1. Introduction
Since it was suggested that the rise of warm events in the Pacific Ocean could be triggered by westerly wind anomalies over the western side of the basin (e.g., Barnett 1977; Gill and Rasmussen 1983), increasing efforts have been made to understand the mechanisms that rule such anomalies. The occurrence of the strong 1982–83 El Niño event, following the development of westerly wind anomalies to the west, as described by Philander (1990), contributed to turn the attention of the scientific community to the western Pacific. It is now recognized that the western Pacific warm pool (WPWP) is indeed a crucial center-of-action for the El Niño–Southern Oscillation phenomena.
The understanding of the processes over the WPWP was extremely poor just less than a couple of decades ago. Although the existence of a major tropical variability on the timescale of 30–60 days was already recognized [Madden–Jullian oscillation (MJO), first described by Madden and Jullian (1971)], the presence of important air–sea interactions over that region was ignored. In fact, even the knowledge of the hydrographic characteristics of the WPWP was limited at that time. For instance, it was believed that the mixed layer of the western Pacific was very deep; a view surpassed by the analysis of data from the Western Equatorial Pacific Ocean Circulation Study (WEPOCS; e.g., Lukas and Lindstrom 1991). Only the completion of the Tropical Ocean Global Atmosphere Coupled Ocean–Atmosphere Research Experiment (TOGA COARE) provided the opportunity to investigate the warm pool processes in depth.
Indeed, as pointed out by Godfrey et al. (1998), TOGA COARE answered many questions regarding the dynamics of the coupled system over the WPWP. COARE encompassed a rich panorama, comprising:
The evolving structure of the atmosphere (e.g., McBride et al. 1995; Ding and Suni 1995; Lin and Johnson 1996a,b; Frank et al. 1996; Johnson and Lin 1997; Brown and Zhang 1997) and the ocean (e.g., Bradley et al. 1993; Cechet 1993; Eldin et al. 1994; You 1995; Anderson et al. 1996; Cronin and McPhaden 1997; Soloviev and Lukas 1997a,b; Ohlmann et al. 1998; Inall et al. 1998);
The behavior of the surface fluxes, and their relationships with atmospheric and ocean parameters (e.g., Bradley 1994; Fairall et al. 1996a,b; Jabouille et al. 1996; Sun et al. 1996; Song et al. 1996);
The organization of convection and precipitation (e.g., Mapes and Houze Jr. 1995; Mori 1995; Chen et al. 1996; Chen and Houze Jr. 1997; Short et al. 1997; Rickenbach and Rutledge 1998; Hildebrand 1998; Roux 1998; DeMott and Rutledge 1998a,b; LeMone et al. 1998; Lewis et al. 1998).
Also, several aspects of short-term air–sea interactions were explored thanks to the COARE dataset (Wang et al. 1996; Lau and Sui 1997; Feng et al. 1998; Li et al. 1998).
However, despite COARE's multiple achievements, there are still many open questions; among them several are related to the coupling of the atmosphere and the ocean on small timescales. This includes the WPWP oceanic variability on spatial scales of hundreds of kilometers and less, the control of the hydrographic fields by local and remote forcing associated with cloud systems, the possible air–sea feedbacks associated with this variability, etc. Coupling a cloud-resolving model (CRM) to a model of the upper ocean is certainly useful for such investigations. In previous work, using CRM simulations with prescribed sea surface temperatures (SST), Costa et al. (2001) showed that SST changes can influence several aspects of the organization of convection.
This paper aims to investigate atmosphere–ocean interaction over the WPWP using a coupled CRM–UOM (cloud-resolving model–upper-ocean model) for different situations. In section 2, distinct regimes under which convection developed over the WPWP during COARE are discussed. Section 3 describes the modeling tool used in this study. In section 4, a comparison is performed between some atmospheric and oceanic observations and the corresponding simulated fields. Sections 5–7 focus on the description of the properties of the simulated fresh anomalies (freshwater lenses). In section 8, the dataset produced by the coupled CRM–UOM is used to estimate the importance of the contribution from several factors (gustiness, ocean currents, wind speed–air temperature correlation, wind speed–water vapor mixing ratio correlation, and wind speed–SST correlation) to the total magnitude of the surface heat fluxes. A summary is presented in section 9.
2. Different WPWP convective regimes
The most significant large-scale intraseasonal oscillation over the WPWP is the one that corresponds to a 30–60-day cycle (e.g., Madden and Julian 1971; Hendon and Salby 1994; Madden and Julian 1994; Kayano and Kousky 1999). The so-called MJO acts as an important modulator for deep convection (e.g., Hendon and Liebmann 1994; Zhang and Hendon 1997).
Often, deep convection occur in association with westerly wind bursts (WWBs), with a phase lag of the order of a couple weeks between the peak convective activity and the peak near-surface winds (Lin and Johnson 1996a). However, as can be shown from the TOGA COARE observations, deep convection over the WPWP can also develop under conditions other than moderate westerlies associated with WWBs.
Figure 1 is a scatterplot showing the occurrence of values of the zonal wind, for events in which the minimum omega (vertical velocity in pressure coordinates) was less than −10 mb day−1. Each diamond corresponds to one observation (observations were taken each 6 h). Although three of the five events occurred under moderate westerlies (11 November and 22 and 24 December 1992) related to developing WWBs, deep convection was also observed under moderate easterlies (19 January 1993) and under weak easterlies in transition to westerlies, about three weeks before the peak WWB (11 December 1992). In this paper, simulations of convection under both moderate and weak winds are performed.
Figure 2 depicts the time evolution of the COARE Intensive Flux Array (IFA) average zonal wind (2a), near-surface wind speed (2b), vertical velocity in pressure coordinates (2c) and the advective tendencies of potential temperature (2d) and moisture (multiplied by L/cp, 2e) for the period between 0000 UTC 1 December 1992 and 0000 UTC 1 January 1993. The two periods selected for our modeling study are also indicated in Fig. 2.
Case 1 corresponds to the occurrence of deep convection under weak large-scale winds. Most of the 10-day period between 7 and 17 December was characterized by weak westerlies, with an increasing eastward trend starting on 12 December (Fig. 2a). The large-scale surface winds are below 2.5 m s−1 until that date (Fig. 2b). The entire 10-day period is characterized by ascending motion, as indicated in Fig. 2c. The large-scale ascent is strong on 11 December, as also shown in Fig. 1. Accompanying the rising motion, significant large-scale cooling (Fig. 2d) and moistening (Fig. 2e) are present, particularly on 11 December.
Case 2 corresponds, in essence, to one of the cases studied by the Working Group 4 of the Global Energy and Water Cycle Experiment (GEWEX) Cloud System Study (GCSS) and has been described by several authors, including Moncrieff et al. (1997), Su et al. (1999), and Costa et al. (2001). Figure 2a shows that westerly winds increased rapidly during the first 3 days of that period and that a strong shear developed from 22 to 29 December. The correspondent increase in the wind speed close to the surface is shown in Fig. 2b. Figure 2c indicates that successive events of large-scale ascent occurred between 19 and 29 December, particularly on 20, 22, and 24 December. Those events and the correspondent large-scale cooling and moistening (Figs. 2d and 2e) were associated with multiple episodes of deep convection.
