This dissertation utilizes data from four sub-watersheds in the Little River Experimental Watershed, GA to develop models to improve forecast predictions related to the management of surface-water pollution due to non-point source runoff. Non-point source pollution is the primary cause of US surface-water quality impairment and a main transport mechanism for pathogens and other pollutants into receiving surface water bodies (US EPA 2008). In response to pollution reduction and watershed remediation mandates under the Federal Clean Water Act (1972)-particularly the Total Maximum Daily Load (TMDL) program-the role of water quality modeling in effectively rehabilitating impaired waters has taken on greater importance. Consequently, the significance of this study is that it is the first of its kind to incorporate a multi-model approach to address limitations in using single water quality models. In this regard, it builds on water quality engineering research by presenting methods to estimate contaminant concentrations and reduce uncertainty in overall model predictions in impaired water-bodies. Methodologically, the key point of departure in this dissertation is centered on the fact that water quality modeling is the cornerstone of TMDL analyses but the associated prediction uncertainty affects their adequacy in providing reliable contaminant loadings estimates in an impaired water body. As such, utilizing hydrological and water-quality process equations embedded in the two most widely used watershed-scale models, the Soil and Water Assessment Tool (SWAT) and Hydrological Simulation Program-Fortran (HSPF), and observed data from the sub-watersheds mentioned above, the dissertation addresses this limitation by combining results from the two competing models to reduce uncertainty and enhance accuracy of predictions. The study was conducted in two phases. First, HSPF and SWAT-two extensively-used, scientifically-rigorous, US EPA-approved watershed-scale codes-were used to build models of the four study catchments. The models were individually calibrated and shown (based on Nash-Sutcliffe Efficiency (NSE) ratios) to produce reliable simulations of the hydrologic and water quality conditions in the watershed. The second phase of the analysis involved using a multi-model approach to combine model forecasts. Model combination, introduced by Bates and Granger in 1969, has emerged as a viable analytical technique (Claesken and Hjort, 2008; Ajami et al., 2006) and widely-used across disciplines to improve model-forecasting results (Kim et al., 2006; Shamseldin et al., 1997; Granger, 2001; Clemens, 1989; Thompson, 1976; Newbold and Granger, 1974; Dickinson, 1973). After calibration, the model predictions were combined for each catchment using three different methods: the Weighted Average Method (WAM), the Nash-Sutcliffe Efficiency Maximization Method (NSE-max) and an Artificial Neural Network Method (ANN). Comparison of the results of the multi-model formulation with original individual model results showed improved estimates with all three combination methods. The improvement in model accuracy (based on NSE ratios) varied from modest to significant in both hydrologic and water quality variables. These improvements were attributed to a reduction in model structural uncertainty resulting from the ability to capture aspects of some of the more complex watershed interactions from exogenous information provided by the contributing models. It should be noted here, however, that as model availability increases, if additional models (beyond those utilized here) are used with this approach, care should be taken to ensure the credibility of each individual model for simulating the watershed scale processes under review. Limitations of this study include possible bias introduced by the use of deterministic models to estimate probabilistic contaminant distributions, limitations in available data, and the use of a seven-year study period that did not account for possible impacts of shorter periods of extreme hydrologic conditions on the individual model performances and model combination weightings. Recommendations for future research include (a) improving watershed-scale codes to better describe the probability distribution functions characteristic of contaminant distributions and data collection on wildlife species and populations; and investigating the fate and transport processes of pathogenic indicator bacteria deposited in forested areas and the impact of extreme hydrologic conditions on model performance and weighting. Overall, the findings from this dissertation suggest that water quality modeling incorporating a multi-model approach has the potential to significantly improve predictions compared to the predictions obtained when only one model is used. Clearly, the findings reported here have significant implications in improving TMDL analyses and remediation plans by presenting an approach that exploits the strengths of two of the most complete and well-accepted watershed-scale water quality models in the United States. Moreover, the findings of this dissertation auger well for the future of TMDL management in that it provides a more robust and cost effective basis for policy makers to decide on effective management strategies that incorporate acceptable risk, allowable loading and land use.

