Skip to main content
U.S. flag

An official website of the United States government

Publication Date
1 September 2020

A Review of SWAT Applications, Performance and Future Needs for Simulation of Hydro-Climatic Extremes

Subtitle
The majority of the reviewed 111 SWAT studies reported satisfactory performance.
Print / PDF
Powerpoint Slide
Science

The Soil and Water Assessment Tool (SWAT) model (https://swat.tamu.edu/) is one of the most widely used ecohydrological models worldwide and has been tested across an extensive range of watershed scales and environmental conditions ((https://www.card.iastate.edu/swat_articles/). An analysis of SWAT’s performance in replicating extreme flows has been lacking among the approximately 4,000 SWAT-based studies that exist in the current literature. Thus a review was conducted for this study that focused on the ability of SWAT to replicate extreme flows, droughts, floods, or droughts, and floods for 111 studies that reported such flow comparisons. SWAT was compared versus extreme flows in one or more watersheds in 23 countries or transnational systems among the 111 reviewed studies.

Impact

Increasing awareness of climate change and its potential impact on regional hydrological processes has resulted in the need to develop strategies to assess hydro-climatic related policies based on the possible effects of future extreme events. SWAT has been used to assess to estimate future hydro-climatic changes in response to climate model projections in hundreds of studies. Thus it is of considerable importance to evaluate how well the model has replicated extreme flows for existing conditions for different climate and stream system conditions across the globe. Statistical results based on comparisons of SWAT simulated continuous streamflows versus the corresponding measured values for monthly, daily and sub-daily time periods showed that nearly 90% of the Nash-Sutcliffe Modeling Efficiency (NSE) values and over 90% of the R2 statistics exceed 0.5 and 0.6, respectively. These values met the criteria of satisfactory, good, or very good streamflow simulation results as suggested by https://doi.org/10.13031/2013.23153 and https://doi.org/10.13031/trans.58.10715. However, just a small percentage of the studies further validated the model for specific flood events, peak flow, low flow, flow duration curves, and/or different precipitation intensity comparisons.

Summary

The review confirmed that SWAT was able to replicate continuous streamflows for a variety of watershed conditions affected by extreme flow conditions. However, the review also points to several future research needs to improve the ability of SWAT for evaluating extreme hydro-climatic events. These research topics include: (1) a unified SWAT extreme performance assessment framework, (2) SWAT modifications that result in improved replication of peak and low flows, (3) reliability assessment of global and satellite precipitation products that could be used for SWAT extreme simulations, (4) bias correction of CMIP6 and regional climate projections, (5) comparison of the most recent SWAT+ models with older standard SWAT models for evaluation of extreme flow simulations in different types of basins, (6) development of an extreme flow module within an overall SWAT modeling system, and (7) integration of artificial intelligence within SWAT modeling.

Point of Contact
Philip W. Gassman
Institution(s)
Iowa State University
Funding Program Area(s)
Publication