Australia: The Land Where Time Began
Climate - The Atlantic Ocean
According to the authors1 there is a tendency for climate research on the Atlantic to be focused on decadal and longer term variability centred on the processes of deep water formation in the North Atlantic and on the processes of sea ice in the Nordic Seas and the Arctic. The reason for this is that the mean ventilation age of the deep waters in the northern North Atlantic is on the order of decades or less with the associated variability that is measureable. It has been documented that at all scales, from interannual to millennial, variability of climate affects all regions of the Atlantic. Also documented are trends that have been related to climate change, such as anthropogenic forcing.
Atlantic - Tropical variability
Interannual variability studies are focused on the tropical Atlantic, a region where there are several modes, among which are 2 that are intrinsic to the Atlantic Ocean, the Atlantic Meridional Mode (AMM), the Atlantic Niņo and remote forcing from the ENSO in the Pacific Ocean. The AMM, a cross-equatorial mode, and the Atlantic Niņo, a zonal equatorial mode that is dynamically similar to ENSO with a tropical Bjerknes feedback, are both modes intrinsic to the Atlantic Ocean. None of these modes are overwhelmingly dominant as the ENSO is in the Pacific Ocean. Upper ocean variability is linked to these modes. There may be other sources and timescales for variability at intermediate and abyssal depths. The PIRATA array, designed to sample the meridional and zonal modes (Bourlès et al., 2008) regularly monitors the variability in the tropical Atlantic.
The AMM has Sea Surface Temperature (SST) anomalies of opposite sign on opposite sides of the equator - warm SST to the north and cold SST to the south and vice versa. The AMM is also known as the "Tropical Dipole Anomaly" as a result of these opposing anomalies. During positive AMM the Intertropical Convergence Zone (ITCZ) is located to the north of the equator, is displaced towards the north. Alternating highs and lows from the Nordic Seas and the Southern Ocean are included in the full Atlantic hemispheric pattern, though its amplitude reaches it highest point in the tropics, while the North Atlantic Oscillation (NAO), with a spatial pattern resembling that of the AMM, has its highest amplitude in the north. The AMM has a seasonal cycle that peaks in the boreal spring, and variability that is interannual to decadal. Decadal variation in the AMM has been described as a wind-evaporation-SST feedback (Chang et al., 1997; Kushnir et al., 2002) though the feedback is weak (Sutton el al., 2000; Chiang & Vimont, 2004). It appears that to maintain decadal energy, external forcing, such as from the North Atlantic Oscillation (NAO) or the ENSO in the Pacific is necessary.
The Atlantic Niņo, aka the "Atlantic zonal equatorial mode" has typical Bjerknes tropical feedback between the SST of the ocean and the winds of the atmosphere. The Atlantic Niņo timescale is interannual, on the order of 30 months, though there is a considerable amount of randomness. In the central and eastern Atlantic a cold tongue appears in every boreal summer, in the normal seasonal cycle. A large area of the equatorial Atlantic is occupied by this seasonal cold tongue, the coldest temperatures being less than 24oC, which is comparable to the temperature of the seasonal cold tongue in the Pacific. In the Atlantic the warm pool is about 28oC, and is cooler and is more limited spatially than the warm pool, more than 30oC, in the warm pool of the Pacific. The cold tongue is almost obliterated by the warm SST anomalies during the Atlantic Niņo, which is accompanied by a shift to the east and a weakening of the Walker circulation in the Atlantic, in the central Atlantic rising air over the maximum anomaly and strengthening of the Hadley circulation (Wang, 2002).
