ML071290387: Difference between revisions
StriderTol (talk | contribs) (Created page by program invented by StriderTol) |
StriderTol (talk | contribs) (StriderTol Bot change) |
||
Line 440: | Line 440: | ||
140 | 140 | ||
.*-Mt. Ninham 120 k White Plains 100 I 80 I-0.~ | .*-Mt. Ninham 120 k White Plains 100 I 80 I-0.~ | ||
60 I 40 k 20 I b | 60 I 40 k 20 I b | ||
0 0700 0900 1100 1300 1500 1700 1900 2100 2300 Local Hour Figure 13. Ground level 03 concentration levels at three stations north of New York City. (a) Locations of 03 monitors in southern New York State. (b) Time series of 03 concentrations. Note "spikes" as the SBF passes successively farther inland. Plots are based on hourly 03 observations from July 1995. Figure 13 was redrawn after Gaza [1998]. | 0 0700 0900 1100 1300 1500 1700 1900 2100 2300 Local Hour Figure 13. Ground level 03 concentration levels at three stations north of New York City. (a) Locations of 03 monitors in southern New York State. (b) Time series of 03 concentrations. Note "spikes" as the SBF passes successively farther inland. Plots are based on hourly 03 observations from July 1995. Figure 13 was redrawn after Gaza [1998]. |
Latest revision as of 16:24, 13 March 2020
ML071290387 | |
Person / Time | |
---|---|
Site: | Pilgrim |
Issue date: | 04/25/2007 |
From: | Palla R Office of Nuclear Reactor Regulation |
To: | Natasha Greene |
References | |
TAC MD3698 | |
Download: ML071290387 (35) | |
Text
I1Robert Palla - Additional Info/References Page 1 ii From: Robert Palla To: Natasha Greene Date: 04/25/2007 2:26:04 PM
Subject:
Additional Info/References see attached CC: jojones @sandia.gov
1;c:\temp\GWJ00001.TMP Page 1 !
I........... 1 . M P ag 1 Mail Envelope Properties (462F9D3C.A32: 18: 10412)
Subject:
Additional Info/References Creation Date 04/25/2007 2:26:04 PM From: Robert Palla Created By: RLP3 @nrc.gov Recipients Action Date & Time nrc.gov OWGWPO02.HQGWDO01 Delivered 04/25/2007 2:26:16 PM NAG (Natasha Greene) Opened 04/25/2007 2:33:25 PM sandia.gov Transferred 04/25/2007 2:26:38 PM jojones CC (joiones@sandia.gov)
Post Office Delivered Route OWGWPO02.HQGWDO01 04/25/2007 2:26:16 PM nrc.gov sandia.gov Files Size Date & Time MESSAGE 319 04/25/2007 2:26:04 PM Mail Mail Options Auto Delete: No Expiration Date: None Notify Recipients: Yes Priority: Standard ReplyRequested: No Return Notification: None Concealed
Subject:
No Security: Standard To Be Delivered: Immediate Status Tracking: Delivered & Opened
I Robert Palla. - Sea Breeze Journal Article Page 111 From: Mark Lilly To: mrll@nrc.gov; Robert Palla Date: 04/25/2007 9:03:19 AM
Subject:
Sea Breeze Journal Article Bob, I forgot to attach a paper written by Dr. Miller regarding the sea breeze phenomena in the northeast. Here it is.
Also, Dr. Miller suggested that the Boston area experiences about 60-70 sea breeze scenarios a year.
Though most occur during the months of May, June, and July his graduate student has seen some that occur in the winter months during very weak synoptic conditions. The 60-70 sea breeze occurrences DO NOT include synoptically driven onshore flow. The sea breeze occurrences are only driven by mesoscale flows in weak synoptic conditions, according to Dr. Miller.
Sorry for an extra email, just thought these were some interesting things I forgot to mention in the first email.
Mark Lilly
SRobedr Palla - Here is the article Page 1 From: Mark Lilly To: mrll @nrc.gov; Robert Palla Date: 04/25/2007,9:04:09 AM
Subject:
Here is the article I forgot to attach the article.
Mark
SEA BREEZE: STRUCTURE, FORECASTING, AND IMPACTS S. T. K. Miller,1 B. D. Keim, 2' 3 R. W. Talbot,' and H. Mao 1 Climate Change Research Center, Institute for the Study of Earth, Oceans and Space, University of New Hampshire, Durham, New Hampshire, USA Received 15 January2003; revised 9 June 2003; accepted 19 June 2003; published 16 September 2003.
[i] The sea breeze system (SBS) occurs at coastal loca- mechanisms, structure and related phenomena, life cy-tions throughout the world and consists of many spatially cle, forecasting, and, impacts on air quality.
and temporally nested phenomena. Cool marine air INDEX TERMS: 3329 Meteorology and Atmospheric Dynamics: Me-propagates inland when a cross-shore mesoscale (2-2000 soscale meteorology; 3307 Meteorology and Atmospheric Dynamics:
km) pressure gradient is created by daytime differential Boundary layer processes; 3322 Meteorology and Atmospheric Dy-heating. The circulation is also characterized by rising namics: Land/atmosphere interactions; 3339 Meteorology and Atmo-currents at the sea breeze front and diffuse sinking spheric Dynamics: Ocean/atmosphere interactions (0312, 4504); 3399 currents well out to sea and is usually closed by seaward Meteorology and Atmospheric Dynamics: General or miscellaneous; flow aloft. Coastal impacts include relief from oppressive KEYwoRDs: sea breeze hot weather, development of thunderstorms, and Citation: Miller, S. T. K., B. D. Keim, R. W. Talbot, and H. Mao, Sea changes in air quality. This paper provides a review of breeze: Structure, forecasting, and impacts, Rev. Geophys., 41(3), 1011, SBS research extending back 2500 years but focuses doi:10.1029/2003RG000124, 2003.
primarily on recent discoveries. We address SBS forcing
- 1. INTRODUCTION force (PGF) directed from sea to land, and a shallow layer of marine air moves inland in response. Some of
[2] The sea breeze is a local circulation that occurs at the components of the sea breeze system (SBS) (Figure 1) coastal locations throughout the world [Masselink and are the following:
Pattiaratchi,1998]. (Italicized terms are defined in the 1. Sea breeze circulation (SBC) is a vertically rotating glossary, after the main text.) It has been observed from mesoscale cell, with shoreward flow near Earth's surface, polar regions to the equator. The sea breeze can provide rising air currents inland, diffuse sinking currents several relief from oppressive hot weather, trigger thunder- kilometers out to sea, and (usually) seaward return flow storms, provide moisture for fog and may result in either near 900 hPa.
improved or reduced air quality near Earth's surface 2. Sea breeze gravity current (SBG) is the landward
[Barbato, 1975; Kozo, 1982; Hsu, 1988; Simpson, 1994; flow of cool, moist marine air in the lower horizontal Camberlinand Planchon, 1997; Silva Dias and Machado, arm of the SBC.
1997]. It is important to continue improving our under- 3. Sea breeze front (SBF) is the landward edge of the standing of the sea breeze because a large part of the SBG and the SBC, often associated with sharp changes human population lives in major cities in coastal areas. in temperature, moisture, and wind. Its approach may be
[3] The large-scale environment in which the sea marked by the development of fair-weather cumulus breeze occurs has been understood for thousands of clouds (Cu).
years, and the broad outlines of the sea breeze's behav- 4. Sea breeze head (SBH) is the raised head above ior have been understood for at least a hundred years. It and immediately behind the SBF, created by updrafts occurs under relatively cloud-free skies, when the sur- within both the continental and marine air masses. It is face of the land heats up more rapidly than the sea. The about twice as high as the following "feeder" flow behind thermal contrast creates a local-scale pressure gradient the SBF.
- 5. Kelvin-Helmholtz billows (KHBs) are waves that develop along the upper boundary of the SBG during 1
Also at Department of Earth Sciences, University of New periods of low static stability (midday).
Hampshire, Durham, New Hampshire, USA. 6. Convective internal boundary layer (CIBL) is an
'Also at Department of Geography, University of New unstable region within the marine air mass, appearing at Hampshire, Durham, New Hampshire, USA. the coast and growing in depth with distance inland, in 3
Now at Southern Regional Climate Center, Louisiana which low-level pollutants may become trapped and State University, Baton Rouge, Louisiana, USA. concentrated.
Copyright 2003 by the American Geophysical Union. Reviews of Geophysics, 41, 3 / 1011 2003 8755-1209/03/2003RG0001 24$15.00 doi:1 0.1 029/2003RGO001 24 0 1-1 0
1-2 o Miller et aL: SEA BREEZE 41, 3 / REVIEWS OF GEOPHYSICS that the prevailing wind in Athens is from the north or SBC
/C south, and the Etesian wind, a persistent, northerly or northeasterly wind, occurs after the summer solstice.
Aristotle attributed local-scale winds to the unequal BF SBH KHBs CIBL SBG distribution of Earth's moist and dry "exhalations" and variations in local topography [Lee, 1952, p. 165].
[8] Theophratus' De Ventis was written about 300 B.C., -30 years after Meteorologica, and discusses (1) the importance of the Sun in driving the wind and (2) a distinct sea breeze from the south or southwest that Thermal PGF occurred during the period of the Etesian wind. Accord-Figure 1. Sea breeze system (SBS). Details are discussed in ing to Theophratus, "at about the time of the Etesians, the text. winds arise counter to the north wind because of a circling back so that ships move in the opposite direc-tion. These winds are called reverse north winds" (as
[4] The details of the sea breeze's behavior along quoted by Coutant and Eichenlaub [1975, p. 29]). This specific coastlines can be surprising, even to experienced may be the first written reference to an onshore wind meteorologists. For example, while the sea breeze is opposing a synoptic (continental)-scale offshore wind.
initiated during the daytime, it may persist at any time Theophratus also noticed that "everywhere at noon the during the day or night. The leading edge of the sea winds die down because of the Sun's actions, and arise breeze may separate from the feeder flow behind the again with the late afternoon. It happens that the revers-front and move hundreds of kilometers inland as an ing wind [the seabreeze] blows against the land winds independent entity. and the Etesians rise at the same time again" [Coutant
[5] A great deal of new knowledge has been gained and Eichenlaub, 1975, p. 31]. This observation can be over the last 15 years, and an updated overview is explained as opposing forces reaching equilibrium at needed, tying together our improved understanding of about noontime, and Theophratus correctly attributed the sea breeze. This paper will fill that need, focusing on the developing equilibrium to the Sun's influence. Dur-sea breeze forcing mechanisms, structure, life cycle, ing midafternoon, when the land-sea temperature con-forecasting, and the ways in which the sea breeze im- trast reaches its maximum, the sea breeze moves inland pacts local air quality. The glossary contains key terms, as a southwesterly wind, while well offshore the north-and the notation list defines symbols used in this paper. easterly Etesian wind reaches its diurnal maximum
[Coutant and Eichenlaub, 1975].
[9] As far as we know, there is little documented to
- 2. ANCIENT HISTORY demonstrate major advances in the understanding of the sea breeze from about 300 B.C. to the seventeenth
[61 Given their dependence on the wind and tides, it century A.D. This gap in research is a subject best seems likely that early sailors and fishermen understood addressed by historians. Over the past several centuries, the diurnal cycle of the sea breeze long before natural there has been an increasing interest in the sea breeze, scientists began writing about it. Military personnel, which is documented in Jehn's [1973] bibliography of sea particularly those that fought battles at sea, would also breeze. research from the seventeenth century to 1972.
have been keenly interested in the sea breeze. Simpson He lists 16 references on the sea breeze written before
[1994] discusses the story of Athenian military leader 1800, 113 written between 1800 and 1899, and 407 writ-Themistocles, commander of the Greek forces in a naval ten between 1900 and 1972. Among Jehn's references battle with the Persians in 480 B.C. The battle took place are a brief article by Francis Bacon, written in 1664, and in a narrow channel between the island of Salamis and another by Immanuel Kant from 1755 [Jehn, 1973]. A the Greek mainland. Themistocles chose to begin his recent search reveals more than 500 new articles pub-battle with the Persians at the moment when the sea lished on the subject since 1990 alone.
breeze began flowing into the narrow channel. The low, solid design of the Greek ships gave them the advantage in the choppy waters caused by the sea breeze, and they 3. MODERN UNDERSTANDING defeated the Persians.
[7] The first scientist to write about the wind in gen- [to] Contemporary studies of the sea breeze system eral was Aristotle. In Meteorologica, written 150 years have been motivated by its influence on local wind after Themistocles' victory at Salamis, Aristotle stated velocity and air quality [e.g., Angell and Pack, 1965; Yu that "as a rule a considerable area may be expected to be and Wagner, 1970; Barbato, 1975; Neumann, 1977; Shair similarly affected, because neighboring places lie in a et al., 1982; Kalthoff et al., 2002] and convective activity similar relation to the Sun, unless they have some local [e.g., Pielke, 1985; Pielke et al., 1991; Rao and Fuelberg, peculiarity" (as quoted by Lee [1952, p. 169]). He noted 2000; Lericos et al., 2002], as well as by an interest in the
41, 3 / REVIEWS OF GEOPHYSICS Miller et a[.: SEA BREEZE o 1-3 Table 1. Summary of Subjects Discussed in Sea Breeze Literature General Subject Specific Topics Included Forcing the role of sound waves; development of the sea breeze circulation from the mesoscale pressure gradient; more complete physical models Structure prefrontal phenomena; sea breeze's response to the prevailing wind; sea breeze gravity current and its response to Kelvin-Helmholtz billows; sea breeze front and associated phenomena Life cycle sequence of development stages Interactions with external separate sea breeze systems; convective updrafts; urban heat islands; river breezes; meteorological phenomena temperature inversions; horizontal convective rolls; synoptic-scale cold fronts Forecasting occurrence of a sea breeze; wind speed and direction; extent of inland penetration SBS-related air quality impact pollutant transport mechanisms; convective internal boundary layers; sea breeze in mechanisms complex air pollution scenarios more esoteric aspects of its dynamics and structure [e.g., [14] The first is the "upward" theory [Tijm and van Smith, 1982; Kraus, 1992; Reible et al., 1993; Tijm and van Delden, 1999], wherein a seaward flow (sometimes called Delden, 1999]. Some have even studied SBS influence on return flow) develops aloft first, because of the vertical the coastal ocean [Schoellhamer, 1996; Gibbs, 2000] and expansion of the warmer air over land. Landward flow beach morphology [Masselink and Pattiaratchi,1998]. near Earth's surface then develops in response to the
[ii] These researchers have found that the SBS, often seaward flow aloft. The second is called the "sideways" thought to be easily understood, is actually a very com- theory [Simpson, 1994], wherein the low-level onshore plicated phenomenon, producing associated phenomena flow develops first in response to the horizontal expan-or nonlinear interactions on several scales, from the sion of the warmer air over land. Seaward (return) flow meso-3 scale (20-200 km) through the inertial subrange aloft then develops in response to the low-level onshore (where turbulent motions are approximately homoge- flow. The third is called the "mixed" theory [Godske et neous, isotropic, and inviscid) [Stull, 1988; Finkele et al., al., 1957], in which the warmer air over land expands 1995; Cenedese et al., 2000]. Some of the phenomena are both vertically and horizontally, causing the simulta-the raised head of the advancing sea breeze [e.g., Craig neous development of the surface sea breeze and the et al., 1945; Simpson, 1994]; interactions between the sea return current aloft [Tijm and van Delden, 1999].
breeze or sea breeze front and synoptic-scale or other [15] All three versions, in spite of some evidence in mesoscale circulations [e.g., Zhong and Takle, 1992; At- the synoptic surface observations to support them, ne-kins et al., 1995; Briimmer et al., 1995], as well as between glect the process of hydrostatic adjustment. This occurs multiple sea breeze fronts [Clarke, 1984]; the turbulent in the developing SBS and is accompanied by the cre-wake composed of Kelvin-Helmholtz billows seaward of ation of sound (compression) waves, which are gener-the sea breeze front and along the upper boundary of the ated over land when the air expands because of diabatic shallow marine layer [e.g., Chiba et al., 19991; and the heating and then propagate at 300 m s-1 in all directions CIBL that forms within the stable marine layer as it
[Tijm and van Delden, 1999]. Utilizing a nonhydrostatic moves over the hot land surface [e.g., Hsu, 1988].
computer model, Tijm and van Delden [1999] concluded
[12] In sections 3.1-3.6 each of the areas summarized that sound waves play a critical role in establishing the in Table 1 will be explored in detail.
mesoscale pressure gradient. Walsh [1974] concluded earlier that simulations of the sea breeze circulations 3.1. Forcing using nonhydrostatic models differ very little from those produced by hydrostatic models, but the Tijm and van 3.1.1. Role of Sound Waves Delden [1999] results indicate that the small difference
[131 The sea breeze is caused by the temperature between the two model schemes is important, because difference between the hot land and cool sea surfaces. the hydrostratic models do not incorporate sound waves As the difference increases during the day, a pressure and therefore do not include the correct mechanism for gradient is produced at low levels in the atmosphere, establishing the mesoscale pressure gradient.
initiating the sea breeze near Earth's surface [Simpson, [16] Within a few minutes, sound waves traveling up-1994]. The appearance of the mesoscale pressuregradient ward induce an increase in pressure through the entire force (PGF) that drives the SBS may seem relatively atmosphere above the heated land surface. Waves prop-simple to explain, but Tijrn and van Delden [1999] care- 'agating horizontally induce pressure falls over land, and fully examined three different variations found in the surface pressure increases over the sea. The resulting literature and concluded that all three are physically horizontal pressure gradient initiates the landward inconsistent. movement of the marine air mass associated with the
1-4
- Miller et al.: SEA BREEZE 41, 3 / REVIEWS OF GEOPHYSICS SBS. The surface pressure decrease over land is first 14. L 0 seen at the coast, but locations more than 1000 km inland experience a decline in pressure within an hour. P1 These results are supported by a careful analysis of the synoptic surface record in Europe and represent a more physically consistent version of the mixed theory [Tijm 7
7 7
" T I
and van Delden, 1999]. 7 V
7 7 FH 7 /
3.1.2. Sea Breeze Circulation (SBC) and the Bjerknes Circulation Theorem Ocean
[17] Mathematical models, when used in conjunction P0 with field observations, are a powerful means for explor- T1 ing physical mechanisms. The Bjerknes circulation the-orem is a relatively simple model that begins with the Figure 2. Sea breeze and the Bjerknes circulation theorem.
