1 Introduction

Surface fluxes and tropical intraseasonal variability: a reassessment


The authors argue for the hypothesis that interactive feedbacks involving surface enthalpy fluxes are important to the dynamics of tropical intraseasonal variability. These include cloud-radiative feedbacks as well as surface turbulent flux feedbacks; the former effectively act to transport enthalpy from the ocean to the atmosphere, as do the latter. Evidence in favor of this hypothesis includes the observed spatial distribution of intraseasonal variance in precipitation and outgoing longwave radiation, the observed relationship between intraseasonal latent heat flux and precipitation anomalies in regions where intraseasonal variability is strong, and sensitivity experiments performed with a small number of general circulation and idealized models.

The authors argue that it would be useful to assess the importance of surface fluxes to intraseasonal variability in a larger number of comprehensive numerical models. Such an assessment could provide insight into the relevance of interactive surface fluxes to real intraseasonal variability, perhaps making it possible to rule out either theoretical explanations in which surface fluxes are crucial, or those in which they are not.

Surface fluxes and tropical intraseasonal variability: a reassessment


Department of Applied Physics and Applied Mathematics and Department of Earth and Environmental Sciences

Columbia University, New York, NY,


Department of Atmospheric Sciences

Colorado State University, Fort Collins, CO,


Department of Applied Physics and Applied Mathematics

Columbia University, New York, NY,


Department of Atmospheric Sciences

University of Washington, Seattle, WA.

May 10, 2008

1 Introduction

Theoretical understanding of the mechanisms responsible for tropical intraseasonal variability is limited. There are many interesting and plausible ideas in the literature, but there is no agreement on which of them, if any, is correct. The Madden-Julian oscillation (MJO) in particular is arguably the most significant mode of atmospheric variability at any sub-decadal time scale whose essential features — its existence, energetics, spatial and temporal scales — remain so unsatisfactorily explained.

In this study, we use the phrase ”MJO” to refer to the eastward-propagating 30-60 day mode which is dominant in southern hemisphere summer. This mode remains present in northern hemisphere summer, but northern summer also features a northward-propagating mode, manifest in northward-propagating rain bands over the Indian subcontinent and adjacent oceans. We refer to these two modes collectively as ”tropical intraseasonal variability” and treat them to some degree as one phenomenon. We recognize that the eastward- and northward-propagating modes have some significant differences, but argue here that there may be fundamental similarities in their energetics.

General circulation models (GCMs) simulate tropical intraseasonal variability with varying degrees of fidelity. A couple of recent intercomparison studies (linet06, ; zhaet06, ) show that even the best models still have significant flaws in their MJO simulations. At the same time, some members of the current generation of models show considerable improvement over previous generations. While improving GCM simulations is sometimes cited as a motivation for theoretical research into the MJO, no clear relationship exists between the fidelity of GCM simulations and the state of theoretical understanding, perhaps because of the very different levels of complexity of GCMs as compared to the idealized models used by theorists. It is not clear that recent improvements in MJO simulation owe anything to theoretical understanding of the mechanism of the MJO. GCM improvements in MJO simulation often seem to be accidental by-products of broader model development efforts, or results of trial-and-error tuning, or perhaps tuning guided by broader principles not specific to the MJO. For example, any model change which tends to inhibit deep convection tends to increase variability at all timescales, including the intraseasonal timescale. This constitutes improvement for models in which intraseasonal variability is too weak, as is the case with many.

Any steps we can take to narrow the range of mechanisms which are considered possible explanations for tropical intraseasonal variability would be valuable, particularly if some mechanisms can be eliminated convincingly enough to focus the attention of the community on evaluating the remainder. We argue that models of the MJO in which variations in net surface enthalpy fluxes are crucial are more likely to be correct than those in which such variations are unimportant, while recognizing that the pioneering studies which first proposed this idea (Emanuel 1987; Neelin et al. 1987) have proven incorrect in their details. Those details are all inessential to the hypothesis that surface enthalpy fluxes are important to tropical intraseasonal variability.

Three primary pieces of evidence support our argument:

  1. Intraseasonal variance in precipitation and outgoing longwave radiation is observed to be larger over ocean than land. This is true in both hemispheres, even in regions where the climatological mean precipitation is larger over land than ocean. Since the net surface enthalpy flux must vanish over land, the land-sea contrast in intraseasonal variance is consistent with a role for that flux in generating the variance.

  2. Over the oceanic regions of largest intraseasonal variability, intraseasonal variations in net surface enthalpy flux and precipitation are correlated.

  3. In several general circulation models (as well as some idealized models) surface enthalpy fluxes are demonstrably important to the simulated intraseasonal variability. Experiments with these models suggest that the role of surface fluxes is larger in those models whose MJO simulation is better.

While most of this evidence is not new, the GCM results in particular have started to become more convincing, partly because the state of the art in GCM simulations of intraseasonal variability has improved (e.g., Zhang et al. 2006). This, in conjunction with continuing observational and theoretical work, has led us to the position that the case for the importance of surface enthalpy fluxes to observed intraseasonal variability has become stronger than it was a decade ago, and deserves to be systematically re-examined.

In the following section, we provide a highly selective review of some results and ideas, mostly from the theoretical and modeling literature, which are relevant to the hypothesis that surface enthalpy fluxes are important to the dynamics of tropical intraseasonal variability. In section 3, we review some observational results which are consistent with this hypothesis. This is followed in section 4 by a discussion of GCM results, including a presentation of new results from the NOAA GFDL AM2 model which show that the hypothesis appears to have merit in that model, consistent with the results of Maloney and Sobel (2004) who used a version of the NCAR model. In section 5, we discuss the implications of these results and propose avenues for further research.

2 Theory and Modeling: An Unbalanced Review

2.1 Southern summer intraseasonal variability (MJO): Models of Emanuel (1987) and Neelin et al. (1987)

We do not attempt to provide a comprehensive or balanced review of MJO theory. This is beyond our intended scope, and recent reviews have been done by wang_05 () and zhang_05 (). Instead we focus on theories in which variations in surface enthalpy fluxes figure prominently.

Approximately simultaneously, ema87 () and neeet87 () proposed that air-sea interaction could destabilize a moist Kelvin wave, leading to intraseasonal variability in the tropics. The arguments involved linear analysis of idealized moist models in which the temperature structure is assumed to be represented by a first baroclinic mode, and convection is controlled by quasi-equilibrium principles. Essentially, convection acts in these systems to eliminate some local measure of stability of the column to deep convection. The atmosphere is adjusted by convection towards a reference value of the stability measure [e.g., convective available potential energy (CAPE)], which for present purposes may be assumed zero. The dynamics of such models is discussed in more detail in a number of reviews (e.g., Emanuel et al. 1994; Emanuel 1997; Neelin 1997, Stevens et al. 1996; Smith 1997; Arakawa 2004; Emanuel 2007). neeet87 () also performed numerical simulations with a general circulation model (GCM).

We refer to models of the type discussed by Emanuel et al. (1994) as Òquasi-equilibriumÓ models. In its most general sense, the term QE refers to a broader category, including all models in which the convection is assumed close to statistical equilibrium with its forcings. Traditional QE models incorporate additional assumptions, in particular that of a pure first baroclinic mode vertical structure, which can be relaxed without relaxing the assumption of QE per se. In models assuming a first baroclinic mode structure as well as QE, the interaction of deep convection with large-scale dynamics alone does not generate unstable large-scale modes. In the simplest such models convection is assumed to respond instantaneously to large-scale forcing (which can come from large-scale dynamics, radiation, or surface fluxes) so as to remove all instability completely; Emanuel et al. (1994) called this ”strict quasi-equilibrium”. In this case, the interaction of convection and large-scale dynamics reduces the effective stratification and thus the phase speed of convectively coupled gravity and Kelvin waves, but does not stabilize or destabilize them. If instead the convection is assumed to relax the CAPE (or other measure of instability) towards its reference value with a finite timescale, then the interaction damps disturbances, a phenomenon known as ”moist convective damping”. Disturbances in these models cannot become linearly unstable through the interaction of convection with large-scale dynamics alone, but only through feedbacks involving processes which can act as sources of moist static energy (or moist entropy) to the column. The two most important such processes are surface turbulent fluxes and radiative cooling.

Right or wrong, the requirement for moist static energy sources to be involved in any linear instability is an interesting feature of first baroclinic mode quasi-equilibrium models. In extratropical atmospheric dynamics it has proved extremely useful to separate dry adiabatic dynamical mechanisms (e.g., Hoskins et al. 1985) from those in which ”diabatic” processes (defined for a dry working fluid with phase changes of water considered external), which break the conservation of potential temperature and potential vorticity, are fundamentally involved. It is clear that dry adiabatic dynamics are inadequate to describe many important aspects of the tropical atmospheric circulation and its variability, but the relative importance of moist adiabatic dynamics — as opposed to dynamics in which moist diabatic processes (those which break the conservation of moist static energy and moist entropy) are critical — remains unresolved. The analogy to extratropical dynamics, and the overall centrality of quasi-conserved variables in all of physics, suggests that it is fruitful to ascertain the relative importance of moist adiabatic and diabatic processes to intraseasonal variability. This is a separate and more fundamental question than that regarding the validity of first baroclinic mode QE models. Nonetheless those models provide a useful starting point for discussion since they make a clear prediction on the relevance of diabatic processes, as well as being both relatively tractable and based on principles that are at root physically reasonable (convection acts to eliminate instability) even though some of their simplifying assumptions may be too strong for some purposes.

In the models of Emanuel (1987) and Neelin et al. (1987), Kelvin waves are destabilized by the interaction of a convectively coupled wave with surface flux perturbations induced by the wave’s surface wind perturbations. This interaction was called ”wind-evaporation feedback” by Neelin et al. (1987) and ”wind-induced surface heat exchange (WISHE)” by Emanuel (1987). For an eastward moving Kelvin wave in a westward mean flow, a positive surface wind speed anomaly occurs a quarter wavelength ahead of the location where the positive precipitation and vertical velocity anomalies would be in the absence of surface flux anomalies, but in phase with the temperature anomaly. Under the strict quasi-equilibrium assumption, the convection responds immediately to surface flux anomalies, so the surface flux anomaly causes the heating anomaly to shift eastward, putting it partly in phase with the temperature anomaly and destabilizing the wave.

Key features of this theory for intraseasonal variability are that the waves must occur in an easterly mean surface flow, that the winds under the convective phase of the disturbance are easterly, and that the intraseasonal disturbances are Kelvin waves. All of these features have been shown to be inconsistent with observations. It was immediately recognized that the strongest MJO events occurred in regions of mean westerlieswang_88 (); emanuel_88 (). It was then shown that the active phases, featuring enhanced precipitation, occurred in surface westerlies (e.g., Kiladis et al. 1994; Zhang and McPhaden 2000) . Wheeler and Kiladis (1999) then showed that ”convectively coupled” Kelvin waves do exist, but that their spectral signatures are quite distinct from that of the MJO, indicating that the two are different phenomena. These observations showed that the models of Emanuel (1987) and Neelin et al. (1987) are, in their specifics, incorrect as explanations of the MJO.

