loading page

Modeling deep control pulsing flux of native H2 throughout tectonic fault-valve systems
  • +2
  • Frederic Victor DONZE,
  • Lukas Bourdet,
  • Laurent Truche,
  • Camille Dusséaux,
  • Pascale Huyghe
Frederic Victor DONZE
Université Grenoble Alpes, ISTerre, UMR 5275 of CNRS,

Corresponding Author:frederic.donze@univ-grenoble-alpes.fr

Author Profile
Lukas Bourdet
Université Grenoble Alpes
Author Profile
Laurent Truche
ISTerre, UMR 5275 of CNRS, University of Grenoble Alpes, F-38041 Grenoble Cedex 9, France.
Author Profile
Camille Dusséaux
University of Plymouth
Author Profile
Pascale Huyghe
Université Grenoble Alpes
Author Profile


Pulsing seepages of native hydrogen (H2) have been observed at the surface on several emitting structures. It is still unclear whether this H2 pulsed flux is controlled by deep migration processes, atmosphere/near-surface interactions or by bacterial fermentation. Here, we investigate mechanisms that may trigger pulsating fluid migration at depth and the resulting periodicity. We set up a numerical model to simulate the migration of a deep constant fluid flow. To verify the model’s formulation to solve complex fluid flows, we first simulate the morphology and amplitude of 2D thermal anomalies induced by buoyancy-driven water flow within a fault zone. Then, we simulate the H2 gas flow along a 1-km draining fault, crosscut by a lower permeable rock layer to investigate the conditions for which a pulsing system is generated from a deep control. For a constant incoming flow of H2 at depth, persistent bursts at the surface only appear in the model if: (I) a permeability with an effective-stress dependency is used, (II) a strong contrast of permeability exists between the different zones, (III) a sufficiently high value of the initial effective stress state at the base of the low permeable layer exists, and (IV) the incoming and continuous fluid flow of H2 at depth remains low enough so that the overpressure does not “open” instantly the low permeability layer. The typical periodicity expected for this type of valve-fault control of H2 pulses at the surface is at a time scale of the order of 100 to 300 days.