Model of Albania's Bulqizë fault: 10 years of hydrogen cycling lowers slip risk, simulation only
Researchers modeled the natural hydrogen reservoir at Bulqizë, Albania, as a storage site. In simulation, fault reactivation risk fell over ten years of injection and withdrawal.

Lei Hou of the China-UK Low Carbon College at Shanghai Jiao Tong University and colleagues, including Fengshou Zhang of Tongji University and Derek Elsworth of Penn State, modeled the Bulqizë fault-controlled natural hydrogen reservoir in Albania as a storage site. The study was published in Proceedings of the National Academy of Sciences. A ten-year simulation of cyclic injection and production gave a stable throughput of 7.5 × 10^5 m³ per day, with fault reactivation risk decreasing over time pnas.org
Proceedings of the National Academy of Sciencespnas.orgOpen source ↗. All results come from computer simulations: no hydrogen has been pumped into this fault.
Natural reservoir as a storage site
The authors note that none of the conventional candidates (salt caverns, aquifers, depleted hydrocarbon reservoirs) offers direct prior evidence of hydrogen-specific sealing. They propose that subcommercial natural hydrogen reservoirs, where the gas has accumulated over geological timescales, can serve as precertified storage sites.
The coupled geofluid–geomechanical model was built from exploration data. It covers the 5-km vertical extent of the fault zone with depth-dependent temperature and permeability, and treats hydrogen dissolution under two end-member hydrogeological configurations. Under conservative upper-bound assumptions, losses from dissolution and abiotic consumption stayed below 2% of the total inventory. According to the authors, this is consistent with the sustained natural accumulation observed in the reservoir.
Fault stability and capacity
Fault stability was assessed through slip tendency analysis. The model showed two permeability-dependent reactivation modes: localized instability near the injection interval at low permeability, and instability at the fault–caprock interface at high permeability pnas.org
Proceedings of the National Academy of Sciencespnas.orgOpen source ↗. A parametric analysis of 72 combinations of injection rate, location and permeability defined safe operating envelopes and gave optimized storage capacities of 3.5 to 4.8 × 10^8 m³.
In the ten-year cycle, the risk of reactivation decreased progressively. The authors attribute this to pressure homogenization through gas redistribution. Hou told Earth.com that the effect is not automatic and depends on choosing the right injection depth.
The model does not include changes in rock permeability under shifting stress or rock fatigue from repeated pressure swings, which could alter fault behavior beyond ten years. Hou declined to name a date for commercial storage at such a site and said that field trials and engineering costs will set the timeline.



