2005 Progress Report: Measuring and Modeling the Source, Transport and Bioavailability of Phosphorus in Agricultural Watersheds

EPA Grant Number: R830669
Title: Measuring and Modeling the Source, Transport and Bioavailability of Phosphorus in Agricultural Watersheds
Investigators: Lathrop, Richard C. , Armstrong, D. E. , Hoopes, John A. , Karthikeyan, K. G. , MacKay, David Scott , Nowak, Peter , Panuska, John C. , Penn, Michael R. , Potter, Kenneth W. , Wu, Chin H.
Institution: Wisconsin Department of Natural Resources , The State University of New York at Buffalo , University of Wisconsin Madison , University of Wisconsin - Platteville
EPA Project Officer: Packard, Benjamin H
Project Period: December 17, 2002 through December 16, 2005 (Extended to December 16, 2006)
Project Period Covered by this Report: December 17, 2004 through December 16, 2005
Project Amount: $749,307
RFA: Nutrient Science for Improved Watershed Management (2002) RFA Text |  Recipients Lists
Research Category: Water , Water and Watersheds


The objectives of this research project are to:

  1. quantify effects of manure management and crop production systems on runoff phosphorus (P) losses, particularly related to the portion that is biologically available;

  2. determine spatial patterns of sediment and associated P in streams;

  3. determine in-stream fate and transport processes of P, including bioavailable P (BAP);

  4. evaluate and improve modeling tools used to assess P transport in agricultural watersheds over a wide range of spatial scales;

  5. determine relation of P losses with the scale of animal operation; and

  6. integrate outreach into on-going research efforts .

Progress Summary:

Upland research

Field Plot Studies. Sediment size fractionation results were obtained from 46 soil samples covering both the frost-melt (10) and the frost-free (36) periods. These results include sediment mass (%), total phosphorus (TP) concentration (mg/kg) and total volatile solids (VS) in each of the 5 particle size classes: < 2, 2-10, 10-50, 50-500 and > 500 µm (Table 1). In general, an inverse relationship between sediment TP concentration and particle size was observed. A strong relationship between sediment TP concentration and % VS (surrogate for organic matter) was also obtained across all particle size fractions (R2 = 0.92). There were no significant differences in the TP mass fraction for the three different field management treatments (i.e., corn-grain, silage, and silage with manure). These results suggest that, within a range of organic matter content, the particle-mass:TP curve might be a function of soil texture, clay content, and mineralogy, and could be used to quantify TP mass across different particle size classes. An important finding is that between 60 % and 70% of total P losses from the corn management systems studied under conservation tillage could occur in the silt-sized and finer particles.

Table 1. Fractionation of Total Solids (TS), Total P (TP) and Volatile Solids in Different Particle-Size Classes From 46 Soil Samples Collected During the Frost-Melt (FM) and Frost-Free (FF) Periods

Particle Size (μm)

< 2



50 – 500

> 500

FM Period, (N=10)

Median TS (%)






Range *






Median TP Conc. (mg/kg)






Range *






FF Period, (N=36)

Median TS (%)






Range *






Median TP Conc. (mg/kg)






Range *






FM and FF Period

Percent Volatile Solids












* Range equals the 1st and 3rd quartile values.

Aggregates are agglomerations of organic matter, clay-, silt-, and sand-sized particles, which typically have lower specific gravity values than primary particles, thus increasing their transportability in runoff. Consequently, aggregates likely play an important role in the transport and delivery of particulate-bound P from agricultural fields. Aggregate-size distribution was determined for 44 samples (most with multiple peaks) and the resulting peaks were grouped into four size classes (Table 2). It is important to note that two of the four aggregate size classes, class 2 and 4, compared well with those previously identified by U.S. Department of Agriculture (USDA) researchers (30 and 500 µm). The incorporation of additional aggregate size classes will improve the P transport predictive ability of non-point source simulation models.

