Spatial Sequential Modeling and Predication of Global Land Use and Land Cover Changes by Integrating a Global Change Assessment Model and Cellular Automata

作者：Min Cao, Yanhui Zhu, Jinling Quan, Sheng Zhou, Guonian Lü, Min Chen, Mengxue Huang · 发表于：Earth s Future · 年份：2019 · DOI：10.1029/2019ef001228 · 被引用次数：59 · 研究领域：Land Use and Ecosystem Services、Climate change impacts on agriculture、Remote Sensing in Agriculture

Abstract Characterizing land use and land cover change (LUCC) is critical for understanding the interaction between human activities and global environmental changes, such as in biological diversity and the carbon cycle. Both natural cycles and human activities can be better examined with more accurate sources of land use data with higher spatial resolution. More importantly, it is crucial to consider spatial heterogeneity to simulate future changes in LUCC. In this paper, a modeling strategy (hereinafter referred to as GCAM‐CA) that combines a global change assessment model (GCAM) with cellular automata (CA) is proposed. This modeling strategy is designed to sequentially spatialize global LUCCs with 1‐km spatial resolution and 5‐year temporal resolution from 2010 to 2100. The GCAM model is employed to predict the land use and land cover area demands for 283 world regions, which are divided by intersecting 32 geopolitical and socioeconomic regions and 18 agroecological zones. The spatialization rules of CA is trained separately for each world region to distinguish global land use and land cover types. The different spatialization rules and trends in land use and land cover demand for each of the 283 regions reflect the spatial heterogeneity in the GCAM‐CA model. We implement and validate the model for the simulation from 2000 to 2010. Next, the model is used to simulate three future scenarios, REF, G26, and G45, demonstrating that the GCAM‐CA model is effective for future glo...