Missisquoi Bay, Lake Champlain is a eutrophic, northern shallow freshwater bay that experiences toxic cyanobacteria blooms during the summer months, largely as result of high nutrient (P and N) loading from the agricultural watershed. The sediments, which contain minerals that readily sorb P, can act as a sink or source of water column nutrients. Phosphorus, both inorganic and some organic forms, sorbs to metal oxides at neutral pH in the sediment, thus P release into overlying and pore water can be significantly affected by the reduction and subsequent solubilization of these oxides. This study addresses novel aspects of nutrient cycling in lake sediments as part of a larger study to better understand the link between phosphorus forms, mobility, and cyanobacteria blooms. These aspects include: 1) diel and seasonal sediment redox fluctuations and 2) the role of organic P (Porg) in overall P mobility within sediments as a function of depth and time. Missisquoi Bay sediment porewater redox chemistry was monitored across diel and seasonal cycles over the course of two summers (May-October, 2007 and 2008) by using in-situ voltammetry. Redox chemistry was monitored at the sediment-water interface (SWI) continuously over diel cycles, and the vertical concentration profiles of several key redox species (O2, Mn2+, Fe2+, and FeS(aq)) were obtained from cores collected at different times. The sediments were then analyzed for Total P (TP), Reactive P (RP), Porg, Mn, Fe, Ca, Al, Total Organic C and N. A bloom did not occur in Missisquoi Bay during the summer of 2007, but did in summer of 2008, providing an opportunity to compare the sediment chemistry between non-bloom and bloom conditions. Increasingly anoxic SWI conditions across summer 2008 were observed but the SWI remained oxic for the duration of summer 2007. Significant changes in diel cycle redox chemistry at the SWI were also detected in both summers. Reactive P in the surface sediments decreased across the 2008 season but not in 2007. A strong correlation found between RP and RFe (operationally defined as Fe(III)OOH) suggests that a significant portion of sediment P (30-40%) is closely associated with Fe(III)OOHs, which are susceptible to reduction in anoxic conditions. Phosphorus mobility from the sediment into the water column can be limited by the amount of Fe(III)OOH at the surface, thus P flux from the sediments would be greatest when reducing conditions promote solubilization of these minerals. Completely anoxic surface sediments were only observed during the presence of a bloom, explaining the loss of RP in the surface sediments in 2008 in the late summer. Organic P species represent 18-26% of the P in sediments and the lack of a definite, consistent trend of Porg fractionation across the season suggests that there is variable mobility and degradation of these complex organic compounds on small timescales. The loss of RP from the sediment in 2008 could have contributed to an estimated water column P increase on the order of thousands of μg/L, which in addition to measured increases in NH4+ gradients and subsequent N flux estimates in the upper sediment, could have sustained the bloom for an extended period of time. The relationship between the bloom and reducing sediment conditions suggest that bloom dynamics enhance nutrient release from the sediments, allowing for proliferation and sustainability of the bloom.

In shallow lake systems, phosphorus (P) availability to cyanobacteria populations is often controlled by the release (internal loading) or sequestration of sediment P. This study provides novel insight into the feedbacks between the water column and benthic P pools across multiple time scales and explain how these dynamics influence chemical partitioning of P in lake sediment. Phosphorus partitioning in seasonal sediment core time series collected from a shallow eutrophic bay of Lake Champlain were quantified with SEDEX and enzyme hydrolysis selective extraction schemes. Time series extraction data were interpreted with concurrent water column physical and biogeochemical monitoring data to examine the relationship between water column dynamics and P partitioning of near-surface sediments in this intensively monitored system. Nonmetric multidimensional scaling analysis (NMDS) indicates that both sediment and water column time series cluster seasonally, linking water column variables such as pH, thermal stratification, and dissolved oxygen concentrations to the behavior of sediment P pools over the course of a year. Iron (FeP), exchangeable (Ex-P), calcium carbonate bound P (Ca-P) pools, and enzyme labile P were highly dynamic, especially in spring and summer. The SEDEX concentration data indicated that the sediment was mainly composed of inorganic bound P (De-P), but FeP and Ex-P pools proportionally varied most between sampling dates. Remarkably, while highly dynamic on an intra-annual timescale, the sediment ultimately returned to similar P concentration and chemical partitioning by late fall. The hysteretic nature of this interaction between water column dynamics and sediment P inventory/partitioning was clearly driven by systematic seasonal changes in water column physical, chemical, and ecological conditions governed by northern Vermont’s climate and the physical configuration of the bay and its watershed. This study provides novel insight into the unique challenges associated with improving water quality in lake systems impacted by internal loading of legacy P.

