| Summary: | Mid- to high-latitude Asia (MHA) is one of the regions with the strongest warming trend and it is also a region where ecosystems are most sensitive to climate variability. However, how the vegetation in the region will change in the future remains uncertain. Using observation-based Leaf Area Index (LAI) and meteorological data and the multiple regression method, this study analyzes the response of vegetation in the MHA to climate elements during 1982–2020. Then, machine learning prediction models based on the Random Forest (RF) and Extreme Random Tree (ERT) algorithms are built and validated. Based on the calibrated meteorological fields from 17 Coupled Model Intercomparison Project Phase 6 (CMIP6) models under intermediate (SSP2-4.5) and high (SSP5-8.5) emission scenarios and the machine learning models, the LAI over the MHA in 2021–2100 is projected. The historical long-term increasing trends of LAI in the MHA since 1982 are found to be mainly caused by the increasing near-surface air temperature, while the interannual variations of LAI are also greatly affected by precipitation and surface downward solar radiation, especially in summer. The LAI over most of the MHA shows a significant increasing trend in the future, except over some dry areas, and the increasing trends are stronger under the SSP5-8.5 scenario than under the SSP2-4.5 scenario.
|
|---|