experimental study of the vertical structure of the lower troposphere sea

8/22/2019 Experimental Study of the Vertical Structure of the Lower Troposphere Sea

1/12

AUGUST 2002 1181H E L M I S E T A L .

2002 American Meteorological Society

Experimental Study of the Vertical Structure of the Lower Troposphere over a Small

Greek Island in the Aegean Sea

C. G. HELMIS, C. JACOVIDES, D. N. ASIMAKOPOULOS, AND H. A. FLOCAS

Division of Applie d Physic s, Department of Physic s, University of Athens, Greece

(Manuscript received 20 June 2001, in final form 27 November 2001)

ABSTRACT

An experimental campaign was carried out on a small Greek island that is characterized by complex terrain;its aim was to study the local characteristics of the vertical structure of the atmospheric boundary layer (ABL).

The instrumentation was installed close to the shoreline and consisted of a 13-m-high meteorological mastinstrumented at three levels, and a high-range vertical monostatic sodar. Tethered balloon flights were carriedout for 3 days under different atmospheric conditions. The analysis of the available data revealed interestingfeatures of the vertical structure of the atmosphere over the island, with the development of a convective internalboundary layer (IBL) within the first 150 m above the ground, while the marine boundary layer (MBL) formedat higher altitudes, up to 450 m. Buoyant oscillations appear within the MBL in the form of gravity waves withfrequencies of 7 min. Theoretical calculations of the IBL height verified the experimental results. During thenight, a complex wind flow forms in the lower 250300 m, resulting from the development of katabatic flowsand topographic channeling.

1. Introduction

Scopelos is a small island in the northwestern Aegean

Sea, which is situated in the eastern Mediterranean Ba-sin between Greece and Turkey, extending a few hun-dred kilometers. An experimental campaign was carriedout in summer 1996 near the shoreline at the north-eastern part of Scopelos; the goal was the study of theairflow regime at a location where the new local airportwas proposed to be constructed. During summer, a weaksynoptic wind prevails over the Aegean Sea under thefrequent presence of anticyclonic circulation, allowingthe generation of local flows, such as sea and land breez-es and katabatic winds. Furthermore, the developmentof an internal boundary layer (IBL) is evident over theexamined area. Close to the shoreline an IBL is me-chanically induced, which is in response to an abruptchange in roughness between the sea (upwind) and theland (downwind). In addition, a thermal (convective)IBL develops during the daytime as the air moves fromthe cool sea toward the much warmer land, producingchanges in the surface heat fluxes. Thus, the developedflow over the coastal region combines features of the

Corresponding author address: Prof. C.G. Helmis, Division ofApplied Physics, Department of Physics, Building Phys-5, Universityof Athens, 157 84 Athens, Greece.E-mail: [email protected]

local marine boundary layer (MBL) and the thermallyinduced local flows.

Observations of the marine atmosphere structure, per-formed by Mandics and Owens (1975) and Gaynor andMandics (1978), revealed the existence of a rich varietyin the vertical structure associated with temperature in-versions that were often perturbed by gravity waves orwind shear. The characteristics of the IBL were exten-sively studied on a theoretical and experimental basisby Stunder and Sethuraman (1985), Venkatram (1986),Hsu (1986), Bergstrom et al. (1988), and Garratt (1990).According to these studies, the characteristics of the IBLare greatly determined by the roughness of the shoreline,the maritime air stability, the temperature difference be-tween the sea and the land, and the prevailing back-

ground flow. The depth of the convective IBL increaseswith distance downwind of the shoreline, while the in-fluence of the sea is expected to minimize at a suffi-ciently large distance from the shoreline, where the IBLhas achieved an equilibrium with the ABL.

