Spatial variability of methane production and methanogen communities in a reservoir:

importance of organic matter source and quantity

A thesis submitted to the

Graduate School

of the University of Cincinnati

In partial fulfillment of the

requirements for the degree of

Master of Science

in the Department of Biological Sciences

of the College of Arts and Sciences

By

Megan E. Berberich

B.S. Biological Sciences, 2014

University of Louisville, Louisville KY

November 2017

Research Advisory Committee: Committee Chair: Dr. Ishi Buffam Dr. Jake J. Beaulieu Dr. Trinity L. Hamilton Abstract

Freshwater reservoirs are an important source of the greenhouse gas methane (CH4) to the atmosphere, but global emission estimates are poorly constrained (13.3 – 52.5 Tg C yr-1) due, in part, to extreme spatial variability in emission rates within and among reservoirs. While morphological characteristics, including water depth, contribute to the variation in emission rates, spatial heterogeneity of biological methane production rates by sediment dwelling methanogenic archaea may be another important source of variation.

An important constraint on CH4 production rates is the availability of organic matter

(OM). Laboratory experiments have shown that both the quantity and quality of OM influences production rates. For example, CH4 production rates have been shown to respond strongly to algal-derived OM, a highly labile OM source. It is unclear, however, whether this pattern persists at the field scale where other sources of organic matter, such as sediment loads from the watershed, may play an important role in CH4 generation. We measured methane production rates, sediment OM source, OM quantity, and methanogen community composition at fifteen sites in a temperate, eutrophic reservoir to assess OM drivers of spatial variability in CH4

-2 production rates. Areal CH4 production rates (g CH4 m ) were highest in the riverine portion of the reservoir below the main inlet where OM quantity (g OM cm-2) was greatest, presumably due to high sedimentation rates. The pattern of high CH4 production rates in the riverine portion of

-1 the reservoir persisted even when rates were normalized to OM quantity (g CH4 g OM), suggesting that not only was OM more abundant in the riverine zone, it was more readily utilized by methanogens. Sediment stable isotopes and elemental ratios indicated a greater proportion of allochthonous OM in the riverine zone than other areas of the reservoir, suggesting that watershed derived OM is an important driver of CH4 production in the system. Methanogens

ii were abundant at all sampling sites but the functional diversity of methanogens was highest in the riverine zone. Variation in functional diversity of methanogens likely reflects differences in decomposition processes or OM quality across the reservoir. In contrast to previous reports of water column primary productivity as a key predictor of CH4 emission rates in reservoirs, we found that measures of OM quantity best explained variation in CH4 production rates within the reservoir and that the highest production rates occurred at sites with a strong contribution of terrestrial OM. This indicates that while OM source is important, the total OM quantity, regardless of source, is the primary driver of CH4 production rates.

iii (This page intentionally left blank.)

Acknowledgements

I owe a great deal of thanks to my entire committee for the tremendous amount of time, energy and resources that each member dedicated to supporting me and this research.

Specifically, I thank my advisor Dr. Ishi Buffam for always making time, and for providing a research environment that promoted independence while still providing direction. Ishi allowed and encouraged me to pursue my research interests, and his mentorship and commitment are what allowed this project to develop and grow into one that challenged and excited me. My committee members Dr. Jake Beaulieu and Dr. Trinity Hamilton were both extremely active in their involvement in this project, and provided direction and encouragement in addition to intellectual, financial, field and lab support. The combined and individual strengths of each of my three committee members afforded me the opportunity to gain experience that I would not have had without this collaboration. I thank Trinity for her patience and expertise in guiding me through molecular techniques, and for her constant availability and insight. I thank Jake for his detailed and methodical approach to conducting research and addressing problems, and for his invaluable counsel. I thoroughly enjoyed working with each member, and am grateful for the opportunity to have developed relationships and worked with such outstanding and dedicated scientists and people.

I extend thanks to Dr. Sarah Waldo, who was a constant source of encouragement and provided valuable intellectual insight. Sarah’s support and perspective helped drive this research, and I am very grateful for her involvement and willingness to provide mentorship. I also would like to specifically thank Dr. Jeff Havig for his critical role in much of the isotopic analysis; Jeff gave his personal time and energy to ensure that my samples were analyzed. Further, Jeff was always willing to explain a method, discuss science, offer guidance, or just chat. I thank all those who helped in collecting and processing samples and data, including Kit Daniels, Dr. Sarah

Waldo, Madison Duke, Dr. Xuan Li, Karen White, Dr. Joel Allen, Dr. Mike Elovitz, and undergraduate helpers Caroline Tran and Kaitlin Henn. I especially want to thank undergraduate

Madison Duke for her commitment and eagerness during the summer of 2016. I thank my lab mates Jeremy Alberts, Mark Mitchell, Alicia Goldschmidt, Chelsea Hintz and Sarah Handlon for their encouragement and enjoyable company.

I would like to thank my family, especially my mom, dad, stepdad, stepmom and brother, and new and old friends for their unwavering confidence in me and support of my goals. Finally,

I thank Rob for his patience and perspective, and for driving me to pursue my interests.

ii Table of Contents

Abstract ...... ii

Acknowledgements ...... i

List of Tables and Figures ...... ii

Chapter 1: Background ...... 1

1.1 Methane: The global context ...... 1

1.2 Methanogenesis and methanogens ...... 2

1.3 Organic matter source characterization ...... 6

Chapter 2: Spatial variability of methane production and methanogen communities in a reservoir: importance of organic matter source and quantity ...... 8

2.1 Introduction ...... 8

2.2 Methods...... 15

2.3 Results ...... 24

2.4 Discussion ...... 35

References ...... 43

Appendices ...... 50

Note on reproducibility and open data access ...... 50

Appendix 1: Supplemental Data from Harsha Lake ...... 52

Appendix 2: Select Detailed Methods ...... 62

Appendix 3: Sediment Traps ...... 68

Appendix 4: July Slurries – Absence of Methane Production ...... 74

List of Tables and Figures

Tables

Table 2.1. Physical and chemical water column properties for each of the sampling sites...... 26

Table 2.2. Bulk sediment and porewater characteristics for each of the 15 sites...... 29

Table 2.3. Results of model hypothesis testing...... 34

-3 -1 Table 2.4. Best predictors of methane production rates (µmol CH4 cm day ) from univariate linear regression analysis...... 35

Figures

Figure 1.1. Steps involved in the anaerobic degradation of organic matter. From Conrad (1999)...... 4

Figure 1.2. Phylogeny of methanogens. From Borrel et al. (2011)...... 5

Figure 1.3. Figure from Bianchi & Canuel (2011). Schematic representation of various tools for characterizing OM...... 6

Figure 1.4. Figure from Meyers (1994). Plot of stable carbon isotopes and C/N ratios that are typical of terrestrial vs. aquatic organic matter sources...... 7

Figure 2.1. Simplified concept diagram of this study...... 13

Figure 2.2. Predicted variation in sediment composition, methane production rates, and microbial community structure across the three reservoir zones...... 14

Figure 2.3. Harsha Lake sampling sites and site categories...... 16

Figure 2.4. Potential methane production rates from sediment slurries in each of the reservoir zones...... 25

Figure 2.5. Comparison of the quantity of OM among reservoir zones...... 27

Figure 2.6. Plot of �15N vs. N/C elemental ratios of terrestrial and aquatic OM sources, and the sediment mixtures for each core across the three reservoir zones (A). The proportion of autochthonous (aquatic) OM found in each core grouped by reservoir zones (B) ...... 28

ii Figure 2.7. Optical indices of dissolved organic matter in sediment porewater...... 28

Figure 2.8. Abundance of 16S archaeal gene (A) and mcrA gene (total methanogens) (B) among the three reservoir zones...... 30

Figure 2.9. Alpha diversity of bacterial and archaeal 16S rRNA sequences for a subset of the samples...... 31

Figure 2.10. Heat maps showing relative abundance of methanogen orders. Values are normalized to total number of sequences of all methanogens, and percentages are averages of all samples within a group...... 31

Figure 2.11. Heat maps showing relative abundance of methanogen genera. Row labels show order and genus (order; genus)...... 32

Figure 2.12. Carbon isotope signature in CH4 (A) and the apparent fractionation factors (B) across reservoir zones. The blue horizontal line (B) indicates the cutoff between hydrogenotrophic (above) and acetoclastic (below) methanogenesis...... 33

iii Chapter 1: Background

This chapter provides background information that is relevant to understanding the context of this research, the types of data, and methods used.

1.1 Methane: The global context

Methane (CH4) is an important greenhouse gas (GHG) with over 25 times the warming potential of CO2 on a 100-year time scale (Myhre et al. 2013). Since preindustrial times, it is responsible for about 20% of warming and has risen in atmospheric concentrations by 150% (Ciais et al.

2013; Kirschke et al. 2013; Myhre et al. 2013). Atmospheric CH4 contributes to the production of tropospheric ozone, stratospheric water vapor, and CO2, all of which enhance warming (Saunois et al. 2016).

Atmospheric CH4 increases when methane emissions exceed methane sinks. CH4 emitted to the atmosphere originates from biogenic, thermogenic and pyrogenic sources derived from human activities or natural processes (Neef et al. 2010). Biogenic methane results from microbial degradation of organic matter under anaerobic conditions (Conrad 1996) (see section 1.2).

Wetlands are the largest natural biogenic CH4 source, while important human-caused biogenic methane sources include rice paddies, landfills and waste, ruminants, and freshwater reservoirs

(human-made lakes) (Ciais et al. 2013). Thermogenic methane is created when organic matter is subjected to high heat and pressure over geological time scales, resulting in the creation of natural gas (which is primarily composed of CH4) and other fossil fuels. Natural thermogenic sources include marine methane hydrates, geothermal vents, and volcanoes. Unintentional

1 leakage and incomplete combustion from the extraction, production, transportation, and use of fossil fuels is a main anthropogenic methane emission source (Ciais et al. 2013). Pyrogenic sources of methane result from incomplete burning of biomass and biofuels, and can be both naturally occurring and anthropogenic.

Two approaches are used to quantify global emissions: top-down and bottom-up. The top-down approach estimates GHG exchange using initial estimates of surfaces fluxes to optimize consistency with measured atmospheric GHG concentrations, while the bottom-up approach uses ecosystem-scale empirical carbon flux measurements to scale up to global emissions (Kondo et al. 2015). A recent estimate of global methane emissions from the top-down approach is ~554

-1 Tg CH4 yr (Prather et al. 2012), while estimates of global methane emissions from the bottom-

-1 up approach are usually greater, and range from 542 – 852 Tg CH4 yr (Myhre et al. 2013).

These estimates include all emission sources, of which the estimated range of anthropogenic sources is between 50% and 65% (Ciais et al. 2013). The main emission source that contributes to the discrepancy between top-down and bottom-up estimates is natural emissions, particularly emissions from freshwater systems (Kirschke et al. 2013). Better estimates of emissions from these sources will reduce uncertainty in their global contribution to methane emissions (Kirschke et al. 2013).

1.2 Methanogenesis and methanogens

Biological methane production, or methanogenesis, is carried out by methanogenic archaea during the terminal step of organic matter degradation in anoxic environments, including lake sediments (Ferry 1992). The anaerobic decomposition of organic matter is a multi-step process

2 that involves fermenting and anaerobic oxidizing bacteria breaking down complex molecules to subsequently smaller molecules that methanogens then convert to CH4 (Angelidaki et al. 2011)

(Figure 1.1). Methanogens can utilize a limited number of substrates, namely acetate, CO2, and methylated compounds such as methanol, methylamines, and dimethylsulfide. There are three pathways by which methane is biologically produced that correspond to these initial substrates:

1) Acetoclastic methanogenesis: CH3COOH à CO2 + CH4

2) Hydrogenotrophic methanogenesis: 4H2 + CO2 à CH4 + 2H2O

3) Methylotrophic methanogenesis: 4CH3OH à 3CH4 + CO2 + 2H2O.

Acetoclastic and hydrogenotrophic methanogenesis are considered the dominant methanogenic pathways in freshwater environments, where theoretically ~67% of methanogenesis should be from acetate and ~33% should be from hydrogen (Figure 1.1), assuming that the starting material is derived from carbohydrates (an abundant component of OM in aquatic sediments) (Conrad

1999).

Presumably, variation in abundance and composition of methanogens contributes to observed biogenic methane production rates. However, in studies reporting methanogen community abundance and methane production, some have observed a correlation between community size and methane production while others report methane production rates that vary regardless of methanogen abundance (Schwarz et al. 2009, West et al. 2012).

3

Figure 1.1. Steps involved in the anaerobic degradation of organic matter. From Conrad (1999).

Taxonomically, methanogens can be divided into 6 known orders: Methanomicrobiales,

Methanocellales, Methanosarcinales, Methanococcales, Methanopyrales, and

Methanobacteriales (Figure 1.2) (Borrel et al. 2011). Of these orders, only Methanosarcinales has representatives that are capable of acetoclastic methanogenesis, found in the genera

Methanosarcina and Methanosaeta (Liu & Whitman 2008). Methanosaeta spp. are obligate acetate-utilizers, while the Methanosarcina genus contains representatives that can utilize acetate, methylated compounds, and CO2 in some cases (Liu & Whitman 2008). Other genera in the order Methanomicrobiales are methylotrophic methanogens, and include

Methanomethylovorans and Methanolobus, both of which are known to occur in freshwater lakes

(Borrel et al. 2011). Methylotrophic methanogenesis is only performed by members of the order

4 Methanosarcinales, with the exception of some species of one genus in the order

Methanobacteriales capable of this metabolic pathway.

Figure 1.2. Phylogeny of methanogens. From Borrel et al. (2011).

All of the metabolic pathways used by methanogens utilize the same enzyme complex in the terminal step of methane release: methyl coenzyme-M reductase (MCR) (Luton et al. 2002). It is thought that this enzyme complex is only found in methanogens (Thauer 1998), making it useful in the detection and identification of methanogens (Luton et al. 2002). Specifically, the gene that encodes the alpha subunit of MCR (mcrA) has been used extensively to detect methanogens

(Luton et al. 2002).

5 1.3 Organic matter source characterization

Determination of organic matter (OM) source is complicated due to OM processing and

transformation, especially in aquatic environments. Therefore, a variety of tools are used to

characterize OM (Figure 1.3), a combination of which were used in the research presented in

Chapter 2. While several metrics can be used to describe OM, including elemental composition,

OM percent or concentration, and the age of OM (using radiocarbon dating), it is often useful to

determine the original source of the organic matter. This is done for purposes such as December 28, 2010 Time: 09:06am chapter2.tex constraining the carbon cycle in specific ecosystems, tracing organic pollutants, measuring

biogeochemical processes, and understanding ancient ecosystems.

20 ■ Chapter 2

Bulk organic matter Decreasing fraction of bulk organic matter city f TOC, TN

Elemental composition C:N, C:P, N:P

Bulk stable isotopes δ13C, δ15N, δ D

Functional group composition 1H-, 13C-, 15N-, and 31P- NMR

Biomarker composition pigments, fatty acids, amino acids

Increasing information and source speci Compound specifc isotope analysis δδδ13C, 15N, D, ∆14C of biomarkers

FigureFigure 1.3. Figure 2.1. Schematic from Bianchi illustrating & Canuel how (2011). approaches Schematic for characterizing representation organicof various matter tools differ for characterizing in their source OM and howspecificity the tools (left) vary and in thedetermining proportion source of organic specificity matter they(left) represent and the fraction (right). of OM they represent (right).

matter sources in aquatic systems using tools such as elemental, isotopic (bulk and compound specific), The methodsand chemical used biomarker to characterize methods. OM These source tools each differ have in their their ability strengths both to and identify limitations. specific sources For of organic matter and to represent bulk organic matter (fig. 2.1). This chapter provides a general overview example,of the in strengths Figure and1.3 weaknessesBianchi and of differentCanuel (2011) approaches indicate used forthat characterizing to gain more organic information matter sources about in aquatic environments. the source of OM, the fraction of OM that is being described by that information decreases. In

environments2.3 Bulk with Organic complex Matter and Techniques diverse OM, this may be limiting; however, conversely, the

The abundance and ratios of elements important in biological cycles (e.g., C, H, N, O, S, and P) provide the basic foundation of information about6 organic matter sources and cycling. For example, concentrations of total organic carbon (TOC) provide the most important indicator of organic matter, since approximately 50% of most organic matter is composed of C. When bulk TOC measurements

are combined with additional elemental information, as in the case of atomic C to N ratios ((C : N)a or mol : mol), basic source information can be inferred about algal and vascular plant sources (see

review in Meyers, 1997). The broad range of (C : N)a ratios across divergent sources of organic matter in the biosphere demonstrate how such a ratio can provide an initial proxy for determining source

information (table 2.1). Differences in (C : N)a ratios for vascular plants (>17) and microalgae (5 to 7) are largely due to differences in contributions from structural components. A large fraction of the organic matter in vascular plants is composed of carbon-rich biochemicals, such as carbohydrates and lignin. The most abundant carbohydrates supporting this high carbon content in vascular plants are structural polysaccharides,suchascellulose, hemicellulose,andpectin (Aspinall, 1970). In contrast, algae tend to be protein-rich and carbohydrate-poor because of both an absence of these structural components and higher contributions (relative to biomass) of protein and nucleic acids (fig. 2.2). However, these generalizations do not hold for all plants. A recent survey of the elemental composition of a variety of plant types in San Francisco Bay (USA) showed that C : N ratios were highly 1 variable, ranging from 4.3 to 196 (mol C mol− N), with the highest ratios measured in terrestrial and marsh vascular plants and lower ratios in aquatic plants (Cloern et al., 2002). Despite the general observation that the C : N ratios of terrestrial and marsh vascular plants were higher than for aquatic plants, C : N ratios ranged from minima of 10–20 to maxima of 40 to >100, suggesting large variability at the species level in biochemical composition. Seasonal variations in C : N ratios for individual plants information gained by describing bulk OM has limitations in the specificity of source description. While it is generally preferred to use techniques that provide specific information about OM sources (i.e. using biomarkers), especially in the field of organic geochemistry, there are often practical limitations to these methods, including cost and expertise required to analyze samples and interpret results. Therefore, techniques that characterize the bulk OM (as opposed to a fraction of the OM) are widely used.

The elemental composition of bulk OM and various ratios of elements can be used to discern

OM sources (Bianchi & Canuel 2011). For example, C/N atomic ratios can indicate if OM is derived from terrestrial plants or algal sources (Meyers 1994). Terrestrial plants contain more carbon (C) -containing structural material than algae, resulting in higher C/N ratios relative to those found in algae or microbial sources (Figure 1.4). An additional tool that is often used in combination with elemental composition is the stable isotope composition of various elemental components of OM, including �13C, �15N, and �D.

Figure 1.4. Figure from Meyers (1994). Plot of stable carbon isotopes and C/N ratios that are typical of terrestrial vs. aquatic organic matter sources.

7 Chapter 2: Spatial variability of methane production and methanogen communities in a reservoir: importance of organic matter source and quantity

2.1 Introduction

Inland waters are a globally significant source of methane (CH4) (Bastviken et al. 2011;

Holgerson & Raymond 2016; Stanley et al. 2016), a greenhouse gas (GHG) with over 25 times the warming potential of carbon dioxide (CO2) (Myhre et al. 2013). Methane emissions from reservoirs are of particular interest due to the increasing number of these systems, their land coverage area, and their high biogeochemical processing rates (Downing et al. 2006; Harrison et al. 2009; Zarfl et al. 2015). Additionally, emissions from these human-made ecosystems are categorized as ‘anthropogenic’ by the Intragovernmental Panel on Climate Change (IPCC)

(Prairie et al. In Press) and signatories to the UNFCCC may therefore be required to report these emission in future anthropogenic emission inventories.

-1 Global CH4 fluxes from reservoirs remain poorly constrained (13.3 – 52.5 Tg C year ; Deemer et al. 2016), due in part to uncertainty in reported emission rate estimates for individual reservoirs.

The temporal and spatial resolution of sampling strategies varies tremendously among studies and differences in measurement approaches can add additional variability to emission rate estimates. Intrareservoir spatial variability has received considerable attention recently with several studies reporting ‘hot spots’ of CH4 emission near the river-reservoir transition zone (e.g.

Beaulieu et al. 2014, 2016; DelSontro et al. 2011). While physical properties can affect gas transport through the water column and release from the water surface (e.g. McGinnis et al.

8 2006), the spatial heterogeneity of methane production in sediments also likely contributes to the spatial variation in fluxes from the water surface.

The variation in methane emissions from reservoirs adds additional uncertainty in global CH4 fluxes. Identification of the best predictors of CH4 emission rates could offer a cross-system metric for constraining estimates of global emissions from reservoirs. Age and latitude explain some variation in emission rates; tropical and newly created reservoirs exhibit higher emission rates than boreal and older reservoirs (Barros et al. 2011). However, more recent work has shown that emission rates from temperate, midlatitude reservoirs can be much higher than both their age and latitude would predict (e.g. DelSontro et al. 2010, Maeck et al. 2013, Beaulieu et al. 2014,), possibly because carbon (C) derived from algal blooms in these productive reservoirs stimulates

CH4 production. This hypothesis is consistent with a recent report that the best predictor of CH4 emission rates from reservoirs across the globe is trophic status and chlorophyll a (Deemer et al.

2016), suggesting that CH4 production is closely coupled to autochthonous production.

Algal biomass may provide a labile C source for organisms involved in OM decomposition in anoxic sediments, including those responsible for methane production (methanogens). The anaerobic degradation of organic matter is a multi-step process that requires three major physiological groups of organisms to successively convert complex OM molecules to smaller substrates: primary fermenting bacteria, anaerobic oxidizing bacteria, and finally, methanogenic archaea (Angelidaki et al. 2011). Laboratory studies have demonstrated that OM additions to lake sediments provide a substrate for methanogens and increase methane production rates

(Schwarz et al. 2009; West et al. 2012, 2015a), but the response to algal derived OM is greater

9 than that to the same mass of terrestrial OM (West et al. 2012). This is consistent with reports that sediment methane production rates across different lakes are positively correlated with lake productivity (West et al. 2015b). However, the relationship between chlorophyll a, a proxy for lake productivity, and CH4 production within a lake or reservoir has not been shown.

Additionally, the established relationships rely on water column measurements of productivity or chlorophyll a, whereas the relationship between sediment CH4 production and measures of algal- derived (autochthonous) C found in the sediment have not been investigated.

An alternative hypothesis to explain the link between lake or reservoir productivity and CH4 emissions and sediment production is that methanogenesis rates are driven by the quantity of organic matter (rather than OM source). Eutrophic and highly productive systems that receive high sediment loads from the watershed may support increased CH4 emissions simply due to the large amount of OM available for decomposition and subsequent methane production. In fact, sediment accumulation and ebullative emission rates have been shown to positively correlate in a small temperate reservoir (Maeck et al. 2013). Under this hypothesis, water column productivity or chlorophyll a would still be a useful proxy for CH4 production; these measures would reflect trophic status and would relate to the quantity of OM loading to the sediment regardless of source (autochthonous (algal-derived) or allochthonous (terrestrial-derived) OM).

Allochthonous OM may be another important OM source fueling methanogenesis. Reservoirs can have exceptionally high rates of carbon burial relative to lakes (Tranvik et al. 2009), and the large watershed to surface area ratio of reservoirs results in high sediment loading (Thornton

1990; Knoll et al. 2014; Hayes et al. 2017). Although the current paradigm is that organic matter

10 derived from terrestrial sources is older and less labile than newer algal and microbial OM sources, there is growing evidence that this “recalcitrant” carbon can fuel respiration in aquatic ecosystems (see Guillemette et al. 2017 and McCallister & delGiorgio 2012). Although CH4 emission rates from reservoirs have been shown to correlate with algal productivity at the global scale, the relative importance of allochthonous and autochthonous OM as a CH4 precursor across the productivity gradient remains unclear.

