Grantee Research Project Results
Final Report: Development of Alternative Approaches to Assessing the Impact of Pollutants on Environmental Systems: Part 2: Management of Environmental Quality in Urban Watershed Ecosystems
EPA Grant Number: CR830912Title: Development of Alternative Approaches to Assessing the Impact of Pollutants on Environmental Systems: Part 2: Management of Environmental Quality in Urban Watershed Ecosystems
Investigators: Driscoll, Charles T. , Effler, Steven W.
Institution: Syracuse University , Upstate Freshwater Institute, Inc.
EPA Project Officer: Packard, Benjamin H
Project Period: June 27, 2003 through June 30, 2007 (Extended to June 28, 2009)
Project Amount: $2,993,493
RFA: Targeted Research Grant (2002) Recipients Lists
Research Category: Targeted Research
Objective:
Syracuse University (SU), in collaboration with Upstate Freshwater Institute, the New York Indoor Environmental Quality (NYIEQ) Center, Tetra Tech Inc., Smith College, Cornell University and the Adirondack Lakes Survey Corporation (ALSC), conducted an interdisciplinary research program strategically targeted at developing the fundamental science, enabling technologies and technology transfer pathways for intelligent environmental quality systems (i-EQS) appropriate for aquatic ecosystems. Such systems would integrate advances in monitoring and modeling to yield dramatic improvements in information available for decision-making concerning the management and control of water quality in a variety of aquatic ecosystems, including urban and forested settings that are subjected to multiple disturbances.The research program initiated a unified approach to develop an i-EQS that was designed to monitor and improve understanding of environmental quality across a broader range of spatial and temporal scales than are possible with traditional approaches. Central to our vision is the coupling of monitoring activities with models through intelligent control and data management. By linking monitoring with modeling it is possible to obtain insight on the role of disturbances in regulating the structure and function of ecosystems and the interactions of multiple disturbances. This approach was applied to two distinct aquatic ecosystems: the Seneca River watershed in Central New York, which is a predominately urban ecosystem, and the Adirondack region of New York, which is predominately forested. For each ecosystem, the objectives include: (1) development of reliable quantitative and timely information concerning water quality status; (2) improved scientific understanding of the processes and drivers that regulate water quality and ecological conditions; and (3) development of models that effectively integrate the scientific understanding of ecosystems and monitoring information, to provide a quantitative representation of processes and their interactions, as well as reliable predictions. For the Seneca River ecosystem, we investigated the interactions of elevated nutrient inputs (from domestic waste and agricultural activities), industrial wastes, and the invasion of zebra mussels, an exotic bivalve, Dreissena polymorpha. For the Adirondack region, our research was focused on the deposition and cycling of mercury within two lake and forest ecosystems that are representative of the numerous lakes in this region, including the interactions between atmospheric deposition of mercury and strong acids with the bioaccumulation of mercury in fish.
Summary/Accomplishments (Outputs/Outcomes):
Seneca River Ecosystem: Urban areas, with their dense populations and concentrated industrial activities, place great demands and stress on proximate aquatic resources and ecosystems. Associated waste discharges require adequate assimilative capacity in receiving waters. Surface waters serving as water supplies for these urban areas represent striking contrasts. The pressure of ever-increasing growth and associated needs to maximize utilization and protection of our urban aquatic resources requires innovations in monitoring, information transfer and dissemination, and mathematical modeling. In response to these needs, we envision an intelligent urban environmental system (i-UES) that monitors water quality of an array of aquatic systems within an urban ecosystem, and supports control decisions for the use of these resources while protecting ecosystem and human health. The i-UES would include: (1) a network of robotic water quality monitoring units deployed in both water supply surface waters and receiving waters; (2) near-real-time (NRT) data delivery, analysis and model forecasting capabilities; and (3) a robust communication system linking sensor and model output to automated control sites or resource managers to support actions. The capabilities of the i-UES can be expanded where the results of related process studies are available and as technologies improve. The overall goal of the research was to advance an i-UES for an urban network within the Seneca River watershed through the development, testing and implementation of a robotic monitoring network, a related data transfer system and mathematical models, the conduct support process studies and limnological analyses, and the demonstration of NRT modeling.
