The origin of along-shelf pressure gradient in the Middle Atlantic Bight

May 13, 2010

Long-term mean depth-averaged shelf circulation on the Middle Atlantic Bight (MAB) shelf is mainly alongshelf and southwestward. Noticeably, the circulation has a seasonal and inter-annual variability, southwestward in winter and spring, and weaker or even reversed in summer and fall. Momentum balance considerations suggest that forcing for these seasonal fluctuations of shelf currents is the along-shelf pressure gradient (ASPG) whose origin, however, remains a subject for debate. The large-scale wind pattern over the North Atlantic, Gulf Stream path shift, the activity of Gulf Stream warm-core rings, Labrador Sea transport, and river discharge may contribute to the ASPG setup. Sixteen years (1993-2008) of satellite data from AVISO and data-assimilated model reanalysis, as well as tide-gauge sea level data are analyzed. It is shown that the 16-year mean ASPG is primarily produced by the total river discharge along the east coast of North American and Labrador Sea transport. The southwestward propagation of planetary and topographic Rossby waves and warm-core rings mainly contribute to the seasonal/inter-annual variations in ASPG. The warm-core rings affect ASPG by changing sea surface height near shelf break. To explore how eddies influence shelf water, we conducted an idealized numerical study about warm-core rings. The model results show that the warm-core rings increase sea surface height in the north when they move towards the middle of the MAB, and they produce sea surface sloping down to the north when they move further south. The influence of large-scale wind pattern on ASPG is also examined. The change of wind can not directly affect the ASPG, but the wind is important for the activity of warm-core rings.

1. 
   Introduction
   

The shelf circulation in the MAB has been investigated through observation and modeling studies over several decades (e.g. Beardsley and Boicourt 1981; Chapman 1986; Linder and Gawarkievicz 1998; Flagg et al. 2006; Lentz 2008a). The depth-averaged mean currents over the MAB shelf flow along-isobath were observed at 0.03-0.1 m s^-1 (Lentz 2008a). The mean currents are westward on the New England shelf, and southwestward in the middle of MAB to follow isobath. South of the Chesapeake Bay, the mean currents are offshore towards the open ocean. The observations also exhibit the along-shelf mean currents increase with distance offshore (Beardsley et al. 1976).

The primary driving force for the depth-averaged mean southwestward currents in the MAB is an along-shelf pressure gradient (ASPG) (Beardsley and Boicourt 1981). Stommel and Leetmaa (1972) did the first theoretical modeling study for a steady mean winter circulation. They proposed an ASPG/sea surface slope of order of 10^-7 is required to drive the flow southwestward. Csanady (1976) used a similar dynamic model and also concluded that an ASPG/sea surface slope must exist to account for the observed circulation within the MAB. Lentz (2008a) modified Csanady’s model and used a constant sea surface slope, 3.710^-8, to produce the depth-averaged along-shelf flow southwestward. His study reinforced the key role of the ASPG on the southwestward along-shelf flow.

Then, the question is asked ‘what mechanism produces the mean ASPG over the shelf’. The depth-averaged along-shelf density gradient is about zero estimated from the density profiles in the historical hydrographic database (Lentz 2008a). The sea surface slope creates the ASPG. The deep-water cyclonic gyre found between the continental shelf and the Gulf Stream may drive an ASPG at the shelf break. However, the penetration of the large-scale circulation to the shelf is unclear (Wang 1982; Chapman 1986). The role of large-scale circulation to the ASPG needs further investigation.

