Hydrogen underfoot – a more than 200-year supply

Modelling of geologic hydrogen in the Earth’s subsurface suggests that less than 2% of the resource could supply the projected need for net zero for around 200 years.

New modelling by US Geological Survey (USGS) scientists Geoffrey Ellis and Sarah Gelman suggest that the potential in-place hydrogen resource in the Earth’s subsurface – aka ‘white hydrogen' in the hydrogen colour palette – is between one billion and 10 quadrillion metric tons, with the most probable level around 5.6 trillion metric tons.

Given that most of this hydrogen is likely to be impractical to recover, just a small fraction, around 100 billion metric tons – 1.8% – would supply the projected hydrogen needed to reach net zero carbon emissions for approximately 200 years, the scientists state.

This amount of hydrogen contains more energy, around 14 quadrillion MJ and almost double than all proven natural gas reserves on Earth of around 8.4 quadrillion MJ.

Have you read?
Low-emissions hydrogen projects growing but policy support still lags
Can clean hydrogen power the future?

In a paper in the journal Science Advances, the scientists point to the recent discovery of a substantial accumulation of natural hydrogen in Mali that has challenged the long-held view that such fields do not exist.

They also note a growing recognition among geoscientists that suitable exploration tools have not been deployed in the appropriate locations to truly evaluate the resource potential of natural hydrogen in the Earth’s subsurface.

Mass balance model

The USGS model is based on a ‘mass balance’ approach with inputs including the annual generation of natural hydrogen, fraction of hydrogen detained in traps, residence time in reservoirs, proportion of biotic and abiotic loss and the rate of anthropogenic production.

The calculated outputs of the model are the amount of hydrogen stored in reservoirs at a given time and the flux to the atmosphere, primarily from volcanic and hydrothermal settings and with most thought to be taken up by soils.

The scientists comment that given the uncertainties in the model construction and inputs, the results should be viewed as a first-order approximation of the magnitude of the potential in-place geologic hydrogen resource.

Moreover, it makes no predictions about the distribution of the hydrogen in the subsurface, which is critical for the economic viability of any potential resource.

However, given what is known about the distribution of petroleum and nonpetroleum fluids such as helium and CO2 in the subsurface, it is likely that recovery of most subsurface hydrogen can be expected to be in accumulations that are too deep, too far offshore or too small to be economically recovered.

But even if a small amount of the most probable predicted in-place resource was recoverable, this could represent a substantial resource.

They also add that their estimate is only for natural hydrogen potentially stored in accumulations in the subsurface.

Additionally, it is possible that natural hydrogen production could be stimulated to increase the rate of generation or induce generation in settings where it has the potential but is not naturally doing so.

Is the hydrogen renewable?

The final issue the scientists address is whether the hydrogen is renewable, i.e. that the generation rates are rapid enough to offset the potential extraction rates from the reservoirs.

Their modelling suggests a probable global renewable hydrogen production rate of about 5 million metric tons per year, which would meet only less than 1% of the projected worldwide 2050 demand.

However, the model does not account for potential geologic hydrogen that might be produced as it is generated or moves through the subsurface, which would be a renewable resource.

In conclusion the researchers point to the need for further research in the field, with a better understanding of the rates and controls on geologic hydrogen consumption in subsurface accumulations and more accurate estimates of the rates of natural hydrogen generation to improve model predictions of the resource potential.

The realisation of potential natural hydrogen resources will require a more advanced understanding of the processes that lead to the accumulation of hydrogen in the subsurface as well as optimised methods for finding these resources, they conclude.