Research On The Spatial And Temporal Trend Of Vegetation Phenology Of Tibetan Plateau From 1982 To 2017

Vegetation phenology has been regarded as a simple and effective indicator to reflect the response of vegetation to global climate change.Dynamics of the spatial and temporal pattern of vegetation phenology plays an important role in identifying and predicting the response to ecosystem to the frequent occurrence of mutations and gradients in meteorological factors.Tibetan Plateau ecosystem is mainly characterized by highly heterogeneity,sensibility and vulnerability.Apart from it,there is lack of ecological monitoring stations and meteorological stations in the northwest of the region.Thus,applying remote sensing data to retrieve vegetation phenology can be an increasing way to make related research in Tibetan Plateau as well as cover the shortage of limitation of data observation.However,it was found that discrepancy existed in phenology metrics using greenness vegetation indices from different kinds of satellite sensors and a variety of determining methods,which aroused abroad controversy in the validity of monitoring over the Tibetan Plateau.Furthermore,it has had influence on understanding of response to global climate change in the alpine ecosystem.In this study,we compared the regularity of the general distribution of phenology metrics derived from Normalized Difference Vegetation Index(NDVI)and Solar-Induced Chlorophyll Fluorescence(SIF),respectively.To explore possible similarities and differences of vegetation phenology between two types of data sources in Tibetan Plateau,we also investigated the spatial distribution pattern and temporal trends in different scales.At last,we analyzed the sensibility of vegetation phenology on main meteorological factors in Tibetan Plateau.The main results are as follows:The spatial pattern of vegetation phenology determined by two kinds of data were consistent in the Tibetan Plateau.Plants tend to start growing earlier and end up earlier in the northeast part of the region and later in the southwest,and it may cause a lengthening growing season from northwest to southeast.There were significant differences in the general distribution of phenology metrics derived from two types of data.The discrepancy of vegetation phenology from NDVI and that from SIF in various types of vegetation in Tibetan Plateau was noticeable.As for SOS,the difference in broad-leaved forests was the smallest and that in the grasslands was the largest.As for EOS and LOS,the difference in the wildness was the smallest and that in the shrubs was the largest.In the Tibetan Plateau,SOS derived from NDVI showed a delayed trend from 1982 to 2013,which was not significant.From 2001 to 2017,SOS derived from SIF showed a significantly advancing trend,with about 6.4 days advance per decade.EOS from NDVI and SIF experienced a trend of significant advance,with about 1.6 days and 3.2 days per decade,respectively.The substantial spatial heterogeneity of trends in SOS was stronger than that in EOS,with a notably delay in the southwest region and widespread advancing trend in the other areas from both two data sources.A delay of EOS was found in the southwest,while it presented no trend in any other regions.There was a significant delay along with increasing altitude in SOS,with more intense change from SIF(0.9 d/100m)than from NDVI(0.4 d/100m).While EOS had little significant trend as altitudes elevated except for EOS from NDVI.LOS decreased significantly with elevation rise,and the trend from SIF and NDVI is consistent.Local characteristics of SOS in different parts of the Tibetan Plateau was involved with its unique terrain and vegetation types,so its response to meteorological factors was also diverse.From 1982 to 2013,SOS in the southwest of the Tibetan Plateau and Hoh Xil mountain was positively correlated with summer temperature and that in the northwest border and the eastern three river sources region was negatively correlated with summer temperature and that in the north of the Tibetan plateau and the southeast of Hoh Xil mountain and Tibetan Plateau was positively correlated with summer precipitation.We mainly analyzed the spatio-temporal trend and we also explored spatial heterogeneous pattern response to meteorological factors tentatively.And we found that the influence to main results from two types of data,which was mainly related to the selection of data acquisition.In this study,it highlights the significance of SIF for deriving vegetation phenology on the Tibet Plateau.Thus,the basic data and effective scientific proof will be provided to characterize the pattern of vegetation phenology in the Tibetan Plateau through this study,as well as reinforce the understanding of response of climate change to alpine ecosystem.