Sedimentary Analysis of Fire History and Paleohydrology

Home , Paleohydrology

SEDIMENTARY ANALYSIS OF FIRE HISTORY AND PALEOHYDROLOGY,

EAGLE LAKE, CALIFORNIA

______

A Thesis

Presented

to the Faculty of

California State University, Chico

______

In Partial Fulfillment

of the Requirement for the Degree

Master of Science

Geosciences

Hydrology/Hydrogeology Option

______

David Benjamin Calo

Summer 2014 SEDIMENTARY ANALYSIS OF FIRE HISTORY AND PALEOHYDROLOGY,

EAGLE LAKE, CALIFORNIA

A Thesis

David Benjamin Calo

Summer 2014

APPROVED BY THE DEAN OF GRADUATE STUDIES AND VICE PROVOST FOR RESEARCH:

______Eun K. Park, Ph.D.

APPROVED BY THE GRADUATE ADVISORY COMMITTEE:

______Sharon A. Barrios, Ph.D. David L. Brown, Ph.D., Chair

Graduate Coordinator ______Susan Riggins, Ph.D.

______Adrienne Edwards, Ph.D.

______Don Hankins, Ph.D. DEDICATION

To all those who dare to dream.



iii ACKNOWLEDGMENTS

This project would have been impossible if not for the support of my friends and family. Stephanie tolerated my odd work hours, trips to Eagle Lake and various labs, and helped me cope with the stress from numerous setbacks. My family funded parts of this thesis when all hope was lost. The constant support of Drs. David Brown, Susan

Riggins, Don Hankins, and Adrienne Edwards was an essential part of this project. I would like to thank Mike Torresan and Dr. Brian Edwards of the USGS Pacific Coastal

Marine Science Center for use of their Sedimentology Lab and Geotek MSCL-S, and to

Dr. Eric Ayers of CSU, Chico for his assistance using the CSU, Chico Department of

Geological and Environmental Sciences Bruker S2 Ranger. I would like to thank Dr.

Susan Zimmerman of the Center for Accelerator Mass Spectrometry at the Lawrence

Livermore National Laboratory, for her assistance in obtaining age-dates. Lastly, some funding for this project was provided by the Geological and Environmental Sciences

Department, California State University, Chico.

iv TABLE OF CONTENTS

PAGE

Dedication...... iii

Acknowledgments ...... iv

List of Tables...... vii

List of Figures...... viii

Abstract...... ix

CHAPTER

I. Introduction...... 1

Purpose ...... 1 Objectives...... 3 Fire Interconnections...... 4 Sedimentary Tracers...... 15

II. Study Area...... 22

Eagle Lake...... 22

III. Methods...... 29

Sediment Core Extraction and Processing ...... 29 Lithostratigraphy ...... 31 Geochronology ...... 31 Fire History...... 34 Paleohydrology...... 36 Zonation Analysis...... 37

v CHAPTER PAGE

IV. Results...... 39

Geochronology ...... 39 Lithostratigraphy ...... 40 Fire History...... 44

V. Discussion...... 49

Paleohydrology of Eagle Lake ...... 49 Fire History of Eagle Lake ...... 51 Regionality ...... 53

VI. Conclusion...... 55

References Cited...... 57

Appendices

A. Core Photographs for ELNEC1-1 ...... 67 B. Geochemical Dataset Used to Determine PCA Scores ...... 70

vi LIST OF TABLES

TABLE PAGE

1. Radiocarbon Age-Dates and Calibrated Age-Dates of Core ELNEC1-1 ...... 40

2. Magnetic Susceptibility (MS) Summary Table of Core ELNEC1-1...... 43

3. Macro-charcoal Concentration (CC) Summary Table of Core ELNEC1-1...... 44

4. Charcoal Influx (CHAR) Summary Table of Core ELNEC1-1 ...... 45

5. Fire Frequency (FF) Summary Table of Core ELNEC1-1...... 47

6. Fire-episode Magnitude (FEM) Summary Table of ELNEC1-1...... 48

vii LIST OF FIGURES

FIGURE PAGE

1. Sediment Transport...... 10

2. Effect of Fire on Plants and Soils...... 12

3. Hydrophobic Soils and Hillslope Debris Flows ...... 13

4. Solar Radiation and Wind Before and After Fire...... 14

5. Fire Induced Dry Ravel ...... 16

6. Geography of Eagle Lake, California...... 23

7. Hydrography of the Eagle Lake Watershed ...... 24

8. Average Annual Water Surface Elevation (1420 - 2012) of Eagle Lake, CA ...... 27

9. AMS Sludge and Sediment Sampler ...... 30

10. Linear Age-depth Model for Core ELNEC1-1...... 40

11. Lithostratigraphic Log of Core ELNEC1-1...... 41

12. Zonation Analysis of Principle Component 1 (PC1) of the X-ray Fluorescence (XRF) Dataset for Core ELNEC1-1 ...... 42

13. Charcoal Concentration (CC) and Magnetic Susceptibility (MS) of Core ELNEC1-1 ...... 43

14. Charcoal Influx (CHAR) of Core ELNEC1-1...... 45

15. Fire Frequency (FF) and Fire-episodes (Peaks) of Core ELNEC1-1...... 46

16. Fire-episode Magnitude (FEM) of Core ELNEC1-1...... 48

viii ABSTRACT

SEDIMENTARY ANALYSIS OF FIRE HISTORY AND PALEOHYDROLOGY,

EAGLE LAKE, CALIFORNIA

David Benjamin Calo

Master of Science in Geosciences

Hydrology/Hydrogeology Option

California State University, Chico

Summer 2014

This project used a high-resolution sedimentary technique to characterize the fire history of Eagle Lake, California from present to 18 kyrs BP, and considered the ecohydrologic interconnection of fire. In order to understand global fire history trends, and the influence of climatic and anthropogenic factors, continued mapping and analysis of fire history as an earth systems process is essential. A previous study analyzed the pa- leoenvironment of Eagle Lake and recommended a high resolution paleofire study be performed at Eagle Lake.

Eagle Lake is an endorheic lake located in central Lassen County, California.

Fire history and paleohydrology were analyzed using macro-charcoal, magnetic susceptibility data, chemostratigraphy (bulk geochemical data), radiochronology (14C), and lithology from a 1 m sediment core (ELNEC1-1). Fire history was characterized by ix entering macro-charcoal data and age-date information into CharAnalysis. Paleohydrol- ogy data was characterized using lithology and magnetic susceptibility data.

Chemostratigraphy was used to enhance lithostratigraphy via zonation analysis. Radio- carbon age-dating was used to establish a geochronology of core ELNEC1-1.

The fire history of the Eagle Lake can be divided into three periods: 18 to 10 kyrs BP, 10 to 2.8 kyrs BP, and 2.8 kyrs BP to present. Hydrologic activity during the first period (18 to 10 kyrs BP) was low, limiting fuel availability, and supporting small, frequent fires. Hydrologic activity during the second period (10 to 2.8 kyrs BP) was initially low due to limited fuel availability, supporting small frequent fire. The beginning of the second period marks the end of the glacial period. After approximately 8 kyrs BP, hydrologic activity increased to moderate/high levels, and fuel availability increased, supporting larger moderately frequent fires until 4.2 kyrs BP. From 4.2 to 1.5 kyrs BP, fire frequency and magnitude were low due to neoglaciation. Hydrologic activity during the third period (2.8 kyrs BP to present) was moderate and decreased to present, allowing for low to moderately frequent fires with moderate magnitude through to period end. The climate of Eagle Lake prior to 15 kyrs BP was cold and moist, from 15 to 8.5 kyrs BP it was warm and dry, from 8.5 kyrs to 4 kyrs BP it was warm and wet, and from 4 kyrs BP to present it was cool and wet. Indigenous populations clearly influenced the fire history of Eagle Lake, moderated fire frequency over the Holocene, and set fires for forest understory management, wildlife management, manufacturing, and agricultural purposes.

Despite evidence of indigenous occupation, it is difficult to distinguish anthropogenic influences and climatic trends.

CHAPTER I

INTRODUCTION

Purpose

Fire has influenced the landscape composition of Earth for the last 420 million years, affecting hydrology, geomorphology, and ecosystem dynamics across various spatiotemporal scales (Shakesby and Doerr, 2006; Whitlock et al., 2010). At broad scales, climate, vegetation, and anthropogenic factors are the primary controls that govern ignition, biomass availability, and fire suppression (see Fire drivers section). Fire alters vegetative composition through succession dynamics (Marlon et al., 2013). Fire alters erosion patterns by exposing rock faces, altering soil composition, and modifying watershed hydrology (precipitation storage, evapotranspiration, and snowmelt timing)

(Shakesby and Doerr, 2006). Fire-generated sediments can contain elevated levels of charcoal, magnetic compounds, radionuclides, and various other compounds (see

Sedimentology section). During and after fire events, fire-induced compounds, as well as radionuclides, are transported via eolian and hydrologic transport, and preserved in lake sediments (see Sedimentology section).

The region of interest for this fire history study is northern California. Starting in the late 1800s, settler-set controlled fires and fire suppression caused fire history to break from climatic equilibrium, inducing a fire deficit in the western United States

(Marlon et al., 2012), and these same principles of suppression were applied elsewhere.

1 2

Between 2000 and 2010, vegetated landscapes (globally) experienced an increase in the number of large uncontrolled fires – even those with fire-suppression programs (Marlon et al., 2013). It should be noted that in some ecosystems, indigenous fire regimes have kept fire intensity low by intentionally setting frequent fires, which maintained a fine age- class mosaic and mitigated overall fire size (Bird et al., 2008). In temperate zones, like northern California, larger fires have been attributed to a combination of climate change, rural depopulation, and fire suppression (Marlon et al., 2013). In order to understand global fire history trends, and the influence of climatic and anthropogenic factors, continued mapping and analysis of fire history as an earth systems process is essential

(Bowman et al., 2009; Marlon et al., 2012).

Dendrochronology and sedimentary charcoal time-series are the two primary records of long-term fire history (Whitlock et al., 2010). Typically, dendrochronology can characterize fire history to approximately 1 kyrs BP, and can provide annual data, but is temporally limited by stand age, and species characteristics (ring growth, fire resiliency, and decay resistance) (Conedera et al., 2009). The sedimentary approach to fire history surpasses the temporal limitations of dendrochronology, with potential temporal ranges extending back to the appearance of terrestrial plant life during the

Silurian (420 Myrs BP), and potential temporal resolutions of ≥1 yr per observation interval (Bowman et al., 2009; Conedera et al., 2009). The main drawbacks of the sedimentary approach is that there are the many methods available (Rhodes, 1998), and that extrapolation of spatial fire dynamics would require a sediment budget (e.g., Blake et al., 2009). The sedimentary method was selected for its temporal advantage over dendrochronology.

Sedimentary tracers are byproducts of an interconnection between hydrology, ecology, and fire (hereafter the ecohydrologic interconnection with fire). Previous paleofire studies used lake sediment cores to characterize fire histories with temporal ranges in excess of 10 kyrs BP (e.g., Minckley et al., 2007), making fire paleofire tracers ideal for characterizing long-term fire histories. An analysis of peer-reviewed literature found that only four articles used paleofire tracers to determine fire history in northern

California, north of Chico, CA, and most were in the Klamath Mountains; only one article focused on northeastern California as of March, 2013. Most current paleofire studies used high resolution sampling intervals (e.g., Minckley et al., 2007).

Consequently, this study used paleofire tracers to characterize long-term fire history in northeastern California using high-resolution sampling (hereafter referred to as a high resolution fire history). Paleohydrology was also considered to enhance the analysis of fire history.

Objectives

In northern California, a few paleofire studies have shown spatially variable regional fire frequencies, displaying both increases and decreases over the past 1 kyrs

(e.g., Mohr et al., 2000; Minckley et al., 2007; Briles et al., 2008). Minckley et al. (2007) noted that the fire frequency of the northwestern Great Basin increased over the past 6 kyrs. Over the past 1 kyrs, lakes in western Great Basin, adjacent to the California-Great

Basin (CGB) drainage divide, exhibited more dramatic fire frequency increases than those opposite the divide (e.g., Minckley et al., 2007; Hallett and Anderson, 2010). Do geographically similar lakes exhibit similar fire history trends? Insufficient data was

4 available to conclusively answer this question. Given these observations, additional fire history characterization along the CGB drainage divide was warranted.

Wigand et al. (1995) indicated that the vegetative composition and charcoal flux of Eagle Lake, CA has varied over the past 15 kyrs. Given the unexplored variability and geographic location of Eagle Lake in the CGB, this site was an ideal candidate for a high resolution paleofire analysis. Most paleofire tracer studies analyze sediment cores for macroscopic charcoal (macro-charcoal), magnetic susceptibility (MS), and radiochronology. It was also found that x-ray fluorescence (XRF) could enhance lithostratigraphic data (lith-logs) (e.g., Kylander et al., 2012). A few other paleofire studies in northeastern California have characterized super fire regimes; for comparison, the fire history of Eagle Lake, CA was characterized as close to the regional and millennial scale as possible, using macro-charcoal, MS data, radiochronology, XRF data, and lithology from a lake sediment core. The paleohydrology of Eagle Lake was interpreted using MS and lithostratigraphy.

Fire Interconnections

Fire is a complex landscape process. Fire history can be difficult to characterize even with historic records because they may be missing, incomplete, or of poor quality (e.g., Lloret and Mari, 2001). Paleoenvironmental data is important because it enables the analysis of Earth’s history beyond historical records. Fire history characterized via paleofire tracers can be used for a broad paleoenvironmental analysis of watershed hydrology, geomorphology, and ecology. An understanding of fire history terminology, fire drivers, and sedimentary tracers is integral to understanding fire history.