3. Coupled model
Several researchers have reported the occurrence of very pronounced small-scale variability in the thermal and haline structure of the upper layers of the western Pacific. Aircraft measurements described by Hagan et al. (1997) have shown significant horizontal SST gradients, in scales similar to that of cloud systems. From ship data, Soloviev and Lukas (1997a) have indicated the existence of sharp discontinuities in the oceanic fields in association with precipitation-produced freshwater lenses.
However, observations of the small-scale oceanic structure are very limited. Satellite and aircraft measurements can provide a reasonable view of the skin–SST variability, but it does not provide information on the haline and below-surface thermal structures. On the other hand, ship and buoy observations can furnish detailed information on the dynamical and thermohaline vertical profiles. Nonetheless, scarce point measurements provide little insight on the horizontal variability of the SST. Hence, the use of multidimensional ocean models provides a means to investigate the properties of inhomogeneous features in the upper ocean.
From the point of view of the atmosphere, the same kind of observational restriction is found. With the exception of measurements via radar (and, under certain circumstances, aircraft), the observed data are, in general, not suitable to investigate detailed horizontal and vertical structures down to the meso- and cloud scales. Consequently, several characteristics of clouds and cloud systems can only be investigated using cloud-resolving models. According to Moncrieff et al. (1997), since CRMs resolve individual convective cells, one can understand the collective effects of clouds directly through cloud-resolving simulations.
Moreover, using a CRM, one can properly force an ocean model with heterogeneous fluxes of heat, freshwater, and radiation at the fluid interface. In turn, because the SST is important to the organization of convection (see, e.g., Costa et al. 2001), the use of varying SSTs, as calculated by a UOM, to force a CRM, is a realistic approach. These were the motivations to build a modeling system comprising a CRM and a two-dimensional UOM with two-way coupling.
The atmospheric model used in this study is a two-dimensional, cloud-resolving version of the Regional Atmospheric Modeling System (RAMS; e.g., Pielke et al. 1992). RAMS is a nonhydrostatic model with several different options of numerical schemes and physical parameterizations. Microphysical processes are represented by a one-moment bulk parameterization (Walko et al. 1995), in which the water substance is divided into up to eight categories: vapor, cloud water, rainwater, pristine ice, snow, aggregates, graupel, and hail. Hydrometeor diameters are assumed to follow a gamma distribution. The two-stream radiation scheme is coupled with the cloud microphysics parameterization, as described by Harrington (1997) and Olsson et al. (1998).
The choice of a 2D instead of a 3D CRM was obviously based on the computational issue, although it is not completely clear how the additional degree of freedom impacts the atmospheric fields in cloud-resolving simulations of tropical convection. Although Donner et al. (1999) showed that surface heat and moisture fluxes may differ in 2D versus 3D simulations, Grabowski et al.'s (1998) results suggest that two- and three-dimensional simulations exhibit similar statistics. In addition, Costa et al. (2001) obtained realistic cloud mass fluxes, precipitation, and surface heat fluxes using a 2D CRM, while 3D simulations by Su et al. (1999) overestimated both sensible and latent heat transport, which suggests that a better surface parameterization is actually more important than the use of a 3D model, at least to attain realistic surface fluxes.
Forcing terms were added to certain prognostic equations to represent the large-scale tendencies of momentum, temperature, and moisture. The large-scale forcing is applied in a form similar to that proposed by Grabowski et al. (1996). The contribution of the large-scale to the horizontal velocity is calculated using a relaxation technique, in which the nudging terms are calculated from the averaged velocity components and are uniform through the model domain. For the temperature and moisture fields, forcing terms involving the contribution from both large-scale horizontal and vertical advection are added to the tendencies in the ice–liquid potential temperature and water vapor mixing ratio. The data depicted in Figs. 2a, 2d, and 2e, and the observed meridional wind (not shown) were used in this procedure.
The ocean model is a 2D version of the Princeton Ocean Model, described by Blumberg and Mellor (1987). The model contains prognostic equations for the zonal and meridional currents, potential temperature, salinity, turbulent kinetic energy, and turbulent length scale. A splitting procedure is adopted to resolve the fast-varying external mode and the slow-varying internal mode. The turbulence closure submodel is the level-2.5 scheme in the hierarchy described by Mellor and Yamada (1974). Longwave radiative fluxes are evaluated at the boundary. Shortwave radiation is allowed to penetrate the ocean and is treated as a source term of heat in each level. The present radiation scheme is quite simple: no spectral dependence is considered and a single exponential attenuation is calculated for the whole radiative flux. No large-scale forcing was imposed on the ocean model. Other model limitation is the lack of a proper representation of wave effects.
The coupling formulation is similar to the one described by Hodur (1997). The models interchange momentum, heat, and water substance. Information regarding the downward short- and longwave radiation at the ocean surface and the SSTs allows the calculation of the radiative fluxes. In its present version, the ocean currents were not considered in the calculation of the surface heat and momentum fluxes.
In all simulations presented in this paper, 512 horizontal grid points and a horizontal grid spacing of 1 km were used in both the atmospheric and ocean models. The atmospheric vertical grid comprised 50 levels with a variable grid spacing of 100 m close to the surface to 500 m by the model top (21 km). The ocean vertical grid contained 58 vertical levels with the grid spacing varying from 2 cm at the surface to 5 m at the model bottom (240 m). The coupled model domain is located at the equator. The atmospheric model was initialized with the average soundings over the COARE IFA on 0000 UTC 7 December 1992 (case 1) and 0000 UTC 19 December 1992 (case 2). To simulate case 1, the ocean model was initialized using data collected aboard R/V Kexue No. 1, which was located at 3.963°S, 156.053°E at 0000 UTC 7 December 1992. For the simulation of case 2, the ocean model was initialized using current, temperature, and salinity data aboard the R/V Moana Wave, which remained approximately stationary at 1.75°S, 156°E from 20 December 1992 to 11 January 1993.
4. Comparison with the observations
As many studies have already shown, cloud-resolving models constrained by a imposed forcing often produce results in good agreement with the observed fields (e.g., Grabowski et al. 1996; Su et al. 1999; Costa et al. 2001). As in Costa et al.'s (2001) uncoupled simulations, the atmospheric component of the coupled model described in the previous section produced results in very good agreement with several observed fields, such as the domain-averaged winds, the surface heat fluxes, and the total precipitation. In contrast, because the ocean model was much less constrained than the atmospheric model, the agreement between observed and modeled oceanic fields was less than its counterpart for the atmosphere.
a. Case 1
Most of the results from the atmospheric model are in good agreement with the observations. Figure 3 depicts the time evolution of the surface precipitation (3a) and surface sensible and latent heat flux (3b) as calculated from the observations (solid lines) and simulated by the coupled CRM (dashed lines). Qualitatively, the model represented well the evolution of all three fields. The agreement between the 9-day average modeled and observed fields for the period between 8 December and 16 December is also good, particularly for the sensible heat flux, as shown in Table 1. The moist bias probably contributed to the underestimation of the latent heat flux.
