Revised and Resubmitted to Journal of Climate November 2005 Abstract

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

1. Introduction

The tropical ocean has been studied most extensively with regard to its role in the El Niño – Southern Oscillation (ENSO) phenomenon. During the decade (1985-1994) of the Tropical Ocean-Global Atmosphere (TOGA) program, a concerted international research effort was undertaken primarily within the tropical Pacific Ocean with the goal of better describing ENSO as a coupled ocean-atmosphere phenomenon, understanding its underlying dynamics, as well as exploiting its predictability to develop forecast systems using coupled ocean-atmosphere models. The immensely successful TOGA led to improvements in our understanding of how the dynamics of the equatorial wave guide plays the role of ocean memory in the ENSO cycle. It has led to an understanding of how this oceanic mechanism, working in concert with ocean-atmosphere feedbacks, gives rise to a class of coupled modes that are so critical to our understanding of ENSO physics. This understanding forms the theoretical basis for predicting ENSO and associated climate fluctuations on seasonal-to-interannual time scales. What emerged from these studies during the TOGA decade was an appreciation that forecasting ENSO was indeed possible, and that accurate estimation of the upper ocean state in the tropical Pacific was critical for such a forecast. This appreciation led to the development of the TOGA observing system in the tropical Pacific, and activity in developing initialization and assimilation schemes to exploit this valuable information about the ocean state, as well as continuing efforts to improve models and coupling strategies that facilitate the development of coupled model prediction systems. There is now a large body of literature devoted to issues concerning the role of the tropical Pacific Ocean in ENSO, including the monograph by Philander (1990) and a collection of papers in a special volume of the Journal of Geophysical Research (JGR, 1998) dedicated to reviewing the progress of ENSO research in prediction during the TOGA decade. This review focuses on recent progress that has been made after TOGA and during the early years of the ongoing Climate Variability and Prediction (CLIVAR) Program.

The role of tropical oceans in climate goes far beyond ENSO. The global energy balance requires strong poleward heat transport out of the tropics through both atmospheric and oceanic processes. Conditions in the tropical oceans make them highly effective in this heat transport. This is achieved through a number of oceanic processes. Among them are the entrainment of colder subsurface waters into the surface layer that enables the equatorial ocean to absorb atmospheric heat input, and the subsequent transport of these surface waters to the extratropics. This so-called warm water formation and escape (WWFE) process (Csanady 1984) in the upper tropical oceans is in sharp contrast to the deep and bottom water formation process in the polar oceans, both of which form a critical part of the global heat budget. Together with the atmospheric circulation, the tropical ocean circulation creates a unique environment for ocean-atmosphere interactions which are critical elements of the global heat budget. It is in the tropics where the warmest surface water resides, supporting atmospheric deep convection that is itself highly sensitive to changes in sea surface temperatures (SSTs). It is in this region where the ocean mixed layer is tightly coupled to the subthermocline ocean, allowing changes in the subsurface ocean to have a direct impact on SST. It is in this region where a sharp thermocline is formed and maintained by the large-scale ocean circulation, making changes in ocean circulation an important player in global climate variability. It is in this region where the poleward oceanic heat transport is at its maximum, making it a critical part of the global heat budget. As will be discussed, the conditions that make all this possible are not mere accidents. Indeed, one of the major achievements of CLIVAR research has been a significant advance in our understanding of ocean dynamics and coupling processes that determine the structure of the tropical oceans and their direct connection to the subtropics.

There is no doubt that ENSO is the most spectacular and profound climatic phenomenon in the tropical ocean-atmosphere coupled system, whose impact reaches far beyond the tropical Pacific. Considerable progress has been made in the understanding of its variability and predictability over the past two decades. But it is by no means true that ENSO is the only mode of climate variability in the tropical coupled system. Nor does it mean that our understanding of this phenomenon is complete. While the effort is continuing to improve our understanding of ENSO dynamics and to refine the ENSO prediction system during the early years of CLIVAR, the scope and dimensions of this research have expanded considerably to explore the role of other climate phenomena in the tropical coupled system. These recent studies clearly indicate the need to improve our understanding of these other climate phenomena in order to extend our success in ENSO-based seasonal climate prediction to the global domain.