Seasonal variations in the depth-averaged along-shelf currents were also noticed. The seasonal variations differ in different sub-regions of MAB (Lentz 2008b). Over the southern flank of Georges Bank, a seasonal variation in along-shelf flow is found with maximum southwestward flow in September (Butman and Beardsley 1987; Brink et al. 2003; Flagg and Dunn 2003; Shearman and Lentz 2003). A well-defined seasonal along-shelf flow in the mid and outer shelf, however, was not observed in several previous studies (Mayer et al. 1979; Beardsley et al. 1985; Aikman et al. 1988). Along Oleander line, Flagg et al. (2006) observed a shelfbreak jet offshore of 100m-isobath was stronger southwestward in fall and winter and weaker in spring and summer. In the inner shelf near the mouth of Long Island Sound, Ullman and Codiga (2004) and Codiga (2005) found the along-shelf flow reached its maximum in summer but at its minimum in winter. The currents measured by ADCP at the location of station 5 on the Coastal Ocean Bio-optical Buoy (COBY) transect (75.029W, 37.833N) shows maximum southwestward in spring and weak in summer and fall. From analysis of 27 long current records with many of them concentrated in the New England Shelf, Lentz (2008b) found that the seasonal variation in alongshore currents has amplitude of a few centimeters per second with maximum southwestward flow in spring onshore of the 60m isobath for the residual alongshore flow, after removing the wind-driven component. He suggested the seasonality of along shelf currents are primarily caused by the cross-shelf density gradient induced by freshwater discharge. The role of ASPG in seasonality of along-shelf currents remains unclear. To improve the understanding of the dynamics of shelf circulation and ASPG, a high resolution, long-term realistic simulation in MAB is required.

In this study, we carried out a hindcast experiment for 16-year (1993-2008) to approach this problem. We will examine the mean, seasonal, and inter-annual variability of along-shelf flow and ASPG. The analysis of tide-gauge sea level and the simulation results indicate a significant seasonal variation in the ASPG. We investigate the driving mechanisms for the along-shelf currents and ASPG, including the wind stress curl over the North Atlantic, Gulf Stream latitude shift, the activity of Gulf Stream warm-core rings, Labrador Sea transport, and river discharge.

The manuscript is organized as follows: Section 2 presents a description of the observation datasets used in the study. Sections 3 describe the numerical experiment. In section 4, we investigate the mean, seasonal, and inter-annual variations in the along-shelf circulation and ASPG in the MAB. The influence of wind, river, Gulf Stream and Labrador Sea transport are discussed in section 5. We conclude in section 6.

2. 
   Data
   

The gridded sea surface heights above geoids and their corresponding absolute geostrophic velocities for the period 1993 to 2008 were produced by Ssalto/Duacs and distributed by Aviso, with support from Cnes (http://www.aviso.oceanobs.com/duacs/). This dataset has a temporal resolution of 7 days and spatial resolution of . A detail description of the dataset is in Le Traon et al. (1998).

The Cross-Calibrated, Multi-Platform ocean surface wind velocity data (CCMP) used is assimilated data products including satellite surface winds from Seawinds on QuikSCAT, Seawinds on ADEOS-II, AMSR-E, TRMM TMI and SSM/I, measurements of ships and buoys, and 10-m winds from the ERA-40 Re-analysis. The CCMP is distributed by the Physical Oceanography Distributed Active Archive Center of the Jet Propulsion Laboratory (JPL), Pasadena, California. The wind dataset contains 6-hourly gridded variantional analysis with horizontal resolution. The winds are referenced to a height of 10 meters above sea surface.

2. 
   The Numerical Model
   

Our interest is on the shelf and shelf-edge transports (across the 200 m isobath). Therefore, the Gulf Stream and eddies are assimilated in deep ocean regions only (water depth H > 1000 m) using satellite sea-surface height anomaly data from AVISO (www.aviso.oceanobs.com). We use the Princeton Regional Ocean Forecast System (PROFS; http://www.aos.princeton.edu/WWWPUBLIC/PROFS/) to hindcast the ocean state. The model integration and analysis are for 19931999. PROFS has been extensively tested against observations and also used for process studies in the Gulf of Mexico [Oey et al. 2005a and b, where a list of recent publications is given]. The Mellor and Ezer’s [1991; see also Ezer and Mellor, 1994] scheme is used to assimilate the AVISO data. In this scheme, the SSH anomaly is projected into the subsurface density field using correlation functions pre-computed from the model’s eddy statistics derived from a non-assimilated 15-year model run. The method is simple, yet it yields fairly accurate upper-layer structures (z = 0 to approximately 800 m) of mesoscale currents and eddies [Oey et al. 2005a; Lin et al. 2006; Yin and Oey, 2007]. No assimilation is done in deep layers for z < 800 m and as mentioned above in regions where the topography is shallower than 1000 m. In these regions, the simulated currents rely entirely on the model’s dynamics.

