Revised and Resubmitted to Journal of Climate November 2005 Abstract




Скачать 371.88 Kb.
НазваниеRevised and Resubmitted to Journal of Climate November 2005 Abstract
страница5/12
Дата25.10.2012
Размер371.88 Kb.
ТипДокументы
1   2   3   4   5   6   7   8   9   ...   12

1) Tropical Ocean Processes



The pronounced seasonal variation of the tropical Atlantic Ocean circulation is, at least partially, responsible for the distinct seasonal manifestation of TAV on seasonal-to-interannual time scales. During the boreal winter/spring, entrainment cooling is weak and the averaged mixed layer depth within the deep tropics is shallow, making the mixed layer temperature highly sensitive to surface heat flux perturbations. It is during these seasons that the wind-induced latent heat flux, via either remote influences or local feedback, is most effective in producing SST anomalies in the tropical Atlantic. In addition, during the boreal spring nearly uniform warm SSTs cover the entire tropical Atlantic. This makes the underlying atmosphere highly sensitive to small cross-equatorial perturbations (Chiang et al 2002), and thus the meridional mode tends to dominate TAV during this season. Enfield et al. (1999) report a marginally significant coherence with an antisymmetric phase only during this season. During boreal summer seasonal entrainment is at its maximum, the thermocline is shallowest in the equatorial eastern Atlantic with a strong zonal gradient of SST and strong equatorial trades. This oceanic setting is highly favorable for dynamic interaction between the zonal wind and SST along the equator, and thus the zonal mode appears more dominant. The eastern tropical Pacific Ocean undergoes a similar seasonal variation, except that the equatorial thermocline variation is not nearly as strong as that in the equatorial Atlantic. Whether this difference in the subsurface seasonal variation contributes to the different seasonal phase-locking between the Pacific ENSO and the Atlantic zonal mode remains to be explored.


The evolution of the meridional mode can be described in three phases: 1) initiation phase, 2) development phase, and 3) decay phase. The initiation phase typically begins in the boreal winter during which both ENSO and NAO are at their peak. As previously mentioned, the latent heat flux change induced by the change of the northeasterly trade winds provides the major source of forcing for the northern tropical Atlantic SST. The SST response can be roughly described by a 1-D mixed layer driven by surface heat fluxes as the seasonal entrainment cooling begins to weaken and the mixed layer depth begins to shoal. The development phase usually takes place in the boreal spring when the entrainment cooling diminishes and the mixed layer depth is at its seasonal minimum.


During this phase, the thermodynamic feedback between surface heat flux and SST, such as WES, can further enhance the SST signal, causing the development of a strong anomalous cross-equatorial SST gradient. The role of subsurface ocean dynamics appears to be rather passive and does not seem to have a major impact on SST. This was first demonstrated by Carton et al. (1996). Alexander and Scott (2002) show that an atmospheric GCM coupled to a 1-D ocean mixed layer model is capable of capturing some of basic features of the observed SST evolution associated with the remote influence of ENSO during boreal spring. Chang et al. (2003) conducted ensembles of seasonal predictions using NCAR's CCM3 coupled to a slab ocean by initializing the model with observed SST in December and found that the simple coupled model was quite skillful in forecasting boreal spring SST anomalies in much of the north tropical Atlantic. Subsequent analyses of these experiments by Saravanan and Chang (2004) and Barreiro et al. (2004) show that the high skill of the model comes from the combined effect of local thermodynamic coupling and the remote influence of ENSO. Collectively, these recent studies lend strong support to the view that the formation of the meridional mode during the boreal spring results from remote atmospheric influences and regional thermodynamic feedbacks, both of which can be captured by a simple 1-D mixed layer ocean.