Based on the above, the atmospheric environment inMediterranean coastal regions is influenced by local cir-culation flows due to sealand contrast, which is morepronounced in the islands. Although extensive researchconcerning the IBL is published, only a few studiesexamine the IBL over islands. During nighttime thewinds over the island are low, and the development ofstrong katabatic flows combined with land-breeze flow


2/12

1182 VOLUME 19J O U R N A L O F A T M O S P H E R I C A N D O C E A N I C T E C H N O L O G Y

FIG. 1. Topography of Scopelos Island. The numbers represent the maximum height ofthe mountain ranges. The bullet indicates the experimental site.

are substantially evident. During the daytime, differentsea-breeze systems, which develop over the island, in-teract and produce convergence zones accompanied byupward motion (Melas et al. 2000). Also, the strong

influence on the marine atmospheric boundary layer(ABL) structure of isolated islands was demonstratedby Petenko et al. (1996). According to their study, thisinfluence reveals the occurrence of inversion layers ata distance of several tens of kilometers from the shore,which modifies the stratification prevailing in the MBL.Thus, experimental or theoretical studies focusing onthe modification of flow in coastal zones as a result ofstability and roughness changes are still of major im-portance. This subject becomes more interesting in asmall-scale sea, such as the Aegean, which is charac-terized by numerous scattered small islands and is sur-rounded by the Greek (in the north and west) and Turk-

ish mainland (in the east).In this respect, the objective of the present paperis to study the development of the IBL forming overthe coast of a small Aegean island and the marineinfluence. Furthermore, the vertical structure of thelower troposphere is investigated during the daytimeand nighttime, under weak synoptic conditions, al-lowing the development of local flows, or when on-shore moderate synoptic winds blow from the north-ern sector of the open sea. Finally, an effort was madeto test basic empirical relationships for the estimationof IBL height, taking into account the roughness andscalar flux changes.

2. Experimental site and instrumentation

Scopelos is positioned at latitude 39N and longi-tude 24E, with a total area of 96 km 2 , and is char-acterized by complex topography (Fig. 1). The in-strumentation was installed at a location in the north-eastern part of the island, at 40 m ASL, close to theshoreline (150 m downwind). This site is also sur-rounded by a mountainous complex with maximumheight 500600 m at the western and eastern sides,forming a small valley (canal) oriented from north tosouth with an elevation of 200 m at its southern region(see Fig. 1).

The experiment covered a 10-day period (2029 July1996) where low or moderate background wind prevails.The instrumentation consisted of 1) a meteorologicalmast equipped with thermometers at three levels (2, 7,

and 13 m), cup anemometers at two levels (7 and 13m), and a wind vane at 7 m, allowing measurements ofwind and air temperature; and 2) a 1600-Hz high-rangevertical monostatic sodar, capable of measuring in realtime the vertical component of the wind and the thermalturbulent structure of the atmosphere up to a height of800 m, with a minimum discernible height of 50 m. Itshould be mentioned that the sodar was developed atthe Department of Applied Physics of the University ofAthens (DOAP-UOA) and provided variable operatingparameters to best suit the experimental conditions (As-imakopoulos and Helmis 1994).

Also, tethered balloon flights were carried out at the


3/12


FIG. 2. Mean sea level pressure distribution (every 5 hPa) and frontal analysis at 1400LST 27 Jul 1996.

same location for three days (22, 24, and 27 July) underdifferent meteorological conditions, using a speciallydesigned meteorological package by DOAP-UOA, pro-

viding measurements of wind speed and direction, tem-perature, and humidity up to the height of 900 m witha sampling rate of 1 s (Soilemes et al. 1993). The clas-sical psychrometric method is used to measure dry- andwet-bulb temperatures using two identical negative tem-perature coefficient (NTC) thermistors, while a cup an-emometer and an electrically controlled magnetic com-pass were used to measure wind speed and direction,respectively. Data acquisition was obtained by a micro-computer placed in the airborne package.

Information on the synoptic-scale atmospheric cir-culation during the experimental period was derivedfrom the European Meteorological Bulletin archive.

The surface synoptic circulation during the first 2 daysof the examined period results in moderate northerlywinds, while on 22 and 23 July strong windsete-siansprevail, blowing from the northeasterly sector.In the following days, from 24 to 28 July, anticycloniccirculation predominates over the whole Greek area (seeFig. 2), producing weak synoptic flow.

