HYDROLOGICAL STUDIES AND SURFACE CURRENTS 

IN THE COASTAL AREA OF THE BA Y OF BISCA Y.

Miguel Ibañez Artica

Instituto Geográfico Vasco

LABURPENA

ITSAS AZALEKO UR-KORRONTEEN IKERKET AK BIZKAIKO GOLKOAN

Lan honetan, Bizkaiko Golkoaren hego sortalde inguruko korrenteei buruz 1978.an lortutako ezaguerak azaltzen dira. 1.500 plastikozko txartel bota dira itsasora, illero eta puntu ezberdinetatik jaurtiaz. Urtean zear bi korronte era nagusi dabiltza; bata, negu garaikoa, W ~ E dator espainiako kantauri itsasertz parean barrena, eta azkenik. frantziaT<o itsas ertza jarraituaz S ~ N okertzen da; azkartasun handienean azaroan, abenduaren lehen egunetan eta otsailean dimardu. Udaran, ekainaren erditik urriaren amaierara, korronteak harat-honuntz dabiltza haizearen eraginpean E ~ W batzutan, W ~ E bestetan. Abuztu eta urrian E ~ W korrontea da garaile. Itsas azaleko uren berotasuna eta korrontearen arteko eragin bat aurkitzen da; udaran. bereziki, Euskadi inguruko itsas urek jasatan duen berotzea zerbait ulertarazten duena. Berotze hau ur hauetan aurkitzen diren Mediterraneo erako bentiko bizidun multzoak hazteko bear duten bal. dintza garrantzizkoa da.

ESTUDIOS HIDROLOGICOS y DE CORRIENTES SUPERFICIALES EN EL AREA COSTERA DEL GOLFO DE VIZCAYA

En el presente trabajo se exponen los resultados obtenidos sobre las corrientes superficiales en el extremo SE del Golfo de Vizcaya durante 1978. A tal fin se han utilizado 15.000 tarjetas plastificadas en total, que se han lanzado al mar desde diferentes puntos de la costa mensualmente. Se manifiestan durante el año dos tipos fundamentales de corrientes, el primero característico de los meses de invierno consiste en una circulación homogénea que discurre paralela a la costa cantábrica española en dirección WE para más tarde ascender SN paralelamente a la costa francesa, estas corrientes presentan su máxima intensidad en los meses de febrero. noviembre y sobre todo en los primeros días de diciembre. En los meses de verano, desde mediados de julio hasta finales de octubre, las corrientes son de tipo oscilatorio, dependiendo de la dirección de los vientos dominantes en cada momento, alternando corrientes EW con W E. En algunos meses como agosto y octubre predomina la corriente E W. Comparando con los valores medios de las temperaturas superficiales del agua de mar, se observa una relación entre las corrientes y las temperaturas que contribuye en parte a explicar el fenómeno de calentamiento del agua frente a la costa vasca en los meses de verano, factor muy importante que condiciona el desarrollo de comunidades bentónicas litorales de tipo mediterráneo en esta costa.

ABSTRACT

HYDROLOGICAL STUDIES AND SURFACE CURRENTS IN THE COASTAL AREA OF THE BA Y OF BISCA Y.

The results obtained on the surface currents in the extreme SE of the Bay of Biscay during 1978 are given in the present work. For this end. 15,000 plastic drift cards were thrown into the sea from different coastal points. During the year two fundamental types of currents are manifest. The first characteristic of the winter months consists of a homogeneous circulatjon which circulates parallel to the Spanish Cantabrian coast in the direction W ~ E to ascend later S ~ N parallel with the French coast. These currents show their maximum intensity in the months of February. November and in the first days of December. In the summer months, from the middle of July to the end of October, the currents are oscillatery. Depending all the time on the direction of the prevajling winds, the currents alternate E ~ W with W ~ E. In some months like August and October, the current E ~ W prevails . Comparing the average values of the temperatures of the surface sea water, a relation between the currents and temperatures is observed and which contrjbutes in part to explain the phenomenon of the warming-up of the water in front of the Basque coast in the summer months, a very important factor which conditions the development of the benthic shore communities of a Mediterranean type on this coast.

