Date of Award

Fall 1-1-2025

Document Type

Dissertation

Degree Name

Doctor of Philosophy (PhD)

Department

Geology and Geophysics

First Advisor

Fedorov, Alexey

Abstract

The double-ITCZ bias is one of the most persistent and long-standing model deficiencies in General Circulation Models (GCMs), affecting simulations of tropical precipitation patterns and future climate projections. This bias manifests as an unrealistic secondary Intertropical Convergence Zone (ITCZ) forming south of the equator, particularly in coupled ocean-atmosphere models. Since the ITCZ plays a crucial role in shaping global precipitation patterns, addressing this bias is essential for improving climate model reliability. CMIP6 projections indicate that the double-ITCZ feature may become more pronounced under global warming, raising the question of whether this enhancement represents a true climate change response or a model artifact. This dissertation investigates the mechanisms responsible for the double-ITCZ bias, examines its behavior across models, and evaluates its projected strengthening in future simulations. First, we analyze the ocean-atmosphere interactions that shape the ITCZ by identifying a positive feedback mechanism involving the North Equatorial Countercurrent (NECC) and ITCZ. This WASP (Wind curl - Advection - SST - Precipitation) feedback suggests that the NECC actively modulates the ITCZ rather than passively responding to wind forcing. Using CESM2, we demonstrate that a stronger NECC enhances the ITCZ, implying that models that underestimate the NECC may contribute to the double-ITCZ bias. This provides a novel explanation for why climate models struggle to simulate the ITCZ correctly. To test the robustness of this relationship across different models, we extend our analysis to 13 CMIP6 models and examine how the ITCZ and NECC are linked across various climate simulations by comparing historical (1850-2014) and high-emission scenario (SSP585). Under global warming, we find that the NECC weakens, leading to an exaggerated double-ITCZ feature in CMIP6 projections. These results raise concerns that the projected changes in tropical precipitation patterns may be influenced by preexisting biases rather than genuine climate responses. In the meantime, a faster Equatorial Undercurrent (EUC) is found driven by reduced friction despite the weaker trade winds. To determine whether the enhancement of the double-ITCZ in a warming climate is a real climate signal or a model artifact, we conduct a flux-adjusted CESM2 experiment. By introducing corrections to the heat flux in the ocean component, we significantly reduce the double-ITCZ bias in the preindustrial control simulation. When we impose a CO2 doubling experiment, our bias-corrected model produces a much smaller reduction in interhemispheric precipitation asymmetry, with only half the increase in precipitation in the southeastern Pacific compared to the uncorrected model. This suggests that current GCMs may overestimate the future enhancement of the double-ITCZ due to preexisting biases. By systematically investigating the mechanisms behind the double-ITCZ bias and testing correction methods, this dissertation advances our understanding of climate model biases and enhances the accuracy of future climate projections.

Download

Share

COinS