Development of a Coupled Regional Model and Its Application to the Study of Interactions between the West African Monsoon and the Eastern Tropical Atlantic Ocean

Samson M. Hagos Department of Earth and Atmospheric Sciences, Cornell University, Ithaca, New York

Search for other papers by Samson M. Hagos in
Current site
Google Scholar
PubMed
Close
and
Kerry H. Cook Department of Earth and Atmospheric Sciences, Cornell University, Ithaca, New York

Search for other papers by Kerry H. Cook in
Current site
Google Scholar
PubMed
Close
Full access

Abstract

A regional ocean–atmosphere coupled model is developed for climate variability and change studies. The model allows dynamic and thermodynamic interactions between the atmospheric boundary layer and an ocean mixed layer with spatially and seasonally varying depth prescribed from observations. The model reproduces the West African monsoon circulation as well as aspects of observed seasonal SST variations in the tropical Atlantic. The model is used to identify various mechanisms that couple the West African monsoon circulation with eastern Atlantic SSTs. By reducing wind speeds and suppressing evaporation, the northward migration of the ITCZ off the west coast of Africa contributes to the modeled spring SST increases. During this period, the westerly monsoon flow is expanded farther westward and moisture transport on to the continent is enhanced. Near the end of the summer, upwelling associated with this enhanced westerly flow as well as the solar cycle lead to the seasonal cooling of the SSTs. Over the Gulf of Guinea, the acceleration of the southerly West African monsoon surface winds contributes to cooling of the Gulf of Guinea between April and July by increasing the entrainment of cool underlying water and enhancing evaporation.

Corresponding author address: Samson Hagos, RSMAS/MPO, University of Miami, 4600 Rickenbacker Causeway, Miami, FL 33149. Email: shagos@rsmas.miami.edu

Abstract

A regional ocean–atmosphere coupled model is developed for climate variability and change studies. The model allows dynamic and thermodynamic interactions between the atmospheric boundary layer and an ocean mixed layer with spatially and seasonally varying depth prescribed from observations. The model reproduces the West African monsoon circulation as well as aspects of observed seasonal SST variations in the tropical Atlantic. The model is used to identify various mechanisms that couple the West African monsoon circulation with eastern Atlantic SSTs. By reducing wind speeds and suppressing evaporation, the northward migration of the ITCZ off the west coast of Africa contributes to the modeled spring SST increases. During this period, the westerly monsoon flow is expanded farther westward and moisture transport on to the continent is enhanced. Near the end of the summer, upwelling associated with this enhanced westerly flow as well as the solar cycle lead to the seasonal cooling of the SSTs. Over the Gulf of Guinea, the acceleration of the southerly West African monsoon surface winds contributes to cooling of the Gulf of Guinea between April and July by increasing the entrainment of cool underlying water and enhancing evaporation.

Corresponding author address: Samson Hagos, RSMAS/MPO, University of Miami, 4600 Rickenbacker Causeway, Miami, FL 33149. Email: shagos@rsmas.miami.edu

1. Introduction

The purpose of this paper is twofold. The first purpose is to describe the development of a coupled regional ocean–atmosphere–land model. The impetus for the development of this model is to advance our understanding of how monsoon circulations, which develop as a result of land–sea contrast, modify and are influenced by SSTs. The interaction comes about primarily through the local surface heat balance and the application of wind stresses on the ocean surface, which can induce upwelling as well as horizontal advection of water within the ocean mixed layer. Thus, a basic understanding of this land–atmosphere–ocean interaction does not require a simulation of the global-scale ocean circulation but would benefit from a higher-resolution simulation that better represents coastlines, topography, upwelling regions, and surface winds, and provides a representation of the monsoon that is better than what is typically generated in coupled GCMs.

The second purpose is to apply the coupled regional ocean–atmosphere–land model to improve our understanding of the West African monsoon system, including its role in determining the pronounced observed seasonality of SSTs in the eastern tropical Atlantic. The West African monsoon system is chosen for this study because this system features strong atmosphere–land surface coupling and also because this monsoon circulation is thought to play an important role in determining the seasonality of eastern tropical Atlantic SSTs. Also, previous experience in modeling this monsoon system in a regional atmospheric model (Vizy and Cook 2002; Hsieh and Cook 2005; Hagos and Cook 2007) demonstrates that the regional model is able to capture the monsoon system more realistically than the current generation of coupled GCMs (Cook and Vizy 2006).

The following section provides background on the evidence for coupling among the land, ocean, and atmosphere within the West African monsoon system and identifies outstanding problems. This review includes a discussion of our current understanding of SST seasonality in the eastern Atlantic, and the extent to which this seasonality is captured in modeling studies. Section 3 describes the coupled atmosphere–ocean–land regional model, and in section 4 the application of the model over West Africa and the adjacent Atlantic Ocean is presented.

2. Background

a. The West African monsoon system

The West African summer monsoon is a dominant component of the regional hydrological cycle on which the livelihood of a growing population is dependent. The system comprises surface flow across the Guinean and western coasts on to the continent. Figure 1 shows the mean summer [June–September (JAS)] surface winds from the National Centers for Environmental Prediction–National Center for Atmospheric Research (NCEP–NCAR) reanalysis project (NNRP; Kalnay et al. 1996) and precipitation from Tropical Rainfall Measuring Mission (TRMM)-3B42V6 (Huffman et al. 2001). During the summer, the confluence of the southwesterly monsoon and the northeasterly Harmattan winds is as far as 20°N and the precipitation maximum is at about 10°N in the continental interior, with local maxima over Sierra Leone (8°N) and the Cameroon highlands. During this period, the Atlantic marine ITCZ, along with the confluence of the southeasterly and northeasterly surface winds, is at its northernmost latitude and the easterly surface winds over the central Atlantic are weak.

The West African monsoon surface winds are primarily driven by temperature contrast between the land and the adjacent ocean surfaces. However, the associated surface heat and momentum fluxes, moisture transport, and precipitation introduce various feedback processes that render the system inherently nonlinear. For example, both the low-level westerly winds across the western coast of Africa and the African easterly jet (AEJ) aloft, which maintain the zonal moisture transport, are sensitive to the north–south land surface temperature gradient. This temperature gradient in turn is tied to the precipitation through soil moisture content (Cook 1999).

This land–atmosphere coupling operates at various time scales. For example, it is implicated in the rapid drying of the Sahara at the end of the African humid period (Patricola and Cook 2007). On seasonal time scales, the relatively long memory of soil moisture has been suggested as a possible source of predictability (Douville et al. 2001). In addition, these land surface processes are known to enhance the response of the precipitation to remote forcing on decadal time scales (Zeng et al. 1999).

