HR: 0830h
AN: GC31B-0182    [PDF]
TI: Detecting Thermohaline Circulation Changes from Ocean properties
AU: * Hu, A
EM: ahu@ucar.edu
AF: National Center for Atmospheric Research, 1850 Table Mesa Dr., Boulder, CO 80305 United States
AU: Meehl, G A
EM: meehl@ucar.edu
AF: National Center for Atmospheric Research, 1850 Table Mesa Dr., Boulder, CO 80305 United States
AU: Han, W
EM: whan@colorado.edu
AF: University of Colorado, Campus Box 311, Boulder, CO 80309 United States
AB: Significant changes of the thermohaline circulation (THC) are likely to cause abrupt earth climate change. Here we examined several mechanisms controlling the THC variation using a coupled climate system model (NCAR's CCSM2.0). Three experiments are analyzed, a control run (present; CON), and two forced runs: a fresh water hosing run with 0.1 Sv additional fresh water uniformly distributed in northern North Atlantic (NA) between 50 and 70$^{\circ}$N (past; HOS), and a transient climate run with 1% CO$_{2}$ increase per year (future; TRC). In HOS, THC is weakened to 11.4 Sv from 15.7 Sv in CON by the end of the 100-year hosing, then recovers to its full strength 80 years after the hosing was turned off. The tropical and subtropical Atlantic ocean average over the upper 1000 meter (T$_{at1000}$) warms by 0.3$^{\circ}$C by the end of the hosing, implying a change of 14 Sv in THC for a 1$^{\circ}$C variation of T$_{at1000}$. In TRC, the THC weakens to 14.1 Sv at 2XCO$_{2}$ and to 12.6 Sv at 4XCO$_{2}$. T$_{at1000}$ increases by 1.1$^{\circ}$C at 4XCO$_{2}$, where most of the T$_{at1000}$ changes can be attributed to the CO$_{2}$ effect. Therefore, the same regression relationship between THC and T$_{at1000}$ do not hold in this experiment as in HOS. EOF analysis of the meridional streamfunction (MSF) in the Atlantic reveals a seesawing pattern of the rate of North Atlantic Deep Water (NADW) and Antarctic Bottom Water (AABW) formation in the first EOF for HOS. When the NADW increases (stronger THC), the AABW decreases. The same is true for CON. Changes in AABW is related to sea ice volume changes in southern ocean. This seesawing pattern of THC agrees with proposed THC oscillation in the past 1500 years. In TRC, as CO$_{2}$ increases, the northern NA and the periphery of the Antarctic become warmer and fresher, reducing the surface water density, and resulting in a decrease in both NADW and AABW formation. Other mechanisms examined here are the mean sea surface salinity (SSS) and temperature (SST) contrast between NA and North Pacific (NP), and the steric height gradient (SHG) between 30$^{\circ}$S and 60$^{\circ}$N in Atlantic. In HOS and CON, it shows a higher SSS contrast related to a stronger THC, but opposite in TRC. However, a colder NP (warmer NA) is related to a stronger THC for both forced runs. The SHG in Atlantic gives the most consistent result among these 3 runs, however, the linear regression shows a 17 Sv change in THC vs a change of one cm/deg-lat in SHG for CON and HOS, but a number increased to 29 for TRC. EOF analyses of the global SST indicate that the first EOF in CON, explaining 14% of the total variance, is a ENSO related pattern with a 2.6-year frequency. In the forced runs, this pattern becomes the second EOF with a frequency of 3.3- to 4.5-year, explaining 11% and 3% of the total variance for HOS and TRC, respectively. The first EOFs are a general cooling in Northern Hemisphere and warming in Southern Hemisphere in HOS (explaining 12% of the variance) and a global warming in TRC (explaining 70% of the variance). The general conclusion is that the proposed mechanisms used to detecting the THC strength are held for past and current climate condition, but not perfectly held for the future (at least in NCAR's CCSM2.0). The increase in atmospheric CO$_{2}$ level seems changed the behavior of the THC, and causes a breakdown of many teleconnections between THC and others.
DE: 1635 Oceans (4203)
DE: 4215 Climate and interannual variability (3309)
SC: Global Climate Change [GC]
MN: 2003 Fall Meeting