Journal of Coastal Research 1119-1129 Fort Lauderdale, Florida Fall 1997

Field Measurements of Longshore Sand Transport and Control Processes on a Steep Meso-Tidal Beach in Portugal

Paolo Ciavola]", Rui Taborda']; Oscar Ferreira] and joao Alveirinho Dias] t Unidade de Ciencias e t Departamento de Geologia Tecnologias dos Recursos Universidade de Lisboa Aquaticos Cidade Universitaria Uni versidade do Algarve Faculdade de Ciencias Campus de Gambelas Bloco C2, 5° Piso 8000 Faro, Portugal Campo Grande 1700 Lisboa, Portugal

ABSTRACT _

CIAVOLA, P.; TABORDA, R.; FERREIRA, 0., and DIAS, J.A., 1997. Field Measurements of Longshore Sand Trans­ .tllllllll:. port and Control Processes on a Steep Meso-Tidal Beach in Portugal. Journal ofCoastal Research, 13(4),1119-1129. Fort Lauderdale (Florida), ISSN 0749-0208. ~.~ A field experiment was carried at Culatra Beach in Algarve (Southern Portugal) to determine longshore transport ~ ~ 7.J rates and sand mixing depth on a steep (slope 0.11) meso-tidal beach. The experiment was undertaken over one and -. b--- a half tidal cycles using sand tracers in conjunction with wave and current monitoring. Variation of mean significant wave height during the experiment was limited (0.34-0.37 m) with mean zero-up crossing periods of 5.1-5.8 sec. Mean longshore current velocities in the breaker zone reached a peak in the second tide (0.28 m sec 1), while they were one order of magnitude smaller during the first (0.02 m sec 1) and third tide (0.04 m sec 1). The increase in current speed was due to a moderate wind that was blowing along shore during the second tide. Average advection velocity of the tracer cloud and longshore currents showed a good correlation, leading to calculation of much larger transport rates for the second tide (1.38X 10 2 m'' sec I) than for the other two (0.23X 10 2 m" sec I). Average depth of sand mixing of 10.6 cm in the beach face was 29% of breaking wave height and showed a marked uni-modal distribution, with maximum of 15 em in the breaker zone. Previously published empirical formulae do not predict satisfactorily this behavior in depth of sand mixing that seems to be peculiar of steep beaches under plunging waves. Empirical formulae were used to compute theoretical longshore transport and compare it with field observations. They all underestimated measured transport rates of about one order of magnitude, thus confirming that the morphodynamics of steep beaches are characterized by relatively high sediment transport even in relatively low energy regimes.

ADDITIONAL INDEX WORDS: Meso-tidal beach, plunging waves, fluorescent sand tracers, mixing depth, longshore transport, longshore wind, empirical models.

INTRODUCTION for predicting longshore sand transport mainly based on mod­ ifications of the original work of KOMAR and INMAN (1970). The prediction of littoral sand transport has important ap­ The main problem in using these empirical notations is that plications both in earth sciences and coastal oceanography they contain an empirical coefficient which shows large vari­ and in the past 40 years efforts by oceanographers, sedimen­ ability due to environmental factors (KOMAR, 1988) that in­ tologists and coastal engineers have greatly improved the un­ clude beach characteristics such as grain size (DEAN et al., derstanding of the physical forces that control the process. Net rates of longshore transport under carefully monitored 1982; KAMPHUIS et al., 1986) and slope (KAMPHUIS et al., physical conditions have been previously studied using large­ 1986). Values of this empirical coefficient as proposed in the scale sediment traps (DEAN et al., 1982; KAMPHUIS et al., international literature varied by a factor of four since intro­ 1986), portable sediment traps (KRAUS, 1987), dispersion of duction (BODGE, 1989). sand tracers (INGLE, 1966; KOMAR and INMAN, 1970; INMAN Recent reviews of field data on longshore sediment trans­ et al., 1980; KRAUS et al., 1982; TABORDA et al., 1994) and port have concluded that there are large uncertainties re­ beach profiles (BEREK and DEAN, 1982). Many of these beach garding the distribution of sediment movement across the experiments have led to the production of empirical formulae surf zone (BODGE, 1989), and the behavior of beaches for transport rates exceeding 0.2 X 106 mvyear, significant wave 96023 received 25 March 1996; accepted in revision 8 August 1996. height larger that 1.8 m, grain size coarser than 0.6 mm and Contribution n. A76 to the Projecto DISEPLA (Sediment Dynamics beach slope steeper than 0.06 (SCHOONEES and THERON, of the Portuguese Shelf), 1993). The lack of information on steep beaches seems to be I Present address: Dipartimento di Scienze, Geologiche e Paleonto­ logiche, Universita di Ferrara, Corso Ercole I d'Este, 32, 44100 Fer­ of particular importance, since their hydrodynamic behavior rara, Italy. is radically different from that of low-gradient beaches 1120 Ciavola el al.

Height (em) 500','------, Berm Top 400 Tra cer Injection 30 _ 1 Rig ~----.... 20

100 Mean Sea Level o

-100 I, ii ', I o 10 20 30 40 50 60 70 80 Distance Irom benchmark (m) Figur e I . Ind ex map of th e Alga rve barrier island sys tem and continen­ ta l she lf (depth contours are in met er s below Mean Sea Level), Figure 2. Beach profile at Culatra Beac h at th e beginning of th e exper­ imen t and location of tracer injection a nd ocea nogra phic rig.

