Page Banner

United States Department of Agriculture

Agricultural Research Service

Research Project: Spatial Modeling of Agricultural Watersheds: Water and Nutrient Management and Targeted Conservation Effects at Field to Watershed Scales

Location: Agricultural Systems Research Unit

Title: Simulating unstressed crop development and growth using the Unified Plant Growth Model (UPGM)

Authors
item McMaster, Gregory
item Ascough, James
item Edmunds, Debora
item Wagner, Larry
item Fox, Jr, Fred
item DeJonge, Kendall
item Hansen, Neil -

Submitted to: Environmental Modeling and Assessment
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: February 3, 2014
Publication Date: April 2, 2014
Citation: Mcmaster, G.S., Ascough Ii, J.C., Edmunds, D.A., Wagner, L.E., Fox, F.A., Dejonge, K.C., Hansen, N.C. 2014. Simulating unstressed crop development and growth using the Unified Plant Growth Model (UPGM). Environmental Modeling and Assessment. 19(5):407-424. DOI 10.1007/s10666-014-9402-x.

Interpretive Summary: Since development of the Erosion Prediction Impact Calculator (EPIC) model in 1989, many versions of the plant growth component have been incorporated into other erosion and crop management models and subsequently modified to meet model objectives (e.g., WEPS, WEPP, SWAT, ALMANAC, GPFARM). As a result, different versions of the same base model now exist. This has resulted in a number of problems including: 1) hindering the ability of users to understand differences among the plant growth components of the models, and therefore limiting their appropriate use, 2) improvements in one plant growth component have not been implemented into the other plant growth components, and 3) end users such as the USDA Natural Resource Conservation Service (NRCS) Conservation Delivery Streamlining Initiative (CDSI) must simulate different resource concerns using different models and databases providing required inputs. The objectives of this work are 1) describe the architecture of a standalone Unified Plant Growth Model (UPGM) derived from the WEPS plant growth model that is being used as the foundation to merge enhancements from different EPIC-based plant growth models, and 2) describe and evaluate new phenology, seedling emergence, and canopy height sub-models from the Phenology Modular Modeling System (PhenologyMMS) that have been incorporated into UPGM. A 6-yr irrigated maize (Zea mays L.) study from northeast Colorado was used to calibrate and evaluate UPGM running both the original and new phenology, seedling emergence, and canopy height sub-models. A variety of model statistics indicated the new sub-models usually resulted in better simulation results than the original sub-models. Both the new phenology and seedling emergence sub-models respond to varying water deficits, increasing the robustness of UPGM. Future research will continue work on UPGM to incorporate existing enhancements from other EPIC-based plant growth models to unify them into one model.

Technical Abstract: Since development of the EPIC model in 1989, many versions of the plant growth component have been incorporated into other erosion and crop management models and subsequently modified to meet model objectives (e.g., WEPS, WEPP, SWAT, ALMANAC, GPFARM). This has resulted in different versions of the same base model. The objectives of this work are 1) describe the architecture of a standalone Unified Plant Growth Model (UPGM) derived from the WEPS plant growth model that is being used as the foundation to merge enhancements from different EPIC-based plant growth models, and 2) describe and evaluate new phenology, seedling emergence, and canopy height sub-models from the Phenology Modular Modeling System (PhenologyMMS) that have been incorporated into UPGM. A 6-yr irrigated maize (Zea mays L.) study from northeast Colorado was used to calibrate and evaluate UPGM running both the original and new phenology, seedling emergence, and canopy height sub-models. A variety of model statistics indicated the new sub-models often resulted in better simulation results than the original sub-models. For example when comparing original and new sub-models, respectively, for predicting canopy height, the RMSE was 53.7 and 40.7 cm, index of agreement (d) was 0.84 and 0.92, relative error (RE) was 26.0 and -1.26%, and normalized objective function (NOF) was 0.47 and 0.33. Only the new sub-models predict leaf number, with mean values for four years of 2.43 leaves (RMSE), 0.78 (d), 18.38 % (RE), and 0.27 (NOF). Both the new phenology and seedling emergence sub-models respond to varying water deficits, increasing the robustness of UPGM. Simulating grain yield, final above ground biomass, harvest index, and leaf area index showed little difference when running the original or new sub-models. Future research will continue work on UPGM to incorporate existing enhancements from other EPIC-based plant growth models to unify them into one model.

Last Modified: 10/20/2014
Footer Content Back to Top of Page