RESUMÉ

Dans le présent travail sont exposés les résultats obtenus sur les courants de surface a l'extremité S.E. du Golfe de Gascogne pendant 1978. A cette fin on a utilisé un total de quinze mille cartes plastifiées qui on été lancées en mer mensuellement a partir de différents points de la cote. Deux types fondamentaux de courants se manifestent pendant I'année: le premier, caractéristique des mois d'hiver, constitue une circulation homogene qui court parallele a la cote cantabrique espagnole en direction W ~ E pour remonter ensuite S ~ N parallelement a la cote francaise; ces courants présentent leur intensité maximum surtout dans les premiers jours de décembre et aux mois de février et novembre. Aux mois d'été, depuis la mi-juillet jusqu'a la fin octobre, les courants sont de type oscillatoire, en fonction de la direction des vents dominants a chaque moment; une alternance de courants E ~ W et W ~ E se produit. Dans certains mois, comme aoOt et octobre, le courant E-)0 W prédomine. Si I'on compare avec les valeurs moyerines des températures de surface de I'eau de mer, on observe un rapport entre les courants et les températures qui contribue en partie a expliquer le phénomene d'échauffement de l'eau face a la cote basque aux mois d'été, facteur tres important qui conditionne le développement de communautés benthiques littorales de type méditerranéen sur cette cote.

Traditionally. the method used to measure the direction and speed of surface currents has been to use drifting floats.

The first observations made in the Bay of Biscay date from the end of the last century (GARSTANG. 1898) who observed a drift of floating bottles going towards the SE with a speed of between 8 and 30 nautical miles a day.

For this type of study, first of all glass bottles with a message inside were used. As these bottles went without ballast, it was noticed that in many cases they suffered a translation due more to the wind than to surface currents (CARRUTHERS, 1927) .

KURC (1956) , used bottles with ballast which now have been substituted by red low density polythene cards (DUNCAN, 1965) in Australia, South Africa and USA.

These cards have a long life and in many cases have crossed the Ocean from South Africa to Central and North America (SHANNON et al., 1973) .the only inconvenience, as we have been able to verify in collecting some of these cards from the USA, is that they are almost totally covered in a layer of oil with the result that in many cases it is difficult to decipher (1).

Fig. 1: Type of drigt-card used

A priori, we decided on four fixed points for throwing the cards along the coast of Guipuzcoa: Fuenterrabia. San Sebastián. Orio y Zumaya, occasionally they were thrown off other points such as Biarritz. Mundaca. Santander. Vigo, etc. ..

The throwing was done from the coast itself or from the sea near the coast. In the very mome.nt in which the tide began to fall immediately after high tide. bundles of 250 cards were thrown monthly in each of the four fixed points.

During the course of the campaign, some of the fixed points for throwing were modified, for example, in Orio at the beginning we selected the mouth of the river Oria for throwing, but an elevated number of cards returned to the same beach in Orio. Aftewards we threw from «Anarri. with identical results. This made us think that there was a local whirlpool in front of Orio wich retained the cards in a restricted area near the coa.st, for which reason from July onwards we eliminated this throwing point.

In a recent study (CONTRERAS, 1966, 1977) it is possible to appreciate the complexity of interpreting the surface currents near the coast. For example, cards thrown on the same day appeared at the same time to the West of the coast of Santander, Asturias, or to the North East on the French coast. These irregular results give an idea of the complexity of interpreting them with any great certainty.

In the present work we are covering a period of one year and we have taken into consideration maps of the translation of the cards and simultaneously maps of the prevailing winds at this period of time.

Judging from the maps of the card movements, we can see that in January (Fig. 2) they were always going to the N-NE which coincides with a dominance in more than 50 % of the days with winds proceeding from NW (winds from NE and SE were present for a few days) .

In February (Fig. 3) the cards were translated very far North reaching as far as Sables d'Olonne in France, and it happened that the cards which went furthest were precisely thGse which were thrown furthest from the extreme S. Eastern point of the Bay of Biscay (Fuenterrabia). Perhaps the current generated parallel to the coast has less intensity in this Southern extremity.

During the month of February, in the first fortnight N. Westerly winds prevailed (rn more than 50 % of the days) while in the latter fortnight winds from the SW prevailed 100% of the time, which also correlates directly with the results already obtained.

In March (Fig. 4) the translation of the cards was also E-NE but was a moderate translation. During this month winds from NW prevailed again, storms being very frequent at this time.