(HUNTLEYand BOWEN, 1975). On steep beaches the surf zone tend s to be narrow and swash processes and wav e character­ in files corres ponding to a duration of 10.24 minutes, thus istics (e.g, plunging) become particularly important for sa nd giving 3072 data points for each time series. remob ilization (JACKSON and NORDSTROM , 1993 ). The pressure tran sducer was calibrated in th e laboratory Th e field experiment described in this pap er took place to esta blish a lin ear relationship between pressure and water over one and a half tid al cycles on 7 and 8 October 1993 on level and to determine the offset typi cal of the sen sor as sug­ th e beach of Cul atra Island in Southern Portugal (Figu re 1). gested by DAVIDSON (1992).A Fast Fourier Tr an sformation Th e experiment was part of a field study (LUAR-Culatra '93 ) was applied to each time series following the methods of undertaken between 7 and 12 October 1992 by several Por­ EARLE and BISHOP (1984) before calculating significant wave tuguese and European universities (Algarve, Lisbon, South­ height (Hs) and mean zero-up crossing period (T) with spec­ ampton, East Angli a, Liverpool, Bordeaux) with the support tral methodologies using routines availab le in th e popular of local and national authorities (lnstituto Hidrografico, MATLAB@J environment (digital signal processing toolbox ). Parque Natural da Ria Formosa, Capitania do Porto de Faro­ Angles of wave approach at breaking point were calculated Olh ao ), by refr action of deep water waves hindcasted for the time of Culatra Islan d is part of the 60 km long barrier isla nd sys­ th e experime nt using a wind stress model (PIRES and Ro­ tem of th e Algarve, that exists under meso-tid al conditions DR IGUES, 1988 ) and validation by field observations. Tim es (about 4 m maximum spring ran ge), and moderate to high when the rig was ins ide th e surf zone were also recorded in wave ene rgy regime (PILKEYet al., 1989). Th e studied beach th e field. Voltage read ings of longshore and cross-shore cur­ had at th e time of the experime nt a reflecti ve profil e (Figure rents collected in th e field were converted into speed units 2) with a stee p foreshore (slope was 0.11), falling th erefore (rn sec ') using a laboratory calibration carried out at th e within a category wh ere publi shed field st udies a re particu­ Univers ity of Southampton following th e methodology of larly scarce ( S C H OO N ~; E S and THERON, 1993 ). G UZA et al. (1988). Hourly average wind speed and direction for th e time of METHO DOLOGY th e experi ment wa s supplied by the weather station of the A multi-instrument rig was deployed on th e beach on 7 Instituto de Meteorologia located at the nearby Faro airport October 1993 at about 17 em above local Mean Sea Level­ (15 km distant, see Figure 1), being collecte d by an anemom­ MSL (Figu re 2). Th e rig included a Sensym LX piezoresistive ete r located at 17 m above MSL. pressure tran sducer with onboa rd a mplificat ion a nd temper­ A composite sediment sample of about 120 kg was collected ature compensation to record waves as in HARDI STY (1988 ). at th e field site on the lower foreshore and a sub-sa mple of Currents were recorded by employing two discus type bi-axi al 120 g was an a lyzed for particle size determination by dry electromagnetic Valeport 800 current meter s mounted verti­ sieving using a set of meshes ranging betw een 4 mm (- 2 Phi ) cally in th e structure at 17 ern (EMCM-1) and 45 ern and 0.063 mm (4 Ph il, followin g the methodology of INGRAtVI (EMCM-2) above sea bed . Data from three successi ve high (1971). Mean gr ain size, sorting and skewness of th e sa nd tid es were logged on th e morning of 7 October 1993 (05:20­ population was calculated using th e FOLK'S (1974) gr aphical 08:23), hereafter indicated as 7/10am , evening of 7 October parameters. After collection th e composite sa mple was dri ed 1993 (17:40-21:28), hereafter 7/10pm, a nd morning of 8 Oc­ in the open air and tagged usin g an orange fluorescent paint, tober 1993 (07:09-09:03), hereafter 8/10am. Data loggin g was according to criteria outlined by previous investigators (IN­ continuous during each tid e using an IBM compa tible desk­ GLE, 1966; YASSO, 1966 ). The tagged sa nd was th en dri ed top PC, running with a n analog to digital conve rte r card at again in th e open air and sieved afterwards using a 2 mm (0 a frequ ency of 5 Hz and seg mentation of th e informa tion was Phi) mesh to screen for aggregates. A sub-sample (120 g) of

Journal of Coastal Resear ch, Vol. 13, No. 4, 1997 Field Measurements of Longs hore Tr an sport in Portu gal 1121

Tabl e 1. Significant Wave Height (HJ and Zero-up Crossing Period (T') measured du ring the experiment . 0.36 Wave Spectra 0.32 Mini- Maxi- 7/lOam mum mum Mean 0.28 7/1 Opm Maximum Mean T, T, T, Si lOam Energy 0.24 Tide H, (m) H, (m) (sec) (sec) (sec) Density (m 2 m z) 0.20 7 October am 0.44 0.37 5 6.5 5.8 7 October pm 0.38 0.34 4.1 6.1 5.1 0.16 8 October am 0.41 0.37 4.6 5.5 5.1 0.12 0.08 the tracer sa nd was th en ana lyzed for grain size determina­ 0.04 tion to assess statistical similarity between th e original sa nd population and the tagged one. On 7 October 1993 th e tracer sand was injected at low tide on the lower foreshore at an hor izontal distance of 16 m from Figu re 3. Wave energy spectra for the th ree tides measure d at hig h the berm top (Figure 2) imm ediately before starting to collect wa ter outsi de the breake r zone. The spect ra are ty pica l of a 10.24 min oceanographic data. The method of injection is comparable to data run . th at used by previous authors (KING, 1951; WILLIAMS, 1971; CORBAU et al., 1994; TABORDA et al., 1994; DOLPHIN et al., 1995) by digging a shallow trench with an area of 2 m" and 10, 10-15 ern, etc.) and a weighted position of th e total ma ss average depth of 3 cm. In order to obtain a n independ ent cente r was calculated according to the tra cer concen tration assessment of th e thickness of th e mixing sa nd layer (KOMAR in each inte rval. The horizontal tran slation of th e cente r of and INMAN , 1970; KRAu s et al., 1982; KRA us, 1985; SUNA­ mass was refer red to the injectio n point for th e first tide, MURA and KI{Aus , 1985; J ACKSON a nd NORDSTROM , 1993) a while the positi on mea sured at th e previous low tid e was control hole 30 cm deep was dug next to the injection trench used for th e second and third counts (KRAUS et al., 1982 ). and filled with tracer sand up to surface level (KING, 1951; Average advection speed of th e tracer was calculated by KOMAR and INMAN, 1970; WILLIAMS, 1971; TABORDA et al., dividing th e distanc e covered by th e center of ma ss during 1994). At each low tide four beach transects wer e studied for half tidal peri od for the interval bet ween two consec utive changes in beach profile using a standard th eodolite, being high tid es (12 hours and 30 minutes). A line of cores was located at 0, 20, 40 and 60 m eastwards of th e injection site. collected at 400 m from th e injecti on point orthogona lly from Following field observations of th e gross direction of long­ th e berm to provide information on cross-shore changes in shore dri ft , sha llow cores (about 25 ern long) of th e beach face th e th ickn ess of th e moving sa nd layer as previously done by were collecte d at low tid e along a grid of sample sta tions east­ KOMAR and INM AN (1970 ). Total sa nd tra ns port ra te was wards of th e injection site. No cores were collected where in th en obtained by multiplyin g the tracer ad vection speed by situ inspection of th e sa nd surface revealed th at th e site was th e area of th e moving sa nd layer (KOMAHand INMAN, 1970). not covered by water at high tide. The cores wer e collected using 30 ern long PVC tubes th at RESULTS were split lengthwise. Each core provided vertical sa mples equivalent to 5 cm intervals. Sa mples wer e put into plasti c Physical Processes bags and back in the laboratory oven dried at 40°C a nd care­ During th e first semi-tida l cycle th at was studied on 7 Oc­ fully desaggr egated using an Agate mortar before undertak ­ tober (7/10a m) wave mot ion was wea k, with H, ran ging be­ ing counting und er an UV source. The total number of grains tween 0.33-0.44 m and T, between 5.0 and 6.5 sec (Table l ). counte d in eac h sample was normali zed by th e sa mple weight Mean H, and T, measured in the breaker zone were respec­ to obtain an absolute mass valu e (number of gra ins per unit tively 0.37 m and 5.8 sec. Sout hweste rly waves were breaking weight of sediment). In order to assess tracer advection be­ at an esti mate d angl e of 5° with th e shoreline, plunging di­ havior a Lagran gian a pproach was adopted (MADSEN, 1987 ), rectly onto th e beach face. The wave spectru m measured at commonly kn own as the Spatial Integration Method (KOMAR high tid e indicates th e presence of long period waves (T= 14 and INMAN , 1970 ; INMAN et al., 1980; fu{AUS et al., 1982; sec) together with shorter peri od waves (Figu re 3). Longsh ore MADSEN, 1987; TABORDA, 1993; TABORDA et al., 1994 ): currents were weak and predominantly eastwards , with EMCM-1 recording a n ave rage value 0.01 m sec- 1 within th e y = L Pidi (1) breaker zone (Figu re 4a) and EMCM-2 recording 0.02 m sec : ' L Pi (Figu re 4b). Maximum tidal eleva tion was 0.84 m above MSL where Y is the location of th e mass center of the tracer cloud and th ere was no significant wind (Figu re 5). along th e two-dimen sional sa mpling grid, Pi is the mass of Wave motion during th e second half tidal cycle on 7 Octo­ tra cer recovered at each grid point and d, is th e distan ce of ber (7/10pm) did not change significantly , with H, ran ging th e grid point from th e injecti on site . For each sa mpling ex­ betw een 0.28- 0.38 m and T, of 4.1-6.1 sec, giving for the ercise th e position of the mass center was calculated inde­ break er zone valu es of mean H, of 0.34 m a nd T, of 5.1 sec pend ently for every depth interval within th e cores (0-5, 5- (Ta ble 1). Southwesterl y waves a pproached the shoreli ne at

