Change Of Vegetation Phenology Based On NDVI And Its Relationship With Climatic Factors

Vegetation phenology is a sensitive indicator of climate change,and its changes directly reflect the response of vegetation growth to climate change.Research of vegetation phenology is important for understanding the impact of globle climate change on terrestrial ecosystems.Based on the GIMMS NDVI 3g v1 dataset,we used the Polyfit-Maximum?PM?method to extract start date of vegetation growing season?SOS?,end date of vegetation growing season?EOS?and length of vegetation growing season?LOS?from 1982 to 2015,and the spatiotemporal variations of SOS,EOS and LOS were analyzed.Combined the concurrent meteorological dataset,the relationship between SOS and climatic factors?temperature and precipitation?were analyzed,and the response mechanism of SOS to climate change were explored.1.Trends of vegetation phenology and its shift patterns in different climatic zones in the middle and eastern EurasiaFrom 1982 to 2015,SOS across the middle and eastern Eurasia and for the four climatic zones?Cold zone,Cold-Temperate zone,Temperate zone and Tibetan Plateau?all showed an advanced trend,EOS all showed a delay trend,and LOS all showed an extended trend.At climatic zone scale,the trends of vegetation phenology have changed,and its patterns are different.The turning point?TP?of SOS trends in Cold zone,Temperate zone and Tibetan Plateau occurred in the middle or late 1990s.After TP,the advanced trends of SOS in Cold zone,Temperate zone and Tibetan Plateau showed accelerated,stagnant and reversal patterns,respectively.The TP of EOS trends in Cold-Temperate zone and Tibetan Plateau occurred in1987 and 1989,respectively.The TP of EOS trends in Cold zone and Temperate zone occurred in 1995 and 1994,respectively.After TP,the delay trends of EOS in the Cold zone,Cold-Temperate zone,Temperate zone and Tibetan Plateau slowed down or disappeared.LOS trends and its changes are controlled by SOS and EOS,but SOS is the dominant factor.Discrepanies in pattern of vegetation phenology trend reflect the long-term and short-term characteristics of climate change and its spatial heterogeneity.2.Changes of spring phenology and sensitivity analysis in the temperate grassland and desert zones of ChinaThe SOS in the temperate grassland and desert zones of China was in general advanced trend,with an average rate of 0.14 d·a^-1from 1982 to 2015.SOS in meadow,steppe and desert vegetation have advanced by rates of 0.08 d·a^-1,0.17 d·a^-1and 0.10 d·a^-1,respectively.Spatially,68%of natural vegetation pixels showed an advanced trend in SOS,whereas 32%of pixels displayed a delay trend.Preseason mean temperature is the main factor driving variations of SOS,and the mean temperature during 3 months before vegetation growing season has a greater effect on SOS.SOS was most correlated with mean temperature during 40 d preseason period and cumulative precipitation during 200 d preseason period,respectively.From 1982 to2015,a warming of 1?in preseason?40 d?mean temperature would significantly advance SOS by 1.31 d over the temperate grassland and desert zones of China,whereas an increase of10 mm in preseason?200 d?cumulative precipitation would advance SOS by 0.44 d.Preseason precipitation determines the spatial patterns of temperature sensitivity of SOS and precipitation sensitivity of SOS.Spatially,SOS is more sensitive to preseason mean temperature in areas with more preseason precipitation;SOS is more sensitive to preseason precipitation in areas with less preseason precipitation.3.Changes of spring phenology and its temperature sensitivity in the temperate forest zone of ChinaThe SOS in the temperate forest zone of China mainly occur from mid-April to early May,and preseason?November-April?temperature is the primary controlling factor of SOS spatial pattern.SOS over in the temperate forest zone of China has advanced by 0.09 days per year during 1982 to 2015.SOS in needle-leaf forest,broadleaf forest,shrub,steppe and grass all showed an advanced trend,and SOS in needle-leaf and broadleaf mixed forest,meadow and swamp all showed a delay trend.The mean temperature during 2-3 months before vegetation growing season has the most significant effect on interannual variation of SOS.The preseason?56 d or 72 d?mean temperature by 1?would lead to an advance in SOS of 2.00 days.Preseason cumulative precipitation has less effect on the interannual variation of SOS,and is heterogeneous.Increase in preseason cumulative precipitation would lead to a delay SOS in deciduous needle-leaf forest,needle-leaf and broadleaf mixed forest,broadleaf forest,meadow and swamp,and an advanced SOS in shrub,steppe and grass.Results also show that preseason temperature and preseason precipitation determine the spatial variation of temperature sensitivity of SOS.The temperature sensitivity of SOS became significantly stronger with rising in preseason mean temperature or increasing in preseason cumulative precipitation.