Unfortunately, there were no measurements of ocean fields from the R/V Kexue after 11 December, which prevents a proper validation of the ocean model data for most of the simulated period. Between 8 and 11 December, there is a reasonable agreement between the observed and modeled temperature at a depth of 2 m, including an important diurnal variation and a general cooling trend (Fig. 4a). Nonetheless, the agreement between the model domain-averaged 2-m salinity and the observations from the research vessel is not good (Fig. 4b). Although the model was able to represent some aspects of the evolution of the salinity field (e.g., the significant freshening of the upper ocean from 8 to 9 December), the differences between the modeled and observed fields are large at certain times. In part, this is due to the crucial difference between an area-averaged model variable and a point observation. The sudden and large variations in the salinity field suggest that occasionally the vessel was sitting over fresh anomalies. Conversely, it might suggest that the surface forcing was not the only mechanism controlling the upper-ocean budget, particularly concerning the salinity field.
b. Case 2
As in the previous case, the results from the atmospheric model were in good agreement with the observations, particularly the surface heat fluxes and the precipitation, which are important to force the ocean model. The comparison between those fields and their observed counterparts is depicted in Fig. 5 and Table 2. The agreement between the model precipitation and surface latent heat flux (evaporation) and the observations was better in case 2 than in case 1. The model did, however, overestimate the sensible heat flux. Concerning other atmospheric model fields, the results were similar to the ones obtained by Costa et al. (2001) using an uncoupled CRM, including the development of a cold, dry bias (not shown).
Figure 6 depicts the time evolution of the observed 2-m temperature and salinity (measured by the R/V Moana Wave) and the horizontally averaged 2-m temperature and salinity predicted by the ocean model for case 2.
Despite its limitations, the ocean model was able to reproduce important observed features of the 2-m temperature (Fig. 6a). With the exception of the exaggerated diurnal warming on 20 December due to the overestimated downward solar radiation associated with the atmospheric model spinup, the departure from the observed field is always smaller than 0.3°C (the difference at 0000 UTC 29 December is 0.15°C). The observed cooling trend between 20 and 28 December was well represented. The temperature drop observed by the vessel was 0.69°C, compared with 0.59°C predicted by the model. Although a small difference, the warm ocean bias, must have contributed to the overestimation of the surface sensible heat flux along with the cold bias in the air temperature.
The agreement in the salinity field was not as good, and the model showed a fresh bias of 0.17 psu at the end of the simulated period. As seen in Fig. 6b, despite the large flux of freshwater associated with the heavy precipitation over the COARE IFA, the salinity field showed an increasing trend, which can be associated with other factors, such as horizontal advection of more saline water to the location of the vessel. Nevertheless, despite the amplitude difference (that can be attributed to the contrast between a point measurement and the model domain average), the major freshening events were represented by the model. Overall, the agreement between the modeled oceanic fields and the observations is better in case 2 than in case 1. The surface forcing was stronger in case 2, therefore, it can dominate the heat and freshwater budget of the upper ocean in this case. In contrast, in case 1, for which the surface forcing was weaker, larger-scale transport of heat and salt could play a more significant role.
5. Horizontal dimension and duration of simulated freshwater anomalies
Tropical convection is often organized in mesoscale systems comprising multiple convective cells that produce heavy precipitation and a stratiform region that produces lighter precipitation. The complex structure of convection accounts for a large spatial variability in the surface precipitation fields. Freshwater anomalies are generated at the tropical ocean surface by rainfall (Soloviev and Lukas 1997b; Wijekera et al. 1999), producing a correspondent variability in the sea surface salinity (SSS) field. The amplitude of the SSS perturbations is particularly large at newly formed freshwater anomalies, especially the ones generated behind cores of intense convective precipitation. Since they are often formed from cooler rainwater, freshwater anomalies also correspond to cold anomalies in the SST field, at least in the early stages of their life cycles. Those anomalies are haline-stratified, and, since the salinity effect dominates the temperature effect, exhibit strong vertical stability and a significant inhibition to vertical mixing, which therefore modifies the transport processes near the ocean surface. As a consequence, the further evolution of the SST field is also influenced by the presence of freshwater anomalies.
To illustrate the formation of freshwater anomalies from precipitation and their influences on the SST field, Figs. 7 and 8 show Hovmöller diagrams of the surface precipitation, the SSS, and the SST for cases 1 and 2, respectively. In both cases, the complex structure of the precipitation fields produced a spectrum of freshwater anomalies with a variety of horizontal dimensions and lifetimes. It is apparent that nearly all freshwater anomalies correspond to cold SST anomalies during most of their life cycles.
The initial characteristics of a freshwater anomaly are strongly dictated by the behavior of its parent storm. The horizontal dimension of a newly formed freshwater anomaly depends on the propagation speed of the precipitating system. Fast-propagating systems generate widespread anomalies, while near-stationary storms tend to produce localized salinity disturbances. The lifetime of the freshwater anomalies depends strongly on the near-surface winds. Because strong winds induce vertical mixing in the ocean mixed layer, the duration of freshwater anomalies becomes shorter. On the other hand, calm winds favor long-lived upper-ocean salinity perturbations. Figures 7 and 8 illustrate how freshwater anomalies behave under the influence of different atmospheric regimes. Case 1 was generally characterized by slowly propagating precipitating systems and weak large-scale winds that generated relatively narrow, long-lived freshwater anomalies. Simulated perturbations as large as 0.2 psu persisted for more than 48 h. In contrast, case 2 exhibited fast-moving precipitating systems and moderate large-scale winds, which produced wider salinity disturbances with shorter lifetimes.
6. Statistical thermohaline behavior of simulated freshwater anomalies
As discussed by Webster (1994), who analyzed data from WEPOCS experiment (Lukas and Lindstrom 1991), three major physical processes account for the variability of the SST and the SSS over the WPWP: diurnal heating, precipitation, and entrainment/evaporation (one could add the transport of heat and salt by the currents). The diurnal warming and cooling of the ocean surface layer is often more pronounced when convection is suppressed and the winds are light. When such a process dominates, the SST experiences a diurnal variation often greater than 1°C while the SSS remains almost constant, due to little evaporation (precipitation is usually near zero in those situations). During periods of active convection, the diurnal cycle is usually very small due to the blocking of solar radiation. However, the variability of both SSS and SST is often large, due to the presence of precipitation-formed disturbances, in which both the salinity and the temperature are reduced. Finally, strong westerly winds (often associated with suppressed convection) induce simultaneously surface evaporation and deep mixing. The latter allows cooler, saltier water to penetrate the ocean mixed layer from below. Both processes contribute to lower the SST and to an increase in the SSS. The two first processes differ from the third in the timescales in which they operate. Both the diurnal warming and the precipitation generate shallow layers, in which the stability is governed by thermal or haline stratification, respectively. In those two cases, the lifetime of the SST–SSS disturbances is of the order of one day or less. By contrast, the cooling and salting by evaporation and entrainment operate on a deeper vertical scale and a longer timescale.