The previously developed watershed hydrological and water quality model for St. Louis Bay watershed by Kieffer (2002) was refined and calibrated. The aspects of model development refinement included development of fertilization-related nutrient input parameters, evaluation of nutrient input methods, development of plant uptake-related nutrient input parameters, non-cropland simulation using PQUAL module, and recalibration of hydrology in Jourdan River. The related information of typical cropland management practice based on consultation from Mississippi State University Extention Service personnel was integrated into the watershed model. In addition, the Mississippi Department of Environmental Quality (MDEQ) observed water quality data were analyzed to evaluate the appropriateness of current watershed delineation and assess the health of the stream based on the MDEQ proposed numerical water quality target. The refined watershed model was calibrated in Wolf Rover and Jourdan River using both USGS and MDEQ observed water quality data. The concentrations of water quality constituents calculated from the developed watershed model will be provided as boundary conditions for the developed Bay hydrodynamic and water quality model for Total Maximum Daily Load studies.

The use of best management practices has become common in recent years, leading to the need for hydrologic models to predict their behavior and effectiveness. A vegetated filter strip at Mississippi State University was used to test two models: the Hydrologic Simulation Program-FORTRAN Best Management Practices Editor (HSPF BMPrac) and the System for Urban Stormwater Treatment and Analysis Integration (SUSTAIN). Water samples were taken during the Spring of 2011 and tested for sediments and nutrients; HSPF was used for computing flows, sediments, and nutrients. The filter strip was not effective at pollutant removal with removal efficiency rates of 68.1, 91.7, 86.3, and 115.4 percent for total suspended solids (TSS), total nitrogen (TN), total phosphorus (TP), and dissolved phosphorus (DP) respectively. Calibration of HSPF was successful for TSS with a R2 value of 0.52; nutrients were not as successful with R2 values of 0.11 and 0.43 for TN and TP.

Modeling ground-water hydrology is critical in low-gradient, high water table watersheds where ground-water is the dominant contribution to streamflow. The Hydrological Simulation Program-FORTRAN (HSPF) model has two different subroutines available to simulate ground water, the traditional ground-water (TGW) subroutine and the high water table (HWT) subroutine. The HWT subroutine has more parameters and requires more data but was created to enhance model performance in low-gradient, high water table watershed applications. The objective of this study was to compare the performance and uncertainty of the TGW and HWT subroutines when applying HSPF to a low-gradient watershed in the Coastal Plain of northeast North Carolina. One hundred thousand Monte Carlo simulations were performed to generate data needed for model performance comparison. The HWT model generated considerably higher Nash-Sutcliffe efficiency (NSE) values while performing slightly worse when simulating the 50% lowest and 10% highest flows. Model uncertainty was assessed using the Average Relative Interval Length (ARIL) metric. The HWT model operated with more average uncertainty throughout all flow regimes. Based on the results, the HWT subroutine is preferable when applying HSPF to a low-gradient watershed and the accuracy of simulated stream discharge is important. In situations where a balance between performance and uncertainty is called for, the choice of which subroutine to employ is less clear cut. / Master of Science

