Frontal Air-Sea Interaction Experiment Research Articles

A Comparison between Observations and MM5 Simulations of the Marine Atmospheric Boundary Layer across a Temperature Front

Simulations, made with the fifth-generation Pennsylvania State University (PSU)‐National Center for Atmospheric Research (NCAR) Mesoscale Model (MM5), of the response of the marine atmospheric boundary layer (MABL) as air moves over a sharp SST front are compared with observations made during the Frontal Air‐Sea Interaction Experiment (FASINEX) in the North Atlantic subtropical convergence zone. The purpose of undertaking these comparisons was to evaluate the performance of MM5 in the vicinity of an SST front and to determine which of the planetary boundary layer (PBL) parameterizations available best represents MABL processes. FASINEX provides an ideal dataset for this work in that it contains detailed measurements for scenarios at the two extremes: wind blowing from warm to cold water normal to a 28C SST front and the converse, wind blowing from cold to warm water. For the wind blowing from warm to cold water, there is a pronounced modification of the near-surface wind field over the front, in both model results and aircraft observations. The decrease of near-surface wind speed and stress is due to a stable internal boundary layer (IBL) induced by the SST front, restricting exchange of mass and momentum between the surface and upper part of the MABL. For the cold-to-warm case, the relatively strong vertical mixing through the entire MABL over warm water dampens the response of the near-surface winds and surface stress to the SST front. The properties observed by the aircraft are simulated quite well in both cases, suggesting that MM5 captures the appropriate boundary layer physics at the mesoscale or regional scale.

Ekman layers and two‐dimensional frontogenesis in the upper ocean

A two‐dimensional modified semi geostrophic model is used to study the evolution of oceanic fronts in the presence of vertical mixing. The parameterization used for mixing is an elevated constant value of vertical viscosity and friction; hence the Ekman layer acts as a surrogate mixed layer, with, however, vertical shear allowed everywhere. An initial condition representative of the observed fields in Frontal Air‐Sea Interaction Experiment (FASINEX) is used, and its modification in the presence of vertical mixing alone is investigated. Without external forcing an Ekman layer results because of the presence of vertical geostrophic shear at the surface. The maximum density gradient moves toward the dense side of the front driven by this flow. Convergence of Ekman flow results in downward bowing of the isopycnals beneath the surface expression of the front, a feature reminiscent of density sections taken during FASINEX. This feature is not evident when a barotropic convergence field is applied in the absence of mixing or in fully nonlinear simulations of the evolution of fronts with baroclinic instability. An analytic theory suggests that the maximum density gradient will increase over time when vertical mixing alone is present. In the presence of negative uniform wind stress in the direction opposite to the surface geostrophic flow the front moves toward the denser water, and the jet uniformly decreases in strength. In the presence of a positive uniform wind in the direction of the surface geostrophic flow the surface jet initially weakens but then strengthens again as the wind‐driven Ekman flow opposes the frictionally driven Ekman flow and the cross‐front density gradient increases.

Near‐inertial internal wave interactions with mesoscale fronts: Observations and models

Moored observations of near‐inertial currents and buoyancy fluctuations from the Frontal Air Sea Interaction Experiment (FASINEX) may be understood as nearinertial internal waves reflecting off sheared background currents. The FASINEX moored array measured upper ocean currents, temperature, and conductivity between January and June 1986 in a region of the Sargasso Sea heavily populated by strong upper ocean fronts. The background jets associated with these features have typical cross‐front scales of 20km, maximum currents of 0.5 m s−1, and relative vorticities of ±0.3ƒ. In the vicinity of fronts, observed near‐inertial currents have short horizontal scales, show no clear horizontal phase structure, and trace out elongated current ellipses with elevated across‐front over along‐front variance. During these periods, inertial band‐pass buoyancy fluctuations often correlate with major‐axis currents. WKB ray‐tracing models of propagating, monochromatic plane waves exhibit horizontal phase propagation and correlations between minor‐axis currents and buoyancy and thus cannot explain these results. Models of near‐inertial internal waves reflected or trapped by a barotropic frontal jet produce mode‐like structures with short horizontal scales, little horizontal phase propagation, and anisotropic current ellipses aligned primarily cross front. Model buoyancy may correlate with one or both current ellipse components, depending on the incidence angle of the wave and the degree to which it reflects. Completely reflected waves with horizontal scales similar to those of the front and near‐normal incidence angle produce current ellipses and correlation patterns consistent with the observations.