The observations are not, however, inconsistent with the general notion that surface flux anomalies may be important to the dynamics of the MJO, but only with the specific linear models proposed by Emanuel (1987) and Neelin et al. (1987). If the disturbance is something other than a linear Kelvin wave, the requirements for mean easterly flow and net easterly flow in regions of active convection no longer apply. Some studies with nonlinear models have identified such ”nonlinear WISHE” as being important in simulated MJO-like disturbances (ray01, ; malsob04, ).

In the two decades since the publication of Emanuel (1987) and Neelin et al. (1987), much more work has been done with idealized moist models which aim to explain either the MJO, other aspects of tropical intraseasonal variability (such as the northward-propagating mode found in northern hemisphere summer, discussed further below), or other parts of the convectively coupled wave spectrum. In the case of the MJO in particular, none of these has been broadly accepted as providing a satisfactory explanation of the essential mechanisms zhang_05 (); wang_05 ().

2.2 Northern summer intraseasonal variability

In northern summer, intraseasonal variability modulates the Asian and western Pacific monsoons. Spectra of atmospheric variability exhibit two significant peaks in the intraseasonal range: one at 10-20 days and one at 30-60 days (Go05, ). The 10-20-day mode is characterized by convective disturbances which propagate from the western Pacific warm pool and the maritime continent towards the northern Bay of Bengal and South Asia. These disturbances have been associated with equatorial Rossby waves deviated northward by the mean monsoon flow (CG04, ). The 30-60-day mode is characterized by the northward propagation of approximately zonally-oriented rain bands from to (W&06, ). This northward propagation is sometimes accompanied by eastward propagation (WR90, ; LW02, ). Nevertheless, the northward propagating mode appears to be an independent regional mode of variability, rather than simply a local response in the South Asian region to the eastward-propagating disturbances (JL05, ), though this is still controversial in some quarters (e.g., Sperber and Annamalai 2008). We focus here on this northward-propagating mode, assuming that the eastward-propagating mode is essentially similar to the southern summer MJO.

Given the nearly zonal orientation of the rain bands and their nearly meridional direction of propagation, a number of studies have assumed that longitudinal structure is inessential to the dynamics of this mode, and modeled it axisymmetrically (WC80, ; GS84, ; GS90, ; NSG92, ; SGW93, ; JLW04, ; DW05, ; BS07a, ; BS07b, ). These studies have obtained linearly unstable northward propagating modes which resemble the observed one to varying degrees. In earlier studies, land-atmosphere interaction was proposed as crucial to the northward propagation (WC80, ; W83, ). However, northward propagating modes were also later obtained in aquaplanet simulations (GS84, ; NSG92, ). The northward propagation has been attributed in several recent studies to dynamical mechanisms that involve low-level convergence north of the propagating rainband. This convergence is caused via Ekman pumping under a maximum of barotropic vorticity which itself leads the maximum convection (JLW04, ; Go05, ; BS07a, ; BS07b, ). The mechanisms explaining the generation of this barotropic vorticity maximum are still debated (JLW04, ; DW05, ; BS06, ; BS07b, ).

The question of what destabilizes the mode is distinct from that of what causes its propagation. In the model of BS07a (); BS07b (), interactive surface fluxes were found to be important to the instability of the northward propagating mode. They used the quasi-equilibrium model developed by sobel_neelin_06 (), which has a barotropic mode and prognostic boundary layer in addition to a first baroclinic mode in the free troposphere. Because of this more complex vertical structure, the set of possible dynamical mechanisms in this model is broader than that in the pure first baroclinic mode QE models. It is possible for linear instability to occur in this model without surface flux feedbacks. Nonetheless, Bellon and Sobel (2008a,b) found that WISHE is critical to the linear instability of the northward-propagating mode in the parameter regime which appears most justified based on observations. As usual with idealized models, one can easily challenge various details of this model (which has some similarities to earlier ones (e.g. Jiang et al. 2004) as well as some differences). The results of Bellon and Sobel (2008a,b) just show that it is possible to construct a plausible model of the northward-propagating mode of intraseasonal variability — one based on physics that is within the broad envelope of what is commonly found in idealized models of tropical atmospheric dynamics, and also broadly consistent with observations — in which surface flux feedbacks are essential.

2.3 The near-equivalence of surface fluxes and radiation in quasi-equilibrium

The primary radiative effects of the high clouds associated with deep convection are a cooling of the surface due to reflection and absorption of shortwave radiation and a warming of the atmosphere due to the greenhouse effect in the longwave and the absorption of shortwave. To the extent that these effects have similar magnitudes, so that they cancel at the top of the atmosphere, they lead to a cooling of the ocean and equal warming of the atmosphere. This is equivalent to a surface flux, as far as the vertically integrated energy budget of the atmosphere is concerned. QE theory provides a useful context in which to frame this equivalence.

If the vertical structure of the atmospheric flow is assumed fixed (for example, a first baroclinic mode), and if we assume steady state and neglect horizontal advection, the budget of moist static energy requires that the large-scale vertical motion, or net vertical mass flux, is proportional to the net convergence of the vertical flux of moist static energy into the tropospheric column (e.g., Neelin and Held 1987; Raymond 2000; Neelin 1997; Neelin 2007; Sobel 2007) . The latter is the sum of the net turbulent latent and sensible surface heat fluxes plus the vertically integrated radiative heating of the troposphere (or minus the radiative cooling).

The proportionality factor which relates the energy flux to the mass flux is known as the gross moist stability (GMS), following neehel87 (). There is no very good theory for the value of the GMS, though some observational estimates have been made (yuet98, ; back_bretherton_06, ). The first baroclinic mode assumption is restrictive, perhaps even qualitatively misleading in some circumstances (e.g., Sobel 2007), but no better idea of comparable simplicity has yet appeared. In general, the GMS need not be a constant or a simple function of the temperature and humidity profiles alone (as in first baroclinic mode QE theory), because it is quite sensitive to the vertical profile of the divergent circulation (sobel_07, ). Since the latter can vary dynamically on a range of space and time scales, the GMS can as well. In simulations in a GCM with simplified physics fri07a () the GMS is strongly influenced by properties of the convective parameterization fri07b ().

For our immediate purpose, what matters most is that GMS be positive on intraseasonal time scales, so that increases in net vertical energy flux convergence into the column lead (with a lag that is either negligible or at least short by comparison to the intraseasonal timescale; storage on timescales of a few days does not significantly complicate the argument) to increases in vertical mass flux, which in turn imply increases in deep convection. This is a weaker constraint than usually assumed in QE theory, though the difference is one of degree rather than kind. Even the positivity of the gross moist stability is questionable in observations, particularly in the eastern Pacific ITCZ (Back and Bretherton 2006), but it appears to be a reasonable assumption in the Indian and western Pacific regions.

We assume that the difference in the cloud field between convectively active and suppressed precipitation regimes consists predominantly of the presence vs. absence of high clouds. Satellite observations have shown that, in the mean these clouds produce perturbations in the net radiative energy flux at the top of the atmosphere which are small compared to their largely cancelling shortwave and longwave components  (ramanathan_et_al_89, ; harrison_et_al_90, ; hartmann_et_al_01, ). Lin and Mapes (2004) found that this cancellation is less close on intraseasonal time scales, with MJO-related shortwave anomalies being larger than longwave ones by as much as 30%. This is a significant difference, but still the cancellation substantially exceeds the remainder. The implication is that any anomalous radiative heating of the atmosphere due to these clouds, whether occurring in the longwave or shortwave bands, is approximately compensated by anomalous radiative cooling of the ocean. In the vertically integrated moist static energy budget, cloud-radiative heating anomalies due to deep convection are essentially similar to convectively induced perturbations to turbulent surface heat fluxes, as both amount to a net transfer of energy from ocean to atmosphere in a convectively active phase. When convection is active, there is a net decrease of radiative energy flux into the ocean, accompanied by a significantly smaller change in the top-of-atmosphere balance. Thus we use the phrases ”surface fluxes” or ”surface flux feedbacks” to include radiative cooling feedbacks.

2.4 Ocean coupling

A substantial body of work over the last decade or so argues that intraseasonal SST variability is not only driven by the atmosphere, through intraseasonal variations in surface energy fluxes, but that SST variability also influences the atmosphere through the influence of SST anomalies on column stability and deep convection. To the extent that these feedbacks are significant, intraseasonal variability is coupled. Most GCM studies addressing this in the context of the MJO have shown some enhancement of the simulated variability in experiments with atmospheric models coupled to a mixed-layer ocean models, as compared to models with fixed SST(Waliser-Lau-Kim-1999:influence, ; kemball-cook_et_al_02, ; zheng_et_al_04, ; fu_et_al_07, ), although at least one study found no enhancement (Hendon-2000:impact, ) and and others found small enhancements (e.g., Maloney and Sobel 2004) or mixed results, with differences in the mean climate between coupled and uncoupled runs complicating the interpretation (inness_slingo_03, ). This evidence suggests that the MJO is enhanced by coupling, but is not fundamentally dependent on coupling for its existence. In virtually all models tested in this way, a simulated MJO is present to some degree without coupling.

Observations suggest that coupling has a qualitatively similar impact on intraseasonal variability of the Asian monsoon in northern hemisphere summer, including northward-propagating rainbands and SST variability in the Arabian sea and Bay of Bengal (Vecchi-Harrison-2002:monsoon, ; W&06, ; roxy_tanimoto_07, ). In GCM studies, ocean coupling enhances northward-propagating intraseasonal variability in the Indian Ocean to varying degrees (e.g. Zheng et al. 2004, Seo et al. 2007, Fu et al. 2007, Fu and Wang 2004, Kemball-Cook et al. 2002). One recent study with an idealized axisymmetric model suggests that the SST variability is largely passive, being forced by the atmosphere but having only a modest impact on the atmospheric mode BSV08 ().

The question of the importance of ocean coupling is related to but not the same as that of the importance of surface fluxes to the dynamics of intraseasonal variability. If ocean coupling is important, surface fluxes must be involved, since only through those fluxes can the ocean influence the atmosphere. The converse is not true: an important role for surface fluxes does not necessarily imply that coupling is important. Surface flux feedbacks can operate in models which assume fixed SST. Such models do not satisfy a surface energy budget, but their surface fluxes can still vary interactively and influence the atmosphere. Coupling can either amplify or damp intraseasonal variability, depending on the phasing of the SST anomalies relative to anomalies in atmospheric variables. For example, Shinoda et al. (1998) found that observed SST anomalies slightly reduced the amplitude of MJO-related surface latent heat fluxes compared to what they would have been for fixed SST.

Our interest here is in the role of surface fluxes in the dynamics of atmospheric intraseasonal variability. Ocean coupling, while also arguably important, is secondary in this discussion. However, as discussed next, the nature of the underlying surface is important to the extent that it must have sufficiently large heat capacity to allow substantial fluctuations in the net surface enthalpy flux on the intraseasonal time scale.

