Drought forecasting based on the remote sensing data using ARIMA models

  • Authors:

  • Ping Han;Peng Xin Wang;Shu Yu Zhang;De Hai Zhu

  • Affiliations:

  • Department of Physics, College of Science, China Agricultural University, West Campus, Beijing, 100193, PR China;Department of Geographic Information Engineering, College of Information and Electrical Engineering, China Agricultural University, East Campus, Beijing, 100083, PR China;Remote Sensing Information Center for Agriculture of Shaanxi Province, Xi'an, 710015, PR China;Department of Geographic Information Engineering, College of Information and Electrical Engineering, China Agricultural University, East Campus, Beijing, 100083, PR China

  • Venue:
  • Mathematical and Computer Modelling: An International Journal
  • Year:
  • 2010

Quantified Score

Hi-index 0.98

Visualization

Abstract

Regarded as a near real time drought monitoring method, the VTCI index based on remote sensing data is applied to the drought forecasting in the Guanzhong Plain. ARIMA models are used in the VTCI series, and forecast its changes in the future. A new way of modeling for the spatio-temporal series is used in the VTCI series. The time series of 36 pixels are studied firstly for their fitting models. Then the ARIMA model fitting for the whole area is determined. The AR(1) model are chosen to be the best model used in each pixel of the whole area, and the forecast is done with 1-2 steps. The results show that forecasting accuracy is better, 1 step is better than 2 steps. The historical VTCI data are simulated by AR(1) models. Comparing the simulating data with the historical data, the results show that the simulating accuracy is better. Most of the simulating errors are small. All results demonstrate that AR(1) model developed for VTCI series can be used for the drought forecasting in the Guanzhong Plain.