Improve the Prediction of TC Track and Intensity by Machine Learning Based Ensemble Techniques
Abstract
Tropical cyclones (TCs) are among the most destructive environmental hazards on earth, causing hundreds of deaths and billions of dollars of damage every year. Of the $2.6 trillion in damage associated with the U.S. billion-dollar weather events in 1980-2023, $1.4 trillion was due to TCs. With strong winds, high waves, and heavy rainfall, TCs can pose significant threats to personnel, naval vessels (ships and aircraft), assets, and infrastructure. TCs can disrupt communication systems that the navy heavily relies on for command and control. Accurate TC forecasting on the path and intensity with lead time is vital to navy as it helps to plan and execute maneuvers to avoid severe conditions, helps to plan missions effectively and position resources strategically, allows for proactive decision-making and reducing the impact of these natural disasters on naval operations. Early detection and accurate tracking of tropical cyclones are essential for minimizing the impact on human lives and property. This project focuses on improving the prediction of TC#s track and intensity with long lead time (>48 hours) by machine learning (ML) based ensemble techniques. We plan to (1) analyze all the TCs in North Atlantic basin using the historical storm data, reanalysis data, and operational forecast data; (2) use statistical models to approximate historical relationships between storm behavior and storm-specific features; and (3) eventually develop ML based ensemble techniques that combine numerical weather prediction models (NWPs) with statistical models. The anticipated outcome of this research includes the pattern and categories of all the TCs (in time segments) with the aid of statistical models, and a system of combined statistical-deterministic models with ML ensemble techniques that produces prediction and forecasting on the track and intensity with comparable and/or improved accuracies when compared with the operational official forecast errors. North Carolina Agricultural and Technical State University (NC A&T) is an Historically Black College and University (HBCU) with over 80% African Americans in undergraduate student population. As America#s largest HBCU with over 13K students# enrollment in 2023 and the toprated public HBCU, NC A&T is a preeminent doctoral, research, and land-grant institution. The research team led by the PI for this project consists of two undergraduate students, two M.S. graduate students, and one Ph.D. student. All five students come from underrepresented populations, majoring in mathematics, statistics, physics, or applied science and technology. Students in this project will receive rigorous training, fascinating exploration and research experience in meteorology, computational science, data analytics, and applications in naval operations. After the training in this program, students will be ready for navy interns and be prepared to pursue their future career in the navy. This project will help the PI and her colleagues to integrate research and education by advancing discovery and understanding while at the same time promoting teaching, training, and learning.
Document Details
- Document Type
- DoD Grant Award
- Publication Date
- Jun 13, 2024
- Source ID
- N000142412351
Entities
People
- Liping Liu
Organizations
- North Carolina Agricultural and Technical State University
- Office of Naval Research
- United States Navy