3 00 cal yr BP when SSTs became more stable and ENSO variability increased drastically . Biomass burned at coastal sites is currently near 5000 year historical lows, and like northern sites, more fire events and high severity fires were identified at coastal sites. Unlike coastal sites, trends in b iomass burned at inland sites remained relatively stable throughout the record and is currently at a 5000 year historical low. During the middle of the MCA (~1 1 00 cal yr BP ) biomass burned and fire activity dec reased dramatically at inland sites where at coastal sites it d eclined only slightly . F ire events and high severity fires are more prevalent at coastal sites due to increased precipitation and increased fuels availability in comparison to the drier inland sites . Inland sites do not appear to follow trends in SSTs or ENSO events to the extent that coastal sites do likely because of the lack of coastal influences on precipitation and moisture availability . The notable gap in high severity fires from ~2 5 00 cal yr BP until ~1 6 00 cal yr BP is also evident in this compa rison of coastal -to -inland sites. The coastal -to -inland precipitation gradient present in the Klamath Siskiyou Ecoregion and along the western coast of the US is well documented ( Briles et al, 2008; Walsh et al, 2015 ); however , the Klamath -Siskiyou Mountains are at a transition (~41N) between climate regimes to the north and south, which result in contrasting fire regimes over a short distance. Northern sites are cool and wet , resulting in more fuels and more frequent high severity fires than drier southern sites.

PAGE 62

52 In western the United States , despite over a century of fire suppression , frequency, area burned, and wildfire size have all increased significantly since the mid -1980 ’s ( Stephens, 2005; van Mantgem et al, 2013; Westerling et al. 2006 ). Biomass burned in the w estern US has been relatively stable during the previous 5,000 years, compar ed to the prior 8,000 years of increasing biomass burned (Marlon et al. 2012) . Biomass burned reconstructions for the P acific Northwest from northern Oregon to southern British Columbia differ substantially from those for the western US . For example, b iomass burned increased from 5 000 cal yr BP until the end of the MCA and substantially decreased during the last 1000 years with only a slight increase a round 350 cal yr BP in the Pacific Northwest (Walsh et al. 2015) . The n orthwest US coast biomass burned reconstruction include s sites in Washington, Oregon , and Northern California that span the north/south precipitation gradient . Biomass burned in the n orthwest coast was stable from the beginning of the record until the onset of the MCA when it increased slightly for about 500 years then decreased dramatically. These trends highlight increased fires during the MCA. The scaling of composite records helps identify key locations within the western US where there are transitions in fire regimes , useful for developing fire management plans for diverse forests. Spatial Variability of Fire in the Klamath-Siskiyou Ecoregion and Implications for Fire Mana gement The identification of the historical range of variability of diverse forests requires long-term historical data sets (Gavin et al. 2007; Kaufmann et al., 2007) . P aleoenvironmental records ha ve the potential to greatly influence future ecosystem an d fire management strategies, which is invaluable particularly given i ncreasing wildfire activity , coupled with the increasing urban -wildland interface across the western United States. It is important to note that the records and results here should not be used as predictive tools , but rather for informing about controls that have resulted in changing fire conditions and the spatial tends across the region . Many current forest management plans (e.g., Healthy Forest Initiative , US Forest Service fire management plans, and CAL FIRE ) rarely consider

PAGE 63

53 the changing role of fire across diverse landscapes and forests, but rather focus on how to reduce hazards and risks associated with wildland fire , and improve environmental resilience to fire . For example, t he 2010 Strategic Fire Plan for California suggest s that emerging research be used in management strategies, but does not specifically mention the types of research useful for developing better management plans ( Dixon et al, 2010). The Klamath National Forest has created watershed based fire management plans within the forest, and while they do take into consideration vegetation types within the watershed, little to no consideration is given to precipitation gradients in the region and how fire will respo nd to the warmer and drier climate predicted in the future. Historically, northern forest s in the K la math -Siskiyou Mountain E coregion, that receive higher amounts of precipitation than southern forests , had fire regimes th at burned frequently and at high severity . Current Klamath -Siskiyou forests are a product of LIA conditions and fire suppression at lower elevations where ample biomass exists. As conditions become warmer and drier in the future, th ese forests will likely burn more frequently and at hi gher severity than they have historically . Lower , drie r sites that historically burned more often than wetter sites, but at lower severity, are experiencing fire activity already beyond their historical range of variability . The spatial variability of fire is high er in the Klamath -Siskiyou Ecoregion than elsewhere along the n orthwest US coast , as documented through the gradients analyzed in this study, and forest managers and management plans need to consider how fire may change along these gradients in the future . In addition to informing future ecological and fi re management strategies, the information in the thesis addresses the debate on fire severity in the western US. Some studies have shown that high-severity fires have increased in comparison to historical trends (Colombaroli and Gavin, 2010; Higuera et al., 2014 ; Miller et al., 2009 ). On the other side of the debate, Hanson and Odion (2014) suggest that modern high-severity fires have not increased in proportion, area, or patch size in western North America and conclude that climate change might not have as large of an impact on fire severity in the future as the other s have projected (Hanson and Odion, 2014) . This study indi cates a decrease

PAGE 64

54 in high severity fires during cool and wet periods, like those ~25 00 to 16 00 cal yr BP and during the LIA, and increased hi gh severity fires during war m dr y periods , such as the MCA and prior to 25 00 cal yr BP . Within the limitations of this study it does not appear as though high severity fires ha ve increased in frequency in the recent past in the Klamath -Siskiyou Ecoregion; however, prior to the LIA, during warm and dr y periods, there were period s of more severe and widespread fire and these can be expected if conditions continue to warm and bec ome drier than before . Future research A goal of this study was to determine a way of d ifferentiating high severity fire from moderate-to -low severity fires. The study showed that it is possible to identify high severity , canopy replacing fires by using high-resolution charcoal record s in conjunction with changes in pollen ratios during fire events in comparison with ratios during non-fire periods . However, a clear method of identifying all high severity fires with the limited number pollen samples available was not determined. To get a complete history of fire severity for an individual site, pollen samples around each fire event need to be analyzed. F uture research on fire severity should include additional pollen processing at the same depth an d immediately following fire events to determine fire severity for all identified fire events and vegetation response post -fire . A dditional sampling of non-fire time periods will enable a more robust non-fire to fire pollen ratio , and a moving average record could be more representative of normal variation through time than a single average . This would allow for the spatial variability of high severity fires to be determined along elevational gradients in addition to coastal -to -inland and latitudinal g radients. It would also be advantageous to determine pollen zones at each site so that fuel types and connectivity can be examined for their effects on fire severity similar to the study conducted in the Cascades of northern Oregon (Minckley and Long, 2016). Examining fire pollen ratios after known high severity fires would increase the validity of the understory -to -pollen ratio method used in this study and described by Minckley and Shriver (2011).

PAGE 65

55 Modern calibration studies from recent fires would al so help develop more accurate reconstructions of fire severity. Additionally, there is a lack of data in the southern portion of this study area. To better understand the latitudinal gradient proposed by this study, pollen and charcoal data from new si tes would be useful in increasing our understanding of how differences in precipitation along these gradients have influenced vegetation and fire regimes in the Klamath -Siskiyou Ecoregion. To increase knowledge of the main climatic factors that influence the spatial variability of fire in the Klamath -Siskiyou Ecoregion a dditional informatio n regarding climate -fire relationship is needed. How fire activity interact s with interannual and decadal variability in climate, like the effects of El Nio / Southern Oscillations and Pacific Decadal Oscillation s especially along the gradients discussed in this study will enable better predictions of wildfire severity, occurrence, and extent (Hessl et al. 2010).