The Subinertial Momentum Balance of the North Atlantic Subtropical Convergence Zone

Abstract The upper-ocean to forcing by pressure gradients and wind stress is examined using observations from the Frontal Air–Sea Interaction Experiment. A moored way acquired time series of winds. upper-ocean currents, temperatures, and salinities between winter and late spring in a region of the Sargasso Sea known for the presence of upper-ocean fronts. These fronts have timescales of 10 days and dominate current variance, while winds varied with the 4-day timescale of passing weather systems. Employing a frequency domain regression model. it is found that gestrophy accounts for most of the low-frequency (>100 h)current variance in the seasonal pycnoclino, but wind-forced shear becomes important nearer the surface. In particular, currents oriented in the typical NE–SW alonfront direction display geostrophic balance, while those perpendicular to them do not. Wind forcing can produce geostrophic currents indirectly through Ekman pumping. and knowledge of the geostrophic shear is required to distinguish be...

The Upper-Ocean Response to Surface Heating

Abstract Moored observations of atmospheric variables and upper-ocean temperatures from the Long-Term Upper-Ocean Study (LOTUS) and the Frontal Air-Sea Interaction Experiment (FASINEX) are used to examine the upper-ocean response to surface heating. FASINEX took place between January and June 1986 at 27°N, 7°M while LOTUS took place between May and October 1982 at 34°N, 7°W. The frequency-domain transfer function between rate of change of heat and the net surface heat flux is consistent with a one-dimensional heat balance between heating and convergence of vertical turbulent heat flux at timescales longer than the inertial. The observations satisfy the vertically integrated one-dimensional heat equation and indicate that the response to surface heating has been successfully isolated. Within the internal waveband, upward phase propagation in the response is inconsistent with a one-dimensional balance and the vertically integrated heat balance fails. The internal waveband response is explained as a balance ...

Fractal nature of the planetary boundary layer depth in the trade‐wind cumulus regime

The fractal nature of the planetary boundary layer depth in the trade wind cumulus regime was explored. The boundary layer depths were derived from discontinuities in airborne lidar backscatter profiles obtained during the Frontal Air‐Sea Interaction EXperiment (FASINEX) in February 1986. Scale‐invariant behaviour was observed over horizontal distances ranging from 50 m to 50 km. The degree of intermittency and non‐stationarity of the signals were calculated using standard multifractal statistical methods.

Evolution of the instability of a mixed‐layer front

Observations from the Frontal Air‐Sea Interaction Experiment (FASINEX), indicating the presence of small‐scale cold‐core features in the North Atlantic Subtropical Convergence Zone, motivated a recent linear analysis of the instability of a geostrophically balanced mixed‐layer front. The results of that analysis suggested that the instability would preferentially form small‐scale cold‐core eddies at finite amplitude. In the present study, the finite‐amplitude evolution of the fastest growing mode of this nongeostrophic baroclinic instability is investigated numerically. The linear prediction of cold‐core eddy formation is confirmed by the nonlinear calculation. There are large horizontal and vertical heat and potential vorticity fluxes associated with the developing disturbance. The heat flux is confined above the thermocline, in the region of sloping frontal isotherms that provide the energy source for the instability, but the potential vorticity fluxes are maximum 50–75 m deeper and reach into the thermocline. A tongue of low‐potential‐vorticity fluid is advected 50–75 m downward along isopycnal surfaces from the cold side of the front into the thermocline at the mixed‐layer base. The small‐scale potential vorticity structure has similarities to estimates of the upper ocean potential vorticity field obtained previously from FASINEX observations. The calculations illustrate the role that frontal instabilities may play in the flux of heat and potential vorticity from the mixed layer into the thermocline. The evolution of the disturbance resembles baroclinic wave life cycles obtained in atmospheric models.