The crucial distinction between those two groups of processes was shown in our modeling study and illustrated in Fig. 9, which represents scatterplots of SST and SSS for cases 1 (Fig. 9a) and 2 (Fig. 9b). As in Webster's (1994) Fig. 14, theoretical axes are superimposed on the scatterplots in Fig. 9, representing particular physical processes. The diurnal warming and cooling of the ocean surface is represented by the points lying on an almost horizontal line (axis A) at the top of both panels of Fig. 9. The points representing reduced SST and SSS in both panels are generally associated with precipitation (axis B). Webster's axis C is absent for two reasons: first, the regime in which evaporation and entrainment is the dominant process (trade wind regime accompanying strong near-surface westerlies) was not simulated; second, it requires a longer timescale to operate than the one associated with the other two processes.
It is clear from Fig. 9 that precipitation-produced freshwater anomalies are often cooler than its surroundings. In fact, as shown in Fig. 10, the spatial correlation between SST and SSS tends to be close to one just after events of strong rainfall, as occurred between 11 and 14 December (Fig. 10a) and between 20 and 22 December and at 24 December (Fig. 10b). Especially under stronger near-surface winds, if precipitation is not present or is weak, the correlation between the SST and the SSS decreases rapidly with time. That occurred, for instance, late on 23 December, when convection was suppressed and the SST–SSS correlation decreased from 0.90 to 0.21 (minimum value, early on 24 December). A more dramatic example occurred early on 28 December, when the SST–SSS correlation changed sign, falling from 0.75 to −0.27 in 9 h. During that period, the modeled precipitation was zero or almost zero (Fig. 5a), freshwater anomalies were mostly absent (Fig. 8), and the evaporation was very significant, with a latent heat flux of more than 100 W m−2 (Fig. 5b).
7. Thermal, haline, and dynamic structure of simulated freshwater anomalies
Observations from TOGA COARE, as described by Wijesekera et al. (1999), indicate that fresh anomalies have a complex structure of both their thermohaline and dynamic characteristics. The following are some of the features of a precipitation-produced freshwater disturbance described by Wijesekera et al. (1999):
The fresh anomaly, or freshwater lens, deepens with time, as a result of vertical transport;
Temperature and salinity are highly correlated within the lens, consistent with its formation by cool precipitation;
A jet (horizontal velocity anomaly) develops at the lens, in the direction of the wind, associated with the trapping of the momentum transferred from the atmosphere due to the haline stratification;
In relation to the jet, downwelling (upwelling) occurs at the downwind (upwind) edge of the lens.
Despite the limitations associated with the lack of the third dimension, the coupled model was able to simulate those processes. Figure 11 depicts a newly formed fresh anomaly and its parent storm at 1900 UTC 22 December, corresponding to 2 h after the beginning of the precipitation event. The rainfall-generated freshwater lens is initially shallow. In Fig. 11, salinity values of 34.0 psu, corresponding to a fresh anomaly of about 0.1 psu are found to less than 6 m of depth. The initial geometry of the lens is strongly dictated by the precipitation field.
Some aspects of the time evolution of the same freshwater lens are shown in Fig. 12. The two panels in Fig. 12 are Hovmöller diagrams depicting the SSS and the vertical velocity at a depth of 22.5 m (Fig. 12a), and the SST and the surface zonal current (Fig. 12b).
As expected, after the end of the precipitation event, the amplitude of the fresh anomaly at the surface decreased with time, mostly due to subgrid vertical transport (Fig. 12a). Concurrently, the lens contracted in the horizontal. During its whole life cycle, the freshwater lens was cooler than the environment (Fig. 12b). The location and time of the SST and SSS minima coincided. A significant westerly anomaly developed in the surface zonal current, with a peak eastward current of more than 0.16 m s−1 (Fig. 12b). Such an anomaly was caused by the trapping of the westerly momentum transferred from the atmosphere in the shallow, salinity-stratified layer. In association with the anomaly in the horizontal current, downwelling developed at the downwind edge of the lens, while upwelling developed at the upwind edge. At a depth of 22.5 m the downwelling was as strong as 4 × 10−4 m s−1 (more than 30 m day−1), as shown in Fig. 12a.
Some characteristics of the same simulated fresh anomaly in its mature stage are shown in Fig. 13.This figure depicts the salinity and vertical velocity (Fig. 13a) and the ocean temperature and zonal current (Fig. 13b) at 0400 UTC 23 December (9 h after its formation). The comparison between Figs. 13a and 11 indicates the deepening of the fresh lens. In Fig. 13a, the 34.0 psu contour has deepened to a depth of about 16 m (compared to less than 6 m earlier). The base of the fresh anomaly is tilted, with a deeper fresh layer on its downwind (right) edge and a shallower layer on its upwind (left) edge. Due to the significant downwelling, freshwater is allowed to penetrate deep into the mixed layer on the downwind side of the lens. As shown in Fig. 13a, water with a salinity of less than 34.1 psu was found down to a depth of about 40 m. To the left of the lens, the increased SSS indicates the upwelling of deeper, saltier water. Figure 13b shows that the freshwater lens was significantly cooler than the environment, even several hours after its formation. Cool water is transported downward on the downwind side of the lens and warmer water is carried upward on the upwind side. As a consequence, the isotherms and the isohalines are tilted downward in the direction of the surface wind. The existence of a jet coinciding with the fresh, cool anomaly is evident in Fig. 13b, with a positive (eastward) perturbation zonal current of about 7 cm s−1. Below the lens, the perturbation current changes sign, becoming negative (westward), the reversal caused by a zonal pressure gradient associated with the vertical motions in the opposite sides of the lens.
8. Concerning the parameterization of surface fluxes in large-scale models
It is well known that the passage of a convective system brings important changes to the atmospheric boundary layer and the surface fluxes. According to Young et al. (1995), over the TOGA COARE region, a significant decrease in air temperature, significant increases in the wind speed, sensible and latent heat fluxes, and decreases in the SST, water vapor mixing ratio in the atmospheric surface layer, and sea surface saturation mixing ratio accompany the convective wakes. Saxen and Rutledge (1998) showed that different types of organization of convection are able to produce greatly enhanced surface fluxes during the convectively active phase.
The present coupled model is a useful tool to evaluate the importance of each of these subgrid-scale processes. The entire CRM domain can be conceived as a proxy for a grid column of a GCM, with the mesoscale and convective scales resolved by the CRM representing subgrid-scale processes in the GCM. Therefore, the nonbar variables in Eqs. (1)–(5) are going to be interpreted as CRM gridpoint values, and the bar variables as CRM domain averages.