Robotic Monitoring Network:
Robotic monitoring platforms have successfully collected important water quality information on components of this study network and delivered the data, in near-real-time (NRT), to an array of stakeholders at www.ourlake.org. These activities have served to: (1) educate the public; (2) provide key information to managers; (3) advance the protocols for maintenance and operation of robotic monitoring networks; and (4) demonstrate the attributes of the robotic platforms for addressing research issues related to the functioning of aquatic ecosystems. Research and ecosystem management capabilities documented in the peer reviewed literature that integrated the robots have addressed: (1) the impacts of zebra mussel metabolism on the water quality of a river; (2) the diagnostic value of robotic monitoring of turbidity in a polluted lake; (3) the utility of a reference lake to assess progress for a lake targeted for rehabilitation; and (4) the effective integration of robots into a "dual discharge scheme" that would allow temporal variation of the point of waste discharge. The Onondaga Lake robot is now providing critical data to managers and stakeholders as part of the monitoring program for the Superfund site and to track changes in water quality in response to the implementation of advanced treatment of nutrients at the Syracuse Municipal Wastewater Treatment Facility. This robot, together with an Onondaga Creek robot, were central to recent (2008) work that established the seasonal plunging of this stream in the lake; described in a manuscript published in the Journal of the American Water Resources Association (Effler et al. 2009). This was critical information to management programs for these systems. Specifically, it lead to discontinuation of an expensive infrastructure project on Onondaga Creek ($150 million) that would not have resulted in the intended environmental benefit.
Mathematical Models:
Mathematical models are being developed to serve as quantitative predictive tools to support research activities and management applications. Models serve to integrate multi-disciplinary information and test our understanding of ecosystems and processes. Substantial advancements in the modeling program were made in 2007 and 2008 that will help to guide managers in ongoing efforts to rehabilitate Onondaga Lake, as well as contribute to modeling efforts for other perturbed systems. These advancements have included: (1) the set-up and testing of a onedimensional stratification/transport model for Onondaga Lake; (2) the set-up and testing of a two-dimensional stratification/transport model for the lake; (3) the set-up and testing of a threedimensional stratification/transport model for the lake; (4) the set-up and testing of a two dimensional wave model for the lake; (5) the development, testing and application of an optics model for the lake; (6) the set-up and application of radiative transfer model to predict remote sensing reflectance (Rrs) from measurements of backscattering and absorption; (7) the set-up and preliminary testing of a one-dimensional nutrient-phytoplankton model for the lake; and (8) a post-audit analysis and upgrade of a two-layer nitrogen model for the lake. These advancements and their utility are summarized below.
(1) One-dimensional stratification/transport model. The adopted model is UFILS4, which evolved from CE-THERM-R1 (U.S. Army Corp). This is an integral, or mixed layer, dynamic model. The model was demonstrated to perform well for 19 years of observations in simulating both temperature and salinity. The model was applied in a probabilistic format to evaluate the effects of natural meteorological variations on the timing of oxygen loss from the hypolimnion. It is expected to serve as the transport framework for the primary water quality management model for the lake. An associated manuscript was produced that will appear in Fundamental and Applied Limnology (O’Donnell et al. 2009).
(2) Two-dimensional stratification/transport model. The adopted model was the transport submodel (W2) of CE-QUALW2 (U.S. Army Corp). This model segments the lake longitudinally and vertically. W2 is the most widely used two-dimensional transport model. It has been set-up and preliminarily tested for six years for Onondaga Lake, and is being used to evaluate the interplay with potential vertical transport of mercury from the hypolimnion of the lake. Moreover, it may also be used to evaluate deployment location options for hypolimnetic additions of oxygen or nitrate, under consideration to abate mobilization of mercury from the pelagic sediments.
(3) Three-dimensional stratification/transport model. The adopted model was ELCOM (from Centre for Water Research, University of Western Australia). This selection was based on the specific capability of simulation of the behavior of density currents. The model was preliminarily set-up and applied in 2008 (Owens 2008) to simulate the plunging behavior for Onondaga Creek, the primary urban stream entering the lake. The model is presently being tested against the results of a dye release study conducted in 2009 and the turbidity patterns imparted from runoff events, also resolved in 2009 (Owens et al. 2009). This tool will be used to evaluate Superfund site issues for the lake, including: (1) the positioning of deployment locations for either oxygen or nitrate in the hypolimnion to abate Hg release from the sediments; and (2) potential turbidity impacts from dredging activities. This model will also be used for further domestic waste management issues related to the effects of Onondaga Creek plunging on effective phosphorus loading and public health effects. A manuscript on the testing of the model against the dye and turbidity patterns is under preparation (Journal of the American Water Resources Association, Owens et al. 2009).
