link to main US Geological Survey website
125 Years of Science for America - 1879 to 2004
link to Western Coastal and Marine Geology website

CHAPTER 6. SEA BREEZE

Peter Hamilton

Skip Navigation Links


| 6.1. Introduction | 6.2. Diurnal Period Current and Temperature Fluctuations | 6.3. Variability of the Cross-Shelf Sea Breeze Circulations | 6.4. References |

Home

Chapter 1. Introduction and Background

Chapter 2. Methodology

Chapter 3. Surfzone Bacteria Patterns

Chapter 4. Subtidal Circulation Pathways

Chapter 5. Newport Canyon Transport Pathway

Chapter 6. Sea Breeze

Chapter 7. Tidal Transport Pathways

Chapter 8. Sediment Resuspension and Transport near the OCSD Outfall

Chapter 9. Nearshore Circulation and Transport Pathways

Chapter 10. Spatial and Temporal Patterns of Plume Tracers

Chapter 11. Temporal and Spatial Patterns for Surfzone Bacteria before and after Disinfection of the OCSD Outfall

Acknowledgements

6.1. Introduction

In the analysis of the Phase II measurements (Hamilton et al., 2001), large-magnitude rotary currents were observed on the outer shelf through the upper 30 m of the water column. The signal period was exactly 24 hours, and because the surface tide at this period (S1) is very small and the 24-hour period is longer than the local inertial period (21.7 hours), the motions were attributed to direct forcing by the sea breeze. Freely propagating internal waves only exist for periods shorter than the inertial period, 2π/f, where f is the Coriolis parameter. The sea breeze in the southern California Bight is a persistent feature of the coastal atmosphere throughout the year, with the strongest winds occurring in the summer. Winds from the meteorological buoy (HB07) at the shelf break and two local airports (Orange County (John Wayne) and Long Beach) are given in Figure 6-1 for July and August. John Wayne Airport winds are from a site about 9 km from the shoreline, directly inshore from the mouth of the Santa Ana River. The Long Beach site is about 30 km WNW of John Wayne Airport, and about 4.5 km from the shoreline, with an east-west direction in this part of San Pedro Bay. Station locations for instruments are shown in Figure 6-2, with additional stations shown in Figure 6-3.

Sea breezes normally arise from differential heating between the land and coastal sea, and the landward wind, which is at maximum in the late afternoon, is perpendicular to the coastline and drives a density current that moves cool moist marine air over the land (Simpson, 1995). The wind components at John Wayne Airport clearly show the dominance of the cross-shoreline flows with maximum onshore winds in the afternoon (Figure 6-1). (Note that the time zone of the series is GMT so 7 hours need to be subtracted to get local (PDT) time). Wind measurements at Plant #2 close to the beach were not reliable, because of sensor problems during this study. However, wind measurements at Plant #2 in the summer of 2002 show that the daily beach winds are primarily directed across the shoreline, as afternoon beach goers may verify (Largier, personal communication). The cross-shore wind component at the buoy (HB07) is similar to that at John Wayne Airport, except with diminished magnitudes. The alongshore wind, however, is the larger component, and the maximum downcoast winds generally occur 1 or 2 hours later than the onshore component peak. Beardsley et al. (1987) discuss the occurrence of daily alongshore wind fluctuations off the northern California coast, and show that they are not caused by normal sea breezes but rather by a daily modulation of the marine layer inversion, partly caused by the barrier of the coastal mountains. The maximum downcoast daily winds occur in mid-afternoon off northern California, and the cross-shore winds are either very weak or not present. This is clearly not the case off Huntington Beach because of the late afternoon maximum of the strong sea-breeze system that is unimpeded by the low relief of the Los Angeles coastal region. Therefore, the dynamics of the sea breeze that produces strong daily alongshore winds on the outer shelf are not clear at present. It is speculated that modulations of the height of the marine layer by the sea-breeze system may be responsible. The coincidence of the daily atmospheric pressure minimum with the maximum alongshore wind could support this.

