Location: Bowdoin / Environmental Studies / Many Rivers, One Estuary / Posters and Abstracts

Environmental Studies

Posters and Abstracts

Bioenergetics modeling of Atlantic cod in the Kennebec-Androscoggin marine system
Authors: Miguel Barajas, Dr. Karen Wilson & Dr. Theo Willis from the University of Southern Maine

The Wisconsin Bioenergetics Model (Hanson et al., 1997) was used to simulate yearly prey consumption for Atlantic cod, Gadus morhua, in the Kennebec-Androscoggin nearshore marine system using diet composition data of age-3 cod. Simulations were run by fitting the model to observed growth from January 2010 to December 2010. Alewives, Alosa pseudoharengus, a historically significant prey source in cod diets, have been negatively impacted by anthropogenic effects such as dam construction, pollution, and overharvesting (Atkins, 1887; Robbins and Lewis, 2008). Simulations were run in which the amount of alewives consumed by cod were reduced by 25, 50, 75, and 100%, to determine the degree to which growth would be suppressed. Another simulation was run to determine the amount of non-alewife prey that cod would need to consume to maintain original growth rates. The baseline simulation showed that alewives accounted for 24% of the biomass consumed and 39% of the energy consumed. Reducing 100% of alewives in cod diets reduced growth by 101%. To maintain original growth rates, cod would need to increase annual consumption of non-alewife prey by more than 26%. These bioenergetics modeling results shed light on the importance of specific prey items to the growth rates of fish and how the absence of particular prey items can drastically limit fish growth.

Ecological and Economic Recovery of the Kennebec and Androscoggin Rivers, Estuary, and Nearshore Marine Environment

Catherine Johnston,  Riker Wikoff, Daniel  Lowinger
Mentors: Professor John Lichter, Biology and Professor Guillermo Herrera, Economics

Human activities in the Kennebec and Androscoggin watersheds led to a collapse of the ecosystem in the second half of the twentieth century. Since then, water quality was able to rebound much faster than biotic components of the system. Improvements are still underway for populations of submerged aquatic vegetation, macroinvertebrates, and fish. Biology research this summer focused on mapping vegetation and surveying anadromous fish populations in the Kennebec estuary. The mapping data will be added to a time series describing vegetation change over the past 50 years in Merrymeeting Bay, the confluence of the Kennebec, Androscoggin, and four smaller rivers. Juvenile shad and river herring travel through the bay on their way to the Atlantic and submerged aquatic vegetation provides habitat for these important species. Weekly beach seines provided information on use of the bay by juvenile alosids and other fish species. Preserved stomach contents will be used to study juvenile alosid diet and promote an understanding of food web interactions in the river system. In addition, economics research examined the valuation methods used in river restoration, including hedonic pricing theory, travel cost method, and benefits transfer. The economics research culminated in running focus groups on the subject of Maine rivers with participants from across the state. The results from the focus groups will be used to develop a contingent valuation survey that values different scenarios of river restoration.

Use of Archival Datasets to Determine Seasonal Changes in Androscoggin Water Quality: 1943-2004
Authors:  Jennifer Lindelof, Department of Geology, Bates College, jlindelo@bates.edu, Beverly Johnson, Department of Geology, Bates College, bjohnso3@bates.edu, Mathieu Duvall, Bates Imaging and Computing Center, Bates College, mduvall@bates.edu, Katherine Stefko, Muskie Archives and Special Collections Library, Bates College, kstefko@bates.edu

