During the last ice age lasting from ~115,000 to ~12,000 year ago, the glaciers and large ice sheets covered up to ~25% of the total land area (it is only ~11% today) and occupied some of continental shelves. Barents Sea shelf currently submerged 230 m below sea surface on average was one such fully ice-covered territory. A >1 km thick ice sheet was resting and sliding on bedrocks and scraped off 130,000 km3 of them sculpturing its present topography1. To put this in perspective, the volume of material removed from the shelf during the last ice age is equal to the volume of water discharged by the Amazon river over a period of ~20 years.
The last ice age was not a one-of-a-kind. More than 30 of them occurred on the Barents Sea shelf during the last 2.58 million years known as the Quaternary Period. Altogether, Quaternary ice sheets removed > 1 km thick layer of rocks2. This explains why, old (250 – 150 million years) sedimentary rocks appear on or very close to the seafloor today. Some of these rocks happen to contain hydrocarbons such as natural gas and oil.
Typically, hydrocarbon accumulations are locked in deep geological structures. Such structures consist of porous reservoir rock surrounded with impermeable sealing rocks and require drilling to retrieve oil or gas. During hydrocarbon production, the unwanted losses of oil and gas should be amenable. However, hydrocarbon leakage from reservoirs onto the Earth surface may also occur through natural conduits which are problematic to find, quantify and, especially, control.
Due to the strong erosion of the Barents Sea shelf in Quaternary Period and beyond, the seals of hydrocarbon accumulations may be locally damaged leading to naturally uncapped hydrocarbon reservoirs leaking their content into the seawater. In other words, glacial erosion might have created large and uncontrollable natural “boreholes”. Are these natural "boreholes" leaking hydrocarbons today and how widespread the leakage is? Until now, a common perception was that hydrocarbon reservoirs across the Barents Sea shelf were excavated and drained some time during the Quaternary Period3. Reservoirs left intact against all the odds must be tight and do not let hydrocarbons to escape in significant quantities. Does this mean that hydrocarbon in Barents Sea shelf are securely locked in remaining structures or had leaked in geological past? Is the ongoing leakage not likely?
CAGE – Center for Arctic Gas Hydrate, Environment and Climate from the Arctic University of Norway joined forces with the Norwegian Petroleum Directorate to investigate whether accumulations of hydrocarbons naturally excavated by glacial erosion remain leaking gas and oil in the water and under which geological circumstances. We focused on the part of the shelf north of 74o 30´N, which is not open for petroleum activity and has not been surveyed extensively in the past. Several research expeditions later, we report that the extent and the magnitude of the present-day hydrocarbon escape from glacially uncapped accumulations is beyond surprising.
During several expeditions, we used multibeam sonars to map the seafloor and to detect bubbles of gas in the water. Altogether we scanned 660 km2 of the seafloor (a size of 11 Manhattan Islands in New York, USA or ~80,000 football fields) and water overlying it and identified >4,000 streams of gas bubbles called seeps rising through the water towards the sea surface. The gas bubbles contained plenty of greenhouse gas methane with some admixture of ethane and propane. Some areas exhibited “forests” of gas seeps with up to 200 of them within 1 km2 of the seafloor area. The extent and magnitude of gas seepage is significant compared to other known methane release regions across the continental shelves worldwide.
Not only methane gas liberates into the water. Oil does too. On a sunny day with calm sea, oil droplets rising through the 300 m thick water column (oil is buoyant in sea water) and hitting its surface are well visible with a naked eye. Once a droplet reaches the surface, it forms a stain of a size of large pancake with rich rainbow colors. The “pancakes” merge with each other forming an oil slick, which later gets stretched, twirled and ripped with winds and waves. The slicks we found were large enough (up to 10 km long) to be well visible from space. The satellite images were very helpful to find oil slicks in dark nights and at not so favourable weather conditions and guided the expeditions to a proper location for sampling.
Documenting the oil seepage from space did not satisfy our curiosity. We wanted to observe this phenomena from a closer distance and collect samples. A remotely operated underwater vehicle equipped with cameras, powerful yet precise mechanic arms and a suite of sensors came in handy. Through the lens of the underwater robot, the seafloor appears as dark, cold, vast plain of muddy sediments with sporadic stones, tube worms and flatfish. Oil droplets and gas bubbles steadily detach from the bottom with frequency almost regular as clockwork and begin their journey through several hundred meters of water towards the sea surface. It looks peaceful until the delicate muddy plain is disturbed with robot´s powerful mechanic arms. The clouds of oil droplets and gas bubbles burst from spots where we take samples from. The retrieved bottom sediments contain smears of sticky, dense dark brown oil.
Given that some of the Arctic and North Atlantic continental shelves bear significant quantity of oil and natural gas4 and at places experienced glacial erosion on a scale similar to the Barents Sea shelf, it is possible that oil and gas escape from naturally uncapped hydrocarbon reservoirs is far more pervasive phenomenon than previously thought. Luckily, the fluxes of greenhouse methane gas are naturally constrained by microorganisms living on the seafloor and in the sea water which consume major portion of the gas and are capable of degrading oil too. Resonating with this, we report that content of methane in surface water is an order of magnitude lower than in near-bottom water layer. Whether such depleted, yet possibly widespread and long-lasting fluxes of methane through the sea water and to the air make a significant contribution to the atmospheric methane budgets remains a question deserving attention.
1 Patton, H. et al. The extreme yet transient nature of glacial erosion. Nature Communications 13, 7377 (2022). https://doi.org:10.1038/s41467-022-35072-0
2 Lasabuda, A. P. E. et al. Cenozoic uplift and erosion of the Norwegian Barents Shelf – A review. Earth-Science Reviews 217, 103609 (2021). https://doi.org:https://doi.org/10.1016/j.earscirev.2021.103609
3 Ostanin, I., Anka, Z. & Di Primio, R. Role of Faults in Hydrocarbon Leakage in the Hammerfest Basin, SW Barents Sea: Insights from Seismic Data and Numerical Modelling. Geosciences 7, 28 (2017).
4 Gautier, D. L. et al. Assessment of Undiscovered Oil and Gas in the Arctic. Science 324, 1175-1179 (2009). https://doi.org:10.1126/science.1169467