Jo urnal of Coasta l Research, Vol. 13, No. 4, 1997 1122 Ciavo la et al.

an angle of about 50, and similarly to the previous tide, plung­ (a) ing waves were breaking directly onto th e foreshore. Exam­ ina tion of the wave spectrum at high tide (Figure 3) points Lower Current Meter (EMCM-l ) out ana logies with th e previous dataset , particularly the ex­ 0.4 istence of a swell with a peri od of about 13 sec. The lower current meter (EMCM-1) stopped working at 18:20, probably 03 becau se it became buried und er th e sea bed. However , prior C ur rent Speed to failure, longshore curren t readings were in the order of (m tse e) 0.2 0.24-0.35 m sec" (Figu re 4a). The upp er current meter Instrum ent Failure (EMCM-2) provided instead more reliable data until starting 0.1 to come out of water at 20:25, giving ave rage readings of an eastward longshore cur rent of 0.28 m sec" (Figure 4b). A o.o~ . ;~ bri sk longshore wind was blowing from a westerl y direction 1 2 3 4 5 6 7 8 9 10 II (azimuth 260-3000 N) with hourly average speeds of up to 28 Data Runs km hr - 1 (Figure 5). Maximum tid al elevation was 0.64 m above MSL. The wave climate during th e third hal f tidal cycle on 8 Oc­ ( b) tober (8/10am) was characte rized by southweste rly waves with H, ranging 0.34-0.41 m and T z of 4.6-5.5 sec (Table 1). Upper Current Meter (EMCM-2) Plunging wave s were breaking directly onto th e beach face 0.4 i ... I at an estimate d angle of 5° with th e shoreline and average wave conditions at breakers were H, of 0.37 m and T, of 5.1 0.3 . ~--- Current sec. The wave spectrum shows that the swell was still present Speed with a per iod of 13 sec but higher frequency waves had in­ (m tsee) 0.2 creased in energy (Figu re 3). Longshore currents wer e eas t­ wards with a magnitude comparable to th e first tide since 0.1 the ave rage recorded by EMCM-1 within the surf zone was o. o ~-v-~ ~ ~ 0.03 m sec:" (Figure 4a) and by EMCM-2 was 0.06 m sec:" l • -:of (Figure 4b). During the collectio n of oceanographic data a 2 3 4 5 6 7 8 9 10 11 Data Runs weak easterl y wind (azimuth 80- ll0° N) was blowing with ave rage hourly speed up to 9 km hr" (Figure 5). Maximum Fi gure 4. Long sh ore current speeds mea sure d during the experiment by tid al elevation was 0.72 m above MSL. the lower (4a ) a nd upper (4b) curre nt meters. Notic e that the lower cur­ ren t me ter (EMCM-ll did not pr ovide a cont inuous record durin g the second tide beca use it beca me buried under the se a bed . Longsh ore cur­ Beach Characteristics and Sand Transport rent rea dings are av eraged over eve ry 10.24 mi n data ru n. The beach foresh ore maintained a steep profile with an av­ erage slope of O.ll during th e time of the experiment. Com­ parison between surveys carried out on 7 and 8 October 1993 sh owed weak accretion above MSL and erosion below MSL

360 - -0-- Direction 30 ~ 320 ...... c:::::=::::J Speed 25 5 280 ...... ------...... o rn Z 24D / ...... --. 20 -g o ~ 200 .. 15 ~ 8, 160 ,.,. '5 120 102. :: e E 80 5 ~ 4D o { I'' I'Ilw l 0 1 23 4 5 67 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 2425 26 27 28 29 Time (h)

Figure 5. Wind da ta collected at th e meteorologica l station of Fa ro Airport. Directions and speeds arc hourl y a verages. Time is indica ted in hours from th e beginning of th e bea ch experimen t, monitored tides a re indica ted with underlined numbers. N.B. Interval s when there was no wind a rc indicated by a n ull value in th e graph of wind direction.

J ourn al of Coastal Research, Vol. 13, No. 4, 1997 Field Mea surements of Longsho re Tra nsport in Portugal 1123