This study was conducted to evaluate the relative roles of trophic dynamics and nutrient concentrations in the development of cyanobacteria blooms. The motivation for this research was to gain insights into how food webs respond to ecosystem-scale changes, using Lake Champlain as a case study. I sought to link field-based observations with experimentally derived data on mechanisms to better understand the processes that drive cyanobacteria blooms. My research addressed three specific topics: (1) associations among phytoplankton and nutrient concentration trends over time, (2) the impacts of planktivory by invasive fish on the ambient zooplankton community, and (3) the role of herbivore zooplankton grazers in determining the composition of the phytoplankton community. I found little evidence of a strong association between nutrient concentrations and phytoplankton community composition during summer months in shallow bays of Lake Champlain prone to annual cyanobacteria blooms. Fish diet analysis indicated that invasive white perch (Morone americana) and alewife (Alosa pseudoharengus) selectively graze on large zooplankton, which has likely led to substantial declines in zooplankton biomass. I used these results to inform the design of a mesocosm study, which tested the effects of zooplankton grazing on phytoplankton and provided support for the theory that large zooplankton grazing pressure changes the size structure, abundance and composition of phytoplankton. High nutrient concentrations support increased levels of ecosystem productivity, but cascading trophic dynamics are additional forces that are likely contributing to the determination of phytoplankton community composition. Collectively, my research suggests that in shallow bays of Lake Champlain, selective grazing by invasive planktivorous fish is shifting the size structure of the zooplankton grazer community and has likely contributed to conditions that favor dominance by cyanobacteria in summer.

Stonecats (Noturus flavus, Rafinesque 1818) are designated as a Vermont state-listed endangered species because their known distribution is limited to two systems, the LaPlatte and Missisquoi rivers. The restricted distribution and lack of knowledge on abundance in either river is cause for concern in the continued survival of these populations. Based on the capture numbers and large size range of individuals, we predicted that the population in the LaPlatte River, which provides quality benthic habitat, is stable. However, the Missisquoi River population has the potential for increased intermittent mortalities from two sources, lampricide (3-trifluoromethyl-4-nitrophenol) treatment every four years and dewatering during drought conditions. In 2012, 2013, and 2014 we captured, PIT tagged (> 90 mm total length), and VIE marked all Stonecats collected using backpack electrofishing and minnow traps in the LaPlatte and Missisquoi rivers. A total of 1252 were PIT tagged in the LaPlatte River and 125 in the Missisquoi River. First we estimated survival and seniority of Stonecats in the LaPlatte River, using the Pradel model in Program Mark, and derived an average annual lambda of 0.9826. The population estimates from the LaPlatte River were modeled in a population viability analysis (PVA). Few Stonecats were captured in the Missisquoi River, so we used the PVA model to estimate the extinction rates with increased intermittent mortalities on 4-, 6-, and 8-year cycles to predict the long-term viability of the population. With an initial number of 2000 individuals, the population became extinct 100% of the time with an increase in mortality of 0.1 on a 4-year cycle. Our results indicate that the LaPlatte River population is stable, but the Missisquoi River population, in the area affected by lampricide, is not. These results are informative for developing future management scenarios, however, our approach has uncertainty that can only be addressed through obtaining more data on the Missisquoi River population.