2.5 Single column dynamics

Single column dynamics inferred from observations

The MJO is commonly defined as having large spatial scales. However, plots of intraseasonal variance in quantities related to deep convection also show relatively small-scale features as described in section 3. These smaller-scale features appear to be related to the nature of the underlying surface, and thus dynamically to be a result of the local interaction of that surface with the atmosphere. A simple framework within which to grasp these local interactions may be the idealized dynamics of a single column, consisting of the atmosphere and ocean in a relatively small horizontal area.

The single-column view is taken in the observational study of Waliser-1996:formation (), who showed in a composite analysis that an oceanic “hot spot”, defined as a region of at least km in which the sea surface temperature (SST) exceeds 29.75 for at least a month, typically appears after period of calm surface winds and clear skies. Anomalously strong surface winds and enhanced high cloudiness develop after the time of peak SST. Associated surface latent heat flux and shortwave radiative flux anomalies lead to the decay of the hot spot.

The conceptual model resulting from Waliser’s study, as well as similar ones articulated by subsequent studies (fasullo_webster_99, ; stephens_et_al_04, ), describe coupled oscillations occurring in a single column. They leave open to what extent the oscillation can be self-contained in a single region, as opposed to being fundamentally driven by the passage of large-scale disturbances. Observations suggest that the latter is a better description of the MJO (e.g., Hendon and Glick 1997; Zhang and Hendon 1997), as well as the northward-propagating Asian monsoon mode, since they have large-scale spatiotemporal structure and propagation within which regional-scale features are embedded. Nonetheless, the single-column view is useful for understanding some aspects of the controls on deep convection, and may be particularly relevant to understanding the smaller-scale regional features shown below.

A simple coupled single-column quasi-equilibrium model

Sobel and Gildor (2003, SG03) presented an explicit single column dynamical model of a simple atmosphere coupled to a slab mixed-layer ocean of constant depth. This model was designed to capture the behavior described by the observational studies described in section 2.5.1 above. Their model incorporates the standard assumptions of first baroclinic mode QE theory, as expressed in the “quasi-equilibrium tropical circulation model” (QTCM) formulation neezen00 (); zeng_et_al_00 (), plus a few additional assumptions. The most important additional assumption is that the local temperature profile — that is, not only the vertical structure of the temperature field, but also its value — is fixed. In this ”weak temperature gradient” (WTG) approximation (e.g., Held and Hoskins 1985; Neelin and Held 1987; Mapes and Houze 1995; Zeng and Neelin 1999; Sobel and Bretherton 2000; Sobel et al. 2001; Majda and Klein 2003) large-scale vertical motion is assumed to occur as needed in order for adiabatic cooling to balance diabatic heating, and the local temperature profile is held close to that of surrounding regions by a process which is essentially geostrophic adjustment with a small Coriolis parameter bretherton_smolarkiewicz_89 (). This parameterization of large-scale dynamics allows the precipitation to vary strongly in response to variations in SST, surface turbulent fluxes, and radiative cooling. In the absence of a large-scale circulation — for example, in radiative-convective equilibrium — large precipitation variations cannot occur, because any changes in convective heating have to be balanced by changes in radiative cooling, which cannot become too large. By employing the WTG approximation rather than an assumption of no large-scale circulation (as in, for example, Hu and Randall 1994, 1995), the SG03 model allows physically plausible, if highly parameterized, interactions between convection, radiation and large-scale dynamics to occur in a single column.

SG03 also assumed that deep convective clouds induced radiative perturbations which reduced the longwave cooling of the atmosphere and shortwave warming of the ocean surface in equal measure, so that the cloud-radiative perturbations play a role very similar to that of surface flux perturbations, as discussed above. SG03 very crudely represented the effect of surface flux feedbacks by assuming them to be local and lumping them together with radiative feedbacks, which in turn were parameterized as proportional to precipitation. They did this by increasing the proportionality coefficient, r, relating radiative anomalies to precipitation anomalies compared to that estimated from observations (bresob02, ; lin_mapes_04, ), arguing that the parameterized flux anomalies represented those in both radiative and wind-induced turbulent surface fluxes. Recent work suggests such increases are justifiable. Araligidad and Maloney (2008) found that intraseasonal latent heat flux anomalies alone are about 20% of precipitation anomalies in the west Pacific warm pool (e.g., , which appears at broadly consistent with the wind speed-precipitation relationship found on daily time scales by Back and Bretherton (2005).

With a proportionality coefficient of around 0.25 or greater — corresponding to a net 25 W m transfer of energy from ocean to atmosphere for each 100 W m of column-integrated latent heating — and other parameters set at typical control values, the model of SG03 is linearly unstable to free oscillations which qualitatively resemble those found in the observational studies (Waliser 1996; Fasullo and Webster 1999; Stephens et al. 2004). The growth rate of the oscillations is sensitive to several parameters in the model, including the surface enthalpy flux feedback parameter , the mixed layer depth, the time scale for convective adjustment, and the GMS (which is assumed constant), but the period is robustly in the intraseasonal range.

SG03 argued that their model on its own was not adequate to represent the MJO, as its single-column structure makes it incapable of capturing the MJO’s horizontal structure and propagation. They argued instead that their model might better depict how a small horizontal area responds to the passage of an MJO disturbance, with that disturbance viewed as an external forcing. With small enough to render the model stable — such as is appropriate if it represents radiative feedbacks alone — the model solution is the forced response of a damped oscillator. SG03 imposed the forcing through the atmospheric temperature field, which they took to have a sinusoidal variation with intraseasonal period. Maloney and Sobel (2004) instead imposed an intraseasonally fluctuating surface wind speed forcing in their application of the SG03 model, making it appropriate to set , representing radiation only.

Figure 1: Precipitation amplitude vs. mixed layer depth, SG03 model, from Maloney and Sobel (2004).

A parameter of particular interest in this model is the mixed layer depth. The amplitude of the model oscillations in precipitation as a function of mixed layer depth is presented in Figure 1, reproduced from Maloney and Sobel (2004). In this curve, the precipitation amplitude has a maximum at a particular value of the mixed layer depth, around 10-20 meters. It falls off slowly as mixed layer depth increases past the maximum, and rapidly as the mixed layer depth approaches zero. A similar (if weaker) amplitude maximum was found in the GCM results of Maloney and Sobel (2004) and is supported by a recent analysis of spatial and seasonal variability in the amplitude of intraseasonal variability compared to that in mixed layer depth (bellenger_duvel_07, ).

The amplitude decrease for mixed layers deeper than that at which the maximum occurs indicates that in this model, ocean coupling can (modestly) enhance intraseasonal variability, since infinite mixed layer depth corresponds to fixed SST. This decrease was also found in the GCM study of Watterson (2002), and is implied in those studies which find stronger intraseasonal variability in coupled models than in atmospheric models over fixed SST.

The vanishing of the response as mixed layer depth goes to zero reflects the fact that surface enthalpy fluxes are critical to the oscillations in this model. As the mixed layer depth approaches zero, the net surface enthalpy flux must also vanish. This kills the oscillation because the gross moist stability is positive, requiring net moist static energy input into the column (which is equivalent to net surface enthalpy flux under our assumptions) in order to generate circulation anomalies.

A mixed layer of zero depth, or ”swamp”, with zero heat capacity but an infinite moisture supply, may be thought of as a crude representation of a land surface in a tropical region during its monsoon season, although it is not a good representation of land surface processes in general. When soil becomes subsaturated, variations in the Bowen ratio (ratio of sensible to latent heat flux) can result in ”soil moisture memory” by which the land-atmosphere interactions have intrinsic time scales of up to several months. This effect appears most important in semi-arid regions (koster_suarez_01, ). In active monsoon regions, soil moisture memory is less important, and we assume that modeling the land surface dynamics by a swamp, with zero heat capacity but infinite available moisture, is adequate for purposes of understanding the qualitative dynamics of the coupled system. Treating each horizontal location as represented by an independent single column model under WTG and SQE (SG03), we arrive at the prediction that the amplitude of intraseasonal variability in deep convection can vary locally depending on surface type, and should be small over land and larger over ocean.

3 Observations

3.1 Intraseasonal variance maps


Figures 2 and 3 show maps of 30-90 day variance in precipitation from the TRMM 3B42 precipitation data set and outgoing longwave radiation (OLR) from the NOAA interpolated OLR data set, respectively, for the months November-April and May-October. Similar maps are shown in previous studies (e.g., Weickmann et al. 1985; Zhang and Hendon 1997; Vincent et al. 1998; Fasullo and Webster 1999; Sperber 2004). Daily-averaged TRMM precipitation data during 1998-2005 averaged to a grid are used. The TRMM 3B42 product we use here incorporates several satellite measurements, including the TMI and TRMM precipitation radar to calibrate infrared measurements from geostationary satellites (adler_et_al_00, ). The daily-averaged NOAA interpolated OLR product is used during 1979-2005 on a grid liebmann_smith_96 (). Unless otherwise stated, intraseasonal bandpass filtering is conducted using two 60-point non-recursive digital filters with half-power points at 30 and 90 days.

Figure 2: Intraseasonal variations in rainfall for a) November-April and b) May-October ().
Figure 3: Intraseasonal variations in OLR for a) November-April and b) May-October ().

During southern hemisphere summer, intraseasonal variability is dominated by the canonical MJO, which has very large horizontal scales (Figs. 2a and 3a). However, the variance maps also exhibit prominent smaller-scale patterns. These small-scale patterns consist primarily of enhanced intraseasonal variance over the oceans and reduced variance over land. Particularly striking is the land-sea contrast in the Maritime Continent region. This land-sea variance contrast is also prominent in northern summer ( 2b and 3b), and is evident down to the smallest scales resolved by the data.

Figure 4: Climatological rainfall for a) November-April and b) May-October ().
Figure 5: Climatological OLR for a) November-April and b) May-October ().

Figures 4 and 5 show the climatological mean precipitation and OLR, respectively for May-October and November-April. The patterns of the southern and northern hemisphere monsoons are evident. In May-October (4b and 5b), the climatological precipitation resembles the intraseasonal variance in its horizontal structure, with maxima in rainfall over the oceans and minima over land. In November-April, however, the same is not true (4a and 5a). Climatological convection maximizes over the large islands of the maritime continent region, while intraseasonal variance minimizes there. This tendency is most striking when examining the OLR product, although neither the patterns of intraseasonal variance nor those of climatological precipitation shown above is sensitive to the choice of data set. Similar patterns are apparent, for example, in the CMAP Xie-Arkin-1997:global () precipitation data set (not shown).

A proposed explanation of observed variance patterns

Despite the extreme simplicity of the SG03 model, the small-scale features in the observed patterns of intraseasonal precipitation variance are consistent with it, and thus with the assumptions of convective quasi-equilibrium and WTG which it incorporates. This is true in at least two important respects:

  1. The fine-scale structure in precipitation variance suggests that, despite the large-scale structure of the flow features associated with intraseasonal variability, it may be appropriate to consider variations in convection in terms of a local picture, which can be captured by a single-column model using WTG [or perhaps also by other single-column parameterizations of large-scale dynamics (e.g., Bergman and Sardeshmukh 2004; Mapes 2004)].

