Thalassas greek voice meaning...”of the sea” Cover photograph: taken by Miguel Ángel Nombela Castaño

THALASSAS is included in the following DATABASES:

THE BOWKER INTERNATIONAL SERIALS DATABASE (Urich’s International Peridicals Directory). USA.

ÍNDICE ESPAÑOL DE CIENCIA Y TECNOLOGÍA (I.C.Y.T.) ESPAÑA

FAO: FISHERY INFORMATION. DATA AND STATISTICS SERVICE ITALIA.

MS. MEDIA SERVICE GMBH. ALEMANIA.

This issue is published thanks to the financial help of Este volume publícase cunha axuda da

WEB PAGE: http://webs.uvigo.es/thalassas/

© Universidade de Vigo, 2005

Edita: Servizo de Publicacións Universidade de Vigo. Campus das Lagoas, Marcosende 36310 Vigo. España.

Imprime: Gamesal • Torrecedeira, 92 - 36.208 Vigo. España

I.S.S.N.: 0212-5919 Dep. Leg.: C379-83 Nº 21 (1) - 2005 Volume 21(1), July, 2005 THALASSAS AN INTERNATIONAL JOURNAL OF MARINE SCIENCES

Scientific Committee

ALFREDO ARCHE MIRALLES TOMOHIRO KAWAGUCHI Instituto de Geología Económica. Department of Environmental Health Sciences C.S.I.C., Madrid (Spain) The Norman J. Arnold School of Public Health University of South Carolina (USA) ANTONIO CENDRERO UCEDA D.C.I.T.T.Y.M. Facultad de Ciencias ROLAND PASKOFF Universidad de Cantabria, Santander (Spain) University Lumiére, Lyon (France)

DARÍO DÍAZ COSÍN NORBERT P. PSUTY Departamento de Zoología Center for Coastal and Environmental Studies Facultad de Biología, University of New Jersey (U.S.A.) Universidad Complutense de Madrid (Spain) RICARDO RIGUERA VEGA GRAHAM EVANS Departamento de Química Orgánica Department of Geology. Imperial College Universidade de Santiago de Compostela (Spain) The London University (United Kingdom) RAFAEL ROBLES PARIENTE FERNANDO FRAGA RODRÍGUEZ Instituto Español de Oceanografía, Madrid (Spain) Instituto de Investigacións Mariñas C.S.I.C., Vigo (Spain) AGUSTÍN UDÍAS VALLINA Departamento de Geofísica JOSÉ MARÍA GALLARDO ABUÍN Facultad de Física, Instituto de Investigacións Mariñas. Universidad Complutense de Madrid (Spain) C.S.I.C., Vigo (Spain) CARMINA VIRGILI RODÓN FEDERICO ISLA Departamento de Estratigrafía Centro de Geología de Costas Facultad de Geología, Universidad de Mar del Plata (Argentina) Universidad Complutense de Madrid (Spain)

JESÚS IZCO SEVILLANO TAKESHI YASUMOTO Departamento de Bioloxía Vexetal Department of Chemistry Facultade de Farmacia, Agricultural Faculty, Universidade de Santiago de Compostela (Spain) University of Tohoku (Japan)

Editorial Committee

Honour President: Federico Vilas • [email protected]

Editor-in-Chief: Manuel J. Reigosa Roger • [email protected]

Editor Electronic Version: Daniel Rey • [email protected]. Editor, Marine Geology: Guillermo Francés • [email protected]. Editor, Marine Biology: Francisco Ramil • [email protected]. Editor, Marine Physics: Carlos Souto • [email protected]. Editor, Multidisciplinary Aspects of Marine Sciences:Cástor Guisande • [email protected]. Secretary and Treasurer: Miguel Á. Nombela • [email protected]. Editor, General: Luis González • [email protected]. Redaction Council: Luisa Andrade • [email protected]. Redaction Council: Gabriel Rosón • [email protected]. INSTRUCTIONS FOR THE AUTHORS For the paper and electronic version, electronic submission is encouraged. Please clarify (in the e-mail text) which software was used or is needed to open the Manuscript. Please send as attachments cover letter in Word for Windows or any other usual wordprocessor, manuscript, graphics and data. For software submission, please contact previously with the electronic version Editor, Daniel Rey ([email protected]).

