Discrepancies in vegetation phenology trends and shift patterns in different climatic zones in middle and eastern Eurasia between 1982 and 2015.

Yaobin Li,Fengxue Gu,Shirong Liu,Yuandong Zhang

doi:10.1002/ece3.5408

Abstract

Changes in vegetation phenology directly reflect the response of vegetation growth to climate change. In this study, using the Normalized Difference Vegetation Index dataset from 1982 to 2015, we extracted start date of vegetation growing season (SOS), end date of vegetation growing season (EOS), and length of vegetation growing season (LOS) in the middle and eastern Eurasia region and evaluated linear trends in SOS, EOS, and LOS for the entire study area, as well as for four climatic zones. The results show that the LOS has significantly increased by 0.27 days/year, mostly due to a significantly advanced SOS (−0.20 days/year) and a slightly delayed EOS (0.07 days/year) over the entire study area from 1982 to 2015. The vegetation phenology trends in the four climatic zones are not continuous throughout the 34‐year period. Furthermore, discrepancies in the shifting patterns of vegetation phenology trend existed among different climatic zones. Turning points (TP) of SOS trends in the Cold zone, Temperate zone, and Tibetan Plateau zone occurred in the mid‐ or late 1990s. The advanced trends of SOS in the Cold zone, Temperate zone, and Tibetan Plateau zone exhibited accelerated, stalled, and reversed patterns after the corresponding TP, respectively. The TP did not occurred in Cold‐Temperate zone, where the SOS showed a consistent and continuous advance. TPs of EOS trends in the Cold zone, Cold‐Temperate zone, Temperate zone, and Tibetan Plateau zone occurred in the late 1980s or mid‐1990s. The EOS in the Cold zone, Cold‐Temperate zone, Temperate zone, and Tibetan Plateau zone showed weak advanced or delayed trends after the corresponding TP, which were comparable with the delayed trends before the corresponding TP. The shift patterns of LOS trends were primarily influenced by the shift patterns of SOS trends and were also heterogeneous within climatic zones.

Highlights

| INTRODUCTIONPhenology of surface vegetation is highly sensitive to regional and global climate change (Cleland, Chuine, Menzel, Mooney, & Schwartz, 2007; Fu et al, 2015; Menzel & Fabian, 1999; Menzel et al, 2006; Wang et al, 2016), and its interannual changes can strongly affect the carbon balance, as well as nitrogen and water cycles of global ecosystems (Cornelissen et al, 2007; Piao et al, 2015; Richardson et al, 2010; Shen et al, 2014)
In another way, provides continuously spatial and temporal informations on a variety of surface vegetation at regional or global scales, and numerous studies have investigated the variations of vegetation phenology in different regions over different time periods based on satellite‐measured Normalized Difference Vegetation Index (NDVI) datasets (Fu et al, 2014; Myneni, Keeling, Tucker, Asrar, & Nemani 1997; Shen et al, 2018; Wang et al, 2016; White et al, 2009; Wu & Liu, 2013; Zhang, Tarpley, & Sullivan 2007)
Upon further analysis of the linear trends during the two periods before and after the corresponding Turning points (TP), we found that the delayed trends of end date of vegetation growing season (EOS) in the four climatic zones were stalled and even reversed during the later period

Summary

| INTRODUCTION

Phenology of surface vegetation is highly sensitive to regional and global climate change (Cleland, Chuine, Menzel, Mooney, & Schwartz, 2007; Fu et al, 2015; Menzel & Fabian, 1999; Menzel et al, 2006; Wang et al, 2016), and its interannual changes can strongly affect the carbon balance, as well as nitrogen and water cycles of global ecosystems (Cornelissen et al, 2007; Piao et al, 2015; Richardson et al, 2010; Shen et al, 2014). Traditional ground‐based observations are widely used for the species level and for small‐scale areas because detailed information can be documented at the observation; these observations are limited by the location and number of observation sites and spatial scopes (Cleland et al, 2007; Studer, Stöckli, Appenzeller, & Vidale, 2007; Wu & Liu, 2013; Zhu et al, 2012) Remote sensing, in another way, provides continuously spatial and temporal informations on a variety of surface vegetation at regional or global scales, and numerous studies have investigated the variations of vegetation phenology in different regions over different time periods based on satellite‐measured Normalized Difference Vegetation Index (NDVI) datasets (Fu et al, 2014; Myneni, Keeling, Tucker, Asrar, & Nemani 1997; Shen et al, 2018; Wang et al, 2016; White et al, 2009; Wu & Liu, 2013; Zhang, Tarpley, & Sullivan 2007). A new generation of GIMMS NDVI 3g dataset has been extended to 2015

| MATERIALS AND METHODS

| DISCUSSION

Findings

| CONCLUSIONS