Different values of the heat fluxes were calculated using formulas (1)–(5) substituting the fields predicted by the CRM. The wind speed, air temperature, and water vapor mixing ratio were taken at the first atmospheric level (about 50 m). Both cT and cq were substituted by 1.4 × 10−3. The use of this value of the transport coefficients in the bulk equation (3) was shown to produce values of both sensible and latent heat flux very close to the ones calculated by the surface parameterization of the CRM.
Table 3 shows the 9-day average value of the heat fluxes for cases 1 (between 8 and 16 December) and 2 (between 20 and 28 December), according to formulas (1)–(5), and the percent difference from Ql and Qs [calculated using Eq. (3)]. As expected, the effects of the surface currents and small-scale processes are more noticeable in the weak wind case. Furthermore, the results indicate that in a relatively long-term average (several days), the effects of the wind speed–water vapor mixing ratio and wind speed–SST correlations are negligible. The positive correlation between the air temperature and the wind speed accounted for an average enhancement of 3.0% in the sensible heat flux in case 1. However, because the Bowen ratio is small, its average effect on the total surface heat flux can be neglected too. The influence of the surface currents is typically small, accounting for an average reduction of less than 3.0 W m−2 in the total surface heat flux, again in case 1. In contrast, the effects of gustiness are significant in the multiday average, particularly if the large-scale winds are weak, producing an increase of 44.0 and 12.5 W m−2 in the total surface heat flux in cases 1 and 2, respectively.
The time evolution of the difference between Qs,c, Qs,nt, Qs,nsst, and Qs,ave and Qs for cases 1 and 2 is shown in Fig. 14. Similarly, the difference between Ql,c, Ql,nq, Ql,nsst, and Ql,ave and Ql for cases 1 and 2 is shown in Fig. 15. The figures indicate that, despite their small average value, those differences are relatively large at specific times and, under certain circumstances, the ocean currents, and the wind speed–air temperature, wind speed–water vapor mixing ratio, and wind speed–SST correlations have to be considered in the calculation of the surface heat fluxes.
The ocean currents usually contribute to reduce the surface fluxes, since they tend to align with the surface winds. They accounted for a maximum reduction of 1.1 W m−2 in the sensible heat flux and 7.7 W m−2 in the latent heat flux. This effect is more pronounced after strong precipitation events and during daytime. In both cases, the momentum transport from the atmosphere is limited to a shallow layer that accelerates in the direction of the wind. In a few circumstances, particularly at nighttime, the effect of the currents is to enhance the fluxes, which suggests the existence of a current component opposing the wind. This occurs in association with the mixing of the upper-ocean layers with the mixed layer below, which eventually has a bulk current in a direction other than the atmospheric surface flow.
The effects of the wind speed–air temperature and wind speed–water vapor mixing ratio are usually small, related to changes in the heat fluxes of less than 5 W m−2. The most evident exception was the strong convective event that occurred under a weak wind regime (case 1, 11 December). During that episode, the correlation between the wind speed and the air temperature accounted for a maximum increase of 12.5 W m−2 in the sensible heat flux and the wind speed–mixing ratio correlation contributed to an increase of 14.8 W m−2 in the latent heat flux.
The variability of the SST and its subgrid-scale correlation with the surface wind speed can have important effects on the surface latent heat flux. In case 1, this correlation acted to reduce the latent heat flux up to 3.9 W m−2. In case 2, the coupled system showed a more complex behavior, with the wind speed–SST correlation changing sign several times. The maximum variation in the latent heat flux in case 2 was 9.3 W m−2. The effects of the SST variability on the sensible heat flux were significantly smaller (maximum reduction of 1.0 W m−2, case 1).
Among the different phenomena analyzed, gustiness is the one that has the greatest contribution to the surface heat fluxes, particularly when convection is active. In case 1, the small-scale circulations accounted for an enhancement of up to 47.2 W m−2 in the sensible heat flux and 261.4 W m−2 in the latent heat flux. In case 2, because the large-scale winds were larger, the contribution from gustiness to the surface heat fluxes was smaller, with a maximum enhancement of 22.7 W m−2 in the sensible heat flux and 121.7 W m−2 in the latent heat flux.
9. Summary
In a “standard” cycle of the MJO, deep convection usually develops under moderate low-level westerly winds, prior to the peak in a WWB. Nonetheless, during the TOGA COARE IOP, it was observed that deep convective systems could form under different low-level wind regimes. In this paper, two convectively active periods were studied, using a two-dimensional cloud-resolving model coupled to an upper-ocean model described in section 3. In one case (7–16 December 1992), the lower troposphere exhibited weak westerlies and the environmental shear was relatively small. The second case (19–28 December 1992) was characterized by the presence of moderate eastward winds in low levels and relatively strong westward winds in the upper levels, producing a significantly sheared environment.
In general, the coupled model produced results in good agreement with the observations. In both cases, the model atmospheric component produced temperatures cooler than the observations. In case 1, the model generated a moist bias, while conversely, a dry bias developed in case 2. The differences between the modeled precipitation and surface heat fluxes and their counterparts calculated from the COARE dataset in the two simulated periods are both small. As expected the modeled oceanic fields showed a greater departure from the observations, mostly due to the lack of the large-scale advective tendencies of temperature, salinity, and momentum.
The coupled model produced a large variability in the simulated upper-ocean fields (salinity, temperature, and currents), in association with the variable atmospheric forcing. Modeled freshwater anomalies exhibited a behavior similar to the ones observed over the WPWP, as the one described in detail by Wijesekera et al. (1999), including: 1—deepening with time; 2—high temperature–salinity correlation; 3—development of a jet in the direction of the wind; 4—downward motion at the downwind edge and upward motion at the upwind edge of the fresh anomaly.
Finally, simple calculations of the surface heat fluxes were made, using the data generated by the coupled model. The effects of the ocean currents, the wind speed–air temperature, wind speed–water vapor mixing ratio, and wind speed–SST correlations and gustiness were evaluated using a bulk formulation. The results showed that, among the different effects analyzed, gustiness produced the greatest changes in the surface heat fluxes. However, although exhibiting small average values in both multiday periods, the other effects can also be significant under certain conditions. Therefore, one has to consider the issue of the parameterization of such small-scale effects in large-scale models, such as GCMs. Recent studies (e.g., Qian et al. 1998) presented formulations of parameterizations of the effects of the cold, dry downdrafts associated with precipitation on the atmospheric near-surface layer and the surface fluxes. Those parameterizations account for the negative correlations between wind speed and air temperature and between wind speed and water vapor mixing ratio near the surface. In contrast, little attention has been given to the role of the ocean surface currents and SST inhomogeneities. In our simulations, those two effects together produced changes in the total surface heat fluxes of the same order of magnitude (or slightly less) than the changes associated with the cooling and drying of the boundary-layer by convective downdrafts. The present results suggest that the development of a parameterization of such ocean surface processes is needed to obtain more accurate calculations of surface fluxes in GCMs. Since the momentum transfer from the atmosphere is limited to a shallow layer if the upper ocean is highly stable, the influence of the ocean surface currents has a markedly strong diurnal variation, and is often amplified in the presence of precipitation. In addition, a significant negative (or even positive) correlation can be established between the SST and the near-surface winds, especially during strong precipitation events. This is the basis of a parameterization of ocean surface processes to be presented in future work.
