Limits...
Modeling the Impacts of Spatial Heterogeneity in the Castor Watershed on Runoff, Sediment, and Phosphorus Loss Using SWAT: I. Impacts of Spatial Variability of Soil Properties.

Boluwade A, Madramootoo C - Water Air Soil Pollut (2013)

Bottom Line: Overall, there was no significant difference in runoff simulation across the five configurations including the reference.This may be attributable to SWAT's use of the soil conservation service curve number method in flow simulation.Therefore having high spatial resolution inputs for soil data may not necessarily improve predictions when they are used in hydrologic modeling.

View Article: PubMed Central - PubMed

Affiliation: Department of Bioresource Engineering, Macdonald Stewart Building, McGill University, 21111 Lakeshore Road, Sainte-Anne-de-Bellevue, Quebec H9X 3V9 Canada.

ABSTRACT
Spatial accuracy of hydrologic modeling inputs influences the output from hydrologic models. A pertinent question is to know the optimal level of soil sampling or how many soil samples are needed for model input, in order to improve model predictions. In this study, measured soil properties were clustered into five different configurations as inputs to the Soil and Water Assessment Tool (SWAT) simulation of the Castor River watershed (11-km(2) area) in southern Quebec, Canada. SWAT is a process-based model that predicts the impacts of climate and land use management on water yield, sediment, and nutrient fluxes. SWAT requires geographical information system inputs such as the digital elevation model as well as soil and land use maps. Mean values of soil properties are used in soil polygons (soil series); thus, the spatial variability of these properties is neglected. The primary objective of this study was to quantify the impacts of spatial variability of soil properties on the prediction of runoff, sediment, and total phosphorus using SWAT. The spatial clustering of the measured soil properties was undertaken using the regionalized with dynamically constrained agglomerative clustering and partitioning method. Measured soil data were clustered into 5, 10, 15, 20, and 24 heterogeneous regions. Soil data from the Castor watershed which have been used in previous studies was also set up and termed "Reference". Overall, there was no significant difference in runoff simulation across the five configurations including the reference. This may be attributable to SWAT's use of the soil conservation service curve number method in flow simulation. Therefore having high spatial resolution inputs for soil data may not necessarily improve predictions when they are used in hydrologic modeling.

No MeSH data available.


Related in: MedlinePlus

Plot of the comparison between averages of monthly simulated and observed discharge (April 2001–December 2002)
© Copyright Policy - OpenAccess
Related In: Results  -  Collection


getmorefigures.php?uid=PMC3824274&req=5

Fig8: Plot of the comparison between averages of monthly simulated and observed discharge (April 2001–December 2002)

Mentions: Runoff. Graphical comparisons between the model simulation and observation of surface flow between April 2001 and December 2002 are shown in Fig. 8. For the five different configurations, no significant differences are seen (Fig. 8). This may be attributable to the SCS–curve number (CN) method implemented in SWAT. Surface runoff is estimated using the SCS runoff equation. This is an empirical model and could contribute to the variability of surface runoff. Ye et al. (2009) found that the SCS–CN method weakens the discrepancy between the different resolutions of soil heterogeneities and strongly affects the similarity in flow prediction. In other words, the CN threshold determined by the soil hydrologic groups are ranked based on soil permeability (Ye et al. 2009; Mishra and Singh 2003). Using this group across all soil types often masked out soils that have notable differences in physical characteristics.Fig. 8


Modeling the Impacts of Spatial Heterogeneity in the Castor Watershed on Runoff, Sediment, and Phosphorus Loss Using SWAT: I. Impacts of Spatial Variability of Soil Properties.

Boluwade A, Madramootoo C - Water Air Soil Pollut (2013)

Plot of the comparison between averages of monthly simulated and observed discharge (April 2001–December 2002)
© Copyright Policy - OpenAccess
Related In: Results  -  Collection

Show All Figures
getmorefigures.php?uid=PMC3824274&req=5

Fig8: Plot of the comparison between averages of monthly simulated and observed discharge (April 2001–December 2002)
Mentions: Runoff. Graphical comparisons between the model simulation and observation of surface flow between April 2001 and December 2002 are shown in Fig. 8. For the five different configurations, no significant differences are seen (Fig. 8). This may be attributable to the SCS–curve number (CN) method implemented in SWAT. Surface runoff is estimated using the SCS runoff equation. This is an empirical model and could contribute to the variability of surface runoff. Ye et al. (2009) found that the SCS–CN method weakens the discrepancy between the different resolutions of soil heterogeneities and strongly affects the similarity in flow prediction. In other words, the CN threshold determined by the soil hydrologic groups are ranked based on soil permeability (Ye et al. 2009; Mishra and Singh 2003). Using this group across all soil types often masked out soils that have notable differences in physical characteristics.Fig. 8

Bottom Line: Overall, there was no significant difference in runoff simulation across the five configurations including the reference.This may be attributable to SWAT's use of the soil conservation service curve number method in flow simulation.Therefore having high spatial resolution inputs for soil data may not necessarily improve predictions when they are used in hydrologic modeling.

View Article: PubMed Central - PubMed

Affiliation: Department of Bioresource Engineering, Macdonald Stewart Building, McGill University, 21111 Lakeshore Road, Sainte-Anne-de-Bellevue, Quebec H9X 3V9 Canada.

ABSTRACT
Spatial accuracy of hydrologic modeling inputs influences the output from hydrologic models. A pertinent question is to know the optimal level of soil sampling or how many soil samples are needed for model input, in order to improve model predictions. In this study, measured soil properties were clustered into five different configurations as inputs to the Soil and Water Assessment Tool (SWAT) simulation of the Castor River watershed (11-km(2) area) in southern Quebec, Canada. SWAT is a process-based model that predicts the impacts of climate and land use management on water yield, sediment, and nutrient fluxes. SWAT requires geographical information system inputs such as the digital elevation model as well as soil and land use maps. Mean values of soil properties are used in soil polygons (soil series); thus, the spatial variability of these properties is neglected. The primary objective of this study was to quantify the impacts of spatial variability of soil properties on the prediction of runoff, sediment, and total phosphorus using SWAT. The spatial clustering of the measured soil properties was undertaken using the regionalized with dynamically constrained agglomerative clustering and partitioning method. Measured soil data were clustered into 5, 10, 15, 20, and 24 heterogeneous regions. Soil data from the Castor watershed which have been used in previous studies was also set up and termed "Reference". Overall, there was no significant difference in runoff simulation across the five configurations including the reference. This may be attributable to SWAT's use of the soil conservation service curve number method in flow simulation. Therefore having high spatial resolution inputs for soil data may not necessarily improve predictions when they are used in hydrologic modeling.

No MeSH data available.


Related in: MedlinePlus