Table 2. Summary of Aggregate-Size Distribution Analysis

Class ID










Median (μm)










Range (μm) *










Peaks in Class










Frequency %





* Range = 1st quartile (25%) and 3rd quartile (75%).

Results from this investigation are currently being used to validate the Wisconsin Phosphorus Index (WPI), a field-scale nutrient management tool, while the particle-mass:TP curve will be used to improve the particulate TP delivery routines used in the WPI. Research results will be disseminated using a variety of methods, including the more traditional peer-reviewed journal publications and conference presentations, as well as direct contacts with researchers actively working in this area. One option under consideration is to use a direct email notification to key individuals directing them to a Web site indexed by research sub-group and topic.

Subwatershed-Scale Studies. Due to significantly less than average precipitation in 2005, a study was conducted within the “small farm” watershed to assess the potential for colluvial storage zones and unchanelled landscape depressions to alter the delivery of sediment and P to receiving streams. A more complete understanding of these depressional zones is needed to accurately manage for P delivery from upland agricultural areas to downstream receiving waters in watersheds dominated by such landscape features. These results have proven particularly insightful to meet our project’s overall objectives for understanding P transport processes.

Our study of colluvial storage zones focused on three primary objectives. First, radioactive 137Cs inventories in soils were used to quantify erosion and deposition rates associated with different landscape features. Although various methods are utilized to predict watershed degradation and aggradation rates from areal 137Cs inventories, these methods are rarely extended to the examination of sediment-bound constituent movement. Secondly, erosion and deposition patterns were then examined in relation to P sorption characteristics of sediments from erosional zones and ultimately delivered to and deposited within depositional locations. The purpose of this objective was to determine the immobility of P at particular positions within the watershed, and ultimately to understand the degree to which depressional zones act as sinks or sources of P in spillage or subsurface drainage pathways. Finally, the potential for different forms of P movement from depressional zones was explored. Erosion is commonly regarded as the primary factor influencing P transport because P is less readily desorbed from soil particles once it has been adsorbed, so P movement is typically associated with particulate P (PP) forms. At scales where the sediment and P transfer continuums are interrupted by colluvial sinks and depressional zones, however, exceptions to this condition may exist.

Vertical soil profile distributions of 137Cs demonstrated clear signatures of sediment erosion and deposition throughout the site. Coupling these results with an examination of P accumulation patterns (as measured by Bray 1P and TP analyses) further linked sediment immobilization to patterns of P enrichment within depressional zones. These conditions indicate that P movement is primarily driven by erosive processes that mobilize particulate-bound P from upslope locations to depositional sites.

Spatial patterns of P accumulation exhibited similarity with sediment accumulation patterns, supported by the positive correlation between Bray-1P levels and 137Cs inventories at the six coring locations (r = 0.61). The most significant factor influencing P levels, however, appeared to be the legacy effect of manure application close to the main animal housing center (Table 3). The importance of transport factors was reflected by the significance of the sampling location’s slope (p < 0.005) and the nearest depressional zone “sink” point (p < 0.001). The inverse relationships of both these predictors with regards to Bray-1P values reflected the accumulation of P at depressional zone landscape positions. Finally, the significance of proximity to the preferred locations for initiating manure spreading (p < 0.050) suggests a further influence of management factors.

Table 3. Results From Two Stepwise Regressions for Bray-1P Values Determined From 126 Soil Sampling Points at KA Subwatershed



R2 = 0.34

R2 = 0.38


Independent variable†








Pearson r

Distance to barn




< 0.001



< 0.001






< 0.005



< 0.005


Distance to nearest start point




< 0.050





Distance to nearest sink point









†The variables “distance to the nearest start point” and “distance to the nearest sink point” were correlated, and therefore not used within the same model.

