environmentalresearchweb blog

« Maui as a microcosm of modern water and energy pressures | Main | Who wants nuclear power? (Part 2) »

Glaciers and the Atlantic Multidecadal Oscillation

In a newly-published paper, Matthias Huss and colleagues squeeze a bit more information out of the well-studied records of mass balance from the Swiss Alps. We know more about alpine glaciers than those of any other region, except perhaps Scandinavia, so it is surprising and interesting to learn that there is yet more to be said.

Huss and his co-authors compile a surprising amount of quantitative information. As well as conventional mass-balance measurements made over the past 60-odd years, they have found scattered records, and made geodetic measurements of surface elevation changes from maps, back to 1910, and have also brought measurements of meltwater discharge to bear on the problem.

They do an impressive job of tying these diverse kinds of information together with a model of glacier responses to climatic forcing. They simulate the climate, day by day, with precipitation and temperature data from weather stations extrapolated to the elevations at which the glacier ice is found. Their model of the response to temperature, for example, is a so-called temperature-index model. There is abundant evidence that, when it gets hotter, the glaciers yield more meltwater, and we can use temperature to predict meltwater in a reproducible way.

Their main contribution, however, is the light thrown on the evolution of glacier mass balance in the Alps over recent, and maybe near-future, decades. The authors show that there is a clear signal of the Atlantic Multidecadal Oscillation embedded in the 20th-century history of alpine mass balance.

I have trouble keeping up with the acronyms in this field. ENSO (El Niño-Southern Oscillation) and NAO (North Atlantic Oscillation) are easy because they have become familiar, but the AMO was new to me. This oscillation of sea surface temperature in the North Atlantic has a period, or quasi-period, of 60-70 years, and like all of the others (except perhaps for ENSO) it is an empirical fact rather than an understood physical phenomenon. That is, we can see it in the climatic records, but we have no explanation of why it is there. Is it real? Good question.

I don’t think there is any question, though, that we can now see the AMO in the response of the glaciers of the Swiss Alps. As the authors point out, this signal may have predictive value. Large as the year-to-year variability is, we know that the glaciers of the Alps have been suffering particularly badly in recent years. The link to the AMO suggests that things may not be quite so bad in the next couple of decades. Of course we have to assume that the AMO is a phenomenon that is “real” enough that it will evolve as its past course suggests, but the fact is that we don’t know enough about it yet.

On the other hand, there are some ideas out there. The A for “Atlantic” might be an important clue, for example, suggesting a link with the oceanic meridional overturning circulation, in which the shallow northward flow in the North Atlantic sinks at high latitude and returns southward at depth — taking dissolved atmospheric gases with it.

Wang and Dong offer an intriguing new twist on this angle. They show that the AMO pattern can be seen in the atmospheric concentration of carbon dioxide, measured at Hawaii and the South Pole since 1958. You have to subtract both a noticeable seasonal cycle (the vegetated landmasses take up CO₂ in northern summer and release it in northern winter) and a long-term trend (the more affluent and numerous we get, the more gas we emit). What remains is a definite low in CO₂, reaching about 2 parts per million below the 1958-2008 average, between the late 1960s and the mid-1990s. This matches in time persuasively with a cool phase of the AMO.

We are still far from a physical explanation of why the North Atlantic climate oscillates. Nobody knows why the oscillation takes about 60-70 years. We can’t even be sure which way the arrow of causality is pointing. Is the CO₂ low because the ocean surface is cooler and better able to take it up? Or is the ocean cooler because there is less CO₂? In fact, the arrow of causality could well be chasing its own tail. But these links between ocean temperature, greenhouse gas and the behaviour of glaciers are worth exploring because of the prospect that we might be able to turn guesswork about the future into predictability.

Posted by Graham Cogley on June 14, 2010 3:00 PM | Permalink