In the Pacific sector, considerable recent research has been directed towards investigating the cause of decadal changes in the “mean state” of the coupled system and its relationship to ENSO. Although the observational record is too short to provide a “canonical” description of mean state changes, empirical studies show that the pattern of the low-frequency variation, albeit similar to ENSO in many aspects, has some distinctive characteristics. Among the distinctions are that its center of action in the tropics shifts westward towards the western Pacific and Indian Oceans. Also, its meridional extent is considerably broader than the canonical ENSO pattern and its variability in the tropics is linked to interdecadal climate fluctations over the north Pacific during boreal winter (Deser et al. 2004). This seasonal feature points to the fact that Pacific interdecadal variability involves not only tropical coupled dynamics, but also processes that involve interactions/exchanges between the tropics and extratropics. In particular, the oceanic aspect of this tropical-extratropical exchange process and its role in determining the nature of the interdecadal changes in the Pacific coupled system has received focused attention in recent years.

In the tropical Atlantic, much of the recent research has been focused on two dominant patterns of coupled ocean-atmosphere variability that are closely linked to major climate fluctuations in the region. The so-called Atlantic zonal mode is often viewed as the Atlantic counterpart of the Pacific ENSO and varies primarily on interannual time scales, whereas the meridional mode seems unique to the tropical Atlantic coupled system and varies on multiple time scales. Collectively, these phenomena are termed as Tropical Atlantic Variability (TAV). TAV is tightly phase locked to the Atlantic seasonal cycle with distinctive seasonality. Although Pacific ENSO plays a dominant role in the remote forcing in TAV, those two phenomena can occur without remote influence from ENSO, suggesting that they are inherent to the tropical Atlantic coupled system. The role of the ocean in TAV remains largely unexplored. Recent studies suggest that processes controlling the zonal mode may be confined within the equatorial ocean, but those affecting the meridional mode may extend well beyond the tropics, possibly including tropical-extratropical ocean exchanges and interactions with the Atlantic Meridional Overturning Circulation (MOC).

In the Indian Ocean, recent anomalous events that took place during 1994 and 1997 led to the discovery of a new ocean-atmosphere coupled phenomenon now widely known as the Indian Ocean Dipole (IOD) mode (Saji et al. 1999; Yamagata et al. 2003a 2004) or the Indian Ocean Zonal (IOZ) mode (Webster et al. 1999). The IOD mode has shown to affect not only climate fluctuations in the Indian Ocean sector but also around the globe. Its discovery has generated vigorous interest in the study of ocean-atmosphere coupled dynamics over the Indian Ocean sector and the role of IOD in global climate variability. A prominent feature of the IOD is the characteristic east-west dipole pattern in SST anomalies which is in turn coupled to large thermocline anomalies. The atmospheric response to the SST dipole pattern is observed in wind and outgoing longwave radiation (OLR) (Behera et al. 1999,2003a; Yamagata et al. 2002) and sea level pressure (Behera and Yamagata 2003). These oceanic and atmospheric conditions imply that the Bjerknes-type (Bjerknes 1969) feedback mechanism is responsible for the IOD evolution. Therefore, the ocean memory is expected to play an important role through equatorial wave adjustment but this needs to be investigated by more observation.

The purpose of this paper is to provide an overview of recent studies concerning the role of the oceans in a wide spectrum of climate phenomena within the tropical coupled system. For the sake of convenience, the discussion is divided geographically into three tropical ocean basins. For a comparative view of the mean state, seasonal cycle and interannual variability among the tropical oceans, the readers are referred to Wang et al. (2004). We begin our discussion of the tropical Pacific Ocean in section 2, then move to the tropical Atlantic Ocean in section 3, and finally visit the tropical Indian Ocean in section 4. For each basin, instead of giving a general discussion of the ocean circulation, we focus on the oceanic processes that are most relevant to the specific phenomena in the region after a brief description of the phenomena and the feedback mechanisms. This division does not imply that the phenomena within each tropical ocean basin are independent of those in other basins. On the contrary, many of these phenomena are interrelated and interact among each other. In section 5, we discuss the challenges that lie ahead.

