Clouds exert a significant impact on global temperatures and climate change. Cloud-radiative feedback (CRF) is one of the major sources of climate change uncertainty. Understanding CRF is therefore crucial for accurate climate projections. Biases like the double-ITCZ problem in Global Climate Models (GCMs) hampers precise climate projections. Here, we employ a bias-corrected downscaling method to constrain the cloud feedback uncertainties in the tropical and sub-tropical Atlantic region. We use high-resolution regional climate model (RCM) simulations with resolved deep convection, driven by debiased driving fields from three different global climate models (GCMs). Our simulations project a narrower range of the CRF over the tropical and sub-tropical Atlantic (-0.7 to 0.6 $Wm^{-2}K^{-1}$) compared with the driving GCMs (-0.9 to 1.0 $Wm^{-2}K^{-1}$). Furthermore, using linear regression between equilibrium climate sensitivity (ECS) and the CRF over the studied domain, a slightly narrower range of ECS (3.05 - 4.87 K) is inferred after the downscaling, compared to the GCMs’ result (2.98 - 5.15 K). The narrowed range is mainly due to a reduced range of stratocumulus feedback.   Our study highlights the potential of bias-corrected downscaling in constraining the uncertainty of simulations and estimates of cloud feedback and equilibrium climate sensitivity. The results advocate for further simulations with additional RCMs and domains for a more comprehensive analysis.

The long-existing double-ITCZ problem in Global Climate Models (GCMs) hampers accurate climate simulation. Using a regional climate model (RCM) over the tropical and sub-tropical Atlantic with a horizontal resolution of 12 km and explicit convection, we develop a bias-correction downscaling methodology to remove GCM biases. The methodology is adapted from the pseudo-global warming (PGW) approach, typically used to exert the climate-change signal to a reanalysis-driven RCM simulation. We show that the double ITCZ problem persists with conventional dynamical downscaling, but with our bias-corrected downscaling, the double ITCZ problem is removed. Detailed analysis attributes the main cause of the double ITCZ problem of the selected GCM to the sea surface temperature (SST) bias. Compared to the GCM’s AMIP simulations, RCMs with higher resolution allow explicit deep convection and enable a better simulation of tropical convection and clouds. The developed methodology is promising for constraining climate sensitivity by removing double-ITCZ biases.

Non-hydrostatic km-scale weather and climate models show significant improvements in simulating clouds, especially convective ones. However, even km-scale models need to parameterize some physical processes and are thus subject to the corresponding uncertainty of parameters. Systematic calibration has the advantage of improving model performance with transparency and reproducibility, thus benefiting model intercomparison projects, process studies, and climate-change scenario simulations. In this paper, the regional atmospheric climate model COSMO v6 is systematically calibrated over the Tropical South Atlantic. First, the parameters’ sensitivities are evaluated with respect to a set of validation fields. Five of the most sensitive parameters are chosen for calibration. The objective calibration then closely follows a methodology extensively used for regional climate simulations. This includes simulations considering the interaction of all pairs of parameters, and the exploitation of a quadratic-form metamodel to emulate the simulations. In the current set-up with 5 parameters, 51 simulations are required to build the metamodel. The model is calibrated for the year 2016 and validated in two different years using slightly different model setups (domain and resolution). Both years demonstrate significant improvements, in particular for outgoing shortwave radiation, with reductions of the bias by a factor of 3 to 4. The results thus show that parameter calibration is a useful and efficient tool for model improvement. Calibrating over a larger domain might help improve the overall performance, but could potentially also lead to compromises among different regions and variables, and require more computational resources.