Land is on the left, ocean is on the right. The integration path presence of a cross-shore mesoscale PGF and repro-for equation (1) is indicated by arrows along the perimeter. T duces the SBC from an initially stationary atmosphere.
indicates the average temperature through the vertical col-Circulation is a scalar quantity that represents a macro- umns over the land or ocean surfaces. Dashed diagonal lines scopic measure of rotation over a finite area of fluid in are isopycnals, with denser air at lower right. P, and P1 are the two dimensions [Holton, 1992]. Mathematically, circula- pressure on the surface and at the top of the circulation cell, tion about a closed contour in a fluid is defined as the respectively. Redrawn from Holton [1992]. Reprinted with line integral about the contour of the component of the permission from Elsevier Science.
velocity vector that is locally tangent to the contour
[Holton, 1992]. The Bjerknes circulation theorem is given by DU Rln (pi) (T2 - T1)
DCý f dP Dt 2(H + L) (4)
Dt P-
-, (1)
Which is an expression for the mean acceleration of where DJDr indicates the material derivative in the fixed the wind over time, as a result of the SBC. Realistic values for the right-hand side of equation (4) are p 0 =
reference frame, Ca is circulation, P is pressure [Pa], and p is density [kg m- 3]. 1000 hPa, P, = 900 hPa, (T2 - T1) = 10 K, L = 20
[18] The application of the circulation theorem to the km, and H = 1 kin, which yields an acceleration of 7.2 SBC begins with equation (1) and substitutes p = PIRT X 10- 3 m S-2. Beginning at rest, after 1 hour1.157407e-5 days <br />2.777778e-4 hours <br />1.653439e-6 weeks <br />3.805e-7 months <br /> the mean (from the ideal gas law), where T is temperature [K] and wind speed around the perimeter is 25.9 m s-' or -50 R is the gas constant for dry air (287 J kg-' K-'). By knots [Holton, 1992].
integrating around the closed path beginning on the land [20] Wind speeds actually produced by the SBS are surface (Figure 2, lower left) and using the fact that the generally much lower than those suggested by the circu-line integral about a closed loop of a perfect differential lation theorem. Simpson [1994] suggested that surface is zero, one obtains wind speeds of 6 or 7 m s-' can reasonably be expected.
Other authors [e.g., Masselink and Pattiaratchi, 1998]
DCa (Po , have suggested that speeds as high as 10 m s-' are
- T1), (2) common. The main reason for this overestimate is that Dt = Rln (P) in the initial formulation the Bjerknes circulation theo-where Po represents atmospheric pressure near Earth's rem assumes friction (in its various forms) is too small to surface [Pa], p, represents atmospheric pressure near be of importance. Another weakness of this application, the top of the circulation cell [Pa], and T indicates the perhaps just as important, is that it ignores the along-average temperature through the vertical column [Hol- shore dimension.
ton, 1992].
[19] To extract the mean wind speed (U) associated 3.1.3. More Complete Models and Topography with the SBS, one uses [21] The following factors affecting the SBS should be included in a complete physical model [Simpson, 19941:
- C ' (1) diurnal variation of the ground temperature, (2)
U =2(H + L) (3) diffusion of heat, (3) static stability, (4) Coriolisforce(s),
(5) diffusion of momentum, (6) topography, and (7) where H is the height of the circulation and L is its prevailing wind. The first two are essential for the sea cross-shore length (Figure 2). Combining equations breeze to occur, and the third is a factor in its inland (2) and (3) yields movement [Simpson, 1994]. The Coriolis force, while
41, 3 / REVIEWS OF GEOPHYSICS Miller et a].: SEA BREEZE
- 1-5 unimportant for the first 6 hours6.944444e-5 days <br />0.00167 hours <br />9.920635e-6 weeks <br />2.283e-6 months <br />, is responsible for and Pielke, 1991]. The thermal forcing over a narrow producing the horizontal rotation of the SBS over time peninsula or island (width <100 kin) is insufficient for and thus limits the extent of its inland penetration developing a deep, well-organized mesoscale circulation,
[Pearce, 1955; Neumann, 1977; Anthes, 1978; Simpson, and both sea breeze systems are weak. For landmasses 1996]. Momentum diffusion is important for producing with widths between 100 and 150 km the thermal forcing the observed near-surface wind profile, is the most im- is strong enough to develop deep systems, and the land-portant brake on the developing circulation, and pre- mass is still narrow enough so that the two opposite vents the SBS from producing the very high wind speeds systems converge at the center and produce a region of predicted by the Bjerknes circulation theorem [Anthes, deep convection. Landmasses >150 km across are too 1978; Simpson, 1994]. SBS response to prevailing (syn- wide for the two opposing systems to reach each other optic-scale) wind is discussed in section 3.2.2. before sunset, and the associated convergence region in
[22] Topography, including the size and shape of the the center of the landmass is weakened [Xian and Pielke, landmass and the details of the coastline, is another 1991]. When the two opposing sea breeze systems reach important factor in the SBS. Many early models inves- each other during the evening hours (when the ambient tigating SBS behavior assumed straight coastlines and atmosphere is stable), undular bores may result [Clarke, flat topography [e.g., Estoque, 1962] but included terms 1984].
corresponding to the other six items listed. More recent authors discuss the behavior of the SBS in more realistic 3.1.3.4. Complex Terrain physical settings. Most of their conclusions are necessar- [26] Complex terrain, with or without nonzero synop-ily specific to the geographical region examined, al- tic-scale flow, may produce several separate sea breeze though some important general conclusions are ob- systems along different portions of the coastline. The tained. appearance of these independent systems may not be simultaneous, and they may not ultimately reach the 3.1.3.1. Seaward Facing Slopes same intensities. Inland topographic features channel
[23] If a hill is located just inland of a flat coastal the low-level flow, creating areas of enhanced conver-plain, with its slope facing the sea, the influence of the gence and upward vertical motion. Upward vertical mo-hill on the SBS is determined by the hill's surface tem- tion is also enhanced where different sea breeze systems perature cycle. When the temperature of the slope varies converge at points inland [Melas et al., 1998, 20001.
with the same diurnal period as that of the coastal plain, the SBS is amplified and occurs earlier than it would if 3.1.3.5. Temperature Characteristics of Land and the hill were not present. If the slope acts as a barrier Sea Surfaces only (i.e., is enshrouded in fog or clouds), the SBS is [27] Stronger SBS-associated wind speeds occur with confined to the coastal plain. Asai and Mitsumoto [1978] stronger cross-shore temperature gradients. Schumann originally described this effect with a computer model, et al. [1991] noted this effect in the sea breeze of Alcoa and Banta et al. [1993] observed it on the west coast of Bay, South Africa, where the strongest sea breeze occurs the United States using a pulsed Doppler lidar. in areas with nearby dune fields. During the daytime the sand dunes create a region of superheated air immedi-3.1.3.2. Bays ately overhead, setting up a very strong land-sea tem-
[241 When a bay is located on an otherwise straight, perature difference. Coastal upwelling, which brings east-west coastline with the ocean to the south, the bay colder water to the surface, also enhances the cross-causes the sea breeze to bow landward relative to the shore temperature gradient. On the east coast of a straight sections of coastline on either side. Within the continent in the Northern Hemisphere, upwelling (driv-convergence zone between the sea breeze and the sea- en by Ekman pumping) occurs when there is an along-ward synoptic-scale flow, there are regions of enhanced shore wind component, with low pressure to the west upward and downward vertical motion related to the and high pressure to the east [Apel, 1987]. Franchitoet al.
position of the bay but distributed asymmetrically. In the [1998] observed that the SBS of coastal Brazil is stronger Northern Hemisphere, enhanced convergence and ver- during periods of upwelling and weaker when there is no tical motion occur to the west of the bay, where the PGF upwelling.
and Coriolis force act in opposition to each other, and weakened convergence and vertical motion occur on the 3.2. Structure east side of the bay, where the two forces act in the same [28] The SBS consists of phenomena occurring on direction [McPherson, 1970]. several spatial scales (Figure 1). A summary of these includes the initiating sound waves discussed by Tijm 3.1.3.3. Narrow Landmasses and van Delden [1999], the sea breeze forerunner and
[251 When two sea breeze systems form on opposite similar prefrontal waves [Geisler and Bretherton, 1969; sides of a narrow landmass, the interactions that result Sha et al., 1993], the closed cell of circulation predicted when they converge at the center of the landmass are by the Bjerknes circulation theorem [Holton, 1992], the partially determined by the width of the landmass [Xian sea breeze gravity current and all of its associated phe-
1-6
- Miller et a[.: SEA BREEZE 41, 3 / REVIEWS OF GEOPHYSICS nomena [e.g., Simpson, 1997], the sea breeze front and be a closed system under some circumstances [Estoque, all of its associated phenomena [e.g., Kraus et al., 1990; 1962; Intrieri et al., 1990; Banta et al., 1993]. The upper Kraus, 1992; Reible et al., 1993], and smaller-scale phe- limit of the circulation varies from a few hundred meters nomena, such as a convective internal boundary layer, to 1 or 2 km, depending on synoptic conditions [Frizzola within the landward moving marine air mass [Hsu, 1988; and Fisher,1963; Barbato, 1975; Intrieriet al., 1990; Xian Zhong and Takle, 1992; Rao and Fuelberg, 2000]. One and Pielke, 1991]. On the other hand, Tijm et al. [1999]
must also include interactions between all of these phe- found that the seaward return flow may overcompensate nomena and with phenomena originating from other for the low-level landward flow by as much as 30% (with processes in both the atmosphere and ocean [e.g., Briim- respect to mass), creating a landward "retum-return" mer et al., 1995; Gibbs, 2000]. flow above -4 km. Weak sinking currents have been observed in situ and in model results over a broad area 3.2.1. Prefrontal Phenomena several tens of kilometers offshore, supporting a closed
[29] In addition to the sound waves that initiate the SBC with at least partial mass conservation [Estoque, cross-shore mesoscale PGF, another type of transitory 1962; Neumann and Mahrer, 1971; Elliott and O'Brien, wave precedes the arrival of the sea breeze front on land. 1977; Asimakopoulos et al., 1999; Chiba et al., 1999].
The sea breeze forerunner is actually a cluster of waves [33] The prevailing wind (PW) plays a large role in that is initiated when the cross-shore thermal contrast determining the shape and shore-relative location of the appears above the land-sea interface. The waves with the SBC. The PW is important in determining whether or largest amplitude arrive at points inland first, moving at not a sea breeze will be detectable on land (a very strong speeds much faster than the sea breeze front. Over time, offshore PW will prevent the SBC from reaching the progressively smaller-scale waves arrive inland. Observ- coast), as well as exerting an influence on its behavior ers on Earth's surface will experience a wind from the [e.g., Finkele et al., 1995]. Mizuma's [1995, 19981 obser-direction of the sea consisting of local (continental) air vational studies of SBC behavior on the coast of Japan before the arrival of the sea breeze front. The model include descriptions of several manually identified sea results from Geisler and Bretherton [1969] indicate that breeze modes, which are dependent on the shape of the with reasonable values of internal (eddy) viscosity and local coastline and the direction and strength of the PW.
surface friction the forerunner reaches as far inland as Several other studies have examined SBC response to 60 km. the PW both generically and along specific coastlines
[30] Another type of prefrontal wave may occur in the [e.g., Fisher, 1960; Estoque, 1962; Pearson et al., 1983; late evening, when the sea breeze front interacts with an McKendry and Roulet, 1994; Planchon and Cautenet, inland nocturnal stable layer. These waves consist of 1997; Melas et al., 1998; Asimakopoulos et al., 1999; small overturning cells with a characteristic wavelength Clappieret at., 2000; Melas et al., 2000; Laird and Kris-of -10 kin, propagate inland at approximately 3.5 m s-1, tovich, 2001; Oliphant et al., 2001; Lericos et al., 2002].
and rapidly dissipate in the ambient flow field [Sha et al.,
1993]. 3.2.2.1. Generic Categories of Sea Breeze Circulation 3.2.2. SBC Response to the Synoptic-Scale Wind [34] The SBC falls into four PW-dependent catego-
[31] The SBC begins when a landward component is ries: pure, corkscrew, backdoor, and synoptic [Adams, added to the net low-level wind vector in response to the 1997]. The fourth stretches the definition of the sea mesoscale thermal PGF, resulting in a divergence in the breeze from a locally driven circulation to any wind cross-shore wind component over the sea. The shore- blowing from sea to land, and therefore this category will parallel zone in which the new wind component is added not be addressed here. The remaining three are de-grows horizontally in the cross-shore direction, expand- scribed below in the context of the east coast of a ing more rapidly seaward than landward [Finkele et al., Northern Hemisphere continent.
1995]. A shallow layer of marine air begins flowing [35] A pure sea breeze [Adams, 1997] occurs under toward the land, converging with continental air and light PW conditions, when the existing geostrophic forc-creating a region of upward vertical air currents (Figure ing points seaward at a right angle to the coastline, and 1). closely matches the general description given above
[32] A seaward return flow usually develops aloft, (Figures 3a-3c). This type of sea breeze is preceded by above the upper boundary of the marine air, creating a calm conditions, when the synoptic-scale PGF is briefly shear zone where the flow changes directions from land- in equilibrium with the locally created thermal PGF.
ward to seaward [Finkele et al., 1995; Camberlin and Ohashi and Kida [2001] noted its presence in Japan, Planchon, 1997; Tijm et al., 1999; Oliphant et al., 2001]. ahead of an advancing SBG and below 800 in above This return flow is required if the SBC is to be consid- ground level (AGL). Chiba et al. [1999], using an instru-ered closed, i.e., mass conservative. Several authors, mented helicopter, observed that the calm region ex-however, report either the complete absence, of this tends to -10 km offshore. The SBC expands throughout return flow or its insufficiency at counterbalancing the the day, with the landward extreme reaching as far low-level onshore flow, implying that the SBC may not inland as -130 km [Adams, 1997].
41, 3 / REVIEWS OF GEOPHYSICS Miller et a[.: SEA BREEZE
- 1-7
~"," ~~'K
/ #/
/ ',
a b e
Figure 3. Development of pure, corkscrew, and backdoor sea breezes. Land is on the left; ocean is on the right. North is into the page. Bold red arrows indicate wind at the top of the planetary boundary layer. Thin red arrows indicate near-surface wind not related to sea breeze circulation (SBC). Blue arrows indicated SBC-related near-surface wind. Bold dashed blue line indicates inland limit of SBC; thin dashed blue line indicates seaward extent. (a, b, and c) Development of pure sea breeze. Dashed red line indicates inland extend of calm region that extends offshore to the first dashed blue line. (d, e, and f) Development of corkscrew sea breeze. (g, h, and i) Development of backdoor sea breeze, AfterAdams [1997].
[36] The corkscrew sea breeze [Adams, 1997] occurs ponents (Figures 3g-3i). In this case the along-shore when the PW has both along-shore and cross-shore component is southward on the east coast of the hypo-components (Figures 3d-3f). The along-shore compo- thetical Northern Hemisphere landmass. Buys-Ballot's nent is northward on the east coast of the hypothetical law implies that lower pressure is over the ocean, and Northern Hemisphere landmass. According to Buys-Bal- higher pressure is over the land [Lutgens and Tarbuck, lot's law, a northward wind component implies lower 2001]. This, combined with the cross-shore variation in pressure over land and higher pressure over water [Lut- the coefficient of surface friction, results in an area of gens and Tarbuck, 2001]. When differences in the coef- low-level convergence near the coast. This surface con-ficient of surface friction are taken into account, a north- vergence prevents air from sinking from aloft and inhib-ward wind component and lower pressure over land its the sea breeze's progress toward land. This also imply an area of low-level divergence near the coast. Air implies that a backdoor sea breeze can only reach the from aloft sinks into the divergence zone and assists with coast with a stronger thermal PGF, relative to a pure or the initiation of the sea breeze. Since the thermal PGF corkscrew sea breeze. As with the corkscrew sea breeze, and the synoptic-scale PGF are not entirely oriented arrival of a backdoor sea breeze is marked by a gradual along the same dimension, the former does not have to shift in wind direction from northwest to northeast. In overcome the full magnitude of the latter, and therefore this case the wind veers (rotates in a clockwise direction) there is no calm period prior to the onset of the sea rather than backs. The backdoor sea breeze also rotates breeze. This also implies that a corkscrew sea breeze can helically rather than in a simple loop, although in this reach the coast with a weaker thermal PGF, relative to a case the along-shore geostrophic wind component causes pure sea breeze. The arrival of the corkscrew sea breeze the helix to spiral southward rather than northward. The is marked by a gradual backing (rotation in the counter- other important differences between the corkscrew and clockwise direction) of the wind from southwest to backdoor sea breezes are that the latter is much weaker, southeast. Like the pure sea breeze, the corkscrew sea arrives on land later in the day, and may manifest in breeze circulates in a vertical cell, but the along-shore pulses rather than strong, steady flow [Adams, 1997].
component of the synoptic wind causes the circulation to take on a helical shape rather than a simple loop [Ad- 3.2.2.2. Examples of Sea Breeze Circulation ams, 1997]. [38] A series of case studies carried out on the coast of
[37] The backdoor sea breeze [Adams, 1997] also oc- New England provide real-world comparisons to the curs when the PW has along-shore and cross-shore com- idealized sea breeze scenarios described in section
1-8 a Miller et aL: SEA BREEZE 41, 3 / REVIEWS OF GEOPHYSICS La~titudle VT
- VT NHl' J '"X II:EV '
42.5
&N.H E
- X X T X, xft 7&"
["*]~
\I LC x _ '
~ý x
^ *.x, - -<+**+
, -. x
<.* X Xr*-,+
La , +xx X x ' x x ,x FN nl , 2002.,s Xbreez s XX a I.x ex seXrb x ,x Xx
,XC . x x X XX
/ _ "X rn+mL x, X along the relatively straight section of coastline from Lawrence, Massachusetts (KLWM), through north of Portland, Maine (KPWM) (see inset, lower right). Standard National Oceanic and Atmospheric Administra-tion surface weather observing sites are labeled in black lettering with their International Civil Aviation Organization Identifiers. Stations added by the University of New Hampshire to improve the spatial resolution are shown in red. Data from all stations were combined and interpolated onto the 10-km rectangular grid shown in blue.
