Multi-Model Evaluation of the Global Marine Methane Hydrate Inventory

Lead Research Organisation: University of Leeds
Department Name: School of Earth and Environment


Methane is trapped in seafloor sediments of many of the world's continental margins in the form of hydrate - an ice-like solid that efficiently stores methane. Hydrates represent an immense carbon store, estimated to be comparable to conventional oil and gas reserves. As waters near the sea-floor warms, the thermal conditions that permit hydrate to exist may deteriorate. This may lead to the breakup of hydrate within the sediment with the subsequent release of methane. Methane is a powerful greenhouse gas, its release into the atmosphere leads to further warming - this could lead to a positive feedback loop in which further hydrate is disassociated.

Hydrate also acts to cement and strengthen sediment. As hydrate breaks-down, the surrounding sediment may weaken. This can lead to a structural failure of the sediment column and potentially causing a submarine landslide. The geological record contains examples where hydrate disassociation has been implicated in causing large-scale sediment failure and submarine landslides. One of these is the Storegga slide off the Norwegian coast, when around 8,000 years ago a submarine landslide occurred, triggerring a tsunami that reached as far as northern England. Many areas around the Arctic and Atlantic regions remain potentially susceptible to slope failure. The hazard associated with associated tsunamis (i.e. catastrophic flooding) motivates a desire to improve our knowledge of the hydrate inventory.

The CMIP5 project is designed to compare the worlds leading state-of-the-art climate models in their ability to model the present climate and make predictions of future climate change. CMIP5 therefore provides a unique opportunity to develope our understanding of the current and future conditions at the sea floor. Using this insight, this project aims to create improved reconstructions of the current hydrate inventory and model how this may change under future climate change.

Planned Impact

International scientific impact will be achieved by :

- the publication of a high-ranking journal paper "CMIP5 multi-model evaluation of the current and future evolution of the marine hydrate inventory" in time for inclusion and consideration within IPCC AR5. Manuscript and 1 page summary of critical aspects of the work will be sent to lead author of IPCC AR5 Working group I at earliest opportunity.

- To address earth science community present oral and poster presentations at the Annual Geophysical Union December 2011 and the European Geophysical Union April 2012. Both conferences have wide media coverage (including international news agencies). Conferences will also provide means to arrange project meetings with interested partners and parties (Bruce Buffett and David Archer).

To increase project visibility and impact we will work in close co-operation with METHANE.NET. A brief summary of our work (with a focus the transient simulation of the inventory through climate change scenarios) will be written for Fire in the Ice, a hydrates R&D program newsletter.

We will communicate the significance of our research and key findings to Dr Dan Bernie of the Met Office, who works on understanding dangerous climate change and provides advice to government on its implications. He is a key part of the AVOID programme, and provides scientific advice to inform government mitigation policy.


10 25 50
Description 12 climate models contributed to the research. Using the modelled historical record (i.e. 1860 to 2005), climate models were evaluated against the WOA05 historical record of bottom water conditions, performance weights were generated so that a series of RCP and ECP multi-model mean datasets could be generated.

We modelled the propagation of bottom water thermal anomalies through the sediment column from 1860 to +5000 yrs and computed the volume of the hydrate stability zone throughout the continental margins where hydrates are expected to reside. This was the first time that this had been conducted for a CMIP multi-model ensemble for any AR-5 future scenario over the global domain.

We integrated an envelope of linear sea-level models into the hydrate stability zone (HSZ) calculation to produce a series of predictions of how the global HSZ volume (gHSZv) will respond to anthropogenic warming and sea-level change. This was the first time that sea level had been quantitatively assessed in a coupled manner. Previously, it had been generally assumed that sea level would not be significant driver; here we show quantitatively its contribution to future gHSZv reduction and hydrate dissociation.

We generate a series of estimates of the global hydrate inventory using a 1D hydrate model and the CMIP5 model pre-industrial bottom water conditions (as well as a multi-model mean).

Both the gHSZv and global hydrate inventory compare favourably with previous estimates, confirming that our work sits alongside and is compatible with established research.

These global inventories were then convolved with the prediction of future HSZ change to produce a series of hindcast simulations of the evolution of global hydrate volume. We carry this out for each climate model, for each RCP and ECP scenario, and for each sea- level model. We collate these datasets along with knowledge of model performance (i.e. using the model performance weights) to generate multi-model mean responses for each R/ECP scenario (and sea level model).

We also used a simpler hydrate model that had been used in previous research, which assumed constant sediment pore-space fill, which we convolved with the HSZ data, to produce a second series of multi-model mean hydrate volume temporal-evolution datasets.

From these multi model means we derive decadal scale predictions of the global mean hydrate dissociation rate (one for each scenario, considering fixed sea level) and investigate the temporal evolution of hydrate dissociation in terms of overlying water depth, sediment depth and latitudinal distribution. Our results correlate well with other studies including the recent Arctic-only study of Biastoch which consider the response to the AR-4 2× CO2 1% per year experiment in terms of dissociation rates and the geographiphical distribution of 2000-2100 Arctic dissociation. We are able to expand on this study by being able to consider the full temporal response (at a decadal level) from 2000 to 2300 for RCP 4.5, 6.0 and 8.5 future scenarios. A subsequent paper will present the global study in terms of temporal-geographical distribution of hydrate dissociation.

We demonstrated how predictions of hydrate dissociation are particularly sensitive to the modelled vertical distribution of hydrate within the sediment column. This potentially sheds light on an ambiguity within published research in which submarine hydrate are seen as either a purely geological-timescale issue (models predict deep-seated hydrates i.e. hydrate concentrated at great depth within the sediment) or a more immediate concern (i.e. Models predict methane hydrate distributed closer to the upper part of the sediment column).

Within RCP and ECP timescales (i.e. 2100 and 2300) we were able to predict peak global dissociation rates and we demonstrate within the submitted paper that previous estimates of global sea-floor methane fluxes arising from AR-4 2× CO2 1% per yr scenarios may be more overestimated than previously thought.

We have not modelled methane hydrates associated with subsea permafrost (this was not within our initial project plan) due to complex physics and uncertain boundary conditions. Nevertheless, we have generated multi-model mean datasets of the evolution of Arctic (and global) bottom water temperatures on a monthly, seasonal, yearly and decadal (global) timescale, so that this can be studied at a future time, when adequate models have been developed.
Exploitation Route Policy makers (UK-gov i.e. Deepwater and Arctic oil drilling note SNSC-5981, all international gov via IPCC AR5 WG 1) and regulators (HSE Horizon scanning intelligence group Policy makers (UK-gov i.e. Deepwater and Arctic oil drilling note SNSC-5981, all international gov via IPCC AR5 WG 1) and regulators (HSE Horizon scanning intelligence group, British Geological Survey);

Public Sector and other Nongovernmental organisations (British Geological Survey+ other similar international institutions i.e. USGS)

Academic Beneficiaries -(The following UK and international communities: Earth System Modelling, Geoscience, carbon capture and storage community, ocean drilling)
Sectors Energy,Environment