During the month of April (Fig. 5) , the translation of the cards continued in the same direction reaching greater distances than in the previous month. The winds were similar to those in March but with fewer stormy days.

In Maya stability seemed to exist in the waters of the extreme SE of the Bay of Biscay with modera~e migration of the cards to the E-NE (Fig. 6) .In this month the prevailing winds were from NW, but the percentage of days with winds proceeding from NE increased. (The recoveries made to the west of the throwing point, Orio, were made at the end of 1-3 months, for which reason these recoveries have no significant value for this month).

During June (Fig. 7) winds from NW continued to prevail but winds from SE appeared with some frequency. The cards continued drifting E-NE.

In July the direction of winds was similar to the previous month and the cards recovered in the first half of the month are from the East of the throwing points.

The situation changed drastically in the days following when the cards were recovered West of the throwing points. This is correlated with the variation in the norm of prevailing winds which began at the end of this month (Fig. 8).

In the month of August the direction of the prevailing winds changed. SE winds prevailed followed by winds from NE and NW.

During this month the majority of the cards were translated Westwards in relation to the prevailing winds (Fig. 9) .

But sporadically when winds from the NW prevailed, the cards migrated in the opposite direction going to the E-NE.

This explains the irregularities found by CONTRERAS (1966) due to the fact that on this occasion the throwing of the cards.was made in

These changes in the direction of the current were more evident in the month of September. In effect, the cards thrown from Mundaca (Viz. caya) in the first days of the month appeared on the French coast. In these first days NW winds prevailed. Whereas on the contrary, cards thrown later fr-om San Sebastián always appeared to the West of the point of throwing. (Fig. 10) .

A case which merits special consideration is that of cards thrown on the 20th of September in San Sebastián. After 9-10 days. 14 cards appeared on Baquio beach (Vizcaya) .During these days SE winds prevailed. but what calls the attention is a fact that we have observed on numerous occasions. namely that the cards travel together and in cases like the present are found far away on one and the same beach. This explains the accumulation of floating material which may be of natural origin (wood, branches) or anthropic (plastics and oil) which occasionaIly appear on our beaches.

A few days later the cards thrown on the 25th from Santander went in the opposite direction towards Vizcaya, Winds from the NW prevailed at this time.

In October the direction of the winds varied a lot. During the first few days of the month. winds from the SW prevailed and produced a translation of the cards for the most part towards the West. although in general the distances covered were short (Fig. 11). (The recoveries made very far North were made in December. for which reason these recoveries have no significant value for this month) .

In November and December, the dominant winds were from S.W. (this was a constant pattern) producing a circulation towards the E-N.E. typical of the winter months.

The cards appeared again to the East of the throwing points and above all on the French coast.

The translation of the cards during November was similar to the month of February. reaching as far as St. Gilles and St. Hilaire de Riez.

The oceanic circulation described in the present work explains the appearance of seaweeds like Fucus vesiculosus, Ascophyllum nodosum, Halidrys siliquosa and Himanthalia e/ontaga on our beaches on various occasions. Although these types do not live on our coasts for the reasons given in a previous work (IBAÑEZ. 1978) , they are transported from the NW of the Peninsula.

Spores of these seaweeds arrive in the same way but do not develop except for very exceptionally in some places and in some years.

The study of the type of currents in the Bay of Biscay carried out over a period of various years could result interesting wi~h the aim of 

It is important in these cases to know the drift of the surface water in order to know the direction which the «black tide» will take (PASQUAY & BONNOT, 1971) .

In our coast catastrophes of this type have not yet occurred, but the risk could increase notably with the functioning of the enormous new harbour of Bi Ibao where the great tankers wi II enter .

Another matter of minor ecological repercussion but of great aesthetic importance is the problem of non-recyclable objects, above all plastic containers of many domestic products which are normally dumped in the sea, directly or via the rivers.

These plastics normally come to rest on the French coast producing the logical denunciations and revulsion of the local inhabitants.

In the month of February great quantities of these plastic containers accumulated on Oleron Island and nearly all were of Spanish origen (this was easily seen from the label) .During other months the accumulation is produced on the extensive beaches of the Landes.

It is essential that the town councils take steps by means of publi. city campaigns directed at all levels of society with the objective of cutting out some of this waste and increasing the degree of responsability and civic pride of the people.