The Mellor and Yamada’s (1982) turbulence closure scheme modified by Craig and Banner (1994) to effect wave-enhanced turbulence near the surface is used. To account for mixing in stable stratification (e.g., internal waves; MacKinnon and Gregg, 2003), Mellor’s (2001) modification of a Ridchardson-number-dependent dissipation is introduced. A fourth-order scheme is used to evaluate the pressure-gradient terms [Berntsen and Oey, 2010] and, in combination with high resolution and subtraction of the mean -profile, guarantees small truncation errors of O(mm/s) [c.f. Oey et al. 2003].

Initial and Boundary Conditions, and Forcing:

The NWAOM is first run for 15 years, forced by monthly climatological NCEP surface fluxes. The World Ocean Atlas data (“Levitus” data) from NODC [http://www.nodc.noaa.gov/OC5/WOA05/pr_woa05.html] was used for initial condition as well as boundary condition along the eastern open boundary at 55^oW. The transport across 55^oW is also specified [Oey et al. 2003]. This 15-year run establishes a statistically equilibrium ocean field, as verified by examining the domain-averaged kinetic energy and eddy potential energy time series (not shown). This run is then continued by applying NCEP six-hourly reanalysis winds from Jan/01/1992 through 1999, during which the SODA inputs are also used along 55^oW. Surface heat and evaporative fluxes are relaxed to monthly climatological values with a time scale of 100 days. A combination of flow-relaxation and radiation conditions described in Oey and Chen [1992a,b] are used to specify the SODA variables and the M2-tidal forcing from Oregon State University [http://www.oce.orst.edu/research/po/research/tide/index.html;]. The nested NWAOM field is then used to initialize the finest MAOR nested grid, in which similar open-boundary conditions are also applied.

To calculate wind stresses, we use a bulk formula with a high wind-speed limited drag coefficient that curve-fits data for low-to-moderate winds (Large and Pond, 1981) and data for high wind speeds (Powell et al. 2003):

C_d 10³ = 1.2, |u_a|  11 m s^-1;

= 0.49 + 0.065 |u_a|, 11 < |u_a|  19 m s^-1;

= 1.364 + 0.0234 |u_a|  0.00023158 |u_a| ², 19 < |u_a|  100 m s^-1

(2)

where |u_a| is the wind speed.¹ According to this formula, C_d is constant at low winds, is linearly increasing for moderate winds, reaches a broad maximum for hurricane-force winds, |u_a|  30~50 m s^-1, and then decreases slightly for extreme winds. It is necessary to use a C_d formula that accounts for high winds since the study period (1993-1999) includes a few hurricanes. Donelan et al. (2004) suggest that the C_d-leveling at high wind may be caused by flow separation from steep waves. Moon et al. (2004) found that C_d decreases for younger waves that predominate in hurricane-forced wave fields. Bye and Jenkins (2006) attribute the broad C_d-maximum to the effect of spray, which flattens the sea surface by transferring energy to longer wavelengths.

2. 
   Results
   

Figure 2a shows the 16-year (1993-2008) mean of sea surface height (SSH) from the NWAOM simulation. A fine contour interval of 0.005 m is used to show the SSH variation over the shelf and slope. A cyclonic gyre is seen in the Gulf of Maine. The cyclonic flow branches eastward and south-southwestward off Cape Cod. The eastward outer branch flows anticyclonically over the Georges Bank and merges with the inner, weaker branch over the outer shelf southeast of Cape Cod. Figure 2a shows three local high pressure cells in this vicinity, one over the Georges Bank, one south of the Bank over the 1000 m isobath, and a weaker one directly south of Cape Cod. It is clear that here is the region where the sea level begins to slope downward along the entire length of the MAB shelf from Cape Cod to Cape Hatteras. The SSH-contours are cross-shelf for water depths shallower than about 100 m, but over the shelf break and slope they are aligned along the isobaths.

Figure 2b plots the SSH along the 50 m isobath (at the “dots” of Fig.2a). This again shows the generally downward sea-level tilt from the northern station off Cape Cod (x = 730 km) to the southern station off Cape Hatteras (x = 0); the corresponding ASPG  3~5×10^-8 which is in excellent agreement with Lentz’s (2008) estimate of 3.7×10^-8 based on long-term observations. Figure 2b shows that the ASPG is larger  8.4×10^-8 between the east end of Long Island (ELS) and Delaware (DEL), where a linear sea-level slope is seen. This stronger ASPG seems to be consistent with Scott and Csanady’s (1976) estimate  1.44×10^-7 off Long Island based on a 25-day time series in September 1975. However, Figure 2c shows that there are strong seasonal as well as inter-annual fluctuations. The standard deviation is 8.6×10^-8 and maximum and minimum peaks are 2×10^-7, larger in some years (2003~2007) and smaller in others (1997~1999). There is a clear seasonal pattern of maximum (positive) ASPG in winter and minimum in summer; the exceptions are 1993, 1997 and 1998.