2. Pacific ENSO and Decadal Changes in ENSO

Although its impact is global, the origin of ENSO resides in the tropical Pacific. The interaction between the atmosphere and ocean within the tropical Pacific basin plays a fundamental role in determining the characteristics of ENSO. Numerous studies have been devoted to documenting and understanding its evolution, its apparent preferred time scale, its phase-locking to the annual cycle, its irregularity and its impact. For a detailed account of the history of ENSO, the readers are referred to many review papers and monographs, in particular, the excellent text by Philander (1990) and the special volume of the Journal of Geophysical Research (JGR, 1998). There is little dispute that ENSO is a genuine ocean-atmosphere phenomenon borne out of active interaction between the two components of the climate system. Its occurrence can not be explained by either atmospheric or oceanic processes alone. In the following we first give a brief summary of our general understanding of this phenomenon, and then focus on the more recent studies that are mostly concerned with ENSO’s predictability and its low-frequency variability.

As a useful approximation, ENSO can be described as a climate perturbation around the “mean” state of tropical Pacific coupled system. The mean state consists of an east-to-west gradient in SST and overlying Trade winds (Walker circulation) of the tropical atmosphere that support a positive dynamical feedback: the temperature difference along the equator reinforces the strength of the trade winds by favoring large scale ascent and atmospheric heating over the western equatorial Pacific and large scale descent and atmospheric cooling over the eastern equatorial Pacific. In turn, the easterly wind stress acting on the ocean surface causes the thermocline to rise and the cold subsurface water to upwell in the east. The Trade winds and associated equatorial upwelling maintain the climatological distribution in the tropical Pacific SST: i.e., a western warm pool and eastern cold tongue structure in the equatorial Pacific. Because of the positive feedback, a modest change in either the equatorial SST or in the trade winds can trigger a chain reaction in the coupled system. For instance, if there is a weakening in the equatorial trade wind (a westerly wind anomaly), the equatorial upwelling will decrease. This causes a relaxation in the west-to-east slope of the thermocline and decreases the west-to-east sea surface temperature contrast. Since the Walker circulation is maintained by this surface temperature gradient, the weakening in the zonal temperature gradient will cause further weakening of the trade winds, which in turn causes further warming in the eastern equatorial Pacific. This feedback mechanism, known as Bjerknes' hypothesis, is the key element responsible for the development of warm (and cold) ENSO episodes.

The canonical picture of ENSO based on a variety of observations is entirely consistent with Bjerknes' hypothesis. The robust features accompanying warm ENSO events include: 1) quasi-stationary warm SST anomalies in the eastern and central Pacific; 2) a relaxation of the trade winds associated with positive SST anomalies in the eastern and central equatorial Pacific at event onset; 3) a deepening in the east and a shoaling in the west of the thermocline along the equator, that lead the SST changes; 4) a tendency for the anomalously deep thermocline in the eastern/central Pacific to return to climatological values prior to the peak of the ENSO event; 5) an increase in the trades in the far western Pacific one to two seasons prior to the onset of the ENSO event. These features underscore the tight coupling between the atmosphere and ocean during ENSO evolution.