As we indicated in a previous work (IBAÑEZ, 1978) numerous authors have shown the specjal thermal characteristics which occur in the summer months in the extreme SE of the Bay of Blscay and whjch appear to be responsable for the changes in the benthic coastal populations, producing some coastal ecosystems which have more affinity with the Mediterranean than with the Atlantic coast of Brittany or of Galicia.

This thermal anomaly was explained by some oceanographers (LE DANOIS, 1934. 1938; LETACONNOUX, 1953, 1954, 1955) as the result of a warm counter current which emanates from the waters of Cadiz. arrjving towards May at Finjsterre travelljng onwards to the Cantabrjan sea where it remajns «trapped» during the summer.

This theory noted by MOLINA (1972) does not appear now to be the most probable.

Interest in studying the surface temperatures normally goes along with studies in the behaviour of the Albacore (Germo alalunga) a specjes very sensitive to thermal variations, and which normally js concentrated in isotherms 18.5-19.5° C (although, according to the fishing area thjs concentration can be found in 16.7-17.6° C to the West of the Iberian peninsula or between 19.7-20.2° C off Cape Peñas according to ALLAIN & ALONCLE, 1968) 2.

The upwelling which begins jn the month of May off the Galician coast has been studied and the mechanism responsable for this is relatively well known (MOLINA. 1972) .

Its importance in the behaviour of sardjnes was demonstrated by MARGALEF & ANDREU (1958) .

However. the mechanism' for heating up the waters in the extreme SE of the Bay of Biscay does not appear to be so clear .

SERVAIN (1977) studjed among various parameters the average monthly temperatures of the surface waters of the sea over a period of 120 years (1854-1973) (Fig. 16).

We can see how in April a nucleus of warmer water appears in the

This situation modifies itself again in August and September. so that in October the situation observed in April and May returns again. In December this thermal inversion disappears completely in the extre. me SE of the Bay of Biscay.

It seems probable that the cause of this warming-up of the waters

has to be interpreted as a relative cessation of these waters and perhaps the correct way to interpret this is not as an abnormal warming-up in the extreme SE of the Bay of Biscay, but as an abnormal cooling-down due to the upwelling in the extreme NW in the Iberian Peninsula.

In effect. it is a significant fact that during April, May and June the currents were in the direction E.NE. This mass of warmer water continues in the same direction, while in July currents appear going towards the West. The mass of warm water moves in this direction. In the summer months we have seen how the surface currents off the Basque coast fluctuate from E to W and from W to E, but the distances covered by the cards are inferior to those which are manifest in some of the winter months.

This could be interpreted as a fluctuation of the mass of water from one side to the other, but without moving beyond the extreme SE of the Bay of Biscay.

It seems evident, on the face of the previous results, that the littoral currents which affect the extreme SE of the Bay of Biscay are produced by the prevailing winds.

These general currents caused by the wind mask the tidal currents which normally gyrate towards the right on the French Basque coast (SERV. HYDROGR. OCEANOGR. MARINE N.o 550, 1968) .

From January to the middle of July (during 1978) the ocurrents were parallel with the coast in the direction SW ~ NE. The speed and intensity of these currents reached their height in the first days of December. February and November .

From the middle of July to the end of October the situation was very variable and currents from W to E alternated continually with currents from E to W producing an oscillatory movement of the mass of surface water .

During 1978in July. August and October the prevailing current was E ~ W. while in September the current E ~ W and W ~ E had the same intensity.

These summer variations were also observecj, in the period JulyAugust in 1977 at a depth of 10m.. two miles off Txjmistarri (near San Sebastián) using a current reading meter (TEJEDOR L., J. C. SANTAS & F. RUIZ. 1978) .It was further observed in this work that the intensity of the current at a depth of 40m was half of that observed at a depth of 10m.

So. we can suppose that if the water of the surface layer at a depth of Om "oscillates» for a month with a maximum amplitude of 120 miles (in a direction parallel with .the coast) .at a depth of 60m the mass of water would oscillate, during this time, with a maximum amplitude of 15 miles, and would remain nearly stationary below a depth of 75m.

Lastly, in November a uniform circulation began again towards the NE. We can sum up by saying that during the year 1978, two circulatory

movements of the surface coastal waters were produced. One typical of '-. the winter months in which the water moved parallel with the Cantabrian coast and later ascended the French coast. The second was an alternating movement during the summer months in which the current osciIlated with the prevailing winds from E to W and vice versa. (In some of the summer months currents prevailed from E to W).

