Subseasonal Forecasts of Opportunity Identified by an Explainable Neural Network
In one sentence. Neural network identifies subseasonal forecasts of opportunity over the North Atlantic at 22-day lead, with layer-wise relevance propagation revealing tropical predictors.
Abstract
Abstract Midlatitude prediction on subseasonal timescales is difficult due to the chaotic nature of the atmosphere and often requires the identification of favorable atmospheric conditions that may lead to enhanced skill (“forecasts of opportunity”). Here, we demonstrate that an artificial neural network (ANN) can identify such opportunities for tropical‐extratropical circulation teleconnections within the North Atlantic (40°N, 325°E) at a lead of 22 days using the network's confidence in a given prediction. Furthermore, layer‐wise relevance propagation (LRP), an ANN explainability technique, pinpoints the relevant tropical features the ANN uses to make accurate predictions. We find that LRP identifies tropical hot spots that correspond to known favorable regions for midlatitude teleconnections and reveals a potential new pattern for prediction in the North Atlantic on subseasonal timescales.
Labels
- Topic
- Sub-seasonal to seasonal, Explainability, uncertainty & physics
- Type
- New model or method
- Method
- Classical ML (trees, GP, linear, kernels)
- Time scale
- 2 weeks–2 months
- Domain
- Regional
- Data
- ERA5 / reanalysis
- Status
- Research
- Identifiers
- DOI:10.1029/2020gl092092 OpenAlex:W3160924831
- Added
- 2026-10-22
Labels and summary are generated by Claude and reviewed weekly. Wrong? Edit this entry.