Climate change may cause responses in many human and natural systems, influencing human health (disease outbreaks, heat/cold stress), agriculture (yield, pest outbreaks, crop timing), physical systems (glaciers, icepack, streamflow), and biological systems (distributions/abundances of species, timing of events). In intensely human-managed systems, the direct effects of climate change may be either buffered or so completely confounded with other factors that they become impossible to detect. Conversely, in systems with little human manipulation, the effects of climate change are most transparent. Systems for which we have a good process-based understanding of the effects of climate and weather events, and have had minimal human intervention, may act as indicators for the more general effects of climate change in systems and sectors where they are less readily studied.
The cryosphere is very sensitive to climate change because of its proximity to melting. Consequently, the size, extent, and position of margins of various elements of the cryosphere (sea ice, river and lake ice, snow cover, glaciers, ice cores, permafrost) are frequently used to indicate past climates and can serve as indicators of current climate change (Bradley and Jones, 1992; Fitzharris, 1996; Everett and Fitzharris, 1998). In particular, former glacier extent is indicative of past glacials and the Little Ice Age. At high latitudes and high altitudes, ice cores have provided high-resolution annual (and, in some cases, seasonal) records of past precipitation, temperatures, and atmospheric composition. These records stretch back for many hundreds of years, well before the instrumental period, so they have proven to be very valuable in documenting past climates. Borehole measurements provide data on permafrost warming. Later freeze-up and earlier breakup of river and lake ice is measurable at high latitudes.
Interpretation of climate change resulting from changes in the cryosphere is seldom simple. For example, in the case of glaciers, glacier dynamics and extent are influenced by numerous factors other than climate. Different response times are observed for the same climate forcing, so some glaciers can be in retreat while others are advancing. Changes in glacier size can be caused by changes in temperature or in precipitationor even a nonlinear combination of both. Similarly, changes in sea ice can be a result of changes in ice dynamics (winds, currents) as much as thermodynamics (temperature). Thus, attribution of the exact nature of climate change from changes in the cryosphere is quite complicated.
With many measures of the cryosphere, there frequently is large interannual variability. This makes determination of possible anthropogenic climate trends difficult to distinguish from the natural noise of the data. Another problem is that high-resolution records usually are not available, except from polar or high-altitude ice cores. Changes in the extent of sea ice and seasonal snow are best observed with satellites, but such records are relatively short (from about 1970), so long-term climate change is difficult to distinguish from short-term, natural variability. The first records of cryospheric extent and changes often come from documentary sources such as old diaries, logbooks of ships, company records, and chronicles (Bradley and Jones, 1992). Although these sources are fraught with difficulty of interpretation, they clearly demonstrate climate changes such as the medieval warm period and the Little Ice Age (see Section 5.7, Chapter 16, and Section 19.2).
Other reports in this collection