One exception to this is the coastal region in the vicinity of the Guinea Dome where simple mixed layer models consistently underestimate the strength of the SST variability (Alexander and Scott 2002; Barriero et al. 2004). The OGCM study of Carton et al. (1996) show that the SST variability does depend on alongshore fluctuations in wind stress via coastal upwelling. Visbeck et al. (1998) and Chang et al. (2001) show that much of the SST variability in this region is related to the NAO-induced fluctuations in the northeasterly trade winds which affect the SST through changing the surface latent heat flux and upwelling. The relative importance of the heat flux-induced and upwelling-induced SST changes has not been quantified.


Ocean dynamics are likely to play a more important role during the decay phase of the meridional mode, which occurs during the boreal summer when the southeasterly trade winds begin to regain their strength and the seasonal entrainment cooling is rapidly developing. The seasonal deepening of the mixed layer makes surface heat fluxes a less effective influence, while the shoaling thermocline in the eastern equatorial Atlantic makes the subsurface ocean processes more tightly coupled to the surface processes. Experiments with an AGCM coupled to a slab ocean indicate that without active ocean dynamics the thermodynamic feedbacks tend to exaggerate the coupled variability (e.g., Saravanan and Chang 2004; Barriero et al. 2004), suggesting that the role of the ocean in the tropical Atlantic is mainly damping, thereby providing a negative feedback to counteract the thermodynamic air-sea feedback.


There are three potential mechanisms through which this negative feedback can be achieved: 1) transport of temperature anomalies by the mean circulation (e.g. Chang et al. 1997, 2001, Seager et al. 2001) and by surface Ekman flow (Xie, 1999), 2) transport of the mean temperature gradient by circulation anomalies (e.g. Joyce et al. 2004), and 3) nonlinear heat transport by circulation anomalies acting on temperature anomalies (e.g. Jochum et al. 2004). Chang et al. (2001) present an argument emphasizing the role of the advection of anomalous SST by the mean cross-equatorial ocean transport within the context of a hybrid coupled model. By analyzing the model upper ocean heat budget, they show that while stronger (weaker) trades increase (decrease) the surface heat loss and cool (warm) the ocean, creating an anomalous cross-equatorial SST gradient, advection of the anomalous SSTs by the northward cross-equatorial mean current will always tend to weaken the anomalous SST gradient. In reality, the South Equatorial Current and the North Brazil Current carry approximately 13 Sv water across the equator (Schott et al. 2002a), which can potentially play an important role in this negative feedback process. Chang et al. (2001) further argue that the negative ocean feedback can introduce a delay in the coupled loop, which may set a time-scale for the meridional mode. Seager et al. (2001) also found the dominance of the horizontal advection of anomalous temperatures by the mean meridional currents, but reported no evidence of any obvious phase lag introduced by the oceanic advection.


Joyce et al. (2004) recently presented an analysis based on both observations and model simulations that argues for the importance of the transport of the mean temperature gradient by circulation anomalies. They show that the wind driven response of the tropical Atlantic thermocline depth is such that it opposes the cross-equatorial SST gradient. Briefly, the mechanism proposed by Joyce et al. (2004) can be described as follows: In response to a northward cross-equatorial SST gradient anomaly, a southerly wind anomaly develops in the deep tropical Atlantic. This anomalous wind veers to the east upon crossing the equator, which in turn generates a negative wind stress curl anomaly just north of the equator (Fig. 10). The curl then drives cross-equatorial Sverdrup transport in the ocean which is in the opposite direction of the meridional wind, i.e., from the warm to the cold side of the equator. Since the mean SST gradient in the tropical Atlantic is northward, the southward cross-equatorial Sverdrup flow tends to advect warm water from the anomalously warm Northern Hemisphere to the anomalously cold Southern Hemisphere, and thus damp the anomalous SST gradient. Joyce et al. (2004) further show that this negative ocean feedback can introduce a lag of approximately one year into the coupled system. Therefore, potentially it can play a role of setting a time scale for the meridional mode. One intriguing aspect of this mechanism is the potential linkage between the cross-equatorial SST gradient and ocean circulation changes. Joyce et al. (2004) show that the cross-equatorial wind variability is connected to the change in the NECC.