Observed Sea Surface Height

To corroborate the above model results, we use an independent dataset, coastal tide gauges, to estimate the ASPG. Sea surface slope variation is analyzed by calculating the EOF of the 16-year sea level data at tide gauge stations shown in Fig.1. Only those data that overlap in time at the tide-gauge stations shown in Fig.1 (excluding Bermuda and Wilmington NC) are used in the EOF analysis. To calculate the ASPG, we multiplied the EOF mode 1 with the standard deviation of principal component 1 (PC1) and normalized the PC 1 with its standard deviation. Figure 3a shows the EOF mode 1 which explains 67% of the total variance. From south to north, the sea level remains nearly constant, about 0.08 m, from Duck Pier to Atlantic City, and then gradually increases to about 0.02 m at Halifax. Figure 3b shows the corresponding principal component. The PC1 is generally positive and maximum in winter (Dec~Feb) and negative and minimum in summer through fall. Thus, sea level in winter (summer~fall) increases (decreases) northward consistent with the modeled seasonal fluctuations shown in Figure 2. After linear regression of EOF mode 1 (Fig. 3a), a sea-level slope of +4.810^-8 is estimated from Halifax to Duck Pier (Fig. 3c). So, the sea level variations of mode 1 are in the range of 110^-7. This is smaller than but remarkably consistent with the model-predicted range of 210^-7 shown in Fig.2c. There are also inter-annual variations (Figure 3b), most notably in 1993~1996 when the PC1 was mostly positive, meaning that sea-level tends to slope more steeply northward, and in 1996~1999 when the reverse occurred.

The depth-averaged mean along-shelf current varies with location (Figure 4a). Over the southern flank of Georges Bank and the Nantucket Shoals, the mean along-shelf currents flow westward. On the shelf between Hudson Shelf Valley and Long Island, the mean along-shelf is weak (<0.005 m s^-1). South of Hudson Shelf Valley, the along-shelf mean flow becomes large (>0.01 m s^-1). At the location just north of Cape Hatteras, the mean flow turns to flow eastward in the south off Chesapeake Bay. This spatial variation in mean flow is generally consistent with observations (Lentz, 2008).

Time series of 3-month-mean depth-averaged along-shelf current averaged along the 50 m isobath is shown in Figure 4c (the cross-shelf currents are very weak and are not shown). The along-shelf current fluctuates from about -0.06 m s^-1 in winter-spring to about 0.04 m s^-1 in summer-fall. The along-shelf mean value is 0.015 m s^-1 which is approximately 2-3 times weaker than Lentz’s value. Part of the discrepancy is due to the spatial averaging (i.e. along 50 m isobath); also, a higher-resolution experiment gives a larger magnitude comparable to that observed (Oey et al. 2010). The seasonal fluctuations of the along-shelf current are correlated with the ASPG from model (Figure 4b). The correlations are about -0.50, above the 95% significance level, 0.20.

2. 
   Discussion
   

_bx = ^ox  gH/x, (1)

where  is surface elevation, H is water depth, _bx and ^ox are the x-component bottom and wind stresses respectively. For the mean wind stress in MAB, Lentz (2008) shows that the RHS of (1) is negative (ASPG overcomes wind stress, both are positive), so that if _bx is parameterized as ru where from the model r  3×10^-4 m s^-1 is the linear bottom friction coefficient, then u is also negative (i.e. equatorward).

Many mechanisms may affect the sea surface slope titling in MAB, including the Gulf Stream latitude shift, the wind stress over the North Atlantic, the southwestward propagating (planetary and topographic) Rossby waves and Gulf Stream warm-core rings, Labrador Sea transport, and river discharge. The NWAOM simulation results described in section 4 show a clear mean ASPG sloping downward to the south. The value, 8.4×10^-8 is consistent with estimated ASPG from observations by Lentz (2008a).We examine the dominant mechanism for the mean ASPG setup first.