  2. The fact that intraseasonal precipitation variance maximizes over the ocean and minimizes over the land is consistent with an important role for interactive variations in the net surface enthalpy flux in generating the variance, since such variations can have significant amplitude over ocean but not over land.

The variations in net enthalpy flux most likely have turbulent and radiative components, corresponding to wind-evaporation and cloud-radiative feedbacks. It is not possible to determine which is more important on the basis of either the idealized model of SG03 or the patterns of variance alone. The observational analysis of Waliser (1996) suggests that the two components may be of comparable magnitude, while Hendon and Glick (1997) suggest that the relative importance of the two may vary with location.

Alternative explanations

An alternative explanation for the patterns shown in Figs. 2 and  3 is that the patterns are controlled by orographic effects rather than by land-sea contrasts in surface enthalpy fluxes. For example, in May-October, the patterns of intraseasonal precipitation variance (especially away from the maritime continent) resemble the patterns of mean rainfall, which are certainly influenced by orography. The orographic influence most likely is largely due to the dynamical forcing of upslope flow as monsoon winds impinge on mountain ranges, and is thus dynamically distinct from the thermodynamic effects of land-sea contrasts. Focusing on the Indian and Southeast Asian regions, one maximum (in both variance and mean rainfall) occurs over and just upstream of the Western Ghats, while another occurs over the Bay of Bengal, upstream of the mountains on the Cambodian coast. It might be argued that the variance minimum in between (fig. 2b), over the Indian subcontinent, owes its existence to the minimum in mean rainfall, and that the latter minimum owes its existence to orography. Southern India lies in the rain shadow of the Western Ghats, making it drier than the regions upstream and downstream. In general, where mean rainfall is smaller variability will also be smaller. A related argument follows from the analysis of Hoyos and Webster (2007), who present evidence both that much of the total precipitation falling in the Asian monsoon is associated with intraseasonal events, and that the precipitation distribution in these events is modulated by orography.

While orography undoubtedly influences the rainfall patterns shown above, orographic effects alone cannot explain all aspects of the intraseasonal variance maps shown in Figures 2 and  3. We contend that these patterns can be explained more generally by the land-sea difference in heat capacity, which in turn suggests a role for surface fluxes. This is particularly apparent when we consider the November-April and May-October results together, and look for the most general explanation for both. Consider the maritime continent region in November-April. Grossly speaking, intraseasonal variance maximizes over ocean and minimizes over land. Mean rainfall does the opposite, particularly over the largest Indonesian islands (though there is also more complex structure within individual islands which is surely influenced by topography). The large mountains on these islands most likely play a role in inducing the mean precipitation maxima (e.g., Qian 2008). If the structure of the intraseasonal variance were determined by the structure of the mean rainfall, we would expect to see variance maxima over these large islands, coincident with the mean rainfall maxima, but instead these are regions of relatively low variance. The dominance of surface type over orography in the determination of the intraseasonal variance patterns is also suggested by the pattern over and around northern Australia, where intraseasonal variance also maximizes over ocean to a greater extent than mean rainfall does. Northern Australia lacks significant orography, so it seems almost certain that this difference is due to land-sea contrast. Even in May-October, a primary role for land-sea difference in heat capacity rather than orography is suggested by the maximum variance to the east of the Philippines, which lies neither over nor immediately upstream of any mountains.

Besides orography, another explanation might be that convection over land is dominated by the diurnal cycle, and this disrupts the variability at intraseasonal time scales (e.g., Wang and Li 1994). This can be interpreted as essentially the same explanation that we present in section 2e, but in different language. If we force the SG03 single column model (for example) with insolation which varies at diurnal frequencies, the response maximizes at a very small mixed layer depth (not shown). Over land, due to the small heat capacity of the surface, the preferred frequency for coupled single-column oscillations is much higher than over ocean. Thus, the system responds more strongly to diurnal forcing and less strongly to intraseasonal forcing.

3.2 Correlation between surface latent heat flux and precipitation

Surface fluxes can be important to intraseasonal variability only if anomalous surface fluxes are able to influence the occurrence or intensity of deep convection. If this is the case, we might reasonably expect surface fluxes and precipitation to covary in space and time. The degree of covariance has been assessed in a couple of recent studies. Back and Bretherton (2004) showed that there is a small but significant correlation on the daily time scale between surface wind speed (which plays the dominant role in controlling the surface turbulent flux variations over tropical oceans) and precipitation, with the correlation being stronger for regions of high column water vapor content.

Figure 6: Scatterplot of intraseasonal (15-90 day filtered) TRMM 3B42 precipitation vs. TAO buoy latent heat flux at S, E, from Araligidad and Maloney (2008). Regression and correlation coefficients are indicated on the plots, and the black points represent binned averages, with bars depicting the 90% confidence limits about those averages.

On the intraseasonal time scale, surface latent heat flux and precipitation (or quantities related to it, such as OLR) have been found to be locally correlated. The peak correlation is typically found when latent heat flux lags convection by a week or so, though that optimal lag varies slightly from one study to the next (e.g., Hendon and Glick 1997, Shinoda et al. 1998, Woolnough et al. 2000) . Araligidad and Maloney (2008) demonstrated a strong instantaneous correlation ( 0.7) between November-April 30-90 day QuikSCAT wind speed and TRMM precipitation within the west Pacific regions of strong intraseasonal precipitation variance shown in Figures 2 and 3. Araligidad (2007) demonstrated similar strong correlations in the Indian Ocean during both summer and winter. Consistent with a strong covariance of precipitation and wind-driven fluxes, Araligidad and Maloney (2008) showed a significant correlation between intraseasonal TRMM precipitation anomalies and Tropical Atmosphere Ocean buoy latent heat flux anomalies. For example, Figure 6 is derived from Araligidad and Maloney (2008) and shows a scatterplot of intraseasonal latent heat flux versus precipitation anomalies at 8S, 165E during November-April of 1999-2005, within the band of strongest intraseasonal precipitation and OLR variance of Figures 2 and 3. If only the wind-driven portion of the latent heat flux anomaly is retained in this analysis, the correlation is about 0.1 higher, indicating that intraseasonal anomalies in air-sea humidity difference (forced primarily by SST variations) act to reduce the correlation of latent heat flux and precipitation versus if the wind-driven component were acting alone.

A positive covariance between precipitation and latent heat flux is encouraging regarding the ability for wind-evaporation feedbacks to support the MJO, in that it suggests that surface latent heat fluxes can influence convection. However, for surface flux feedbacks to drive the MJO, it is essential is for wind-induced fluxes to engender a positive covariance of intraseasonal tropospheric temperature and diabatic heating. Such a positive correlation would indicate eddy available potential energy (EAPE) generation, with subsequent conversion of EAPE to eddy kinetic energy supporting the large-scale MJO circulation against dissipation. It is difficult to diagnose these energy conversions accurately from observations, but the evidence from studies done to date suggests that the phase relationships are consistent with a role for surface enthalpy fluxes in the instability of the MJO. Hendon and Salby (1994) used satellite observations of OLR and tropospheric temperature to show that heating and temperature are positively correlated over the region of strong intraseasonal convective activity, being almost perfectly in phase in the Indian ocean where the MJO is growing in amplitude. Similar results were found in a more recent observational study by Yanai et al. (2000). Since surface latent heat flux lags precipitation only by a small amount (compared to the 30-60 day period of the mode), and radiative heating is exactly in phase with precipitation (within the accuracy of the observational estimates) the total surface enthalpy flux anomaly is positioned to induce convective heating anomalies with the correct phase to generate EAPE, particularly in the growing phase of the MJO life cycle.

4 GCM results

4.1 Previous work

The hypothesis that interactive surface energy flux feedbacks are essential to the dynamics of intraseasonal variability is testable in numerical models, under a perfect model assumption. This can be done by overriding the parameterizations which determine the net surface enthalpy flux, or its individual components, and forcing the fluxes to equal those from a climatology. The climatological fluxes can be taken from a control run of the same model having interactive fluxes. Prescribing fluxes in this way renders the surface flux feedbacks inactive, as the surface fluxes in the model are no longer a function of the instantaneous model variables. If surface flux feedbacks are essential to the model’s intraseasonal variability, that variability should be eliminated, or at least significantly reduced in amplitude. A number of variations on these experiments may be useful, such as one in which surface wind speed, rather than the turbulent surface fluxes themselves, is prescribed. Wind speed can either be set to a climatology or, in a simpler but less clean experiment shown below, to a spatially and temporally constant value.

To our knowledge, experiments of this type have been done only with a couple of recent-generation general circulation models using realistic basic state SST. Maloney (2002) performed an experiment in which the surface wind speed was set to its climatological value in the computation of surface turbulent heat fluxes. As intraseasonal variations in these fluxes are largely controlled by wind speed variations — the WISHE feedback — this eliminated most (but not all) intraseasonal flux variations. The surface latent heat flux itself was set to its climatological seasonal cycle in one simulation in Maloney and Sobel (2004). In that study, eliminating surface flux feedbacks significantly reduced the amplitude of the simulated MJO, indicating an important role for surface flux feedbacks. The MJO was not totally eliminated, but this could have been either due to fact that variations in the other components of the net surface energy flux (the radiative fluxes, and the turbulent flux variations due to air-sea humidity difference alone) were not suppressed, or due to dynamics unrelated to surface flux variations. In Maloney (2002), on the other hand, the elimination of WISHE actually increased the amplitude of eastward-propagating wind and precipitation variability. The complete disagreement between the results of these two studies is at first perplexing. The models used in them were rather similar. Both used the relaxed Arakawa-Schubert convection parameterization in successive versions of the NCAR Community Atmosphere model; the two models differed primarily in the treatments of convective downdrafts and cloud microphysics. However, the resulting relationships between intraseasonal convection and the large-scale anomalous circulation in these two models was significantly different, with enhanced convection occurring in anomalous easterlies and suppressed latent heat fluxes in the model of Maloney (2002), and in anomalous westerlies and enhanced latent heat fluxes (as observed) in the study of Maloney and Sobel (2004). Thus, removing wind-evaporation feedback might be expected to have different effects in these two models.

A few earlier GCM studies also tested the importance of surface turbulent and cloud-radiative feedbacks to intraseasonal variability (e.g. Hayashi and Golder 1986) , but given the considerable advances in simulation capability in the last two decades, it may be most productive to focus on results from more recently developed models. In some relatively recent GCM studies using zonally-symmetric SST distributions strong sensitivity to WISHE has been found (e.g. Hayashi and Golder 1997, Colon et al. 2002). It might be argued that the differences between the basic state wind fields in these calculations and the observed wind fields renders their relevance to real intraseasonal variability somewhat indirect. On the other hand, we do not understand that variability well enough to be sure what the role of the basic state is.