The Atlantic Niņo has a lower amplitude and a geographically a smaller impact than the ENSO in the Pacific. The authors1 suggest that the simplest answer is that the Atlantic m is much narrower than the Pacific, therefore in the east thermocline variation of depth and the associated SST anomalies are weaker (Jin, 1996). As the Atlantic mean western warm pool is much narrower than that in the Pacific, anomalies in the Atlantic are also weaker than in the Pacific. The ENSO of the Pacific reaches eastward in to the tropical Atlantic (Wang, 2002). The Walker circulation in the Pacific shifts eastward during an El Niņo warm event, with air that is ascending moving to the central and eastern equatorial Pacific. The descending branch of this anomalous Walker circulation is in the central Atlantic with strongest effects on SST in the tropical North Atlantic. In the tropical Atlantic SST anomalies lag an El Niņo warm event in the Pacific by 5-6 months.
Variability - decadal and multidecadal
It is common to interpret North Atlantic variability in terms of the North Atlantic Oscillation (NAO) and East Atlantic Pattern (EAP), internal atmospheric modes, on short timescales that have important decadal and longer term variability that might involve feedbacks with the ocean. The NAO is closely related with the AO (NAM). Decadal climate variability in the south is associated with the SAM. A longer timescale of the natural mode of the Atlantic overturning circulation that is associated with surface temperature throughout the North Atlantic is represented by the Atlantic Multidecadal Oscillation (AMO).
The NAO is one of the most vigorous, as well as best described natural climate modes of the Earth. In the mean, the pressure difference between the subtropical (Bermuda) high and subpolar (Iceland) low in the lower atmosphere force the westerly winds of the North Atlantic. The location and strength of the westerly winds change when the pressure systems change strength or shift. The traditional NAO index is the pressure difference between Portugal and Iceland, though other indices are also used. The pressure difference is large and the westerlies are shifted to the north, relative to their mean position, that is, with maximum strength between Iceland and Portugal, and vice versa, when the NAO is positive. The NAO variability is roughly decadal and includes seasonal to multidecadal timescales. From the 1970s to 19990s a high NAO with strong westerlies, and a cold subpolar gyre, and warm Nordic Sea and Gulf Stream region dominated. From the 1950s to the 1960s a low NAO dominated.
The circulation of the North Atlantic and the production and properties of water masses are affected by shifts in the NAO. Associated with the NAO, the Gulf Stream and its separation point move slightly, though measurably, to the north, with increased transport, lagging the NAO by several years (Curry & McCartney, 2001; Visbeck et al., 2003). Also during high NAO, there is a shift to the west and intensification of the subpolar gyre circulation to the north of about 50oN (Flatau et al., 2003; Häkkinen & Rhines, 2004). The properties of the North Atlantic are highly variable from top to bottom because it forms intermediate and deep water. Labrador Sea Water (LSW), Greenland Sea Deep Water (GSDW) and Eighteen Degree Water (EDW) all vary with the NAO (Dickson et al., 1996). When the subpolar gyre and the Labrador Sea are cold, during positive NAO, the production of LSW is strong and anomalously cool. During high NAO the Greenland Sea is warmer, and there is a weakening of the production of Greenland Sea Deep Water is weakened and warmer. During periods of high NAO production of EDW is also weaker, reaching a point in the mid-1970s when it almost ceased and in the 1990s there was a shift to lower densities (Dickson et al., 1996; Talley, 1996b).
In the northern North Atlantic decadal variability is also associated with the East Atlantic Pattern (EAP) (Barnston & Livezey, 1986; Josey & Marsh, 2005). Increased precipitation associated with increasing EAP appears to be linked to decades long freshening of the subpolar gyre. The EAO and NAO are independent. In the Atlantic the EAP is the second empirical orthogonal function (EOF) of climate variability, and the NAO can be defined as the first EOF. The EAP has a zero crossing about 35oN that is further to the south than the NAO, and a symmetric shape about the equator. For the Atlantic, it appears to be the lowest order symmetric (sine-like) meridional mode.