To explore the dominant mechanism accounting for the mean ASPG setup, we conducted three sensitivity tests, including simulation with river only, simulation with river and upstream transport, and simulation with wind only. For the simulation with river only, 16-year mean sea surface height shows slightly increase to the north (Figure 5b). The estimated mean ASPG is about 2.1×10^-8, smaller than the ASPG from the standard run (Figure 5a). A reduced transport (1/3 of transport of the NWAOM run) is added to the east boundary of the river-only run. A doubled mean ASPG, 4.3×10^-8, is obtained (Figure 5c), compared to the river only run. On the contrary, a simulation with wind only produced a negative mean ASPG, -7.2×10^-8 (Figure 5d). These results clearly indicate the mean ASPG setup is mainly caused by the river discharge and upstream transport. And the large-scale winds can not build a mean ASPG sloping down to the south.

Next, the contributions of aforementioned five mechanisms to the seasonal and inter-annual variations in ASPG are discussed.

It has been studied for many years that the seasonal Gulf Stream mean path shifts south of the slope sea contributes to the seasonal variations in currents near shelf break and slope sea(e.g. Bane 1988; Dong and Kelly, 2003). The EOF analysis of surface velocity and SST anomalies in the slope sea by Molino and Joyce (2008) also shows the Gulf Stream path shifts influence southwestward slope currents both on seasonal scale and inter-annual scale. In general, when the Gulf Stream moves away from the slope in winter-spring, the slope currents are strengthened southwestward, but are weak even reversed in summer-fall when the Gulf Stream is close to the slope. However, the influence of the Gulf Stream mean path shifts on the shelf circulation and ASPG has not been fully studied. The Gulf Stream mean path shift is hard to directly affect the shelf water, except near Cape Hatteras, where the Gulf Stream may contribute to the slightly sloping down northward of sea level north of Cape Hatteras (Figs. 2b and 3a). The 3-month running average, zonally averaged Gulf Stream position anomalies (relative to the 16 year mean position, 1993-2008) from AVISO satellite SSH dataset show that the Gulf Stream moved south in winter-spring, but north in summer-fall by about 0.6⁰ latitude (Figure 6b). The maximum lag correlation between the ASPG (Figure 6a) and the Gulf Stream path shift is about 0.57 with 4-month lag (Table 1). However, there is no significant correlation for the one-year average GS path shifts and APSG. This result suggests that the Gulf Stream may influence the ASPG seasonal variations with 4-month delay. In other words, the northward movement of the Gulf Stream corresponds to an ASPG sloping down to the south after about 4 months. The reason for the delay and ‘remotely’ forced by the Gulf Stream may be caused by the activity of warm-core rings, discussed later in section Warm-core rings.

The 16-year 3-monthly mean wind stress curl is estimated over the Northwest Atlantic from 60W towards the 200m isobath and from 35N to 42N. As expected, the wind stress curl shows a significant seasonal cycle, strong positive in winter and weak negative in summer (Fig. 6c). Its correlation with ASPG is low and not significant. This implies the large scale wind pattern can not directly influence the variability of ASPG. Noticeably, the wind curl significantly correlates with GS path shifts and eddy kinetic energy (EKE) north of the Gulf Stream mean path with a few months lag (Table 1). The wind stress curl also correlates well with the upstream transport.

The warm-core ring activities are evaluated though the estimates of EKE north of the 16-year Gulf Stream mean path from the AVISO geostrophic current anomaly (hereafter, N-EKE). The seasonal evolution of EKE is averaged over the region from 75W to 55W and from the north of the Gulf Stream mean path to 42.5N (Figure 7a). The N-EKE is high in summer and fall but low in winter and spring (Figure 7c). The maximum EKE in October is induced by the movement of the Gulf Stream since Gulf Stream is in its northernmost location at this time. So, the maximum N-EKE/activity of warm-core rings occurs in summer, which is consistent with findings by Zhai et al. (2008). This implies the warm-core rings are more active in summer than in winter.

