There are many unanswered questions with regards to sources of subseasonal predictability and windows of forecast opportunity. Real-time and reforecasts from operational model databases have been instrumental in facilitating research on subseasonal prediction, however, these large databases have limitations, as the underlying models do not easily allow researchers to run their own experiments to elucidate sources of predictability and test hypotheses. We describe here the newly developed subseasonal prediction systems for research purposes based on the Community Earth System Model, version 2 (CESM2) with the Community Atmosphere Model, version 6 (CAM6), and with the Whole Atmosphere Community Climate Model, version 6 (WACCM6). We describe here the details of the forecasting systems and the historical reforecasts carried out with the two prediction systems.
We show that the CESM2-based prediction systems show very similar subseasonal prediction skill to the NOAA CFSv2 model and a little lower than that of ECMWF for surface temperature, precipitation, MJO and the NAO, and hence are credible models for subseasonal prediction research. In addition, CESM2(WACCM6) system provides a novel capability to explore the predictability of the stratosphere and mesosphere and their impacts on the surface. The real-time forecasts from these models are contributing to the NOAA experimental weeks 3-4 outlook.
We describe here the newly developed CESM2(CAM6) and CESM2(WACCM6) subseasonal prediction systems for research on sources of subseasonal predictability. We show that both models have subseasonal prediction skill of surface temperature, precipitation, MJO, and the NAO comparable to the NOAA CFSv2 operational model and hence are credible models for future prediction research.