Regional Variability in Vegetation Greenup Times: A Fine-Grained Analysis Using Ensemble Machine Learning Models

Vegetation dynamics and phenology play an important role in inter-annual vegetation changes in terrestrial ecosystems and are key indicators of climate-vegetation interactions, land use/land cover changes and variation in year-to-year vegetation productivity. Satellite remote sensing data have been widely used for vegetation phenology monitoring over large geographic domains using various types of observations and methods over the past several decades. This study utilizes the Enhanced Vegetation Index 2 (EVI2) from MODIS data(MCD12Q2 Dataset) to analyze some key phenological stages of regional vegetation including Greenup and Dormancy trends from 2001 to 2021, as well as to highlight regional discrepancies and forecast future Greenup times. The findings reveal significant regional variations in Greenup time, For example, the greening of the Northeast Plain of China occurs earlier than that of the Ukrainian Plain and the Mississippi Plain. In addition, Random Forest, AutoRegressive Integrated Moving Average(ARIMA) and Long Short-Term Memory (LSTM) models were employed to present Greenup times have regionally advanced, albeit with varying degrees over the past two decades. These insights are pivotal for regional agricultural planning and ecosystem management, particularly in mitigating climate change impacts.