Home / Our Science / Energy & Resources / Gas Hydrates / Current Research / Pockmarks and their possible link to glacial gas hydrate destabilization

Pockmarks and their possible link to glacial gas hydrate destabilization

Project: Uncorking the hydrate bottle: Release of methane from melting gas hydrates during glacial cycles on the Chatham Rise, New Zealand

Funder: Royal Society of New Zealand Marsden Fund (2011-2014)


  • Ingo Pecher (University Auckland & GNS Science)
  • Andrew Gorman (University of Otago)
  • Bryan Davy (GNS Science)
  • Helen Neil (NIWA)
  • Graham Rickard (NIWA)
  • Hossam Abuel-Naga (University of Auckland)
  • Christian Berndt (IfM-Geomar, University of Kiel, Germany)
  • Jörg Bialas (IfM-Geomar, University of Kiel, Germany)
  • Richard Coffin (Naval Research Laboratory, USA)
  • George Moridis (Lawrence Berkeley National Laboratory, USA)
  • Matthew Reagan (Lawrence Berkeley National Laboratory, USA)


Large quantities of methane, a potent greenhouse gas, are stored beneath the seafloor in frozen gas hydrates. Hydrate stability depends on pressure and temperature and thus, vast amounts of methane may be released from melting hydrates during climate change with potentially significant environmental implications (Kennett et al., 2000) (Kvenvolden, 1993) (Solomon et al., 2009).

A >20,000 km2 region on the southern flank of the Chatham Rise east of New Zealand is covered by seafloor depressions (SDs) that we interpret as gas-escape features [Davy et al., 2010]. We have observed three classes of SDs (Figure):

  • Sub-circular features ~150 m across and 4-8 m deep in water depths of 500-700 m. These are interpreted as pockmarks.
  • Irregular SDs ~1-5 km across and 50-150 m deep in water depths of 800-1100 m. These are similar to giant pockmarks observed elsewhere [Judd and Hovland, 2007].
  • Large, circular SDs, 8-11 km across and 80-100 m deep in water depths of 800-1100 m.

The shallower pockmarks are close to the current top of gas hydrate stability. Parasound sub-bottom profiles show a reflection that may constitute a bottom simulating reflection (BSR) at the base of gas hydrate stability (BGHS), suggesting a link between the pockmarks and gas hydrates. The Parasound data also reveal buried pockmarks at horizons that appear to mark glaciation peaks.

Figure: Key observations (after Davy et al., 2010).
a) Location map. Shaded: Known extent of pockmark region. STF: Sub-Tropical Front. R: Escarpment.
b) Pockmarks in 500-700 m water depth. Parasound profile; correlation of reflections with global d18O stack (left). Buried pockmarks are correlated with to d18O maxima from glacial-stage sealevel lowstands. Inset: Seafloor morphology.
c) Example of large seafloor depressions.
d) Example of largest seafloor depressions with ~10 km diameters. Three such features have been identified so far. R: Escarpment that seems to be linked to the large features

We hypothesise that gas release from dissociating hydrates during glacial-interglacial cycles led to the formation of the seafloor depressions. The shallow pockmarks, currently in water depths of 500-700 m, may have formed during glacial-stage lowstands by the movement of the seafloor out of the gas hydrate stability zone as sealevel fell and, possibly, bottom-water warmed. Similar events may have occurred around past lowstands causing buried pockmarks. The two types of SDs between 800 and 1100 m may be a result of gas hydrate dissociation at the BGHS. The size differences between the SDs hint at two different formation mechanisms. Their formation may be linked to changes in pressure-temperature conditions from both sealevel lowering and bottom-water warming that led to different rates of upward movement of the BGHS.

The project is part of an international initiative involving IfM-Geomar (Germany), Lawrence Berkeley National Laboratory (USA), and the U.S. Naval Research Laboratory. We have recently concluded two research voyages, using the University of Otago’s R/V Polaris II (February 2012) and the German R/V Sonne (January-February 2013).

R/V Polaris II voyage 12PL015 (10-25 February 2012) focussed on 12PL015 collecting multi-beam and Boomer data in the vicinity of the smallest types of pockmarks as well as initial sediment coring.

The Chatham Rise Methane Pockmarks (CHRIMP) survey on the German R/V Sonne (9 January – 1 March 2013) led by Geomar, University of Kiel, focussed on acquiring 2-D and 3-D seismic data across the pockmarks, followed by sediment coring.

The following studies are now being conducted:

  • Analysis of seismic data to study formation mechanisms of pockmarks, investigate their link to gas hydrates, and study deeper structures that may have affected pockmark formation
  • Geochemical analyses to seek for evidence of past venting
  • Paleoceanographic analyses to reconstruct past oceanographic conditions.
  • Oceanographic modelling to investigate past and possible future oceanographic conditions
  • Modelling of the response of the gas hydrate system in the study area to past and possible future oceanographic conditions, including constraints on the amount of released methane

Our studies should provide key results for understanding the causes and implications of the release of methane from the vast oceanic gas-hydrate reservoir in response to climatic changes.


  • Davy, B., I. A. Pecher, R. Wood, L. Carter, and K. Gohl (2010), Gas Escape Features off New Zealand – Evidence for a Massive Release of Methane from Hydrates?, Geophys. Res. Lett., 37, L21309.
  • Judd, A., and M. Hovland (2007), Seabed fluid flow, 475 pp., Cambridge University Press, Cambridge (UK).
  • Kennett, J. P., K. G. Cannariato, I. L. Hendy, and R. J. Behl (2000), Carbon Isotopic Evidence for Methane Hydrate Instability During Quaternary Interstadials, Science, 288, 128-133.
  • Kvenvolden, K. A. (1993), Gas hydrates - geologic perspective and global change, Rev. Geophys., 31, 173-187.
  • Solomon, E. A., M. Kastner, I. R. MacDonald, and I. Leifer (2009), Considerable methane fluxes to the atmosphere from hydrocarbon seeps in the Gulf of Mexico, Nature Geoscience, 2(8), 561-565.