3. Results and discussion

a. The vertical structure of the atmosphere

Figure 3 shows the vertical profiles of wind directionand speed, temperature, and mixing ratio during three

flights of the tethered balloon on 27 July 1996 at 08170910, 10271119, and 12211319 LST. The prevailingwind blows from the northerly/northeasterly sector up

to a height of 900 m during the whole morning (Fig.3a), indicating the maritime air character. The potentialtemperature profiles show (Fig. 3b) that the air is rel-atively stable, maintaining the same characteristicsabove the height of 450 m during all flights. It shouldbe noted that similar characteristics were observed inother studies over coastal areas (Luhar et al. 1998),where the combined superadiabatic and neutral layer,typically 250 m deep, was capped by a fully stable strat-ification at higher levels. However, different character-istics are evident at the lower levels. While a ground-based inversion is initially present (not shown), after1030 LST an unstable surface layer develops that in-

tensifies in depth during the next 2 h, associated withthe developed IBL. The formation of the IBL is evidentfrom the potential temperature profiles, where the lowerpart (below 100 and 200 m, respectively, of the two lastflights) corresponds to the IBL and the upper part (upto 450 m with intense stability factor) of the MBL flow(Fig. 3b). The depth of the MBL is more evident in themixing ratio profiles (Fig. 3d), where there is an abruptchange (from 14 to 10 g kg1 ) of the mixing ratio valuesfor heights below and above 450 m.

According to the wind speed profiles (Fig. 3c), a speed-up effect is evident in the lower levels (less than 200 m)that is probably attributable to the sloping terrain (about


4/12


FIG. 3. Vertical profiles of (a) wind direction, (b) potential temperature, (c) wind speed, and (d) mixingratio during the flights of the tethered balloon on 27 Jul 1996 at 08170910, 10271119, and 12211319LST.

15) close to the shoreline. At higher levels (more than500 m) the wind speed is reduced to the corresponding

background flow. It is noteworthy that during the fourthflight, after 1221 LST, the air is significantly cooler, ascompared with the previous flight (see Fig. 3b) in thelowest 400 m. This, in relation to the substantial increaseof the wind speed in this layer (Fig. 3c)and the mixingratio (Fig. 3d)implies the strong advection of cold airmasses from the sea over the inland.

Figure 4 shows the acoustic sounder fascimile record,which displays the backscattered signal echo intensity(convective turbulence) as a function of height and time,for the period 06001424 LST 27 July. It should benoted that strong echo (black) regions correspond tointense thermal turbulence in contrast to weak echo

(white) regions. During the morning (Fig. 4a) a stablelayer exists with an inversion layer above at a height

that varies between 200 and 400 m. This is due to theinfluence of the island on the marine ABL, where thedevelopment of elevated inversion layers modified theusually unstable or neutral stratification that prevailsover the ocean (Petenko et al. 1996). Later (after 0830LST) an organized thermal plume activity is present atlower levels, from the ground up to 200 m, due to thedevelopment of the IBL. This is associated with the localthermal heating of the land. The height of the inversionlayering rises continuously up to a height of 450 m.After 1030 LST (see Fig. 4b) the developed sea-breezecell intensifies the flow from the sea, giving an intenseinversion with strong echo returns at heights up to 400


5/12


FIG. 3. (Continued)

m. The lower part of the layer (the IBL) that extendsup to 150 m and the stable layer above (higher than 400

m) retain their characteristics.Wavelike activity combined with double layering be-

comes evident on the fascimile record in the 200400-m layer between 1000 and 1400 LST (see Fig. 4b). Itshould be noted that the flights of the tethered balloonbetween 10271119 and 12211317 LST are evident onthe fascimile record as straight lines up to the height of800 m. The characteristics of these disturbances are in-vestigated by performing a Fourier analysis of the ver-tical velocity field at different levels within the heightrange of the sodar. More specifically, fast Fourier trans-form spectral analysis was employed using 512 pointswith segment and frequency averaging. Figure 5 shows

the vertical velocity power spectrum at the height of270 m, as it was estimated during the period from 1000

to 1400 LST. It can be seen that there is a spectral densitymaximum at the frequency of 2.5 103 Hz that cor-responds to time periods of about 7 min. It should benoted that the second spectral density maximum cor-responds to time periods of 45 min and is attributed tolarge-scale variations.

To further investigate the nature of these distur-bances, the Richard son nu mber (Ri) and BruntVais-ala frequenc y (N) are estimated for the last two flightsof the tethered balloon and for the periods 10271119and 12211319 LST on the basis of tethered balloondata for different layers (200400 m), using the re-lations


6/12


FIG. 4. The sodar fascimile record on 27 Jul 1996 from (a) 06001000 and (b) 10001424 LST.


7/12


FIG. 5. Vertical velocity power spectrum at 270-m height for the time period 10001400LST 27 Jul 1996 as derived from fast Fourier transform spectral analysis.

