EGU25-7577, updated on 14 Mar 2025
https://doi.org/10.5194/egusphere-egu25-7577
EGU General Assembly 2025
© Author(s) 2025. This work is distributed under
the Creative Commons Attribution 4.0 License.
Poster | Friday, 02 May, 14:00–15:45 (CEST), Display time Friday, 02 May, 08:30–18:00
 
vPoster spot 2, vP2.2
Understanding Marine Heat Waves in the Chesapeake Bay: Drivers, Variability, and Predictive Insights Using Machine Learning
Cyrus Li1, Noah Xiong2, and Jian Zhao3
Cyrus Li et al.
  • 1Ocean Lakes High School Math And Science Academy, Virginia Beach VA, United States of America, cyruslee2023@gmail.com
  • 2Ocean Lakes High School Math And Science Academy, Virginia Beach VA, United States of America, noahxiongweihe@gmail.com
  • 3University of Maryland Center for Environmental Science, Cambridge MD, United States of America, jianzhao@umces.edu

Marine heat waves (MHWs) pose significant threats to coastal ecosystems, with particularly severe impacts in shallow waters where their magnitude is often amplified. The Chesapeake Bay, the largest estuary in the United States, is highly vulnerable to these events, which have increased in frequency and duration in recent decades. MHWs in the Chesapeake Bay have critical implications for its ecological balance, including effects on fish populations, habitat degradation, and water quality. Despite their growing prevalence, the underlying causes of these events and the factors regulating their variability remain poorly understood. Our study employs machine learning approaches to elucidate the drivers of marine heat waves in the Chesapeake Bay and to quantify their contributions to these extreme temperature events. By incorporating a comprehensive set of potential predictors, including local air temperature, wind forcing, river discharge, and Atlantic Ocean temperature, the model reveals the key mechanisms driving the onset, intensity, and persistence of MHWs in the Chesapeake Bay. Advanced feature selection techniques isolate the most relevant variables, while model outputs are validated against observed data to ensure accuracy and robustness. Our results suggest that local air temperature and ocean temperature anomalies from the Atlantic Ocean are dominant in triggering MHWs. These findings shed light on the complex interactions between atmospheric, hydrological, and oceanographic processes in shaping extreme thermal events in estuarine systems.

How to cite: Li, C., Xiong, N., and Zhao, J.: Understanding Marine Heat Waves in the Chesapeake Bay: Drivers, Variability, and Predictive Insights Using Machine Learning, EGU General Assembly 2025, Vienna, Austria, 27 Apr–2 May 2025, EGU25-7577, https://doi.org/10.5194/egusphere-egu25-7577, 2025.