Abstract

This paper examines the strong, summertime northerly low-level jet (LLJ) that frequently exists along the California coast. The persistent synoptic-scale pressure distribution (North Pacific high to the west, thermal low to the east) and baroclinity created by the juxtaposition of the heated continent and the cool marine layer produce the mean structure of this LLJ. Strong diurnal thermal forcing, coupled with topographic influences on the flow, modulate the jet structure, position, and intensity. A mesoscale model is used to examine many of the complex facets of the LLJ flow dynamics. Several sensitivity studies, in addition to a control experiment, aid in this investigation. Principal findings of this study include the following.

1. The pronounced east–west slope of the marine planetary boundary layer (MPBL) is not due primarily to colder SST values along the coast.

2. Dynamically forced low-level coastal divergence, coupled with synoptic-scale divergence, appears to be dominant in determining MPBL inversion slope and profoundly impacts the coastal stratus distribution.

3. Maximum baroclinity occurs in midafternoon, whereas the LLJ maximum occurs in the evening. An analytical treatment of the dynamics shows that diurnal variation of the jet-level baroclinity, coupled with inertial and friction effects, explain this jet timing.

4. In a no-terrain simulation, the jet is broader, somewhat weaker, and tilts more to the west than in our control case.

5. A deeper boundary layer occurs over the location of the Central Valley of California in the no-terrain simulation than in the more realistic control run. Consequently, a delay in time of maximum baroclinity aloft occurs in the no-terrain case, and the LLJ maximum occurs later as compared to the control.

6. The core of the jet, which resides in the inversion capping the MPBL, lowers and moves toward the coast during the day and lifts and moves farther away from the coast at night.

7. Meso-β-scale structure of the LLJ along the coast is forced by the topography of points and capes. Thee mesoscale model simulation has supercritical flow, showing expansion fan characteristics, in the MPBL around Cape Mendocino.

8. Model results are consistent with mountain wave theory in that a near-surface wind speed maximum and pressure minimum are modeled on the lee side of Cape Mendocino.

9. The LLJ maxima in the lee of points and capes produce local maxima in surface stress. The position of these wind stress maxima correlate well with the location of cold pools observed in the SST, implying that locally enhanced, wind-forced upwelling plays a major role in the creation of such cold SST patches.

Abstract

The model we describe involves a unique strategy in which a high vertical resolution grid is nested within the coarse vertical resolution grid of a regional numerical weather prediction (NWP) model. Physics computations performed on the high vertical resolution grid involve time-dependent solution of second-order turbulence equations, the transfer equations for long- and shortwave radiation, and moist thermodynamic calculations which include liquid water content and fractional cloudiness. The dynamical computations involving advection, pressure gradient, and Coriolis terms are performed on the regional model grid. The two grids interact fully each model time step.

This approach represents an extension into NWP of the general practice of supplying coarse large-scale dynamical forcing to high-resolution boundary layer models. Aside from the computational savings of performing dynamical calculations only at the coarser resolution, we also avoid difficulties which can arise with high vertical-resolution dynamical computations in regions of significant topography. This model can, however, be easily made to take on the appearance of a standard, nonnested model by specifying everywhere one fine grid paint per coarse grid layer.

Several preliminary model forecasts are presented. The first is a 36-hour forecast over the Mediterranean and adjacent regions during midsummer. This provides a good test of the model's ability to develop a realistic cool marine mixed layer over the Mediterranean, while properly treating the extreme diurnal variations in the boundary layer over North Africa. Our second numerical forecast takes place in a much more active synoptic regime involving a wintertime frontal passage at a weather station ship in the North Atlantic.

Abstract

A one-dimensional turbulence model has been coupled with the large-scale fields of a hemispheric model so as to produce a high-resolution marine boundary layer forecast system. Model initialization is performed either by use of individual ship soundings or from standard fields of the hemispheric model. Detailed boundary layer forecasts in specified oceanic regions are desirable for many purposes, but large-scale model forecasts with such high resolution are computationally impractical. This paper presents results from approximately 90 different 24 h forecasts at the location of four different ocean station vessels.

We statistically compare model forecast profiles of temperature and moisture with verifying soundings, and also evaluate persistence as a forecast. Results consistently show a significant improvement of the model forecasts relative to persistence. The one-way influence driving force provided by large-scale time derivative terms derived from the hemispheric model is found to be very important to this coupled forecast system.

Abstract

The Naval Research Laboratory’s Coupled Ocean–Atmosphere Mesoscale Prediction System (COAMPS) is used in conjunction with satellite observations and data from the Coastal Waves 1996 experiment to investigate the dynamics of unusual wave clouds that occur upwind and offshore of orographic features along the California coast. Results indicate that supercritical flow within the marine boundary layer, interacting with blocking coastal orography, is forced to decelerate and an atmospheric bow shock forms. The location and orientation of the COAMPS forecast shock matches well with the leading edge of the wave clouds in satellite imagery, and the modeled jump in boundary layer depth across the shock is in good agreement with the aircraft observations. In the parameter space of Froude number and jump strength that develops within the flow (observed and modeled), the shock manifests itself as an undular bore.