3.2.2.1. The study area shown in Figure 4 is nearly ideal [39] The 3 August 2002 event (Figure 5) most closely for comparison to the hypothetical sea breezes shown in resembled the pure sea breeze. A high-pressure center Figure 3, because the central 90% of the coastline is was approaching southern New England from the west, reasonably straight and can be approximated by a setting up a shore-perpendicular ambient flow. At 1200 straight line rotated 30' clockwise from true north. UTC (subtract 4 hours4.62963e-5 days <br />0.00111 hours <br />6.613757e-6 weeks <br />1.522e-6 months <br /> to convert to local daylight time These sea breeze events were observed during the New (LDT)) the flow throughout most of the study area was England Air Quality Study (NEAQS) 2002 summer cam- northwesterly (Figure 5a). At 1600 UTC a region of light paign, which involved researchers from the U.S. Na- winds, cyclonic rotation, and low-level convergence de-tional Oceanic and Atmospheric Administration veloped in the southern half, -10 km offshore (Figure (NOAA), the University of New Hampshire (UNH), 5b). By 1900 UTC, there was a well-developed sea and others (Atmospheric Investigation, Regional Mod- breeze with a clearly defined inland convergence zone in eling, Analysis, and Prediction project (AIRMAP), The the southern half and wind flow crossing the coast at a New England Air Quality Study, Draft Science Plan, right angle (Figure 5c). The sea breeze was beginning to October 2001, available at http://www.al.noaa.gov/ develop in the north as well. By 2300 UTC the wind neaqs/). For NEAQS, researchers at UNH's Climate direction within the SBC veered to a more southerly Change Research Center created a mesoscale surface direction, causing the system to take on characteristics of weather network by adding a series of small, automated the corkscrew sea breeze (Figure 5d). At latitudes where stations to NOAA's existing network surface stations Coriolis is important, the sea breeze evolves to take on (Figure 4). The enhanced network of stations permits corkscrew characteristics under the influence of rota-the resolution of meteorological features as small as 10 tional forces.
km across, particularly over land. Hourly reports from [40] The 21 July 2002 event (Figure 6) most closely the combined network were interpolated onto a rectan- resembled the corkscrew sea breeze from its inception.
gular grid using standard techniques [Barnes, 1964], and A cold front was approaching New England from the the regridded data were examined for patterns of wind west, and a high was to the south, setting up shore-flow near Earth's surface. parallel, southwesterly flow. At 1300 UTC the wind
41, 3 / REVIEWS OF GEOPHYSICS Miller et aL: SEA BREEZE
- 1-9 44 43,5 43 42.5 Latitude
[°N]
43.'
4; 42.'
-71.5 -71 -70.5 -70 -71.5 -71 -70.5 -70 5 ms Longitude
[°E]
Figure 5. Pure sea breeze event 3 August 2002: (a) 1200 UTC (0800 LDT), (b) 1600 UTC, (c) 1900 UTC, and (d) 2300 UTC.
throughout the study area was southwesterly and shore- out most of the study area was northeasterly and shore-parallel (Figure 6a). By 1400 UTC the wind in the parallel (Figure 7a). By 1500 UTC, veering added a offshore area backed from southwest to southeast, de- shoreward component throughout the southern half; veloping a noticeable shoreward component and creat- therefore there was no well-defined line of convergence ing a pronounced convergence zone at the coast (Figure between the ambient flow over the continent and the 6b). The SBC reached its peak at 2100 UTC, when the flow within the developing SBC (Figure 7b). Continued leading edge of the SBC extended inland nearly 65 km veering rotated the wind from northeast to southeast by (Figure 6c) and thereafter began to diminish. By 0300 2000 UTC, resulting in shore-perpendicular flow UTC on 22 July, veering wiped out the convergence zone, throughout the study area (Figure 7c). By 2300 UTC the and the SBC was no longer readily identifiable (Figure demise of the SBC was marked by the collapse of orga-6d). nized flow on this scale (Figure 7d).
[411 The 24 July 2002 event (Figure 7) most closely [42] These case studies indicate that the generic mod-resembled the backdoor sea breeze described in section els introduced in the previous section are probably too 3.2.2.1. A high-pressure center was approaching north- simple for use other than in teaching basic concepts. In ern New England from the west, setting up shore-paral- addition to along-shore variations, such as differing lel, northeasterly flow. At 1200 UTC the wind through- times of onset of the sea breeze, entire sea breeze events
I 1 -10 9 Miller et al.: SEA BREEZE 41, 3 / REVIEWS OF GEOPHYSICS Latitude
[*N] 44 rlI ~ ~
-4 'f I r ,.-., - . ,,, I 45.5 , I , \V I,j i,,-
4L., ,, *
- C , d
-1[ -1 - 5 1-70 -70 5 ms Longitude
[*E]
Figure 6. Corkscrew sea breeze event 21-22 July 2002: (a) 1300 UTC (0900 LDT), (b) 1400 UTC, (c) 2100 UTC, and (d) 0300 UTC.
evolve in ways not predicted by the generic models. The it. Instead, the wind direction became random at the end Coriolis force, which favors veering, is a dominant fac- of the SBC's life, driven by smaller-scale, local forces.
tor. In the 3 August (pure) sea breeze case, veering The Coriolis force's influence on the SBS is discussed in induced by the Coriolis force transformed the SBC from greater detail in sections 3.4, 3.5.2.2, and 3.5.3.
a pure into a corkscrew system. In the 21 July (cork-screw) case, backing occurred early in the life of the SBC, before Coriolis had time to become an important 3.2.3. Sea Breeze Gravity Current (SBG) and Kelvin-factor. (The Coriolis force requires several hours to Helmholtz Billows become an important factor, a fact that can be derived [43] The low-level landward flow of marine air is one by scaling the equations of motion.) The corkscrew sea example of a large class of phenomena called gravity or breeze was terminated when the Coriolis force wiped density currents. Gravity currents are primarily horizon-out the cross-shore wind component. In the 24 July tal flows in fluids that can be generated by a density (backdoor) case, veering (driven by thermodynamic dif- difference of only a few percent. Other examples of ferences between land and sea) initiated the sea breeze, gravity currents are thunderstorm outflows, turbidity cur-but the Coriolis force prevented backing from destroying rents, avalanches, and pyroclasticflows [Simpson, 1997].
41, 3 / REVIEWS OF GEOPHYSICS Miller et aL: SEA BREEZE
- 1 -11 43.5 Latitude
[°N]
-71.5 -71 -70.5 -70 -71.5 -71 -70.5 -70 5 mss1 -
Longitude
[*E]
Figure 7. Backdoor sea breeze event 24-25 July 2002: (a) 1200 UTC (0800 LDT), (b) 1500 UTC, (c) 2000 UTC, and (d) 0000 UTC.
[44] The microscale (<2 km) horizontal thermal con- laboratory and the field [e.g., Kuelegan, 1957; Simpson, trast across the boundary between marine and continen- 1969; Physick, 1980; Chiba, 1993]. Kuelegan's [1957] in-tal air masses can be very sharp, and the boundary often terest was the flow of salt water through locks and into takes on a frontal nature similar to a synoptic-scale cold freshwater channels, for which he established the follow-front [Chiba, 1993; Finkele et al., 1995; Simpson, 1997]. ing relationship:
The leading edge of the marine air mass develops a raised head, because of the updraft created by low-level convergence between the marine and continental air JUI = k, -gd, (5) masses [Finkele et al., 1995]. The height of the SBH is p approximately twice that of the feeder flow originating where UI is the speed at which the denser fluid front offshore [Simpson et al., 1977], but a headwind (the advances into the less dense fluid [m s-1], k is a constant ambient offshore flow) flattens the SBH [e.g., Frizzola (taken as 0.78), p is the density of the denser fluid [kg and Fisher, 1963]. The depth of the onshore flow behind m- 3], Ap is the density difference between the two fluids the front ranges between about 300 and 2500 m [e.g., [kg m- 3], g is the gravitational acceleration (9.81 m S-2),
Barbato, 1975]. Simpson [1969] noted that its average and d is the height of the SBH [m].
depth in Great Britain was -700 m. [46] Simpson [1969] adapted equation (5) to describe
[45] The speed with which a density current advances the rate at which the SBG propagates inland. To calcu-into the ambient fluid has been studied extensively in the late the constant k, he replaced the height of the SBH
1-12 9 Miller et al.: SEA BREEZE 41, 3 / REVIEWS OF GEOPHYSICS (d) with its mean height d, taken as 700 m (from a climatology of 54 sea breezes recorded over a 6-year period in Britain). He also replaced Ap/p with AT/T [AT and T in kelvins]. This work yielded a k value of 0.62, which Schoenberger [1984] reported is equally effective when compared to radar observations of the propaga-tion speed of a land breeze on the lower peninsula of Michigan.
[47] Later, Simpson and Britter [1980] determined that the SBG's rate of inland progression is slowed by -3/5 of the opposing component of the PW or Figure 8. Simpson's [1997] shadowgraph of a laboratory gravity current. Pure water is on the left, and slightly salty (denser) water enters from the right. Visible in the photograph IUI k y/Tgd - 0.59u , (6) are the current's nose (slightly elevated leading edge), raised head, and trailing Kelvin-Helmholtz billows (as described in where ug is the cross-shore geostrophic wind component text). Reprinted with the permission of Cambridge University
[m s- 1 ] resulting from the synoptic-scale PGF. Press.
[48] A shear zone develops on the upper boundary of the marine air mass, between the low-level onshore flow and the seaward return flow aloft. KHBs develop in the analogs of Simpson's laboratory gravity current features shear zone over land (and are further enhanced by (Figure 8) are also present in the SBG [e.g., Craig et al.,
thermodynamic instability induced by insolation during 1945; Chiba et al., 1999; Wood et al., 1999].
the middle part of the day), causing mixing between marine air and continental air and creating a turbulent 3.2.4. Sea Breeze Front (SBF) wake behind the SBH [Simpson, 1969;Atkins et al., 1995; [50] The leading edge of the SBG is often associated Chiba et al., 1999]. The development of KHBs in late with a strong cross-shore temperature contrast, and it morning causes a friction-like force on the upper bound- can take on characteristics similar to those of a synoptic-ary of the air mass that slows the sea breeze's inland scale cold front. As a front, the SBF is subject to fron-progression [e.g., Sha et al., 1991]. KHBs appear as togenesis and frontolysis, bifurcation (separation of the vortex rolls in regions of strong shear when the Richard- regions of maximum temperature gradient and maxi-son number (Ri), the ratio of static stability to kinetic mum low-level convergence), variations in intensity (in energy of shear, is <0.25. This may be calculated by terms of cross-shore width and the magnitude of the various gradients involved), and variations in slope.
g ao Strong vertical velocities and the SBH are associated 0 a0 with the SBF (Figure 9). The region of the SBG behind Ri = au2(7)
((z) where g is the gravitational acceleration (9.81 m s-2), 0 is the potential temperature [K], u is the horizontal flow component [m s-1], and z is the vertical dimension [m].
It can be seen from equation (7) that regions where the vertical temperature gradient vanishes, such as in a well-mixed layer resulting from convective overturning,
-- iB are prone to the development of KHBs. Simpson and Britter [19801 suggested that the KHBs along the upper ................ C boundary of the sea breeze marine layer may be 500-1000 m in length. Nielsen [1992], using aircraft, observed KHBs with wavelengths between 1000 and 2000 m on Figure 9. Simplified two-dimensional flow associated with a the upper boundary of a New England coastal front, a gravity current in presence of surface friction [after Simpson, mesoscale phenomenon that closely resembles a sea 1997; Sha et al., 1993]. Ambient fluid (red) is on the left and breeze. above. Denser fluid (blue) is invading from the right. Black arrows indicate direction of fluid flow. Mixing between the two
[49] Simpson [1994, 1997] conducted experiments that fluids occurs both beneath the "nose" of the denser fluid documented the three-dimensional (3-D) structure of (stippled, below line C) and in turbulent wake region behind gravity currents, using laboratory tanks and two bodies the head of the gravity current (between lines A and B). Figure of water of slightly different densities. His work repro- 9 was redrawn from Simpson [1997] with permission from duced several aspects of the SBG, including the SBH, the author.
front, and KHBs. Field studies have confirmed that
41, 3 / REVIEWS OF GEOPHYSICS Miller et aL: SEA BREEZE 9 1-13 the SBF and beneath the SBH can sometimes separate the quantity Q, in the x-z plain and moving the remaining from the feeder flow and propagate inland as a cutoff LHS terms to the RHS, the relation can be written as vortex or undular bore. Sea breeze fronts vary in intensity in the along-shore direction, even along nearly homoge- aQX au aw neous coasts. =at - Qx - - Qz +KV 2 Qx. (11)
[55] Reible et al. [1993] began with equation (11) and 3.2.4.1. Frontogenesis Function then introduced scaling arguments to approximate the
[51] Sea breeze frontogenesis (the formation or inten- values of the terms on the RHS. Kraus et al. [1990]
sification of the sea breeze front) can be defined as an carried the argument further by using Qx = -CpPaOx and increase in the magnitude of the cross-shore potential Q, = -c,,pO, where cp is the coefficient of specific heat temperature gradient. A secondary set of processes in- at constant pressure [J kg-' K-'], Pa is the density of air volves the cross-shore water vapor gradient [e.g., Kraus [kg m- 3], and Ox and 0. are the derivatives of the poten-et al., 1990]. A kinematic function describing the processes tial temperature field [K m-1] in the x and z directions (not including the release of latent heat) that contribute to
[Halliday et al., 1993]. These substitutions and algebraic the thermal frontogenesis can be derived by beginning with manipulation lead to the three-dimensional diffusion relation:
aQ aQ aQ aQ a2Q a2Q dtx - au 0,- - 0 + K VwOx, at -ax ay +W az =KK+V-x Kyy 2 (12)
-+u dt ax 0 ax a2Q which is the kinematic frontogenesis function. Term 1 on
+ Kz aZ2 (8) the RHS describes cross-shore confluence (or deforma-tion); term 2 describes rotation of vertical temperature where Q is the heat content (energy) per unit volume [J gradients into the horizontal plane; and term 3 describes m- 3], the first term on the left-hand side (LHS) is the cross-shore variations in turbulent heat diffusion [Kraus time-dependent (Eulerian) term, and the remaining et al., 1990].
three terms on the LHS describe the horizontal and [56] With the ocean to the right (toward positive x vertical advection of heat via wind components u, v, and values), sea breeze frontogenesis is indicated if the sum w [m s-1 ]. The three terms on the right-hand side (RHS) of the RHS is less than zero, and frontolysis (the de-describe the turbulent diffusion of heat in three dimen- struction or weakening of the sea breeze front) is indi-sions, with coefficients K,, Ky, and Kz [M 2 s-1] [Anderson cated if the sum of the RHS is greater than zero. In the et al., 1984]. All seven terms have overall units of power former case the cross-shore potential temperature gra-per unit volume [W m- 3]. dient is tending toward more negative values with time,
[52] By assuming, (1) a north-south coastline with and in the latter it is tending toward more positive values ocean on the right, (2) a weak north-south temperature with time.
gradient, (3) a wind primarily from the west (v - 0), and [57] There are two ways to improve this result. First, (4) K, - Kz, relation (8) can be rewritten as Neumann [1977] noted that within a few hours; even a system that begins with exclusively cross-shore flow will OQ aQ aQ (a 2Q a 2Q\ develop important flows and gradients in the along-
-at + u -x + w -y= K_-+W z2. (9) shore direction, because of the rotation of the SBS under Coriolis and other forces. An improved kinematic
[53] From here one may follow a line of reasoning frontogenesis function will account for horizontal rota-similar to that of Miller [1948] and derive a function de-tion. Second, the assumption that the coefficient of scribing sea breeze frontogenesis. Since equation (9) de- turbulent diffusion is the same in both the horizontal scribes heat, not heat gradient, the first logical step is to and vertical dimensions may not be correct [e.g., Ander-take the partial derivative in the cross-shore (x) direction. son et al., 1984]. An improved kinematic frontogenesis Assuming that K is approximately constant and following function will permit anisotropic coefficients. Both of the chain rule for derivatives, equation (9) becomes these points indicate a need for a 3-D thermal fronto-genesis function, even if the front initially forms parallel OQx au aQX aw aQx to the coastline.
at a Qx+ u x+-xQzax +aw =KV2Qx [58] If one begins with equation (8) and follows a (10) derivation similar to that shown above, keeping all three 2
spatial dimensions and allowing KH : K, (horizontal where V is the second-order derivative in the x-z plane and vertical coefficients, respectively), one arrives at the and subscripts (e.g., Qx) indicate the partial derivative of following:
the function (Q) in the direction indicated (x, z).
[54] By noting that the first, third, and fifth terms on the dOx Ou av aw O [ 0201 LHS of equation (10) constitute the material derivative of dt ax ax aOx +axIK ax']
1-14 9 Miller et A: SEA BREEZE 41, 3 / REVIEWS OF GEOPHYSICS ahead of the kinematic front. The region between the
[K 1"], (13) two fronts (called the thermodynamic zone) is the ana-
+a[x a2 +a[K aZ2f log of frontal transition zones in synoptic-scale cold which is similar to equation (12). In equation (13) the fronts, and it is widest where the SBF has merged with RHS terms correspond to the following: HCRs [Atkins et al., 1995].