TABLE 1. Estimated values of the Richardson number (Ri), BruntVaisala fre quency (N), and period (T) of gravity waves, as derivedfrom tethered balloon data.

DateTime

(LST)Layer

(m) RiN

(rad s1)T

(min)

24 Jul 1996 12201247 200400300400

1213.2

8.64 103

9.11 10312.111.5

27 Jul 1996 1027111912211319

200400200400300400

13.43.83.2

1.46 102

1.57 102

1.81 102

7.26.65.7

g zRi , (1)

2 u z1/2

g N , (2) [ ] z

where is the mean potential temperature, g is the grav-itational acceleration, and / z and u/ z are the po-tential temperature gradient and wind shear, respective-ly. Table 1 presents the estimated values of the Rich-ardson number (Ri), the corresponding BruntVaisal afrequency (N), and the period (T) of the gravity waves,as derived from the tethered balloon data. From thistable, it becomes evident that Ri is characterized bypositive values, much larger than unity, suggesting thatbuoyant oscillations could appear in the form of gravitywaves with frequencies equal to or less than the BruntVaisala frequency N and periods 2/N (Arya 1988).This is in accordance with the wave activity of the ther-mal structure of the ABL observed on the fascimilerecord, and the estimated periods are consistent with theperiod derived from the Fourier analysis of the verticalvelocity field (see Fig. 5). It is worth mentioning that

similar disturbances are observed on the fascimile re-cord of the radar on 24 July. The estimation of the sameparameters for the period 12201247 LST, with the aidof the available balloon data for the same layers, resultsin relatively longer wave periods, of the order of 10min, as in Table 1.

Figure 6 displays the profiles of the variance of thevertical velocity component and the structure pa-2

w

rameter of the temperature obtained by the sodar on2cT

27 July for the period 12301310 LST. According tothis figure, the variance increases, as it is expected,2w

up to 120 m and then decreases significantly up to theheight of 200 m. At 200 and 300 m, the variance appearsrelatively constant, due to the presence of the elevatedinversion in this layer (see Fig. 3b). On the other hand,the substantial decrease of the temperature structure pa-rameter in the first 150 m above the ground supports2c

T

the surface thermal heating with strong mechanical mix-ing, as expected. The subsequent increase up to theheight of 250 m is associated with the above-mentionedinversion layer. Similar observations and characteristicshave been reported from other experimental studiesclose to the shoreline (Helmis et al. 1987, 1995). Itshould also be mentioned that the observed minimum

of profile at a height of 200 m (Fig. 6) is associated2w

with high wind speeds at heights up to 200 m (see Fig.3c, the 12211319 LST flight). This behavior is in gen-eral agreement with Yoshiki (1997), who found that fora small island the mechanical turbulence generally dom-inates for high wind speeds, and the developed IBL isidentified by a low level jet, when combined with aminimum of

w.

b. The development of the IBL

The top of the mechanically induced IBL may beidentified with changes in the velocity shear / z oru


8/12


FIG. 6. Vertical velocity variance and temperature structure parameter at 12301310 LST 27 Jul2 2 cw T

1996, estimated by the sodar.

TABLE 2. Estimated mechanically induced IBL parameters.

DateTime(LST)

u*

(m s1)z

02

(m)h

i

(m)

22 Jul 199624 Jul 199627 Jul 1996

100010051030

0.890.340.67

1.20.81.05

34.230.832.8

with a sharp discontinuity in slope when plotted in log-linear form. The depth h

iof the IBL that is developed

due to the roughness changefrom a smooth to a roughsurfaceunder neutral conditions can be estimated us-ing the following empirical formula (Kaimal and Fin-nigan 1994):

n

h xi

A , (3)1 z z02 02

where n 0.8, z02 is the roughness length downwind,and x is the downwind distance (fetch). The constant ofproportionality A

1depends on the roughness change and

varies between 0.35 and 0.75. It can be approximated as

A 0.75 0.03 M,1

(4)

where M ln(z01

/z02

), and z01

is the upwind roughnesslength.

Also, the vertical profile of the wind within the IBLunder neutral conditions is given by the well-knownlogarithmic law

u* z2u(x) ln , (5)

k z02

where k is the von Karman constant (0.4), and u* 2

isthe downwind friction velocity.