The wind record at Long Beach Airport shows both cross-shore and alongshore components. The alongshore component is in phase with the alongshore component at the buoy (Figure 6-1). However, the Long Beach station is situated where the coastline curves and it may be influenced by the topography of the Palos Verde coastal hills just to the north. It does indicate that fluctuating daily alongshore winds may occur over most of the San Pedro shelf.

The regular character of the daily sea-breeze winds is evident from Figure 6-1. There are few studies of the coastal ocean response to daily period winds. An exception is Rosenfeld (1988), who described the surface-intensified daily currents on the northern California shelf. The distribution of current amplitudes in the surface mixed layer, with depth and distance offshore, was attributed to direct forcing by the daily alongshore winds. A difference with the San Pedro currents, described subsequently, is that the northern California measurements showed only small phase changes with depth, whereas on the San Pedro shelf, the phase changes are large (Hamilton et al., 2001). This has important implications for cross-shelf transports in the upper layers.

Lerczak et al. (2001) discuss the generation of diurnal period internal waves off Mission Beach, California. They attribute the observed daily current fluctuations to propagating internal waves forced by the sea breeze. Their observations have similarities to the San Pedro currents in that the current vectors rotate clockwise and decay with depth. However, the Mission Beach diurnal currents were more intermittent than for San Pedro, and extended to the bottom in water depths of 70 m or more. San Pedro diurnal currents are generally negligible below 30 m. The arguments for the existence of propagating internal waves require that the effective Coriolis parameter be decreased by anticyclonic vorticity of the low-frequency current field so that the effective inertial period is longer than 24 hours. The effective local inertial frequency, fe, is given by Mooers, 1975.


where z is the relative vorticity of the flow. In coastal regions, z is usually approximated by dV/dx, the cross-shelf gradient of the alongshore component of the current. Off Mission Beach changes in fe account for the intermittency of the observed wave field as the background currents change (Lerczak et al., 2001). Mission Beach is further south than San Pedro, and, thus, relatively small decreases in effective f, (~8%) allow the propagation of internal waves.

The ratio z/f was calculated by least-square fitting a plane to the east and north gradients of the velocity components, at a depth of 5 m, using multiple moorings. The method is similar to that used by Chereskin et al. (2000) for a spatial array of moored instruments. For the outer and inner shelf, moorings HB05, HB06, HB07, HB09, HB10 and HB11, and HB03, HB05, HB09, and HBNC, respectively, were used. The results are very similar to just estimating dV/dx between moorings HB05 and HB07, and HBNC and HB05, respectively. If the ratio z/f is less than -0.18, then freely propagating 24-hour-period internal waves are dynamically possible. Above this value, the waves are evanescent. The resulting time series are given in Figure 6-4, along with amplitudes of the diurnal period currents and wind, at HB06, HB03, and the HB07 buoy, respectively. The diurnal amplitudes are calculated by complex demodulation with a 4-day filter and 2-day running means removed. On the inner shelf, apart from a couple of events, fe is greater than 1 cpd, yet the diurnal current amplitudes are quite large throughout the summer until the beginning of October. Wind amplitudes also remain fairly constant separated by short calm periods. On the outer shelf, fe falls below 1 cpd in July and in the second half of September. The diurnal current amplitudes at HB06 show some relative increases during these periods and the event between July 10 and 19 corresponds to a period of calm winds. Thus, there is some evidence that diurnal internal waves could exist on occasion and perhaps enhance the directly forced daily current fluctuations. However, for the majority of the summer and almost always on the inner shelf, fe is greater than 1 cpd, yet vigorous diurnal current fluctuations are persistent. Therefore, the generation and maintenance of surface-layer diurnal currents for the San Pedro shelf seems to be more complex than by propagation of internal waves, as is the case further south where the continental shelf is narrow (Lerczak et al., 2001).

