Effects of satellite LAI data on modelling land surface temperature and related energy budget in the Noah-MP land surface model

Abstract

• Satellite LAI scheme can provide more accurate vegetation dynamic information. • Satellite LAI scheme improved LST simulation in Noah-MP land surface model. • Mechanism of LAI on LST simulation was revealed by energy budget analysis in model. Leaf area index (LAI) is essential for modelling the land surface temperature (LST) and related energy budget in the land surface models (LSMs). Due to the lack of a parameterization scheme based on satellite LAI data, it is unclear whether satellite LAI data can improve the performance of Noah with multi-parameterization (Noah-MP) LSM to model LST in China land data assimilation system (CLDAS). Here, by using satellite LAI data reconstructed by the advanced deep learning method, we established a new vegetation scheme based on satellite data for the Noah-MP LSM in the CLDAS. We quantitatively evaluated simulation effects of different vegetation schemes including the parameter table scheme (TAB), the dynamic vegetation scheme (DVEG), and the satellite LAI data scheme (OBS), on modelling LAI, LST and their related energy budget in China from 2016 to 2018. We found that except for the cropland, the LAI of the vegetation in growing season observed by satellite was lower than that simulated by the Noah-MP LSM. Validations based on the automatic weather stations in the vegetation-covered area showed that compared with the TAB and DVEG scheme, the OBS scheme improved the negative bias and high unbiased root mean square error (ubRMSE) of simulated LST. Compared with the TAB and DVEG scheme, the optimal improvements of the OBS scheme for annual mean bias and annual mean ubRMSE were 0.32 °C and 0.17 °C, respectively. We further revealed the mechanism of LAI on LST simulation in the Noah-MP LSM by decomposed temperature metric method. With the adoption of the new satellite LAI-based scheme, the absorbed radiation (SW), sensible heat flux (H), latent heat flux (LE), and ground heat flux (G) simulated by the Noah-MP LSM tend to change in different directions, leading to an overall increase in the net energy absorbed by the surface, which is consistent with the LST changes. The LAI changes of dense vegetation affect the LST simulation mainly through the competition of the absorbed shortwave radiation and turbulent heat fluxes. Due to the influence of exposed soil, the LAI changes of sparse vegetation mainly affect the LST simulation mainly through the interaction of the absorbed shortwave radiation and ground heat flux. Our study highlighted that the application of satellite LAI data has the potential to improve simulations of multiple land surface processes in the Noah-MP LSM.