Reservoirs are ideal ecosystems for investigating the relationship between OM dynamics and

CH4 production. They are comprised of three functional zones – riverine, transitional, and lacustrine – characterized by differences in depth, thermal stratification, sediment composition, primary productivity, and water velocity (Thornton et al. 1990). The quantity of OM in the sediment is expected to vary across reservoir zones as well as the relative contribution of autochthonous and allochthonous OM to the sediments. This natural gradient in OM quantity and quality (or source) can be used as an experimental template to study how differences in organic matter dynamics affect heterogeneity in methane production, and evaluate if algal-derived C in sediments is related to CH4 production.

Methanogens play a significant role in the global carbon cycle yet their composition and activity

(methane production) are rarely studied together in freshwater environments. Further, although organic matter can shape microbial communities in freshwater environments (Fagervold et al.

2014), little is known about the response of the methanogenic community to variations in organic matter source or quantity in these ecosystems. Differences in the types and amounts of OM available in an ecosystem are likely reflected in the composition and/or activity of the microbial

11 community. Taxonomic groups of methanogens vary in their ability to utilize different substrates

(Liu & Whitman 2008); thus, variation in substrate availability in freshwater sediment likely affects methanogen community structure and activity.

To address the role of carbon source (allochthonous vs. autochthonous) and quantity on sediment

CH4 production and methanogen communities, we investigated methanogenesis, sediment OM characteristics, and methanogen communities across Harsha Lake, a eutrophic reservoir located in the midwestern United States. The surface CH4 emission rates of this reservoir have been quantified previously (Beaulieu et al. 2014; Beaulieu et al. 2016), and extensive water quality sampling by the USEPA informed our understanding of the productivity of the system. Both methane emissions and summer chlorophyll a concentrations are highest near the main tributary of the reservoir (Nietch, unpublished data). These data informed our sample design and conceptual framework (Figure 2.1).

The objectives of this study were to: (1) establish if methane production rates, bulk sediment characteristics, sediment OM composition, and sediment methanogen communities exhibit distinct variation among reservoir zones that reflect the previously-reported patterns in methane emissions and chlorophyll a concentrations, (2) determine the relative influence of OM source and OM quantity on sediment methane production rates, and (3) identify the variables that best predict methane production rates from a suite of sediment, water column, and methanogen community measurements.

12

CH emissions to atmosphere Allochthonous carbon 4

WAT E RSHE D

Autochthonous carbon

CH transport and CH oxidation 4 4 Sediment processing and deposition

WAT E R COL UMN

Methanogen abundance and community structure CH production Sediment OM 4 Microbe mediated decomposition

SEDIMENT

Figure 2.1. Simplified concept diagram of this study. Black text boxes represent sediment inputs (not quantified here). Light gray text boxes represent methane transport (not quantified here). White text boxes are the pieces that will be discussed throughout the paper or in supporting information, with an emphasis on the sediment processes.

We addressed these objectives by collecting triplicate sediment cores from fifteen sites across three reservoir zones in Harsha Lake to determine methanogenesis rates and methanogen community composition, and to characterize the OM source and physical properties of the sediment using a variety of techniques. Our predicted spatial patterns for these variables are illustrated in Figure 2.2. Briefly, we expected that methanogenesis rates would be highest in the riverine zone, where sedimentation rates and algal abundance would be greatest, and be lowest in the lacustrine zone. Methanogen abundance was expected to increase with high CH4 production rates, and methanogen community composition would vary spatially along the riverine-lacustrine gradient.

13 Flow

Riverine Zone Transitional Zone Lacustrine Zone Site: Shallow, not Site: Medium depth, Site: Deep, stable thermally stratified, high weak thermal thermal stratification, turbidity stratification, decreasing low turbidity Sediment deposition: turbidity Sediment deposition: High rates, Sediment deposition: Low rates, mostly allochthonous and Medium rates, autochthonous inputs autocthonous inputs decreasing allochthonous Sediment: Higher % Sediment: Low % OM, inputs OM than other zones, dense Sediment: Increasing % least dense CH production: High 4 OM, less dense CH production: Low rates 4 CH production: rates Microbial community 4 Decreasing rates Microbial community structure: Potentially Microbial community structure: Low higher methanogen structure: Less methanogen abundance, abundance and methanogen abundance, fewer acetoclastic acetoclastic diversity and representatives, diversity representatives, diversity representatives unique to and representatives and representatives zone unique to zone unique to zone DAM

Figure 2.2. Predicted variation in sediment composition, methane production rates, and microbial community structure across the three reservoir zones.

We used the data to evaluate the following three hypotheses relating organic matter dynamics to rates of methanogenesis using an information-theory approach:

H1: Organic matter source alone is the best predictor of CH4 production rates in Harsha Lake.

H2: Organic matter quantity alone is the best predictor of CH4 production rates in Harsha Lake.

H3: The combination of OM source and quantity is the best predictor of CH4 production rates in Harsha

Lake.

Based on direct and indirect evidence of the mechanistic relationship between algal-derived carbon and methanogenesis rates, we hypothesized that the source of OM would best predict methanogenesis rates (H1). Specifically, we expected that autochthonous OM content would have a positive relationship with high production rates. We anticipated that OM characteristics, water column depth, and chlorophyll concentrations would best explain CH4 production rates.

Collectively, the results presented in this thesis contribute to our understanding of environmental

14 factors that influence spatial variation in methane production and methanogen community composition in freshwater sediments.

2.2 Methods

Site description. William H. Harsha Lake is a reservoir in southwest Ohio that was built on the

East Fork of the Little Miami River in 1978. The primary functions of this reservoir include flood control, drinking water supply, recreation, and wildlife habitat. It has a surface area of 7.9 km2, is seasonally stratified, and reaches 32.8 m at its maximum depth. Harsha Lake’s watershed is 882 km2, with agriculture (including corn, soybean and pasture) as the dominant land-use

(Beaulieu et al. 2016).

Sample collection. Triplicate sediment cores, water samples, and water column measurements were collected from 15 sites across the reservoir (Figure 2.3). Sites were selected to span the length of the reservoir and encompass a range of water depths, temperature, oxygen, productivity, and inputs to the sediment. Sites were categorized into reservoir zones (riverine, transitional or lacustrine) based on thermal stratification from temperature-depth profiles measured at each site in July 2016, when stratification was expected to be strongest. Sites with strong thermal stratification, including a dramatic thermocline and hypolimnion thickness >1 m, were defined as lacustrine. Sites with weak stratification or a hypolimnion thickness <1 m were categorized as transitional, and sites that were not stratified were defined as riverine. July temperature-depth profile data can be found in Appendix 1 (Supplemental Table 4). Sediment and water sampling was conducted on three separate dates that occurred within a 7-day window in May 2016. On each date (May 19, 23, 26, 2016) three sediment cores were collected from five sites distributed across the three reservoir zones using a K-B Corer (Wildco ®, Yulee, FL,

15 U.S.A.). Epilimnion (0.1 m) and hypolimnion (0.5-2 m above sediment) water samples were collected at each site using a Niskin bottle. At each site, depth profiles of water temperature, pH, dissolved oxygen, specific conductivity were measured using a ProDSS multiparameter sonde

(YSI, Yellow Springs, OH, U.S.A.), and depth profiles of chromophoric dissolved organic matter

(CDOM), in vivo chlorophyll a, and turbidity were measured using a C3 Submersible

Fluorometer (Turner Designs, Sunnyvale, CA, U.S.A.). Secchi depth and depth profiles of photosynthetically active radiation (PAR) were also measured.

Figure 2.3. Harsha Lake sampling sites and site categories. Green sites (n = 4) represent the riverine sites, blue sites (n = 5) represent transitional sites, and red sites (n = 6) are the lacustrine sites. Site categorization was based on thermal stratification. Riverine sites did not stratify, transitional sites had weak or transitory stratification, and lacustrine sites had strong thermal stratification. Sites were categorized post hoc, and stratification data was based on vertical temperature profiles from July 2016.

Water sample processing. Water samples collected from each site were stored on ice or refrigerated until they were processed (within 24 hours). Water samples were analyzed for chlorophyll a, dissolved nutrients, and total suspended solids (TSS).

16 Sediment processing. Sediment cores were sectioned and processed within 24 hours of collection. The top 5 cm of each core was extruded, homogenized, and subsampled for potential

CH4 production rate assays, sediment characterization, porewater collection and nucleic acid extraction in a glove box under N2 atmosphere. Sediment slurries were prepared in the glove box by adding 15 mL of sediment and 15 mL of lake water to a 120 mL serum bottle using syringes.

Slurries were capped with a rubber stopper, crimp sealed, and wrapped in aluminum foil.

Subsamples of sediment for DNA extraction and sediment characterization were aliquoted into the appropriate containers and frozen at -80˚C (for DNA extraction) or -20˚C (for sediment characterization) until further analysis. Samples for porewater collection were refrigerated (4˚C) until the porewater was extracted (within 24 hours of core subsampling).

Sediment slurries – potential CH4 production rates. After removing the capped slurries from the glove box, they were shaken vigorously for 2 minutes and purged with N2 gas for 5 minutes. All slurries were stored in the dark at room temperature (~23˚C) during the 9-day incubation. 11 mL gas samples were taken on days 1, 2, 3, 5, 7 and 9. After taking the gas sample, 11 mL of N2 gas was returned to the serum bottle to maintain the same pressure. Gas samples were analyzed on a gas chromatograph (Bruker 450, Massachusetts, U.S.A.) equipped with a flame ionization detector (FID). Methane production rates for the slurries were calculated by accounting for dilution during sampling, then determining the change of moles of methane over time. Methane production rates were expressed either normalized to sediment volume, sediment dry mass, or mass of sediment organic matter.

13 13 � C in CO2 and CH4 – methane production pathway. Stable isotope ratios of carbon (� C) in

CO2 and CH4 were measured from gas sampled on the last day (day 9) of a subset of the microcosms to discern between dominant methane production pathways. Analysis was carried

17 13 out at the UC Davis Stable Isotope Facility. � C – CO2 was measured on a ThermoScientific

GasBench system interfaced to a ThermoScientific Delta V Plus isotope ratio mass spectrometer

13 (IRMS) (ThermoScientific, Bremen, Germany), and � C – CH4 was measured on a

ThermoScientific Precon concentration unit system interfaced to a ThermoScientific Delta V

Plus IRMS (ThermoScientific, Bremen, Germany). Isotope fractionation occurs in many chemical reactions, including methanogenesis. The two main methane production pathways, hydrogenotrophic and acetoclastic, differ in the degree to which they exhibit isotope fractionation; carbon isotope fractionation is greater in hydrogenotrophic methanogenesis relative to acetoclastic methanogenesis (Whiticar et al. 1986). To accurately calculate the percentage of each methane production pathway, the � 13C of the methyl group on acetate is required, as well as isotopic fractionation factors that are specific to the study system of interest

(Conrad 2005). These values can be difficult to obtain, and are often poorly constrained.

13 However, the apparent fractionation factor (aC) requires only � C in CO2 and CH4 and can be used to estimate the dominant methanogenesis pathway (Whiticar et al. 1986; Conrad 2005). It was calculated as:

( ) � = . ( )

Apparent fractionation factor values greater than 1.065 indicate that hydrogenotrophic methanogenesis is the dominant pathway, while aC < 1.055 are indicative of acetoclastic methanogenesis (Whiticar et al. 1986; Whiticar et al. 1999; Conrad 2005).

Sediment characterization. Sediment was dried at 60˚C for 3 days or until the weight was constant, and the difference between the wet and dry sediment weight was used to determine water content. Dried sediment was ground with a ceramic mortar and pestle prior to all other

18 analyses. Measured sediment characteristics are reported in Table 2.2. Elemental analysis and stable isotope analysis (13C and 15N) were conducted using an elemental analyzer connected to an isotope ration mass spectrometer (EA-IRMS). Analysis with EA-IRMS was conducted at the UC

Davis Stable Isotope Facility (Elementar Vario EL Cube, Elementar Analysensysteme GmbH,

Hanau, Germany interfaced to a PDZ Europa 20-20 isotope ratio mass spectrometer, Sercon Ltd.,

Cheshire, UK) or the Stable Isotope Geochemistry Lab in the Department of Geology in the

University of Cincinnati (Costech Instruments Elemental Analyzer periphery interfaced to a

Thermo Scientific Delta V Advantage Isotope Ratio Mass Spectrometer).

An analytical intercomparison between laboratories verified that the different instruments used gave consistent elemental and isotopic results. Sediment was fumigated with hydrochloric acid prior to C and 13C analysis to removed carbonates (Harris et al. 2001). Density was calculated from the weight of a known volume of sediment. Organic matter content was calculated by determining weight loss after ignition (550˚C, 4 hours). Elemental and isotopic composition of sediment was used to determine the proportion of autochthonous and allochthonous organic matter (see Mixing model).

Porewater analyses. Porewater was extracted from a subsample of sediment for determination of volatile fatty acids (VFA), dissolved organic carbon (DOC) and for excitation emission matrix

(EEMs) fluorescence. We extracted porewater by centrifuging sediment in 50 mL tubes at 7800 rpm. The porewater was filtered at 0.45 µm prior to measurement of optical absorbance and fluorescence using a scanning spectrofluorometer. The optical data were used to calculate FI

(fluorescence index), BIX (biological index), RFE (relative fluorescence efficiency) and

SUVA254 (specific ultraviolet absorbance at 254 nm), which provide information about the composition and source of dissolved organic matter. FI is calculated as the ratio of emission

19 wavelengths 470 nm / 520 nm at excitation 370 (McKnight et al. 2001). BIX is the ratio of emission wavelengths 380 nm / 430 nm at excitation 310 nm (Huguet et al. 2009). RFE is the ratio of fluorescence at excitation 370 nm and emission 460 nm to the absorbance at 370 nm

(Downing et al. 2009). SUVA254 is the absorbance at 254 nm divided by DOC concentration

(Weishaar et al. 2003). A summary of these and other optical properties is described by Hansen et al (2016).

DNA extraction. DNA was extracted from a ~500 mg (wet weight) sediment subsample from each core using a MoBio PowerSoil ® DNA Isolation Kit (MoBio, Carlsbad, CA, U.S.A.) following the manufacturer’s protocol. DNA concentration was determined using a Qubit dsDNA HS Assay kit and a Qubit 3.0 Fluorometer (Invitrogen, Burlington, ON, Canada). DNA was used in Microbial community analysis and qPCR.

Microbial community analysis. The V4 region of the 16S rRNA gene was amplified and sequenced using the paired-end Illumina MiSeq sequencing platform with the primers 515f and

806rB (Caporaso et al. 2012; Apprill et al. 2015). All sequencing was performed at the Center for Bioinformatics & Functional Genomics at Miami University (Oxford, OH, USA). The resulting sequences were subjected to quality control, alignment to the SILVA (v128) database, chimera check, and classification of operational taxonomic units (OTUs) at a sequence identity of 97% using the mothur software (v1.39.5) (Schloss et al. 2009) following the MiSeq SOP protocol from Kozich et al. 2013 (accessed: 2/28/17). All sequences from two sediment cores

(one from site 5 and one from site 7) were removed during quality control steps in mothur and were not included in the results. One OTU from an unclassified methanogen order with only one identified sequence was excluded from analysis. Figures for visualizing community data were

20 generated using phyloseq v.1.20.0 (McMurdie & Holmes 2013) and ampvis v.1.27.0. (Albertson et al 2015). qPCR. Quantitative polymerase chain reactions (qPCR) were performed to determine the abundance of mcrA (a biomarker for methanogens), and archaeal 16S rRNA genes on a StepOne

Plus ™ Real-Time PCR System. mcrA encodes a protein necessary for methanogenesis and has been widely used as a marker for methanogens (Luton et al. 2002). mcrA was quantified using the degenerate primers mcrA-F(5’-GGTGGTGTMGGATTCACACARtAYGCWACAGC-3’) and mcrA-R (5’-TTCATTGCRTAGTTWGGRTAGTT-3’) (Luton et al. 2002). Each 20 µL

SYBR qPCR reaction contained 2x SYBR Green PCR Universal Master Mix (Applied

Biosystems), 2 µL of DNA (pre-diluted to 2 ng/µL), and 500 nM of the forward and reverse primers using the following cycling conditions: 40 cycles of 15s at 95˚C followed by 60s at

55˚C. A TaqMan probe and primer set developed by Yu et al. (2005) were used quantify total archaea. TaqMan qPCR assays (20 µL final volume) consisted of 2x TaqMan PCR Universal

Master Mix (Applied Biosystems), 500 nM of the forward and reverse primers, 200 nM of the

TaqMan probe, and 2 µL of DNA (pre-diluted to 2 ng/µL) and were performed using the following cycling conditions: 45 cycles of 15s at 95˚C followed by 60s at 60˚C. The standard curve for the SYBR green assay was constructed using an mcrA clone (Promega pGEM® -T

Easy Vector with JM109 High Efficiency Competent Cells) from environmental DNA. Briefly, mcrA was PCR-amplified from a Harsha Lake sediment DNA sample, purified (Wizard® PCR

Preps DNA Purification System, Applied Biosystems), inserted into a vector, transformed into

JM109 High Efficiency Competent Cells, and then the vector was isolated (Wizard® Plus SV

Minipreps DNA Purification System, Applied Biosystems). Manufacturer’s instructions were followed for each step. Standard curves for archaeal 16S rRNA genes were constructed from a

21 PCR-amplified 16S rRNA gene from a pure archaeal culture and purified (Wizard® PCR Preps

DNA Purification System). The vector and purified PCR product were quantified as described above. The copy number of each gene was normalized by the concentration of DNA and the grams of sediment (wet weight) used in the DNA extraction.

Statistical methods and data analysis. All statistical analyses were performed using R version

3.4.0 (R Core Team 2017). One-way nested ANOVAs were used to evaluate differences in sediment characteristics, methane production, and microbial community characteristics among reservoir zones.

Mixed effects models with site as a random factor were performed using the nlme package

(Pinheiro et al. 2017) as described in Zuur et al. (2009). This model structure nests replicate cores within each of the 15 sampling sites, thereby accounting for the likelihood that measurements from replicate cores are more likely to be related to each other than to measurements from cores at other sites. To evaluate the relationships between organic matter source or quantity and methane potential production rates, an information-theory approach was used (Anderson 2008). Models were generated to represent the following working hypotheses: 1) methane production rates are best explained by OM source, 2) methane production rates are best explained by OM quantity, and 3) methane production rates are best explained by the combination of OM source and quantity. All models used methane production rates normalized by volume as the response variable. Model structure and output are described in the results section. AIC scored corrected for the number of estimated parameters (AICc) were used in model ranking. Akaike weights (wi) representing the relative likelihood of the model, evidence ratios (Ei,j), and AICc values were calculated as described by Anderson (2008). Correlation among independent variables was assessed using variance inflation factor (VIF) scores, and

22 variables with scores higher than 3 were excluded from models (Zuur et al. 2010). The best model for each hypothesis was generated using the "top-down" strategy for model selection

(Diggle et al. 2002). All models were compared against the "null" (intercept-only) model.

Methane production rates were predicted from a large set of variables that included site characteristics (depth, secchi depth, etc.), water chemistry, bulk sediment characteristics, porewater characteristics (DOC concentration and optical properties), and methanogen community abundance and composition. The regsubsets() command within the R package leaps

(Miller 2017) provided the output. The response and predictor variables were averaged for each site (to reach n=15). Only results for 1 predictor variable are presented, as using more predictor variables than this could lead to overfitting due to small sample size. Although many of the variables are auto correlated, VIF scores were not calculated for this test; the goal was to determine the best individual variables that predict CH4 production from many types of variables.

Mixing model for determination of autochthonous proportion of OM. The mixing model was generated using the MixSIAR package version 3.1.7 (Stock & Semmens 2013) in R. To determine the proportion of autochthonous (aquatic) and allochthonous (terrestrial) organic matter found in each sediment sample, a Bayesian mixing model was used. Terrestrial and aquatic end members were collected, and elemental N/C ratios and �15N were analyzed along with sediment samples using EA-IRMS. Terrestrial samples were taken from the surrounding watershed and included stream-bank soil, leaf litter, corn field soil, and corn stalk litter from tributary streams and a field along the perimeter of the reservoir. Aquatic end members came from epilimnion water samples from each of the 15 sites in late May (collected as described above). Water samples were filtered through 0.7 µm GF/F filters and dried at 55˚C before

23 packing into capsules for elemental and isotopic analysis. Epilimnion water samples from all sites were included in the determination of N/C ratios and �15N values for the aquatic end member to account for potential compositional variation of algae and cyanobacteria across the reservoir. Further, chlorophyll a (as determined by both in vivo fluorescence and extracted pigment absorbance) was high in all sites, indicating algal and/or cyanobacterial growth (see

Results for more detail).

2.3 Results

Site characteristics. The water column depth at the sampling sites ranged from 2.5 – 31 m (Table

2.1). The depth of the water column averaged 3.6 m in the riverine zone, 9.6 m in the transitional zone, and 20.3 m in the lacustrine zone. All sites were nutrient rich, with epilimnion total nitrogen (TN) concentrations ranging from 1210 – 2040 µg/L and total phosphorus (TP) concentrations ranging from 166 – 232 µg/L. The riverine zone had the highest average TN and

TP concentrations (1745 and 210 µg/L, respectively), and the lacustrine zone had the lowest average concentrations (1380 and 180 µg/L, respectively). Similarly, chlorophyll a, measured by in vivo fluorescence and from extracted pigment absorbance was highest in the riverine zone

(4128 RFUB and 47.6 µg/L, respectively). Secchi depth increased from the riverine zone

(average 0.33 m) to the lacustrine zone (average 0.78 m). The full suite of chemical and physical measurements at each site can be found in Appendix 1 (Supplemental Tables 2, 4, and 5).

Methane production among reservoir zones. Potential CH4 production rates normalized by slurry volume and by grams of organic matter were greater in riverine zone compared to the

-3 -1 transitional or lacustrine zones (µmol CH4 cm day F(2,12) = 12.88, � = 0.001, and CH4 g OM dry sediment-1 day-1 F(2,12) = 10.27, � = 0.003) (Figure 2.4). The mean areal methane

24 -2 -1 production rate for all sediment cores was 3.79 µmol CH4 m day (range: 0.79 - 8.56 µmol CH4 m-2 day-1. Methane production rates normalized by dry weight of the sediment slurry were not statistically different among the zones (F(2,12) = 2.037, � = 0.2).

A B C )

) 1

) − 1 2.0 a b b 3 a a a a b b − d 1

− d 1

d − 3

0.4 − 1 ● M −

● ●

m ● g

● O ●

c l ●

l g o

l o ●

m ● 1.5 ● o m µ ● ● ●

● m µ ● ● ●

( 0.3 µ

●

● ( ●

s 2 ● ● ● ( s

● e ● ●

● ● t s e ● ● ● ● ● ● t ● ●● ● ● a ● e ● t a ●

● R ● ● ● a ● ●

R ● ● ● ● ● n ● 1.0 ● ● R ●● ● ●

n

● o ● ● ● ●

● i ●

n ● o 0.2 t ● ● i ● t

● o

● c ● ● i

c ● ● t ● ● u ●● ● ● ● c u ● ● ●

d ● u d ● ● ●

● o ● ● ● ● r d o ● 1 ● ● ●

r ●

● ● o ●● P ● ●

● ● r ● ● P ● ● ● 0.5 e ● P

e

● n ● ● 0.1 ● e n ●● ● ● a ● ● n a ● h ●

● ● t a h ●

t ● ● ● e

● h t e ● ● ● ● M e M M riverine transitional lacustrine riverine transitional lacustrine riverine transitional lacustrine

Figure 2.4. Potential methane production rates from sediment slurries in each of the reservoir zones, normalized to sediment volume (A), sediment mass (B), and sediment organic matter (C). Each dot represents the potential methane production rate calculated from 1 sediment core by a sediment slurry assay. Different lowercase letters indicate significant differences between reservoir zones.