4.2 Results with the GFDL AM2

In this section we present results from new simulations with the Geophysical Fluid Dynamics Laboratory’s Atmospheric Model 2.1 (AM2.1), which is the atmospheric component of the coupled climate model CM2.1.

Model description

With the exception of the modification to the convection scheme that we describe below, the model used here is identical to that presented by the GFDL Global Atmospheric Model Development Team (andet04, ). It has a finite volume dynamical core, with horizontal resolution, and 24 vertical levels. The model is run over realistic geography and climatologically varying SSTs. The simulations are run for 11 years, with statistics taken over the last 10 years.

The convection scheme is a version of the Relaxed Arakawa-Schubert (RAS) scheme (moosua92, ). In the RAS scheme, convection is represented by a spectrum of entraining plumes, with a separate plume corresponding to each model level that can be reached by convection. The entrainment rates in these plumes are then determined by the requirement that the levels of neutral buoyancy of the plumes correspond to model levels. The convection scheme in AM2 also uses the modification of toket88 (), in which convection is not allowed to occur when the calculated entrainment rates are below a critical value determined by the depth of the subcloud layer , with . Thus with larger values of the Tokioka parameter , convection is prevented from reaching as deeply. Inspired by the results of toket88 () and linet08a () showing that larger values of the Tokioka parameter lead to stronger and more realistic MJO variability, we change from its standard AM2 value of to the control value of in order to increase the MJO variance in the model. We also show results using the standard AM2 value of the Tokioka parameter, in which the simulated MJO is weak.

In order to identify the importance of WISHE to the model MJO, we construct no-WISHE simulations by replacing the wind speed dependence in the surface flux formulation with a constant value representative of typical values over the tropics, everywhere. This modification is very simple to implement, as it only requires changing one line of code. However, in addition to preventing intraseasonal variations in turbulent surface fluxes (the intended effect) this modification also alters the climatology of the model, since the model’s actual climatological surface wind speed is not constant in either space or season. We therefore must confirm that the climatological precipitation distribution in our no-WISHE simulation does not become so different from that in the control model (or in observations) as to render the experiment irrelevant. The annual mean precipitation distributions for the control and no-WISHE cases are plotted in Figure 7. This figure shows that while there are some important changes in the precipitation distribution when WISHE is removed (e.g., less precipitation in the NW Pacific and more precipitation in the Indian Ocean), the distributions remain qualitatively similar enough to merit comparison of the MJO characteristics.

Figure 7: Annual mean precipitation distributions for the control and no-WISHE simulations GFDL AM2 with Tokioka parameter (see text for details).


Figure 8: Lag-correlation of averaged, 30-90 day filtered zonal wind at 850 hPa (U850) against the time series of the same field at , in GFDL AM2 with Tokioka parameter . The top panel shows results from the NCEP/NCAR reanalysis, the middle shows results from the model, and the bottom shows results from the model with no WISHE (see text for details).

As a first measure of the strength of the MJO with and without WISHE in these simulations, we show in Figure 8 lag-regression plots for 30-90 day filtered equatorial ( averaged) zonal wind at 850 (U850) for the months of November-April for observations, the control case (with ), and the control case without WISHE. Regression coefficients are scaled by the 1 sigma value of the reference time series, and stippling indicates where the correlation coefficient is significantly different from zero at the 95% confidence level. The control case MJO propagation is quite similar to observations in many aspects, including implied phase speeds of approximately 5 , large variance over the Indian and Pacific Ocean, and faster propagation over the central/eastern Pacific, though the regression coefficients are generally a bit weaker than those in observations, especially just to the west of the reference point. When WISHE is suppressed, the amplitude of the MJO is reduced signficantly. Only over the Pacific does any significant correlation exist away from the reference point. This clearly demonstrates that the MJO in this model is strongly influenced by WISHE.

As an alternative measure of the MJO intensity, we examine the intraseasonally averaged (30-90 days) wavenumber spectrum for U850, separated into eastward and westward propagating components. The ratio of eastward to westward variance at wavenumber 1 is often used as a measure of the MJO strength (see, e.g., zhaet06 ()). The ratio of eastward to westward variance at wavenumber 1 in the control case is 2.60, which is stronger than nearly all the atmosphere-only models in the zhaet06 () study, although weaker than observations, which have a value of 3.5. When WISHE is removed, this E/W ratio is reduced to 1.16, corroborating the result that WISHE is fundamentally important to the MJO in this model.

As a test of robustness to changes in model physics, we examine the same MJO diagnostics for the standard version of AM2.1, in which the Tokioka parameter . Examining lag correlations for this configuration in Figure 9a, one can clearly see weaker MJO propagation everywhere as compared to the Tokioka-modified control case in Figure 8b. Correlations are significantly weaker, especially in the Pacific basin. When WISHE is removed in this model configuration (Figure 9b), the model MJO is little affected. There is a small indication of decreased MJO correlations in the Indian Ocean and immediately downstream of the reference point, but these changes are subtle. The ratio of eastward to westward intraseasonal variance at wavenumber 1 for U850 is reduced from 1.31 to 1.08 when WISHE is removed, indicating a small decrease in MJO amplitude with WISHE in this diagnostic. Generally speaking, removing WISHE has a small effect on the MJO in this model configuration; but then, the MJO is weak to begin with.

Figure 9: Lag-correlation of U850 in GFDL AM2 , as in the lower two panels of fig. 8, but with Tokioka parameter .

The results from the model with bring to two the number of recent-generation models in which WISHE has been found to be important to the MJO in simulations with realistic basic states, the other being that used by Maloney and Sobel (2004). Both models use versions of the RAS convective parameterization, so they are not entirely unrelated, but most other aspects of the two models are different. In the models of Maloney (2002) and the AM2.1 with , WISHE is not important to the simulated MJO. On the other hand, these two models have MJO simulations which resemble observations less closely than do those of Maloney and Sobel (2004) and the AM2.1 simulations with . At least for this small sample of models, a better simulation of the MJO seems to be associated with an increased role for WISHE.

The simulations discussed above address only the role of the surface latent heat flux, and in the case of Maloney (2002) and the AM2.1 calculations described here, only the wind-induced component of that flux. A few , studies with full-physics GCMs over realistic of continents and sea surface temperature have assessed the role of radiative flux perturbations in simulated MJO dynamics. Lee et al. (2001) found in an aqua-planet GCM that overactive cloud-radiative feedbacks degraded the MJO simulation by inducing spurious small-scale disturbances; reasonable changes to the model’s physical parameterizations mitigated this degradation. Other studies have been done in more idealized frameworks. Raymond (2001) found that cloud-radiative feedbacks were essential to the MJO simulated in his intermediate-complexity model, with surface turbulent fluxes also playing a significant role. Grabowski (2003) found in aqua-planet simulations with what is now called the ”multiscale modeling framework” (MMF) that cloud-radiative feedbacks were not important to his simulated MJO disturbances, while surface latent flux feedbacks were essential to the disturbances’ development. Lin et al. (2008)  found no effect of cloud-radiative feedbacks on the MJO in a model in which the MJO was weak to begin with.

5 Discussion

5.1 The crux of the matter, in theoretical context

The claim that surface enthalpy fluxes are essential to the dynamics of tropical intraseasonal variability is not new, going back at least 20 years to the studies of Emanuel (1987) and Neelin et al. (1987). We believe that, given the lack of broad agreement on the mechanisms of the MJO (despite decades of intense effort) and the evidence from both observations and GCMs discussed above to support the hypothesis that surface fluxes are important, the time has come to reassess this hypothesis in a more focused way.

We have not provided a comprehensive discussion of all theories for tropical intraseasonal variability. However, it should be uncontroversial to state that in many of these theories, interactive surface fluxes either are not essential or are absent altogether. We propose that it would be useful to divide the current set of theories into two subsets, one in which feedbacks involving surface fluxes (including radiative fluxes) are essential and one in which they are not, and then attempt to eliminate one subset via focused numerical modeling studies, perhaps combined with further analysis of observations.

From a purely conceptual point of view, whether surface fluxes are essential to intraseasonal variability is a fundamental question. In extratropical dynamics, it has been found useful to divide the set of possible dynamical processes into those which are dry adiabatic and those which are not. If a phenomenon can be understood using adiabatic models, it is advantageous to do so. Similarly, it is natural when discussing tropical dynamics to divide the large set of possible processes into those which involve only deep convection and large-scale dynamics — that is, those which can be represented by moist adiabatic dynamics — and those in which diabatic processes external to both deep convection and large-scale dynamics, namely turbulent surface fluxes and radiative cooling, are involved. It has been a goal of theoretical tropical meteorology for several decades to determine whether the interaction of convection and large-scale dynamics alone can generate large-scale variability [as in early CISK models, as well as in more recent models with more complex physics (e.g., Mapes 2000; Majda and Shefter 2001; Kuang 2008)] or whether interaction with diabatic processes external to convection and large-scale dynamics is necessary (as in first baroclinic mode QE models). Determining whether interaction with turbulent surface fluxes and radiation is essential to observed intraseasonal variability in particular would be a major step forward in our understanding of the tropical atmosphere.

5.2 Proposal for further model intercomparison

Recent model intercomparisons (zhaet06, ; linet06, ), have been performed which summarize the state of the art in simulating the MJO in general circulation models. Without restating the results of these studies in detail, the MJO simulations in the latest generation of models are on average superior to those in previous generations (Slingo-Sperber-Boyle-et-al-1996:intraseasonal, ; Sperber-Slingo-Inness-et-al-1997:maintenance, ) in simulating eastward-propagating zonal wind variability in the tropics with a dominance of eastward vs. westward power, though even the best models still have deficiencies in their MJO simulations. While keeping model deficiencies in mind, we propose that interested modeling groups perform experiments like those described above, in which the total surface enthalpy flux, and ideally also its individual components, are set to climatology, eliminating feedbacks involving those fluxes. These experiments are likely to yield unambiguous information about the dynamics of a model’s intraseasonal variability. The negative of the quantitative change in the strength of the intraseasonal variability in these experiments provides a direct estimate of the role of the eliminated feedbacks in the dynamics of the simulated variability in the control simulation. Besides GCMs with parameterized physics, these experiments can also be done in models with resolved convection such as the multiscale modeling framework (randall_et_al_03, ; khairoutdinov_et_al_05, ; khairoutdinov_et_al_08, ) or global cloud resolving models (Miura et al. 2007). Both of these technologies are showing great promise in simulating the MJO, and sensitivity experiments to determine the roles of surface turbulent and radiative flux feedbacks in their results would be particularly valuable.

Because simulations of intraseasonal variability are imperfect in all models, such experiments will not yield unambiguous information about the dynamics of intraseasonal variability in the real atmosphere. It is entirely possible that any given model, or even an entire generation of models (given the broad similarities of approach found in common physical parameterizations in climate models), is getting something close to the right answer for the wrong reasons, so that the results of these experiments would be misleading. It is perhaps also equally probable that different models will yield different results from these experiments.

