Observational Evidence for Desert Amplification Using Multiple Satellite Datasets

Nan Wei,Geng Xia,Yongjiu Dai,Wenjian Hua,Liming Zhou

doi:10.1038/s41598-017-02064-w

Abstract

Desert amplification identified in recent studies has large uncertainties due to data paucity over remote deserts. Here we present observational evidence using multiple satellite-derived datasets that desert amplification is a real large-scale pattern of warming mode in near surface and low-tropospheric temperatures. Trend analyses of three long-term temperature products consistently confirm that near-surface warming is generally strongest over the driest climate regions and this spatial pattern of warming maximizes near the surface, gradually decays with height, and disappears in the upper troposphere. Short-term anomaly analyses show a strong spatial and temporal coupling of changes in temperatures, water vapor and downward longwave radiation (DLR), indicating that the large increase in DLR drives primarily near surface warming and is tightly associated with increasing water vapor over deserts. Atmospheric soundings of temperature and water vapor anomalies support the results of the long-term temperature trend analysis and suggest that desert amplification is due to comparable warming and moistening effects of the troposphere. Likely, desert amplification results from the strongest water vapor feedbacks near the surface over the driest deserts, where the air is very sensitive to changes in water vapor and thus efficient in enhancing the longwave greenhouse effect in a warming climate.

Highlights

Recent studies have identified another warming pattern that land surface air temperature (LSAT) in mid- and low- latitudes is amplified over deserts, referred to as desert amplification
Zhou et al.[10] and Zhou[13] attributed desert amplification mostly to the enhanced downward longwave radiation (DLR) at the surface associated with stronger water vapor feedbacks over drier ecoregions in a warming climate
Desert amplification is strongest near the surface, gradually decays with height, and mostly disappears in the upper troposphere (UT)

Summary

Satellite Datasets

Desert amplification identified in recent studies has large uncertainties due to data paucity over remote deserts. Inference is analogous with the long-term trend analyses from observational and reanalysis data[9,10,11,12] and from historical and projected simulations[13] These results confirm that desert amplification could be driven primarily by the increases in DLR, which are closely linked to the strongest water vapor feedbacks over the driest climate zones as discussed next. In a warming climate the largest overall warming is expected to be seen in the tropical UT especially over deep convection regions where the warming profile should follow moist adiabat to be consistent with the strongest positive water vapor feedback and a corresponding negative lapse rate feedback as reflected in theoretical studies and climate simulations[1, 13, 16, 38,39,40] This feature is supported by our short-term temperature anomaly analysis. These two types of uncertainties in MSU-derived satellite data likely introduce spurious cooling signals to tropospheric temperature trends and weaken the warming effects throughout most of the troposphere

Conclusions

Data and Methods

Findings

Additional Information