The relationship between the microwave radar cross section and both wind speed and stress: Model function studies using Frontal Air‐Sea Interaction Experiment data

The Frontal Air‐Sea Interaction Experiment (FASINEX) provided a unique data set with coincident airborne scatterometer measurements of the ocean surface radar cross section (RCS) (at Ku band) and near‐surface wind and wind stress. These data have been analyzed to study new model functions which relate wind speed and surface friction velocity (square root of the kinematic wind stress) to the radar cross section and to better understand the processes in the boundary layer that have a strong influence on the radar backscatter. Studies of data from FASINEX indicate that the RCS has a different relation to the friction velocity than to the wind speed. The difference between the RCS models using these two variables depends on the polarization and the incidence angle. The radar data have been acquired from the Jet Propulsion Laboratory airborne scatterometer. These data span 10 different flight days. Stress measurements were inferred from shipboard instruments and from aircraft flying at low altitudes, closely following the scatterometer. Wide ranges of radar incidence angles and environmental conditions needed to fully develop algorithms are available from this experiment.

Observations of Superinertial and Near-Inertial Wind-driven Flow

Abstract The superinertial and near-inertial wind-driven flow in the western North Atlantic is examined using data from two recent experiments. The Frontal Air–Sea Interaction Experiment (FASINEX) took place at 27°N, 70°W during 1986. The Long-Term Upper-Ocean Study (LOTUS) took place at 34°N, 70°W during 1982. Each experiment included moored measurements of meteorological variables that allowed estimation of the wind stress and oceanic currents. The directly wind-driven flow is isolated from other sources of variability, such as internal waves and mooring motion, using a transfer function between geocentric acceleration and wind stress. The transfer function is examined in rotary spectral bands bounded by periods of 36 and 12 h, and 12 and 2 h. For surface-moored observations, wind-driven mooring motion is found to cause a response that extends at least to 1000 m (much deeper than the frictional layer of direct wind forcing). Once this artifact is removed, the directly wind-driven flow is identified. Thi...

Linear instability of a mixed‐layer front

The linear instabilities of a geostrophically balanced frontal jet are computed numerically in an effort to understand the origin of intense small‐scale cold‐core features observed in the North Atlantic subtropical convergence zones during the Frontal Air‐Sea Interaction Experiment (FASINEX). A simple analytic continuously‐stratified basic state is used to represent the observed frontal structure. The most unstable linear modes have e‐folding timescales of 2–3 days, alongfront wavelengths near 70 km, and significant asymmetry across the front. The results provide indirect evidence that baroclinic instability of the frontal jet is a possible source of the cold‐core features observed during FASINEX.

The heat budget in the North Atlantic subtropical frontal zone

The heat budget in the North Atlantic subtropical frontal zone is examined using moored measurements of horizontal velocity, temperature, and surface heat flux obtained during the Frontal Air‐Sea Interaction Experiment (FASINEX). The three moorings used in this calculation defined a triangle with base of 19 km and height of 28 km, and were densely instrumented from the surface to 160‐m depth. Empirical orthogonal functions (EOFs) are employed to identify components of current and temperature variability with scales that are resolved by the observations. A least squares fit to the EOF‐filtered data is used to derive maps of velocity and temperature. The temperature field is dominated by the passage of fronts, with the highest variability at 160 m. Horizontal temperature gradients at 160 m peak at values approaching 1 × 10−4 °C m−1 and are indicative of a tilting of the seasonal thermocline. Velocities are directed primarily along isotherms and are swifter at the surface than at depth. The heat budget of the upper 160 m, for periods longer than 48 hours, is essentially a balance between‐rate of change of heat and horizontal advection; the correlation between these terms is 0.7. A simple explanation is the northward advection by a mean current of a series of fronts. The effect of surface heating becomes apparent in the upper 40 m, notably at the diurnal frequency and its first harmonic. An imbalance in the mean (102 days) heat budget can be rectified by a mean downwelling velocity of 5 × 10−5 m s−1 at 160 m.