(4) Two-dimensional wave model. The adopted model was the Donelan/GLERL framework that is based on the wave energy equation. Direct measurements of wave pressure supported testing of the model. The wave model and local sediment trap studies were integrated to guide a simple off-shore transport model that supported estimates of mercury resuspension from contaminated industrial shoreline waste deposits. The results support the planned remediation of these deposits as part of the Superfund cleanup project. This study was documented in a journal manuscript, Journal of Environmental Engineering, ASCE (Owens et al. 2009).
(5) Optics Model. A probabilistic mechanistic model for Secchi disc (SD) and the light attenuation coefficient (kd), that represents the effects of multiple constituents on the regulating processes of absorption and scattering, was developed and tested for the lake’s long-term optics record. This work was published in Fundamental and Applied Limnology (Effler et al. 2008). The model serves to establish realistic expectations for potential future improvements and represents the important roles of inorganic particle inputs and Daphnia grazing.
(6) Radiative Transfer Expression. A widely applied radiative transfer expression that estimates remote sensing reflectance (Rrs) based on the backscattering (bb) and absorption coefficients (a) was tested . This testing was supported by direct measurements of bb, a, and Rrs in Otisco, Owasco, Skaneateles and Onondaga Lakes. Good closure between these independent measures was achieved through this expression. This supports the credibility of the independent measurements, thereby opening the way to project changes in Rrs in response to changes in a or bb from management actions. The results of the analysis were presented at a professional meeting (Kalanack et al. 2008).
(7) and (8) Water Quality Modeling. The one-dimensional water quality (nutrientphytoplankton) model uses UFILS1 as the transport framework and simulate the dynamics of nutrient cycling and phytoplankton biomass. It has been set-up and preliminary tested for several years. This model, or a modified version of it, is expected to serve as the primary framework to support ongoing deliberations for phosphorus control programs for the lake. Post audit analyses with the two layer nitrogen model served to evaluate and refine our understanding of kinetic processes regulating the levels of important forms of nitrogen. These insights will be integrated into larger ongoing modeling initiatives.
Remote Sensing:
To advance the development of mechanistic approaches for remote sensing for inland waters, the research team focused on partitioning the contributions of various constituents to light absorption and scattering and assessment of the related emergent light flux signatures. These efforts addressed several of the network systems and yielded insights on optical characteristics. The study systems represented a rich range with respect to magnitudes and dynamics in these optical features. A related manuscript on the dynamics of components of the absorption coefficient (a) in Onondaga Lake will be published in Fundamental and Applied Limnology (Perkins et al. 2009). Direct in-situ measurements of absorption and scattering in the water columns of the study lakes were demonstrated, through application of the radiative transfer expression (described above), to close (quantitatively consistent) well with paired direct measurement of remote sensing reflectance, the signal available for remote sensing reflectance. This closure opens the way to develop new and improved mechanistic optics models to support remote sensing for inland (case 2) waters. These approaches may have broad application for the growing interest and initiatives for advancing remote sensing capabilities for inland waters. Direct measurements of remote sensing reflectance are presently being made from the Onondaga Lake robot.
Process Studies and Limnological Analyses:
Mercury. Mercury fluxes to three of the network lakes have been assessed through a combination of mercury and lead-210 analyses of sediment cores. The recently reported (peerreviewed publication) findings demonstrated a range of anthropogenic effects; two lakes demonstrated increases from land use effects; while all were affected by regional atmospheric inputs (Bookman et al. 2008). Paleolimnological analyses of Otisco Lake depicted increased sedimentation rates starting in the early 1900s, shifts in the contributions of inorganic carbon (e.g., calcite) to sedimentation, and the effect of application of copper sulfate as an algicide.
This work is summarized in Bookman et al. (in press).
