基于CA-Markov的土地利用时空变化与生境质量预测——以宁夏中部干旱区为例

探究干旱半干旱区土地利用结构动态变化特征,评估并预测生境质量,可为区域生态规划和恢复提供决策性依据。本研究以搭载OLI和ETM+传感器遥感影像和气象数据、社会经济数据等为基础,采用土地利用转移流概念、InVEST模型和CA-Markov模型研究2000-2030年宁夏中部干旱区土地利用和生境质量时空变化特征及规律,对未来情境进行模拟和预测。研究结果表明：宁夏中部干旱区土地系统结构变化与生态建设规划高度一致,具有黄河流域两侧集中分布的特征,土地系统综合动态度逐期减小,活跃度具有稳中有变、部分较活跃的特征。土地系统信息熵特征值呈逐年降低趋势。随着时间变化土地系统受人类活动干扰强度低,自我调整程度高,修复能力强,系统向稳定状态转化。2000-2015年研究区生态环境质量呈“U”型特征,以优秀等级为主。2015-2030年土地系统结构将发生显著变化,生境质量优秀、差和良好等级面积逐渐扩大,生态系统呈现稳定向好的趋势。本研究的多模型集成应用可为区域土地规划和生态恢复建设提供理论依据和支撑。

Predicting spatiotemporal changes in land use and habitat quality based on CA-Markov: A case study in central Ningxia, China

The ecological impacts of land use have become a contested topic in ecological environmental research. An understanding of the land-use structure characteristics in arid and semi-arid areas allows for habitat quality assessment and prediction and provides a decision-making guide for regional ecological planning and restoration. Remote sensing apparatus equipped with an operational land imager and enhanced thematic mapper plus sensors, meteorological data, and socio-economic data were used to explore the temporal and spatial changes of land use and habitat quality in the arid area of central Ningxia, China, from 2000 to 2030. Land-use transfer flow, the InVEST model, and the CA-Markov model were used to simulate the habitat characteristics and laws and generate predictions. The land-use transfer flow, which explored the dynamic changes of the land system structure, showed significant results and explained the land-use transfer laws and spatial distribution characteristics. The land system structure changes were consistent with the ecological construction plan and had a concentrated distribution on both sides of the Yellow River basin. The land system''s comprehensive dynamics gradually decreased and stabilized. The entropy value showed an annual downward trend, indicating that over time, with less human disturbances, the land can self-adjust and self-repair, creating a stable state. From 2000 to 2015, the ecological environment quality showed "U" shape characteristics, indicating an excellent habitat quality grade. From 2015 to 2030, the land system structure will undergo significant changes; the excellent, good, and poor habitat quality areas will gradually expand, and the ecosystem will stabilize and improve. The multi-model integration application can thus provide support for regional land planning and ecological restoration.