Abstract The environmental restoration of terrestrial ecosystems helps to protect the natural world and enhances sustainable land resource development. Modern and efficient approaches for the conservation of ecological functions must be established for more severe land degradation. In this paper, artificial intelligence assisted intelligent planning framework has been proposed to manage the environmental restoration of the terrestrial ecosystem. Facilitating balance of ecosystem service provision, demand, and using machine learning to dynamically build Biological Retreat Configuration (BRCs) that helps better to apprehend the influence of urban growth on environment-related procedures. Such factors can be used as a theoretical reference in the combination of commercial development and eco-friendly conservation. The BRC of the metro area of Changsha Zhuzhou Xiangtan (CZX) has been developed in this study to classify ecological sources using the Bayesian network model efficiently. Using the Least Collective Resistance (LCR) model and circuit theory, the environmental passage and environmental strategy points were established. The BRC was developed by integrating seven environmental factors with 35 ecological policy points. The results showed that the supply and demand of organic unit services (EUS) were spatially decoupled with the deterioration in locations with a significant EUS trend. The urban agglomeration's environmental sources and ecological corridors have been primarily located in forests and waters. The terrestrial environmental pathway has been scattered around the outer edge of the region, while the aquatic green corridor has been extended over the whole town. The environmentally sensitive areas were located primarily around the borders of the growing region and the intersections between land development and forest area. Finally, environmental components have been mainly identified in existing zones of biological defense, which support the effectiveness of Machine Learning (ML) in green sources forecasting and offer novel insight into the development of urban BRCs. The proposed approach has proven to be effective for the planning of assessing environmental restoration in terrestrial ecosystems.

中文翻译：

---

人工智能辅助陆地生态系统环境恢复智能规划框架

摘要 陆地生态系统的环境修复有助于保护自然世界，促进土地资源的可持续开发。对于更严重的土地退化，必须建立保护生态功能的现代有效方法。本文提出了人工智能辅助的智能规划框架来管理陆地生态系统的环境恢复。促进生态系统服务供应、需求的平衡，并使用机器学习动态构建生物撤退配置 (BRC)，有助于更好地理解城市增长对环境相关程序的影响。这些因素可以作为商业开发与生态保护相结合的理论参考。本研究开发了长沙株洲湘潭都市区的 BRC（CZX），以有效地使用贝叶斯网络模型对生态源进行分类。利用最小集体阻力（LCR）模型和电路理论，建立环境通道和环境策略点。BRC 是通过整合 7 个环境因素和 35 个生态政策点而制定的。结果表明，有机单位服务 (EUS) 的供需在空间上与具有显着 EUS 趋势的地点的恶化脱钩。城市群的环境源和生态廊道主要分布在森林和水域。陆地环境路径已经散布在该地区的外围，而水上绿色长廊则延伸至全镇。环境敏感区主要位于种植区的边界和土地开发与森林区的交汇处。最后，主要在现有生物防御区中确定了环境成分，这支持机器学习 (ML) 在绿色资源预测中的有效性，并为城市 BRC 的发展提供新的见解。所提议的方法已被证明对评估陆地生态系统环境恢复的规划是有效的。环境成分主要在现有的生物防御区中确定，这支持机器学习 (ML) 在绿色资源预测中的有效性，并为城市 BRC 的发展提供新的见解。所提议的方法已被证明对评估陆地生态系统环境恢复的规划是有效的。主要在现有生物防御区中确定了环境成分，这支持机器学习 (ML) 在绿色资源预测中的有效性，并为城市 BRC 的发展提供新的见解。所提议的方法已被证明对评估陆地生态系统环境恢复的规划是有效的。