Fire History Terminology

The analysis of fire history is critical to determine fire trends. Fire regimes are collections of ecologically relevant metrics that characterize fire systems, occurring over defined spatiotemporal scales that are based on fire history and environmental data

(Conedera et al., 2009; Whitlock et al., 2010). Fire regimes can be used to compare difference periods of fire history. In the context of paleofire data, fire regimes operate over centuries to millennia, at regional to global spatial scales, and are driven by climate, vegetation, and other natural or anthropogenic landscape variables (Whitlock et al.,

2010). Fire regimes are therefore defined according to particular study needs (Conedera et al., 2009). Whitlock et al. (2010) suggested that, in the context of paleofire data, fire regime definitions should consider only high-temporal ranges, so scales in thousands of years were favored. Additionally, since lacustrine environments are a product of watershed dynamics, the watershed extents were used as the landscape extent. Therefore, to analyze the ecohydrologic interconnection of fire, this study focused on the fire regimes as close to regional and millennial scales as possible.

Fire Drivers

Fire drivers are components of fire occurrence, and different sets of fire drivers operate at different spatiotemporal scales (Whitlock et al., 2010). Each fire driver influences the occurrence of fire, and the drivers exhibit dependent and/or independent characteristics between one another (Whitlock et al., 2010; Marlon et al., 2013). The interactions between fire and its drivers can exhibit lag-times from years to millennia

(Marlon et al., 2012), making the temporal range of observation critical. To tie in with

6 similar studies in northeastern California (e.g., Minckley et al., 2007; Beaty and Taylor,

2009), super fire regimes variables were considered.

Climate-fire link. The climate-fire link is a top-down fire driver. It influences net primary productivity and biomass availability (fuel) via long-term temperature, precipitation, evaporative and ignition trends, with fire typically more frequent during drought conditions, and less frequent during humid conditions (Whitlock et al., 2010;

Marlon et al., 2013). The climate also has a direct influence on patterns of erosion (see

Sedimentology section). At regional scales, climate is affected by shifts in solar radiation and atmospheric composition, which influence latitudinal temperature gradients and land- sea temperature contrasts that drive global circulation patterns and surface-climate anomalies (Marlon et al., 2013). Compared to glacial-interglacial climate changes,

Holocene climate change has been subdued, but has still affected biome-scale vegetative compositions and fire regimes (e.g., boreal forest extent) (Marlon et al., 2013). In western

North America, climatic variability is primarily driven by the Pacific Decadal Oscillation

(PDO), which affects the amount of precipitation delivered to the Pacific Coast (Minnich,

2006; Marlon et al., 2012). Inter-annual climate variability is primarily driven by the El

Niño-Southern Oscillation (ENSO) and temperature, precipitation, snowpack, and snowmelt timing (Marlon et al., 2012).

Biome-fire link. The biome-fire link is a bottom-up fire driver, since primary productivity is responsible for the production of combustible biomass (Whitlock et al.,

2010). Fire and vegetative processes establish biome composition, which affects rates of erosion and vegetative microclimates via land cover dynamics (devegetation/ revegetation) (see Sedimentology section). Moderate to high severity fire can also destroy

7 root systems and disrupt seedbanks (Moody and Martin, 2009). This disruption to the seedbank may prolong the revegetation process, and maintains elevated erosion rates (see

Dry ravel section). Fire can also speed vegetative establishment by removing allelopathic substances, by creating mineral seedbeds, or by removing impermeable seed coats

(Moody and Martin, 2009). Net primary productivity and biomass combustion are affected by combination of vegetative and climatic variables, so the fire-vegetation link is also mediated by climatic conditions (Marlon et al., 2013).

The climate-biome link influences the composition, spatial distribution, and structure of vegetative communities via succession and the alteration of net primary productivity (Marlon et al., 2013). For example, at grassland-forest interfaces, cycles of biomass buildup, drying, and combustion affect the spatial extent of forests and grasslands mediated by fire disturbance, which influences succession (Falk et al., 2007;

Marlon et al., 2013). Independent of the climate-biome link, the biome-fire link alters community composition through post-fire successional dynamics and community structure (Marlon et al., 2013). For example, the buildup of fire resistant species and biomass, as well as stand-age and tree density, can mediate fire cycles and fire spread characteristics (Marlon et al., 2013). Fire tends to occur in environments with intermediate climatic and vegetative conditions, where neither the climate nor the biome are limiting factors (Marlon et al., 2013).

Anthropogenic-fire link. Humans have influenced fire history for thousands of years by altering ignition rates, fuel availability, and land cover (Whitlock et al., 2010;

Hurteau and Brooks, 2011). At fine scales, humans have broad impacts on fire history through agricultural practices, the introduction of non-native species, ignition (accidental

8 and intentional), fire suppression, and indirectly through climate change (e.g., longer fire seasons) (Westerling, 2006; Whitlock et al., 2010; Hurteau and Brooks, 2011; Marlon et al., 2012). It should be noted that the indigenous Californians have altered fire regimes

(Lake, 2013), that indigenous northeastern Californians had the ability to set fires

(Simmons et al., 1997; McGuire, 2007), and fire suppression occurred after non- indigenous settlement. Fire tends to be more frequent in areas with intermediate population densities, and tends to be less frequent in sparsely populated (extreme environments) and densely populated areas (due to fragmentation and fire suppression)

(Marlon et al., 2013). Humans alter vegetative composition, shift biomes, affect biomass availability, fragment landscapes, and alter moisture regimes (Whitlock et al., 2010).

Humans affect fire history most where ignitions are naturally rare, are not limited by climate, and where biomass alterations increase the vegetative susceptibility to fire

(Whitlock et al., 2010).

Currently, no methods exists to distinguish between anthropogenic and natural fires within the paleo-record (Marlon et al., 2013); however, in conjunction with historical records, archeological evidence, and GIS, paleo-records can provide valuable insight into human-fire interactions, and can help disentangle underlying human and climatic trends (Whitlock et al., 2010; Marlon et al., 2012; Lake, 2013). In temperate forests, during the Holocene, climate was the primary driver of vegetation and fire history in areas that lacked human settlements capable of affecting fire and vegetative composition (Whitlock et al., 2010). Northeastern California has been occupied for at least the last 11.45 kyrs ± 340 yrs BP (McGuire, 2007). Indigenous Californians used fire to clear forest understory, manage wildlife, for manufacturing, and for agricultural

9 purposes – in particular basket weaving materials (Anderson, 2006). In the western

United States, from 500 CE to 1800 CE, fire frequency remained in equilibrium with the climate, but sharply declined from fire frequency trends beginning in 1870 CE (based on climate alone), indicating disequilibrium with climatic variability due to anthropogenic influences (including the decimation of native populations) (Marlon et al., 2012; Lake,

2013). The fire frequency decline is now described as a regional fire deficit (Marlon et al., 2012).

Sedimentology

During and after fire events, fire-induced compounds are transported and stratigraphically preserved in lake sediments (Minckley et al., 2007; Blake et al., 2009;

Conedera et al., 2009; Shakesby, 2011). In lakes, typically only the upper sediment layer

(1-2 mm) is well oxygenated (Wetzel, 2001), which can enhance preservation within sediments (provided prolonged lake desiccation is absent). Lacustrine sediments can be classified as either organic or inorganic (clastic and chemical) (Schnurrenberger et al.,

2003; Conedera et al., 2009), with fire derived sediments typically containing increased concentrations of paleofire tracers such as charcoal, magnetic particles, radionuclides, and various other compounds (Amiro et al., 1996; Johansen et al., 2001; Blake et al.,

2009; Conedera et al., 2009). Fire derived sediment composition varies widely, and is dependent on source materials (Smith et al., 2011). Following fire, pulses of erosion are common, and are due to decreased vegetated land cover (Sugihara and Barbour, 2006).

Generally, fire intensity and severity are proportional to increased erosion

(Conedera et al., 2009), and increased fire-derived sediment concentrations – including paleofire tracers useful for tracking fire (Blake et al., 2009; Wilkinson et al., 2009).

Erosion typically increases as a function of precipitation, wind, and devegetation

(Nichols, 2009; Owens et al., 2010). Biome trends toward more or less vegetation would also increase or decrease erosion respectively. Post-fire erosion occurs in two transport phases: (1) primary transport during to shortly after fire, and (2) secondary transport during non-fire years (Figure 1) (Conedera et al., 2009). Primary transport occurs via runoff, eolian transport, and dry ravel (Shakesby and Doerr, 2006; Conedera et al., 2009).

Secondary transport occurs via runoff, rain-splash, eolian transport, and subaqueous redistribution within the lake (re-suspension, mixing, slumping) (Figure 1) (Selby, 1993;

Shakesby and Doerr, 2006; Higuera et al., 2007; Conedera et al., 2009).

Figure 1. Sediment transport. Displays sediment transport before and after fire.

Source: Adapted from Conedera, M., Tinner, W., Neff, C., Meurer, M., Dickens, A.F., and Krebs, P., 2009, Reconstructing past fire regimes: Methods, applications, and relevance to fire management and conservation: Quaternary Science Reviews, v. 28, p. 555–576.

Following fire, it is thought most sediment would initially remain close to or at the source area. Sediment is transported away from the source area by water, wind, dry ravel, and, rarely, mass movements (Shakesby, 2011). Sediment remaining at the source area may be transported during secondary transport. In many cases, sediment sinks within watersheds can capture sediments prior to reaching lacustrine environments (Owens et al., 2013). Sediment storage is driven by vegetation, bioturbation, and topography, with possible sinks including pit and mound topography (Selby, 1993), bedload storage

(DeBano, 2000; Wilkinson et al., 2009), pore-space entrainment (micro and macro), dry ravel buildup, and other structures (Shakesby and Doerr, 2006; Shakesby, 2011). The potential for sediment storage increases in large watersheds (Owens et al., 2013).

Therefore, secondary transport is a mixture of sediment from sinks and sources.

Hydrologic transport. The primary mode of sediment transport after fire is hydrologic transport (Shakesby and Doerr, 2006). Fire of sufficient magnitude and duration affects hillslope processes via the creation of hydrophobic surfaces (Figure 2), covered by erodible ash and sediment, which varies spatially on the landscape (DeBano,

2000). Fire increases runoff (overland flow, interflow) and channel flow from within fire- impacted watersheds due to land cover change (defoliation, and litter layer removal)

(Ward et al., 2003; Valeron and Meixner, 2009; Smith et al., 2011). After fire, runoff mobilizes sediments, which accumulates on lower slopes or in sediment sinks, enters drainage networks, or are deposited directly in lakes (Selby, 1993; Doerr et al., 2006;

Shakesby and Doerr, 2006; Conedera et al., 2009); additionally, when saturation excess overland flow occurs, rill networks and micro-debris flows may occur (Figure 3) (Doerr et al., 2006; Shakesby and Doerr, 2006). Low-density sediments, such as charcoal, soot,

Figure 2. Effect of fire on plants and soils.

Source: Adapted from DeBano, L.F., 2000, The role of fire and soil heating on water repellency in wildland environments: A review: Journal of Hydrology, v. 231, p. 195–206.

etc., would likely float on top of overland flow at first. When high rates of erosion due to runoff persist, revegetation may be prolonged (Wang et al., 2013).

Erosion from forested watersheds is typically low (Doerr et al., 2006); however, post-fire sediment yields have been reported to increase from approximately 1 –

1500 times the rate of comparable unburnt sites (Smith et al., 2011). Sediment yields from hillslope runoff can be a significant portion of post-fire sediment influx, and correlates well with elevated paleofire tracer concentrations (Conedera et al., 2009).

However, other hillslope processes can alter erosion rates independent of fire (mass- wasting, bioturbation, devegetation, etc.) (Conedera et al., 2009; Owens et al., 2010). For instance, patches of fire-induced hydrophobic surfaces intermixed with wettable patches,

Figure 3. Hydrophobic soils and hillslope debris flows.

Source: Adapted from DeBano, L.F., 2000, The role of fire and soil heating on water repellency in wildland environments: A review: Journal of Hydrology, v. 231, p. 195–206.

or macropores (e.g., fractured rock, bioturbation), can act as infiltration sinks, subduing expected runoff (Blake et al., 2009; Malkinson and Wittenberg, 2011). Further, sediment sinks can delay transport up to hundreds of years (Selby, 1993; Conedera et al., 2009) or more.

Eolian transport. In arid to semiarid environments, fire can increase eolian transport (Sankey et al., 2009), with low-density fire-generated sediments highly susceptible to eolian transport (Shakesby and Doerr, 2006). Wind can increase erosion rates up to 70 times unburned rates during strong wind events (Shakesby and Doerr,

2006). During particularly severe crown-fires, strong convective currents and vortices can loft and transport centimeter-sized charred particles kilometers from the source area

(Pisaric, 2002; Tinner et al., 2006). Eolian transport can also transport sand sized clastic particles across the landscape, and can carry silts/clays as suspended load during high speed winds (~30 m-s-1) (Pisaric, 2002; Nichols, 2009). Eolian transport is most common in areas with sparse vegetation (Nichols, 2009), with post-fire eolian transport most significant when vegetated land cover has been reduced (Figure 4). Eolian transport would become more significant during shifts toward a less vegetated biome.

Figure 4. Solar radiation and wind before and after fire.

Source: Adapted from Van Wagtendonk, J.W., 2006, Fire as a physical process, in Sugihara, N.G., van Wagtendonk, J.W., Fites-Kaufman, J., Shaffer, K.E., and Thode, A.E., Fire in California’s ecosystems: Los Angeles, California, University of California Press, p. 38–57.