Printed version

Regular papers should be addressed to Editor-in-Chief, Manuel J. Reigosa, preferably by electronic means ([email protected]) or alternatively by post to (original plus three copies plus electronic version in diskette or zip): Thalassas, Manuel Reigosa, Editor-in-Chief, Facultade de Ciencias de Vigo, Campus de As Lagoas / Marcosende, 36310 Vigo (SPAIN),

Please indicate clearly the programs needed to read the article. Microsoft Word for Windows will be preferred and will help the speed of publication, but other wordprocessors can be also accepted. The e-mail message (and the diskette, if postal submission is done) should contain the following information: 1.Cover letter 2.Title page 3.Abstract 4.Article 5.Information about the wordprocessor and programs needed to see the paper.

You can also contact directly with the Editors, this is especially encouraged before submitting review papers or letters.

Articles

Thalassas is an international journal that accepts original papers, review papers and short notes about every aspect of marine sciences, especially when a multidisciplinary approach is followed. Languages accepted are English, Spanish, French and Portuguese, although all the publications will be done in English. For Manuscripts in a language other than English, the Journal could help to correct the paper once it is translated into English by the authors.

Authors are allowed to post their accepted papers in their own Web pages. Thalassas will, in any case, provide free to all the scientific community, a version of the published papers to download from the Thalassas Web page.

Revision of papers will be done using electronic facilities (that is, referees would receive by e-mail the papers under revision and should answer no later than two weeks after receiving the article by e-mail or fax).

Authors can apply for a picture or graphics to be used as a full colour cover image for the paper version. Please state it when you submit your paper.

Full-length papers:

Are original previously unpublished works about any aspect of marine sciences. The title should be indicative of contents, and no longer than 60 characters. The first page should include the names of authors and complete affiliations, including e-mail addresses and Web page addresses if any. They will include an abstract of between 100 and 300 words, followed by less than 15 keywords (both included in the abstract and additional). Materials and Methods sections will be followed by Results and Discussion. Those sections can be put together if this fits the content of the Manuscript. Manuscripts should be written in simple sentences, conforming to accepted Scientific Standard English. Texts should be clear considering the great scope of the audience (this is not a very specialised journal, covering a broad range of disciplines, although always related to marine or coastal ecosystems). After those sections, Acknowledgements and References should follow. The style of citation will be as follows:

Journal articles: Author AA, Author BB (year). Title of article, Complete name of Journal, number: pages

Book articles: Author AA, Author BB (year). Title of article. In: A Editor, B Editor, eds, Title of Book, Ed, number, Vol number, Publisher, City, pages. Theses Author AA (year) Title of thesis. University, City. The citations should be arranged in the text from earliest to most recent year, alphabetised by name within the same year. In the references list, order by author (s) name, after by year. Finally, figures, tables and captions for figures and tables should be included. Review papers: Those papers will be published mainly by invitation. But suggestions are also welcome. If you feel that you can contribute with a review, please contact Editor-in-Chief or any of the Editors by e-mail. Technical papers: Are especially welcome for the electronic version, but if the interest is appreciated by the editors, they can also be published in the paper version. The structure of the article should follow the same recommendations as full-length papers. Letters: Correspondence intended to be published in the paper version should not exceed two printed pages. For the electronic version (that will be the preferred for letters because the speed of publication and the possibility of several responses) no page limit is applied, although the shorter the best. Meeting reports and Conference Proceedings: For meeting reports of Conferences about Marine Sciences, the coordinator of the Congress who wish an abstract to appear in Thalassas (either in electronic or paper versions) should contact Editor-in-Chief ([email protected]). Congress Proceedings could also be published as special numbers of the journal. Meteorological and Oceanographic Buoy of the Physical Oceanography Group of the University de Vigo GOFUVI (http://eddy.uvigo.es/Principal.html) Photograph taken by Miguel Ángel Nombela Castaño.

