skip to main content
US FlagAn official website of the United States government
dot gov icon
Official websites use .gov
A .gov website belongs to an official government organization in the United States.
https lock icon
Secure .gov websites use HTTPS
A lock ( lock ) or https:// means you've safely connected to the .gov website. Share sensitive information only on official, secure websites.


This content will become publicly available on December 7, 2026

Title: Machine Learning Approaches to Identifying Tropical Waves That Develop into Hurricanes
African Easterly Waves (AEWs) are synoptic-scale atmospheric disturbances that serve as precursors to tropical cyclones (TCs) in the North Atlantic and North Africa. As climate changes, TC activities are increasingly frequent, leading to exponentially growing socio-economic losses. So understanding the physical mechanisms governing the tropical cyclogenesis (TCG) of AEWs remains a crucial problem. Competing theoretical frameworks, including baroclinic instability, barotropic instability, and moisture-vortex instability (MVI) have been proposed, but their relative importance and temporal evolution during storm development remain unclear. In this study, machine learning algorithms are used to empirically analyze the governing mechanisms of AEW development based on 40 years of reanalysis data (1979-2018). We develop a computer vision framework utilizing convolutional neural networks (CNNs) and transformer architectures to identify developing AEWs (DAEWs) from non-developing AEWs (NDAEWs) based on wave-centered composites of key thermodynamic and dynamic variables for storm development. The model results suggest that the MVI framework is a critical factor for high classification accuracy in distinguishing developers from non-developers.  more » « less
Award ID(s):
2411723
PAR ID:
10658032
Author(s) / Creator(s):
;
Publisher / Repository:
NeurIPS Climate Change AI Workshop
Date Published:
Format(s):
Medium: X
Location:
San Diego, CA
Sponsoring Org:
National Science Foundation
More Like this
  1. Abstract While considerable attention has been given to how convectively coupled Kelvin waves (CCKWs) influence the genesis of tropical cyclones (TCs) in the Atlantic Ocean, less attention has been given to their direct influence on African easterly waves (AEWs). This study builds a climatology of AEW and CCKW passages from 1981 to 2019 using an AEW-following framework. Vertical and horizontal composites of these passages are developed and divided into categories based on AEW position and CCKW strength. Many of the relationships that have previously been found for TC genesis also hold true for non-developing AEWs. This includes an increase in convective coverage surrounding the AEW center in phase with the convectively enhanced (“active”) CCKW crest, as well as a buildup of relative vorticity from the lower to upper troposphere following this active crest. Additionally, a new finding is that CCKWs induce specific humidity anomalies around AEWs that are qualitatively similar to those of relative vorticity. These modifications to specific humidity are more pronounced when AEWs are at lower latitudes and interacting with stronger CCKWs. While the influence of CCKWs on AEWs is mostly transient and short lived, CCKWs do modify the AEW propagation speed and westward-filtered relative vorticity, indicating that they may have some longer-term influences on the AEW life cycle. Overall, this analysis provides a more comprehensive view of the AEW–CCKW relationship than has previously been established, and supports assertions by previous studies that CCKW-associated convection, specific humidity, and vorticity may modify the favorability of AEWs to TC genesis over the Atlantic. 
    more » « less
  2. Eddies in the northwestern tropical Atlantic Ocean play a crucial role in transporting the South Atlantic Upper Ocean Water to the North Atlantic and connect the Atlantic and the Caribbean Sea. Although surface characteristics of those eddies have been well studied, their vertical structures and governing mechanisms are much less known. Here, using a time-dependent energetics framework based on the multiscale window transform, we examine the seasonal variability of the eddy kinetic energy (EKE) in the northwestern tropical Atlantic. Both altimeter-based data and ocean reanalyses show a substantial EKE seasonal cycle in the North Brazil Current Retroflection (NBCR) region that is mostly trapped in the upper 200 m. In the most energetic NBCR region, the EKE reaches its minimum in April–June and maximum in July–September. By analyzing six ocean reanalysis products, we find that barotropic instability is the controlling mechanism for the seasonal eddy variability in the NBCR region. Nonlocal processes, including advection and pressure work, play opposite roles in the EKE seasonal cycle. In the eastern part of the NBCR region, the EKE seasonal evolution is similar to the NBCR region. However, it is the nonlocal processes that control the EKE seasonality. In the western part of the NBCR region, the EKE magnitude is one order of magnitude smaller than in the NBCR region and shows a different seasonal cycle, which peaks in March and reaches its minimum in October–November. Our results highlight the complex mechanisms governing eddy variability in the northwestern tropical Atlantic and provide insights into their potential changes with changing background conditions. 