Beginning in 1930, scientists began studying the condition of the Androscoggin River due to both its economic significance and rising concern over deteriorating water quality.  In 1943, Dr. Walter A. Lawrance was appointed Rivermaster by the Maine Supreme Court with the task of mitigating the effects of water pollution though effluent caps and legislation.  Dr. Lawrance sampled water at sites from Berlin, NH to Lisbon Falls, ME from the summer of 1943 to the summer of 1977, testing a range of water quality parameters including dissolved oxygen (DO), biological oxygen demand (BOD), water temperature, pH, and methylene blue stability among others. The goal of this project was to convert materials from the Lawrance Papers, into electronic format for ease of analysis and use on the web. Presented here is the first compilation of DO data from Turner Center Bridge through datasets available from Muskie Archives and the Maine Department of Environmental Protection. Data analysis indicates a major shift in dissolved oxygen over the tenure of Dr. Lawrance, with DO levels on average 4 ppm higher than initial values. This increasing trend appears to be correlated to shifts in both the manufacturing process as well as major policy shifts such as the Clean Water Act.  The long-term dataset will be presented in the context of both policy shifts and environmental significance.

Dynamics of Carbon Export from Maine Watersheds to the Gulf of Maine
Matthew Ramos ’12, mramos@bowdoin.edu
Mentor: Professor Phil Camill

Coastal ecosystems are highly influenced by the presence of organic matter in the form of carbon, nitrogen and phosphorus. Dissolved and particulate organic carbon (DOC/POC) are produced in terrestrial ecosystems and transported to the coast by rivers and streams through runoff processes. Carbon mobilization and transport is expected to change drastically in the near future due to changes in land use, temperature, and precipitation. The changing flux of carbon to the coast will alter the biogeochemistry of coastal ecosystems affecting their functionality, including the severity and frequency of harmful algal blooms and the productivity of local fisheries. To better understand how the changing flux of carbon will alter coastal ecosystems in Maine, we must first investigate the processes that factor into carbon mobilization and transport. I’m looking at how carbon export from terrestrial ecosystems into Maine’s major river systems (Androscoggin/Kennebec, Penobscot, and St. John) changes from spring thaw to summer, how direct measurements of DOC/POC correlate with optical proxies, estimating the flux of carbon to the Gulf of Maine, and how these values compare to what is known from satellite-based estimates of carbon delivery to the coastal ocean.

Selected abstracts from the 2010 Maine Water Conference

Edward Ames1, Andrew Bell2, Cory Elowe2, Catherine Johnston2, Eileen Johnson3, John Lichter3, Elsie Thomson2
1 Coastal Studies, Bowdoin College, Brunswick, ME.
2 Environmental Studies Program, Bowdoin College, Brunswick, ME.
4 Environmental Studies Program, Bowdoin College, Brunswick, ME; jlichter@bowdoin.edu;.ejohnson@bowdoin.edu

Bountiful fish populations in the Gulf of Maine once supported a thriving fishing industry. However, overfishing and poor management has left these populations at historic lows. Little is known about the movement and distribution of fish species throughout the gulf making development of sustainable management plans difficult. We use descriptions of historic fish populations to gain insight about the habitat and seasonal abundance distribution of seven fish species in the Gulf of Maine. We investigated the correlations between spatial characteristics such as substrate type and depth and historical fishing areas. Specifically, our analysis compared the location of substrate types collected from historic sources with current day substrate location and characteristics. Comparing substrate data from earlier historic records with current substrate data will help identify critical habitat characteristics of different fish species as well as continue to add to our understanding of fish movement and life history. We investigated strategies to characterize the substrate composition of the entire Gulf of Maine using sediment data from a variety of sources. We also investigated how current scientific sediment descriptions compared with the historic descriptions of sediment from fishermen. Lastly, we developed a grid system with which we hope to use spatial statistics to investigate whether characteristics such as depth and substrate can be used to predict the distribution of particular fish species. These advances will aid in future management and recovery plans for the Gulf of Maine.

Andrew Bell, Henry Berghoff, Cory Elowe, Holly Jacobson, Catherine Johnston, Benjamin Towne, John Lichter
Department of Biology, Bowdoin College, Brunswick, ME; jlichter@bowdoin.edu.

