Forecasting anthracnose-twister disease using weather based parameters: geographically weighted regression focus

Abstract

The aim of this study is to identify the environmental factors that may influence the onion anthracnose-twister disease incidence and severity. In this study, Geographically Weighted Regression (GWR) analysis was used to identify the dominant environmental factors that might influence the occurrence of Anthracnose-twister disease of onion using Geographic Information System approach. The onion disease records were acquired from the Institute of Climate Change and Environmental Management. The weather parameters such as relative humidity, cumulative rainfall and temperature were acquired from the National Aeronautics and Space Administration website while the river parameters were generated from Sentinel-2 images. This study has identified the ‘distance to river’ and ‘rainfall’ factor as the two (2) important factors that may influence the occurrence of the disease. The predictive surface map generated from GWR model was able to predict the occurrence of the disease in onion field by as much as 86% in the study area. The results of the study can be used to forecast the occurrence of anthracnose-twister disease in the onion fields the future.