Sediment characteristics among reservoir zones. Sediment was the densest in the riverine zone, least dense in the lacustrine zone, and intermediate in the transitional zone (Table 2.2). The quantity of organic matter was greatest in the riverine zone for both the solid fraction of OM (g

OM cm-3) and the dissolved fraction of OM (DOC, g mL-1) (Figure 2.5), but the differences among zones for the solid fraction of OM were not significant (F(2,12) = 1.92, p = 0.19). The dissolved fraction of OM was greater in the riverine zone than either the transitional or lacustrine zones (F(2,12) = 5.43, p = 0.021).

Stable isotope and elemental composition data indicated that the source of OM varied significantly among reservoir zones in the solid fraction (Figure 2.6). The proportion of

25

g/L) 46 TP TP 172 210 232 244 227 287 210 185 182 216 201 203 220 211 199 216 194 220 166 167 194 185 215 188 195 248 177 254 210 231 199 213 180 191 B.D. µ (

g/L) TN 1210 1488 2000 2240 2040 2280 1730 1600 1410 1475 1400 1288 1490 1510 1490 1490 1400 1520 1380 1710 1330 1710 1750 1249 1250 4575 1320 1270 1250 1330 1745 1902 1438 1457 1380 1974 µ (

) a 1743 1994 2102 2176 2036 2170 1926 2068 2197 2133 2182 2174 1894 2076 2184 2091 1954 2140 1983 2146 1951 2105 1937 1910 1995 1968 1881 1898 1831 1891 1952 2102 2082 2123 1930 1986 CDOM (RFUB

in vivo) in vivo) ) a ( a 457 526 346 455 437 430 352 495 346 779 387 436 286 303 288 418 255 286 237 225 232 362 257 6958 8750 8054 2333 1068 3620 3035 1078 2827 4128 2368 1021 1877 (RFUB Chlorophyll a ------g/L) 4.8 3.2 1.8 1.7 5.4 1.8 0.8 1.9 0.5 0.5 3.0 0.9 74.2 63.9 31.6 18.6 30.4 16.4 29.2 32.0 16.0 19.9 47.6 12.2 21.8 24.3 µ ( Chlorophyll

pH 7.91 7.56 8.20 7.81 8.98 7.58 8.90 8.47 8.80 7.45 8.83 7.35 7.85 7.39 8.29 7.39 8.12 7.36 8.78 7.15 8.56 7.08 9.25 7.28 9.25 7.28 7.90 7.04 8.57 6.81 8.50 7.86 8.38 7.39 8.72 7.11 3.0 3.8 1.0 8.2 9.5 0.9 76.1 58.7 90.3 74.5 39.3 84.0 35.5 31.7 85.5 31.5 30.4 12.0 82.7 76.6 21.1 10.3 174.2 163.5 133.7 139.5 147.3 108.7 122.4 143.0 232.7 198.0 133.3 126.0 113.0 152.0 (% Saturation) Dissolved Oxygen (˚C) 27.3 26.7 22.9 16.3 24.1 19.1 22.7 21.9 27.0 17.2 26.7 17.2 22.8 17.1 20.4 16.8 19.8 16.9 21.4 16.7 21.7 15.6 27.7 14.7 26.5 11.0 19.5 10.4 21.6 10.3 24.3 21.0 23.3 17.0 23.1 13.1 Temperature Temperature 0.1 1.5 0.1 2.0 0.1 3.0 0.1 4.0 0.1 8.0 0.1 8.0 0.1 8.0 0.1 9.0 0.1 0.1 0.1 0.1 0.1 0.1 0.1 10.0 10.5 12.0 13.5 19.0 24.0 24.5 Sample Depth (m) Sample Location epilimnion epilimnion epilimnion epilimnion epilimnion epilimnion epilimnion epilimnion epilimnion epilimnion epilimnion epilimnion epilimnion epilimnion epilimnion epilimnion epilimnion epilimnion hypolimnion hypolimnion hypolimnion hypolimnion hypolimnion hypolimnion hypolimnion hypolimnion hypolimnion hypolimnion hypolimnion hypolimnion hypolimnion hypolimnion hypolimnion hypolimnion hypolimnion hypolimnion 0.22 0.38 0.33 0.40 0.52 0.68 0.73 0.53 0.80 0.60 0.53 1.04 0.87 1.00 0.67 0.33 0.65 0.78 Secchi Depth (m) Zone riverine riverine riverine riverine riverine lacustrine lacustrine lacustrine lacustrine lacustrine lacustrine Reservoir Reservoir lacustrine transitional transitional transitional transitional transitional transitional Physicaland waterchemical properties forcolumn of each sampling the sites.

2.5 3.0 4.0 5.0 9.0 9.0 9.0 3.6 9.6 (m) 10.0 11.0 12.0 14.0 16.5 22.5 26.0 31.0 20.3 1. Site DepthSite 2. T1 T2 T3 T4 T5 L1 L2 L3 L4 L5 L6 R1 R2 R3 R4 Site Table RFUB are blank-subtracted raw fluorescence units from a submersible fluorometer. submersible a from units raw fluorescence RFUB blank-subtracted are Cells with no values were not measured for the given parameter. given for the measured not were no values with Cells detection. below were B.D. values a

26 autochthonous solid-fraction OM (calculated from the mixing model) was lowest in the riverine zone, intermediate in the transitional zone, and highest in the lacustrine zone (F(2,12) = 9.23, p =

0.004) (Figure 2.6). Optical properties of the dissolved OM showed some evidence of differences in DOM source across reservoir zones (Figure 2.7). Fluorescence index (FI) values are usually between 1.2 and 1.8 in natural waters (Hansen et al. 2016). Higher FI values indicate DOM derived from microbial sources (e.g. bacteria and algae) and lower values indicating DOM derived from terrestrial OM (McKnight et al. 2001; Cory et al. 2010). FI values were higher in the riverine zone than the lacustrine or transitional zones (F(2,12) = 14.05, p = 0.0007) (Figure

2.7). The biological index (BIX) can indicate autotrophic productivity: a BIX of 0.6-0.7 indicates a low autochthonous component, 0.7-0.8 an intermediate autochthonous component, 0.8-1 a strong autochthonous component, and values >1 indicate aquatic bacterial origin (Huguet et al.

2009). BIX did not vary among reservoir zones (F(2,12) = 1.41, p = 0.28), and site averages ranged from 0.594 to 0.713 (Table 2.2).

A B a a a a 0.06 b b ● 50 ● ● ●

● ● ● )

0.05 ● 1 ● ● 40 − 3 ●

− ● L ● ● ● ● ● ● m ● g ● ● ● c ● ● ● ● ● m ● ● M (

● ● ● O ● ●

● ● ● C

g 30 0.04 ● O ● ● ● ● ● ● D ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● ● 20 ● ● ● ● ● ●● 0.03 ● ●

●● ● riverine transitional lacustrine riverine transitional lacustrine

Figure 2.5. Comparison of the quantity of OM among reservoir zones. Panel A represents the amount of OM found in the bulk sediment, normalized to sediment volume. Panel B represents the dissolved fraction of OM found in the sediment porewater. One riverine value with an exceptionally high DOC concentration was eliminated from the plot for readability. Different lowercase letters indicate significant differences between reservoir zones.

27

A B a b c

M 0.60 O ● ● ● ● s

u ● ● ● o ● n ● o ● ● ● ● h

t ● ● ●

h ●● c ● o

t 0.58 ●

u ●

a ● ● ● f ● ● o ● ● ●

n ● ● ● ● o ●

i ● ● ● ● t ● ● ● r

o ●

p ●

o 0.56 ● r P ● ● ● ● riverine transitional lacustrine

Figure 2.6. Plot of �15N vs. N/C elemental ratios of terrestrial and aquatic OM sources, and the sediment mixtures for each core across the three reservoir zones (A). The center dot for the two sources represents the mean values (n = 13 for terrestrial sources, n = 16 for aquatic sources), and error bars represent the standard deviation. Panel B depicts the proportion of autochthonous (aquatic) OM found in each core grouped by reservoir zones calculated from the stable isotope mixing model. Different lowercase letters indicate significant differences among reservoir zones.

A B a b b 0.80 a a a 1.80

● ●

● )

) 0.75

x ● ●

● x ●

e 1.75 ● e d ● d n ● ● i ● n

i

● e l c a ● ● ● ● ● ● ● n ● ● c 0.70 ● ● i ● ●

e ●

1.70 ● g ● c ●● ● ● o ● s ● l ● ● ●● e o ● ● i r ● ● b o ● ● ● ● ● ● ● u

l ● ( ●

f ●● ● ● 0.65 ● ● ● ● ● ● ● ● ● X

( 1.65 I ● ● ● ● ● I ● B ● F ● ● ● ● ● ● ● ● ● ●● ● ● 0.60 ● ● 1.60 ● ● ● ● ● ● riverine transitional lacustrine riverine transitional lacustrine

Figure 2.7. Optical indices of dissolved organic matter. Higher FI values (A) indicate DOM from microbial sources, and higher BIX values (B) indicate a stronger autochthonous component of the OM. Different lowercase letters indicate significant differences between reservoir zones.

28 254 0.027 0.042 0.055 SUVA 0.029 (0.003) 0.016 (0.013) 0.030 (0.009) 0.032 (0.009) 0.033 (0.000) 0.038 (0.001) 0.044 (0.003) 0.050 (0.006) 0.046 (0.001) 0.061 (0.024) 0.065 (0.010) 0.050 (0.017) 0.035 (0.004) 0.060 (0.011) 0.069 (0.015)

0.681 0.640 0.664 index) 0.649 (0.004) 0.694 (0.071) 0.694 (0.024) 0.687 (0.007) 0.651 (0.008) 0.638 (0.014) 0.628 (0.021) 0.666 (0.004) 0.614 (0.021) 0.698 (0.004) 0.687 (0.005) 0.594 (0.010) 0.612 (0.010) 0.700 (0.002) 0.713 (0.045) BIX (biological 1.73 1.66 1.64 index) 1.73 (0.007) 1.73 (0.038) 1.72 (0.016) 1.74 (0.024) 1.69 (0.015) 1.66 (0.014) 1.65 (0.026) 1.66 (0.008) 1.64 (0.011) 1.69 (0.012) 1.66 (0.041) 1.60 (0.013) 1.61 (0.018) 1.65 (0.006) 1.66 (0.037)

FI (fluorescence 57.32 22.80 22.52

28.7 (2.2) 39.3 (5.5) 26.8 (9.3) 23.9 (1.5) 24.6 (2.2) 21.5 (2.8) 24.8 (2.0) 19.3 (1.4) 22.9 (3.6) 19.2 (0.5) 20.9 (1.8) 26.7 (7.0) 26.1 (1.6) 17.9 (4.2) DOC (mg/L) 134.5 (152.2)

0.564 0.574 0.587 Proportion Proportion 0.557 (0.005) 0.555 (0.009) 0.571 (0.002) 0.573 (0.003) 0.569 (0.005) 0.573 (0.005) 0.577 (0.010) 0.562 (0.004) 0.586 (0.008) 0.591 (0.002) 0.583 (0.002) 0.597 (0.003) 0.584 (0.013) 0.575 (0.012) 0.590 (0.006) autochthonous

1.45 1.23 1.16 (g/mL) Density Density 1.66 (0.26) 1.36 (0.08) 1.45 (0.06) 1.35 (0.02) 1.20 (0.03) 1.20 (0.03) 1.27 (0.01) 1.21 (0.01) 1.26 (0.01) 1.29 (0.01) 1.23 (0.03) 1.08 (0.02) 1.13 (0.03) 1.17 (0.01) 1.05 (0.02)

6.52 9.03 11.02 weight) 6.02 (0.45) 6.93 (0.44) 5.23 (0.99) 7.91 (0.16) 9.02 (0.68) 9.32 (0.12) 8.41 (0.32) 7.73 (0.26) 9.34 (0.11) 9.62 (0.38) OM % (dry 10.67 (0.58) 10.85 (0.21) 11.49 (0.77) 11.73 (1.12) 12.90 (0.26)

N/C 0.107 0.139 0.137 0.104 (0.007) 0.102 (0.020) 0.114 (0.011) 0.108 (0.010) 0.127 (0.018) 0.134 (0.018) 0.140 (0.027) 0.145 (0.024) 0.148 (0.019) 0.129 (0.018) 0.135 (0.022) 0.127 (0.004) 0.135 (0.012) 0.151 (0.008) 0.144 (0.009) sites. Values are are Values sites.

C/N 9.47 7.34 7.38 8.9 (0.8) 9.3 (0.8) 8.0 (1.1) 7.5 (0.9) 7.3 (1.3) 7.0 (1.1) 6.8 (0.9) 7.8 (1.0) 7.5 (1.1) 7.9 (0.2) 7.5 (0.6) 6.6 (0.4) 7.0 (0.4) 9.7 (0.7) 10.0 (1.7)

N 15 5.48 5.92 6.21 5.30 (0.05) 5.25 (0.34) 5.68 (0.10) 5.68 (0.05) 5.75 (0.25) 5.86 (0.22) 6.01 (0.40) 5.73 (0.10) 6.26 (0.07) 6.25 (0.18) 6.12 (0.19) 6.38 (0.02) 6.14 (0.34) 6.06 (0.30) 6.33 (0.13) of suite comprehensive the For % N 0.221 0.289 0.366 0.191 (0.010) 0.224 (0.023) 0.216 (0.031) 0.254 (0.005) 0.286 (0.019) 0.287 (0.004) 0.269 (0.009) 0.344 (0.015) 0.260 (0.000) 0.293 (0.009) 0.336 (0.012) 0.311 (0.009) 0.366 (0.010) 0.438 (0.024) 0.456 (0.005) org C 13 -25.82 -25.84 -25.86 -26.19 (1.10) -25.44 (0.28) -25.84 (0.52) -25.44 (0.47) -25.79 (0.51) -25.72 (0.26) -25.82 (0.41) -26.31 (0.31) -25.32 (0.49) -25.38 (0.48) -26.43 (0.46) -25.94 (0.74) -26.16 (1.14) -25.78 (1.25) -25.95 (0.41) org

2.08 2.12 2.68 % C

1.84 (0.18) 2.22 (0.18) 1.90 (0.26) 2.37 (0.15) 2.27 (0.22) 2.16 (0.25) 1.98 (0.40) 2.41 (0.28) 1.78 (0.24) 2.29 (0.24) 2.51 (0.30) 2.46 (0.03) 2.72 (0.19) 2.90 (0.28) 3.17 (0.19) ) -3 Averages for each zone are in bold. bold. in are zone each for Averages

cm -1 1.23 0.56 0.62 SD). 1.36 (0.30) 1.31 (0.11) 1.16 (0.02) 0.97 (0.16) 0.55 (0.01) 0.36 (0.18) 0.67 (0.16) 0.23 (0.06) 0.86 (0.10) 0.72 (0.06) 0.26 (0.09) 0.65 (0.14) 0.66 (0.22) 0.54 (0.19) ( 1.10 (0.05) production rates rates production mol day mol

4 µ ( CH

mean Bulk sediment and porewater characteristics for each of the 15

2. Zone riverine riverine riverine riverine riverine 2. Reservoir Reservoir lacustrine lacustrine lacustrine lacustrine lacustrine lacustrine lacustrine transitional transitional transitional transitional transitional transitional T1 T2 T3 T4 T5 L1 L2 L3 L4 L5 L6 R1 R2 R3 R4 Table as reported measurements,see Appendix 1. Site

29 Methanogen communities among reservoir zones. Abundance of the mcrA gene and a region of the 16S rRNA gene targeting archaea were determined using qPCR. Copies of mcrA ranged from

1.4 x 101 to 3.8 x 104 ng DNA-1 g sediment-1 (wet weight), with a mean of 3.7 x 103. There were no differences in copy number for either the mcrA gene (F(2,11) = 0.76, � = 0.5) or the archaeal

16S rRNA gene (F(2,11) = 1.81, � = 0.2) among reservoir zones (Figure 2.8). Alpha diversity was reported using the Shannon diversity index for a subset of the samples (Figure 2.9). Several samples had low coverage and were excluded from the final results. From 16S rRNA gene sequencing, 5708 archaeal operational taxonomic units (OTUs) were observed across 44 sediment cores. Of the 5708 archaeal OTUs, 939 were assigned to methanogens (16.5%).

A B

40000 a a a 15000 a a a

1 ●

1 ● − − t ●

t ● n n e e m i m i d

d 30000 e e ● ● s s

10000 ● g g

1

● 1

− ● − A A N N 20000 D D

●

g ● g n n ●

● ● ● A C ● r 5000 c R m A

10000 ● ● s s ● ● ●● ● e e ●

● i i ● ● ● ● ● p p ● ● ● ● ● ● ●●● ● ● o o ● ● ● ● ● ● c c ● ● ● ● ● ● ● ● ● ● ● ● ● ●● ● ● ● ● ● ● ● ● ● ● ●●● ●● ● ● ● 0 ● ● 0 ● ● ● ● ● ● riverine transitional lacustrine riverine transitional lacustrine

Figure 2.8. Abundance of 16S archaeal gene (A) and mcrA gene (total methanogens) (B) among the three zones. Different lowercase letters indicate significant differences between reservoir zones.

Sequences affiliated with four of the six known orders of methanogens were recovered from the

Harsha sediments: Methanomicrobiales (606 OTUs), Methanosarcinales (218 OTUs),

Methanobacteriales (105 OTUs), and Methanocellales (9 OTUs). On average, methanogens represented 34.8% of total archaea when calculated from 16S rRNA sequence data, and 33.7% of

30 A B

● ● ● 7.3 ● ●

●

● ) ) ● ● a a

i ● ● e r 7.2 ● 3.9 ● ● ● a e

t ● ● h

c ●

● c ● ●

a ● r B A 7.1 ● ● ● ● (

● (

y y t t i i 3.6 s ● ● s r r 7.0 ● ● e e ● ● v v i ● i ● d ● d

● ● n ● n o 6.9 ● o 3.3 ● n n ● n n ● ● a a ● ● ● h ● h 6.8 ● ● S S ●

● 3.0 ● riverine transitional lacustrine riverine transitional lacustrine

Figure 2.9. Alpha diversity of bacterial and archaeal 16S rRNA sequences. Reported values are only a subset of the samples.

Methanomicrobiales 53.3 77 76.2

Methanosarcinales 34.1 16.1 16.9

Methanobacteriales 15.6 9.6 7.4

Methanocellales 0.9 0.3 0.3 riverine lacustrine transitional Figure 2.10. Heat maps showing relative abundance of methanogen orders. Note that values are normalized to total number of sequences of all methanogens (not to total archaea), and that percentages are averages of all samples within a group.

total archaea when calculated from qPCR data. Representative OTUs from the order

Methanomicrobiales accounted 53.3%, 77% and 76.2% of total methanogen sequences from the riverine, transitional, and lacustrine zones, respectively (Figure 2.10). The order

31

Methanosarcinales comprised 34.1%, 16.1% and 16.9% of riverine, transitional and lacustrine

methanogens. The 15 most abundant methanogen genera are shown in Figure 2.11. The two most

abundant methanogen genera in all three zones were Methanoregula and Methanosaeta.

Methanobacterium was relatively abundant in all three zones, whereas Methanosarcina and

Methanospirillum composed a greater proportion of methanogen genera in the riverine than in

other zones.

Methanomicrobiales; Methanoregula 47.7 69.5 66.9 Methanosarcinales; Methanosaeta 19.6 15.2 16.6 Methanobacteriales; Methanobacterium 15.1 9.6 7.3 Methanomicrobiales; Methanolinea 2.8 6.4 4.8 Methanosarcinales; Methanosarcina 15.4 0.5 0.4 Methanomicrobiales; Methanoline 1.4 1.7 3.6 Methanomicrobiales; Methanospirillum 4.3 0.8 1.3 Methanocellales; Rice_Cluster_I 0.9 0.1 0.3 Methanomicrobiales; SMS−sludge−7_unclassifie 0.3 0.1 0.5 Methanosarcinales; Candidatus_Methanoperedens 0.4 0.8 0 Methanomicrobiales; Methanomicrobiales_unclassifie 0 0.4 0.4 Methanosarcinales; Methanolobus 0.8 0 0 Methanomicrobiales; Methanomicrobiales_unclassified 0.4 0.1 0.2 Methanosarcinales; Methanomethylovorans 0.6 0 0 Methanobacteriales; Methanobrevibacter 0.4 0 0 riverine lacustrine transitional

Figure 2.11. Heat maps showing relative abundance of methanogen genera. Row labels show order and genus (order; genus). Note that values are normalized to total number of sequences of all methanogens (not to total archaea), and that percentages are averages of all samples within a group.

32 13 Methane production pathway. � C of CH4 ranged from -56.7 to -48.0‰ (Figure 2.12A).

Apparent fractionation factors (aC) were consistent with acetoclastic methanogenesis across all zones (range 1.040 – 1.050). The highest aC was observed in the lacustrine zones while the riverine zone had the lowest aC (Figure 2.12B), suggesting an increased signal from hydrogenotrophic methanogenesis in the lacustrine zone.

A B 1.0525 a a a a a a )

C α −48 ● 1.0500 ● (

●

● r o t

● c ● a

● f

4 1.0475

n ● ● H o

i ● t ● C −51 ●

a ● ● ● n C ● ● ● ● ● ● o ● 13 ●● ● i ● t δ ● ● 1.0450 ● ● ● ● ● c ● ● ● ● ● ●● a ● ● r f

●

t ● ●

−54 n e

● r ● a 1.0425 ● ●

● ● p ●

p ● ● ●

● A

● ● ● ● −57 1.0400 riverine transitional lacustrine riverine transitional lacustrine

Figure 2.12. Carbon isotope signature in CH4 (A) and the apparent fractionation factors (B) across reservoir zones. The blue horizontal line (B) indicates the cutoff between hydrogenotrophic (above) and acetoclastic (below) methanogenesis. Methane production pathway was estimated at 10 of the 15 sampling sites (n = 30 sediment cores). Different lowercase letters indicate significant differences between reservoir zones.

Influence of OM source and quantity on methane production rates. Results of AIC hypothesis testing to discriminate the importance of OM source, quantity, and the combination of source and quantity indicated that the best model corresponds to the hypothesis that methane production rates are best explained by the combination of OM source and quantity (H3), with a model probability of 0.989 (Table 2.3). The best predictor variables for the “OM source” hypothesis

(H1) were the proportion of autochthonous OM and BIX (biological index calculated from DOM fluorescence). DOC concentration, g OM per sediment slurry, and the interaction between them

33 created the best model for the “OM quantity” hypothesis. All of these variables comprised the best model for the “OM source + quantity” hypothesis. All variables, except for the proportion of autochthonous OM, had positive correlations with methane production rates, regardless of the model. Note that ranks, model probabilities, and evidence ratios (Ei,j) are relative to the set of models and dependent on the data used to create the models. Despite the quantity model having a lower model probability and rank than the source model, the marginal R2 was higher, indicating more explained variance. The combined source and quantity model had the highest marginal R2 of the three models.

-3 Table 2.3. Results of model hypothesis testing. The response variable was methane production rates (µmol CH4 cm day-1). Arrows next to predictor variables indicate a positive (↑) or negative (↓) relationship with the response variable. K is the number of estimated parameters, AICc is a second order AIC score used to account for the number of estimated parameters. Di values are AICc differences (compared to the best model). Model probabilities (wi), otherwise called Akaike weights, are estimates of the probability of the model being the best K-L (Kullback-Leibler) model, given the data and set of competing models (see Anderson 2008). The evidence ratios (Ei,j) measure the strength of evidence of the hypotheses and represent the relative likelihood of hypothesis i vs. j. In this case they 2 represent the likelihood of the best model (H3) relative to the other models. Marginal R describes the proportion of variance explained by fixed factors alone, and the conditional R2 describes the proportion of variance explained by both fixed and random effects.

