基于临界慢化模型和长时间序列叶面积指数的植被及其恢复力遥感监测研究——以三峡库区为例

基于临界慢化模型，利用长时间序列叶面积指数(GLASS LAI)数据，进行时间序列分解后，计算了LAI及其时间自相关指数作为指标，对三峡库区植被及其恢复力进行监测，通过案例模型对临界慢化模型精度进行了验证，分析了三峡库区植被及其植被恢复力的时空分布特征，探索基于临界慢化模型的植被恢复力遥感定量估算方法的适用性。结果表明：(1)2000—2018年三峡库区LAI平均值为3.4,重庆段LAI较低，湖北段LAI较高；三峡库区LAI整体呈上升趋势，重庆段LAI呈现降低趋势，显著下降区域占重庆段面积的21.75%,湖北段LAI呈现升高趋势，显著上升区域占湖北段面积的21.22%;(2)2000—2018年三峡库区重庆市北碚区、大渡口区、渝北区植被恢复力较低，宜昌市兴山县、夷陵区、点军区植被恢复力较高；(3)模型精度方面，在两个地质灾害扰动事件中案例模型结果与临界慢化模型结果呈现较高的一致性。本文对三峡库区2000—2018年的植被恢复力进行了定量估算，同时通过案例模型对临界慢化模型在恢复力监测上的有效性进行了验证，为三峡库区制定相应生态环境管理决策提供理论基础，为保障西南地区生态安全提供决策依据...

Remote sensing assessing of vegetation and its resilience based on critical slowing down model and GLASS LAI: A case study in the Three Gorges Reservoir Area

Vegetation resilience is the ability of vegetation to recover from disturbances without shifting to an alternative state or losing function and services, which is critical to maintain ecosystem quality and stability. Assessing vegetation resilience has become an urgent requirement to deal with ecosystem degradation under the climate change and influence of anthropogenic. However, large scale vegetation resilience measurement is fraught with difficulty, since the lack of remote sensing production and limitation of measurement model. Here, we used GLASS LAI and critical slowing down model to monitor the pattern of vegetation and vegetation resilience in the Three Gorges Reservoir Area (TGRA). We also analyzed the accuracy of critical slowing down model using case model to discuss the feasibility in remote sensing vegetation resilience monitoring. In general, LAI autocorrelation as an indicator monitored the vegetation resilience of each district in the TGRA. We found that the average LAI was 3.4 and LAI showed an increasing trend during 2000-2018 in the TGRA. LAI of Chongqing section showed a decrease trend while that of Hubei section showed an increase trend. Spatially, the area of significant decline accounted for 21.75% of the Chongqing section and the significantly increased area accounted for 21.22% of the area of Hubei section. For vegetation resilience, Hubei section showed stronger resilience than Chongqing section. Within TGRA, Beibei, Dadukou and Yubei exhibited low vegetation resilience while Xingshan, Yiling and Dianjun exhibited high vegetation resilience. In terms of model accuracy, the results of the case model and the critical slowing down model were with a high consistency in two geohazard disturbances. Overall, when assessing vegetation resilience in a large scale using long-term remote sensing data, the critical slowing down was able to offer reasonable indicators. Furthermore, our results indicated that anthropogenic factors had the negative effects on vegetation resilience. Confirmation with ground data will be needed to validate these results and to better understand the biological processes determining vegetation restoration ability.