Analysis of the variability of Atlantic tropical cyclone forecast errors

For each active tropical cyclone (TC) in the Atlantic basin, the National Hurricane Center (NHC) issues public products that rely upon their track and intensity forecasts. The TC track forecast cone and surface wind speed probabilities graphics, among other products, additionally take into account recent error characteristics of past official NHC TC forecasts to convey historical uncertainty in the forecast. However, such error information is derived from the entire basin-wide set of Atlantic TCs using only official NHC forecasts. Due to differences in the predictability of TCs by location (Aberson 1998) and environmental characteristics (Kehoe et al. 2007), the historical confidence of a particular forecast may be better compared to specific analog storms. In this study, TCs are divided into subsets with similar characteristics, such as current and forecasted intensity, location, and time of year. Further, similar analyzes are performed on model data to augment understanding in the range of forecast error. By relating a particular forecast to prior forecasts in similar areas and intensities (instead of all prior storms), the uncertainty of a TC's likely track and intensity evolution by the errors of forecast analogs is better constrained and can be used to more accurately assess the TC's potential impact risk.