6.2. Diurnal Period Current and Temperature Fluctuations

The depth profiles of current velocity for each ADCP mooring were analyzed for the diurnal frequency band using EOFs. The longest common period of 84 days that included the near-shore moorings was used for the spectra. The profiles of amplitude and phase for the main transect moorings HB07, HB05, HB03, and HBN2 are given in Figure 6-5. At the outer moorings, the U and V amplitudes are about equal and V leads U by approximately 90°, which indicates that the current vectors rotate clockwise through 360° every 24 hours. The amplitude profiles show two maxima, separated by a minimum. The first maximum is at or near the surface and the second between 15 and 20 m, with the minimum at about 5 to 10-m depth. Below about 30 m at HB07, the diurnal amplitudes become very small. The other noteworthy feature of the depth profiles is the 180° phase difference between the upper and lower maxima. The upper and lower amplitude maxima and the 180° phase shift also apply to the cross-shelf (U) component at the inner shelf moorings HB03 and HBN2. The depth change of phase for the U components is particularly rapid at the 5- to 10-m depth of the amplitude minima for HB05, HB03, and HBN2. The upper and lower maxima also decrease in magnitude from the shelf break to the near shore, as expected, because there is no volume flux through the shoreline. The implications of these results indicate that when the surface sea-breeze-driven current is towards shore, there is an offshore flow at about 10- to 20-m depth, which only extends down to about 30 m, and vice versa. This implies, through continuity, that there is a daily cycle of up and down vertical velocities, which have a maximum at between 5 and 10 m, and occur across the shelf, but are probably most intense on the inner shelf. Thus, the sea-breeze-forced flows are a mechanism where material at about 20- to 30-m depth could be transported across the shelf, and upwelled into the surface layer, and vice versa.

Cross-shelf fluctuations of velocity are the dominant influence on the temperature field, and, therefore, it is expected that the observed temperature fluctuations in the upper 20 m and nearshore will be predominantly diurnal. Figure 6-6 shows a short period of the 5-, 10-. and 15-m temperature records from the main transect, along with winds and sea level. At the nearshore moorings, HBN2 and HB03, the temperature data have a clear daily signal with the coldest water occurring when winds are slack in the early hours of the morning. Moreover, the largest amplitude fluctuations occur at the 5-m level corresponding to the minimum in cross-shelf velocity amplitudes at these sites. Similarly, at HB05, the largest amplitude of the temperature fluctuations is at 10 m, which is also the depth of the minimum cross-shelf velocity amplitude. At the latter station, there is a little more evidence of semidiurnal fluctuations, but the majority of the record still has a diurnal character. How much the daily heating and cooling cycle influences these temperature records is not yet clear, but because the maximum amplitude fluctuations are subsurface rather than at the surface points to a strong influence from cross-shelf advection. The air temperature record at the buoy (Figure 6-1) shows relatively small 1 to 2°C day-night fluctuations, and generally the subsurface fluctuations exceed this range and are unlikely to be accounted for by surface-heat fluxes. At the shelf-break mooring, HB07, the 15-m level shows more of a semidiurnal tidal character. This level, at this time, is in the thermocline and the fluctuations seem to be more in character with the internal M2 tide. This is analyzed in Chapter 7. The 5-m level has weak fluctuations, and this reflects the level temperature surfaces in the upper water column at the shelf break (Figure 4-16) and the lack of sea-breeze-forced vertical velocities. Vertical velocity fluctuations are only likely to be important where the sea-breeze-forced currents interact with the seabed in shallower water.

The depth distribution of diurnal period amplitudes and phases of the U component indicate that, at any instant, the onshore-offshore flows cancel out. Thus, the diurnal depth-mean cross-shelf velocities are very small (<1 cm/s) and can be accounted for by the diurnal astronomical tide. This is also true for the along-shelf (V) component at the shelf break, but on the inner shelf, the top to bottom phase difference for V is small (~30 to 60°) and linear with depth (Figure 6-5). The V amplitudes also decrease linearly with depth, which is not the behavior expected of purely tidal flow at diurnal periods. Therefore, it appears that the alongshore diurnal velocities are also influenced by the wind, even though the sea breeze is predominantly cross-shore at the beach. It is, therefore, difficult to separate out the diurnal tide from the sea-breeze currents in the nearshore.

