Supplementary material to “Rapid Climate Change and the Role of the Ocean Circulation”
Meric Srokosz, National Oceanography Centre, Southampton, UK; Eric Wolff, British Antarctic Survey, Cambridge, UK; Valborg Byfield, National Oceanography Centre, Southampton, UK
Citation:
Srokosz, M., E. Wolff, and V. Byfield (2006),
Rapid climate change and the role of the ocean circulation,
Eos Trans. AGU, 88(13), 152.
[Full Article (pdf)]
Meeting Report: Rapid Climate Change International Science Conference
24–27 October 2006, Birmingham, UK
Most scientists accept that climate change is going to happen as a result of forcing by increased concentrations of greenhouse gases. In contrast, there is much more controversy over whether rapid climate change—occurring on timescales much faster than the forcing—will be overlain on it. This is an area of active research. In particular, the role of the North Atlantic branch of the meridional overturning circulation (MOC) in possible future rapid changes is much debated. In the UK, the Natural Environment Research Council initiated a major 5 year programme called “RAPID” to investigate this issue, and RAPID has also engaged a number of international partners. The recent Rapid Climate Change International Science Conference was held to review progress and discuss the scientific challenges that exist in this area, bringing together results from paleo, modelling, and observational studies. Over 200 scientists attended. This report highlights some of the work presented at the meeting.
From a paleo perspective, it is clear that rapid changes, the Dansgaard-Oeschger events, did occur during the last glacial period. However, most speakers focussed on a more recent event, when the Earth was in an interglacial climate. The so-called 8.2 ka event—characterised by a sharp drop in temperatures—is thought to have occurred when an outburst from a North American ice-dammed lake (Lake Agassiz) flooded the North Atlantic with freshwater and dramatically slowed the MOC. The event has been considered a possible analogue for potential effects of future freshening of the North Atlantic from increased precipitation and ice melt. Mark Chapman (University of East Anglia, UK) presented results from the analysis of a high-resolution sediment core from the Gardar Drift, which showed the first clear evidence of the impact of the 8.2 ka event on North Atlantic circulation. Others presented evidence from paleo records, such as Newfoundland peat bog data (Tim Daley, University of Southampton, UK) and other archives around the North Atlantic. Taken together with other evidence, these suggest that the nature of the event is becoming clearer, but that the relationship between the timing of the various observed changes and the outburst from the lake should be investigated further in order to define the targets for climate modelling exercises more clearly. The impacts appear to be confined largely to the North Atlantic and surrounding landmasses.
Much interest was generated by the first year (2004–2005) of results from the MOC observing system deployed by the RAPID programme at 26.5°N in the Atlantic, presented by Harry Bryden (National Oceanography Centre, UK). Earlier work, based on 5 ship sections at the same latitude in 1957, 1981, 1992, 1998, and 2004, had suggested a possible slowdown of the circulation, but the observed changes could also have been due to natural variability. Results from the observing system were consistent with those from the ship section in 2004 and exhibited flow compensation; that is, over periods longer than 15 days the different components of the flow going northward and southward balance. This shows that the system is measuring the flow correctly and can be used to monitor changes. A single year of data are insufficient to draw any conclusions about possible trends in the circulation, and the aim is to sustain the observing system over a period of a least a decade (initial funding from NERC, NSF, and NOAA is for 4 years deployment).
