February 1, 2023
Research Highlight

Understanding the Origin of Tropical Precipitation Bias in Climate Models

Identifying links between large-scale and local-scale model biases in tropical precipitation

Satellite image of clouds above the ocean and land

In contrast to observations, climate models simulate excessive precipitation and pronounced intertropical convergence zones in both hemispheres. This bias has persisted for decades despite generations of model development.

The Science                                 

Tropical precipitation in climate models presents significant biases in both the large-scale pattern (i.e., a double intertropical convergence zone [ITCZ], bias) and local-scale characteristics (i.e., drizzling bias with too frequent drizzle/convection). By untangling the coupled system and analyzing the biases in precipitation, clouds, and radiation, researchers find that local-scale drizzling bias in atmospheric models can lead to large-scale double-ITCZ bias in coupled models. This happens because the drizzling bias induces convective-regime-dependent biases in precipitation and the cloud radiative effect.

The Impact

Despite generations of development, the double-ITCZ bias has persisted in climate models. This impairs the simulation of tropical climate and variability as well as representations of tropical-extratropical interactions. A major obstacle to fixing the double-ITCZ bias has been understanding its origin. This work shows that local-scale drizzling bias can lead to large-scale double-ITCZ bias through both direct and indirect pathways. The results indicate that fixing the local-scale drizzling bias is critical for correcting the double-ITCZ bias. The study also connects local- and large-scale precipitation biases, which are generally studied separately in the literature.


By analyzing biases in precipitation, clouds, and radiation in both atmospheric and coupled models, this study establishes a physical linkage from the local-scale too frequent drizzling/convection bias to the large-scale pattern bias of double-ITCZ. Directly, the local-scale precipitation bias contributes to the hemispherically symmetric wet bias in atmospheric models without sea surface temperature (SST) biases by increasing light precipitation in regime of moderate convection and suppressing heavy precipitation in the strong convection regime. Indirectly, local-scale characteristic bias drives the hemispherically asymmetric precipitation bias in coupled simulations through influences on the cloud radiative effect (CRE) and consequently SST. Specifically, the local-scale drizzling bias induces a positive CRE bias in the stratocumulus (convective) regime. As the stratocumulus region is climatologically more pronounced in the southern tropics, the CRE bias is deemed to be hemispherically asymmetric, driving warm and wet biases in the southern tropics when coupled to the ocean. Together, the direct and indirect effects lead to the full double-ITCZ bias in coupled models. The results thus connect the two outstanding model biases at local- and large-scales. They suggest that correcting the local-scale characteristic bias of too frequent drizzling/convection as a critical step for fixing the large-scale double-ITCZ bias. The drizzling and double-ITCZ biases are not alleviated in models with higher resolution, implying that either large-eddy simulations or fundamental improvements in small-scale representations are needed.

PNNL Contact

L. Ruby Leung, Pacific Northwest National Laboratory, ruby.leung@pnnl.gov


This research was supported by the Department of Energy, Office of Science, as part of research in the Regional and Global Model Analysis program area, Earth and Environmental System Modeling Program.

Published: February 1, 2023

W. Zhou, L. R. Leung, and J. Lu. 2022. “Linking large-scale double-ITCZ bias to local-scale drizzling bias in climate models,” Journal of Climate, 35(24), 4365–4379. [DOI: 10.1175/JCLI-D-22-0336.1]