关键词: 资源禀赋, 发展需求, 元胞自动机, 人工神经网络, 最小累积阻力模型, 五大连池风景区

Abstract: With the development of economy, there are different degrees of unbalanced matches between resource endowments and development intentions in different regions. It is necessary to quantify the decision-making behavior of land-use subjects through one or more feasible methods. The spatial pattern can be optimized using model simulation, landscape pattern, and ecological function evaluation. Taking Wudalianchi Scenic Area as the research area, based on the land use data of Wudalianchi in 2000 and 2018, we selected NDVI, elevation, slope, soil organic carbon, distance to road, and population density as driving factors, coupled with Markov model, artificial neural network (ANN), cellular automata model (CA), minimum cumulative resistance model (MCR) and ecological footprint algorithm to simulate land use change. It was estimated that ecological healthcare function under the ecological protection scenario based on the ecological footprint algorithm would increase by 1.87% compared to the natural state in 2030, with the help of air negative ion supply function accounting. Counting through the landscape pattern index showed that the scattered and juxtaposed index and landscape segmentation index would decrease and the patch aggregation degree and landscape connectivity would increase after setting restricted areas. The focus of Wudalianchi’s future development plan is to weigh the proportion of cultivated land, woodland, and grassland, and to take into account the tourism development of the scenic spot and the protection of ecological functions.

Key words: resource endowment, development need, cellular automata, artificial neural network, minimum cumulative resistance model, Wudalianchi Scenic Area.