人与自然双重作用下的长时间序列的若尔盖湿地NPP时空演变研究

From the importance of the carbon cycle in the study of global climate changes, and the great “carbon sinks” function of the wetland, the Zoige wetland, a alpine wetland in China with the largest carbon store, was taken as the study area for its vulnerability importance and the seriously degraded situation which has caused the alpine wetland to turn into a “carbon source” from a “carbon sink”. Aiming at a better protection of the wetland ecosystems and the effective recovery of the degraded ones, this study is carried out to distinct the relative functions of the human and nature in the spatial and temporal variations of the Zoige wetland. While the status quo of the NPP research in the study area is that 1)the existing studies of NPP on spatial and temporal scales are relatively insufficient, 2)the driving factors of NPP being considered are not so comprehensive, only the average status of the climate factors being discussed, and 3)the analysis methods are less quantitative. Based on the above, the GIMMS - AVHRR NDVI and MODIS NDVI data sets are chosen as the data source to build the long time series of NPP of Zoige wetland. The CASA model, one of the widely accepted energy use efficiency model, is being as the tool for the NPP evaluation research. On the basis of the thoroughly analysis of climate changes and human activities of the Zoige wetland region, the spatial and temporal variations of the long time series of NPP in different time scales（inter-annual and monthly scale ） are discussed. Besides, the driving factors of this process are quantitatively analysized both considering the average status and variability properties of the climate factors. Further, the relative effects of climate change and human activities in the process are quantitatively defined and the spatial and temporal variations of the relative effects is also discussed. This study can not only more comprehensively and scientificly reveal the spatial and temporal variation patterns and its driving forces of the Zoige wetland NPP, but can better promote the understanding of the relationship between climate and NPP. That will help to protect the current wetland system of the Zoige Plateau and restore these degraded wetland system, as well as to provide some more reliable theoretical references for the “carbon sink” function of the Zoige Plateau wetland being efficiently played.

碳循环在全球变化研究非常重要，湿地作为重要碳库对其研究亦很必要。若尔盖湿地是我国碳储量最大的高寒湿地，气候和人类活动的双重作用使其由“碳汇”向“碳源”转变，形势严峻。为有效促进区域湿地生态系统保护及退化湿地生态系统的恢复，本研究以揭示若尔盖湿地NPP时空演变进程中人与自然的作用量为目标，基于该区现有NPP研究时空尺度不够完善，驱动力分析指标不够全面，分析方法定量性不足的现状，在分析若尔盖湿地气候变化和人类活动双重作用的背景基础上，构建不同时间尺度长时间的NPP序列，分析区域NPP的空间分异特征及其与环境因子的定量关系，并从综合考虑气候因子平均状况及变异性角度探讨区域NPP时间演变规律及其驱动力，对这一进程的气候变化和人类活动相对作用进行定量区分并分析其时空格局与规律。不仅利于对NPP-气候关系的理解，利于区域退化生态系统的恢复与重建的有效性和针对性及“碳汇”功能的发挥。

若尔盖湿地是我国碳储量最大的高寒湿地，气候和人类活动的双重作用使其由“碳汇”向“碳源”转变。本研究以揭示若尔盖湿地NPP时空演变进程中人与自然的作用为目标，构建长时间的NPP序列，分析区域NPP时空演变格局及驱动力，并实现气候变化和人类活动相对作用的定量区分。研究发现：.1）区域气温呈上升趋势，降水呈下降趋势，蒸发量也呈增大趋势，整体气候朝暖干化方向发展。各季节降水呈明显减少趋势，气温和蒸发则以增加为主，只在秋季有变化。总体呈相对湿润度指数下降趋势，且1995年发生显著趋干的突变，并存在三种主要尺度的周期变化。.2) 提出了长时间序列时空融合模型（LTSTAFM）和长时间序列线性融合模型（LTLFM）两种长时间序列高时空分辨率NDVI数据集构建方法。两模型所产生的NDVI的结果差异不十分显著，但LTSTAFM模型更适于河流、耕地、起伏山地等扰动较大区域的时间序列NDVI的融合重建，而LTLFM模型则在草地等均质、变化相对规律的区域表现更佳。.3）基于重构的NDVI时间序列，选用改进的CASA模型估算和验证了若尔盖湿地1982-2013年植被NPP，得出若尔盖湿地年均NPP为215gC∙m^(-2)左右，与实测值较为一致。.4）若尔盖湿地近32年来多年平均NPP的范围在31 gC∙m^(-2)∙a^(-1)~640 gC∙m^(-2)∙a^(-1)之间。植被NPP在若尔盖湿地的东北部若尔盖县、西南部的阿坝县和红原县的南部边缘一带偏高，中部次之，西北部的玛曲县、北部碌曲县和东南部的红原县边缘一带最低。NPP总量呈减少变化趋势。影响植物生产力的气候因子主要为日照时数和降水量。在年尺度上，最主要因素为日照时数。而在生长季内，则主要是降水量。.5）若尔盖湿地近32年来人类活动干扰作用强烈。20世纪80-90年代，人类活动干扰不是造成湿地植被净初级生产力变化的主导因素。随着时间推移，21世纪00-10年代，人类活动对湿地植被净初级生产力的干扰程度持续上升，其影响区域较20世纪90年代相比，玛曲县人类活动区域有所转移。. 该研究结果不仅利于对NPP-气候关系的理解，更利于区域退化生态系统的恢复与重建的有效性和针对性及“碳汇”功能的发挥。