wa ve period ('I'm), the dim ensionless fall velocity m = Hbm/ 300 wsTm ) describing th e beach state during the experiment 250 - 7.10 .93 1 (WRIG HT and SHOHT, 1984) is 1.59 for the first tid e, 1.67 for 200 1== 8 .10 .93 150 th e second and 1.81 for the th ird one, which together wit h Height 100 the mean spring-tide range (about 3 m) clas sify th e beach as (em ) 50 Mean Sea Lev el _ reflective with low tide terrace (MASSE LINK and SHORT, 1993; 0 -, MASSELINK and HEGGE, 1995 ), a conceptual model that -50 - -100 agrees well with th e morphology of th e profile surveyed on -150 8.10 .199 3 (Figure 6) despite that th e latter did not exte nd far 50 55 60 65 70 75 80 85 90 enough to include th e terrace. Ind eed yearly monitoring at a Distance from benchmark (m) nearby site (Praia de Faro) shows that this is th e most fre­ Figure 6. Beach profiles at th e study site car ried out on 7 and 8 October quent beach state for th e barrier-i sland beach es (TOME MAR­ 1993. TINS et al., 1996). On the first low tid e in the evening of 7 Octob er, 24 cores were collected every ten meters from th e injection point in a of about 20 em (Figure 6). During the second tide small beach longshore direction at distances between 0 and 20 m from th e cusps developed, with an average spacing of 20 m on the fore­ berm top. Th e last core was collect ed at a longshore distance shore . In ord er to define the energy regim e of the bea ch, typ­ of 100 m. On the low tide in th e morning of 8 October, 14 ical during the experiment, the surf scaling parameter (E) cores were collected every 50 m along a line at 20 meters from was calcul ated (GUZA and INMAN, 1975 ): the berm top , th e last core being 700 m eastward of the in­ jecti on site . On th e low tide in th e afte rn oon of 8 October 11 (aw2 ) E =--- (2) cores were collected every 100 m at a distance betw een 15 2 g tan J3 and 20 m from th e berm top , with th e last sam ple located at where a is the wave amplitude (H12), w is the wav e radian 1000 m from the injection point. Th e discrepancy betwee n the frequency (2nff), g is the gravitational constant and J3 is th e number of cores recovered at different samplings is du e to beach slope. Values of (E) range from 1.78 for the first tid e, th e fact that sites un covered at high tid e were not consi dered . 2.20 for the second and 2.31 the third one, describing th e Th e position of th e high tid e mark varied during the experi­ morphodynamic behavior of the beach as reflective (GUZA ment du e to changes in tidal excurs ion, beach cusp topogra­ and INMAN , 1975). As recently MASSELINK and HEGGE phy and wave and wind set-up. Th e longshore distance be­ (1995 ) pointed out, (E) is interchangeable with the surf sim­ tween sa mpling transects varied du ring the t hree tid es (every ilarity parameter m of BATTJES(1974) considering that 10 m for the first one , every 50 m for the second one and ~= (n/E )1 /2 . Values of (~ ) for the first 0 .33), second (1.19) and every 100 m for th e third one) in order to obt ain the bes t third tid e 0 .17) pr edict br eaking waves as plunging br eakers, coverage of the area of tracer dispersion consid ering time con­ thus in agreement with field observations. straints . Information on cross-shore variations of t he depth The beach consists of moderately sorted fine to medium of sand mixing were obtained studying in detail a line of core s sand (M,=0.26 mm) with predominant quartz grains. As­ located at 400 m from th e injection point where five cores sessment of grain size distribution before and after coating were collected a t 0, 5, 10, 15 and 20 m from the berm top. with fluorescent paint did not show significant changes (Fig­ Th e majority of tracer recovered in each core , calculated us­ ure 7). Using th e sediment fall velocity (w.) together with the ing a cut-off percentage of 80 % (KHA US et al., 1982), was on measured modal wave height at breakers (Hbm) and modal average distributed within th e interval 0- 15 em in all t hree

100 90 - -.--- Tracer Sand ---f'r-- Beach Sand / 80 ,- / 70 " 60 ,-.' I 50 I I 40 I I ,- 30 /" ", 20 ", 10 ~", o !:r---O--'c...--*==~ ~;.==+:--I----+---+---+---+---+-~ -2 -1.5 -1 -0.5 o 0.5 1.5 2 2.5 3 3.5 4 Dia meter (Phi)

Figure 7. Cumulati ve distribu t ions of grain size in the original and tagged beach sa mples.

J ourn al of Coastal Resear ch, Vol. 13. No. 4. 1997 1124 Cia vola et at.

Distance from top of berm (m) o 2 4 6 8 10 12 14 16 18 20 22 24 26 o I' ! 4iiiiiIlIII:' I I I I I I I I I I

-5 Breakers at Low Tide

-10

-15 1 Mixing Depth (ern)

Fig ure 8. Changes in dep th of activity along a shore-normal transect loca ted at 400 rn from the injection site. The rep resented pattern was constant throughout th e ex periment.

tides . Depth of sand mixin g varied between 2.5 ern in the Zm = 0.027 n, (3) swash zone and 15 em at th e br eak er lin e (Figu re 8). The control hole dug near the injection site and filled up wit h where Zm is the average sa nd mixing depth and H, is th e tracer sand measured a depth of sa nd mixing during the first significant wave height at break ing point. If th e meth od is tide of 13.5 ern, showing the refore good agreement wit h th e applied using the wave data from th e LUAR-Cul atra '93 ex­ results obtained from the cores th at were used for calculating periment, predicted values of avera ge mixing-depth are 1.0 the thickness of the sand layer in motion. From th e graph in ern for tide 7/10am, 0.92 ern for 7/10pm and 1.0 cm for Fig ure 8, it can be seen that the average mixing depth (Zm) 8/10am, thus significantly less th an th e ave rage calculated in the five cores is 10.6 em, while th e maximum values are using th e val ues of Figure 8 (10.6 ern). Moreover, the distri­ obse rve d in the breaker zone (15 ern). bu tion of depth of activa tion in the surf zone at Culatra is Observed ave rage longsh ore velocity of tracers advection uni-modal (Figure 8) and not bi-modal as obse rved by previ­ during th e first tide was 0.96 X10- 3 m sec", lead ing to a dis­ ous studies on gentle beaches (INMAN et al., 1980; KRAus, placement of 43 m of th e mass center of th e tracer cloud . Th e 1985), which show maxima near th e br eak er line and the second tide registe red larger values, with an average advec­ shoreline. Accord ing to J ACKSON a nd NORDSTROM (1993), tion speed of 5.69X 10- 3 m sec:" and a ma ss center displ ace­ break ers plungin g onto stee p foreshores are converted di­ ment of 256 m. Values for the third tid e reversed to levels rectly into swash a nd cannot produ ce variations in activation simi lar to th e first tide, with an ave ra ge advection speed of landward of th e break er line. 0.94 X10 - 3 m sec ! an d a displ acement of 43 m. Calculate d Disp arity bet ween the predictions of KRAus (1985) and rates of total longsh ore tran sport obviously reflect th e data field observations on stee p beaches had been noted in previ­ 2 presen ted above, wit h 0.23 X10- m" sec! for th e first tid e, ous studies (J AC KSON and NORDSTROM, 1993; SHERMAN et 2 1 2 1.38 X10 - m" sec- for the secon d tid e and 0.23X 10- m' al., 1994; TABORDAet al., 1994). Th ere is ind eed some debate 1 sec - for th e third one . in th e lit erature on th e predominant fact ors controlling the depth of sa nd activation at a give n site . Alt hough it is rec­ DISCUSSION ogn ized th at there is a direct relationsh ip of some kind with Sand Mixing Depth breaker height (KING, 195 1; WILLIAMS, 1971 ; KRAus, 1985), The usage of sa nd tracers for empi rica l calculations ofl ong­ it is also evident that th ere is a great variability when data shore sand transport is dependent up on a relia ble assessment from differen t sites are compare d (J AC KSON and NORD­ of the thickness of the moving sa nd layer (KRAUS, 1985; Su­ STROM , 1993).The field database of KRA US et al. ( 982) and NAMURA an d KRA US, 1985). Previous au tho rs have tried to KRAus (1985 ) contained only one beach (Hirono) with char­ ta ke direct measurements in the field using plu g holes filled acteristics compa rable to Culatra. Results from stee p beaches up with tagged sand (KING, 1951 ; WILLIAM S, 1971 ; SONU, in low-wave energy environments have proved th at mean 1972 ; TABORDA et al., 1994) , distribution of tracer sand in depths of acti vation te nd to be greater than in gentle beaches beach cores (KOMAR an d INMAN , 1970; GAUGHAN, 1978; (JACKSON and NORDSTROM, 1993 ; SHERMAN et al., 1994) and KRAUS et al., 1982; KRAUS, 1985; SUNAMURA and KRAUS, of the ord er of 22% of the breake r height according to SHER­ 1985; SHERMAN et al., 1990; SHERMAN et al., 1994; DOLPHIN MAN et al. (1994 ). The ave rage value measured at Culatra et al., 1995) an d depth- of-disturbance rods (GREENWOOD and (10.6 em) provides a relationsh ip with th e wave height at HALE, 1980; J ACKSON an d NORDSTROM, 1993; ))OLPlIlN et break ers (assuming a value of 0.36 m as an average for all al., 1995) . An empirical method for the calculation of th e av­ three tides) of29%, thus showing close agreement with SHER­ erage mixing-depth on a non-barred beach is th at of KRAUS MAN et al. (1994) . In a sim ila r way, th e maximum value ob­ (985): served (15 em) provides a ratio with the br eak ing wave height