- 1. Term 1 is a cross-shore confluence (or deforma- [62] Atkins et al. [1995] noted that only those cases tion, as in equation (12)). with an offshore ambient wind exhibited the kinematic
- 2. Term 2 is the rotation of along-shore temperature front. The lack of a clearly defined kinematic front may gradients into the cross-shore direction resulting from a also be responsible for the apparent absence of a front in cross-shore variation in the along-shore wind compo- some of the sea breeze events noted elsewhere [e.g.,
nent. Along-shore gradients can result from rotation of Helmis et al., 1995].
the entire SBC, interactions between the advancing SBF and convective cells in the continental air mass, and the 3.2.4.3. Strength of the SBF lobe and cleft structures observed on the front itself [63] Strong sea breeze fronts are those with (1) a
[Neumann, 1977; Simpson, 1997; Stephan et al., 1999]. significant temperature contrast between the two air
- 3. Term 3 is the rotation of vertical temperature masses, (2) a sharp temperature gradient (occurring gradients into the horizontal plane (as in equation (12)). over a narrow horizontal distance), or (3) both. The
- 4. Terms 4 and 5 are the cross-shore variation in strongest fronts, most clearly identified by the fronto-cross- and along-shore eddy-driven heat diffusion. genesis function, will also be those where the thermody-
- 5. Term 6 is the cross-shore variation in eddy-driven namic and kinematic fronts are in the same place. At its diffusion of heat in the vertical dimension. simplest level these are the narrowest fronts. For narrow
[59] Kraus [1992], using data recorded by instru- fronts to occur, strong low-level convergence is needed.
mented aircraft, computed the contribution to overall Strong low-level convergence can occur with (1) strong, frontogenesis resulting from terms 4, 5, and 6 and de- opposing offshore winds, (2) strong onshore winds termined that their contributions are of the same order within the marine air mass, or (3) both.
of importance as the convergence and shear terms, at [64] Helmis et al. [1987] conducted a field study of the least in very sharp fronts. These terms can be either SBF in Athens, Greece, utilizing surface observations, frontolytic or frontogenetic. an acoustic minisounder, and an atmospheric turbulence probe. They classify fronts into two groups, narrow and 3.2.4.2. Frontal Bifurcation wide, that are dependent on the synoptic-scale wind
[60] The sea breeze frontogenesis function (13) is a direction. Narrower fronts with stronger temperature diagnostic tool, but there are a number of problems with gradients occur when the synoptic-scale wind is offshore, attempting to use it for practical meteorology. Among and wider fronts with weaker temperature gradients them are the absence of a term for variations in along- occur when the synoptic-scale wind is parallel to the shore gradients (which come into play in term 2), the coast.
absence of a term for shear due to along-shore variation [65] Chiba [19931 studied the SBF in Japan through a in the cross-shore wind component (which will result in full annual cycle and reports the width of the front as a the rotation of horizontal temperature gradients), and function of the ratio Uc/UM. Uc is the offshore wind the implicit assumption that the strongest temperature speed in the continental air mass (observed before fron-gradients and convergence zones associated with sea tal passage), and UM is the onshore wind speed within breeze fronts occur in the same place. The latter as- the marine air mass (observed after frontal passage).
sumption is often not correct. The wind speeds are recorded on a 21-m tower, 2 km
[61] More than one SBF may develop in a single SBS inland from the coast. During the cold season (October
[e.g., Kraus et al., 1990; Atkins et al., 1995; Lapworth, through March) the ratio has typical values between 0.2 2000]. A field study of the SBF in eastern Florida, using and 0.8, and during the warm season (April through Doppler radar stations and an instrumented aircraft, September) it has values between 0.7 and 1.3. As the discerned a difference between the thermodynamic and ratio increases from 0.7 to 2, the width of the SBF kinematic fronts [Atkins et al., 1995]. The thermodynamic increases from 130 to 1120 m, with an overall average of front is the location within the planetary boundary layer 560 +/- 230 m. This indicates that in the study area of (PBL) where the mean thermodynamic properties of the Chiba [1993] the width of the front is inversely propor-air begin to differ from those of the ambient (continen- tional to the relative strength (magnitude) of the on-tal) air, and it is created by interactions between the SBC shore wind within the marine air mass and directly and horizontal convective rolls (HCRs). The kinematic proportional to the offshore wind within the continental front is the location of maximum near-surface (100 m air mass.
AGL) wind convergence. In an ideal case the two fea-tures are in the same place, and equation (13) diagnoses 3.2.4.4. SBF Slope strong frontogenesis. In cases of onshore ambient flow [66] The advancing SBF slopes backwards over the the thermodynamic front may be as much as 15 km marine air at an angle of between 10' and 60' from the
41, 3 / REVIEWS OF GEOPHYSICS Miller et al.: SEA BREEZE
- 1-15 horizontal [Wood et al., 1999; Lapworth, 2000]. Theoret- lift, and is the updraft responsible for the SBH. Addi-ical studies indicate the slope should be 60' near Earth's tional, smaller-scale updrafts are visible behind the SBF, surface, with smaller values aloft [Xu, 1992]. These find- beneath the landward moving marine air mass, caused ings are mirrored in the laboratory tank experiments by surface friction and thermal instability. A prefrontal carried out by Simpson [1994, 1997]. Friction on Earth's downdraft is visible inland of the primary updraft and is surface reduces the forward momentum of the advanc- associated with prefrontal gravity waves [Sha et al.,
ing front in the lowest few hundred meters (forcing the 1993]. This landward downdraft is part of a repeating frontal surface toward the vertical), but aloft, in the pattern of updrafts and downdrafts in the continental air absence of surface friction, the front moves inland more mass, associated with convective activity on a horizontal quickly, resulting in a leading nose with a much shal- scale of -2 km [Stephan et al., 1999]. Another downdraft lower slope above (Figure 8). is visible immediately seaward of the cold updraft within
[67] Wood et al. [1999], from measurements made of the sea breeze head, and together these form a closed the SBF in instrumented aircraft, reported a slope of 300 vertical circulation [Kitada, 1987]. Pollutants can be-from the horizontal between 200 and 500 m AGL in one come trapped within this postfrontal roll vortex and be case but only 15O-20' in another. They speculated that transported for great distances inland [Kitada, 1987].
the latter front had a shallower slope because it was Numerous updrafts and downdrafts are also visible in moving across flat terrain, while the former was moving the turbulent mixing region associated with the KHBs.
up a shallow incline, which reduced the front's forward [70] Before sea breeze frontal passage (fropa), down-momentum. An ambient opposing wind has a similar drafts occur 60% of the time, while updrafts dominate effect on the SBF. Helmis et al. [1987] observed that after fropa [Chiba, 1993]. Observed downdrafts have had when the sea breeze advances into stronger opposing average magnitudes of -0.5 m s- , while updrafts were synoptic-scale winds, it exhibits a shallower slope (small- between 0.5 and 1.0 m s-' [Chiba, 1993; Stephan et al.,
er angles from the horizontal), and a sea breeze moving 1999]. The magnitude of the vertical velocities is also into ambient shore-parallel wind regimes exhibits a dependent on the strength of the frontal zone and the steeper slope (larger angles from the horizontal). rate at which the SBF advances inland [Helmis et al.,
1987]. Vertical velocities of 1.0 to 1.5 m s-1 have been 3.2.4.5. Vertical Velocities at the SBF observed with sharp frontal zones, while weaker fronts
[68] A scale analysis of the frontogenesis function produced updrafts and downdrafts only one third as reveals that the vertical shear term (involving cross- large [Helmis et al., 1987]. Fast moving fronts produce shore variations in the vertical component of velocity) is greater vertical velocities than slow moving fronts [Hel-of the same approximate magnitude as the horizontal mis et al., 1987]. The direction of the ambient flow is also terms. Vertical currents initiated or enhanced by the important, with updrafts as large as 2.0 m s-1 occurring SBF can cause convective clouds, including thunder- at the SBF when the ambient flow is offshore [Helmis et storms. (For more about the initiation of thunderstorms al., 19871. When the SBF interacts with HCRs, the and other convective activity by the sea breeze front, see largest upward vertical currents at the front occur where Pielke [1985], Pielke et al. [1991], Rubes et al. [1993], the frontal surface intersects an HCR rotating in the Atkins et al. [1995], Weckwerth et al. [1996], Silva Dias same sense as the postfrontal roll vortex [Atkins et al.,
and Machado [1997], Weckwerth et al. [1997], Carbone et 1995]. Regions of enhanced vertical velocities are al. [2000], and Rao and Fuelberg [2000].) Vertical cur- marked by cumulus cloud development [Atkins et al.,
rents are also a factor in the dilution of pollutants in the 1995].
PBL and are therefore of interest to air pollution mete-orologists and engineers. Aviators are concerned about 3.2.4.6. Formation of Undular Bores low-level wind shear resulting from sudden changes in [71] During the late evening hours the SBF, SBH, and vertical velocity. Therefore there are ample theoretical postfrontal roll vortex can separate from the feeder flow and practical reasons for studying the changes in the and continue moving inland independently as an undular atmosphere's vertical motions brought about by the bore or cutoff vortex [Kitada, 1987; Simpson, 1994]. The SBF. cutoff vortex is a solitary wave that propagates along the
[69] Figure 9 illustrates a simplified two-dimensional top of a low-level maritime or nocturnal radiation inver-flow structure in a gravity current with bottom friction, sion [Smith et al., 1982] and is -1 km high and -20 km representing a situation analogous to the sea breeze [Sha across in the cross-shore direction [Sha et al, 1993]. At et al., 1993; Simpson, 1997]. Arrows indicate the direc- ground level the approach of a cutoff vortex is marked tion of the fluid flow. The primary updraft is in the warm by an abrupt increase in atmospheric pressure and tem-air immediately adjacent to the SBF. This updraft is perature and a change in the wind speed and direction.
caused by the physical wedge of the denser marine air Its arrival may or may not be followed later by the arrival undercutting the lighter continental air. Another updraft of the sea breeze proper [Simpson et al., 1977; Simpson, is in the cold air, immediately behind the SBF. This 1996]. It often continues moving inland after the low-updraft is caused by low-level convergence and mass level onshore flow has ended, and it therefore is no conservation, with thermal instability adding additional longer part of a gravity current [Abbs and Physick, 1992].
1-16
- Miller et al.: SEA BREEZE 41, 3 / REVIEWS OF GEOPHYSICS 3.2.4.7. Along-Shore Variations in the SBF
[74] The cross-shore gradients of potential tempera-ture, water vapor, and wind components associated with the SBF usually vary in magnitude in the along-shore direction. On a nearly homogeneous coast these varia-tions are about 10 times smaller than the variations in the cross-shore direction. Along-shore variations in the SBF increase as the coastline becomes more complex.
One cause of these variations is collisions between the SBF and convective elements in the continental air mass
[Stephan et al., 1999]. The interaction between sea breeze fronts and convective activity is discussed in greater detail in section 3.3.
Figure 10. Australian Morning Glory, as seen from the wing 3.3. Interactions With External Meteorological of a glider -1 km above ground level. (Photograph courtesy of R. White.)
Phenomena
[751 The SBS interacts with external meteorological phenomena at all stages of its life cycle. Within a short period of time following sunrise on days conducive to the formation of the SBS, cellular convection begins over
[72] Simpson et al. [1977] suggest that a cutoff vortex land in response to surface insolation and low-level sometimes forms in southern England, shortly before destabilization [Mitsumoto et al., 1983]. The cells have an sunset, but the most dramatic example of the phenom- aspect ratio of 1:1 and horizontal scales of 1-4 km and enon is the Morning Glory in the Gulf of Carpentaria are sometimes visible as fair-weather cumulus [Lyons, region of northern Australia [e.g., Smith et al., 1982]. 1972; Mitsumoto et al., 1983; Stephan et al., 1999]. A The Morning Glory occurs around sunrise, during the mixed layer of uniform thickness forms over land as a latter half of the austral dry season (mid-August through result of the convection, and the smaller cells merge into mid-November), and is associated with the sea breeze of larger ones and attenuate [Mitsumoto et al., 1983]. The the previous day. In addition to an abrupt pressure SBG moves into the well-mixed layer and interacts with increase (sometimes exceeding 1 hPa in a few minutes) the onshore convective disturbances, which, in turn, con-and shift in wind direction it is usually accompanied by tribute to variations in the SBF's shape and speed one or more roll clouds roughly 1 km in height and many [Stephan et al., 1999].
kilometers long (Figure 10), as well as intense wind [761 The SBS may also interact with HCRs that de-squalls at the Earth's surface [Smith et al., 1982]. velop in the convectively unstable continental air mass
[73] Numerical model studies have determined two [Atkins et al., 1995]. During periods of ambient offshore methods by which the sea breeze may contribute to the flow the HCR rotation axes are nearly perpendicular to formation of undular bores. When two SBFs collide, or the SBF. As an HCR collides with the SBF, its axis of when one is overtaken by another, a hump of marine air rotation is lifted upward from updrafts associated with is generated, eventually resulting in the conditions fa- the front. During periods of ambient onshore flow the vorable for the formation of bores. In this case, two HCR rotation axes are approximately parallel to the undular bores are created that move with approximately SBF. In this case, collisions between HCRs and the SBF the same velocities as the original SBFs [Clarke, 19841. constitute the merging of the two like-sign vortices and Undular bores may also be created when the sea breeze result in the strengthening of the SBF. Both scenarios penetrates inland and interacts with a nocturnal temper- result in the enhancement of cumulus clouds at the ature inversion [Sha et al., 1993]. The interaction causes points of collision [Atkins et al., 1995].
the SBH to detach and move inland, cut off from the [77] If the SBS develops near a coastal city, it will low-level feeder flow. Eventually, the cutoff vortex dis- interact with the city's urban heat island (UHI). The sipates from inertial damping and radiative energy loss presence of the UHI increases the sea breeze velocity
[Sha et al., 1993]. Air pollutants emitted from ground during its growing stages [Yoshikado, 1992], and the sources within the marine layer can become trapped acceleration increases as the size of the inland urban within the closed circulation behind the leading edge of area increases [Ohashi and Kida, 2002]. The UHI also the cutoff vortex [Kitada, 1987]. The vortex then serves causes the SBC to last longer [Ohashi and Kida, 2002],
as a mechanism that transports the pollutants inland. and the interaction may result in unusually regular and Downward vertical currents at the rear of the vortex persistent lines of clouds, such as the Kampachi Street move pollutants from the top of the PBL to the points cloud line in downtown Tokyo [Kanda et al., 2001].
closer to Earth's surface, even in thermally stable con- [78] The SBS may interact with other thermally driven ditions [Kitada, 19871. (The sea breeze's influence on air breezes, generated at points inland. Zhong and Takle quality is discussed in greater detail in section 3.6.) [1992] conducted an observational study of SBS interac-
41, 3 / REVIEWS OF GEOPHYSICS Miller et al.: SEA BREEZE
- 1-17 tions with river breezes in eastern Florida. The Indian gravity current develops a raised head [Finkele et al.,
River is a shore-parallel feature about 10 km inland, and 1995].
it produces the largest of the river breezes in their study [83] During stage two, early mature, insolation begins area. These authors found that the collision of the In- to decline, and the KHBs decay, removing the top fric-dian River breeze (spreading outward perpendicular to tion and allowing the inland progression of the sea the river's axis) with the Atlantic sea breeze creates breeze to accelerate [Clarke, 1984; Sha et al., 1991; areas of enhanced low-level convergence on both sides Buckley and Kurzeja, 1997]. During stage three, late of the river. Smaller water features nearer the coastline mature, insolation drops toward zero, the land-sea ther-produce minor atmospheric circulations that interact mal difference driving the SBC vanishes, the magnitude with the sea breeze, enhance low-level convergence, and of the resulting landward force vector drops to zero, and speed up the inland movement of the SBF [Zhong and supply of new marine air to the head of the sea breeze is Takle, 1992]. cut off. However, the SBF remains sharp, and the center
[79] The low-level landward flow associated with the of the SBC cell shifts from a point near the coast to the SBS interacts with the offshore temperature inversion landward edge of the marine air mass [Clarke, 1984].
separating the marine boundary layer from the air above After sunset, radiationalcooling increases, which reduces it [Barkan and Feliks, 1993]. A field study in a harbor vertical mixing and the associated updrafts and reduces south of Tel Aviv, Israel, found that the marine inversion the height of the SBH. The SBH continues to move moved downward approximately adiabaticallyduring the inland [Clarke, 1984].
day, when the SBC was operating, and moved upward [84] During stage four, early degenerate, the SBH, during the night, when the land breeze was in effect now separated from the feeder flow and continuing to
[Barkan and Feliks, 1993]. Model studies have shown move inland independently, may interact with a night-that the portion of the marine inversion that varies time radiational inversion or other low-level features diurnally extends >100 km offshore [Feliks, 1993]. and form a sea breeze cutoff vortex or undular bore
[80] The SBS also interacts with synoptic-scale fea- [Clarke, 1984]. The most noted example of this phenom-tures. Briimmer et al. [1995] observed the interaction enon is the Morning Glory, which is discussed in section between the SBS and a synoptic-scale cold front (SSCF) 3.2.4.6 [Smith et al., 1982; Simpson, 1994]. Kitada [1987]
that approached the northern coast of Germany from found that the undular bore can act as a pollutant the North Sea. Before the relatively weak, shore-parallel transport mechanism, with descending currents at the SSCF arrived at the coast, the SBC was initiated be- rear of the cutoff vortex moving pollutants aloft down to tween the marine air mass ahead of the SSCF and the Earth's surface.
continental air mass on shore. At about noon, as the [85] In the final stage, late degenerate, the circulation SSCF approached the shore from the northwest, the in the SBH is no longer closed [Clarke, 1984]. The SBF moved inland ahead of it. The onshore wind behind leading edge flattens out, and Coriolis force rotates the the SBF and ahead of the SSCF was so strong that the flow and limits further penetration inland. At this stage SSCF was subjected to a divergent wind field, resulting the remnant of the sea breeze may also interact with in the forward transfer of the thermal contrast associ- developing nocturnal features, such as low-level jets and ated with the SSCF to the SBF. At points inland later in other gravity flows [Clarke, 1984; Garratt and Physick, the day, a single very strong frontal passage was ob- 1985; Buckley and Kurzeja, 1997].
served [BrUmmer et al., 1995].