H Hypothesis Model

H1 source OM autochthonous OM (↓) + BIX (↑) -1 H2 quantity OM log(DOC) (↑) * g OM slurry (↑) -1 H3 source OM + quantity OM autochthonous OM + log(DOC) * g OM slurry + BIX

2 2 H K log(L ) AIC AICc Rank ∆i Model probability w i E i,j Marginal R Conditional R

H1 5 13.53 -17.06 -15.56 2 9.6 0.0080 123.8 0.33 0.85

H2 6 13.76 -15.52 -13.37 3 11.8 0.0027 370.6 0.48 0.87

H3 8 22.55 -29.09 -25.20 1 0.0 0.9893 0.70 0.89

34 Variables that best predict methane production rates. Individual variables that best predicted

-3 -1 CH4 production rates (µmol CH4 cm day ) in univariate linear regression analysis included the fluorescence index (FI) of DOM, 15N of the sediment, secchi depth (m), and C/N ratios (Table

2.4). For a complete list of variables that were included, see Appendix 1 (Supplemental Table 6).

-3 -1 Table 2.4. Best fifteen predictors of methane production rates (µmol CH4 cm day ) from univariate linear regression analysis. Variables include physical and chemical characteristics of sampling sites, sediment and porewater measurements, and methanogen abundance and community composition.

Rank Predictor variable Adjusted R2 Residual sum of squares (RSS) 1 FI (fluorescence index) of DOC 0.741 0.389 2 15N (sediment) 0.741 0.389 3 Secchi depth (m) 0.734 0.399 4 C/N ratio (sediment) 0.734 0.399 5 N/C ratio (sediment) 0.721 0.420 6 Slurry dry weight (g sediment) 0.666 0.502 7 Proportion autochthonous OM in sediment 0.643 0.537 8 Percent of order Methanosarcinales of total methanogens 0.572 0.642

9 NO2- (nitrite) in epilimnion 0.572 0.642 10 Mass OM per slurry (g) 0.482 0.778 11 % water (sediment) 0.479 0.783 12 log(DOC) 0.469 0.797 13 Percent of genus Methanosarcina of total methanogens 0.450 0.826 14 Sediment bulk density (g/mL) 0.387 0.920

15 SUVA254 0.385 0.923

2.4 Discussion

Studies on sediment methane production in lakes and reservoirs generally are based upon sediment sampled from the deepest portion of the system (e.g. Kelley & Chynoweth 1981;

Schulz & Conrad 1995; Schwarz et al. 2008; West et al. 2012), and do not include sediment sampled from a range of sites with different properties. By including spatial variability, we were able to identify important within-reservoir trends in methane production rates, sediment

35 characteristics, and methanogen composition. Categorizing sites into reservoir zones aided in conceptualizing the differences in function across the reservoir. Further, this approach provides a more generalizable framework for comparisons with other reservoirs.

Spatial variability of methane production. Sediment CH4 production rates, expressed on an areal or volumetric basis, were higher in the riverine than transitional or lacustrine zones (Figure

2.4A), which is consistent with previous reports of higher CH4 emission rates in the riverine zone compared to other areas of this reservoir (Beaulieu et al. 2014, 2016). Published areal surface

CH4 emission rates from the Harsha Lake riverine zone (Beaulieu et al. 2016) and our measured riverine zone sediment CH4 production rates from this study were in remarkable agreement (33.0

-2 -1 -2 -1 mg CH4 m h and 30.8 mg CH4 m h , respectively). However, CH4 surface emission rates in the riverine zone in Harsha can be six-fold (Beaulieu et al. 2016) to 1-2 orders of magnitude

(Beaulieu et al. 2014) higher than other areas of the reservoir, while sediment CH4 production rates were only two-fold higher in the riverine zone compared to other zones. This could be due to methane oxidation that is likely to occur in lacustrine and transitional zones where the water column is deeper and more strongly stratified, resulting in a larger difference between CH4 production and emission in those zones.

When expressed per gram dry mass, sediment CH4 production rates did not differ among the three reservoir zones. This discrepancy is due to the density of the sediment, which was 18-25% greater in the riverine than other reservoir zones. Therefore, production rates measured in the incubation vessels, which were filled with equal volumes of sediment slurry, scaled with the total mass of sediment added to the vessel, rather than the volume. However, it is likely that not all sediment components support CH4 production. While inorganic fractions of the sediment may provide surface area for microbiota attachment, the organic fraction is the source fueling

36 methanogenesis. When normalized to sediment OM, the riverine zone had higher CH4 production rates than the other reservoir zones (Figure 2.4C), possibly because the sediment OM was utilized more effectively by the microbial community, either due to the quality of the OM, the composition of microbial communities involved in decomposition, or the functioning of these communities.

Spatial variability of sediment characteristics and sediment OM. Annual records of algal and cyanobacterial biomass indicate blooms are more intense in the riverine section of Harsha Lake

(historical data collected and maintained by USEPA). Thus, we predicted a higher proportion of autochthonous OM in the riverine zone. Consistent with annual records, we observed the highest chlorophyll levels in the riverine zone during our sampling (Table 2.1). Contrary to our prediction, however, the proportion of autochthonous OM in the sediment was lowest in the riverine zone and highest in the lacustrine zone (Figure 2.6). This likely reflects deposition in the riverine zone of sediment from the surrounding watershed. In fact, sediment traps that were stationed at 4 of the 15 sampling sites for 6 weeks indicated that riverine sedimentation rates can be ~10-35% higher than in other areas of the reservoir (see Appendix 3).

Organic matter and organic carbon content of sediments is often expressed as a percentage; however, due to the density variation across Harsha Lake sediments, the amount of OM per volume is a more accurate representation of the OM content. While percentage of OM was lower in the riverine zone than the other zones, the grams of OM per cm-3 of sediment was higher, potentially resulting in more substrate availability for methanogenesis, per unit volume. The composition of OM can be described using elemental C/N (or N/C) ratios; C/N ratios can indicate OM source, with higher C/N ratios resulting from terrestrial OM and lower C/N ratios resulting from algal and microbial sources. C/N ratios were positively correlated to CH4

37 production rates, which differs from other results from lake sediments that indicate a negative correlation between CH4 production rates and C/N ratios (Duc et al. 2010). This indicates that in

Harsha Lake, there is a link between terrestrial OM (higher C/N ratios) and CH4 production.

Spatial variability of methanogen communities. Similar to results of other studies in freshwater systems, we found no correlation between methanogen abundance and CH4 production rates, likely because methanogen activity, rather than abundance, determines rates of methanogenesis

(West et al. 2012; Chaudhary & Blaser 2017). The dominant methanogen genera identified in

Harsha Lake were comparable to methanogen communities reported in other freshwater ecosystems (Borrel et al. 2011).

While the dominant methanogenic genera were similar across zones, the functional diversity of the riverine zone was higher compared to the other zones due to the presence of generalist acetate utilizing methanogens that were observed only in low proportion in other reservoir zones.

The two main biochemical pathways of methane production (acetoclastic and hydrogenotrophic) are phylogenetically distributed within the 6 known orders of methanogens (Liu & Whitman

2008; Sakai et al. 2008). One order, Methanosarcinales, is the only order that has representatives capable of acetoclastic (acetate-utilizing) methanogenesis (Liu & Whitman 2008). One of the biggest differences in methanogen communities among reservoir zones was the abundance of

Methanosarcina, a genus of the Methanosarcinales order, which composed 15% of methanogens at the riverine site, but less than 1% of transitional and lacustrine methanogens. Methanosarcina can utilize acetate, H2, and methylated compounds via several metabolic pathways to produce

CH4 (Liu & Whitman 2008). These methanogens may be more abundant in the riverine zone due to the more diverse mix of inorganic substrates, autochthonous OM, and allochthonous OM

38 relative to the lacustrine zone with more homogenous substrate. Anaerobic degradation of organic matter is a multistep process that relies upon fermenting bacteria and anaerobic oxidizing bacteria to provide substrates for methanogens (Angelidaki et al. 2011). These methanogens were recovered from the zone where we observe higher proportions of allochthonous OM relative to autochthonous OM. Thus, variation in methanogen taxa and methane production rates could be dictated by substrate availability. Specifically, Methanosarcina is the only genus with representatives capable of utilizing methylated compounds; the OM in the riverine zone may be conducive to the production of these methylated compounds. Overall, we saw higher methanogen functional diversity in the zone with the highest methane production rates and high

OM diversity. One hypothesis for future testing is that the type of organic matter may be more diverse and complex in riverine sites due to the mixture of algal and terrestrial inputs; this OM is subsequently broken down into substrates that metabolically diverse Methanosarcina spp. can convert to methane. A second hypothesis is that the OM in the riverine zone is being degraded more completely in the presence of oxygen before it reaches the anoxic sediment layers, resulting in high substrate availability for methanogenesis. Sites in this zone do not stratify, and in fact CH4 production rates were higher in sites with the highest hypolimnion oxygen.

Methane production pathway. Acetoclastic and hydrogenotrophic methanogenesis are considered the dominant methanogenic pathways in freshwater environments, where theoretically 66% of methanogenesis should be acetoclastic and 33% should be hydrogenotrophic, based upon stoichiometry (Conrad 1999). Though our method for estimating the methanogenesis pathway only provides a coarse estimate of the production pathway, our results indicated that all or most methanogenesis in the Harsha Lake sediments was acetoclastic, despite our observation of a hydrogenotrophic methanogen being the most abundant methanogen

39 in all zones. It could be that though hydrogenotrophic methanogens comprise most of the methanogenic communities, they are not the most active.

Evaluating the influence of OM source and quantity on methane production rates.

For modeling purposes, we elected to use rates normalized by volume, as these rates are easily scalable to area relfect variations in sediment density. The organic matter source model had a slightly higher model probability than the organic matter quantity model, but the amount of variance explained by the OM quantity model was greater than that of the OM source model.

Regardless, the model that combined both source and quantity had a much higher model probability than either of the models that addressed quantity or source individually. The predictor variables that were included in the model with source and quantity included the proportion of autochthonous OM, BIX (biological index) of DOM, DOC concentration and the grams of OM per slurry. The inclusion of DOC concentration and the BIX optical property in the model indicates that the dissolved fraction of organic matter was important for methane production in terms of both quantity and source. BIX, an indicator of autochthonous DOM (Huguet et al. 2009;

Hansen et al. 2016), had a positive relationship with methane production, suggesting that algal- derived OM fuels methanogenesis in Harsha Lake. However, the proportion of autochthonous

OM in the sediment was negatively correlated to production rates; the riverine sites had lower proportions of autochthonous OM and the highest methane production rates. There are several potential explanations for these seemingly contradictory results:

1. Algal material is more readily and rapidly degraded and therefore less is incorporated

into the bulk sediment. In the riverine zone, the labile algal material fuels high methane

40 production rates and the high sedimentation rates result in sediments with a higher

proportion of terrestrial OM.

2. There is more algal-derived OM in the riverine sediments per volume, and describing it

as a proportion may not accurately reflect the composition. For example, if we take

density of the sediment into account and describe the value per volume based on the

isotope mixing model, there is slightly more autochthonous OM in riverine sediments

(differences not significant). This OM fraction maybe be ultimately converted to CH4.

14 Without more direct evidence, such as C-CH4 which would discriminate between modern autochthonous C and older terrestrial C sources, we cannot definitively say which source of OM is fueling methanogenesis with our data set. Regardless, results from model testing indicate that both quantity and source of organic matter affect methane production.

Variables that best predict methane production rates. The individual best predictors of methane production rates included FI (fluorescence index), 15N, secchi depth, and C/N ratio. Three of these first four variables can describe the source of OM. Fluorescence index is another optical property of DOM that indicates the relative contributions of terrestrial and microbial OM to

DOC. One predictor variable that stands out for potential as a proxy for CH4 production rates is secchi depth. This is quickly, easily and inexpensively measured; thus, a robust cross system approach should be developed to validate if it can be established as a reliable proxy.

The results from this analysis should be interpreted with caution, because many of the variables co-vary. For example, 15N is highly co-correlated with other variables including freshness index and N/C ratios (Table 2.4). Further, this analysis does not show the general trends in site

41 characteristics, sediment composition and microbial community composition that we observed.

Future directions should incorporate co-variation of variables.

Conclusions and summary. The riverine zone in Harsha Lake is functionally different than the lacustrine and transitional zones and is characterized by dense sediments with low percent organic matter that is heavily influenced by terrestrial inputs and high sedimentation rates.

Methane production rates in this zone are higher than the other reservoir zones. While methanogens do not vary in abundance across the reservoir, the riverine zone contains a higher proportion of the metabolically diverse genus Methanosarcina that may be present in higher proportions due to the nature of the organic matter present at these sites. We found that both the quantity and source of OM together explain methane production rates better than individually.

Several lines of evidence indicate that algal derived organic matter likely plays a role in the production of methane, particularly when considering DOC optical indices, but that terrestrial organic matter also may contribute. Studies that directly measure the source of carbon that is converted to CH4, such as incorporating radiocarbon measurements to evaluate the age of the C source, are needed to confirm if one source contributes more to methanogenesis, and how that might vary spatially in a reservoir.

42 References

Albertsen M., Karst S.M., Ziegler A.S., Kirkegaard R.H., Nielsen P.H. (2015) Back to Basics - The Influence of DNA Extraction and Primer Choice on Phylogenetic Analysis of Activated Sludge Communities. PLoS ONE. 10(7): e0132783. Anderson D.R. (2008). Model Based Inference in the Life Sciences: A Primer on Evidence. Springer, New York, NY. Anderson N.J., Dietz R.D., Engstrom D.R. (2013) Land-use change, not climate, controls organic carbon burial in lakes. Proc. R. Soc. London, B, DOI: 10.1098/rspb.2013.1278. Anderson N. J., Bennion H., Lotter A.F. (2014) Lake eutrophication and its implications for organic carbon sequestration in Europe. Global Change Biol. 20(9): 2741–2751. Angelidaki I., Karakashev D., Batstone D.J., Plugge C.M., Stams A.J. (2011) Biomethanation and its potential. In Methods in methane metabolism. Part A: methanogenesis. Ragsdale S.W., editor; Rosenzweig A.C., editor. Methods in Enzymology. Vol. 494: 327–351. London, UK. Apprill, A., McNally S., Parsons R. & Weber L. (2015) Minor revision to V4 region SSU rRNA 806R gene primer greatly increases detection of SAR11 bacterioplankton. Aquat. Microb. Ecol. 75: 129–137. doi:10.3354/ame01753. Barros N., Cole J.J., Tranvik L.J., Prairie Y.T., Bastivken D., Huszar V.L.M. et al. (2011) Carbon emission from hydroelectric reservoirs linked to reservoir age and latitude. Nat. Geosci. 4: 593-596. Bastviken D., Tranvik L.J., Downing J.A., Crill P.M. & Enrich-Prast A. (2011) Freshwater Methane Emissions Offset the Continental Carbon Sink. Science 331: 50. Beaulieu J.J., Smolenski R.L., Nietch C.T., Townsend-Small A. & Elovitz M.S. (2014) High Methane Emissions from a Midlatitude Reservoir Draining an Agricultural Watershed. Environ. Sci. Technol. 48: 11100-11108. Beaulieu J.J., McManus M.G., Nietch C.T. (2016) Estimates of reservoir methane emissions based on a spatially balanced probabilistic-survey. Limnol. Oceanogr. 61: S27-S40. Bianchi T.S. & Canuel E.A. (2011) Chemical Biomarkers in Aquatic Ecosystems. Princeton University Press, Princeton, NJ. Borrel G., Jézéquel D., Biderre-Petit C., Morel-Desrosiers N., Morel J-P., Peyret P., Fonty G., Lehours A-C. (2011) Production and consumption of methane in freshwater lake ecosystems. Research in Microbiologoy 162: 832–847. Caporaso J.G., Lauber C.L., Walters W.A., Berg-Lyons D., Huntley J., Fierer N., Owens S.M., Betley J., Fraser L., Bauer M., Gormely M., Gilbert J.A., Smith J., Knight R. (2012) Ultra-high- throughput microbial community analysis on the Illumina HiSeq and MiSeq platforms. The ISME Journal 6: 1621–1624.

43 Chaudhary P.P., Rulík M., Blaser M. (2017) Is the methanogenic community reflecting the methane emissions of river sediments?-comparison of two study sites. Microbiology Open 7:947–9. Ciais P., Sabine C., Bala G., Bopp L., Brovkin V., Canadell J., Chhabra A., DeFries R., Galloway J., Heimann M., Jones C., Le Quéré C., Myneni R.B., Piao S., and Thornton P., (2013) Carbon and Other Biogeochemical Cycles. In: Climate Change 2013: The Physical Science Basis. Contribution of Working Group I to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change [Stocker, T.F., D. Qin, G.-K. Plattner, M. Tignor, S.K. Allen, J. Boschung, A. Nauels, Y. Xia, V. Bex and P.M. Midgley (eds.)]. Cambridge University Press, Cambridge, United Kingdom and New York, NY, USA.

Clow D. W., Stackpoole S. M., Verdin K. L., Butman D. E., Zhu Z., Krabbenhoft D. P., & Striegl R. G. (2015) Organic carbon burial in lakes and reservoirs of the conterminous United States. Environ. Sci. Technol. 49: 7614–7622.

Conrad R. (1996) Soil Microorganisms as Controllers of Atmospheric Trace Gases (H2, CO, CH4, OCS, N2O and NO). Microbiological Reviews 60: 609–640. Conrad R. (1999) Contribution of hydrogen to methane production and control of hydrogen concentrations in methanogenic soils and sediments. FEMS Microbiology Ecology 28: 193–202. Conrad R. (2005) Quantification of methanogenic pathways using stable carbon isotopic signatures: a review and a proposal. Org. Geochem. 36: 739–752. Cory R. M., Mcneill K., Cotner J.P., Amado A., Purcell J.M., and Marshall A.G. (2010) Singlet oxygen in the coupled photochemical and biochemical oxidation of dissolved organic matter. Environ. Sci. Technol. 44: 3683–3689.

Deemer B.R., Harrison J.A., Li S., Beaulieu J.J., DelSontro T., Barros N., Bezerra-Neto J.F., Powers S.M., dos Santos M.A., Vonk J.A. (2016) Greenhouse gas emissions from reservoir water surfaces: A new global synthesis. BioScience 66: 949-964. DelSontro T., McGinnis D.F., Sobek S., Ostrovsky I. & Wehrli B. (2010) Extreme Methane Emissions from a Swiss Hydropower Reservoir: Contribution from Bubbling Sediments. Environ. Sci. Technol. 44: 2419-2425.

DelSontro T., Kunz M.J., Kempter T., Wüest A., Wehrli B., Senn D.B. (2011) Spatial Heterogeneity of Methane Ebullition in a Large Tropical Reservoir. Environ. Sci. Technol. 45: 9866–9873. Diggle P.J., Heagerty P., Liang K.Y., Zeger S.L. (2002) The Analysis of Longitudinal Data. Second Edition. Oxford University Press, Oxford, England. Downing, J.A., Prairie, Y.T., Cole, J.J., Duarte, C.M., Tranvik, L.J., Striegl, R.G. et al. (2006). The global abundance and size distribution of lakes, ponds, and impoundments. Limnol. Oceanogr. 51: 2388-2397.

Downing B. D. and others. (2009) Quantifying fluxes and characterizing compositional changes

44 of dissolved organic matter in aquatic systems in situ using combined acoustic and optical measurements. Limnol Oceanogr-Meth 7: 119–131.

Duc N.T., Crill P., Bastviken D. (2010) Implications of temperature and sediment characteristics on methane formation and oxidation in lake sediments. Biogeochemistry 100:185–196.

Fagervold S.K., Bourgeois S., Pruski A.M., Charles F.C.O. et al. (2014) River organic matter shapes microbial communities in the sediment of the Rhone prodelta. The ISME Journal 8:2327– 2338.

Ferry J.G. (1992) Biochemistry of Methanogenesis. Critical Reviews in Biochemistry and Molecular Biology 27(6): 473-503.

Guillemette F., Bianchi T.S., Spencer R.G.M. (2017) Old before your time: Ancient carbon incorporation in contemporary aquatic foodwebs. Limnol Oceanogr 33:2–19. Hansen A.M., Kraus TEC, Pellerin BA, Fleck JA, Downing BD, Bergamaschi BA. (2016) Optical properties of dissolved organic matter (DOM): Effects of biological and photolytic degradation. Limnol. Oceanogr. 61:1015–1032. Harris D., Horwáth W.R., & van Kessel C. (2001) Acid fumigation of soils to remove carbonates prior to total organic carbon or carbon-13 isotopic analysis. Soil Sci. Soc. Am. J. 65: 1853-1856. Harrison J. A., Maranger R., Alexander R. B., Giblin A., Jacinthe P.-A., Mayorga E., Seitzinger S.P., Sobota D.J., and Wollheim W. (2009) Controls and significance of nitrogen retention in lakes and reservoirs, Biogeochemistry 93: 143–157. Hayes N. M., Deemer B. R., Corman J. R., Razavi N. R. and Strock K. E. (2017) Key differences between lakes and reservoirs modify climate signals: A case for a new conceptual model. Limnol. Oceanogr. 2: 47–62.

Holgerson M.A. & Raymond P.A. (2016) Large contribution to inland water CO2 and CH4 emissions from very small ponds. Nature Geoscience 9: 222-226. Huguet A., Vacher L., Relexans S., Saubusse S., Froidefond J.M., Parlanti E. (2009) Properties of fluorescent dissolved organic matter in the Gironde Estuary. Organic Geochemistry 40: 706- 719. Kelley C.A. & Chynoweth D. P. (1981) The contributions of temperature and of the input of organic matter in controlling rates of sediment methanogenesis. Limnol. Oceanogr. 26: 891–897. doi:10.4319/lo.1981.26.5.0891. Kirschke S. et al. (2013) Three decades of global meth- ane sources and sinks. Nat. Geosci. 6: 813–823.

Knoll L. B., Vanni M.J., Renwick W.H., & Kollie S. (2014) Burial rates and stoichiometry of sedimentary carbon, nitrogen and phosphorus in Midwestern US reservoirs. Freshwater Biology 59(11): 2342-2353.

45 Kondo M., Ichii K., Takagi H., and Saskawa M. (2015) Comparison of the data-driven top-down and bottom-up global terrestrial CO2 exchanges: GOSAT CO2 inversion and empirial eddy flux upscaling. J. Geophy. Res. Biogeosci. 120: 1226-1245.

Kozich J.J., Westcott S.L., Baxter N.T., Highlander S.K., Schloss P.D. (2013) Development of a dual-index sequencing strategy and curation pipeline for analyzing amplicon sequence data on the MiSeq Illumina sequencing platform. Applied and Environmental Microbiology 79(17): 5112-20. Liu Y. & Whitman W.B. (2008) Metabolic, Phylogenetic, and Ecological Diversity of the Methanogenic Archaea. Annals of the New York Academy of Sciences 1125: 171–189. Luton P.E., Wayne J.M., Sharp R.J., Riley P.W. (2002) The mcrA gene as an alternative to 16S rRNA in the phylogenetic analysis of methanogen populations in landfill. Microbiology 148: 3521–3530. Maeck A., DelSontro T., McGinnis D.F., Fischer H., Flury S., Schmidt M., Fietzek P., Lorke A. (2013) Sediment Trapping by Dams Creates Methane Emission Hot Spots. Environ. Sci. Technol. 47: 8130–8137.

Meyers P.A. (1994) Preservation of elemental and isotopic source identification of sedimentary organic matter. Chemical Geology 114: 289–302. McCallister S. L., & del Giorgio P.A. (2012) Evidence for the respiration of ancient terrestrial organic C in northern temperate lakes and streams. Proc. Natl. Acad. Sci. USA 109: 16963– 16968. doi:10.1073/pnas.1207305109

McGinnis D. F., Greinert J., Artemov Y., Beaubien S. E., W€uest A. (2006) The fate of rising methane bubbles in stratified waters: How much methane reaches the atmosphere? J. Geophys. Res. 111(C9): C09007.

