HR: 08:15h
AN: PP21E-02 INVITED     [Abstracts]
TI: A Signature of Persistent Natural Thermohaline Circulation Cycles in Observed Climate
AU: * Knight, J
EM: jeff.knight@metoffice.gov.uk
AF: Hadley Centre, FitzRoy Road, Exeter, EX1 3PB United Kingdom
AU: Allan, R
EM: rob.allan@metoffice.gov.uk
AF: Hadley Centre, FitzRoy Road, Exeter, EX1 3PB United Kingdom
AU: Folland, C
EM: chris.folland@metoffice.gov.uk
AF: Hadley Centre, FitzRoy Road, Exeter, EX1 3PB United Kingdom
AU: Vellinga, M
EM: michael.veliinga@metoffice.gov.uk
AF: Hadley Centre, FitzRoy Road, Exeter, EX1 3PB United Kingdom
AU: Mann, M
EM: mann@meteo.psu.edu
AF: Dept. of Meteorology, Pennsylvania State University, 503 Walker Building, University Park, PA 16802-5013
AB: An ensemble of simulations of 20th Century climate using the HadCM3 coupled climate model forced with estimates of natural and anthropogenic forcings is compared with instrumental temperature data sets. The results show that while external climate forcing can account for most of 20th century climate change, there are also significant multidecadal climate fluctuations that are not produced by forcings. A major part of this variability corresponds to the `Atlantic Multidecadal Oscillation' (AMO), which has been identified in observations, and represents coherent fluctuations in temperature throughout most of the Northern Hemisphere. Using a 1400 year calculation with HadCM3 without external forcings, we show the model produces a quasi-periodic internal mode with a pattern, amplitude and characteristic time scale similar to that of the observed AMO. Further, the model implies the AMO is a genuine long-lived quasi-cyclical climate phenomenon related to large-scale oceanic heat transport variations associated with changes in the strength of the thermohaline circulation (THC) of about 2 Sv (10%). In the simulation, stronger cross-equatorial temperature gradients are associated with the anomalous northward ocean heat transport during a warm AMO phase. This causes a northward displacement of the mean ITCZ, leading to more rainfall and the development of anomalously fresh water in the tropical North Atlantic. These sustained anomalies slowly propagate to the subpolar North Atlantic in about 5 decades, where they act to slow the THC. The results also confirm observed links between the AMO and multidecadal variability in north-east Brazil and Sahel precipitation, and Atlantic hurricane formation. In addition, the simulated link between temperature and the THC allows an estimate of possible past changes in THC strength. Our results imply that the THC has undergone distinct strong and weak phases in the 20th century and has strengthened over recent decades. We also produce a forecast of the natural component of future THC change that shows a likely decline in the next 35 years to levels similar to the lowest levels reconstructed in the 20th century. This would accelerate anthropogenic THC weakening and the associated change in the AMO would offset Northern Hemisphere warming.
DE: 1616 Climate variability (1635, 3305, 3309, 4215, 4513)
DE: 1626 Global climate models (3337, 4928)
DE: 3339 Ocean/atmosphere interactions (0312, 4504)
DE: 4513 Decadal ocean variability (1616, 1635, 3305, 4215)
DE: 9325 Atlantic Ocean
SC: Paleoceanography and Paleoclimatology [PP]
MN: Fall Meeting 2005