Vorticity and Vertical Circulation at an Ocean Front

Abstract Density and velocity data with 4-km horizontal resolution from a survey of a front during FASINEX (Frontal Air-Sea Interaction Experiment) are combined to describe the structure of the top 300 m of the ocean. The geostrophic velocity field is derived and is used to examine the relative importance of stratification, relative vorticity, and twisting terms in Ertel's potential vorticity Q. Tenfold isopycnic changes in Q are found across a horizontal scale of only 10 km. These changes are confined to isopycnals that outcrop from the seasonal thermocline into the mixed layer. The ageostrophic velocity field is quantified by solution of the omega equation, and vertical velocities of up to 40 m day−1 are found. Small (40 km) surface-trapped (top 200 m) eddies are found to play a crucial role in the transport and effective diffusion of properties across the thermocline out of the mixed layer.

Ocean surface topography measured by the Geosat radar altimeter during the Frontal Air‐Sea Interaction Experiment

An objective analysis (OA) technique was used to analyze Geosat Geodetic Mission residual sea height anomalies (RSHA) for the Frontal Air‐Sea Interaction Experiment (FASINEX), conducted in the western North Atlantic. Preprocessing of the Geosat data included removal of a mean sea surface and orbit error and corrections for tides, wave height, inverse barometer, and ionospheric and dry tropospheric effects. Geosat data were not corrected for water vapor. Central to the analysis was the computation of a spatially anisotropic, time‐dependent autocorrelation function directly from the Geosat data. Results show a single autocorrelation function peak, having a diameter of ∼270 km at 0‐day time lag. A 15‐month time series of RSHA topography was constructed at 10‐day intervals, forming a set of 47 OA maps. These maps show a field of topographic features propagating to the west at ∼4 km d−1. The features exhibited spatial and temporal scales of ∼200 km and ∼50 days, respectively. The root‐mean‐square (rms) of the OA‐derived feature amplitudes fluctuated slowly between 0.05 and 0.09 m with a period of ∼200 days over the 15‐month interval. These fluctuations appeared to be related to the passage of large‐scale, northeast‐southwest trending sea surface temperature (SST) anomaly features reported for FASINEX, which were surface manifestations of large‐scale, first‐mode baroclinic Rossby waves. The rms amplitude of the OA‐derived features was maximal (minimal) during periods of negative (positive) SST anomaly, when the depths of the mixed layer and main thermocline in the upper ocean shoaled (deepened). The fluctuations also appeared to be correlated with the structure of the field. Periods of decreased feature amplitudes corresponded to a “closely packed” structure, while during times of increased amplitudes, the field structure became “zonal” with regions of alternating sign. During FASINEX (January–June 1986) the field was characterized by the closely packed structure and decreasing feature amplitudes. This corresponded to the period of positive SST anomaly and high upper ocean heat content reported for FASINEX.

