岷江上游土地利用与生态系统服务价值的动态变化

岷江上游是中国西南典型生态脆弱区之一，研究岷江上游土地利用与生态系统服务价值（Ecosystem Services Values，ESV）的动态变化，对区域生态保护，实现可持续发展具有重要意义。基于岷江上游2000、2005、2010和2015年四期遥感解译数据，利用CA-Markov模型预测了2035年土地利用格局，并引入空间异质系数、社会经济调整系数和资源稀缺系数对生态系统服务价值进行修正，构建适用于研究区的ESV评估模型，分析了岷江上游2000-2035年土地利用及ESV的时空动态变化。结果表明：1）岷江上游土地利用类型以林地和草地为主。建设用地和耕地面积持续增加，草地、水域和未利用地呈现波动变化；2）2015-2035年和2000-2015年相比，ESV增幅下降，土地提供生态系统服务的能力有所降低。维持生物多样性、保持土壤、气体调节、水文调节和气候调节是岷江上游生态系统的主要功能；3）岷江上游中东部及河谷地带是ESV低值及损失冷点集中区，西部地区是ESV高值及增加热点集中区；4） ESV对系数缺乏弹性，林地和草地面积的变化是引起生态系统服务价值变化的主要敏感因子。研究结果对研究区土地利用格局优化以及推行生态文明建设具有重要参考意义。

Dynamic changes of land use and ecosystem services value in the upper reaches of the Minjiang River

The upper reaches of the Minjiang River (URMR) are one of the typical ecologically fragile areas in southwest China. Studying the dynamic changes of land use pattern and the ecosystem service values (ESV) in the URMR is highly significant for the ecological protection and sustainable development of the region. Based on four phases of remote sensing data obtained during 2000, 2005, 2010, and 2015 in URMR, we predicted the land use pattern in 2035 by using the CA-Markov model. The ecosystem service value was revised by introducing the coefficients of spatial heterogeneity, socioeconomic adjustment, and resource scarcity. Subsequently, an ESV evaluation model for the study area was developed. The spatiotemporal changes in land use and ESV in the URMR were analysed for the period 2000 to 2035. The results indicate that:1) the two main types of land use in the URMR are forest and grassland. The area of construction land and cultivated land exhibited an increasing trend, whereas grassland, water bodies, and unutilized land showed fluctuations. 2) Compared with the period 2000-2015, the growth rate of ESV and the ability of land to sustain ecological services declined in 2015-2035. The maintenance of biodiversity and soil, and the regulation of gas, hydrology, and climate were found to be the main functions of the ecosystem in the URMR. 3) Cold spots with ESV loss and low ESV were concentrated in the middle-eastern part of the URMR and the valley area, whereas hot spots with ESV increment and high ESV were concentrated in the western area. 4) ESV was relatively inelastic with respect to the coefficient, and changes in forest and grassland areas were the most sensitive factors that generate changes in the ecosystem service values. Our results have significant importance for realizing a reasonable land use pattern of the study area that facilitates the construction of an ecological civilisation.