Various air–sea interaction mechanisms are believed to couple the monsoon dynamics with the regional SSTs (Okumura and Xie 2004; Li and Philander 1996). By partially controlling surface winds and the associated moisture transport, SSTs over the Atlantic Ocean have significant influence on variability at interannual to decadal time scales. The two-way interaction of the West African monsoon with eastern tropical Atlantic SSTs, specifically the role of the West African monsoon dynamics in the seasonal SST cycle and the associated feedback onto the monsoon, may contribute to better understanding interannual variability and improve prediction.

b. Seasonality of SSTs in the eastern tropical Atlantic

Sea surface temperature variability over eastern tropical Atlantic is marked by a strong annual cycle. During boreal spring, the sun is directly over the Gulf of Guinea, the trade winds are weak, and a band of high SSTs lies along the equator from 10°S to 5°N. As the year progresses, the trade winds along the equator intensify and a rapid decline of SSTs over the eastern equatorial Atlantic follows. For example, at 10°W, equatorial SSTs reach 28°C in April and drop below 23°C in July and August. This seasonal warming and cooling is highly asymmetric, with the latter taking only three months and the former seven months (Xie and Carton 2004). Over the northeastern tropical Atlantic, on the other hand, April–August is a period of rapid warming.

Identifying the role of the West African monsoon winds in this pronounced seasonal SST cycle has been a subject of extensive study. Mitchell and Wallace (1992) proposed that the onset of the summer monsoon is instrumental in initiating the rapid cooling over the equatorial region. According to their observational analysis of SSTs, surface winds, and outgoing longwave radiation (OLR), the intensification of the northward wind stress across the equator induces a remote response, bringing colder water to the surface just south of the equator to form the characteristic equatorial cold tongue. They also hypothesize that the marked equatorial asymmetry in the annual mean SST distribution, with warmer waters in the northern tropics, may partly be due to the continental geometry. For example, the fact that the African coastline is parallel to the trade winds near the equatorial region, they argue, favors upwelling and advection of cold water, while the north–south alignment of the west coast of northern Africa favors weakening of the easterly trades.

Subsequent studies using atmospheric GCMs and simple ocean models support this general hypothesis. By forcing the Geophysical Fluid Dynamics Laboratory (GFDL) R30 GCM by SSTs with and without the seasonal cycle, Li and Philander (1997) showed that the enhancement of the meridional winds and the associated surface stress are indeed forced by the warming of the African land surface, and the feedback due to the seasonal SST cycle does not have strong influence. In the same study, using the Cane–Zebiak ocean model (Zebiak and Cane 1987) coupled with a one-layer atmospheric model, they also show that about 60% of the seasonal SST variations over the Gulf of Guinea can be attributed to variations in evaporation, while other processes such as changes in thermocline depth and cloud feedback play secondary roles.

In contrast, recent studies suggest a more active role of the ocean surface dynamics and associated feedbacks. An atmospheric GCM study by Okumura and Xie (2004), for example, suggests that the southerly flow across the equator is itself enhanced by the cooling. According to their results, the land surface warming drives cross-equatorial southerlies, which induce oceanic upwelling south of the equator that cools the eastern equatorial Atlantic. This cooling further intensifies the southerly monsoon flow in the Gulf of Guinea and accelerates the northward shift of the precipitation band and monsoon onset.

The strong seasonality of the tropical Atlantic Ocean SST has proven remarkably difficult to simulate using coupled GCMs. In an intercomparison study of 23 models, Davey et al. (2002) demonstrate that the seasonal SST range of most models with no flux adjustment is small compared to the observations. However, they show that many of the models correctly place the maximum in the SST cycle over the eastern tropical Atlantic. DeWitt (2005) shows that a weak zonal wind stress along the equator may be the cause of the incorrect zonal equatorial SST gradient in the ECHAM GCM coupled with the Modular Ocean Model (MOM). In his study, the ocean model was forced by observed and simulated surface winds and the bias in the SST gradient is primarily due to the coupled model’s weak zonal wind stress. In their recent analysis of the Community Climate System Model, version 3 (CCSM3) zonal SST gradient problem, Deser et al. (2006) point to the poor representation of the West African monsoon by its atmospheric component [Community Atmosphere Model, version 3 (CAM3)] even when forced by observed SSTs. They argue that the fact that the model puts the summer Sahel precipitation too far inland might affect the air–sea interaction and introduce the bias. Most recently, Richter and Xie (2008) studied the problem using an ensemble of coupled GCMs. They show that the models with such a bias also suffer a bias in surface winds (westerly bias) and precipitation, with excessive precipitation over Africa and drier conditions over South America in spring. Therefore, understanding the processes through which atmospheric processes over land influence ocean surface dynamics and thermodynamics is likely an important step toward correcting these biases. In summary, while the seasonality of SST variability over the Gulf of Guinea and the influence of the monsoon have been studied extensively, the interaction of the monsoon with variability of northeastern Atlantic SSTs has not received as much attention. This application of a coupled regional climate model to study the interaction of the monsoon with both northeastern Atlantic and Gulf of Guinea SSTs provides an appropriate context for both evaluating the regional coupled modeling and improving our understanding of the land–air–sea processes associated with monsoon dynamics.

3. Coupled regional climate model development

a. Model and experiment design

The atmospheric component of the coupled regional climate model (CRCM) is an adaptation of the fifth-generation Pennsylvania State University–National Center for Atmospheric Research Mesoscale Model, version 3.7 (PSU–NCAR MM5-V3.7; Grell et al. 1994), also described and applied for African monsoon and variability studies in Vizy and Cook (2002) and Hagos and Cook (2007). The model is nonhydrostatic and solves the equations governing momentum, temperature, pressure, moisture, and liquid water on σ surfaces. Physical parameterizations were tested to produce a realistic representation of the circulation and precipitation over the model domain. The Kuo scheme (Kuo 1974) is used to represent convection, the radiation scheme is adapted from that of the NCAR Community Climate Model version 2 (Joseph et al. 1976; Kiehl and Briegleb 1991), and boundary layer processes are represented using the medium-range forecasting (MRF) planetary boundary layer scheme (Troen and Mahrt 1986).

The NCEP Oregon State Air Force Hydrological Laboratory (Noah) land surface model is used to represent land surface processes (Chen et al. 1996). The surface skin temperature is determined by a linearized surface energy balance, and heat transfer across the four soil layers is represented by the diffusion equation. Soil heat capacity and conductivity are functions of the volumetric soil water content, which is governed by Richardson’s equation. Land use types are specified according to observations and held fixed throughout the model integration.

In the ocean component of the model, the evolution of SSTs is governed by conservation of energy (see Navarra 1999; Sterl and Hazeleger 2003) according to
i1520-0442-22-10-2591-e1

In (1) the left-hand side is the net heat gain by the mixed layer (Qnet) where T is the temperature of the ocean mixed layer, which is assumed to be isothermal. Here ρ and Cp are density and specific heat capacity of water, h is the mixed layer depth, Ve is the Ekman current velocity, T is the temperature at the base of the mixed layer, and we is the vertical entrainment velocity. For brevity, the first two terms on the RHS of (1) will be referred to as Qadv and Qent, respectively. The Qrad is the net radiative heating, and Qlat and Qsen are the latent and sensible heat fluxes from the ocean to the atmosphere, respectively. The last term represents heat transfer by vertical diffusion, where κ is the diffusion coefficient.

As in the Cane–Zebiak model and Zebiak and Cane (1987), the dynamics of the Ekman current, Ve = (ue, υe), where ue and υe are the zonal and meridional components, is governed by a three-way balance among the mean friction in the mixed layer, the Coriolis force, and the surface wind stress τ = (τx,τy) according to
i1520-0442-22-10-2591-e2
i1520-0442-22-10-2591-e3
Here r is the mean friction coefficient in the mixed layer and f is the Coriolis parameter. The vertical entrainment velocity associated with the deepening of the mixed layer and wind-driven divergence we is given by
i1520-0442-22-10-2591-e4

If the mixed layer is thickening, it is entraining cold water. If it is getting shallower, SSTs are not affected. Therefore, Γ = 1 if the term in the bracket is positive and Γ = 0 otherwise.

