Locust remote sensing monitoring methods based on landsat8 satellite data

Regional locust does great harm to agricultural production. Real-time monitoring of the development of locust is of great significance for the locust control. We took three counties in northern Chifeng City of Inner Mongolia as the study area. Firstly, we classified locust host plants by using the sophisticated remote sensing classification algorithm on the Landsat8 OLI data, overlapped with prior locust distribution regions, and distinguished the locust suitable bases regions. Then, we retrieved some important locust habitat parameters, such as leaf area index, land surface temperature and soil moisture by using Landsat8 satellite data. Meanwhile, the synchronous investigation data, land cover data, historical locust hazard data were combined for analysis and modeling. Finally, we used stepwise regression analysis to obtain the relationship between locust density and leaf area index, land surface temperature and soil moisture. The model results showed a high accuracy with R² of 0.50 and RMSE of 3.17. It is indicated that the Landsat8 satellite data has a certain potential in locust remote sensing monitoring, and the research provides an important reference for similar studies.