Monitoring of hydrological situation in the western part of Songnen Plain based on fusion index
Based on the Google Earth Engine(GEE)platform and Landsat remote sensing data,a study was conducted on the hydrological situation in the western region of the Songnen Plain from 2011 to 2020.By integrating multiple vegetation hydrological indices such as normalized difference vegetation index(NDVI),modified normalized water body index(mNDWI),automatic water extraction index(AWEIsh),and utilizing the ReliefF importance feature selection algorithm and CART decision tree model,dynamic monitoring of water distribution in the region from 2011 to 2020 was conducted.The results indicate that compared to existing single index algorithms,the fusion index has significant advantages in remote sensing water body recognition,and can effectively detect water bodies obscured by vegetation.The water area in the western part of the Songnen Plain had experienced a decrease followed by a fluctuating increase over the past 10 years,showing an overall decreasing trend.As of 2020,the identified water area was 24 118.05 km².The result of land type conversion analysis showed that the receding water bodies were mainly transformed into vegetation,with a net outflow area of approximately 5 388.78 km².By comparing the main precipitation changes and the impact of human activities,it was found that the main reason for the reduction of water bodies in the western part of the Songnen Plain was human interference.This study provides a reference for water resources management and ecological protection in the Songnen Plain.
万方数据
维普