At every time step, the heat and momentum flux terms (Qrad, Qlat, Qsen, and τ) are calculated in the atmospheric PBL scheme and communicated to the ocean model. In the ocean model, the net heat flux (QradQlatQsen), ue, and we are calculated. Heat transport by advection and entrainment are then calculated, and the diffusion term ∂2 T/∂z2 is approximated by (TT)/h2. Once the SST tendency [LHS of (1)] is calculated, the SST is updated and fed back to the PBL parameterization where it is used to derive the heat and momentum flux terms for the next time step in the atmospheric model.

In the ocean model, spatial derivatives at an ocean point surrounded by other ocean points are calculated using central finite differencing. If the ocean variable is not defined at a neighboring point, that is, if a neighboring point is on land, forward or backward difference is used. Simple forward differencing is used to advance SSTs in time.

The seasonal cycle of the ocean mixed layer depth is incorporated using 12-hourly data obtained by interpolating the Naval Research Laboratory monthly climatology of mixed layer depth (Kara et al. 2003). In accordance with the definition of mixed layer depth in this dataset, the temperature at the base of the mixed layer is set to be T = Tobs − 0.8K, where Tobs is the climatological annual mean SST from Reynolds and Smith (1995). The vertical diffusion coefficient κ is 10−5 m2 s−1 (as suggested by Bryan and Lewis 1979), and Cp is 4218 J K−1 kg−1. To avoid singularity at the equator, the Rayleigh friction coefficient in the Ekman transport equations is set to r = 0.5 day-1 (as in Sterl and Hazeleger 2003 and Okumura and Xie 2004).

The model simulations are run over a rectangular domain enclosed by 30°E–65°W and 30°S–30°N for 1 yr, and another test run of 2-yr length was performed to test the stability of the model. The 90-km grid spacing used in these simulations allows for resolution of the important features over a domain large enough to cover West Africa and the tropical Atlantic Ocean. There are 23 vertical σ levels and the model time step is 90 s. The top of the atmosphere is fixed at 50 hPa and an upper radiative boundary condition is used. The effects of snow cover are neglected. The Noah land surface model (LSM) is initialized by soil moisture and temperature fields at the 10- and 200-cm levels from the NCEP–NCAR reanalysis project. While several experiments were performed to analyze the stability of the model and its representation of observed features, the main results presented here pertain to a 1-yr-long experiment with the above described setup. It starts on 15 October, with the first 2 weeks discarded as a spinup period.

b. Evaluation

Figure 2 displays the mean JAS 10-m wind and precipitation from the CRCM simulation. Comparison of Figs. 1 and 2 shows that the model reproduces the northeasterly and southeasterly winds over the tropical Atlantic well. In both the reanalysis and the coupled model output, the surface winds are westerly over the northeastern tropical Atlantic (near 10°N) and directed toward the continent; however, in the NNRP, the westerly surface winds into the continent are weaker. In situ measurements suggest that the NNRP underestimates this westerly low-level flow. For example, field studies from the Global Atmospheric Research Program Atlantic Tropical Experiment (GATE) showed these shallow westerly winds contained inside the frictional boundary layer (depth of 2 km) between 0° and 20°N attain maximum speeds of 5 m s−1 at around 8°N (Grodsky et al. 2003). The GATE observations indicated that these shallow westerly winds occur over vast areas of the tropical Atlantic sector.

The mean summer precipitation in the model (Fig. 2) is compared with the TRMM data in Fig. 1. The precipitation maxima over the coast of Africa (15°W) and the Cameroon highlands (10°E) are well simulated, but the precipitation along the coast of Guinea is relatively excessive. The precipitation over the central Atlantic is shifted north by about 5° and is weaker over the western tropical Atlantic. These biases in precipitation are related to SST biases in the model, which are the subject of the subsequent discussion.

Here the performance of the model in capturing the monsoon onset and the mean summer precipitation is evaluated. The modeled daily latitudinal migration of precipitation averaged over West Africa (10°E–10°W, Fig. 3a) is compared with various observational datasets. The 1998–2006 mean summer precipitation from TRMM (3B42V6) is displayed in Fig. 3b. Figures 3c,d display the 1998–2004 mean summer precipitation from the Global Precipitation Climatology Project (GPCP; Huffman et al. 1997) and from the Famine Early Warning System (FEWS; Herman et al. 1997) data.

The model captures the seasonal migration of the precipitation well. Specifically, the magnitude of the simulated precipitation is comparable to the observations, and the date of the simulated monsoon jump, the day at which the 10-day running mean precipitation along 10°N is greater than that along the coastline (marked by the dashed lines), is within the range of the observations. The meridional extent of the model precipitation in May is narrower than in the observations, and the modeled precipitation over the Gulf of Guinea is on average 2 mm day−1 drier than the observations.

The simulated mean annual SST is shown in Fig. 4a. The shading in Fig. 4a denotes the difference between the simulated and observed SSTs. Over most of the central Atlantic, the mean simulated SSTs are within 1 K of the observations. In general, there is a warm bias of up to 3 K over the eastern Atlantic and a cold bias of similar magnitude over the western tropical Atlantic. This model shares the bias in the east–west SST gradient with many state-of-the-art global models discussed in the introduction. The distribution and magnitude of these biases are remarkably similar to those found in the CCSM3 (Chang et al. 2007), suggesting a common cause despite the differences in the model setup, especially in their oceanic components.

Figures 4b,c show the evolution of simulated and observed area-averaged SSTs over the northeastern Atlantic (NETA; the region enclosed by 10°–22°N and 18°–38°W in Fig. 4a) and the Gulf of Guinea (GOG; 10°S–3°N and 10°E–10°W in Fig. 4a). The model reproduces the April–July rapid warming over the NETA region with a bias of about 0.5 K. The rapid observed April–July cooling of GOG SSTs during that period is captured in the simulation, but as the summer season progresses the bias becomes significant (about 2 K). During the summer, the model accounts for about half of the seasonal change in SSTs over the GOG. To assess the stability of the model, an additional test run is performed for a longer time period. The evolution of SSTs in this test run is shown in Fig. 5. Over both NETA and GOG the bias is fairly constant from year to year but shows significant seasonality in that it is largest in cold seasons, once again pointing to the model’s short comings in correctly representing the cooling processes.

In summary,

  • The model simulates surface winds, summer precipitation, and monsoon seasonal cycle reasonably well.

  • It reproduces the seasonal SST variability over northeastern tropical Atlantic with a mean bias of less than 1 K.

  • Even though the model has significant biases of about 2 K over the Gulf of Guinea through the summer, the reasonable simulation of April–July SST variations by the CRCM suggests that the April–July rapid cooling over the Gulf of Guinea is partly related to the monsoon dynamics. This connection is further analyzed in the following section.

  • Sensitivity tests show that the overall bias in the zonal SST gradient, which the model shares with coupled GCMs, is at least partly related to the relatively weak wind stress. Appropriate flux corrective steps could be used for multiyear simulations depending on the purpose of the study.

