Ross Dixon, PhD Defense
Department of Atmospheric and Oceanic Sciences, University of Wisconsin-Madison
On the Saharan Heat Low Bias in CMIP5 Models and its Relationship with Tropical Precipitation and Global Energy Transport Biases
Room 811 AOSS, February 6, 2017, 3:30 PM
One of the grand challenges for the climate science community is understanding the fundamental mechanisms that determine the strength and location of tropical precipitation. Representing the West African monsoon (WAM) is a major challenge in climate modeling because of the complex interaction between local and large scale mechanisms. The representation of the Saharan Heat Low (SHL), a key aspect of West African climate, is relatively unexplored in the literature. Comparison of the SHL representation in both CMIP5 simulations and reanalyses shows large biases in the strength and location of the climatological SHL. Both coupled (CMIP) and uncoupled (AMIP) ocean/atmosphere models that place the SHL farther to the north are associated with increased precipitation across the Sahel. Further, the northward SHL placement is also associated with a northward shift in the Atlantic ITCZ in coupled CMIP models, but an eastward shift in uncoupled AMIP models.
This statistical relationship informs three causal hypotheses investigated with perturbation experiments using the Community Earth System Model (CESM). We find that when the Atlantic ITCZ is forced locally, there is no coherent response in West African climate. However, when the Atlantic ITCZ is forced through altering the cross equatorial energy transport, the SHL and Sahel precipitation respond to the large scale forcing. Similarly, when the SHL strength is forced directly, there is a weak but robust increase in Sahel precipitation and a northward shift in the Atlantic ITCZ. These experiments show that global scale biases have an important impact on biases local to the West African monsoon.
We then investigate how extratropical heating anomalies are communicated to the tropics resulting in changes in both the Atlantic ITCZ and West African monsoon, by perturbing an idealized climate model. Extratropical forcing is applied to an aquaplanet which is coupled to a slab ocean and driven by seasonally varying insolation. This reveals an asymmetry in the seasonality of ITCZ shifts that depends on the direction of the forcing. SSTs are then fixed in the tropics and subtropics, reinforcing the necessity of atmosphere-ocean coupling to the propagation of extratropical forcing into the tropics. We develop a framework for exploring how this occurs by turning mechanisms that allow atmosphere-ocean coupling off in the model and exploring the transient response to the forcing.