In order to pinpoint the origins of precipitation biases in the most recent version of global climate models a team of researchers funded in part by CPO’s Climate Variability & Predictability (CVP) program developed a new framework to study the moisture budgets of the newest generation of global climate model simulations (CMIP6 and ERA5).
Previous generations of CMIP climate models have a known bias towards excess precipitation over tropical oceans and dry biases over tropical land. These biases can lead to uncertainty in model outcomes of projected changes for future climate. Researchers theorize that the model precipitation biases may be related to how the models represent convection, aka the vertical transport of heat and moisture in the atmosphere. By modifying how convection is parameterized, or represented, in a model, researchers have shown they can reduce precipitation biases.
A key relationship between convection and environmental humidity can be approximated by the relationship between precipitation and precipitable water. Precipitable water is the volume of water in a column of the atmosphere if all the water in that column were precipitated as rain. In a previous research paper, some team members demonstrated how precipitable water is an important metric for understanding model biases in simulations of monsoon precipitation in the Indian Ocean. This new research, published in the Journal of Climate, focuses on the precipitation-precipitable water relationship in terms of moisture budget across the entire tropics, as shown in historical and projected climate model simulations.
The team conceptualized a theoretical measure, normalized critical precipitable water, that separates tropical precipitation into two regimes: widespread drizzle (evaporation-controlled) or heavy rainfall (water-controlled). Compared to ERA5 models, most of the CMIP6 historical simulations more frequently operate in the drizzle regime and, compared to observations, overestimate precipitation over dry (high-evaporation) oceanic regions while underestimating precipitation over large tropical land masses and wet oceanic regions near the equator. This normalized critical precipitable water framework can help improve climate models by both serving as a theoretical tool for interpreting convection effects in climate models and by providing insight into model biases in the spatial distribution of rainfall in the tropics.