4. Interaction of the West African monsoon circulation with the northeastern tropical Atlantic Ocean

As noted in the previous section, the model’s performance in simulating the mean SST and seasonal cycle varies spatially. In particular, the model simulates the northeastern tropical Atlantic SST seasonal cycle considerably better than it does the Gulf of Guinea seasonal SST cycle. Here we examine the dynamic and thermodynamic interactions of the monsoon with the northeastern tropical Atlantic.

A surface energy budget analysis shows that the primary balance over the NETA region is between radiative heating and evaporation cooling. Figure 6a shows this balance. Entrainment cooling is small during spring. During the fall and winter, evaporative cooling dominates over radiative heating and there is a net loss of heat from the mixed layer. During the spring and summer, radiative heating exceeds evaporative and entrainment cooling and there is a net heat flux into the ocean mixed layer.

Most of the seasonal variation in the mixed layer heat content is related to seasonal variations in radiative heating. However, the roles of the suppressed evaporation and sensible cooling between May and August and increased entrainment cooling between July and September are significant. In particular the combined contribution from the suppression of evaporation and sensible cooling becomes comparable to that of radiative heating in July. By the end of the summer, entrainment cooling becomes the primary process associated with the SSTs’ decline. At their respective maxima, contributions from the suppression of evaporation and increased entrainment cooling are comparable to that of radiative heating. This suggests an important role of the surface winds in the seasonal SST cycle. These results agree with results of a numerical study by Carton and Zhou (1997), which shows that much of the annual SST cycle over the northern tropical Atlantic is related to reduced evaporation because of weakened surface winds. Energy budget analyses using the Pilot Research Array in the Tropical Atlantic (PIRATA) moored buoy have also shown similar results (Foltz et al. 2003).

It is important to note the role of the prescribed seasonal cycle of the mixed layer depth in the cycle of SST. On the one hand, by modulating the heat capacity of the water column, the prescribed variations of mixed layer depth affect SST seasonal cycle. For example, the impact of entrainment cooling on SST between July and September is enhanced by the relatively shallow mixed layer during that period. The entrainment associated with the deepening of the mixed layer [the first term in RHS of (4)] contributes to the variations in the heat content. Hereafter, aspects of the variations in mixed layer heat content involving direct transfer of heat and momentum between the ocean mixed layer and atmospheric boundary layer mixed layer are considered.

The seasonal cycle of surface wind stress from the model and observations are considered. Figure 7 shows wind stress over NETA from the CRCM (Fig. 7a), Quick Scatterometer (QuikSCAT) 2000–04 climatology (Fig. 7b) (Klinger et al. 2006), and Goddard Satellite Retrievals, Version 2 (Fig. 7c) (GSSTF2; Chou et al. 2003). In both the model and the observations the rapid decline of surface wind stress between April and July is apparent. Much of the variability in surface stress is in the meridional component. However, there are some significant differences. The seasonal variability of the model is about 50% of both observations, and the stress reaches its peak later than both observations. That explains some of the warm bias in the model SST through the underestimate of entrainment; some test runs (not shown) performed by artificially doubling stress reduce the warm bias both over NETA and GOG. As discussed in the introduction, the period of transition between northern spring and summer is marked by the northward migration of the line of confluence of the northeasterly and southeasterly winds. Figures 8a and 8b show the evolution of surface pressure and the meridional surface winds over the NETA from the CRCM and the NNRP, respectively. In both panels, the northward migration of the southerly winds (the −5 m s−1 contour, for example) is apparent, and it leaves the NETA (which is denoted by the dashed lines) with low wind speed. This is related to the northward migration of the ITCZ low pressure. This is demonstrated by the tilt of the 1016-hPa and 0 m s−1 contour in Fig. 8a.

In addition to influencing heat transfer between the atmosphere and the ocean, the winds over the NETA apply stress to the ocean surface, which induces entrainment. As shown in the energy budget analysis, this entrainment is the primary cooling mechanism at the end of the summer. Figure 9 shows the surface winds and the associated wind-driven ocean currents in August (the month in which entrainment over NETA is at its maximum). Because of the Coriolis effect, the strong westerly flow onto the continent near 10°N drives southeasterly currents (the North Equatorial Counter Currents). This results in divergence and entrainment cooling to the north of the surface wind maximum. The same effect of the westerly flow is also identified in an observational and numerical study by Grodsky et al. (2003). The southeasterly surface winds south of the equator drive the modeled South Equatorial Currents. Among the limitations of the model is the fact that the North Brazil Current (NBC) and North Equatorial Current (NEC) are essentially absent, likely because the modeled stresses are too weak.

The primarily nondivergent westerly surface flow onto the continent is part of the cyclonic circulation associated with the African heat low (Fig. 10). During the summer, increased soil moisture and reduced surface temperature enhances the temperature gradient between the Sahel and the Sahara and leads to the northward expansion of this westerly low-level flow. Its westward extension also exhibits significant seasonal variation. The direction of the zonal surface winds over the NETA depends on the competition between the North Atlantic high, which favors easterly flow, and the African heat low, which favors westerlies. During the summer, enhanced land–sea contrast enables the strengthening and westward expansion of the westerly winds (cf. Figs. 10a and 10b).

SST seasonal cycle controls the availability of moisture for transport into the continent. Figure 11 shows the zonal wind and mixing ratio off the west coast of Africa in April and July. In summer, the increased mixing ratio coupled with the increased westerly flow into the continent supplies moisture for the inland monsoon precipitation. By enhancing the saturation mixing ratio, the seasonal warming increases the amount of moisture available for transport into the continent.

5. Interaction of the West African monsoon circulation with Gulf of Guinea SSTs

Because the model fails to fully account for seasonal SST variations over the GOG region, comparison of the energy budget terms could lead to erroneous conclusions. However, enumeration of the important processes simulated provides some information on the monsoon–ocean interaction in this region. The seasonal cooling over the Gulf of Guinea is accompanied by an enhancement of the surface wind stress. Figure 12 shows the evolution of the area-averaged wind stress from the model, QuikSCAT and GSSTF2, and the zonal and meridional components over the GOG. The rapid increase in wind stress between April and July coincides with the decrease of GOG SSTs. Most of the wind stress change is related to the meridional component, which increases by about 50% in both the model and the observed climatologies. Note once again the overall underestimate of stress by the model when compared with both observations. In the model surface stress is proportional to the square of the surface wind speed and hence is very sensitive to variations in surface winds.

This enhanced stress over GOG due to the southerly winds induces entrainment. Figure 13 shows the surface winds and the associated wind-driven ocean currents in May, when the entrainment is at its maximum. Over the southeastern tropical Atlantic, there is little meridional variation in the magnitude of the surface winds. For example, the wind speed at 0°, 0° is about 6 m s−1 southerly and that over 0°, 10°S is also about 6 m s−1, but southeasterly (Fig. 13a). But because of the variation of the Coriolis parameter with latitude, the response of the ocean surface to the stress exerted at the two latitudes is different. On the equator, the primary balance is between the surface stress and friction in the mixed layer. Therefore, the ocean currents are in the same direction as the surface winds (Fig. 13b). At 10°S, on the other hand, friction and Coriolis forces are comparable and the surface currents are oriented at an angle of about 45° to the surface winds. This results in the divergence of ocean currents and upwelling. This process is also identified in a numerical study by Philander and Pacanowski (1981). Similarly, the southerly winds near the coast induce easterly currents and cooling (shaded) because of coastal upwelling along the southern coast of Africa. The relatively shallow mixed layer thickness over the region also contributes to the higher sensitivity of SSTs to variations in the net heat flux (Fig. 13b). Increased evaporation and decreases in radiative cooling also contribute to the accelerated cooling in the model.

