GIS-based multi-criteria analysis model for identifying probable sites of locust outbreak in Xinjiang, China

Abstract

Locusts are a kind of primary pests that cause severe damage to the agriculture in Xinjiang, northwest of China. Early forecasting probable sites of locust outbreaks are very important for rangeland management and agricultural protection. This study promoted a GIS-based model combining with multi-criteria analysis to predict the possible area where locust might outbreak. Factors including monthly average temperature, monthly relative humidity, elevation, slope, NDVI and soil PH value were used in this model. The results showed that the locusts were mainly distributed in the north and west part of Xinjiang, which was highly consistent with the actual locust distribution. The average accuracy was 84.37%, and the highest accuracy that appeared in Urumqi reached 97.25%. The average empowering weight method is more suitable for this study as the accuracies are both higher than 90% in 2011 and 2012. Hence, this model was able to predict the probable sites of locust outbreak in Xinjiang, which would provide valuable information to locust control and prevention authorities.