ORCID
https://orcid.org/0009-0006-9039-5430
Date of Award
Spring 2025
Language
English
Embargo Period
4-30-2025
Document Type
Master's Thesis
Degree Name
Master of Science (MS)
College/School/Department
Department of Atmospheric and Environmental Sciences
Program
Atmospheric Science
First Advisor
Ryan Torn
Committee Members
Kristen Corbosiero
Keywords
Hurricane, Weather, Atmosphere, Numerical Weather Prediction, Ensemble
Subject Categories
Atmospheric Sciences
Abstract
Dynamically based ensemble prediction systems have gained considerable attention because they can provide a greater range of possible forecast outcomes and quantify the uncertainty in forecasts. In turn, forecasters can convey clearer messages to the public on the range of forecast scenarios and display inherent uncertainty in weather forecasts, which can be difficult to do with deterministic forecasts. Although global ensemble prediction systems have demonstrated skill in their probabilistic track predictions, the information contained within them is not always fully utilized beyond the mean forecast and standard deviation (i.e., spread). One way that these ensembles could be better employed is in the creation of the “Tropical Cyclone (TC) Forecast Track Cone”, which has been used by the National Hurricane Center (NHC) to convey the potential track forecast uncertainty. Currently, the forecast cone is based on the 66.7% of all Official Forecast errors averaged over the last five years; therefore, it does not reflect the confidence and uncertainty that NHC forecasters have in any individual forecast. Given the skill of global EPS’s, there is an opportunity to retain the current cone definition but dynamically adjust the size and shape based on the uncertainty of that individual forecast.
This study evaluates a dynamic cone that adjusts its size based on ensemble spread, creating different cone sizes for low, medium, and high spread forecasts while maintaining the 66.7% definition. An elliptical cone further refines this approach by extending the shape in the direction of greatest ensemble spread to better capture directional uncertainty. During the 2023 Atlantic hurricane season, which was characterized by large forecast errors, the dynamic and elliptical cones outperform the static cone; however, these approaches had a smaller advantage during the 2024 season, which had lower forecast errors. Randomly sampling forecasts from the 2018–2024 seasons confirmed that both dynamic and elliptical cones had the best track fall closer to the 66.7% benchmark compared to the static cone. In the eastern Pacific basin, there are smaller differences in forecast errors between all training bins, which in turn lead to mixed testing results for the dynamic and elliptical cones. Nevertheless, when taking a random sample, both dynamic and elliptical cones still consistently performed closer to the 66.7% benchmark compared to the static cone. The frequency of the low and high spread cones displayed distinct geographic and intensity patterns. In the Atlantic basin, low spread cases are more common in tropical regions with more intense TCs, while high spread cases are more likely north of 35°N and with weaker TCs. In the eastern Pacific basin, low spread cases are more frequent in the eastern portion of the basin and with stronger storms, while high spread cases predominate in the western portion and with weaker systems. These findings demonstrate that it is possible to utilize this dynamic uncertainty approach while maintaining the current NHC cone definition.
License
This work is licensed under a Creative Commons Attribution 4.0 International License.
Recommended Citation
Barletta, Michael N., "A Dynamically Adapting Forecast Cone Based on Ensemble Spread" (2025). Electronic Theses & Dissertations (2024 - present). 206.
https://scholarsarchive.library.albany.edu/etd/206
