Identifying Influential Climate Factors of Land Surface Phenology Changes in Songnen Plain of China Using Grid-based Grey Relational Analysis

The variation of land surface phenology responds to climate change and controls many feedbacks of vegetation to the climate system. In this study, spatial and temporal patterns of land surface phenology in Songnen Plain of China were investigated between 1983 and 2012 using the third generation NDVI (Normalized Difference Vegetation Index) dataset from GIMMS (Global Inventory Modelling and Mapping Studies). The contribution of climate factors on the start (SOS), end (EOS) and length of growing season (LOS) were identified by grid-based grey relational analysis. Our results indicate that average SOS and EOS dates in Songnen Plain showed clear spatial patterns. Delayed SOS accounted for 12.7% of the study area with a highly significant increased rate of 0.54 days per year (p<0.05). 12.2% of the total land presented prolonged EOS at highly significant rates of 0.38 days per year. Areas showing shortened and delayed LOS occupied 17.9% and 14.2%, respectively. Spatially, the most influential factor affecting SOS changes was sunshine duration, the relative humidity in winter and spring. EOS changes were mostly associated with spring sunshine hours, relative humidity and temperature in summer. Sunshine duration and relative humidity were the main influential factors in regulating the variations in LOS. The response of the land surface phenology to climate factors showed different spatial patterns in different seasons. The results may provide a basic reference for studying the ecosystem functioning in temperate semi-humid and semi-arid region.