PAGE 66

56 CHAPTER VI CONCLUSIONS Th is study report s on the pyrogeography of the Klamath -Siskiyou Ecoregion. Charcoal and pollen records from nine lakes were examined along elevational, coastal -to -inland , and latit udinal gradients to determine the var ia bility in fire regimes and controls . In addition, fire severity was explored to determine how it has changed along the se gradients . The research addressed the following questions : 1) What is the spatial and temporal variability of fire during the last 5000 years in the KlamathSiskiyou Ecoregion? Fire act ivity not only varied temporally , but also along elevational , coastal -to -inland , and latitudinal gradients in the Klamath -Siskiyou Ecoregion. While long -term fire activity is largely governed by changes in solar radiation influencing ocean temperatures, pressure systems and the position of the jet stream and storm tracks , there is evidence that native people influenced fire regimes at very small local scales and at low elevations. As vegetation moved along elevational gradients and fuel loads and moisture levels changed , so did fire activity and fuel loads. Sites more sensitive to moisture fluctuations such as lower and more southern/ inland sites burned more frequently and were driven more by fuel availability than climate . H igher, more coastal, and northe rn sites burned less frequently and had more climate driven fire regimes. Interestingly, fire activity and biomass burned increased at most sites during the Medieval Climate Anomaly (MCA) and decreased during the Little Ice Age (LIA) , suggesting regional coherency to drought conditions and cool wet conditions of those events, respectively. There is also evidence of a widespread fire across the northern Klamath Siskiyou Ecoregion during the MCA (~900 cal yr BP). At no other time in the last 5000years did this level of coherency exist across all sites and all composite records, suggesting that drivers of fire were operating across a much boarder region than they had in the past . Biomass burned is at historical lows

PAGE 67

57 in the ecoregion, when compared to the las t 5000 years, and primarily at northern and inland locations , due to LIA conditions and fire suppression. Fire managers should consider the complexity at which fires have burned in the region historically when developing future management plans. 2) H ow has fire severity changed during the last 5000 years in the Klamath -Siskiyou Ecoregion? The resolution of the pollen records used in this study are not high enough to distinguish fire severity trends at individual sites , but when they are combined, and examined along latitudinal and coastal -to -inland gradients trends can be inferred . Southern sites had less frequent high severity fires than northern sites, due to increased precipitation and increased fuel availability . Coastal sites were like northern sites and recorded more high severity fires during the MCA than inland site s . Southern and inland locations, receiving lower precipitation than the other locations , recorded less frequent higher severity fires. The highest concentration of high severity fires occurred p rior to 2 5 00 cal yr BP . N o high severity fires were identified from 25 00 cal yr BP to ~ 1600 cal yr BP ; which corresponds with a drop in SSTs through present day and suggests that the climate during this time was cool and wet enough to limit high severity fires. H owever, they increased through the MCA and became less severe during the LIA. Fire severity in the Klamath -Siskiyou Ecoregion has not increased in the last 500 years . Higher resolution pollen records are needed to reconstruct individual site fire severity and to enhance the composite analyses through the region. Regional historical fire severity reconstructions helps to inform the debate on changing patterns of fire severity in the western US . Trends can be observed over much longer timelines with paleoeclogical data than is possible with other methods of measuring fire severity (i.e dendrochronology , and remotely sensed data) . These longer timelines increase knowledge of how fire severity responded to warmer drier conditions in the past, like during the MCA, and have the potential to contribute greatly to ecological and fire management strategies to better protect the forest in the Klamath -Siskiyou Ecoregion in the face of future warmer and drier conditions.

PAGE 68

58 REFERENCES Ali A. A., Carcaillet C., Bergeron Y. (2009). Long -term fire frequency variability in the eastern Canadian boreal forest: the influences of climate vs. local factors. Global Change Biology 15, 1230 – 1241 Agee, J. K. (1993). Fire ecology of Pacific Northwest forest. Washington DC: Island Press. Alexander, E.B., Coleman, R.G., Keeler Wolfe, T., Harrison, S.P., 2007. Serpentine geoecology of western North America. Oxford University Pr ess. Baker, W. L. (2015). Are high-severity fires burning at much higher rates recently than historically in dry -forest landscapes of the western USA? PLoS ONE, 10(9), 1– 26. Baker, W. L. (1992). The landscape ecology of large disturbances in the design an d management of nature reserves. Landscape Ecology , 7 (3), 181– 194. Blaauw, M., & Christen, J. A. (2011). Flexible paleoclimate age -depth models using an autoregressive gamma process. Bayesian Analysis , 6(3), 457– 474. Blarquez, O., Vannire, B., Marlon, J . R., Daniau, A. L., Power, M. J., Brewer, S., & Bartlein, P. J. (2014). Paleofire: An R package to analyze sedimentary charcoal records from the Global Charcoal Database to reconstruct past biomass burning. Computers and Geosciences, 72(September), 255 – 26 1. Bartlein, P. J., Anderson, K. H., Edwards, M. E., Mock, C. J., Thompson, R. S., Webb, T. I., & Whitlock, C. (1998). Paleoclimate simulations for North America over the past 21,000 years: Features of the Simulated Climate and Comparisons with Paleoenvir onmental Data. Quaternary Science Reviews, 17, 549– 585. Barrett, S.; Havlina, D.; Jones, J.; Hann, W.; Frame, C.; Hamilton, D.; Schon, K.; Demeo, T.; Hutter, L.; Menakis, J. (2010). Interagency Fire Regime Condition Class Guidebook. Version 3.0, https://ww w.frcc.gov/. Barron, J. A., Heusser, L., Herbert, T., & Lyle, M. (2003). High -resolution climatic evolution of coastal northern California during the past 16,000 years. Paleoceanography , 18(1). Bond, W.J., & van Wilgen, B.W. (1996) Prescribed burning in C alifornia wildlands vegetation management. University if California Press, Berkley, California, USA. Briles, C. E., Whitlock, C., & Bartlein, P. J. (2005). Postglacial vegetation, fire, and climate history of the Siskiyou Mountains, Oregon, USA. Quaternary Research, 64(1), 44 – 56. Briles, C. (2008) Holocene Vegetation and fire history of the floristically Diverse Klamath Mountains, Northern California, USA. PhD Dissertation Briles, C. E., Whitlock, C., Bartlein, P. J., & Higuera, P. (2008). Regional and loc al controls on postglacial vegetation and fire in the Siskiyou Mountains, northern California, USA. Palaeogeography, Palaeoclimatology, Palaeoecology , 265(1 – 2), 159– 169.

PAGE 69

59 Briles, C. E., Whitlock, C., Skinner, C. N., & Mohr, J. (2011). Holocene forest develo pment and maintenance on different substrates in the Klamath Mountains, northern California, USA. Ecology , 92(3), 590– 601. Briles, C. E. (2017). Controls on Mountain Plant Diversity in Northern California: A 14,000Year Overview. Annals of the American As sociation of Geographers , 107(2), 238 – 249. Brown, D. P., & Comrie, A. C. (2004). A winter precipitation “dipole” in the western United States associated with multidecadal ENSO variability. Geophysical Research Letters, 31(9), 1 – 4. Christen, J.A., 1994. Summarizing a set of radiocarbon determinations a robust approach. Applied Statistics -Journal of The Royal Statistical Society Series C , 43 (3), 489 – 503. Christen, J.A. & Fox, C. (2010 ). A general purpose sampling algorithm for continuous distributions (the t -walk). Bayesian Ana lysis, 4(2), 263-282 Cook, E.R. & Krusic, P.J., (2004). The North American Drought Atlas.LamontDoherty Earth Observatory and the National Science Foundation. Cook, E.R., Woodhouse, C. A., Eakin, C. M., Meko, D. M., & Stahle, D. W. (2004). Long -term aridity changes in the western United States. Science, 306(5698), 1015– 1018. Colombaroli, D., & Gavin, D. G. (2010). Highly episodic fire and erosion regime over the past 2,000 y in the Siskiyou Mountains, Oregon. Proceedings of the National Academy of Sciences, 107(44), 18909. Crawford, J. N., Mensing, S. A., Lake, F. K., & Zimmerman, S. R. H. (2015). Late Holocene fire and vegetation reconstruction from the wester n Klamath Mountains, California, USA: A multidisciplinary approach for examining potential human land-use impacts. Holocene, 25(8), 1341– 1357. Daniels, M. L., Anderson, R. S., & Whitlock, C. (2005). Vegetation and fire history since the Late Pleistocene from the Trinity Mountains, northwestern California, USA. The Holocene, 15(7), 1062 – 1071. DellaSala, D. A., S. R. Reid, T. J. Frest, J. R. Strittholt, & D. M. Olson. (1999). A global perspective on the biodiversity of the Klamath-Siskiyou ecoregion . Natural Areas Journal 19:300 – 319. Deng, W., Liu, X., Chen, X., Wei, G., Zeng, T., Xie, L., & Zhao, J. (2017). A comparison of the climates of the Medi eval Climate Anomaly, Little Ice Age, and Current Warm Period reconstructed using coral records from the Northern China Sea. Journal of Geophysical Research: Oceans, 122, 264– 275. Dixon, S. L., Giocomini, P., Andre, M., Bradshaw, Ll., Nakamura, G., et al. (2010). 2010 Strategic Fire Plan For California. State Board of Forestry and Fire Protection , 47. Retrieved from http://cdfdata.fire.ca.gov/pub/fireplan/fpupload/fpppdf668.pdf Dobrowski, S. Z. (2011). A climatic basis for microrefugia: The influence of terrain on climate. Global Change Biology , 17(2), 1022 – 1035.

