본문 바로가기
  • Home

Experimental Predictions of Crop Yields Using Time-Series Modeling of Climate Reanalysis Data: A Case of Iowa, USA, 1960-2009

  • Journal of the Korean Cartographic Association
  • Abbr : JKCA
  • 2016, 16(2), pp.115-126
  • Publisher : The Korean Cartographic Association
  • Research Area : Social Science > Geography > Geography in general > Cartography
  • Published : August 31, 2016

Kim, Na Ri 1 Yang-Won LEE ORD ID 1

1부경대학교

Accredited

ABSTRACT

Global warming can bring about changes in crop yield by directly affecting meteorological parameters such as temperature and precipitation. Previous studies based on the climate change scenarios had difficulties in detailed simulation owing to the problem of spatial resolution of GCM (general circulation model). The researches using time-series modeling rarely incorporated climate factors info the crop yield prediction. In this study, we conducted experimental predictions of corn and soybean yields by time-series modeling of downscaled climate reanalysis data. We built a database for the climate dataset and governmental yield statistics for the period of 1960-2009 for the 99 counties in Iowa State. Then we carried out 10 sets of the next-year prediction for corn and soybean yields using VAR (vector autoregression) and ARIMA (autoregressive integrated moving average) methods. The VAR and ARIMA were able to predict the next-year yields with the errors of 16-18% and 11-14%, respectively. In addition, soil properties such as topsoil pH, subsoil clay fraction and subsoil sodicity were closely related to the actual yields and the prediction accuracies.

Citation status

* References for papers published after 2022 are currently being built.

This paper was written with support from the National Research Foundation of Korea.