Large, abrupt, and widespread climate changes with major impacts have occurred repeatedly in the past, when the Earth system was forced across thresholds. Although abrupt climate changes can occur for many reasons, it is conceivable that human forcing of climate change is increasing the probability of large, abrupt events. Were such an event to recur, the economic and ecological impacts could be large and potentially serious. Unpredictability exhibited near climate thresholds in simple models shows that some uncertainty will always be associated with projections. In light of these uncertainties, policy-makers should consider expanding research into abrupt climate change, improving monitoring systems, and taking actions designed to enhance the adaptability and resilience of ecosystems and economies.

Uncertainty about sea-level rise is dominated by uncertainty about iceberg calving, mass loss from glaciers or ice sheets by fracturing. Review of the rapidly growing calving literature leads to a few overarching hypotheses. Almost all calving occurs near or just downglacier of a location where ice flows into an environment more favorable for calving, so the calving rate is controlled primarily by flow to the ice margin rather than by fracturing. Calving can be classified into five regimes, which tend to be persistent, predictable, and insensitive to small perturbations in flow velocity, ice characteristics, or environmental forcing; these regimes can be studied instrumentally. Sufficiently large perturbations may cause sometimes-rapid transitions between regimes or between calving and noncalving behavior, during which fracturing may control the rate of calving. Regime transitions underlie the largest uncertainties in sea-level rise projections, but with few, important exceptions, have not been observed instrumentally. This is especially true of the most important regime transitions for sea-level rise. Process-based models informed by studies of ongoing calving, and assimilation of deep-time paleoclimatic data, may help reduce uncertainties about regime transitions. Failure to include calving accurately in predictive models could lead to large underestimates of warming-induced sea-level rise. ▪ Iceberg calving, the breakage of ice from glaciers and ice sheets, affects sea level and many other environmental issues. ▪ Modern rates of iceberg calving usually are controlled by the rate of ice flow past restraining points, not by the brittle calving processes. ▪ Calving can be classified into five regimes, which are persistent, predictable, and insensitive to small perturbations. ▪ Transitions between calving regimes are especially important and with warming might cause faster sea-level rise than generally projected. Expected final online publication date for the Annual Review of Earth and Planetary Sciences, Volume 51 is May 2023. Please see http://www.annualreviews.org/page/journal/pubdates for revised estimates.

Climate change is now well recognized as a potentially significant factor in the human future, affecting ecological systems, agriculture, health, and settlement patterns. The scientific consensus, periodically assessed by the Intergovernmental Panel on Climate Change, foresees a long‐term atmospheric warming of a few degrees over a century. The accumulating knowledge of the world's climate history, however, suggests that any implied stability of trend may be illusory. The evidence comes from ice and seabed cores and tree rings, which can yield year‐by‐year information on atmospheric conditions and time‐series of mean temperatures extending back for millennia. These data show many episodes of a sudden rise or fall in temperature in particular regions, sometimes of 10°C or more in ten years, with the new mean lasting for decades or centuries. There is thus the likelihood—even the inevitability—of comparably large and abrupt changes occurring in coming years, a very different prospect from the “greenhouse warming” scenario projected by current large‐scale climate models. Moreover, the two phenomena may be connected: a continued slow temperature rise may at some point trigger an abrupt shift in climate regime through mechanisms such as the effect on ocean circulation. In 2002 a committee set up by the US National Research Council reviewed what is known about this subject in a report, Abrupt Climate Change: Inevitable Surprises (Washington, DC: National Academy Press). The Council is the operating arm of the US National Academy of Sciences; the Committee on Abrupt Climate Change comprised mostly oceanographers and climate experts and was chaired by Richard B. Alley, Professor of Geosciences, Pennsylvania State University, a glaciologist and author of The Two‐Mile Time Machine: Ice Cores, Abrupt Climate Change, and Our Future (Princeton, 2000). A recent spate of media attention to the subject has been occasioned less by the NRC report itself than by a dramatic scenario exercise derived from it, prepared under US Defense Department auspices, and by release of a disaster film with an instant‐ice‐age theme. The Pentagon study, An Abrupt Climate Change Scenario and Its Implications for United States National Security, authored by Peter Schwartz and Doug Randall, was issued in October 2003. It explores the hypothetical geopolitical consequences of a repetition in the near future of an event experienced 8,200 years ago (as recorded in an ice core from central Greenland): a sudden cooling of some 2–3° C, lasting for a century, punctuating climatic conditions broadly similar to those of the present day. In the scenario, the collapse of ocean heat‐conveying currents causes rapid cooling in the northern hemisphere and warming in the southern hemisphere. Outcomes include resource wars, large population movements, and “a significant drop in the human carrying capacity of the Earth's environment.” In February 2004 the NRC reasserted the more measured voices of its committee by issuing a four‐page summary of the 2002 report, under t...