Spatial and Temporal Evaluation of Hydrological Response to Climate and Land Use Change in Three South Dakota Watersheds

Abstract

This study analyzed changes in hydrology between two recent decades (1980s and 2010s) with the Soil and Water Assessment Tool (SWAT) in three representative watersheds in South Dakota: Bad River, Skunk Creek, and Upper Big Sioux River watersheds. Two SWAT models were created over two discrete time periods (1981‐1990 and 2005‐2014) for each watershed. National Land Cover Datasets 1992 and 2011 were, respectively, ingested into 1981‐1990 and 2005‐2014 models, along with corresponding weather data, to enable comparison of annual and seasonal runoff, soil water content, evapotranspiration (ET), water yield, and percolation between these two decades. Simulation results based on the calibrated models showed that surface runoff, soil water content, water yield, and percolation increased in all three watersheds. Elevated ET was also apparent, except in Skunk Creek watershed. Differences in annual water balance components appeared to follow changes in land use more closely than variation in precipitation amounts, although seasonal variation in precipitation was reflected in seasonal surface runoff. Subbasin‐scale spatial analyses revealed noticeable increases in water balance components mostly in downstream parts of Bad River and Skunk Creek watersheds, and the western part of Upper Big Sioux River watershed. Results presented in this study provide some insight into recent changes in hydrological processes in South Dakota watersheds. Editor's note: This paper is part of the featured series on SWAT Applications for Emerging Hydrologic and Water Quality Challenges. See the February 2017 issue for the introduction and background to the series.

Rationale and Efficacy of Assimilating Remotely Sensed Potential Evapotranspiration for Reduced Uncertainty of Hydrologic Models

“…The Upper Wabash model that uses SWAT's weather generator (WGN; Table ) to obtain energy‐related weather inputs shows relatively poor fit scores for streamflow, even though the model is calibrated at multiple sites (Figure ). Relying on WGN generally yields reasonable outputs where required data are discontinuous or sparsely available (Alighalehbabakhani et al, ; Evenson et al, ; Paul et al, ; Price et al, ). Accordingly, the WGN‐based model for Upper Wabash, despite underperforming with R 2 = 0.56 and NSE = 0.5 in daily streamflow simulation, can be vetted acceptable based on the criteria suggested by SWAT developers (e.g., Moriasi et al, ).…”

“…are discontinuous or sparsely available (Alighalehbabakhani et al, 2017;Evenson et al, 2016;Paul et al, 2017;Price et al, 2014). Accordingly, the WGN-based model for Upper Wabash, despite underperforming with R 2 = 0.56 and NSE = 0.5 in daily streamflow simulation, can be vetted acceptable based on the criteria suggested by SWAT developers (e.g., Moriasi et al, 2015).…”

Modelling the impacts of grassland to cropland conversion on river flow regimes in Skunk Creek watershed, Upper Midwest United States

“…The remaining land uses are water and wetlands (7%), urban (6.5%), and other land uses (0.35%) as shown in Figure b. Between 1992 and 2001, grassland conversion to cropland in the watershed was estimated to be 3% (Paul et al, ; Rajib et al, ). Skunk Creek is a main tributary to the Big Sioux River, which is a tributary of the Missouri River.…”

“…In this context, there is a need to explicitly examine the influence of grassland–cropland conversion on river flow regimes in order to provide insight that can be useful for formulation of watershed management strategies in the Upper Great Plains. Skunk Creek watershed in Eastern South Dakota is a representative catchment of the Upper Great Plains that undergoes rapid reduction of grassland due to increased cultivated crop production (e.g., Paul, Rajib, & Ahiablame, ; Reitsma et al, ; Wimberly et al, ). The main objective of this study was to approximate changes in annual flow regime as a result of grassland conversion to cultivated cropland in Skunk Creek watershed.…”

Featured Series Introduction: SWAT Applications for Emerging Hydrologic and Water Quality Challenges

“…The goal of this featured series is to demonstrate the versatility of SWAT in addressing traditional hydrologic problems such as Best Management Practices (BMP) implementation, land use/ climate change impacts, model/data uncertainty, and emerging issues such as the impact of natural gas development and bio-cropping scenarios on hydrology. In this issue, the use of SWAT to assess the role of climate and land use changes on hydrology of three South Dakota watersheds is presented (Paul et al, 2016). The study by Radcliffe and Mukundan (2016) demonstrates the role of precipitation data quality on SWAT model calibration and validation.…”