In a recent numerical study, Jochum et al. (2004) suggest that the nonlinear heat transport induced by Tropical Instability Waves (TIWs) may also play a role in negative feedback. It has been known for some time that the TIWs are an important part of the equatorial heat budget (e.g., Philander and Hurlin 1988, Weisberg and Weingartner 1988). Jochum et al. (2003, 2004) show that in the Atlantic the TIWs can exist from May to January and can potentially perturb the seasonal cycle of the ITCZ during boreal spring. Philander and Delecluse (1983) show that an increased southerly wind across the equator associated with a northward SST gradient leads to a strengthening of the EUC which should lead to stronger TIW activity and an increased southward heat flux, again a negative feedback.


In reality, it is likely that all the above-described mechanisms act simultaneously to weaken the cross-equatorial SST gradient. Their relative importance, however, may depend on season and location, and have not been fully explored. It remains to be seen if any of these mechanisms alone are of sufficient strength to explain the decay of the SST gradient.


The dynamic processes that govern the equatorial SST changes in the Atlantic Ocean appear to be more complex and less understood than those involved in ENSO. The observations made during SEQUAL/FOCAL clearly indicate that the anomalous warming in the summer of 1984 was proceded by weaker-than-normal trade winds which led to the usually deep thermocline depth in the eastern equatorial basin (Philander 1986). The anomalous weak trade winds caused the zonal pressure gradient along the equator to disappear in January-February, 1984 (Katz et al. 1986), which in turn drove the anomalous eastward current just south of the equator, contributing to the 1984 warm event (Hisard et al. 1986). This oceanic response resembles, in many respects, conditions in the Pacific during a warm ENSO event. Hisard et al. (1986) note that although the events such as that of 1984 are rare, similar oceanic conditions were also observed during 1963. An updated analysis of the relationship between the thermocline and SST for the Atlantic zonal mode by Vauclair and du Penhoat (2001) identifies two more warm events, 1994-95 and 1997-1998, where the equatorial warming appears to be El Niño-like, in the sense that there is a relaxation of the equatorial easterlies and a deepening of the equatorial thermocline prior to boreal summer warming. But perhaps more interestingly, the study shows the relationship in the Atlantic is more fragile and statistically less significant than in the Pacific.


Not all the warm events follow the El Niño-like relationship. Indeed, the overall correlation between the SST and thermocline depth anomalies has a maximum value of 0.4 in the eastern equatorial cold tongue region, which is substantially lower than the corresponding value for ENSO. The correlation between the zonal wind anomaly in the western equatorial basin and equatorial thermocline variation is also low, achieving a maximum value of 0.2 when the zonal wind leads the thermocline by 4 months. Furthermore, the EOF mode that characterizes the thermocline variability associated with the zonal mode only explains less than 10% of total thermocline variance in the tropical Atlantic. These observational analyses suggest that the Bjerknes type of feedback appears to operate sporadically for certain events in the equatorial Atlantic, but its strength is much weaker than in the Pacific. The basin size may be the ultimate cause of the differences between the Pacific and Atlantic in the strength of the Bjerknes feedback and the relative importance of thermocline depth and upwelling changes in the subsurface-to-SST feedback; theoretical studies show that the equatorial mode is more unstable in the larger Pacific basin (Hirst 1986; Battisti and Hirst 1989).


The simple coupled model experiments carried out by Zebiak (1993) provide further insight into the similarities and contrasts between the dominant ocean processes controlling the Pacific ENSO and the zonal mode. In the Pacific, the ENSO related SST variability in the eastern equatorial Pacific is significantly forced by the mean upwelling acting on the anomalous vertical temperature gradient and therefore thermocline variability and large-scale equatorial ocean dynamics are crucial in determining the onset of the SST anomaly. In the Atlantic, the primary driving force for the SST anomaly in the eastern equatorial region comes from anomalous upwelling acting on mean vertical temperature gradient. The mean upwelling, though it also contributes to forcing SST anomalies, is much less significant.