McKnight D.M., Boyer E.W., Westerhoff P.K., Doran P.T., Kulbe T., Andersen D.T. (2001) Spectrofluorometric characterization of dissolved organic matter for indication of precursor organic material and aromaticity. Limnology and Oceanography 46: doi: 10.4319/lo.2001.46.1.0038. McMurdie P.J. & Holmes S. (2013) phyloseq: An R package for reproducible interactive analysis and graphics of microbiome census data. PLoS ONE 8(4): e61217. Miller A. (2017). leaps: Regression Subset Selection. R package version 3.0. https://CRAN.R- project.org/package=leaps. Mulholland P. J. & Elwood J.W. (1982) The role of lake and reservoir sediments as sinks in the perturbed global carbon cycle. Tellus 34: 490–499. Myhre G., Shindell D., Bréon F.M., Collins W., Fuglestvedt J., Huang J., Koch D., J.-F. Lamarque J.F., Lee D., Mendoza B., Nakajima T., Robock A., Stephens G., Takemura T. and Zhang H. (2013) Anthropogenic and Natural Radiative Forcing. In: Climate Change 2013: The Physical Science Basis. Contribution of Working Group I to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change [Stocker, T.F., D. Qin, G.-K. Plattner, M. Tignor,

46 S.K. Allen, J. Boschung, A. Nauels, Y. Xia, V. Bex and P.M. Midgley (eds.)]. Cambridge University Press, Cambridge, United Kingdom and New York, NY, USA.

Neef L., van Weele M. & van Velthoven P. (2010) Optimal estimation of the present-day global methane budget. Glob. Biogeochem. Cycles 24: GB4024.

Pinheiro J., Bates D., DebRoy S., Sarkar D. & R Core Team (2017). nlme: Linear and Nonlinear Mixed Effects Models. R package version 3.1-131. https://CRAN.R-project.org/package=nlme. Prairie et al. In Press. Greenhouse gas emissions from reservoirs: what does the atmosphere see? Ecosystems. Prather M. J., Holmes C.D., and Hsu J. (2012) Reactive greenhouse gas scenarios: Systematic exploration of uncertainties and the role of atmospheric chemistry. Geophys. Res. Lett. 39: L09803.

R Core Team (2017). R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. https://www.R-project.org/. Sakai S., Imachi H., Hanada S., Ohashi A., Harada H., and Kamagata Y. (2008) Methanocella paludicola gen. nov., sp nov., a methane-producing archaeon, the first isolate of the lineage ‘Rice Cluster I’, and proposal of the new archaeal order Methanocellales ord. nov. Int. J. Syst. Evol. Microbiol. 58: 929–936. Saunois M. et al. (2016) The global methane budget 2000–2012. Earth Syst. Sci. Data 8:697– 751. Schloss P.D. et al. (2009) Introducing mothur: Open-source, platform-independent, community- supported software for describing and comparing microbial communities. Appl. Environ. Microbiol. 75(23): 7537-41. Schulz S. & Conrad R. (1995) Effect of algal deposition on acetate and methane concentrations in the profundal sediment of a deep lake Lake Constance. FEMS Microbiol. Ecol. 16: 251–260. Schwarz J.I.K., Eckert W. & Conrad R. (2008) Response of methanogenic microbial community of a profundal lake sediment (Lake Kinneret, Israel) to algal deposition. Limnol. Oceanogr. 53: 113-121. Sickman J. O., C. L. DiGiorgio, M. L. Davisson, D. M. Lucero, and B. Bergamaschi. (2010) Identifying sources of dissolved organic carbon in agriculturally dominated rivers using radiocarbon age dating: Sacramento-San Joaquin River Basin, California. Biogeochemistry 99: 79–96. doi: 10.1007/s10533-009-9391-z

Stanley E. H., Casson N. J., Christel S. T., Crawford J. T., Loken L. C. & Oliver S. K. (2016). The ecology of methane in streams and rivers: Patterns, controls, and global significance. Ecological Monographs 86: 146–171.

Stock B.C. & Semmens B.X. (2013) MixSIAR GUI User Manual. Version 3.1. https://github.com/brianstock/MixSIAR.

47 Thauer R.K. (1998) Biochemistry of methanogenesis: a tribute to Marjory Stephenson. Microbiology 144: 2377-2406.

Thornton K. W., Kennedy R. H., Carroll R. H., Walker W. W., Gunkel R. C., and Ashby S. (1980) Reservoir sedimentation and water quality- an heuristic model, p. 654–661. In H. G. Stefan [ed.], Proceedings of the symposium on surface water impoundments. Amer. Soc. Civil Engr. Thornton K. W., Kimmel B. L., and Payne F. E. [eds.]. (1990) Reservoir limnology: Ecological perspectives. John Wiley & Sons. Tranvik L.J., Downing J.A., Cotner J.B., Loiselle S.A., Striegl R.G. Ballatore T.J. et al. (2009). Lakes and reservoirs as regulators of carbon cycling and climate. Limnol. Oceanogr. 54: 2298– 2314. Weishaar J. L., Aiken G. R., Bergamaschi B. A., Fram M. S., Fujii R., and Mopper K. (2003) Evaluation of specific ultraviolet absorbance as an indicator of the chemical composition and reactivity of dissolved organic carbon. Environ. Sci. Technol. 37: 4702–4708. Wen X., Yang S., Horn F., Winkel M., Wagner D., Liebner S. (2017) Global Biogeographic Analysis of Methanogenic Archaea Identifies Community-Shaping Environmental Factors of Natural Environments. Front Microbiol 8: doi 10.3389/fmicb.2017.01339. West W.E., Coloso J.J. & Jones S.E. (2012) Effects of algal and terrestrial carbon on methane production rates and methanogen community structure in a temperate lake sediment. Freshwater Biol. 57: 949-955. West W.E., McCarthy S.M., Jones S.E. (2015a) Phytoplankton lipid content influences freshwater lake methanogenesis. Freshwater Biology 60: 2261–2269. West W.E., Creamer KP, Jones SE. (2015b) Productivity and depth regulate lake contributions to atmospheric methane. Limnology and Oceanography 61: S51-S61. Whiticar M.J., Faber E., & Schoell M. (1986) Biogenic methane formation in marine and freshwater environments: CO2 reduction vs. acetate fermentation – Isotope evidence. Geochimica et Cosmochimica Acta 50: 693-709. Whiticar M.J., (1999) Carbon and hydrogen isotope systematics of bacterial formation and oxidation of methane. Chemical Geology 161: 291–314. Yu Y., Lee C., Kim J., Hwang S. (2005) Group-specific primer and probe sets to detect methanogenic communities using quantitative real-time polymerase chain reaction. Biotechnol. Bioeng. 89: 670–679. Zarfl C., Lumsdon A., Berlekamp J., Tydecks L., Tockner K. (2015) A global boom in hydropower dam construction. Aquat. Sci. 77 (1): 161−170.

48 Zuur A.F., Ieno E.N., Walker N.J., Saveliev A.A., Smith G.M. (2009) Mixed Effects Modelling for Nested Data. In: Mixed effects models and extensions in ecology with R. Statistics for Biology and Health. Springer, New York, NY. Zuur A.F., Ieno E.N., Elphick C.S. (2010) A protocol for data exploration to avoid common statistical problems. Methods in Ecology and Evolution 1: 3–14.

49 Appendices

Note on reproducibility and open data access

An effort has been made to make this study transparent and reproducible. Therefore, all data contained in the appendices, as well as raw and processed data presented in the body of the thesis can be found at the Center for Open Science’s Open Science Framework (OSF) website by searching the project’s unique 5-digit ID (59hnj) or visiting https://osf.io/59hnj/. Code and documentation for calculations, figures, and statistical analyses can be accessed the project’s

GitHub repository (link found on OSF project website).

50

The following is a summary of the information contained in each appendix:

Appendix 1: This appendix contains all site and sonde measurements from field sampling in

May and July 2016. Additionally, it contains sample metadata, water column chemistry not reported in Chapter 2 of the thesis, and a list of all variables that were measured.

Appendix 2: This appendix provides additional details for select methods from both field sampling and lab work, including details of sediment core retrieval and sediment slurry incubations.

Appendix 3: Appendix 3 contains sediment trap methods and summary results, including sediment trap construction details, dates of deployment, sedimentation rates, and composition of the sediment.

Appendix 4: Appendix 4 contains details of sediment slurry incubations from July of 2016.

These slurry incubations did not result in methane production, and some attempts at troubleshooting the failed slurries are described.

51 Appendix 1: Supplemental Data from Harsha Lake

This appendix includes tables with sample metadata, tables with all sonde and site measurements from both May and July sampling, and any supplemental information referenced in the main body of the thesis.

Supplemental Table 1. List of samples from May sampling of Harsha Lake (n = 46).

Sample ID Site Number Core Rep. Core ID Site Name Latitude Longitude Date Sampled Site Depth (m) Reservoir Zone H001 01 A 01A EFL 39.03657 84.13824 Thursday, May 19, 2016 16.5 lacustrine H002 01 B 01B EFL 39.03657 84.13824 Thursday, May 19, 2016 16.5 lacustrine H003 01 C 01C EFL 39.03657 84.13824 Thursday, May 19, 2016 16.5 lacustrine H004 02 A 02A EMB 39.01978 84.13150 Thursday, May 19, 2016 22.5 lacustrine H005 02 B 02B EMB 39.01978 84.13150 Thursday, May 19, 2016 22.5 lacustrine H006 02 C 02C EMB 39.01978 84.13150 Thursday, May 19, 2016 22.5 lacustrine H007 02 D 02D EMB 39.01978 84.13150 Thursday, May 19, 2016 22.5 lacustrine H008 03 A 03A ENN 39.02106 84.09780 Thursday, May 19, 2016 9 transitional H009 03 B 03B ENN 39.02106 84.09780 Thursday, May 19, 2016 9 transitional H010 03 C 03C ENN 39.02106 84.09780 Thursday, May 19, 2016 9 transitional H011 04 A 04A EUS 39.02695 84.09118 Thursday, May 19, 2016 2.5 riverine H014 04 B 04B EUS 39.02695 84.09118 Thursday, May 19, 2016 2.5 riverine H017 04 C 04C EUS 39.02695 84.09118 Thursday, May 19, 2016 2.5 riverine H020 05 A 05A ECP 39.01210 84.09890 Thursday, May 19, 2016 9 transitional H021 05 B 05B ECP 39.01210 84.09890 Thursday, May 19, 2016 9 transitional H022 05 C 05C ECP 39.01210 84.09890 Thursday, May 19, 2016 9 transitional H023 06 A 06A HWD 39.02025 84.14464 Monday, May 23, 2016 26 lacustrine H024 06 B 06B HWD 39.02025 84.14464 Monday, May 23, 2016 26 lacustrine H025 06 C 06C HWD 39.02025 84.14464 Monday, May 23, 2016 26 lacustrine H026 07 A 07A HCW 39.01400 84.12667 Monday, May 23, 2016 11 transitional H027 07 B 07B HCW 39.01400 84.12667 Monday, May 23, 2016 11 transitional H028 07 C 07C HCW 39.01400 84.12667 Monday, May 23, 2016 11 transitional H029 08 A 08A HND 39.01588 84.11209 Monday, May 23, 2016 9 transitional H030 08 B 08B HND 39.01588 84.11209 Monday, May 23, 2016 9 transitional H031 08 C 08C HND 39.01588 84.11209 Monday, May 23, 2016 9 transitional H032 09 A 09A HST 39.00447 84.09761 Monday, May 23, 2016 10 transitional H033 09 B 09B HST 39.00447 84.09761 Monday, May 23, 2016 10 transitional H034 09 C 09C HST 39.00447 84.09761 Monday, May 23, 2016 10 transitional H035 10 A 10A HTM 39.02551 84.08060 Monday, May 23, 2016 3 riverine H036 10 B 10B HTM 39.02551 84.08060 Monday, May 23, 2016 3 riverine H037 10 C 10C HTM 39.02551 84.08060 Monday, May 23, 2016 3 riverine H038 11 A 11A EOF 39.02041 84.15400 Thursday, May 26, 2016 31 lacustrine H039 11 B 11B EOF 39.02041 84.15400 Thursday, May 26, 2016 31 lacustrine H040 11 C 11C EOF 39.02041 84.15400 Thursday, May 26, 2016 31 lacustrine H041 12 A 12A HCE 39.01313 84.11840 Thursday, May 26, 2016 14 lacustrine H042 12 B 12B HCE 39.01313 84.11840 Thursday, May 26, 2016 14 lacustrine H043 12 C 12C HCE 39.01313 84.11840 Thursday, May 26, 2016 14 lacustrine H044 13 A 13A HEB 39.01831 84.10303 Thursday, May 26, 2016 12 lacustrine H045 13 B 13B HEB 39.01831 84.10303 Thursday, May 26, 2016 12 lacustrine H046 13 C 13C HEB 39.01831 84.10303 Thursday, May 26, 2016 12 lacustrine H047 14 A 14A HRM 39.02144 84.09072 Thursday, May 26, 2016 5 riverine H048 14 B 14B HRM 39.02144 84.09072 Thursday, May 26, 2016 5 riverine H049 14 C 14C HRM 39.02144 84.09072 Thursday, May 26, 2016 5 riverine H050 15 A 15A HEF 39.02751 84.08593 Thursday, May 26, 2016 4 riverine H051 15 B 15B HEF 39.02751 84.08593 Thursday, May 26, 2016 4 riverine H052 15 C 15C HEF 39.02751 84.08593 Thursday, May 26, 2016 4 riverine

52 Supplemental Table 2. May site sonde measurements.

2 3 6 2 1 mol/s/m 82 59 72 94 55 58 45 19 58 47 18 40 25 85 52 49 13 6.6 3.5 0.1 9.5 1.3 2.2 4.5 3.5 1.4 2.4 0.5 4.9 1.7 165 120 760 245 255 185 130 103 370 770 425 270 222 100 58.5 41.5 38.5 43.5 37.5 13.5 1.85 0.45 0.62 0.15 0.05 30.5 24.5 15.5 72.5 0.66 0.08 0.01 12.5 7.65 0.06 72.5 0.65 0.08 0.01 µ 103.5 127.5 177.5 147.5 112.5 182.5 142.5 622..5 PAR 0 1 2 3 4 5 0 1 2 3 4 5 0 1 2 3 4 5 0 1 2 0 1 2 3 4 5 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.5 2.5 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.5 2.5 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.5 2.5 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.5 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.5 2.5 PAR depth (m) 0.90 0.85 0.85 0.50 0.55 0.50 0.15 0.25 0.25 0.65 0.70 0.70 1.050 1.075 1.000 Secchi depth Secchi (m) ------Temp. ˚C (Turner) 1937.12 1950.88 1960.43 1969.30 1991.71 2000.31 2030.41 2039.69 2033.66 1980.14 2069.69 2109.53 2003.70 1909.79 1994.90 2003.66 2041.10 2051.64 2078.20 2070.06 2068.01 2050.73 2041.64 2061.49 2060.84 2089.36 2203.80 2185.47 2069.31 2082.66 2040.50 1961.41 1967.92 2197.00 2209.83 2238.02 2239.94 2218.52 2189.00 2182.48 2133.17 2004.10 1742.90 1850.53 1993.83 2181.59 2181.89 2232.83 2236.98 2202.97 2216.06 2243.84 2174.13 CDOM RFUB (Turner) 91.21 89.08 95.79 98.43 98.13 106.99 112.72 106.06 106.24 111.89 172.61 136.65 166.39 261.95 333.57 100.17 104.67 100.82 103.43 103.14 102.83 104.86 109.46 157.02 206.53 207.18 216.50 237.65 218.09 219.27 245.91 249.31 255.55 165.83 162.14 170.96 184.74 186.83 220.68 239.80 298.05 474.44 620.60 534.09 356.81 157.89 157.00 154.45 160.89 177.57 194.92 210.48 235.27 Turbidity RFUB (YSI) 417.70 444.90 389.03 449.61 400.95 356.95 269.92 275.24 267.80 282.91 256.70 259.65 255.52 254.71 285.53 380.79 529.74 564.56 476.80 391.25 370.22 334.54 325.48 275.11 261.80 259.28 261.73 271.41 246.04 243.42 244.67 235.00 237.21 429.51 446.18 387.74 382.79 394.76 383.57 329.52 352.20 348.78 457.12 667.80 525.86 495.20 527.11 555.15 539.95 459.66 468.86 453.29 346.48 ChlA RFUB in vivo (Turner) 5.10 3.90 3.90 5.40 5.40 6.20 6.60 7.40 7.80 8.60 8.30 8.70 9.40 3.00 3.90 4.40 5.10 5.60 5.60 5.90 6.50 7.00 8.50 5.30 6.20 8.80 9.80 9.60 8.40 4.50 4.44 4.90 5.00 4.80 6.10 9.20 9.00 20.00 11.10 12.50 11.90 10.30 11.50 15.40 14.90 13.50 11.30 53.30 48.30 60.00 54.20 Turbidity FNU (YSI) 9.25 9.25 9.07 8.66 8.06 7.91 7.75 7.61 7.51 7.42 7.37 7.33 7.30 7.28 9.25 9.25 9.01 8.52 7.97 7.75 7.67 7.54 7.48 7.44 7.40 7.36 7.34 7.33 7.33 7.31 7.28 8.80 8.76 8.48 8.22 8.02 7.75 7.60 7.45 7.91 7.68 7.61 7.56 8.83 8.65 8.38 8.10 7.77 7.59 7.42 7.35 pH (YSI) S/cm (YSI)S/cm µ 191.90 198.20 205.30 212.70 212.20 211.20 210.00 209.60 209.60 211.80 212.90 214.40 214.00 212.40 204.50 206.70 213.30 225.20 218.70 210.00 209.00 208.30 209.50 214.20 220.80 226.60 229.90 221.40 206.90 195.10 185.90 243.00 247.00 247.40 241.20 242.40 235.00 235.40 238.30 258.40 253.70 258.00 245.60 240.50 245.40 253.20 236.40 225.40 225.80 231.30 233.20 Sp.Cond. 9.00 4.70 1.80 1.30 1.00 5.30 2.10 1.20 1.00 0.90 0.60 8.20 6.80 4.40 3.00 7.20 4.60 3.80 69.00 49.40 32.00 31.00 23.00 16.90 52.20 30.10 30.10 27.00 23.10 17.20 11.10 82.00 30.60 15.80 76.10 74.20 63.20 58.70 72.30 39.00 19.80 232.70 173.40 131.00 198.00 173.50 117.00 139.50 119.00 147.30 106.70 D.O. % Sat. (YSI) 27.70 26.80 25.90 22.10 20.10 18.20 17.70 17.10 16.80 16.30 16.00 15.50 15.20 14.70 26.50 26.00 25.00 21.80 19.10 17.70 17.20 16.90 16.70 16.50 16.10 15.90 15.60 15.10 13.50 12.10 11.00 27.00 26.60 25.00 21.70 20.30 18.80 17.70 17.20 27.30 27.00 27.00 26.70 26.70 25.80 24.10 21.70 19.50 18.20 17.50 17.20 Temp. ˚C (YSI) 1 2 3 4 5 6 7 8 9 1 2 3 4 5 6 7 8 9 1 2 3 4 5 6 7 8 1 1 2 3 4 5 6 7 10 11 12 13 10 11 12 13 14 15 16 17 18 19 0.1 0.1 0.1 0.1 0.5 1.5 0.1 Measurement Depth Measurement (m) 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 2.5 2.5 2.5 2.5 2.5 2.5 2.5 2.5 2.5 2.5 2.5 2.5 2.5 16.5 16.5 16.5 16.5 16.5 16.5 16.5 16.5 16.5 16.5 16.5 16.5 16.5 16.5 16.5 16.5 16.5 22.5 22.5 22.5 22.5 22.5 22.5 22.5 22.5 22.5 22.5 22.5 22.5 22.5 22.5 22.5 22.5 22.5 22.5 22.5 22.5 Site DepthSite (m) riverine riverine riverine riverine riverine riverine riverine riverine riverine riverine riverine riverine riverine lacustrine lacustrine lacustrine lacustrine lacustrine lacustrine lacustrine lacustrine lacustrine lacustrine lacustrine lacustrine lacustrine lacustrine lacustrine lacustrine lacustrine lacustrine lacustrine lacustrine lacustrine lacustrine lacustrine lacustrine lacustrine lacustrine lacustrine lacustrine lacustrine lacustrine lacustrine lacustrine lacustrine lacustrine lacustrine lacustrine lacustrine transitional transitional transitional transitional transitional transitional transitional transitional transitional transitional transitional transitional transitional transitional transitional transitional transitional transitional transitional transitional transitional transitional transitional transitional transitional transitional transitional transitional transitional transitional transitional transitional transitional transitional Reservoir Zone Reservoir EFL EFL EFL EFL EFL EFL EFL EFL EFL EFL EFL EFL EFL EFL EFL EFL EFL EUS EUS EUS EUS EUS EUS EUS EUS EUS EUS EUS EUS EUS ECP ECP ECP ECP ECP ECP ECP ECP ECP ECP ECP ECP ECP ECP ECP ECP ECP ENN ENN ENN ENN ENN ENN ENN ENN ENN ENN ENN ENN ENN ENN ENN ENN ENN EMB EMB EMB EMB EMB EMB EMB EMB EMB EMB EMB EMB EMB EMB EMB EMB EMB EMB EMB EMB Site Name Site 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 4 4 4 4 4 4 4 4 4 4 4 4 4 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 Site NumberSite