Rivers deliver significant macronutrients and sediments to lakes that can vary substantially throughout the year. These nutrient and sediment loadings, exacerbated by winter and spring runoff, impact aquatic ecosystem productivity and drive the formation of harmful algae blooms. The source, extent and magnitude of nutrient and sediment loading can vary drastically due to extreme weather events and hydrologic processes, such as snowmelt or high flow storm events, that dominate during a particular time period, making the temporal component (i.e., time over which the loading is estimated) critical for accurate forecasts. In this work, we developed a data-driven framework that leverages the temporal variability embedded in these complex hydrologic regimes to improve loading estimates. Identifying the "correct" time scale is an important first step for providing accurate estimates of seasonal nutrient and sediment loadings. We use water quality concentration and associated 15-minute discharge data from nine watersheds in Vermont's Lake Champlain Basin to test our proposed framework. Optimal time periods were selected using a hierarchical cluster analysis that uses the slope and intercept coefficients from individual load-discharge regressions to derive improved linear models. These optimized linear models were used to improve estimates of annual and "spring" loadings for total phosphorus, dissolved phosphorus, total nitrogen, and total suspended loads for each of the nine study watersheds. The optimized annual regression model performed ~20% better on average than traditional annual regression models in terms of Nash-Sutcliffe efficiency, and resulted in ~50% higher cumulative load estimates with the largest difference occurring in the "spring". In addition, the largest nutrient and sediment loadings occurred during the "spring" unit of time and were typically more than 40% of the total annual estimated load in a given year. The framework developed here is robust and may be used to analyze other units of time associated with hydrologic regimes of interest provided adequate water quality data exist. This, in turn, may be used to create more targeted and cost-effective management strategies for improved aquatic health in rivers and lakes.

Globally, habitat fragmentation has had a major impact on the conservation and management of many species and is one of the primary causes of species extinction. Habitat fragmentation is loosely defined as a process in which a continuous habitat is reduced to smaller, disconnected patches as the result of habitat loss, restriction of migration or the construction of barriers to movement. Aquatic systems are particularly vulnerable to habitat fragmentation, and today an estimated 48% of rivers are fragmented worldwide. My dissertation evaluates how habitat fragmentation has influenced the populations of four different species of fish in the Lake Champlain basin. In chapter 1 I summarize the current state of habitat fragmentation research, I broadly describe habitat fragmentation, review how habitat fragmentation pertains to population genetics, and describe the legacy of habitat fragmentation in the Lake Champlain basin. In chapters 2, 3 and 4 I evaluate and discuss the impact of nine lake causeways on the population structure of slimy sculpin (Cottus cognatus), rainbow smelt (Osmerus mordax), and lake whitefish (Coregonus clupeaformis). The genetic effects of causeways are limited. However, causeways appear to have had a significant influence on rainbow smelt demographics, and the genetic structure observed in lake whitefish may be a product of reduced effective population size resulted from commercial harvest in the late 1800s. In chapter 5 I evaluate how the basin-wide population of tessellated darters (Etheostoma olmstedi) is naturally structured throughout Lake Champlain and three different major tributaries and evaluates the effect that different types of habitat fragmentation (dams, causeways, and natural fall lines) have on tessellated darter populations. Tessellated darters appear to be highly structured by river drainage but not by dams, causeways or fall lines. My dissertation highlights how comparative population genetic studies can be used to identify patterns of isolation within large populations. My results stress the value of reporting both the presence and absence of barrier induced population sub-structuring.