Once sediments are deposited within depressional zones, these colluvial materials appear to be acting primarily as P sinks, with most sediment and P remaining close to their deposited locations. Total P levels and inventories were greatest in the depressional zones relative to their respective surrounding hillslopes. Mean TP levels for the 020 cm soil depth increment were significantly (p < 0.001) greater for depressional zones (956 mg kg-1) than erosional locations (599 mg kg-1). Furthermore, the TP inventory (mg TP cm-2) for the 020 cm increment was also found to be significantly related to 137Cs inventory, although this relationship was not as strong (R2 = 0.40, F = 4.09, p < 0.10).

Although P delivery is associated with sediment movement, the interruption of the sediment delivery continuum by depressional zones implies that the primary pathway of P migration in such areas may be associated with conditions that foster dissolved P (DP) transport. However, results suggest that P saturation in these soils is related primarily to P sorption maxima that are likely influenced by soil texture and particle size. For instance, greater water-extractable P (WEP) levels at erosional locations are attributable to the presence of coarser texture of sediments due to the selective removal of fine-textured particles possessing fewer sorption sites. This explanation is supported by the lower values of sorption maxima (PSmax) for cores in the erosional zones, and is reflected in the positive correlation between WEP and slope (r = 0.40).

Results suggest that by using manure to protect eroded hillslopes from further erosion, soils may become saturated with P, which then becomes more vulnerable to being transported in dissolved forms in runoff. For instance, soil interflow layers may be less able to buffer DP movement as rainfall-runoff travels through the watershed flow-system and comes into contact with soils where WEP and equilibrium P concentrations at zero-sorption (EPC0) levels are elevated within the effective depth of rainfall-runoff interaction (~510 cm below surface). The likelihood of readily available P delivery from these eroded soils is higher, and may supply P to overland runoff. Heightened focus has traditionally been on sediment-bound P transport, but as DP migrates through watersheds with high EPC0, these soils will not buffer P movement, and dissolved phase for P delivery may be more dominant than previously observed.

Where runoff flows into a depressional zone, P may reside until spillage events occur. Runoff data from monitored spillage events, for instance, showed that dissolved reactive phosphorus (DRP) concentrations were greater than EPC0. Furthermore, results also indicate that DRP is being desorbed from suspended sediments, as EPC0 from suspended sediments was equal to 4.38 mg L-1 for a successfully monitored May 22, 2004 event. These findings present two implications for managing water quality. First, the elevation of runoff DRP beyond EPC0 may indicate that P movement from other locations is not being buffered well by depressional zone soils. Secondly, this condition is further evidence that depositional zones may act as both sinks and sources of P. This condition arises because long-term sediment storage does not necessarily imply P storage, due to saturation of depressional zone soils with P.

Stream Research

Stream Survey Work. The stream survey work on our primary study site (Dorn Creek) was completed in 2005. A large spatial variability of TP in surficial sediments was found with the greatest TP values occurring in wetland depositional areas. In addition, stream sediment sampling was done throughout the entire Lake Mendota watershed. The lowest TP values were observed at downstream sites in urbanized portions of the subwatersheds. Sediment deposits at these sites had larger median particle size (i.e., greater sand content) compared to the silty deposits in the agricultural, upper portions of the watershed. Preliminary estimates of the P and sediment inventory for the entire watershed indicate approximately 10% of both the total phosphorus and sediment mass was present in the uppermost 5 cm of the channel deposits (Table 4). It is estimated that the upper 5 cm of deposits is approximately equal to the annual average P loading to Lake Mendota.

Table 4. Inventory of Sediment and Phosphorus Within the Stream Channels of the Lake Mendota Watershed



Mass of Sediment
(Tonne DW)

Mass of P

Avg sed conc (ppm)

Dorn Creek





Pheasant Branch





Six Mile Creek





Token Creek





Yahara River










Based on the analysis of stream sediment data, we have concluded that the single best predictor of sediment P is the percentage of water in the sediment (percentage water = mass of water/bulk mass of saturated sediment). Higher water content is associated with silty, unconsolidated sediments in depositional areas, whereas low water content is associated with compacted sands in high-gradient stream reaches. Sediment P concentrations in wetlands were typically higher than stream P concentrations and were handled separately.