Forced ocean response during the Frontal Air‐Sea Interaction Experiment

A central hypothesis of the Frontal Air‐Sea Interaction Experiment (FASINEX) was that the strong gradients in the upper ocean associated with ocean fronts would spatially modulate the response of the ocean to local atmospheric forcing. To test this hypothesis, measurements of near‐surface meteorology and upper ocean variability were made for 5 months from an array of moorings and, for one of those 5 months, from two ships. A number of fronts passed through the moored array, and the moored meteorological and current meter data were used to examine the response of the upper ocean to atmospheric forcing at subinertial and near‐inertial frequencies. At low frequencies an Ekmanlike, frictionally driven component in the mixed layer and a more deeply penetrating component thought to be forced by Ekman pumping were found. Strong near‐inertial response was found in the mixed layer. The near‐inertial motion showed considerable spatial variability in the seasonal thermocline; here it was evident that the structure of the front influenced the downward propagating, near‐inertial internal waves. Data were also obtained by profiling instruments deployed along cross‐frontal sections. These data were used to examine the spatial and temporal variability of the internal waves, microstructure, and dissipation, and the relation of that variability to oceanic fronts and atmospheric events. The spectral densities of the components of the internal waves that rotated clockwise with increasing depth were higher in regions of negative relative vorticity than those predicted by Garrett and Munk (Munk, 1981), and estimates of clockwise and total shear were highest in regions of negative relative vorticity. In the strong shear of the ocean front, significant anisotropy of the internal wave field was observed and kinetic energy dissipation levels relative to the rate of work by the wind were higher than previously reported.

Climatological perspectives, oceanographic and meteorological, on variability in the subtropical convergence zone in the northwestern Atlantic

We discuss the large‐scale setting for the Frontal Air‐Sea Interaction Experiment (FASINEX). A climatology of the west‐central North Atlantic, derived from various data sets, is presented, and the meteorological and oceanographic context for FASINEX is thereby established. While the atmospheric and oceanic processes of interest during the intensive phases of FASINEX are not resolved in this large‐scale study, we are able to show how these scales produce an environment favorable to the smaller‐scale processes of frontogenesis that are the subject of other contributions in this issue.

Descriptive oceanography during the Frontal Air‐Sea Interaction Experiment: Medium‐ to large‐scale variability

Medium‐ and large‐scale oceanographic variability in the Sargasso Sea is examined during the Frontal Air‐Sea Interaction Experiment (FASINEX), focusing primarily on processes that influence the formation of subtropical fronts. From Fall to Spring the mean meridional gradient of meridional Ekman transport in the Subtropical Convergence Zone (STCZ) enhances the meridional sea surface temperature (Ts) gradients between 26° and 32°N. In the presence of this enhanced mean gradient, baroclinic eddies with zonal wavelengths of ≈800 km and periods of ≈200 days exert the dominant influence on the formation of subtropical fronts at medium and large scales. These eddies generate westward propagating Ts anomaly features with the same dominant wavelengths and periods. They are confined between 26° and 32°N and have amplitudes that occasionally exceed ±1°C. Ts fronts tend to be found within bands ≈200 km wide that roughly follow the periphery of these anomaly features. Deformation in the horizontal eddy current field is primarily responsible for the existence of these frontal bands. The migration of the strong front originally bracketed by the FASINEX moored array was related to the westward propagation of the larger‐scale eddy/anomaly/frontal‐band pattern. The moored array was located within a warm‐anomaly feature during most of the experiment, which produced exceptionally warm conditions in the upper ocean. These anomalies are confined between 26° and 32°N, not only because the relatively large seasonal mean Tsy there allows horizontal eddy currents to force strong anomalies, but also because the baroclinic eddies with wavelengths of ≈800 km and periods of ≈200 days are confined to the STCZ. Large meridional variability exists in many properties of the eddy field, much of which can be traced to the influence of the Sargasso Sea mean current field on eddy variability.

Overview of the Frontal Air‐Sea Interaction Experiment (FASINEX): A study of air‐sea interaction in a region of strong oceanic gradients

From 1984 to 1986 the cooperative Frontal Air‐Sea Interaction Experiment (FASINEX) was conducted in the subtropical convergence zone southwest of Bermuda. The overall objective of the experiment was to study air‐sea interaction on 1‐ to 100‐km horizontal scales in a region of the open ocean characterized by strong horizontal gradients in upper ocean and sea surface properties. Ocean fronts provided both large spatial gradients in sea surface temperature and strong jetlike flows in the upper ocean. The motivation for and detailed objectives of FASINEX are reviewed. Then the components of the field program are summarized. Finally, selected results are presented in order to provide an overview of the outcome of FASINEX.

