Modeling North Atlantic Tropical Cyclone Counts Directly From Sea Surface Temperature Maps

Annual North Atlantic tropical cyclone (TC) counts are frequently modeled as a Poisson process with a state‐dependent rate. Current models based on Poisson regression can explain roughly 50% of the annual variance using three climate indices: El Niño/Southern Oscillation, average sea surface tempera...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Geophysical research letters Jg. 52; H. 12
Hauptverfasser: Wesley, Daniel, Mann, Michael E., Jain, Bhuvnesh, Twomey, Colin R., Christiansen, Shannon
Format: Journal Article
Sprache:Englisch
Veröffentlicht: Washington John Wiley & Sons, Inc 28.06.2025
Schlagworte:
Annual variations
Atmospheric circulation
Atmospheric forcing
Atmospheric oscillations
Circulation indexes
Climatic indexes
Cyclones
El Nino
El Nino phenomena
El Nino-Southern Oscillation event
Global temperatures
Hurricanes
North Atlantic Oscillation
Ocean-atmosphere system
Poisson density functions
Regression analysis
Sea surface temperature
Southern Oscillation
Statistical analysis
Statistical models
Surface temperature
Tropical cyclones
ISSN:0094-8276, 1944-8007
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:Annual North Atlantic tropical cyclone (TC) counts are frequently modeled as a Poisson process with a state‐dependent rate. Current models based on Poisson regression can explain roughly 50% of the annual variance using three climate indices: El Niño/Southern Oscillation, average sea surface temperature (SST) in the main development region of the North Atlantic, and the North Atlantic oscillation atmospheric circulation index. We introduce a new method, based on the Elastic Net (EN) that predicts TC counts directly from global SST maps. We show it achieves performance on par with current models, without requiring manually constructed indices. To understand the performance of the EN we argue that, when TC counts are generated by independent Poisson draws, statistical models are subject to a lower limit on prediction error. We estimate this limit and show that it is saturated by both current models and our new method. Plain Language Summary Our work studies the relationship between the total number of North Atlantic tropical cyclones (TCs) each year and climate factors, such as sea surface temperature (SST). Many researchers have worked on this problem and have typically needed to invent climate indices by hand to use in explaining TC counts. We introduce a method that predicts TC counts directly from global SST maps, and show it performs as well as previous methods. The new method is in principle sensitive to much more information about global climate and does not require hand‐crafted climate indices. We also argue that we may have reached a limit of predictive power and speculate that current models already extract all available information from climate factors to explain TC counts. Key Points We introduce a statistical model to explain interannual variability in tropical cyclone (TC) counts directly from global sea surface temperature maps, and show it performs as well as existing models We present evidence that extending current statistical models by nonlinear interactions between predictors does not reduce model error We explain these results by arguing that if TC counts are generated by a Poisson process there is a lower limit to model error. We estimate the limit and show it is saturated by existing models
Bibliographie:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:0094-8276
1944-8007
DOI:10.1029/2025GL115213