INDEX

9-16 J. J. Alonso, P. Villares, M. J. González, M. Arias & B. Marín Ocean Tides and Fractal Geometry: Tidal Station Stability

17-28 J. Moreno, F. Fatela, C. Andrade, J. Cascalho, F. Moreno & T. Drago Living Foraminiferal Assemblages from the Minho and Coura Estuaries (Northern Portugal): A Stressful Enviroment

29-33 H. Matthews-Cascon, H. A. Pereira Alencar, S. Guimarães Rabay & R.M.S. Mota Sexual Dimorphism in the Radula of Pisania pusio (Linnaeus, 1758) (Mollusca, Gastropoda, Buccinidae)

35-44 J. Ferreira & M. Cachão Calcareous Nannoplankton from the Guadiana Estuary and Algarve Continental Shelf (Southern Portugal): An Ecological Model

45-52 I. Abrantes, F. Rocha & J. A. Dias Spatial Distribution and Composition of Suspended Sediments in Ria de Aveiro Lagoon

53-58 F. Rocha, T. Drago, F. Naughton & T. Silveira Palaeoenvironmental Analysis of Douro Estuary based on Mineralogical Parameters

59-65 C. Litulo Fecundity and Size at Sexual Maturity of the Fiddler Crab Uca vocans (Linnaeus, 1758) (Brachyura: Ocypodidae)

67-76 A. Machado, F. Rocha, C. Gomes, J.A. Dias, M.F. Araújo & A. Gouveia Mineralogical and Geochemical Characterisation of Surficial Sediments from the Southwestern Iberian Continental Shelf

77-84 R. Coelho, L. Bentes, C. Correia, J.M.S. Gonçalves, P.G. Lino, P. Monteiro, J. Ribeiro & K. Erzini Age, Growth and Reproduction of the Axillary Seabream, Pagellus acarne (Risso, 1827), from the South Coast of Portugal

Thalassas, 2005, 21 (1): 9-16 An International Journal of Marine Sciences

OCEAN TIDES AND FRACTAL GEOMETRY: TIDAL STATION STABILITY

J. J. ALONSO(1)(*), P. VILLARES(1), M. J. GONZÁLEZ(2), M. ARIAS(1) & B. MARÍN(1)

Keywords: Mean Sea Level analysis, fractal, analysis of tides

ABSTRACT INTRODUCTION

Fractal Geometry is one of the most powerful tools A fractal is a non usual geometrical object with a in the research of complex dynamical systems of high, dimension greater then its topological one (Guzmán et fractional or unknown dimensions. It allows the al. 1993). They have been also named mathematical computation of the dimensions of attractors of the monsters (Pla 1994) because they present very different phase space of the dynamical system and their geometrical properties than usual geometrical sets. The stability. The authors apply some fractal concepts on vector and functional spaces of non integer or tidal elevation data sets considering them as the transitional dimensions are included in fractals. In 1919 projection of the temporal evolution of the dynamic Hausdorff built up the main tools to compute measures system of the tide in the vertical axis. At the same time on that sets in the known as Hausdorff measures and the Mean Sea Level is considered the attractor of the dimensions. In the twenties, Besicovitch developed dynamic in such way that the system evolves around it. their geometrical properties (tangency, projections and The three considered tidal stations are Ceuta, Malaga densities) and he called them strange sets. His original and Algeciras. In addition a theoretical standard from and revolutionary ideas are the actual basis of the the tidal potential RATGP95 has been used for Geometrical Theory of the Measure. In the seventies comparison. B.B. Mandrelbrot made an important number of works naming that sets as fractals, making popular the theory and pointing out its possibilities. A nice resume of fractals and its application on Geophysics can be found in Mandrelbrot (1989). Fractals were applied on many fields of the Earth Science (i.e.): Geology (Turcotte (1)Applied Physics Dept. (Physical Oceanography). 1989; Todoeschuck & Jensen 1989; García & Otárola University of Cadiz. 1994), Seismology (Crossley & Jensen 1989), Physical (2) Applied Mathematics Dept. University of Cadiz. Oceanography (Provenzale et al. 1989). Nowadays the Avda República Saharaui s/n, Puerto Real, 11510, Cadiz, Spain. fractals are the central point of many mathematical (*)E-mail: [email protected] researches with wide applications.