3.5. Forecasting 3.4. Daily Life Cycle [86] The sea breeze is a standard forecast problem in
[81] The life cycle of the SBS consists of five stages: coastal locations. Weather forecasters usually work un-immature, early mature, late mature, early degenerate, der rather severe time constraints and therefore prefer and late degenerate [Clarke, 1984; Buckley and Kurzeja, simple, straightforward methods to complicated meth-1997]. These five stages apply to all three of the sea ods. There are several SBS forecasting techniques in the breeze categories described above. The land breeze, a literature for specific locations that rely on meteorolog-near mirror of the sea breeze, evolves through a similar ical variables included in standard surface and upper air sequence of stages [Holmer and Haeger-Eugensson, observations. Unfortunately, local irregularities in the 1999]. coastline and inland topography can have a strong effect
[82] During stage one, immature, the SBC begins as a on the SBS, so techniques developed for use in one divergence in the cross-shore wind component over the location may not work well in another [e.g., McPherson, sea in response to the local-scale thermal PGF [Clarke, 1970; Simpson, 1994; Miller and Keim, 2003].
1984]. The SBC expands more rapidly seaward than [87] Forecasting the sea breeze consists of at least landward [Finkele et al., 1995]. Marine air moves on- three major components: (1) a simple yes/no (on shore as a gravity current [Simpson, 1997], and the whether or not it will occur), (2) the wind speed and leading edge takes on frontal characteristics [e.g., Kraus direction, and (3) the distance of inland penetration et al., 1990; Reible et al., 1993]. Convergence in the during the day. The first question, whether or not a sea cross-shore wind component develops over land, and the breeze will be observed at a given location, is also a
1 -18
- Miller et al.: SEA BREEZE 41, 3 / REVIEWS OF GEOPHYSICS function of the last question. All three are discussed critical value of 7. Once again, the difference between separately in sections 3.5.1, 3.5.2, and 3.5.3, respectively. Simpson's critical value and that reported by previous authors is attributed to the method used for obtaining 3.5.1. Occurrences JU1. In this case, rather than an observed surface wind or calculated geostrophic wind speed, Simpson [1994] used
[88] A number of empirical studies examined the oc-the 1000-m wind speed obtained with a pilot balloon currence of the sea breeze as a function of land temper-ascent.
ature, land-sea temperature difference, the strength of
[92] About 60% of the forecast error by the L-B index the opposing background (synoptically driven) wind, or some combination of these. The simplest approach uses is associated with the failure of the expected lake breeze to occur; that is, it overpredicts the lake breeze [Biggs the temperature information only. For example, Mc-and Graves, 1962; Lyons, 1972]. Laird and Kristovich Kendry and Roulet [1994] reported that all sea breezes
[2001] examined this and related problems with the observed on the western shore of James Bay were asso-ciated with inland temperatures >20'C. index using a 15-year climatology of the region sur-
[89] For it to be detected inland, the sea breeze must rounding Lake Michigan. They tested three variations on the L-B index, by computing it as originally described often blow upwind in opposition to the ambient flow.
Biggs and Graves [1962] developed an early lake breeze by Biggs and Graves [1962] (using JUI, the wind speed irrespective of wind direction), by substituting the ob-forecasting index that compares the background wind to served cross-shore wind component U, for U, and by the land-sea temperature contrast:
substituting IUJ, the magnitude of U(, for IUI. The best results are obtained using U*, and the worst results are lCuAT (14) obtained with U].
[93] Laird and Kristovich [2001] also reported three where , is the lake breeze (L-B) index and represents the additional important findings. The first is that, although ratio of the inertial force (pU 2/2) and buoyancy force the lake breeze is driven by land-lake temperature dif-(pg3AT, where P3 is the coefficient of expansion of air). ferences (AT), large values of AT are not required, and UIgis the near-surface wind speed [m s-'], Cp is the in 70% of the lake breeze events in their climatology, specific heat coefficient of dry air at constant pressure daytime maximum AT values were <12'C. Simpson (1.004 J g-' K-'), and AT is the difference between the [1994] reports that temperature differences as small as inland air temperature and the temperature of the water 5°C are sufficient. A related result is that the lake breeze surface (TIand - Twater) ['C]. Both 1U1 and AT are deter- is always associated with relatively light ambient wind mined by using a point far enough inland so that it is not conditions; for example, for 95% of all lake breeze affected by the sea breeze [Simpson, 1994]. If s is large, events the average daytime wind speed for the unaf-then the inertial force is large, and th'ere will not be a fected inland station was <-5 m s-1. Three quarters of all lake (sea) breeze. If e is small, then the buoyant force is lake breeze events occurred when the cross-shore wind relatively large, and the lake (sea) breeze can move component had a magnitude (IUxI) of 2 m s-1 or less.
inland. Using a site on the west end of Lake Erie, Biggs The third additional finding is that, in agreement with and Graves [1962] found that the critical e value is 3.0. If Walsh [1974], changes in location alone do not signifi-a transition zone between 2.7 and 3.2 is permitted, the cantly affect the critical value of the index, meaning that L-B index correctly predicts the occurrence/nonoccur- for locations with similar topography and coastline rence of the lake breeze 97% of the time [Biggs and shape the same critical L-B index value applies.
Graves, 1962].
[90] Lyons [19721 recalibrated the L-B index for Chi- 3.5.2. Surface Wind cago, Illinois (on the southwest shore of Lake Michi- [94] Once the forecaster has decided a sea breeze will gan), using a climatological record obtained during 10 occur, the next questions are of speed and direction.
summer months over a 2-year period (1966-1968). There are relatively straightforward methods for quan-Rather than using an observed wind recorded at an titatively estimating both of these in the literature. Speed inland surface site for the value of U], Lyons [1972] and direction are considered separately in sections calculated the geostrophic wind speed using the 1200 3.5.2.1 and 3.5.2.2.
UTC surface analysis. He reports a critical F of 10.0, above which the lake breeze does not occur, which 3.5.2.1. Speed accurately forecasts the lake breeze 90-95% of the time. [95] One method of estimating the wind speed asso-The difference between the critical value of Lyons [19721 ciated with the sea breeze is the Bjerknes circulation and that reported by Biggs and Graves [1962] is attrib- theorem, but it has already been shown that this greatly uted to the method used to determine LX. overestimates the surface wind (see section 3.1.2.). An
[91] Lake breezes and sea breezes are driven by the alternative approach is to develop climatological records same mesoscale, thermally induced pressure gradients. of the strength of the sea breeze and the temperature Simpson [1994] repeated the L-B index calculations for contrast between the inland air and the sea surface and the sea breeze at Thorney Island, Britain, and found a then fitting a regression line relating the two data sets.
41, 3 / REVIEWS OF GEOPHYSICS Miller et al.: SEA BREEZE 9 1-19 This is very straightforward, although it obviously must k 08 p gd (17) be repeated for each forecast site. Mathews [1982] de- us'? ý 0 .8 7 -pd veloped a forecast rule for a Royal Australian Naval Air Station (RANAS) using this method. From a climatol- where k is a constant equal to 0.62 and d (the height of ogy of surface observations recorded over two summers the SBH) can be replaced by a climatological mean (d) at the Nowra RANAS, southeastern Australia, Mathews without the loss of significant accuracy [Simpson, 1969].
[1982] selected those days when the only important contribution to the 10-m wind was from the sea breeze. 3.5.2.2. Direction Wind speed was plotted as a function of the difference [99] Two preliminary questions of interest are the between the air temperature at the RANAS and tem- following: (1) How does the surface wind direction perature of the nearby ocean surface. A regression line change from the synoptically driven direction to the sea resulted in breeze direction? (2) What is the initial direction of the sea breeze once it is established? McKendry and Roulet
[1994] addressed the first question in a field study con-IU = ' T, (15) ducted on the southwest shore of James Bay, where the coastline is fairly straight and oriented from south-where JUI is the magnitude of the sea breeze component southeast to north-northwest. They report that with a
[knots] and AT is the land-sea temperature difference synoptic-scale wind that has a southerly component, the
[oC]. sea breeze is associated with backing (counterclockwise
[96] Another approach is to relate wind speed within rotation) of the wind diiection from the southern to the the marine air mass to the speed of an advancing of SBG eastern quadrant. For a synoptic-scale wind with a
[e.g., Simpson and Britter, 1980]. Keulegan's [1957] rela- northerly component the sea breeze is associated with tion (5) describes the rate at which a gravity current veering (clockwise rotation) of the wind from the north-advances into the ambient fluid. Simpson [1969] adapted ern to the eastern quadrant.
this for forecasting the rate at which the SBG advances [100] McKendry and Roulet's [1994] results are logical inland, and Simpson and Britter [1980] modified the when considered in the context of simple vector addi-relation to account for the opposing prevailing wind tion. Assuming that the Coriolis force is unimportant (at (equation (6)) and determined the relationship between least initially) in the sea breeze, then the mesoscale flow the speed of the advancing sea breeze front and the wind will be parallel to the thermally induced pressure gradi-velocity ahead of and behind it: ent toward lower pressure. In the case of a relatively uncomplicated coastline this implies a sea breeze wind Ufront - 0.8 7 usB - 0.59u., (16) vector directly perpendicular to the coast. For the net where Ufront is the speed at which the sea breeze ad- wind direction at any location, one must add the sea vances inland, Us, is the wind speed within the marine breeze wind vector to the synoptically produced wind air mass, and ug is the cross-shore geostrophic wind vector to determine the net wind. In the case of a component resulting from the synoptic-scale pressure northerly synoptic-scale wind the growing sea breeze pattern [Reible et al., 1993]. The two constants in equa- vector will gradually deform the net wind vector seen by tion (16) were determined by Simpson and Britter [1980] a stationary observer in such a way that the net appears in laboratory studies using tanks of water. to rotate in a clockwise direction. With a southerly
[97] Finkele [1998], from a model study compared synoptic-scale wind the opposite is true.
with in situ observations from aircraft, reported average [ioi] Mathews [1982] discussed vector addition in a speeds for the growth of the SBC of 1.6 m s-1 in the practical technique for forecasting the direction of the landward direction (slowing somewhat in the middle of sea breeze once it is established. The strength of the the day) and compared this to a typical value of 3.4 m cross-shore sea breeze component is calculated by one of s-1 in the seaward direction. The former is comparable the methods discussed in section 3.5.2.1, and its direc-to Ufront. Simpson et al. [1977], from field studies, re- tion is assumed to be directly perpendicular to the local ported that a typical value for Ufront in England is about coastline, with some small degree of variation. Then the 3 m s-', and Clarke [19551 reported Ufront values as high sea breeze vector components are added to the gradient as 7 m s-1 in Australia. Several studies have reported wind components, and the resultant wind is determined.
variations in the speed of the front's inland propagation, [102] Once the sea breeze is established, the associ-slowing during the midday hours and regaining speed ated wind direction may continue to rotate as the day toward late afternoon, which is attributed to the devel- progresses. Several field studies report a diurnal rotation opment of Kelvin-Helmholtz billows along the top of the in the sea breeze [e.g., Pearce, 1955; Fisher, 1960; Zhong inland marine air mass and variations in soil moisture and Takle, 1992; McKendry and Roulet, 1994]. Within the
[Physick, 1980; Sha et al., 1991]. same hemisphere north or south of the equator the
[98] Assuming that equation (6) represents the scalar sense of the rotation may be either clockwise or the speed at which the SBF moves inland, one can combine reverse and therefore cannot be accounted for by the equations (6) and (16) to obtain Coriolis force alone. Neumann [1977] derived an expres-
1-20
- Miller et al.: SEA BREEZE 411, 3 / REVIEWS OF GEOPHYSICS Table 2. Varying Extent of the Sea Breeze's Inland Penetration in Different Parts of the World Where Distance Inland, km Reference Beaufort Sea coast, Alaska, United States 15 Kozo [1982]
Chicago, Illinois, United States (lake breeze) 40 Lyons [1972]
Indonesia 60-80 Hadi et al. [2002]
Southern England 100 Simpson et al. [1977]
James Bay, Ontario, Canada 100 McKendry and Roulet [1994]
Southeastern United States 150 Buckley and Kurzeja [1997]
Southeastern Spain 150 Kottmeier et al. [2000]
Australia 200 Simpson [1994]
sion describing the time evolution of the wind direction [104] Consider the east coast of a continent in the associated with the sea breeze for a north-south oriented Northern Hemisphere with no important variations in coastline. The x and y axes correspond to the cross- and the shape of the coastline and no significant topography.
along-shore dimensions, respectively, and the expression If Coriolis clearly dominates, equation (18) becomes takes the following form:
=-fI (20) at f+1V1 [pF vapm+
-x+f(uug + vvg), (18) at and the wind direction rotates clockwise (anticycloni-where the LHS describes the rate and sign of the rota- cally). Once rotation begins, then v (the along-shore tion [s-1] and the three terms on the RHS are associated component of the net wind vector) is nonzero. If the with (from left) (1) the Coriolis force, (2) thermally synoptic-scale pressure field is exceedingly weak, then induced mesoscale PGF, and (3) synoptic-scale PGF. the two components of the geostrophic wind are approx-The angle between the wind vector and x axis is repre- imately zero, and equation (18) becomes sented by ct [radians], f is the Coriolis parameter (=
21lsiny, where Dl is the rotational velocity of Earth and act V IxapJ (21)
(pis latitude) [s-l], JUl is the scalar wind speed (net) [m at - -f pul I ax I s-1], u and v are the cross- and along-shore components With the ocean on the right, one can assume that in a sea of the total wind vector [m s-1], respectively, ug and vg breeze environment, apm/ax is positive. Therefore, ignor-are the cross- and along-shore components of the ing the Coriolis force for the moment, a positive (north-geostrophic wind vector [m s-1], respectively, p is density ward) along-shore component (v) implies counterclock-
[kg m-3], and apm/ax is the mesoscale pressure gradient wise (cyclonic) rotation, and a negative along-shore resulting from the cross-shore temperature contrast [Pa component implies clockwise (anticyclonic) rotation. Put mI]. Note that the third term on the RHS of equation another way, if the locally induced wind has a southward (18) can also be written as component, then the rotation imposed on the wind di-f(uug + vvg) 1 (apL aPL (19) rection will be counterclockwise, but if the locally in-duced wind has a northward component, then the two IUI2 - pIUj2 axa V ay U), terms in the RHS of equation (21) are at odds, and the where apL/ax and apL/y represent the synoptic-scale pres- net rotation imposed on the wind direction will depend sure gradients in the cross- and along-shore dimensions, on their relative magnitudes. Given that density (p) is respectively. A scale analysis reveals that the three terms nearly constant (-1.2 kg m- 1 [List, 1968]) and f is on the RHS of equation (18) are within an ord&r of constant for a fixed latitude, the terms that one must magnitude of each other. Using data recorded in Nova determine are v, IU], and apmOax.
Scotia, southern Australia, Washington state (United States), Scotland, and Great Britain, Simpson [1996] 3.5.3. Extent of Inland Penetration concluded that terms 1 and 2 are of comparable impor- [105] Forecasting whether the sea breeze will occur at tance but that topographical influences can often be an inland location is a problem of combining guidance responsible for completely reversing the rotation im- from an index, such as equation (14), with an additional posed by all three terms. estimate of how far inland the sea breeze will penetrate
[103] If the synoptic-scale pressure field is nonzero at during the day. Field studies have documented the ex-the forecast location, then the best approach for deter- tent of the sea breeze's inland penetration in different mining the rotation imposed on the wind is to calculate parts of the world (Table 2), but determining a set of the values of all three terms on the RHS of equation generic governing principles is a very complex problem.
