Review and development of ASABE Engineering Practice 621: Guidelines for calibrating, validating, and evaluating hydrologic and water quality models

Authors: Harmel, Daren; Baffaut, Claire; DOUGLAS-MANKIN, KYLE - Us Geological Survey (USGS)

Submitted to: Transactions of the ASABE
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 4/20/2018
Publication Date: 8/29/2018
Citation: Harmel, R.D., Baffaut, C., Douglas-Mankin, K. 2018. Review and development of ASABE Engineering Practice 621: Guidelines for calibrating, validating, and evaluating hydrologic and water quality models. Transactions of the ASABE. 61(4):1393-1401. https://doi.org/10.13031/trans.12806.
DOI: https://doi.org/10.13031/trans.12806

Interpretive Summary: Information to support application of hydrologic and water quality (H/WQ) models abounds, yet modelers commonly use arbitrary methods to conduct modeling and document and report model results. Consistent methods are needed to: 1) increase model accuracy and precision, and 2) more effectively interpret and communicate model performance to decision-makers and other modeling stakeholders. Based on this need, the Natural Resources and Environmental Systems – Hydrology Committee (NRES-21) of the American Society of Agricultural and Biological Engineers initiated a series of annual rap sessions in 2010 and formed Process and Communication committees that developed manuscripts and reviewed published literature to support potential development of an ASABE standard. Following the publication of 33 articles in two Special Collections in Transactions of ASABE and a Position Paper in Environmental Modeling and Software, the ASABE Engineering Practice 621 “Guidelines for Calibrating, Validating, and Evaluating Hydrologic and Water Quality (H/WQ) Models: ASABE 621 JUN2017” was approved in 2017. The Engineering Practice establishes consistent terminology and definitions and presents recommendations for selecting the most appropriate model; compiling and processing input data and calibration, validation, and evaluation data; determining measures for evaluating model performance; parameterizing and calibrating the model; evaluating model performance; and documenting the modeling process and results. It is not meant to establish prescriptive requirements but rather recommends “good” modeling practices to enhance calibration, validation, evaluation, and communication of H/WQ models.

Technical Abstract: Information to support application of hydrologic and water quality (H/WQ) models abounds, yet modelers commonly use arbitrary, ad hoc methods to conduct, document, and report model calibration, validation, and evaluation. Consistent methods are needed to improve model calibration, validation, and evaluation to: 1) increase model accuracy and precision, and 2) more effectively interpret and communicate model performance to decision-makers and other modeling stakeholders. Based on this need, the Natural Resources and Environmental Systems – Hydrology Committee (NRES-21) of the American Society of Agricultural and Biological Engineers initiated a series of annual rap sessions in 2010 and formed Process and Communication committees that developed manuscripts and reviewed published literature to support potential development of an ASABE standard. Following the publication of 33 articles in two Special Collections in Transactions of ASABE and a Position Paper in Environmental Modeling and Software, the ASABE Engineering Practice 621 “Guidelines for Calibrating, Validating, and Evaluating Hydrologic and Water Quality (H/WQ) Models: ASABE 621 JUN2017” was approved in 2017. The Engineering Practice establishes consistent terminology and definitions and presents recommendations for selecting the most appropriate model; compiling and processing input data and calibration, validation, and evaluation data; determining measures for evaluating model performance; parameterizing and calibrating the model; evaluating model performance; and documenting the modeling process and results. It is not meant to establish prescriptive requirements but rather recommends “good” modeling practices to enhance calibration, validation, evaluation, and communication of H/WQ models.