• Bao, J. W., , J. M. Wilczak, , J. K. Choi, , and L. H. Kantha, 2000: Numerical simulations of air–sea interaction under high wind conditions using a coupled model: A study of hurricane development. Mon. Wea. Rev., 128 , 21902210.

    • Search Google Scholar
    • Export Citation
  • Bao, J. W., , S. A. Michelson, , and J. M. Wilczak, 2002: Sensitivity of numerical simulations to parameterizations of roughness for surface heat fluxes at high winds over the sea. Mon. Wea. Rev., 130 , 19261932.

    • Search Google Scholar
    • Export Citation
  • Beardsley, R. C., , C. E. Dorman, , C. A. Friehe, , L. K. Rosenfeld, , and C. D. Winant, 1987: Local atmospheric forcing during the Coastal Ocean Dynamics Experiment I. A description of the marine boundary layer and atmospheric conditions over a northern California upwelling region. J. Geophys. Res., 92 , 14671488.

    • Search Google Scholar
    • Export Citation
  • Beardsley, R. C., , A. G. Enriquez, , C. A. Friehe, , and C. A. Alessi, 1997: Intercomparison of aircraft and buoy measurements of wind and wind stress during SMILE. J. Atmos. Oceanic Technol., 14 , 969977.

    • Search Google Scholar
    • Export Citation
  • Beg-Paklar, G., , V. Isakov, , D. Koračin, , V. Kourafalou, , and M. Orlic, 2001: A case study of bora-driven flow and density changes on the Adriatic shelf (January 1987). Contin. Shelf Res., 21 , 17511783.

    • Search Google Scholar
    • Export Citation
  • Burk, S. D., , T. Haack, , and R. M. Samelson, 1998: Mesoscale simulation of supercritical, subcritical and transcritical flow along coastal topography. J. Atmos. Sci., 56 , 27802795.

    • Search Google Scholar
    • Export Citation
  • Burk, S. D., , T. Haack, , L. T. Rogers, , and L. J. Wagner, 2003: Island wake dynamics and wake influence on the evaporation duct and radar propagation. J. Appl. Meteor., 42 , 349367.

    • Search Google Scholar
    • Export Citation
  • Chan, J. C. L., , Y. Duan, , and L. K. Shay, 2001: Tropical cyclone intensity change from a simple ocean–atmosphere coupled model. J. Atmos. Sci., 58 , 154172.

    • Search Google Scholar
    • Export Citation
  • Chen, C., , R. C. Beardsley, , S. Hu, , Q. Xu, , and H. Lin, 2005: Using MM5 to hindcast the ocean surface forcing fields over the Gulf of Maine and Georges Bank region. J. Atmos. Oceanic Technol., 22 , 131145.

    • Search Google Scholar
    • Export Citation
  • Dorman, C. E., 1985: Hydraulic control of the northern California marine layer. Eos, Trans. Amer. Geophys. Union, 66 , 914.

  • Dorman, C. E., , and C. D. Winant, 1995: Buoy observations of the atmosphere along the west coast of the United States, 1981–1990. J. Geophys. Res., 100 , 1602916044.

    • Search Google Scholar
    • Export Citation
  • Dorman, C. E., , and C. D. Winant, 2000: The marine layer in and around the Santa Barbara Channel. Mon. Wea. Rev., 128 , 261282.

  • Dorman, C. E., , L. Armi, , J. M. Bane, , and D. P. Rogers, 1998: Sea surface mixed layer during the 10–11 June 1994 California coastally trapped event. Mon. Wea. Rev., 126 , 600619.

    • Search Google Scholar
    • Export Citation
  • Dorman, C. E., , D. P. Rogers, , W. Nuss, , and W. T. Thompson, 1999: Adjustment of the summer marine boundary layer around Pt. Sur, California. Mon. Wea. Rev., 127 , 21432159.

    • Search Google Scholar
    • Export Citation
  • Dorman, C. E., , T. Holt, , D. P. Rogers, , and K. Edwards, 2000: Large-scale structure of the June–July 1996 marine boundary layer along California and Oregon. Mon. Wea. Rev., 128 , 16321652.

    • Search Google Scholar
    • Export Citation
  • Dudhia, J., 1993: A nonhydrostatic version of the Penn State/NCAR mesoscale model: Validation tests and simulations of an Atlantic cyclone and cold front. Mon. Wea. Rev., 121 , 14931513.

    • Search Google Scholar
    • Export Citation
  • Edwards, K. A., 2000: The marine atmospheric boundary layer during Coastal Waves 96. Ph.D. thesis, University of California, San Diego, 132 pp.

  • Edwards, K. A., , A. M. Rogerson, , C. D. Winant, , and D. P. Rogers, 2001: Adjustment of the marine atmospheric boundary layer to a coastal cape. J. Atmos. Sci., 58 , 15111528.

    • Search Google Scholar
    • Export Citation
  • Fett, R. W., , and S. D. Bury, 1981: Island barrier effects as observed by satellite and instrumented aircraft, and simulated by a numerical model. Mon. Wea. Rev., 109 , 15271541.

    • Search Google Scholar
    • Export Citation
  • Friehe, C. A., , R. C. Beardsley, , C. D. Winant, , and J. P. Dean, 1984: Intercomparison of aircraft and surface buoy meteorological data during CODE 1. J. Atmos. Oceanic Technol., 1 , 7986.

    • Search Google Scholar
    • Export Citation
  • Grell, G. A., , J. Dudhia, , and D. R. Stauffer, 1995: A description of the fifth-generation Penn State/NCAR Mesoscale Model (MM5). National Center for Atmospheric Research Tech. Note TN-398, 122 pp.

  • Haack, T., , S. D. Burk, , C. Dorman, , and D. Rogers, 2001: Supercritical flow interaction within the Cape Blanco–Cape Mendocino orographic complex. Mon. Wea. Rev., 129 , 688708.

    • Search Google Scholar
    • Export Citation
  • Halliwell, G. R., , and J. S. Allen, 1987: The large-scale coastal wind field along the west coast of North America, 1981–1982. J. Geophys. Res., 92 , 18611884.

    • Search Google Scholar
    • Export Citation
  • Johnson, E. R., , and G. G. Vilenski, 2004: Flow patterns and drag in near-critical flow over isolated orography. J. Atmos. Sci., 61 , 29092918.

    • Search Google Scholar
    • Export Citation
  • Jones, I. S. F., , and Y. Toba, 2001: Wind Stress over the Ocean. Cambridge University Press, 307 pp.

  • Kessler, R. C., , and S. G. Douglas, 1991: Numerical study of mesoscale eddy development over the Santa Barbara Channel. J. Appl. Meteor., 30 , 633651.

    • Search Google Scholar
    • Export Citation
  • Koračin, D., , and C. E. Dorman, 2001: Marine atmospheric boundary layer divergence and clouds along California in June 1996. Mon. Wea. Rev., 129 , 20402056.

    • Search Google Scholar
    • Export Citation
  • Koračin, D., , C. E. Dorman, , and E. P. Dever, 2004: Coastal perturbations of marine layer winds, wind stress, and wind stress curl along California and Baja California in June 1999. J. Phys. Oceanogr., 34 , 11521173.

    • Search Google Scholar
    • Export Citation
  • Lagerloef, G. S. E., , and R. L. Bernstein, 1988: Empirical orthogonal function analysis of Advanced Very High Resolution Radiometer surface temperature patterns in Santa Barbara Channel. J. Geophys. Res., 93 , 68636873.

    • Search Google Scholar
    • Export Citation
  • Large, W. G., , and S. Pond, 1981: Open ocean momentum flux measurements in moderate to strong winds. J. Phys. Oceanogr., 11 , 324336.

  • Lopez, J. W., , and L. H. Kantha, 2000: A data-assimilative numerical model of the northern Indian Ocean. J. Atmos. Oceanic Technol., 17 , 15251540.

    • Search Google Scholar
    • Export Citation
  • Ly, L. N., , P. Luong, , J. D. Paduan, , and D. Koracin, 1999: Response of the Monterey Bay Region to wind forcing by an atmospheric model. Proc. Third Conf. on Coastal Atmospheric and Oceanic Prediction and Processes, New Orleans, LA, Amer. Meteor. Soc., 76–79.

  • Mass, C. F., , and M. D. Albright, 1989: Origin of the Catalina eddy. Mon. Wea. Rev., 117 , 24062436.

  • Mellor, G. L., , and T. Yamada, 1974: A hierarchy of turbulence closure models for planetary boundary layers. J. Atmos. Sci., 31 , 17911806.

    • Search Google Scholar
    • Export Citation
  • Pickett, M. H., , and J. D. Paduan, 2003: Ekman transport and pumping in the California Current based on the U.S. Navy’s high-resolution atmospheric model (COAMPS). J. Geophys. Res., 108 .3327, doi:10.1029/2003JC001902.

    • Search Google Scholar
    • Export Citation
  • Powers, J. G., , and M. T. Stoelinga, 2000: A coupled air–sea mesoscale model: Experiments in atmospheric sensitivity to marine roughness. Mon. Wea. Rev., 128 , 208228.

    • Search Google Scholar
    • Export Citation
  • Powers, J. G., , M. T. Stoelinga, , and W. S. Boyd, 1997: Distributed numerical weather prediction via satellite. Bull. Amer. Meteor. Soc., 78 , 27552770.

    • Search Google Scholar
    • Export Citation
  • Rogers, D., and Coauthors, 1998: Highlights of Coastal Waves 1996. Bull. Amer. Meteor. Soc., 79 , 13071326.

  • Rogerson, A. M., 1999: Transcritical flows in the coastal marine atmospheric boundary layer. J. Atmos. Sci., 56 , 27612779.

  • Samelson, R. M., 1992: Supercritical marine-layer flow along a smoothly varying coastline. J. Atmos. Sci., 49 , 15711584.

  • Shafran, P. C., , N. L. Seaman, , and G. A. Gayno, 2000: Evaluation of numerical predictions of boundary layer structure during the Lake Michigan Ozone Study. J. Appl. Meteor., 39 , 412426.

    • Search Google Scholar
    • Export Citation
  • Skyllingstad, E., , P. Barbour, , and C. E. Dorman, 2001: The dynamics of northwest summer winds over the Santa Barbara Channel. Mon. Wea. Rev., 129 , 10421061.

    • Search Google Scholar
    • Export Citation
  • Thompson, W. T., , S. D. Burk, , and J. Rosenthal, 1997: An investigation of the Catalina eddy. Mon. Wea. Rev., 125 , 11351146.

  • Winant, C. D., , C. E. Dorman, , C. A. Friehe, , and R. C. Beardsley, 1988: The marine layer off northern California: An example of supercritical channel flow. J. Atmos. Sci., 45 , 35883605.

    • Search Google Scholar
    • Export Citation
  • View in gallery

    June 1999 MM5 7-m wind speed and direction. This is the geographical coverage of the 9-km MM5 grid domain. The cause of the wind speed structure around Point Conception is the focus of this article.

  • View in gallery

    (a) SSM/I satellite example of strong winds about Point Conception on 9 Jun 1999. (b) SSM/I satellite example of moderate winds about Point Conception on 19 Jun 1999. (c) SSM/I satellite example of weak winds about Point Conception on 23 Jun 1999.