This anomalous upwelling is primarily induced by local surface Ekman current divergence in response to changes in local winds. Therefore, in the equatorial Atlantic local Ekman feedback, in which the zonal SST gradient drives wind anomalies that induce upwelling, may play a more important role than the Bjerknes feedback which relies on subsurface temperature changes induced by the equatorial thermocline adjustment to wind changes in the western basin. In the eastern tropical Pacific, it has been proposed that the Ekman feedback may play a role in the seasonal onset of the cold tongue (Mitchell and Wallace 1992, Chang and Philander 1994), but is less important on interannual time scales (Koberle and Philander 1994), while in the equatorial Atlantic this mechanism seems to dominate both the seasonal cycle (Carton and Zhou 1997) and the zonal mode (Zebiak 1993). If these modeling results hold, it would mean that the subsurface memory mechanism that is so critically important for the evolution of ENSO (see Neelin et al. 1998, for a reveiw) may be less effective in the zonal mode.


Carton and Huang (1994), however, point out that the role of subsurface ocean dynamics can be different in different warm events. By contrasting responses of an OGCM to the observed wind changes during 1983-84 and 1987-88, they show that the subsurface ocean plays a preconditioning role for the warm event in 1984 in the sense that there was a build-up of a positive heat content anomaly in the western tropical basin during the summer and fall of 1983, but such a role was not evident for the warm event in 1988. Delecluse et al. (1994) attempt to quantify the role of the 1982-83 El Niño in the 1984 Atlantic warm event by conducting a set of AGCM experiments to isolate the ENSO-forced atmospheric variability from the total fields over the tropical Atlantic sector and then conducted a set of OCGM experiments forced with output from the AGCM. This study shows that the strong 1982-83 El Niño helped to precondition the Atlantic warming that occurred in 1984, but that local feedback is required to retain both the strength and phase of the equatorial warming in the Atlantic. Based on this finding, they propose that Pacific ENSO may influence the onset of the zonal mode more than one year in advance. Vauclair and du Penhoat (2001) argue that the warm events that occurred in summer of 1995 and 1998 may also be linked the corresponding El Niño events. However, these authors also note that the event in 1988 appears to be independent of the 1986-87 El Niño. At the moment, it is not clear why El Niño affects some Atlantic warm events but not the others. Nor is it clear what causes such a long lag (more than a year) between the warming in the equatorial Pacific and Atlantic.


As noted earlier, the decay phase of the meridional mode coincides with the onset phase of the zonal mode, raising the possibility that the two phenomena may be interrelated. Servain et al. (1999, 2000) report a significant correlation between the cross-equatorial SST gradient variation and the zonal slope of 20ºC-isotherm depth variation along the equator in both observed data and OGCM simulations during 1980-1997. There are, however, two caveats associated with this result. First, the high correlation reported by Servain et al. (1999, 2000) appears to be only robust for the 1980s and 1990s. An analysis of an ocean model simulation forced with the NCEP reanalysis winds for the period 1949 to 2000 by Murtugudde et al. (2001) reveals that the correlation between the cross-equatorial SST gradient variation and the zonal slope of 20°C-isotherm depth variation are much lower for the period before 1980. Murtugudde et al. (2001) conjecture that the 1976 ``climate shift'' has an impact on TAV. Prior to 1976, TAV was dominated by the meridional mode, and was less affected by the Pacific ENSO, and the connection between the zonal and meridional modes was weak. After 1976, TAV is dominated by the zonal mode, the ENSO influence is stronger, and the connection between the two Atlantic modes is also stronger.


If this conjecture could be proven, it would suggest that the connection between the meridonal and zonal modes may be felt through the remote influence of ENSO. Second, as discussed earlier, in contrast to ENSO the Atlantic equatorial SST anomaly is only relatively weakly correlated to the thermocline changes. Therefore, a high correlation between the cross-equatorial SST gradient variation and the zonal slope of 20ºC-isotherm depth variation along the equator does not necessarily mean that a strong cross-equatorial SST gradient anomaly would lead to a strong equatorial SST anomaly. Obviously, much remains to be studied on the relationship between the two TAV modes. Of particular interest is their relationship during the boreal spring and summer when the two phenomena co-exist.