The maximum correlation for 3-month running average of ASPG and N-EKE is about 0.41 with 5-month lag (Figure 6, Table 1). This implies the seasonal variations in along-shelf sea surface slope correspond to the southwestward propagation of warm-core rings. Also, the correlation for one-year low pass filtered N-EKE and ASPG is relatively high, 0.52, meaning the warm-core rings can influence ASPG in inter-annual time scale. To examine how the southwestward propagation of warm-core rings affects shelf water in MAB, an idealized study about warm-core ring propagation was conducted. The simulation has the same domain as NWAOM, but external forcings are removed, e.g. winds, river, data assimilation, and transport at boundaries. Three warm core rings with diameter 125 km were injected over the open ocean in the northwest of model domain every 360 days. The simulation was run for 6 years.

Figure 8a shows the mean sea surface height (SSH) averaged over 6 years for the idealized simulation. The SSH is positive over the middle Atlantic Ocean. Large SSH from northeast to southwest indicate the path of the southwestward propagation of the warm-core rings. Maximum SSH is obtained near the continental shelf break north of Cape Hatters, where the warm-core rings tended to be trapped. A mean ASPG sloping down northward (-1.5×10^-8) is estimated using linear regression (Figure 8b), the same as the estimate of ASPG in the NWAOM case. The variations in ASPG with time are shown in Figure 8c. The red lines represent the times when new warm-core rings were injected into the model. The seasonal and inter-annual variations in ASPG are manifest. These variations are related to the location of the warm-core ring.

When a ring approaches the shelf of middle MAB, the sea surface height is increasing over the shelf break region, nearly parallel to the isobaths (e.g. Figure 9c and 9d). This creates a cross-slope pressure gradient towards the shelf, and therefore the northeastward geostrophic slope current anomalies. The ASPG is slightly positive, corresponding to the positive variation in ASPG at day 1610 (Figure 8c). The shelf circulation is close to zero. When the ring moves further south, the sea surface height over the shelf tends to slope down to the north. Figure 9a and 9b show an example of SSH and surface currents at day 1200, when the ASPG has a large negative variation (Figure 8c). North of the ring, the shelf water is advected to the open ocean, and south of the ring, the open ocean water is brought into the shelf region. This effect causes the along-shelf currents to be northeastward. The sea surface height decreases to the shoreline and to the north (Figure 9b). The variations in the surface slope induced by the warm-core rings are in the order of magnitude of 10^-8, comparable with the observed variability. So the Gulf Stream warm-cores rings that impinge upon the southern portion of the MAB shelf break contribute to the seasonal and inter-annual variations in SSH, and subsequently the pressure gradient.

Noticeably, the Gulf Stream path change correlates with N-EKE from Table 1. The influence of the Gulf Stream path change on the ASPG may partially be induced through the activity of warm-core rings. This is corroborated by the similar time lags of Gulf Stream path change and N-EKE with ASPG (Table 1). The time lag mainly arises from the time spent for the southwestward propagation of warm-core rings.

The Labrador Sea transport is estimated along the south of Nova Scotia towards 1000 m isobath. The transport is 3-monthly averaged from 1993 to 2008 (Figure 6e). Its mean value is about -4.1 Sv and its standard deviation is about 1.9 Sv. In spring, the transport is strong southwestward, and becomes much weaker in fall. After the one-year running average, the transport shows inter-annual variability. For example, from 93 to 96, the transport was weak southwest, but from 1996 to 1997, the transport became stronger. This variation is consistent with Labrador Sea transport variability in Dong and Kelly Fig. 4a (2003). From Table 1, the transport correlates well with wind stress curl (R=-0.65). This indicates the transport variations are mainly produced by wind change, since the total transport at NWAOM east boundary is fixed. The temperature and salinity anomaly over the shelf of MAB is also well correlated with the transport. The colder and fresher shelf water corresponds to the strong southwestward transport. However, the correlations for the seasonal and inter-annual variations in transport and ASPG are low, which indicate the change of transport can not directly influence the seasonal and inter-annual variations in ASPG.

The freshwater discharge from 17 rivers along the east coast from Cape Hatteras to the St. Lawrence system was used in NWAOM. The 3-monthly average of total river input has a clear seasonality, peak in spring due to snow melting and precipitation (Fig. 6f). From table 1, the correlation between river discharge and the ASPG is about 0.19 without time lag (above the 95% significant level, 0.13), indicating the influences of rivers. However, the influence of rivers on ASPG seasonality/inter-annual variability is much less than the N-EKE and GS path change according to the correlations.