PAGE 70

60 Franklin, J. F., & C. T. Dyrness (1988), Natural Vegetation of Oregon and Washington Rep., Oregon State University Press, Oregon State University, Corvallis, Oregon. Fried, J.S., Torn, M. S., & Mills, E. (2004). The Impact of Climate Change on Wildfire Severity: A Regional Forecast for Northern California. Climatic Change , 64, 169– 191. Gavin, D. G., Hallett, D. J., Hu, F. S., Lertzman, K. P., Prichard, S. J., Brown, K. J., Lynch, J.A., Bartlein, P., & Peterson, D. L. (2007). Forest fire and climate change in western North America: insights from sediment charcoal records. Frontiers in Ecology and the Environment , 5(9), 499– 506. Gershunov, A., T. P Barnett, & D. R. Cayan. (1999). North Pacific inte rdecadal oscillation seen as factor in ENSO related North American climate anomalies. EOS Transactions 80 (3): 25– 30. Graham, N. E., M. K. Hughes, C. M. Ammann, K. M. Cobb, M. P. Hoerling, D. J. Kennett, J. P. Kennett, B. Rein, L. Stott, P. E. Wigand, and T. Xu. (2007). Tropical Pacific – mid latitude teleconnections in medieval times. Climate Change 83:241 – 285. Grove, A. T. (2001). The “Little Ice Age” and its geomorphological consequences in Mediterranean Europe. Climatic Change , 48(1), 121 – 136. Hanson, C. T., & Odion, D. C. (2014). Is fire severity increasing in the Sierra Nevada, California, USA? International Journal of Wildland Fire, 23(2009), 1 – 8. Harrison, S., Damschen, E. I., & Grace, J. B. (2010). Ecological contingency in the effects of climatic warming on forest herb communities. Proceedings of the National Academy of Sciences, 107(45), 19362– 19367. Harrison, S ., Damschen, E., Fernandez-Going, B., Eskelinen, A., & Copeland, S. (2014). Plant communities on infertile soils are less sensitive to climate change. Annals of Botany , 116(6), 1017 – 1022. Hessl, A. E., Mckenzie, D., & Schellhaas, R. (2010). Drought and Pacific Decadal Oscillation Linked to Fire Occurrence in the Inland Pacific Northwest. Ecological Applications , 14 (2), 425– 442. Higuera PE, Peters ME, Brubaker LB and Gavin DG (2007) Understanding the origin and analysis of sediment -charcoal records with a simulation model. Quaternary Science Reviews 26: 1790 – 1809 Higuera, P.E., L.B. Brubaker, P.M. Anderson, F.S. Hu, T.A. Brown. (2009). Vegetation mediated the impacts of postglacial climatic change on fire regimes in the south central Brooks Range, Alaska. E cological Monographs, 79: 201 -219. Higuera, P. E., Whitlock, C., & Gage, J. a. (2011). Linking tree -ring and sediment -charcoal records to reconstruct fire occurrence and area burned in subalpine forests of Yellowstone National Park, USA. The Holocene, 21(2 ), 327 – 341. Higuera, P. E., Briles, C. E., & Whitlock, C. (2014). Fire -regime complacency and sensitivity to centennial through millennial-scale climate change in Rocky Mountain subalpine forests, Colorado, USA. Journal of Ecology , 102(6), 1429– 1441.

PAGE 71

61 Jim enez-Moreno, G., Anderson, R. S., Desprate, S., Grigg, L. D., Grimm, E. C., & Al., E. (2010). Millennial-scale variability during the last glacial in vegetation records from Europe. Quaternary Science Reviews, 29(21– 22), 2839– 2864. Jones, G.D., Bryant, V. M., Lieux, M.H., Jones, S.D., and Lingren, P.D. (1995). “Pollen of the southeastern United States: with emphasis on melissopalynology and entomopalynology.” American Association of Stratigraphic Palynologists Foundation, College Station, TX. Kaufmann, M.R. , D. Binkley, P.Z. Ful, M. Johnson, S.L. Stephens, & T.W. Swetnam. (2007). Defining old growth for fire -adapted forests of the western United States. Ecology and Society 12(2): 15. Kapp, R.O., Davis, O.K., and King, J.E. (2000). “Ronald O. Kapp’s pollen and spores.” American Association of Stratigraphic Palynologists Foundation, College Station, TX. Kelly, R., Chipman, M. L., Higuera, P. E., Stefanova, I., Brubaker, L. B., & Hu, F. S. (2013). Recent burning of boreal forests exceeds fire regime limits of the past 10,000 years. Proceedings of the National Academy of Sciences, 110(32), 13055– 13060. Kru ckeberg, A.R., (2002). Geology and plant life: the effects of landforms and rock types on plants. University of Washington Press. Seattle, WA. p. 362. Mann, M. E., Zhang, Z., Rutherford, S., Bradley, R. S., Hughes, M. K., Shindell, D., Ammann, C., Faluvegi , G., & Ni, F. (2009). Global signatures and dynamical origins of the Little Ice Age and Medieval Climate Anomaly. Science, 326(5957), 1256– 60. Mantua, N. J., & Hare, S. R. (2002). The Pacific Decadal Oscillation. Journal of Oceanography . Marlon, J. R., Bartlein, P. J., Carcaillet, C., Gavin, D. G., Harrison, S. P., Higuera, P. E., Joos, F., Power, M.J., & Prentice, I. C. (2008). Climate and human influences on global biomass burning over the past two millennia. Nature Geoscience, 1, 697– 702. Marlon, J. R., Bartlein, P. J., Gavin, D. G., Long, C. J., Anderson, R. S., Briles, C. E., Walsh, M. K. et al. (2012). Long-term perspective on wildfires in the western USA. Proceedings of the National Academy of Sciences, 109(9), 535 – 543. Martin, R.E., & Sapsis, D. B., (1992). Fires as agents of Biodiversity: Pyrodiversity promoted biodiversity. In Harris, R.R., Erman, & Kerner, H.M (Editors), Proceedings of the Symposium on Biodiversity of Northwest California. University of California, Berkley, CA, Wildland Resour ces Center Repot No. 29, pp. 150-157. Mcafee, S. A., & Wise, E. K. (2016). Intra -seasonal and inter -decadal variability in ENSO impacts on the Pacific Northwest. International Journal of Climatology , 36 (1), 508 – 516. Miller, J. D., Safford, H. D., Crimmins , M., & Thode, A. E. (2009). Quantitative evidence for increasing forest fire severity in the the Sierra Nevada and the southern Cascade Mountains , California, and Nevada, USA. Ecosystems, 12(1), 16 – 32.