Journal of Coasta l Resear ch, Vol. 13, No.4, 1997 Field Measu rements of Longs hore Tr an sport in Portugal 1125

1-r------~

.1 [JD~~D&, 0 Tracer 0 Speed .01 (m/sec) f). f). ~ f). • Komar & Inman , 1970 IP •0 Ingle, 1966 f).• • :1 • ,em • f). Kra us ct aI., 1982 • • t •• 0 Taborda ct al., 1994 • • THIS STUDY . cxx:n +--~~---r~~...... ,.--~~~~~rr_-~_~~~~ .01 .1 1 10 Longshore Current Speed (m/sec)

Figure 9. Plot of tr acer advection speed versus average longshore current speed measured during the exper iment and by previous authors.

of 41%, very simila r to the 40% obtained by WILLIAM S (1971) on Honk Kong beaches. (5)

Longshore Sand Transport where (EC n\ is the wave energy flux wit hin th e breaker zone , (X b is the br eaker a ngle, VI is th e longsh ore current in Th e obtained field measurements of sediment transport of­ the surf zone (here tak en as an average), u., is th e maximum fer an opportunity to test empirical formulae for th e predic­ wa ve orbita l velocity at br eak ers comp uted using lin ear wave tion of longsh ore dri ft although th e single point measure­ th eory and K' is an empirical dim en sionless coefficient th at ments of flow oscillations are spa tially limited. However, KOMAR and INMAN (1970) calcula te d as 0.28 and KRAUS et since the foreshore at Culatra had a consta nt slope and no al. (1982) as 0.21. Since for small (X b the te rm (cos (X b ) can be bar was present, mean flows acro ss th e nearshore should not consi dered eq ual to one, uncertainties in th e det ermination show th e la rge spatial variations observed on barred beaches. of the break er angle have lim ited effects on th e determination Although criticisms have been mad e of tracer methodology of II(KOMAR , 1988). due to its sensi tivity to variations in th e transporting system Alth ough Equ ati on 5 is often pr esen ted in the form directly and inability to reach equilibrium conditions (SCHOONEES related to t he longsh ore compone nt of wave power at break­ and THERON, 1993 ), th e techn ique is nevertheless simple to use for estima ting total longshore drift within an entire semi­ ers (KOMAR and INMAN , 1970), whe never exte rnal factors tidal cycle. In th e Culatra experiment th e advection speed of such as tid es, cell circulatio n or local winds are present, such the tracer cloud correlated with the acce leration in longshore a formula cannot be applied (KOMAR, 1976; ALLEN, 1985; Ko­ current speed th at was measured du ring the second tide. Th e MAR, 1988). Th e differen ce in magnitud e of longsh ore current results are consiste nt wit h th ose obta ined by other authors that was observ ed in the break er zone on 7/l0pm (average l using compa ra ble methodologies (Figure 9), with the excep­ speed of 0.28 m sec ) can only be explained by the effect of tion of th e data of INGLE (1966), wh ere advection speeds are the local longshore wind blowing at the time of the measure­ great er. However, careful re-interpretation of that data set ments in th e sa me directi on as th e current, since no appre­ may prove fru itful as th e transport rate was computed using cia ble cha nge in wave cha rac te ristics was observed, thus jus­ only surface dispersion of th e tracer . tifying the application of Equation 5. HUBERTZ (1986) had For computations of th eoretical tran sport rates it is useful indee d noticed at th e CERC's Field Research Fac ility (Duck, to express rat es of littoral drift as immersed tran sport rates Carolina, USA) th at longshore wind can increase the long­ (II) as given in INMANand BAGNOLD (1963): shore current up to a factor of three. Th ere is however the problematic issu e of what value of K' II (p, - p)(1 - p)Q = (4) should be used , since th ere are environmental controls (e.g. where P, is the sa nd density, p is the sea wate r den sity, p is grain diamet er , settling velocity , beach slope, wave steep­ t he sa nd porosity (0.4) and Q is the volume of sand transport. ness) that should be accounted for (KOMAR , 1988). Pr evious If one uses S.1. un its for th e den sities [kg m-I] and for the papers (ALLEN, 1985 ; ALLEN, 1988 ) have discussed t he vari­ volumes [m"], (II) will be expressed as [Newton sec - I], which ability of th e empirical coefficients in the two equations of have a physical meaning and can be related to the forces dr iv­ KOMAR & INMAN (1970) desp ite attem pts of resea rchers to ing the sediment alongsh ore (BAGNOLD, 1963; INMAN and relate the m to mean sa nd diameter (DF:AN et al., 1982), beach BAGNOLD , 1963): slope, signi ficant break er height and bre aker index (KAM-

J ournal of Coas ta l Research, Vol. 13, No. 4, 1997 1126 Cia vola et al.

D Field Measurements D k'=0.21 (Kraus et al., E!ll k'=0.28 (Komar & Inman, 1982) 1970)

[] k'=0.45 (after Dean et al., [ill k'=0.34 (after Kamphuis 1982) er al., 1986)

1000.00

100.00 " '" -­'"c: o 10.00 ~ OJ ;Z 1.00

0.10 v I I 71J0am 71J0pm SIlOam Tide Figure 10. Comparison betw een observe d and predicted values of Immer sed Weight Sand Tr ansport Rate (I,). The predictions are computed using different K' coefficients in Equation 5.

PHUISet al., 1986). Val ues offield observations of'I, calculated diction than the energetics model , since the orbital wave ve­ using Equation 4 are compared in Figu re 10 with theoretical locity in Equation 5 is difficult to measure. However, since in predi ctions using in Equation 5 the K ' constants of 0.28 (Ko­ the Culatra study large non-wave driven currents were pres­ MAR and INMAN, 1970) and 0.21 (KRAus et al., 1982), or K' ent during the second tide, for uniformity of data comparison values of 0.45 calculated using the methodologies of DEANet the energetic ap proach is preferred. al., (1982 ) an d 0.34 of KAMPHUIS et ai.(1986). All solutions The difference between field observations at Culatra Island predict a transport rate an order of magnitude les s than that and theoretical predictions oflongshore transport may be due observed, thus within the accuracy of prediction that ALLEN to limited field information in the literature on steep beaches (1988) believes to be typical of the methodology. In a field (SCHOONEES and THERON, 1993). As it can be seen in Figure study comparable to Culatra Island, ALLEN(1985) concluded 11, there is some scatter if all data on reflective foreshores that the energy flux ap proach offers better accuracy of pre- (slope larger than 0.05) are plotted together, which becomes

Field Database of Longshore Transport on Reflective Beaches 10000 i, , I • Culatra(ThisStudy) '" Cantodo Marco (Taborda at aI., 1994) o EIMorano (Komar and Inman, 1970) 1000 !J. Hirono 1·2 (Kraus at aI., 1982) Immersed o Torrey Pinas(Inmanat aI., 1980) Weight x Sandy Hook1bl , 1b2, le, 2b (Allan, 1985) [Nt/s ec] 100 x • x x x x • x • 10 x x x XX x

10 100 1000 10000 (EC n)v I I U m [Wattlm]

Figure 11. Measured Immerse d Weight Sand Tr ansport Rat es (I,) obtained du ring field studies of reflective beaches (slope larger than 0.05). Notice th e restricted information available at th e lower and upp er ends of th e scale. Th e lin e represents Equation 5 fitted with a K' coefficient of 0.28.

