Development and application of a weather data service client for preparation of weather input files to a crop model

Abstract

Crop yield prediction has been made using a crop growth model that relies on four categories of input data including soil, crop, management, and weather. Most crop models are single column models, which require individual weather inputs for each site of interest. The objectives of this study were to develop a weather data service client that prepares weather input files for a crop growth model and to examine its application to yield prediction at a national scale. The weather data service client downloads daily weather data from the web-based weather data service portal operated by the Korea Meteorological Administration (KMA). The client also prepares weather input files for the ORYZA 2000 model at minimum effort. In total, 4950 input files were prepared to predict rice yield in 2011 and 2012 using the weather data service client. To prepare nearly 5000 weather input files, it would take more than a month for a skilled person to download weather data from the KMA database and to reorganize those data to the input data format for the ORYZA 2000 model manually. Using the weather data service client, several hours were enough to prepare all the input files without error associated with manual preparation as well as with minimum effort and labor.