  • View in gallery

    500-hPa analysis at 1200 UTC for a (a) strong wind case on 9 Jun, (b) moderate wind case on 19 Jun, and (c) weak wind case on 22 Jun. Source is NCEP reanalysis provided by the NOAA/OAR/ESRL PSD (information online at http://www.cdc. noaa.gov/).

  • View in gallery

    Map of the 1-km MM5 grid domain and topography, with indicated hourly surface locations and geographical points. Circles are NDBC buoys, Xs are platforms, squares are land stations. NOJO is in a low pass on the Santa Ynez Mountain ridge.

  • View in gallery

    Buoy wind speeds at selected (top) east–west stations along the Santa Barbara Channel and (bottom) north–south stations along the western edge of the Southern California Bight. Area wind strengths for each day, posted below the lower frame, are noted in Table 2.

  • View in gallery

    (a) June 1999 B11-measured winds and nearest-point MM5-simulated winds for domain 3 (1-km resolution). (b) June 1999 B54-measured winds and nearest-point MM5-simulated winds for domain 3 (1-km resolution).

  • View in gallery

    Diagram of marine layer hydraulic response to Point Conception.

  • View in gallery

    Wind speed decreases cross coast in a compression bulge on the north side of the Point Conception topographic complex. Average June 1999 wind speeds are plotted relative to distance from the coast at PURI. Circles are measured values and the bar is one standard deviation wide. The PSAL coastal station has nearly identical values and distance as PURI, but is left off for clarity; “X” is nearest MM5 domain 3 value. Station designations are in Table 3.

  • View in gallery

    MM5-simulated sea surface winds for 0000 UTC 10 Jun, under strong winds. This is the geographical coverage of the 3-km MM5 grid domain. Every fourth wind vector is shown for clarity.

  • View in gallery

    MM5-simulated sea surface winds for 0000 UTC 20 Jun, under moderate winds.

  • View in gallery

    Locations of the profiler and sounding stations noted in Table 6. The Vandenberg balloon sounding station on the north side of Point Conception has two locations, usually one for the morning and one for the evening sounding. The San Diego balloon sounding and the Miramar radar profiler appear as a single symbol, as they are at the same site. The locations of buoys used in Fig. 5 are also shown.

  • View in gallery

    MM5-simulated domain 3 vertical section, extending east–west across the coast near PURI, valid at 0000 UTC 10 Jun. This profile extends through a compression bulge. The alongshore winds for the compression bulge are plotted in Fig. 4. The wind speeds decrease and the marine layer depth increases in the last 30 km of the coast.

  • View in gallery

    MM5-simulated domain 3 north–south vertical section across the western mouth of the Santa Barbara Channel over B54 at 00 UTC 10 Jun. Santa Rosa Island is on the left and the Santa Ynez Mountains are on the right. The leeside mountain flow is limited to the Santa Ynez Mountain upper slope. The main wind speed maximum in the center of the channel near the sea surface is an expansion fan. The island on the south side significantly extends the high-speed wind zone to the south.

  • View in gallery

    MM5-simulated domain 3 north–south vertical section across the mid–Santa Barbara Channel at 0000 UTC 10 Jun. This left end of this section extends between the two large islands of Santa Rosa and Santa Cruz and is west of B53. The Santa Barbara coast is on the right.

  • View in gallery

    Sea level pressure analysis during a strong wind case on 0000 UTC 10 Jun 1999. There is a well-supported local pressure minimum over the Santa Barbara Channel. This feature and a weak ridge of higher pressure on the east end of the Santa Barbara Channel are found during almost all strong and moderate wind speed conditions in summer. The weak pressure maximum over the coast north of Point Conception is due to the deepened marine layer in a compression bulge. The number posted is the observed sea level pressure minus 1000 hPa.

  • View in gallery

    Pressure differences for selected buoys (top) along and (bottom) across the Santa Barbara Channel. The pressures mostly increase to the east of B54, are against the eastbound flow over the channel, and are generally unrelated to the measured winds at B54 and B53 (Fig. 5).

  • View in gallery

    MM5-simulated, 3-km grid domain, 1500-m level valid at 0000 UTC 10 Jun 1999 during a strong wind case. Modest northerly winds cross the Santa Ynez Mountains.

  • View in gallery

    MM5-simulated, 3-km grid domain, 1500-m level valid at 0000 UTC 20 Jun 1999 during a moderate wind case. Weak northerly winds cross the Santa Ynez Mountains.

  • View in gallery

    Surface winds at stations along the Santa Barbara Channel coast. The strong offshore winds unique to (top) GAVW are due to its location at the Gaviota Canyon mouth. More typical are the weak and mostly onshore winds experienced by the better-exposed surface stations such as GAVE (only 2 km to the east of GAVW) and the coastal plane stations of ECAP, WCAM, and SBA, as well as other mountain slope stations that are not shown.

  • View in gallery

    (top) Pressure differences across the western Santa Ynez Mountains between B11 and SBA along a path that approximates the direction of the approaching marine layer winds. (middle) Along this approximate path are NOJO, the Santa Ynez ridge crest station, and the lee Santa Ynez foothill station GOLE. (bottom) B54 shows that the high-speed winds in the Santa Barbara Channel are unrelated to the cross-mountain pressure gradient or the leeslope winds. Note that the B54 winds are rotated to the mean wind direction of 306° in order to facilitate viewing.

  • View in gallery

    Vertical soundings around the Santa Ynez Mountains for a strong wind case (1200 UTC 9 Jun–0000 UTC 20 Jun) and a moderate wind case (1200 UTC 19 Jun–0000 UTC 20 Jun). VBG is off the western end of the Santa Ynez Mountains and GOL is on the coastal plane in the immediate lee, while LAX is to the southeast. The vectors point downwind. All cross-mountain ridge flow of any consequence is above 700 m and occurs in the afternoon.

  • View in gallery

    June 1999 satellite-observed sea surface temperature (found online at http://coastwatch.pfel.noaa.gov/). The general temperature pattern within 200 km of Point Conception is unrelated to the wind maxima (Fig. 1).

All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 35 35 10
PDF Downloads 30 30 13

Response of the Summer Marine Layer Flow to an Extreme California Coastal Bend

View More View Less
  • 1 Scripps Institution of Oceanography, and San Diego State University, San Diego, California
  • 2 Desert Research Institute, Reno, Nevada
© Get Permissions
Full access

Abstract

A summer wind speed maximum extending more than 200 km occurs over water around Point Conception, California, the most extreme bend along the U.S. West Coast. The following several causes were investigated for this wind speed maximum: 1) synoptic conditions, 2) marine layer hydraulic flow effects, 3) diurnal variations, 4) mountain leeside downslope flow, 5) sea surface temperature structure, and 6) island influence. Synoptic conditions set the general wind speed around Point Conception, and these winds are classified as strong, moderate, or weak. The strong wind condition extends about Point Conception, reaching well offshore toward the southwest, and the highest speeds are within 20 km to the south. Moderate wind cases do not extend as far offshore, and they have a moderate maximum wind speed that occurs over a smaller area in the western mouth of the Santa Barbara Channel. The weak wind speed case consists of light and variable winds about Point Conception. Each category occurs about one-third of the time. Atmospheric marine layer hydraulic dynamics dominate the situation after the synoptic condition is set. This includes an expansion fan on the south side of the point and a compression bulge on the north side. The expansion fan significantly increases the wind speeds over a large area that extends to the southwest, south, and east of Point Conception, and the maximum wind speed is increased for the strong and moderate synoptic cases as well. The horizontal sea surface temperature pattern contributes to the sea surface wind maximum through the Froude number, which links the potential temperature difference between the sea surface temperature and the capping inversion temperature with marine layer acceleration in an expansion fan. A greater potential temperature difference across the top of the marine layer also causes more energy to be trapped in the marine layer, instead of escaping upward. The thermally driven flow resulting from differential heating over land in the greater Los Angeles, California, coastal and elevated area to the east is not directly related to the wind speed maximum, either in the Santa Barbara Channel or in the open ocean extending farther offshore. The effects of the thermally driven flow extend only to the east of the Santa Barbara Channel. The downslope flow on the south side of the Santa Ynez Mountains that is generated by winds crossing the Santa Ynez Mountain ridge contributes neither to the high-speed wind maximum in the Santa Barbara channel nor to that extending farther offshore. Fifth-generation Pennsylvania State University–NCAR Mesoscale Model (MM5) simulations do support a weak leeside flow in the upper portions of the Santa Ynez Mountains. The larger Channel Islands have a significant effect on the marine layer flow and the overwater wind structure. One major effect of the Santa Barbara Channel Islands is the extension of the zone of high-speed winds farther to the south than would otherwise be the case.

Corresponding author address: Dr. Clive E. Dorman, Integrative Oceanography, Scripps Institution of Oceanography, La Jolla, CA 92093-0209. Email: cdorman@ucsd.edu

Abstract

A summer wind speed maximum extending more than 200 km occurs over water around Point Conception, California, the most extreme bend along the U.S. West Coast. The following several causes were investigated for this wind speed maximum: 1) synoptic conditions, 2) marine layer hydraulic flow effects, 3) diurnal variations, 4) mountain leeside downslope flow, 5) sea surface temperature structure, and 6) island influence. Synoptic conditions set the general wind speed around Point Conception, and these winds are classified as strong, moderate, or weak. The strong wind condition extends about Point Conception, reaching well offshore toward the southwest, and the highest speeds are within 20 km to the south. Moderate wind cases do not extend as far offshore, and they have a moderate maximum wind speed that occurs over a smaller area in the western mouth of the Santa Barbara Channel. The weak wind speed case consists of light and variable winds about Point Conception. Each category occurs about one-third of the time. Atmospheric marine layer hydraulic dynamics dominate the situation after the synoptic condition is set. This includes an expansion fan on the south side of the point and a compression bulge on the north side. The expansion fan significantly increases the wind speeds over a large area that extends to the southwest, south, and east of Point Conception, and the maximum wind speed is increased for the strong and moderate synoptic cases as well. The horizontal sea surface temperature pattern contributes to the sea surface wind maximum through the Froude number, which links the potential temperature difference between the sea surface temperature and the capping inversion temperature with marine layer acceleration in an expansion fan. A greater potential temperature difference across the top of the marine layer also causes more energy to be trapped in the marine layer, instead of escaping upward. The thermally driven flow resulting from differential heating over land in the greater Los Angeles, California, coastal and elevated area to the east is not directly related to the wind speed maximum, either in the Santa Barbara Channel or in the open ocean extending farther offshore. The effects of the thermally driven flow extend only to the east of the Santa Barbara Channel. The downslope flow on the south side of the Santa Ynez Mountains that is generated by winds crossing the Santa Ynez Mountain ridge contributes neither to the high-speed wind maximum in the Santa Barbara channel nor to that extending farther offshore. Fifth-generation Pennsylvania State University–NCAR Mesoscale Model (MM5) simulations do support a weak leeside flow in the upper portions of the Santa Ynez Mountains. The larger Channel Islands have a significant effect on the marine layer flow and the overwater wind structure. One major effect of the Santa Barbara Channel Islands is the extension of the zone of high-speed winds farther to the south than would otherwise be the case.