PAGE 72

62 Minckley, T. A., & Shriver, R. K. (2011). Vegetation responses to changing fire regimes in a rocky mountain forest. Fire Ecology, 7(2), 66 – 80. Minckley, T. A., & Long, C. J. (2016). Paleofire severity and vegetation change in the Cascade Range, Oregon, USA. Quaternary Research , 85(2), 211– 217. Mock, C. J. (1996). Climatic controls and spatial variations of precipitation in the western United States. Journal of Climate . Mohr, J. A., C. Whitlock, and C. N. Skinner. (2000). Postglacial vegetation and fire history, Eastern Klamath Mountains, California. Holocene 10:587– 601. Moy, C. M., Seltzer, G. O., Rodbell, D. T., & Anderson, D. M. (2002). Variability of El Nio/Southern Oscillation activity at millennial timescales during the Holocene epoch. Nature , 420(November). Nagel TA, Taylor AH (2005) Fire and persistence of montane chaparral in mixed conifer forest landscapes in the northern Sierra Nevada, Lake Tahoe Basin, California, USA. The Journal of the Torrey Botanical Society, 132, 442 – 457. Newman, M., Alexander, M. A., Ault, T. R., Cobb, K. M., Deser, C., Di Lorenzo, et al., (2016). The Pacific Decadal Oscillation, Revisited. Journal of Climate , 29(12), 4399– 4427. Odion, D. C., & Hanson, C. T. (2006). Fire severity in conifer forests of the Sierra Nevada, California. Ecosystems, 9(7), 1177– 1189. Olson, D., DellaSala, D. a., Noss, R. F., Strittholt, J. R., Kass, J., Koopman, M. E., & Allnutt, T. F. (2012). Climate Change Refugia for Biodiversity in the Klamath-Siskiyou Ecoregion. Natural Areas Journal , 32(1), 65 – 74. Power, M. J., Marlon, J., Oritz, P. J., Bartlein, P. J., Harrison, S. P., Mayle, F. E., Ballouche, A., et al. (2008). Changes in fire regimes since the Last Glacial Maximum: an assessment based on a global synthesis and analysis of charcoal data. Climate D ynamics, 30, 887– 907. PRISM Climate Group, Oregon State University, http://prism.oregonstate.edu, created 4 Feb 2004. Ramsey, C. (2007). Deposition Models for Chronological Records, Quaternary Science Reviews, 1 – 42. Reimer, P.J., Bard, E., Bayliss, A., Be ck, J.W., Blackwell, P.G., Bronk, R.C., Buck, C.E., Cheng, H., Edwards, R.L., et al. (2013). IntCal13 and Marine13 radiocarbon age calibration curves 0-50000 years cal BP. Radiocarbon 55, 18691887. Safford, H. D., Miller, J., Schmidt, D., Roath, B., & Par sons, A. (2007). BAER soil burn severity maps do not measure fire effects to vegetation: A comment on Odion and Hanson (2006). Ecosystems, 11(1), 1– 11.

PAGE 73

63 Sommers, W. T., Coloff, S. G., & Conard, S. G. (2011). Synthesis of Knowledge: Fire History and Climate Change. JFSP Synthesis Reports. Paper 19, 190p. Retrieved from http://www.firescience.gov/JFSP_fire_history.cfm Steel, Z. L., Safford, H. D., & Viers, J. H. (2015). The fire frequency -severity relationship and the legacy of fire suppression in California forests. Ecosphere, 6(1), 8. Stephens SL. (2005). Forest fire causes and extent on United States forest service lands. International Journal of Wildland Fire, 14:213 – 22. Stephens, S. L., Moghaddas, J. J., Edminster, C., Fiedler, C. E., Haase, S., Harrington, M., Youngblood, A. et al. (2009). Fire Treatment Effects on Vegetation Structure, Fuels , and Potential Fire Severity in Western U.S. Forests. Ecological Applications , 19 (2), 305 – 320. Stephens, S. L., Agee, J. K., Ful, P. Z., North, M. P., Romme, W. H., Swetnam, T. W., & Turner, M. G. (2013). Managing Forests and Fire in Changing Climates. Science, 342(6154), 41– 2. Taylor, A. H., & Beaty, R. M. (2005). Climatic influences on fire regimes in the northern Sierra Nevada mountains, Lake Tahoe Basin, Nevada, USA. Journal of Biogeography , 32(3), 425 – 438. Taylor, A. H., & Skinner, C. N. (1998). Fire history and landscape dynamics in a late -successional reserve, Klamath Mountains, California, USA. Forest Ecology and Management , 111, 285– 301. Taylor, A. H., & Skinner, C. N. (2003). Spatial Patterns and Controls on Historical Fire Regimes and Forest Structure in the Klamath Mountains. Ecological Applications , 13(3), 704 – 719. van Mantgem, P. J., Nesmith, J. C. B., Keifer, M., Knapp, E. E., Flint, A., & Flint, L. (2013). Climatic stress increases forest fire severity across the western United States. Ecology Letters, 16(9), 1151– 1156. Walsh, M. K., Whitlock, C., & Bartlein, P. J. (2010). 1200 year s of fire and vegetation history in the Willamette Valley, Oregon and Washington, reconstructed using high-resolution macroscopic charcoal and pollen analysis. Palaeogeography, Palaeoclimatology, Palaeoecology , 297(2), 273– 289. Walsh, M. K., Marlon, J. R., Goring, S. J., Brown, K. J., & Gavin, D. G. (2015). A Regional Perspective on Holocene Fire– Climate – Human Interactions in the Pacific Northwest of North America. Annals of the Association of American Geographers , 105(6), 1135 – 1157. Westerling, A. L., Hidalgo, H. G., Cayan, D. R., & Swetnam, T. W. (2006). Warming and earlier spring increase western US forest wildfire activity. Science, 313(5789), 940– 943. Westerling, A. L. (2016). Increasing western US forest wildfire activity: sensitivity to changes in the timing of spring. Philosophical Transactions of the Royal Society of London. Series B, Biological Sciences, 371(1696), 717– 728. White, A., Briles, C., & Whitlock, C. (2015). Postglacial vegetation and fire history of the southern Cascade Range, Oregon . Quaternary Research , 84(3), 348 – 357.

PAGE 74

64 Whitlock, C., & Larsen C. P. S. (2001). Charcoal as a fire proxy. Pages 75 – 97 in J. P. Smol, H. J. B. Birks, and W. M. Last, editors. Tracking environmental change using lake sediments. Volume 3. Terrestrial, algal, and siliceous indicators. Kluwer Academic, Dordrecht, The Netherlands. Whitlock, C., Skinner, C. N., Bartlein, P. J., Minckley, T., & Mohr, J. A. (2004). Comparison of charcoal and tree-ring records of recent fires in the eastern Klamath Mountains, Califor nia, USA. Canadian Journal of Forest Research, 34(10), 2110 – 2121. Whittaker, R. H. (1960). Vegetation of the Siskiyou Mountains, Oregon and California. Ecological Monographs , 30(3), 279 – 338.

PAGE 75

65 APPENDIX A : ANALYSISES PERFORMED Analysis performed Program the file was used in Puropose of the analysis File name of data set or file folder Reference/ Online Location Chronology R: Bacon package Age-depth models All_Lakes_Bacon.xls http://www.chrono. qub.ac.uk/blaauw/b acon.html Peak detection in sediment charcoal CharAnalysis Fire history reconstruction CharAnalysis File Folder https://github.com/ phiguera/CharAnal ysis Z -score charcoal composites R: Paleofire package Biomass burning/ fire frequency reconstructions Paleofire File Folder https://cran.r project.org/web/pac kages/paleofire/ind ex.html Pollen percent calculations Microsoft Excel Vegetation history reconstructions Pollen File Folder Canopy: understory ratios Microsoft Excel Fire severity type characterization All Lakes_Last5k_Ratios .xls Minckley and Shriver 2011 & 2016

