Predictability of Atlantic Tropical Cyclone Activity on Sub-seasonal Time Scales

The prediction of tropical cyclone (TC) activity is currently confined to weather forecasts within five days or activity over an entire season. However, we expect that there is predictability and utility in forecasts on sub-seasonal time scales. This study focuses on lead times between 2-4 weeks. The ensembles of the Community Climate System Model (CCSM4) were used to assess various factors affecting TC activity in the Atlantic including vertical wind shear, humidity and sea level pressure. It was found that the variability of Atlantic TC activity in the CCSM4 model corresponded with the state of the El Nino Southern Oscillation (ENSO). Additionally, other variables in CCSM4 hindcasts such as area averaged vertical wind shear, vorticity and precipitation, were also compared to actual accumulated cyclone energy (ACE) for Septembers in select years. Favorable conditions in the model were found to be positively correlated with ACE. This approach has been extended to forecasts in the Sub-seasonal Experiment (SUBX) project in 2017 to investigate the ability of the predictive models to capture TC activity and relevant environmental parameters in real time.