基于遥感生态指数的塔里木垦区生态环境质量评价

【目的】调查塔里木垦区近十年间生态环境质量的变化情况,为政府制定农业发展和生态格局规划提供参考。【方法】以Landsat卫星数据为基础,通过ENVI软件得到湿度指标(WET)、绿度指标(NDVI)、干度指标(NDBSI)、热度指标(LST),利用遥感生态指数(RSEI)模型和主成分分析(PCA)结果对塔里木垦区2010年、2015年、2020年生态环境质量做出评价。【结果】2010年、2015年、2020年,塔里木垦区RSEI数值分别为0.301、0.292、0.315,区域生态环境质量总体处于较差等级,但RSEI数值呈现先下降后升高、总体升高的趋势。塔里木垦区生态环境改善和优化的面积占比9.10%,主要分布在恰拉水库南侧及大西海水库东侧;变差和恶化的面积为2.32%,集中在大西海水库西侧。【结论】2010—2020年,塔里木垦区生态环境质量总体处于较差等级,但随着时间推移有所提高。对水体掩膜内的MNDWI进行正规化处理,并将其压缩至0.8～1的区间内,以此来填补RSEI中缺失的水域信息,可以提高模型评价效果。

Evaluation of Ecological Environment Quality in Tarim Reclamation Area Based on Remote Sensing Ecological Index

【objective】It is to find out the changes in the quality of the ecological environment in the Tarim Reclamation Area in the past ten years,and to provide references for the government to formulate agricultural development and ecological pattern planning.【Methods】Based on Landsat satellite data,using ENVI software preprocessing to acquire humidity,greenness,dryness and thermal indexes and using the improved remote sensing ecological index(RSEI)model and principal component analysis results,the ecological environment quality of the Tarim Reclamation Area for the years of 2010,2015 and 2020 was evaluated.【Result】In the years of 2010,2015 and 2020,the RSEI of the Tarim Reclamation Area was 0.301,0.292 and 0.315,respectively.The quality of the regional ecological environment was generally at a poor level,but the RSEI showed a pattern of“decline first,increase then,and increase overall”.9.10%of the ecological environment in the Tarim Reclamation Area was improved and optimized,mainly distributing on the south side of Qala Reservoir and east side of Daxihai Reservoir;and the proportion of the deteriorated area was 2.32%,mainly concentrated on the west side of the Daxihai Reservoir.【Conclusion】From 2010 to 2020,the ecological environment quality of the Tarim Reclamation Area is at a poor level,but it has improved over time.The MNDWI in the water mask is normalized to the range of 0.8~1,to insert the missing information on water areas in the initial RSEI.This could improve the evaluation effect of the model.