9 J. J. Alonso, P. Villares, M. J. González, M. Arias & B. Marín

There are roughly three types of fractal sets: means of the simple box-counting method (García & a) Common mathematical entities with the same Otárola 1994), it is possible the computation of some topological and fractal dimensions, b) Regular self- fractal measures for one dimensional time series. In similar entities with a fractal dimension greater than any case, and because in the case of the tides, there are their topological one and c) Irregular self-similar several time scales embedded in the data, it is more entities with a fractal dimension much greater than proper to use the term of multifractal instead fractal. their topological one. The usual geometrical entities lie Perhaps this is the most remarkable geometrical on the first group: zero dimensional point, one property of the nature. dimensional line, two dimensional plane surface, and so. The most interesting ones are the two last groups There is only one reference about the application where the attractors and strange attractors lie in. of multifractal on tidal records. Alonso (1998) analyzed two high precision earth tides series with a However in Applied Physics it is no usual dealing noise level of pico-gals from two superconducting with perfect mathematical entities. The mathematical gravity meters placed at Brussels (Royal Observatoire models of the Fractal Geometry applied on real world de Belgique) and in the Black Forest Observatory physical processes have four characteristics (Guzmán (Germany). A new parameter named stability of the et al. 1993; Pacheco 1994): a) They are difficult to tidal station was defined giving an additional compute because they need massive computation, b) It parameter for classifying the earth tides gravity is a method for classification and not for prediction, stations. One of his main conclusions was that very c) The role of the Geometry depends on the range of good quality time series, easily to predict by means of values of the physical observable and its temporal harmonic analysis, can be fractally very different. evolution and d) They use to present several Being possible to scope very deep into the geometrical temporal/spatial scales imbedded in the same record. structure of the time series and determine which one Undoubtly the proper reading of the result will depend has the best quality. Because fractals are extremely on the physics of the phenomenon, on its formulation powerful tools for classification, it is possible to in terms of fractals and on the physical intuition of the conclude from the numerical results that the station researcher. under study can present problems, even if the results of other classical procedures (i.e. the harmonic analysis) Although the basic applications of fractals were militate against the fractal conclusions. This was fairly developed to treat two dimensional objects (images) by demonstrated in Alonso (1998) for earth tides and now the study is extended on ocean tides.

The work has been organized as follows. The simple mathematical apparatus of fractal dimensions is presented in Section 2. The proper modification of some fractal concepts needed to apply the theory on tides is presented in the same section. Section 3 is devoted to describe the numerical experiments and to discuss the results. Finally the conclusions are drawn in section 4.

FRACTAL CONCEPTS AND THEIR APPLICATION ON TIDAL RECORDS

There are two main concepts in the large body of the Fractal Theory that are necessary to discuss and explain in order to apply on tides. Firstly it is the Figure 1. Areas of the plot of the generalized correlation integral for the computation concept of dimension. This can be defined in many of the fractal dimension. ways but only few definitions could be valid in the

