Samaresh Das

Pollutants Removal by Vegetative Filter Strips Planted with Different Grasses

Over the last few years, increasing occurrence of deadly pathogens and presence of nvarious pollutants (nutrients, pesticides, other chemicals, and sediments) above the prescribed limit nin water systems, clearly indicate alarmingly deteriorating quality of water resources. As a result, nfarming systems that are known to be the main non-point or diffuse pollution source are being reviewed microscopically. Vegetative Filter Strip (VFS) is considered to be one of the best nmanagement practices (BMPs) for effective control sediment and nutrient transport over agricultural nlands. Many laboratory and field scale studies have also indicated the limited usefulness of VFS to ncontrol movement of bacteria in surface runoff. However, design of VFS under field conditions still nremains a challenge due to variation in upland hydrological parameters and factors effecting nmovement of pollutants through VFS such as type of vegetation cover and density, width of strip, and nland slope. Determination of trapping efficiency of VFS for bacteria is more complex due to the ncomplex interaction of various factors governing the die-of and re-growth of bacteria under field ncondition, and release of bacteria from soil reserve. An extensive field experiment is being conducted nat the research farm of University of Guelph in Southern Ontario, Canada, to evaluate to neffectiveness of VFS under different vegetation cover, ground slope, width of filter strip, and in nvarious seasons. Concentration of sediment reduced an average by 88.3% and almost 94.3% nsediment mass was trapped in various filter strips. Higher trapping efficiencies for mass were nobserved for sediment bound nutrients (94.5% and 93.9% for N and P, respectively) compared to nsoluble forms (57.0% and 77.3% for N and P, respectively). Results for bacteria (Total Coliforms, nFecal Coliforms, and E. Coli) through VFSs were encouraging but not conclusive. In the present npaper, experiment and results of the study are presented and discussed in details.

Comparing the Performance of SWAT and AnnAGNPS Model in a Watershed in Ontario

Safe distribution of water from source to tap is a major concern in Ontario after the Walkerton tragedy. The Nutrient Management Act and Source Water Protection Act in Ontario are being planned to protect the quality of water resources by controlling pollutant transport from upland contributing areas. To achieve the objective of providing safe, clean, and affordable water supply and aquatic ecosystem, it is essential to assess the contribution of pollution from different sources to the stream. While source water protection is a major issue in Ontario, a number of watershed models are being used by different agencies to mitigate the water quality and quantity problems. The Guelph Watershed Research Group at the University of Guelph is also assessing the performance of different watershed models to best apply in Ontario conditions. This study has investigated the hydrology and sediment output from two widely used watershed models (SWAT and AnnAGNPS). The models are daily time step, watershed scale, pollutant-loading model developed to simulate long-term runoff, sediment, and chemical transport from agricultural watersheds. The models were run for a period of ten years (1991-2000) and the output data for the first five years (1991-1995) were used to calibration and that for the last five years data (1996-2000) for validation. The results of the study indicate that the models performed fairly well in simulating the runoff and sediment yield for Ontario conditions. For annual water balance, the model underpredicted the evapotranspiration, but the average difference was under acceptable range (3.1% for AnnAGNPS and 11.2% of SWAT). For surface runoff, the average difference between the observed and simulated annual runoff for the calibration period was 2.1% for AnnAGNPS, and 4.5% for SWAT model. The deviations for validation period were 1.8% for AnnAGNPS and 18.8% for the SWAT model. The Nash-Sutcliffe coefficients for monthly outputs were 0.79 for AnnAGNPS and 0.7 for AnnAGNPS in the calibration phase and 0.69 for AnnAGNPS and 0.57 for SWAT model for the validation phase. A significant challenge in calibrating and validating the sediment portion of the model was a shortage of observed sediment data, which might have lead to the over prediction of sediment yield by both the models during validation period. A detailed evaluation of SWAT model on the watershed is still in progress and the output for SWAT might be improved once the work is completed.

ASSESSING THE PERFORMANCE OF VARIOUS HYDROLOGICAL MODELS IN THE CANADIAN GREAT LAKES BASIN

These days various types of models are available for non-point Source (NPS) pollution modeling nand management. These models can help in identifying critical areas and selecting best nmanagement practices to reduce pollutant loads to streams. However, the way the existing nmodels handle spatio-temporal variability of parameter and description of the natural processes nmake straight forward application difficult. The objective of the present study is to assess the nperformance of various models to simulate streamflow and sediment yield in Ontario conditions. nThe NPS models considered include Soil and Water Assessment Tool (SWAT), Hydrological nSimulation Program-FORTRAN (HSPF), Annualized Agricultural Non-Point Source n(AnnAGNPS); and Areal Non-Point Source Watershed Environmental Response Simulation n(ANSWERS-2000). The comparisons of the models result indicate an important difference in the ndifferent components of water quality and quantity that is addressed in this paper.

Application of AnnAGNPS Model under Ontario condition

Watershed scale model helps to find out the contributing areas and best describe the best management practices to prevent the contribution. ANNualized AGricultural Non-Point Source (AnnAGNPS) model was applied to a watershed in Southern Ontario to compare and validate the hydrology and sediment component from the non-point sources. The model was run for six years (1990 to 1995) and the outputs for the first 3 years were used to calibrate and the last 3 years were use for validation purpose. Scarcity of the observed sediment data leaded to compare the model output to compare with the matched events. Both the runoff and sediment yield compared fairly well with the observed data indicating that the model had an acceptable performance in simulation of runoff and sediment.

