Aurora Hunter: A Two-Stage Framework for Probabilistic Visibility Forecasting
Abstract
Aurora visibility at a given location requires two physically distinct conditions to hold at once: aurora occurring overhead, governed by solar wind-magnetosphere coupling, and observing conditions that permit detection, governed by cloud cover and moonlight. Approaches that conflate the two weaken the space-weather signal and limit cross-site generalizability. We develop Aurora Hunter, a two-stage cascade that separates occurrence prediction from observing-condition assessment. Stage 1 uses 43 physics-driven features to predict P(aurora identifiable in all-sky images) via gradient-boosted trees trained on joint Tromso+Kiruna data (about 16,600 hours, 2015-2023). Stage 2 models P(unobscured image class | identifiable) with logistic regression on 15 observing-condition features, trained on hours with identifiable aurora. The cascade P(visible) = P(identifiable) x P(unobscured | identifiable) achieves retrospective ROC-AUC of 0.958 (Tromso test, 2019-2020) and 0.933 (independent Kiruna, 2024), improving on the occurrence stage used alone by +0.095 and +0.097. Transfer to Skibotn, the one station withheld entirely (2022-2025), is limited. SHAP analysis identifies magnetic local time, the Kp x nightside interaction and the three-hour mean Kp as the dominant features (44% of attribution), consistent with auroral oval physics. Hemisphere-wide occurrence maps combine the Stage 1 amplitude, on a clear-sky basis, with the Feldstein oval parameterization. By providing location-specific visibility probabilities with measured reliability rather than coarse geomagnetic indices, the framework links space weather research to practical observation planning. An operational proof of concept is available at https://aurora-hunter.onrender.com
Related
- Probabilistic Machine Learning for Short-Term Atmospheric Seeing Forecasting
- IViS: Interferometric Visibility-domain inversion Software. A GPU-accelerated Python framework for joint deconvolution with ASKAP
- PROSWIN: Probabilistic Solar Wind Speed Forecasting Using Deep Distributional Regression From Solar Images
- IViS: Interferometric Visibility-domain inversion Software - A GPU-accelerated Python framework for joint deconvolution with ASKAP
- A Probabilistic Framework for Incorporating Helioseismic Far-Side Active Regions into SFT Models
BibTeX
@misc{ge2026aurora,
title = {{Aurora Hunter: A Two-Stage Framework for Probabilistic Visibility Forecasting}},
author = {Zongyuan Ge and Chenwaner Zhang and Haoyang Li and Hantai Zhang and Wei Zhou and Wenxin Gu and Zhaoming Wang},
howpublished = {arXiv},
year = {2026},
month = sep,
doi = {10.48550/arxiv.2605.24038},
eprint = {2605.24038},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2605.24038},
}
RIS
TY - UNPB
TI - Aurora Hunter: A Two-Stage Framework for Probabilistic Visibility Forecasting
AU - Zongyuan Ge
AU - Chenwaner Zhang
AU - Haoyang Li
AU - Hantai Zhang
AU - Wei Zhou
AU - Wenxin Gu
AU - Zhaoming Wang
PY - 2026
DA - 2026/09/21
PB - arXiv
DO - 10.48550/arxiv.2605.24038
UR - https://arxiv.org/abs/2605.24038
AB - Aurora visibility at a given location requires two physically distinct conditions to hold at once: aurora occurring overhead, governed by solar wind-magnetosphere coupling, and observing conditions that permit detection, governed by cloud cover and moonlight. Approaches that conflate the two weaken the space-weather signal and limit cross-site generalizability. We develop Aurora Hunter, a two-stage cascade that separates occurrence prediction from observing-condition assessment. Stage 1 uses 43 physics-driven features to predict P(aurora identifiable in all-sky images) via gradient-boosted trees trained on joint Tromso+Kiruna data (about 16,600 hours, 2015-2023). Stage 2 models P(unobscured image class | identifiable) with logistic regression on 15 observing-condition features, trained on hours with identifiable aurora. The cascade P(visible) = P(identifiable) x P(unobscured | identifiable) achieves retrospective ROC-AUC of 0.958 (Tromso test, 2019-2020) and 0.933 (independent Kiruna, 2024), improving on the occurrence stage used alone by +0.095 and +0.097. Transfer to Skibotn, the one station withheld entirely (2022-2025), is limited. SHAP analysis identifies magnetic local time, the Kp x nightside interaction and the three-hour mean Kp as the dominant features (44% of attribution), consistent with auroral oval physics. Hemisphere-wide occurrence maps combine the Stage 1 amplitude, on a clear-sky basis, with the Feldstein oval parameterization. By providing location-specific visibility probabilities with measured reliability rather than coarse geomagnetic indices, the framework links space weather research to practical observation planning. An operational proof of concept is available at https://aurora-hunter.onrender.com
ER -