Vegetation structural shift tells environmental changes on the Tibetan Plateau over 40 years

Area of Science:

• Ecology
• Remote Sensing
• Deep Learning

Background:

• Understanding grassland structural changes on the Tibetan Plateau is vital for ecosystem functioning and environmental change detection.
• Regional-scale data on grassland community structure has been limited by methodological constraints.

Purpose of the Study:

• To characterize formation-based plant community structure in alpine grasslands at the Tibetan Plateau regional scale for the first time.
• To compare historical changes using multivariate data fusion and deep learning against the Vegetation Map of China.
• To identify drivers of grassland structural changes, differentiating between climate and human influences.

Main Methods:

• Utilized multivariate data fusion and deep learning techniques.
• Characterized formation-based plant community structure in alpine grasslands.
• Compared current data with the historical Vegetation Map of China.

Main Results:

• The proportion of alpine meadows increased from 50% to 69% over 40 years, correlating with warming and wetting trends.
• Dominance of Kobresia pygmaea and Stipa purpurea formations strengthened to 76% and 92%, respectively.
• Stipa purpurea distribution is climate-driven, while Kobresia pygmaea distribution is influenced by human activities.

Conclusions:

• Grassland changes on the Tibetan Plateau over the past 40 years are formation-dependent.
• This study provides a new perspective on the drivers and spatial heterogeneity of grassland changes.
• The research innovates large-scale vegetation study paradigms by incorporating structural information.