Soil moisture plays an important role in eolian transport, with soil moisture generally proportional to soil cohesion (Sankey et al., 2009); however, fire induced soil hydrophobicity increases the amount of soil moisture necessary for sediment cohesion

(Sankey et al., 2009), making fire-generated sediments more erodible. In general atmospheric moisture content is inversely proportional to erosivity (Sankey et al., 2009);

15 however, when temperatures are low, high relative humidity can increase soil erosivity since cold air holds less moisture (Sankey et al., 2009). Additionally, the removal of vegetated land cover by fire increases the amount of solar radiation reaching the ground, which elevates ground surface temperatures, and decreases soil moisture content within the vegetative microclimate (Van Wagtendonk, 2006).

Dry ravel. In the western United States, the post-fire mobilization of dry sediments downslope due to gravity (dry ravel) is commonly reported (Shakesby and

Doerr, 2006). Dry ravel occurs most on steep, south-facing slopes in the Northern

Hemisphere (most solar radiation per day) that consist primarily of weathered bedrock and organic matter (Shakesby and Doerr, 2006). In California, chaparral areas, with

Mediterranean climates, have long dry seasons; during which dry ravel is most likely to occur (Shakesby and Doerr, 2006). As sediments are transported, they may be temporarily trapped by sediment sinks. In chaparral areas, sediment builds up behind vegetation on the landscape, and are released when the vegetative support structure is destroyed (Figure 5) (Shakesby and Doerr, 2006). Revegetation mediates transport by dry ravel, with revegetation capable of halting dry ravel by the second growing season, but can take a decade or more when high rates of ravel persist (Shakesby and Doerr, 2006).

Dry ravel would also would become more significant during shifts toward a less vegetated biome.

Sedimentary Tracers

There are a number of sedimentary tracers that can be used to characterize fire history and paleohydrology, but only a few are common. In general, the use of tracers

Figure 5. Fire induced dry ravel.

Source: Adapted from Shakesby, R.A., and Doerr, S.H., 2006, Wildfire as a hydrological and geomorphological agent: Earth-Science Reviews, v. 74, p. 269–307, doi: 10.1016/j.earscirev.2005.10.006.

from lake sediments requires extraction, processing, and analysis, which can be time consuming and expensive (Conedera et al., 2009). Sediment analysis can be quantitative or qualitative (Brunelle and Anderson, 2003; Conedera et al., 2009), and observation can occur at multiple scales (e.g., Rhodes, 1998; Minckley et al., 2007). The temporal resolution for sediment core analysis is limited to ≥1 year due to secondary transport and lake re-mixing of sediments (Selby, 1993; Conedera et al., 2009; Owens et al., 2013).

Sedimentary analyses such as elemental geochemistry, magnetic susceptibility, and geochronology are most commonly used in conjunction with paleofire tracers. The paleofire tracers then can be used to determine sedimentology and hydrologic characteristics (e.g., Minckley et al., 2007; Breckenridge et al., 2010; Ortega et al., 2010;

Cuven et al., 2011).

Paleofire Tracers

Paleofire tracers are ideal when the area lacks old growth trees, when lake sediments are available, and when fire history time-span is a priority (e.g., Walsh,

Whitlock, et al., 2010). Paleofire tracers accumulate stratigraphically, and studies using paleofire tracers take advantage sediment core stratigraphy to provide chronologic evidence for the characterization of fire history (e.g., Minckley et al., 2007); making undisturbed sediments important. Further, it is possible to integrate paleofire tracers with dendrochronologies when available (e.g., Beaty and Taylor, 2009). Different sediment processing methods can create functional differences between studies (e.g., Schlachter and Horn, 2009), so the most common methods will be used to make this study compatible with future studies. Due to their common usage among studies analyzing sediment cores (e.g., Minckley et al., 2007; Hallett and Anderson, 2010; Walsh, Pearl, et al., 2010), macro-charcoal analysis and geochronology were used to characterize the fire history of Eagle Lake.

Macroscopic charcoal. Minckley et al. (2007) and Walsh, Whitlock, et al.

(2010) employ methods typical of most macro-charcoal studies. Charcoal particles >100

μm are typically considered macro-charcoal (Whitlock and Larsen, 2001). Charcoal is one of the slowest decaying components of the carbon cycle because it is geochemically resistant to oxidation (Ascough et al., 2011). Macro-charcoal as an indicator of past fire events is advantageous because it can be analyzed under low-level magnification, can quantify biomass burning, and because fire regime metrics can be quantified by processing macro-charcoal data with CharAnalysis (e.g., Minckley et al., 2007). In some cases, charcoal can be found in pristine condition and identified to the species level even

18 after thousands of years (Ascough et al., 2011). Charcoal sources can be local, extra-local

(slightly beyond the watershed), regional, continental, or global in origin (Conedera et al.,

2009). The dispersal distance of macro-charcoal due to fire is typically localized (1 – 103 m), but can be transported kilometers from the source area (Higuera et al., 2007; Peters and Higuera, 2007). Once charcoal is deposited, it is temporally and contextually relevant long after land cover has changed, and even after conversion to sedimentary rock (Scott et al., 2000; Conedera et al., 2009).

Typically, charcoal can be used to determine fire history over a temporal range of 103 – 104 yrs (e.g., Brunelle and Anderson, 2003; Minckley et al., 2007; Walsh,

Whitlock, et al., 2010). However, sedimentary charcoal has two spatial limitations: (1) spatial fire dynamics cannot be quantified below the minimum transport range without source area tracking (e.g., Blake et al., 2009; Wilkinson et al., 2009), and (2) spatial fire dynamics cannot be quantified above the maximum transport range without multiple study sites (e.g., Walsh, Whitlock, et al., 2010). Spatial limitations can be overcome using multiple study sites (e.g., Walsh, Whitlock, et al., 2010), and by estimating metrics using a reference dendrochronology dataset (e.g., Higuera et al., 2011). Sedimentary charcoal is temporally limited as well, since maximum temporal resolution can only be obtained with annually varved sediments (Conedera et al., 2009). The maximum resolution of unvarved sediments is temporally limited by sedimentation rate and age- date accuracy (Conedera et al., 2009). Lastly, macro-charcoal analysis lacks standard methods, and may be prone to misidentification, double counting, etc. (Rhodes, 1998;

Schlachter and Horn, 2009).

Paleohydrology Tracers

Like fire history, paleohydrology is difficult to quantify, but broad trends can be characterized using magnetic susceptibility and geochemical data (e.g., Breckenridge et al., 2010; Ortega et al., 2010). In general, the use of paleohydrologic indicators requires many of the same procedures as with paleofire indicators, so they can be easily paired with paleofire tracers to enhance knowledge of particular transport pathways via sedimentologic analysis (e.g., Ortega et al., 2010). For instance, higher sedimentation rates are likely during wet climates, while lower sedimentation rates are likely during dry climates. Due to their use in other studies (Breckenridge et al., 2010; Ortega et al., 2010;

Kylander et al., 2012), lithostratigraphy, magnetic susceptibility, and chemostratigraphy were used to characterize the paleohydrology of Eagle Lake.

Magnetic susceptibility. Magnetic minerals are produced in soils by natural and anthropogenic processes, and typically exist as fine grains, mineral coatings, concretions, and aerosols (clays-silts) (Blake et al., 2006; Maher, 2011). The most common magnetic minerals are hematite, goethite, maghemite and magnetite, and account for the primary magnetic signature of soils (Blake et al., 2006); Magnetic susceptibility (MS) can be used to determine terrestrial depositional flux (e.g., Walsh,

Whitlock, et al., 2010), which can be used to identify periods of increased hydrologic transport (e.g., Ortega et al., 2010) since terrigenous materials typically have higher metal concentrations. Magnetic susceptibility is advantageous because it can be used to characterize sediment (e.g., Collins and Walling, 2004; Blake et al., 2006), and because it can be analyzed non-destructively (whole core scans). The major limitation of MS is that

20 other processes can potentially alter soil magnetism (Blake et al., 2006; Conedera et al.,

2009).

Chemostratigraphy. Lithologic descriptions are the traditional way of characterizing sediments. Geochemical analysis can provide compositional data for both igneous and sedimentary rocks (chemostratigraphy; e.g., Walther, 2009). The upper continental crust is composed of approximately 98% oxides, containing Si, Al, Fe, Ca,

Na, K, Mg, Ti, P, and Mn, with other trace elements (Walther, 2009), and geochemical analysis of sediments can provide elemental level data (ppm, percentages, etc.). Major

(bulk) geochemical analysis can be used to detect hydrologic shifts (e.g., Ortega et al.,

2010), stratigraphic shifts (tephras, etc.) (e.g., Kylander et al., 2012), etc. Common laboratory techniques for bulk geochemical analysis include atomic absorption XRF

(Walther, 2009). XRF is non-destructive and requires little preparation (Artiola et al.,

2004; Shackley, 2011), making it easy to pair chemostratigraphy with other analyses; enhancing support for transport pathways or hydrologic shifts. For example, in sedimentary environments, Al, Ti, Fe, Ca and K typically indicate allogenic sediments

(Ortega et al., 2010), so XRF data can be easily paired with lithology or MS data. XRF is only limited by instrumental precision and cannot yield data below the elemental level

(e.g., isotopic data). Given that lithologic descriptions can be time consuming and facies changes can be missed, chemostratigraphy was used to enhance lithostratigraphy.

Chronology

Chronology is an important aspect of paleoenvironmental analysis using sediment cores, and there are a number of methods for creating geochronologies (varve counting, radiochronology, etc.; Jull et al., 2006; Minckley et al., 2007; Power et al.,

2010). When creating a geochronology for a sediment core, radiochronology via cosmogenic isotope analysis is most common (e.g., Minckley et al., 2007). Notable cosmogenic radionuclides used for age-dating are 14C, and 10Be (Faure and Mensing,

2005). Cosmogenic radionuclides primarily occur when cosmic rays bombard particles

(O2, N2, etc.) in the upper atmosphere, causing nuclear spallation and the formation of secondary particles that react in the lower atmosphere, which are deposited at the Earth’s surface (Faure and Mensing, 2005). Cosmic rays may also form cosmogenic radionuclides at the Earth’s surface (Faure and Mensing, 2005).

Radiochronology can be used to create age-depth models for sediment cores

(e.g., Walsh, Whitlock, et al., 2010), and can even be used with unvarved sediments

(Conedera et al., 2009). The most common radionuclide analyzed in paleofire studies is radiocarbon (14C) (e.g., Minckley et al., 2007). When analyzing radiocarbon (14C), extra care should be taken to correct for atomic bomb-contamination, the Suess effect (dilution of atmospheric 13C and 14C by 12C from fossil fuels beginning about 1830 CE) (Faure and

Mensing, 2005), and the possibility of radiocarbon contamination (Buchholz et al., 2000) to obtain reliable age-dates. Once collected, it is important that age-datable materials remain isolated, since radiochronology relies on the assumption that the sample is an isolated system (Faure and Mensing, 2005). General limitations of radiochronology are the need for a closed system, and the need for sufficient age-datable material (Faure and

Mensing, 2005). Finally, when determining age-dates, a degree of uncertainty is warranted due to systematic errors, cosmogenic depositional flux, etc. (Jull et al., 2006).

CHAPTER II

STUDY AREA

Eagle Lake

Eagle Lake is an endorheic, alkaline lake located in central Lassen County, approximately 25km northwest of Susanville, CA, with an elevation of approximately

1555 m (Figure 6) (Gester, 1962; Carmona-Catot et al., 2012). Eagle Lake has been described as being within the territory of various native tribes (Simmons et al., 1997;

McGuire, 2007), was known to native populations (Atsugewi and Mountain Maidu) as

Acapsu’kati or Hinch’esimim momdanim (reflection lake) (Gester, 1962; Simmons et al.,

1997), and is the second largest naturally occurring lake completely within California

(Carmona-Catot et al., 2012). Eagle Lake has a surface area of approximately 100 km2, has an irregularly shaped shoreline with a perimeter of approximately 100 km, and is located along the eastern edge of the Eagle Lake Watershed (ELW) (Figure 7) (Gester,

1962). The ELW has an area of approximately 1000 km2 (Carmona-Catot et al., 2012).

The Eagle Lake Watershed lies east of the CGB drainage divide, with the ELW’s western edge coincident with the CGB drainage divide (Figure 6). Precipitation at Eagle Lake ranges from 46 – 152 cm-yr-1 and is predominantly snowfall (Carmona-Catot et al.,

2012). Eagle Lake consists of three interconnected bays: two northern bays

(approximately 5 – 6 m deep), and a southern bay (approximately 10 – 20 m deep)

(Carmona-Catot et al., 2012).

22 23

Figure 6. Geography of Eagle Lake, California. 24

Figure 7. Hydrography of the Eagle Lake watershed. 25

Geologic Setting

Physiographically, the ELW is within the Great Basin and adjacent to the

Cascade and the Sierra-Nevada Mountains (Figure 6). The ELW is a geologic hybrid of all three regions, containing lava flow deposits, small volcanic cones, low-velocity streams, and north-south faulting (Gester, 1962). The geologic composition of the ELW is primarily andesite, dacite, rhyolite, and basalt (Gester, 1962). Older lake deposits also surround Eagle Lake (Grose et al., 1992). Aquifers within the ELW consists of

Quaternary lake deposits and Holocene basalt flows (DWR, 2004). Based on wells logs, the Quaternary lake deposits are approximately 5 – 10 m thick, and consist of thinly bedded clays, silts, and sands (DWR, 2004). The Holocene basalt flows in the ELW are highly jointed, and range from approximately 10 – 150 m in thickness (DWR, 2004). The jagged surface expression of the ELW Holocene basalts distinguishes them from older basalt flows, with individual flows separated (top and bottom) by columnar jointing of 1

– 2 m in thickness (DWR, 2004). The likely origin of Eagle Lake is the blockage of

Willow Creek by lava flows (e.g., Gester, 1962; Harding, 1965), given that lava deposits bound the ELW (Grose et al., 1992) where Pine Creek and Willow Creek may once have connected. Eagle Lake likely formed during the Pliocene or early-Pleistocene (Gester,

1962).