Stream channel slope can be readily determined from digital elevation models of watersheds. In an effort to refine the sediment and P inventory of the Lake Mendota watershed, a relationship between the best predictor of P, percentage water, and stream slope is being evaluated. Outliers have been identified as sites that are subject to higher stream velocities during runoff events (urban areas) and transitional deposition sites. It is hypothesized that at transitional deposition sites, a surficial layer several centimeters thick of fine grained, high P sediment is deposited during smaller runoff events and remains until a significant storm resuspends and transports the sediment downstream.

Sediment Chemistry Work. We are investigating the levels and dynamics of bioavailable P (BAP) in the sediments of Dorn Creek. Our goal is to determine whether the stream sediments are a sink for BAP transported from the upland portion of the watershed or a conduit for transport to downstream Lake Mendota. BAP is measured as NaOH-extractable inorganic P, a relationship developed earlier using algal bioassays. Other P fractions (chemical forms) of P in the sediments were also measured. Another aim of this research is to determine the controlling factors of BAP concentrations in Dorn Creek. To explore the sediment properties controlling BAP concentrations, we measured iron, aluminum, magnesium, calcium, nitrogen, and carbon concentrations, and porosity of sediment samples. Sediment cores along the stream were obtained at sites chosen to represent contrasts in depositional environments and associated sediment characteristics. The sites varied in accumulated unconsolidated sediment depth, stream dynamics (stream velocity, width, and depth), and surrounding watershed characteristics.

Amounts of sediment BAP varied among sites, as inventories of BAP in the top 10 cm of surface sediment ranged from 140 to 14,000 mg P cm-2. The variations among sites generally reflected differences in the site characteristics. BAP inventories were highest in sites where sediment accumulation was favored due to characteristics such as lower stream velocities, decreased gradient, or diffuse flow. Inventories were lowest in areas more conducive to sediment scouring. Variations may also be related to the adjacent local watershed characteristics (e.g., wetland versus cropland or livestock holding areas) and associated local P transport to the stream. The concentrations of BAP were related closely to concentrations of total P, averaging 48%, but ranging from 7.7% to 62%. Concentrations of BAP in stream sediments were closely related to the concentrations of sediment total P (r2 = 0.97) and inorganic P (r2 = 0.98) levels. Total P was related to Fe (r2 = 0.74) and Al (r2 = 0.57) concentrations but not to Ca (r2 = 0.093) or Mg (r2 = 0.018) concentrations. Total P, inorganic P, and BAP correlated strongly to sediment Fe concentrations for nearly all the Dorn Creek samples, reflecting the association of BAP with Fe oxyhydroxide phases. Several chemical constituents showed moderate (organic C, total N, Fe, Al) to weak (total P, organic P) relations to porosity, an indicator of sediment texture.

To explore the influence of discharge events on P transport by the stream, we compared P levels carried by the stream at six locations during three events: August 2004 (rainfall = 5.5 cm), March 2005 (snowmelt), and July 2005 (rainfall = 3.0 cm). During the August 2004 event, stream BAP concentrations rose to very high levels (2.5 mg/L), falling to background baseflow levels of about 0.09 mg/L after two days. In contrast, during the March snowmelt event, stream BAP concentrations rose to levels comparable to smaller (2.0 cm to 2.2 cm) fall rain events (1.4 mg/L), but took a much longer time period (17 days) before returning to near base flow levels. Implications are that winter snowmelts are important in the overall annual transport of BAP through Dorn Creek.