Corresponding author address: Dr. Clive E. Dorman, Integrative Oceanography, Scripps Institution of Oceanography, La Jolla, CA 92093-0209. Email: cdorman@ucsd.edu

1. Introduction

The summer sea surface wind field along the U.S. West Coast has appeared in a number of modern studies with sufficient resolution to detect the maximum wind speed around Point Conception, California, including those based upon surface buoy and land measurements (Halliwell and Allen 1987; Dorman and Winant 2000; Dorman et al. 2000), aircraft measurements (Rogers et al. 1998), satellite-based measurements (Koračin et al. 2004), and modeling simulations (Burk et al. 1998; Pickett and Paduan 2003; Koračin et al. 2004). All of the studies covering Point Conception indicate a wind speed maximum extending around and well offshore of the Point Conception coast (Fig. 1). Why is this wind speed maximum here? What are the causes of this wind feature, and how is this related to other smaller-scale wind structures that are not resolved in the fifth-generation Pennsylvania State University–National Center for Atmospheric Research (PSU–NCAR) Mesoscale Model (MM5) analysis shown in Fig. 1? It is the purpose of this manuscript to investigate these questions. We hypothesize that there are several causes for this wind maximum and for other mesoscale wind structures over the coastal waters around Point Conception; they are listed in Table 1. The general large-scale coastal winds about southern California are established by the synoptic-scale conditions. However, high-speed winds are usually limited to a portion of California, such as northern California or the region between Cape Mendocino and Point Conception. Within this geographical variation, there is a range of marine layer conditions, such as the faster wind speeds, marine layer depth, and inversion strength.

The summer meteorological conditions along the California coast are reviewed in Koračin et al. (2004). The northerly summer wind flow is driven by the large-scale anticyclone off northern California and by the thermal low pressure over the southwestern United States. Warm, dry subsiding air from the anticyclone forms a contrast with the cold ocean water, and even more so with upwelled ocean water. The high heat capacity and moist ocean surface help to form and dominate an atmospheric marine layer by direct contact. Above the marine layer is subsiding air in an air temperature inversion that is generally lowest and strongest along the central California coast.

California’s summer dense marine layer, capped by an air temperature inversion, is a medium for hydraulic dynamics. The coastal mountains extending above the layer effectively form a coastal wall. The marine air acts as a single-layer incompressible fluid that responds hydraulically when the layer speed is sufficiently fast. If the Froude number is greater than 1, or supercritical, the speed of the layer is faster than the speed of long gravity waves in the layer and so that they cannot travel upstream. A supercritical marine layer moving around a corner, where the wall bends away from the flow, would accelerate and thin in the lee. Simple frictionless flow would continue on indefinitely as a uniformly faster, shallower layer in the form an expansion fan. If the wall turns into the flow downstream, the supercritical flow would slow and thicken in a compression jump that would extend out into the flow forming an oblique angle with the coast (Dorman 1985; Winant et al. 1988).

The supercritical layer structure responding to a bending away wall or cape is much more realistic when frictional effects are included (Samelson 1992). The expansion fan is limited to an area in the immediate lee of the bend. Downstream of the maximum wind speed, friction slows the layer and may be assisted by the coastline turning into the flow to form a compression jump where the flow deepens and slows over a short distance. Including the effects of rotation and moderate external pressure gradient forcing only makes relatively small structural changes.

If the Froude number of the inbound flow is greater than 0.5, but less than 1, the marine layer accelerates and thins around a cape, becoming supercritical over a limited area that includes a supercritical expansion fan (Rogerson 1999). This is the transcritical case that extends hydraulic effects to almost any California cape during the summer where the wind speed is not light. On the downstream side of the expansion fan, friction and an outward coastal projection will cause slowing and thickening of the layer either over a short horizontal distance in a compression jump or, even more slowly, over an extended horizontal distance, depending upon conditions. While the imposed large-scale pressure gradient can have a significant effect on the extent of the hydraulic features, the flow downstream is subcritical. Changing inbound layer conditions or deepening the downstream subcritical layer may cause the deeper layer to the south to move upstream toward the cape as a trapped event.

The large-scale bend in California at Cape Mendocino causes a large-scale summer expansion fan that extends along the coast, past Point Conception, and hundreds of kilometers offshore (Edwards 2000; Edwards et al. 2001). Within this California-scale expansion fan, each cape has smaller-scale hydraulic features with a compression bulge on the upwind side and an expansion fan on the downwind side. The alongcoast extent of this large-scale expansion is supported by buoy and coastal station winds and soundings (Dorman et al. 2000).

Partial blocking of the inbound marine layer forces slowing and deepening on the upwind side of a cape. If the inbound flow is supercritical, there will be a hydraulic jump (Rogers et al. 1998; Dorman et al. 1999); if not, a smoother change will form a compression bulge. Mesoscale modeling (Koračin and Dorman 2001) found a compression bulge on the upwind side of every major cape in southern Oregon and California, which is supported by satellite cloud observations.

Cape Blanco and Cape Mendocino in northern California are sufficiently close that their hydraulic features interact with each other (Haack et al. 2001). In addition, the Cape Mendocino compression bulge can become detached, moving northward in response to a diurnally decreasing Froude number.

Turning to Point Conception, the topography of the surrounding marine layer is a manifestation of the hydraulics of the layer to the degree that it is controlled by single-layer flow dynamics responding to coastal topography. A dense network of surface stations and measurements, soundings, and radar profilers suggests that this is largely a transcritical flow. Further, the measured pressure gradient across the expansion fan wind maximum is constant with a strong supercritical expansion (Dorman and Winant 2000). Using different mesoscale numerical models, Skyllingstad et al. (2001) and Koračin et al. (2004) found that the wind field about Point Conception fit the description of an expansion fan. Deep, inbound marine layers and strong winds are supercritical and shallow, inbound layers and weaker winds are transcritial (Skyllingstad et al. 2001).

Other dynamical effects have been proposed for the general area around the Southern California Bight that are directly related to the topography. Mountain lee effects have been suggested for downslope, leeside flow that is caused by a stronger northerly flow crossing the Santa Ynez Mountains (Mass and Albright 1989; Thompson et al. 1997). A westward-moving, trapped Kelvin-type wave on the western slope of the Santa Ynez Mountains has been proposed by Kessler and Douglas (1991), which manifests itself as westward-progressing diurnal wind reversals along the Santa Barbara coast.

A significant outstanding topographic structure at Point Conception is the extreme angular bend in the California coast, which approaches 90° and is backed by the Santa Ynez coastal mountain range. The coast then turns to the south, forming the Southern California Bight. The east side of the Southern California Bight is a north–south coast, extending from Los Angeles to past San Diego. An additional topographic complexity is the mountainous Channel Islands, which are parallel to the coastline along the south side of Point Conception. To the north, the central California coast has an orientation of 320°. The greater Point Conception topographic bend itself is a large triangular cape, with the west side oriented due north and the south side extending 80 km due east.

To understand the complex wind structure near the greatest bend in the California coastline, the 3D nature of the upwind compression bulge and the lee expansion fan in the marine layer were investigated. Satellite, buoy, and sounding data and numerical mesoscale model simulations were used for this study.

2. Observed characteristics of the spatial and temporal variability of the winds in the Southern California Bight

Synoptic atmospheric conditions, including the location of a weak summer trough over the coast and the center of the eastern Pacific subtropical anticyclone, are related to the location, area, and maximum strength of the wind speed maximum along the U.S. West Coast during the summer (Koračin et al. 2004). Usually there is one large area of wind maximum along the coast of southern Oregon and California, which tends to extend a few hundred kilometers offshore and along the coast with the fastest speeds at the coast. Often, the wind maximum is along northern California only; it may also extend along northern and central California past Point Conception, or it may be only along southern central California, extending past Point Conception.

The wind speed maximum horizontal scale and variations are confirmed by Special Sensor Microwave Imager (SSM/I) satellite wind analysis (Koračin et al. 2004). Although individual SSM/I satellites made two passes per day (in the early morning and early evening), coverage of any specific area is likely to have occurred less frequently than this, because there are occasional gaps between the tracks. It should be noted that there is a land mask that eliminates data that are closer to the coast than about 50 km, but this is not a problem for identifying a wind speed maximum that extends a few hundred kilometers offshore. SSM/I data were retrieved from Remote Sensing Systems (see information online at htt://www.remss.com), which is sponsored by the National Aeronautics and Space Administration’s (NASA’s) Earth Science Research, Education and Applications Solution Network (REASoN) Distributed Information Services for Climate and Ocean Products and Visualizations for Earth Research (DISCOVER) Project. An example of SSM/I coverage of a strong wind case touching Point Conception is shown in Fig. 2a for 9 June 1999. In this case, winds greater than 10 m s−1 extend from 33° to 43°N. SSM/I coverage of a moderate wind case for Point Conception on 19 June 1999 is shown in Fig. 2b. Winds higher than 10 m s−1 extend from Point Conception to beyond central California, with greater speeds in a more northerly location. A weak wind case is shown for 23 June 1999 in Fig. 2c, where weak winds less than 3 m s−1 extend along central and southern California.

A large-scale, 500-hPa synoptic structure is associated with each of the Point Conception wind states (Fig. 3a). The strong wind case is linked to a broad trough over the West Coast that extends to southern California. The moderate wind case has an anticyclone centered near 28°N, 130°W with an east–west ridge extending over southern California (Fig. 3b). The weak wind case has two states. In the first 6 days of June, it was a closed cyclone over north California or Oregon. Later in the month, it was an anticyclone that was centered near 37°N, 134°W, with a weak cyclone over southern California (Fig. 3c).

The conclusion is that the synoptic-scale condition determines the general strength of the winds over a large area around Point Conception. This may be expressed in terms of whether the inbound surface flow is strong, moderate, or weak. This is apparent when considering selected representative stations (Fig. 4 and Table 3) to examine the trend of buoy-measured wind speeds along two zones (Fig. 5). The upper frame of Fig. 5 contains three buoy wind speed time series along a west–east zone that includes the Santa Barbara Channel. The lower frame contains stations in a north–south zone that starts on the north side of Point Conception and ends on the southern end of the wind jet on the southwestern edge of the Southern California Bight. All of the buoys are robustly correlated with the synoptic changes.

The National Data Buoy Center (NDBC) buoys in Fig. 5 show the essential nature of the general wind field around the extended Point Conception area, including the Southern California Bight. The inbound marine air moving past the central California coast is represented by B11; B54 represents the maximum winds speed in the area and in the lee of Point Conception. The flow continues east and always decelerates before reaching B53, where there is a large diurnal range. Some eastbound flow on both sides of the Channel Islands continues east and southeast, slowing to B25, which is typical of the weak and more variable winds in the inner Southern California Bight.

Some of the inbound flow continues south, past Point Conception and the Channel Islands, to form a jet with weakening speeds to the south on the western edge of the Southern California Bight. This is represented by B11, B54, and B47 (Fig. 5). When there is a strong, large-scale wind speed maximum, the strong winds extend from an area well north of Point Conception to a distance past B47, well to the south of the point. The fastest winds occur at B54, where the speeds remain high all day, with only a small diurnal range.

Moderate winds at B54 and around Point Conception have a wind maximum that covers a smaller area than that which is associated with strong winds. All of the western buoys have slower speeds, but the highest speed is always at B54. In addition, during moderate wind conditions, there is a large diurnal range at B54, with a relatively low-speed minimum occurring around sunrise and a sharp peak in the afternoon.

