Sahel Rainfall Variability as simulated by the CAM4 Model and its Associated Atmospheric Dynamics

LDEO Publication: 
Yes
Publication Type  Conference Paper
Year of Publication  2013
Authors  Pomposi, C.A; Kushnir, Y.; Giannini, A.
Conference Name  American Geophysical Union Fall Meeting
Abstract  

The West African monsoon supplies the Sahel region of Africa with the majority of its annual rainfall, which can range from ~300mm to 700mm in dry or wet years, respectively. It has been shown in numerous studies that monsoonal rains in this region (peaking in July-September), are largely influenced by global ocean temperatures (Giannini et al. 2003, and others). However, models display inconsistent ability to simulation of 20th century Sahel rainfall variability and trends even if given observed ocean conditions. It has been previously noted that differences among models in simulating such trends may be due to deviations in the model’s representation of a particular ocean basin’s relationship with Sahel rainfall (Caminade and Terray 2010). It is therefore of worth to continue with sensitivity studies in which observed SSTs force atmospheric models and the resulting precipitation and dynamical fields are studied for consistency both across models and checked for validity with observations. We use this insight as motivation for the present study. We seek to understand the relationship between the global oceans and summertime Sahel precipitation in the National Center for Atmospheric Research’s (NCAR) Community Atmosphere Model, version 4 (CAM4). The atmospheric model is forced with observed SSTs from 1901-2002 and the resulting precipitation field is studied on multi-decadal timescales and compared to observations. The 16-member ensemble runs' statistics are compared to observations to validate the model. More detailed statistical techniques such as Maximum Covariance Analysis (MCA) are used to understand whether the key modes of variability between global SSTs and Sahel precipitation are similar in observations and the SST-driven model. It is found that on multi-decadal timescales, the ensemble-averaged precipitation maintains the overall characteristics in space and time of the 20th century observed precipitation fields. Furthermore, modes of covariability between the Atlantic and Indian Ocean basins with Sahel rainfall are represented, as expected based on observations. Further analysis is performed to better understand the dynamical mechanisms by which the global oceans influence atmospheric circulation patterns that account for Sahel rainfall variability during the 20th century. Specifically, we seek to understand the differences between dry and wet years in the model and determine the major discrepancies in the moisture budgets of dry versus wet years. We thus focus on the precipitation, evaporation, regional wind patterns, and moisture fluxes for dry and wet periods.

URL  http://abstractsearch.agu.org/meetings/2013/FM/sections/A/sessions/A11G/abstracts/A11G-0136.html