Research

Paper

TESTING March 03, 2026

Probabilistic and Alarm-Based Evaluation of a b-Value-Driven Deep Learning Earthquake Forecast

Authors

Jonas Köhler, Wei Li, Johannes Faber, Georg Rümpker, Nishtha Srivastava

Abstract

We evaluate the forecasting performance of a deep learning model, originally introduced as a pattern-extraction framework, that operates on the spatiotemporal evolution of seismic b-values in a short-term forecasting context. Model output is rescaled to account for training on balanced datasets and evaluated relative to a spatial base-rate model using the Brier Skill Score (BSS). Absolute skill values are small, but mean BSS values are consistently positive, including at locations where Mw geq 5 earthquakes occurred during the test period, indicating information content beyond historical seismicity alone. Alarm-based evaluation using Molchan diagrams shows elevated event capture rates at low alarm fractions (5.88 percent of events captured at 1 percent area under alarm), indicating discrimination exceeding random and purely spatial reference models under constrained alarm conditions. Comparison with ETAS-derived triggered probabilities further reveals a weak positive correlation, suggesting partial sensitivity of the model output to seismic regimes characterized by enhanced clustering and recent activity, while remaining distinct from classical aftershock-based descriptions. Together, these results indicate that spatiotemporal variations in b-values contain a persistent, though limited, signal relevant to probabilistic earthquake forecasting, yielding marginal but consistent improvements over baseline models across complementary evaluation frameworks.

Metadata

arXiv ID: 2603.03079
Provider: ARXIV
Primary Category: physics.geo-ph
Published: 2026-03-03
Fetched: 2026-03-04 03:41

Related papers

Raw Data (Debug)
{
  "raw_xml": "<entry>\n    <id>http://arxiv.org/abs/2603.03079v1</id>\n    <title>Probabilistic and Alarm-Based Evaluation of a b-Value-Driven Deep Learning Earthquake Forecast</title>\n    <updated>2026-03-03T15:24:00Z</updated>\n    <link href='https://arxiv.org/abs/2603.03079v1' rel='alternate' type='text/html'/>\n    <link href='https://arxiv.org/pdf/2603.03079v1' rel='related' title='pdf' type='application/pdf'/>\n    <summary>We evaluate the forecasting performance of a deep learning model, originally introduced as a pattern-extraction framework, that operates on the spatiotemporal evolution of seismic b-values in a short-term forecasting context. Model output is rescaled to account for training on balanced datasets and evaluated relative to a spatial base-rate model using the Brier Skill Score (BSS). Absolute skill values are small, but mean BSS values are consistently positive, including at locations where Mw geq 5 earthquakes occurred during the test period, indicating information content beyond historical seismicity alone. Alarm-based evaluation using Molchan diagrams shows elevated event capture rates at low alarm fractions (5.88 percent of events captured at 1 percent area under alarm), indicating discrimination exceeding random and purely spatial reference models under constrained alarm conditions. Comparison with ETAS-derived triggered probabilities further reveals a weak positive correlation, suggesting partial sensitivity of the model output to seismic regimes characterized by enhanced clustering and recent activity, while remaining distinct from classical aftershock-based descriptions. Together, these results indicate that spatiotemporal variations in b-values contain a persistent, though limited, signal relevant to probabilistic earthquake forecasting, yielding marginal but consistent improvements over baseline models across complementary evaluation frameworks.</summary>\n    <category scheme='http://arxiv.org/schemas/atom' term='physics.geo-ph'/>\n    <published>2026-03-03T15:24:00Z</published>\n    <arxiv:primary_category term='physics.geo-ph'/>\n    <author>\n      <name>Jonas Köhler</name>\n    </author>\n    <author>\n      <name>Wei Li</name>\n    </author>\n    <author>\n      <name>Johannes Faber</name>\n    </author>\n    <author>\n      <name>Georg Rümpker</name>\n    </author>\n    <author>\n      <name>Nishtha Srivastava</name>\n    </author>\n  </entry>"
}