Neither will such experiments provide any direct information about how to improve the simulation of intraseasonal variability in any given model. The importance of surface flux feedbacks to the dynamics of intraseasonal variability may not be related in any simple way to any particular property of the physical parameterizations of a model, nor to any other aspect of its construction (e.g., resolution or the dynamical core). These feedbacks are arguably a high-level, or ”emergent” property of a given model. Even in the relatively simple models used in theoretical studies, it is often not apparent what determines the importance of surface flux feedbacks. We might expect it to be even less obvious in comprehensive GCMs.

If we were fortunate, the importance of surface flux feedbacks in a model would be related to that model’s ability to simulate intraseasonal variability, as is the case in the very small sample of models discussed in section 4. We can imagine a scatter plot in which one axis is the fidelity of modeled intraseasonal variability to that observed (how to quantify this is a separate problem which we do not address here); the other axis is the importance of surface flux feedbacks to intraseasonal variability in the same model, as quantified by minus the change in MJO amplitude in an experiment where the surface flux (either total, or a given component) is set to climatology; and each point represents one model. A significant positive slope to the best-fit regression line would suggest that surface flux feedbacks are important to dynamics of intraseasonal variability in the real atmosphere, while a significant negative slope would suggest the opposite. Lack of any significant slope, of course, would be an ambiguous result.

In any case, knowledge of the role of surface fluxes in simulated intraseasonal variability would be useful to model developers. It seems likely that any increase in physical understanding of the dynamics of the modeled intraseasonal variability, such as quantification of the role of surface flux feedbacks, would help to guide in the formulation of hypotheses about how to improve a model. For example, an active role for surface fluxes in regulating intraseasonal variability may compel modelers to further develop parameterizations coupling mesoscale perturbations of moist entropy and gustiness to the boundary layer (e.g. Jabouille et al. 1996; Redelsperger et al. 2000), where they may significantly affect surface fluxes during MJO events.

5.3 Theoretical challenges

A determination that interactive surface fluxes are essential to the dynamics of intraseasonal variability would not constitute a complete theory for that variability. Even if we were able to resolve the importance of surface fluxes, questions that would remain unanswered include (among others): What is the relative importance of turbulent vs. radiative fluxes? How should the physics of deep convection and other unresolved processes be parameterized in order to yield the correct feedback between the fluxes and the large-scale dynamics of the mode? Are the large-scale dynamics essentially linear or nonlinear? Is the structure of the basic state critical? What sets the phase speed of the disturbances? What is the role of ocean coupling? Theorists currently struggle with all of these questions. Proving or disproving the hypothesis that surface fluxes are essential to tropical intraseasonal variability would tell us that the ultimate energy source for the disturbances is the ocean mixed layer, and in doing so would eliminate a large subset of theories, but much theoretical work would be left to do.

Assuming that the wind-induced component of the surface turbulent fluxes is important, the dynamics by which these fluxes interact with the dynamics of eastward-propagating MJO disturbances must be different than that envisioned by Neelin et al. (1987) and Emanuel (1987). The real MJO is not a Kelvin wave (though it retains aspects of Kelvin wave dynamics), and the basic state surface winds in regions of strong tropical intraseasonal variability are westerly (e.g. Inness and Slingo 2003; Maloney and Esbensen 2007). One possibility is that nonlinear WISHE, rather than its linear counterpart, is acting. Studies which present numerical simulations with idealized or intermediate-complexity models (ray01, ; sugiyama_08a, ; sugiyama_08b, ) as well as comprehensive models (Maloney and Sobel 2004) provide suggestions of how this might work, but we do not have a simple analytical prototype model for nonlinear WISHE. Another possibility is that the dynamics are still fundamentally linear, but that fundamental changes to other aspects of the original E87 and N87 models (e.g., the identification of the MJO as a Kelvin wave)allow the requirement of mean easterlies to be relaxed. In the idealized model of Wang and Xie (1998), for example, the combination of coupling to a mixed layer ocean and parameterized radiative feedbacks is linearly destabilizing in the presence of mean westerlies.

In the case of the northern summer northward-propagating mode, there may be no fundamental theoretical problem. In at least one idealized model, a mode resembling that observed is linearly destabilized by WISHE (Bellon and Sobel 2008a, b).

6 Summary

We have argued that feedbacks involving the total surface enthalpy flux are important to the dynamics of tropical intraseasonal variability, possibly providing the primary energy source for intraseasonal disturbances. Observational evidence presented in support of this argument consists of maps of intraseasonal variance in precipitation and OLR as well as local correlations between precipitation and surface fluxes on intraseasonal time scales. Modeling evidence consists of results from several GCMs as well as idealized models in which surface flux feedbacks are demonstrably important if not essential to simulated intraseasonal variability.

Our understanding of the mechanisms responsible for intraseasonal variability is still poor after decades of study. We have argued that the time has come for a more systematic evaluation of the role of surface enthalpy fluxes, given all the tools at hand, with the aim of eliminating from consideration either those hypotheses in which surface fluxes are important or those in which they are not. Given the evidence presented here, the increasing fidelity with which comprehensive numerical models simulate intraseasonal variability, and the relative straightforwardness of assessing the importance of surface flux feedbacks in those models, we have argued that it would be particularly useful if a larger number of interested modeling groups were to perform the necessary assessments. Such efforts, combined with targeted observational and theoretical work, might enable the field to move forward in a more coordinated and productive way towards a better understanding of tropical intraseasonal variability.


This paper was written while the first author spent a sabbatical year as a visitor at the Centre for Australian Climate and Weather Research (CAWCR), Bureau of Meteorology (BoM), Melbourne, Australia. Some of the writing was done during a relatively brief but very stimulating visit (during the peak of an active MJO event) to the BoM’s Northern Territory Regional Office in Darwin. He (AHS) is grateful to the BoM for its hospitality during this year, and to its scientists Harry Hendon, Matt Wheeler, and Hongyan Zhu (Melbourne) and Sam Cleland and Lori Chappell (Darwin) for stimulating discussions on the topic of this paper. This work was supported by the Climate and Large-Scale Dynamics Program of the National Science Foundation under grants ATM-0632341 (EDM) and ATM-054273 (AHS), by the Precipitation Measurement Mission program of the National Aeronautics and Space Administration under grant NNX07AD21G (AHS) and by award NA05OAR4310006 from the National Oceanic and Atmospheric Administration, U.S. Department of Commerce (EDM). The statements, findings, conclusions, and recommendations do not necessarily reflect the views of NSF, NASA, NOAA, or of the Department of Commerce.


  1. Corresponding Author Address: Adam Sobel, Columbia University, Dept. of Applied Physics and Applied Mathematics, 500 W. 120th St., Rm. 217, New York, NY 10027 USA. E-mail: ahs129@columbia.edu