As the Atlantic has longer term variability, a potential relationship to variability in the MOC is of interest. The Atlantic SST is used to quantify variability at longer than decadal timescales, the Atlantic Multidecadal Oscillation (AMO) or "Atlantic Multidecadal Variability" is an index of the Atlantic SST. The AMO anomaly index is the average SST anomaly of the entire North Atlantic, 0-70oN, that is detrended and the application of a 10-year running mean (Enfield, Mestas-Nuñez & Trimble, 2001). The North Atlantic as a whole is warm when the index is positive, while the South Atlantic is cool, therefore it is an interhemispheric mode. On the NOAA ESRL (2009b) web site the monthly values of the index since 1856 are available, listed as an updated Kaplan SST product (Kaplan et al., 1998). The timescale of the AMO is 65-80 years with a range of several 0.1oC. The SST record has only 2 cycles, though longer palaeoclimate records also show an AMO (Delworth & Mann, 2,000). The AMO can also be reproduced with coupled ocean-atmosphere models in which meridional overturning circulation changes are included.
The "bipolar seesaw", an interhemispheric mode, has been introduced to explain timescales that are much longer, such as millennial variability in palaeoclimate records that occurred during the Younger Dryas interval at the end of the last glaciation (Broecker, 1998). Included in these records are signals in the far Northern and far Southern Hemispheres that are out of phase with each other. A climate mode, such as the AMO, that has north-south structure, could be used to explain these. There would be enhanced transport to the north of heat into the subpolar gyre and Nordic Seas, with a strong MOC, and the SST would then be higher there. Also, warm water from the Southern Hemisphere would be moved to the Northern Hemisphere by northward cross-equatorial flow of warm water. There is also a multidecadal timescale in the NAO (Delworth & Mann, 2000; Visbeck, 2002), as is shown by the use of a 10-year running mean of its index. There is also decadal variability in the EAP that resembles somewhat the AMO index after about 1970. The EAP has been associated with the Great Salinity Anomalies.
Atlantic Ocean - Variability of properties
Changes of the MOC in the Atlantic, natural or anthropogenic, reflect the climate of the Earth and have the potential to affect it (Vellinga & Wood, 2002). Coordinated programs have been in place since the late 1990s to monitor overflows of the Nordic Seas, the Labrador Sea, the Strait of Gibraltar and the meridional overturn in the North Atlantic at several latitudes with most resources across 24oN. The larger context for the changes are being provided by broad-scale observations, and its variations are widely monitored.
In this section the authors1 focused on the salinity of the surface water because it provides a control on the depth and density of the mixed layer. North Atlantic subpolar Sea Surface Salinity (SSS) have been significant over the past century. After a fresh period that was centred around 1910, at 60oN salinity was relatively high until the 1970s, following which there was a decline of salinity that remained low until about 2000, the continuing freshening being of great interest because it had the potential to slow the MOC in the North Atlantic (Curry et al., 2003; Dickson et al., 2003). The salinity trend reversed after 2000, with the salinity increasing throughout the subpolar gyre. This has now joined the trend of increasing salinity over the past 50 years in the remainder of the Atlantic.
When decadal timescales are observed within the long-term salinity records there are clear freshwater pulses that occurred in the 1970s, 1980s, and 1990s. These, including the event of 1910, have been called Great Salinity Anomalies (GSAs; Dickson, Meinke, Malmberg & Lee, 1988; Belkin, 2004). Around the northern North Atlantic and the Nordic Seas coherent patterns that are time-lagged are formed by these GSAs of low salinity. The GSAs of the 1970s emerged from the Fram Strait into the East Greenland Current in 1968 (Dickson et al., 1988). The freshwater pulse appeared to move down around Greenland, into the Labrador Sea, then further on into the North Atlantic Current, and into the subpolar gyre, and about 10 years later returned to the Nordic Seas. In the 1980s and the 1990s low salinity anomalies occurred, that had similar propagation patterns, both events entering the Labrador Sea from the Canadian Archipelago where they originated. In the northern Labrador Sea the GSAs are related closely the extent of sea ice in Davis Strait, the northern entrance from the Arctic into the Labrador Sea (Deser, Reverdin & Timlin, 2002).
