Subseasonal Forecasts of TC activity in the Western North Pacific in the Navy-ESPC

Skillful subseasonal forecasting of tropical cyclones (TCs) has long been a lofty goal for meteorologists as TCs are small, convective features that make them especially hard to predict in this time frame that is already well-known for being hard to predict.

In this study we evaluate the utility of Navy’s Earth System Prediction Capability (Navy-ESPC) forecasts of TC activity in the Western North Pacific basin in the subseasonal timeframe. We use TempestExtremes to track forecasted TCs. We find that, while the Navy-ESPC overproduces TCs and there are some errors associated with initialization, the model generally captures the spatial climatology of TCs very well and can even simulate relatively intense TCs. Navy-ESPC forecasts of TCs were added to a statistical-dynamical model that also incorporated MJO and ENSO parameters to forecast basin-wide TC activity. There were large improvements to the statistical-dynamical model when forecasted accumulated cyclone energy (ACE) was incorporated into the model at two-week lead times. There were also slight improvements in skill at weeks three and four. An analysis of forecasted TC position also suggests that the model may have some skill in forecasting TC track out to week three. These results are promising and suggest that regional and skillful TC forecasting in the subseasonal range is possible.