Sedimentology. Wigand et al. (1995) logged a sediment core from Eagle Lake and identified a beach unit from 20 – 30cm (7.02 kyrs ± 50 yrs BP), a silt unit from 48 –

53cm (15.36 kyrs ± 210 yrs BP), and silts overlying beach unit at 140cm (15.4 kyrs ± 230 yrs BP). Up gradient from Eagle Lake (McCoy Flat, Figure 7), the sedimentation rate at was 1 m per 10 kyr during the Pleistocene, and 1 m per 1 kyr (or less) during the

Holocene (Wigand et al., 1995). Within Eagle Lake, the sedimentation rate during the

Pleistocene was 1 m / 100 yrs, suggesting that the rate within Eagle Lake was higher during the Pleistocene than the Holocene (Wigand et al., 1995); most likely caused by land cover change (Wigand et al., 1995). During the Pleistocene the climate of the intermountain west was cold and dry, resulting in sparse vegetation and increased erosion via dry ravel (Wigand et al., 1995). During the Holocene, temperature and precipitation increased, increasing vegetation and decreasing erosion (Wigand et al., 1995). Holocene erosion increases were due to increased precipitation, temporal distribution of precipitation, and other activities (e.g., logging and fire) (Wigand et al., 1995).

Hydrologic Setting

Eagle Lake is a natural alkali lake with three interconnected bays, and is fed by precipitation, surface water, and groundwater inflow (DWR, 2004; Carmona-Catot et al., 2012). The main tributary of Eagle Lake is Pine Creek, which accounts for 65% of surface water inflows (Carmona-Catot et al., 2012). Minor tributaries to Eagle Lake include Cleghorn Creek, Papoose Creek, and Merrill Creek (Figure 7) (Gester, 1962;

Carmona-Catot et al., 2012). Surface water within the ELW is primarily spring fed, with groundwater recharged primarily from infiltration into young upland volcanics in the western ELW (DWR, 2004; Carmona-Catot et al., 2012). Generally, groundwater in the

ELW flows in an east-northeast direction, with groundwater from the ELW recharging the headwaters of Willow Creek (Susan River and Honey Lake tributary) (Gester, 1962;

Harding, 1965; DWR, 2004). Since Honey Lake was a tributary to ancient Lake Lahontan

(Russell, 1885), Eagle Lake is associated with the Great Basin. Due to this, the Lahontan

Region was used to define the extent of the Great Basin in Figure 6.

The water level of Eagle Lake has fluctuated throughout its history, as evident by wave cut terraces, beach sands, mollusk shells, and older lake deposits surrounding

Eagle Lake (Gester, 1962; Grose et al., 1992). Over the past 600 yrs, the water elevation of Eagle Lake has fluctuated (approximately) from a low of 1552 m to a high of 1562 m

(Figure 8). Older lake deposits (Qelo) exhibit a maximum highstand of approximately

Figure 8. Average annual water surface elevation (1420 - 2012) of Eagle Lake, CA. Note: Blue line indicates water levels estimated via dendrochronology, and the red line indicates surveyed elevations.

Source: Adapted from Harding, S.T., 1965, Recent variations in the water supply of the western great basin: Berkeley, California, University of California, Water Resources Center Archives 16, 226 p. (post- 1955 data obtained from CA DWR and Owen Bateson of Spaulding, CA).

1567 m (Grose et al., 1992). Wigand et al. (1995) thought the lake level decreased approximately 7 kyrs BP, during the Holocene minimum, and conjectured that the lake level rose again approximately 5 kyrs BP. With regard to geochemistry, the alkalinity of

Eagle Lake is inversely proportional to lake level flux, with alkalinity increasing during dry periods (Carmona-Catot et al., 2012). The hydrology of the ELW was undisturbed

28 until the construction of the Bly Tunnel (1923), which connected Eagle Lake to Willow

Creek between 1923 – 1936 (Harding, 1965; Moyle et al., 1991); however, in 1986 a valved concrete plug (installed in the Bly Tunnel by the BLM) allowed outflows until

2012 (Harding, 1965; Collum, 2012). All previous lake level fluctuations prior to the construction of the Bly Tunnel are assumed to have been due to natural variation.

Paleoecology

The ELW has been forested for more than 30 kyrs (Wigand et al., 1995).

Human occupation of northeastern California occurred beginning 11.45 kyrs ± 340 yrs

BP (McGuire, 2007). Juniper and sagebrush were dominant during the Pleistocene

(Wigand et al., 1995). Sagebrush, pine, and rabbitbrush have been dominant throughout the Holocene (Wigand et al., 1995). Near the end of the Pleistocene (~15.5 kyrs BP), a severe regional drought took place (Wigand et al., 1995). Using a graph by Wigand et al.

(1995), it was determined that charcoal abundance was low throughout the Pleistocene.

The decrease in shrub species at 10 kyrs BP may indicate the establishment of a pine forest surrounding Eagle Lake (Wigand et al., 1995). Using a graph by Wigand et al.

(1995), it was determined that charcoal abundance increased approximately 10 kyrs BP.

By 8 kyrs BP, the decreased production of local pollen, and vanishing of grasses and some riparian indicators, indicates the warming and drying (middle Holocene drought)

(Wigand et al., 1995). Over the previous 2 kyrs, periods of increased charcoal abundance have occurred compared to present (Wigand et al., 1995). Finally, the decrease in charcoal abundance at present may reflect recent fire suppression efforts (Wigand et al.,

1995).

CHAPTER III

METHODS

Sediment Core Extraction and Processing

The methods used here were similar to those used by Brunelle and Anderson

(2003), Walsh, Whitlock, et al. (2010), Hallett and Anderson (2010), and Minckley et al.

(2007); deviations were used where appropriate. Eight sediment cores (5cm ID x 1 m) were extracted from multiple off-shore locations. Typically, lake sediment core extraction uses a Livingstone sediment core sampler and a coring platform (e.g., Brunelle and

Anderson, 2003; Walsh, Whitlock, et al., 2010). In this study, substitutions were made for both a floating platform and Livingstone sediment core sampler, as they were unavailable. Sediment cores were extracted using a 4ft (2in ID) multi-staged AMS sludge and sediment sampler (AMS sampler), with a polyurethane tube inside a segmented stainless steel shaft, allowing for cores to be capped and stored within the polyurethane tubes (Figure 9). No coring platform was available, so only shallow (<2 m deep) near shore locations could be cored.

Sampling was done during the summer of 2012, when the water surface elevation of Eagle Lake was low (1553 m), which allowed coring areas that were inundated with an additional 1 – 10 m water in previous years (Figure 8). Greater water column depth reduces wave agitation of bed sediments (Nichols, 2009), and increased distance from shore was assumed to decrease anthropogenic disturbance; therefore,

29 30

Figure 9. AMS sludge and sediment sampler.

deeper coring locations, farther from shore, were preferred. A small inflatable raft made traversing the shoreline with heavy equipment easier. After extraction, the cores were stored in refrigeration (3° C) (e.g., Minckley et al., 2007). Prior to splitting the cores were analyzed for core magnetic susceptibility (1cm increments). The cores were split axially with a Geotek core splitter, photographed, and their lithostratigraphy was described (e.g.,

Walsh, Whitlock, et al., 2010). Prior to observation, cores were smoothed with an osmotic knife. After splitting, one half of the split core was stored for archival purposes.

After describing the basic lithology, core ELNEC1-1 was divided into 1 cm intervals (group A). Sediment group A (3 cm3 per 1 cm core interval) was collected, and

31 placed samples in whirl-paks. Sediment sample group B was (1 cm3 per 1 cm core interval) was subsampled from group A. The remaining 2 cm3 of sediment in group A was analyzed via XRF. Core splitting, photography, logging of basic lithostratigraphy, and logging of magnetic susceptibility was performed at the USGS Pacific Coastal

Marine Science Center (USGS PCMSC) in Menlo Park, CA. Analysis of sediments via

XRF was performed at California State University, Chico. Microscopic analysis of sediments was also performed. Archived core halves and unused whole cores can be found at the USGS PCMSC in Menlo Park, CA (USGS, Field Activity ID: L-14-12-CA).

Lithostratigraphy

After extraction and processing, one core was selected, and the lithostratigraphy was described (see Appendix A). ELNEC1-1 was selected because it is the longest core (107 cm), and because the overall matrix appeared to be primarily felsic, making it easier to identify charcoal (fewer black lithics overall). ELNEC1-1 was visually inspected, photographed, and the following criteria were logged: color, moisture content, grain size, and macrofossils; potential beach units were noted. A Geological

Society of America rock-color chart was used to describe sediment color. The GSA rock- color chart was based on the Munsell Color System. A grain size chart was used to distinguish sediment size classes. Core lithostratigraphy was enhanced using macro- charcoal, magnetic susceptibility, geochronology, and chemostratigraphy.

Geochronology

Three shells were extracted from different stratigraphic layers in core

ELNEC1-1, and were age-dated via radiochronology. Typically, stratigraphic levels

32 selected for age-depth model development in sediment cores have been end-caps and central locations (e.g., Brunelle and Anderson, 2003; Walsh, Whitlock, et al., 2010). It should be noted that legacy radiocarbon contamination is a potential hazard in any lab that may have been used for isotope tracing, is often a serious concern in bioscience labs, and must be eliminated to ensure reliable age-dates (Buchholz et al., 2000). Basic precautions were taken to minimize potential radiocarbon contamination. While sieving for macro-charcoal, all shells were collected, their stratigraphic position was noted, and they were stored in individual whirl-paks. To halt potential bacterial growth, a weak

H2O2 solution was added to each whirl-pak along with the shells. Samples for age-dating were selected at depths of 15 cm (shell), 43 cm (shell), and 107 cm (sediment) from top of core. The lithostratigraphy of ELNEC1-1 was correlated with Wigand et al. (1995) to provide an estimate of expected shell age-dates. Radiocarbon age-dating was performed at the Center for Accelerated Mass Spectrometry (CAMS), Lawrence Livermore National

Laboratory (LLNL). The age-dates were used to fit an age-depth model for ELNEC1-1

(e.g., Ali et al., 2009).

Radiocarbon Swipe Testing

Before samples were age-dated, all lab spaces were assessed for radiocarbon contamination using an excess radiocarbon testing kit (swipe kit). The swipe kit was provided by CAMS, and included vinyl gloves, 9 whatman grade GF/A glass microfiber filters (1.6 µm, Ø2.4 cm, wrapped in foil), 9 resealable glass vials, and directions. All labs used for sample preparation were analyzed for excess radiocarbon: (1) a USGS sedimentology lab in Menlo Park, CA, (2) a CSUC sedimentology lab in Chico, CA, and

(3) a personal sedimentology lab in Chico, CA. To avoid cross-contamination, a dirty

33 hands-clean hands protocol was implemented, such that gloves were changed after each swipe test, and nothing was touched before grabbing the filter paper. During the swipe tests, all surfaces were treated as if contaminated, and only permanent surfaces were tested for excess radiocarbon (e.g., door knobs, tables, sinks, etc.). To perform a swipe, the filter paper was picked up with gloved hands, saturated in isopropyl alcohol, swiped against the surface of interest several times, placed in a vial, and labeled. The first swipe was saturated in isopropyl alcohol and kept as a blank. If a surface was visibly dirty, it was cleaned with isopropyl alcohol prior to swiping. After swipe testing was completed, the filters were mailed to CAMS for analysis.

Radiocarbon Analysis

Once the swipe testing returned negative results, age-datable materials from

ELNEC1-1 were sent to CAMS, where they were pretreated and analyzed. Samples were cleaned using an acid base acid (ABA) treatment described in Zimmerman et al. (2010).

The carbonate shells were etched using a Dremel (slow setting), and approximately 10 mg of material was extracted. For bulk sediment, 10 mg of material was taken. The ABA method was performed by soaking the raw material in a bath of 1 N HCl for an hour

(90 °C), then a bath of 1 N NaOH for an hour (90 °C), then in a bath of 1 N HCl (90 °C), and finally rinsing with Milli-Q water. CO2 was released from the samples by placing aliquots (~2 mg) of treated material (plus CO and S) in evacuated quartz tubes and combusting at 900 °C. The CO2 was collected, cryogenically purified, transferred to separate graphitization reactors, and was reduced to with graphite-iron mixture

(elemental carbon + iron catalyst + excess hydrogen). The graphite-iron mixture was pressed into individual aluminum target holders and measured via accelerator mass

34 spectrometry (AMS) by CAMS. The AMS ionized the samples via cesium ion bombardment, eliminated 14N and other isobaric molecules, and determined the activity of 14C (Faure and Mensing, 2005). CAMS reported uncalibrated radiocarbon age-dates.

The age-dates were calibrated with CALIB (version 7.0.1, http://calib.qub.ac.uk/calib/) by Stuiver and Reimer (1993) (e.g., Rius et al., 2011) using the IntCal13 by Reimer et al.

(2013), as the calibration dataset.

