Abstract
In this study, several metrics and diagnostics are proposed and implemented to systematically explore and diagnose climate model biases in short-range hindcasts and quantify how fast hindcast biases approach to climate biases with an emphasis on tropical precipitation and associated moist processes. A series of 6-day hindcasts with NCAR and the U.S. Department of Energy Community Atmosphere Model, version 4 (CAM4) and version 5 (CAM5), were performed and initialized with ECMWF operational analysis every day at 0000 UTC during the Year of Tropical Convection (YOTC). An Atmospheric Model Intercomparison Project (AMIP) type of ensemble climate simulations was also conducted for the same period. The analyses indicate that initial drifts in precipitation and associated moisture processes (“fast processes”) can be identified in the hindcasts, and the biases share great resemblance to those in the climate runs. Comparing to Tropical Rainfall Measuring Mission (TRMM) observations, model hindcasts produce too high a probability of low- to intermediate-intensity precipitation at daily time scales during northern summers, which is consistent with too frequently triggered convection by its deep convection scheme. For intense precipitation events (>25 mm day−1), however, the model produces a much lower probability partially because the model requires a much higher column relative humidity than observations to produce similar precipitation intensity as indicated by the proposed diagnostics. Regional analysis on precipitation bias in the hindcasts is also performed for two selected locations where most contemporary climate models show the same sign of bias. Based on moist static energy diagnostics, the results suggest that the biases in the moisture and temperature fields near the surface and in the lower and middle troposphere are primarily responsible for precipitation biases. These analyses demonstrate the usefulness of these metrics and diagnostics to diagnose climate model biases.
Critical Southern Ocean climate model biases traced to atmospheric model cloud errors