Spatial variation of DRP was monitored during the July 2005 event. Hours 3 and 6 held the highest DRP concentrations during the event timeframe, implying that P concentrations spike shortly following the beginning of a discharge event. Local variation is highlighted by Sites C and D, which are 30 meters from one another, separated by a cattle operation. Site D, the more downstream site of the two, had higher DRP concentrations at three of the five sampling periods, reflecting differences in source inputs between the sites. In general, mid-stream sites located in agricultural areas exhibited higher concentrations than upstream or downstream sites located in wetland areas. This pattern and the high degree of variation among sites at similar time periods following the rain event suggest that local watershed characteristics play an important role in stream water P concentrations. Additionally, sediment disturbance and P release at high flow may augment DRP concentrations carried by the stream water.

We are also investigating sediment dynamics and sources using radionuclide data (210Pb, 7Be, 137Cs, 40K) from sediment cores, including data from cores taken during the summer of 2005. The isotope data will allow us to investigate short- and long-term mixing, sedimentation rates, P burial rates, and sediment age in zones of sediment accumulation. The 210Pb and 7Be data should provide information on sediment mixing and age due to their respective long and short half-lives, while 137Cs and 40K will be useful in examining sediment sources.

Stream Flow and Sediment Processes. During 2005, we focused on the study of stream flow and sediment processes. Measurements of real-time water levels and suspended sediment concentrations at the main study site of riffle-pool sequences were made, continuing the work in 2004. In addition, the scope of the study was widened significantly in 2005 to cover more stream sites in Dorn Creek and to include measurements of sediment storage and morphology changes, sediment bed characteristics, and P transport processes. Sites 1 and 2 (boundaries of Reach A) are located in the intermittent flow portion of the stream. Sites 3 and 4 (boundaries of Reach B) are located downstream of a wetland area in a very flat portion of the stream. Sites 5 and 7 (boundaries for Reach C) are located in a steep, boulder-filled section of the creek. Site 6, located between Sites 5 and 7, continues to serve as the main study site of riffle-pool sequences. Sites 8 and 9, located at the head of the Dorn Creek’s major downstream wetland, form the boundaries of Reach D and have characteristics of heavy sediment deposits.

Monitoring of discharge and suspended sediment was continued at Site 6 throughout the season. To investigate the effects of the upstream wetland on sediment transport patterns in the stream, a new instrument package was deployed at Site 2, upstream of the wetland. Previous data in 20032004 showed that the duration of a hydrograph for a rainfallR1year,24hours = 1.5” is generally approximately 5 7 days at Site 6. In 2005, our recorded data show that the duration of hydrographs for a rainfall R1year,24hours= 1.0” at Site 2 reduces to approximately 1 day, suggesting the important function of the wetland. In addition, the multiple modal characteristics of the sediment graph exhibited at Site 6 was not apparent at Site 2, suggesting that the wetland could serve to store water and sediments.

In 2005, solute transport processes that mimic dissolved P were studied using injections of a fluorometric tracer dye, Rhodamine WT. Three reaches with varying gradients— B, C, and D— were selected and dye concentration breakthrough curves were recorded at the upstream and downstream ends of each. Results were analyzed for mean travel time. Transient storage (TS) parameters were estimated using a One-Dimensional Transport with Inflow and Storage model with Parameter estimation (OTIS-P). These parameters were used to model solute transport in a 2.4 km reach of Dorn Creek with and without TS . The model results indicated that those with TS can have 26% longer residence time in the study reach than those with no TS . In other words, the considerable increase in residence times in Dorn Creek due to TS has great potential to play a significant role in nutrient cycling in streams.

To elucidate the sediment bed and storage characteristics of Dorn Creek, cross sectional channel profiles were determined at 5 locations ( Sites 1 and 3, and three sections at Site 6) approximately every 2 weeks during the 2005 field season. A sixth location (Site 9) was added in September. This monitoring allowed the identification of temporal changes in sediment storage in the stream channel, as well as an estimate of the magnitude of these changes. Variations as large as 5 cm were found, which was in good agreement with depth estimates of recently deposited sediments determined by isotope analyses conducted by other project investigators. Our results indicate that soft (unconsolidated) sediment deposits are primarily located in stream reaches with minor slope, and that steep areas store little to no sediment.

