Evaluation of surface albedo over the Tibetan Plateau simulated by CMIP5 models using in‐situ measurements and MODIS

Abstract

Surface albedo plays a key role in the energy and water cycles, and reasonable parameterizations of surface albedo will be greatly helpful to improve the simulation of radiation partition in climate models. In-situ measurements of albedo from five sites over the Tibetan Plateau (TP) and MODIS albedo product are used to evaluate monthly, annual, and seasonal variations of the surface albedo simulated by 24 Global Climate Models (GCMs) archived by the Coupled Model Intercomparison Project Phase 5 (CMIP5). Potential factors contributing to the bias of simulated albedo were investigated. The results show that the monthly albedo of 24 GCMs varied across models, and the difference among the models was smaller in the June-July-August period than the December-January-February period. The ensemble mean albedo of the 24 GCMs was more consistent with the in-situ measurements than the individual model. The albedo calculated from the BNU-ESM, GFDL-CM3, INM-CM4, MIROC4h, MIROC-ESM, and MIROC-ESM-CHEM models coincided with the observations from June to September, increasing rapidly to above 0.4 from November. However, the annual cycle of surface albedo was insignificant when simulated by some models, such as CanESM2, CSIRO-Mk3.6.0, CMCC-CMS, IPSL-CM5A-MR, and MPI-ESM-MR. Additionally, the surface albedo was overestimated by CMIP5 multi-model ensemble mean-with a smaller amplitude of daily albedo-compared with the values estimated from in-situ observations and MCD43A3. In all 24 models, snow albedo parameterization schemes were found to variably fit the attenuation of snow albedo over time. The cold temperature, relatively more precipitation, fresh snow density, and albedo in these models led to large biases in snow depth and ablation.The surface albedo tends to increase without the influence of blowing snow and withered grass on the model's albedo.

Global Daily Actual and Snow‐Free Blue‐Sky Land Surface Albedo Climatology From 20‐Year MODIS Products

“…The CMIP6 overestimates the albedo in the Tibetan Plateau and boreal regions, which is mainly due to the overestimation in winter. Such albedo overestimation in the Tibetan Plateau also occurred in CMIP5 (An et al, 2021). The Weather Research and Forecasting (WRF) model yields a similar overestimation in the Tibetan Plateau, resulting in a cold bias in the surface air temperature simulation, which is mainly due to the simulated precipitation biases and overparameterization of the snow albedo (Meng et al, 2018).…”