What question did this study set out to answer?

The study aims to create a quantitative framework for analyzing geogenic gas emissions and tectonic activities in arid regions to improve pre-seismic hazard assessment.

March 3, 2026Open Access

DESERTAS: The Desert Breathes — A Quantitative Framework for Decoding Geogenic Gas Emissions, Tectonic Pulse Detection, and Pre-Seismic Geochemical Forecasting in Arid Cratons

Puntos clave

The study aims to create a quantitative framework for analyzing geogenic gas emissions and tectonic activities in arid regions to improve pre-seismic hazard assessment.
Developed an AI-driven framework named DESERTAS to quantify geogenic gas emissions.
Analyzed data from 2,491 Desert Rock-Gas Units across 36 stations in various arid cratons.
Integrated eight parameters into the Desert Rock-Gas Intelligence Score for assessment.
Validated findings over 22 years of monitoring to ensure reliability.
Achieved a classification accuracy of 90.6% using the Desert Rock-Gas Intelligence Score.
Detected pre-seismic radon with a rate of 93.1% and a low false alert rate of 5.4%.
Provided a mean pre-seismic lead time of 58 days for events of magnitude ≥ 4.0.
Demonstrated a strong correlation (r = +0.904) between radon pulses and the DRGIS.

Resumen

DESERTAS (Desert Emission Sensing & Energetic Rock-Tectonic Analysis System) introduces the first mathematically integrated, AI-driven geophysical framework for the systematic quantification of geogenic gas emissions from rock fissures in hyperarid environments — the Desert Rock-Gas Intelligence Score (DRGIS). Built on eight physically orthogonal parameters (ΔΦₜh, Ψcrack, Rnₚulse, Ωₐrid, Γgeo, Heᵣatio, βdust, Sᵧield), DESERTAS transforms the continuous geochemical breath of desert rock fractures into a quantitative diagnostic tool for pre-seismic hazard assessment. Validated against 2, 491 Desert Rock-Gas Units (DRGUs) spanning 36 monitoring stations across 7 arid craton systems (Saharan, Arabian Shield, Kaapvaal, Yilgarn, Atacama-Pampean, Tarim Basin, Scandinavian Shield) over 22 years (2004–2026). Key results: - DRGIS Classification Accuracy: 90. 6% - Pre-seismic Radon Detection Rate: 93. 1% · False Alert Rate: 5. 4% - Mean Pre-Seismic Lead Time: 58 days before M ≥ 4. 0 events - Maximum Lead Time: 134 days (Saharan Shield, 2019) - Rnₚulse/DRGIS Correlation: r = +0. 904 (p < 0. 001, n = 2, 491) - Heᵣatio Source Discrimination: 99. 1% · depth ±800 m - βdust Particulate Transport: detectable at 340 km downwind - AI Ensemble vs. single-parameter: +18. 2% improvement Submitted to Nature Geoscience, March 2026. Code: https: //gitlab. com/gitdeeper07/desertas Dashboard: https: //desert-as. netlify. app PyPI: https: //pypi. org/project/desertas/1. 0. 0/

Leer artículo completoexternamente

Me gusta

Guardar

Ver artículo completo