When weak winds occur, they extend throughout the area about Point Conception and the Southern California Bight and at all buoys. This may be associated with southerly flow in the inner bight. This is the only case where other buoys may have faster speeds than B54.

The large-scale synoptic wind strength for each day of June 1999 is tallied in Table 2, where each day is established as strong, moderate, or weak. This is based upon several factors. A strong day is defined as having speeds >10 m s−1 at B54 and relatively fast winds at B11 and B47, while the diurnal range at B54 is small (<2 m s−1) and the satellite SSM/I large-scale coastal wind maximum includes Point Conception. A day is considered to be moderate if the speeds at B54, B11, and B47 are less than 10 m s−1, there is a large diurnal variation (>3 m s−1) at B54, and the satellite SSM/I analysis has a moderate-speed zone about Point Conception. Finally, a weak speed day is defined as a day when the B54, B11, and B47 speeds are less than 4 m s−1, the wind directions are from the south at some buoys, and the SSM/I satellite winds are weak about Point Conception. Each category occurs about ⅓ of the time.

3. Model

a. Model setup

The MM5 is a community model, used worldwide, that was developed jointly by NCAR in Boulder, Colorado, and PSU (Dudhia 1993; Grell et al. 1995). The MM5 has been used in a variety of research and application studies, including studies on atmospheric dynamics along the California coast (Koračin and Dorman 2001), and has been used as a driver for the ocean models (Powers et al. 1997; Bao et al. 2000, 2002; Lopez and Kantha 2000; Powers and Stoelinga 2000; Ly et al. 1999; Beg-Paklar et al. 2001; Chan et al. 2001; Chen et al. 2005). For the purpose of this study, the MM5 was run in a nonhydrostatic mode, with 9-km horizontal resolution and an integration step of 27 s, for all of June 1999. The month of June was selected because its conditions are typical of summer, with strong northerly winds and significant synoptic variations. The model domain consisted of 149 × 191 grid points in the horizontal direction and 35 sigma levels in the vertical direction. The model domain was represented as a Lambert-conformal map projection and was centered at 35.15°N, 120.65°W. The topography was read from the 30-s-resolution global terrain and the land use files. The first-guess synoptic fields for every 12 h were obtained from the U.S. National Centers for Environmental Prediction (NCEP) Global Data Assimilation System archive. The synoptic information includes virtual temperature, geopotential height, U and V wind components, and relative humidity, shown on a global grid with a horizontal resolution of 2.5° in both the latitudinal and longitudinal extensions. This synoptic information was horizontally interpolated onto the model grid by a two-dimensional, 16-point overlapping parabolic fit. All available upper-air and surface stations were assimilated into the synoptic fields by objective analysis, using a model grid that was extended horizontally by 180 km on all sides. The NCEP 12-h lateral boundary conditions were used to run MM5 for the entire period. The model options include mixed phase microphysics, the parameterization of shortwave and longwave radiation, including cloud–radiation effects, and the Grell cumulus parameterization (Grell et al. 1995). A surface energy balance algorithm and a five-layer soil model were used to compute surface heat and moisture fluxes and to predict surface temperature. Gayno–Seaman turbulence parameterization (Shafran et al. 2000) was chosen, which provides the turbulence kinetic energy as a prognostic variable based on the level 2.5 turbulence closure (Mellor and Yamada 1974). The geographical coverage of the gridpoint separation domain for 9 km is the area of Fig. 1 (for 3 km it is the area of Fig. 9; see below). Figure 4 shows the geographical location of the 1-km grid domain, along with the topography and geographical locations that will be referred to in the forthcoming text.

b. Model evaluation

The MM5 has been evaluated for many applications (information online at http://www.mmm.ucar.edu/mm5), and it appears to be a reliable modeling tool for atmospheric studies. However, it is still important to evaluate the model results in predicting the complex coastal dynamics of this particular case. Koračin and Dorman (2001) and Koračin et al. (2004) used MM5 to simulate winds and related fields along the California coast for all of June 1996; they evaluated the modeled winds using data from seven NDBC meteorological buoys (model simulations correctly resulted in correlation coefficients ranging from 0.61 to 0.82) and coastal wind profilers, with over 18 000 comparison points. They also used satellite visual cloud image and infrared data in order to evaluate indirectly the wind and divergence field near the coast and offshore. Their study shows that the MM5 is a sufficiently accurate tool to predict the main characteristics of the marine layer dynamics along the U.S. West Coast.

Here we examine the accuracy of the MM5 as it was set up for this investigation, using NDBC buoys as a reference. The accuracy of buoy-measured winds was established for northern California NDBC buoys during the summer (Friehe et al. 1984) and winter (Beardsley et al. 1997) by aircraft flying at about 30-m elevation over the buoy. The adjustment to a common height was done with the Large and Pond (1981) bulk parameterization. The agreement was good, with an aircraft-minus-buoy wind speed average difference and standard deviation of 0.6 ± 0.8 m s−1. It should be noted that the scatter has been linked to unresolved variables characterizing the ocean wave field, as well as to the difference between the wind and wave field vectors and other factors (reviewed in Jones and Toba 2001).

We selected 17 buoys, platforms, and coastal stations (Table 3) representing a variety of areas relative to Point Conception (over water, the north coast, and the east coast) and compared their data with the model results at corresponding points. The buoy locations are shown in Fig. 4. The main statistics of the comparison (Table 4) reveal that the stations fall into three geographical groups. The overwater stations on buoys or the edge of islands have high correlations with model simulations (0.68–0.87) and faster winds (5.1–9.7 m s−1). The north coastal stations generally have more moderate correlations with the simulations (0.56–0.77) and moderate wind speeds as well. The east coast stations and platforms stand out as being much more poorly correlated with the model simulations (0.38–0.65), and these have the weakest winds (2.3–5.0 m s−1).

Model results can also be evaluated by performing an empirical orthogonal function (EOF) analysis (Lagerloef and Bernstein 1988). This organizes the multiple stations into mean and major variability patterns. The EOFs were carried out on various collections of stations so as to examine the roles of the geographical groupings (Table 5). When all of the stations are taken together, the measured modes 1 and 3 are seen to be substantially different for MM5. However, when only the overwater stations and the north coastal stations (station groups in Table 3) are taken, the variance in mode 1 is significantly increased at the expense of the others, so that the measured and MM5 values are much more similar. The measured east coast stations weaken the EOF in the first mode, which is not picked up in the MM5 simulations. These east coast stations have light winds that are poorly correlated with each other and with the overwater stations (Dorman and Winant 2000).

Some of the results from the statistical and EOF investigations are amplified in the time series plots of a measured station, plotted with the simulations for the nearest model grid point for each of the three domains. B11 is typical of the overwater stations (Fig. 6a). The wind speeds for the simulations track the measured diurnal and the synoptic scales very closely, both in timing and magnitude, so that there are few modest differences that are detectable to the naked eye. The simulated wind direction follows that measured, but with a small but persistent offset that is reflected in the persistent difference in the north component. The simulations for each of the domains for the nearest point differ among themselves, but slightly rather than significantly. A somewhat similar result occurs for buoy B54, because the model speed follows the measured speed closely while the model-simulated direction is offset by about 20° (Fig. 6b). Otherwise, the most substantial difference is in the simulated diurnal peak, which was excessive for 12–22 June.

4. Synoptic effects on the flow structure

The examination of the synoptic pressure field during the strong wind cases indicates that the setup of the high and low pressure centers induces 25%–50% stronger sea level pressure gradients than normal across the central California coast. The most definitive structure above sea level is around 700 hPa (not shown). During the strong wind cases, the 700-hPa height gradient over the central California coast from 38° to 34°N is uniform but modest, with isoheights extending in a northwest direction offshore. During moderate and weak wind cases, the 700-hPa height gradient weakens significantly to the south along the central California coast, frequently terminating into a near zero-gradient feature, such as a saddle in the vicinity of 34°N.

In summary, synoptic pressure gradients significantly control the marine flow with frequent occurrences of high winds in the western side of the Point Conception expansion fan. The measurements and modeling demonstrate that small variations in the synoptic flow direction can regulate the strength of the flow in the western end of the Santa Barbara Channel. The synoptic pressure gradients appear to have an insignificant bearing on the generally weak flows in the eastern end of the channel.

5. Hydraulic effects

a. A primer on hydraulic dynamics

As noted in the introduction, an atmospheric marine layer moving past a cape such as Point Conception is expected to have distinctive hydraulic effects if the inbound layer is supercritical or transcritical with a Froude number (Fr) greater than 0.5. The Froude number may be computed by
i1520-0493-136-8-2894-e1
where C is the layer speed, g is the acceleration of gravity, H is the layer depth, θ is the layer potential temperature, and θt is the potential temperature of the inversion top. Note that the denominator is the speed of a long wave in the layer that is the fastest possible layer gravity wave. To compute the Froude number, we use the Vandenberg sounding (VBG) for the height, potential temperature, and strength of the inversion. The layer speed is assumed to be that of B23 multiplied by 1.15 because almost all aircraft soundings show that a buoy measurement is an underestimate by at least this and often more. The 0000 and 1200 UTC values were computed for June 1999. The result is that the Froude number is 0.90 for the strong cases, 0.67 for the moderate cases, and 0.26 for the weak cases. The Froude number standard deviations for the strong, moderate, and weak cases are 0.22, 0.15 and 0.21, respectively, so that there is a marginally significant Froude number difference between the three states. Eleven of the June 1999 strong case Froude numbers were greater than 0.8, while only two of the moderate case Froude numbers were greater than 0.8. Further, all of the strong cases and all but two of the moderate cases had Froude number greater than 0.5. Finally, the Froude number difference between the strong and moderate cases is largely due to faster speeds while the layer depths are similar. While not too fine a point should be made about how representative these rough computations might be, the measured inbound flow is at least transcritical so that hydraulic effects are expected and the strong cases tend to have higher Froude numbers than the moderate cases.

The structure of the hydraulic flow effects around the major West Coast capes of Cape Blanco, Cape Mendocino, Point Arena, and Point Sur have been directly measured over water by research aircraft (Beardsley et al. 1987; Rogers et al. 1998; Dorman et al. 1999). Measurements at buoys, coastal stations, and sounders provide time series at fixed points that support the aircraft measurements (Dorman et al. 2000; Dorman and Winant 1995, 2000). Based upon these measurements and consistent with this manuscript, the basic structure and key features of a transcritial or supercritical marine layer interacting with Point Conception are presented in a cartoon with the islands left out for clarity (Fig. 7). A cold, dense, incompressible marine layer acting as a fluid is caped by an air temperature inversion. This stable layer is less than the coastal mountain height, which effectively acts as a wall.

As the layer approaches the point, the layer reacts as a small-area compression bulge by slowing, thickening, and generating a sea level high pressure. The inner portion of the marine layer reaching the first corner begins to respond as an expansion fan. It accelerates around the corner and into the western portion of the Santa Barbara Channel where the surface speed is the fastest, the layer is the thinnest, and the sea surface pressure is the lowest. Momentum keeps the layer moving toward the east, but against the pressure gradient and surface friction. In this deceleration zone, the layer slows and thickens and the sea level pressure increases over an extended distance and not as a hydraulic jump.