On the innermost grid (Δx = ⅓ km), long, lineal variations in the wind, temperature, and moisture fields are forecast to develop on the subcritical side of the shock front and the modeled wavelength of these perturbations is close to the observed ∼4 km wavelength of the cloud lines. Their cellular structure and the quadrature between the vertical velocity and potential temperature fields strongly suggest that these are trapped internal gravity modes. Further, solutions to the Taylor–Goldstein equation for stationary waves, using a model-computed Scorer parameter profile, provide a comparable estimate of ∼3 km for a trapped, resonant wavelength.

The subkilometer forecasts presented are the highest-resolution real data forecasts with COAMPS to date. Time-dependent outer boundary conditions are supplied to COAMPS by the Naval Operational Global Atmospheric Prediction System. The nonhydrostatic nature of the COAMPS model is essential to forecasting these nonhydrostatic, trapped waves.

Abstract

On 28 August 2002, a visually striking sequence of events appeared in satellite imagery showing a coastally trapped disturbance (CTD) propagating northward along the coast of California against a northerly background flow. As a narrow tongue of coastal stratus indicative of the CTD propagated northward, a long, linear set of wave clouds developed ahead of the advancing CTD and angled away from the coast. The CTD and cloud lines moved northward over the next ∼6 h and, as they approached Cape Mendocino (CM), the leading edge of the CTD clouds rolled up into a cyclonic mesoscale eddy—with the wave clouds being wrapped into the eddy. The CTD abruptly stalled and failed to round CM. Further, a second cyclonic mesoscale eddy formed southwest of Point Arena (PA).

Although there has been extensive study of the propagation phase of CTDs, relatively little attention has been paid to the cessation of their propagation wherein mesoscale eddy development is not uncommon. Using the U.S. Navy's Coupled Ocean–Atmosphere Mesoscale Prediction System (COAMPS), run in an operational manner, numerous observed features of this case are forecast, including: (i) the cold, shallow, cloud-filled, northward-propagating CTD; (ii) the development, linear structure, orientation, and movement of an oblique hydraulic jump–like (“shock”) feature; (iii) a southerly wind shift associated with the CTD that precedes the advancing cloud tongue by several hours in both the observations and the model; (iv) the modeled CTD that rounds PA, but fails to round CM; and (v) the formation of modeled cyclonic mesoscale eddies near both CM and PA. North of PA, however, a phase error develops in which the modeled CTD propagation is too slow.

The model forecast cloud tongue behaves as a gravity current, and similar to earlier observational and modeling studies of CTDs, the model forecasts a bore propagating in the stratified atmosphere immediately above the marine boundary layer. Supercritical flow is forecast in the accelerating northerly flow rounding CM, and when the advancing bore interacts with this high Froude number region a pronounced oblique shock develops and the CTD stalls. Vorticity is enhanced along this shock due to vertical stretching and potential vorticity is generated within the shock. Additionally, juxtaposition of the CTD's southerly flow with the background northerly flow creates a vortex sheet–like shear zone along the offshore flank of the CTD, with the horizontal gradient of absolute vorticity changing signs, which is a necessary condition for classic barotropic instability.

Abstract

A vertically mesoscale regional numerical weather prediction model is used to simulate an arctic front. The front was observed during the Arctic Cyclone Expedition of 1984. The regional model employs a unique vertical nesting scheme in which the dynamics computations are performed on a low vertical-resolution (coarse) grid and the physics computations are performed on a high vertical resolution (fine) grid nested within the coarse grid. Turbulent fluxes are parameterized using a second-order closure approach. The model forecast compares favorably with the observations. Moreover, the model develops detailed mesoscale and boundary layer structure that verifies against the observations when initialized using only sparse, synoptic-scale data.

A control experiment is run in which identical, high vertical resolution is used on both the dynamics and the physics grids. Several additional simulations are performed in order to demonstrate the utility and accuracy of the vertical nesting methodology. With the typical nested configuration (14 coarse grid levels, 24 fine levels), the evolution of the front is nearly identical to the control. When the resolution is degraded to 14 points on both grids, significant structural differences in the boundary layer arise.

The terms of the frontogenetic forcing function are evaluated in each of the experiments. In all of the simulators, the horizontal deformation is the dominant frontogenetic effect while the tilting term is the dominant frontolytic term for this arctic front, just as it is for midlatitude cold fronts. The diabatic term is predominantly frontolytic with the strongest heating occurring in the cold air as the front moves off the ice edge and out over the Barents Sea. In an experiment in which surface sensible and latent heat fluxes are deleted, a slightly stronger front having more pronounced ageostropic circulation develops.