10 Ocean Tides and Fractal Geometry: Tidal Station Stability

Figure 2. Several dimensions to take into account in the analysis of the dimension of the attractor (Mean Sea Level) in the phase space. frame that will be stated. The second one is the concept The estimation of the Dq dimension of an attractor of self-similarity or scale invariant. This is more in the phase space generated by a time series is difficult to establish but it is the basis for the computed easily by means of temporal differences application of the Geometric Theory of the Measure together with the Takens' theorem (Takens 1981). The applied on the Mean Sea Level (MSL) oscillations. general expression was due to (Grassberger & Procaccia 1984): 1. Dimensions (1) From natural the dimension of a vector or functional space is the minimum number of Being ε the size step and Cq(ε) the generalized coordinates needed to locate any point in that space. integral of correlation defined as: For the real physical world there are zero dimensional points, one dimensional lines, two dimensional plane (2) surfaces, three dimensional objects and the four dimensional Minkowski space-time. In addition it is possible to choose any proper system of coordinates to Where q is the type of dimension, M is the so study a given problem. Changing the system of called embedding dimension, yk is the k-thm orbit used coordinates can be understood as a distortion of the to state of the Takens' matrix and θ(·) is the step or body under study or a distortion of the same space Heaviside function. A nice and simple introduction to keeping the dimensions of the problem. the problem and to its proper reading can be found in Broomhead & King (1985). It is possible to compute The computation of the dimension for simple and the dimensions for any arbitrary integer value of q, but common geometrical objects has no problem (Guzmán the reading with q>2 is not clear and nor stated. et al. 1983). However there are many geometrical objects with a priori unknown dimension. As an In the fractal problems there are three types of example it is possible to point out the attractors in the dimensions. With q=0 the Hausdorff dimension is phase space of a given dynamical system. The obtained, D0, and it is the same that obtained by the dynamical systems can have several attractors in the box-counting method (Guzmán et al. 1983). The phase space, each one with their own dimension. problem with the application of D0 lies in that it does Hence it is necessary to adopt other methods to not take into account the density of points, giving the estimate the dimension of geometrical objects. same weight to boxes with one point or full of them.

11 J. J. Alonso, P. Villares, M. J. González, M. Arias & B. Marín

With q=1 the information dimension is computed, D1, that introduces the correction for the density of points. However it does not take into account the correlation between orbits. In order to deal with time series the best choice is q=2 and obtain the correlation dimension, D2. This takes into account the correction of the density of points and the correlation between orbits, giving an unbiased value of dimension except by the same length of the data set. Because we are dealing with long time series it is possible to consider that the resulting estimations will be unbiased. However the authors will present the results for the three described dimensions and their consequences.

Independently of the value of q, some considerations must be done. A typical example of the plot of Cq(ε) vs. ε, in linear scale, is presented in Figure 1. It is clear than as a function of the length of the data set, the range of ε and the physics of the process; the linear regression can be carried out with higher statistical confidence. The value of the q-dimension will depend of where the regression is performed in Figure 1. The segment number 3 corresponds to very high values of e and the segment number 1 to very small values. Only the segment 2 will provide valuable information about the geometrical properties of the set and it corresponds to optimum values of the size step. Region 2 is usually called scaled region. In practice the regions 1 and 3 use to be very small, however the starting and ending points of the regression depend on the experience of the researcher.

2. Self-similarity of tidal records

The tide is the observed result of a Figure 3. multidimensional dynamical system. The number of Some plots about the restoration of the phase space parameters needed to specify the location of the free of the dynamical system of the tide. surface in the time t is 3N+1, being N the number of resolved waves taking into account the amplitudes, resolved. Such number will depend on the length of the phase lags and frequencies. So the working space is a record and on the Rayleigh's criterion of the harmonic 3N+1 parametric one. All of them are resumed in the analysis. The Eq. 3 describes the projection of the Fourier decomposition of the vertical tide: phase space in the vertical of a forced oscillator around an equilibrium point, the mean sea level. That point, if (3) it is stable (no motion) must have a null geometrical dimension and it would be a common geometrical Where η(t) is the location of the free surface object. Because the tide is composed by many respect to the mean sea level z0, Ai is the amplitude, ωi embedded time scales, the tide is a multifractal. The is the frequency and φi is the phase lag. The infinity scheme of the dimensions involved in the problem of must be understood as all the waves that can be the vertical tide is presented in Figure 2. It is possible

