IPCC Fourth Assessment Report: Climate Change 2007
Climate Change 2007: Working Group I: The Physical Science Basis What Is Known About the Mechanisms of Transitions Into Ice Ages?

Successful simulation of glacial inception has been a key target for models simulating climate change. The Milankovitch theory proposes that ice ages were triggered by reduced summer insolation at high latitudes in the NH, enabling winter snowfall to persist all year and accumulate to build NH glacial ice sheets (Box 6.1). Continental ice sheet growth and associated sea level lowering took place at about 116 ka (Waelbroeck et al., 2002) when the summer incoming solar radiation in the NH at high latitudes reached minimum values. The inception took place while the continental ice volume was minimal and stable, and low and mid-latitudes of the North Atlantic continuously warm (Cortijo et al., 1999; Goni et al., 1999; McManus et al., 2002; Risebrobakken et al., 2005). When forced with orbital insolation changes, atmosphere-only models failed in the past to find the proper magnitude of response to allow for perennial snow cover. Models and data now show that shifts in the northern treeline, expansion of sea ice at high latitudes and warmer low-latitude oceans as a source of moisture for the ice sheets provide feedbacks that amplify the local insolation forcing over the high-latitude continents and allow for growth of ice sheets (Pons et al., 1992; Cortijo et al., 1999; Goni et al., 1999; Crucifix and Loutre, 2002; McManus et al., 2002; Jackson and Broccoli, 2003; Khodri et al., 2003; Meissner et al., 2003; Vettoretti and Peltier, 2003; Khodri et al., 2005; Risebrobakken et al., 2005). The rapid growth of ice sheets after inception is captured by EMICs that include models for continental ice, with increased Atlantic Meridional Overturning Circulation (MOC) allowing for increased snowfall. Increasing ice sheet altitude and extent is also important, although the ice volume-equivalent drop in sea level found in data records (Waelbroeck et al., 2002; Cutler et al., 2003) is not well reproduced in some EMIC simulations (Wang and Mysak, 2002; Kageyama et al., 2004; Calov et al., 2005). When Will the Current Interglacial End?

There is no evidence of mechanisms that could mitigate the current global warming by a natural cooling trend. Only a strong reduction in summer insolation at high northern latitudes, along with associated feedbacks, can end the current interglacial. Given that current low orbital eccentricity will persist over the next tens of thousand years, the effects of precession are minimised, and extremely cold northern summer orbital configurations like that of the last glacial initiation at 116 ka will not take place for at least 30 kyr (Box 6.1). Under a natural CO2 regime (i.e., with the global temperature-CO2 correlation continuing as in the Vostok and EPICA Dome C ice cores), the next glacial period would not be expected to start within the next 30 kyr (Loutre and Berger, 2000; Berger and Loutre, 2002; EPICA Community Members, 2004). Sustained high atmospheric greenhouse gas concentrations, comparable to a mid-range CO2 stabilisation scenario, may lead to a complete melting of the Greenland Ice Sheet (Church et al., 2001) and further delay the onset of the next glacial period (Loutre and Berger, 2000; Archer and Ganopolski, 2005).