Atmosphereocean interaction

The TOGA-COARE has provided firm evidence that active air-sea interaction occurred during two MJO events in late 1992 and early 1993. The coupled structure of the MJO and oceanic mixed layer variability were documented in the mid-1990s (see Section 10.6 and Chapter 7). Stimulated by these observations, there has been a surge of theoretical and numerical model studies of the nature of the air-sea interaction in the Indo-Pacific warm oceans and its role in the development of ISO.

In theoretical model studies, Wang and Xie (1998) introduced a simple coupled atmosphere-ocean model suitable for study of the coupled instability of the warm pool climate system. This model emphasizes oceanic mixed layer physics and its thermodynamic coupling (through surface heat exchanges) with transient atmospheric motion. The fastest growing coupled mode in the model has a planetary zonal scale and an intraseasonal timescale, as well as a realistic SST-convection relationship. The wind-evaporation/entrainment feedback plays a primary role in generating the coupled instability, while the contribution of cloud-SST coupling becomes significant only when the wind effect is weak (see Section 10.6). Recently, Sobel and Gildor (2003) introduced a simple model for the evolution of SST in a localized region of a warm ocean. The model consists of a 0-D atmosphere coupled to an ocean mixed layer. For plausible parameter values, the steady state of the system can oscillate with periods ranging from intraseasonal to sub-annual. The basic mechanism for the instability and oscillation comes from cloud-radiative and wind-evaporation feedback, which agree qualitatively with the model results of Wang and Xie (1998). In their model, however, these two processes play the same roles. This latter conclusion may be due to the neglect of atmospheric dynamics. In the presence of atmospheric dynamics, the regions of active interaction associated with these two feedback processes would have a spatial phase shift with their roles being different as discussed by Wang and Zhang (2002).

In numerical modeling studies, Flatau et al. (1997) used an atmospheric global circulation model with a parameterized, empirical relationship between wind speed and SST tendency to examine the effect of SST feedback on the equatorial convection on an aqua-planet. The model MJO-like fluctuations were slowed down and became more organized compared to those with fixed SST distribution.

Waliser et al. (1999), using a GCM coupled to a slab ocean mixed layer, showed that air-sea coupling improves the simulation of the MJO. The improvement includes increased MJO variability, a closer match of the timescale of oscillation with observations, reduced eastward phase speed in the eastern hemisphere, and an increased seasonal signature in the MJO with more events occurring in the December-May period. The subsequent numerical model studies have generally confirmed the positive contributions of the air-sea interaction in enhancing eastward propagating MJOs and the northward propagation of boreal summer ISOs. For instance, analysis of the European Centre for Medium-range Weather Forecast-Hamburg atmospheric model (ECHAM4) and its coupled version with the University of Hawaii 2.5-layer tropical ocean model, Kemball-Cook et al. (2002) found that upon coupling, pronounced northward propagation of convection and circulation anomalies appear in the May-June Asian monsoon season.

However, it has also been recognized that in order for air-sea interaction to enhance ISO, realistic simulation of the mean state in the coupled model appears to be necessary (Gualdi et al., 1999; Hendon, 2000). Kemball-Cook et al. (2002) found that their coupled model failed simulating the August-October ISO in the western North Pacific, because the mean SST in the coupled model was too cold and the monsoon vertical easterly shears were absent. Inness and Slingo (2003) also found the air-sea coupling improves eastward propagation of the convection across the Indian Ocean; but there was no eastward propagation in the western Pacific, because the errors in the mean low-level zonal wind component in the west Pacific prevented the MJO from propagating into this region (Inness et al., 2003).

In the coupled model of Waliser et al. (1999), the SST variation is primarily due to changes in latent heat flux and to a lesser degree, changes in the surface shortwave flux. However, the slab ocean model might be too simple to address the question of what causes intraseasonal SST variations. The cause of the SST variability has been further studied using a coupled atmosphere-ocean general circulation model by Cubucku and Krishnamurti (2002), who found that the intraseasonal SST oscillations in the warm pool are primarily caused by the tendency of solar radiation, while evaporative cooling was of secondary importance.

How the SST anomalies (SSTAs) feedback to the MJO is a key, and a more complex, issue. Waliser et al. (1999) showed that, in their coupled model, the enhanced SST to the east of the convection reinforces meridional convergence associated with the frictional moisture convergence. The resulting increase in moist static energy helped destabilize the disturbance and maintain it against dissipation more effectively relative to the case without coupling. Kemball-Cook et al. (2002) concurred with this idea and they attributed the improved northward propagation of monsoon ISO to the increased low-level convergence into the regions where a positive SST anomaly exists (i.e., ahead of the convective anomaly). Fu and Wang (2004) pointed out that the positive SSTAs ahead (north) of the convection organize convection through destabilization of the moist Rossby waves and local adjustment of the atmospheric convection to the SSTAs, thus enhancing the northward propagating ISO.

The observed positive SSTAs tend to lead convective anomalies by about one-quarter of a wavelength. What creates this phase lag is controversial. Woolnough et al. (2001) investigated the response of convection to an idealized imposed mobile intraseasonal SSTA in an atmospheric GCM on an aqua-planet. The convection was found to organize on the spatial and temporal scales of the imposed SSTAs and the location of the maximum in precipitation relative to the SSTAs is in good agreement with observations. They suggested that the free-tropospheric humidity plays a critical role for determining the location and magnitude of the precipitation response. On the other hand, Wu et al. (2002) compared an observed strong case of the MJO and its counterparts, simulated by 10 different atmospheric GCMs forced with the same observed weekly SST. In the observations, the positive SSTAs develop upstream of the main convection center while in the simulations, the forced component is in phase with the SST.