(18). For a few special cases, however, the solution is By combining the results of several studies, Simpson somewhat simpler. [1994] found that the magnitude of the ambient cross-
41, 3 / REVIEWS OF GEOPHYSICS Miller et aL: SEA BREEZE
- 1-21 shore wind component is an important control. With static stability. Conversely, where onshore synoptic-scale winds, inland penetration is lim-ited to 30 or 40 kin, possibly because the associated f < (o (23) land-sea temperature contrast is weak. Conversely, an the sea breeze is in the form of internal waves that offshore wind of 7 m s-.' at 1000 m AGL limits inland extend an "infinite" distance inland. In equations (22) movement to 20 km [Simpson, 1994]. and (23), f is the magnitude of the local Coriolis param-
[106] Chiba et al. [1999] conducted a field study utiliz- eter (= 2flsinp), and w is the cycle of heating and ing an instrumented helicopter and expanded on Simp- cooling (= 27r day'). fl is the angular rotation of Earth, son's [1994] general principles. From the results of sev- and (p is the latitude. It follows that the dividing line is eral sea breeze case studies in Japan, Chiba et al. [1999] where 4-rrsinp day-' - 27r day- 1 , or sinp ->1/2, which is determined that the inland intrusion distance at midday true if yp - 30'. For regions at and above 30' latitude, is between 10 and 25 km, depending on the magnitude of Rotunno [1983] concluded that the range of sea breeze's the cross-shore synoptic-scale flow, atmospheric stratifi- inland extent (L) near the ground is defined by cation, and geographical features in the area. Specifi-cally, they noted the following: (1) With a strong off- Nh shore gradient and an unstable coastal layer the sea L (24) breeze's inland penetration is limited to 10 km; (2) when the atmosphere is stably stratified and the ambient flow where N is the Brunt-Vaisala frequency (=
is onshore at 6 m s-', the sea breeze reaches as far as 15 \/(g9/)(l0/az)) [s-'] [Rogers and Yau, 1989] and h is the km inland; and (3) with high-pressure and light, offshore vertical scale of the heating [m]. Dalu and Pielke [1989]
850-hPa winds overhead the sea breeze reaches as far as stated that friction limits both intensity (speed) and the 25 km inland. inland extent. At the equator, where the Coriolis force is
[107] Field studies produce excellent results on the zero, the sea breeze is entirely limited by friction. Within local behavior of the sea breeze, but computer models 300 of the equator the "infinitely propagating" waves are better able to differentiate the many overlapping suggested by Rotunno [1983] may not appear because physical phenomena in the SBS and therefore are a friction limits the sea breeze. By combining conclusions better tool for determining the generic principles gov- of the studies noted above with results of McPherson erning the sea breeze's inland extent. Some model stud- [1970], Asai and Mitsumoto [1978], Mitsumoto et al.
ies focus on various phenomena that reach inland prior [1983], Sha et al. [1991], Zhong and Takle [1992], Arritt to the arrival of the sea breeze proper. Tijm and van [1992], and Melas et al. [1998], one can construct a Delden's [19991 sound waves reach over 1000 km inland. summary of the controls on SBC inland propagation, Geisler and Bretherton's [1969] sea breeze forerunner shown in Table 3.
may reach as far inland as .60 km. Most modelers have focused on the SBC or the SBG. 3.6. Air Quality Impact Mechanisms
[108] Estoque [1962], in a model study assuming a [11o] Meteorological and topographical features in straight coastline and a hydrostatic atmosphere, con- coastal regions create complex low-level wind circula-cluded that an ambient offshore wind of 5 m s-1 limits tions. High concentrations of air pollutants are released SBC inland penetration to 18 km, while similar thermal into these circulations by industrial facilities (e.g., pet-forcing with calm ambient winds permits SBC inland rochemical plants, nuclear power stations, etc.), which penetration of 32 km. The former result closely parallels are typically in or near urban areas and are often located Simpson's [1994] observation-based conclusions. In a near coastlines. Understanding the downwind behavior comprehensive series of model studies, Arritt [1992] of these pollutants could assist in mitigating significant noted that an offshore ambient wind of -1 m s-1 was health problems [e.g., Shearerand Kaleel, 1982; Clappier conducive to the deepest inland, penetration, peaking at et al., 2000]. The sea breeze has been implicated as
>60 km. Finkele [1998] reported that the inland extent either a direct control [e.g., Kitada, 1987; Hsu, 1988; of the sea breeze is greater under light (2.5 m s-') Physick and Abbs, 1992] or as one contributing factor to ambient offshore wind conditions than in moderate (5 m more complex air quality scenarios [Gaza, 1998; Seaman s-1) ones, but the seaward extent is similar in both and Michelson, 2000; Gangoiti et al., 2002].
situations. For strong offshore geostrophic winds (7.5 m [1n] One example of a direct control is Kitada's s- 1) the SBC is entirely over the sea. [1987] inland transport of pollutants by the postfrontal
[109] Pearce [1955], using an early 3-D numerical roll vortex. Another example is the convective internal model, determined that inland motion of the sea breeze boundary layer, also known as the thermal internal is ultimately limited by Coriolis-induced rotation. Ro- boundary layer. Cool, stable marine air encounters a tunno [1983] expanded on this, stating that where thermodynamically unstable situation when it advects f ,(22) over a hot land surface. As the marine air mass moves inland and conducts heat from the land surface, convec-the sea breeze is confined to a finite distance from the tive currents begin to develop, and modified air is trans-coast determined by the vertical scale of heating and the ported vertically. As the air moves farther inland, con-
1-22
- Miller et al.: SEA BREEZE 41, 3 / REVIEWS OF GEOPHYSICS Table 3. Summary of Controls on the Inland Penetration of the Sea Breeze Control Description Magnitude and direction of the Seaward wind component of 7 m s-' will limit inland penetration to 10 km or less.
synoptic-scale cross-shore wind Light seaward wind component or calm ambient wind is most favorable for deep component penetration. Seaward ambient wind component of 1 m s-' may result in inland propagation of more than 60 km. Strong shoreward ambient wind is associated with weaker land-sea temperature contrasts and limits inland propagation to -40 km.
Land-sea temperature contrast There must be a positive land-sea temperature difference. Very strong temperature differences are not required. Minimum of 5°C is required for "deep" penetration.
Coriolis Coriolis prevents SBC from extending inland "to infinity" by rotating associated wind until it is parallel to the coast.
Surface friction Surface friction slows rate of inland movement and therefore extent SBC may reach before sunset, when thermal forcing is removed.
Top friction (KHBs) KHBs slow rate of inland movement. KHBs form on SBH in low-stability conditions and propagate backward (toward sea) along top of SBG.
Thermodynamic stability in the Sea breeze may initially advance more readily into a well-mixed ambient air mass (often continental air mass marked by fair-weather cumulus).
Instability also gives rise to KHBs, which slow sea breeze's inland movement and therefore reduce the degree of penetration.
Complex relationship between stability and penetration is discussed by Rotunno [1983],
who predicts greater inland penetration in high-stability atmospheres but also in atmospheres with large vertical heating scale.
Shape of coastline Inland bays along an otherwise straight coastline provide low-friction pathways for sea breeze to advance inland.
Topography Terrain features can either assist or obstruct sea breeze's inland movement.
If sea-facing hillside is relatively close to coast and is heated on the same diurnal cycle as adjacent coastal plain, resulting uphill wind current can assist the sea breeze's inland movement. If hill is in shadows, or shrouded in fog or clouds, it acts as an obstruction.
Valleys funnel sea breeze, accelerating associated wind speeds, which carry marine air farther inland than would be the case over comparatively featureless terrain.
Interactions with other mesoscale If sea breeze encounters other small-scale circulations, it is modified, and extent of its systems inland penetration may be either increased or reduced. For very weak sea or lake breeze, collision with thunderstorm downdraft may be enough to completely stop inland progression.
Sea breeze's speed and direction of movement (and therefore its inland penetration) may be modified by collisions with other sea breeze systems, lake breezes, or river breezes generated at points inland.
vective currents reach progressively greater heights. The 1998]. Plumes mixed to the surface may be vertically vertical overturning results in a CIBLnear Earth's sur- transported hundreds of meters aloft in updrafts associ-face, whose upper limit increases nonlinearly with dis- ated with convection or lines of low-level convergence, tance from the coast. Above the CIBL the remaining then divided into multiple branches which move in many unmodified marine air acts as a cap that prevents mixing different directions. Some may eventually be caught up between destabilized marine air below and the continen- in the seaward return flow, transported many tens of tal air above [Hsu, 1988; Stull, 1988; Prabhaet al., 20021. kilometers offshore, and recirculated. landward within
[112] The CIBL is frequently the cause of persistent the SBC cell. The pollutant concentration within the air pollution problems in coastal cities. The onset of the SBC can grow throughout the day as fresh emissions are CIBL results in a dramatic reduction of the atmo- fumigated downward into the CIBL, adding to older sphere's mixing depth, and locally generated pollutants pollutants already in the SBS [Shair et al., 1982; Panelon trapped in the shallow surface layer can quickly reach Coastal Meteorology, 1992; Abbs and Physick, 1992].
unhealthy levels (Figure 11) [e.g., Barbato, 1975]. Pol- [113] The variation in the height of the CIBL with lutants such as sulfur dioxide (SO 2) and ozone (03) can distance from the coast is the subject of several field be rapidly mixed to the surface when they encounter the studies. Raynor et al. [1979] studied the CIBL over Long convective currents within the CIBL [Barbato, 1975; Island, New York, under a variety of prevailing wind and Gangoiti et al., 2002]. Fumigation occurs when air pollu- stability conditions. They found that the height of the tion plumes emitted in stable regions of the marine air CIBL (as a function of cross-shore distance) increases mass encounter the top of the CIBL and are rapidly sharply near the coast but increases more gradually mixed down to the surface (Figure 12) [Lyons et al., inland. Druilhet et al. [1982], using an instrumented air-1981; Hsu, 1988; Abbs and Physick, 1992; Sawford et al., craft, found that height of the CIBL increased as a
41, 3 / REVIEWS OF GEOPHYSICS Miller et a].: SEA BREEZE 9 1-23 70T 60+
Onset 50-40-
.0 C.,
30-20+
10-Retreat 0
0 '01'02'03'04'05'06'07'08'09' 10'11* 12'13'14'15' 16'17'18' 19'20'21 '22'23 '24 Local Hour Figure 11. Atmospheric S0 2 concentration levels at two stations in Boston, Massachusetts. The concentra-tion of SO, jumps dramatically following the passage of the sea breeze front (SBF) (indicated by "onset")
because of fumigation induced by the convective internal boundary layer (CIBL). Plots are based on hourly SO 2 observations from September 1972 [after Barbato, 1975].
power of 0.4 of the cross-shore distance along the [114] In general, CIBL heights (h) grow as a function French Mediterranean coast. Melas et al. [2000] con- of the square root of distance from shore (x), but there ducted a study of sea breeze behavior on the island of are many modifying factors that scale the rate of growth.
Sardinia and reported CIBL heights of -50 m near the From first principles, Venkatram [1977] derived coastline, which increase gradually to over 500 m several kilometers inland. Prabha et al. [2002] reported CIBL h 2CD(Oland
= - 0sea)X heights between 150 and 200 m AGL, measured by a N = y(1 - 2F) (25) mini-SODAR positioned 5 km inland in the tropics. Liu and Chan [2002] studied internal boundary layers over where h and x are in meters, C, is the drag coefficient ("-
Hong Kong and reported CIBL heights of up to 700 m 12 x 10-3), ", is the lapse rate above the boundary layer AGL, reaching almost to the top of mountains in the or over the water surface [K m- 1 ], F is the entrainment complex terrain of the island. coefficient (0-0.22), and 01and and 0,ea are the potential temperatures over the land and sea surfaces [K], respec-tively. Hsu [1988] evaluated the effectiveness of equation (25) by utilizing data from several field studies and Fumigation reported that the observed CIBL heights and the theo-Phenomenon retical predictions are in substantial agreement.
[1s] Melas and Kambezidis [1992] evaluated equation (25) and several alternative models using data observed in Athens, Greece. They reported that the following relation, based on Gryning and Batchvarova [1990], is the most accurate predictor of CIBL height:
h = (1 - 0.002La) !2 CDCT(! + 2A) X xAO/y
+ llln (-La), (26) where CT is heat flux coefficient (-- CD) and A is a Figure 12. Convective internal boundary layer formed by constant approximately equal to 2. La is given by advection of cool marine air over a warm land surface. Here "h" is the depth of the CIBL and varies as a function of 2 distance from the coast (equations (25) and (26)). Details are W(CoUh)
La (27) given in text [from Hsu, 1988]. Reprinted with permission from k Tgo--CAO' Elsevier Science.
1-24 9 Miller et al.: SEA BREEZE 41, 3 / REVIEWS OF GEOPHYSICS where Uh is the cross-shore wind component at the upper [1987] describes the Appalachian lee trough (APLT),
part of the CIBL [m s-1], k is the von Kirman constant which forms along the eastern seaboard when the 700- to (0.40), g is the gravitational acceleration (9.81 m s-2), To 500-hPa flow is southwesterly to northwesterly. The is the surface temperature [K], and A0 is 0 land - 0sea [K] APLT causes the winds below 850 hPa to turn south-
[Melas and Kambezidis, 1992]. Luhar et al. [1998] evalu- westerly, resulting in shore-parallel flow along the east ated equation (26) using data recorded on the east coast coast, which, in turn, results in long-distance transport of of Australia. They found that it effectively predicted the 03 and its precursors from the northeast urban corridor observed CIBL height unless the atmosphere was neu- to rural areas farther up the coast [Pagnotti, 1987].
trally stable, then a more complex model was needed. [120] The role played by the SBF in this scenario is
[116] A number of additional studies have advanced documented in a case study of a July 1995 03 pollution the understanding of how the CIBL evolves. For exam- event [Gaza, 1998]. Gaza [1998] used hourly observa-ple, Raynor et al. [1979], using an instrumented aircraft, tions of ground level 03 concentrations at three loca-found that the time rate of growth of the CIBL was tions on a line running north from New York City. 03 initially slower with stable lapse rates over the sea than concentrations spiked at each location as the SBF passed with neutral or unstable lapse rates. They also found that through (Figure 13). A horizontal contour analysis of 03 the height (h) of the CIBL does not grow to infinity with concentrations revealed a front-parallel region of highly distance inland (as suggested by equations (25) and (26)) polluted air moving inland with the SBF. From addi-but reaches an equilibrium height that is dependent on tional case studies in the same region it became appar-downwind (inland) conditions. A higher CIBL is associ- ent that higher ground level 03 levels occurred on days ated with low wind speeds and stronger land surface when the SBF was associated with well-defined low-level heating [Raynor et al., 1979]. Federovich et al. [2001], convergence. The highest 03 levels occurred when the SBF combining wind tunnel and numerical models, reported moved inland and merged with the APLT [Gaza, 1998].
that the sign of the wind shear at the top of the CIBL affects the CIBL's rate of growth. Positive shear, where the wind momentum above the top of the CIBL is 4.
SUMMARY
greater than within the CIBL, impedes the growth of the CIBL, when compared to the case when no wind shear is [121] The sea breeze is a mesoscale wind that occurs at present. Negative shear, where the wind momentum many coastal locations throughout the world. It develops within the CIBL is greater than above it, enhances the when solar radiation and the .differing rates at which growth of the CIBL. land and water change temperature create a mesoscale
[117] A third example of a direct control by the sea pressure gradient force pointing toward the land. Cool breeze on air quality was studied by Physick and Abbs marine air moves toward the land as a gravity current, is
[1992]. They observed the behavior of pollutants re- lifted vertically at the sea breeze front, and often forms leased by power stations in the Latrobe Valley, south- a closed circuit via a return flow aloft and a diffuse eastern Australia, over a 5-day period. The east-west region of descending currents several tens of kilometers Latrobe Valley extends more than 130 km inland and is out to sea.
bordered by mountains as high as 2000 m. An easterly [122] There are historical records documenting hu-sea breeze regularly penetrates >100 km up the valley, man understanding of the sea breeze system going back where it collides with an SBC moving inland from the almost 2500 years, and it is still a rapidly evolving field.
south in the late afternoon. Observations indicated that The SBS has been intensely studied since the beginning the associated marine layer was -1500 m deep. Westerly of the twentieth century, and more than 500 new papers winds prevailed above 3000 m AGL during the study. have been published on the subject since 1990 alone,
[118] The power stations are located -90 km up the describing research using surface observations, balloon valley. The easterly SBG reaches this point by late af- soundings of various kinds, aircraft, radar, lidar, and ternoon, and the polluted valley air is lifted aloft and computer models. Modern researchers have been moti-into the westerly return flow above the SBG. The marine vated by its influence on local wind velocity and air air behind the SBF is cleaner than the valley air because quality, convective activity, and other aspects of its dy-(1) there are no additional pollution sources closer to namics and structure. Others have studied the sea the coast, (2) the top of the CIBL within the marine air breeze's influence on the coastal ocean and beaches.
is well below the top of the easterly SBG, preventing the These authors have found that the SBS is a 'highly downward convective mixing of pollutants, and (3) the complex collection of nested phenomena, ranging in size combination of wind velocities aloft and offshore dis- from near synoptic to microscale. The SBS consists not tances traveled makes recycling of the pollutants from only of the mesoscale sea breeze circulation (SBC),
the subsiding end of the SBC impossible [Physick and crudely described by the Bjerknes circulation theorem, Abbs, 1992]. but sound waves and other transitory wave-like phenom-
[119] The SBS can also become a factor in more ena that either initiate or are initiated by nonlinear complex air pollution scenarios. One of these occurs responses of the atmosphere to diabatic heating. The regularly on the east coast of the United States. Pagnotti SBC itself, sometimes called the sea breeze proper, is
41, 3 / REVIEWS OF GEOPHYSICS Miller et al.: SEA BREEZE
- 1-25 42 41.5 Latitude
['N] 41 40.5 40 a
Longitude
['E]
140
.*-Mt. Ninham 120 k White Plains 100 I 80 I-0.~
60 I 40 k 20 I b
0 0700 0900 1100 1300 1500 1700 1900 2100 2300 Local Hour Figure 13. Ground level 03 concentration levels at three stations north of New York City. (a) Locations of 03 monitors in southern New York State. (b) Time series of 03 concentrations. Note "spikes" as the SBF passes successively farther inland. Plots are based on hourly 03 observations from July 1995. Figure 13 was redrawn after Gaza [1998].
distorted by nonzero ambient flow, forming helical circu- where the former is the leading edge of the marine air's lations with both cross- and along-shore components. The influence and the latter is the point of maximum low-level SBC may not be a closed (mass conservative) circulation. convergence between the marine and continental air
[123] The sea breeze gravity current (SBG) is the masses. Strong vertical currents are associated with the low-level landward flow of cool, moist marine air. The front, creating a raised sea breeze head (SBH) above the leading edge of the SBG is usually marked by a sea SBF that may reach twice the height (up to 2200 m) of the breeze frontal zone (SBF) that resembles a synoptic- feeder flow behind the SBF. The SBF and a vertically scale cold front. The sea breeze front (SBF) may be bifur- rotating postfrontal vortex within the marine air mass can cated into a thermodynamic front and a kinematic front, sometimes separate from the feeder flow and move inland
1-26 o Miller et al.: SEA BREEZE 41, 3 / REVIEWS OF GEOPHYSICS independently as an undular bore. The most dramatic CD drag coefficient.
example of this is the Australian Morning Glory. CT heat flux coefficient.
[124] Along the upper boundary of the SBG, Kelvin- d vertical depth of sea breeze.
Helmholtz billows (KHBs) develop when the Richard- d mean vertical depth of sea breeze.
son number is smaller than -0.25. Sea breeze-related )/Dt material derivative.
KHBs have wavelengths between 500 and 1000 m and f Coriolis parameter.
are responsible for mixing the two air masses in the F entrainment coefficient.
turbulent wake zone behind the SBH. KHBs also gen- g gravitational acceleration.
6rate a friction-like force along the top of the SBG that h height of CIBL.
slows its inland progression during midday. A number of H horizontal scale.
other instability phenomena can interact with the SBS, k constant, von Karman constant.
including vertical convection cells and horizontal con- *L vertical scale.
vective rolls within the continental air mass. P, Po,, P, atmospheric pressure.