PAGE 76

66 APPENDIX B: CHRONOL O GY FOR ALL SITES

PAGE 77

67 APPENDIX C : POLLEN PERCENTAGES AND CANOPY-TO UNDERSTORY RATIOS Bolan Lake Fire Ratio Depth Age Fire Adapted Canopy Fire Sensitive Canopy Fire Adapted Shrubs Fire Sensitive Shrub All Canopy: Understory Sensitive Canopy: Understory Adapted Canopy: Understory 1 47 29.40 20.72 5.30 0.48 0.79 0.95 0.69 8 4 33.24 26.24 6.41 0.29 0.80 0.98 0.68 25 132 16.83 9.06 5.50 0.32 0.63 0.93 0.51 45 309 22.12 13.27 4.13 0.00 0.79 1.00 0.69 65 522 19.48 19.58 5.19 0.00 0.77 1.00 0.58 85 843 30.93 13.40 6.19 0.52 0.74 0.93 0.67 105 1154 23.12 12.01 6.31 1.50 0.64 0.78 0.57 125 1474 26.69 8.90 7.98 0.31 0.62 0.93 0.54 145 1787 24.29 9.89 6.50 0.56 0.66 0.89 0.58 165 2077 20.00 9.65 5.41 0.24 0.68 0.95 0.57 185 2406 28.49 15.62 4.93 1.37 0.75 0.84 0.70 205 2719 23.75 9.06 5.63 1.25 0.65 0.76 0.62 225 3056 23.62 12.88 5.21 0.00 0.75 1.00 0.64 245 3377 29.51 23.77 5.46 0.55 0.80 0.96 0.69 265 3690 24.65 15.30 5.67 0.85 0.72 0.89 0.63 285 4076 22.30 16.39 2.30 0.66 0.86 0.92 0.81 305 4481 26.27 16.72 5.07 0.60 0.77 0.93 0.68 325 4843 31.58 7.24 7.24 0.66 0.66 0.83 0.63 345 5186 34.21 10.20 7.89 1.64 0.65 0.72 0.63 Campbell Lake Fire Ratio Depth Age Fire Adapted Canopy Fire Sensitive Canopy Fire Adapted Shrubs Fire Sensitive Shrub All Canopy : Understory Sensitive Canopy: Understory Adapted Canopy: Understory 0 52 45.86 28.99 0.89 3.25 0.90 0.80 0.96 4 17 40.23 23.51 2.55 3.12 0.84 0.77 0.88 8 23 49.86 13.97 2.47 3.29 0.83 0.62 0.91 14 228 46.69 18.37 3.01 2.41 0.85 0.77 0.88 18 344 42.86 18.96 2.60 2.86 0.84 0.74 0.89 22 437 48.44 19.56 0.67 1.11 0.95 0.89 0.97 26 518 47.90 8.24 1.34 4.54 0.81 0.29 0.95 30 578 45.34 14.51 2.59 2.33 0.85 0.72 0.89 34 612 32.58 10.66 2.46 4.71 0.72 0.39 0.86

PAGE 78

68 38 647 44.05 18.72 1.32 2.20 0.89 0.79 0.94 42 680 44.72 9.52 1.04 4.14 0.83 0.39 0.95 46 713 43.44 14.48 1.17 2.74 0.87 0.68 0.95 50 760 37.44 10.76 2.91 4.71 0.73 0.39 0.86 54 848 44.96 20.16 0.82 4.63 0.85 0.63 0.96 58 925 41.52 13.45 1.17 3.31 0.85 0.60 0.95 62 1019 40.98 13.66 3.09 1.03 0.86 0.86 0.86 66 1105 44.29 13.67 1.63 5.51 0.78 0.43 0.93 70 1194 49.72 12.43 2.82 3.77 0.81 0.53 0.89 74 1269 39.77 14.19 3.26 4.65 0.74 0.51 0.85 78 1342 45.64 14.22 2.06 5.50 0.78 0.44 0.91 82 1419 39.44 9.05 1.61 4.83 0.77 0.30 0.92 86 1489 32.06 8.18 1.45 2.24 0.83 0.57 0.91 90 1553 35.12 10.12 3.51 4.96 0.68 0.34 0.82 94 1632 43.63 12.74 2.17 4.34 0.79 0.49 0.91 98 1699 44.83 9.92 1.03 7.02 0.74 0.17 0.95 102 1768 42.59 16.08 1.04 2.71 0.88 0.71 0.95 106 1837 48.88 9.13 1.42 3.25 0.85 0.48 0.94 110 1902 49.10 14.03 2.61 2.40 0.85 0.71 0.90 118 2059 33.77 13.29 1.96 3.70 0.79 0.56 0.89 126 2228 39.56 13.56 2.89 3.78 0.78 0.56 0.86 134 2393 27.41 28.70 3.52 1.85 0.83 0.88 0.77 142 2568 40.49 10.49 2.68 3.90 0.77 0.46 0.88 150 2740 41.25 10.94 2.81 6.25 0.70 0.27 0.87 158 2910 52.45 11.76 2.70 2.70 0.85 0.63 0.90 166 3086 57.28 12.58 1.49 4.97 0.83 0.43 0.95 174 3267 49.91 7.10 4.09 3.37 0.77 0.36 0.85 182 3464 43.55 16.94 2.96 2.69 0.83 0.73 0.87 190 3666 38.38 11.18 5.48 1.97 0.74 0.70 0.75 198 3871 40.19 11.92 5.61 3.50 0.70 0.55 0.76 206 4058 42.40 11.66 4.42 3.53 0.74 0.53 0.81 214 4225 43.34 10.32 2.63 3.38 0.80 0.51 0.89 222 4380 46.49 10.46 1.49 3.74 0.83 0.47 0.94 230 4538 42.90 7.50 3.43 4.40 0.73 0.26 0.85 234 4627 49.17 10.47 1.16 4.82 0.82 0.37 0.95 242 4782 44.06 12.47 3.42 4.02 0.77 0.51 0.86 250 4950 39.85 8.23 0.91 4.75 0.79 0.27 0.96 258 5111 34.08 11.80 0.67 6.68 0.72 0.28 0.96 Fish Lake Fire Ratio

PAGE 79

69 Depth Age Fire Adapted Canopy Fire Sensitive Canopy Fire Adapted Shrubs Fire Sensitive Shrub All Canopy : Understory Sensitive Canopy: Understory Adapted Canopy: Understory 6 47 78.00 2.75 14.25 1.75 0.67 0.22 0.69 11 28 72.87 2.58 16.02 2.58 0.60 0.00 0.64 20 9 81.10 0.79 11.29 2.89 0.70 0.57 0.76 27 47 77.04 4.23 7.25 3.63 0.76 0.08 0.83 36 143 85.35 1.41 6.20 1.69 0.83 0.09 0.86 39 195 79.52 2.66 9.04 3.19 0.74 0.09 0.80 46 314 80.38 2.99 7.36 4.90 0.74 0.24 0.83 56 368 78.03 2.54 7.61 5.07 0.73 0.33 0.82 66 416 72.32 0.78 14.36 6.27 0.56 0.78 0.67 76 467 79.13 1.63 10.84 2.44 0.72 0.20 0.76 81 496 71.20 0.52 19.63 4.71 0.49 0.80 0.57 86 531 79.95 1.03 10.54 2.57 0.72 0.43 0.77 101 662 76.47 1.07 9.09 4.81 0.70 0.64 0.79 111 758 77.11 1.32 9.47 6.32 0.66 0.66 0.78 126 904 70.99 1.10 17.68 4.14 0.54 0.58 0.60 141 1048 76.88 0.84 10.31 4.46 0.68 0.68 0.76 151 1139 75.59 1.31 16.27 2.62 0.61 0.33 0.65 166 1280 75.86 1.06 14.59 1.86 0.65 0.27 0.68 176 1371 77.84 0.53 11.35 3.43 0.68 0.73 0.75 187 1448 75.13 1.05 13.09 3.40 0.64 0.53 0.70 201 1538 73.09 1.00 16.36 2.90 0.59 0.49 0.63 215 1599 71.85 1.87 15.01 2.41 0.62 0.13 0.65 226 1642 76.52 0.79 10.29 2.11 0.72 0.45 0.76 241 1692 76.08 2.14 12.90 1.34 0.69 0.23 0.71 251 1725 70.69 1.28 17.99 2.06 0.56 0.23 0.59 262 1762 74.27 0.53 18.83 1.59 0.57 0.50 0.60 276 1808 70.90 2.38 16.93 1.06 0.61 0.38 0.61 292 1877 71.86 0.82 21.31 0.55 0.54 0.20 0.54 301 1920 71.05 1.05 17.89 1.58 0.57 0.20 0.60 309 1956 73.85 1.08 17.79 1.08 0.60 0.00 0.61 326 2052 78.97 0.77 13.85 1.28 0.68 0.25 0.70 346 2213 82.71 1.06 5.32 1.06 0.86 0.00 0.88 391 2552 72.51 0.79 18.59 0.79 0.58 0.00 0.59 411 2677 72.41 1.33 18.04 2.12 0.57 0.23 0.60 446 2865 75.82 0.82 13.74 1.37 0.67 0.25 0.69 Hobart Lake Fire Ratio

