基于地理探测器的西南岩溶槽谷区近20年

植被覆盖度变化及其与驱动因子关系的研究是开展大规模植被恢复效益评估的基础，而目前对西南槽谷地区植被绿化的时空变化模式及其驱动因素的认识并不明晰，不利于进一步生态建设。基于2000—2018年西南槽谷区NDVI、气温、降雨、DEM、土地覆被和人口密度数据，应用线性趋势回归分析和Hurst指数分析等方法以及地理探测器模型，探究了西南槽谷区植被覆盖的时空变化特征和驱动因素，预测未来变化趋势。结果表明：（1）近19年来NDVI范围介于0.79~0.84，总体呈波动上升趋势，岩溶区植被NDVI （0.003 17/a）年增长率显著高于非岩溶区（0.002 60/a）。（2） Hurst指数分析结果表明，西南槽谷区的植被NDVI主要以上升趋势为主，但其中64.31%在未来呈现退化趋势，植被保护形势较为严峻。（3）岩溶区植被NDVI与温度变化之间存在负相关关系（R=-0.040），而非岩溶区为正相关关系（R=0.013）。残差分析结果表明，人类活动促进89.60%的区域植被NDVI增长。（4）在整个岩溶槽谷地区，土地覆被类型和气温是影响植被NDVI的主要驱动因素，解释率在25%以上，各因子的交互作用明显高于单因子作用。总体来看，人类活动对槽谷区植被恢复具有明显的积极影响。

Variation Characteristics and Influencing Factors of NDVI in Karst Trough Area of Southwest China in Recent 20 Years Based on Ggeo-detector

The study on the change of vegetation coverage and its relationship with the driving factors is the basis of benefit evaluation on large-scale vegetation restoration, but the spatial-temporal change patterns of vegetation and its driving factors in the southwest karst trough-valley region are still not clear at present, which is not conducive to further ecological construction. This study explored the spatio-temporal characteristics and driving factors of vegetation coverage change based on the data of NDVI, air temperature, rainfall, DEM, land cover and population density in the southwest trough valley area from 2000 to 2018 and to predict the future trends. The methods of linear trend regression analysis and Hurst index analysis as well as the geographic detector model were applied. The results showed that: (1) In recent 19 years, the range of NDVI was 0.785~0.838, which showed a fluctuating upward trend, and the annual growth rate of vegetation NDVI (0.003 17/a) in karst area was significantly higher than that in non-karst area (0.002 60/a). (2) Hurst index analysis showed that the vegetation NDVI in the study area mainly showed an upward trend, but 64.31% of them would show a degradation trend in the future, and the situation of vegetation protection was more severe. (3) There was a negative correlation between NDVI and temperature change (R=-0.040) in karst area, but a positive correlation (R=0.013) in non-karst area. The results of residual analysis showed that human activities promoted the NDVI growth of 89.60% of regional vegetation. (4) In the whole karst trough-valley area, land cover type and air temperature were the main driving factors affecting NDVI, the interpretation rate was over 25%, and the interaction of each factor was obviously higher than that of a single factor. On the whole, human activities had a significant positive impact on the vegetation restoration in the southwest trough valley area.