Optimization of key land surface albedo parameter reduces wet bias of climate modeling for the Tibetan Plateau
image:
Fig. 1. (a)The optimized “soil color” map for the Tibetan Plateau, (b) Comparison of the interannual variations of the surface albedo for remote sensing product, WRF-CTL, and WRF-OPT simulations over the Tibetan Plateau. WRF-CTL represents the control simulation with the default soil color in the WRF model, WRF-OPT is the simulation with the optimized soil color parameter
view moreCredit: ©Science China Press
As the world’s highest plateau, the Tibetan Plateau receives intensified summer solar radiation due to its high altitude and low air density. This creates low-pressure systems and cyclonic circulation in the lower atmosphere, driving monsoon water vapor into the plateau’s interior from its eastern and southern boundaries. Accurate representation of land surface heat source is crucial for reliable climate predictions.
Surface albedo, a key factor in land surface energy balance, is typically parameterized in climate models using a “soil color” parameter, where darker soils (higher values) correspond to lower albedo, and vice versa. However, uncalibrated global “soil color” parameter in the climate models have led to underestimated summer surface albedo on the Tibetan Plateau, contributing to overestimated precipitation.
Researchers from the National Institute of Natural Hazards (Ministry of Emergency Management of China), Southwest University, Tsinghua University, and Institute of Tibetan Plateau Research (Chinese Academy of Sciences), introduced an optimized “soil color” map into the Weather Research and Forecasting model, enhancing land surface albedo and temperature simulations in the plateau over a 10-year period from 2011 to 2020. This weakened land-atmosphere interactions, reducing sensible heat and evapotranspiration. Such adjustments increased lower tropospheric geopotential height, suppressing moisture flux convergence, and limiting water vapor flow into the plateau. Consequently, the wet bias in precipitation estimates dropped from 52% to 36% as compared to the IMERG precipitation product, with improved accuracy over 66% of rain gauge stations. The study further revealed that the reduction in moisture flux convergence contributes about 77% to the summer precipitation decrease.
The study demonstrates that underestimating surface albedo partially contributes to the wet bias in precipitation simulation and highlights the potential of optimizing land surface parameters using cost-effective satellite remote sensing to improve climate modeling.
See the article:
Ma X, Zhao L, Sun J, Chen J, Wang Y, Zhou J, Liu J, Lu H, Yang K. 2025. Optimization of key land surface albedo parameter reduces wet bias of climate modeling for the Tibetan Plateau. Science China Earth Sciences, 68(8): 2653-2662, https://doi.org/10.1007/s11430-025-1635-0
Fig. 2. Spatial distribution of the differences in (a) sensible and (b) latent heat fluxes between two experiments (WRF-OPT minus WRF-CTL; W m-2) over 10 summer seasons from 2011 to 2020. Dots denote areas where the difference are statistically significant (t-test with a 90% confidence interval); Spatial distribution of the difference in (c) geopotential height (gpm) at 500-hPa (shaded) and wind (m s-1; vectors), and (d) vertical integral of water vapor flux (kg m-1 s-1)
Fig. 3. Spatial distribution of the difference in (a) precipitation and (b) moisture flux convergence between two experiments (WRF-OPT minus WRF-CTL; mm d-1) over 10 summer seasons from 2011 to 2020. Dots denote areas where the difference are statistically significant (t-test with a 90% confidence interval). (c) Comparison of the interannual variations of the summer precipitation for the IMERG precipitation product, WRF-CTL, and WRF-OPT simulations over the Tibetan Plateau hinterland (30°N–35°N, 90°E–95°E).
Credit
©Science China Press
Journal
Science China Earth Sciences
Method of Research
Computational simulation/modeling