53 Supplemental Table 2 (cont.). May site sonde measurements.

2 mol/s/m 9.60 4.80 2.38 0.58 0.14 5.60 1.02 0.44 0.12 7.48 3.27 1.32 0.20 0.04 3.32 0.77 2.23 0.07 µ 79.00 73.00 57.00 42.00 24.20 94.22 65.22 51.27 24.70 13.30 89.45 79.61 54.47 23.59 55.90 36.60 82.50 31.30 14.00 525.00 291.40 245.10 198.00 160.00 125.00 105.00 695.60 364.40 294.70 233.20 196.09 160.67 734.50 385.10 304.00 230.20 170.96 160.40 139.67 104.18 747.00 360.00 187.80 126.00 667.00 323.00 185.00 PAR 0 1 2 3 4 5 0 1 2 3 4 5 0 1 2 3 4 5 0 1 2 3 0 1 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.5 2.5 0.1 0.2 0.3 0.4 0.5 0.7 1.2 1.5 2.5 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.5 2.5 0.1 0.3 0.5 0.8 0.1 0.3 0.5 0.8 1.5 2.5 PAR depth (m) 1.00 1.00 1.00 0.80 0.80 0.80 0.70 0.75 0.75 0.50 0.60 0.50 0.40 0.40 0.35 Secchi depth Secchi (m) 19.50 18.60 18.10 17.83 17.69 17.53 17.09 16.93 16.75 16.67 16.57 16.25 15.76 15.43 14.67 13.75 12.75 11.75 11.41 11.19 10.99 10.94 10.84 10.81 10.66 19.87 18.93 18.49 18.36 18.27 18.15 17.54 17.34 17.22 17.05 23.09 19.38 18.59 18.40 18.31 17.91 17.66 17.46 17.28 21.08 19.01 18.59 18.48 18.27 17.91 17.62 17.20 17.01 22.63 20.41 17.93 17.12 16.64 16.52 Temp. ˚C (Turner) 1881.28 1945.73 1991.72 2002.83 2007.97 2014.61 2106.25 2089.03 2127.75 2122.89 2132.98 2188.73 2196.13 2178.05 2198.09 2017.61 1939.45 1931.74 1915.28 1927.32 1915.53 1912.32 1918.76 1911.94 1898.48 1953.57 1985.94 1995.96 1984.95 1985.61 2026.87 2103.79 2158.24 2148.54 2139.76 1893.83 2077.87 2103.48 2080.52 2058.01 2101.07 2106.01 2062.96 2075.51 2183.56 2181.89 2176.26 2176.05 2182.98 2213.29 2183.61 2130.06 2090.75 2101.76 2231.23 2280.77 2259.67 2218.06 2176.20 CDOM RFUB (Turner) 88.81 86.96 85.39 83.58 89.54 96.92 94.15 91.83 95.17 103.40 124.36 126.71 153.88 127.98 121.74 130.49 141.82 151.18 174.46 189.57 212.35 230.44 246.02 238.00 233.56 228.04 223.65 244.04 248.31 102.51 122.12 130.58 143.43 148.00 161.62 111.24 114.50 118.37 125.52 142.22 123.81 172.60 259.78 278.14 143.78 135.90 140.32 143.33 146.29 183.60 208.74 292.11 334.83 310.61 308.64 353.66 374.69 427.82 434.83 Turbidity RFUB (YSI) 810.84 624.77 505.66 467.36 431.37 297.22 267.38 284.38 250.01 249.51 255.53 262.19 253.66 250.54 234.36 233.31 231.14 230.04 232.12 224.86 237.94 222.98 222.38 224.90 981.90 814.60 688.55 422.28 332.02 310.23 285.70 779.40 831.44 639.75 578.99 429.28 399.03 387.30 718.94 504.74 415.30 436.33 346.17 471.63 543.37 490.33 470.43 455.45 1078.20 1068.05 1404.17 1217.42 1468.52 1019.90 2333.28 2008.41 1200.19 1093.17 1031.06 ChlA RFUB in vivo (Turner) 7.50 7.00 7.00 7.00 7.10 7.40 9.90 7.50 7.80 7.80 8.90 9.50 7.90 9.60 8.40 9.70 9.50 10.10 10.60 12.20 10.60 11.00 11.80 12.20 14.20 15.70 18.10 19.80 20.30 20.00 19.60 19.60 18.90 20.20 20.10 11.50 11.70 12.60 10.00 10.80 11.40 13.70 21.50 22.80 11.00 11.00 11.10 11.10 12.10 14.30 16.20 21.90 28.10 23.70 26.00 28.80 30.70 32.80 38.70 Turbidity FNU (YSI) 7.90 7.71 7.58 7.53 7.49 7.46 7.41 7.33 7.27 7.25 7.23 7.21 7.18 7.15 7.13 7.15 7.15 7.14 7.11 7.09 7.08 7.08 7.07 7.07 7.04 8.12 8.02 7.88 7.71 7.62 7.59 7.53 7.45 7.39 7.36 7.85 7.91 7.72 7.63 7.58 7.55 7.49 7.44 7.39 8.29 8.14 7.85 7.69 7.62 7.58 7.49 7.45 7.39 8.20 8.18 8.06 7.97 7.89 7.81 pH (YSI) S/cm (YSI)S/cm µ 244.20 244.10 243.30 242.60 242.50 242.70 247.90 246.40 250.90 246.80 243.90 246.70 243.60 241.80 242.40 226.50 222.30 215.70 215.10 211.60 218.80 221.70 225.90 226.40 228.20 246.20 245.30 245.70 245.70 245.90 246.00 250.10 256.00 255.00 256.10 257.80 255.60 255.00 254.30 253.00 252.60 258.00 266.10 276.50 249.10 255.00 255.70 255.20 255.10 255.30 259.70 264.60 257.90 322.60 327.90 319.50 319.70 320.20 322.00 Sp.Cond. 9.90 5.70 9.50 82.70 67.00 59.10 55.00 53.10 51.40 32.70 29.70 25.70 23.40 21.40 14.60 10.20 18.10 23.70 23.70 23.70 26.70 30.50 30.70 33.30 25.10 85.50 80.00 70.50 68.50 65.30 60.90 48.20 37.20 35.80 31.50 84.00 71.80 64.80 62.30 60.10 50.80 45.70 39.40 35.50 73.50 65.20 63.60 59.10 49.10 43.00 34.30 31.70 90.30 86.50 81.20 78.60 76.60 74.50 108.70 D.O. % Sat. (YSI) 19.5 18.4 17.9 17.6 17.4 17.3 16.9 16.7 16.6 16.5 16.3 16.0 15.6 15.4 14.4 13.6 12.5 11.3 11.0 10.9 10.7 10.6 10.6 10.5 10.4 19.8 18.7 18.3 18.2 18.1 18.0 17.6 17.2 17.0 16.9 22.8 19.1 18.3 18.2 18.1 17.7 17.4 17.2 17.1 20.4 18.7 18.3 18.3 18.1 17.8 17.4 17.1 16.8 22.9 20.5 17.2 16.7 16.4 16.3 Temp. ˚C (YSI) 1 2 3 4 5 6 7 8 9 1 2 3 4 5 6 7 8 9 1 2 3 4 5 6 7 8 1 2 3 4 5 6 7 8 1 2 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 0.1 0.1 0.1 0.1 0.1 0.5 1.5 2.5 Measurement Depth Measurement (m) 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 3 3 3 3 3 3 3 3 26 26 26 26 26 26 26 26 26 26 26 26 26 26 26 26 26 26 26 26 26 26 26 26 26 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 10 Site DepthSite (m) riverine riverine riverine riverine riverine riverine riverine riverine lacustrine lacustrine lacustrine lacustrine lacustrine lacustrine lacustrine lacustrine lacustrine lacustrine lacustrine lacustrine lacustrine lacustrine lacustrine lacustrine lacustrine lacustrine lacustrine lacustrine lacustrine lacustrine lacustrine lacustrine lacustrine transitional transitional transitional transitional transitional transitional transitional transitional transitional transitional transitional transitional transitional transitional transitional transitional transitional transitional transitional transitional transitional transitional transitional transitional transitional transitional transitional transitional transitional transitional transitional transitional transitional transitional transitional transitional transitional transitional transitional transitional transitional transitional transitional transitional transitional transitional transitional transitional transitional transitional transitional Reservoir Zone Reservoir HST HST HST HST HST HST HST HST HST HST HST HST HST HST HST HST HST HND HND HND HND HND HND HND HND HND HND HND HND HND HND HND HND HCW HCW HCW HCW HCW HCW HCW HCW HCW HCW HCW HCW HCW HCW HCW HCW HND HTM HTM HTM HTM HTM HTM HTM HTM HWD HWD HWD HWD HWD HWD HWD HWD HWD HWD HWD HWD HWD HWD HWD HWD HWD HWD HWD HWD HWD HWD HWD HWD HWD HCW Site Name Site 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 6 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 7 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 8 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 9 10 10 10 10 10 10 10 10 Site NumberSite

54 Supplemental Table 2 (cont.). May site sonde measurements.

2 mol/s/m 4.54 1.96 0.91 0.44 0.13 0.03 4.86 2.28 0.93 0.52 0.12 0.05 1.78 0.63 0.21 0.03 0.01 8.54 5.01 3.18 2.30 1.61 1.07 0.31 0.09 0.04 0.01 9.31 5.58 3.55 2.27 1.79 1.13 0.77 0.21 0.05 0.02 0.00 µ 89.96 57.42 41.63 36.24 30.51 26.43 23.17 20.96 15.57 11.75 61.14 42.71 29.62 22.00 16.99 15.77 20.36 11.50 65.20 45.21 35.65 28.10 24.00 64.00 23.28 21.72 19.46 15.45 26.06 16.25 35.29 14.01 202.10 756.30 375.20 268.90 720.80 103.51 673.60 231.00 128.79 333.20 125.17 PAR 0 1 2 3 4 5 0 1 2 3 4 5 0 1 2 3 4 5 0 1 2 3 0 1 2 3 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.5 2.5 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.5 2.5 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.5 2.5 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.5 2.5 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.5 2.5 PAR depth (m) 0.70 0.70 0.60 0.50 0.60 0.50 0.60 0.60 0.60 0.40 0.40 0.40 0.30 0.30 0.40 Secchi depth Secchi (m) 21.69 21.70 20.60 19.33 18.28 17.83 17.57 17.10 16.86 16.63 16.44 16.19 15.45 14.49 13.24 12.39 11.67 11.23 10.91 10.56 21.69 21.31 20.13 18.77 18.05 17.74 17.37 17.21 17.00 16.65 16.52 16.23 15.85 21.57 21.31 21.01 20.94 19.44 18.08 17.73 17.41 17.23 17.03 16.92 22.94 22.80 22.50 22.36 22.13 22.13 24.23 22.71 21.64 21.23 20.39 18.72 Temp. ˚C (Turner) 1830.94 1836.92 1898.66 1940.97 1965.00 1982.35 1995.06 2018.80 2045.41 2083.07 2126.08 2126.13 2153.63 2105.68 1973.55 1915.70 1889.18 1901.23 1893.94 1890.89 1951.05 1940.97 2027.53 2079.64 2084.88 2087.28 2130.68 2131.61 2150.23 2122.05 2085.21 2107.69 2104.97 1982.84 2004.76 2022.83 2024.48 2125.94 2178.25 2153.57 2161.82 2178.22 2204.45 2145.68 1926.06 1983.09 2027.92 2034.47 2044.43 2068.46 2035.62 2122.16 2238.05 2235.12 2163.80 2169.95 CDOM RFUB (Turner) 90.11 87.66 81.88 80.24 80.69 71.80 74.01 90.03 93.82 98.81 89.84 87.43 90.41 92.53 95.20 96.35 103.29 108.83 118.94 132.79 153.60 184.99 208.43 218.78 225.01 221.63 220.09 114.28 113.12 140.96 227.45 254.56 244.43 292.22 102.48 101.08 105.01 122.66 117.81 150.87 148.15 161.72 163.65 187.99 170.52 203.71 192.33 154.31 139.81 155.01 247.11 333.34 290.41 275.37 356.12 413.89 Turbidity RFUB (YSI) 938.22 618.86 445.59 312.36 251.87 238.34 254.16 256.73 245.38 242.85 228.64 224.81 225.59 225.07 219.13 231.88 469.51 412.60 309.64 286.32 268.00 274.49 296.23 282.36 287.78 462.70 367.11 302.88 298.80 298.59 302.79 801.66 525.46 437.40 2827.37 3767.69 3299.49 1744.65 3035.05 5238.49 3313.75 1211.87 3620.42 4411.24 4716.45 5731.16 2461.10 8749.73 9500.89 9093.25 8536.01 8053.70 6957.69 3681.90 1091.94 10795.12 ChlA RFUB in vivo (Turner) 6.80 7.00 6.60 6.60 6.30 6.00 6.10 7.30 7.50 8.30 9.00 9.50 7.90 7.50 7.60 7.50 7.70 8.60 9.50 9.60 7.80 8.10 8.30 8.20 9.40 9.60 11.00 12.90 15.60 17.50 18.90 18.60 18.40 18.80 12.50 18.40 21.70 20.50 23.70 12.10 12.40 13.70 13.60 15.90 14.10 17.00 14.40 12.30 11.70 12.60 20.40 30.30 22.20 24.50 32.60 33.30 Turbidity FNU (YSI) 8.57 8.52 8.20 7.67 7.61 7.40 7.28 7.22 7.15 7.12 7.08 7.05 7.01 6.99 6.98 6.94 6.91 6.89 6.87 6.81 8.56 8.47 8.19 7.51 7.38 7.26 7.57 7.31 7.17 7.09 7.07 7.07 7.08 8.78 8.45 8.33 8.29 7.90 7.63 7.48 7.35 7.25 7.21 7.15 8.90 8.65 8.57 8.55 8.53 8.47 8.98 8.52 8.01 7.87 7.77 7.58 pH (YSI) S/cm (YSI)S/cm µ 182.70 182.70 186.60 186.70 185.80 184.80 184.60 184.10 183.90 184.20 185.10 185.00 183.10 178.30 171.00 167.40 165.50 165.70 170.40 176.10 189.20 186.00 191.60 192.20 197.30 193.10 196.30 196.30 205.40 236.30 244.70 251.30 241.40 191.90 192.30 192.50 192.90 200.90 207.50 207.50 207.10 221.40 218.20 245.00 198.80 209.80 205.60 202.50 201.20 205.50 225.10 249.20 260.70 265.00 265.50 263.10 Sp.Cond. 8.20 9.20 0.90 90.70 67.70 46.00 45.20 41.60 30.00 25.80 20.80 16.40 13.60 15.90 21.40 23.40 25.70 22.10 80.00 52.00 41.50 34.40 27.50 24.90 20.70 20.50 21.00 21.30 12.00 47.80 41.90 35.60 26.00 28.00 20.20 30.40 95.10 80.80 69.50 57.80 39.30 133.30 132.30 143.00 123.00 122.40 116.20 112.20 111.20 163.50 144.20 142.80 142.70 139.00 133.70 174.20 D.O. % Sat. (YSI) 21.60 21.50 20.30 19.10 18.10 17.60 17.30 16.90 16.60 16.40 16.30 16.00 15.20 14.10 12.90 12.10 11.40 10.90 10.70 10.30 21.70 20.90 19.70 18.60 17.80 17.50 17.20 17.00 16.80 16.40 16.30 16.10 15.60 21.40 21.40 20.80 20.80 18.80 17.70 17.50 17.10 17.00 16.80 16.70 22.70 22.60 22.20 22.10 21.90 21.90 24.10 22.20 21.50 20.80 20.40 19.10 Temp. ˚C (YSI) 1 2 3 4 5 6 7 8 9 1 2 3 4 5 6 7 8 9 1 2 3 4 5 6 7 8 9 1 2 3 1 2 3 10 11 12 13 15 17 19 21 23 25 10 11 12 10 0.1 0.1 0.1 0.1 1.5 2.5 0.1 1.5 2.5 Measurement Depth Measurement (m) 5 5 5 5 5 5 5 5 5 5 5 5 5 5 5 4 4 4 4 4 4 4 4 4 4 4 4 4 4 4 31 31 31 31 31 31 31 31 31 31 31 31 31 31 31 31 31 31 31 31 14 14 14 14 14 14 14 14 14 14 14 14 14 14 14 14 14 12 12 12 12 12 12 12 12 12 12 12 12 12 12 12 12 12 Site DepthSite (m) riverine riverine riverine riverine riverine riverine riverine riverine riverine riverine riverine riverine riverine riverine riverine riverine riverine riverine riverine riverine riverine riverine riverine riverine riverine riverine riverine riverine riverine riverine lacustrine lacustrine lacustrine lacustrine lacustrine lacustrine lacustrine lacustrine lacustrine lacustrine lacustrine lacustrine lacustrine lacustrine lacustrine lacustrine lacustrine lacustrine lacustrine lacustrine lacustrine lacustrine lacustrine lacustrine lacustrine lacustrine lacustrine lacustrine lacustrine lacustrine lacustrine lacustrine lacustrine lacustrine lacustrine lacustrine lacustrine lacustrine lacustrine lacustrine lacustrine lacustrine lacustrine lacustrine lacustrine lacustrine lacustrine lacustrine lacustrine lacustrine lacustrine lacustrine lacustrine lacustrine Reservoir Zone Reservoir EOF EOF EOF EOF EOF EOF EOF EOF EOF EOF EOF EOF EOF EOF EOF EOF EOF EOF EOF EOF HEF HEF HEF HEF HEF HEF HEF HEF HEF HEF HEF HEF HEF HEF HEF HCE HCE HCE HCE HCE HCE HCE HCE HCE HCE HCE HCE HCE HCE HCE HCE HCE HEB HEB HEB HEB HEB HEB HEB HEB HEB HEB HEB HEB HEB HEB HEB HEB HEB HRM HRM HRM HRM HRM HRM HRM HRM HRM HRM HRM HRM HRM HRM HRM Site Name Site 11 11 11 11 11 11 11 11 11 11 11 11 11 11 11 11 11 11 11 11 12 12 12 12 12 12 12 12 12 12 12 12 12 12 12 12 12 13 13 13 13 13 13 13 13 13 13 13 13 13 13 13 13 13 14 14 14 14 14 14 14 14 14 14 14 14 14 14 14 15 15 15 15 15 15 15 15 15 15 15 15 15 15 15 Site NumberSite

55

Supplemental Table 3. List of samples from July sampling of Harsha Lake (n = 45).

Sample ID Site Number Core Rep. Core ID Site Name Latitude Longitude Date Sampled Site Depth (m) Reservoir Zone H053 01 A 01A EFL 39.03669 84.13814 Tuesday, July 19, 2016 16.5 lacustrine H054 01 B 01B EFL 39.03669 84.13814 Tuesday, July 19, 2016 16.5 lacustrine H055 01 C 01C EFL 39.03669 84.13814 Tuesday, July 19, 2016 16.5 lacustrine H056 02 A 02A EMB 39.01978 84.13158 Tuesday, July 19, 2016 22.5 lacustrine H057 02 B 02B EMB 39.01978 84.13158 Tuesday, July 19, 2016 22.5 lacustrine H058 02 C 02C EMB 39.01978 84.13158 Tuesday, July 19, 2016 22.5 lacustrine H059 03 A 03A ENN 39.02114 84.09769 Tuesday, July 19, 2016 9 transitional H060 03 B 03B ENN 39.02114 84.09769 Tuesday, July 19, 2016 9 transitional H061 03 C 03C ENN 39.02114 84.09769 Tuesday, July 19, 2016 9 transitional H062 04 A 04A EUS 39.02692 84.09108 Tuesday, July 19, 2016 2.5 riverine H063 04 B 04B EUS 39.02692 84.09108 Tuesday, July 19, 2016 2.5 riverine H064 04 C 04C EUS 39.02692 84.09108 Tuesday, July 19, 2016 2.5 riverine H065 05 A 05A ECP 39.01219 84.09872 Tuesday, July 19, 2016 9 transitional H066 05 B 05B ECP 39.01219 84.09872 Tuesday, July 19, 2016 9 transitional H067 05 C 05C ECP 39.01219 84.09872 Tuesday, July 19, 2016 9 transitional H068 06 A 06A HWD 39.02036 84.14469 Thursday, July 21, 2016 26 lacustrine H069 06 B 06B HWD 39.02036 84.14469 Thursday, July 21, 2016 26 lacustrine H070 06 C 06C HWD 39.02036 84.14469 Thursday, July 21, 2016 26 lacustrine H071 07 A 07A HCW 39.01414 84.12653 Thursday, July 21, 2016 11 transitional H072 07 B 07B HCW 39.01414 84.12653 Thursday, July 21, 2016 11 transitional H073 07 C 07C HCW 39.01414 84.12653 Thursday, July 21, 2016 11 transitional H074 08 A 08A HND 39.01611 84.11194 Thursday, July 21, 2016 9 transitional H075 08 B 08B HND 39.01611 84.11194 Thursday, July 21, 2016 9 transitional H076 08 C 08C HND 39.01611 84.11194 Thursday, July 21, 2016 9 transitional H077 09 A 09A HST 39.00467 84.09756 Thursday, July 21, 2016 10 transitional H078 09 B 09B HST 39.00467 84.09756 Thursday, July 21, 2016 10 transitional H079 09 C 09C HST 39.00467 84.09756 Thursday, July 21, 2016 10 transitional H080 10 A 10A HTM 39.02556 84.08050 Thursday, July 21, 2016 3 riverine H081 10 B 10B HTM 39.02556 84.08050 Thursday, July 21, 2016 3 riverine H082 10 C 10C HTM 39.02556 84.08050 Thursday, July 21, 2016 3 riverine H083 11 A 11A EOF 39.02058 84.15414 Tuesday, July 26, 2016 31 lacustrine H084 11 B 11B EOF 39.02058 84.15414 Tuesday, July 26, 2016 31 lacustrine H085 11 C 11C EOF 39.02058 84.15414 Tuesday, July 26, 2016 31 lacustrine H086 12 A 12A HCE 39.01314 84.11881 Tuesday, July 26, 2016 14 lacustrine H087 12 B 12B HCE 39.01314 84.11881 Tuesday, July 26, 2016 14 lacustrine H088 12 C 12C HCE 39.01314 84.11881 Tuesday, July 26, 2016 14 lacustrine H089 13 A 13A HEB 39.01844 84.10314 Tuesday, July 26, 2016 12 lacustrine H090 13 B 13B HEB 39.01844 84.10314 Tuesday, July 26, 2016 12 lacustrine H091 13 C 13C HEB 39.01844 84.10314 Tuesday, July 26, 2016 12 lacustrine H092 14 A 14A HRM 39.02142 84.09069 Tuesday, July 26, 2016 5 riverine H093 14 B 14B HRM 39.02142 84.09069 Tuesday, July 26, 2016 5 riverine H094 14 C 14C HRM 39.02142 84.09069 Tuesday, July 26, 2016 5 riverine H095 15 A 15A HEF 39.02756 84.08594 Tuesday, July 26, 2016 4 riverine H096 15 B 15B HEF 39.02756 84.08594 Tuesday, July 26, 2016 4 riverine H097 15 C 15C HEF 39.02756 84.08594 Tuesday, July 26, 2016 4 riverine

56

Supplemental Table 4. July site sonde measurements.

