Although our knowledge of the boundary layer over the open
ocean, and over homogenous land surfaces is incomplete, it is
sufficient to make significant progress in our understanding of
the connection between surface exchange processes and larger scale
problems [ Rogers, 1995]. In contrast, our knowledge of the
coastal boundary layer, whether over the land or the sea, which
is dominated by horizontal heterogeneous forcing, is deficient in
many areas and limited to relatively few studies. Results
derived from homogeneous open ocean or land conditions are not
generally applicable to the coast. Davidson et al. [1992]
have shown that accurate coastal wind prediction requires knowledge
of the local responses to a given synoptic forcing scenario. They
reported that the surface roughness effect on wind stress for a
15 m s
wind is about 25% higher in coastal waters than in
open ocean regions. Most parameterizations of the surface
fluxes, however, are derived from homogeneous land conditions, modified
for open ocean conditions [e.g., Smith, 1988]. While most
of the research on wind-wave interactions has occurred in coastal
regions, the complexity of these interactions makes the
interpretation of bulk surface transfer coefficients particularly
difficult [ Geernaert, 1990]. Smith et al. [1992] demonstrated
a relationship between wave age and surface roughness height, which
contributes to the expression for the wind drag coefficient. In
turn, the wave age, which depends on the fetch and water depth,
is sensitive to the wind direction relative to the coastline and
bottom boundary [e.g., Geernaert and Smith, 1994].
The coastal atmospheric boundary layer is inherently heterogeneous, dominated by variations in topography, large temperature gradients and large changes in roughness. The coastal ocean is characterized by large variations in sea surface temperature (SST) and roughness and a nonequilibrium sea state. These conditions produce interactions between sea-breezes, mesoscale eddies, and terrain-generated winds that cause complex flow patterns [ Wilczak et al., 1991; Douglas and Kessler, 1991; Ulrickson, 1992].
Upwelled water is often much colder than the ambient surface
water, so sharp temperature fronts form between the colder water
nearshore and the warmer water offshore. The coherent SST
features that are observed throughout the coastal ocean in
regions of upwelling [ Davis, 1985; Kelly, 1985;
Dewey et al., 1991; Largier et al., 1993; Flament et
al., 1994] present a complex structure to the atmosphere. These
are regions of particularly intense air-sea interactions because of
the large inhomogeneities and nonequilibrium conditions. The
structure of the atmospheric boundary layer becomes increasingly
complex in the vicinity of ocean fronts. Changes in stability allow
the development of internal boundary layers that affect the
interaction of the surface with the overall boundary layer.
These result in complex wind stress, water vapor, cloud and
boundary layer depth patterns [e.g., Friehe et al., 1991;
Rogers et al., 1995a, b]. The effect of a modest
temperature front (2
C in 5 km) was demonstrated by
Friehe et al. [1991] in the open
ocean where a 50% reduction in the wind stress was observed as
warm air flowed over cooler water. Coherent wind circulations
develop in response to these surface temperature fronts and in
response to the inhomogeneity introduced by the land-sea
boundary. Rogers et al. [1995b] observed the development
of a stable internal boundary layer (IBL) when warm air flowed from
the land over a cooler sea. Gravity waves, associated with
mountain lee waves, propagated along the direction of the mean
wind shear within the IBL. These waves modulated the wind field
and surface stress, on a scale of about 20 km, producing changes
comparable to those observed by Friehe et al. [1991] in the
vicinity of an ocean front. Riordan and Lin [1992] have
shown that coastal winds, immediately offshore of North and South
Carolina, often exhibit a mesoscale confluent and diffluent
pattern that is governed by the configuration of the coast and
the SST field.
The wind-driven cross-shelf circulation is particularly important because it drives upwelling and downwelling and is primarily responsible for the variability of the surface temperature and flow fields that feedback to the atmosphere. Upwelling occurs when an equatorward alongshore wind on a west coast produces an offshore surface flow driven by turbulent stresses (Ekman transport). This offshore flow is compensated by an onshore flow deeper in the water column and a compensating vertical velocity that results in irreversible incorporation of cold nutrient rich water into the surface layer. Dalu and Pielke [1990] investigated the behavior of coastal currents and of upwelling forced by winds of different spatial and temporal structures. Their results show that the intensity and duration of the wind-driven current depends on the duration of the wind. If the wind stress is periodic, upwelling occurs only when the period of forcing is longer than a characteristic time scale, which is the sum of the inertial period and the friction e-folding time. Upwelling occurs in a horizontal region of the order of the Rossby radius of deformation (see below). The horizontal gradient of the wind stress, however, can be more important than the deformation radius in determining the extent of the upwelling region [ Dalu and Pielke, 1990].
The effects of topography on the current, and local
thermally-driven atmospheric circulation on the wind flow, are
important in determining the wind-driven circulation [ Davis
and Bogden, 1989]. Results obtained from the Northern California
Coastal Circulation Study [ NCCCS, 1990] indicate that the
coldest water along the coast does not always coincide with upwelling
favorable winds. They observed the coincidence of coldest
surface temperatures and persistent northward shelf flow and no
correlation between the currents and wind. A possible cause of
this effect is Cape Mendocino, which may determine the flow in
the atmosphere and forces the convergence of shelf currents,
which result in offshore transport in the ocean. Munchow
and Garvine [1993] showed that the local wind and lateral buoyancy
fluxes from estuaries are the two major forcing mechanisms on the
continental shelf of the Mid Atlantic Bight on the eastern
seaboard of the United States. Their results indicate a linear
superposition of the wind and buoyancy-forced motions. The
current has a mean flow of about 10 cm s
that generally
opposes the upwelling local favorable winds. These winds, however,
force important cross-shelf flows.
Boundary layer stratus and stratocumulus clouds are persistent features of cool, upwelling coastal regions, such as the west coast of the United States. The shallow depth of the boundary layer and strength of the inversion indicate that cloud development is very sensitive to small changes in boundary layer structure [e.g., Dorman, 1994]. Betts [1990] demonstrated that a large diurnal signature in coastal stratocumulus occurs when daytime heating raises cloud base and thins the cloud layer, decoupling it from the surface. This results in an increase in the wind stress and heat flux divergence between the top of the subcloud layer and the surface [e.g., Betts and Boers, 1990; Rogers and Koracin, 1992]. The diurnal variation in cloud amount is also accompanied by changes in cloud-top height, cloud liquid water path, and effective droplet radius [ Minnis et al., 1992]. Skupniewicz et al. [1991] used a numerical model and sodar observations to determine the boundary layer structure in the vicinity of coastal stratocumulus clouds. They showed that, although a sea-breeze developed and propagated landward, baroclinic effects (density variations) associated with the cloud resulted in a circulation that caused the cloud edge to remain stationary offshore. Felsch and Whitlatch [1993] developed a forecast scheme to predict stratus surges along the central California coast associated with coastally-trapped waves [e.g., Winant et al., 1988]. The periodic clearing of large areas of clouds off the coast of California has implications for climate studies because of the effect of stratocumulus on the global albedo. Kloesel [1992] has shown that synoptic-to-mesoscale clearing episodes are correlated with ridging of the Pacific subtropical anticyclone in the United States Pacific Northwest region that results in offshore flows.