This situation of dynamic equilibrium of the surface water of the extreme SE of the Bay of Biscay. permits this warming-up during the summer months.

So, we can see that a direct relationship could exist between the type of surface current and the temperature. Vertical dlstribution of temperature

In accordance with the work carried out by the Ministry of Public Works, on the vertical distribution of temperature, the thermocline off the Guipuzcoan coast is present in July and August and disappears at the end of October .

This disappearance occurs first of all in the spots nearest the coast. possibly due to more turbulence and the effect of the tides in these points of lesser depth (Fig. 17).

In a wider perspective and as result of the oceanographic expeditions PHYGAS 41-44 (FRUCHAUD-LAPARRA. LE FLOCH, LE ROY. LE TAREAU & MADELAIN. 1976) we can see in the Bay of Biscay a great homogeneity in the vertical structures to a depth of 600m, the "ceiling" of the layer of water originating from the Mediterranean.

During the winter months. the T /S values are:

Atlantic water layer T=11.6°C S = 35.6 por mil

 Mediterranean water layer T = 11.6° C S = 36.4 por mil 

Sea bed layer T = 4° C S = 35.0 por mil

In figure 18 we can see pockets of Mediterranean water near the Spanish coast clearly defined between depths of 900-1,1 OOm.

These pockets are less evident in the interior of the Bay of Biscay (Fig. 19) .

According to this information we can observe that a wide coastal area (to more than 100m in depth) possesses salinity values equal to or less than 35.5%0 which clearly show an influence of continental waters in the coastal shore water.

Anexo no disponible  en versión internet : (registers corresponding to the cards recovered)

ALLAIN Ch. & H. ALLONCLE: Migrations du Germon en fonction des variations thermiques du milieu entre le secteur portugais et le S.O. de l'Irlande (Campagnes de «La Pelagia» et de la «Thalassa» du 18 mai au 30 septembre 1968). Science et peche Bull. Inst. peches Marit. 176 (1968) 1-15.

BAONZA, E. & A. PLATA: Estudio del movimiento de los efluentes vertidos al mar por los emisarios de la Costa del Sol mediante el marcado con is6topos radiactivos. Agua 88 (1975) 30-35.

CARRUTHERS, J. N.: Further investigations upon the water movements in the English Ch(Innel. Drift-bottle experiments in the summers 1927, 1928, 1929, with critical notes on drift-bottle experiments in general. J. Mar. Biol. Ass. U. K. 17 (1930-31), 241-275.

CONTRERAS, L.: Estudios sanitarios sobre las poluciones del agua, aire y suelo de Guipúzcoa. Observaciones recogidas durante los años 1964, 1965 y 1966. Munibe 28 (1967) n.o 1/46, 189-206.

CONTRERAS, L.: Estudios sanitarios en la costa guipuzcoana. Publ. Caja Ahorros Municipal. San Sebastián (1977), 177 pp.

DUNCAN, C. P.: Disadvantages of the 01son drift card and description of a newly designed card. J. Mar. Res. 33 (1965, n.O 3) 233-236.

FISCHER~PIETE, E.: La distribution des principaux organismes intercotidaux Nord-Iberiques en 1954-1955. An. Inst. Oceanogr. 40 (1963, n.o 3) 165-312.

FLOS J ., G. IGUINIZ, X. IRIBAR, X. TOLDRA y M. ZABALA: Informe preliminar del estudio oceanográfico de la costa guipuzcoana entre Guetaria y Fuenterrabía. Informe manuscrito (1976).

FRUCHAUD-LAPARRA B., J. LE FLOCH, C, LE ROY, J. Y. LETAREAU & F. MADELAIN: Etude hydrologique et variations saisonnieres dans le proche Atlantique en 1974. Rapp. Scient. Tech. CNEXO 30 (1976) 108 pp.

GARST ANG W .: Report on the surface drijt of the English Channel and neighbouring seas during 1897. J. Mar. Biol. Ass. U.K. 5 (1898).

GUERIN O. M. & J. P. PINOT: La diffusion des eaux de la Loire dans le Golfe de Gascogne et les risques de pollution littorale d'apres les satellites artificiels. Rev. Int. Ocean. Med. 47 (1977) 69-72.

