黑河中游地区耕地景观演变及社会经济驱动力分析

Dynamic of Farmland Landscape and Its Socioeconomic Driving Forces in the Middle Reaches of the Heihe River

选取1987~2010年的土地利用数据,运用变化轨迹分析方法以及景观指数表征黑河中游地区耕地景观的演变过程,采用偏最小二乘回归模型分析耕地景观演变的驱动力。结果表明:从耕地的变化轨迹来看,研究期间耕地的扩张是黑河中游地区耕地景观演变的主要趋势;从空间上来看,扩张型耕地和缩减型耕地分布趋向集中,主要集中在稳定型耕地边缘,同时耕地扩张的方向趋向河流;从驱动因素分析来看,总人口、城市化、产业发展、机电井数等是促使耕地景观演变的主要社会经济驱动因素,但是各个驱动因素的驱动作用均有差异。

It is a complex procedure to identify, record and monitor land cover change on the earth＇ s surface. Most of the existing Spatio-temporal methods are based on multi-temporal land cover maps or temporal trend analysis of spatial pattern indices. Change trajectory analysis explores the dynamic of landscape pattern in the temporal scale, will not separate the landscape change process, and is an effective method of spatio-temporal modelling. The rapid expansion, abandonment and mis-management of farmland would result in soil salinization and desertification in the inland river basin in the arid region, northwest China. In the countrywide context of continuous decrease of farmland, the evident increase intensifies the imbalance between supply and demand of water resources in the arid region. In order to better understand the Spatio-temporal pattern of farmland change in the middle reaches of Heihe River, this paper used change trajectory analysis and landscape metrics to explore the evolution process of farmland landscape and analyzed its socio-economic driving forces by partial least squares regression （PLSR） models. Spatial patterns of farmland change were quantified by employing distribution-related landscape metrics, such as indices of interspersion（IJI）, connectivity （COHESION） and area-weighted patch fractal dimension index isolation （FRAC-AM）, percentage of landscape （PLAND）, Euclide- an Nearest Neighbor Index （ENN-AM） to analyze characteristics of farmland change based on the remote-sensing images of 1987, 1992, 2001 and 2010. Results showed that the expansion of farmland was one of the main trends of land use change in the middle reaches of the Heihe River. The area of farmland expansion was significantly larger than that of abandonment. The proportion of farmland occupied in study area increased from 15.67% in 1987 to 19.74% in 2010, with a net increase of 43 758.9 hm2 for arable land and converted mainly from wetland and unused land. The expanded and reduced farmland exhibited a concentrated pattern, tending to distribute at the edge of the stable farmland, and farmland expanded towards the river. The total population, urbanization rate, industry development and number of electromechanical well were the primary social and economic driving factors prompting the farmland landscape evolution, but the effects of driving factors differed for different types of farmland. The interaction among driving factors influenced the dynamic of farmland landscape. The water resources exploitation directly influenced the change of social-economic factors such as total population, urbanization rate, industry development. The study showed that the proposed research method effectively visualized and quantified the spatio-temporal dynamics of farmland change and its driving forces.