  1. Adler, R. F., G. J. Huffman, D. T. Bolvin, S. Curtis, and E. J. Nelkin, 2000: Tropical rainfall distributions determined using TRMM combined with other satellite and rain gauge information. J. Appl. Met. , 39,2007–2023.
  2. Anderson, J. L., and Co-authors, 2004: The new GFDL global atmosphere and land model AM2-LM2: Evaluation with prescribed SST simulation. J. Climate, 17,4641–4673.
  3. Arakawa, A., 2004: The cumulus parameterization problem: Past, present and future. J. Climate, 17,2493–2525.
  4. Araligidad, N. M., and E. D. Maloney, 2008: Wind-driven latent heat flux and the intraseasonal oscillation. in press, Geophys. Res. Lett.
  5. Back, L. E., and C. S. Bretherton, 2005: The relationship between wind speed and precipitation in the Pacific ITCZ. J. Climate, 18,4317–4328.
  6. Back, L. E., and C. S. Bretherton, 2006: Geographic variability in the export of moist static energy and vertical motion profiles in the tropical Pacific. Geophys. Res. Lett., 33,doi:10.1029/2006GL026672.
  7. Bellenger, H., and J. P. Duvel, 2007: Intraseasonal convective perturbations related to the seasonal march of the Indo-Pacific monsoons. J. Climate, 20,2853–2863.
  8. Bellon, G., and A. H. Sobel, 2008a: Instability of the axisymmetric monsoon flow and intraseasonal oscillation. Journal of Geophysical Research, 113,D07109, doi:10.1029/2007JD008968.
  9. Bellon, G., and A. H. Sobel, 2008b: Poleward-propagating intraseasonal monsoon disturbances in an intermediate-complexity axisymmetric model. Journal of the Atmospheric Sciences, 65,470–489.
  10. Bellon, G., A. H. Sobel, and J. Vialard, 2008: Ocean-atmosphere coupling in the monsoon intraseasonal oscillation: a simple model study. Journal of Climate, p. in press.
  11. Bellon, G., and J. Srinivasan, 2006: Comment on ”Structures and mechanisms of the northward propagating boreal summer intraseasonal oscillation”. Journal of Climate, 19,4738–4743.
  12. Bergman, J. W., and P. D. Sardeshmukh, 2004: Dynamic stabilization of atmospheric single-column models. J. Climate, 17,1004–1021.
  13. Bretherton, C. S., and P. K. Smolarkiewicz, 1989: Gravity waves, compensating subsidence and detrainment around cumulus clouds. J. Atmos. Sci., 46,740–759.
  14. Bretherton, C. S., and A. H. Sobel, 2002: A simple model of a convectively coupled Walker Circulation using the weak temperature gradient approximation. J. Climate, 15,2907–2920.
  15. Chatterjee, P., and B. N. Goswami, 2004: Structure, genesis and scale selection of the tropical quasi-biweekly mode. Quarterly Journal of the Royal Meteorological Society, 130,1171–1194.
  16. Drbohlav, H.-K., and B. Wang, 2005: Mechanism of the northward-propagating intraseasonal oscillation: Insights from a zonally symmetric model. Journal of Climate, 18,952–972.
  17. Emanuel, K. A., 1987: An air-sea interaction model of intraseasonal oscillations in the tropics. J. Atmos. Sci., 44,2324–2340.
  18. Emanuel, K. A., 1988: Reply to Bin Wang. J. Atmos. Sci., 45,3528–3530.
  19. Emanuel, K. A., 2007: Quasi-equilibrium dynamics of the tropical atmosphere. The Global Circulation of the Atmosphere. Princeton University Press.
  20. Emanuel, K. A., J. D. Neelin, and C. S. Bretherton, 1994: On large-scale circulations in convecting atmospheres. Quart. J. Roy. Meteor. Soc., 120,1111–1143.
  21. Fasullo, J., and P. J. Webster, 1999: Warm pool SST variability in relation to the surface energy balance. J. Climate., 12,1292–1305.
  22. Frierson, D. M. W., 2007a: Convectively coupled Kelvin waves in an idealized moist general circulation model . In press, J. Atmos. Sci.
  23. Frierson, D. M. W., 2007b: The dynamics of idealized convection schemes and their effect on the zonally averaged tropical circulation. In press, J. Atmos. Sci.
  24. Fu, X., B. Wang, D. E. Waliser, and L. Tao, 2007: Impact of atmosphere-ocean coupling on the predictability of monsoon intraseasonal oscillations. J. Atmos. Sci., 64,157–174.
  25. Fuchs, Z., and D. J. Raymond, 2002: Large-scale modes of a nonrotating atmosphere with water vapor and cloud-radiation feedbacks. J. Atmos. Sci., 59,1669–1679.
  26. Gadgil, S., and J. Srinivasan, 1990: Low frequency variation of tropical convergence zone. Meteorology and Atmospheric Physics, 44,119–132.
  27. Goswami, B. N., 2005: South Asian monsoon. Intraseasonal variability in the Atmosphere-Ocean Climate system, W. K. M. Lau, and D. E. Waliser, Eds., pp. 19–61. Springer, Berlin, Heidelberg.
  28. Goswami, B. N., and J. Shukla, 1984: Quasi-periodic oscillations in a symmetric general circulation model. Journal of the Atmospheric Sciences, 41,20–37.
  29. Grabowski, W. W., 2003: MJO-like coherent structures: sensitivity simulations using the cloud-resolving convection parameterization (CRCP). J. Atmos. Sci., 60,847–864.
  30. Harrison, E. F., P. Minnis, B. R. Barkstrom, V. Ramanathan, R. D. Cess, and G. G. Gibson, 1990: Seasonal variation of cloud-radiative forcing derived from the Earth Radiation Budget Experiment. J. Geophys. Res., 95,18687–18703.
  31. Hartmann, D. L., L. A. Moy, and Q. Fu, 2001: Tropical convection and the energy balance at the top of the atmosphere. J. Climate, 14,4495–4511.
  32. Hayashi, Y., and D. G. Golder, 1986: Tropical intraseasonal oscillations appearing in a GFDL general circulation model and FGGE data. Part I: Phase propagation. J. Atmos. Sci., 43,3058–3067.
  33. Hayashi, Y., and D. G. Golder, 1997: United mechanisms for the generation of low- and high-frequency tropical waves. Part I: Control experiments with moist convective adjustment. J. Atmos. Sci., 54,1262–1276.
  34. Held, I. M., and B. J. Hoskins, 1985: Large-scale eddies and the general circulation of the troposphere. Advances in Geophysics, 28,3–31.
  35. Hendon, H. H., 2000: Impact of air-sea coupling on the Madden-Julian oscillation in a general circulation model. J. Atmos. Sci., 57,3939–3952.
  36. Hendon, H. H., and J. Glick, 1997: Intraseasonal air-sea interaction in the tropical Indian and Pacific oceans. J. Climate, 10,647–661.
  37. Hendon, H. H., and M. L. Salby, 1994: The life cycle of the Madden-Julian oscillation. J. Atmos. Sci., 51,2225–2237.
  38. Hoskins, B. J., M. E. McIntyre, and A. W. Robertson, 1985: On the use and significance of isentropic potential vorticity maps. Quart. J. Roy. Meteor. Soc., 111,877–946.
  39. Hoyos, C. D., and P. J. Webster, 2007: The role of intraseasonal variability in the nature of Asian monsoon precipitation. J. Climate, 20,4402–4424.
  40. Hu, Q., and D. Randall, 1994: Low-frequency oscillations in radiative-convective systems. J. Atmos. Sci., 51,1089–1099.
  41. Hu, Q., and D. Randall, 1995: Low-frequency oscillations in radiative-convective systems. Part II: An idealized model. J. Atmos. Sci., 52,478–490.
  42. Inness, P. M., and J. M. Slingo, 2003: Simulation of the Madden-Julian oscillation in a coupled general circulation model. Part I: Comparison with observations and an atmosphere-only GCM. J. Climate, 16,345–364.
  43. Jabouille, P., J.-L. Redelsperger, and J. P. Lafore, 1996: Modification of surface fluxes by atmospheric convection in the TOGA COARE region. J. Atmos. Sci., 11,2116–2135.
  44. Jiang, X., and T. Li, 2005: Reinitiation of the boreal summer intraseasonal oscillation in the tropical Indian Ocean. Journal of Climate, 18,3777–3795.
  45. Jiang, X., T. Li, and B. Wang, 2004: Structures and mechanisms of the northward propagating boreal summer intraseasonal oscillation. Journal of Climate, 17,1022–1039.
  46. Kemball-Cook, S., B. Wang, and X. Fu, 2002: Simulation of the intraseasonal oscillation in the ECHAM-4 model: the impact of coupling with an ocean model. J. Atmos. Sci., 59,1433–1453.
  47. Khairoutdinov, M., C. DeMott, and D. A. Randall, 2008: Evaluation of the simulated interannual and subseasonal variability in an AMIP-style simulation using the CSU multiscale modeling framework. J. Climate, 21,413–431.
  48. Khairoutdinov, M., D. A. Randall, and C. DeMott, 2005: Simulations of the atmospheric general circulation using a cloud-resolving model as a superparameterization of physical processes. J. Atmos. Sci., 62,2136–2154.
  49. Khouider, B., and A. J. Majda, 2006a: A simple multicloud parameterization for convectively coupled tropical waves. Part I: Linear analysis. J. Atmos. Sci., 63,1308–1323.
  50. Khouider, B., and A. J. Majda, 2006b: Multicloud parameterizations with crude vertical structure. Theoret. Comp. Fluid Dyn., 20,351–375.
  51. Kiladis, G. N., G. A. Meehl, and K. M. Weickmann, 1994: Large-scale circulation associated with westerly wind bursts and deep convection over the western equatorial Pacific. J. Geophys. Res., 99,18527–18544.
  52. Koster, R. D., and M. J. Suarez, 2001: Soil moisture memory in climate models. J. Hydromet., 2,558–570.
  53. Kuang, Z., 2008: A moisture-stratiform instability for convectively coupled waves. J. Atmos. Sci., 65,834–854.
  54. Lawrence, D. M., and P. Webster, 2002: The boreal summer intraseasonal oscillation: Relationship between northward and eastward movement of convection. Journal of the Atmospheric Sciences, 59,1593–1606.
  55. Lee, M.-I., I.-S. Kang, J.-K. Kim, and B. E. Mapes, 2001: Influence of cloud-radiation interaction on simulating tropical intraseasonal oscillation with an atmospheric general circulation model. J. Geophys. Res., 106,14219–14233.
  56. Liebmann, B., and C. A. Smith, 1996: Description of a complete (interpolated) outgoing longwave radiation dataset. Bull. Amer. Meteor. Soc., 77,1275–1277.
  57. Lin, J., and coauthors, 2006: Tropical intraseasonal variability in 14 IPCC AR4 climate models. Part I: Convective signals. J. Climate, 19,2665–2690.
  58. Lin, J., and B. E. Mapes, 2004: Radiation budget of the tropical intraseasonal oscillation. J. Atmos. Sci., 61,2050–2062.
  59. Lin, J.-L., D. Kim, M.-I. Lee, and I.-S. Kang, 2008a: Effects of cloud-radiative heating on AGCM simulations of convectively coupled equatorial waves. in press, J. Geophys. Res.
  60. Lin, J. L., D. Kim, M.-I. Lee, I.-S. Kang, and D. M. W. Frierson, 2008b: The impacts of convective parameterization and moisture triggering on AGCM-simulated convectively coupled equatorial waves. In press, J. Climate.
  61. Majda, A. J., and R. Klein, 2003: Systematic multiscale models for the tropics. J. Atmos. Sci., 60,393–408.
  62. Majda, A. J., and M. Shefter, 2001: Waves and instabilities for model tropical convective parameterizations. J. Atmos. Sci., 58,896–914.
  63. Maloney, E. D., 2002: An intraseasonal oscillation composite lifecycle in the NCAR CCM3.6 with modified convection. J. Climate, 15,964–982.
  64. Maloney, E. D., and S. K. Esbensen, 2007: Satellite and buoy observations of boreal summer intraseasonal variability in the tropical northeast Pacific. Mon. Wea. Rev., 135,3–19.
  65. Maloney, E. D., and A. H. Sobel, 2004: Surface fluxes and ocean coupling in the tropical intraseasonal oscillation. J. Climate, 17,4368–4386.
  66. Mapes, B. E., 2000: Convective inhibition, subgrid-scale triggering energy, and stratiform instability in a toy tropical wave model. J. Atmos. Sci., 57,1515–1535.
  67. Mapes, B. E., 2004: Sensitivities of cumulus-ensemble rainfall in a cloud-resolving model with parameterized large-scale dynamics. J. Atmos. Sci., 61,2308–2317.
  68. Mapes, B. E., and R. Houze, 1995: Diabatic divergence profiles in western Pacific mesoscale convective systems. J. Atmos. Sci., 52,1807–1828.
  69. Miura, H., M. Satoh, T. Nasuno, A. T. Noda, and K. Oouchi, 2007: A Madden-Julian oscillation event realistically simulated by a global cloud-resolving model. Science, 318,1763–1765, doi: 10.1126/science.1148443.