Sediment Traps. Long-term and seasonal patterns in the deposition of suspended solids, total (TSS) and the fixed (FSS non-volatile; e.g., inorganic) and volatile (VSS e.g., organic) components were assessed for the 1980-2008 interval. The fixed suspended solids component was partitioned according to CaCO3 and non-CaCO3 contributions. These results and related interpretations will be published in the peer reviewed literature Lake and Reservoir Management (Hurteau et al. 2009). Decreases in the deposition of these constituents occurred associated with the closure of an industry and decreases in phosphorus loading from a domestic waste facility. These data are important for the lake’s overall sediment budget and in considerations related to the “monitored natural recovery” approach for rehabilitation of this Superfund site.
Recently, long-term and seasonal patterns in the deposition of carbon, nitrogen, phosphorus, and chlorophyll were determined for the 1980-2008 interval. These fluxes quantify an important feature of the cycling of these constituents for the lake’s domestic waste problems. Nonphytoplankton forms of phosphorus were demonstrated to contribute importantly to the phosphorus pool of the lake’s upper waters. These trap results and related evaluations were presented at a professional meeting (Matthews et al. 2009); a related manuscript is under preparation.
Zebra Mussels. The utility of zebra mussels as a biological monitor of mercury is under evaluation, through the analysis of collections from mercury impacted and three reference lakes within the urban network. Collections were made in Oneida, Skaneateles, Otisco and Onondaga lakes. Tissue concentrations were the highest by far in contaminated Onondaga Lake. However, the second highest levels were in Skaneateles Lake, the most oligotrophic of the network study lakes. Related analyses and interpretations of the salient findings were presented at a professional meeting (Blackwell et al. 2009) and will be published in peer-reviewed literature.
Adirondack Ecosystem: Mercury cycling in lake-watershed ecosystems involves the interactions of complex biogeochemical processes. Recent studies have indicated that the concentrations of mercury in water and fish are strongly influenced by atmospheric deposition of mercury and strong acids, and the geochemical and biological processes occurring in lakes and their surrounding watersheds. Several mechanisms may contribute to the linkage between the enrichment of fish mercury and atmospheric deposition. Sulfate inputs may enhance conversion of ionic mercury to methyl mercury by methylation mediated by sulfate-reducing bacteria. Acidification may decrease lake biological productivity, resulting in higher concentrations of mercury in aquatic biota. Finally increases in concentrations of dissolved organic carbon may increase binding of ionic mercury and/or methyl mercury, thereby decreasing bioavailability of mercury. We envision that these processes affect the response of fish mercury concentrations to recent and future declines in atmospheric mercury and strong acid deposition within remote watersheds.
The overall research objective in this component was to evaluate the response of fish mercury to changes in atmospheric deposition of mercury and strong acids by investigating mercury in lakewatersheds in the Adirondack region of New York and applying a complex mercury biogeochemical cycling model. Detailed mercury cycling studies of atmospheric mercury deposition and fate were conducted at two intensive lake-watersheds in the Adirondacks, with a primary focus on the Arbutus Lake watershed at Huntington Forest, New York. Sunday Lake water chemical parameters differ from that of Arbutus Lake, providing a comparison of the effects of water chemistry on watershed mercury flux. These studies included investigation of upland flux processes and watershed fate, including an assessment of event flux. The upland study involved a characterization of atmospheric mercury deposition, throughfall, plant leaf tissue and litter, soil and soil solutions, and evasive flux. Investigation of watershed flux process involved an extensive wetland porewater and surface water sampling in addition to soil, vegetation, and evasive flux analysis. Mercury dynamics were also assessed within each lake. Event sampling was also conducted to assess their influence on mercury flux processes. Field experimentation was conducted over a two-year period, ending in August 2006.
Mercury accumulated in plant leaf tissue over the growing season, with concentrations varying by species. Litterfall constituted the primary annual input to upland forest soils, approximately double the throughfall input (Bushey et al. 2008). Litter decomposition studies, supported by soil and soil water depth profile results, suggest that mercury largely remained in the surficial soil layer (Bushey et al. in review). Soil water drainage mercury losses, as determined through model estimation of water flux, were low, with concentrations of total mercury and methyl mercury near the analytical detection limit. Evasive mercury flux dominated the ecosystem losses, demonstrating the importance of understanding temporal variability in assessing upland flux processes.