PAGE 80

70 Depth Age Fire Adapted Canopy Fire Sensitive Canopy Fire Adapted Shrubs Fire Sensitive Shrub All Canopy : Understory Sensitive Canopy: Understory Adapted Canopy: Understory 1 59 59.62 11.12 1.69 0.84 0.93 0.86 0.94 16 6 64.07 8.29 1.54 2.15 0.90 0.59 0.95 32 57 49.58 6.48 1.67 3.35 0.84 0.32 0.93 48 126 53.75 3.33 0.42 2.92 0.89 0.07 0.98 64 187 65.19 8.92 1.70 1.70 0.91 0.68 0.95 80 250 59.13 3.04 3.04 9.57 0.66 0.52 0.90 96 314 60.51 11.02 0.46 2.30 0.93 0.66 0.98 119 402 53.26 9.44 1.32 4.83 0.82 0.32 0.95 135 459 60.50 6.73 0.69 1.72 0.93 0.59 0.98 151 520 71.03 9.78 0.38 1.92 0.94 0.67 0.99 167 555 41.49 21.72 0.39 0.39 0.98 0.96 0.98 183 611 63.72 10.08 0.00 2.42 0.94 0.61 1.00 199 686 52.01 16.16 1.49 4.48 0.84 0.57 0.94 202 706 65.04 7.82 0.41 4.11 0.88 0.31 0.99 217 836 65.56 7.92 1.28 2.57 0.90 0.51 0.96 218 840 66.66 5.52 2.29 3.05 0.86 0.29 0.93 233 916 60.03 18.70 0.73 0.73 0.96 0.92 0.98 234 917 56.33 16.35 1.54 1.54 0.92 0.83 0.95 249 978 68.12 8.77 1.99 0.80 0.93 0.83 0.94 250 982 61.12 11.56 0.90 2.40 0.91 0.66 0.97 266 1076 46.31 18.50 0.90 0.30 0.96 0.97 0.96 282 1173 57.70 11.51 1.24 1.24 0.93 0.80 0.96 298 1269 67.84 11.31 1.26 0.50 0.96 0.91 0.96 314 1362 67.30 10.28 1.23 1.23 0.94 0.79 0.96 330 1463 62.74 10.60 1.21 2.12 0.91 0.67 0.96 346 1558 58.36 13.77 2.32 1.99 0.89 0.75 0.92 362 1654 47.02 19.71 1.08 3.25 0.88 0.72 0.95 378 1749 56.50 15.55 1.18 1.97 0.92 0.78 0.96 394 1852 52.79 11.54 0.84 1.26 0.94 0.80 0.97 410 1958 55.36 15.41 1.19 0.79 0.95 0.90 0.96 426 2068 57.16 9.52 1.98 1.19 0.91 0.78 0.93 444 2188 52.70 9.70 2.43 2.43 0.86 0.60 0.91 460 2294 51.94 11.12 2.52 1.26 0.89 0.80 0.91 476 2405 56.07 12.52 1.99 3.18 0.86 0.59 0.93 492 2530 52.85 13.58 0.38 1.13 0.96 0.85 0.99 508 2658 58.91 12.63 0.71 0.36 0.97 0.95 0.98 524 2785 66.59 9.27 1.18 2.36 0.91 0.59 0.97

PAGE 81

71 540 2908 63.90 8.09 0.62 1.56 0.94 0.68 0.98 556 3031 41.63 19.46 1.59 1.59 0.90 0.85 0.93 572 3153 42.21 18.88 1.53 1.14 0.92 0.89 0.93 588 3279 74.08 7.99 0.65 1.30 0.95 0.72 0.98 604 3399 51.40 14.19 0.33 1.67 0.94 0.79 0.99 620 3528 51.57 10.37 0.68 0.68 0.96 0.88 0.97 636 3654 50.13 11.63 0.00 0.00 1.00 1.00 1.00 652 3781 60.21 9.26 0.32 0.65 0.97 0.87 0.99 668 3907 55.51 10.48 1.98 2.37 0.88 0.63 0.93 684 4034 60.45 6.83 0.95 3.18 0.88 0.37 0.97 706 4208 61.91 6.65 0.40 0.81 0.97 0.78 0.99 722 4333 58.08 7.58 1.35 0.67 0.94 0.84 0.95 738 4454 56.36 6.46 1.17 3.13 0.87 0.35 0.96 754 4554 56.52 6.50 1.33 0.33 0.95 0.90 0.95 770 4657 64.19 7.83 1.12 2.98 0.89 0.45 0.97 786 4764 72.14 5.79 1.35 1.62 0.93 0.56 0.96 808 4907 67.13 7.68 1.54 0.61 0.94 0.85 0.96 824 5014 64.56 8.86 2.04 1.36 0.91 0.73 0.94 Kelly Lake Fire Ratio Depth Age Fire Adapted Canopy Fire Sensitive Canopy Fire Adapted Shrubs Fire Sensitive Shrub All Canopy : Understory Sensitive Canopy: Understory Adapted Canopy: Understory 11 67 42.86 15.82 4.59 0.51 0.84 0.94 0.81 15 123 31.97 11.15 7.43 0.37 0.69 0.94 0.62 23 233 30.18 17.68 6.10 0.00 0.77 1.00 0.66 27 283 35.14 23.55 0.39 1.16 0.95 0.91 0.98 31 331 35.56 22.46 4.28 0.27 0.85 0.98 0.79 37 408 26.98 13.23 5.82 0.00 0.75 1.00 0.65 39 431 36.59 14.29 8.01 2.09 0.67 0.74 0.64 47 534 44.33 8.59 8.25 0.69 0.71 0.85 0.69 51 590 39.11 8.12 5.90 0.00 0.78 1.00 0.74 55 639 41.20 4.87 7.49 0.75 0.70 0.73 0.69 63 728 47.89 9.96 9.96 0.38 0.70 0.93 0.66 65 751 41.83 9.13 5.77 0.48 0.78 0.90 0.76 69 796 56.62 10.14 6.76 0.00 0.82 1.00 0.79 73 848 46.43 19.94 2.38 0.00 0.93 1.00 0.90 77 897 45.51 18.84 2.61 0.58 0.91 0.94 0.89 85 996 41.95 9.73 4.36 0.00 0.84 1.00 0.81 91 1072 25.43 12.50 5.17 0.00 0.76 1.00 0.66 95 1125 45.07 9.87 3.29 0.00 0.89 1.00 0.86