An EOF analysis for the diurnal frequency band was used to characterize the surface current patterns and their relationship to the wind (Figure 6-7). Because phase can change quite rapidly with depth, only the current records that were closest to the surface were used, and because wind and currents have different units, the cross spectra were normalized. The surface currents and winds are almost in phase, with the currents lagging by less than one hour. Thus, when the sea breeze is strongly downcoast at the buoy and simultaneously landward at John Wayne Airport (Figure 6-1), the surface currents across the shelf are also downcoast. Maximum offshore flow at the surface then occurs about 6 hours later (~2300 PDT), and maximum onshore flow about 18 hours later (~1100 PDT). The situation is reversed for the lower layer (10- to 20-m depth), where maximum on- and offshore currents occur around 2300 and 1100 (PDT), respectively. The occurrence of nearshore cold water in the early hours of the morning is consistent with maximum subsurface onshore flows occurring a few hours earlier with upwelling of colder water into the near surface. The close matching of the current phase at HBN2 and HBN3 with the offshore surface currents, where diurnal astronomical tidal currents are weak, supports the notion that these nearshore, along-shelf diurnal currents are primarily wind driven. The current hodographs are roughly circular on the outer shelf, becoming more elliptical with the major axis directed along-shelf in the nearshore, as was pointed out previously in the discussion of the velocity components. The other important point to be made is that these are fluctuations with substantial magnitudes, even in shallow water. They have comparable magnitudes at the surface to the subtidal flows (Figure 4-6).


TOP OF PAGE



6.3. Variability of the Cross-shelf Sea Breeze Circulations

The cross-shelf structure of the sea-breeze-forced currents has been analyzed using the longest possible time series (July 20 to October 12). In Chapter 4, it was shown that the subtidal circulation changes the structure of the cross-shelf temperature fields and strongly influences the presence of colder water in the near shore. Since the diurnal fluctuations are modulations on the longer time-scale changes, the analysis of the cross-shelf diurnal period transport uses the same four periods that were defined by the low-frequency current modes. The mean temperature sections for the four periods are given in Figure 4-16; these fields are subtracted from the temperature records for each period before input into the diurnal frequency domain EOF analysis. The onshore-offshore displacements of the 14, 16 and 18°C isotherms for the four periods are given in Figures 6-8 and 6-9. In these plots, if the isotherm moves further offshore than 3 km from HB01 (i.e., HB05), then this water is below the depth where the diurnal currents are significant. In period 1, the coldest water nearshore (isotherms closest to shore) occurs at neap tide, when tidal period oscillations of the isotherms are very small, despite strong diurnal winds. This appears to be coincidence since it does not hold in the following periods. It is possible that strong uptilting of the isotherms because of large downcoast subtidal flows (Figure 4-14), generates a surface front that inhibits cross-shelf, upper-layer fluxes at these times. When oscillations appear, they are clearly diurnal, despite the spring tides, with the coldest water occurring at the time of minimum sea-breeze winds. In period 2, the upper-layer mean temperature field has fairly level isotherms and this is reflected in the spreading of the depths and positions of the isotherms in Figure 6-8. Again the isotherm fluctuations are dominated by diurnal fluctuations with some semidiurnal influence on the 14°C isotherm. The changing mixtures of semidiurnal and diurnal influences with depth can cause interesting behavior. For example, between July 30 and August 2, the 14 and 16°C isotherm positions generally move in opposite directions (Figure 6-8). The oscillations have similar characteristics in period 3 (Figure 6-9), except that, as the 14°C isotherm moves deeper, the semidiurnal fluctuations become more prominent. This period has times when the sea breeze reduces in strength (August 18-21, and August 28-September 1). Coincidently, the amplitudes of the diurnal fluctuations of temperature and cross-isobath current are reduced and the semidiurnal fluctuations become more prominent. The fourth period occurs while the subtidal isotherms are quite strongly tilted up towards the coast, though the thermocline is much deeper than in period 1. However, similar to parts of period 1, period 4 fluctuations are relatively weak, though apparently not related to the neap tide.