Fire History

The fire history of Eagle Lake was characterized by stratigraphically quantifying macro-charcoal extracted from core ELNEC1-1 in 1 cm intervals, creating an age-depth model using shells from core ELNEC1-1, and using CharAnalysis to create a charcoal influx (CHAR, particles/cm2/yr) time-series, and by entering CHAR time-series into CharAnalysis. CharAnalysis (http://charanalysis.googlepages.com/) is designed to characterize fire history using CHAR data and geochronology information (see

CharAnalysis section); moreover, CharAnalysis has been used in multiple fire history studies (e.g., Walsh, Whitlock, et al., 2010).

Macro-charcoal Processing

The macro-charcoal concentration (particles/cm3) of core ELNEC1-1 was stratigraphically quantified by analyzing sediment group B (1 cm3 per 1 cm core interval)

(e.g., Walsh, Whitlock, et al., 2010), which were subsampled from sediment group A (3 cm3 per 1 cm core interval) (e.g., Walsh, Whitlock, et al., 2010). Samples from sediment group B were soaked for 24 hours in a .08 M (50g/L) solution of sodium hexametaphosphate, washed through a 125 μm sieve, placed in a weak bleach solution

(5% by weight) for 1 hour, re-washed through a 125 μm sieve, transferred to a gridded petri dish, and observed under magnification (e.g., Walsh, Whitlock, et al., 2010); after observation, they were washed into filter paper, and stored in whirl-paks. Charcoal samples were restricted to >125 μm as that size class was thought primarily be from local fires (<1 km) and, to a lesser degree, from regional fires (<20 km) (Higuera et al., 2007;

Peters and Higuera, 2007). Once in gridded petri dishes, macro-charcoal was identified and counted under magnification (e.g., Brunelle and Anderson, 2003). A Barska

AY11234 trinocular microscope was used, and particles were examined at 10 – 45x.

Particles were identified as macro-charcoal if they were black, opaque, angular (Whitlock and Larsen, 2001), and were easily discernable at 20x magnification. Only the 105 cm of core ELNEC1-1 were analyzed; the remaining 2 cm were saved for radiocarbon analysis.

CharAnalysis

CharAnalysis (http://charanalysis.googlepages.com/) uses charcoal concentration (particles/cm3) and geochronology data to determine statistically significant local fire events (Higuera et al., 2009; Walsh, Whitlock, et al., 2010). CharAnalysis creates a charcoal flux (CHAR, particles/cm2/yr) time-series to model background CHAR

(Cback) and interpolated CHAR (Cint), which are then used to obtain peak CHAR (Cpeak)

(Ali et al., 2009; Higuera et al., 2009; Walsh, Whitlock, et al., 2010). To obtain Cint, the

CHAR data was interpolated using a time-interval matching the average temporal resolution (e.g., Higuera et al., 2009; Ali et al., 2009). Cback (background) represents baseline charcoal accumulation over time (Brunelle and Anderson, 2003). Cpeak (peak) represents temporal locations where Cint exceeds Cback, and each peak marks one or more fire events (Brunelle and Anderson, 2003; Higuera et al., 2009). Cback was estimated by

36 applying a robust locally weighted scatter plot smoothing (lowess) regression to the

CHAR data, with a smoothing window of 1000 yr. Peaks were determined using the

Equation 1 (Higuera et al., 2009).

Cpeak = Cint – Cback (1)

The peaks consist of Cnoise and Cfire (Ali et al., 2009), with Cnoise (noise) representing variability due to sediment mixing, sampling, and other noise sources, and

Cfire (fire events) representing charcoal influx from local fire events (Ali et al., 2009). In order to distinguish between fire and non-fire events, CharAnalysis separated Cfire from

Cnoise using a Gaussian model and a locally defined threshold (Walsh, Whitlock, et al.,

th th 2010). CharAnalysis determined the local threshold by considered the 95 , 99 , and

th 99.9 percentile of Cnoise (Higuera et al., 2009); if the peak was significantly different than the noise, it was kept. Additionally, CharAnalysis screened peaks, and they were eliminated if their charcoal count was not significantly different than the minimum charcoal count of the previous 75 yrs (Walsh, Pearl, et al., 2010). Fire-episode frequency

(fire frequency) and fire return interval (FRI) were smoothed using a lowess function

(Higuera, 2009) over 2 kyr intervals; with fire-episode frequency being the sum of fire episodes per 2 kyr interval, and mean FRI the average number of years between fire episodes (Walsh, Pearl, et al., 2010). Fire-episode magnitude (particles/cm2) was also defined as the total charcoal influx per peak (Walsh, Pearl, et al., 2010).

Paleohydrology

MS and XRF data was obtained to interpret the paleohydrology of Eagle Lake

(e.g., Breckenridge et al., 2010; Ortega et al., 2010; Cuven et al., 2011). MS and XRF

37 data was used to analyze sediment transport pathway shifts (allogenic vs. autogenic).

High MS readings indicate elevated terrigenous deposition, while low MS readings indicate increased biogenic or eolian deposition. XRF analysis provides bulk geochemical data, and was analyzed using zonation analysis to identify geochemical shifts. Bulk Kylander et al. (2012) bulk geochemistry shifts occur when the sediment source changes, providing evidence for transport pathway shifts. Volumetric MS data (SI units) was analyzed and collected in 1cm intervals by running core ELNEC1-1 through a

Geotek MSCL equipped with an 8 cm Bartington loop sensor (MS2C) (e.g.,

Breckenridge et al., 2010). Bulk geochemistry (XRF) was obtained from sediment group

A using a Bruker S2 Ranger, at CSU, Chico. The paleohydrologic record of Eagle Lake was interpreted from an MS time-series and a zonation analysis of a principle component

XRF time-series.

Zonation Analysis

Zonation analysis was performed by: (1) abstracting the bulk geochemical data (Mg, Al, Si, Ti, and Fe) of core ELNEC1-1 using the principle component analysis

(PCA) function of Minitab, (2) running the PCA scores through Equation 2, and using the output of Equation 2 to identify zonal breaks (e.g., Davis, 2002) in core ELNEC1-1. If the zone interface was adjacent to a facies change identified in the lith-log (see Appendix

A), the zone interface was adjusted to match the facies change. A short window was used for Equation 2 because it is better to use for the detection of smaller zones (Davis, 2002).

More robust equations exist for zonation analysis (e.g., Davis, 2002), but Equation 2

(Davis, 2002) was chosen for its simplicity.

(2)

Principle Component Analysis

Principle component analysis is a method of describing a multivariate dataset via the eigenvectors of a correlation matrix, and can be used to reduce multivariate datapoints to a single datapoint (Davis, 2002). Eigenvectors are coordinates that define the principal semiaxes of a five-dimensional hyperellipsoid, and are known as principle components (Davis, 2002). In PCA, eigenvectors are used to determine which principle component describes the maximum variance of the multivariate dataset (Davis, 2002).

Each principle component describes a set of potentially correlated multivariate dataset as a linear dataset in principle component scores (Davis, 2002). Because none of the original variables can account for total variation, at least one principle component accounts for more variation than any single variable (Davis, 2002); therefore, it is possible to reduce the complexity of a multivariate dataset and still maintain a high level of variability. At least one principle component must include less variation than the multivariate dataset

(Davis, 2002), so care must be taken when selecting the appropriate principle component.

XRF analysis provided multivariate geochemical data (elements); therefore, PCA was used to reduce bulk geochemical data to a single time-series with Minitab.

CHAPTER IV

RESULTS

Geochronology

Initial swipe testing of all lab spaces (personal, USGS, and CSUC) used for sample processing returned results negative for radiocarbon contamination. Age-datable materials were shipped to CAMS on 01/10/2014 for radiometric dating. Radiocarbon age calibration with CALIB used the 2-sigma calibration method (e.g., Whitlock et al., 2008), since it returned high relative probabilities (Table 1). Using the calibrated age-dates

(calendar yrs BP), an age-depth model was developed using linear regression (Figure 10).

It was assumed the core-top was at present. The lower values of the age-date ranges provided by CALIB were used to develop the age-depth model. Linear regression provided the best model fit (R2 = 0.9874). Only one sample returned a date with a relative probability of less than 1. The age-depth model fit the older datapoints better than the earliest data point (Figure 10), and the earliest datapoint had the lowest confidence (Table

1). Additional quality assurance analysis of the age-date from bulk sediment (107) suggested that the age-date may have been influenced by older charcoal or younger carbonate present within the sample; however, the original age-date are similar to age- dates reported by Wigand et al. (1995), and were preferred.

39 40

TABLE 1. RADIOCARBON AGE-DATES AND CALIBRATED AGE-DATES OF CORE ELNEC1-1 14C Standard Calibrated age-date Relative Depth (cm) 14C age dev. range probability 13 665 ±30 630 - 674 0.527 43 7435 ±30 8187 - 8336 1 107 16730 ±130 19850 - 20532 1 Note: Radiocarbon age-date in yrs BP, and calibrated age-date ranges in calendar yrs BP. Radiocarbon age-dates calibrated with CALIB (version 7.0.1) by Stuiver and Reimer (1993). IntCal13 set by Reimer et al. (2013) was used as the calibration dataset.

Figure 10. Linear age-depth model for core ELNEC1-1.

Lithostratigraphy

Figure 11 shows the lithostratigraphy for ELNEC1-1. Core photographs for

ELNEC1-1 are in Appendix A. ELNEC1-1 was a dark yellowish brown (10y 4/2) fluid mud between 0 and 5 cm, was a light olive grey (5y 5/2) muddy sand from 5 – 30 cm, was a dusky yellowish brown (10y 2/2) muddy sand, intermixed with shells, from 30 – 43

Figure 11. Lithostratigraphic log of core ELNEC1-1. Note: ELNEC1-1 collected from Eagle Lake on 7/23/2012. Not to scale. Dash dot line indicates age-date locations.

cm, was dark yellowish brown (10y 4/2) mud from 43 – 89 cm, and was a dusky yellowish brown (10y 2/2) muddy sand, intermixed with shell fragments, between 89 and

107 cm; some gradation was apparent between units. The two dusky yellowish brown

(10y 2/2) muddy sand units are similar to those described by Wigand et al. (1995). The upper beach unit ranges from approximately 8 kyrs BP at its base, to 5 kyrs BP at its top

(Figure 11). The lower beach unit ranges from approximately 20 kyrs BP to 16 kyrs BP

(Figure 11). Zonation analysis identified four geochemical zones using GZ-1, GZ-2, GZ-

3, and GZ-4 (Figure 12). The sedimentation rates for GZ-1 through GZ-4 (respectively) are 5.42 cm/kyr, 3.58 cm/kyr, 1.74 cm/kyr, and 0.89 cm/kyr.

Figure 12. Zonation analysis of principle component 1 (PC1) of the x-ray fluorescence (XRF) dataset for core ELNEC1-1. Note: Zonation analysis performed by computing the generalized distance (D2) of principle component one. Geochemical dataset used to determine PCA scores found in Appendix B. Green lines indicate geochemical zone divisions.

MS has fluctuated over time (Figure 13; Table 2). MS values between 0 and

25 are low, between 25 and 50 are moderate, and greater than 50 are high. GZ-1 had moderate average MS and variability (range), with only GZ-2 having a greater average

MS. GZ-2 had high average MS and variability. GZ-3 had low average MS and

Figure 13. Charcoal concentration (CC) and magnetic susceptibility (MS) of core ELNEC1-1. Note: Grey bars indicate beach units based on lithology. Age-dates from linear age-depth model found in the Geochronology section. Green lines indicate geochemical zone divisions.

variability. GZ-4 had low to moderate average MS and variability, with an average MS greater than GZ-3 but less than GZ-1. Overall, starting at base of core, MS was low, increased to a regional high in GZ-4 at 99 cm, decreased to a minimum in GZ-3 at 71 cm, increased to a regional high in GZ-2 at 37 cm, and decreased to a regional low in GZ-1, at top of core. Charcoal concentration (CC) also changed over time (Figure 13; Table 3).

TABLE 2. MAGNETIC SUSCEPTIBILITY (MS) SUMMARY TABLE OF CORE ELNEC1-1 Zone Depth Range (cm) High MS Low MS MS Range Average MS GZ-1 0 - 21 45.29 6.92 38.37 25.76 GZ-2 22 - 60 115.74 4.40 111.33 47.27 GZ-3 61 - 95 15.10 4.40 10.69 7.43 GZ-4 96 - 107 28.30 8.81 19.50 16.91 Note: MS value are unitless but are in SI parameters.

TABLE 3. MACRO-CHARCOAL CONCENTRATION (CC) SUMMARY TABLE OF CORE ELNEC1-1 Zone Depth Range (cm) High CC Low CC CC Range Average CC GZ-1 0 – 21 4183.00 924.00 3259.00 2327.33 GZ-2 22 – 60 2784.00 110.00 2674.00 1400.95 GZ-3 61 – 95 662.00 67.00 595.00 257.86 GZ-4 96 – 107 748.00 347.00 401.00 512.00 Note: CC units in particles/cm3

CC values between 0 and 1000 are low, between 1000 and 2000 moderate, and above

2000 high. GZ-1 has high average CC and variability. GZ-2 had moderate average CC and high variability, with values greater than GZ-3 and GZ-4 but less than GZ-1. GZ-3 had the low average CC and low variability, with only GZ-4 exhibiting lower variability.

GZ-4 had low average CC and the lowest variability.