Jo urnal of Coastal Research, Vol. 13, No. 4, 1997 Field Measurements of Longshore Transport in Portugal 1127 more evident if the theoretical line of Equation 5 is plotted transport rates estimated using sediment budget analysis with a K' of 0.28 (KOMAR and INMAN, 1970). The data from from wave models and field observations might be much larg­ EI Moreno Beach (KOMAR and INMAN, 1970) show a more er than the one order of magnitude measured in this study linear trend than the rather large database of ALLEN (1985) or pointed out in previous reviews. The universal applicabil­ which comes from several locations along the ocean side of ity of coefficients determined by other authors should not be Sandy Hook (NJ, USA). INMAN et al. (1980) even admit large assumed, because of the empirical nature of such parameters. uncertainty in their data, publishing not single values of Each beach type is believed to behave in an almost unique transport but ranges for each experiment. way regarding longshore sand transport and steep reflective The available field database becomes even more restricted beaches are certainly the least understood. when only beaches with slope larger than 0.08 are consid­ ered. Published studies that are directly comparable to Cu­ ACKNO~EDGEMENTS latra Beach are only EI Moreno Beach in California (KOMAR and INMAN, 1970) and Hirono Beach in Japan (KRAus et al. The authors are grateful to all the Portuguese and Euro­ 1982), where however average (D.')o) grain size was coarser pean institutions who participated to the field study or made (about 0.60 mm). Results of a study carried out in Portugal resources available for its realization. We are also deeply in­ by TABORDA et al. (1994) at Canto do Marco Beach debted with all the people who helped in the field sharing (slope=0.08 and D.')o=0.60 mm ) in high wave energy condi­ with us the pleasure of working on the beaches of the Ria tions are also presented in Figure 11 and seem to agree re­ Formosa Natural Park. A special mention to Ms. I. Siborro markably well with the theoretical prediction. for carrying out laboratory analyses, Dr. H. Pires for running the wind-stress model, Mr. S. Marsh, Mr. M. Wilkin and Dr. G. Voulgaris for helping us to process the oceanographic data. CONCLUSIONS Data analysis was supported by the EU-HeM Project "Coast­ The field study carried out at Culatra Island confirmed a al Environments (E-Atlantic Physical Processes)" n. ERB/ good correlation between speed of advection of tracer sands CHRXlCT94/0541 while fieldwork activities were partially fi­ and longshore currents. Despite an almost constant wave re­ nanced by the EU-MASTII Project "C-STAB". Useful com­ ments to the structure of the manuscript were made by Dr. gime throughout the experiment (Hs=0.28-0.44, a=5°) there was a change of one order of magnitude in longshore current C.D.R. Evans (British Geological Survey, UK) and Mr. A. and transport during the second tide (7/10pm). This is ex­ Klein (University of Vale do Itajai, Brazil). The final version plained as due to a change in the azimuth and speed of local of the paper was improved thanks to careful reviews by Dr. wind, that was blowing longshore with an average speed of J.R. Allen (Boston, USA) and by an anonymous JCR referee.

28 km hr 1 during that part of the experiment. Such a phe­ nomenon has confirmed that is preferable to use the energetic liTERATURE CITED model (BAGNOLD, 1963; INMAN and BAGNOLD, 1963), rather than the wave power approach (KOMAR and INMAN, 1970), ALLEN, J.R., 1985. Field measurements of longshore sediment trans­ port: Sandy Hook, New Jersey, USA. Journal ofCoastal Research, for the prediction of longshore sand transport if field mea­ 1(3), 231-240. surements of longshore current are available. ALLEN, J.R., 1988. Nearshore sediment transport. Geographical Re­ A disparity has been identified between field measure­ view, 78(2), 148-157. ments of sand transport and predictions using empirical for­ BAGNOLD, R.A., 1963. Mechanics of marine sedimentation. In: HILL, M.N., (ed.), The Sea, vol. 3. New York: Wiley-Interscience, pp. mulae (KOMAR and INMAN, 1970; DEAN et al., 1982; KRAus 507-528. et al., 1982; KAMPHUIS, et al., 1985), proving that the accu­ BATT.JES, J.A., 1974. Surf similarity. In: Proceedings of the 14th racy of the methodology is not better than one order of mag­ Coastal Engineering Conference (Copenhagen, Denmark, ASCE), nitude (ALLEN, 1988). This could be due to several factors, pp. 466-480. such as scarcity of similar data in the literature and errors BEREK, E.P. and DEAN, R.G., 1982. Field investigation of longshore transport distribution. In: Proceedings of the 18th Coastal Engi­ in measuring the effective sand transport rate. Field obser­ neering Conference (Cape Town, South Africa, ASCE), pp. 1620­ vations of the depth of activity due to wave action have con­ 1639. firmed the limitation of empirical formulations (KRAUS et al., BODGE, K.R., 1989. A literature review of the distribution of long­ 1982; KRAUS, 1985) when applied to steep beaches under shore sediment transport across the surf zone. Journal of Coastal Research, 5(2), 307-328. plunging wave conditions, as this had been noted by JACK­ CORBAU, C.; Howx, H.; TESSIER, B.; DE RESSEGlJIER, A., and CHAM­ SON and NORDSTROM (1993) and SHERMAN et al. (1994). LEY, H., 1994. Evaluation du transport sedimentaire sur une plage The present study does not introduce new empirical nota­ macrotidale par tracage fluorescent, Dunkerque Est, France. tions to replace those in the literature, since it is only based Compte e Resumes Academie de Sciences de Paris, 319, 1503-1509. on one particular beach within quite a restricted wave energy DAVIDSON, M.A., 1992. Implementation of linear wave theory in the frequency domain for the conversion of sea bed pressure to surface spectrum and experience has shown the risks of generaliza­ elevation. School of Civil and Structural Engineering Report n. 92­ tion. The main conclusion is that it is necessary to improve 008. Plymouth, UK: University of Plymouth, 20p. the understanding of the sedimentary behavior of steep re­ DEAN, R.G.; BEREK, E.P.; GABLE, C.G., and SEYMOUR, R.J., 1982. flective beaches, where, because of the short surf zone, wave Longshore transport determined by an efficient trap. In: Proceed­ ings of 18th Coastal Engineering Conference (Cape Town, South energy dissipation takes place over small distances, leading Africa, ASCE), pp. 954-968. to relatively high alongshore sediment transport rates. If this DOLPHIN, T.J.; HUME, T.M., and PARNELL, K.E., 1995. Oceano­ concept is accepted, discrepancies between mean annual graphic processes and sediment mixing on a sand flat in an en-