Application of CANWET and HSPF for TMDL Evaluation under Southern Ontario Conditions

The CANWET (Canadian ArcView Nutrient and Water Evaluation Tool) and HSPF (Hydrologic Simulation Program - FORTRAN) models were applied to Upper Canagagigue Creek watershed of the Grand River basin in southern Ontario, Canada, for hydrology and sediment evaluations. Both the models have similarity in structure where CANWET is simpler, both in algorithms and use, than HSPF. The outputs of both the models for water budgeting components were compared on annual, seasonal, and monthly basis. The water budget components, evapotranspiration, surface runoff, and subsurface runoff produced by both the models were comparable on annual and seasonal time steps; however, there were some discrepancies in monthly and daily simulations. The seasonal, monthly, and daily Nash-Sutcliffe efficiency coefficient with observed stream flows were 0.83, 0.81, and 0.48 for HSPF, respectively, and 0.80, 0.67, and 0.24 for CANWET, respectively. The monthly and daily simulations by HSPF model were better since HSPF algorithm has more control on temporal variation in parameters sensitive for hydrologic simulations. The sediment simulations by both the models were consistently close for erosion and sediment yield on annual basis. However, superiority in predictions for total suspended sediment yield of one model over the other could not be concluded because of lack of observed data. The daily load of sediment modeled by HSPF followed flow peaks and available observed sediment data points.

Evaluation of Nutrient Component of AnnAGNPS Model in a Watershed in Ontario

Abstract: The water impairment with nutrient losses from non-point source pollution (NPS) is a major concern in different watersheds in Ontario where agricultural and live stock production are the major activities. The Annualized Agricultural Non-Point Source (AnnAGNPS) model has been applied in a watershed situated in the Southern Ontario to investigate model simulated N (N) and P (P) output. The model was run from 1991 to 1995 and the transport of dissolved and attached part of both the nutrients with runoff and clay particles was investigated. Since very limited observed nutrient data was available in the studied watershed and thus it was difficult to obtain a reasonable comparison between the simulated and observed values. The study results show that the dissolved and attached N has followed the similar trend of transport with the runoff and clay particles respectively. The model underpredicetd the total N and overpredicted the total P. The R2 and correlation coefficient was 0.75 and 0.19 for N and 0.82 and 0.23 for P respectively. The uncertainty of the input parameters and excessive use of manure were assumed to be the main reason behind the discrepancy. However, it was also obvious from the study that the model performance may be improved with proper calibration and further investigation on the fate and transport of the nutrient components.

Application of SWAT to Meet Water Quality Requirements for Canadian Conditions–A Study in Grand River Watershed

The objective of the present study was to evaluate SWAT (Soil and Water Assessment Tool) hydrological model for Canagagigue Creek watershed, a tributary of Grand River. This drainage basin is facing a growing challenge of maintaining water quality under the increasing rural/agricultural and urban/industrial activities. Currently, water quality in the Grand River is rated form fair-to-good. Winter freezing and spring thawing plays the dominating effect on various hydrological processes in this region of Canada. Detailed evaluation of the SWAT model indicated that this model could potentially be used for simulation of flows and sediment yield in the watershed. Analysis of daily, monthly, and yearly flows and sediment yield illustrated good match between simulated and observed values. Efforts were also made to map critical areas in the watershed. The result further indicated that 7% of the watershed area is responsible for 11% of total sediment yield from the watershed. The model was also tested for various scenarios of management practices in the watershed for the selection of best management practices (BMPs). Detailed results of the study and functionality of the model under the limited availability of data are presented and discussed in the paper.

A Model Evaluation and Modification Strategy

Abstract: Watershed models have been used as management tools for an ever increasing array of water resource issues such as source water protection and non-point source pollution management. This paper outlines a strategy that has been adopted for the evaluation and improvement of watershed models. The strategy involves model calibration for annual, seasonal, monthly, and daily flows, and includes careful consideration of the rapid and slow response components for each flow variable. This evaluation step clarifies how well or poorly a model captures the most significant hydrologic processes and flow paths in the watershed for the time period under consideration. It also helps to flag algorithms requiring modification for reliable use of the model. An example implementation of the strategy is presented and discussed.

Hydrological and water quality modeling in the Ontario River basins: comparison of model results

The applicability and validity of hydrological and water quality models has to be critically evaluated before they can be used in a basin different from where they were originally developed. Variations in physiographic characteristics and climate regime will affect the choice of a suitable hydrological model as models vary in the assumption and simplification of the natural process. These entail evaluation and if necessary modification of the original model assumptions, processes descriptions and structure to suit the river basin in consideration. The objective of this study is to investigate the applicability of widely used hydrological and water quality models under the Ontario condition in Canada. In this study the ANNualized AGricultural Non-Point Source (AnnAGNPS) and the Areal Non-Point Source Watershed Environmental Response Simulation (ANSWERS-2000) are considered. First, the uncalibrated models were applied to the Canagagigue Creek, a tributary of the Grand River basin in Ontario, Canada for a period of 1998–1999 on a daily basis. Based on parameter sensitivity analysis, the models were calibrated. Finally, the performance of the models were assessed and evaluated for their ability to simulate streamflows and sediment yield.