12 Ocean Tides and Fractal Geometry: Tidal Station Stability

Table 1. The Fractal Geometry gives extremely simple and Classification of attractors. powerful tools to evaluate the strength of the attractor that must be understood as the averaged numerical derivative of the equation of motion around the location of the attractor. Only in this way it is possible to estimate the rate of motion of the attractor or the mean sea level. Because of this, the dimension of the attractor can be thought as a parameter to characterize the stability of the tidal station. The recovering of the phase space from the experimental data of the Ceuta tidal station is presented in Figure 3 for several values to define a two dimensional local mean sea level of embedding dimensions. If m=24 is considered, the surface, the direction of the projection is one attractor is well defined in the centre of the scatter plot. dimensional, the position of the free surface is If the embedding dimension is higher, m=336 or 672, determined with a 3N+1 dimension and the attractor of the dynamic of the attractor is occulted in the plots but the system, the mean sea level, has an unknown the attractor is present, governing the dynamic in the dimension that must be estimated. phase space.

The attractor is characterized by the temporal The concept that acts a key-stone in the fractal derivative of the equation of motion in a small segment analysis of tidal records is the self-similarity or scale around itself (Guzmán et al. 1993). For the case of invariance. From the mathematical model of Eq 3, the tidal records the solution of the problem is easy tidal record must be considered self-similar. This is because it is usual to assume that Eq (3) reproduce valid for infinity Fourier time series. However if this is quite well the experimental observations. Denoting by truncated, as it is the case of tidal records, any change y' to the temporal derivative of η(t), the attractor can be in the scale will act as a filter. Because of this the classified as presented in Table 1 as super-attractive, minimum and maximum values of the size step, ε, attractive, indifferent and repulsive. This works quite must be defined. It can not start with ε→0. Only in this well if no instrumental or environmental noises are way it is possible to adapt the self-similarity to records acting or with analytical functions as the Tide of any geophysical variable affected by tides. Generating Potential (Alonso 1998). In addition the mathematical decomposition of Eq (3) is not perfect in The next problem is considering the multifractal any case when dealing with tides because the number property of the nature. Because the tide is a of resolved waves is much smaller that the existing multifractal it is necessary to compute the dimension ones. Hence the classification of Table 1 must be with several values of embedding dimension. In order considered with an unknown error in the computation to show this the typical tidal time scales have been of the derivatives (Alonso 1998). considered: 6, 12, 24, 336 y 672 hours, corresponding

Table 2. Fractal dimensions for the Algeciras tidal station.

13 J. J. Alonso, P. Villares, M. J. González, M. Arias & B. Marín

Table 3. Fractal dimensions for the Ceuta tidal station.

to fourth diurnal, semidiurnal, diurnal, fortnightly and RATGP95 tide generating potential (Roosbeek 1985). monthly tidal waves. In addition it must be pointed out Such a value is 5·10-6mgals/year, practically zero, as it that the input data for fractal analysis must be pikes corresponds to a quasi infinity Fourier time series. This free and non detided. The main reason for the first is conclusive value points out that if a time series is clear and for the second is that the driving force is perfectly harmonic, the fractal dimension of the eliminated and the attractor can not be observed attractor must tend to zero. If the obtained value is (Alonso 1998). higher there is some undetermined problem with the time series. It is worth worthy hoping that the fractal NUMERICAL EXPERIMENTS AND dimension of ocean tidal records be so small than for DISCUSSION earth tides (Alonso 1998) because the water level recorders do not suffer the permanent and exhaustive Three tidal stations have been considered: quality controls of the superconducting gravity meters. Algeciras, Ceuta and Malaga. The time series passed the usual quality controls in order to correct pikes and For the Algeciras tidal station a quite constant instrumental errors. The available time span is detailed value of fractal dimension is observed for m=6 close to in Tables 2 to 4. The analyzed tidal stations have the 4 mm/year. For the main periodicity, m=12, the trend typical mid-latitude semidiurnal periodicities. or fractal dimension is over 6 mm/year. The value of

