Changes In Vegetation Phenology And Productivity And Their Responses To Climate Factors On The Loess Plateau During 2001-2016

The relationship between vegetation phenology and productivity in response to regional climate change is one of the core contents of global climate change research.The Loess Plateau is an ecologically fragile and climate-sensitive region in China.The analysis of the characteristics of vegetation change and its response to climate change on the Loess Plateau is conducive to the better implementation of the policy of returning farmland to forest(grassland)and the utilization of vegetation resources in the region,and is also conducive to the stable development of local agriculture and animal husbandry.Based on the MODIS NDVI data from 2001 to 2016,this study extracted the beginning and end periods of the vegetation growing season on the Loess Plateau and calculated the length of the vegetation growing season by using the dynamic threshold method.Combined with the vegetation productivity data and meteorological observation data in the same period,This paper systematically analyzed the spatiotemporal variation characteristics of vegetation phenology and productivity in the Loess Plateau and their spatiotemporal responses to different climatic factors.Based on the multi-year variation trend of vegetation phenology and productivity,the interaction between them was discussed.The main conclusions of this study are as follows:(1)The results show that the start of the vegetation season on the Loess Plateau from 2001 to 2016 is mainly in early April,the earliest start date is less than 90 d and the latest is more than 160 d.The end date of the growing season is generally in late October,the end date is less than 280 days at the earliest and 310 days at the latest.The average length of the growing season is 162 days,the shortest length of the growing season is less than 120 days and the longest is more than 200 days.The beginning of the growing season showed an advanced trend with a rate of 6 d/10a,and the end of the growing season showed a delayed trend with a rate of 2.8 d/10a.The length of the growing season showed an extended trend with a rate of 8.8 d/10a.The annual mean value of vegetation productivity on the Loess Plateau is 515.17 g C/m2,and the spatial distribution pattern is high in the southeast and low in the northwest.The vegetation productivity increased at a rate of 12.76 g C/(m~2·a),and the growth trend gradually decreased from southeast to northwest.(2)In terms of time pattern,the overall increase of pre-season precipitation and sunshine duration contribute to SOS advance of vegetation.The influence of temperature on SOS of vegetation is complex.As a whole,the increase of temperature before last season is favorable for vegetation to begin to grow,but the average temperature in October and February,the maximum temperature in October-December and the minimum temperature in November and January will delay the arrival of vegetation growing season.The EOS of vegetation was most significantly affected by precipitation in July.The increase of precipitation in July was conducive to postponing the end date of the vegetation growing season,while the increase of sunshine duration in most months would lead to the early end of the vegetation growing season.An increase in the average temperature in July and the maximum temperature in September would lead to an earlier EOS,while an increase in the minimum temperature at night in September would delay the vegetation EOS.(3)In terms of spatial pattern,the relationship between the beginning of vegetation growing season and climate factors is mainly negative.Vegetation SOS in the Loess Plateau is greatly affected by temperature,in which average temperature,daytime temperature and night temperature dominate the area of the beginning of vegetation growing season,accounting for 26.03%,36.74%and 15.35%,respectively.Vegetation SOS is mainly distributed in the southwest of the Loess Plateau,which is controlled by pre-season precipitation and sunshine duration.There was no significant negative correlation between vegetation EOS and sunshine duration,but positive correlation between vegetation EOS and other climatic factors.Among them,33.43%of vegetation EOS is controlled by pre-season rainfall,and the land cover type in these areas is mainly woodland.The vegetation EOS of 18.2%was controlled by sunshine factor,which was consistent with the spatial distribution of rainfall.Vegetation EOS was controlled by the three temperature factors in the same area,and there was no obvious spatial distribution rule.(4)The increase of average temperature on the Loess Plateau is not conducive to the accumulation of vegetation productivity.The possible reason is that the increase of temperature will lead to faster evapotranspiration,which leads to water stress on vegetation.Vegetation productivity is positively correlated with the other four climatic factors,and there is an obvious spatial distribution rule.Specifically,the GPP of vegetation in the northern part of the Loess Plateau is mainly affected by precipitation,while the GPP of vegetation in the southern part of the Loess Plateau is mainly affected by temperature.The maximum temperature has a great influence on the southeastern part of the Loess Plateau,while the minimum temperature mainly affects the southwest part of the Loess Plateau.The area dominated by sunshine duration accounted for3.87%,which was mainly distributed in the alpine regions of Qinghai in the form of pixels.(5)The analysis results of the relationship between vegetation phenology and productivity showed that the extension of vegetation growing season contributed to the increase of vegetation productivity in 74%of the regions,while the reduction of LOS in 23.5%of the regions contributed to the increase of GPP.The advance of the beginning of vegetation growing season can promote the increase of spring vegetation productivity in 71%of the regions,and the increase of spring vegetation productivity in 21.5%of the regions is caused by the delay of vegetation SOS.The vegetation with delayed EOS and increased GPP accounted for 67.3%of the study area,while the vegetation with advanced EOS and increased GPP accounted for 28.8%of the study area.