To estimate the relative roles of variations in pressure over land and ocean in forcing the acceleration of these southerly winds, the seasonal evolution of the surface pressure is considered. Figures 14a and 14b show the evolution of the mean surface pressure (shaded) and the meridional velocity (contours) from the CRCM and the NNRP, respectively. They show that, while the African heat low is in place for most of the year, the meridional wind speed near the equator varies with pressure over the Gulf of Guinea. In both the CRCM and the NNRP, the meridional wind speed reaches its maximum near (and slightly ahead) of the surface pressure to the immediate south.

6. Discussion

In this study the development of a regional coupled model and its application to ocean–atmosphere–land interactions over West Africa and the adjacent Atlantic Ocean are presented. The model includes both dynamic and thermodynamic air–sea interactions as well as a land surface scheme. It is a coupling of the PSU–NCAR MM5 regional climate model with an ocean mixed layer model with observed spatially and seasonally varying mixed layer depth.

The ability of the model to simulate the monsoon dynamics and precipitation and the SST seasonal cycle are evaluated. The model simulates the monsoon precipitation as well as its seasonal cycle well, and the associated surface winds compare well with the NCEP–NCAR reanalysis. Without any flux corrective measures the model captures the full seasonal cycle of SSTs over the northeastern tropical Atlantic, but the simulated seasonality over the Gulf of Guinea is smaller than in the observations. In particular, the modeled summer SSTs over the Gulf of Guinea are warmer by up to 3 K.

Air–sea interactions over the northeastern tropical Atlantic, off the west coast of Africa, are further investigated. While the primary factor for the seasonal SST cycle in this region is solar heating, suppression of evaporation during early summer and entrainment cooling at the end of the summer play important roles, comparable to the radiative heating variation. The suppression of evaporation is mainly due to the northward migration of the ITCZ and the associated reduction of meridional wind speed. The seasonal SST warming weakens the North Atlantic high, weakening the associated anticyclonic easterly winds and favoring the westward expansion of the cyclonic westerly flow associated with the African heat low. This, coupled with the seasonal northward migration of the mixing ratio maximum, enhances moisture transport into the continent. As the summer progresses, the stress associated with the westerly winds induces southerly ocean currents. These currents result in upwelling to the immediate north of the westerly flow and initiate the gradual decline of SSTs. The low-level flow across the western coast is known to be significantly correlated with precipitation over the Sahel on interannual time scales. For example, Grist and Nicholson (2001) show stronger low-level westerly winds during wet years in the western Sahel. Over the Gulf of Guinea the start of the monsoon season is accompanied by the acceleration of southerly winds across the equator. Because of variations of the Coriolis force with latitude, these winds drive upwelling (and cooling) to the immediate south of the equator. In addition, evaporation associated with these accelerated surface winds also significantly contributes to the seasonal cooling.

The model shares SST bias with several of the current coupled GCMs. Diagnostic analyses suggest the bias is related to a problem in the models in simulating the observed surface stress, which is related to large-scale heating gradients involving the distribution of precipitation (Richter and Xie 2008). Even though the development of coupled models that faithfully represent the observed precipitation distribution as well as surface wind stress and SST cycle continues to be a major challenge, analysis of the three-way interactions among Atlantic SSTs, the low-level monsoon flow, and precipitation is an important step toward better understanding of precipitation variability on interannual time scales and beyond.

Acknowledgments

The authors thank Dr. E. K. Vizy, who provided many of the analysis tools used in this study. This work was supported by NSF Award ATM-0415481.

REFERENCES

  • Bryan, J. P., and L. J. Lewis, 1979: A water mass model of the world ocean. J. Geophys. Res., 84 , 25032517.

  • Carton, J. A., and Z. X. Zhou, 1997: Annual cycle of sea surface temperature in the tropical Atlantic Ocean. J. Geophys. Res., 102 , 2781327824.

    • Search Google Scholar
    • Export Citation
  • Chang, C. Y., J. A. Carton, S. A. Grodsky, and S. Nigam, 2007: Seasonal climate of the tropical Atlantic sector in the NCAR Community Climate System Model 3: Error structure and probable causes of errors. J. Climate, 20 , 10531070.

    • Search Google Scholar
    • Export Citation
  • Chen, F. K., and Coauthors, 1996: Modeling of land surface evaporation by four schemes and comparison with FIFE observations. J. Geophys. Res., 101 , 72517268.

    • Search Google Scholar
    • Export Citation
  • Chou, S-H., E. Nelkin, J. Ardizzone, R. M. Atlas, and C-L. Shie, 2003: Surface turbulent heat and momentum fluxes over global oceans based on the Goddard Satellite Retrievals, Version 2 (GSSTF2). J. Climate, 16 , 32563273.

    • Search Google Scholar
    • Export Citation
  • Cook, K. H., 1999: Generation of the African easterly jet and its role in determining West African precipitation. J. Climate, 12 , 11651184.

    • Search Google Scholar
    • Export Citation
  • Cook, K. H., and E. K. Vizy, 2006: Coupled model simulations of the West African monsoon system: Twentieth- and twenty-first-century simulations. J. Climate, 19 , 36813703.

    • Search Google Scholar
    • Export Citation
  • Davey, M., and Coauthors, 2002: STOIC: A study of coupled model climatology and variability in tropical ocean regions. Climate Dyn., 18 , 403420.

    • Search Google Scholar
    • Export Citation
  • Deser, C., A. Capotondi, R. Saravanan, and A. Phillips, 2006: Tropical Pacific and Atlantic climate variability in CCSM3. J. Climate, 19 , 24512481.

    • Search Google Scholar
    • Export Citation
  • DeWitt, D. G., 2005: Diagnosis of the tropical Atlantic near-equatorial SST bias in a directly coupled atmosphere–ocean general circulation model. Geophys. Res. Lett., 32 , L01703. doi:10.1029/2004GL021707.

    • Search Google Scholar
    • Export Citation
  • Douville, H., F. Chauvin, and H. Broqua, 2001: Influence of soil moisture on the Asian and African monsoons. Part I: Mean monsoon and daily precipitation. J. Climate, 14 , 23812403.

    • Search Google Scholar
    • Export Citation
  • Foltz, G., S. A. Grodsky, J. A. Carton, and M. McPhaden, 2003: Seasonal mixed layer heat budget of the tropical Atlantic. J. Geophys. Res., 108 , 3146. doi:10.1029/2002JC001584.

    • Search Google Scholar
    • Export Citation
  • Grell, G. A., J. Dudhia, and D. R. Stauffer, 1994: A description of the fifth-generation Penn State/NCAR mesoscale model (MM5). NCAR Tech. Note NCAR/TN-398+STR, 122 pp.