PAGE 82

72 103 1226 51.09 9.35 5.61 0.31 0.82 0.94 0.80 107 1278 32.46 12.72 5.92 0.00 0.77 1.00 0.69 111 1332 41.97 6.48 5.35 0.00 0.80 1.00 0.77 119 1437 42.80 8.33 3.79 0.76 0.84 0.83 0.84 127 1538 48.38 9.42 3.25 0.32 0.88 0.93 0.87 137 1663 37.30 12.85 5.64 0.00 0.80 1.00 0.74 143 1740 38.41 8.28 4.30 0.99 0.80 0.79 0.80 151 1839 36.84 4.86 6.07 0.40 0.73 0.85 0.72 159 1935 38.91 10.29 2.57 0.32 0.89 0.94 0.88 167 2034 42.99 14.01 4.78 0.32 0.84 0.96 0.80 175 2148 38.00 17.43 5.71 0.00 0.81 1.00 0.74 183 2258 36.05 9.09 7.84 0.00 0.70 1.00 0.64 199 2494 40.82 12.60 4.93 0.27 0.82 0.96 0.78 207 2609 38.90 8.73 9.73 0.25 0.65 0.94 0.60 215 2726 41.18 14.82 11.53 0.24 0.65 0.97 0.56 223 2841 31.91 8.21 6.99 0.00 0.70 1.00 0.64 231 2948 36.96 23.91 6.21 0.93 0.79 0.93 0.71 239 3097 29.47 13.17 10.34 0.94 0.58 0.87 0.48 247 3257 34.14 10.27 5.74 0.91 0.74 0.84 0.71 255 3427 33.55 14.47 7.24 0.33 0.73 0.96 0.65 263 3593 33.63 12.39 7.37 1.47 0.68 0.79 0.64 271 3811 35.92 12.64 8.33 0.00 0.71 1.00 0.62 279 4063 36.03 12.27 7.31 0.52 0.72 0.92 0.66 287 4319 42.57 5.28 6.27 0.66 0.75 0.78 0.74 Miller Lake Fire Ratio Depth Age Fire Adapted Canopy Fire Sensitive Canopy Fire Adapted Shrubs Fire Sensitive Shrub All Canopy : Understory Sensitive Canopy: Understory Adapted Canopy: Understory 1 57 18.47 24.44 2.61 0.00 0.89 1.00 0.75 8 12 22.12 20.06 4.05 0.67 0.80 0.93 0.69 12 67 20.77 10.32 4.13 0.26 0.75 0.95 0.67 16 110 20.30 21.99 2.46 0.35 0.88 0.97 0.78 24 223 14.15 13.87 0.76 0.25 0.93 0.96 0.90 28 281 17.25 14.59 3.50 1.17 0.74 0.85 0.66 32 324 24.59 16.67 2.01 0.00 0.91 1.00 0.85 42 399 22.30 18.85 2.82 0.70 0.84 0.93 0.78 44 412 16.36 16.10 2.81 0.77 0.80 0.91 0.71 52 468 18.25 19.34 2.81 0.62 0.83 0.94 0.73 60 518 17.86 17.62 6.02 0.60 0.69 0.93 0.50 62 532 18.08 17.37 3.16 0.00 0.84 1.00 0.70

PAGE 83

73 68 573 17.98 14.36 4.33 0.27 0.75 0.96 0.61 76 625 25.61 5.73 2.80 0.51 0.81 0.84 0.80 84 676 11.49 16.04 3.03 0.51 0.77 0.94 0.58 92 730 24.40 11.84 5.00 0.87 0.72 0.86 0.66 100 780 12.05 17.96 4.92 0.58 0.69 0.94 0.42 108 833 23.19 15.12 2.59 0.78 0.84 0.90 0.80 116 887 21.13 17.05 3.44 0.86 0.80 0.90 0.72 124 946 18.67 12.27 6.22 1.21 0.61 0.82 0.50 132 1006 18.99 12.57 6.86 1.43 0.58 0.80 0.47 140 1076 19.68 10.80 2.91 0.42 0.80 0.93 0.74 156 1272 18.88 19.80 5.22 1.53 0.70 0.86 0.57 172 1464 12.92 17.24 4.46 0.50 0.72 0.94 0.49 188 1654 19.01 8.82 3.67 0.18 0.76 0.96 0.68 204 1847 27.86 21.32 1.78 0.59 0.91 0.95 0.88 220 2030 26.35 12.28 3.29 2.40 0.74 0.67 0.78 236 2233 23.94 14.02 4.08 0.25 0.80 0.96 0.71 252 2513 23.94 12.27 4.04 0.00 0.80 1.00 0.71 260 2649 10.34 11.76 1.67 0.83 0.80 0.87 0.72 268 2782 18.43 3.90 2.14 1.33 0.73 0.49 0.79 276 2918 6.86 4.15 3.79 0.72 0.42 0.70 0.29 284 3055 27.86 11.19 2.62 0.24 0.86 0.96 0.83 292 3185 12.44 10.60 2.07 0.26 0.82 0.95 0.71 300 3320 24.61 14.80 3.48 0.58 0.81 0.92 0.75 308 3454 17.94 9.30 1.83 0.00 0.87 1.00 0.82 316 3586 29.02 14.89 3.91 0.30 0.83 0.96 0.76 324 3723 13.95 12.34 3.49 0.54 0.73 0.92 0.60 332 3857 34.75 14.96 2.12 1.24 0.87 0.85 0.89 340 3944 22.85 10.92 2.61 0.95 0.81 0.84 0.79 348 4020 21.31 13.25 4.21 0.90 0.74 0.87 0.67 364 4174 24.76 17.07 1.88 0.00 0.91 1.00 0.86 372 4256 18.71 21.79 5.41 0.32 0.75 0.97 0.55 Ogaromtoc Lake Fire Ratio Depth Age Fire Adapted Canopy Fire Sensitive Canopy Fire Adapted Shrubs Fire Sensitive Shrub All Canopy : Understory Sensitive Canopy: Understory Adapted Canopy: Understory 1 59 78.61 1.03 12.37 0.52 0.72 0.33 0.73 10 46 70.00 1.28 18.72 2.56 0.54 0.33 0.58 20 26 75.42 0.96 14.46 1.93 0.65 0.33 0.68 40 72 88.43 0.77 4.88 0.77 0.88 0.00 0.90 60 220 90.89 0.52 2.60 1.04 0.92 0.33 0.94

PAGE 84

74 70 296 86.17 0.99 4.94 2.96 0.83 0.50 0.89 80 383 81.39 1.74 7.44 3.23 0.77 0.30 0.83 100 562 89.09 0.78 4.42 2.08 0.87 0.45 0.91 120 775 81.98 2.28 6.09 4.06 0.78 0.28 0.86 135 865 83.25 0.25 8.12 4.82 0.73 0.90 0.82 150 953 80.10 2.07 7.49 2.84 0.78 0.16 0.83 166 1046 83.10 1.19 10.00 2.62 0.74 0.38 0.79 180 1126 87.19 1.01 3.77 3.02 0.86 0.50 0.92 190 1185 81.44 0.00 7.22 3.09 0.78 1.00 0.84 200 1242 84.12 0.50 7.20 2.73 0.79 0.69 0.84 208 1287 81.25 1.50 8.25 3.75 0.75 0.43 0.82 220 1361 75.73 1.58 9.50 1.85 0.74 0.08 0.78 231 1428 81.09 1.49 8.21 3.48 0.75 0.40 0.82 250 1553 81.65 1.81 7.24 2.84 0.78 0.22 0.84 259 1601 80.90 1.26 7.79 2.76 0.77 0.38 0.82 270 1662 82.37 1.76 10.83 1.51 0.74 0.08 0.77 280 1720 84.29 1.05 7.59 1.31 0.81 0.11 0.83 290 1781 84.18 0.97 10.22 1.95 0.75 0.33 0.78 306 1907 83.91 1.73 8.66 2.48 0.77 0.18 0.81 310 1940 83.42 0.75 8.29 2.26 0.78 0.50 0.82 320 2027 81.30 0.52 9.61 2.34 0.75 0.64 0.79 332 2133 80.15 1.96 10.78 3.19 0.71 0.24 0.76 350 2293 88.43 1.54 4.37 1.03 0.89 0.20 0.91 367 2549 76.50 1.25 16.25 1.50 0.63 0.09 0.65 383 2673 72.10 2.22 9.88 8.15 0.61 0.57 0.76 400 2856 81.23 0.77 6.43 1.54 0.82 0.33 0.85 410 3037 78.88 1.46 10.92 3.16 0.70 0.37 0.76 428 3363 88.40 1.39 7.19 1.62 0.82 0.08 0.85 Sanger Lake Fire Ratio Depth Age Fire Adapted Canopy Fire Sensitive Canopy Fire Adapted Shrubs Fire Sensitive Shrub All Canopy : Understory Sensitive Canopy: Understory Adapted Canopy: Understory 1 46 45.99 6.94 7.81 0.87 0.72 0.78 0.71 5 14 45.58 9.12 5.63 0.54 0.80 0.89 0.78 9 19 40.85 14.08 4.23 0.56 0.84 0.92 0.81 13 56 36.11 10.90 4.06 1.28 0.80 0.79 0.80 17 95 26.14 5.61 5.44 0.35 0.69 0.88 0.66 21 144 24.83 4.70 8.89 0.17 0.53 0.93 0.47 25 202 29.14 7.58 7.19 0.40 0.66 0.90 0.60 29 252 42.29 11.97 6.12 0.27 0.79 0.96 0.75

