Different calibration procedures for flows estimation using SWAT model

Abstract

This paper evaluates the calibration effectiveness for estimating monthly average flow using the SWAT model in a watershed located at Espírito Santo state in Brazil. Six methodologies were tested, which differ from each other by varying calibration parameters, size of historical series and number of sub-basins associated with the process. SWAT-CUP was used for the calibration tests and its effectiveness was analyzed by three objective functions: Percent bias (PBIAS), Nash-Sutcliffe efficiency (NSE), and Ratio of the Root Mean Square Error of measured data (RSR). Tests with long historical flow data (29 years long) did not generate satisfactory outputs. However, the best results were obtained while calibrating all parameters for all sub-basins with a data series of only 3 years long. From the results, we highlight the importance of a better description of the methodologies used in the calibration process, since different results can be obtained by the same input data.

KEYWORDS:

• SWAT-CUP
• Modeling
• watershed
• sensitivity analysis
• water production

Additional information

Funding

This work was supported by the Fundação de Amparo à Pesquisa do Estado de São Paulo [grant number 2019/24292-7].

Notes on contributors

Luana Lavagnoli Moreira

Luana Lavagnoli Moreira PhD student in at the Institute of Hydraulic Research, Federal University of Rio Grande do Sul (UFRGS), Brazil. She is also an environmental engineer and master of science in environmental engineering, both from the Federal University of Espírito Santo (UFES), Brazil. Currently, she is interested in field topics related to hydrology, modeling, water flow monitoring and geoprocessing.

Dimaghi Schwamback

Dimaghi Schwamback Master of Science student in hydraulics and sanitation engineering at the University of São Paulo (USP), Brazil. He is also an environmental engineer graduated from the Federal University of Espírito Santo (UFES), Brazil. Currently, he is interested in field topics related to soil loss, hydrology, infiltration, geoprocessing, and land cover.

Daniel Rigo

Daniel Rigo Professor in the environmental engineering department of the Federal University of Espírito Santo (UFES), Brazil. He is also a civil engineer graduated from the UFES, Master of Science in civil engineering and PhD in ocean engineering, both from the Federal University of Rio de Janeiro (UFRJ), Brazil. Currently, he is interested in field topics related to decision support systems, water quality, computational modeling, and water resources management.