-Iournal of Coastal Research, Vol. 13, No.4, 1997 1128 Ciavola et al.

closed sea , Manukau Harbour, New Zeal and. Marin e Geology, 128, measurem ents of sedime nt tran sport rat es in the surf zone . Jour­ 169- 181. nal of Coastal Research . 3(2), 139-152. EAllLE, M.D. an d BISHOP, J .M., 1984. APractical Guide to Ocean K/{A LJ S, N.C.; ISOBE, M.; I(ii\Ri\SHI, H.; SASAKI, T.O. , an d HOHlKA­ Wave Measurement and Analysis. Marion, Maine: Endeco, Inc., WA , K, 1982 . Field experi ments on longshore sa nd transport in 78p . the sur f zone. In: Proceedin gs of ISth Coasta l Engine ering Confer­ FOLK, RL., 1974 . Petrology of S edimenta ry Rock s. Austin, Texas : ence (Cape Town , South Africa. ASCE ). pp. 970-988. Hemphill, 183p. MADSEN. O.S.• 1987. Use of tracers in sedime nt transport st udies. GAlJ(;IIAN, M., K , 1978. Depth of disturbance of sand in surf zones . In: Proceedin gs ofCoastal Sediments 'S7 (New Orl ean s, Louisian a, In: Proceedings of16th Coastal Engineering Conference (Ha mburg, ASCE ), pp. 424-435. German y, ASCE ), pp. 1513-1530 . MASSELINK, G. and H ~;(;( ; E , B., 1995. Morphodynam ics of meso- and Glll';I';NWOOD, B. and HALE, 1'., B., 1980. Depth of act ivity, sediment macrotidal beaches: exa mples from centra l Queensland, Australi a . flux and morphological cha nge in a barred nearshore env ironment. Marin e Geology. 129, 1- 23. In: MCCAN N, S.B. (ed.), The Coastline of Canada , Pap er 80-10, MASSELlNK, G. a nd SIIOllT, AD., 1993. The effect of tide ra nge on Hali fax , Canada: Geological Survey of Cana da, pp. 89-109. beach morphodynamics and morphology: A conceptua l beach mod­ GUZA , R T. and INMA N, D.L. , 1975. Edg e waves and beach cus ps. el. Jou rnal of Coastal Research , 9(3), 785-800. Journal of Geophy sical Research, 80 , 2997- 3012. PILKEY, O.H ., JR; NEAL, W.J .; MONTI';II{O , J.H., and DIi\S, J .M.A., GUZA, RT.; Cr.n-ro», M.C., and REZVANl , F. , 1988 . Field intercom­ 1989. Algarve barrier islands: a non coas ta l-pla in syste m in Por­ pa rison of electromagnetic current met ers. Journal ofGeophysical tu gal. Journal of Coastal Research , 5(2), 239-261. Research. 93, 9302-9314 . PIIU;S, H. and ROIlIUmJES, A, 1988. Modele de Agita rao Maritima HAIWI STY, J ., 1988. Mea su rem ent of shallow water wave direction MAR 211. Monografia de Met eorologia e Geofisica No.3, Lisbon, for longshore sedime nt tran sport. Geo-Marine Letters, 8, 35-39. Portugal: Instituto Nacional de Meteorologia e Geofisica , 21p (in HUBERTZ,J .M., 1986 . Observations oflocal wind effects on longsho re Portuguese ), currents. Coas tal Engineering. 10, 275-288. SCIIOONEES, J.S. and TIIE RO N, A ,K, 1993. Review of th e field-d ata HUNTLEY, D.A. and Bow tcrc, AJ., 1975. Comparison of th e hydro­ base for longshore sedime nt tran sport. Coastal Engineering. 19. dyn amics of steep and shallow beaches. In : HAIL, J .R, and CARR, 1-21). A.P .• (eds ), Nearshore S ed iment Dynamics and Sedi mentation. SIIERMA N, D.J .; BAUER, B.O.; NOIWSTIWM, KF., and ALLEN, J.R, New York: Wiley-In ter science, pp . 69-109. 1990 . A tracer study of sedime nt tran spo rt in the vicinity of a IN(iLE, .J.C., 1966 . The Movement of Beach Sand. Amsterdam, HoI­ groin. Journ al of Coastal Research. 6(2), 427-438. land:Elsevier, 22Ip. SHEHMA N. D.J. ; NOIWSTHOM , KF.; J ACKSON, N.L.; and ALLEN, J .R, IN(;RA;\l, R.I.., 1971. Sieve Analy sis. In: CARVER, RE., (ed.), Proce­ 1994. Sediment mixing-depths on a low-en ergy re flective beach. d ures in S edi ment ary Petrology. New York: Wiley-Interscienc e, pp. Jou rnal of Coastal Research , 10(20 ), 297- 305. 49-67. SONLJ ,C.J .• 1972 . Bimodal composition and cyclic cha rac teristics of IN:\I AN, D.L. and BAGNOLD , RA, 1963. Lit toral Proc esses. In : HILL, beach sedime nt in continuous ly cha nging profil es. Journ al ofSed­ M.N., red.), Th e S ea, vol. 3, New York : Wiley-Interscien ce, pp . imenta ry Petrology, 42, 852- 857. 529- 553. SUNi\M URA, T. and KllAUS, N.C., 1985. Predi ction of average mixing INM AN, D.L.; ZAMPOL, J.A. ; WHITE, T.E., HANt:s , D.M.; WALDORF, depth of sedime nt in the surf zone. Marine Geology, 62, 1-12. B.W., a nd KASTENS, KA, 1980 . Field measurem en ts of sand mo­ TABORIlA , R , 1993. Modelacao da Dinarni ca Sedimentar Induzida tion in th e sur f zone. In : Proceedings of 17th Coastal Engineering Conference (Sydney , Australi a , ASCE), pp. 1215-1234. pela Ondulacao na Plataforma Continental Portuguesa, unpub­ .]ACKSO:--: , N.L. an d NOlWSTROM , KF., 1993. Depth of activation of lish ed Master's Thes is, Departm ent of Geology , Uni versity of Lis­ sedime nt by plunging br eak ers on a stee p sandy beach. Marine bon , 126 p (in Portuguese ). Geology. 115, 143-1 51. TABORIlA , R ; F EHH EIRA , 6.; DIAs, J .; ALVEIIU NIIO, and MOITi\, 1'., KAMI'llLJ IS, J. W.; DAVIES, M.H. ; NAIRN, RB., and SAY AO , O.J. , 1986 . 1994. Field observations of longshore sa nd trans port in a high en­ Calculati on of littoral sa nd tran spo rt rate. Coasta l Engineering, ergy environme nt. In: Proceedings of Littoral 94. Second Interna­ 10,1-21. tion al Symposium of th e Eur opean Coastal Zone Association for KIN(i, C.AM., 1951. Depth of disturban ce of sa nd on sea beache s by Scien ce & Technology, Lisbon, Portugal , EUROCOAST, pp. 479­ waves. Journ al of Sedimen tary Petrology, 21, 131-140. 487 . KO:\IAll, P.D. , 1976 . Nears hore curren ts an d sedime nt tra nsport, and TOME MARTINS. J.; F~; HI{I<;IHA , 6.; CIAVOLA, P., a nd DIAS, J . A., the resul ting beach configu ration. In: STANLEY, D.J. , and SWIFT, 1996. Monitoring of profil e chan ges at Prai a de Faro (Alga rve): a D.J .P., (eds r, Marine S ediment Transport and Environmental tool to predi ct and solve probl ems. In: TAus sIK, J . and MITCIIELL Management, New York: Wiley, pp. 241- 254. J .(eds r, Partn ership in Coastal Zone Management . Cardiga n,Unit­ KOM All, P.D., 1988. En vironmental Controls on Littoral Sand Tran s­ ed Kingdom: SAMARA, pp. 611)-622. port. In: Proceed ings of 21st Coastal Engi neering Conference (Ma ­ WILLIAM S, AT., 1971. An analysis of some factors involved in the lag a , Spa in , ASCEI, pp. 1238-1252 . depth of disturbance of beach sand by waves. Mar ine Geology , 11, KOMi\R, P.O. a nd IN:\Ii\N, D.L., 1970 . Long shore sand trans port on 145-1 58. beaches. Journal of Geophysical Research , 75, 5514- 5527. WllIGIIT, L.D. and SHOHT, 1984. Morphodyn amic va riability of surf KHAU S, N.C., 1985. Field expe rime nts on vertical mixing of sand in zones a nd beache s: a synt hesis. Marin e Geology, 56, 93-118. th e surf zone. Journal of Sedimentary Petrology , 55, 3-14. Yi\SSO, W.E., 1966. Formulation and use of f1uorescen t tracers coat­ Klli\US, N.C., 1987. Application of portable trap s for obt aining poin t ings in sedime nt tran sport studies . S edimentology. 6, 287- 301.

