Volume 40, Issue 7 p. 3360-3379
RESEARCH ARTICLE

Multimodel ensemble projection of meteorological drought scenarios and connection with climate based on spectral analysis

Yog Aryal

Corresponding Author

Yog Aryal

Department of Civil and Architectural Engineering, University of Wyoming, Laramie, Wyoming

Correspondence

Yog Aryal, Department of Civil and Architectural Engineering, University of Wyoming, Laramie, WY 82071.

Email: [email protected]

Search for more papers by this author
Jianting Zhu

Jianting Zhu

Department of Civil and Architectural Engineering, University of Wyoming, Laramie, Wyoming

Search for more papers by this author
First published: 11 November 2019
Citations: 14

Abstract

Potential change in the hydrological cycles may lead to changes in extreme events such as drought due to the changes in intensity, frequency, and seasonality of precipitation and evaporative demand. In this study, projected drought scenarios and associated uncertainties in the late 21st century over the Continental United States are analysed based on seven regional climate models (RCMs) contributing to the Coordinated Regional Climate Downscaling Experiment. Meteorological drought scenarios are analysed based on standardized precipitation evapotranspiration index (SPEI). Here, we propose a new multimodel ensemble approach based on the similarity of spectral power of different frequency components of the SPEI time series. The approach combines output from different RCMs based on their ability to produce the observed power spectra as well as convergence towards the average spectra from all participating RCMs in the future climate. Furthermore, to understand the SPEI variability associated with climate teleconnection, continuous wavelet transform-based spectral coherency between the SPEI and two climate indices: the El Niño–Southern Oscillation (ENSO) and the Pacific Decadal Oscillation (PDO) is analysed. Results show that among-model uncertainty is the dominant source of uncertainties in the projected drought scenarios with contributions as large as 97% to the total uncertainty. Fifteen to 20% more frequent droughts are projected in future climate due to a decrease in mean SPEI as well as an increase in the variability of SPEI. Observed SPEI variability is strongly associated with ENSO variability while PDO modulates the strength of correlation between the SPEI and the ENSO. Spectral analysis of future SPEI shows that the increase in the SPEI variability is due to an increase in interannual variability as well as an enhanced ENSO teleconnection over the study area. The results show the increased role of climatic variability and hence the enhanced predictability of drought scenarios in the future climate.