Acknowledgments
This research was supported by the Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq, Brazil) and by the National Oceanic and Atmospheric Administration under Grant NA67RJ0152. Valuable contributions from Dr. George N. Kiladis and three anonymous reviewers helped to improve the manuscript.
REFERENCES
Anderson, S., R. Weller, and R. Lukas, 1996: Surface buoyancy forcing and mixed layer of the western Pacific warm pool: Observations and one-dimensional model results. J. Climate, 9 , 3056–3085.
Barnett, T. P., 1977: Attempt to verify some theories of El Niño. J. Phys. Oceanogr, 7 , 633–647.
Blumberg, A., and G. Mellor, 1987: A description of a three-dimensional coastal ocean circulation model. Vol. 4, Coastal and Estuarine Sciences, N. Heaps, Ed., Amer. Geophys. Union, 1–16.
Bradley, E. F., 1994: Variability of meteorological variables and air–sea heat, moisture and momentum fluxes during TOGA-COARE. Eos, 75 , 42.
Bradley, E. F., J. S. Godfrey, P. A. Coppin, and J. A. Butt, 1993: Observations of net heat flux into the surface mixed layer of the western equatorial Pacific Ocean. J. Geophys. Res, 98 , 22521–22532.
Brown, R. G., and C. Zhang, 1997: Variability of midtropospheric moisture and its effects on cloud-top height distribution during TOGA COARE. J. Atmos. Sci, 54 , 2760–2774.
Cechet, R. P., 1993: Diurnal SST cycle in the western equatorial Pacific: Measurements made aboard the R/V Franklin during TOGA-COARE and pre-COARE cruises. Bull. Aust. Meteor. Oceanogr. Soc, 6 , 26–32.
Chen, S. S., and R. A. Houze Jr., 1997: Diurnal variation of deep convective systems over the tropical Pacific warm pool. Quart. J. Roy. Meteor. Soc, 123 , 357–388.
Chen, S. S., R. A. Houze Jr., and B. E. Mapes, 1996: Multiscale variability of deep convection in relation to large-scale circulation in TOGA COARE. J. Atmos. Sci, 53 , 1380–1409.
Costa, A. A., W. R. Cotton, R. L. Walko, R. A. Pielke Sr., and H. Jiang, 2001: SST sensitivities in multi-day TOGA COARE cloud-resolving simulations. J. Atmos. Sci, 58 , 253–268.
Cronin, M., and M. J. McPhaden, 1997: The upper ocean heat balance in the western equatorial Pacific warm pool during September–December 1992. J. Geophys. Res, 102 , 8533–8553.
DeMott, C. A., and S. A. Rutledge, 1998a: The vertical structure of TOGA COARE convection. Part I: Radar echo distributions. J. Atmos. Sci, 55 , 2730–2747.
DeMott, C. A., and S. A. Rutledge, 1998b: The vertical structure of TOGA COARE convection. Part II: Modulating influences and implications for diabatic heating. J. Atmos. Sci, 55 , 2748–2762.
Ding, Y. H., and A. Sumi, 1995: Large-scale atmospheric circulation features during TOGA-COARE IOP. J. Meteor. Soc. Japan, 73 , 339–351.
Donner, L. J., C. J. Seman, and R. S. Hemler, 1999: Three-dimensional cloud-system modeling of GATE convection. J. Atmos. Sci, 56 , 1885–1912.
Eldin, G., T. Delcroix, C. Henin, K. Richards, Y. du Penhoat, J. Picault, and P. Rual, 1994: The large-scale structure of currents and hydrology along 156°E during the COARE Intensive Observation Period. Geophys. Res. Lett, 21 , 2681–2684.
Fairall, C. W., E. F. Bradley, J. S. Godfrey, G. A. Wick, J. B. Edson, and G. S. Young, 1996a: Cool-skin and warm-layer effects on sea surface temperature. J. Geophys. Res, 101 , 1295–1308.
Fairall, C. W., E. F. Bradley, D. P. Rogers, J. B. Edson, and G. S. Young, 1996b: Bulk parameterization of air–sea fluxes for Tropical Ocean Global Atmosphere Coupled Ocean–Atmosphere Response Experiment. J. Geophys. Res, 101 , 3746–3764.
Feng, M., P. Hacker, and R. Lukas, 1998: Upper ocean heat and salt balances in response to a westerly wind burst in the western equatorial Pacific during TOGA COARE. J. Geophys. Res, 103 , 10289–10311.
Frank, W. M., H. Wang, and J. L. McBride, 1996: Rawinsonde budget analyses during the TOGA COARE IOP. J. Atmos. Sci, 53 , 1761–1780.
Gill, A. E., and E. M. Rasmussen, 1983: 1982–83 climate anomaly in the equatorial Pacific. Nature, 306 , 229–234.
Godfrey, J. S., R. A. Houze Jr., R. H. Johnson, R. Lukas, J-L. Redelsperger, A. Sumi, and R. Weller, 1998: Coupled Ocean–Atmosphere Response Experiment (COARE): An interim report. J. Geophys. Res, 103 , 14395–14450.
Grabowski, W., X. Wu, and M. Moncrieff, 1996: Cloud-resolving modeling of tropical cloud systems during Phase III of GATE. Part I: Two-dimensional experiments. J. Atmos. Sci, 53 , 3684–3709.
Grabowski, W., X. Wu, M. W. Moncrieff, and W. D. Hall, 1998: Cloud-resolving modeling of cloud systems during phase III of GATE. Part II: Effects of resolution and the third spatial dimension. J. Atmos. Sci, 55 , 3264–3282.
Hagan, D., D. Rogers, C. Friehe, R. Weller, and E. Walsh, 1997: Aircraft observations of sea surface temperature variability in the tropical Pacific. J. Geophys. Res, 102 , 15733–15747.
Harrington, J. Y., 1997: The effects of radiative and microphysical processes on simulated warm and transition season Arctic stratus. Ph.D. dissertation, Atmospheric Science Paper 637, Colorado State University, 289 pp.
Hendon, H. H., and B. Liebmann, 1994: Organization of convection within the Madden–Julian oscillation. J. Geophys. Res, 99 , 8073–8083.
Hendon, H. H., and M. L. Salby, 1994: The life cycle of the Madden–Julian oscillation. J. Atmos. Sci, 51 , 2225–2237.
Hildebrand, P. H., 1998: Shear-parallel moist convection over the tropical ocean: A case study from 18 February 1993 TOGA COARE. Mon. Wea. Rev, 126 , 1951–1976.
Hodur, R. M., 1997: The Naval Research Laboratory's Coupled Ocean/Atmosphere Mesoscale Prediction System (COAMPS). Mon. Wea. Rev, 125 , 1414–1430.