Within the watershed study, wetlands strongly influenced mercury flux, particularly for methyl mercury, relative to that for the upland stream. Porewater collected within the wetland exhibited high mercury concentrations, particularly for the methyl mercury. Dissolved organic carbon and sulfate influenced stream and porewater concentrations of mercury species with wetland waters demonstrating seasonal effects. As in the upland ecosystem, evasive flux was dependent upon environmental conditions, particularly incident radiation and temperature, reinforcing the temporal dependence of measurement. Event results demonstrated the importance of flow conditions on sampling time, as flux increased substantially for elevated discharge conditions relative to base conditions. Dissolved organic carbon and total suspended solids were both important drivers for increasing mercury flux. However, volume-weighted methyl mercury concentrations increased only for the upland watershed, demonstrating the importance of hydrologic conditions on mercury flux. Results from these detailed studies have been summarized in a series of papers (Bushey et al. 2008; Selvendiran et al. 2008; Selvendiran et al. 2009a,b; Demers et al. in review).
In this research we also sought to apply the understanding developed from the intensive studies of lake-watershed ecosystems to sites throughout the Adirondack region. This initiative involved a resurvey of 25 lakes for concentrations of mercury in the water column and fish that were previously surveyed in 1993. For each of the 25 resurveyed lakes, patterns of water column mercury species and fish mercury were analyzed to evaluate if changes in lake concentrations of mercury species or fish mercury occurred. Changes in water chemistry and fish mercury concentrations varied by lake. Twelve lakes showed a decrease in perch mercury, six lakes showed an increase, and in seven lakes perch mercury did not change. These data suggest that there are four key variables influencing the change in perch mercury concentrations in the Adirondacks: watershed area, elevation, change in pH, and change in fish condition. The results from our study have led us to hypothesize that as the acidity in lakes is attenuated, the lakes may become more productive and water quality conditions less stressful to fish leading to increasing in fish condition. As fish body condition and growth rates improve, fish can exhibit “growth dilution” of tissue contaminants leading to lower fish mercury concentrations. This work is summarized in Dittman et al. (in 2009).
Journal Articles on this Report : 17 Displayed | Download in RIS Format
Other project views: | All 47 publications | 18 publications in selected types | All 18 journal articles |
---|
Type | Citation | ||
---|---|---|---|
|
Adams RM, Twiss MR, Driscoll CT. Patterns of mercury accumulation among seston in lakes of the Adirondack Mountains, New York. Environmental Science & Technology 2009;43(13):4836-4842. |
CR830912 (Final) |
Exit |
|
Bookman R, Driscoll CT, Engstrom DR, Effler SW. Local to regional emission sources affecting mercury fluxes to New York lakes. Atmospheric Environment 2008;42(24):6088-6097. |
CR830912 (2008) CR830912 (Final) |
Exit Exit Exit |
|
Bushey JT, Driscoll CT, Mitchell MJ, Selvendiran P, Montesdeoca MR. Mercury transport in response to storm events from a northern forest landscape. Hydrological Processes 2008;22(25):4813-4826. |
CR830912 (2008) CR830912 (Final) |
Exit |
|
Bushey JT, Nallana AG, Montesdeoca MR, Driscoll CT. Mercury dynamics of a northern hardwood canopy. Atmospheric Environment 2008;42(29):6905-6914. |
CR830912 (2008) CR830912 (Final) |
Exit Exit Exit |
|
Denkenberger JS, Driscoll CT, Effler SW, O’Donnell DM, Matthews DA. Comparison of an urban lake targeted for rehabilitation and a reference lake based on robotic monitoring. Lake and Reservoir Management 2007;23(1):11-26. |
CR830912 (2008) CR830912 (Final) |
Exit |
|
Denkenberger JS, O’Donnell DM, Driscoll CT, Effler SW. Robotic monitoring to assess impacts of zebra mussels and assimilative capacity for a river. Journal of Environmental Engineering 2007;133(5):498-506. |
