Incorporating Tropical Cyclogenesis Guidance Tools (Probabilistic and Spacial) for Verification at the National Hurricane Center

It has now been more than a decade since the National Hurricane Center (NHC) started providing probabilistic guidance of tropical cyclone (TC) genesis events in the Tropical Weather Outlook (TWO), regularly issued four times a day during the hurricane season. In each outlook, areas of interest are assigned probabilities in 10 percent increments and expressed in categories from low (0-30%), medium (40-60%), and high (70-100%). Improving guidance tools used for forecasting genesis timing and location has long been highlighted as a critical objective in the hurricane forecast improvement project (HFIP). However, the platform used to generate TC track and intensity forecasts, the Automated Tropical Cyclone Forecasting (ATCF) system, currently does not incorporate or save similar guidance tools available online for probabilistic TC genesis forecasting. The lack of genesis guidance aids in the ATCF currently precludes a more detailed investigation of TC genesis verification, especially between current human forecaster and model output, like commonly done for existing TC track and intensity at NHC.

This study aims to leverage recent work that provides probabilistic genesis guidance for areas of interest using numerical model output at the same cadence as the regularly issued TWOs. One such method is called the TC logistic genesis guidance (TCLOGG) which computes genesis probabilities of given features using logistic regression equations for a slew of publicly available global models (GFS, ECMWF, CMC, UKMET, NAVGEM). These models can also be combined to provide consensus based probabilistic guidance, not unlike how existing TC track and intensity consensus guidance aids are created. Another potential guidance addition is the development of an ensemble global TC genesis index (EGGI) which, in addition to probabilistic information, also composes spatial regions on TWO timescales where genesis is anticipated. While NHC forecasters already use TCLOGG, EGGI, raw ensemble TC probabilities, and other publicly available web-based tools to compose human generated TC genesis probabilities, having these products available in ATCF would assist the forecast process in addition to better verifying model output against human based probabilities for genesis areas.

This presentation will cover progress made on incorporating a set of model genesis guidance tools that can be incorporated in existing “e-deck” ATCF files, presenting examples of how the guidance can be used to enhance NHC forecaster intuition, while also providing a backbone to begin more systematic verification of genesis probabilities at NHC. Examples of how genesis guidance performed for 2023 both spatially and for individual TC events will also be presented, with the goal to make this a regularly verified dataset, alongside TC track and intensity forecasting in the annual NHC verification reports.