Skip to main content
ARS Home » Pacific West Area » Maricopa, Arizona » U.S. Arid Land Agricultural Research Center » Plant Physiology and Genetics Research » Research » Publications at this Location » Publication #249303
Title: Evaluating the accuracy of VEMAP daily weather data for application in crop simulations on a regional scale

Author
item WU, WEI - Southwest University
item LIU, HONG-BIN - Southwest University
item HOOGENBOOM, GERRIT - University Of Georgia
item White, Jeffrey

Submitted to: European Journal of Agronomy
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 11/15/2009
Publication Date: 12/2/2009
Citation: Wua, W., Liu, H., Hoogenboom, G., White, J.W., 2009. Evaluating the accuracy of VEMAP daily weather data for application in crop simulations on a regional scale. European Journal of Agronomy 32, pp. 187-194.

Interpretive Summary: Weather plays a critical role in determining crop yields, but analyses of how weather affects crops is often constrained by difficulties in locating accurate data for long periods. The Vegetation/Ecosystem Modeling and Analysis Project (VEMAP) provides daily weather variables on a 0.5 latitude–longitude grid across the US. To compare the effects of different sources of daily weather data, the VEMAP (1961–1990) for the state of Georgia were compared with data from 52 individual ground stations of the National Weather Service Cooperative Observer Program (COOP). Additionally, crop grain yields of soybean (Glycine max) were simulated using a computer model. Weather data from the sources were generally comparable, but deviations showed strong seasonal trends. Simulations of soybean yield using the two data sources were correlated (r = 0.68, p < 0.01). Analyses using hybrid sets of weather data (e.g., combining VEMAP precipitation with COOP temperature and solar radiation) indicated that the variation of simulated yield was mainly associated with the differences in rainfall. The results showed that the VEMAP daily weather data were adequate for use with crop models for long-term climate change research. These results thus suggest that VEMAP-type data are a useful source of long-term daily weather data when other sources are not available.

Technical Abstract: Weather plays a critical role in eco-environmental and agricultural systems. Limited availability of meteorological records often constrains the applications of simulation models and related decision support tools. The Vegetation/Ecosystem Modeling and Analysis Project (VEMAP) provides daily weather variables on a 0.5 latitude–longitude grid across the conterminous USA. Daily weather data from the VEMAP (1961–1990) for the state of Georgia were compared with data from 52 individual ground stations of the National Weather Service Cooperative Observer Program (COOP). Additionally, simulated crop grain yields of soybean (Glycine max) were compared using the two data sources. Averaged daily maximum and minimum temperatures (Tmax and Tmin, respectively), solar radiation (SRAD), and precipitation (PPT) differed by 0.2 °C, -0.2 °C, 1.7 MJ m-2 d-1, and 0 mm, respectively. Mean absolute errors (MAEs) for Tmax, Tmin, SRAD, and PPT were 4.2 °C, 4.4 °C, 4.4 MJ m-2 d-1, and 6.1 mm, respectively, and root mean squared errors (RMSEs) for Tmax, Tmin, SRAD, and PPT were 5.5 °C, 5.9 °C, 5.8 MJ m-2 d-1, and 13.6 mm, respectively. Temperature differences were lowest during summer months. Simulations of grain yield using the two data sources were strongly correlated (r = 0.68, p < 0.01). The MAE of grain yield was 552 kg ha-1. The RMSE of grain yield was 714 kg ha-1. Hybrid analyses indicated that the variation of simulated yield was mainly associated with the differences in rainfall. The results showed that the VEMAP daily weather data were able to be adequately applied to crop growth simulation at spatial and temporal scales, especially for long-term climate change research. Overall, the VEMAP weather data appears to be a promising source for crop growth modeling concerned with scale to 0.5° coordinate grid.