全球变化知识资源中心

Vegetation plays an important role in terrestrial ecosystems and affects both water and carbon cycles. It is crucial to predict vegetation phenology accurately to understand the Earth's ecosystems better. In this study, we evaluated the temperate grassland phenology simulated by the Community Land Model 4.5 for the Northeast China Transect with the Moderate Resolution Imaging Spectroradiometer leaf area index (LAI) data set. Results show that although the CLM could partly capture the seasonal patterns of LAI for temperate grasslands, shortcomings lie in a delayed end date of growing season and a problematic senescence existed in early summer. The reason is attributed to the low temperature threshold set for leaf offset and an inaccurate carbon allocation strategy for its display and storage. With the adjusted key phenology parameters, the modeled temperate grassland phenology has been substantially improved for this system: An overestimation of the growing-season LAI as well as the annual gross primary productivity can be constrained. Meanwhile, the modified phenology would effectively change the partition of evapotranspiration in a more reasonable pattern.