Merrymeeting Bay is a freshwater tidal ecosystem in Midcoast Maine that supports a diverse and complex food web. Historically, migratory waterfowl and anadromous fish thrived in the bay. Human activities led to a collapse of the ecosystem in the second half of the twentieth century. Since then, water quality of the bay was able to rebound much faster than the biotic components of the bay. Improvements are still underway for populations of submerged aquatic vegetation, macroinvertebrates, and fish. Our research in the summer of 2010 focused on surveying populations of the bay and Lower Kennebec to begin to understand the current state of the ecosystem and the implications for human use. Stationary nets and beach seines were used to collect a time series of data for fish in the bay. Benthic macroinvertebrate sampling sought to investigate a relationship between invertebrate density and diversity and presence or absence of vegetation. A Trimble GeoXM GPS unit was used to create a map of submerged aquatic vegetation in the bay and in future years, comparison with this map can reveal the extent of increased recruitment of plants. Continuing research in the bay will build on the foundation of knowledge established by this project and can contribute information that can be used to promote recovery. Although Merrymeeting Bay may never recover to the ecosystem it once was, advancements can be made through local environmental awareness and involvement and continued investigation of the changes and improvements that occur in the future.

Andrew Cardomone, Bowdoin College
Dharni Vasudevan
Bowdoin College, Brunswick, ME; dvasudev@bowdoin.edu.

The Androscoggin River has historically had varying levels of inorganic phosphate (o-P) input from many sources, including pulp and paper mills, agricultural runoff, and wastewater treatment plants. Since the passage of the Clean Water Act, water quality has greatly improved. As the river becomes cleaner, however, questions remain over whether the o-P currently bound to the sediment will reenter the water and adversely affect the ecosystem. To examine this dynamic, several sediment samples were collected at two locations along the river: Gulf Island Pond (GIP) and Merrymeeting Bay (MMB). GIP is an impoundment in Lewiston, Maine while Merrymeeting Bay is a freshwater, tidal ecosystem at the mouth of the Androscoggin. Sediments from each site were used in sorption experiments from which the equilibrium phosphorus concentration (EPC) was extrapolated. The EPC value has been extensively used to establish aqueous phosphorus concentrations at which there is no net phosphorus release or uptake by the sediment. At all locations, the EPC value was significantly less than measured aqueous o-P concentrations, suggesting the sediments will act as a sink (uptake) of o-P into the future. Furthermore, GIP sediments had higher o-P sorption capacities, surface areas, and extractable iron (ex-Fe) and aluminum (ex-Al) content compared to MMB sediments, indicating the capacity to act as a o-P sink is likely determined by surface area, ex-Fe, and ex-Al content. The relationship between sediment-Fe content and o-P retention also suggests that under anoxic conditions, the reduction of Fe in the sediment could cause o-P to be released into the river.

Barbara Arter, Mathew Dietert
Diadromous Species Restoration Research Network, University of Maine Orono, ME; barbara.s.arter@umit.maine.edu

Covering 8,570 square miles, the Penobscot River is Maine's largest and New England's second largest watershed. Unfortunately, centuries of dam construction have blocked the migration of diadromous fish to their upstream spawning and rearing habitats, as well as altered the structure and function of fish assemblages throughout the river. The Penobscot River Restoration Project is a multi-million dollar endeavor to restore nearly 1,000 miles of sea-run fish habitat by removing two large hydro-electric dams in the lower part of the river and providing improved fish passage at a third dam upstream. In 2008, the Penobscot River Restoration Trust and agency and academic researchers began conducting studies and environmental monitoring on the river in order to establish pre-dam removal conditions that will allow managers to document restoration outcomes. This group of approximately 30 researchers makes up the Penobscot Science Exchange, which is a collaboration with the Diadromous Species Restoration Research Network (DSRRN), a five-year, NSF-funded collaborative research effort to advance the science of diadromous fish restoration. This poster provides descriptions of research projects currently being conducted on the Penobscot in conjunction with the dam removals. Projects include shortnose sturgeon movement and spawning, bird assemblages, sea lamprey movement in tributaries, iron drainage impacts to water quality, alewife population structure and migration, marine-freshwater food web linkages, sea lamprey and Atlantic salmon interactions, and dam removal impacts on fish assemblages.