Fire History

Charcoal

CHAR values (particles/cm2/yr) less than 1000 indicate low CHAR, values between 1000 and 2000 indicate moderate CHAR, and values greater than 2000 indicate high CHAR; borderline values were considered on a case by case basis. From 18 kyrs BP to 8.5 kyrs BP CHAR was low (Figure 14). From 8.5 to 6.8 kyrs BP CHAR was moderate. From 6.8 to 1.3 kyrs BP CHAR was high (Figure 14). From 1.3 kyrs BP,

CHAR continued to decrease to a low at present (Figure 14). A maximum occurred at 2.8 kyrs BP, and a minimum occurred at 14.2 kyrs BP (Figure 14). Figure 14 displays the

CHAR time-series for Eagle Lake. Table 4 summarizes high, low, range (variability) and average CHAR for Eagle Lake in 1 kyr increments. Charcoal influx to Eagle Lake has shifted over the past 18 kyrs (Figure 14; Table 4).

Figure 14. Charcoal influx (CHAR) of core ELNEC1-1. Note: Time-scale is based on CharAnalysis interpolation of age-depth model developed in Geochronology section.

TABLE 4. CHARCOAL INFLUX (CHAR) SUMMARY TABLE OF CORE ELNEC1-1 Time period High CHAR Low CHAR CHAR Range Average CHAR 0 - 1 kyrs BP 1669.18 924.00 745.18 1320.52 1 - 2 kyrs BP 2999.15 1312.47 1686.69 2164.23 2 - 3 kyrs BP 4103.72 2070.83 2032.90 3020.00 3 - 4 kyrs BP 3343.86 2563.22 780.64 2909.84 4 - 5 kyrs BP 2360.81 1696.34 664.47 1991.21 5 - 6 kyrs BP 2397.93 1771.75 626.19 2128.53 6 - 7 kyrs BP 2714.41 1981.39 733.01 2223.74 7 - 8 kyrs BP 1589.25 1161.85 427.40 1367.71 8 - 9 kyrs BP 1119.20 703.43 415.77 955.46 9 - 10 kyrs BP 865.46 418.26 447.20 616.17 10 - 11 kyrs BP 441.45 122.84 318.60 324.15 11 - 12 kyrs BP 213.18 83.64 129.54 141.92 12 - 13 kyrs BP 269.98 111.57 158.41 154.26 13 - 14 kyrs BP 243.94 114.67 129.27 181.28 14 - 15 kyrs BP 414.20 94.33 319.86 274.07 15 - 16 kyrs BP 327.98 173.79 154.19 213.98 16 - 17 kyrs BP 585.13 175.40 409.72 393.33 17 - 18 kyrs BP 652.45 375.03 277.42 495.28 Note: Units in particles/cm2/yr. Time-scale is based on CharAnalysis interpolation of age-depth model developed in Geochronology section.

Fire Frequency

FF values (fire-episodes/2 kyrs) less than 1 indicate low FF, values between 1 and 2 indicate moderate FF, and values greater than 2 indicate high FF; borderline values were evaluated on a case by case basis. From 18 to 16.2 kyrs BP FF decreased from moderate levels to a near-minimum (Figure 15). From 16.2 to 10.7 kyrs BP FF increased

Figure 15. Fire frequency (FF) and fire-episodes (peaks) of core ELNEC1-1. Note: Fire-episodes (Cpeak) represents temporal locations where Cint exceeds Cback, marking one or more fire events (Brunelle and Anderson, 2003; Higuera et al., 2009). Time-scale is based on CharAnalysis interpolation of age-depth model developed in Geochronology section.

(ignoring intermediary flux) to a maximum (Figure 15). From 10.7 to 2.6 kyrs BP, FF decreased to a minimum (Figure 15). From 2.6 kyrs BP, FF increased (ignoring intermediary flux) to moderate FF levels at present (Figure 15). Figure 15 displays the fire-episode frequency (FF) for Eagle Lake. Table 5 summarizes FF using high, low, range (variability) and average FF (AFF) for Eagle Lake in 1 kyrs increments. Major fire frequency shifts have occurred surrounding Eagle Lake over the past 18 kyrs (Figure 15;

Table 5).

TABLE 5. FIRE FREQUENCY (FF) SUMMARY TABLE OF CORE ELNEC1-1 Time period High FF Low FF FF Range Average FF 0 - 1 kyrs BP 1.93 1.75 0.19 1.85 1 - 2 kyrs BP 1.81 1.00 0.81 1.41 2 - 3 kyrs BP 0.78 0.35 0.43 0.49 3 - 4 kyrs BP 1.41 0.59 0.83 1.00 4 - 5 kyrs BP 1.90 1.59 0.31 1.80 5 - 6 kyrs BP 1.83 1.57 0.26 1.73 6 - 7 kyrs BP 1.59 1.35 0.24 1.43 7 - 8 kyrs BP 2.41 1.90 0.52 2.13 8 - 9 kyrs BP 3.00 2.59 0.41 2.84 9 - 10 kyrs BP 3.42 3.00 0.42 3.16 10 - 11 kyrs BP 3.90 3.58 0.31 3.78 11 - 12 kyrs BP 3.76 2.29 1.46 3.09 12 - 13 kyrs BP 2.06 1.35 0.71 1.66 13 - 14 kyrs BP 2.38 1.33 1.05 1.77 14 - 15 kyrs BP 2.49 2.08 0.41 2.34 15 - 16 kyrs BP 1.92 0.85 1.07 1.43 16 - 17 kyrs BP 0.90 0.51 0.39 0.65 17 - 18 kyrs BP 1.84 1.00 0.84 1.42 Note: Units in fire-episodes/kyr. Time-scale is based on CharAnalysis interpolation of age-depth model developed in Geochronology section.

Fire Magnitude

Average FEM (AFEM) values (particles/cm2) less than 100 indicate low

AFEM, values between 100 and 300 indicate moderate AFEM, and values greater than

300 indicate high AFEM; borderline values were evaluated on a case by case basis. From

18 to 8 kyrs BP AFEM remained low (Figure 16). From 8 to 6 kyrs BP, AFEM remained moderate. From 6 to 4 kyrs BP, AFEM was at a maximum. From 4 to 2 kyrs BP, AFEM was at a minimum. From 2 kyrs to present, AFEM remained low. Figure 16 displays fire- episode magnitude (FEM) for Eagle Lake, from the present to 8 kyrs BP. Table 6 summarizes FEM by listing average FEM in 2 kyrs increments, from present to 18 kyrs

BP. Fire-episode magnitude has fluctuated over the past 18 kyrs (Figure 16).

Figure 16. Fire-episode magnitude (FEM) of core ELNEC1-1. Note: Respective peaks are identified in Figure 15. The fire-episode magnitude values are averaged over 1 kyr intervals to obtain average FEM. Time-scale is based on CharAnalysis interpolation of age-depth model developed in Geochronology section.

TABLE 6. FIRE-EPISODE MAGNITUDE (FEM) SUMMARY TABLE OF ELNEC1-1 Time period Average FEM 0 - 2 kyrs BP 114.56 2 - 4 kyrs BP 0.00 4 - 6 kyrs BP 463.67 6 - 8 kyrs BP 259.17 8 - 10 kyrs BP 60.00 10 - 12 kyrs BP 32.21 12 - 14 kyrs BP 14.45 14 - 16 kyrs BP 27.62 16 - 18 kyrs BP 29.90 Note: Units in particles/cm2. Time-scale is based on CharAnalysis interpolation of age-depth model developed in Geochronology section.

CHAPTER V

DISCUSSION

Paleohydrology of Eagle Lake

MS data and lithology were best suited to describe broad hydrologic shifts in watershed dynamics. MS data showed two periods of elevated hydrologic activity at

Eagle Lake, which were followed by low hydrologic activity adjacent to mud-beach unit interfaces. The first period of elevated hydrologic activity occurred 16 to 20 kyrs BP, and the second from 4 to 8 kyrs BP. At Eagle Lake, periods of high MS coincide with elevated CHAR, increased fire frequency, and lake level shifts (indicated by beach unit formation). Likewise, at Eagle Lake, periods of low MS coincide with low CHAR levels, low fire frequency, and lake level shifts (indicated by mud unit formation). Major shifts in MS also occurred adjacent to geochemical zone shifts, and changes in sediment size. A lag-time between periods of high MS and high CHAR events is apparent. Elevated hydrologic activity allows fuel build up, likely contributing to peak CHAR events. Once environmental moisture decreases below a particular threshold, biomass cannot regenerate quickly enough to maintain elevated CHAR. An attempt to cross-reference palynology data from Wigand et al. (1995) was made, but the data was no longer available. Future studies should incorporate pollen analysis into their study to analyze how biomass shifted with environmental conditions.

49 50

Where beach units occurred, and MS was elevated, the sediment source shifted because hydrologic conditions allowed for the transport of larger terrestrial materials surrounding Eagle Lake. Where mud units occurred, and MS was low, the sediment source shifted due to hydrologic conditions, allowing for significantly more authigenic deposition and transport of sediments via wind and dry ravel than hydrologic transport. Hydrologic to eolian transport shifts indicate time periods where major hydroclimate shifts occurred, and where the water level of Eagle Lake would have decreased; likewise, shifts from eolian to hydrologic transport indicate lake level gains or stabilization. Given the data available, precise lake level fluctuations were unable to be determined.

Based on MS and lithostratigraphy, hydrologic activity was moderate from 20 kyrs to 16 kyrs BP. From 16 kyrs BP, hydrologic activity transitioned into a lower phase, and reached a minimum at 13.5 kyrs BP. From 13.5 kyrs BP, hydrologic activity remained low until 8.5 kyrs BP. The 15 – 8.5 kyrs BP period is characteristic of a severe drought. The period from 20 – 10 kyrs BP was a post-glacial transitional period (Gester,

1962). From 8.5 kyrs BP, hydrologic activity increased to moderate to high levels by 7.5 kyrs BP, and likely marks the end of the glacial period. From 7.5 kyrs BP, hydrologic activity continued to increase, and reached a maximum at 6.8 kyrs BP. From 6.8 kyrs BP, hydrologic activity remained high until 4 kyrs BP. From 4 kyrs BP, hydrologic activity remained moderate until present, but decreased overall. The period from 4 – 2 kyrs BP was a neoglacial period (Wigand et al., 1995). The lake level would have been at its peak around 6.8 kyrs BP, and at its lowest around 13.5 kyrs BP (Figure 14). Future studies should analyze varved sediments to develop a higher resolution paleohydrology dataset.

Fire History of Eagle Lake

The fire history of the Eagle Lake Watershed has been dynamic over the past

18 kyrs, but can generally be divided into three periods based on CHAR. The first period occurred from approximately 18 to 10 kyrs BP, the second period from approximately 10 to 2.8 kyrs BP, and the third period from approximately 2.8 kyrs BP to present.

Anthropogenic fire influences have been present in northeastern California for the past

11.45 kyrs ± 340 yrs. Temperature fluctuation may account for some fire regime fluctuation, but was not evaluated as part of this study.

First Period

Throughout the first period, CHAR production and fire-magnitude were low

(CHAR minimum at 14.2 kyrs BP), while fire frequency reached a maximum (10.6 kyrs

BP) during the period. Hydrologic activity during the first period was low, limiting fuel availability, and supporting small, frequent fires. This period can be characterized as a major drought period due to low hydrologic activity, low charcoal production, and elevated fire frequency due to the dry climate. Indigenous occupation of northeastern

California occurred 11.45 kyrs ± 340 yrs BP (McGuire, 2007), and fire has likely been used for hunting since (Anderson, 2006). After human arrival, fire frequency declined from maximum, CHAR increased without evidence of excess precipitation, and fire magnitude remained low, marking a transitional period most mixed with human influences and warming at end of the glacial period. It is likely indigenous populations were low enough that the Eagle Lake watershed was infrequently visited, and that the fire regime of the area was dominated primarily by climatic factors.

Second Period

During the second period, CHAR production increased throughout the period to a maximum (2.8 kyrs BP), fire frequency decreased to a minimum (2.8 kyrs BP), and fire magnitude increased to a maximum (5.2 kyrs BP) and was followed by a minimum within approximately 2 kyrs. Hydrologic activity during the second period was initially low, and fuel availability was limited, supporting small frequent fire. After approximately

8 kyrs BP, hydrologic activity increased to moderate to high levels, and fuel availability increased significantly, supporting larger moderately frequent fires. After approximately

6 kyrs BP, hydrologic activity remained high to moderate, and CHAR production remained high until 4.2 kyrs BP. After 4.2 kyrs BP, fire magnitude and frequency declined to low levels, while CHAR production remained high, suggesting that by period end, more frequent and intense fires had fragmented the vegetative community and charcoal typically entrained on the landscape was being released at greater rates. The period from 4 – 2 kyrs BP is also classified as a neoglacial period, and may have decreased vegetative regrowth as well. Indigenous populations were also present for the entirety of the second period, and Eagle Lake was frequented by hunting parties

(Simmons et al., 1997), which would likely have set fires in the area (Anderson, 2006), potentially moderating the fire frequency. An indigenous fire regime may account for the fire frequency stability prior to the neoglacial low.

Third Period

Throughout the third period, CHAR decreased high to low levels, fire frequency increased from low to moderate levels, and fire magnitude increased but remained low. Hydrologic activity during the third period was moderate and decreased

53 from 2.8 kyrs BP to a low at present, allowing for low to moderately frequent fires with moderate magnitude through to period end. Indigenous populations likely prevented

CHAR from returning to high levels given that elevated environmental moisture persisted and fire frequency remained relatively steady. The third period resembles a hybrid of the first and second periods. It is unknown to what extent humans influenced the fire history of Eagle Lake during this period. Based on the fire deficit described in other studies due to fire suppression, it is likely a fire deficit would be apparent if the fire history of Eagle

Lake were evaluated over the past few hundred years using dendrochronology or written records (if available); however, the data obtained is at too coarse a resolution to evaluate such a phenomenon. If hydrologic activity continues to decline, the future fire history of

Eagle Lake may look similar to Pleistocene fire activity, with smaller more frequent fires.