  70. Moorthi, S., and M. J. Suarez, 1992: Relaxed Arakawa-Schubert: A parameterization of moist convection for general circulation models. Mon. Wea. Rev., 120,978–1002.
  71. Nanjundiah, R., J. Srinivasan, S. Gadgil, and P. Webster, 1992: Intraseasonal variation of the Indian summer monsoon. Part II: Theoretical aspects. Journal of the Meteorological Society of Japan, 70,529–550.
  72. Neelin, J., and J. Yu, 1994: Modes of tropical variability under convective adjustment and the Madden-Julian Oscillation. Part I: Analytical theory. J. Atmos. Sci., 51,1876–1894.
  73. Neelin, J. D., 1997: Implications of convective quasi-equilibrium for the large-scale flow. The Physics and Parameterization of Moist Atmospheric Convection, pp. 413–446. Kluwer.
  74. Neelin, J. D., 2007: Moist dynamics of tropical convection zones in monsoons, teleconnections, and global warming. The Global Circulation of the Atmosphere. Princeton University Press.
  75. Neelin, J. D., and I. M. Held, 1987: Modeling tropical convergence based on the moist static energy budget. Mon. Wea. Rev., 115,3–12.
  76. Neelin, J. D., I. M. Held, and K. H. Cook, 1987: Evaporation-wind feedback and low-frequency variability in the tropical atmosphere. J. Atmos. Sci., 44,2341–2348.
  77. Neelin, J. D., and N. Zeng, 2000: A quasi-equilibrium tropical circulation model – formulation. J. Atmos. Sci., 57,1741–1766.
  78. Qian, J.-H., 2008: Role of the diurnal cycle in precipitation distribution over the maritime continent. Part I: Why precipitation is mostly concentrated over islands. submitted, J. Atmos. Sci.
  79. Ramanathan, V., R. D. Cess, E. F. Harrison, P. Minnis, B. R. Barkstrom, E. Ahmad, and D. Hartmann, 1989: Cloud-radiative forcing and climate: Results from the Earth Radiation Budget Experiment. Science, 243,57–63.
  80. Randall, D. A., M. Khairoutdinov, A. Arakawa, and W. W. Grabowski, 2003: Breaking the cloud parameterization deadlock. Bull. Amer. Meteor. Soc., 84,1547–1564.
  81. Raymond, D. J., 2000: Thermodynamic control of tropical rainfall. Quart. J. Roy. Meteor. Soc., 126,889–898.
  82. Raymond, D. J., 2001: A new model of the Madden-Julian Oscillation. J. Atmos. Sci., 58,2807–2819.
  83. Roxy, M., and Y. Tanimoto, 2007: Role of SST over the Indian ocean in influencing intraseasonal variability of the indian summer monsoon. J. Met. Soc. Japan, 85,349–358.
  84. Seo, K.-H., J.-K. E. Schemm, W. Wang, and A. Kumar, 2007: The boreal summer intraseasonal oscillation simulated in the NCEP climate forecast system: the effect of sea surface temperature. Mon. Wea. Rev., 135,1807–1827.
  85. Shinoda, T., and H. H. Hendon, 1998: Mixed layer modeling of intraseasonal variability in the tropical western Pacific and Indian oceans. J. Climate, 11,2668–2685.
  86. Shinoda, T., H. H. Hendon, and J. Glick, 2008: Intraseasonal variability of surface fluxes and sea surface temperature in the tropical western Pacific and Indian oceans. J. Climate, 11,1685–1702.
  87. Slingo, J. M., K. R. Sperber, J. S. Boyle, J. P. Ceron, M. Dix, B. Dugas, W. Ebisuzaki, J. Fyfe, D. Gregory, J. F. Gueremy, J. Hack, A. Harzallah, P. Inness, A. Kitoh, W. K. M. Lau, B. Mcavaney, R. Madden, A. Matthews, T. N. Palmer, C. K. Park, D. Randall, and N. Renno, 1996: Intraseasonal oscillations in 15 atmospheric general circulation models: Results from an AMIP diagnostic subproject. Clim. Dyn., 12,325–357.
  88. Smith, R. K., 1997: On the theory of CISK. Quart. J. Royal Met. Soc., 123,407–418.
  89. Sobel, A. H., 2007: Simple models of ensemble-averaged tropical precipitation and surface wind. The Global Circulation of the Atmosphere. Princeton University Press.
  90. Sobel, A. H., and C. S. Bretherton, 2000: Modeling tropical precipitation in a single column. J. Climate, 13,4378–4392.
  91. Sobel, A. H., and C. S. Bretherton, 2003: Large-scale waves interacting with deep convection in idealized mesoscale model simulations. Tellus, 55,45–60.
  92. Sobel, A. H., and H. Gildor, 2003: A simple time-dependent model of SST hot spots. J. Climate, 16,3978–3992.
  93. Sobel, A. H., and J. D. Neelin, 2006: The boundary layer contribution to intertropical convergence zones in the quasi-equilibrium tropical circulation model framework. Theor. Comp. Fluid Dyn., 20,323–350.
  94. Sobel, A. H., J. Nilsson, and L. M. Polvani, 2001: The weak temperature gradient approximation and balanced tropical moisture waves. J. Atmos. Sci., 58,3650–3665.
  95. Sperber, K. R., 2004: Madden-Julian variability in NCAR CAM2.0 and CCSM2.0. Climate Dyn., 23,259–278.
  96. Sperber, K. R., J. M. Slingo, P. M. Inness, and W. K. M. Lau, 1997: On the maintenance and initiation of the intraseasonal oscillation in the NCEP/NCAR reanalysis and in the GLA and UKMO AMIP simulations. Clim. Dyn., 13,769–795.
  97. Srinivasan, J., S. Gadgil, and P. J. Webster, 1993: Meridional propagation of large-scale monsoon convective zones. Meteorology and Atmospheric Physics, 52,15–35.
  98. Stephens, G. L., P. J. Webster, R. H. Johnson, R. Engelen, and T. L’Ecuyer, 2004: Observational evidence for the mutual regulation of the tropical hydrological cycle and tropical sea surface temperatures. J. Climate., 17,2213–2224.
  99. Stevens, B., D. A. Randall, X. Lin, and M. T. Montgomery, 1997: A comment on: On large-scale circulations in convecting atmospheres by Emanuel, Neelin and Bretherton. Quart. J. Royal Met. Soc., 123,1771–1778.
  100. Sugiyama, M., 2008a: Moisture mode in the tropics. Part I: Analysis based on the weak temperature gradient approximation. submitted, J. Atmos. Sci.
  101. Sugiyama, M., 2008b: Moisture mode in the tropics. Part II: Nonlinear behavior on an equatorial -plane. submitted, J. Atmos. Sci.
  102. Tokioka, T., K. Yamazaki, A. Kitoh, and T. Ose, 1988: The equatorial 30-60 day oscillation and the Arakawa-Schubert penetrative cumulus parameterization. J. Meteor. Soc. Japan, 66,883–901.
  103. Tompkins, A. M., 2001: Organization of tropical convection in low vertical wind shears: the role of water vapor. J. Atmos. Sci.., 58,529–545.
  104. Vecchi, G. A., and D. E. Harrison, 2002: Monsoon breaks and subseasonal sea surface temperature variability in the Bay of Bengal. J. Climate, 15,1485–1493.
  105. Vincent, D. G., A. Fink, J. M. Schrage, and P. Speth, 1998: High- and low-frequency intraseasonal variance of OLR on annual and ENSO timescales. J. Climate, 11,968–986.
  106. Waliser, D. E., 1996: Formation and limiting mechanisms for very high sea surface temperature: Linking the dynamics and the thermodynamics. J. Climate, 9,161–188.
  107. Waliser, D. E., K. M. Lau, and J. H. Kim, 1999: The influence of coupled sea surface temperatures on the Madden-Julian oscillation: A model perturbation experiment. J. Atmos. Sci., 56,333–358.
  108. Wang, B., 1988: Comments on ”An air-sea interaction model of intraseasonal oscillation in the tropics. J. Atmos. Sci., 45,3521–3525.
  109. Wang, B., 2005: Theory. Intraseasonal Variability in the Atmosphere-Ocean Climate System, p. chapter 10. Praxis Publishing.
  110. Wang, B., and T. Li, 1994: Convective interaction with boundary-layer dynamics in the development of a tropical intraseasonal system. J. Atmos. Sci. , 51,1386–1400.
  111. Wang, B., and H. Rui, 1990: Synoptic climatology of transient tropical intraseasonal convection anomalies: 1975–1985. Meteorology and Atmospheric Physics, 44,43–61.
  112. Wang, B., P. J. Webster, K. Kikuchi, T. Yasunari, and Y. Qi, 2006: Boreal summer quasi-monthly oscillation in the global tropics. Climate Dynamics, 27,661–675.
  113. Wang, B., and X. Xie, 1998: Coupled modes of the warm pool climate system. Part I: The role of air-sea interaction in maintaining Madden-Julian oscillation. J. Climate., 11,2116–2135.
  114. Watterson, I. G., 2002: The sensitivity of subannual and intraseasonal tropical variability to model ocean mixed layer depth. J. Geophys. Res., 107,doi:10.1029/2001JD000671.
  115. Webster, P. J., 1983: Mechanism of monsoon low-frequency variability: Surface hydrological effects. Journal of the Atmospheric Sciences, 40,2110–2124.
  116. Webster, P. J., and L. C. Chou, 1980: Low-frequency transitions in a simple monsoon system. Journal of the Atmospheric Sciences, 37,368–382.
  117. Weickmann, K. M., G. R. Lussky, and J. E. Kutzbach, 1985: Intraseasonal (30-60 day) fluctuations of outgoing longwave radiation and 250 mb streamfunction during northern winter. Mon. Wea. Rev.., 113,941–961.
  118. Wheeler, M., and G. N. Kiladis, 1999: Convectively coupled equatorial waves: Analysis of clouds and temperature in the wavenumber-frequency domain. J. Atmos. Sci., 56,374–399.
  119. Woolnough, S. J., J. M. Slingo, and B. J. Hoskins, 2000: The relationship between convection and sea surface temperature on intraseasonal timescales. J. Climate, 13,2086–2104.
  120. Xie, P. P., and P. A. Arkin, 1997: Global precipitation: A 17-year monthly analysis based on gauge observations, satellite estimates, and numerical model outputs. Bulletin of the American Meteorological Society, 78,2539–2558.
  121. Yanai, M., B. Chen, and W.-W. Tung, 2000: The Madden-Julian oscillation observed during the TOGA COARE IOP: Global view. J. Atmos. Sci., 57,2734–2396.
  122. Yu, J. Y., C. Chou, and J. D. Neelin, 1998: Estimating the gross moist stability of the tropical atmosphere. J. Atmos. Sci., 55,1354–1372.
  123. Zeng, N., and J. D. Neelin, 1999: A land-atmosphere interaction theory for the tropical deforestation problem. J. Climate, 12,857–872.
  124. Zeng, N., J. D. Neelin, and C. Chou, 2000: A quasi-equilibrium tropical circulation model – implementation and simulation. J. Atmos. Sci., 57,1767–1796.
  125. Zhang, C., 2005: Madden-Julian oscillation. Rev. Geophys., 43,doi:10.1029/2004RG000158.
  126. Zhang, C., M. Dong, H. H. Hendon, E. D. Maloney, A. Marshall, K. R. Sperber, and W. Wang, 2006: Simulations of the Madden-Julian Oscillation in four pairs of coupled and uncoupled models. Clim. Dyn., 27,573–592.
  127. Zhang, C., and M. McPhaden, 2000: Intraseasonal surface cooling in the equatorial western Pacific. J. Climate, 13,2261–2276.
  128. Zhang, C. D., and H. H. Hendon, 1997: Propagating and standing components of the intraseasonal oscillation in tropical convection. J. Atmos. Sci., 54,741–752.
  129. Zheng, Y., D. E. Waliser, W. F. Stern, and C. Jones, 2004: The role of coupled sea surface temperatures in the simulation of the tropical intraseasonal oscillation. J. Climate, 17,4109–4134.
Comments 0
Request Comment
You are adding the first comment!
How to quickly get a good reply:
  • Give credit where it’s due by listing out the positive aspects of a paper before getting into which changes should be made.
  • Be specific in your critique, and provide supporting evidence with appropriate references to substantiate general statements.
  • Your comment should inspire ideas to flow and help the author improves the paper.

The better we are at sharing our knowledge with each other, the faster we move forward.
The feedback must be of minimum 40 characters and the title a minimum of 5 characters
Add comment
Loading ...
This is a comment super asjknd jkasnjk adsnkj
The feedback must be of minumum 40 characters
The feedback must be of minumum 40 characters

You are asking your first question!
How to quickly get a good answer:
  • Keep your question short and to the point
  • Check for grammar or spelling errors.
  • Phrase it like a question
Test description