Another minor feature not shown to reduce complexity is a narrow, weak wind zone that initiates where Point Conception joins the coast, extending eastward, and widening, so as to be a few kilometers in width off Santa Barbara (Dorman and Winant 2000). This overwater weak zone is believed to be related to the minimum depth that the marine layer can have at an extreme bend with the increased drag over a narrow coastal plane.

b. Measured surface winds

There are an unusually high number of hourly surface stations (Fig. 4) that are sufficient to resolve the major mesoscale features and to test the effects of hydraulic dynamics in the vicinity of Point Conception. In all but light winds, the southbound marine layer approaches from central California, slowing modestly at B11. Most of the flow continues southward, with some acceleration around the point, and reaches its fastest velocity near B54, which is in the western mouth of the Santa Barbara Channel. The flow continues east, decelerating significantly before reaching B53 and slowing even more at the eastern end of the Santa Barbara Channel. Further, the cross-channel pressure difference is related to the along-channel winds, which fit a supercritical hydraulic flow model, and not geostrophic flow (see Fig. 22 of Dorman and Winant, 2000).

Hydraulic theory expects a marine layer compression bulge on the upwind side of Point Conception. This feature appears in Fig. 8, which presents the average June 1999 wind speeds relative to distance from the coast. Under the fast-wind regimes (Table 2), this flow slows by about half at the coastal station of PURI, and the diurnal change is largely manifested as a shift in direction in the early morning. The surface measurements, the simulations of sea level winds, and the cross-coast structure in the marine layer are consistent with cross-coast slowing in a hydraulic compression bulge on the upwind side of a cape.

Continuing over the water, the marine layer flow accelerates around Point Conception past stations ARGO, IRIN, and B63, and reaches a maximum measured wind speed at B54, representing the center of the expansion fan. Under strong wind regimes (Fig. 5), B54 winds remain high and the diurnal variation is small. Under moderate regimes, B54 winds peak to high values in the afternoon, but decrease in the morning to less than 25% of the day’s maximum. The marine layer continues eastward down the Santa Barbara Channel but decelerates to about half of its value by B53 (Fig. 5). In all wind regimes, B53 wind speed peaks in the afternoon and decreases to very weak values in the morning. The layer slows as it moves farther east, moving past GAIL, and then either due east over the coastal plane or to the southeast over B25. Winds at the coastal stations along the Santa Barbara coast are weak and completely uncorrelated with higher speed winds over the more central Santa Barbara Channel (Table 4). The overwater stations are consistent with a marine layer hydraulic expansion fan centered on B54 for strong and moderate winds.

c. Simulated surface winds

MM5-simulated near-surface winds are presented to show the conditions over areas that are not represented by measurements, especially those farther offshore. Both a strong event and a moderate event are presented in order to examine the wind patterns over water. The MM5 simulations of the wind fields over water are well supported by surface measurements.

For the strong and moderate cases, the fastest winds of the diurnal cycle are near 0000 UTC (Figs. 9 and 10). During these events, the inbound marine layer is from the northwest, because it is aligned with the central California coast. Both the strong and moderate cases show additional slowing over water within 20 km of the north side of Point Conception, which is associated with the hydraulic, upwind compression bulge noted earlier.

Both cases behave as a hydraulic expansion fan in which acceleration begins within 10 km the western end of Point Conception, turns and accelerates into the western portion of the Santa Barbara Channel, and then decelerates in the middle or eastern end of the channel. In all cases, the highest speeds are on the western end of the channel, and the fastest speeds are associated with the strongest wind cases (Table 4).

The majority of the inbound marine layer flow continues to the southeast, accelerating past Point Conception and the Channel Islands. Significant turning to the east occurs well south and east of the western end of the Channel Islands. The effect is that that the greatest wind speeds in the offshore area occur 100 km to the south and 20 km to the west of the western tip of the Point Conception complex for the strong wind case (Fig. 9). For the moderate wind case (Fig. 10), a portion of the maximum is also west of the western tip and is more than 80 km to the south. At that point the flow slows and turns farther east. This persistent offshore wind maximum is consistent with a broader expansion fan, but it cannot be said to be linked to leeside flow over the Santa Ynez Mountains.

In addition to the closed-in hydraulic features revealed by the direct measurements, the MM5 simulations support a larger-scale expansion fan as the marine layer responds to Point Conception, one that includes a wind speed maximum well south and offshore of the point and the Channel Islands.

d. Marine layer top–inversion base height

It was reported above that the topography of the marine layer is a manifestation of the hydraulics of the layer to the degree that it is controlled by single-layer flow dynamics responding to a cape. To test the extent to which this holds true for the greater Point Conception complex, an analysis was performed of the 10 sounding stations in the area for June 1999 (Tables 6 and 7, Fig. 11). A marine air temperature inversion is clearly present at all sounding stations along the open coast (Table 7). The exception is the SIM station, which is in an isolated valley that has different characteristics from the open coast. The marine layer depths are posted as having the two extremes of 1200 [0400 Pacific standard time (PST)] and 0000 (1600 PST) UTC, so as to characterize the essence of the marine layer topography and to allow the inclusion of the two operational balloon stations. There is a distinct diurnal trend in the northern stations, with most having a maximum inversion height near sunrise and a minimum in the afternoon.

The air temperature inversion base heights along the Southern California Bight support those of the hydraulic model. As posted in Table 7, the inversion base is low at the inbound marine layer air station (DIA) and higher at the bulge station (VBG); it is followed by a decrease at the station on the edge of the expansion fan area in the Santa Barbara Channel (GOL), and then is greatly elevated at the sheltered downwind stations of LAX and PTL. For both types of days, the highest inversion base height occurs in the early morning, and all stations have greater heights for the strong wind cases and lower heights for the moderate wind cases, with the sole exception of the upwind (inbound marine layer) station of DIA. This is true for the inversion base heights on a strong and a moderate wind day shown in Table 7, as well as for the others that are not shown.

e. Hydraulic properties of the flow according to numerical simulations

Because most of the profile measurements are on the mainland, the hydraulic effects on the cross-coast marine layer structure are investigated by MM5 simulations. A fast wind case at 0000 UTC 10 June is selected, as this is associated with the highest wind speeds at a time when the hydraulic and leeside effects should be the greatest. The profiles are across the coast at three contrasting locations, each extending from the main coast.

The first cross section is through the compression bulge on the upwind side of Point Conception (Fig. 12). The right side starts near the station PURI (location in Fig. 4) and extends to the southwest. The air temperature inversion structure (near the 288 isotherm) and isotherms elevate beginning 30 km from shore, forming a raised inversion bulge in the 20 km nearest to shore. The surface wind slows down as it approaches the coast over the last 30 km, reaching its slowest speeds at the beach.

The second cross section is downwind, around the south side of Point Conception and across the expansion fan in the vicinity of B54 (Fig. 13). The section starts in the Santa Ynez Mountains and extends to the south, near B54 and across Santa Rosa Island. The top of the marine layer, represented by the 293 isotherm, is down to 300 m in the center of the channel and rises up to the north and south. There are low wind speeds in the lower half of the mountain slope and in the inner coastal waters (this is confirmed by surface stations). The fast winds near the sea surface start about 8 km south of the Santa Barbara Coast, extending across the channel and deepening toward the south.

The third cross section, which is farther east through B53, is in the deceleration zone (Fig. 14). There is a shallower and narrower high-speed zone over the water.

The results from the surface observations, soundings, and simulations support the presence of significant hydraulic effects during strong and moderate wind cases. This includes a hydraulic compression bulge on the upwind side, a lee expansion fan, and deceleration in the eastern end of the Santa Barbara Channel.

6. Effects of the diurnal variation on the inhomogeneous wind structure

We next consider the possible role of thermally driven circulation on the overwater wind maxima. A separate issue is the coincident diurnal wind speed cycle covering the eastern Pacific, which extends far beyond the conventional “sea breeze” (Koračin and Dorman 2001). The smaller-scale, thermally driven circulation is generally considered to be due to the differential heating and cooling of land to the east of Point Conception, such as the greater Los Angeles coastal plain and the surrounding elevated areas. Of course, this heating reduces the air density and decreases the pressure, generating a pressure gradient force from west to east. The signature of this effect would be a sea level pressure gradient increasing in magnitude from sunrise to late afternoon and coincident with a downgradient wind. Furthermore, the winds would develop over the heated land in the morning, and would, with time, extend to the west and increase in velocity. However, this model is directly inconsistent with the observed pressure and wind fields over the Southern California Bight. The surface pressure has a local minimum in the Santa Barbara Channel for individual observation times (Fig. 15), with strong and moderate winds as well as the monthly mean (not shown). Based upon the surface pressure analysis, the winds in the eastern half of the Santa Barbara Channel are in the opposite direction, as would be expected from the observed sea level pressure gradient.

Other evidence supporting the hypothesis that the diurnal variation is not a dominant factor is that the measured surface winds over water are fastest in the western mouth of the Santa Barbara Channel and are much weaker to the east. In the mean, and for most individual observations, the measured eastbound winds decrease toward the east when moving from B54 to B53, to the platforms, and then to the coastal land stations (see Fig. 5 for B54 and B53; see also, Dorman and Winant 2000). Further, the pressure gradient, as represented by the pressure differences between B54–B53 and B54–GAIL, is mostly negative after 6June (Fig. 16), which is the opposite of what would be expected if simple heating of the land creating a negative pressure gradient to the east was a significant factor. Finally, during the strong wind cases, the diurnal variation at B54 and the western Santa Barbara Channel over the water stations is small, but it is always very large at the stations further east, which is inconsistent with thermally driven flow driving the overwater wind speed maximum in the western Santa Barbara Channel.

The conclusion is that the diurnal heating over the greater Los Angeles area does not cause the high-speed winds in the Santa Barbara Channel. In addition, they cannot be responsible for the wind speed maximum extending farther offshore.

7. The effects of coastal topography on the marine flow

a. Flow over topography

To consider the possible role of mountain lee effects, we first examine the lower middle atmosphere in the vicinity of Point Conception. The 1500-m elevation was chosen as a reference point, because it is well above all of the coastal topography but still low enough to register flow directions that are influenced by the temperature/density field at lower elevations. During the fast wind case on 9–10 June, the 1500-m flow was from the north or northwest over the entire area (Fig. 17). There was a weak, cross-ridge flow over the east–west-oriented Santa Ynez Mountains backing the Santa Barbara coast and channel, which should be conducive to weak leeside effects. The diurnal changes at the south slope surface stations are modest (not shown).

There is a major change in the midlevel structure, so that by 0000 UTC 20 June (Fig. 18), the 1500-m winds have a cyclonic pattern that is centered on a weak wind area just off the coast at Los Angeles. At this time, winds are from the north over the Santa Barbara Channel. This should induce weak Santa Ynez Mountain leeside effects. These two cases will be examined in terms of the surface station measurements and simulations.

b. Surface flows

The south-facing coastal slope of the Santa Ynez Mountains on the Santa Barbara coast should be the site of leeside, downslope flow, if it exists. To the east of Point Conception and the coastal area on the north side of the Santa Barbara coast, there is a weak wind zone that is quite different in nature than that which exists over the majority of the Santa Barbara channel, despite the fact that the latter is just a few kilometers away. The low Santa Barbara Channel coastal plain is well sampled along the coast, with six automated meteorological stations; ECAP, WCAM, and SBA are typical (Fig. 19), and GAVE, GOLE, and EMMA are similar (the latter three are not shown). Most wind speeds are weak, 1–2 m s−1 onshore in the afternoon, and offshore in the early morning. Strong, cross-shore, northerly winds are unusual for these stations individually and are rare for the group of stations together. Furthermore, the wind regime on the Santa Barbara Coast is uncorrelated and unrelated to that in the central Santa Barbara Channel (Dorman and Winant 2000).