IBAÑEZ M.: Características biogeográficas de la Costa Vasca. Lurralde 1 (1978) 121-

127.

KNAUSS J. A.: Drogues and neutral bouyant floats en The Sea V.2 (1965) 303-305.

Interscience Pub1. John Wiley N. Y.

KURK G.: Observations sur la dérive des flotteurs lancés par le «Président Théodore Tissier» pendant les campagnes de 1951 a 1954. Rev. Trav. Inst. peches Marit. 20 (1956, n.o 3) 225-261.

LE DANOIS, E.: Les transgressions océaniques. Revue des Travaux 7 (1934, n° 4), 

LE DANOIS, E.: L'Atlantique. Histoire et vie d'un Ocean. Ed. Albin Michel. Paris

(1938).

LETACONNOUX, R.: Les conditions du milieu sur les pecheries de germon en Juin 1952. Cons. Int. Expl. Mer. Ann. Biol. 9 (1953).

LET ACONNOUX, R.: Conditions hydrologiques en juin sur les pecheries de Germon. Cons. Int. Expl. Mer. Ann. Biol. 10 (1954).

LETACONNOUX, R.: peche du Germon: Campagne effectuée par le «Président Théodore Tissier» en mai-juin 1955. Science et peche. 27 (1955) 3-4.

MARGALEF, R. & B. ANDREU: Componente vertical de los movimientos del agua en la ría de Vigo y su posible reJación con la entrada de la sardina. Invest Pesq. 11 (1958).

MOLLER, L.: Oberkommando der Kriegsmarine Temperatur, Salzgehalt, Dichte und Strümungen an der Oberfliiche und in den Tiefen der Biskayabucht (Atlas). Institut für Meereskunde (1941) Berlin.

PASQUAY, J. N. & J. BONNOT: Utilisation de cartes-flotteurs pour l'étude des dérives de surfaces et application a la prévision des pollutions cotieres. La Houille Blanche. 8 (1971) 769-778.

SERVAIN, J.: Rapport preliminaire sur l'étude des donnees climatologiques historiques du proche Atlantique. Rapport n.O 5. Univ. Bretagne Occidentale. (1977).

SERVICE HYDROGRAPHIQUE: Courants de marée dans la Manche et sur les cates francaises de l'Atlantique. Serv. Hydrogr. Oceanogr. Marine 550 (1968) 281-283.

SHANNON L. V., G. H. ST ANDER & J. A. CAMBELL: Oceanic circulation deduced from plastic drift cards. Invest. Rep. Sea Fish. brch. S. Afr. 108 (1973) 1-31.

SOULE, RF. & «M. OGURI: Circulation patterns in Los Angeles -Long Beach Harbor. Drogue study atlas and data report. Allan Hancock Found. USC.;SG-6-72 (1972) 113 pp.

TEJEDOR L., J. C. SANTOS & F. RUIZ: Estudios oceanográficos. Rev. Obras Públicas. 3157 (1978) 393-401.

TR~GLOS, Y.: Analyse automatique des températures de surface de la mer dans le Golfe de Gascogne et le proche Océan. Météor. Natl. Notes de l'établissement d'études et de recherches météorologiques 302 (1972).

(1) In specific studies on local currents in harbours and estuaries, coloured substances such as Fluoresceine or Potassium Dichromate are used, although they have the inconvenience that, after a féw minutes, the dilution is such that is impossible to locate it visually. Another system is using special floats. drogues (KNAUS. 1965; SOULE & OGURI. 1972) or with radioactive isotopes such as Br82 which semi-disintegrates after 36 hours (BAONZA & PLATA, 197q).

At the present moment the French National Meteorological Office is doing an automatic analysis programme of temperatures with the help of computers (TREGLOS. 1973).

It is now possible to use infrared photography from satellites NOAA and LANDSAT to obtain all the surface temperatures simultaneously of extensive oceanic areas (GUERIN & PINOT, 1977).

(2) The first studies of the temperature of the surface water in the Bay of Biscay were carried out for military reasons by the German Navy in 1941 (L. MOLLER), this data was considered &ecret till the end of World War II, and was used by the French (FISCHER-PIETTE, 1963) to explain the biogeographical distribution of the most common shore species in the Basque coast.