Journal Article

Comparative analysis of ecological sensitivity assessment using the coefficient of variation method and machine learning

Lv, Baohong, Wang, Changming, Zhang, Zefang

2024

Overview

Ecological sensitivity is an essential indicator for measuring the ecological environment’s level, and its assessment results have significant reference value for regional ecological environment protection and resource development and utilization. Taking Xifeng County as the study area, we selected a total of 12 assessment factors in terms of ecological environment, geological environment, and human environment, including average annual rainfall, average annual temperature, average annual wind speed, river density, vegetation coverage, soil erodibility, elevation, slope, geological disaster susceptibility, road density, land use, and night light index, and explored the spatial distribution patterns of ecological sensitivities and the characteristics of the differences in the study area based on the coefficient of variation method and machine learning. The results show that the overall spatial distribution pattern of ecological sensitivity in Xifeng County shows a low sensitivity in the north and a high sensitivity in the south. 41.90% of the regional ecological sensitivity intensity is classified as very high and high sensitivity, mainly distributed in mountainous and hilly areas, while 35.51% is classified as very low and low sensitivity, mainly distributed in plains and low mountainous areas. The assessment results are consistent with the actual situation, enriching the ecological sensitivity assessment methods and models.

• Publisher: Springer International Publishing,Springer Nature B.V
• Subject: Accuracy / Annual rainfall / Annual rainfall data / Annual temperatures / Atmospheric Protection/Air Quality Control/Air Pollution / China / Coefficient of variation / Comparative analysis / Conservation of Natural Resources - methods / Distribution patterns / Earth and Environmental Science / Ecology / Ecosystem / Ecosystems / Ecotoxicology / Environment / Environmental economics / Environmental impact / Environmental Management / Environmental Monitoring - methods / Environmental protection / erodibility / Geology / humans / Land use / Learning algorithms / Luminous intensity / Machine Learning / Monitoring/Environmental Analysis / Monte Carlo simulation / Mountain regions / Mountainous areas / Mountains / Neural networks / Plant cover / rain / Regional analysis / Regional development / Resource development / resource management / rivers / Sensitivity analysis / Soil erodibility / Soil erosion / Soil temperature / Spatial distribution / Support vector machines / temperature / Vegetation / Wind speed