2. 
   Summary
   

The study indicates that the total freshwater discharge and upstream transport from Labrador Sea mainly contribute to the mean ASPG setup. The seasonal/inter-annual variations in ASPG are mainly induced by the activity of warm-core rings. To explore how eddies influence shelf water, we conducted an idealized numerical study about warm-core rings. The model results show that the warm-core rings increase sea surface height in the north when they move towards the middle of the MAB, and they produce sea surface sloping down to the north when they move further south. The influence of large-scale wind pattern on ASPG is also examined. The change of wind can not directly affect the ASPG, but the wind is the major mechanism accounting for the GS path change (Dong and Kelly, 2003) and the seasonality of EKE (Zhai et al. 2008).

Aikman, F., III, H.W. OU, and R.W. Houghton, 1988: Current variability across the New England continental shelf-break and slope. Continental Shelf Research, 8, 625-651.

Beardsley, R.C., and W.C. Boicourt, 1981: On estuarine and continental-shelf circulation in the Middle Atlantic Bight. Evolution of Physical Oceanography: Scientific Surveys in Honor of Henry Stommel, B. A. Warren and C. Wunsch, Eds., The MIT press, 198-233.

Beardsley, R.C., D.C. Chapman, K.H. Brink, S.R. Ramp, and R. Schlitz, 1985: The Nantucket Shoals Flux Experiment (NSFE79). Part I: A basic description of the current and temperature variability. Journal of Physical Oceanography, 15, 713-748.

Berntsen, J. and L.-Y. Oey, 2010: Estimation of the internal pressure gradient in σ-coordinate 482 ocean models: comparison of second-, fourth-, and sixth-order schemes. Ocean Dyn. 60, 483 317-330. DOI 10.1007/s10236-009-0245-y.

Brink, K.H., R. Limeburner, and R.C. Beardsley, 2003: Properties of flow and pressure over Georges Bank as observed with near-surface drifters. Journal of Geophysical Research, 108,8001, doi:10.1029/2001JC001019.

Butman, B., and R.C. Beardsley, 1987: Long-term observations on the southern flank of Georges Bank. Part I: A description of the seasonal cycles of currents, temperature, stratification, and wind stress. Journal of Physical Oceanography, 17, 367-384.

Bye, J. A. T., and A. D. Jenkins, 2006: Drag coefficient reduction at very high wind speeds, Journal of Geophysical Research, 111, C03024, doi:10.1029/2005JC003114.

Codiga, D.L., 2005: Interplay of wind forcing and buoyant discharge off Montauk Point: Seasonal changes to velocity structure and a coastal front. Journal of Physical Oceanography, 35, 1068-1085.

Chapman, D.C., 1986: A simple model of the formation and maintenance of the shelf/slope front in the Middle Atlantic Bight. Journal of Physical Oceanography, 16, 1273-1279

Csanady, G.T., 1976: Mean circulation in shallow seas. Journal of Geophysical Research, 81, 5389-5399.

Donelan, M. A., B. K. Haus, N. Reul, W. J. Plant, M. Stiassnie, H. C. Graber, O. B. Brown, and E. S. Saltzman, 2004: On the limiting aerodynamic roughness in very strong winds. Geophys. Res. Letters, 31, L18206, doi:10.1029/2004GL019460.

Dong S., and K.A. Kelly, 2003: Seasonal and interannual variations in geostrophic velocity in the Middle Atlantic Bight. Journal of Geophysical Research, 108(C6), 3172, doi:10.1029/2002JC001357.

Ezer, T., and G.L. Mellor, 1994: Diagnostic and prognostic calculations of the North Atlantic circulation and sea level using a sigma coordinate ocean model. Journal of Geophysical Research, 99(C7), 14 159-14 171.

Flagg, C.N., and M. Dunn, 2003: Characterization of the mean and seasonal flow regime on Georges Bank from shipboard acoustic Doppler current profiler data. Journal of Geophysical Research, 108, 8002, doi:10.1029/2001JC001257.

Flagg, C.N., D.-P. Wang, H.T. Rossby, and R.L. Benway, 2009: A study of the currents of the outer shelf and upper slope from a decade of shipboard ADCP observations in the Middle Atlantic Bight. Journal of Geophysical Research, 111, C06003, doi:10.1029/2005JC003116.