Bed sediments were sampled on a monthly basis at Sites 1, 3, 6 and 9. Sediment samples were taken using Shelby tube cores, which were separated into 4 depths. Particle size and available phosphorus (AP) analyses were carried out by the University of Wisconsin Soil Testing Laboratory. Results indicate that median AP values increased near the bed surface, suggesting that much of the surficial sediment P is active. AP also increased with silt content, which is likely responsible for the majority of the suspended load carried in the stream. Overall data suggest that both in-stream processes relating to the transport and storage of sediments in Dorn Creek may have a significant impact on the delivery of P to downstream waters.

Modeling Activities

The goals of the modeling activities are to integrate the insights from the field investigations on the source and transport of BAP and develop more effective tools for decision making. Given the breadth of potential outcomes from such work, a set of focusing questions centered on a common theme was developed.

Problem Statement. Digital Elevation Models (DEMs) are widely used in distributed hydrologic modeling. In general, interior depressions within catchments are viewed as errors in the DEM, even though they are hydrologically significant features. Natural depressions in catchments are capable of trapping surface runoff and associated sediment, but they are difficult to identify and represent, especially in ungauged basins. Watershed models based on DEMs cannot directly resolve small-scale topographic depressions, which control the release of solid/dissolved P. Our modeling work examined the errors associated with the removal of such depressions on predictions from hydrologic models – Soil and Water Assessment Tool (SWAT) and Agricultural Policy/Environmental eXtender (APEX). The following questions are being addressed:

  1. Do sediment delivery ratios capture depression dynamics, and if not, what alternatives are available?

  2. Are these alternatives useful in nested watersheds?

  3. Do these alternatives scale to larger watersheds, especially those lacking detailed observations?

  4. What costs, in terms of predictive uncertainty, are associated with more explicit representation of storage-release dynamics?

There are four objectives in this study:

Objective 1: Representing topographic depressions
Objective 2: Parameter scaling up
Objective 3: BAP extensions to APEX
Objective 4: Uncertainty analysis

Representing Topographic Depressions. Natural depressions in catchments are capable of trapping surface runoff and associated sediment. Interior sinks or depressions are usually assumed to be errors in DEMs . This makes natural depressions difficult to identify and represent, especially in ungauged basins. Can watershed scale hydrological models still represent natural depression function when the depressions are filled (i.e., ignored) to build the flow connectivity of the drainage network? Our working hypothesis is that sediment transport parameters (Sed = a*Vb) can be used as proxies for the functioning of surface depressions to obtain the correct sediment response. The alternative is to explicitly prescribe depressions as reservoirs with more geometric details of depressions if the hypothesis failed. To test this hypothesis, we utilized data obtained from two monitored subwatersheds (KA1 and KA2) as well as the larger Dorn Creek. Manual calibration was applied to KA1 and KA2 data to find out the optimal relevant parameters. These parameters were then applied on Dorn Creek where further calibration was conducted.

Preliminary results indicated that the APEX model overestimated the runoff in KA1 and KA2 because the KA1 and KA2 have smaller contributing areas in comparison to those predicted from the 3 meter resolution DEM. This was due to the presence of natural topographic depressions, which substantially reduce the effective drainage areas of the KA1 and KA2 outlets. The reservoir sub-model within the APEX was used to represent these depressions.

APEX was then scaled up to the Dorn Creek watershed by dividing the watershed into 89 sub-watersheds and adding a reservoir to the outlet point in each sub-watershed. APEX was run for an 8-year period at daily time steps and total sediment storage (deposition minus re-entrainment) was monitored along the main channel. Preliminary results suggest that the modeling scheme can accurately capture relative magnitudes of sediment deposition estimated from stream channel surveys conducted by other project investigators in 2003 and 2004. The next work is that reservoirs will be randomly set up in the subareas to get more accurate sediment depositions compared to the measured sediment data in the Dorn Creek watershed.