D RIASSU NTO [J 1<:' st.uo svolto un cs pc rimcnto di campagna sulla spiaggia di Culat ra rPort oga llo meridionale ), per dcterminarc il ta sso di trasp orto di sa bbic lungocosta o la profond ita di rimaneggiamento del sedimento di rondo SlI una spiaggia rnosot idal e (marca semidiurnal eon elcvata pendenza (tani> 0,11 1. L'ospori mont.o e stato cffottuato durant e un ciclo c mezzo di marea utili zzando delle sa bbie marca tc come traccia nti assiemc a misur e del moto ondoso e corrc ntorncu-ichc. La variazione dcll'altczza signifi eativa media delle onde durante l'esperirn cnto e stata limitata (0,34- 0,37 m ), mentrc il periodo medio di zero-up crossing l' stato ncll'ordine di 5,1-5,8 see. La volocit.i media delle correnti lungocost a ha raggiu nto un picco massimo duran te la seconds maroa (0,28 m sec I ) pe r In prescn zn di un vento moderato che soffiava lun gocosta , mentre durante la prima (0,02 m sec 1) C la ter za marea (0,04 m sec I) i valori di corrcnte sono stn t i nettamentc inferio ri. La vclocita media di trasporto dell a sabbia t racciunte 6 risulta ta fort cm cnt e correlata con qu ella delle correnti lungocosta con una vclocita di trasporto molto maggiore per la seconda marc-a ( 1 . :l H~ 10 ' m' see 'I chc per lc altrc due 10,23 x 10 ' m' see I ). La profondita di rim an eggiarncnto del scdirncnto di fondo nella porziono int crt.idalc della spiaggiasi (' mantcnuta costantcmcntc su un valore media di 10,6 em durant e tutto l'csperiment o. Essa c risultata circa 2W1r dcll'alto zzu d'ond a ncl punto di frangenza can una dist nbuzio nc unimod alc c con puntc rnassime di 15 em nell a zona dci fra ngont i. Lc formulazioni cmpirichc pubhlicatc da preceden t: a ut ori non riproducono in man icra soddis faconte talc comportamcnto dell a profondita di rimaneggiamento su spiaggc a pcndenza clcvata sotto l'intlucn za di fra ngon t: di tipo plunging. E'stato

Journal of Coasta l Research, Vol. 1:3, No.4, 1997 Field Measurements of Longshore Transport in Portugal 1129 inoltre calcolato il tasso teorico di trasporto lungocosta e comparato con i dati di campagna. Tutti i cal coli teorici sottoestimano le osservazioni rcali di circa un ordine di grandezza. Lo studio mette in evidenza come le spiagge a pendenza elevata presentino un tasso di trasporto alto anche in condizioni energetiche relativamente basse.

D RESUMO D Realizou-se em 7 e 8 de Outubro de 1993 uma experiencia de campo, na Ilha da Culatra (Algarve), tendo como objectivo a dcterrninacao de taxas pontuais de transporte longilitoral e de valores de profundidade de rernobilizacao da areia numa praia mesotidal com forte pendor (tan~ = 0.11). A experiencia decorreu durante urn ciclo e meio de mare, tendo sido utilizada a metodologia de areias marcadas por fluorescencia em simultaneo com a aquisicao de dados de ondas e correntes. As medias das alturas significativas e as medias dos periodos obtidos durante a experiencia tiveram variacao reduzida, de 0.34 m a 0.37 m e de 5.1 sec a 5.8 sec respectivamente. A velocidade media das correntes longilitorais na zona de rebentacao atingiu urn pico na segunda mare (0.28 m sec 1), enquanto que na primeira (0.02 m sec 1) e na terceira mares (0.04 m sec 1) as velocidades obtidas foram uma ordem de magnitude inferiores. 0 aumento da velocidade da corrente deveu-se a existencia de vento moderado no sentido da corrente, durante a segunda mare. As velocidades medias obtidas para 0 transporte do tracador mostram uma relacao directa com a velocidade da corrente, permitindo 0 calculo de uma taxa de transporte sedimentar muito superior para a primeira mare (1.38x 10 ~ m' sec 1) do que para as outras duas mares (O.23X 10 ~ m' sec 1). A profundidade de rernobilizacao da areia na face da praia foi relativamente constante ao longo da experiencia, com urn valor medic de 10.6 ern (29 lk da altura significativa na rebentacao) e urn maximo de 15 em, na zona de rebentacao. Varias formulas ernpiricas frequentemente utilizadas nao reproduzem satisfatoriamente os elevados valores obtidos no campo para a profundidade de remobilizacao, facto que parece ser caracteristico de praias com declive acentuado e com ondas na rebentacao do tipo "plunging". Foram ainda testadas as f6rmulas empiricas mais frequentemente utilizadas para 0 calculo do transporte longilitoral, tendo-se comparado os seus resultados com as observacoes de campo. Todas as f6rmulas utilizadas subestimaram a taxa de transporte em diferencas de valores pr6ximos de uma ordem de grandeza, confirmando que a morfodinamica de praias com declive acentuado ecaracterizada pelo elevado transporte sedimentar, mesmo que em regime de baixa energia.

Journal of Coastal Research, Vol. 13, No.4, 1997