[125] Forecasting the sea breeze is a three-part prob- Q heat.
lem, consisting of determining (1) a simple yes/no for the R specific gas constant for air.
shoreline, (2) the wind speed and direction behind the t time.
SBF, and (3) how far inland the sea breeze will pene- T, T 1 T 2 temperature.
trate. A simple index comparing the offshore wind com- T1 72 mean temperature.
ponent within the continental air mass to the magnitude u eastward component of net wind vector.
of the cross-shore temperature gradient is very effective U mean wind speed.
at answering the first question. The wind speed associ- IUI scalar wind speed.
ated with the sea breeze can be calculated by a number fron, rate of advance of sea breeze front.
of different methods, all of which are also dependent on ug eastward component of geostrophic wind, the cross-shore temperature gradient, and the wind di- cross-shore wind speed in continental air rection can be determined by an equation that accounts mass.
for rotation of the SBC under the Coriolis force, net U1, cross-shore wind component at upper part along-shore wind component, and synoptic-scale PGF. of CIBL.
In the absence of the latter two the entire SBC, and USB wind velocity behind sea breeze front.
therefore the wind direction, will rotate anticyclonically v northward component of net wind vector.
over time. Determining the extent of the SBC's inland vg northward component of geostrophic penetration is a complex problem, dependent on land- wind.
sea temperature contrast, the Coriolis force, surface and w vertical component of net wind vector.
top friction, stability, and other factors. While theoreti- x cross-shore dimension.
cal relationships exist to determine the maximum possi- y along-shore dimension.
ble inland penetration, accurate forecasts are still depen- z vertical dimension.
dent on a field study for the site in question. a angle.
[126] The sea breeze exerts both direct and indirect P3 expansion coefficient of air.
controls on the fate of pollutants released into the e lake breeze index.
coastal atmosphere. Examples of direct controls are the (p latitude.
landward transport of pollutants by elements of the SBS, -y vertical temperature lapse rate.
trapping and concentration of pollutants in shallow con- K, KH, vective internal boundary layers, and replacement of Kv coefficient of heat diffusion.
polluted continental air with relatively clean marine air. 0 potential temperature.
The east coast of the United States is the site of a more , Pa air density.
complex air pollution problem, involving the sea breeze (1) diurnal cycle of heating and cooling and a larger mesoscale trough that forms east of the Appalachian Mountains. A significant portion of the world's population resides in coastal areas, often side by side with industrial sites, ensuring that the sea breeze GLOSSARY will be the subject of increasingly sophisticated research for some time to come. Adiabatic: A process in which a system (or parcel of air) does not exchange energy with its surroundings.
Anticyclonic: Horizontal circulation associated Notation with areas of high atmospheric pressure and clockwise (counterclockwise) rotation in the Northern (Southern)
C. circulation. Hemisphere.
cp specific heat coefficient of air (constant Appalachian lee trough (APLT): A trough of low pressure). pressure that forms east of the Appalachian mountains
41, 3 / REVIEWS OF GEOPHYSICS Miller et al.: SEA BREEZE 9 1-27 on days when the 700-500 hPa flow has a significant level pressure on Earth's surface is about 1013 hectopas-westerly component. cals.
Backing: Counterclockwise rotation of the wind di- Horizontal convective roll (HCR): Overturning cy-rection. lindrically shaped wind structures whose axis of rotation Buys-Ballot's law: A law describing the geostrophic is parallel to Earth's surface.
wind. If you stand with your back to the wind in the Hydrostatic: Theoretical condition wherein the Northern Hemisphere, lower pressure is on your left, vertical pressure gradient force is balanced by gravity.
and higher pressure is on your right. The relation is There is no net vertical acceleration.
reversed in the Southern Hemisphere. Inviscid: An inviscid (or ideal) fluid is one in which Circulation: A scalar quantity that represents a all surface forces exerted on the boundaries of each macroscopic measure of rotation over a finite area of small element of the fluid act normal (perpendicular) to fluid in two dimensions. these boundaries.
Coastal upwelling: See upwelling. Isobar: A line of equal atmospheric pressure.
Convection: Vertical heat transfer by fluid flow. Isotropic: Having the same properties in all direc-This occurs when the atmosphere is thermodynamically tions.
unstable, i.e., when warmer (lighter) air underlies colder Kelvin-Helmholtz billow (KHB): Wave structures (denser) air. seaward of the sea breeze head along the top of the sea Convective internal boundary layer (CIBL): Well- breeze gravity current.
mixed region within the marine air mass over land, Kinematic: Pertaining to pure motion without re-adjacent to Earth's surface. gard to force, momentum, or energy.
Coriolis force: An apparent force causing objects Kinematic (sea breeze) front: The location of max-moving over the surface of Earth to be deflected to the imum near-surface (100 m AGL) wind convergence.
right in the Northern Hemisphere and to the left in the L-B index: Lake breeze index.
Southern Hemisphere. Mesoscale: Size scale of meteorological phenom-Cu: Fair-weather cumulus. ena between 2 and 2000 km across. The meso-ct scale Cutoff vortex: See undular bore. consists of phenomena between 200 and 2000 km across, Cyclonic: Horizontal circulation associated with ar- the meso-P3 scale consists of phenomena between 20 and eas of low atmospheric pressure and counterclockwise 200 km across, and the meso-y scale consists of phenom-(clockwise) rotation in the Northern (Southern) Hemi- ena between 2 and 20 km across.
sphere. Mesoscale pressure gradient force (mesoscale PGF):
Diabatic: A process in which a system (or parcel of Pressure gradient force created by the local density air) exchanges energy with its surroundings. difference between the marine and continental air Diffusion: The transport of matter solely by ran- masses. It points from the sea to the land.
dom motions of individual molecules (also known as Microscale: Size scale of meteorological phenom-molecular diffusion) or due to the effects of turbulent ena less than 2 km across.
motion (also known as dispersion). Morning Glory: Pronounced undular bore that oc-Ekman pumping: Upward or downward vertical curs in northern Australia.
motion in the surface ocean resulting from the balance New England Air Quality Study (NEAQS): Joint of the Coriolis force and wind stress on the sea surface. project of the U.S. National Oceanic and Atmospheric Frontogenesis: The formation or amplification of a Administration, the University of New Hampshire, and front due to an increase in the magnitude of the associ- others to understand the interrelationships in New En-ated gradients. Results from the convergence of two gland's air quality, meteorology, and climate phenom-dissimilar air masses. ena.
Frontolysis: The dissipation or weakening of a front Nocturnal stable layer: Shallow, thermodynami-due to a decrease in the magnitude of the associated cally stable region of the atmosphere in contact with gradients. Earth's surface that is created by the emission of infra-Fumigation: Rapid downward mixing of an ele- red radiation during hours of darkness.
vated plume of air pollution as it crosses from a stable Planetary boundary layer (PBL): Region of the tro-layer into a turbulent mixed layer, such as a convective posphere directly influenced by Earth's surface.
internal boundary layer. Potential temperature (0): The temperature a par-Geostrophic wind: Theoretical wind resulting from cel of air would have if it were transported adiabatically balance between the horizontal pressure gradient and to 1000 hPa.
Coriolis forces. Pressure gradient force (PGF): Horizontal force Gravity current: A flow driven by horizontal den- pointing from higher pressure to lower pressure, perpen-sity contrasts. dicular to the isobars.
Hectopascal (hPa): Unit of pressure equivalent to Prevailing wind (PW): Wind flow resulting from 100 Pascals or one millibar. The mean atmospheric sea synoptic-scale systems.
1-28
- Miller et al.: SEA BREEZE 41, 3 / REVIEWS OF GEOPHYSICS Pyroclastic flow: Downslope flow of hot volcanic Urban heat island (UHI): Region of atmosphere ash and cinders. above and near urban areas that, because of the pres-Radiational (or radiative) cooling: The nocturnal ence of urban structures, is warmer than the air above emission of infrared radiation from Earth's surface. It the surrounding countryside.
typically occurs on calm, clear nights and results in a Veering: Clockwise rotation of the wind direction.
nocturnal stable layer.
Richardson number (Ri): The ratio of static (ther-
[127] ACKNOWLEDGMENTS. We would like to thank modynamic) stability to kinetic energy of shear.
Roll Vortex: See Undular Bore. Frank Colby and James Koermer for their editorial assistance with this work. Figure 10 was reprinted from www.dropbear.com, Sea breeze circulation (SBC): Vertically rotating, with permission from Russell White. This review was produced closed circuit of wind flow caused by a mesoscale cross- for the Atmospheric Investigation, Regional Modeling, Anal-shore temperature gradient. ysis, and Prediction (AIRMAP) Project and was supported by Sea breeze front (SBF): The landward edge of the NOAA grants NA07RP0475 and NA17RP1488.
sea breeze system. The SBF consists of two main com- [128] Kendal McGuffie was the Editor responsible for this ponents: The thermodynamic SBF (SBFth) and the ki- paper. He thanks three technical reviewers and one cross-nematic SBF (SBFkn). The former is characterized a disciplinary reviewer.
strong cross-shore temperature gradient and marks the landward edge of the sea breeze gravity current. The latter is characterized by a change in wind direction and REFERENCES speed and marks the landward edge of the sea breeze Abbs, D. J., and W. L. Physick, Sea-breeze observations and circulation. modelling: A review, Aust. Meteorol. Mag., 41, 7-19, 1992.
Sea breeze gravity current (SBG): Landward move- Adams, E., Four ways to win the sea breeze game, Sailing World, March, 44-49, 1997.
ment of cool marine air near Earth's surface. Anderson, D. A., J. C. Tannehill, and R. H. Pletcher, Compu-Sea breeze head (SBH): The raised wave-like struc- tationalFluid Mechanics and Heat Transfer, 599 pp., Taylor ture at the leading edge of the sea breeze gravity current. and Francis, Philadelphia, Pa., 1984.
Sea breeze system (SBS): Includes all aspects of the Angell, J. K., and D. H. Pack, A study of the sea breeze at group of phenomena collectively called the sea breeze. Atlantic City, New Jersey using tetroons as Lagrangian tracers, Mon. Weather Rev., 93, 475-493, 1965.
Synoptic-scale cold front (SSCF): The boundary Anthes, R. A., The height of the planetary boundary layer and between two synoptic-scale air masses, with the colder the production of circulation in a sea breeze model, J.
(denser) air mass advancing onto an area of Earth's Atmos. Sci., 35, 1231-1239, 1978.
surface previously dominated by the warmer (lighter) air Apel, J. R., Principlesof Ocean Physics, 634 pp., Academic, San Diego, Calif., 1987.
mass. Arritt, R. W., Effects of the large-scale flow on characteristic Synoptic scale: Size scale of meteorological phe- features of the sea breeze, J. Appl. Meteorol., 32, 116-125, nomena more than 2000 km across. 1992.
Thermal pressure gradient force (thermal PGF): Asai, T., and S. Mitsumoto, Effects of an inclined land surface on the land and sea breeze circulation: A numerical exper-See mesoscale pressure gradient force. iment, J. Meteorol. Soc. Jpn., 56(6), 559-570, 1978.
Thermodynamic (sea breeze) front: The location Asimakopoulos, D. N., C. G. Helmis, K. H. Papadopoulos, within the PBL where the mean thermodynamic prop- J. A. Kalogiros, P. Kassomenos, and M. Petrakis, Inland erties of the air begin to differ from those of the ambient propagation of sea breeze under opposing offshore wind, (continental) air. Meteorol. Atmos. Phys., 70,97-110, 1999.
Atkins, N. T., R. M. Wakimoto, and T. M. Weckwerth, Ob-Thermodynamic stability: The ability of a fluid to servations of the sea-breeze front during CaPE, part II:
become turbulent or laminar because of the effects of Dual-doppler and aircraft analysis, Mon. Weather Rev., 123, buoyancy, a measure of the potential for convective 944-969, 1995.
overturning in a region of the atmosphere. Banta, R. M., L. D. Olivier, and D. H. Levinson, Evolution of the Monterey Bay sea-breeze layer as observed by pulsed Turbidity current: Gravity current in which a lim- doppler radar, J. Atmos. Sci., 50, 3959-3982, 1993.
ited volume of turbid (muddy) water moves relative to Barbato, J. P., The sea breeze of the Boston area and its effect the surrounding water because of the current's greater on the urban atmosphere, Ph.D. dissertation, Boston Univ.,
density. Boston, Mass., 1975.
Turbulence: Irregular, unpredictable fluctuations Barkan, J., and Y. Feliks, Observations of the diurnal oscilla-tion of the inversion over the Israeli coast, Boundary Layer in fluid media, generally occurring in all three spatial Meteorol., 62, 393-409, 1993.
dimensions. Barnes, S. L., A technique for maximizing details in numerical Undular bore: A solitary wave that forms when the weather map analysis, J. Appl. Meteorol., 3, 396-409, 1964.
leading edge of the SBG (beneath the SBH) separates Biggs, W. G., and M. E. Graves, A lake breeze index, J. Appl.
from the remainder of the marine air mass and propa- Meteorol., 1, 474-480, 1962.
Briummer, B., B. Hennemuth, A. Rhodin, and S. Thiemann,
.gates inland independently. Interaction of a cold front with a sea-breeze front: Obser-Upwelling: The vertical transport of subsurface wa- vations, Tellus, Ser. A, 47, 383-402, 1995.
ter to points nearer the sea surface. Buckley, R. L., and R. J. Kurzeja, An observational and nu-
41, 3 / REVIEWS OF GEOPHYSICS Miller et a].: SEA BREEZE
- 1-29 merical study of the nocturnal sea breeze, part I: Structure a blocking anticyclone in west-central Europe, Atmos. En-and circulation, J. Appl. Meteorol., 36, 1577-1598, 1997. viron., 36, 1349-1361, 2002.
Camberlin, P., and 0. Planchon, Coastal precipitation regimes Garratt, J. R., and W. L. Physick, The inland boundary layer at in Kenya, Geogr. Ann., Ser. A, 79A(1-2), 109-119, 1997. low latitudes: II, Sea-breeze influences, Boundary Layer Carbone, R. E., J. W. Wilson, T. D. Keenan, and J. M. Hacker, Meteorol., 33, 209-231, 1985.
Tropical island convection is the absence of significant Gaza, R. S., Mesoscale meteorology and high ozone in the topography. part I: Life cycle of diurnally forced convec- northeast United States, J. Appl. Meteorol., 37, 961-977, tion, Mon. Weather Rev., 128, 3459-3480, 2000. 1998.
Cenedese, A., M. Miozzi, and P. Monti, A laboratory investi- Geisler, J. E., and F. P. Bretherton, The sea-breeze forerunner, gation of land and sea breeze regimes, Exp. Fluids, suppl., J. Atmos. Sci., 26, 82-95, 1969.
S291-S299, 2000. Gibbs, M. T., Detecting a response to weak sea breezes in the Chiba, 0., The turbulent characteristics in the lowest part of New South Wales coastal ocean, N. Z. J Mar. Freshwater the sea breeze front in the atmospheric surface layer, Res., 34, 669-680, 2000.
Boundary Layer Meteorol., 65, 181-195, 1993. Godske, C. L., T. Bergeron, J. Bjerknes, and R. C. Bundgaard, Chiba, 0., F. Kobayashi, G. Naito, and K. Sassa, Helicopter Dynamic Meteorology and Weather Forecasting,800 pp., Am.
observations of the sea breeze over a coastal area, J. Appl. Meteorol. Soc., Boston, Mass., 1957.
Meteorol., 38, 481-492, 1999. Gryning, S. E., and E. Batchvarova, Analytical model for the Clappier, A., A. Martilli, P. Grossi, P. Thunis, F. Pasi, B. C. growth of the coastal internal boundary layer during on-Krueger, B. Calpini, G. Graziani, and H. van den Bergh, shore flow, Q. J. R. Meteorol. Soc., 116, 187-203, 1990.
Effect of sea breeze on air pollution in greater Athens area, Hadi, T. W., T. Horinouchi, T. Tsuda, H. Hashiguchi, and S.
part I: Numerical simulations and field observations, Fuhao, Sea-breeze circulation over Jakarta, Indonesia: A J.App!. Meteorol., 39, 546-562, 2000. climatology based on boundary layer radar observations, Clarke, R. H., Some observations and comments on the sea Mon. Weather Rev., 130, 2153-2166, 2002.
breeze, Aust. Meteorol. Mag., 11, 47-68, 1955. Halliday, D., R. Resnick, and J. Walker, Fundamentals of Clarke, R. H., Colliding sea-breezes and the creation of inter- Physics, 1306 pp., John Wiley, Hoboken, N. J., 1993.
nal atmospheric bore waves: Two-dimensional numerical Helmis, C. G., D. N. Asimakopoulos, D. G. Deligiorgi, and studies, Aust. Meteorol. Mag., 32, 207-226, 1984. D. P. Lalas, Observations of sea-breeze fronts near the Coutant, V., and V. L. Eichenlaub, TheophrastusDe Ventis, 105 shoreline, Boundary Layer Meteorol., 38, 395-410, 1987.
pp., Univ. of Notre Dame Press, Notre Dame, Ind., 1975. Helmis, C. G., K. H. Papadopoulos, J. A. Kalogiros, A. T.
Craig, R. A., I. Katz, and P. J. Harney, Sea breeze cross Soilemes, and D. N. Asimakopoulos, Influence of back-sections from pyschrometric measurements, Bull. Am. Me- ground flow on evolution of Saronic Gulf sea breeze, At-teorol. Soc., 26(10), 405-410, 1945. mos. Environ., 29, 3689-3701, 1995.
Dalu, G. A., and R. A. Pielke, An analytical study of the sea Holmer, B., and M. Haeger-Eugensson, Winter land breeze in breeze, J. Atmos. Sci., 46, 1815-1825, 1989. a high latitude complex coastal area, Phys. Geogr., 20(2),
Druilhet, A., A. Herrada, J. P. Pages, J. Saissac, C. Allet, and 152-172, 1999.
