Austral summer precipitation biases over tropical South America in five CMIP5 earth system models

• Published in: International Journal of Climatology
• Main Authors: Vasconcellos, Fernanda Cerqueira, Deng, Yi, Zhang, Henian, Martins, Guilherme
• Format: Article in Journal/Newspaper
• Language: English
• Published: Wiley 2020
• Subjects: Austral; North Atlantic
• Online Access: http://dx.doi.org/10.1002/joc.6595; https://api.wiley.com/onlinelibrary/tdm/v1/articles/10.1002%2Fjoc.6595; https://onlinelibrary.wiley.com/doi/pdf/10.1002/joc.6595; https://onlinelibrary.wiley.com/doi/full-xml/10.1002/joc.6595; https://rmets.onlinelibrary.wiley.com/doi/pdf/10.1002/joc.6595

Abstract This study examined the historical (1980–2005) austral summer precipitation in Tropical South America (SA) simulated by five Earth System Models (ESM) from the Coupled Model Intercomparison Project Phase 5 (CMIP5). All simulations show a band of maximum precipitation eastward of the observed one and without typical NW‐SE orientation. This displacement suggests models have trouble in reproducing the South Atlantic Convergence Zone (SACZ). Due to this and additional model problems in simulating the Atlantic Intertropical Convergence Zone (ITCZ), all models exhibit negative precipitation biases at the extreme northern SA, including Guianas, Suriname, and north of Brazil and positive precipitation biases at part of northeastern Brazil. For extreme northern SA, models, in general, underestimate intense precipitation and overestimate weak rainfall. Analysis of the moisture flux divergence over the northern coast of SA in models suggested that the precipitation bias could primarily stem from model misrepresentations of moisture availability for convection. Further analyses indicate that the moisture flux biases are, in turn, tied to a negative sea surface temperature (SST) bias in the tropical North Atlantic, inducing stronger northeasterly trade winds. Thus, more intense moisture flux goes to the inner continent. Consequently, an anomalous divergence of moisture flux and less precipitation occur near the coast. Despite some differences in energy budget and cloudiness, models results for wind, precipitation, SST, and latent heat flux suggest problems at WES feedback. In the GFDL‐ESM and MIROC‐ESM models, the negative SST bias was also partly associated with a lower incident shortwave (SW) radiation over the tropical North Atlantic. This SW bias was tied to a positive bias of cloud cover over tropical North Atlantic, at least in the GFDL‐ESM.