ENSO events last approximately 12-18 months and occur every two to seven years with large variation in strength. Further analysis indicates that for most ENSO events the maximum warming in the eastern equatorial Pacific occurs in December and January. This property has been referred to as the phase-locking of ENSO to the annual cycle. Linear (and non-linear) wave dynamics of the equatorial wave guide are crucially important in giving rise to the quasi-oscillatory nature of ENSO. Oceanic Kelvin and Rossby waves propagate energy and momentum received from the wind stress, providing the oceanic memory that is so important to ENSO. (Similar waves exist in the atmosphere, but their propagation rates are far greater than the oceanic counterparts. Therefore, the adjustment time scale of the tropical atmosphere to changes in SST is much shorter (10 days or less) than the adjustment time scale of the equatorial ocean (approximately six months) to changes in wind stress. The short adjustment time of the atmosphere suggests the assumption (good to first order) that the atmosphere is in a statistical equilibrium with the SST on time scales longer than a few months.) Thus, the memory of the state of the climate system primarily resides in the ocean.

The free oceanic Kelvin and Rossby waves can be strongly modified by air-sea coupling. The Bjerknes feedback can destabilize these waves, giving rise to unstable coupled modes that resemble the slow westward propagating oceanic Rossby mode and the eastward propagating oceanic Kelvin mode. These coupled modes represent the extrema of the continuum of unstable coupled atmosphere-ocean modes. In fact, the coupling between the atmosphere and ocean generates a new breed of modes whose characteristics depend on the time scale of dynamical adjustment of the ocean relative to the time scale of SST change due to the air-sea coupling (Hirst 1986, 1988; Neelin 1991; Neelin and Jin 1993; Jin and Neelin 1993a,b). Stability analysis of a simple ENSO model linearized around a given mean state reveals a rich variety of structures in the coupled modes in a space spanned by parameters characterizing the two time scales (Neelin 1991; Neelin and Jin 1993; Jin and Neelin 1993a,b; Neelin et al. 1994; Neelin et al. 1998). The coupled mode of most relevance to ENSO in reality appears to reside in a parameter regime where the time scales associated with the local air-sea interaction are comparable to the dynamical adjustment time of the tropical Pacific Ocean.

The evolution of the coupled mode in this parameter regime can be described in two phases. During the development phase, the Bjerknes positive feedback dominates which causes the anomalies to grow. During the decay phase, the equatorial wave adjustment process of the ocean plays a role of delayed negative feedback. Rossby wave packets carry off-equatorial thermocline anomalies of opposite sign to the equatorial anomaly generated by the Bjerknes feedback to the western boundary where they are reflected into equatorial Kelvin waves which subsequently propagate eastward along the equator. These negative feedbacks counteract the Bjerknes positive feedback and cause the system to have perpetual turnabouts from warm to cold states and back again. The time scale that is associated with the ocean wave adjustment provides the "memory" of the coupled system that is essential for the oscillations in this ENSO paradigm.

Mathematically, the behavior of the ENSO mode can be described by a heuristic differential-delay equation in terms of sea-surface temperature anomaly. Therefore, this coupled mode is widely known as the “delayed-oscillator mode” (Battisti and Hirst 1989; Schopf and Suarez 1988, 1990). While the delayed oscillator appears to explain many features of both observations and models, it has not been directly verified through observation; in particular, the implied symmetry of cold and warm states seems unrealistic (Kessler 2002a). At the low frequencies of ENSO, determining wave reflection efficiency and other necessary components of the theory are extremely difficult (Zang et al. 2002), but the work of Schopf and Suarez (1990) pointed out that a wave reflection efficiency of as little as 15% could still sustain a viable oscillating system. The importance of Rossby wave dynamics and western boundary reflection was also noted by Schneider, et al. (1995).

A related prototype ENSO model, known as the recharge oscillator, has been proposed by Jin (1997), in which the role of the equatorial Kelvin and Rossby waves in the ocean adjustment is described as the slow recharge/discharge of the equatorial heat content. This view makes use of the fact that the Rossby/Kelvin wave transit time across the basin is short in comparison with the period of ENSO, and therefore one may make a quasi-steady approximation, replacing the Rossby waves with Sverdrup balance, as described by Anderson and Gill (1975). This removes the wave transit time as a natural time scale of the problem, replacing it with a free parameter of the system. The free parameter determines the time it takes to recharge the off-equatorial heat content in response to changes in off-equatorial wind-stress curl. A major advantage of this approach is that the essential mechanism, interior Sverdrup flow, is more amenable to measurement, and monitoring of the low frequency convergence of the tropical thermocline can be undertaken (Johnson and McPhaden 1999; McPhaden and Zhang 2002, 2004). The challenge to such monitoring, however, occurs because the hard-to-measure western boundary current transports may counteract the interior mass convergence.