Frequency-domain EOF analyses were performed for the diurnal period band using all the temperature records for each of the four periods. A subset of the velocity records, using the same positions as the subtidal analyses (Chapter 4), but excluding any records below 60 m on HB08, were used for separate EOF analyses. The subset of the results for the main transect are shown in Figures 6-10, 6-11, 6-12, and 6-13 where the first modes from the temperature and the cross-shelf (U) velocity component are displayed as contoured cross-sections. Note that the U and V component velocities were part of the same velocity analysis, but the V components are not shown. The first modes explain between 67 and 85% of the total variance of all the records included in these analyses. The results for period 1 (Figure 6-10) show maximum amplitude temperature fluctuations occur on the inner shelf centered on about 10-m depth. The amplitudes become very small below 30 m for both temperature and velocity and the phase contours are horizontal. The U-component amplitudes show the double maxima at all stations, with the upper concentrated close to the surface and the lower at about 15 m on the outer shelf and becoming shallower as the water depth shoals. Cross-shelf velocity amplitudes decrease towards the shore. Phase differences between the near surface and the bottom, or 25 m, whichever is less, are approximately 180°, showing that the flows at the depths of the maxima are in opposition. This cross-shelf distribution of diurnal velocity components explains the large temperature amplitudes on the inner shelf because of enhanced vertical velocities required by continuity. Horizontal advection of temperature could also contribute to the large nearshore amplitudes because of the cross-shelf temperature gradients in the upper layer, established by the background flows. It is noted that in Figure 6-10, temperature measurements at HB01 are from surface and bottom only so it is possible that larger mid-depth temperature amplitudes are present in shallower water than indicated by the contours. Vertical phase differences for temperature are not large over the depth range of significant cross-shore velocities, and this is also consistent with direct pumping of the isotherms. The same general patterns hold for the other three periods; however, there are some significant differences in magnitudes and the details of the cross-shelf amplitudes and phases.

In period 2 (Figure 6-11), the amplitudes are less than in period 1 and the subsurface U-component maximum is not as well defined in the offshore stations. In this case, with more nearshore temperature measurements available, the maximum in temperature amplitudes extends to at least the 10-m isobath. In period 3 (Figure 6-12), the subsurface maximum in the U components is reestablished, but at deeper depths than in period 1. Temperature amplitudes are small; however, in this period the upper layers are only weakly stratified (Figure 4-16). The higher temperature amplitudes extend further offshore than in the earlier periods and this appears to be a consequence of the deeper penetration depth of the sea-breeze-forced cross-shelf flows. The same trend is continued in period 4 (Figure 6-13), where the subsurface maximum in cross-shelf velocity has moved deeper and the upper layer stratification is further reduced in the offshore. The temperature amplitude maximum has also moved further offshore, indicating that the region where vertical velocities are important is deeper, and has been displaced offshore to between HB05 and HB03. The increasing depth of the subsurface maximum at the outer shelf stations through periods 2, 3, and 4 indicate that upper-layer stratification may have an important role in limiting the penetration depth of the sea-breeze-forced flows. The depth of the subsurface return flows, in turn, determines the region of the inner shelf where vertical transport becomes important. These differences are illustrated in Figure 6-14 where the U- and V- component amplitudes for HB06 are plotted as profiles. The deepening trend, for periods 2, 3 and 4, is clear for both the U and V component. For period 1, the U-component subsurface maximum is slightly shallower than period 2; however, the V-component maximum is as deep as period 4. Surface maxima show a factor two range with period 2 having the smallest value and period 1 the largest. However, at moorings HB03 and HBN2, period 2 amplitudes are the largest for the four periods (Figure 6-14). This may be a consequence of period 2 having more level isotherms than the other periods so that there is less inhibition by surface, cross-shelf density gradients. HB03 and HBN2 cross-shelf components show the surface and bottom maxima, and the HBN2 along-shelf component linearly decreases with depth for all four periods, similar to the individual mooring analyses given in the previous section. Therefore, the cross-shelf response of the flow field to sea-breeze winds appears to be quite complex, even though the basic flow patterns are quite consistent over the summer. The continuity arguments for enhanced vertical transport on the inner shelf in water depths of 10 to 25 m do not depend on the existence of stratification, and, therefore, could be considered independent of any other internal wave activity.