M. Ravaut, Etude exp6rimentale de ]a couche limite interne Holton, J. R.,An Introduction to Dynamic Meteorology, 511 pp.,
associ6e A ]a brise de mer, Boundary Layer Meteorol., 22, Academic, San Diego, Calif., 1992.
511-524, 1982. Hsu, S. A., CoastalMeteorology, 260 pp., Academic, San Diego, Elliott, D. L., and J. J. O'Brien, Observational studies of the Calif., 1988.
marine boundary layer over an upwelling region, Mon. Intrieri, J. M., C. G. Little, W. J. Shaw, R. M. Banta, P. A.
Weather Rev., 105, 86-98, 1977. Durkee, and R. M. Hardesty, The Land/Sea Breeze Exper-Estoque, M. A., The sea breeze as a function of the prevailing iment (LASBEX), BulL Am. Meteorol. Soc., 71(5), 656-664, synoptic situation, J. Atmos. Sci., 19, 244-250, 1962. 1990.
Federovich, E., F. T. M. Nieuwstadt, and R. Kaiser, Numerical Jehn, K. H., A Sea Breeze Bibliography,1664-1972, Rep. 37, 51 and laboratory studies of horizontally evolving convective pp., Atmos. Sci. Group, Coll. of Eng., Univ. of Tex., Austin boundary layer, part II: Effects of elevated wind shear and 1973.
surface roughness, J. Atmos. Sci., 58, 546-560, 2001. Kalthoff, N., et al., Mesoscale wind regimes in Chile at 30 'S, Feliks, Y., A numerical model for estimation of the diurnal J. Appl. Meteorol., 41, 953-970, 2002.
fluctuation of the inversion height due to a sea breeze, Kanda, M., Y. Inoue, and I. Uno, Numerical study on cloud Boundary Layer Meteorol., 62, 151-161, 1993. lines over an urban street in Tokyo, Boundary Layer Meteo-Finkele, K., Inland and offshore propagation speeds of a sea rol., 98, 251-273, 2001.
breeze from simulations and measurements, Boundary Kitada, T., Turbulence structure of sea breeze front and its Layer Meteorol., 87, 307-329, 1998. implication in air pollution transport-Application of k-e Finkele, K., J. M. Hacker, H. Kraus, and R. A. D. Byron-Scott, turbulence model, Boundary Layer Meteorol., 41, 217-239, A complete sea-breeze circulation cell derived from aircraft 1987.
observations, Boundary Layer Meteorol., 73, 299-317, 1995. Kottmeier, C., P. Palacio-Sese, N. Kalthoff, U. Corsmeier, and Fisher, E. L., An observational study of the sea breeze, J. F. Fiedler, Sea breezes and coastal jets in southeastern Meteorol., 17, 645-660, 1960. Spain, Int. J. Climatol., 20, 1791-1808, 2000.
Franchito, S. H., V. B. Rao, J. L. Stech, and J. A. Lorenzzetti, Kozo, T. L., An observational study of sea breezes along the The effect of coastal upwelling on the sea-breeze circulation Alaskan Beaufort sea coast: Part I, J. Appl. Meteorol., 21, of Cabo Frio, Brazil: A numerical experiment, Ann. Geo- 891-905, 1982.
phys., 16, 866-881, 1998. Kraus, H., Turbulence frontogenesis, Meteorol. Atmos. Phys.,
Frizzola, J. A., and E. L. Fisher, A series of sea breeze 48, 309-315, 1992.
observations in the New York City area, J. Appl. Meteorol., Kraus, H., J. M. Hacker, and J. Hartmann, An observational 2, 722-739, 1963. aircraft-based study of sea-breeze frontogenesis, Boundary Gangoiti, G., L. Alonso, M. Navazo, A. Albizuri, G. Perez- Layer Meteorol., 53, 223-265, 1990.
Landa, M. Matabuena, V. Valdenebro, M. Maruri, J. A. Kuelegan, G. H., An experimental study of the motion of Garcia, and M. A. Millfin, Regional transport of pollutants saline water from locks into fresh water channels, in 13th over the Bay of Biscay: Analysis of an ozone episode under Progress Report on Model Laws for Density Currents, Rep.
1-30 9 Miller et al.: SEA BREEZE 41, 3 / REVIEWS OF GEOPHYSICS 5168, Natl. Inst. of Stand. and Technol., Gaithersburg, Md., Neumann, J., and Y. Mahrer, A theoretical study of the land 1957. and sea breeze circulation, J. Atmos. Sci., 28, 532-542, 1971.
Laird, N. F., and D. A. R. Kristovich, Lake Michigan lake Nielsen., J. W., In situ observations of Kelvin-Helmholtz waves breezes: Climatology, local forcing, and synoptic environ- along a frontal inversion, J. Atmos. Sci., 49, 369-386, 1992.
ment, J. Appl. Meteorol., 40, 409-424, 2001. Ohashi, Y., and H. Kida, Observational results of the sea Lapworth, A., Observations of atmospheric density currents breeze with a weak wind region over the northern Osaka using a tethered balloon-borne turbulence probe system, urban area, I. Meteorol. Soc. Jpn., 79(4), 949-955, 2002.
Q. J. R. Meteorol. Soc., 126, 2811-2850, 2000. Oliphant, A. J., A. P. Sturman, and N. J. Tapper, The evolution Lee, H. D. P., Aristotle VII Meteorologica, 432 pp., Harvard and structure of a tropical island sea/land-breeze system, Univ. Press, Cambridge, Mass., 1952. northern Australia, Meteorol.Atmos. Phys., 78, 45-59, 2001.
Lericos, T. P., H. E. Fuelberg, A. I. Watson, and R. L. Holle, Pagnotti, V., A meso-meteorological feature associated with Warm season lightning distributions over the Florida pen- high ozone concentrations in the northeastern United insula as related to synoptic patterns, Weather Forecast., 17, States, JAPCA, 37(6), 720-722, 1987.
83-98, 2002. Panel on Coastal Meteorology, CoastalMeteorology-A Review List, R. J., Smithsonian Meteorological Tables, 527 pp., Smith- of the State of the Science, 99 pp., Natl. Acad. Press, Wash-sonian Inst. Press, Washington, D. C., 1968. ington, D. C., 1992.
Liu, H. P., and J. C. L. Chan, Boundary layer dynamics asso- Pearce, R. P., The calculation of a sea-breeze circulation in ciated with a severe air-pollution episode in Hong Kong, terms of differential heating across the coastline, Q. J. R.
Atmos. Environ., 36, 2013-2025, 2002. Meteorol. Soc., 81, 351-381, 1955.
Luhar, A. K., B. L. Sawford, J. M. Hacker, and K. N. Rayner, Pearson, R. A., G. Carboni, and G. Brusasca, The sea breeze The Kwinana coastal fumigation study: II-Growth of the with mean flow, Q. J. R. Meteorol. Soc., 109, 809-830, 1983.
thermal internal boundary layer, Boundary Layer Meteorol., Physick, W. L., Numerical experiments on the inland penetra-89, 385-405, 1998. tion of the sea breeze, Q. J. R. Meteorol. Soc., 106, 735-746, Lutgens, F. K., and E. J. Tarbuck, The Atmosphere an Intro- 1980.
duction to Meteorology, 484 pp., Prentice-Hall, Old Tappan, Physick, W. L., and D. J. Abbs, Flow and plume dispersion in N. J., 2001. a coastal valley, J. Appl. Meteorol., 31, 64-73, 1992.
Lyons, W. A., The climatology and prediction of the Chicago Pielke, R. A., Sea breeze-induced mesoscale systems and se-lake breeze, J. Appl. Meteorol., 11, 1259-1270, 1972. vere weather, NASA Proj. Prog. Rep. NASA-CR-174317, Lyons, W. A., E. R. Sawdey, J. A. Schuh, R. H. Calby, and C. S. NASA, Greenbelt, Md., 1985.
Keen, An updated and expanded coastal fumigation model, Pielke, R. A., A. Song, P. J. Michaels, W. A. Lyons, and R. W.
paper presented at 74th Annual Meeting, Air Pollut. Con- Arritt, The predictability of sea-breeze generated thunder-trol Assoc., Philadelphia, Pa., 21-26 June 1981. storms, Atmosfera, 4, 65-78, 1991.
Masselink, G., and C. B. Pattiaratchi, The effect of sea breeze Planchon, 0., and S. Cautenet, Rainfall and sea-breeze circu-on beach morphology, surf zone hydrodynamics and sedi- lation over south-western France, Int. J. Climatol., 17, 535-ment resuspension, Mar. Geol., 146, 115-135, 1998. 549, 1997.
Mathews, J. H., The sea-breeze-Forecasting aspects, Aust. Prahba, T. V., R. Venkatesan, E. Mursch-Radlgruber, G. Ren-Meteorol. Mag., 30, 205-209, 1982. garajan, and N. Jayanthi, Thermal internal boundary layer McKendry, I., and N. Roulet, Sea breezes and advective effects characteristics at a tropical coastal site as observed by a in southwest James Bay, J. Geophys. Res., 99(D1), 1623- mini-SODAR under varying synoptic conditions, Proc. In-1634, 1994. dian Acad. Sci. Earth Planet Sci., 111(1), 63-77, 2002.
McPherson, R. D., A numerical study of the effect of a coastal Rao, P., and H. E. Fuelberg, An investigation of convection irregularity on the sea breeze, J. Appl. Meteorol., 9, 767-777, behind the Cape Canaveral sea-breeze front, Mon. Weather 1970. Rev., 128, 3437-3458, 2000.
Melas, D., and H. D. Kambezidis, The depth of the internal Raynor, G. S., S. Sethuraman, and R. M. Brown, Formation boundary layer over an urban area under sea-breeze con- and characteristics of coastal internal boundary layers dur-ditions, Boundary Layer Meteorol., 61, 247-264, 1992. ing onshore flows, Boundary Layer Meteorol., 16, 487-514, Melas, D., I. Ziomas, 0. Klemm, and C. S. Zerefos, Anatomy 1979.
of the sea-breeze circulation in Athens area under weak Reible, D. D., J. E. Simpson, and P. F. Linden, The sea breeze large-scale ambient winds, Atmos. Environ., 32, 2223-2237, and gravity-current frontogenesis, Q. J. R. Meteorol. Soc.,
1998. 119, 1-16, 1993.
Melas, D., A. Lavagnini, and A. Sempreviva, An investigation Rogers, R. R., and M. K. Yau, A Short Course in Cloud Physics, of the boundary layer dynamics of Sardinia Island under 290 pp., Butterworth-Heinemann, Woburn, Mass., 1989.
sea-breeze conditions, J. Appl. Meteorol., 39, 516-524, 2000. Rotunno, R., On the linear theory of the land and sea breeze, Miller, J. E., On the concept of frontogenesis, J. Meteorol., 5, J. Atmos. Sci., 40, 1999-2009, 1983.
169-171, 1948. Rubes, M. T., H. J. Cooper, and E. A. Smith, A study of the Miller, S. T. K., and B. D. Keim, Synoptic-scale controls on the Merritt Island, Florida sea breeze flow regimes and their sea breeze of the central New England coast, Weather effect on surface heat and moisture fluxes, NASA Contract.
Forecast., 18, 236-248, 2003. Rep. 4537, 141 pp., NASA, Greenbelt, Md., 1993.
Mitsumoto, S., H. Ueda, and H. Ozoe, A laboratory experi- Sawford, B. L., A. K. Luhar, J. H. Hacker, S. A. Young, I.-H.
ment on the dynamics of the land and sea breezes, J. Atmos. Yoon, J. A. Noonan, J. N. Carras, D. J. Williams, and K. N.
Sci., 40, 1228-1240, 1983. Rayner, The Kwinana coastal fumigation study: I-Pro-Mizuma, M., General aspects of land and sea breezes in Osaka gram overview, experimental design, and selected results, Bay and surrounding area, J. Meteorol. Soc. Jpn., 73(6), Boundary Layer Meteorol., 89, 359-384, 1998.
1029-1040, 1995. Schoellhamer, D. H., Factors affecting suspended-solids con-Mizuma, M., General aspects of land and sea breezes in centrations in south San Francisco Bay, California, J. Geo-western Seto Inland Sea and surrounding areas, J Meteorol. phy. Res., 101(C5), 12,087-12,095, 1996.
Soc. Jpn, 76(3), 403-418, 1998. Schoenberger, L. M., Doppler radar observation of a land-Neumann, J., On the rotation rate of the direction of sea and breeze cold front, Mon. Weather Rev., 112, 2455-2464, 1984.
land breezes, J. Atmos. Sci., 34, 1913-1917, 1977. Schumann, E. H., W. K. Illenberger, and W. S. Goschen,
41, 3 / REVIEWS OF GEOPHYSICS Miller et aL: SEA BREEZE
- 1-31 Surface winds over Algoa Bay, South Africa, S. Afr. J. Sci., Stull, R. B., An Introduction to Boundary Layer Meteorology, 87, 202-207, 1991. 670 pp., Kluwer Acad., Norwell, Mass., 1988.
Seaman, N. L., and S. A. Michelson, Mesoscale meteorological Tijm, A. B. C., and A. J. van Delden, The role of sound waves structure of a high-ozone episode during the 1995 in sea-breeze circulation, Q. J. R. Meteorol. Soc., 125, 1997-NARSTO-Northeast Study, J.AppL. Meteorol., 39, 384-398, 2018, 1999.
2000. Tijm, A. B. C., A. A. M. Holtstag, and A. J. van Delden, Sha, W., T. Kawamura, and H. Ueda, A numerical study on Observations and modeling of the sea breeze with the sea-land breezes as a gravity current: Kelvin-Helmholtz return current, Mon. Weather Rev., 127, 625-640, 1999.
billows and inland penetration of the sea-breeze front, J. Venkatram, A., A model of internal boundary-layer develop-Atmos. Sci., 48, 1649-1665, 1991. ment, Boundary Layer Meteorol., 11, 419-437, 1977.
Sha, W., T. Kawamura, and H. Ueda, A numerical study of Walsh, J. E., Sea breeze theory and applications, J. Atmos. Sci.,
nocturnal sea breezes: Prefrontal gravity waves in the com- 31, 2012-2026, 1974.
pensating flow and inland penetration of the sea-breeze Weckwerth, T. M., J. W. Wilson, and R. M. Wakimoto, Ther-cutoff vortex, J. Atmos. Sci., 50, 1076-1087, 1993. modynamic variability within the convective boundary layer Shair, F. H., E. J. Sasaki, D. E. Carlan, G. R. Cass, W. R. due to horizontal convective rolls, Mon. Weather Rev., 124, Goodin, J. G. Edinger, and G. E. Schacher, Transport and 769-784, 1996.
dispersion of airborne pollutants associated with the land Weckwerth, T. M., J. W. Wilson, R. M. Wakimoto, and N. A.
breeze-sea breeze system, Atmos. Environ., 16, 2043-2053, Crook, Horizontal convective rolls: Determining the envi-1982. ronmental conditions supporting their existence and char-Shearer, D. L., and R. J. Kaleel, Critical review of studies on acteristics, Mon. Weather Rev., 125, 505-526, 1997.
atmospheric dispersion in coastal regions, Rep. PNL-4292, Wood, R., I. M. Stromberg, and P. R. Jonas, Aircraft obser-41 pp., Pac. Natl. Lab., Richland, Wash., 1982. vations of sea-breeze frontal structure, Q. J. R. Meteorol.
Silva Dias, M. A. F., and A. J. Machado, The role of local Soc., 125, 1959 -1995, 1999.
circulations in summertime convective development and Xian, Z., and R. A. Pielke, The effects of width of landmasses nocturnal fog in Sdo Paulo, Brazil, Boundary Layer Meteo- on the development of sea breezes, J. Appl. Meteorol., 30, rol., 82, 135-157, 1997. 1280-1304, 1991.
Simpson, J. E., A comparison between laboratory currents and Xu, Q., Density currents in shear flows-A two-fluid model, J atmospheric density currents, Q. J. R. Meteorol. Soc., 95, Atmos. Sci., 49, 511-524, 1992.
758-765, 1969. Yoshikado, H., Numerical study of the daytime urban effect Simpson, J. E., Sea Breeze and Local Wind, 234 pp., Cambridge and its interaction with the sea breeze, J.AppL. Meteorol., 31, Univ. Press, New York, 1994. 1146-1164, 1992.
Simpson, J. E., Diurnal changes in sea-breeze direction, Yu,-T. W., and N. K. Wagner, Diurnal variations of onshore J. Appl. Meteorol., 35, 1166-1169, 1996. wind speed near a coastline, J. Appl. Meteorol., 9, 760-766, Simpson, J. E., Gravity Currents in the Environment and the 1970.
Laboratory, 244 pp., Cambridge Univ. Press, New York, Zhong, S., and E. S. Takle, An observational study of sea- and 1997. land-breeze circulation in an area of complex coastal heat-Simpson, J. E., and R. E. Britter, A laboratory model of an ing, J. Appl. Meteorol., 31, 1426-1438, 1992.
atmospheric mesofront, Q. J. R. Meteorol. Soc., 106, 485-500, 1980.
Simpson, J. E., D. A. Mansfield, and J. R. Milford, Inland B. D. Keim, Southern Regional Climate Center, Louisiana penetration of sea-breeze fronts, Q. J. R. Meteorol. Soc.,
State University, Baton Rouge, LA 70803, USA. (bkeim@srcc.
103, 47-76, 1977.
Smith, R. K., N. Crook, and G. Roff, The Morning Glory: An Isu.edu) extraordinary atmospheric undular bore, Q. J. R. Meteorol. H. Mao, S. T. K.Miller, and R. W. Talbot, Climate Change Soc., 108, 937-956, 1982. Research Center, Institute for the Study of Earth, Oceans and Stephan, K., H. Kraus, C. M. Ewenz, and J. M. Hacker, Space, University of New Hampshire, Durham, NH 03824, Sea-breeze front variations in space and time, Meteorol. USA. (hmao@typhoon.sr.unh.edu; stm@cisunix.unh.edu; Atmos. Phys., 70, 81-95, 1999. bob@resolution.sr.unh.edu)