Long-time series ecological environment quality monitoring and cause analysis in the Dianchi Lake Basin, China

As the core area in the transformation of Kunming into an international center city, studying the changes in ecological environment quality and causes of the Dianchi Lake Basin is of great significance for its future optimization of the landscape pattern. This study is based on the Google Earth Engine (GEE) platform to calculate the Remote Sensing Ecological Index (RSEI) of the Dianchi Lake Basin from 1990 to 2020. Then we used Mann-Kendall mutation detection to obtain the time points when significant changes in RSEI occurred. Finally, the Geodetector and MGWR models were combined to analyse the driving factors of ecological quality changes in the Dianchi Lake Basin. The results show that: (1) The ecological environment quality of the Dianchi Lake Basin showed an increasing trend from 1990 to 2020, with the mean value of RSEI increasing from 0.49 to 0.52. (2) According to the results of the mutation test, the years 1990, 1993, 2006, 2015, and 2020 were used as the time points for monitoring the ecological environment quality of the Dianchi Lake Basin over a long time series. The ecological environment quality of the study area in the past 30 years was mainly in an improved state, accounting for 49.43%. The ecological deterioration areas are mainly located in the northeastern part of Xishan District (north of Caohai Lake), the southwestern part of Guandu District, Kunyang Town in Jining District, and the northern areas of Jincheng and Shangsuan Town. (3) The single factor detection results show that elevation and slope have the strongest influence on RSEI. The q-value of average annual temperature has changed the most, from the 6th to the 3rd place. This indicates that the urban heat island effect and the expansion of construction land have had a greater impact on the quality of the local ecological environment in recent years. The multi-factor interaction test shows that the influence of each factor on RSEI is enhanced after the interaction. (4) The MGWR regression results show that the actual scales of action of the factors are inconsistent, with the most significant spatial heterogeneity in the Percentage of cropland area. Based on the above findings, it can provide data to support the future urban planning of the Dianchi Lake Basin. It also provides a new means of integrating Geodetector and MGWR into the study of ecological quality analysis.