The recession of mountain glaciers was used in IPCC (1990) to provide qualitative support to the rise in global temperatures since the late 19th century. Work on glacier recession has considerable potential to support or qualify the instrumental record of temperature change and to cast further light on regional or worldwide temperature changes before the instrumental era. Two types of data from glaciers contain climatic information: (i) mass balance observations and (ii) data on the geometry of glaciers, notably glacier length. More comprehensive information is now becoming available and worldwide glacier inventories have been updated (e.g., IAHS (ICSI)/UNEP/UNESCO, 1999). Note that changes in the Greenland and Antarctic ice sheets are discussed in Chapter 11.
We first discuss mass balance observations. The specific mass balance is defined as the net annual gain or loss of mass at the glacier surface, per unit area of the surface. The mass balance averaged over an entire glacier is denoted by Bm. Systematic investigations of glacier mass balance started after 1945, so these records are shorter than the instrumental climate records normally available in the vicinity. In contrast to frequently made statements, Bm is not necessarily a more precise indicator of climate change than is glacier length. Time-series of Bm do contain year-to-year variability reflecting short-term fluctuations in meteorological quantities but of concern on longer time-scales is the effect of changing glacier geometry. A steadily retreating glacier will get thinner and the mass balance will become more negative because of a slowly increasing surface air temperature due to a lowering surface that is not reflected in a large-scale temperature signal. Climatic interpretation of long-term trends in of mass balance data requires the use of coupled mass balance-ice flow models to separate the climatic and geometric parts of the signal. Such studies have only just begun. However, mass balance observations are needed for estimating the contribution of glacier melt to sea level rise, so are discussed further in Chapter 11.
A wealth of information exists on the geometry of valley glaciers. Glacier records are very useful for studies of Holocene climate variability (e.g., Haeberli et al., 1998; and Section 2.4). Written documents going back to the 16th century exist that describe catastrophic floods caused by the bursting of glacier-dammed lakes or arable land and farms destroyed by advancing glaciers, e.g., in 18th century Norway (Østrem et al., 1977). A large amount of information is available from sketches, etchings, paintings and old photographs of glaciers, though many show the same glaciers (Holzhauser and Zumbühl, 1996). About fifty glaciers have two or more useful pictures from distinctly different times. In many cases geomorphologic evidence in the form of terminal moraines and trimlines can be used as reliable complementary information to construct the history of a glacier over the last few centuries. Systematic mapping of glaciers started only 100 years ago and has been limited to a few glaciers. The most comprehensive data are of length variations. Glacier length records complement the instrumental meteorological record because (i) some extend further back in time; (ii) some records are from remote regions where few meteorological observations exist; (iii) on average, glaciers exist at a significantly higher altitude than meteorological stations.
Figure 2.18: A collection of twenty glacier length records from different parts of the world. Curves have been translated along the vertical axis to make them fit in one frame. The geographical distribution of the data is also shown, though a single triangle may represent more than one glacier. Data are from the World Glacier Monitoring Service (http://www.geo.unizh.ch/wgms/) with some additions from various unpublished sources.
The last point is of particular interest in the light of the discrepancy between recent tropical glacier length reduction and lack of warming in the lower troposphere since 1979 indicated by satellites and radiosondes in the tropics (Section 2.2.3). Long-term monitoring of glacier extent provides abundant evidence that tropical glaciers are receding at an increasing rate in all tropical mountain areas. This applies to the tropical Andes (Brecher and Thompson, 1993; Hastenrath and Ames, 1995; Ames, 1998), Mount Kenya and Kilimanjaro (Hastenrath and Kruss, 1992; Hastenrath and Greischar, 1997) and to the glaciers in Irian Jaya (Peterson and Peterson, 1994).
Relating mass balance fluctuations to meteorological conditions is more complicated for tropical glaciers than for mid- and high latitude glaciers, and it has not been demonstrated that temperature is the most important factor. Nevertheless, the fast glacier recession in the tropics seems at first sight to be consistent with an increase in tropical freezing heights of 100 m over the period 1970 to 1986 as reported by Diaz and Graham (1996), corresponding to an increase of 0.5°C at tropical high mountain levels, which they also link to increases in tropical SST since the mid-1970s (Figure 2.10). However, although Gaffen et al. (2000) found a similar increase over 1960 to 1997, they found a lowering of freezing level over 1979 to 1997 which, at least superficially, is not consistent with glacier recession.
Figure 2.18 shows a representative selection of glacier length records from different parts of the world and updates the diagram in IPCC (1990). It is clear from Figure 2.18 that glacier retreat on the century time-scale is worldwide. The available data suggest that this retreat generally started later at high latitudes but in low and mid-latitudes the retreat generally started in the mid-19th century.
On the global scale, air temperature is considered by most glaciologists to be the most important factor reflecting glacier retreat. This is based on calculations with mass balance models (Greuell and Oerlemans, 1987; Oerlemans, 1992; Fleming et al., 1997; Jóhannesson, 1997). For a typical mid-latitude glacier, a 30% decrease in cloudiness or a 25% decrease in precipitation would have the same effect as a 1°C temperature rise. Such changes in cloudiness or precipitation can occur locally or even regionally on a decadal time-scale associated with changes in circulation, but global trends of this size on a century time-scale are very unlikely. As mentioned in the SAR, Oerlemans (1994) concluded that a warming rate of 0.66 ± 0.20°C per century at the mean glacier altitude could explain the linear part of the observed retreat of 48 widely distributed glaciers.
Glaciers are generally not in equilibrium with the prevailing climatic conditions and a more refined analysis should deal with the different response times of glaciers which involves modelling (Oerlemans et al., 1998). It will take some time before a large number of glaciers are modelled. Nevertheless, work done so far indicates that the response times of glacier lengths shown in Figure 2.18 are in the 10 to 70 year range. Therefore the timing of the onset of glacier retreat implies that a significant global warming is likely to have started not later than the mid-19th century. This conflicts with the Jones et al. (2001) global land instrumental temperature data (Figure 2.1), and the combined hemispheric and global land and marine data (Figure 2.7), where clear warming is not seen until the beginning of the 20th century. This conclusion also conflicts with some (but not all) of the palaeo-temperature reconstructions in Figure 2.21, Section 2.3 , where clear warming, e.g., in the Mann et al. (1999) Northern Hemisphere series, starts at about the same time as in the Jones et al. (2001) data. These discrepancies are currently unexplained.For the last two to three decades, far more records have been available than are shown in Figure 2.18. Many are documented at the World Glacier Monitoring Service in Zürich, Switzerland (e.g., IAHS (ICSI)/UNEP/UNESCO, 1998) The general picture is one of widespread retreat, notably in Alaska, Franz-Josef Land, Asia, the Alps, Indonesia and Africa, and tropical and sub-tropical regions of South America. In a few regions a considerable number of glaciers are currently advancing (e.g., Western Norway, New Zealand). In Norway this is very likely to be due to increases in precipitation owing to the positive phase of the North Atlantic Oscillation (Section 2.6), and in the Southern Alps of New Zealandand due to wetter conditions with little warming since about 1980. Finally, indications in the European Alps that current glacier recession is reaching levels not seen for perhaps a few thousand years comes from the exposure of radiocarbon-dated ancient remains in high glacial saddles. Here there is no significant ice flow and melting is assumed to have taken place in situ for the first time in millennia (e.g., the finding of the 5,000-year-old Oetzal "ice man").
Other reports in this collection