    • Search Google Scholar
    • Export Citation
  • Grist, J. P., and S. E. Nicholson, 2001: A study of the dynamic factors influencing the rainfall variability in the West African Sahel. J. Climate, 14 , 13371359.

    • Search Google Scholar
    • Export Citation
  • Grodsky, S. A., J. A. Carton, and S. Nigam, 2003: Near surface westerly wind jet in the Atlantic ITCZ. Geophys. Res. Lett., 30 , 2009. doi:10.1029/2003GL017867.

    • Search Google Scholar
    • Export Citation
  • Hagos, S. M., and K. H. Cook, 2007: Dynamics of the West African monsoon jump. J. Climate, 20 , 52645284.

  • Herman, A., V. B. Kumar, P. A. Arkin, and J. V. Kousky, 1997: Objectively determined 10-Day African rainfall estimates created for famine early warming systems. Int. J. Remote Sens., 18 , 21472159.

    • Search Google Scholar
    • Export Citation
  • Hsieh, J. S., and K. H. Cook, 2005: Generation of African easterly wave disturbances: Relationship to the African easterly jet. Mon. Wea. Rev., 133 , 13111327.

    • Search Google Scholar
    • Export Citation
  • Huffman, G. J., and Coauthors, 1997: The Global Precipitation Climatology Project (GPCP) combined precipitation dataset. Bull. Amer. Meteor. Soc., 78 , 520.

    • Search Google Scholar
    • Export Citation
  • Huffman, G. J., R. F. Adler, M. Morrissey, D. T. Bolvin, S. Curtis, R. Joyce, B. McGavock, and J. Susskind, 2001: Global precipitation at one-degree daily resolution from multisatellite observations. J. Hydrometeor., 2 , 3650.

    • Search Google Scholar
    • Export Citation
  • Joseph, J. H., W. J. Wiscombe, and J. A. Weinman, 1976: The delta-Eddington approximation for radiative flux transfer. J. Atmos. Sci., 33 , 24522459.

    • Search Google Scholar
    • Export Citation
  • Kalnay, E., and Coauthors, 1996: The NCEP/NCAR 40-Year Reanalysis Project. Bull. Amer. Meteor. Soc., 77 , 437471.

  • Kara, A. B., P. A. Rochford, and H. E. Hurlburt, 2003: Mixed layer depth variability over the global ocean. J. Geophys. Res., 108 , 3079. doi:10.1029/2000JC000736.

    • Search Google Scholar
    • Export Citation
  • Kiehl, J. T., and B. P. Briegleb, 1991: A new parameterization of the absorptance due to the 15-micron band system of carbon dioxide. J. Geophys. Res., 96 , 90139019.

    • Search Google Scholar
    • Export Citation
  • Klinger, B. A., B. Huang, B. Kirtman, P. Schopf, and J. Wang, 2006: Monthly climatologies of ocean friction velocity cubed. J. Climate, 19 , 57005708.

    • Search Google Scholar
    • Export Citation
  • Kuo, H. L., 1974: Further studies of the influence of cumulus convection on large-scale flow. J. Atmos. Sci., 31 , 12321240.

  • Li, T., and S. G. H. Philander, 1996: On the annual cycle of the eastern equatorial Pacific. J. Climate, 9 , 29862998.

  • Li, T., and S. G. H. Philander, 1997: On the seasonal cycle of the equatorial Atlantic Ocean. J. Climate, 10 , 813817.

  • Mitchell, T., and J. M. Wallace, 1992: The annual cycle in equatorial convection and sea surface temperature. J. Climate, 5 , 11401156.

    • Search Google Scholar
    • Export Citation
  • Navarra, A., 1999: Beyond El Niño Decadal and Interdecadal Climate Variability. Springer, 374 pp.

  • Okumura, Y., and S-P. Xie, 2004: Interaction of the Atlantic equatorial cold tongue and African monsoon. J. Climate, 17 , 35893602.

  • Patricola, C. M., and K. H. Cook, 2007: Dynamics of the West African monsoon under mid-Holocene precessional forcing: Regional climate model simulations. J. Climate, 20 , 694716.

    • Search Google Scholar
    • Export Citation
  • Philander, S. G. H., and R. C. Pacanowski, 1981: The oceanic response to cross equatorial winds (with application to coastal upwelling in low latitudes). Tellus, 33 , 201210.

    • Search Google Scholar
    • Export Citation
  • Reynolds, R. W., and T. M. Smith, 1995: A high-resolution global sea surface temperature climatology. J. Climate, 8 , 15711583.

  • Richter, I., and S. Xie, 2008: On the origin of equatorial Atlantic biases in coupled general circulation models. Climate Dyn., 31 , 587598.

    • Search Google Scholar
    • Export Citation
  • Sterl, A., and W. Hazeleger, 2003: Coupled variability and air–sea interaction in the South Atlantic Ocean. Climate Dyn., 21 , 559571.

    • Search Google Scholar
    • Export Citation
  • Troen, I., and L. Mahrt, 1986: A simple model of the atmospheric boundary layer: Sensitivity to surface evaporation. Bound.-Layer Meteor., 37 , 129148.

    • Search Google Scholar
    • Export Citation
  • Vizy, E. K., and K. H. Cook, 2002: Development and application of a mesoscale climate model for the tropics: Influence of sea surface temperature anomalies on the West African monsoon. J. Geophys. Res., 107 , 4023. doi:10.1029/2001JD000686.

    • Search Google Scholar
    • Export Citation
  • Xie, S-P., and J. A. Carton, 2004: Tropical Atlantic variability: Patterns, mechanisms, and impacts. Ocean–Atmosphere Interaction and Climate Variability, Geophys. Monogr., Vol. 147, Amer. Geophys. Union, 121–142.

    • Search Google Scholar
    • Export Citation
  • Zebiak, S. E., and M. A. Cane, 1987: A model El Niño–Southern Oscillation. Mon. Wea. Rev., 115 , 22622278.

  • Zeng, N., J. D. Neelin, K-M. Lau, and C. J. Tucker, 1999: Enhancement of interdecadal climate variability in the Sahel by vegetation interaction. Science, 286 , 15371540.

    • Search Google Scholar
    • Export Citation

Fig. 1.
Fig. 1.

1998–2006 mean JAS surface winds (m s−1) from NNRP and precipitation (mm day−1) from TRMM (3B42V6).

Citation: Journal of Climate 22, 10; 10.1175/2008JCLI2466.1

Fig. 2.
Fig. 2.

JAS surface winds (m s−1) and precipitation (mm day−1) from CRCM.

Citation: Journal of Climate 22, 10; 10.1175/2008JCLI2466.1

Fig. 3.
Fig. 3.

Daily precipitation (mm day−1) from (a) the CRCM simulation, (b) TRMM (3B42V6, 1998–2006 mean), (c) GPCP (1998–2004 mean), and (d) FEWS (2000–06 mean). Averaged between 10°E and 10°W and all are on 1° × 1° grid.

Citation: Journal of Climate 22, 10; 10.1175/2008JCLI2466.1

Fig. 4.
Fig. 4.

(a) Annual mean SST (K) from CRCM and its deviation from NNRP (shaded) and area-average SST cycle over (b) NETA and (c) GOG drawn in (a).

