May 22 - July 31
Analyzing How Teleconnections Influence Extreme Rainfall Events in the Central U.S. Using Causal Networks
Christopher L. Germano, Jorge A. Celis, and Maria Madsen
What is already known:
- Extreme Weather in Central U.S.: The Central United States frequently experiences extreme weather events, including heavy rainfall and flooding, which cause significant damage to infrastructure and agriculture.
- Influence of Teleconnections: Large-scale climate anomalies like the El Niño-Southern Oscillation (ENSO) and North Atlantic Oscillation (NAO) significantly impact weather patterns in the Central U.S.
- Increasing Heavy Rainfall Trends: Previous studies have shown a positive trend of increasing heavy rainfall events throughout the century in the Central U.S.
- Limitations of Traditional Models: Traditional statistical models struggle to distinguish correlation from causation in complex, nonlinear climate systems.
What this study adds:
- Causal Network Analysis: Utilizes causal network-based machine learning models to form direct links between teleconnections and extreme rainfall to further understand their interactions.
- Integration of Teleconnection Indices: Incorporates indices of teleconnections like ENSO, NAO, PNA, EPO, WPO, and AO into causal network analysis to assess their impact on extreme rainfall events in the Central U.S.
- Focus on Specific Regions: Concentrates on the Ohio Valley and Upper Midwest, regions particularly susceptible to extreme rainfall, providing targeted insights into their unique climatological characteristics.
- Advanced Analytical Methods: Leverages advanced causal discovery algorithms and statistical methods to estimate the strength and direction of causal relationships between atmospheric variables.
Abstract:
Extreme rainfall events in the central U.S. have caused significant damage, underscoring the need for a better understanding of precipitation changes and their drivers. This study utilizes ERA5 data from 1950 to 2023, focusing on the November to May period, to examine changes in winter and spring precipitation and the influence of teleconnections. Our analysis reveals notable increases in precipitation, particularly in spring, with the most recent period showing the highest values. Employing causal discovery methods, we identify significant causal relationships between teleconnections and extreme rainfall events. The Pacific-North American Pattern (PNA) and the Eastern Pacific Oscillation (EPO) emerge as key teleconnections with strong direct connections to extreme rainfall in the Upper Midwest and Ohio Valley. Furthermore, the study uncovers a complex network of interactions among various teleconnections, suggesting that the combined effects of multiple teleconnections must be considered to fully understand their impact on precipitation. These findings provide insights into the complex dynamics influencing precipitation patterns, hoping to aid in better prediction and management of extreme rainfall events.