Although only the D2 fractal dimension must be D2 is quite high for m=24, 1.29 mm/year, and D1 for computed (section 2), the three dimensions (q=0,1,2) m=336 and m=672 with more than 1cm/year. However will be considered in order to compare the results and they do not must be taken into account because the to show that visual inspected well behavioural tidal diurnal waves are very weak. The value of D2 is quite records can be very different and with low quality small for the rest of embedding dimensions. It must be being better do not use them. The dimensions were pointed out that all the computed values are quite estimated by the use of Eq (2) with a linear regression different, even in their magnitude. This is no normal at from Eq (1) selecting q=0,1,2 consecutively. Results all and implies that there must be some problems with are presented in Tables 2 to 4. Due to the multifractal this time series with the presence of some undesirable property of the tide five embedding dimensions were phenomena. selected from the physics of tides: 6, 12, 24, 336 and 672 hours. If several years long time series were The values of the fractal dimensions for Ceuta and available, the embedding dimension could take the Malaga tidal stations are perfect. The mean sea level values corresponding to six month or one year as doest not have practically motion and the fractal showed in Alonso (1998). dimensions are one order of magnitude lesser than the computed for Algeciras tidal station. The results are The theoretical reference value has been taken quite homogeneous with the same order of magnitude from Alonso (1998) as the fractal dimension of a also. The exception, pointed out above, is the value of synthetic time series of 21 years long from the D0 for the embedding dimension of 24 in the Malaga

14 Ocean Tides and Fractal Geometry: Tidal Station Stability

Table 4. Fractal dimensions for the Malaga tidal station.

tidal station. This can be neglected from the analysis station presents any unknown kind of problem (i.e. from without problem because the diurnal periodicity has maintenance related to the clock of the water level not importance in that place. recorder, calibration, etc) that must be deeply implying that the data records in such place must not be used for When comparing all the computed values for the the determination of the Mean Sea Level in the area. three tidal stations the results for Algeciras are the anomalous among them. They are one order of ACKNOWLEDGMENTS magnitude greater. This does not mean that the mean se level is rising at the north coast of the Strait of This work has been supported by REN2000-0549- Gibraltar while at the south are very stable. However C02-01 and CGL2004-01473/CLI CICYT projects. they imply the presence of instrumental effects that recommend do not use the time series for the analysis REFERENCES of the mean sea level. It is necessary to make a more Alonso, J.J., 1998, On the fractal dimension of Earth tides and exhaustive control of the Algeciras tidal station. characterization of gravity stations. Bulletin d'Informations de Marees Terrestres, 129, pp 9963-9973. These results point out that the tidal records at Broomhead, D.S:, and King, G.P., 1985, Extracting Qualitative Ceuta and Malaga can be used to study the mean sea Dynamics from Experimental Data, Physica 20D, 217-236. Crossley, D.J. and Jensen, O.G., Fractal Velocity Models in Refraction level while the tidal record at Algeciras must be put in Seismology, in Fractals in Geophysics, Eds C.H. Scholz & B.B. doubt seriously. Mandelbrot, pp 61-76. García, J. & Otárola, F., 1994, El uso de la geometría fractal en las Ciencias de la Naturaleza. Revista de la S.A.E.M., nº 28, vol 10 CONCLUSIONS (1). Grassberger, P. and Procaccia, I., 1984, Dimensions and Entropies of The authors have computed the fractal dimensions Strange Attractors from a Fluctuating Dynamics Approach, of three tidal stations of the South of Spain. The Fractal Physica 13D, 34. Guzmán, M., Martín, M.A., Morán, M. & Reyes, M., 1993, Geometry allows the contraction of the dynamic of the Estructuras fractales y sus aplicaciones, Ed Labor. multidimensional dynamical system of tides around a Mandelbrot, B.B., 1989, Multifractal Measures, Especially for single attractor point identified on the Mean Sea Level Geophysicist, in Fractals in Geophysics, Eds C.H. Scholz & B.B. Mandelbrot, pp 5-42. of the analyzed time series. Among the three well A. Provenzale,A.R. Osborne,A.D. Kirwan andL. Bergamasco,1989. known fractal dimensions (Hausdorff, information and Fractal Drifter Trajectories in the Kuroshio Extension. Tellus, correlation dimensions) the most proper one to be 41A:416-435. applied on time series with organized motion in it is Pacheco, J.M., 1994, Fractales y Oceanografía. Revista de la S.A.E.M., nº 28, vol 10 (1). the D2 dimension. Pla, J., 1994, Los fractales. Revista de la S.A.E.M., nº 28, vol 10 (1). Roosbeek, F., 1995, A Tide Generating Potential at Nanogal Level, From the numerical results the Algeciras tidal Bull. Inf. Mareés Terrestres, 121, pp 9032-9035. Takens, F., 1981, Detecting Strange Attractors in Turbulence, Lecture station presents anomalous values respect the Malaga Notes in Mathematics, D.A. Rand and L.S. Young eds. Springer and Ceuta tidal stations. Hence the Algeciras tidal Verlag.