Citation: Journal of Climate 22, 10; 10.1175/2008JCLI2466.1

Fig. 5.
Fig. 5.

Area-average SST cycle over (a) NETA and (b) GOG from the test run.

Citation: Journal of Climate 22, 10; 10.1175/2008JCLI2466.1

Fig. 6.
Fig. 6.

(a) The energy budget terms and (b) their deviation from their respective annual averages (W m−2) over NETA.

Citation: Journal of Climate 22, 10; 10.1175/2008JCLI2466.1

Fig. 7.
Fig. 7.

Surface stress (N m−2) from (a) CRCM, (b) QuikSCAT, and (c) GSSTF2 area averaged over NETA.

Citation: Journal of Climate 22, 10; 10.1175/2008JCLI2466.1

Fig. 8.
Fig. 8.

(a) Surface meridional wind speed (contours, m s−1) and surface pressure (shaded, hPa) from (a) CRCM and (b) NNRP. Both averaged over NETA (10°E to 10°W). Distance between shades is 2 hPa.

Citation: Journal of Climate 22, 10; 10.1175/2008JCLI2466.1

Fig. 9.
Fig. 9.

(a) Surface winds (m s−1) wind speed (contours) and (b) wind-driven currents (m s−1) and entrainment cooling (shaded in W m−2) in August.

Citation: Journal of Climate 22, 10; 10.1175/2008JCLI2466.1

Fig. 10.
Fig. 10.

Mean (a) April and (b) July nondivergent component of surface wind (m s−1) and streamfunction.

Citation: Journal of Climate 22, 10; 10.1175/2008JCLI2466.1

Fig. 11.
Fig. 11.

Zonal wind (m s−1) and mixing ratio (g kg−1) at 20°W in (a) April and (b) July.

Citation: Journal of Climate 22, 10; 10.1175/2008JCLI2466.1

Fig. 12.
Fig. 12.

Surface stress (N m−2) from (a) CRCM, (b) QuikSCAT, and (c) GSSTF2 area averaged over GOG.

Citation: Journal of Climate 22, 10; 10.1175/2008JCLI2466.1

Fig. 13.
Fig. 13.

(a) Surface winds (m s−1) wind speed (contours) and (b) wind-driven currents (m s−1) and entrainment cooling (shaded in W m−2) in May.

Citation: Journal of Climate 22, 10; 10.1175/2008JCLI2466.1

Fig. 14.
Fig. 14.

(a) Surface meridional wind speed (contours; m s−1) and surface pressure (shaded; hPa) from (a) CRCM and (b) NNRP. Both averaged over GOG (10°E–10°W). The spacing between shades is 2 hPa.

Citation: Journal of Climate 22, 10; 10.1175/2008JCLI2466.1

Save
  • Bryan, J. P., and L. J. Lewis, 1979: A water mass model of the world ocean. J. Geophys. Res., 84 , 25032517.

  • Carton, J. A., and Z. X. Zhou, 1997: Annual cycle of sea surface temperature in the tropical Atlantic Ocean. J. Geophys. Res., 102 , 2781327824.

    • Search Google Scholar
    • Export Citation
  • Chang, C. Y., J. A. Carton, S. A. Grodsky, and S. Nigam, 2007: Seasonal climate of the tropical Atlantic sector in the NCAR Community Climate System Model 3: Error structure and probable causes of errors. J. Climate, 20 , 10531070.

    • Search Google Scholar
    • Export Citation
  • Chen, F. K., and Coauthors, 1996: Modeling of land surface evaporation by four schemes and comparison with FIFE observations. J. Geophys. Res., 101 , 72517268.

    • Search Google Scholar
    • Export Citation
  • Chou, S-H., E. Nelkin, J. Ardizzone, R. M. Atlas, and C-L. Shie, 2003: Surface turbulent heat and momentum fluxes over global oceans based on the Goddard Satellite Retrievals, Version 2 (GSSTF2). J. Climate, 16 , 32563273.

    • Search Google Scholar
    • Export Citation
  • Cook, K. H., 1999: Generation of the African easterly jet and its role in determining West African precipitation. J. Climate, 12 , 11651184.

    • Search Google Scholar
    • Export Citation
  • Cook, K. H., and E. K. Vizy, 2006: Coupled model simulations of the West African monsoon system: Twentieth- and twenty-first-century simulations. J. Climate, 19 , 36813703.

    • Search Google Scholar
    • Export Citation
  • Davey, M., and Coauthors, 2002: STOIC: A study of coupled model climatology and variability in tropical ocean regions. Climate Dyn., 18 , 403420.

    • Search Google Scholar
    • Export Citation
  • Deser, C., A. Capotondi, R. Saravanan, and A. Phillips, 2006: Tropical Pacific and Atlantic climate variability in CCSM3. J. Climate, 19 , 24512481.

    • Search Google Scholar
    • Export Citation
  • DeWitt, D. G., 2005: Diagnosis of the tropical Atlantic near-equatorial SST bias in a directly coupled atmosphere–ocean general circulation model. Geophys. Res. Lett., 32 , L01703. doi:10.1029/2004GL021707.

    • Search Google Scholar
    • Export Citation
  • Douville, H., F. Chauvin, and H. Broqua, 2001: Influence of soil moisture on the Asian and African monsoons. Part I: Mean monsoon and daily precipitation. J. Climate, 14 , 23812403.

    • Search Google Scholar
    • Export Citation
  • Foltz, G., S. A. Grodsky, J. A. Carton, and M. McPhaden, 2003: Seasonal mixed layer heat budget of the tropical Atlantic. J. Geophys. Res., 108 , 3146. doi:10.1029/2002JC001584.

    • Search Google Scholar
    • Export Citation
  • Grell, G. A., J. Dudhia, and D. R. Stauffer, 1994: A description of the fifth-generation Penn State/NCAR mesoscale model (MM5). NCAR Tech. Note NCAR/TN-398+STR, 122 pp.

    • Search Google Scholar
    • Export Citation
  • Grist, J. P., and S. E. Nicholson, 2001: A study of the dynamic factors influencing the rainfall variability in the West African Sahel. J. Climate, 14 , 13371359.

    • Search Google Scholar
    • Export Citation
  • Grodsky, S. A., J. A. Carton, and S. Nigam, 2003: Near surface westerly wind jet in the Atlantic ITCZ. Geophys. Res. Lett., 30 , 2009. doi:10.1029/2003GL017867.

    • Search Google Scholar
    • Export Citation
  • Hagos, S. M., and K. H. Cook, 2007: Dynamics of the West African monsoon jump. J. Climate, 20 , 52645284.

  • Herman, A., V. B. Kumar, P. A. Arkin, and J. V. Kousky, 1997: Objectively determined 10-Day African rainfall estimates created for famine early warming systems. Int. J. Remote Sens., 18 , 21472159.

    • Search Google Scholar
    • Export Citation
  • Hsieh, J. S., and K. H. Cook, 2005: Generation of African easterly wave disturbances: Relationship to the African easterly jet. Mon. Wea. Rev., 133 , 13111327.

    • Search Google Scholar
    • Export Citation
  • Huffman, G. J., and Coauthors, 1997: The Global Precipitation Climatology Project (GPCP) combined precipitation dataset. Bull. Amer. Meteor. Soc., 78 , 520.