In these theories, the ocean's role lies in the dynamical response of the thermocline throughout the tropics to anomalous wind forcing, transmitting information across the basin through large-scale dynamics. When averaged over periods of a year or more, the delayed and recharge oscillators are fundamentally equivalent, both relying on wind stress-curl-generated thermocline depth and geostrophic flow changes. Both oscillators are self-sustained, with the thermocline depth carrying the memory across both phases of the cycle. Alternative views of the ocean's role emerge in theories that ENSO is due to an advective mechanism (Picaut et al. 1997) or is due to equatorial Kelvin waves only - theories based on “westerly wind bursts” and stochastic forcing. The distinction between these views emerges through a concern over how equatorial wave signals alter the SST, which has important implications for whether simple predictive systems can be built. Niether of these alternative views seeks to determine the time scale of ENSO or its recurrence. Instead, they are concerned with the dynamics during the onset of ENSO. In the next section, we examine the source of ENSO irregularity.

a. ENSO irregularity and predictability limit

The detailed features of any single ENSO event vary considerably from case to case, including when and where the initial warming starts. For example, the 1997/8 event increased surface temperature near Peru by more than 5 ºC. In contrast, in the 1986-87 ENSO event the warming extended eastward only as far the mid-Pacific (near 170ºW) and the maximum temperature was a modest 1ºC above normal. The warming from 1990 to 1994 consisted of three weak warm events and had a persistent “horseshoe-shaped” SST anomaly. This irregularity reflects the complexity of the coupled ocean-atmosphere system and hints at the difficulties in predicting ENSO. Much remains to be done on the fundamental question of the predictability limit for ENSO. Beyond noting that some model systems seem to exhibit predictive skill for ENSO over a short time, we do not know whether this limit is essentially due to lack of model skill, the inability to adequately specify the initial conditions, or to not-yet-understood fundamentals of the physical system.

Theories on the cause of ENSO irregularity can be broadly grouped into three categories: The first view argues for the importance of nonlinearity within the tropical coupled system. The nonlinearity arises from strong air-sea feedback that puts the coupled mode in an unstable dynamic region. In this regime, ENSO can not only be described as a self-sustained oscillator, but it can interact nonlinearly with either the annual cycle (Jin et al. 1994; Tziperman et al. 1994; Chang et al. 1994; Wang et al. 1999) or other coupled modes (e.g., Mantua and Battisti 1995), giving rise to deterministic chaos. The loss of predictability is primarily due to the uncertainty in the initial conditions. This view relies upon fairly robust ocean wave dynamics that provide the underlying timescales for the problem.

The opposing view to this is the stochastic ENSO theory in which "weather" noise (which may include timescales up to the intraseasonal) generated by the internal dynamics of the atmosphere plays a fundamental role in not only giving rise to ENSO irregularity, but also in maintaining ENSO variance. In this view, the coupled mode is in a stable damped regime, and thus ENSO cycle cannot be self-sustained without external noise forcing (Penland and Sardeshmukh 1995; Flügel and Chang 1996; Moore and Kleeman 1999a,b; Kleeman and Moore 1997; Thompson and Battisti 2000 2001; Flügel et al. 2004). All noise forcing is not equal, however, and the spatial structure of some types of high-frequency atmospheric phenomena may project especially well onto optimally growing perturbations
1   2   3   4   5   6   7   8   9   ...   12


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:

База данных защищена авторским правом © 2014
обратиться к администрации
Главная страница