Prediction of Pine Wilt Disease in Jiangsu Province Based on Web Dataset and GIS

Abstract

Abstract80 pine wilt disease occurrence points with geographical coordinates in 2007 and 31 environmental variables from open web datasets were gathered as the main source of information. Four modeling methods of Classification and Regression Trees (CART), Genetic Algorithm for Rule-set prediction (GARP), maximum entropy method (Maxent), and Logistic Regression (LR) were introduced to generate potential geographic distribution maps of pine wood nematode in Jiangsu province, China. Then we calculated three statistical criteria of area under the Receiver Operating Characteristic Curve (AUC), Pearson correlation coefficient (COR) and Kappa to evaluate the performance of the models. The results showed that: CART outperformed other three models; slope, precipitation, seasonal variations (bio15), mean temperature of driest quarter (bio9), north-south aspect (northness), maximum temperature of warmest month (bio5) were the six enforcing environmental factors; future occurrence area of pine wilt disease will be 47.27% of total pine forest, tripling present infected area of the pest.Keywordspine wilt diseasepredictionweb datasetGIS