Agricultural production and human activities in a watershed can expose the watershed to environmental degradation, pollution problems, and a decrease in water quality if resources and activities within a watershed are not managed carefully. In order to best utilize limited resources and maximize the results with respect to time and money spent on nonpoint source (NPS) pollution control and prevention, the environmental risk must be identified so that areas with a higher quantified environmental risk can be targeted. The objectives of the research presented in this master thesis were to develop a procedure to quantify environmental risk of nutrient loadings to surface waters and to demonstrate the procedure on a watershed. A procedure to quantify environmental risk of nutrient loadings to surface waters was developed. The risk is identified as the probability of occurrence of a nonpoint source (NPS) pollution event caused by a runoff event multiplied by the consequences to a biological or chemical endpoint. The procedure utilizes the NPS pollution model ANSWERS-2000 to generate upland pollutant loadings to receiving waters. The pollutant loading impact on stream water quality is estimated using the stream module of Hydrologic Simulation Program FORTRAN (HSPF). The risk is calculated as the product of probability of occurrence of a NPS event and consequences of that event. The risk quantification procedure was applied to a watershed in Virginia. Total phosphorus (TP) loadings were evaluated with respect to resultant in-stream dissolved oxygen (DO) concentration. The TP loadings were estimated in ANSWERS-2000 then the consequences were estimated in HSPF. The results indicated that risk was higher for the smaller, more frequent storms indicating that these smaller, more frequent loading events represent a greater risk to the in-stream water quality and ecosystem than larger events. While the probability of occurrence of lower TP loading was higher because they were caused by smaller, more frequent storms, the consequences were less for the same events. The developed procedure can provide watershed stakeholders and managers with a useful tool to quantify the environmental risk a watershed is exposed to as a result of different land management and development scenarios. The scenarios can then be compared to identify a risk level that is considered acceptable. The procedure can also be used by policymakers to set a cap on the risk a certain activity can expose a watershed to. / Master of Science

In this study eight residential development scenarios are created for the mostly undeveloped Back Creek Watershed outside Roanoke, Virginia. The development scenarios include low, medium (cluster), medium (conventional), and high density development with and without development restrictions. These scenarios represent a large range of development as the land use imperviousness varies from 1% for the baseline condition to 34% for the most developed scenario. The hydrologic model HSPF is used to generate overland and channel flows from 43 years of rainfall. Hydrologic output from HSPF of the various landuse patterns from the eight scenarios are evaluated using Post Processor, a Visual Basic program. The results show that increased development causes a reduction in Back Creek's baseflow and an increase in the occurrence of both high and low flow extreme events. Overall, these results indicate that increased development will increase the variability of flowrate in Back Creek. Stream health impacts from flow variability were also analyzed with the Post Processor. First, hydrologic statistical variables with ecological relationships were used to gage the level of stream health impacts from flow variability. The averaged stream health index for the development scenarios was found to closely follow the amount of development, represented by the percent of impervious landuse. Second, the amount of velocity, depth, and both depth and velocity habitat available for three habitat guild representative species was evaluated for each scenario. The results indicated that increased development would lead to a substantial reduction in available riffle species habitat (represented by the fantail darter) and a moderate reduction in run and pool species habitat (represented by the central stoneroller and smallmouth bass, respectively). Overall, increased development has been found to have a negative impact on stream health. This impact should be considered in any future expansion of the Roanoke suburbs into this watershed. / Master of Science