To examine the coherence between the wind and the surface currents, EOF analyses of these records were performed for each of the four periods. The results are given in Figures 6-15 and 6-16. The relative phases between the wind and current fluctuations are approximately the same for all periods. Comparing the first two periods, the wind fluctuations at the buoy have very similar alongshore magnitudes, with period 1 having slightly larger cross-shore magnitudes. There is, however, quite a substantial difference in the current hodographs over most of the shelf, except for the nearshore at HBN2 and HBN3. A difference between periods 1 and 2 is that the anticyclonic vorticity of the low-frequency flows allows freely propagating internal waves on the outer shelf for substantial parts of period 1, but not in period 2 (Figure 6-4). The presence of propagating waves could enhance the response to the wind and allow the energy to penetrate deeper in the water column as is observed for period 1. Period 3 has less wind energy than period 2, but the current fluctuations are similar or slightly larger. In period 4 (Figure 6-16), the wind fluctuations are similar to period 2, and the current hodographs remain similar to period 3, except in the nearshore where they are again enhanced. Thus, despite relatively constant sea breezes, the response of the currents can have different magnitudes, depending on whether internal waves can propagate, and the stratification of the upper layers that can change the depth profile of the fluctuation amplitudes. It will probably require a model study to sort out the importance of the various influences on the sea-breeze circulation.

Though diurnal period fluctuations have been observed in coastal seas, this is probably the first study where the mechanisms and contributions to cross-shelf transport processes have been comprehensively described and quantified. The importance of sea-breeze-forced flows to nearshore transport processes has probably been underestimated for this region.

6.4 References

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: 1, A description of the marine boundary layer and atmospheric conditions over a northern California upwelling region. Journal of Geophysical Research, v. 92, p. 1467-1488.

Chereskin, T.K., M.Y. Morris, P.P. Niiler, P.M. Kosro, R.L. Smith, S.R. Ramp, C.A. Collins, and D.L. Musgrave, 2000. Spatial and temporal characteristics of the mesoscale circulation of the California Current for eddy-resolving moored and shipboard measurements. Journal of Geophysical Research, v. 105(C1), p. 1245

Hamilton, P., J.J. Singer, E. Waddell, and G. Robertson, 2001. Circulation processes on the San Pedro shelf. Proceeding: MTS 2001 Conference, November, 2001, Honolulu, Hawaii, 8 p.

Largier, John. Personal communication.

Lerczak, J.A., M.C. Hendershott, and C.D. Winant, 2001. Observations and modeling of coastal internal waves driven by a diurnal sea breeze. Journal of Geophysical Research, v. 106, p. 19715-19730.

Mooers, C.N.K., 1975. Several effects of a baroclinic current on the cross-stream propagation of inertial-internal waves. Geophysical Fluid Dynamics, v. 6, p. 245-275.

Rosenfeld, L.K., 1988. Diurnal period wind stress and current fluctuations over the continental shelf off northern California. Journal of Geophysical Research, v. 93, p. 2257-2276.

Simpson, J.E., 1995. Sea-breeze and local winds. Cambridge University Press, 234 p.

TOP OF PAGE


U.S. Department of the Interior, U.S. Geological Survey, Western Region Coastal and Marine Geology
URL of this page: http://pubs.usgs.gov/of/2004/1019/chap6.html
Maintained by: Mike Diggles
Created: September 15, 2004
Last modified: October 06, 2004 (lh)