Decadal Variability in the Southern Midlatitudes of the Pacific and Indian Oceans
J. S. Godfrey (CSIRO Marine Research. GPO Box 1538, Hobart 7001, Australia)
email: godfrey@marine.csiro.au
ABSTRACT
This talk reviews a variety of studies of interdecadal variability in the mid-latitude Pacific and Indian Oceans. Several authors have examined deep hydrographic data in the region, collected roughly 20 years apart; water mass properties have changed on several sections much as expected from global warming. Other authors have examined the more plentiful thermal (BT) data in the southwestern Pacific from recent decades. Their results indicate that the inflow and outflow to the East Australian Current, generally about 25 Sv, changed by 10 Sv or more from the late 1970's to the late 1980's. Some modes of variability correlate with El Nino. SST has risen particularly sharply over the period 1900-1983, in the southern Indian Ocean; COADS data have been analysed for the Indian and Pacific Oceans, and have been used to model SST and circulation anomalies in an effort to understand the origins of the large SST rise. An interdecadal estimate of the Indonesian Throughflow was obtained from sea level records.
INTRODUCTION
Not surprisingly, in view of the relative paucity of data, studies of interdecadal variability in the mid-latitude South Pacific and Indian Oceans are relatively rare compared to similar studies in the northern hemisphere. Nevertheless, significant work has been done. The aim of this talk is to briefly review progress to date. The work can be divided into four categories: (i) Comparisons of hydrographic sections, separated by a few decades. (ii) Use of expendable and mechanical bathythermograph data. (iii) Model studies of multidecadal temperature change in the southern Indian Ocean. (iv) Interdecadal variation in coastal sea level. In the following, we discuss each of these in turn.
Comparisons of hydrographic sections, separated by a few decades:
Bindoff and Church (1992) and Bindoff and McDougall (1993) have examined two pairs of deep hydrographic sections across the Tasman Sea, 22 years apart; the pairs were at 43｡? and 28｡? respectively. They found that, by examining changes in temperature and salinity data on neutral density surfaces, they were able to bypass noise problems associated with transient eddies. They thus found clear, statistically significant evidence that water has cooled and freshened on neutral surfaces, particularly for the Mode Waters. They explained how "pure warming" and/or "pure frshening" of surface waters is expected to lead to such cooling and freshening along neutral surfaces. Following on from their work, Johnson and Orsi (1997) examined a pair of meridional hydrographic sections along 170｡?, from the equator to 60｡?. Similar cooling and freshening on neutral density surfaces, indicative of warming and/or freshening between about 1968/69 and 1990/91 at the outcrop location, was found in Mode and Intermediate Waters south of the centre of the Subtropical Gyre. They were also able to combine more scattered hydrographic station data collected in the late 1960's with data in the early 1990's to generalise their results to the whole South Pacific. After allowing for the different slope of the T-S curve in the southeast Pacific, their results were consistent everywhere with a hypothesis of warming and/or freshening in outcrop regions of Mode and Intermediate Waters. Wong et al. (1998) have extended this style of analysis still further, by comparing WOCE results with scattered historical data, mostly from the 1960's.