During periods of reduced precipitation, flow in low-order, upland streams may be reduced and may stop completely. Under these "low flow" conditions, fecal bacteria directly deposited in the stream dominate in-stream bacteria loads. When developing a Total Maximum Daily Load (TMDL) to address a bacterial impairment in an upland, rural watershed, direct deposit (DD) fecal bacteria sources (livestock and wildlife defecating directly in the stream) often drive the source-load reductions required to meet water quality criteria. Due to limitations in the application of existing watershed-scale water quality models, under low-flow conditions the models can predict unrealistically high in-stream fecal bacteria concentrations. These unrealistically high simulated concentrations result in TMDL bacteria source reductions that are much more severe than what actually may be needed to meet applicable water quality criteria. This study used the Hydrological Simulation Program-FORTRAN (HSPF) to compare three low-flow DD simulation approaches and combinations (treatments) on two Virginia watersheds where bacterial impairment TMDLs had been previously developed and where low-flow conditions had been encountered. The three methods; Flow Stagnation (FS), DD Stage Cut-off (SC), and Stream Reach Surface Area (SA), have all been used previously to develop TMDLs. A modified version of the Climate Generation (CLIGEN) program was used to stochastically generate climate inputs for multiple model simulations. Violations of Virginia's interim fecal coliform criteria and the maximum simulated in-stream fecal coliform concentration were used to compare each treatment using ANOVA and Kruskal Wallis rank sum procedures. Livestock DD bacteria sources were incrementally reduced (100%, 50%, 15%, 10%, 5%) to represent TMDL load reduction allocation scenarios (allocation levels). Results from the first watershed indicate that the FS method simulated significantly lower instantaneous criterion violation rates at all allocation levels than the Control. The SC method reduced the livestock DD load compared to the Control, but produced significantly lower instantaneous criterion violation rates only at the 100% allocation level. The SA method did not produce significantly different instantaneous criterion violation rates compared to the Control. Geometric mean criterion violation rates were not significantly different from the Control at any allocation level. The distributions of maximum in-stream fecal coliform concentrations simulated by the combinations SC + FS and SC + SA + FS were both significantly different from the Control at the 100% allocation level. The second watershed did not produce low-flow conditions sufficient to engage the FS or SC methods. However, the SA method produced significantly different instantaneous violation rates than the Control at all allocation levels, which suggests that the SA method continues to affect livestock DD loads when low-flow conditions are not simulated in the watershed. No significant differences were found in the geometric mean violation rate or distribution of maximum simulated in-stream fecal coliform concentrations compared to the Control at any allocation level. This research suggests that a combination of the SC and FS methods may be the most appropriate treatment for addressing unrealistically high concentrations simulated during low-flow conditions. However, this combination must be used with caution as the FS method may increase the maximum simulated in-stream fecal coliform concentration if HSPF simulates zero volume within the reach. / Master of Science

Sediment transport at the watershed scale in the Bluegrass Region of Kentucky is dominated by surface fine grained laminae, streambed, and streambank erosion; high instream sediment storage; and surface erosion processes. All these processes can be impacted by agricultural, urban, and suburban land-uses as well as hydrologic forcing. Understanding sediment transport processes at the watershed scale is a need for budgeting and controlling sediment pollution, and watershed modeling enables investigation of the cumulative effect of sediment processes and the parameters controlling these processes upon the entire sediment budget for a watershed. Sediment transport is being modeled by coupling the hydrologic model Hydrologic Simulations Program-FORTRAN (HSPF) with an in-house conceptually based hydraulic and sediment transport model. The total yield at the watershed outlet as well as the source fractions from surface fine grained lamina, streambed, and streambank sources; deposition; and biological generation within the streambed are predicted with the sediment transport model. Urbanization scenarios are then run on the calibrated model so as to predict the sediment budget for the South Elkhorn watershed for present and future conditions.

Small lowland agricultural systems promote conditions where benthic biological communities can thrive. These biogeochemical processes have significant impacts on terrestrial ecosystem processes including POC flux and fate, nutrient balances, water quality budges, and aquatic biological functioning. Limited information is available on coupled biological and hydrologic processes in fluvial systems. This study investigates the mixture of biological and hydrologic processes in the benthic layer in order to understand POC cycling in the South Elkhorn system. Further, comprehensive modeling of POC flux in lowland systems has not been performed previously and the behavior of potentially controlling variables, such as hydrologic forcing and seasonal temperature regimes, is not well understood. Conceptual hydraulic and sediment transport models were simulated for the South Elkhorn. Based on data and model results it was concluded that during a hydrologic event, upland and bank sources produce high variability of POC sources. Likewise, over time, the density of hydrologic events influenced accrual of benthic algal biomass in the POC pool. Environmental variables such as temperature and light availability drove seasonal variations of POC in the streambed. Based on model estimates, around 0.29 metric tCkm-2yr-1 of POC is flushed from the system annually with 13 % coming from autochthonous algae.

No description available.