The dynamic processes that contribute to the decay of the zonal mode are generally less understood. For ENSO, the decay of the equatorial SST anomaly in the eastern Pacific has been attributed to the thermocline perturbation generated by Rossby waves of opposite sign to the initial thermocline anomaly. It is not clear to what extent a similar mechanism works for the Atlantic zonal mode. The simple model study by Zebiak (1993) suggests that the dominant processes that contribute to the decay of the SST in the eastern equatorial Atlantic are horizontal advection, particularly the meridional component, and the damping effect of the surface heat flux, but not the thermocline fluctuation. In fact, these damping processes are so strong that the model does not support any self-sustained oscillation in the tropical Atlantic. Nevertheless, Zebiak (1993) concludes that at least in the simple model framework the primary mechanism for the Pacific ENSO and Atlantic zonal mode is basically the same: the so-called delayed oscillator mechanism. Whether or not this conclusion holds in reality needs further investigation.


There have been even fewer studies on the dynamics of the Benguela Niño and Niña. The most recent studies on this subject (Florenchie et al. 2003a,b) suggest that the remote forcing mechanism via equatorial wave dynamics is more plausible than the local feedback mechanism. In particular, the warm and cold events over the past two decades have been linked to the zonal wind stress anomaly in equatorial western Atlantic, in line with the earlier studies by Hirst and Hastenrath (1983) and Picaut (1985), and many Benguela Ninos coincide with strong warm phases of the zonal mode (e.g. 1963, 1984, and 1995).


2) Interactions with extratropical ocean processes


As in the tropical Pacific, the upper circulation of the tropical Atlantic is connected to the extratropical circulation via STCs. What complicates the circulation system in the tropical Atlantic is the effect of the MOC on the wind-driven circulation. As a result of recent observational synthesis efforts (e.g., Schmitz and McCartney 1993; Schmitz 1996; Stramma and Schott 1999; Stramma et al. 2003; Snowden and Molinari 2003; Zhang et al. 2003), a first order description of the three dimensional structure and the pathways of the tropical Atlantic circulation system is emerging (see Snowden and Molinari 2003 for a recent review). There is an estimated total of 21 Sv of water upwelled into the surface layer of tropical Atlantic on an annual average, of which roughly 5 Sv appears to be associated with the Northern Hemisphere STC, 10 Sv is associated with the Southern Hemisphere STC, and the remaining 6 Sv is associated with the MOC (Roemmich 1983, Zhang et al. 2003). The origin of the latter can be traced back into the western Indian Ocean (Sprintall and Tomczak 1993; Tomczak and Godfrey 1994). Among the 10 Sv associated with the Southern Hemisphere STC, 4 Sv of flow can enter the equatorial zone through an interior pathway extending from 10W to the western boundary, and the another 6 Sv of subducted water merges into the northward flowing North Brazil Current/North Brazil Undercurrent. This branch is joined by the 6 Sv of the MOC return flow which is apparently brought into the Atlantic by the Agulhas Current and the associated eddies and carried northwestward by the Benguela Current and the SEC before joining the North Brazil Current/North Brazil Undercurrent south of 10ºS (Stramma and Schott 1999). Therefore, the western boundary pathway is the most effective way for the communication between tropics and extratropics in the Southern Hemisphere. In the Northern Hemisphere, the strength of STC is only a half of that of the Southern Hemisphere STC (Schott et al. 1998; Bourles et al. 1999a,b; Zhang et al. 2003; Wilson et al.1994) and the fate of the subducted water in the Northern Hemisphere, upon crossing 10ºN, is less clear observationally.