Large, W. G., and S. Pond (1981): Open ocean momentum flux measurements in moderate to strong winds, Journal of Physical Oceanography, 11, 324-336

Lentz, S.J., 2008a: Observations and a model of the mean circulation over the Middle Atlantic Bight continental shelf. Journal of Physical Oceanography, 38, 1203-1221.

Lentz, S.J., 2008b: Seasonal variations in the circulation over the Middle Atlantic Bight continual shelf. Journal of Physical Oceanography, 38, 1486-1500.

Le Traon, P.-Y., F. Nadal, and N. Ducet (1998), An improved mapping method of multisatellite altimeter data, Journal of Atmospheric and Oceanic Technology, 103, 8045-8057.

Lin, X.-H., L.-Y. Oey, and D.-P. Wang, 2007: Altimetry and drifter assimilations of Loop Current and eddies, JGR, 112, C05046, doi:10.1029/2006JC003779, 2007

Linder, C.A., and G. Gawarkiewicz, 1998: A climatology of the shelfbreak front in the Middle Atlantic Bight, Journal of Geophysical Research, 103, 18 405- 18 423

MacKinnon, J.A., Gregg, M.C., 2003. Shear and baroclinic energy flux on the summer New England shelf. Journal of Physics and Oceanography 33, 1462–1475.

Mayer, D.A., D.V. Hansen, and D.A. Ortman, 1979: Long-term current and temperature observation on the Middle Atlantic shelf. Journal of Geophysical Research, 84, 1776-1792.

Mellor, G.L., Yamada, T., 1982. Development of a turbulence closure model for geophysical fluid problems. Review of Geophysics and Space Physics 20, 851–875.

Mellor, G.L., 2001. One-dimensional, ocean surface layer modeling: a problem and a solution. Journal of Physics and Oceanography 31, 790–809.

Moon, I.-J., Ginis, I., Hara, T., 2004. Effect of surface waves on Charnock coefficient under tropical cyclones. Geophysical Research Letters 31, L20302.

Oey, L.-Y., and P. Chen, A model simulation of circulation in the north-east Atlantic shelves and seas, J. Geophys. Res., 97, 20,087–20,115, 1992.

Oey, L.-Y., Ezer, T., Lee, H.J., 2005a. Loop Current, rings and related circulation in the Gulf of Mexico: a review of numerical models and future challenges. In: Sturges, W.,Lugo-Fernandez, A. (Eds.), Circulation in the Gulf of Mexico: Observations and Models, Geophysical Monograph Series, vol. 161, 360pp.

Oey, L.-Y., Ezer, T., Forristall, G., Cooper, C., DiMarco, S., Fan, S., 2005b. An exercise in forecasting loop current and eddy frontal positions in the Gulf of Mexico. Geophysical Research Letters 32, L12611.

Oey, L.-Y., Ezer, T., Wang, D.-P., Fan, S.-J., Yin, X.-Q., 2006. Loop Current warming by Hurricane Wilma. Geophysical Research Letters 33, L08613.

Oey, L.-Y., Lee, H.-C, and Schmitz, W.J. Jr., 2003. Effects of Winds and Caribbean Eddies on the Frequency of Loop Current Eddy Shedding: A Numerical Model Study. J. Geophy. Res., 108(C10), doi:10.1029/2002JC001698

Powell, M. D., P. J. Vickery, and T. A. Reinhold, 2003: Reduced drag coefficient for high wind speeds in tropical cyclones. Nature, 422, 279-283.

Stommel H., and A. Leetmaa, 1972: Circulation on the continental shelf. Proceedings of the National Academy of Sciences, 69, 3380-3384.

Ulman, D.S., and D.L. Codiga, 2004: Seasonal variation of a coastal jet in the Long Island Sound outflow region based on HF radar and Doppler current observations. Journal of Geophysical Research, 109, C07S06, doi:10.1029/2002JC001660.

Wang, D.-P., 1982: Effects of continental slope on the mean shelf circulation, Journal of Physical Oceanography, 12, 1524-1526.

Yin and Oey, 2007: Bred-Ensemble Ocean Forecast of Loop Current and Eddies.  Ocean Modelling, 17, 300-326.

Zhai, X., R.J. Greatbatch, and J.-D. Kohlmann, 2008: On the seasonal variability of eddy kinetic energy in the Gulf Stream region. Geophysical Research Letter, 35, L24609, doi:10.1029/2008GL036412,2008.