    • Search Google Scholar
    • Export Citation
  • Huffman, G. J., R. F. Adler, M. Morrissey, D. T. Bolvin, S. Curtis, R. Joyce, B. McGavock, and J. Susskind, 2001: Global precipitation at one-degree daily resolution from multisatellite observations. J. Hydrometeor., 2 , 3650.

    • Search Google Scholar
    • Export Citation
  • Joseph, J. H., W. J. Wiscombe, and J. A. Weinman, 1976: The delta-Eddington approximation for radiative flux transfer. J. Atmos. Sci., 33 , 24522459.

    • Search Google Scholar
    • Export Citation
  • Kalnay, E., and Coauthors, 1996: The NCEP/NCAR 40-Year Reanalysis Project. Bull. Amer. Meteor. Soc., 77 , 437471.

  • Kara, A. B., P. A. Rochford, and H. E. Hurlburt, 2003: Mixed layer depth variability over the global ocean. J. Geophys. Res., 108 , 3079. doi:10.1029/2000JC000736.

    • Search Google Scholar
    • Export Citation
  • Kiehl, J. T., and B. P. Briegleb, 1991: A new parameterization of the absorptance due to the 15-micron band system of carbon dioxide. J. Geophys. Res., 96 , 90139019.

    • Search Google Scholar
    • Export Citation
  • Klinger, B. A., B. Huang, B. Kirtman, P. Schopf, and J. Wang, 2006: Monthly climatologies of ocean friction velocity cubed. J. Climate, 19 , 57005708.

    • Search Google Scholar
    • Export Citation
  • Kuo, H. L., 1974: Further studies of the influence of cumulus convection on large-scale flow. J. Atmos. Sci., 31 , 12321240.

  • Li, T., and S. G. H. Philander, 1996: On the annual cycle of the eastern equatorial Pacific. J. Climate, 9 , 29862998.

  • Li, T., and S. G. H. Philander, 1997: On the seasonal cycle of the equatorial Atlantic Ocean. J. Climate, 10 , 813817.

  • Mitchell, T., and J. M. Wallace, 1992: The annual cycle in equatorial convection and sea surface temperature. J. Climate, 5 , 11401156.

    • Search Google Scholar
    • Export Citation
  • Navarra, A., 1999: Beyond El Niño Decadal and Interdecadal Climate Variability. Springer, 374 pp.

  • Okumura, Y., and S-P. Xie, 2004: Interaction of the Atlantic equatorial cold tongue and African monsoon. J. Climate, 17 , 35893602.

  • Patricola, C. M., and K. H. Cook, 2007: Dynamics of the West African monsoon under mid-Holocene precessional forcing: Regional climate model simulations. J. Climate, 20 , 694716.

    • Search Google Scholar
    • Export Citation
  • Philander, S. G. H., and R. C. Pacanowski, 1981: The oceanic response to cross equatorial winds (with application to coastal upwelling in low latitudes). Tellus, 33 , 201210.

    • Search Google Scholar
    • Export Citation
  • Reynolds, R. W., and T. M. Smith, 1995: A high-resolution global sea surface temperature climatology. J. Climate, 8 , 15711583.

  • Richter, I., and S. Xie, 2008: On the origin of equatorial Atlantic biases in coupled general circulation models. Climate Dyn., 31 , 587598.

    • Search Google Scholar
    • Export Citation
  • Sterl, A., and W. Hazeleger, 2003: Coupled variability and air–sea interaction in the South Atlantic Ocean. Climate Dyn., 21 , 559571.

    • Search Google Scholar
    • Export Citation
  • Troen, I., and L. Mahrt, 1986: A simple model of the atmospheric boundary layer: Sensitivity to surface evaporation. Bound.-Layer Meteor., 37 , 129148.

    • Search Google Scholar
    • Export Citation
  • Vizy, E. K., and K. H. Cook, 2002: Development and application of a mesoscale climate model for the tropics: Influence of sea surface temperature anomalies on the West African monsoon. J. Geophys. Res., 107 , 4023. doi:10.1029/2001JD000686.

    • Search Google Scholar
    • Export Citation
  • Xie, S-P., and J. A. Carton, 2004: Tropical Atlantic variability: Patterns, mechanisms, and impacts. Ocean–Atmosphere Interaction and Climate Variability, Geophys. Monogr., Vol. 147, Amer. Geophys. Union, 121–142.

    • Search Google Scholar
    • Export Citation
  • Zebiak, S. E., and M. A. Cane, 1987: A model El Niño–Southern Oscillation. Mon. Wea. Rev., 115 , 22622278.

  • Zeng, N., J. D. Neelin, K-M. Lau, and C. J. Tucker, 1999: Enhancement of interdecadal climate variability in the Sahel by vegetation interaction. Science, 286 , 15371540.

    • Search Google Scholar
    • Export Citation
  • Fig. 1.

    1998–2006 mean JAS surface winds (m s−1) from NNRP and precipitation (mm day−1) from TRMM (3B42V6).

  • Fig. 2.

    JAS surface winds (m s−1) and precipitation (mm day−1) from CRCM.

  • Fig. 3.

    Daily precipitation (mm day−1) from (a) the CRCM simulation, (b) TRMM (3B42V6, 1998–2006 mean), (c) GPCP (1998–2004 mean), and (d) FEWS (2000–06 mean). Averaged between 10°E and 10°W and all are on 1° × 1° grid.

  • Fig. 4.

    (a) Annual mean SST (K) from CRCM and its deviation from NNRP (shaded) and area-average SST cycle over (b) NETA and (c) GOG drawn in (a).

  • Fig. 5.

    Area-average SST cycle over (a) NETA and (b) GOG from the test run.

  • Fig. 6.

    (a) The energy budget terms and (b) their deviation from their respective annual averages (W m−2) over NETA.

  • Fig. 7.

    Surface stress (N m−2) from (a) CRCM, (b) QuikSCAT, and (c) GSSTF2 area averaged over NETA.

  • Fig. 8.

    (a) Surface meridional wind speed (contours, m s−1) and surface pressure (shaded, hPa) from (a) CRCM and (b) NNRP. Both averaged over NETA (10°E to 10°W). Distance between shades is 2 hPa.

  • Fig. 9.

    (a) Surface winds (m s−1) wind speed (contours) and (b) wind-driven currents (m s−1) and entrainment cooling (shaded in W m−2) in August.

  • Fig. 10.

    Mean (a) April and (b) July nondivergent component of surface wind (m s−1) and streamfunction.

  • Fig. 11.

    Zonal wind (m s−1) and mixing ratio (g kg−1) at 20°W in (a) April and (b) July.

  • Fig. 12.

    Surface stress (N m−2) from (a) CRCM, (b) QuikSCAT, and (c) GSSTF2 area averaged over GOG.

  • Fig. 13.

    (a) Surface winds (m s−1) wind speed (contours) and (b) wind-driven currents (m s−1) and entrainment cooling (shaded in W m−2) in May.

  • Fig. 14.

    (a) Surface meridional wind speed (contours; m s−1) and surface pressure (shaded; hPa) from (a) CRCM and (b) NNRP. Both averaged over GOG (10°E–10°W). The spacing between shades is 2 hPa.

All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 275 91 12
PDF Downloads 117 29 4