4.8 Synthesis
Observations show a consistent picture of surface warming and reduction in all components of the cryosphere (FAQ 4.1, Figure 1), except antarctic sea ice, which exhibits a small positive but insignificant trend since 1978 (Figure 4.23).
Since IPCC (2001) the cryosphere has undergone significant changes, such as the substantial retreat of arctic sea ice, especially in summer; the continued shrinking of mountain glaciers; the decrease in the extent of snow cover and seasonally frozen ground, particularly in spring; the earlier breakup of river and lake ice; and widespread thinning of antarctic ice shelves along the Amundsen Sea coast, indicating increased basal melting due to increased ocean heat fluxes in the cavities below the ice shelves. An additional new feature is the increasingly visible fast dynamic response of ice shelves, for example, the dramatic breakup of the Larsen B Ice Shelf in 2002, and the acceleration of tributary glaciers and ice streams, with possible consequences for the adjacent part of the ice sheets.
One difficulty with using cryospheric quantities as indicators of climate change is the sparse historical database. Although ‘extent’ of ice (sea ice and glacier margins for example) has been observed for a long time at a few locations, the ‘amount’ of ice (thickness or depth) is difficult to measure. Therefore, reconstructions of past mass balance are often not possible.
The most important cryospheric contributions to sea level variations (see Chapter 5) arise from changes in the ice on land (e.g., glaciers, ice caps and ice sheets). In IPCC (2001), the contribution of glaciers and ice caps to sea level rise during the 20th century was estimated as 0.2 to 0.4 mm yr–1 (of 1 to 2 mm yr–1 total sea level rise). New results presented here indicate that all glaciers contributed about 0.50 ± 0.18 mm yr–1 during 1961 to 2003, increasing to 0.77 ± 0.22 mm yr–1 from 1993 to 2003 (interpolation from five-year analyses in Table 4.4). Estimates for both ice sheets combined give a contribution ranging from –0.35 to +0.72 mm yr–1 for 1961 to 2003, increasing to 0 to 0.8 mm yr–1 for 1993 to 2003. A conservative error estimate in terms of summing ranges is given in Table 4.6. Assuming a midpoint mean, interpreting the range as uncertainty and using Gaussian error summation of estimates for glaciers and both ice sheets suggests that the total ice contribution to sea level rise was approximately 0.7 ± 0.5 mm yr–1 during 1961 to 2003 and 1.2 ± 0.4 mm yr–1 during 1993 to 2003.
The large uncertainties reflect the difficulties in estimating the global ice mass and its variability, because global monitoring of ice thickness is impossible (even the total area of glaciers is not exactly known) and extrapolation from local measurements is therefore necessary. A regional extension of the monitored ice masses and an improvement of measurement and extrapolation techniques are urgently required.
In spite of the large uncertainties, the data that are available portray a rather consistent picture of a cryosphere in decline over the 20th century, increasingly so during 1993 to 2003.