Site Number Site Name Reservoir Zone Site Depth (m) Measurement Depth (m) Temp. ˚C (YSI) D.O. % Sat. (YSI) Sp.Cond. µS/cm (YSI) pH (YSI) Turbidity FNU (YSI) Secchi depth (m) PAR depth (m) PAR µmol/s/m2 1 EFL lacustrine 16.5 0.1 28.5 108.6 269.7 8.92 6.2 0.7 0 442.3 1 EFL lacustrine 16.5 1 28 107.5 269.3 9.04 6.4 0.65 0.1 321 1 EFL lacustrine 16.5 1.5 27.9 104.4 269.2 9.04 6.3 0.75 0.2 295.2 1 EFL lacustrine 16.5 2 27.8 102.4 269.3 9.03 6.3 0.3 263.4 1 EFL lacustrine 16.5 2.5 27.8 101.3 269.3 9.02 6.2 0.4 194.37 1 EFL lacustrine 16.5 3 27.1 40.7 281 8.72 5.6 0.5 154.12 1 EFL lacustrine 16.5 3.5 26.7 19 286 8.45 4.9 0.6 124.72 1 EFL lacustrine 16.5 4 26.3 6.5 287.4 8.2 5.2 0.7 91.16 1 EFL lacustrine 16.5 4.5 26.1 1.9 288.9 8.08 4.7 0.8 75.12 1 EFL lacustrine 16.5 5 25.2 1.2 294.4 8.02 3.8 0.9 59.57 1 EFL lacustrine 16.5 5.5 24.6 1 296.3 7.92 4 1 48 1 EFL lacustrine 16.5 6 23.3 0.7 299 7.87 4.5 1.5 28.06 1 EFL lacustrine 16.5 8 18.7 0.5 309.4 7.89 5 2 12.06 1 EFL lacustrine 16.5 9 16.3 0.4 309.2 7.74 6.5 2.5 5.34 1 EFL lacustrine 16.5 10 16 0.4 308.5 7.58 6.2 3 2.58 1 EFL lacustrine 16.5 11 15.6 0.4 308 7.48 8.2 4 0.67 1 EFL lacustrine 16.5 12 14.9 0.4 306.7 7.43 8.1 5 0.21 1 EFL lacustrine 16.5 13 14.5 0.3 303.9 7.38 8.5 1 EFL lacustrine 16.5 14 14 0.3 299.2 7.35 8.1 1 EFL lacustrine 16.5 2 EMB lacustrine 22.5 0.1 29.9 115.5 271.9 9.57 5 0.65 0 680.3 2 EMB lacustrine 22.5 1 28.5 128.8 272.6 9.47 7.1 0.65 0.1 451.2 2 EMB lacustrine 22.5 1.5 28 119.6 272.8 9.32 7.1 0.75 0.2 390.3 2 EMB lacustrine 22.5 2 27.7 93 275.3 9.18 7.1 0.3 298.2 2 EMB lacustrine 22.5 2.5 27.3 56 283.6 8.95 6.2 0.4 251.7 2 EMB lacustrine 22.5 3 27 38.2 288.7 8.76 5.9 0.5 200.7 2 EMB lacustrine 22.5 3.5 26.8 23 293.2 8.53 5.5 0.6 186.22 2 EMB lacustrine 22.5 4 26.4 2.3 297.3 8.35 5.5 0.7 146.37 2 EMB lacustrine 22.5 4.5 25.6 1.5 293.7 8.21 4.9 0.8 96.22 2 EMB lacustrine 22.5 5 25.1 1.2 293.3 8.12 4 0.9 93.35 2 EMB lacustrine 22.5 5.5 24.5 1 295.6 8.02 3 1 76.49 2 EMB lacustrine 22.5 6 24 0.9 297.7 7.99 2.7 1.5 33.46 2 EMB lacustrine 22.5 6.5 23.1 0.8 300.1 7.93 3.5 2 14.52 2 EMB lacustrine 22.5 7 20.1 0.6 309.3 7.88 5.6 2.5 6.1 2 EMB lacustrine 22.5 7.5 18.6 0.4 312.5 7.83 5.4 3 2.51 2 EMB lacustrine 22.5 8 18.2 0.4 313.6 7.75 5.8 4 0.67 2 EMB lacustrine 22.5 9 16.4 0.3 313.5 7.75 8.2 5 0.26 2 EMB lacustrine 22.5 11 15.7 0.3 317 7.67 8.9 2 EMB lacustrine 22.5 13 14.6 0.3 314.7 7.62 12.3 2 EMB lacustrine 22.5 15 12.9 0.2 288.9 7.62 9.6 2 EMB lacustrine 22.5 17 12.1 0.2 272.6 7.57 9.2 2 EMB lacustrine 22.5 19 11.4 0.2 267.4 7.52 12.5 3 ENN transitional 9 0.1 31 156.7 296.5 9.3 9 0.55 0 635.7 3 ENN transitional 9 1 28.2 114 290.6 8.97 8.2 0.65 0.1 389 3 ENN transitional 9 1.5 28 111 290.2 8.78 7.5 0.55 0.2 340.6 3 ENN transitional 9 2 27.9 105.7 290.7 8.69 7.4 0.3 250.9 3 ENN transitional 9 2.5 27.7 70.5 295.6 8.51 7.1 0.4 207.6 3 ENN transitional 9 3 27.3 32.3 300.3 8.2 6.4 0.5 175.67 3 ENN transitional 9 3.5 26.9 16.5 317.1 7.94 9.5 0.6 89.53 3 ENN transitional 9 4 26.5 8 320 7.8 10.3 0.7 67.43 3 ENN transitional 9 4.5 26.1 1.9 313.8 7.62 8.6 0.8 85.44 3 ENN transitional 9 5 25.2 1.2 317.8 7.55 11.1 0.9 63.97 3 ENN transitional 9 5.5 24.8 1 317.4 7.41 14.4 1 51.38 3 ENN transitional 9 6 22.8 0.8 318.8 7.34 14.8 1.5 19.94 3 ENN transitional 9 6.5 21.8 0.7 325.9 7.27 17.3 2 7.83 3 ENN transitional 9 7 20.6 0.6 332.8 7.19 25 2.5 3.06 3 ENN transitional 9 3 1.45 3 ENN transitional 9 4 0.25 3 ENN transitional 9 5 0.04 4 EUS riverine 2.5 0.1 31.7 164.6 301.9 8.81 11.4 0.35 0 525.1 4 EUS riverine 2.5 0.5 30.1 150 304.3 8.82 16.2 0.35 0.1 348.4 4 EUS riverine 2.5 1 28.6 116.7 306.3 8.72 19.9 0.4 0.2 253.4 4 EUS riverine 2.5 1.5 28.4 85.9 312.2 8.39 23.8 0.3 189.89 4 EUS riverine 2.5 0.4 140.61 4 EUS riverine 2.5 0.5 111.57 4 EUS riverine 2.5 0.6 77.57 4 EUS riverine 2.5 0.7 53.51 4 EUS riverine 2.5 0.8 40.36 4 EUS riverine 2.5 0.9 28.43 4 EUS riverine 2.5 1 20.18 4 EUS riverine 2.5 1.5 2.98 5 ECP transitional 9 0.1 32 158 289.5 9 6.9 0.55 0 192.65 5 ECP transitional 9 1 28.9 147.9 287.1 9.04 8.2 0.55 0.1 81.08 5 ECP transitional 9 1.5 28.2 127 288.8 8.89 8.3 0.5 0.2 274.8 5 ECP transitional 9 2 27.9 101 290.7 8.79 8 0.3 211.1 5 ECP transitional 9 2.5 27.7 70.5 294.9 8.62 7.5 0.4 191.5 5 ECP transitional 9 3 27.3 47.7 299.3 8.38 7.2 0.5 149.86 5 ECP transitional 9 3.5 27.1 16.1 303 8.21 6.6 0.6 113.42 5 ECP transitional 9 4 26.8 3.5 301.8 8 6.5 0.7 91.12 5 ECP transitional 9 4.5 25.9 2.1 308.1 7.83 6 0.8 77.64 5 ECP transitional 9 5 25.3 1.6 310.9 7.71 6.1 0.9 68.87 5 ECP transitional 9 5.5 24.2 1.3 314.2 7.64 7.9 1 50.81 5 ECP transitional 9 6 23.6 1.1 316.3 7.52 17.6 1.5 5.02 5 ECP transitional 9 6.5 22.3 0.9 322.4 7.45 14.5 2 1.88 5 ECP transitional 9 7 19.7 0.8 335 7.3 28.8 2.5 0.76 5 ECP transitional 9 3 0.33 5 ECP transitional 9 4 0.08 5 ECP transitional 9 5 0.02

57

Supplemental Table 4 (cont.). July site sonde measurements.

Site Number Site Name Reservoir Zone Site Depth (m) Measurement Depth (m) Temp. ˚C (YSI) D.O. % Sat. (YSI) Sp.Cond. µS/cm (YSI) pH (YSI) Turbidity FNU (YSI) Secchi depth (m) PAR depth (m) PAR µmol/s/m2 6 HWD lacustrine 26 0.1 29.7 138 265.1 9.04 5.7 0.65 0 493.6 6 HWD lacustrine 26 1 29.3 135.9 264.1 9.09 6.3 0.65 0.1 339.6 6 HWD lacustrine 26 1.5 28.8 137.2 264.2 9.1 6.9 0.6 0.2 290 6 HWD lacustrine 26 2 28 103 267.9 9 6.9 0.3 243.7 6 HWD lacustrine 26 2.5 27.5 70.8 274 8.85 6.2 0.4 203.4 6 HWD lacustrine 26 3 27.2 26.3 282.2 8.62 5.3 0.5 169.93 6 HWD lacustrine 26 3.5 26.9 18 284.2 8.32 4.7 0.6 143.23 6 HWD lacustrine 26 4 26.6 12.1 284.3 8.14 4.5 0.7 118.71 6 HWD lacustrine 26 4.5 26.1 1.8 286.5 8 3.9 0.8 101.05 6 HWD lacustrine 26 5 25.3 1.4 288.8 7.93 3.6 0.9 86.06 6 HWD lacustrine 26 5.5 24.3 1.1 291.6 7.87 3.2 1 72.94 6 HWD lacustrine 26 6 23.2 0.9 295.5 7.81 3 1.5 30.92 6 HWD lacustrine 26 7 20.2 0.7 305.2 7.74 3.5 2 13.54 6 HWD lacustrine 26 8 17.9 0.6 306.7 7.65 3.9 2.5 5.92 6 HWD lacustrine 26 9 16.6 0.5 308.3 7.55 4.6 3 2.75 6 HWD lacustrine 26 11 15.6 0.4 302.3 7.48 5.8 4 0.63 6 HWD lacustrine 26 13 14.5 0.5 296.3 7.57 6.4 5 0.23 6 HWD lacustrine 26 15 13.2 0.3 280.5 7.52 7.4 6 HWD lacustrine 26 17 12.2 0.4 271 7.47 8.3 6 HWD lacustrine 26 19 11.2 0.3 263.3 7.42 8.4 6 HWD lacustrine 26 21 10.8 0.2 263.8 7.38 9.9 6 HWD lacustrine 26 22 10.7 0.3 266.2 7.35 10.3 7 HCW transitional 9 0.1 30.4 146.1 268 9.34 5.9 0.5 0 658 7 HCW transitional 9 1 28.9 144.7 267.2 9.31 8.2 0.55 0.1 384.1 7 HCW transitional 9 1.5 28.3 131.9 269.5 9.22 7.8 0.5 0.2 292 7 HCW transitional 9 2 28.1 103.4 271.7 9.08 7.8 0.3 256.6 7 HCW transitional 9 2.5 27.4 71.7 280 8.88 6.1 0.4 205.6 7 HCW transitional 9 3 27 36 285.4 8.57 5.5 0.5 176.7 7 HCW transitional 9 3.5 26.7 18.4 286.4 8.42 4.7 0.7 158.81 7 HCW transitional 9 4 26.3 7.1 288.2 8.18 4.2 0.8 135.54 7 HCW transitional 9 4.5 25.9 2.1 290.1 8.05 3.8 0.9 112.39 7 HCW transitional 9 5 25.3 1.4 294.9 7.95 3.6 1 91.2 7 HCW transitional 9 5.5 24.2 1.1 298.9 7.9 3.6 1.2 82.71 7 HCW transitional 9 6 23.5 0.9 300.2 7.83 3.4 1.5 33.28 7 HCW transitional 9 6.5 22.7 0.9 303.7 7.78 7.5 2 14.97 7 HCW transitional 9 7 21.4 0.7 308.9 7.71 7.3 2.5 6.07 7 HCW transitional 9 3 2.57 7 HCW transitional 9 4 0.67 7 HCW transitional 9 5 0.26 8 HND transitional 9 0.1 30.9 164.4 274.8 9.29 6.3 0.55 0 665.2 8 HND transitional 9 1 29.4 175 274.2 9.31 7.6 0.55 0.1 432.4 8 HND transitional 9 1.5 29 149.4 278 9.12 7.7 0.6 0.2 376.2 8 HND transitional 9 2 28.1 106.6 284.4 8.99 7.8 0.3 326.2 8 HND transitional 9 2.5 27.5 47.5 289.9 8.73 6.8 0.4 261.3 8 HND transitional 9 3 27.1 24.3 294.2 8.34 6.3 0.5 230.7 8 HND transitional 9 3.5 26.9 12.2 296.2 8.16 6 0.6 178.42 8 HND transitional 9 4 26.3 2.4 295.6 7.98 4.9 0.7 152.43 8 HND transitional 9 4.5 25.9 1.4 300 7.83 5.4 0.8 131.69 8 HND transitional 9 5 25.4 1.2 305.2 7.75 6.4 0.9 107.95 8 HND transitional 9 5.5 24.6 0.9 309.2 7.69 18.8 1 88.04 8 HND transitional 9 6 23.2 0.8 321.8 7.62 41.7 1.5 33.14 8 HND transitional 9 6.5 22.4 0.8 318.9 7.56 15.2 2 11.14 8 HND transitional 9 2.5 4.75 8 HND transitional 9 3 1.97 8 HND transitional 9 4 0.41 8 HND transitional 9 5 0.12 9 HST transitional 10 0.1 30.7 161.5 277.2 9.29 7.7 0.45 0 589.2 9 HST transitional 10 1 30.5 168 275.6 9.2 7.9 0.4 0.1 383 9 HST transitional 10 1.5 28.6 125.4 286.5 9.05 7.4 0.45 0.2 307.5 9 HST transitional 10 2 27.8 60 297.7 8.7 6.5 0.3 242.9 9 HST transitional 10 2.5 27.3 11.8 302.9 8.4 5.5 0.4 193.34 9 HST transitional 10 3 27 4.2 303.9 8.16 5.4 0.5 146.5 9 HST transitional 10 3.5 26.8 2.5 304.6 8.04 5.2 0.6 124.63 9 HST transitional 10 4 26.5 1.9 305.8 7.95 5 0.7 102.61 9 HST transitional 10 4.5 26.1 1.5 307.9 7.88 5.3 0.8 96.03 9 HST transitional 10 5 25.1 1.3 311.9 7.83 5.7 0.9 82.93 9 HST transitional 10 5.5 24.1 1.1 316.4 7.77 7.2 1 67.89 9 HST transitional 10 6 23 1 321.5 7.71 8.8 1.5 25.54 9 HST transitional 10 6.5 22 0.8 326.8 7.61 10.6 2 9.62 9 HST transitional 10 7 20.5 0.8 333.6 7.51 16.5 2.5 3.68 9 HST transitional 10 3 1.58 9 HST transitional 10 4 0.38 9 HST transitional 10 5 0.12 10 HTM riverine 3 0.1 33 201.1 335.1 9.13 12 0.35 0 150.61 10 HTM riverine 3 0.4 0.1 64.07 10 HTM riverine 3 0.375 0.3 45.32

58

Supplemental Table 4 (cont.). July site sonde measurements.

Site Number Site Name Reservoir Zone Site Depth (m) Measurement Depth (m) Temp. ˚C (YSI) D.O. % Sat. (YSI) Sp.Cond. µS/cm (YSI) pH (YSI) Turbidity FNU (YSI) Secchi depth (m) PAR depth (m) PAR µmol/s/m2 11 EOF lacustrine 31 0.1 30.2 118.2 249.2 9.01 5.5 0.65 0 556.7 11 EOF lacustrine 31 1 30 117.9 249.2 9.05 5.5 0.7 0.1 296.7 11 EOF lacustrine 31 1.5 30 115.1 249.4 9.04 5.6 0.67 0.2 256.9 11 EOF lacustrine 31 2 29.9 112 249.4 9.02 6 0.3 171.42 11 EOF lacustrine 31 2.5 29.6 99.9 251.2 8.99 6 0.4 163.96 11 EOF lacustrine 31 3 28.6 71 260 8.84 5.3 0.5 158.14 11 EOF lacustrine 31 3.5 28 43.1 265.2 8.6 5 0.6 140.11 11 EOF lacustrine 31 4 27.3 15.6 269.9 8.43 4.2 0.7 120.92 11 EOF lacustrine 31 4.5 26.9 4.4 274.1 8.27 3.5 0.8 107.36 11 EOF lacustrine 31 5 25.7 1.8 276.8 8.09 3.2 0.9 93.17 11 EOF lacustrine 31 5.5 24.7 1.4 280.8 7.99 3.3 1 84.71 11 EOF lacustrine 31 6 23.8 1.1 282.9 7.88 3.1 1.5 42.05 11 EOF lacustrine 31 7 21.1 0.9 292 7.78 3.6 2 17.95 11 EOF lacustrine 31 8 19.1 0.8 295.7 7.72 3.9 2.5 8.72 11 EOF lacustrine 31 9 17.2 0.7 297.2 7.68 4.4 3 4.12 11 EOF lacustrine 31 10 16.4 0.6 293.8 7.63 3.8 4 0.76 11 EOF lacustrine 31 12 15 0.5 286.8 7.61 5.5 5 0.27 11 EOF lacustrine 31 14 13.8 0.5 275.4 7.74 6.6 11 EOF lacustrine 31 16 12.2 0.4 260.5 7.51 7.6 11 EOF lacustrine 31 18 11.3 0.4 254.7 7.45 7.7 11 EOF lacustrine 31 20 10.8 0.4 255.5 7.39 8.7 11 EOF lacustrine 31 22 10.6 0.3 258.8 7.34 9.8 12 HCE lacustrine 14 0.1 31 124.5 260.2 9.2 4.8 0.65 0 236.2 12 HCE lacustrine 14 1 30.4 116 261.4 9.1 5.5 0.6 0.1 161.32 12 HCE lacustrine 14 1.5 30.2 105.6 263.9 9.03 5.3 0.625 0.2 110.49 12 HCE lacustrine 14 2 30 94.1 268.4 8.86 5.1 0.3 100.71 12 HCE lacustrine 14 2.5 28.8 21.9 292.6 8.53 4.7 0.4 82.84 12 HCE lacustrine 14 3 28 4.7 290.4 8.19 5.5 0.5 63.44 12 HCE lacustrine 14 3.5 27.6 4.4 284.7 8.07 5.3 0.6 56.4 12 HCE lacustrine 14 4 26.7 1.9 281 7.98 5.2 0.7 37.91 12 HCE lacustrine 14 4.5 26.4 1.4 282.6 7.9 5.4 0.8 30.28 12 HCE lacustrine 14 5 25.7 1.2 287.6 7.85 5.5 0.9 24.31 12 HCE lacustrine 14 5.5 25 1 287.6 7.81 5.4 1 19.58 12 HCE lacustrine 14 6 23.9 0.9 291.6 7.77 5.4 1.5 8.38 12 HCE lacustrine 14 6.5 22.1 0.8 298.1 7.73 6.7 2 3.98 12 HCE lacustrine 14 7 20.5 0.6 307.5 7.7 8.9 2.5 1.9 12 HCE lacustrine 14 7.5 19.3 0.6 314.1 7.64 9 3 0.89 12 HCE lacustrine 14 8 18.4 0.6 311.7 7.61 9.5 4 0.23 12 HCE lacustrine 14 9 17.1 0.5 310.5 7.56 8 5 0.08 12 HCE lacustrine 14 10 16.4 0.4 311.9 7.52 8.5 12 HCE lacustrine 14 11 15.8 0.4 318.6 7.46 13.3 12 HCE lacustrine 14 12 15.5 0.4 330.7 7.39 13.7 13 HEB lacustrine 12 0.1 30.3 111.2 270.2 8.95 4.9 0.55 0 36.13 13 HEB lacustrine 12 1 30.5 112.3 270.6 8.94 4.9 0.6 0.1 20.8 13 HEB lacustrine 12 1.5 30.5 111.4 270.9 8.94 5 0.6 0.2 14.68 13 HEB lacustrine 12 2 30.2 101.6 273.2 8.91 4.9 0.3 11.78 13 HEB lacustrine 12 2.5 29.2 29.9 287 8.61 4.9 0.4 9.44 13 HEB lacustrine 12 3 28.1 3.3 287.7 8.21 5 0.5 8.17 13 HEB lacustrine 12 3.5 27.8 2.1 289.2 8.01 4.8 0.6 7.13 13 HEB lacustrine 12 4 27.1 1.7 293.1 8.06 4.9 0.7 5.94 13 HEB lacustrine 12 4.5 26.5 1.2 318 7.96 7.6 0.8 5.25 13 HEB lacustrine 12 5 25.5 1 298.5 7.93 15.4 0.9 4.4 13 HEB lacustrine 12 5.5 25 1 298.6 7.9 18.4 1 3.73 13 HEB lacustrine 12 6 23.6 0.9 302.8 7.85 24.6 1.5 1.78 13 HEB lacustrine 12 6.5 22.6 0.8 306.1 7.77 16.2 2 0.86 13 HEB lacustrine 12 7 21.1 0.8 315.3 7.73 11.5 2.5 0.38 13 HEB lacustrine 12 7.8 20 0.6 317.5 7.67 10.3 3 0.19 13 HEB lacustrine 12 8 18.7 0.5 322 7.59 13.8 4 0.04 13 HEB lacustrine 12 8.5 17.5 0.5 328.4 7.53 17.1 5 0.01 13 HEB lacustrine 12 9 16.7 0.4 331.6 7.49 16.8 13 HEB lacustrine 12 9.5 16.6 0.4 331.5 7.44 16.4 13 HEB lacustrine 12 10 16.2 0.4 340.5 7.34 25.8 14 HRM riverine 5 0.1 30.7 100.9 296.9 8.51 19.4 0.35 0 411.5 14 HRM riverine 5 1 30.5 88.2 294.7 8.52 16.6 0.3 0.1 192.88 14 HRM riverine 5 1.5 30.3 82.5 296.2 8.42 16.9 0.275 0.2 120.52 14 HRM riverine 5 2 29.3 27 305 8.3 10.1 0.3 74.43 14 HRM riverine 5 2.5 28.2 5.1 309.7 8.1 10.2 0.4 54.67 14 HRM riverine 5 0.5 39.04 14 HRM riverine 5 0.6 26.74 14 HRM riverine 5 0.7 18.65 14 HRM riverine 5 0.8 11.2 14 HRM riverine 5 0.9 9.55 14 HRM riverine 5 1 7.07 14 HRM riverine 5 1.5 1.74 14 HRM riverine 5 2 0.52 14 HRM riverine 5 2.5 0.2 14 HRM riverine 5 3 0.05 15 HEF riverine 4 0.1 30.8 123.3 319.3 8.85 17.5 3 0 100.65 15 HEF riverine 4 1 30.6 111 321.1 8.77 21 2.5 0.1 42.27 15 HEF riverine 4 2.75 0.2 27.53 15 HEF riverine 4 0.3 19.75 15 HEF riverine 4 0.4 13.4 15 HEF riverine 4 0.5 9.5 15 HEF riverine 4 0.6 6.68 15 HEF riverine 4 0.7 4.4 15 HEF riverine 4 0.8 3.05 15 HEF riverine 4 0.9 2.21 15 HEF riverine 4 1 1.68 15 HEF riverine 4 1.5 0.3 15 HEF riverine 4 2 0.05

59

Supplemental Table 5. All water chemistry from May sampling.

Reservoir Secchi Sample Sample TNH4+ TRP TNO2-3 TNO2 TN TP Chlorophyll a Site Site Depth Zone Depth (m) Location Depth (µg N/L) (µg P/L) (µg N/L) (µg N/L) (µg N/L) (µg P/L) (µg/L) epilimnion 0.1 m 179 118 496 17.4 1750 184.8 - 1 16.5 m lacustrine 1.04 hypolimnion 13.5 m 35 56.4 593 1248.75 214.6 - epilimnion 0.1 m 207 112 516 22.8 1250 187.7 - 2 22.5 m lacustrine 0.87 hypolimnion 19.0 m 30.1 88.3 580 6.95 4575 45.8 - epilimnion 0.1 m 243 107 579 31.3 1410 181.9 - 3 9.0 m transitional 0.52 hypolimnion 8.0 m 234 86.9 466 1475 216 - epilimnion 0.1 m 304 73.9 643 31.8 1210 171.6 - 4 2.5 m riverine 0.22 hypolimnion 1.5 m 290 83.7 630 32.6 1487.5 209.7 - epilimnion 0.1 m 246 116 1400 200.6 - 5 9.0 m transitional 0.68 hypolimnion 8.0 m 163 114 632 9.15 1287.5 203.4 - epilimnion 0.1 m 63.3 115 624 23.5 1320 194.7 15.96 6 26.0 m lacustrine 1.00 hypolimnion 24.0 m 125 125 615 13.4 1270 247.5 0.48 epilimnion 0.1 m 105 125 622 25.8 1400 193.9 16.38 7 11.0 m transitional 0.80 hypolimnion 10.0 m 88.2 126 1130 15.4 1520 220.3 1.80 epilimnion 0.1 m 141 121 625 26.8 1490 219.6 18.58 8 9.0 m transitional 0.73 hypolimnion 8.0 m 181 117 700 19.1 1510 211 1.71 epilimnion 0.1 m 85.3 105 640 34.4 1490 199.2 30.37 9 10.0 m transitional 0.53 hypolimnion 9.0 m 245 99.2 490 19.6 1490 216 5.43 epilimnion 0.1 m 66.9 148 1520 32.1 2000 232.3 4.76 10 3.0 m riverine 0.38 hypolimnion 2.0 m 76.4 146 1470 26.7 2240 243.6 3.16 epilimnion 0.1 m 12.5 81.8 372 23.2 1250 176.9 19.88 11 31.0 m lacustrine 0.67 hypolimnion 24.5 m 168 128 569 19.6 1330 254 0.53 epilimnion 0.1 m 14.5 52.9 254 29.2 1330 167.4 32.04 12 14.0 m lacustrine 0.53 hypolimnion 12.0 m 175 96.6 969 15.9 1710 193.9 1.93 epilimnion 0.1 m 32.6 69.3 407 30.7 1380 165.8 29.16 13 12.0 m lacustrine 0.60 hypolimnion 10.5 m 93.9 93.5 1030 18.4 1710 0.75 epilimnion 0.1 m 23.5 66.6 338 31.7 1730 209.5 63.92 14 5.0 m riverine 0.40 hypolimnion 4.0 m 99.7 71.9 614 43 1600 184.7 31.60 epilimnion 0.1 m 24.3 75.1 520 45.9 2040 226.6 74.16 15 4.0 m riverine 0.33 hypolimnion 3.0 m 239 147 1410 62.8 2280 286.7 1.82

60

Supplemental Table 6. All variables used in the predictive model (n = 56).