Mass animal migrations represent large movements of biomass, energy, and nutrients with predictable patterns and important ecosystem-level consequences. Diel vertical migration (DVM) in aquatic systems, the daily movement of organisms from deeper depths during the day to shallower depths in the water column at night, is widespread in freshwater and marine systems. Recent studies, however, suggest partial migration behavior, whereby only some portion of a population migrates, is the rule rather than the exception in a range of migratory fauna, including those that undergo DVM. Hypotheses to explain why partial migrations occur complicate traditional views on DVM and challenge conventional theories. I address intraspecific variation in DVM behavior of an aquatic omnivore, Mysis diluviana, to test several long-standing assumptions about benthic-pelagic DVM in Mysis. I evaluated the extent of partial DVM and several potential drivers within a Lake Champlain Mysis population. I used traditional net-based field observations, a novel deep-water video camera system, and a laboratory experiment, to compare distributions, demographics, abundance estimates, hunger-satiation state, and feeding behavior, of migrant and non-migrant Mysis across multiple seasons, habitats, and different times of the day. Findings from my dissertation suggest Mysis partial DVM is common, and is associated with body size and demographic differences among individuals. Partial DVM behavior, however, did not correspond to strong differences in feeding preference or hunger-satiation state of individuals. My results contribute toward a more comprehensive understanding of migration theory and mysid biology, by including the often overlooked, but important, benthic habitat component of DVM studies, and fills in several ecological knowledge gaps regarding a key omnivore in many deep lake food webs across North America where Mysis serve as both predators and prey to many organisms.

The advent of steam contributed heavily to the economic transformation of early America, facilitating trade through the transportation of goods along the country’s lakes, rivers, and canals. Serious experimentation with steam navigation began in the last quarter of the 18th century. By the turn of the 19th century, fledgling US steamboat companies vied for control of navigation rights in the country’s northern waterways. The second steamboat to be launched on Lake Champlain, Phoenix, operated as a passenger steamer between 1815 and 1819, when she caught fire and sank in the lake. The intention of this study is to advance our knowledge of early steamboat design and use in the United States through the archaeological investigation of the country’s earliest-known steamboat wreck. As little is known about the development of these early steam vessels, the study of Phoenix offers a unique opportunity to gain new information related to steamboat design in the early 19th century as well as a glimpse into life on the lakes and rivers of North America during this era. The dissertation presents detailed information on Phoenix’s construction, operation, and sinking based on historical and archaeological analysis and interpretation. In combination with the available archival record and analytical comparisons with steamboats of similar size and age, a more comprehensive understanding of the developmental phases of steam travel and its impact on early America can be gained.

Recruitment failure of lake trout (Salvelinus namaycush) in the Great Lakes has been attributed in part to the consumption of alewife (Alosa pseudoharengus) by adult lake trout, leading to Thiamine Deficiency Complex (TDC) and early mortality in fry. The current understanding of thiamine deficiency in lake trout fry is based on information from culture and hatchery settings, which do not represent conditions fry experience in the wild and may influence the occurrence of TDC. In the wild, lake trout fry have access to zooplankton immediately following hatching; previous studies found that wild fry begin feeding before complete yolk-sac absorption. However, hatchery-raised fry are not provided with food until after yolk-sac absorption, long after the development of TDC. Zooplankton are a potential source of dietary thiamine for wild fry in the early life stages that has not previously been considered in the occurrence of thiamine deficiency. We postulated that wild-hatched fry could mitigate thiamine deficiency through early feeding on natural prey. Specifically, we hypothesized 1) feeding should increase thiamine concentrations relative to unfed fry and 2) feeding should increase survival relative to unfed fry. Feeding experiments were conducted on lake trout fry reared from eggs collected from Lake Champlain in 2014 and Cayuga Lake in 2015. A fully crossed experimental design was used to determine the effect of early feeding by lake trout fry in thiamine replete and thiamine deplete treatments before and after feeding. Overall, thiamine concentrations and survival did not significantly differ between fed and unfed fry. Thiamine concentrations increased from egg stage to hatching in both years, suggesting a potential source of thiamine, which had not previously been considered, was available to the lake trout eggs during development.