Beverly J. Johnson, Dava C. Wool
Geology Dept., Bates College, Lewiston, ME; bjohnso3@bates.edu.

The connectivity between fresh water and marine ecosystems is well illustrated by the life cycle of anadromous fish species (e.g., alewife). Alewives mature in marine systems and migrate to freshwater lakes to spawn, after which they return to the ocean. Northern New England rivers have a long history of human alteration (e.g., dams, industry, pollution), which has interfered with anadromous fish runs in freshwater systems. Little is known about the extent to which anadromous fish derived nutrients contribute to freshwater lakes in northern New England, nor the importance of marine derived nutrients prior to human alterations of these river systems. Nitrogen isotopes in sediment cores can track the historic presence of nutrients derived from anadromous fish in the Pacific NW, provided anthropogenic watershed disturbance is minimal (Finney et al. 2000, 2002, 2010). This study investigates nitrogen isotope composition of sediment cores over 500 years from the Androscoggin watershed. Surface cores were analyzed for 15N composition from four lakes: Basin Pond (no known alewives records, serving as a baseline); Thompson Lake, (no modern alewife runs but uncertain historical records); Tripp Pond and Taylor Pond (modern searun alewife runs). Preliminary data indicate δ15N enrichment in uppermost segments of Thompson, Tripp, and Taylor Pond cores, likely reflecting anthropogenic activities independent of the presence/absence of marine derived nutrients. The base of these cores may represent the degree to which MDN contributed to nutrient cycling in the lakes prior to arrival of western Europeans. Age models and spatial comparisons are currently underway.

Kathleen P. Bell1, Lynne Lewis2
1 School of Economics, University of Maine, Orono, ME; kpbell@maine.edu
2 Dept. of Economics, Bates College, Lewiston, ME

River restoration projects pose unique challenges for water resource managers because of, among other factors, uncertainties over project objectives, biophysical and social system responses, societal benefits and costs, and regulatory environments. Biophysical and social science researchers are actively striving to better confront these uncertainties and improve scientific support of water resource management decisions. By using statistical approaches to detect systematic patterns, our social science research strives to lessen uncertainties over public support for river restoration projects. Our empirical analysis is focused on public support for the Penobscot Restoration Project (Maine, USA). We make use of responses to a 2009/2010 survey of Maine households and employ regression analyses to test multiple hypotheses (informed by economics and other social science theories) about drivers of variation in public support. Our results shed light on the relative influence of factors such as proximity of residence to the Penobscot River, income, participation in outdoor recreation activities, perceived benefits and costs, project awareness, and trust in water resource management organizations.

Luke Salvato, Peter Lea
Geology Department, Bowdoin College, Brunswick, ME; plea@bowdoin.edu.

This project provides insights into phosphorus dynamics within Androscoggin Lake, a threatened lake designated as a highest priority lake on Maine’s Nonpoint Source Priority Watersheds List. Androscoggin Lake is unusual because it intermittently receives water and sediment related to back-floods of the Androscoggin River via the connecting Dead River. Compared to other lakes in the region, Androscoggin Lake exhibits relatively high values of phosphorus, part of which is attributed to reversed flow up the Dead River. Another source is internal loading of phosphorus from lake sediment back into the water column, but this mechanism has not been studied in detail. For this project, the lake was investigated at two primary sights, a deep central basin (~12 m deep) and a shallow basin in the northern region of the lake (~4.5 m deep) from June to early August 2010. Data include semi-weekly profiling of temperature, dissolved oxygen, and other variables, along with water sampling, and HOBO submersible temperature loggers. In the deep basin, the epilimnion deepened from ~5 m in June/mid- July to ~8 m in late July and August. Over the same time period, hypolimnetic dissolved oxygen dropped from ~50% to less than 10% saturation. At the shallow site, a well-defined hypolimnion did not develop, and the bottom remained relatively oxygenated (>60% saturation) throughout the study period. Laboratory phosphorus analyses provide insight into the dynamics of internal phosphorous loading under these different conditions.