The exception to the downslope gusts is GAVW, which is at the head of a canyon and at the coast below the narrow, 320-m low point on the mountain ridge extending east–west from Santa Ynez. This steep, short canyon extends down to the sea. The GAVW station does have strong cross-shore winds that are atypical in comparison with the others, such as ECAP and even GAVE, which is on the coastal plain only 2 km to the east (Fig. 4). Very infrequently, there is a wind burst down the GAVW canyon that spreads out over the local coastal plain, briefly affecting GAVE, but not ECAP or the other stations. These infrequent wind bursts are unrelated to the persistent, high-speed winds at B54 and other coastal stations, or the NOJO station at 305-m elevation, which is well exposed to the flow at the head of the same canyon on the Santa Ynez ridge line. Here, the winds are weak and mostly toward the south, 3–6 m s−1, with diurnal shifts in direction. The wind volume down this narrow canyon is so small that it affects only the overwater area within a few kilometers of the mouth.

For significant leeside flow over a major portion of the Santa Ynez Mountains, there should be a strong cross-ridge pressure gradient. This gradient is represented by the pressure difference between B11 on the upwind side and the Santa Barbara (SBA) lee coastal station (Fig. 20, top frame). An imaginary line extending from B11 to SBA is oriented roughly northwest–southeast, which is the direction of the inbound flow on the upwind side of the Point Conception complex and would be the approximate path that the marine air would take if it continued in the same direction. The pressure difference typically diurnally alternates sign such as on 9–30 June. The one occasion where there is a persistent positive pressure difference is on 3–5 June, which is not typical of the remainder of the month (Table 2), and is associated with weak winds in the Santa Barbara Channel (as at B54 in Fig. 20). For reference, the wind vectors for B54 are presented in Fig. 20 in order to show that the cross-ridge pressure gradient and representative lee station winds are quite unrelated to the persistent high-speed winds over water. This is not consistent with significant leeside effects, and the pressure variations are unrelated to the overwater, high wind speed stations in the western mouth of the Santa Barbara Channel, such as B54. In addition, the pressure differences are unrelated to the winds at the Santa Ynez ridge-top station NOJO (Fig. 20), the lee coastal plane stations (Fig. 19) and foothill stations (only GOLE is shown), which would be expected to reflect significant leeside flow if it existed.

The summary of the information from the many surface stations along the Santa Barbara coast and on the Santa Ynez lee slope is that there is no evidence of significant leeside flow. Downslope flow is not related to the overwater winds in the Santa Barbara Channel, and certainly to those extending farther offshore.

c. Wind profiles

On a first impression, it seems that the 1500-m winds should cause significant leeside effects on the south side of the east–west-oriented Santa Ynez Mountains (Figs. 17 and 18). After all, the VBG sounding has a flow below 1500 m from the north or northwest on 9–10 and 19–20 June (Fig. 21). In the GOL sounding, the winds are from the northwest and are between 700 and 1500 m on the afternoons at 0000 UTC 10 and 20 June. In the SIM inland valley sounding (not shown), as well as at LAX just to the south of the east–west-extending coastal/inland mountains, the winds from the northwest are only above 1000 m at 1200 UTC 9 June.

Complicating the interpretation of leeside effects is the GOL sounding station, which is at the foot of the San Ynez Mountain lee and is less than 8 km from the ridgeline (Fig. 4). The measured winds and the vertical velocities below 700 m are very weak (Fig. 21, middle) for the strong and moderate wind events, which is inconsistent with strong, downslope, cross-shore lee flow.

The strong westerly winds and low pressure minimum over the Santa Barbara Channel are present for these days, not varying with the winds above 700 m, which is inconsistent with leeside effects dominating either the wind or the surface pressure field. Finally, the northerly cross-ridge flow should cause a low centered over the inner coastal plane/mountain foothills, rather than over the middle of the Santa Barbara Channel, some 20–30 km farther to the south. The conclusion is that there are some cross-ridge effects, but that they cannot account for the Santa Barbara Channel’s high-speed along-channel winds, or for its pressure minimum.

d. Lee effects according to the simulations

An MM5-simulated section was to the east of Point Conception, extending across the southern slope of the Santa Ynez Mountains (Fig. 13). The top of the marine layer, the 293 isotherm, is down to 300-m elevation in the center of the channel and rises up to the north and south. Weak winds extend from 600-m elevation down across the coastal plain and to 10 km offshore. The model simulates flow to the north, across the coastal plain and then ⅔ of the way up the south side of the Santa Ynez Mountain slope. The weakest winds in the area are 400 m above the coastal plain. Cross-shore flow extends from the east side of the Santa Ynez ridge, over the top, and then down the lee slope to roughly 600-m elevation. This supports weak leeside flow effects, but only in the upper elevations. The simulated next-day 1200 UTC profile (not shown) presents significant differences: the mountain lee speeds are less, and the downward flow stops at about 700 m and spreads out about half way across the channel at a constant level.

The cross section that is farther east, through B53 and the eastern Santa Barbara Channel, is in the deceleration zone (Fig. 14). There is a wider, cross-shore weak wind zone that starts 15 km offshore and extends across the coastal plain with upward flow, ending above the 300-m elevation of the lee mountain shown in this cross section. Significant cross-shore flow is suggested between 400 and 800 m, but it is at least 10 km inland according to this section. Twelve hours later, the section has weak flow, with a small onshore component below 700 m. While there is northerly cross-shore flow above 1000 m, it is weak too.

To summarize, the vertical sections do not support consistent, lower-level, cross-shore flow. Instead, weak cross-shore, downslope, lee mountain flow is limited to the upper elevations, and is well removed from the coastal plain and farther away from the Santa Barbara Channel.

8. The effects of the ocean surface temperature on the inhomogeneous flow field

The sea surface temperature affects the wind speeds in the marine layer through hydraulic dynamics. The Froude number of the layer is proportional to the potential temperature difference between the marine layer and the capping air temperature inversion (Winant et al. 1988). The increase in wind speed of a layer moving around a corner and into an expansion fan is a function of the Froude number. The larger the Froude number, the greater the wind speed increase, all other things being equal. The inbound marine air has a potential temperature that is about 10 K less than the maximum temperature of the capping subsiding air temperature inversion. On the other hand, the near-surface marine air temperature is close to, and tends to follow, the sea surface temperature. Along the central California coast, the sea temperature is about 1 K less than the near–sea surface air temperature (Dorman et al. 1998). Oceanographic upwelling keeps the sea temperature low along the inner 20 km of the central California coast and the western mouth of the Santa Barbara Channel (Fig. 22). The sea surface temperature warms toward the east and the south in the bight. The ocean surface temperature affects the winds in that it maintains the low temperature and high density of the marine layer, which results in a greater Froude number than would otherwise be the case. If there were no upwelled seawater around Point Conception, then the SST would be approximately 5 K warmer and the potential temperature difference across the capping air temperature inversion would decrease by about half; this would reduce the wind speeds by about 40% in the western Santa Barbara Channel.

Another effect of the large potential temperature difference between the marine layer and the air temperature at the top of the inversion is that it restricts vertical mixing. This restriction limits the downward entrainment of slower speed air into the marine layer, which further reduces the layer speed.

The sea surface temperature pattern (Fig. 22) has a structure that is only weakly coincident with the wind speed maximum near Point Conception (Fig. 1). The coldest upwelled waters are within 20 km of the central California coast. In contrast, the main wind field maximum extends more than 200 km to the southwest of Point Conception (Fig. 1) with little spatial relationship to the sea surface temperature. On the other hand, the SST increases by about 4°C toward the east along the Santa Barbara Channel, which could be associated with a minor reduction in wind speeds that is caused by increased exchange with the air above.

These simple models can only be applied to this situation heuristically. Nevertheless, the sea surface temperature is a plausible, weak background factor in maintaining the marine layer and, thus, the wind speed maximum around Point Conception.

9. Modification of the marine flow resulting from the Channel Islands

The three larger Channel Islands on the south side of the Santa Barbara Channel (Fig. 4) essentially form a wall with holes. They have a major effect on the marine layer flow on the south side of the Santa Barbara Channel and over the waters on the south side of the islands. Surface measurements across the Santa Barbara Channel decrease from the center of the channel to the north side of the island (Dorman and Winant 2000), which confirms that the island lee wind speed maxima are isolated from the Point Conception expansion fan wind maximum. Because of their initiation at the island coast, the large wind maxima and minima extending about 30 km to the southeast during strong and moderate wind regimes (Figs. 9 and 10) are believed to be responsible for them. This interpretation is supported by the MM5 simulations, which show strong vertical motions on the south side of the simulated cross section only where there is an island (cf. Fig. 13 with Fig. 14).

Finally, extensive studies of island effects on the atmospheric flow and thermal fields support similar leeside wind structures. Some of the studies include use of aircraft and satellite data with numerical model results (Fett and Bury 1981), idealized predictions and real-time forecasting with radar data (Burk et al. 2003), and shallow-water equation models with satellite data (Johnson and Vilenski 2004).

The effects of the Channel Islands are complex. One significant result is that their presence extends the high-speed zones farther to the south than they would otherwise reach. Thus, they have a significant effect in extending the wind speed maximum, but only in a limited area.

10. Summary and conclusions

There is a large, persistent summer sea surface wind speed maximum about Point Conception, California. While this general maximum extends hundreds of kilometers to the west and south of the point, a smaller absolute maximum is only tens of kilometers from the coast in the immediate, sheltered lee. The unusually long, east–west-oriented southern side of the point that is backed by coastal mountains and with a parallel line of islands to the south makes for great complexity. However, the unusually large number of surface and upper-air stations in the area combined with MM5 simulations offers an opportunity to unravel this case.

The large-scale synoptic setting along the greater California coast sets the stage for the point. This is divided into three equally occurring states of fast, moderate, and weak winds in the area around the point. With each state, there are distinctive patterns at the sea surface and a more limited range of conditions, such as the eastward extent of the central wind maximum in the lee.

The synoptic setting establishes the inbound marine layer properties as wind speed and marine layer depth. The summer atmospheric setting with the California central coast backed by coastal mountains projecting above the marine layer ensures that the inbound marine layer is from the northwest for the strong and moderate settings. For these, marine layer hydraulic dynamics force responses that include slowing and deepening on the upwind side. Expansion fan acceleration occurs as the inshore portion of the marine layer moves by the point while the farther offshore portion of the layer continues to the southeast. The highest wind speeds and shallowest layer are formed by the portion of the layer that turns into the immediate lee.

An unacceptable alternative explanation for the small-scale supermaximum in the western end of the point is that a diurnal thermal low over land to the east causes the wind speed maximum. This directly conflicts with numerous observations, including several strategically located surface and upper-air stations.