Medium‐ to large‐scale atmospheric variability during the Frontal Air‐Sea Interaction Experiment

Shipboard, aircraft, and satellite atmospheric data are examined to determine the representativeness of the Frontal Air‐Sea Interaction Experiment (FASINEX) intensive measurement period, Phase II, in terms of climatology, atmospheric forcing, the general structure of the marine atmospheric boundary layer (MABL), and variability in boundary layer properties affecting air‐ocean dynamics and thermodynamics. With regard to climatology, conditions observed during the intensive period were typical in terms of air‐sea temperature differences, surface pressure patterns, cloud cover, and storm tracks. Storm system variabilities, such as the air temperature behind cold fronts, wind stress maxima occurring after frontal passage, the times for clockwise vector wind shifts associated with the frontal systems, and the systems' speeds, are estimated and emphasized in synoptic‐scale characterizations, since these can be related to observed ocean responses. The local ocean surface variability was observed to have an influence on regional boundary layer properties and on air‐sea interaction parameters even in the presence of the atmospheric storms.

Ocean frontal variability in the Frontal Air‐Sea Interaction Experiment

Oceanographic observations during the Frontal Air‐Sea Interaction Experiment (FASINEX) in the Sargasso Sea indicate that upper ocean fronts in the subtropical convergence are strongly surface‐intensified coherent flow structures which have a preferred orientation and translation speed. Fronts grow as they propagate in this region such that currents roughly double in strength in a time that they translate by twice their width. While currents are dominated by geostrophic momentum balance, the Rossby number of these flows can reach a few tenths. Mixed layer water from the cold side of fronts appears to subduct under warm‐side near‐surface waters, since isolated lenses of water with cold‐side temperature and salinity characteristics are found embedded in the seasonal thermocline on the warm side of fronts. Although plausible vertical speeds were observed, no actual rates of subduction could be estimated. Typical frontal widths, temperature contrasts, horizontal currents, vertical shears, vertical currents, and translation speeds of fronts in the region studied are 20 km, 1°C, 0.5 m s−1, 0.005 s−1, 0.0005 m s−1, and 0.2 m s−1, respectively.

Air‐sea fluxes and surface layer turbulence around a sea surface temperature front

The response of the lower marine atmospheric boundary layer to sharp changes in sea surface temperature was studied in the Frontal Air‐Sea Interaction Experiment (FASINEX) with aircraft and ships measuring mean and turbulence quantities, sea surface temperature, and wave state. Changing synoptic weather on 3 successive days provided cases of wind direction both approximately parallel and perpendicular to a surface temperature front. For the wind perpendicular to the front, both wind over cold‐to‐warm and warm‐to‐cold surface temperatures occurred. For the cold‐to‐warm case, the unstable boundary layer was observed to thicken, with increased convective activity on the warm side. For the warm‐to‐cold case, the surface layer buoyant stability changed from unstable to neutral or slightly stable, and the sea state and turbulence structure in the lower 100 m were immediately altered, with a large decrease in stress and slowing of the wind. Measurements for this case with two aircraft in formation at 30 and 100 m show a slightly increased stress divergence on the cold side. The turbulent velocity variances changed anisotropically across the front: the streamwise variance was practically unchanged, whereas the vertical and cross‐stream variances decreased. Model results, consistent with the observations, suggest that an internal boundary layer forms at the sea surface temperature front. The ocean wave, swell, and microwave radar backscatter fields were measured from several aircraft which flew simultaneously with the low‐level turbulence aircraft. Significant reductions in backscatter and wave height were observed on the cold side of the front.