ScholarWorks@Y-Scholar Hub

초록

The leaf area index (LAI) is a key variable for representing vegetation state, and it is closely related to simulating carbon and water exchanges between land and the overlying atmosphere in land surface and terrestrial ecosystem models. Model simulations are still limited in their representation of vegetation phenological processes and capturing the resulting LAI seasonality. Therefore, this study demonstrated how LAI assimilation into the model improved carbon and water fluxes in different ecosystems over East Asia. We assimilated LAI derived from Moderate Resolution Imaging Spectroradiometer data for seven years (2004–2010) over East Asia into the Community Land Model version 4.5 with a biogeochemistry module (CLM4.5-BGC) by employing the ensemble adjustment Kalman filter method. Results showed that LAI assimilation remarkably improved estimated gross primary production (GPP). In particular, the root mean square error decreased from 97.12 to 48.63 gC/m2/month across the region for June–August. Additionally, while evapotranspiration (ET) was less sensitive to LAI than GPP, the ET components of ground evaporation, canopy evaporation, and canopy transpiration significantly changed after assimilation. The analysis of plant-available soil water showed that LAI assimilation has unique effects on soil moisture depending on the soil layer, climate, and ecosystems. In general, the improvement in ecological prediction skill by LAI assimilation was particularly evident in temperate needleleaf forests where LAI was overestimated most distinctly. This study improves our understanding of the role of LAI assimilation in eco-hydrological processes in different ecosystems with CLM4.5-BGC, which allows for the improvement of model forecasting and more accurate simulation of the effects of LAI state evolution. © 2021 The Author(s)

키워드