PAGE 85

75 35 323 44.83 29.47 3.76 0.00 0.90 1.00 0.85 37 350 37.05 16.71 6.13 1.11 0.76 0.88 0.72 41 404 48.18 11.48 7.00 1.12 0.76 0.82 0.75 45 458 34.26 7.61 11.93 0.51 0.54 0.88 0.48 49 507 37.50 16.19 6.25 0.00 0.79 1.00 0.71 53 553 40.00 7.86 9.29 0.95 0.65 0.78 0.62 57 612 38.12 11.49 6.53 0.78 0.74 0.87 0.71 61 683 36.97 4.70 11.32 0.64 0.55 0.76 0.53 65 772 34.74 6.70 7.69 0.25 0.68 0.93 0.64 69 867 28.24 4.50 12.23 0.54 0.44 0.79 0.40 73 960 32.21 8.11 12.16 0.90 0.51 0.80 0.45 77 1051 43.32 10.08 7.08 0.54 0.75 0.90 0.72 81 1124 42.65 7.60 13.73 0.25 0.56 0.94 0.51 85 1176 47.67 8.81 6.74 0.26 0.78 0.94 0.75 95 1315 33.42 8.29 9.84 0.52 0.60 0.88 0.54 103 1429 38.19 8.04 8.04 0.25 0.70 0.94 0.65 111 1533 38.50 11.74 9.62 0.23 0.67 0.96 0.60 119 1651 47.07 7.56 12.68 0.24 0.62 0.94 0.58 127 1759 34.72 9.07 9.07 1.30 0.62 0.75 0.59 135 1858 35.71 5.56 10.05 0.53 0.59 0.83 0.56 139 1907 31.70 6.70 14.96 0.00 0.44 1.00 0.36 149 2039 38.91 6.11 11.99 0.45 0.57 0.86 0.53 161 2202 29.20 4.42 15.34 1.47 0.33 0.50 0.31 173 2362 35.07 7.58 12.32 0.24 0.55 0.94 0.48 188 2601 36.83 8.57 10.71 0.00 0.62 1.00 0.55 197 2740 35.25 8.53 14.52 0.46 0.49 0.90 0.42 205 2877 27.22 3.52 13.89 0.74 0.36 0.65 0.32 213 3001 38.83 6.91 8.78 0.80 0.65 0.79 0.63 225 3178 41.37 6.35 10.66 0.25 0.63 0.92 0.59 233 3295 37.87 10.40 6.13 0.00 0.77 1.00 0.72 245 3480 39.10 9.02 11.28 0.25 0.61 0.95 0.55 257 3721 23.68 8.70 14.37 0.20 0.38 0.95 0.24 265 3889 35.60 10.99 8.64 0.00 0.69 1.00 0.61 277 4241 40.34 5.80 11.35 0.72 0.59 0.78 0.56 281 4350 25.82 5.82 10.63 0.25 0.49 0.92 0.42 297 4886 32.65 7.14 9.18 0.77 0.60 0.81 0.56 305 5138 28.04 6.30 18.70 0.43 0.28 0.87 0.20 Taylor Lake Fire Ratio

PAGE 86

76 Depth Age Fire Adapted Canopy Fire Sensitive Canopy Fire Adapted Shrubs Fire Sensitive Shrub All Canopy : Understory Sensitive Canopy: Understory Adapted Canopy: Understory 1 48 34.29 14.40 1.57 4.19 0.79 0.55 0.91 4 30 38.73 8.35 3.80 9.37 0.56 0.06 0.82 8 7 28.68 12.22 2.99 7.48 0.59 0.24 0.81 12 58 44.28 14.43 1.49 7.46 0.74 0.32 0.93 16 119 37.56 14.93 0.00 3.73 0.87 0.60 1.00 17 139 45.56 20.60 1.34 2.35 0.89 0.80 0.94 21 206 44.29 21.45 0.93 3.03 0.89 0.75 0.96 24 265 36.03 13.58 0.78 5.74 0.77 0.41 0.96 25 284 34.73 18.72 1.97 6.90 0.72 0.46 0.89 29 371 29.82 20.18 0.66 5.70 0.77 0.56 0.96 32 439 39.23 11.21 3.24 4.42 0.74 0.43 0.85 33 459 29.10 14.44 1.31 5.91 0.72 0.42 0.91 36 515 44.07 17.93 0.30 3.04 0.90 0.71 0.99 37 537 30.47 16.25 1.35 7.00 0.70 0.40 0.91 43 705 32.70 19.53 1.49 4.03 0.81 0.66 0.91 49 840 34.35 14.49 1.87 8.41 0.65 0.27 0.90 53 887 30.59 12.56 3.42 4.11 0.70 0.51 0.80 57 926 31.15 18.06 1.13 4.29 0.80 0.62 0.93 59 947 33.57 12.82 2.80 6.99 0.65 0.29 0.85 71 1062 39.57 14.04 1.28 6.17 0.76 0.39 0.94 75 1098 37.17 17.38 3.48 6.42 0.69 0.46 0.83 79 1147 36.32 20.35 0.44 3.94 0.86 0.68 0.98 83 1212 36.24 10.82 4.00 6.12 0.65 0.28 0.80 87 1292 41.79 11.59 1.45 4.11 0.81 0.48 0.93 91 1376 38.96 17.47 2.61 8.23 0.68 0.36 0.87 95 1456 30.22 13.78 2.00 11.78 0.52 0.08 0.88 99 1552 34.29 12.57 1.71 6.86 0.69 0.29 0.90 103 1639 44.35 11.75 2.44 4.88 0.77 0.41 0.90 107 1726 44.70 16.23 2.98 8.28 0.69 0.32 0.88 111 1817 43.36 15.62 0.47 7.69 0.76 0.34 0.98 115 1915 34.20 9.43 0.94 8.73 0.64 0.04 0.95 119 2004 38.44 18.84 2.01 5.78 0.76 0.53 0.90 127 2157 41.81 9.89 0.56 8.19 0.71 0.09 0.97 135 2291 41.01 13.15 1.93 6.19 0.74 0.36 0.91 143 2433 48.99 11.69 0.90 13.26 0.62 0.06 0.96 151 2586 48.31 12.80 2.17 9.42 0.68 0.15 0.91 159 2734 40.85 17.96 0.00 12.68 0.65 0.17 1.00

PAGE 87

77 167 2881 40.39 9.73 1.70 6.81 0.71 0.18 0.92 175 3029 40.47 13.26 2.33 9.53 0.64 0.16 0.89 183 3179 48.39 13.90 1.99 8.68 0.71 0.23 0.92 187 3260 42.94 20.72 0.60 6.01 0.81 0.55 0.97 195 3432 35.97 8.35 1.71 7.07 0.67 0.08 0.91 203 3596 40.33 11.69 0.95 9.07 0.68 0.13 0.95 211 3746 28.47 12.68 1.91 5.98 0.68 0.36 0.87 219 3897 45.54 13.61 1.98 4.95 0.79 0.47 0.92 227 4048 36.22 10.21 1.73 5.59 0.73 0.29 0.91 235 4200 39.38 11.93 0.95 4.77 0.80 0.43 0.95 243 4363 35.39 9.55 2.25 9.55 0.58 0.00 0.88 251 4506 35.04 6.15 1.54 7.86 0.63 0.12 0.92 259 4645 42.22 10.87 0.21 7.25 0.75 0.20 0.99 269 4905 42.50 7.60 1.56 7.41 0.70 0.01 0.93 277 5121 49.29 8.53 2.84 11.61 0.60 0.15 0.89