A number of modeling studies addressed the issue of freshwater input into the North Atlantic. Thierry Fichefet (Universite Catholique de Louvain, Belgium) and Johann Jungclaus (Max Planck Institute for Meteorology, Germany) examined the effects of Greenland melting on the MOC, an effect excluded from many climate models. Both studies suggest that Greenland melting, even at accelerated rates, would not be critical to the MOC; it would slightly exacerbate the slowdown caused by the heating and freshening of the ocean due to other effects. Paul Spence (University of Victoria, Canada) presented results on the effect of increasing the spatial resolution of the ocean component of a climate model when modeling the 8.2 ka event. Contrary to expectations, the model MOC was found to be more sensitive to freshwater input at much higher spatial resolutions. If this result were reproduced in other models, it would suggest that higher resolution is needed in the ocean component of climate models in order to capture the response of the ocean correctly. Most sophisticated coupled atmosphere–ocean climate models are now predicting that a slowdown, but not a shutdown, will occur (Richard Wood, Hadley Centre, UK). Results presented indicated that a complete shutdown of the MOC could lead to a net cooling in parts of northwest Europe, but that in more realistic scenarios, predicting a slowdown superimposed on greenhouse-induced warming, net warming occurs everywhere over land. Around the North Atlantic the slowing of the MOC reduces but does not reverse the radiatively driven warming. Two papers (Daniela Jacob, Max Planck Institute for Meteorology, Germany; Tim Woollings, University of Reading, UK) examined the use of high-resolution regional models embedded in lower-resolution climate models to study the local impacts of larger-scale changes. The details of how the regional model is embedded in the global one were found to have a significant impact on the results obtained.
In a more theoretical study, Henk Dijkstra (University of Utrecht, Netherlands) used powerful mathematical techniques to study the stability structure of the MOC in an ocean general circulation model (OGCM). Simpler models, most famously the Stommel model, show that depending on the freshwater input, the ocean can be in a single or multiple equilibrium regime. In the latter case a sufficiently large input of freshwater can switch the MOC from an “on” to an “off” state. These new results show, for the first time, that such single and multiple equilibrium states can exist in a realistic OGCM. They also suggest some possible diagnostic quantities (for example, the freshwater flux across 30°S in the Atlantic) that might be used to determine whether the MOC in more complex coupled atmosphere-ocean GCMs lies in the single of multiple equilibrium regime (or, more speculatively, whether the real ocean does).
Two panel discussions were held during the conference. The first addressed the question, How have paleo studies helped us to improve models and predictions of rapid climate change? The second was, What does society want to know about rapid climate change? Participants in the first discussion noted that progress in paleo measurements presents a stimulus and challenge to the modeling community; it is still difficult to use the often complex climate proxies, but recent developments in paleo modeling such as incorporation of isotopic tracers, makes more realistic comparisons and tests feasible. In the second question it became clear that both policy makers and industry (for example, the insurance sector) were interested in more probabilistic assessment of the risks of rapid climate change, a subject addressed by several papers at the conference.
Sir David King (UK Government Chief Scientific Advisor) addressed the more general issue of the impacts and required responses to global warming. He noted that the Stern Report would make clear the economic costs of adapting to climate change in the future would far exceed the cost of immediate mitigation measures. He gave examples of how the UK could reach its target of reducing CO2 emissions by 60% through a combination of strategies such as renewables, energy efficiency, nuclear, carbon capture, and storage. He said that there was no single solution to the problem and highlighted the need for international agreement. He also noted the importance to policy makers of narrowing the range of uncertainties in climate change predictions, and the need for scientists to communicate clearly the probability of various scenarios.
The final session at the conference was focused on some of the challenges facing scientists studying rapid climate change and the role of the MOC in such change. Jochem Marotzke (Max Planck Institute for Meteorology, Germany) posed the question, what is required to predict an MOC collapse? The key elements included a sustained observing system for the MOC, taking advantage of the latest technological developments (for example, full depth ocean gliders once they are available); an (operational) decadal climate prediction system of proven skill; and from a more theoretically perspective, the determination of whether multiple equilibrium states for the MOC exist in high-resolution coupled GCMs (and in reality). These elements are both scientifically and technically challenging and demonstrate that the subject of rapid climate change will be an area of active research for some time to come.
The Rapid Climate Change International Science Conference was held on 24–27 October 2006 in Birmingham, UK (see http://www.noc.soton.ac.uk/rapid/rapid2006/ for more information and abstracts of all the paper presented). The conference was organised as part of the UK Natural Environment Research Council’s Rapid Climate Change programme (RAPID; see http://rapid.nerc.ac.uk/), in conjunction with NSF, NOAA, RCN, NWO, CLIVAR, and the Cape Farewell project.