Site physical and chemiscal characteristics (n = 16) Water chemistry (n = 12) Sediment (n = 12) Porewater (n = 6) Microbial community (n = 10) -1 Site depth (m) TNH4+ (µg N/L) - Epilimnion % Corg DOC (mg L ) Total methanogen abundance (qPCR) 13 Secchi depth (m) TRP (µg P/L) - Epilimnion Corg FI (fluorescence index) Total archaea abundance (qPCR) Temperature (˚C) - Epilimnion TNO2-3 (µg N/L) - Epilimnion % N Freshness index % Methanomicrobiales (order) Dissolved oxygen (% saturation) - Epilimnion TNO2 (µg N/L) - Epilimnion 15N RFE (relative fluorescence efficiency) % Methanosarcinales (order) Specific conductivity (µS cm-1) - Epilimnion TN (µg N/L) - Epilimnion C/N BIX (biological index) % Methanobacteriales (order)

pH - Epilimnion TP (µg P/L) - Epilimnion N/C SUVA254 % Methanoregula (genus) ChlA in vivo (RFUB, fluorescence) - Epilimnion TNH4+ (µg N/L) - Hypolimnion % OM (dry weight) % Methanosaeta (genus) Turbidity (RFUB, fluorescence) - Epilimnion TRP (µg P/L) - Hypolimnion g OM slurry-1 % Methanobacterium (genus) CDOM (RFUB, fluorescence) - Epilimnion TNO2-3 (µg N/L) - Hypolimnion Density (g mL-1) % Methanolinea (genus) Temperature (˚C) - Hypolimnion TNO2 (µg N/L) - Hypolimnion % water % Methanosarcina (genus) Dissolved oxygen (% saturation) - Hypolimnion TN (µg N/L) - Hypolimnion Proportion autochthonous OM Specific conductivity (µS cm-1) - Hypolimnion TP (µg P/L) - Hypolimnion Slurry dry weight (g) pH - Hypolimnion ChlA in vivo (RFUB, fluorescence) - Hypolimnion Turbidity (RFUB, fluorescence) - Hypolimnion CDOM (RFUB, fluorescence) - Hypolimnion

61

Appendix 2: Select Detailed Methods

FIELD

Sediment core sampling. Triplicate cores were collected from 15 sites on two sampling occasions

(May and July 2016), except for in May where 4 cores were collected at site EMB, for a total of

91 sediment cores. Only data from sediment cores collected in May of 2016 were presented in

Chapter 2. See Appendix 4 for a discussion of the sediment slurries from the July 2016 sampling.

Sediment cores were collected using a K-B gravity corer (Wildco, Yulee, FL), and stored on ice until processed. All cores were processed within ~24 hours of collection (see below for core processing). Sediment cores were collected in core tube liners that were 20” long and 2” in diameter. To collect the core, the nose piece of the coring device was unscrewed, a core tube liner was placed inside of the core tube, and the nose piece was screwed back on. The black rubber plunger on the corer head assembly was raised, and the corer was slowly lowered into the water ONLY until the corer head assembly (top part of the corer) was submerged. The messenger was threaded up the cord attached to the head assembly so that there was enough length of cord to reach the sediment (needed to be more than the depth of water at the site). Then the cord was released so the corer could free fall to the sediment. Once there, any slack in the cord was removed and the messenger dropped. The corer was slowly raised back to the surface but the plunger and core tube were left in the water. Then, a size 10 rubber stopper was placed in the end of the nose piece while the tube was still in the water, to prevent the sediment from coming out. The corer was then brought on board the boat. To remove the core tube liner with the sediment core, the black nose piece was unscrewed slowly, and the end of the liner was covered as soon as the nose piece was removed. Then, a white 2” cap was placed inside of the

62 liner (on the bottom), and a different size 10 rubber stopper sealed the bottom of the core. The top was covered with an orange plastic cap, and the core was store upright in a cooler, and kept in the dark and cool until processing.

Water sampling. Epilimnion (0.1 m below water surface) and hypolimnion (0.5 – 2 m above the sediment) water samples were collected from each of the 15 sites by using a 2 or 3 L Niskin bottle. Duplicate samples were taken at 3 of the 15 sites for quality assurance purposes. All samples were triple rinsed with lake water before collection. Water samples for chlorophyll and water chemistry were stored in pre-cleaned 1 L amber Nalgene bottles. Water samples for suspended solid analyses were stored in pre-cleaned 1 L clear Nalgene bottles. Once collected, all water samples were kept dark and cool until further processing.

Dissolved gas sampling. Dissolved gas samples were taken from the epilimnion and hypolimnion from each of the 15 sites. For hypolimnion samples, water was sampled with a Niskin bottle

(using the same hypolimnion water sample as mentioned above). Then a 140 mL syringe with a stopcock was used to collect 120 mL of bubble-free water. 20 mL of air or ultra-high purity

(UHP) helium* was added and the stopcock closed. Samples were shaken for 5 minutes, and then the gas was transferred to a pre-evacuated 12 mL exetainer. During the July sampling, an ambient air sample was also drawn by adding 20 mL of air into a pre-evacuated 12 mL exetainer.

*Note: In May, dissolved gas was displaced with air. In July, dissolved gas was displaced with

UHP Helium.

Field measurements. Several measurements were taken in the field at each site with multiparameter sondes and other sampling equipment. Vertical profile measurements (~1 m

63 depth intervals) of chromophoric dissolved organic matter (CDOM), chlorophyll a in vivo, and turbidity were measured in RFUB (raw fluorescence units, blank substracted) by a C3™

Submersible Fluorometer (Turner Designs, Sunnyvale, CA, U.S.A.). Vertical profile measurements (~ 1 m depth intervals) of temperature, depth, dissolved oxygen, pH, conductivity, pressure and turbidity were measured using a YSI ProDSS (Yellow Springs, OH, U.S.A.).

Photosynthetically active radiation (PAR) was measured using a LiCor light meter (measured every 0.1 m). Secchi depth was also recorded.

LAB – SAMPLE PROCESSING

Sediment core extrusion and processing. Cores were extruded in a glove box under N2 atmosphere. Overlying water was removed with a syringe and temporarily stored in a labeled container, and sediment cores were extruded into three 5 cm sections. Sections were placed into

Fisher 5.5” x 9” sterile sampling bags. The top 5 cm section was the only section that was processed further in the glove box. The other two core sections were placed on ice and later frozen. Subsampling of the top core section also occurred in the glove box and consisted of homogenizing and dividing the sample for molecular analysis, porewater analyses, sediment characterization and sediment slurry incubations. Briefly, each sample was homogenized within the Fisher sampling bag, then 15 mL of sediment was added to a 120 mL serum bottle for slurry incubations using a 3 mL syringe with the tip cut off. With the same syringe, ~ 1-2 mL of sediment was added to a 2 mL cryogenic tube for molecular analysis and stored on ice until later frozen in a -80˚C freezer. A corner of the sampling bag was then cut off and 35-50 mL of sediment was squeezed out into a pre-weighed 50 mL centrifuge tube for porewater analysis and density calculation. The volume of sediment in the 50 mL centrifuge tubes was recorded and the tube was weighed (outside of the glove box) to determine density. The remaining sediment was

64 squeezed into a Fisher 3” x 5” sterile sampling bag and stored on ice then frozen for future sediment characterization (see below). After subsampling, slurry incubations were further prepared by adding 15 mL of overlying core water into the serum bottle containing 15 mL of sediment, shaking for two minutes, then capping with a rubber (butyl) stopper for serum bottles

(Geo-Microbial Technologies, Inc., Ochelata, OK).

Sediment slurry preparation and methane production rate calculation. Once outside the glove box, slurry incubation bottles were sealed with an aluminum crimp seal and covered in aluminum foil to block out light. After this, samples were shaken vigorously for 5 minutes more*, flushed with N2 gas for 5 minutes to removed remaining CH4 and CO2 that may have been in the sediment, and brought to atmospheric pressure**. The time was recorded at this point, and was considered the start of the slurry incubation (“T0”). Slurries were incubated for 9 days and 11 or 12 mL*** gas samples were taken on days 1, 2, 3, 5, 7 and 9. Before taking the gas sample, pressure in the serum bottle was determined and recorded either by measuring volume of displacement after inserting a syringe and needle into the bottle or with a pressure gauge equipped with a Luer-Lok attachment and a needle****. After taking a gas sample, 11 or

12 mL*** of N2 gas was returned to the serum bottle to maintain pressure. Gas samples were analyzed on a Bruker GC equipped with a flame ionizing detector (FID). After accounting for dilution, rates of methanogenesis were calculated using the linear increase in methane concentration in the serum bottle.

*Note: For July slurries, after shaking for 5 min, two 12-mL gas samples were taken from each slurry before flushing. (Added 20 mL of gas into each 12 mL exetainer).

65 **Note: This was not done for all of the slurries – See Appendix 4 for which slurries started at atmospheric pressure (were “bubbled off” as described above), and which were not equilibrated to atmospheric pressure at the start of the incubation period.

***Note: Took 11 mL gas samples for slurries in May and took 12 mL gas samples for slurries in July. Always added the same volume of N2 that was removed to slurries after sampling (if 11 mL samples taken, 11 mL of N2 returned).

****Note: For May slurries the pressure was only recorded using the gas displacement method

(with a syringe). In July, both the pressure gauge and the gas displacement method were used.

LAB – SAMPLE ANALYSIS

Sediment characterization – bulk properties. Bulk sediment was characterized by bulk density, percent water, and percent organic matter (AFDM or LOI). Bulk density was determined by adding a known volume of sediment to a pre-weighed 50 mL centrifuge tube, and weighing the centrifuge tube after the sediment addition. After subtracting the weight of the centrifuge tube, density was reported as g/mL. Percent water of the sediment was determined by adding sediment to a pre-weighed aluminum weighing boat, weighing, then drying the sediment for 3 days at

60˚C, and re-weighing. Percent water is calculated as the difference between the wet weight and the dry weight of the sediment, divided by the wet weight (total weight), after subtracting the weight of the aluminum weighing boat. Percent organic matter was determined by weighing aluminum foil, adding dry sediment and recording the weight, then wrapping the aluminum foil around the sediment to avoid leakage, and ashing in a muffle furnace at 550˚C for 4 hours. After

66 ashing and cooling, the sediment and aluminum packets were reweighed. LOI (percent organic matter) was calculated as the difference between the total sediment weight and the ashed sediment weight (ash is the remaining sediment after ashing), divided by the total sediment weight and multiplied by 100 to get a percent. The weight of the aluminum foil must be subtracted from each weight before calculating LOI.

Sediment characterization – elemental and isotopic composition. Each sample was characterized by its total organic carbon and nitrogen content, as well as by organic carbon and nitrogen stable isotope signatures (13C and 15N). All carbon values are reported as organic carbon; inorganic carbon was removed prior to analyzing via acid fumigation (Harris et al. 2001, see

“References”). Briefly, sediment was weighed into silver capsules that were arranged in a 96 well plate, a small amount of Milli-Q water (30-50 µL) was added to each sample, and the well plate placed in a desiccator (with desiccant removed) with a beaker with 100 mL of concentrated hydrochloric (HCl) acid. After 6-8 hours, the samples were removed and dried at 60˚C overnight, then each capsule was placed into a tin capsule and folded closed. Samples were analyzed on an elemental analyzer with an isotope ratio mass spectrometer (EA-IRMS) at the Stable Isotope

Geochemistry Lab in the Department of Geology at the University of Cincinnati or at the Stable

Isotope Facility (SIF) at the University of California Davis.

67 Appendix 3: Sediment Traps

This appendix details the methods, summary results, and a discussion of sediment traps that were deployed in Harsha Lake in the summer of 2016. The goal of this project was to evaluate sedimentation rates and sediment composition at 4 sites across the reservoir during the summer.

This was part of an independent research project for Madison Duke, a participant of the undergraduate WISE (Women in Science and Engineering) program at UC. She was responsible for constructing the sediment traps and filtering all samples. Together, she and I designed the traps, sampled the traps, and analyzed and discussed the data.

Methods

Sampling sites. The four sites used for sediment traps stations were selected to encompass the spatial variability of reservoir depth and proximity to the main tributary (Figure A3.1). These sites were a subset of the 15 sites used in May of 2016 for sediment core and water sampling.

Buoy deployment at these sites was approved by reservoir managers for a previous study conducted by Jake Beaulieu of the USEPA.

39.050

EFL● 6 EUS 39.025 ● ● 6 EMB● ENN● Zone ● lacustrine Latitude 39.000 ● riverine ● transitional

38.975

−84.15 −84.12 −84.09 Longitude

Figure A3.1. Sites where sediment traps were located during June and part of July 2017. The reservoir zones correspond to the zones described in the main body of the thesis.

68

Sediment trap construction. Sediment traps were constructed using 2” diameter PVC pipe and caps, threaded rod, closed cell foam for floatation, and rope. A cement block and buoy were used in deployment (see Sediment trap deployment and sampling). The collection tubes were constructed by cutting PVC pipe into 25 cm lengths and attaching a PVC cap on one end, resulting in tubes with a height to diameter ratio of 5:1 (Bloesch & Burns 1980). They had a surface area of 19.63 cm2, and a volume of 520 mL (after accounting for the extra volume due to curved bottom of the PVC cap). There were two trap designs; the first design was a simple single-collection-tube design (Figure A3.2A), and the second utilized three collection-tubes per trap (Figure A3.2B). The single-collection-tube design was the initial design, but was later replaced with the three-collection-tube design that allowed for replicates at each site and helped keep the sediment traps floating upright.

A B

Figure A3.2. Single-collection-tube sediment trap (A) and triple-collection-tube sediment trap (B) before deployment. Pictures taken by Madison Duke.

Sediment trap deployment and sampling. The line attached to the sediment traps was secured to a cement block so that length of line between the tops of the sediment collection tubes and the

69 cement block was 1.5 or 2 m (Figure A3.3). The length of line was 2 m for sites EFL, EMB and

ENN, but only 1.5 m for EUS, due to the shallowness of this site (see Table A3.2). The line that was connected to the buoy was attached to the other end of the cement block. To deploy the trap, the collection tubes were filled with deionized (DI) water, then the cement block was slowly lowered into the water using the buoy line until the cement block reached the sediment. Great care was taken to do this slowly to minimize the amount of bottom sediment that was stirred up during deployment and to avoid tipping the sediment trap.

Buoy

Sediment trap

1.5 or 2 m from cement block to top of trap

Cement block

Figure A3.3. Schematic of a deployed sediment trap.

We did not used preservatives (such as formalin, sodium azide, or mercuric chloride) to prevent degradation of the material in the sediment traps; however, traps were sampled weekly to minimize OM decomposition. Sediment traps were deployed for seven weeks during the summer of 2016 (Table A3.1).

70 To retrieve the trap, the buoy line was slowly lifted out of the water. Again, great care was taken to ensure that the sediment trap was always floating above the cement block and not being pulled by the cement block. If the line was pulled up too quickly, the sediment trap could end up sideways and spill the contents. Samples were not collected if the trap was sideways when it reached the surface of the water. To collect the contents of the triple-collection-tube trap, caps were placed in two of the three collection tubes (Figure A3.4A), and contents of the uncapped tube were poured into a 1L pre-cleaned Nalgene bottle. DI water was used to rinse any remaining particles into the Nalgene bottle, and the volume of rinse water was recorded to account for dilution. This process was repeated for each of the three collection tubes, and all samples were stored on ice or at 4˚C in the dark until filtered. If necessary, the tubes were scrubbed with a brush before redeployment. This was usually only needed at the EUS site due to the growth on the inside and outside of the trap (Figure A3.4). Care was taken to not remove or collect the attached growth when collecting the contents of the tubes.

A B

Figure A3.4. (A) Picture of sediment trap with cap used in sampling the triple-collection-tube trap, and (B) an example of the growth that occurred on the sediment trap at the EUS site.

71 Sedimentation rates and composition of sediment trap material. Sedimentation rates were

determined by calculating the total solids (TS) per sediment trap area, and dividing by the

number of days the trap was deployed. Organic matter, chlorophyll, elemental composition (C,

N) and stable isotope composition (13C and 15N) were measured to evaluate the composition of

the sediment trap material. All samples were filtered through three 25 mm 0.7 µm GF/F filters

within 24 hours of sample collection. One filter was used to determine total solids and organic

matter content by loss on ignition (LOI); filters were combusted at 550˚C for 4 hours in a muffle

furnace. A second filter was used to measure total solids, and for a subset of samples, elemental

and stable isotope composition. Elemental analysis and stable isotope analysis were conducted

using an elemental analyzer connected to an isotope ration mass spectrometer (EA-IRMS)

(Elementar Vario EL Cube, Elementar Analysensysteme GmbH, Hanau, Germany interfaced to a

PDZ Europa 20-20 isotope ratio mass spectrometer, Sercon Ltd., Cheshire, UK). The

spectrophotometric method was used to measure chlorophyll a following acetone extraction

(APHA 2012) for the third filter.

Table A3-1. Dates involved in sediment trap sampling and retrieval. All single-tube sediment traps were deployed on 6/2/16. An “X” indicates that a sample was successfully retrieved on the date listed in the column header.

Week 1 Week 2 Week 3 Week 4 Week 5 Week 6 Week 7 Samples (6/10/16) (6/15/16) (6/22/16) (6/29/16) (7/6/16) (7/14/16) (7/19/17) collected

EFL - 1 X X* X X X** 5

EMB - 2 X* X ** 2

ENN - 3 X* X X X X X ** 6

EUS - 4 X* X X X X X** 6 Days between 8 5 7 7 7 8 5 sampling * Deployed three-tube sediment trap on this date after collecting sample from the single tube design. ** Date the sediment trap was removed from Harsha.

72 Results

Sedimentation rates and composition. Sedimentation rates were the highest in the EUS site, while OM percent was lowest in this site. For the full data set, visit https://osf.io/59hnj/.

Table A3-1. Average sedimentation rates, %OM and sedimentation rates of OM for the three sites that were sampled regularly. Site EMB was only sampled twice so was not included here.

Sedimentation rates (mg/cm2/day) OM percent Sedimentation rates (mg OM/cm2/day) EFL 1.53 22.57 0.33 ENN 4.66 18.91 0.82 EUS 53.07 19.37 6.52

Discussion

Sediment trap design could be improved in several ways. First, a better material for floatation is recommended for traps that will be deployed for longer than a month. The closed cell foam that was used to keep the traps upright started to become waterlogged after ~5 weeks, and resulted in the traps struggling to float. Also, the pressure for the deeper sites could have affected the floatation.

Secondly, the sampling procedure from traps with three collection tubes could be streamlined if the tubes were removable. Essentially, each of the tubes (PVC pipe) would be able to be removed, and three new collection tubes would be replaced. This would eliminate the need to have sampling bottles, save time in the field, and allow for thorough cleaning of the tubes in between collections.

73 Appendix 4: July Slurries – Absence of Methane Production

This project originally had an objective to evaluate how methane production, sediment OM composition and methanogen communities changed later in the summer. Our experimental design included repeating sampling at the same sites in July 2016. This sampling was carried out, and the same analyses were performed, but the methane production slurry incubations resulted in no discernable methanogenesis. The data were not included in the thesis for this reason.

However, several attempts were made to troubleshoot this issue. We identified the following categories as causes for the lack of methane production:

1. We did not detect methane production because no methane was being produced in

the reservoir sediment during our sampling in July 2016.

2. We did not detect methane production because, although Harsha Lake sediment

was producing methane, the slurry incubations changed conditions that are

favorable to methanogenesis and the sediment slurry mesocosms somehow

inhibited methane production.

3. We did not detect methane production because, although the slurry mesocosms

were producing methane, there were errors in our sampling process or detection

of methane.

While care was taken to replicate the May sampling, there were several differences in the way the July sediment slurries were prepared. These differences were largely intentional, and the goal of the changes was to improve the method to prepare and sample the sediment slurry mesocosms.

74 Table A4.1 describes the differences in sediment slurry incubation methods from the May and

July 2016 sediment cores. Only differences are listed, so unless mentioned, it can be assumed that all other methods were the same between the two sampling periods.

Table A4.1. Differences in sediment slurry incubations between May and July 2016 sediment sampling. Potential May July consequence

Sediment sampling Re-used (after washing Sediment core tube liners New with mild soap and water) Incubation “start” Taken before purging Could have resulted 2 x 12 mL gas samples Not taken with N gas for stable in negative pressure taken before N gas purge 2 2 isotope analysis in the serum bottle Some (Didn’t equilibrates slurries for Unclear. Some of the sites 1-5; for sites 6-10, May slurries were Slurry incubations equilibrated after All (Equilibrated after equilibrated and experiments equilibrated sample taken on Day 3; N gas purge 2 some were not, but to atmospheric pressure for sites 11-15 (incubation start)) all resulted in CH equilibrated after N 4 2 production gas purge (incubation start)) Chemical used for control Mercuric chloride Chloroform (CHCl3) “kill” slurries* (HgCl2)

Gas sampling

Sample volume 11 mL 12 mL Likely none Pressure gauge could Syringe method and have affected slurries Pressure measurement Syringe method only pressure gauge (perhaps introduced oxygen) * Control “kill” slurries were prepared by adding a poison to kill methanogens to verify that CH4 production was biogenic. The “kill” slurries for May slurry mesocosms were unsuccessful, though, likely due to the binding of the HgCl2 to organic matter present in the sediment.

75 In addition to identifying potential differences in sediment treatment and sediment slurry mesocosm preparation, we sampled sediment cores in October 2016 and prepared slurry incubations to determine if we could detect methane production. On 9/30/16, triplicate sediment cores were sampled from Harsha Lake at EUS (Site #4). Nine total slurry incubations were prepared, with three treatments: control (no treatment), acetate addition, addition of methanogen culture (provided by Dr. Trinity Hamilton). The acetate addition slurries had three final concentrations of acetate: 20mM, 50mM and 100mM. These sediment slurry incubations were not prepared in a glove box, and the sediment cores were left refrigerated for ~2 weeks before extruding the top 5 cm and preparing the mesocosms. Gas sample volume was 12 mL, and the mesocosms were sampled 6 times over ~8 days. Pressure was NOT measured with the pressure gauge, but the “syringe method” of gas displacement was used. While the methane production rates were not calculated, the time courses for each of incubations are shown in figure A4.1 and

A 4.2.

Test Slurries A B C 8000

7000 control 6000

5000

4000

3000

60000 acetate 40000

CH4 (ppm) 20000

10000

8000 methanogen

6000

4000

50 100 150 50 100 150 50 100 150 Time (hours)

Figure A4.1. Methane concentrations (ppm) over time for the October slurry incubations. Note that the Y-axis scale is different for each treatment group (“control”, “acetate” or “methanogen”).

76

Test Slurries A B C

60000 control 40000

20000

0

60000 acetate 40000

CH4 (ppm) 20000

0

60000 methanogen

40000

20000

0 50 100 150 50 100 150 50 100 150 Time (hours)

Figure A4.2. Methane concentrations (ppm) over time for the October slurry incubations. This is the same data as Figure A4.1, but Y-axis scale is the same for each treatment group (“control”, “acetate” or “methanogen”).

Based on the results from these test slurries, it appears that the methanogens were substrate limited; with acetate addition there was a clear increase in methane production. However, in the control slurries, there was still a linear increase in methane concentration over time for at least 2 of the 3 replicates (“A” and “B”, Figure A4.1). The largest differences between this slurry preparation method and that of July is that July mesocosms were prepared in a glove box, the sediment cores were sectioned and subsampled within 24 hours, and the pressure gauge was used to measure pressure in the serum bottle (the pressure gauge was not used for these incubations).

In hindsight, the pressure gauge should have been tested by having a 4th treatment group that varied from the control only in the use of the pressure gauge. However, this is an area for future attention.

77