基于SSP-RCP不同情景的京津冀地区土地覆被变化模拟

Simulation of land cover change in Beijing-Tianjin-Hebei region under different SSP-RCP scenarios

如何实现自然与人文双重驱动下的特大城市群地区土地覆被变化的情景模拟,不仅是当前土地覆被变化研究领域的热点问题,也是城镇化可持续发展研究的核心主题之一。本文在对现有土地覆被变化情景模型缺陷进行分析和修正的基础上,构建了自然要素与人文要素耦合驱动的土地覆被情景曲面建模(SSMLC)方法。结合IPCC 2020年发布的共享社会经济路径(SSPs)与典型浓度路径(RCPs)组合的CMIP6 SSP1-2.6、SSP2-4.5和SSP5-8.5的气候情景数据,以及人口、GDP、交通、政策等人文参数,分别实现了SSP1-2.6、SSP2-4.5和SSP5-8.5情景下的京津冀土地覆被变化的情景模拟。模拟结果表明:SSMLC对京津冀地区土地覆被变化模拟的总体精度为93.52%;京津冀地区的土地覆被在2020—2040时段内的变化强度最高(3.12%/10a),2040年以后的变化强度将逐渐减缓;在2020—2100年间,建设用地增加速度最快,增加率为5.07%/10a。湿地的减少速度最快,减少率为3.10%/10a。2020—2100时段内的京津冀土地覆被在SSP5-8.5情景下的变化强度整体高于在SSP1-2.6和SSP2-4.5情景下的变化强度;GDP、人口、交通和政策等人文因子对京津冀地区耕地、建设用地、湿地和水体的影响强度高于对其他土地覆被类型的影响强度。研究结果证实了SSMLC模型能够有效模拟和定量刻画京津冀地区土地覆被空间分布格局在未来不同情景的时空变化趋势和强度,模拟结果可为京津冀协同一体化的国土空间优化配置与规划、以及生态环境建设提供辅助依据和数据支撑。

How to simulate the scenarios of land cover change, driven by climate change and human acitivities, is not only a hot issue in the field of land cover research, but also one of the core topics in the sustainable urbanization. A new method of scenarios of surface modeling in land cover(SSMLC) was developed to simulate the scenarios of land cover driven by the coupling of natural and human factors. Based on the climatic scenario data of CMIP6 SSP1-2.6, SSP2-4.5 and SSP5-8.5 released by IPCC in 2020 that combined the shared socioeconomic paths(SSPS) and typical concentration paths(RCPs), the observation climatic data,the human data of population, GDP and transportation, the current data of land cover in 2020,and the related policies, scenarios of land cover in Beijing-Tianjin-Hebei(BTH) region are simulated under scenarios of SSP1-2.6, SSP2-4.5 and SSP5-8.5 during the periods of 2040,2070 and 2100, respectively. The simulation results show that the overall accuracy of SSMLC in the BTH region is up to 93.52%. During the period from 2020 to 2100, change intensity of land cover in the BTH region is the highest(an increase of 3.29% per decade) between 2020 and 2040, and it would gradually decrease after 2040. Built-up land would have the highest increasing rate(an increase of 4.741% per decade), and the wetland would have the highest decreasing rate(a decrease of 2.64% per decade) between 2020 and 2100. The change intensity of land cover under the scenario SSP5-8.5 is higher than that under the scenarios of SSP1-2.6 and SSP1-2.6 between 2020 and 2100. The impacts of GDP, population, transportation and policies on the changes in cultivated land, built-up land, wetland and water body are generally higher than on other types of land cover. Moreover, the research results indicate that the SSMLC method could be used to explicitly project the changing trend and intensity of land cover under different sceaniros, which will benefit the spatial optimal allocation and planing of land cover, and could be used to obtain the key data for carrying out the eco-environment conservation measures in the BTH region in the future.