Parameter Scaling Up. Different scale watersheds usually have different model parameters. Based on the Dorn Creek work, we are exploring the parameter scaling up problems to larger monitored stream watersheds in the entire Lake Mendota watershed. Our working hypothesis is that channel slope and contributing drainage areas are proxies for the presence and sediment trapping abilities of depressions. This study is being conducted in two different watershed scales: (1) Dorn Creek and (2) full Mendota watershed including Yahara River and Pheasant Branch, both of which have long-term USGS stream flow, sediment and P monitoring data. In the Dorn Creek watershed, location and geometric properties of reservoirs from catchment scale topographic properties will be explored. Reservoir location (contribution drainage area) and size are keys to controlling sediment deposition. Reservoirs will be randomly set up in the subareas of Dorn Creek to develop proxy parameters in a subset of Dorn Creek sites, tested at remaining Dorn Creek sites, and then applied on Yahara and Pheasant Branch. Preliminary results of this modeling work produced a good relationship between sediment depositions within subarea channel slope within the Dorn Creek watershed.

Journal Articles on this Report : 3 Displayed | Download in RIS Format

Other project views: All 50 publications 9 publications in selected types All 9 journal articles
Type Citation Project Document Sources
Journal Article Cabot PE, Nowak P. Planned versus actual outcomes as a result of animal feeding operation decisions for managing phosphorus. Journal of Environmental Quality 2005;34(3):761-773. R830669 (2003)
R830669 (2004)
R830669 (2005)
R830669 (Final)
R828010 (Final)
  • Abstract from PubMed
  • Full-text: American Society of Agronomy-Full Text HTML
  • Abstract: American Society of Agronomy
  • Other: American Society of Agronomy-PDF
  • Journal Article Cabot PE, Karthikeyan KG, Miller PS, Nowak P. Sediment and phosphorus delivery from alfalfa swards. Transactions of the American Society of Agricultural and Biological Engineers (ASABE) 2006;49(2):375-388. R830669 (2004)
    R830669 (2005)
    R830669 (Final)
  • Abstract: ASABE
  • Journal Article Chen E, Mackay DS. Effects of distribution-based parameter aggregation on a spatially distributed agricultural nonpoint source pollution model. Journal of Hydrology 2004;295(1-4):211-224. R830669 (2003)
    R830669 (2004)
    R830669 (2005)
    R830669 (Final)
  • Full-text: ScienceDirect-Full Text HTML
  • Abstract: ScienceDirect
  • Other: ScienceDirect-PDF
  • Supplemental Keywords:

    Upper Midwest, EPA Region 5,, RFA, Scientific Discipline, INTERNATIONAL COOPERATION, ECOSYSTEMS, Water, Waste, Ecosystem Protection/Environmental Exposure & Risk, Aquatic Ecosystems & Estuarine Research, Water & Watershed, Bioavailability, Aquatic Ecosystem, Water Quality Monitoring, Terrestrial Ecosystems, Environmental Monitoring, Watersheds, anthropogenic stress, bioassessment, anthropogenic processes, watershed classification, nutrient transport, fate and transport, model, ecosystem monitoring, watershed management, biodiversity, nutrient flux, conservation, nitrogen inputs, diagnostic indicators, ecosystem indicators, aquatic ecosystems, bioindicators, watershed sustainablility, water quality, biological indicators, ecosystem stress, watershed assessment, conservation planning, nitrogen uptake, bioavailable phosphorus, transport modeling, ecosystem response, aquatic biota, land use, restoration planning, watershed restoration, biological impairment

    Progress and Final Reports:

    Original Abstract
  • 2003 Progress Report
  • 2004 Progress Report
  • Final Report