Simulation of the spatial pattern of land use change in China: the case of planned development scenario

土地利用变化研究在环境可持续发展研究领域中具有重要的地位,其空间分布格局的变化影响到生物地球化学循环、气候变化、生物多样性等。采用土地利用动态变化模型Dyna-CLUE模拟了在规划情景下中国土地利用变化未来空间分布格局。将土地利用类型分为六大类,即耕地、草地、林地、建设用地、水域和其它用地。驱动因子包括地形地貌、气候、社会交通等方面,对动态驱动因子如气温、降水、人口交通等,考虑了其在未来情景下的发展趋势。基于土地利用类型与驱动因子之间的定量关系和土地利用类型之间的转换规则等,模拟出至2020年中国土地利用分布格局。结果表明,至2020年,中国东南部、黄淮海平原、四川盆地等地区耕地面积将增加,东北、西北等农牧交错区、农林交错区和沙漠边缘耕地面积将会呈轻度减少趋势;林地面积将增加1417.91万hm²,主要发生在中国东北部以及西南部水热条件好的地区;中国草地在面积上保持稳定,空间上中东部、东南地区草地面积减少,内蒙古中部,青海东部,四川盆地北缘区和青藏高原等地面积增加;建设用地增加531.76万hm²,主要发生在中国的东部地区。;Land use change is a key subject in the research of sustainable development in environment. The spatial pattern of land use change closely related to earth system functioning, such as climate warming, biogeochemical circle and landscape biodiversity. In order to improve ecological environment and promote social development, a series of land use policies such as afforestation, restoration of degraded grassland and protection of cultivated land had been formulated. These policies will exert a great influence on the spatial pattern of land use in China. However, the land use policies only provide an overview of the land use changes at the national scale but can't give insight into the changes at the regional and landscape scales. In this paper, the Dyna-CLUE model, which is a dynamic, spatially explicit land use change model had been used to simulate the spatial pattern of land use change in China in the coming decades. The planned development scenario was developed, in which the total area for each land use types in the future were defined as required by the land use policies. The Chinese level land use demands were downscaled to land use pattern at 2 km² resolution. Six land use types were distinguished which are built-up land, arable land, grassland, forest land, water area and other land. The spatial allocation of land uses were simulated based on the location suitability and user-specified decision rules. The driving factors include climatic and economic condition, traffic situation, soil texture, topography and demography. Logistic regression was used to quantify the relation between land use patterns and these drving factors. Climatic factors, traffic and population were defined as dynamic driving factors. In the future, the HadCM3 B2 climatic scenario was adopted to provide climatic data; the spatial pattern of population was simulated by SMPD (surface modeling of population distribution), and the railway and road development plan was made by the government. Other stable driving factors such as topography, soil texture were assumed to remain unchanged in the future 15 years. The performance of the land use change model was validated, showing that this method can simulate the spatial pattern of land use change accurately. The results indicate that the area of cultivated land would keep no less than 120.33 million hectares, however, it would decrease in western region where the land is not suitable for cultivation and would increase in central south China. The forest area would increase by 14.28 million hectares, mainly in northeastern and southwestern China, where the climate is sufficiently hospitable for forest growth. The area of built-up land would increase by 5.3176 million hectares, mainly in eastern and southeastern regions of China which are characterized by high population density and advanced economy. The simulation has the potential to help decision makers and scientists identify the critical regions that need specific consideration. The high spatial resolution of the results enable the assessment of impact of land use change on a large number of environmental indicators, including climate change, carbon sequestration and landscape diversity.