Ruth Curry, Principal Investigator
In the late 1980 's and 1990's, a major shift of deep water production from the Nordic Seas to the Labrador Basin precipitated significant changes in the character of water masses feeding the deep western boundary current (DWBC ) of the North Atlantic . In particular, the subpolar Labrador Sea Water (LSW) properties became colder, fresher, and denser than at any time in the previous fifty years . Also distinctive were the depth of convective penetration, its vertical homogenity , and degree of ventilation with oxygen and other atmospheric tracers. Through its signature shift in temperature -salinity (8-S) and elevated CFC concentrations, the LSW anomaly was first observed to enter the tropical DWBC in 1996 in a decade-long time series of measurements at Abaco (near 26°}. In February 2000, observations revealed a surprisingly strong presence of the LSW anomaly all along the deep western boundary current (DWBC) between 18°N and 13°N. This project was undertaken as a rapid response to those measurements -its purpose was to assess the strength and equatorward progress of this anomaly. Taking advantage of a fortuitous transit of R/V Knorr (from Recife, Brazil to Norfolk , VA in May -June 2000) the leading edge of the water mass was located at 10°N in the DWBC. These measurements provided a rare opportunity to directly determine the speed with which a high latitude climate signal moves through the deep ocean and the degree to which that signal is adulterated in its passage. Such information is key for understanding how Earth's climate system responds to decadal fluctuations such as the North Atlantic Oscillation . The phenomena involved are poorly represented in present climate models, yet are likely to be an important factor determining the predictive skill of such models.
The total costs of the observations were split between NSF and NOAA -NSF supplying two extra days of ship time, and NOAA and NSF each providing half of the funds to
acquire the measurements . Afour-person team accomplished the field work: five short CTD sections (5 stations each} across the DWBC were acquired at the Equator , 6°N, 8°N, 11 °N, and at Puerto Rico (66°W). Water samples were drawn and analyzed for calibration of the salinity and oxygen sensors. Additional samples were frozen at sea, and subsequently analyzed for silicate content by Joe Jennings at Oregon State University. Preliminary reports were submitted to the State Department and all countries granting political clearances for this work.
Comparison of the measurements acquired in 2000 with those from the 1980s and early 1990s has revealed the spatial and temporal development of the tropical LSW signal. A prominent shift in B-S properties at Abaco (26°N) is characterized with values fresher by >0.02 in the DWBC profiles of 1998 compared to the previous two decades. The bend in 8-S curves occurs in the 3.2 - 4.0°C potential temperature range, corresponding to the depth range 1600 - 2300 m. The equatorward progression of that 8-S signature -with similar unmistakable shifts in the 2000 DWBC measurements -was documented at multiple sites along the boundary: at Puerto Rico (18.5°N, 66°W), Barbuda (18°N, 61 °W), Guadeloupe (15°N), Barbados (13°N), and Trinidad (11 °N). The freshening is more pronounced in the upstream direction and slightly weaker downstream towards Trinidad. The signal has not arrived in any appreciable strength equatorward of 10°N: for example in measurements taken at 8°N, 6°N, and in the equatorial channel near 36°W. The horizontal distribution of the 8-S signal is similar in all the post-1997 measurements upstream of 10°N: it is ubiquitous in the DWBC with an abrupt transition to more saline 8-S seaward of the high shear regime. The recent 8-S values offshore from the DWBC revert to a curve nearly identical to the 1980's and early 1990's observations. The interpretation is quite clear . We are witnessing the incursion of the LSW transient into the tropical deep circulation and it is transiting appreciable distances in the DWBC without being mixed away by eddies or diffusion into the adjacent interior . An estimate of the effec tive spreading rate from Abaco to 11 °N, a distance of about 2850 km along the boundary in 4 years, is approximately 2.3 cros - lfaster by a factor of two compared to the effective spreading rates estimated by tracer studies for the upper deep water circulation of the previous two decades. Moreover , the tropical expression of this thermohaline anomaly involves more than a simple shift in water mass characteristics. There have been simultaneous alterations of the DWBC vertical density structure, potential VortlClty, and baroclinic velocity distributions that suggest dynamical implications for the deep circulation.
Funding for a more complete analysis of these data was awarded to the PI this year by NSF. Geostrophic calculations of velocity and transport will be compared with previous measurements acquired during STAGS and WOCE programs. The interpretation of these data may suggest a field measurement program to acquire longer term averages of the velocity field associated with the LSW transient, and/or modeling studies to explore its dynamical consequences. At its present rate of advance, the LSW transient will arrive at the equator circa 2003. Its time history, recorded in repeated hydrography sections upstream near Abaco and Puerto Rico, suggests that isopycnal layer thickening and intensification of its velocity core will develop within three years of its initial incursion. An equatorial potential vorticity anomaly is an interesting prospect because of the potential to incite planetary waves that could perturb the upper ocean dynamics. This would represent an oceanic teleconnection for high-latitude changes to affect the MOC's warm limb. Monitoring this event will provide an opportunity to assess potential linkages between high latitudes and the tropics in the context of climatically relevant (i.e. MOC and NAO) phenomena.