Inall, M. E., K. J. Richards, and G. Eldin, 1998: The changing structure of the upper ocean in the western equatorial Pacific during the Tropical Ocean Global Atmosphere Coupled Ocean–Atmosphere Response Experiment. J. Geophys. Res, 103 , 21385–21400.
Jabouille, P., J-L. Redelsperger, and J. P. Lafore, 1996: Modification of surface fluxes by atmospheric convection in the TOGA COARE region. Mon. Wea. Rev, 124 , 816–837.
Johnson, R. H., and X. Lin, 1997: Episodic trade-wind regimes over the western Pacific warm pool. J. Atmos. Sci, 54 , 2020–2034.
Kayano, M. T., and V. E. Kousky, 1999: Intraseasonal (30–60 day) variability in the global Tropics: Principal modes and their evolution. Tellus, 51A , 373–386.
Lau, K. M., and C-H. Sui, 1997: Mechanisms of short-term sea surface temperature regulation: Observations during TOGA COARE. J. Climate, 10 , 465–472.
LeMone, M. A., E. J. Zipser, and S. B. Trier, 1998: The role of environmental shear and thermodynamic conditions in determining the structure and evolution of mesoscale convective systems during TOGA COARE. J. Atmos. Sci, 55 , 3493–3518.
Lewis, S. A., M. A. LeMone, and D. P. Jorgensen, 1998: Evolution and dynamics of a late-stage squall line that occurred on 20 February 1993 during TOGA COARE. Mon. Wea. Rev, 126 , 3189–3212.
Li, X., C-H. Sui, D. Adamec, and K. M. Lau, 1998: Impacts of precipitation in the upper ocean in the western Pacific warm pool during TOGA COARE. J. Geophys. Res, 103 , 5347–5360.
Lin, X., and R. H. Johnson, 1996a: Kinematic and thermodynamic characteristics of the flow over the western Pacific warm pool during TOGA COARE. J. Atmos. Sci, 53 , 695–715.
Lin, X., and R. H. Johnson, 1996b: Heating, moistening, and rainfall over the western Pacific warm pool during TOGA COARE. J. Atmos. Sci, 53 , 3367–3383.
Louis, J. F., M. Tiedtke, and J. F. Geleyn, 1981: A short history of the PBL parameterizations at ECMWF. Proc. Workshop on Boundary Layer Parameterization, Reading, United Kingdom, ECMWF,. 59–79.
Lukas, R., and E. Lindstrom, 1991: The mixed layer of the western equatorial Pacific Ocean. J. Geophys. Res, 96 , 3343–3357.
Madden, R. A., and P. R. Julian, 1971: Detection of a 40–50 day oscillation of the zonal wind in the tropical Pacific. J. Atmos. Sci, 28 , 702–708.
Madden, R. A., and P. R. Julian, 1994: Observations of the 40–50-day tropical oscillation—A review. Mon. Wea. Rev, 122 , 814–837.
Mapes, B. E., and R. A. Houze Jr., 1995: Diabatic divergence profiles in western Pacific mesoscale cloud systems. J. Atmos. Sci, 52 , 1807–1828.
McBride, J. L., N. E. Davidson, K. Puri, and G. C. Tyrrel, 1995: The flow during TOGA COARE as diagnosed by the BMRC Tropical Analysis and Prediction System. Mon. Wea. Rev, 123 , 717–736.
Mellor, G., and T. Yamada, 1974: A hierarchy of turbulence closure models for planetary boundary layers. J. Atmos. Sci, 31 , 1791–1806.
Moncrieff, M., S. Krueger, D. Gregory, J-L. Redelsperger, and W-K. Tao, 1997: GEWEX cloud system study (GCSS) working group 4: Precipitating convective cloud systems. Bull. Amer. Meteor. Soc, 78 , 831–845.
Mori, K., 1995: Equatorial convection observed by the research vessel Keifu Maru during the TOGA COARE IOP, November 1992. J. Meteor. Soc. Japan, 73 , 491–508.
Ohlmann, J. C., D. A. Siegel, and L. Washburn, 1998: Radiant heating of the western equatorial Pacific during TOGA-COARE. J. Geophys. Res, 103 , 5379–5395.
Olsson, P. Q., J. Y. Harrington, G. Feingold, W. R. Cotton, and S. M. Kreindenweis, 1998: Exploratory cloud-resolving simulations of boundary-layer Arctic stratus clouds. Part I: Warm season clouds. Atmos. Res, 47/48 , 573–597.
Philander, S. G. H., 1990: El Niño, La Niña, and the Southern Oscillation. Academic Press, 289 pp.
Pielke, R., and Coauthors,. 1992: A comprehensive meteorological modeling system—RAMS. Meteor. Atmos. Phys, 49 , 69–91.
Qian, L., G. S. Young, and W. M. Frank, 1998: A convective wake parameterization scheme for use in general circulation models. Mon. Wea. Rev, 126 , 456–469.
Rickenbach, T. M., and S. A. Rutledge, 1998: Convection in TOGA COARE: Horizontal scale, morphology, and rainfall production. J. Atmos. Sci, 55 , 2715–2729.
Roux, F., 1998: The oceanic mesoscale convective system observed with airborne Doppler radars on 9 February 1993 during TOGA-COARE: Structure, evolution and budgets. Quart. Roy. Meteor. Soc, 124 , 585–614.
Saxen, T. R., and S. A. Rutledge, 1998: Surface fluxes and boundary layer recovery in TOGA-COARE: Sensitivity to convective organization. J. Atmos. Sci, 55 , 2763–2781.
Short, D. A., P. A. Kucera, B. S. Ferrier, J. C. Gerlach, S. A. Rutledge, and O. W. Thiele, 1997: Shipboard radar rainfall patterns within the TOGA COARE IFA. Bull. Amer. Meteor. Soc, 78 , 2817–2836.
Soloviev, A., and R. Lukas, 1997a: Observation of large diurnal warming events in the near-surface layer of the western equatorial Pacific warm pool. Deep-Sea Res, 44 , 1055–1076.
Soloviev, A., and R. Lukas, 1997b: Sharp frontal interfaces in the near-surface layer of the ocean in the western equatorial Pacific warm pool. J. Phys. Oceanogr, 27 , 999–1017.
Song, X., C. A. Friehe, and D. Hu, 1996: Ship-board measurements and estimations of air–sea fluxes in the western tropical Pacific during TOGA COARE. Bound.-Layer Meteor, 81 , 373–397.
Su, H., S. S. Chen, and C. S. Bretherton, 1999: Three-dimensional week-long simulations of TOGA COARE convective systems using the MM5 mesoscale model. J. Atmos. Sci, 56 , 2326–2344.
Sun, J., J. F. Howell, S. K. Esbenson, L. Mahrt, C. M. Greb, R. Grossman, and M. A. LeMone, 1996: Scale dependence of air–sea fluxes over the western equatorial Pacific. J. Atmos. Sci, 53 , 2997–3012.
Walko, R., W. Cotton, J. Harrington, and M. Meyers, 1995: New RAMS cloud microphysics parameterization. Part I: The single-moment scheme. Atmos. Res, 38 , 29–62.
