Climate Systems Interaction Group

Relationships among climatological vertical moisture structure, column water vapor, and precipitation over the central Amazon in observations and CMIP5 models

Benjamin R. Lintner, David K. Adams, Kathleen A. Schiro, Alyssa M. Stansfield, Alci&#eacute;lio A. Amorim Rocha, and J. David Neelin

Geophys. Res. Lett., 44, 1981-1989, doi:10.1002/2016GL071923. doi:10.1002/2016GL071923.

Abstract. Bias and spread in Coupled Model Intercomparison Project Phase 5 simulated vertical specific humidity (q) structure are examined and related to both precipitation and column water vapor (cwv) nearManaus, Brazil, site of the recent Green Ocean Amazon campaign. Simulated seasonal mean q profiles aretypically too dry, especially at low levels and during the local dry season, consistent with previously identified surface hydroclimate biases in the Amazon. Multimodel empirical orthogonal function analysis of the models'monthly climatological q profiles indicates two significant modes of ensemble spread in moisture verticalstructure, with the leading mode peaked at low levels and the second mode in the lower free troposphere(LFT). While both modes project onto simulated cwv spread, only the first projects on precipitation,suggesting inconsistent sensitivity of simulated rainfall to LFT moisture. Relative to observations, models withhigh cwv and low-level moisture errors tend to exhibit high precipitation error. Citation:
Lintner, B. R., D. K. Adams, K. A. Schiro, A. M. Stansfield, A. A. Amorim Rocha, J. D. Neelin, 2017: Relationships among climatological vertical moisture structure, column water vapor, and precipitation over the central Amazon in observations and CMIP5 models. Geophys. Res. Lett., 44, 1981-1989, doi:10.1002/2016GL071923.

Acknowledgments. Data were obtained from the following: CMIP5, PCMDI [http://cmip-pcmdi.llnl.gov/cmip5/data_portal.html]; soundings, the University of Wyoming Department of Atmospheric Science [http://weather.uwyo.edu/upperair/sounding.html]; and CMAP, NOAA/OAR/ ESRL PSD, Boulder, Colorado, USA [http://www.esrl.noaa.gov/psd/]. The authors thank Larry D. Oolman of the University of Wyoming for assistance with the soundings, Bryan Raney and Juan Perez Arango for assistance with downloading and formatting CMIP5 output, and Kyle Itterly for useful comments. This work was supported by Department of Energy grants DESC0011069 (B.R.L.) and DE-SC0011074 (K.A.S. and J.D.N.) and National Science Foundation grants AGS-1505198 (B.R.L.) and AGS-1540518 (K.A.S. and J.D.N.). A.S.'s participation was partially facilitated through the Rutgers Aresty Undergraduate Research Assistant program.