Susan Davies1, Nathan Kane2
1 Maine DEP, Augusta, ME; Susan.P.Davies@maine.gov. 2 Bates College, Lewiston, ME.

Maine’s tiered water quality classification system is unique in the United States due to its built-in features that help to protect existing high quality waters and to retain gains that have been made to water quality in more heavily used rivers and streams. Since the legislative restructuring of Maine’s water quality standards and classification law in 1986, Maine has maintained four distinct water quality classifications for rivers (AA, A, B and C) that allow the State to optimize waterbody condition to the highest quality assigned through the public goal-setting process (MRSA Title 38 Ch 3 §464-468). The tiered classifications provide backstops against degradation and promote more precise regulation of freshwater in general. This policy has successfully upgraded many waterbodies to their highest attainable goal condition over the course of the intervening 23 years since passage. While the success of this policy is evident in the exceptional quality of Maine’s freshwater, this poster summarizes the first accurate documentation of the historical water quality progress that these upgrades represent. Created using ArcGIS and the legislative history of water re-classification, this poster helps to demonstrate the impact and extent of the State’s efforts in major river basins since the policy change 23 years ago.

Slade B. Moore1, Alex Abbot2
1 Maine Coastal Program, Maine State Planning Office, Augusta, ME; slade.more@maine.gov.
2 Gulf of Maine Coastal Program, U.S. Fish and Wildlife Service, Falmouth, ME.

Thousands of stream miles in Maine provide habitat for a diverse fauna and recreational/commercial opportunities having cultural and economic significance. Also important to Maine’s economy and the well-being of its people is the extensive road network allowing access to services, goods, and markets throughout the state. Where roads intersect streams, culvert crossings designed without consideration of ecological processes can represent physical barriers to stream connectivity. Barriers fragment habitat, constraining the movements of fish and other aquatic organisms and disrupting stream processes. Some barrier-crossings demonstrate a heightened risk of failure because they inadequately accommodate peak flows, which have increased over recent decades. Culverts that are undersized and perched above stream channels are the most common physical barrier to stream connectivity. Surveys in several Maine watersheds suggest that up to 90% of culvert crossings on perennial streams are ecological barriers for at least part of the year. About 50% of crossings surveyed, are severe, year-round barriers to connectivity on perennial streams. The considerable shift from native stream connectivity to a highly fragmented condition undermines longstanding conservation and recovery efforts focused on species serving key ecological roles and those that are prized by Maine’s people. The magnitude of this challenge requires an innovative, comprehensive effort over decades to incrementally correct barriers with ecologically functional replacement crossings. The Maine Stream Connectivity Work Group focuses on overcoming the technical and practical challenges of correcting barriers by building capacity for restoration, outreach, and coordination among Work Group participants and those responsible for road maintenance.

Zachary Ross, Lynne Lewis
Bates College, Lewiston, ME; llewis@bates.edu.

Literature suggests that in social science survey research, the preferences and opinions of persons who choose to respond to attitudinal questionnaires may be significantly different from those who choose not to respond. If this is true, extrapolations to the general population will be biased and can misinform policy decisions. This poster explores this question further by examining data from a 2010 survey of Penobscot River area homeowners. Early results suggest that the 576 responding households are not representative of the entire 1,904 household sample. Using previously available housing transaction data, a hedonic model was used to estimate effects of housing and environmental characteristics, including location in relation to the Penobscot River, on property values. In preliminary work, a significant difference was noted between the implicit values nonrespondents and respondents placed on proximity to the Penobscot. By adding a dummy variable and interaction term to the model, this behavior was determined to account for all significant differences between the subgroups. These results imply that survey nonrespondents implicitly attribute negative value to the Penobscot River, a result confirmed in the literature. However, it appears that this penalty does not exist among survey respondents. Furthermore, our results imply that behavior toward the river accounts for the entire significant difference between these two samples, an interesting conclusion. As a consequence, caution must be used in drawing larger generalizations from the survey data, as the focus of the survey was on behavior toward the river.