Considerable recent research effort has been devoted to the understanding of the hemispheric asymmetry of the Atlantic STCs. Fratantoni et al. (2000) hypothesize that the presence of the MOC return flow in the Atlantic is the main cause of the asymmetry. Chepurin and Carton (1996) and Jochum and Malanotte-Rizzoli (2001) further argue that there is another factor which has to do with the existence of a potential vorticity barrier created by the Ekman suction associated with the Atlantic ITCZ in the northern tropical latitudes. This vorticity barrier inhibits much of the interior communication, forcing the majority of the water subducted in the north Atlantic to flow westward into the western boundary before turning equatorward by the western boundary current. This combines with the fact that the strength of the equatorward flowing western boundary current along the northeastern coast of the South America is significantly weakened by the MOC return flow (Fratantoni et al. 2000) is responsible for blocking the inflow of North Atlantic waters into the equator.


Hazeleger et al. (2003) confirm, based on a Lagrangian trajectory analysis of a high-resolution global OGCM simulation, that the EUC is mainly ventilated from the south and the main subduction sites are located along the South Equatorial Current. Inui et al. (2002) suggest that the interior communication window between the equatorial ocean and the north subtropical Atlantic can be sensitive to wind stress forcing in the region. Da Silva and Chang (2004) show, based on an analysis of an Ocean Data Assimilation (ODA) product from the Geophysical Fluid Dynamics Laboratory (GFDL), that the seasonal variation of the zonal slope of the thermal ridge along the boundary between the NECC and NEC in response to the seasonal variation of the ITCZ can also have an important impact on the pathways in the NTA.


The asymmetric nature of the tropical Atlantic circulation causes the cold thermocline water from the Southern Hemisphere to upwell in the eastern equatorial Atlantic. The upwelled water subsequently is converted into warm surface water via heat exchange with the atmosphere and partly transported toward the northern high latitudes via surface currents. This unique Warm Water Formation and Escape process (Csanady 1984, and Lee and Csanady 1999) makes the Atlantic the only ocean where the net meridional heat transport is everywhere northward (e.g. Trenberth and Caron 2001). In contrast, the meridional heat transport in the Pacific is poleward in each hemisphere and there is little cross-hemispheric heat transport.


Variability in these tropical-extratropical exchanges via STCs and the MOC likely plays an important role in TAV at decadal and longer time scales. However, at present, the importance of the large-scale ocean circulation can only be made by theoretical argument and numerical model experiment, due to the lack of long-term ocean observations. The meridional mode with its decadal timescales appears to be a likely candidate for involvement (both instrumental and paleo proxy records show an enhancement of spectral power in the 10-13 year frequency band [Mehta 1998; Black et al. 1999]).


Alternatively, some investigators (e. g. Dommenget and Latif 2000, Enfield et al. 1999, Melice and Servain 2003) argue that the ocean does not play a significant role. They argue that the SST variation in each hemisphere may be driven independently by dynamic processes in its own hemisphere. For example, Meclice and Servain (2003) shows that SST anomalies in the north tropical Atlantic are related to the NAO, while anomalies in the south tropical Atlantic are related to low-frequency fluctuations of the southern subtropical high. Occasionally, SST anomalies in each hemisphere line up with opposite sign, giving rise to a strong cross-equatorial gradient. Since extratropical fluctuations, such as the NAO, are predominantly caused by chaotic dynamics inherent to the atmosphere, this scenario would argue that fluctuations of the meridional mode are governed by a red noise process and strong interhemispheric SST anomalies occur by chance. This scenario can be regarded as a null hypothesis for decadal variation of TAV.


Competing hypotheses argue for the importance of the interplay between the regional ocean-atmosphere coupling and oceanic feedback. One proposed mechanism that emphasizes a more active role for ocean circulation in the decadal variation of TAV invokes interactions between the tropical and extratropical ocean circulation via STCs. In the tropical Atlantic, the Southern Hemisphere STC supplies most of the water to the equatorial thermocline, and thus is more likely to bring extratropical anomalies into the equatorial zone. Furthermore, since the Southern Hemisphere STC appears to be nearly steady, at least during the last decade (Stramma et al. 2003), the mechanism proposed by Gu and Philander (1997) may be more relevant and applicable to the south Atlantic.


