| 鄂东北大别山区生态环境质量时空演变及归因分析 |
| 摘要点击 499 全文点击 20 投稿时间：2024-12-27 修订日期：2025-04-10 |
| 查看HTML全文
查看全文 查看/发表评论 下载PDF阅读器 |
| 中文关键词 遥感生态指数（RSEI） 生态环境质量（EEQ） 时空分异 随机森林（RF） 响应阈值 |
| 英文关键词 remote sensing ecological index （RSEI） ecological environmental quality （EEQ） spatio-temporal heterogeneity random forest（RF） response threshold |
| DOI 10.13227/j.hjkx.20260222 |
|
| 中文摘要 |
| 揭示鄂东北大别山区生态环境质量（EEQ）时空演变规律及其驱动机制，是长江中游生态屏障建设的重要科学依据. 研究集成多源遥感、气象及社会经济数据，构建“格局-驱动-阈值”三位一体分析框架，通过融合Theil-Sen Median趋势分析、Mann-Kendall检验与Hurst指数预测，系统解析2001~2021年EEQ时空分异特征. 时空耦合分析表明：①2001~2021年间，研究区多年EEQ均值经历了从0.470上升至峰值0.514后又回落至0.497的波动变化，但整体呈现改善趋势，好转区域面积占比51.95%. ②EEQ等级空间分异显著，良好和优秀一级主要分布在西北部、东北部和西南部地区，生态环境质量“较差”级别主要分布在中部和南部沿江平原地区. ③研究区生态环境质量驱动因素探究的最佳空间尺度为7.5 km，EEQ在不同年份对不同因素的响应程度存在差异. 2001~2021年间，土地利用强度、植被覆盖度、降水量、海拔、温度和建设用地面积占比是研究区EEQ空间分异的主导驱动因素，且土地利用强度的胁迫作用逐年增强，人口密度和人均GDP对EEQ影响较小. EEQ对土地利用强度、温度和建设用地面积占比呈负向非线性响应，与植被覆盖度、海拔和降水量之间呈正向非线性关系. ④结合Sen趋势和Hurst指数得到未来区域EEQ改善区域面积占比为39.72%，恶化区域面积占比30.75%. 研究成果可为动态优化生态保护红线、实施建设用地强度管控及构建气候适应性管理策略提供定量化决策支持，并为同类型山地生态系统可持续治理提供范式参考. |
| 英文摘要 |
| Elucidating the spatio-temporal evolution patterns and driving mechanisms of ecological environment quality （EEQ） in the Dabie Mountain region of northeastern Hubei provides critical scientific support for ecological security barrier construction in the middle Yangtze River Basin. This study develops a triadic analytical framework （pattern-driver-threshold） through the integration of multi-source remote sensing， meteorological， and socioeconomic data， systematically investigating EEQ dynamics from 2001 to 2021 via combined Theil-Sen Median trend analysis， Mann-Kendall testing， and Hurst index prediction. Key findings include：① An overall EEQ improvement trend occurred （51.95% of areas enhanced） despite cyclical fluctuations， with multi-year means evolving from 0.470 in 2001 to a peak of 0.514 before declining to 0.497 in 2021. ② There was pronounced spatial stratification showing “excellent/good” grades concentrated in northwestern， northeastern， and southwestern mountainous zones versus “poor” levels in central-southern river plains. ③ The optimal spatial scale for driver analysis was identified as 7.5 km， with land use intensity （exhibiting growing stress effects）， vegetation coverage， precipitation， elevation， temperature， and built-up area proportion constituting dominant drivers. EEQ demonstrated negative nonlinear responses to land use intensity， temperature， and built-up areas， contrasting with positive relationships to vegetation coverage， elevation， and precipitation， whereas socioeconomic factors （population density， GDP） showed negligible impacts. ④ Projections indicated persistent spatial polarization， with 39.72% of the region predicted for EEQ improvement against 30.75% degradation areas. These results provide quantitative foundations for dynamically optimizing ecological conservation boundaries， regulating construction land intensity， and formulating climate-resilient governance strategies， while establishing a transferable paradigm for sustainable mountain ecosystem management. |
