Comparison of Ensemble Mean and Deterministic Forecasts for Long-Range Airlift Fuel Planning
Abstract
Implementing an ensemble mean forecast to aid in fuel planning for long-range strategic airlift has the potential to improve upon the deterministic forecasts currently used. More accurate wind forecasts could aid in a significant reduction in annual fuel costs for the DoD. This study focuses on the wind forecasts from the Global Forecast System (GFS) deterministic model and the ensemble mean wind forecasts from the Global Ensemble Forecast System (GEFS), Global Ensemble Prediction System (GEPS), and Mesoscale Ensemble Prediction System (MEPS) over a 60-day period from 19 Sep through 17 Nov 2013. The fuel burn and total spread was computed for five great circle flight routes and five aircraft using each model's wind data. The deterministic fuel burn error was then compared to the ensemble mean fuel burn error. For each of the flights investigated at cruise levels 500mb (FL180) and 250mb (FL340) the amount of reserve fuel required to account for the uncertainty in the wind forecasts was typically lower for the ensemble mean forecasts during forecast hours 12 to 48.
Document Details
- Document Type
- Technical Report
- Publication Date
- Mar 27, 2014
- Accession Number
- ADA601404
Entities
People
- Haley A. Homan
Organizations
- Air Force Institute of Technology