Using Eqs. (3)(5), the depth hi

of the mechanically

produced IBL is estimated during different experimentaldates. In our case, the fetch x is about 150 m. Incor-porating the meteorological mast data and Eq. (5), theparameters u

* 2and z

02are graphically calculated using

a log-linear (logz, ) plot (see Table 2). From this table,uit can be seen that the estimated values of h

iare less

than 40 m, while their deviations are relatively small.This is expected, since the depth of the mechanicallyproduced IBL actually depends mainly on the roughnessvalue z

02and the fetch, which in our case is constant.

It should be noted that the power-law formula (3)with n 0.8 has been found by Bradley (1968) andAntonia and Luxton (1971) to be well fitted with theirdata. Besides, Arya (1988) and Stull (1988) state thatthe same formula can be applied to unstable conditionswith a slightly larger value of A i than for the neutralcase, thus resulting in relatively larger values of h

i. The

height hi

of the IBL can also be estimated on the basisof the following widely applied formula (Jegede andFoken 1999):

bh ax ,i (6)

where x is the fetch, and a, b are empirical constants.

For fetch distances between 0.01 and 160 m, Walmsley(1989) had obtained values a 0.75 m and b 0.8m. In our case, the use of (6) led to x 150 m in anh

ivalue of 41.3 m, which is slightly higher than the

graphically estimated values given in Table 2. This isin a close agreement with the results of Walmsley(1989).

In order to include changes in scalar fluxes (for tem-perature and moisture) the development of the convec-tive internal boundary layer (CIBL) after a coldhottransition has to be taken into account. There are severalformulas to determine the depth of the CIBL, takinginto account the mechanically (shear) produced turbu-


9/12


FIG. 7. Vertical profiles of potential temperature during the flights of the tethered balloon: 22 Jul 1996 (10011043 LST), 24 Jul (10051053 and 12021247 LST), and 27 Jul 1996 (10271119 and 12231319 LST).

TABLE 3. Estimated CIBL parameters for selected tetheredballoon flights.

DateTime(LST)

D

(102)T

02 T

01

(C)

/ z(103

C m1)h

(m)

22 Jul 199624 Jul 199624 Jul 199627 Jul 199627 Jul 1996

1000102010051025120212201030105012001240

4.411.691.843.613.49

3.83.15.76.07.2

22.226.74.7

112.159.888.769.689.5

lence at the ground and the convectively driven tur-bulence. A similar simplified formula was successfullyused by Raynor et al. (1979), and Stunder and Sethur-aman (1985):

1/2 T T02 01 h x , (7)

D T1

z

where D (u

* 2/

m) 2 is the drag coefficient, u

* 2is theu

downwind friction velocity,m

is the average mixed-ulayer velocity over the land, x is the fetch, T

02and T

01

are the downwind and upwind surface temperatures, re-spectively, and

1/ z is the potential temperature gra-T

dient in the approach flow, which is also the gradientabove the thermal IBL.

The downwind friction velocity u

*

2is estimated

graphically with the aid of a log-linear plot(logz, ),u

using the meteorological mast data. The potential tem-perature gradient above the thermal IBL is estimated

from the ballooon data flights, taking into account thatthe top of the CIBL can be identified by the point where

the potential temperature gradient shows a suddenchange from an unstable to a stable layer (Venkatram

1986). Figure 7 gives the vertical profiles of the potentialtemperature for all experimental days, using the daytime

balloon flights. It is worth noting that the temperaturegradient, though it has different values above the IBLlayer for the different days, remains almost constant

during all the daytime flights of the same day.The estimated parameters that lead to the calculation

of the CIBL depth h, employing meteorological mast

data and the potential temperature gradient above the

thermal IBL from selected flights, are presented in Table

3. From this table, it can be seen that the h

is greater

that the depth hi of the mechanically induced IBL, as it

was expected, varying approximately from 60 to 110 m.

In order to compare these results with the experi-

mental data, the vertical profiles of the potential tem-

perature for all experimental days are examined (Fig.

