Application and impacts of the GlobeLand30 land cover dataset on the Beijing Climate Center Climate Model

X Shi,S Nie,W Ju,Le Yu

DOI: https://doi.org/10.1088/1755-1315/34/1/012032

2016-01-01

Abstract:Land cover (LC) is a necessary and important input variable of the land surface and climate model, and has significant impacts on climate and climate changes. In this paper, the new higher-resolution global LC dataset, GlobeLand30, was employed in the Beijing Climate Center Climate System Model (BCC_CSM) to investigate LC impacts on the land surface and climate via simulation experiments. The strategy for connecting the new LC dataset and model was to merge the GlobeLand30 data with other satellite remote sensing datasets to enlarge the plant function types (PFT) fitted for the BCC_CSM. The area-weighted up-scaling approach was used to aggregate the 30m-resolution GlobeLand30 data onto the coarser model grids and derive PFT as well as percentage information. The LC datasets of GlobeLand30 and the original BCC_CSM had generally consistent spatial features but with significant differences. Numerical simulations with these two LC datasets were conducted and compared to present the effects of the new GlobeLand30 data on the climate. Results show that with the new LC data products, several model biases between simulations and observations in the BCC climate model with original LC datasets were effectively reduced, including the positive bias of precipitation in the mid-high latitude of the northern hemisphere and the negative bias in the Amazon, as well as the negative bias of air temperature in part of the southern hemisphere. Therefore, the GlobeLand30 data are suitable for use in the BCC_CSM component models and can improve the performance of climate simulations.