Causal-prior-constrained deep learning for rapid-intensification warning of western North Pacific tropical cyclones
Abstract
Rapid intensification (RI) is one of the most dangerous and unpredictable phases of tropical cyclones, and its environmental and structural drivers are difficult to identify and separate in observations. Here we introduce CausalVisionRI, a causal-prior-constrained deep-learning system for 24-h RI warning in the western North Pacific. It integrates infrared remote sensing imagery, atmospheric and oceanic fields, and an observationally discovered causal graph into a unified model. The graph and eight sign-verified effects on 24-h intensity change provide an auditable causal prior for training. Discovered from 522 cyclones in 2000-2017, it identifies causal pathways linking atmospheric and oceanic conditions, the cyclone's dynamical and structural state, and its 24-h intensity change. On a forward holdout covering 2020-2024 and 3,004 records, the five-member ensemble achieves a Brier skill score (BSS) of 0.259 relative to a fixed development climatology, significantly exceeding four reference systems; the three LDA baselines reach at most 0.105. It also attains higher detection rates than a published knowledge-block scheme and a reconstructed contrastive-learning scheme at matched false-detection rates under our USA 30-kt standard and the published RITCF standard. Ablations show that the graph pathway adds 0.032 BSS across the holdout and 0.063 over 2020-2021, with a significant contribution in one ensemble member. A deep-learning model constrained by observationally discovered causal structure thus offers a new route to physically informed RI warning.
Related
- Basin-Specific Changes and Intensity of Tropical Cyclones
- Enhanced Influence of the North Pacific Meridional Mode on the Western North Pacific Tropical Cyclones in a Warming Climate
- Robust Regional Predictability of North Atlantic Tropical Cyclones on Seasonal Timescale
- Can AI Forecast the Extremes? Evaluating Probabilistic Weather Models on Tropical Cyclones
- Rare Tropical Cyclones: Tropical Cyclone Hidaya, A High‐Impact Low‐Latitude Event
BibTeX
@misc{li2026causalpriorconstrained,
title = {{Causal-prior-constrained deep learning for rapid-intensification warning of western North Pacific tropical cyclones}},
author = {Xinrui Li and Yijun Jin and Yifan Dong and Shoujuan Shu},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2610.40304},
eprint = {2610.40304},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2610.40304},
}
RIS
TY - UNPB
TI - Causal-prior-constrained deep learning for rapid-intensification warning of western North Pacific tropical cyclones
AU - Xinrui Li
AU - Yijun Jin
AU - Yifan Dong
AU - Shoujuan Shu
PY - 2026
DA - 2026/10/29
PB - arXiv
DO - 10.48550/arxiv.2610.40304
UR - https://arxiv.org/abs/2610.40304
AB - Rapid intensification (RI) is one of the most dangerous and unpredictable phases of tropical cyclones, and its environmental and structural drivers are difficult to identify and separate in observations. Here we introduce CausalVisionRI, a causal-prior-constrained deep-learning system for 24-h RI warning in the western North Pacific. It integrates infrared remote sensing imagery, atmospheric and oceanic fields, and an observationally discovered causal graph into a unified model. The graph and eight sign-verified effects on 24-h intensity change provide an auditable causal prior for training. Discovered from 522 cyclones in 2000-2017, it identifies causal pathways linking atmospheric and oceanic conditions, the cyclone's dynamical and structural state, and its 24-h intensity change. On a forward holdout covering 2020-2024 and 3,004 records, the five-member ensemble achieves a Brier skill score (BSS) of 0.259 relative to a fixed development climatology, significantly exceeding four reference systems; the three LDA baselines reach at most 0.105. It also attains higher detection rates than a published knowledge-block scheme and a reconstructed contrastive-learning scheme at matched false-detection rates under our USA 30-kt standard and the published RITCF standard. Ablations show that the graph pathway adds 0.032 BSS across the holdout and 0.063 over 2020-2021, with a significant contribution in one ensemble member. A deep-learning model constrained by observationally discovered causal structure thus offers a new route to physically informed RI warning.
ER -