Satellite-based assessment of Phenological Informatics and associated drivers over a biodiversity hotspot in the North Eastern Region of India

Abstract

This study provides a comprehensive assessment of phenological dynamics in the ecologically diverse North Eastern Region (NER) of India (255,083 km²) using MODIS EVI data (2001–2020). We identified 11 climate-induced phenological shift scenarios, comprising eight translational (TS) and three scaling (SS) shifts, and analyzed their key climatic drivers using machine learning. Results indicate an overall advancement in the Start of Growth (SOG) by 0.50 days/year and a delay in the End of Senescence (EOS) by 0.95 days/year, extending the growing season by approximately 0.8 days/year. The most widespread shift, TS4 (delayed EOS, unchanged SOG), affects 12.83 % of NER's forests, while occurrence of TS7 (advancing SOG, delaying EOS) and TS8 (delaying SOG, advancing EOS) highlight forest adaptation to climate variability. Translational shifts predominantly occur below 600m, particularly in moist deciduous (27.57 %) and subtropical broadleaved (24.54 %) forests. Scaling shifts reveal that SS3 (no change in growth rhythm) dominates 36.36 % of the region, while SS1 (increased vigor) accounts for 23.67 %. Climate analysis identifies rainfall (100 %), solar radiation (91.97 %), soil moisture (90.37 %), and vapor pressure deficit (VPD) as key phenology drivers. Land surface temperature and VPD primarily trigger greening in most translational scenarios, while VPD and solar radiation are key factors in senescence. Rainfall is the dominant factor in scaling scenarios. These findings provide crucial insights into climate-phenology interactions in NER, aiding ecosystem management and adaptation strategies for climate resilience.