Indeed, the numerical simulation by Lazar et al. (2002) shows the propagation of thermal anomalies around the subtropical gyre in the south Atlantic, some of which appear to reach the equatorial region after approximately 10 years (Fig. 11). This led to the proposal that the Southern Hemisphere STC, through its action on tropical SST in the equatorial upwelling zone, may provide a way by which slow ocean processes modulate the meridional gradient of tropical SST, and hence the low frequency cycles of the inter-hemispheric SST gradient. One caveat to this is that the anomalies traced to the equator in the model appear to be very weak. Therefore, unless local air-sea feedback is sufficiently strong to amplify the extratropical signal, it seems unlikely that this disturbance alone can have significant impact on the tropical SST. At present, the evidence of amplication of the extratropical signal has not been forthcoming.


The Northern Hemisphere STC, on the other hand, exhibits much more temporal variability, making it a possible candidate for the second type of STC mechanism involving fluctuations in volume transports (Kleeman et al. 1999). However, this STC is much weaker than its Southern Hemisphere counterpart and involves more complex dynamics. One particularly interesting aspect of the Northern Hemisphere STC is that its structure and pathways may be sensitive to the MOC changes (e.g., Fratantoni et al. 2000; Jochum and Malanotte-Rizzoli 2001). Based on the ocean model experiment results of Fratantoni et al. (2000) and Jochum and Malanotte-Rizzoli (2001), it is possible that a decrease in the MOC could cause more symmetric Atlantic STCs which could lead to an increase in the supply of thermocline water from the Northern Hemisphere Equatorial Undercurrent and a resulting change in upper equatorial ocean thermal structure. How viable this mechanism is in term of influencing SST and the overlying atmosphere and how such a feedback loop would work is still very unclear.


There are other proposed mechanisms whereby MOC changes could affect the surface ocean conditions. Yang (1999) proposes that MOC changes can affect the interhemispheric SST changes. His mechanism, which involves MOC modulation of cross-equatorial heat transport, is supported by some numerical simulation experiments. The experiments show that the relatively short adjustment time scale (5 years) that links the high latitude and tropical oceans is set by Kelvin and Rossby wave propagation. The resulting adjustment processes follow closely the theoretical analysis described by Kawase (1987) and Cane (1989). As the wave adjustment takes place along the western boundary and the equator, the maximum SST anomalies simulated by the idealized model appear in a narrow region along the western boundary.


In a recent study that follows Yang’s reasoning, Johnson and Marshall (2002) articulate the importance of the so-called “equatorial buffer” mechanism. The fast propagation of Kelvin waves along the western boundary and then along the equator makes the equatorial region a buffer zone, which acts to limit and delay the response of the Southern Hemisphere to a sudden change in deep water formation in the northern high latitudes. The disconnect between the two hemispheres causes convergence or divergence of heat in the equatorial region during the adjustment period, which can then produce a sizable change in the SST. The tight coupling between the atmosphere and ocean in the deep tropics can potentially amplify the resulting thermal signal and lead to a large-scale response in the coupled system. The extent to which this buffer mechanism works in the real climate system needs to be further examined in a more realistic coupled model framework.


Observational evidence regarding the potential effects of northern deep-water formation anomalies on the tropical circulation and stratification may be forthcoming. A large volume of Labrador Seawater that was formed in the early 1990’s (Molinari et al., 1998; Stramma et al., 2004) is propagating southward along the western boundary of the North Atlantic. It was located east of the Bahamas in 1996 and was just recently detected to have advanced to 8°N (R. Fine, pers. Comm., 2004). Hopefully, an observational system will be in place to study its propagtion in the equatorial zone in comparison with the model predictions as summarized here.