CR830912 (2008) CR830912 (Final) |
Exit |
|
Dittman JA, Driscoll CT. Factors influencing changes in mercury concentrations in lake water and yellow perch (Perca flavescens) in Adirondack lakes. Biogeochemistry 2009;93(3):179-196. |
CR830912 (Final) |
Exit |
|
Dittman JA, Shanley JB, Driscoll CT, Aiken GR, Chalmers AT, Towse JE. Ultraviolet absorbance as a proxy for total dissolved mercury in streams. Environmental Pollution 2009;157(6):1953-1956. |
CR830912 (Final) |
Exit Exit Exit |
|
Dittman JA, Shanley JB, Driscoll CT, Aiken GR, Chalmers AT, Towse JE, Selvendiran P. Mercury dynamics in relation to dissolved organic carbon concentration and quality during high flow events in three northeastern U.S. streams. Water Resources Research 2010;46:W07522, 15 pp., doi:10.1029/2009WR008351. |
CR830912 (Final) |
Exit |
|
Effler SW, Gelda R, Perkins MG, Peng F, Hairston Jr. NG, Kearns CM. Patterns and modeling of the long-term optics record of Onondaga Lake, New York. Fundamental and Applied Limnology/Archiv für Hydrobiolgie 2008;172(3):217-237. |
CR830912 (2008) CR830912 (Final) |
Exit |
|
O'Donnell SM, O'Donnell DM, Owens EM, Effler SW, Prestigiacomo A, Baker DM. Variations in the stratification regime of Onondaga Lake: patterns, modeling, and implications. Fundamental and Applied Limnology/Archiv für Hydrobiolgie 201;176(1):11-27. |
CR830912 (Final) |
Exit |
|
Owens EM, Bookman R, Effler SW, Driscoll CT, Matthews DA, Effler AJP. Resuspension of mercury-contaminated sediments from an in-lake industrial waste deposit. Journal of Environmental Engineering 2009;135(7):526-534. |
CR830912 (2008) CR830912 (Final) |
Exit |
|
Perkins MG, Effler SW, Strait C. Light absorption components in Onondaga Lake, New York, U.S.A. Fundamental and Applied Limnology / Archiv für Hydrobiologie 2010;176(3):209-223. |
CR830912 (Final) |
Exit |
|
Rucinski DK, Auer MT, Watkins Jr. DW, Effler SW, O'Donnell SMD, Gelda RK. Accessing assimilative capacity through a dual discharge approach. Journal of Water Resources Planning and Management 2007;133(6):474-485. |
CR830912 (Final) |
Exit |
|
Selvendiran P, Driscoll CT, Bushey JT, Montesdeoca MR. Wetland influence on mercury fate and transport in a temperate forested watershed. Environmental Pollution 2008;154(1):46-55. |
CR830912 (2008) CR830912 (Final) |
Exit Exit Exit |
|
Selvendiran P, Driscoll CT, Montesdeoca MR, Bushey JT. Inputs, storage, and transport of total and methyl mercury in two temperate forest wetlands. Journal of Geophysical Research 2008;113:G00C01, doi:10.1029/2008JG000739. |
CR830912 (2008) CR830912 (Final) |
Exit |
|
Selvendiran P, Driscoll CT, Montesdeoca MR, Choi H-D, Holsen TM. Mercury dynamics and transport in two Adirondack lakes. Limnology and Oceanography 2009;54(2):413-427. |
CR830912 (2008) CR830912 (Final) |
Exit Exit |
Supplemental Keywords:
robotic monitoring, zebra mussels, nutrient dynamics, eutrophication, modeling, paleolimnology, mercury, atmospheric deposition;, RFA, Scientific Discipline, Water, Ecosystem Protection/Environmental Exposure & Risk, Water & Watershed, Hydrology, Monitoring/Modeling, Environmental Monitoring, Ecology and Ecosystems, Watersheds, ecosystem modeling, model-based analysis, monitoring, watershed, near real time modeling, modeling, integrated watershed model, water quality, robotic monitoring, ecology assessment models, watershed assessment, ecological models, intellegent environment control system, real-time monitoringRelevant Websites:
http://www.lcs.syr.edu/faculty/driscoll/biocomplexity/index.asp
www.ourlake.org
Progress and Final Reports:
Original AbstractThe perspectives, information and conclusions conveyed in research project abstracts, progress reports, final reports, journal abstracts and journal publications convey the viewpoints of the principal investigator and may not represent the views and policies of ORD and EPA. Conclusions drawn by the principal investigators have not been reviewed by the Agency.