Double Intertropical Convergence Zone Bias Research Articles

Convective Entrainment and Large-Scale Organization of Tropical Precipitation: Sensitivity of the CNRM-CM5 Hierarchy of Models

Abstract The spurious double intertropical convergence zone (ITCZ) is a systematic bias affecting state-of-the-art coupled general circulation models (GCMs). Modeling studies show that the ITCZ structure is very sensitive to moist convection parameterization and in particular, to the vertical profile of convective heating and free-tropospheric moistening. To further explore this sensitivity, the authors focus in this study on the influence of lateral entrainment in convective plumes on the simulated tropical precipitation and large-scale circulation. Sensitivity studies to the entrainment parameter were performed in a hierarchy of models (coupled ocean–atmosphere GCM, atmospheric GCM, and aquaplanet GCM), in order to mitigate the double ITCZ problem in the Centre National de Recherches Météorologiques Coupled Global Climate Model, version 5 (CNRM-CM5). The sensitivity of the ITCZ structure to lateral entrainment is robust across our hierarchy of models. In response to increased entrainment, the realistic simulations exhibit a weakening of the southern side of the double ITCZ over the southeastern Pacific Ocean and a better representation of the South Pacific convergence zone (SPCZ). However, as a result of stronger moisture–convection feedbacks, precipitation is overestimated in the center of convergence zones. The change in ITCZ configuration is associated with a more realistic representation of the large-scale vertical regimes, explained by a decreased frequency of weak-to-moderate ascending regimes and an enhanced frequency of subsidence regimes. Mechanisms at play in this circulation change are examined by analyzing the vertically integrated dry static energy budget. This energetic analysis suggests that the feedback between large-scale dynamics and deep convection is crucial in controlling the probability distribution function (PDF) of midtropospheric vertical wind. This PDF, in turn, controls the precipitation distribution and, in particular, the double ITCZ bias.

Rain Reevaporation, Boundary Layer–Convection Interactions, and Pacific Rainfall Patterns in an AGCM

Sensitivity experiments with an atmospheric general circulation model (AGCM) show that parameterized rain reevaporation has a large impact on simulated precipitation patterns in the tropical Pacific, especially on the configuration of the model’s intertropical convergence zone (ITCZ). Weak reevaporation leads to the formation of a “double ITCZ” during the northern warm season. The double ITCZ is accompanied by strong correlation between precipitation and high-frequency vertical motion in the planetary boundary layer (PBL). Strong reevaporation leads to a better overall agreement of simulated precipitation with observations. The model’s double ITCZ bias is reduced. At the same time, correlation between high-frequency (periods < 15 days) vertical motion in the PBL and precipitation is reduced. Experiments with modified physics indicate that evaporative cooling by rain near the PBL top weakens the coupling between precipitation-related heating and vertical motion in high-frequency motions. The strength of high-frequency vertical motions in the PBL was also reduced directly through the introduction of a diffusive cumulus momentum transport (DCMT) parameterization. The DCMT had a visible impact on simulated precipitation in the Tropics but did not reduce the model’s double ITCZ bias in all cases. Further analyses of mass and water vapor budgets, as well as vertical motion statistics, in the ITCZ complex, show that time-mean moisture convergence in the southern ITCZ is largely dominated by high-frequency modes, while in the northern ITCZ time-mean moisture convergence contains large contributions from slower modes. This may explain why the simulated southern ITCZ is more susceptible to parameterization changes that alter high-frequency coupling between moist heating and PBL convergence.