Another explanation for the wind speed maximum is that leeside flow down the southern side of the Santa Ynez Mountains that line the south side of Point Conception. MM5 simulations do support weak leeside flow in the upper portions of the lee mountains. However, the observations from a picket line of surface stations on the lee coast and a radar profiler station also in the immediate lee are in direct conflict with significant downslope, cross-shore flow causing the smaller wind maximum in the immediate lee.

Wind effects could come from two other elements: one is the effect of the sea surface temperatures on the sea surface pressure gradient and then the wind field, but this does not seem to be significant on account of structure mismatches, the other is the Channel Islands, which appear to extend the high-speed winds to the southeast into the Southern California Bight.

However, the winds seem to have a significant effect on the ocean surface temperature, causing a persistent minimum that extends to the south of Point Conception, well past 33°N. This is collocated with southbound-slowing and eastward-turning winds, which result in a positive wind stress curl, upwelling, and cold water. This interpretation is supported by atmospheric numerical modeling (Koračin et al. 2004).

Acknowledgments

We greatly appreciate the data supplied by William Brick, San Diego County Air Pollution Control District, Ed McCarthy, Pacific Gas & Electric Company, Kevin Durkee, South Coast Air Quality Management District, and Mallory Ham, Ventura County APCD. The authors thank Mr. Travis McCord for technical preparation of the mansucript. This manuscript was supported by MMS 14-35-0001-30927, NSF OCE 9907884, and NL123A-A. Author DK acknowledges support from the Office of Naval Research (ONR) Grants N00014-06-1-0028 and N00014-00-1-0524, and the DOD-DURIP-ONR Grant N00014-04-10801.

REFERENCES

  • Bao, J. W., , J. M. Wilczak, , J. K. Choi, , and L. H. Kantha, 2000: Numerical simulations of air–sea interaction under high wind conditions using a coupled model: A study of hurricane development. Mon. Wea. Rev., 128 , 21902210.

    • Search Google Scholar
    • Export Citation
  • Bao, J. W., , S. A. Michelson, , and J. M. Wilczak, 2002: Sensitivity of numerical simulations to parameterizations of roughness for surface heat fluxes at high winds over the sea. Mon. Wea. Rev., 130 , 19261932.

    • Search Google Scholar
    • Export Citation
  • Beardsley, R. C., , C. E. Dorman, , C. A. Friehe, , L. K. Rosenfeld, , and C. D. Winant, 1987: Local atmospheric forcing during the Coastal Ocean Dynamics Experiment I. A description of the marine boundary layer and atmospheric conditions over a northern California upwelling region. J. Geophys. Res., 92 , 14671488.

    • Search Google Scholar
    • Export Citation
  • Beardsley, R. C., , A. G. Enriquez, , C. A. Friehe, , and C. A. Alessi, 1997: Intercomparison of aircraft and buoy measurements of wind and wind stress during SMILE. J. Atmos. Oceanic Technol., 14 , 969977.

    • Search Google Scholar
    • Export Citation
  • Beg-Paklar, G., , V. Isakov, , D. Koračin, , V. Kourafalou, , and M. Orlic, 2001: A case study of bora-driven flow and density changes on the Adriatic shelf (January 1987). Contin. Shelf Res., 21 , 17511783.

    • Search Google Scholar
    • Export Citation
  • Burk, S. D., , T. Haack, , and R. M. Samelson, 1998: Mesoscale simulation of supercritical, subcritical and transcritical flow along coastal topography. J. Atmos. Sci., 56 , 27802795.

    • Search Google Scholar
    • Export Citation
  • Burk, S. D., , T. Haack, , L. T. Rogers, , and L. J. Wagner, 2003: Island wake dynamics and wake influence on the evaporation duct and radar propagation. J. Appl. Meteor., 42 , 349367.

    • Search Google Scholar
    • Export Citation
  • Chan, J. C. L., , Y. Duan, , and L. K. Shay, 2001: Tropical cyclone intensity change from a simple ocean–atmosphere coupled model. J. Atmos. Sci., 58 , 154172.

    • Search Google Scholar
    • Export Citation
  • Chen, C., , R. C. Beardsley, , S. Hu, , Q. Xu, , and H. Lin, 2005: Using MM5 to hindcast the ocean surface forcing fields over the Gulf of Maine and Georges Bank region. J. Atmos. Oceanic Technol., 22 , 131145.

    • Search Google Scholar
    • Export Citation
  • Dorman, C. E., 1985: Hydraulic control of the northern California marine layer. Eos, Trans. Amer. Geophys. Union, 66 , 914.

  • Dorman, C. E., , and C. D. Winant, 1995: Buoy observations of the atmosphere along the west coast of the United States, 1981–1990. J. Geophys. Res., 100 , 1602916044.

    • Search Google Scholar
    • Export Citation
  • Dorman, C. E., , and C. D. Winant, 2000: The marine layer in and around the Santa Barbara Channel. Mon. Wea. Rev., 128 , 261282.

  • Dorman, C. E., , L. Armi, , J. M. Bane, , and D. P. Rogers, 1998: Sea surface mixed layer during the 10–11 June 1994 California coastally trapped event. Mon. Wea. Rev., 126 , 600619.

    • Search Google Scholar
    • Export Citation
  • Dorman, C. E., , D. P. Rogers, , W. Nuss, , and W. T. Thompson, 1999: Adjustment of the summer marine boundary layer around Pt. Sur, California. Mon. Wea. Rev., 127 , 21432159.

    • Search Google Scholar
    • Export Citation
  • Dorman, C. E., , T. Holt, , D. P. Rogers, , and K. Edwards, 2000: Large-scale structure of the June–July 1996 marine boundary layer along California and Oregon. Mon. Wea. Rev., 128 , 16321652.

    • Search Google Scholar
    • Export Citation
  • Dudhia, J., 1993: A nonhydrostatic version of the Penn State/NCAR mesoscale model: Validation tests and simulations of an Atlantic cyclone and cold front. Mon. Wea. Rev., 121 , 14931513.

    • Search Google Scholar
    • Export Citation
  • Edwards, K. A., 2000: The marine atmospheric boundary layer during Coastal Waves 96. Ph.D. thesis, University of California, San Diego, 132 pp.

  • Edwards, K. A., , A. M. Rogerson, , C. D. Winant, , and D. P. Rogers, 2001: Adjustment of the marine atmospheric boundary layer to a coastal cape. J. Atmos. Sci., 58 , 15111528.

    • Search Google Scholar
    • Export Citation
  • Fett, R. W., , and S. D. Bury, 1981: Island barrier effects as observed by satellite and instrumented aircraft, and simulated by a numerical model. Mon. Wea. Rev., 109 , 15271541.

    • Search Google Scholar
    • Export Citation
  • Friehe, C. A., , R. C. Beardsley, , C. D. Winant, , and J. P. Dean, 1984: Intercomparison of aircraft and surface buoy meteorological data during CODE 1. J. Atmos. Oceanic Technol., 1 , 7986.

    • Search Google Scholar
    • Export Citation
  • Grell, G. A., , J. Dudhia, , and D. R. Stauffer, 1995: A description of the fifth-generation Penn State/NCAR Mesoscale Model (MM5). National Center for Atmospheric Research Tech. Note TN-398, 122 pp.

  • Haack, T., , S. D. Burk, , C. Dorman, , and D. Rogers, 2001: Supercritical flow interaction within the Cape Blanco–Cape Mendocino orographic complex. Mon. Wea. Rev., 129 , 688708.

    • Search Google Scholar
    • Export Citation
  • Halliwell, G. R., , and J. S. Allen, 1987: The large-scale coastal wind field along the west coast of North America, 1981–1982. J. Geophys. Res., 92 , 18611884.

    • Search Google Scholar
    • Export Citation
  • Johnson, E. R., , and G. G. Vilenski, 2004: Flow patterns and drag in near-critical flow over isolated orography. J. Atmos. Sci., 61 , 29092918.

    • Search Google Scholar
    • Export Citation
  • Jones, I. S. F., , and Y. Toba, 2001: Wind Stress over the Ocean. Cambridge University Press, 307 pp.

  • Kessler, R. C., , and S. G. Douglas, 1991: Numerical study of mesoscale eddy development over the Santa Barbara Channel. J. Appl. Meteor., 30 , 633651.

    • Search Google Scholar
    • Export Citation
  • Koračin, D., , and C. E. Dorman, 2001: Marine atmospheric boundary layer divergence and clouds along California in June 1996. Mon. Wea. Rev., 129 , 20402056.

    • Search Google Scholar
    • Export Citation
  • Koračin, D., , C. E. Dorman, , and E. P. Dever, 2004: Coastal perturbations of marine layer winds, wind stress, and wind stress curl along California and Baja California in June 1999. J. Phys. Oceanogr., 34 , 11521173.

    • Search Google Scholar
    • Export Citation
  • Lagerloef, G. S. E., , and R. L. Bernstein, 1988: Empirical orthogonal function analysis of Advanced Very High Resolution Radiometer surface temperature patterns in Santa Barbara Channel. J. Geophys. Res., 93 , 68636873.

    • Search Google Scholar
    • Export Citation
  • Large, W. G., , and S. Pond, 1981: Open ocean momentum flux measurements in moderate to strong winds. J. Phys. Oceanogr., 11 , 324336.

  • Lopez, J. W., , and L. H. Kantha, 2000: A data-assimilative numerical model of the northern Indian Ocean. J. Atmos. Oceanic Technol., 17 , 15251540.

    • Search Google Scholar
    • Export Citation
  • Ly, L. N., , P. Luong, , J. D. Paduan, , and D. Koracin, 1999: Response of the Monterey Bay Region to wind forcing by an atmospheric model. Proc. Third Conf. on Coastal Atmospheric and Oceanic Prediction and Processes, New Orleans, LA, Amer. Meteor. Soc., 76–79.

  • Mass, C. F., , and M. D. Albright, 1989: Origin of the Catalina eddy. Mon. Wea. Rev., 117 , 24062436.

  • Mellor, G. L., , and T. Yamada, 1974: A hierarchy of turbulence closure models for planetary boundary layers. J. Atmos. Sci., 31 , 17911806.

    • Search Google Scholar
    • Export Citation
  • Pickett, M. H., , and J. D. Paduan, 2003: Ekman transport and pumping in the California Current based on the U.S. Navy’s high-resolution atmospheric model (COAMPS). J. Geophys. Res., 108 .3327, doi:10.1029/2003JC001902.

    • Search Google Scholar
    • Export Citation
  • Powers, J. G., , and M. T. Stoelinga, 2000: A coupled air–sea mesoscale model: Experiments in atmospheric sensitivity to marine roughness. Mon. Wea. Rev., 128 , 208228.

    • Search Google Scholar
    • Export Citation
  • Powers, J. G., , M. T. Stoelinga, , and W. S. Boyd, 1997: Distributed numerical weather prediction via satellite. Bull. Amer. Meteor. Soc., 78 , 27552770.

    • Search Google Scholar
    • Export Citation
  • Rogers, D., and Coauthors, 1998: Highlights of Coastal Waves 1996. Bull. Amer. Meteor. Soc., 79 , 13071326.

  • Rogerson, A. M., 1999: Transcritical flows in the coastal marine atmospheric boundary layer. J. Atmos. Sci., 56 , 27612779.

