National Weather Service United States Department of Commerce

Proposal Title: Extending the Tropical Cyclone Genesis Index to Global Ensemble Forecasts


Principal Investigator: Dr. Alan Brammer (CIRA/CSU)



Dr. Dustin Grogan (U Albany, SUNY)

Andrea Schumacher (CIRA/CSU)



Dr. Jason Dunion (U Miami/RSMAS/CIMAS)


Dr. Chris Slocum (NOAA/NESDIS/STAR)

National Hurricane Center

Joint Typhoon Warning Center



The prediction of tropical cyclone (TC) genesis remains a difficult challenge for forecasters at the NOAA National Hurricane Center (NHC), NOAA Central Pacific Hurricane Center (CPHC) and Joint Typhoon Warning Center (JTWC). Improving genesis forecast skill by 20% is a current HFIP priority and thus developing reliable genesis products for forecasters is imperative to meet these goals. Recent developments in genesis tools have been well received by NHC for the Atlantic and eastern North Pacific basins. This proposal will extend the capabilities of one of NHC’s operational TC genesis tools, the Tropical Cyclone Genesis Index (TCGI), in an effort to expand the current capabilities and further improve guidance to meet operational and research priorities.

This project seeks to develop a combined statistical–dynamical ensemble-based TC genesis probability forecast model, addressing HFIP priority 1 by developing techniques for TC forecast guidance, as well as developing process-based diagnostics and statistically post-processing ensemble output. The proposed project will extend the existing operational TCGI (Dunion et al. 2013, 2019) in the following ways: (i) utilize input from global ensemble forecasts; (ii) run TC genesis forecasts for pre-genesis disturbances across all basins; (iii) produce products to visualize and diagnose TC genesis forecast uncertainty. The project will run the experimental algorithm in an operational-like environment in real-time providing forecasts for the 2021 and 2022 seasons achieving readiness level 7 and minimizing the cost of operational transition in the future.

The main goal is to provide completely automated, reliable genesis forecasts for all basins. The proposed ensemble TCGI would incorporate ensemble dynamical forecast outputs, combined with the existing statistical genesis algorithm to produce an ensemble of genesis probabilities for both 0-48hr and 0-120hr forecasts. Moreover, extending the existing algorithm to run globally will give valuable real-time information for forecasters at JTWC and CPHC. The ensemble based TCGI model will increase reliability in long-term genesis forecasts and provide more dynamical information to the forecasters regarding the spread and uncertainty in the forecast prior to genesis.