Temporal And Spatial Patterns Of Vegetation Phenology And NPP And Their Responses To Climate Change In Qilian Mountains

The study of vegetation dynamics in terms of vegetation phenology and net primary productivity（NPP）in the context of climate warming is one of the core elements of global change research.The Qilian Mountains,located in the northwest of China,is an important ecological zone of great ecological value.The function and structure of the ecosystem in this area are facing increasingly severe tests.As the dynamics of vegetation in the region are not well characterized,Therefore,analyzing the characteristics of vegetation phenology and NPP changes in the Qilian Mountains and their response to regional climate factors is essential for understanding the mechanisms of ecosystem stability and predicting the carbon cycle of the regional ecosystem,as well as for the rational and effective use of vegetation resources and the sustainable development of the local agriculture and livestock industry in the region.Based on remote sensing datasets and meteorological raster datasets,this study systematically investigates the multi-year characteristics and spatial and temporal trends of vegetation phenology and NPP in Qilian Mountains from 2001 to 2020,as well as their responses to climate factors at different time scales,using dynamic threshold method,Theil-Sen and Mann-Kendall test,persistence and stability analysis,Pearson correlation analysis and partial least squares regression（PLSR）,The main conclusions are as follows:（1）Based on the Double Logistic Dynamic threshold method to extract vegetation phenology datasets from remote sensing datasets,the results show that the Start of growing season（SOS）,End of growing season（EOS）and Length of growing season（LOS）in the Qilian Mountains from 2001 to 2020 differ in terms of average characteristics,spatial and temporal variation and vegetation types.The average characteristics of the entire study area were day154 for SOS,day 273 for EOS,and day121 for LOS;The time variation showed that SOS advanced by 3.56d/10a,EOS delayed by 2.07 d/10a,and LOS extended by 3.30 d/10a.In terms of spatial variation,there is an"altitudinal gradient effect"in the vertical direction of vegetation phenology in the study area,vegetation SOS delayed by 15.67 d/km,EOS advanced by 12.02 d/km,and LOS shortened by 19.24 d/km.In the horizontal direction SOS,EOS and LOS are dominated by advance,delayed and increasing trend from southeast to northwest.In terms of the different vegetation types,the desert vegetation had the largest SOS advance（5.30 d/10a）.The delay of EOS was the largest in broadleaf forest（5.51 d/10a）.The increase of LOS in broad-leaved forest was the largest（8.20 d/10a）.（2）The multi-year average of vegetation NPP in the Qilian Mountains from2001 to 2020 is 2082×10^-4kg C/m²and the multi-year trend is mainly increasing,with an increase of 227.1 10^-4kg C/m²,and the increasing trend covers 97.8%of the total area.In terms of spatial variation,the NPP of vegetation decreases at a rate of405×10^-4kg C/m²/1km with increasing elevation,while the horizontal direction shows a gradual decrease from southeast to northwest.The NPP levels of different vegetation types differ,specifically,NPP were greatest for broadleaf forests(4420×10^-4kg C/m²)and least for desert vegetation(799×10^-4kg C/m²).In addition,the largest increase in NPP was in meadows and the smallest increase was in alpine vegetation.（3）The effects of temperature and precipitation on vegetation phenology and vegetation NPP were more obvious.In terms of vegetation phenology,SOS was mainly positively correlated with the previous year’s winter temperature and significantly negatively correlated with the current year’s spring temperature,of which the negative influence with the current year’s spring mean minimum temperature was particularly prominent,with the area significantly negatively correlated accounting for 30.2%of the total area;EOS was mainly positively correlated with the current year’s summer and autumn temperatures and SPEI,and negatively correlated with precipitation.The area of significant positive correlation with the minimum temperature in autumn of the year accounted for 18.4%of the total area;vegetation NPP specifically showed positive correlation with the temperature and precipitation in spring and summer,with the positive response of NPP to the mean minimum temperature in summer being particularly significant,accounting for 53.8%of the total area.This indicates that the increase in temperature and precipitation in the study area is conducive to the accumulation of NPP in the vegetation in the study area.（4）There is a close relationship between vegetation phenology and NPP.The results of the analysis showed that 67%of the areas showed a negative correlation between SOS and NPP,the advancement of SOS favored the increase of NPP;51%of the areas showed a positive correlation between EOS and NPP,the delay of EOS helped the accumulation of NPP;62.7%of the areas showed a positive correlation between LOS and NPP,the increase of LOS favored the increase of NPP.