Regionality

Over the last 10 kyrs, climate change has been the primary cause of alterations to vegetative land cover in northeastern California (West et al., 2007), and is apparent in the fire history of Eagle Lake, but indigenous populations undoubtedly played a role in shaping the regional fire history too. No permanent indigenous settlement is known to have existed at Eagle Lake; but Clovis points have been found at Eagle Lake (McGuire,

2007) and hunting parties are known to have frequented the area. Plant responses to climatic change ranges from migration, adaptation, extinction, extirpation, and density shifts (West et al., 2007). Compared to other lakes along the edge of the Lahontan hydrologic region (e.g., Lily Lake, Patterson Lake, and Lily Pond), the fire history of the

Eagle Lake Watershed shows some similarities, including fire frequencies ranging

54 between 1 and 8 fire-episodes/kyr (e.g., Minckley et al., 2007), and elevated charcoal occurring at similar time periods as those found in Wigand et al. (1995). The fire history similarities are most likely due to the regional climate, elevation, and topography, with the vegetative community accounting for site specific variations (e.g., CHAR, sediment flux, humans, etc.).

Beaty and Taylor (2009) found that the climate in northern California and southwestern Oregon correlated well in the stratigraphic record, indicating a broad regional climate. The stratigraphic record at Eagle Lake also indicated climate shifts like those described in Beaty and Taylor (2009). Based on Beaty and Taylor (2009), and the hydrology of Eagle Lake, the climate of Eagle Lake prior to 15 kyrs BP was cold and moist, from 15 to 8.5 kyrs BP it was warm and dry, from 8.5 kyrs to 4 kyrs BP it was warm and wet, and from 4 kyrs BP to present it was cool and wet (Beaty and Taylor,

2009). The data uncovered, provides additional support for a severe regional drought during the end of the Pleistocene. In many cases, the response of vegetation to climate change in mountainous areas has been through contraction and expansion along elevational gradients (West et al., 2007), and analysis of pollen data may help enhance temperature shift data (e.g., Beaty and Taylor, 2009). A high-resolution companion pollen dataset does not yet exist for Eagle Lake.

CHAPTER VI

CONCLUSION

This project used a high-resolution sedimentary technique to characterize the fire history of Eagle Lake, California from present to 18 kyrs BP; in addition, the paleohydrology of Eagle Lake was evaluated. Wigand et al. (1995) provided the inspiration for the characterization of fire history at Eagle Lake using a high resolution sedimentary technique. During and after fire events, fire-induced compounds, as well as radionuclides, are transported via eolian and hydrologic transport, and preserved in lake sediments. This study evaluated macro-charcoal, MS data, radiochronology, XRF data, and lithology from a lake sediment core. To match other studies, this study stayed as close to the landscape and millennia scale as possible. It was discovered that the fire history of Eagle Lake varied over the past 18 kyrs and could be divided into three periods.

The three distinguished periods where 18 to 10 kyrs BP, 10 to 2.8 kyrs BP, and 2.8 kyrs BP to present. Hydrologic activity during the first period (18 to 10 kyrs BP) was low, limiting fuel availability, and supporting small, frequent fires. Hydrologic activity during the second period (10 to 2.8 kyrs BP) was initially low due to limited fuel availability, supporting small frequent fire. After approximately 8 kyrs BP, hydrologic activity increased to moderate/high levels, and fuel availability increased, supporting larger moderately frequent fires until 4.2 kyrs BP. After 4.2 kyrs BP, fire frequency and

55 56 magnitude were temporarily low until 1.5 kyrs BP, when moderately frequent and intense fires returned. Hydrologic activity during the third period (2.8 kyrs BP to present) was moderate and decreased to present, allowing for low to moderately frequent fires with moderate magnitude through to period end. Indigenous populations clearly influenced the fire history of Eagle Lake, moderated fire frequency over the Holocene, and set fires for forest understory management, wildlife management, manufacturing, and agricultural purposes. Despite evidence of indigenous occupation, it is difficult to distinguish anthropogenic influences and post-glacial and neoglacial transitions.

Based on the data presented and similarities to other sites, the climate of Eagle

Lake prior to 15 kyrs BP was cold and moist, from 15 to 8.5 kyrs BP it was warm and dry, from 8.5 kyrs to 4 kyrs BP it was warm and wet, and from 4 kyrs BP to present it was cool and wet. The fire history similarities between other sites and Eagle Lake are most likely due to the regional climate, elevation, and topography, with site specific variations likely due to vegetative community differences. Over the past 18 kyrs, climate change appears to have been the primary driver of fire regime change in northeastern

California. Future research should examine how vegetative composition affected the fire history of Eagle Lake. Future macro-charcoal studies should incorporate palynology into the study design to better understand biomass flux, hydrologic shifts, and erosion rates.

Future studies should also attempt to enhance the fire history of Eagle Lake with dendrochronology and delve deeper into the role of indigenous peoples in fire history

REFERENCES CITED

REFERENCES

Ali, A.A., Higuera, P.E., Bergeron, Y., and Carcaillet, C., 2009, Comparing fire-history interpretations based on area, number and estimated volume of macroscopic charcoal in lake sediments: Quaternary Research, v. 72, p. 462–468.

Amiro, B.D., Sheppard, S.C., Johnston, F.L., Evenden, W.G., and Harris, D.R., 1996, Burning radionuclide question: What happens to iodine, cesium and chlorine in biomass fires?: The Science of the Total Environment, p. 93–103.

Anderson, M.K., 2006, The use of fire by Native Americans in California, in Sugihara, N.G., van Wagtendonk, J.W., Fites-Kaufman, J., Shaffer, K. E., and Thode, A. E., eds., Fire in California’s Ecosystems: Los Angeles, California, University of California Press, p. 417–430.

Artiola, J.F., Pepper, I.L., and Brusseau, M., 2004, Environmental monitoring and characterization: San Diego, CA, Elsevier Academic Press, 410 p.

Ascough, P.L., Bird, M.I., Francis, S.M., Thornton, B., Midwood, A.J., Scott, A.C., and Apperley, D., 2011, Variability in oxidative degradation of charcoal: Influence of production conditions and environmental exposure: Geochimica et Cosmochimica Acta, v. 75, p. 2361–2378, doi:10.1016/j.gca.2011.02.002.

Beaty, R.M., and Taylor, A.H., 2009, A 14000 year sedimentary charcoal record of fire from the northern Sierra Nevada, Lake Tahoe basin, California, USA: The Holocene, v. 19, p. 347–358.

Bird, R., Bird, D., Codding, B., Parker, C., and Jones, J., 2008, The “‘fire stick farming’” hypothesis: Australian Aboriginal foraging strategies, biodiversity, and anthropogenic fire mosaics: PNAS, v. 105, p. 14796–14801.

Blake, W.H., Wallbrink, P.J., Doerr, S.H., Shakesby, R.A., and Humphreys, G.S., 2006, Magnetic enhancement in wildfire-affected soil and its potential for sediment- source ascription: Earth Surface Processes and Landforms, v. 31, p. 249–264, doi:10.1002/esp.1247.

Blake, W.H., Wallbrink, P.J., Wilkinson, S.N., Humphreys, G.S., Doerr, S.H., Shakesby, R.A., and Tomkins, K.M., 2009, Deriving hillslope sediment budgets in wildfire-affected forests using fallout radionuclide tracers: Geomorphology, v. 104, p. 105–116, doi:10.1016/j.geomorph.2008.08.004.

58 59

Bowman, D.M.J.S., Balch, J.K., Artaxo, P., Bond, W.J., Carlson, J.M., Cochrane, M.A., D’Antonio, C.M., DeFries, R.S., Doyle, J.C., Harrison, S.P., Johnston, F.H., Keeley, J.E., Krawchuk, M.A., Kull, C.A., et al., 2009, Fire in the earth system: Science, v. 324, p. 481–484, doi:10.1126/science.1163886.

Breckenridge, A., Lowell, T.V., Fisher, T.G., and Yu, S., 2010, A late Lake Minong transgression in the Lake Superior basin as documented by sediments from Fenton Lake, Ontario: Journal of Paleolimnology, v. 47, p. 313–326, doi:10.1007/s10933-010-9447-z.

Briles, C.E., Whitlock, C., Bartlein, P.J., and Higuera, P.E., 2008, Regional and local controls on postglacial vegetation and fire in the Siskiyou Mountains, northern California, USA: Palaeogeography, Palaeoclimatology, Palaeoecology, v. 265, p. 159–169, doi:10.1016/j.palaeo.2008.05.007.

Brunelle, A., and Anderson, R.S., 2003, Sedimentary charcoal as an indicator of late- Holocene drought in the Sierra Nevada, California, and its relevance to the future: The Holocene, v. 13, p. 21–28.

Buchholz, B.A., Freeman, S.P., Haack, K.W., and Vogel, J.S., 2000, Tips and traps in the 14C bio-AMS preparation laboratory: Nuclear instruments and methods in physics research section b: Beam Interactions with Materials and Atoms, v. 172, p. 404–408.

Carmona-Catot, G., Moyle, P.B., and Simmons, R.E., 2012, Long-term captive breeding does not necessarily prevent reestablishment: Lessons learned from Eagle Lake rainbow trout: Reviews in Fish Biology and Fisheries, v. 22, p. 325–342, doi:10.1007/s11160-011-9230-x.

Collins, A.L., and Walling, D.E., 2004, Documenting catchment suspended sediment sources: Problems, approaches and prospects: Progress in Physical Geography, v. 28, p. 159–196, doi:10.1191/0309133304pp409ra.

Collum, K.R., 2012, Notice of field manager’s decision: Bly tunnel bypass flow valve closure: Bureau of Land Management decision record DOI-BLM-CAN050- 2012-14-CX: Susanville, California, BLM.

Conedera, M., Tinner, W., Neff, C., Meurer, M., Dickens, A.F., and Krebs, P., 2009, Reconstructing past fire regimes: Methods, applications, and relevance to fire management and conservation: Quaternary Science Reviews, v. 28, p. 555– 576.

Cuven, S., Francus, P., and Lamoureux, S., 2011, Mid to late Holocene hydroclimatic and geochemical records from the varved sediments of East Lake, Cape Bounty, Canadian high arctic: Quaternary Science Reviews, v. 30, no. 19-20, p. 2651- 2665.

Davis, J., 2002, Statistics and data analysis in geology: New York, NY, John Wiley & Sons, 638 p.

DeBano, L.F., 2000, The role of fire and soil heating on water repellency in wildland environments: A review: Journal of Hydrology, v. 231, p. 195–206.

Doerr, S.H., Shakesby, R.A., Blake, W.H., Chafer, C.J., Humphreys, G.S., and Wallbrink, P.J., 2006, Effects of differing wildfire severities on soil wettability and implications for hydrological response: Journal of Hydrology, v. 319, p. 295–311.

DWR, 2004, Eagle Lake area groundwater basin, California’s groundwater, Bulletin 118: Sacramento, California, California Department of Water Resources.

Falk, D.A., Miller, C., McKenzie, D., and Black, A.E., 2007, Cross-scale analysis of fire regimes: Ecosystems, v. 10, p. 809–823.

Faure, G., and Mensing, T.M., 2005, Isotopes: Principles and applications: Hoboken, New Jersey, John Wiley & Sons, 897 p.

Gester, G.C., 1962, The geological history of Eagle Lake, Lassen County, California: Occasional Papers of the California Academy of Sciences, v. 34, p. 1–29.

Grose, T.L.T., Saucedo, G.J., and Wagner, D.L., 1992, Geologic map of the Eagle Lake quadrangle, Lassen County, California: California Department of Conservation Division of Mines and Geology Open-File Report 92-14, scale 1:100,000.

Hallett, D.J., and Anderson, R.S., 2010, Paleofire reconstruction for high-elevation forests in the Sierra Nevada, California, with implications for wildfire synchrony and climate variability in the late Holocene: Quaternary Research, v. 73, p. 180–190.

Harding, S.T., 1965, Recent variations in the water supply of the western Great Basin: Berkeley, California, University of California, Water Resources Center Archives 16, 226 p.

Higuera, P., 2009, CharAnalysis 0.9: Diagnostic and analytical tools for sedimentary charcoal analysis: Bozeman, Montana, Montana State University.

Higuera, P.E., Brubaker, L.B., Anderson, P.M., Hu, F.S., and Brown, T.A., 2009, Vegetation mediated the impacts of postglacial climate change on fire regimes in the south-Central Brooks Range, Alaska: Ecological Monographs, v. 79, p. 201–219.

Higuera, P.E., Peters, M.E., Brubaker, L.B., and Gavin, D.G., 2007, Understanding the origin and analysis of sediment-charcoal records with a simulation model: Quaternary Science Reviews, v. 26, p. 1790–1809.

Higuera, P.E., Whitlock, C., and 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, v. 21, p. 327–341.

Hurteau, M.D., and Brooks, M.L., 2011, Short- and long-term effects of fire on carbon in US dry temperate forest systems: BioScience, v. 61, p. 139–146, doi:10.1525/bio.2011.61.2.9.

Johansen, M.P., Hakonson, T.E., Whicker, F.W., Simanton, J.R., and Stone, J.J., 2001, Hydrologic response and radionuclide transport following fire at semiarid sites: Journal of Environmental Quality, v. 30, p. 2010–2017.

Jull, A.J.T., Burr, G.S., Beck, J.W., Hodgins, G.W.L., Biddulph, D.L., Gann, J., Hatheway, A.L., Lange, T.E., and Lifton, N.A., 2006, Application of accelerator mass spectrometry to environmental and paleoclimate studies at the University of Arizona: Radionuclides in the Environment, v. 8, p. 3–23, doi:10.1016/S1569-4860(05)08001-0.