  • Samelson, R. M., 1992: Supercritical marine-layer flow along a smoothly varying coastline. J. Atmos. Sci., 49 , 15711584.

  • Shafran, P. C., , N. L. Seaman, , and G. A. Gayno, 2000: Evaluation of numerical predictions of boundary layer structure during the Lake Michigan Ozone Study. J. Appl. Meteor., 39 , 412426.

    • Search Google Scholar
    • Export Citation
  • Skyllingstad, E., , P. Barbour, , and C. E. Dorman, 2001: The dynamics of northwest summer winds over the Santa Barbara Channel. Mon. Wea. Rev., 129 , 10421061.

    • Search Google Scholar
    • Export Citation
  • Thompson, W. T., , S. D. Burk, , and J. Rosenthal, 1997: An investigation of the Catalina eddy. Mon. Wea. Rev., 125 , 11351146.

  • Winant, C. D., , C. E. Dorman, , C. A. Friehe, , and R. C. Beardsley, 1988: The marine layer off northern California: An example of supercritical channel flow. J. Atmos. Sci., 45 , 35883605.

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

June 1999 MM5 7-m wind speed and direction. This is the geographical coverage of the 9-km MM5 grid domain. The cause of the wind speed structure around Point Conception is the focus of this article.

Citation: Monthly Weather Review 136, 8; 10.1175/2007MWR2336.1

Fig. 2.
Fig. 2.

(a) SSM/I satellite example of strong winds about Point Conception on 9 Jun 1999. (b) SSM/I satellite example of moderate winds about Point Conception on 19 Jun 1999. (c) SSM/I satellite example of weak winds about Point Conception on 23 Jun 1999.

Citation: Monthly Weather Review 136, 8; 10.1175/2007MWR2336.1

Fig. 3.
Fig. 3.

500-hPa analysis at 1200 UTC for a (a) strong wind case on 9 Jun, (b) moderate wind case on 19 Jun, and (c) weak wind case on 22 Jun. Source is NCEP reanalysis provided by the NOAA/OAR/ESRL PSD (information online at http://www.cdc. noaa.gov/).

Citation: Monthly Weather Review 136, 8; 10.1175/2007MWR2336.1

Fig. 4.
Fig. 4.

Map of the 1-km MM5 grid domain and topography, with indicated hourly surface locations and geographical points. Circles are NDBC buoys, Xs are platforms, squares are land stations. NOJO is in a low pass on the Santa Ynez Mountain ridge.

Citation: Monthly Weather Review 136, 8; 10.1175/2007MWR2336.1

Fig. 5.
Fig. 5.

Buoy wind speeds at selected (top) east–west stations along the Santa Barbara Channel and (bottom) north–south stations along the western edge of the Southern California Bight. Area wind strengths for each day, posted below the lower frame, are noted in Table 2.

Citation: Monthly Weather Review 136, 8; 10.1175/2007MWR2336.1

Fig. 6.
Fig. 6.

(a) June 1999 B11-measured winds and nearest-point MM5-simulated winds for domain 3 (1-km resolution). (b) June 1999 B54-measured winds and nearest-point MM5-simulated winds for domain 3 (1-km resolution).

Citation: Monthly Weather Review 136, 8; 10.1175/2007MWR2336.1

Fig. 7.
Fig. 7.

Diagram of marine layer hydraulic response to Point Conception.

Citation: Monthly Weather Review 136, 8; 10.1175/2007MWR2336.1

Fig. 8.
Fig. 8.

Wind speed decreases cross coast in a compression bulge on the north side of the Point Conception topographic complex. Average June 1999 wind speeds are plotted relative to distance from the coast at PURI. Circles are measured values and the bar is one standard deviation wide. The PSAL coastal station has nearly identical values and distance as PURI, but is left off for clarity; “X” is nearest MM5 domain 3 value. Station designations are in Table 3.

Citation: Monthly Weather Review 136, 8; 10.1175/2007MWR2336.1

Fig. 9.
Fig. 9.

MM5-simulated sea surface winds for 0000 UTC 10 Jun, under strong winds. This is the geographical coverage of the 3-km MM5 grid domain. Every fourth wind vector is shown for clarity.

Citation: Monthly Weather Review 136, 8; 10.1175/2007MWR2336.1

Fig. 10.
Fig. 10.

MM5-simulated sea surface winds for 0000 UTC 20 Jun, under moderate winds.

Citation: Monthly Weather Review 136, 8; 10.1175/2007MWR2336.1

Fig. 11.
Fig. 11.

Locations of the profiler and sounding stations noted in Table 6. The Vandenberg balloon sounding station on the north side of Point Conception has two locations, usually one for the morning and one for the evening sounding. The San Diego balloon sounding and the Miramar radar profiler appear as a single symbol, as they are at the same site. The locations of buoys used in Fig. 5 are also shown.

Citation: Monthly Weather Review 136, 8; 10.1175/2007MWR2336.1

Fig. 12.
Fig. 12.

MM5-simulated domain 3 vertical section, extending east–west across the coast near PURI, valid at 0000 UTC 10 Jun. This profile extends through a compression bulge. The alongshore winds for the compression bulge are plotted in Fig. 4. The wind speeds decrease and the marine layer depth increases in the last 30 km of the coast.

Citation: Monthly Weather Review 136, 8; 10.1175/2007MWR2336.1

Fig. 13.
Fig. 13.

MM5-simulated domain 3 north–south vertical section across the western mouth of the Santa Barbara Channel over B54 at 00 UTC 10 Jun. Santa Rosa Island is on the left and the Santa Ynez Mountains are on the right. The leeside mountain flow is limited to the Santa Ynez Mountain upper slope. The main wind speed maximum in the center of the channel near the sea surface is an expansion fan. The island on the south side significantly extends the high-speed wind zone to the south.

Citation: Monthly Weather Review 136, 8; 10.1175/2007MWR2336.1

Fig. 14.
Fig. 14.

MM5-simulated domain 3 north–south vertical section across the mid–Santa Barbara Channel at 0000 UTC 10 Jun. This left end of this section extends between the two large islands of Santa Rosa and Santa Cruz and is west of B53. The Santa Barbara coast is on the right.

Citation: Monthly Weather Review 136, 8; 10.1175/2007MWR2336.1

Fig. 15.
Fig. 15.

Sea level pressure analysis during a strong wind case on 0000 UTC 10 Jun 1999. There is a well-supported local pressure minimum over the Santa Barbara Channel. This feature and a weak ridge of higher pressure on the east end of the Santa Barbara Channel are found during almost all strong and moderate wind speed conditions in summer. The weak pressure maximum over the coast north of Point Conception is due to the deepened marine layer in a compression bulge. The number posted is the observed sea level pressure minus 1000 hPa.

Citation: Monthly Weather Review 136, 8; 10.1175/2007MWR2336.1

Fig. 16.
Fig. 16.

Pressure differences for selected buoys (top) along and (bottom) across the Santa Barbara Channel. The pressures mostly increase to the east of B54, are against the eastbound flow over the channel, and are generally unrelated to the measured winds at B54 and B53 (Fig. 5).

Citation: Monthly Weather Review 136, 8; 10.1175/2007MWR2336.1

Fig. 17.
Fig. 17.

MM5-simulated, 3-km grid domain, 1500-m level valid at 0000 UTC 10 Jun 1999 during a strong wind case. Modest northerly winds cross the Santa Ynez Mountains.

Citation: Monthly Weather Review 136, 8; 10.1175/2007MWR2336.1

Fig. 18.
Fig. 18.

MM5-simulated, 3-km grid domain, 1500-m level valid at 0000 UTC 20 Jun 1999 during a moderate wind case. Weak northerly winds cross the Santa Ynez Mountains.

Citation: Monthly Weather Review 136, 8; 10.1175/2007MWR2336.1

Fig. 19.
Fig. 19.

Surface winds at stations along the Santa Barbara Channel coast. The strong offshore winds unique to (top) GAVW are due to its location at the Gaviota Canyon mouth. More typical are the weak and mostly onshore winds experienced by the better-exposed surface stations such as GAVE (only 2 km to the east of GAVW) and the coastal plane stations of ECAP, WCAM, and SBA, as well as other mountain slope stations that are not shown.

Citation: Monthly Weather Review 136, 8; 10.1175/2007MWR2336.1

Fig. 20.
Fig. 20.

(top) Pressure differences across the western Santa Ynez Mountains between B11 and SBA along a path that approximates the direction of the approaching marine layer winds. (middle) Along this approximate path are NOJO, the Santa Ynez ridge crest station, and the lee Santa Ynez foothill station GOLE. (bottom) B54 shows that the high-speed winds in the Santa Barbara Channel are unrelated to the cross-mountain pressure gradient or the leeslope winds. Note that the B54 winds are rotated to the mean wind direction of 306° in order to facilitate viewing.

Citation: Monthly Weather Review 136, 8; 10.1175/2007MWR2336.1

Fig. 21.
Fig. 21.

Vertical soundings around the Santa Ynez Mountains for a strong wind case (1200 UTC 9 Jun–0000 UTC 20 Jun) and a moderate wind case (1200 UTC 19 Jun–0000 UTC 20 Jun). VBG is off the western end of the Santa Ynez Mountains and GOL is on the coastal plane in the immediate lee, while LAX is to the southeast. The vectors point downwind. All cross-mountain ridge flow of any consequence is above 700 m and occurs in the afternoon.

Citation: Monthly Weather Review 136, 8; 10.1175/2007MWR2336.1

Fig. 22.
Fig. 22.

June 1999 satellite-observed sea surface temperature (found online at http://coastwatch.pfel.noaa.gov/). The general temperature pattern within 200 km of Point Conception is unrelated to the wind maxima (Fig. 1).

Citation: Monthly Weather Review 136, 8; 10.1175/2007MWR2336.1

Table 1.

Hypothesized factors responsible for wind structure.

Table 1.
Table 2.

Synoptic variations around three stations and southern California for June 1999 (m s−1): w = weak around Point Conception, central California; all = wind max, north and central California past Point Conception; cen = central California and past Point Conception, starts well before Point Conception complex, highest winds, usually >10 m s−1; sm = small area, localized about Point Conception; Inc = incomplete SSM/I coverage. Asterisks are used to mark the strength of the wind speed maximum in the lee of Point Conception.

Table 2.
Table 3.

Specifics of the meteorological stations. NWS: National Weather Service, SIO: Scripps Institution of Oceanography, SBAPCD: Santa Barbara Air Pollution Control District, and VAPCD: Ventura Air Pollution Control District.

Table 3.
Table 4.

Statistical comparison between station measurements and MM5 simulations for the 3-km grid (m s−1). No.: number of observations, Corr: correlation coefficient, ME: mean error bias, MAE: mean absolute error, RMSE: population root-mean-square error, and RMSVE: root-mean-square vector error.

Table 4.
Table 5.

EOF modes for measured stations and MM5 simulations. See text for explanation of station groups.

Table 5.
Table 6.

Southern California coastal profiler and sounding station characteristics; A: acoustical, B: balloon, and R: radar profiler.

Table 6.
Table 7.

Air temperature inversion base height (m above sea level). SFC is surface-based inversion (station elevation in parentheses).

Table 7.
Save