Abrupt changes - To what extent are tipping points a concern in coping with global change? [Past]
Dorthe Dahl-Jensen
Niels Bohr Institute, University of Copenhagen, Denmark; ddj@gfy.ku.dk
 |
|
Figure 1: High-resolution records from the Greenland ice core NGRIP. Panel A shows stable water isotopes (δ18O) in 20-year resolution. Numbers mark the most prominent of the 25 Dansgaard-Oeschger events. Panels B and C zoom in on two 300-year intervals during the transition from the last glacial to the Holocene. Shown are records of δ18O, deuterium excess and the unsolvable dust at 1-year resolution. The solid lines highlight the transitions in the records; the vertical yellow lines mark the steps in deuterium excess over just a few years. (Figure based on Steffensen et al. 2008). |
Paleoclimatic records can provide information on the operation of the climate system, including the occurrence of tipping points and the risk of abrupt changes. The deep Greenland and Antarctic ice cores are particularly well suited to study abrupt changes, because they provide a detailed and well-dated record of past climate. Prominent examples of abrupt changes are the 25 Dansgaard-Oeschger (DO) events (NGRIP 2004) that occurred during the last glacial cycle (Fig. 1).
The DO events are characterized by abrupt warming followed by a gradual cooling. The isotopic composition of the nitrogen (N2) in air bubbles trapped in Greenland ice and the stable water isotopes of oxygen (18O) and hydrogen (deuterium, D) of the ice itself show that the abrupt warmings represent surface temperature changes in the order of 10-15°C (Landais et al. 2005). Annually dated ice core sections covering the two most recent DO events reveal the actual rapidity of the changes. Some proxies, like the deuterium excess (d=δD-8*δ18O), changed level over just a few years (Steffensen et al. 2008). The deuterium excess reflects the temperature at the moisture uptake region for the precipitation. Its step-like changes in Greenland ice cores suggest that the atmospheric circulation regime shifted substantially and irreversibly basically from one year to the next (Masson-Delmotte et al. 2005). Following the atmospheric regime shift, temperatures over Greenland warmed more gradually over some decades by 10-15°C, as shown by the δ18O record (Steffensen et al. 2008). These observations prove that the climate system did, and therefore can, tip and reorganize internally within years and cause strong and fast regional temperature changes.
How and why did the abrupt climate changes happen? Studies from all latitudes based on ice cores from Polar Regions, marine sediments, stalagmites, corals and other paleoclimatic archives allow us to piece together a broader picture of the DO events and to deduce a sequence of causes and effects. During the cold phases preceding the abrupt warmings, vast volumes of ice were discharged into the ocean from the large glacial ice sheets including the North American Laurentide ice sheet, causing sea level to rise by several tens of meters (Siddall et al. 2003). The overturning circulation and associated northward heat transport in the Atlantic slowed down. This warmed the South and cooled the northern polar region further and may have resulted in a southward shift of the Intertropical Convergence Zone (ITCZ; Partin et al. 2007).
What caused the abrupt warmings? This is less well understood and requires investigation of the (very sparse) near-annually dated records. The studies from the Greenland ice cores suggest that the sudden rearrangement of the northern atmospheric circulation might have been initiated by a sudden shift of the ITCZ in the low latitudes. A sudden decrease of the dust concentration in the ice indicates that the wetness of the source area for the dust (related to the position of the ITCZ) had shifted (Steffensen et al. 2008). Perhaps the warming of the south finally pushed the ITCZ north again?
Can such tipping points of temperature and sea level change happen in the coming decades and centuries? The DO events during the last glacial seemed to be initiated by surges from big glacial ice sheets. Such large Ice sheets are not present nowadays, but other triggers that could cause the system to tip are plausible. Increased precipitation and melting of ice sheets and glaciers could increase the fresh water supply to the Arctic and the North Atlantic Ocean and alter the intensity of the ocean circulation. This would tip the energy distribution between the North and South in a similar way as happened during the glacial DO events. Rapid mass loss of the West Antarctic Ice Sheet, of which major parts lie more than 1 km below the present sea level, could cause an abrupt sea level rise of several meters.
State-of-the-art simulations with complex Earth System models do not project abrupt climate changes for this century. However, based on our understanding of the past DO events we conclude that abrupt changes of temperature and sea level cannot be ruled out entirely for our future world.
selected references
Full reference list online under: http://pastglobalchanges.org/products/newsletters/ref2012_1.pdf
Partin JW et al. (2007) Nature 449: 452-455
Landais AA, Jouzel J, Masson-Delmotte V and Caillon N (2005) Comptes Rendus Geosciences 337: 947-956
Masson-Delmotte V et al. (2005) Science 309: 118-121