Wang, Y., W-K. Tao, and J. Simpson, 1996: The impact of ocean surface fluxes on a TOGA COARE convective system. Mon. Wea. Rev, 124 , 2753–2763.
Webster, P. J., 1994: The role of hydrological processes in ocean–atmosphere interactions. Rev. Geophys, 32 , 427–476.
Wijesekera, H. W., C. A. Paulson, and A. Huyer, 1999: The effect of rainfall on the surface layer during a westerly wind burst in the western equatorial Pacific. J. Phys. Oceanogr, 29 , 612–632.
You, Y., 1995: Salinity variability and its role in the barrier-layer formation during TOGA COARE. J. Phys. Oceanogr, 25 , 2778–2807.
Young, G. S., S. M. Perugini, and C. W. Fairall, 1995: Convective wakes in the equatorial western Pacific during TOGA. Mon. Wea. Rev, 123 , 110–123.
Zhang, C., and H. H. Hendon, 1997: Propagating and standing components of the intraseasonal oscillation in tropical convection. J. Atmos. Sci, 54 , 741–752.
APPENDIX
Coupling Equations
The coupling procedure used in this work is similar to the one described by Hodur (1997). The models interchange information at the boundary through momentum, heat, and water substance fluxes. As in most atmosphere–ocean coupled models, a “salinity flux” is used to mimic the freshening or salting processes due to precipitation or evaporation, respectively. Upward fluxes are assumed as positive.
Scatterplot of the vertical velocity in pressure coordinates against the near-surface zonal wind, showing cases with large-scale ascending motion stronger than −10 mb day−1. Favorable conditions for deep convection during the COARE IOP are shown to have occurred not only under moderate westerlies (as the ones preceding the maximum in a westerly wind burst), but also under easterlies and under weak westerlies
Citation: Journal of the Atmospheric Sciences 58, 22; 10.1175/1520-0469(2001)058<3357:COCRSO>2.0.CO;2
Observations over the COARE IFA during Dec 1992: (a) zonal wind, in m s−1; (b) near-surface wind speed, in m s−1; (c) vertical velocity in pressure coordinates (ω), in mb day−1; (d) large-scale advective temperature tendency, in K day−1, and (e) large-scale advective moisture tendency, multiplied by L/cp, in K day−1. The two periods simulated in this paper are indicated
Citation: Journal of the Atmospheric Sciences 58, 22; 10.1175/1520-0469(2001)058<3357:COCRSO>2.0.CO;2
(a) Precipitation, in mm day−1, and (b) surface heat fluxes, in Wm−2, as calculated from the observations and simulated by the coupled model for case 1
Citation: Journal of the Atmospheric Sciences 58, 22; 10.1175/1520-0469(2001)058<3357:COCRSO>2.0.CO;2
Observed and modeled (a) temperature and (b) salinity at a depth of 2 m for case 1
Citation: Journal of the Atmospheric Sciences 58, 22; 10.1175/1520-0469(2001)058<3357:COCRSO>2.0.CO;2
Same as in Fig. 3 but for case 2
Citation: Journal of the Atmospheric Sciences 58, 22; 10.1175/1520-0469(2001)058<3357:COCRSO>2.0.CO;2
Same as in Fig. 4 but for case 2
Citation: Journal of the Atmospheric Sciences 58, 22; 10.1175/1520-0469(2001)058<3357:COCRSO>2.0.CO;2
Hovmöller diagrams of the (left) surface precipitation, in mm h−1; (center) SSS, in psu; and (right) SST for case 1. The contours at the left and center panels are the 0.1 mm h−1 and the 34.0 psu isolines, respectively
Citation: Journal of the Atmospheric Sciences 58, 22; 10.1175/1520-0469(2001)058<3357:COCRSO>2.0.CO;2
Same as in Fig. 7 but for case 2
Citation: Journal of the Atmospheric Sciences 58, 22; 10.1175/1520-0469(2001)058<3357:COCRSO>2.0.CO;2
Scatterplot of the simulated SSS against the SST for (a) case 1 and (b) case 2. The gray lines represent idealizations of (a) the diurnal heating and cooling and (b) the cooling and freshening associated with precipitation
Citation: Journal of the Atmospheric Sciences 58, 22; 10.1175/1520-0469(2001)058<3357:COCRSO>2.0.CO;2
Spatial correlation between the SST and the SSS, as a function of time, for (a) case 1 and (b) case 2
Citation: Journal of the Atmospheric Sciences 58, 22; 10.1175/1520-0469(2001)058<3357:COCRSO>2.0.CO;2
A newly formed freshwater lens and its parent storm at 1900 UTC 22 Dec. (top) The total condensate mixing ratio (atmosphere) and (bottom) the salinity field (ocean). The upper and lower panels have different vertical scales
Citation: Journal of the Atmospheric Sciences 58, 22; 10.1175/1520-0469(2001)058<3357:COCRSO>2.0.CO;2
Hovmöller diagrams of (a) SSS, in psu, and vertical velocity, in cm s−1, and (b) SST, in °C, and zonal current, in m s−1, for a fresh anomaly. Only part of the model domain (0 < x < 40 km) is shown
Citation: Journal of the Atmospheric Sciences 58, 22; 10.1175/1520-0469(2001)058<3357:COCRSO>2.0.CO;2
Vertical cross section of the same fresh anomaly depicted in Fig. 11, in a later time (0400 UTC 23 Dec), showing (a) salinity, in psu, and vertical velocity, in cm s−1, and (b) ocean temperature, in °C, and zonal current, in m s−1
Citation: Journal of the Atmospheric Sciences 58, 22; 10.1175/1520-0469(2001)058<3357:COCRSO>2.0.CO;2
Time evolution of (a) Qs,c − Qs, (b) Qs,nt − Qs, (c) Qs,nsst − Qs, and (d) Qs,ave − Qs for (left) cases 1 and (right) 2. All values are in Wm−2
Citation: Journal of the Atmospheric Sciences 58, 22; 10.1175/1520-0469(2001)058<3357:COCRSO>2.0.CO;2
Time evolution of (a) Ql,c − Ql, (b) Ql,nt − Ql, (c) Ql,nsst − Ql, and (d) Ql,ave − Ql for (left) cases 1 and (right) 2. All values are in Wm−2
Citation: Journal of the Atmospheric Sciences 58, 22; 10.1175/1520-0469(2001)058<3357:COCRSO>2.0.CO;2
Nine-day averaged surface precipitation, sensible heat flux, and latent heat flux, as calculated from the observations and simulated by the coupled model (case 1)
Same as in Table 1 but for case 2
Surface sensible and latent heat fluxes calculated from the bulk formulas: 3 (Qs and Ql), 2 (Qs,ave and Ql,ave), 1 (Qs,c and Ql,c), 4 (Qs,nt and Ql,nq), and 5 (Qs,nsst and Ql,nsst) for cases 1 and 2. See text for details