Kylander, M.E., Lind, E.M., Wastegard, S., and Löwemark, L., 2012, Recommendations for using XRF core scanning as a tool in tephrochronology: The Holocene, v. 22, p. 371–375.

Lake, F.K., 2013, Trails, fires, and tribulations: Tribal resource management and research issues in Northern California: Occasion, v. 5, p. 1–22.

Lloret, F., and Mari, G., 2001, A comparison of the medieval and the current fire regimes in managed pine forests of Catalonia (NE Spain): Forest Ecology and Management, v. 141, p. 155–163.

Maher, B.A., 2011, The magnetic properties of Quaternary aeolian dusts and sediments, and their palaeoclimatic significance: Aeolian Research, v. 3, p. 87–144, doi:10.1016/j.aeolia.2011.01.005.

Malkinson, D., and Wittenberg, L., 2011, Post fire induced soil water repellency— Modeling short and long-term processes: Geomorphology, v. 125, p. 186–192, doi:10.1016/j.geomorph.2010.09.014.

Marlon, J.R., Bartlein, P.J., Daniau, A.-L., Harrison, S.P., Maezumi, S.Y., Power, M.J., Tinner, W., and Vanniére, B., 2013, Global biomass burning: A synthesis and review of Holocene paleofire records and their controls: Quaternary Science Reviews, v. 65, p. 5–25, doi:10.1016/j.quascirev.2012.11.029.

Marlon, J.R., Bartlein, P.J., Gavin, D.G., Long, C.J., Anderson, R.S., Briles, C.E., Brown, K.J., Colombaroli, D., Hallett, D.J., and Power, M.J., 2012, Long-term perspective on wildfires in the western USA: Proceedings of the National Academy of Sciences, v. 109, p. E535–E543.

McGuire, K.R., 2007, Models made of glass: A prehistory of Northeast California, in Jones, T.L., and Klar, K., eds., California prehistory: Colonization, culture, and complexity: Plymouth, UK, AltaMira Press, p. 165-176.

Minckley, T.A., Whitlock, C., and Bartlein, P.J., 2007, Vegetation, fire, and climate history of the northwestern Great Basin during the last 14,000 years: Quaternary Science Reviews, v. 26, p. 2167–2184, doi:10.1016/j.quascirev.2007.04.009.

Minnich, R.A., 2006, California climate and fire weather, in Sugihara, N.G., van Wagtendonk, J.W., Fites-Kaufman, J., Shaffer, K. E., and Thode, A. E., eds., Fire in California’s Ecosystems: Los Angeles, California, University of California Press, p. 13–37.

Mohr, J.A., Whitlock, C., and Skinner, C.N., 2000, Postglacial vegetation and fire history, eastern Klamath Mountains, California, USA: The Holocene, v. 10, p. 587–601.

Moody, J.A., and Martin, D.A., 2009, Synthesis of sediment yields after wildland fire in different rainfall regimes in the western United States: International Journal of Wildland Fire, v. 18, p. 96, doi:10.1071/WF07162.

Moyle, P.B., Kennedy, T., Kuda, D., Martin, L., and Grant, G., 1991, Fishes of Bly Tunnel, Lassen County, California: Western North American Naturalist, v. 51, p. 267–270.

Nichols, G., 2009, Sedimentology and stratigraphy: Hoboken, NJ: Wiley-Blackwell.

Ortega, B., Vázquez, G., Caballero, M., Israde, I., Lozano-García, S., Schaaf, P., and Torres, E., 2010, Late Pleistocene: Holocene record of environmental changes in Lake Zirahuen, Central Mexico: Journal of Paleolimnology, v. 44, p. 745– 760, doi:10.1007/s10933-010-9449-x.

Owens, P.N., Giles, T.R., Petticrew, E.L., Leggat, M.S., Moore, R.D., and Eaton, B.C., 2013, Muted responses of streamflow and suspended sediment flux in a wildfire-affected watershed: Geomorphology, v. 202, p. 128-139, doi:10.1016/j.geomorph.2013.01.001.

Owens, P.N., Petticrew, E.L., and Perk, M., 2010, Sediment response to catchment disturbances: Journal of Soils and Sediments, v. 10, p. 591–596, doi:10.1007/s11368-010-0235-1.

Peters, M.E., and Higuera, P.E., 2007, Quantifying the source area of macroscopic charcoal with a particle dispersal model: Quaternary Research, v. 67, p. 304– 310.

Pisaric, M.F.J., 2002, Long-distance transport of terrestrial plant material by convection resulting from forest fires: Journal of Paleolimnology, v. 28, p. 349–354.

Power, M.J., Marlon, J.R., Bartlein, P.J., and Harrison, S.P., 2010, Fire history and the global charcoal database: A new tool for hypothesis testing and data exploration: Palaeogeography, Palaeoclimatology, Palaeoecology, v. 291, p. 52–59.

Reimer, P.J., Bard, E., Bayliss, A., Beck, J.W., Blackwell, P.G., Ramsey, C.B., Buck, C.E., Cheng, H., Edwards, R.L., and Friedrich, M., 2013, IntCal13 and Marine13 radiocarbon age calibration curves 0–50,000 years cal BP: Radiocarbon, v. 55, p. 1869–1887.

Rhodes, A.N., 1998, A method for the preparation and quantification of microscopic charcoal from terrestrial and lacustrine sediment cores: The Holocene, v. 8, p. 113–117.

Rius, D., Vannière, B., Galop, D., and Richard, H., 2011, Holocene fire regime changes from multiple-site sedimentary charcoal analyses in the Lourdes basin (Pyrenees, France): Quaternary Science Reviews, v. 30, p. 1696–1709.

Russell, I.C., 1885, Geological history of Lake Lahontan: A Quaternary lake of northwestern Nevada: Monographs of the United States Geological Survey, v. XI, p. 288.

Sankey, J.B., Germino, M.J., and Glenn, N.F., 2009, Relationships of post-fire aeolian transport to soil and atmospheric conditions: Aeolian Research, v. 1, p. 75–85, doi:10.1016/j.aeolia.2009.07.002.

Schlachter, K.J., and Horn, S.P., 2009, Sample preparation methods and replicability in macroscopic charcoal analysis: Journal of Paleolimnology, v. 44, p. 701–708, doi:10.1007/s10933-009-9305-z.

Schnurrenberger, D., Russell, J., and Kelts, K., 2003, Classification of lacustrine sediments based on sedimentary components: Journal of Paleolimnology, v. 29, p. 141–154.

Scott, A.C., Lomax, B.H., Collinson, M.E., Upchurch, G.R., and Beerling, D.J., 2000, Fire across the K–T boundary: Initial results from the Sugarite coal, New Mexico, USA: Palaeogeography, Palaeoclimatology, Palaeoecology, v. 164, p. 381–395.

Selby, M.J., 1993, Hillslope materials and processes: New York, New York, Oxford University Press, 451 p.

Shackley, M.S., editor, 2011, X-ray fluorescence spectrometry (XRF) in Geoarchaeology: New York, NY, Springer New York.

Shakesby, R.A., 2011, Post-wildfire soil erosion in the Mediterranean: Review and future research directions: Earth-Science Reviews, v. 105, p. 71–100, doi:10.1016/j.earscirev.2011.01.001.

Shakesby, R.A., and Doerr, S.H., 2006, Wildfire as a hydrological and geomorphological agent: Earth-Science Reviews, v. 74, p. 269–307, doi:10.1016/j.earscirev.2005.10.006.

Simmons, W.S., Morales, R., Williams, V., and Camacho, S., 1997, Honey Lake Maidu ethnogeography of Lassen County, California: Journal of California and Great Basin Anthropology, v. 19, p. 2–31.

Smith, H.G., Sheridan, G.J., Lane, P.N.J., Nyman, P., and Haydon, S., 2011, Wildfire effects on water quality in forest catchments: A review with implications for water supply: Journal of Hydrology, v. 396, p. 170–192, doi:10.1016/j.jhydrol.2010.10.043.

Stuiver, M., and Reimer, P.J., 1993, Extended (super 14) C database and revised CALIB 3.0 (super 14) C age calibration program: Radiocarbon, v. 35, p. 215–230.

Sugihara, N.G., and Barbour, M.G., 2006, Fire and California vegetation, in Sugihara, N.G., van Wagtendonk, J.W., Fites-Kaufman, J., Shaffer, K. E., and Thode, A. E., eds., Fire in California’s Ecosystems: Los Angeles, California, University of California Press, p. 1–9.

Tinner, W., Hofstetter, S., Zeugin, F., Conedera, M., Wohlgemuth, T., Zimmermann, L., and Zweifel, R., 2006, Long-distance transport of macroscopic charcoal by an intensive crown fire in the Swiss Alps: Implications for fire history reconstruction: The Holocene, v. 16, p. 287–292, doi:10.1191/0959683606hl925rr.

Valeron, B., and Meixner, T., 2009, Overland flow generation in chaparral ecosystems: Temporal and spatial variability: Hydrological Processes, v. 24, p. 65-75, doi:10.1002/hyp.7455.

Van Wagtendonk, J.W., 2006, Fire as a physical process, in Sugihara, N.G., van Wagtendonk, J.W., Fites-Kaufman, J., Shaffer, K. E., and Thode, A. E., eds., Fire in California’s Ecosystems: Los Angeles, California, University of California Press, p. 38–57.

Walsh, M.K., Pearl, C.A., Whitlock, C., Bartlein, P.J., and Worona, M.A., 2010, An 11000-year-long record of fire and vegetation history at Beaver Lake, Oregon, central Willamette Valley: Quaternary Science Reviews, v. 29, p. 1093–1106.

Walsh, M.K., Whitlock, C., and 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, v. 297, p. 273–289.

Walther, J.V., 2009, Essentials of geochemistry: Sudbury, MA, Jones and Bartlett Publishers, 797 p.

Wang, N., Jiao, J.-Y., Du, H.-D., Wang, D.-L., Jia, Y.-F., and Chen, Y., 2013, The role of local species pool, soil seed bank and seedling pool in natural vegetation restoration on abandoned slope land: Ecological Engineering, v. 52, p. 28–36, doi:10.1016/j.ecoleng.2012.12.055.

Ward, A.D., Trimble, S.W., and Wolman, M.G., 2003, Environmental hydrology: Boca Raton, Florida, CRC Press, 475 p.

West, J.G., Woolfenden, W., Wanket, J.A., and Anderson, R.S., 2007, Late Pleistocene and Holocene environments, in Jones, T.L., and Klar, K., eds., California prehistory: Colonization, culture, and complexity: Plymouth, UK, AltaMira Press, p. 11-34

Westerling, A.L., 2006, Warming and earlier spring increase western U.S. forest wildfire activity: Science, v. 313, p. 940–943, doi:10.1126/science.1128834.

Wetzel, R., 2001, Limnology: Lake and river ecosystems: San Diego, CA, Academic Press.

Whitlock, C., Dean, W., Rosenbaum, J., Stevens, L., Fritz, S., Bracht, B., and Power, M., 2008, A 2650-year-long record of environmental change from northern Yellowstone National Park based on a comparison of multiple proxy data: Quaternary International, v. 188, p. 126–138.

Whitlock, C., Higuera, P.E., McWethy, D.B., and Briles, C.E., 2010, Paleoecological perspectives on fire ecology: Revisiting the fire-regime concept: Open Ecology Journal, v. 3, p. 6–23.

Whitlock, C., and Larsen, C., 2001, Charcoal as a fire proxy, in Smol, J.P., Birks, H.J.B., and Last, W.M., eds., Tracking environmental change using lake sediments: Volume 3: Terrestrial, algal, and siliceous indicators, developments in paleoenvironmental research: Dordrecht, The Netherlands, Kluwer Academic, p. 75–97.

Wigand, P.E., Hemphill, M.L., Sharpe, S., and Patra, S., 1995, Eagle Lake Basin, Northern California, paleoecological study: Semi-arid woodland and montane forest dynamics during the late Quaternary in the northern Great Basin and adjacent Sierras: University and Community College System of Nevada, 27 p.

Wilkinson, S.N., Wallbrink, P.J., Hancock, G.J., Blake, W.H., Shakesby, R.A., and Doerr, S.H., 2009, Fallout radionuclide tracers identify a switch in sediment sources and transport-limited sediment yield following wildfire in a eucalypt forest: Geomorphology, v. 110, p. 140–151, doi:10.1016/j.geomorph.2009.04.001.

Zimmerman, S., Guilderson, T., Buckley, B., and Cook, E., 2010, Extension of the southern hemisphere atmospheric radiocarbon curve, 2120–850 years BP: Results from Tasmanian Huon Pine: Radiocarbon, v. 52, p. 887–894.

APPENDIX A

Figure A-1. Photography of the upper section of core ELNEC1-1. Note: scale skewed by angle of photography and adjustment error 68 69

Figure A-2. Photography of the lower section of core ELNEC1-1. Note: scale skewed by angle of photography and adjustment error

APPENDIX B

Note: Raw geochemical data used in principle component analysis (PCA). Data analytically derived using a Bruker S2 Ranger available at CSU, Chico.

Figure B-1. Magnesium (Mg) data for core ELNEC1-1.

Figure B-2. Aluminum (Al) data for core ELNEC1-1. .

71 72

Figure B-3. Silicon data for core ELNEC1-1.

Figure B-4. Titanium (TI) data for core ELNEC1-1.

Figure B-5. Iron (Fe) data for core ELNEC1-1.