The coupled modeling study by Fu and Wang (2004) demonstrates that the air-sea interaction significantly enhances the northward propagation of ISO compared to the forced run in which the same atmospheric model is forced by the daily SST that is produced by the coupled model. They pointed out that the coupled and forced solutions are fundamentally different. Without coupling the SST and convection anomalies are nearly in phase, but in the coupled run the SST-convection has a structure similar to the observed. Neglect of atmospheric feedback makes the forced solution depart from the coupled solution in the presence of initial noises or tiny errors in the lower boundary.

10.3 A GENERAL THEORETICAL FRAMEWORK 10.3.1 Fundamental physical processes

What is the energy source for the MJO disturbances? Given the fact that the vertical structure of the MJO is dominated by the gravest baroclinic mode, it is rational to assume that the MJO is stimulated and sustained by the diabatic heating in the middle troposphere. Wang (1988a) presented a detailed scale analysis for the

Background Circulation (Monsoon, Hadley, Walker)

Background Circulation (Monsoon, Hadley, Walker)

f Moisture V Distribution

Land-Sea Distribution

Climatological S ST

Surface Heat Fluxes

Lower Boundary Forcing

Figure 10.1. Essential physical processes involved in the theoretical modeling of ISO. These processes are numbered in this figure in an order consistent with the reviews in Section 10.2.

Figure 10.1. Essential physical processes involved in the theoretical modeling of ISO. These processes are numbered in this figure in an order consistent with the reviews in Section 10.2.

MJO. He demonstrated that if the observed baroclinic structure of the MJO is maintained, the required divergence is on an order of (5 x 10~7s_1), which has to be sustained by a diabatic heating rate on an order of 2-3mmday_1 because of the relatively small magnitude of the other energy sources in the tropics. This scaling argument is consistent with observations (Krishnamurti et al., 1985) and model results such as the Geophysical Fluid Dynamics Laboratory (GFDL) model (Lau et al., 1988). In this chapter it is assumed that the diabatic heating, especially the latent heat released in precipitation, drives the ISO, albeit some studies have emphasized the roles of extra-tropical forcing (e.g., Hsu et al., 1990; Blade and Hartmann 1993; Lin et al., 2000) in exciting and maintenance of the MJO.

Figure 10.1 presents a schematic summary of the fundamental processes relevant to ISO. Of central importance is the convective interaction with dynamics (CID), following Neelin and Yu (1994). The CID is marked by the ovals with marble shading in Figure 10.1. The CID comprises feedbacks among collective effects of convective heating, low-frequency equatorial Kelvin and Rossby waves, and planetary boundary layer processes including surface heat and momentum exchanges with the lower boundary and moisture feedback with convection and dynamics. These feedback processes integrate the following mechanisms reviewed in Section 10.2: convection-wave convergence feedback, frictional moisture convergence feedback, evaporation-wind feedback, and moisture feedback. The CID is considered essential for understanding the basic internal dynamics of the MJO. The planetary boundary layer processes are of central importance in CID. Dynamically, it provides a large-scale control of the moisture that fuels convection; it contributes to accumulation of moist static energy, and to triggering shallow and deep convections. Thermodynamically, the wind-induced heat flux exchange is key in changing the moist static energy in the boundary layer, thus providing an energy source for the MJO (Emanuel, 1993). Besides the potential role in determining the location and strength of the precipitation heating, the boundary layer friction is also an efficient energy sink for the low-frequency motion.

Other processes include the cloud-radiation feedback, impacts of seasonal mean flows, and air-sea interaction (Figure 10.1). The tropospheric radiative heating associated with the MJO is dominated by long-wave radiative cooling (Lee et al., 2001), which in turn is determined by the properties of clouds. In the following formulation, the simplistic Newtonian cooling is adopted to represent the net radiative heating effect. Keep in mind that, as suggested by Lin and Mapes (2004), in the cloud region, the net radiative heating may be slightly positive, which could enhance latent heating to the first order. In addition, the radiative heating/cooling by short-wave and long-wave radiation at upper-level clouds associated with the MJO convection might be important in destabilizing the atmosphere and leading to deep convection, thus the Newtonian cooling may be an oversimplified parameterization of radiation processes.

The seasonality of the ISO suggests a possible regulation of the moist wave dynamics by seasonal varying background flows. The amplitude of the MJO circulation variations is typically small compared to that of the seasonal variations of the tropical circulations, such as the monsoon circulation. Also, the transient momentum and heat fluxes tend to play a negligible role in determining the tropical mean circulation (Ting, 1994; Hoskins and Rodwell, 1995). Thus, to the first approximation the ISO is treated as a perturbation motion in this chapter. The ISO disturbances are influenced by the seasonal mean circulation and climato-logical distribution of moisture through SST (Figure 10.1). The impact of the background circulation is essential in explaining the seasonal behavior of the MJO (Section 10.5).

While the ISO is, to a large extent, determined by the atmospheric internal dynamics, its coupling to the oceanic mixed layer may have a considerable impact on its behavior. The coupled formulation will be presented in Section 10.6.

Was this article helpful?

0 0
Renewable Energy Eco Friendly

Renewable Energy Eco Friendly

Renewable energy is energy that is generated from sunlight, rain, tides, geothermal heat and wind. These sources are naturally and constantly replenished, which is why they are deemed as renewable.

Get My Free Ebook


Post a comment