Abstract: Taking Shule River Basin as a case, this paper analyzed landscape ecological risk variability by establishing a risk index based on landscape pattern and process using GIS and RS technology. Spatialtemporal processes and aggregation characteristics of ecological risk were analyzed. Minimum cumulative resistance model was used to build river basin ecological security pattern with taking natural reserves, waters, forest lands as the “source”, and ecological security level, elevation and slope as resistance factors to generate resistance surface. The potential corridors and nodes were identified and the optimized ecological security network was established. Results showed that the ecological risk level in the north of Shule River Basin was higher than that in the south. In general, the ecological security level in the Shule River Basin had been improved in recent 30 years. From 1980 to 1995, the areas of Shule River Basin with degraded ecological security level were mainly located in the southwest and the east; the area with degraded ecological security level during 1995-2010 was less than that during 1980-1995. The landscape risk index presented a positive spatial autocorrelation. The spatial autocorrelation degree weakened during the past 30 years, which indicated a reduction in spatial convergence. The optimized ecological network made full use of the existing river system to get through the corridors, and avoid urban industrial land, mining land and transportation lines. Also, the optimized ecological network combined ecological source region, potential corridors, potential nodes and other landscape components, and set up key points in the ecologically fragile regions as the “pedals” of species dispersal.

Key words: maize-soybean relay intercropping, P use efficiency, adsorption-desorption., P application rate, P application depth