Page Banner

United States Department of Agriculture

Agricultural Research Service

Title: Validation Data Bases for Crop Simulators: Design, Assembling Delivery, and Use

Authors
item REDDY, VANGIMALLA
item Boone, M - MISSISSIPPI UNIVERSITY
item PACHEPSKY, YAKOV
item Whisler, F - MISSISSIPPI UNIVERSITY

Submitted to: Agricultural Engineering International Conference
Publication Type: Proceedings
Publication Acceptance Date: December 31, 2000
Publication Date: December 1, 2000

Interpretive Summary: The decision support systems, which are being used as tools in farm management, require data for validation, calibration, and model improvement. There is an enormous amount of data collected to be used for model validation and development, however, so far there are no procedures developed for storage, assembly, delivery, and use of the data. A validation database was developed along with meta data using the data we hove collected more than 10 years from collaborating farmers' fields from more than 15 states for cotton and soybean crops. The data were successfully used in validating crop models, developing new modules, estimating cultivar parameters for the crop models, comparing performance of models for the same crop, finding allometric relationships, establishing variability ranges and scale effects, and developing pedotransfer functions. This paper describes the procedures with examples for design, assembling, delivery, and use of validation databases for users, like scientists' extension workers and farmers, by providing mechanism for use of the validation data sets resulting in improved profits and reduced costs at the farm level.

Technical Abstract: A validation data base is developed along with metadata using the data we have collected during on-farm test of the crop simulation models GOSSYM and GLYCIM for cotton and soybean crops. Choices had to be made about data completeness and homogeneity, metadata design, data hierarchy and relation, retaining information on variability, methods, and scale, and quality control. The data were successfully used in validating crop models, developing new modules, estimating cultivar parameters for the crop models, comparing performance of models for the same crop, finding allometric relationships, establishing variability ranges and scale effects, and developing pedotransfer functions. Organizing field crop growth and development data in an accessible and a manageable manner creates a generic information pool and a multiple use.

Last Modified: 7/28/2014
Footer Content Back to Top of Page