It has been hypothesised that GSAs arise, at least in part, as a response to adjustments of circulation and its fronts, as it is believed to be not likely that anomalies of the magnitude of GSAs could advect all the way around the cyclonic circulation with little amplitude changes for up to 10 years (Sundby & Drinkwater, 2007). When upper ocean salinity anomalies arrive, regardless of the mechanism involved, SST patterns are altered, convection is inhibited, and feedbacks into the climate mode may occur. (Zhang & Vallis, 2006).
In the northern North Atlantic large-scale changes of salinity have been observed at depth. (e.g., in the Labrador Sea Water). During a period of low NAO in the 1960s, the Labrador Sea was warm and saline at all depths. The formation of the LSW was weak and relatively warm and saline. The properties of the LSW had shifted to be fresher, cooler and denser (high NAO), by the 1990s. Throughout the northern part of the North Atlantic freshening occurred at intermediate depths (Dickson et al., 2002). The production of LSW was again interrupted in the later 1990s, with its temperature and salinity beginning to increase (declining NAO; Yashayaev, 2007; Schott, Stramma, Giese & Zantopp, 2009). The spatial structure of salinity didn't change, though in the 1990s salinity had shifted to a fresher range: lowest salinity in the Labrador Sea, and tongues of low salinity extended into the Irminger sea, Iceland Basin and Rockall Trough, and a weak tongue extended to the south around Newfoundland. The overall circulation pattern was therefore mostly preserved.
Throughout the northern North Atlantic and the southern Nordic Seas the freshening of deep waters through the 1990s was also accompanied by overflow of the Nordic Seas (Dickson et al., 2002; Dickson et al.,2003). Changes up stream in the Nordic Seas and the changes in the upper ocean and intermediate waters, that were previously mentioned, as the overflows plunge towards the ocean floor. In terms of velocity, transport and temperature the overflows have been remarkably steady, mainly because of hydraulic control in the straits, which is governed by the large reservoir upstream in the Nordic Seas (Girton, Pratt, Southerland & Price, 2006). Transports, bottom velocities and bottom temperature in the Faroe Bank Channel held steady at 2.1 Sv, >100 cm/sec, and -0.4oC for the period 1995-2005 which was observed directly, and observations from 1948, indirect evidence for similar stability (Olsen, Hansen, Quadfasel & Østerhus, 2008). More variability of overflow velocities, transports and temperatures have been found in Denmark Strait than at Faroe Bank by 4 years of monitoring (Macrander et al., 2005). There was a range of 3.1-3.7 Sv for transports and there was a variation of temperatures by 0.5oC, the higher transports corresponding roughly with colder water. As in Faroe Bank Channel hydraulic control is important, though there are other dynamical processes, including northward flow in the eastern Denmark Strait, that is driven by the wind, that has the ability to modulate the overflow transport.
In the North Atlantic the variable properties in the deep and intermediate waters move to the south into the subtropics and tropics, mostly by way of the Western Boundary Current. It has been documented that there are changes of properties, such as oxygen, at the Grand Banks (~43oN) at Abaco (26.5oN), as well as all the way to the equator with time lags that are appropriate, of 2-10 years from changes at the subpolar sources (Molinari et al., 1998; Stramma et al., 2004 & Bryden et al., 2005b). Variations in DWBC transport, which can be associated with MOC changes, he been difficult to document from sparse decadal hydrographic sampling as large seasonal variability is aliased to longer timescales. As monitoring is now continuous at ~25oN, including the Florida Current and Ekman Transport that is basin-wide, as well as the DWBC, the prognosis for monitoring the interannual variations in the total overturn is good to within about 10% of the total overturn (Cunningham et al., 2007).
|Author: M.H.Monroe Email: firstname.lastname@example.org Sources & Further reading|