While many details remain to be worked out regarding the oceanic connection between the MOC and TAV, there are several recent studies suggesting that a sudden change in the MOC strength can produce a significant response in the tropics. Coupled CGM model studies by Vellinga and Wood (2002) and Dong and Sutton (2002) have shown a southward shift of the ITCZ over the Atlantic, accompanied by a dipole-like SST pattern with a cooler (warmer) temperature in Northern (Southern) Hemisphere (Fig. 12) in response to a collapse of the thermohaline circulation in the Atlantic Ocean, — a pattern which is consistent with the Atlantic meridional mode. Both of these studies are based on a version of the UK Hadley Centre global coupled ocean-atmosphere GCM (HadCM3). While Vellinga and Wood (2002) focus on the quasi-equilibrum response, Dong and Sutton (2002) focus on the adjustment of the coupled system. The latter study finds that the tropical Atlantic response to the change in the high latitudes is established within a relatively short time. Particularly worth noting is the coupled system adjustment when the colder SST reaches the equator and a cross-equatorial SST gradient develops within 4-6 years after the high latitude disturbance was introduced. The SST gradient causes a southward shift of the ITCZ in the tropical Atlantic sector, which apparently initiates an El Niño event in the tropical Pacific. The most recent coupled GCM study by Zhang and Delworth (2005) not only confirms these previous modeling results, but further shows that the Atlantic MOC change can alter weather patterns globally through changes in the tropics and this pattern is consistent with paleo proxy data. While the exact mechanism for the linkage between the high latitudes and tropics in that coupled model still appears unclear, it is possible that the equatorial buffer mechanism may play a role. On the other hand, Chiang et al. (2003) report similar ITCZ response in an AGCM coupled to a slab ocean, raising the question: Does the high-to-low latitude communication occur through an oceanic bridge, or an atmospheric bridge?


d. Summary


Some promising advancement areas for TAV research during the initial phase of CLIVAR are briefly highlighted as follows:
1   2   3   4   5   6   7   8   9   ...   12

Похожие:

Revised and Resubmitted to Journal of Climate November 2005 Abstract iconResubmitted for review to Clinical Biomechanics in August 2005 ms# cb/2005/000090

Revised and Resubmitted to Journal of Climate November 2005 Abstract iconModule specification (Revised November 2008) Title

Revised and Resubmitted to Journal of Climate November 2005 Abstract iconSeventh revised and expanded edition, November, 2007

Revised and Resubmitted to Journal of Climate November 2005 Abstract iconBrushing violets library by author-revised November, 2008

Revised and Resubmitted to Journal of Climate November 2005 Abstract iconBritish Journal of Non-Destructive Testing, November 1990, 32(11) pp 568-577

Revised and Resubmitted to Journal of Climate November 2005 Abstract icon© 2005 by abcm november 6-11, 2005, Ouro Preto, mg

Revised and Resubmitted to Journal of Climate November 2005 Abstract iconPeter N. Nemetz, Ph. D. Professor of Strategy and Business Economics Faculty of Commerce and Business Administration University of British Columbia Journal of International Business Education, 2003, in press. Revised May 11, 2003

Revised and Resubmitted to Journal of Climate November 2005 Abstract icon9th Meeting of The Society for Natural Immunity November 4-8, 2005 nk hawaii Poipu Beach, Kauai

Revised and Resubmitted to Journal of Climate November 2005 Abstract iconResearch topic addressed lsh-2005 1-2 Pharmacogenomics for individualised medicines Proposal abstract

Revised and Resubmitted to Journal of Climate November 2005 Abstract iconАндрей Михайлович Соболев Родился 30 июня 1956 года в г. Свердловске, рсфср
Письма в Астрономический журнал (Letters to Astronomical Journal), 7, 1981, 163-167 (in Russian, English abstract, English version...
Разместите кнопку на своём сайте:
Библиотека


База данных защищена авторским правом ©lib.znate.ru 2014
обратиться к администрации
Библиотека
Главная страница