Spatiotemporal Variations in Satellite-Derived Vegetation Phenological Parameters in Northeast China

Vegetation phenology does not only serve as a key index of terrestrial ecosystem response to worldwide climate change but also has a major influence on plant productivity and the carbon cycle. In the current research, the change of vegetation phenological parameters was studied and the impact exerted by climate change on phenological phases in northeast China for 1982-2014 was explored using the latest edition of the Global Inventory Modeling and Mapping Studies Normalized Difference Vegetation Index (GIMMS NDVI3g) dataset. The results showed that the start of the growing season (SOS) slightly advanced, the end of the growing season (EOS) showed a significant delay, and the length of the growing season (LOS) exhibited a significant prolonging at the regional scale. At the different vegetation types scale, there existed diverse responses of vegetation phenological phases to climate change for forest, grassland, and cultivated land. Significant decreasing trends in the SOS occupied 19.1% of the entire research area, whereas pixels with significantly increasing trends in the SOS accounted for 13.1%. The EOS was delayed in most of the study region (approximately 72.1%). As the result of the variations of SOS and EOS, the LOS was obviously enhanced (p < 0.05) in 29.7% of the research area. According to the correlation of vegetation phenology with climate factors, the SOS had a significant negative relationship with the average temperature in springtime, while the EOS was notably negatively connected to summer total precipitation at the regional scale. At the pixel scale, the correlation of phenological parameters with climate variables showed strong spatial heterogeneities. This study contributes to the comprehension of the responses of vegetation phenology to climate change.