基于人工蜂群CA模型的南京市土地利用变化模拟与预测

以南京市为例，构建人工蜂群元胞自动机（CA_ABC）模型，对2000—2007年的土地利用变化进行模拟以实现CA_ABC模型的校正，并以2007—2015年的土地利用变化为案例，验证该模型的有效性。模拟结果总体精度（OA）2007年为87.79%，2015年为80.61%；模拟结果的品质因数（FOM）2007年为21.23%，2015年为19.25%。基于CA_ABC模型和马尔可夫链预测未来城市土地总量，对南京市2025和2035年的土地利用格局进行了预测，对城市扩张和生态用地被侵占现象进行分析。模型预测结果表明，未来20年的城市扩张主要以牺牲耕地和林地为代价，2025和2035年80%的城市扩张面积来源于对耕地面积的侵占，17%的城市面积扩张是由2015年的林地转换得到的。研究表明，准确模拟、预测未来城市格局及评估城市扩张能够对生态用地侵占，以及为决策者合理规划城市、推动城市可持续发展提供帮助。

Simulation and prediction of land-use change in Nanjing based on artificial bee colony cellular automata model

This paper builds an artificial bee colony cellular automata (CA_ABC) model with Nanjing as the study area, calibrates CA_ABC model by simulating urban expansion from 2000 to 2007, and validates the CA_ABC model from 2007 to 2015 using the actual land-use data. End-state overall accuracies (OA) are 80.61% for 2007 and 87.79% for 2015, while figure-of-merit (FOM) metrics are 19.25% for 2007 and 21.23% for 2015. This paper uses CA_ABC and Markov chain calculating the total urban land amount in Nanjing in 2025 and 2035, predicts the land-use pattern and analyzes the urban encroachment on ecological land. The prediction results show that urban expansion will mainly come at the expense of cultivated land and forest land in the next 20 years. 80% of the urban expansion in 2025 and 2035 will come from the encroachment on cultivated land, and 17% of the urban expansion area is the encroachment on forest land in 2015. This study shows that accurate simulating and predicting the future urban pattern and evaluating the urban encroachment on ecological land are helpful for decision makers reasonably plan for sustainable urban development.