7). On 24 July 1996, the wind blows constantly from

the northern/northeastern sector throughout the whole

morning, while the speed-up effect is also evident below

200 m (not shown). The developed IBL can be seen

during the 10051053 LST flight up to the height of

150 m, where the profile of the potential temperature

changes rapidly from a unstable surface layer to a stable


10/12


FIG. 8. Vertical profiles of (a) wind direction and (b) potential temperature, during the flights of the tethered

balloon at 21342222 and 23130007 LST 24 Jul 1996.

layer. In the next flight, at 12021247 LST, the top ofthe IBL rises to 200 m, as is indicated by the height ofthe inversion. It is characteristic that the gradient of thepotential temperature above 150200 m remains con-stant during the two flights, indicating the maintenanceof the maritime boundary layer characteristics, despitethe depth change of the IBL. On 22 July 1996, relativelystronger northerly background flow prevails as com-pared to the previous days, which is also reflected inthe vertical profile of the wind speed (not shown). The

potential temperature gradient above the IBL is almostthe same, but the D value is more than double that of24 July, when the IBL also develops. In this case, theestimated depth of the IBL is larger than on the previousdays, which is also found in the temperature profiles(Fig. 7). However, it should be noted that the increasedvalue of h

(112.1 m) in Table 3 is due to the large

value of D

rather than to the enhanced thermal inho-mogeneity.

On 27 July, the wind blows again constantly from the


11/12


FIG. 9. The sodar fascimile record from 2030 LST 24 Jul 1996 to 0054 LST 25 Jul 1996.

northeastern sector, due to the developed sea-breeze cell(see Fig. 3), but with greater stability of the MBL ( /z takes values 2 or 3 times more than the previous

cases). Larger values of the T02 T01 differences andhigher

Dvalues form a value of h

of the order of 100

m, which is in agreement with the slope of the tem-perature profile in Fig. 7. It is worth noting that the

observed profiles result from the IBL being combinedwith a speed-up effect, thus inducing greater depth ofthe IBL. Therefore, the comparison between the exper-imental data (potential temperature profiles of Fig. 7)and the estimations of h

shown in Table 3 for the three

experimental days reveals that the experimental valuesof h

are in fairly good agreement with the estimated

values, using Eq. (7).

c. Nocturnal airflow

Figure 8 shows the vertical profiles of the wind di-rection and temperature on 24 July 1996 during the

nighttime for the periods 21342222 and 23130007LST. In this case, southerly/southwesterly moderatewind prevails within the first 250300 m above theground (Fig. 8a), a result of the combined katabaticflows forming along the surrounding mountain rangesand the slope flow channeling close to the experimentalsite (see Fig. 1). At higher altitudes the wind is sub-stantially reduced and turns to the northeasterly sector(Fig. 8a), following the background flow. In the tem-perature profile, a stable layer that is evident above 350m (Fig. 8b) is clearly related to the airflow from theopen sea. At the lower levels, a ground-based inversionis developed, progressively reaching different heights

from 100 to 350 m. This layer seems to be associatedwith the development of the local katabatic and chan-neling flows. This thermal structure is also evident onthe sounder fascimile record (see Fig. 9). In accordancewith these records, the inversion layering developed be-tween 2030 and 2330 LST with variable depth at dif-ferent heights, and it was followed by a combined

ground-based and elevated inversion, which eventuallyformed a strong surface-based activity. This inversionreached the height of 500 m after 2330 LST, maintaininga different stability factor at the various heights.

4. Concluding remarks

The study of the vertical structure of the ABL underdifferent meteorological conditions over the inland re-vealed the strong influence of the isolated island on themarine ABL structure. Thus, interesting characteristicsof the local flows were observed with the modified MBLalongside the developed IBL on the shoreline, in relation

to the depth, the stability factor, and the thermal struc-ture for all studied cases.

During the nighttime, when the background flow isweak, a complex wind flow is formed, resulting fromthe development and combination of katabatic flowsand topographic channeling flows, with strong stabil-ity and depth of the order of 300 m.

During the daytime, the formation of an IBL up to aheight of 100150 m is evident, depending mainly onthe stability of the incoming airflow, and the temper-ature difference between sea and land.

When the background flow permits the development


12/12


of a sea-breeze circulation, the MBL close to theshoreline has a depth of the order of 400 m, withintense structure and strong layering. In many cases,wavelike activity is developed at elevated layers, withperiods of 7 min.

Well-known empirical relationships for the estimation

of the height of the IBL, including changes in rough-ness and scalar fluxes, provided satisfactory results incomparison with the experimental measurements.