    more » « less
  3. Abstract The conditions associated with tropical cyclones undergoing downshear reformation are explored for the North Atlantic basin from 1998 to 2020. These storms were compared to analog tropical cyclones with similar intensity, vertical wind shear, and maximum potential intensity, but did not undergo downshear reformation. Storm-centered, shear-relative composites were generated using ERA5 and GridSat-B1 data. Downshear reformation predominately occurs for tropical cyclones of tropical storm intensity embedded in moderate vertical wind shear. A comparison between composites suggests that reformed storms are characterized by greater low-level and midtropospheric relative humidity downshear, larger surface latent heat fluxes downshear and left of shear, and larger low-level equivalent potential temperatures and CAPE right of shear. These factors increase thermodynamic favorability, building a reservoir of potential energy and decreasing dry air entrainment, promoting sustained convection downshear, and favoring the development of a new center. Significance StatementThe development of a new low-level circulation center in tropical cyclones that replaces the original center, called downshear reformation, can affect the structure and intensity of storms, representing a challenge in forecasting tropical cyclones. While there have been a handful of case studies on downshear reformation, this study aims to more comprehensively understand the conditions that favor downshear reformation by comparing a large set of North Atlantic tropical cyclones that underwent reformation with a similar set of tropical cyclones that did not undergo reformation. Tropical cyclones that undergo reformation have a moister environment, larger surface evaporation, and higher low-level instability in specific regions that help sustain deep, downshear convection that favors the development of a new center. 
    more » « less
  4. Abstract It is well known that African easterly waves (AEWs) can develop into tropical cyclones. However, the processes leading to development are not well understood. To this end, we examine a 38-yr climatology of AEW tracks sorted into developing AEWs (DAEWs) and strong nondeveloping AEWs (SNDAEWs). Wave-centered composites for tracks in the eastern Atlantic (40°–10°W, 5°S–30°N) and West African monsoon regions (10°W–20°E, 5°S–30°N) reveal that DAEWs occur over a more humid background state in both regions. The more humid environment causes DAEWs to exhibit heavier precipitation and wave amplification via vortex stretching. Examination of the column moist static energy (MSE) budget reveals that DAEWs exhibit stronger radiative heating and more moistening via horizontal MSE advection than SNDAEWs. The stronger horizontal MSE advection in DAEWs is due to a northeast shift in the maximum MSE relative to the wave axis, causing the northerlies in the wave to advect a higher MSE into the maximum precipitation. In contrast, MSE is maximum near the center of NDAEWs, making the moistening of the rainfall by horizontal MSE advection weaker. DAEWs exhibit stronger radiative heating per unit of rainfall relative to NDAEWs, suggesting that cloud-radiative feedbacks are stronger in these systems. The sum of horizontal MSE advection and radiative heating explains the buildup in MSE seen over the rainy region of the DAEWs that is not seen in SNDAEWs. These results underscore the importance of moisture, cloud–radiation interactions, and horizontal MSE advection in tropical cyclone (TC) development over these regions. Significance StatementAfrican easterly waves are the most common precursors of tropical cyclones in the Atlantic basin. Despite significant progress in understanding the processes that distinguish waves that develop into tropical cyclones versus those that do not, important gaps in knowledge remain. In this study, we employed a wave-centered compositing scheme and the moist static energy budget to understand the differences between easterly waves that develop and the strongest nondeveloping waves. Our results show that waves that develop into tropical cyclones occur in a more humid environment where less dry air is transported toward the wave’s rainy region. The more humid environment is also associated with stronger rainfall as well as stronger radiative heating in developing waves, the latter which favors the buildup of moisture in developing waves. Our results underscore the importance of water vapor and its horizontal distribution in determining the development of African easterly waves. 
    more » « less
  5. Storm surge flooding caused by tropical cyclones is a devastating threat to coastal regions, and this threat is growing due to sea-level rise (SLR). Therefore, accurate and rapid projection of the storm surge hazard is critical for coastal communities. This study focuses on developing a new framework that can rapidly predict storm surges under SLR scenarios for any random synthetic storms of interest and assign a probability to its likelihood. The framework leverages the Joint Probability Method with Response Surfaces (JPM-RS) for probabilistic hazard characterization, a storm surge machine learning model, and a SLR model. The JPM probabilities are based on historical tropical cyclone track observations. The storm surge machine learning model was trained based on high-fidelity storm surge simulations provided by the U.S. Army Corps of Engineers (USACE). The SLR was considered by adding the product of the normalized nonlinearity, arising from surge-SLR interaction, and the sea-level change from 1992 to the target year, where nonlinearities are based on high-fidelity storm surge simulations and subsequent analysis by USACE. In this study, this framework was applied to the Chesapeake Bay region of the U.S. and used to estimate the SLR-adjusted probabilistic tropical cyclone flood hazard in two areas: One is an urban Virginia site, and the other is a rural Maryland site. This new framework has the potential to aid in reducing future coastal storm risks in coastal communities by providing robust and rapid hazard assessment that accounts for future sea-level rise. 
    more » « less