Author
BULATEWICZ, TOM - Kansas State University | |
JIN, WEI - Kansas State University | |
STAGGENBORG, SCOTT - Kansas State University | |
LAUWO, SIMON - Kansas State University | |
MILLER, M - Kansas State University | |
DAS, SANJAY - Kansas State University | |
ANDRESEN, DANIEL - Kansas State University | |
PETERSON, JEFFREY - Kansas State University | |
STEWARD, DAVID - Kansas State University | |
WELCH, STEPHEN - Kansas State University |
Submitted to: Hydrology and Earth System Sciences
Publication Type: Peer Reviewed Journal Publication Acceptance Date: 8/3/2009 Publication Date: 8/14/2009 Citation: Bulatewicz, T., Jin, W., Staggenborg, S., Lauwo, S., Miller, M., Das, S., Andresen, D., Peterson, J., Steward, D.R., Welch, S.M. 2009. Calibration of a crop model to irrigated water use using a genetic algorithm. Hydrology and Earth System Sciences. 13(8):1467-1483. Interpretive Summary: Technical Abstract: Near-term consumption of groundwater for irrigated agriculture in the High Plains Aquifer supports a dynamic bio-socio-economic system, all parts of which will be impacted by a future transition to sustainable usage that matches natural recharge rates. Plants are the foundation of this system and so generic plant models suitable for coupling to representations of other component processes (hydrologic, economic, etc.) are key elements of needed stakeholder decision support systems. This study explores utilization of the Environmental Policy Integrated Climate (EPIC) model to serve in this role. Calibration required many facilities of a fully deployed decision support system: geo-referenced databases of crop (corn, sorghum, alfalfa, and soybean), soil, weather, and water-use data (4931 well-years), interfacing heterogeneous software components, and massively parallel processing (3.8×10**9 model runs). Bootstrap probability distributions for ten model parameters were obtained for each crop by entropy maximization via the genetic algorithm. The relative errors in yield and water estimates based on the parameters are analyzed by crop, the level of aggregation (county- or well-level), and the degree of independence between the data set used for estimation and the data being predicted. |