Two recent studies, one in Peru and the other in Norway, link glacier retreat, not to climate change as many researchers have done, but to climate variability—the fluctuations in temperature and precipitation across large regions of the world, on time scales of years or decades. These studies add an important level of detail to the role of glacier science in building awareness of climate change. On the one hand, glaciers around the world are shrinking, and rising temperatures, due to the growth of greenhouse gas emissions, are the principal cause of this decline. On the other hand, glaciers do not respond uniformly and homogenously to greenhouse gas concentration and to global mean temperatures; instead, their dynamics are more varied. Indeed, one of the studies shows that some glaciers have periods of growth lasting several years, even though longer-term research indicates that they are shrinking when they are examined on a time scale of decades. The other study shows that even in an area of steady glacier retreat, there can be months with slight growth, though there are no years of net glacier growth.
These two studies are striking, because they examine glaciers which are located in different continents, at different latitudes, and in proximity to different oceans. Moreover, they use different methods, indicating the variety of techniques in glaciology. However, they both point to the influence of major patterns of climate variability on glacier dynamics.
In the study of a Peruvian glacier, published recently in The Cryosphere, Fabien Maussion and his coauthors, all at the University of Innsbruck in Austria, used detailed weather data and glacier mass balance to study the Shallap Glacier in the Cordillera Blanca, a region known to have a strong influence of El Niño. For their study period of 2006-2009, they obtained monthly data on the glacier mass balance, on the energy balance at the surface (including incoming and outgoing shortwave and longwave radiation and heat fluxes) and on a number of meteorological variables (temperature, precipitation, cloud cover, relative humidity, air pressure, wind speed and wind direction). They assessed the mass balance through the use of ablation stakes. They linked these variables to each other and to the El Niño/Southern Oscillation (ENSO), a major form of climate variability in the tropical Pacific. To assess the state of ENSO, they used a standard measure, the anomalies of the sea surface temperatures in a region of the western tropical Pacific, close to the Cordillera Blanca. (This ocean variabilility is the El Niño component of ENSO; there is also an atmospheric component, the Southern Oscillation, which is tied to the difference in atmospheric pressure between Darwin, Australia, where pressure is typically low, and Tahiti, where it is usually high. Since the ocean and atmospheric components of the variability in the tropical Pacific are highly correlated, the researchers used only the ocean component.)
As shown in the figure above, during the drier, warmer El Niño periods, the Shallap Glacier lost mass more rapidly. In the moister, cooler La Niña periods, the loss was slower, and there were a few months of gain, though there were no years of net growth.
In the study of Norwegian glaciers, also published recently in The Cryosphere, Mathias Trachsel and Atle Nesje of the University of Bergen used statistical methods to link three datasets. The first consists of mass balance data for eight Norwegian glaciers, some closer to the ocean and others further inland. The mass balance data begins between 1946 and 1970, depending on the specific glacier, and extends to 2010. It includes both seasonal mass balance (winter, summer) and total annual mass balance. The second is the weather data for the glaciers for the period for which mass balance data is available. The third is climate data for two major forms of climate variability, the North Atlantic Oscillation (NAO) and the Atlantic Multidecadal Oscillation (AMO). The NAO is an atmospheric phenomenon which reflects the variation of two major weather systems, the high-pressure system centered over the Azores and the Icelandic low-pressure system. The NAO is positive when the two systems are both relatively strong, and negative when they are weak. A positive NAO strengthens westerly winds and brings winter storms to northern and central Europe; a negative NAO sends the storms further south. The AMO reflects the variability of the temperature of the surface waters in the North Atlantic Ocean. It is positive when these waters are warmer than normal, and negative when they are cooler. The AMO is associated with precipitation variability across parts of North America, Europe and Africa. These atmospheric and oceanic systems in the North Atlantic are less tightly linked that El Niño and the Southern Oscillation in the tropical Pacific, so these researchers considered them separately.
The researchers found that variability in winter precipitation had a strong influence on the mass balance of maritime glaciers, where annual temperature variability is relatively low, because of the proximity to the ocean. For the continental glaciers further inland, variability in summer temperature had a stronger influence, with greater loss during the warmer summers. These precipitation and temperature variations are associated with the NAO and AMO.
The AMO was negative for a long period, between 1963 and 1996, and this was a period of cooler than normal summers. The NAO was positive, bringing mild, wet winters, for a portion of this time, between 1987 and 1995. This period in the late 1980s and early 1990s was a time of positive mass balance, so the glaciers grew in size—a difference from Peru, where the glaciers shrank every year during the study period, though their rate of change also varied from year to year.
This association is important, because it shows that a general trend towards glacier retreat, caused by global warming, can be masked for a time by climate variability. None of the eight glaciers in the study have grown since the mid-1990s (six have shrunk, and two are the same). However, four of them grew considerably in the period of favorable conditions, so they are larger than they were at the start of the research. The figure in the paper, included here, shows that two had significant net growth and remain steady, while two have lost much of their growth in this period; four of them have grown smaller, though the pace of retreat has varied.
All of these glaciers are highly sensitive to climate variability, since they are relatively warm (in Peru because of the tropical location, in Norway because of the moderating influence of the Gulf Stream); in colder glaciers, net accumulation of snow is relatively slow (because cold air does not hold as much water vapor that can form into precipitation) and net ablation or loss of ice is also slow. The steepness of the mountain ranges in both areas may also contribute to the sensitivity to climate variability, because the ice that forms during a few years of favorable conditions will flow downslope more quickly, where it will be exposed to warmer conditions, where it will melt; glaciers in areas of gentler slopes take longer to respond, so their processes of growth will smooth out the variability from year to year or month or month—much as a small swimming pool will grow warmer or cooler as weather changes at a more noticeable rate than a large lake.
Nonetheless, these studies show the importance of including climate variability as well as climate change in the study of glacier dynamics. Climate change skeptics were quick to pounce on a short period of glacier growth in Norway to challenge the overall global patterns of glacier retreat. Though there have been fewer claims in the last five years of such growth, these studies show the importance of offering more detailed, nuanced accounts of glacier processes. Such information is also of importance to water managers and to local communities.