REFERENCES

Antonia, R. A., and R. E. Luxton, 1971: The response of a turbulentboundary layer to a step change in surface roughness. Part 1:Smooth to rough. J. Fluid Mech., 48, 721761.

Arya, S. P., 1988: Introduction to Micrometeorology. Academic Press,307 pp.

Asimakopoulos, D. N., and C. G. Helmis, 1994: Recent advances onatmospheric acoustic sounding. Int. J. Remote Sens., 15, 223233.

Bergstrom, H., P. Johansson, and A. Smedman, 1988: A study ofwind speed modification and internal boundary layer heights ina coastal region. Bound.-Layer Meteor., 42, 313335.

Bradley, E. F., 1968: A micrometeorological study of velocity profilesand surface drag in the region modified by a change in surfaceroughness. Quart. J. Roy. Meteor. Soc., 94, 361379.

Garratt, J. R., 1990: The internal boundary layerA review. Bound.-Layer Meteor., 50, 171203.

Gaynor, J. E., and P. A. Mandics, 1978: Analysis of the tropical marineboundary layer during GATE using acoustic sounder data. Mon.Wea. Rev., 106, 223232.

Helmis, C. G., D. N. Asimakopoulos, D. G. Deligiorgi, and D. P.Lalas, 1987: Observations of sea-breeze fronts near the shore-line. Bound.-Layer Meteor., 38, 395410.

, J. A. Kalogiros, D. N. Asimakopoulos, K. H. Papadopoulos,and A. T. Soilemes, 1995: Acoustic sounder measurements of

atmospheric turbulent fluxes on the shoreline. Global Atmos.Ocean Syst., 2, 351362.

Hsu, S. A., 1986: A note on estimating the height of the convective

internal boundary layer near shore. Bound.-Layer Meteor., 35,311316.

Jegede, O. O., and T. Foken, 1999: A study of the internal boundarylayer due to a roughness change in neutral conditions observedduring the LINEX field campaigns. Theor. Appl. Climatol., 62,3141.

Kaimal, J. C., and J. J. Finnigan, 1994: Atmospheric Boundary LayerFlows: Their Structure and Measurement. Oxford UniversityPress, 290 pp.

Luhar, A. K., B. L. Sawford, J. M. Hacker, and K. N. Rayner, 1998:The Kwinana coastal fumigation study: IIGrowth of the ther-mal internal boundary layer. Bound.-Layer Meteor., 89, 385405.

Mandics, P. A., and E. J. Owens, 1975: Observations of the marineatmosphere using a ship-mounted acoustic echo sounder. J. Appl.Meteor. , 14, 11101117.

Melas, D., A. Lavagnini, and A. Sempreviva, 2000: An investigation

of the boundary layer dynamics of Sardinia Island under sea-breeze conditions. J. Appl. Meteor., 39, 516524.

Petenko, I. V., A. N. Bedulin, and Y. A. Shurygin, 1996: Sodar ob-servations of the atmospheric boundary layer over the ocean

during ASTEX-91. Bound.-Layer Meteor., 81, 6373.Raynor, G. S., S. Sethuraman, and R. M. Brown, 1979: Formation

and characteristics of coastal internal boundary layers duringonshore flows. Bound.-Layer Meteor., 16, 487514.

Soilemes, A. T., C. G. Helmis, P. G. Papageorgas, and D. N. Asi-makopoulos, 1993: A tethered balloon profiler system. Meas.

Sci. Technol., 4, 11631168.

Stull, R. B., 1988: An Introduction to Boundary Layer Meteorology.Kluwer Academic, 670 pp.

Stunder, M., and S. Sethuraman, 1985: A comparative evaluation ofthe coastal internal boundary layer height equations. Bound.-Layer Meteor., 32, 177204.

Venkatram, A., 1986: An examination of methods to estimate theheight of the coastal internal boundary layer. Bound.-Layer Me-

teor., 36, 149156.

Walmsley, J. L., 1989: Internal boundary layer height formulaeAcomparison with atmospheric data. Bound.-Layer Meteor., 47,251262.

Yoshiki, I., 1997: Planetary boundary layer over a small island asobserved by a Doppler sodar. Bound.-Layer Meteor., 85, 379390.

experimental study of the vertical structure of the lower troposphere sea

Documents