Abstract

The Catalina eddy event of 21 July 1992 is simulated using a mesoscale data assimilation system featuring an optimum interpolation analysis, incremental update, and second-order closure physics. The results are contrasted with other recent modeling studies of the Catalina eddy. Genesis of the eddy occurs when changes on the synoptic scale lead to an intensification of the east–west pressure gradient near the coast, resulting in enhanced northwesterly flow along the coast and over the mountains east of Point Conception. Lee troughing results in an alongshore pressure gradient at the coast with higher pressure to the south. Topographically trapped, ageostrophic southerly flow is then initiated. The combination of southerly flow along the coast with strong northwesterly flow to the west results in formation of a cyclonic eddy in the bight. The zone of southerly flow is characterized by a deep, cool, cloud-topped boundary layer that can considerably alter coastal weather and impact activities involved with aviation, air quality, fire weather, and microwave refractivity. While other recent modeling studies have failed to properly represent boundary layer structure, the data assimilation system used in the present study reproduces these features.

Results show that the model forecast eddy is in relatively good agreement with surface wind observations. The data assimilation system, which consists of the analysis–initialization scheme and the forecast model, retains much of the mesoscale structure of the forecast, while adjusting the position of the eddy to better fit the observations. Within the zone of southerly flow, rapid deepening of the boundary layer is accompanied by the formation of stratus clouds. Through the use of sensitivity studies, the authors demonstrate that the deepening of the boundary layer results from convergence and upward motion forced by the topographic barrier along the coast and that the interaction between clouds and radiation plays a significant role.

Abstract

Supercritical flow interaction occurring in the marine boundary layer between closely spaced coastal capes is investigated with a mesoscale numerical prediction model. As an extension of previous work, the U.S. Navy’s Coupled Ocean/Atmosphere Mesoscale Prediction System (COAMPS) is used to perform idealized model simulations with marine layers of varying upstream Froude number to elucidate the different flow responses for a single convex bend. The impact upon the supercritical flow of introducing a series of closely spaced coastal bends is then investigated. The expansion fan is significantly reduced in magnitude and size by the formation of a compression wave at a blocking, concave bend approximately 150 km downstream. Building upon the idealized marine layer response, real-data forecasts are then examined for several time periods of supercritical flow interaction between Cape Blanco, Oregon, and Cape Mendocino, California.

Observations from the Coastal Waves 1996 (CW96) field program were collected in the vicinity of these capes on several days during June–July of 1996. Aircraft measurements on three CW96 flights provide model validation and show ample evidence of supercritical phenomena, while buoy data along the Oregon and California coastline indicate substantial diurnal variability in the marine environment. GOES-9 satellite imagery reveals preferred regions of clearing in the coastal stratus deck downwind of convex coastal bends, which is consistent with supercritical expansion fan dynamics.

Real-data COAMPS forecasts of summertime marine layer flow between these major capes indicate that the supercritical flow features, and their degree of interaction, vary diurnally. Diurnal oscillations in the upstream Froude number and flow direction driven by the sea–land-breeze circulation enhance or diminish the expansion fan in the lee of Cape Blanco, thereby altering the flow conditions encountering the concave turn at Cape Mendocino. In a manner similar to that produced in the idealized simulations, a compression jump forms due to the impact of highly supercritical flow within the Cape Blanco expansion fan upon the Cape Mendocino terrain. The compression wave becomes detached and propagates northward during the afternoon in response to a reduction in upstream Froude number. This propagating compression wave occurred in all three days of the study. The findings presented here demonstrate that supercritical flow responses about several closely spaced coastal bends cannot be analyzed independently.

Abstract

During the summer months, the California coast is under the influence of persistent northwesterly flow. Several times each summer, this regime is disrupted by coastally trapped wind reversals (CTWR) in which the northwesterly flow is replaced by southerlies in a narrow zone along the coast. Controversy exists as to the physical mechanisms responsible for initiation and maintenance of CTWRs. While it is clear that coastal terrain is important in creating the trapped response, the precise role played by terrain is unclear. In the present study, these issues are investigated using a nonhydrostatic mesoscale model to simulate the 10–11 June 1994 CTWR event. The results show that the model successfully reproduces many of the observed features of this event, including anomalous vertical structure involving the relatively shallow boundary layer with a warm, nearly neutral layer above; the northward propagation of southerly flow in advance of a tongue of coastal stratus/fog; and a substantial reduction in propagation speed due to the sea breeze. Of the several mechanisms that have been proposed in the literature to characterize these events, these results are most consistent with a topographically trapped gravity current. Further investigation, required to verify this hypothesis, is ongoing.

Two sensitivity studies are used to examine the role of terrain in producing and maintaining the CTWR. In the first sensitivity study, the coastline from Pt. Conception to Pt. Reyes is replaced with a straight line and a uniform 840-m-high ridge is placed adjacent to the coast. This simplification permits better isolation of the terrain influence on the mesoscale pressure field and the forcing of the CTWR by the pressure distribution. The results show that adiabatic warming associated with flow over the coastal terrain is required to produce the alongshore pressure gradient, which forces ageostrophic southerly flow, and that, in the absence of bays and gaps in this terrain, southerly flow extends to the location of the minimum pressure. In a second sensitivity study, the height of the ridge along the coast is set to zero. In this simulation there is no mesoscale organization of the southerly flow. Moreover, the structure of the marine boundary layer near the coast is altered by removal of downslope flow and the gravity current characteristics seen in the control and first sensitivity study are absent.