Skip to main content
U.S. flag

An official website of the United States government

Stationary Wave Mediated Remote Control of Amazonian Rainfall by Andean Latent Heating: Implications for Chronic Zonal Dipole Rainfall Biases in CMIP5 Simulations over South America

Presentation Date
Friday, December 14, 2018 at 1:40pm
Location
Walter E Washington Convention Center Hall A-C (Poster Hall)
Authors

Author

Abstract

Accurate simulation of regional precipitation patterns in climate models is especially important over tropical rainforests where radiation and soil moisture impact photosynthesis relevant to global biogeochemistry and terrestrial carbon exchange. But CMIP5 models still suffer from a Wet Andes–Dry Amazon (WADA) bias in the vicinity of the world’s largest rainforest. In complement to other views that have attempted to understand this dipole structure by scrutinizing sea surface temperature over the Atlantic and the Pacific, or by focusing on local atmospheric and terrestrial processes over the Amazon forest, we try to understand the WADA bias from an Andean perspective, by examining the Andean convection – regional circulation feedbacks. Using a wide (180-member) ensemble approach, we conduct sensitivity experiments in the Community Earth System Model (CESM) v1.1 that shed light on the associated mechanisms by revealing an unsteady wave mediated response to artificially muting condensational heating over Andean orography. Comparing to the control group, the experiment group exhibits rainfall reduction over the Andes and eastward propagating rainfall enhancement over the Amazon via a teleconnection bearing signatures of equatorial wave activity with implications for vapor convergence over the forested region. This manifests as a weakening of the Bolivian High–Nordeste Low system, which is known in observations to be conducive to suppression of rainfall over the Mantaro basin and enhancement of convection over Northeastern Brazil. These findings help reveal a previously underappreciated Andean control pathway on Amazonian rainfall biases and help advance understanding of the root causes of the WADA bias feature in the CMIP multi-model mean.

    Category
    Atmospheric Sciences
    Funding Program Area(s)