Vegetation plays a critical role in Earth’s surface systems and serves as a key indicator of global climate change. Traditional methods for studying vegetation trends often fall short in capturing its long-term, nonlinear characteristics. In this study, we utilized advanced techniques such as ensemble empirical mode decomposition and the Breaks for Additive Seasonal and Trend (BFAST) algorithm to reconstruct and analyze the spatiotemporal evolution and abrupt changes in China’s Normalized Difference Vegetation Index (NDVI) data from 1982 to 2018. We also explored the driving forces behind China’s vegetation variations. Our findings revealed significant vegetation browning that had been obscured by positive trends. Notably, there were abrupt changes in vegetation growth in the 1990s and 2000s, affecting approximately 50% and 33.6% of the study area, respectively. Climate emerged as the primary driver of vegetation change in China, exerting a positive influence in the southern regions but a negative effect in the northwest. Initially, human activities had a negative impact on vegetation growth but later turned positive. Specifically, we observed a U-shaped relationship between CO2 emissions and vegetation growth, indicating that while CO2 fertilization initially enhances growth, excessive emissions can hinder further development. Population density had a minor restraining effect on vegetation growth, whereas ecological restoration initiatives had a limited but positive influence on vegetation growth.