15 J. J. Alonso, P. Villares, M. J. González, M. Arias & B. Marín

Todoechuck, J.P. & Jensen, O.G., 1989, Scaling Geology and Seismic Deconvolution, in Fractals in Geophysics, Eds C.H. Scholz & B.B. Mandelbrot, pp 273-288. Turcotte, D.L., Fractals in Geology and Geophysics, in Fractals in Geophysics, Eds C.H. Scholz & B.B. Mandelbrot, pp 171-198.

(Received: March, 11, 2004. Accepted: November, 5, 2004)

16 Thalassas, 2005, 21 (1): 17-28 An International Journal of Marine Sciences

LIVING FORAMINIFERAL ASSEMBLAGES FROM THE MINHO AND COURA ESTUARIES (NORTHERN PORTUGAL): A STRESSFUL ENVIRONMENT

J. MORENO(1), F. FATELA(2), C. ANDRADE(2), J. CASCALHO(3), F. MORENO(4) & T. DRAGO(5)

Keywords: Benthic foraminifera; tidal marsh; estuary; hydrodynamics; low salinity; submersion time; Minho and Coura estuaries.

ABSTRACT Benthic foraminifera clearly reflect the hydrodynamics of the estuary. Upper estuary and marsh assemblages Forthy eight surface sediment samples were are dominated by the low salinity species Miliammina collected in April and July 2002 in the Minho and Coura fusca (Brady), Psammosphera sp. and estuaries, where pronounced daily changes in Haplophragmoides manilaensis Anderson, which are environment conditions were found. The samples confined to the lower-energy sites. Marine species are included bottom estuary and tidal marsh deposits in just found at the lower estuary, close to the mouth. Tidal order to characterize the benthic foraminiferal zonation. marsh assemblages distribution is related with the A total of 34 calcareous and 28 agglutinated species topography, reflecting the duration of submersion time. have been identified. Bottom morphology of the estuary and high precipitation limit the penetration of the salt INTRODUCTION wedge to ~4 km upstream from the coast, during summer high-tide, and flush out salt water during low The work presented in this paper has been tide. These conditions are responsible for a daily developed as a preliminary task of the ENVI- stressful environment with a fluvial dominant signature. CHANGES project. One main goal of this project is the palaeoenvironmental reconstruction of the Minho estuary since the Last Glacial Maximum, based on the study of long sedimentary cores. A study of the present-day assemblages of benthic foraminifera from (1) Centro de Geologia da Fac. Ciências, Univ. Lisboa, Campo Grande, 1749-016 LISBOA, PORTUGAL this estuary has been undertaken in order to establish (2) Centro e Departamento de Geologia da Fac. Ciências, an ecological data base to support further Univ. Lisboa, Campo Grande, 1749-016 LISBOA, PORTUGAL palaeoenvironmental interpretation. (3) Museu Nacional de História Natural (Mineralogia e Geologia), Rua da Escola Politécnica, 58, 1250-102, LISBOA, PORTUGAL The Minho River is located in northern Portugal (4) Departamento de Ciências da Terra, Univ. Minho, Campus (Figure 1), where it defines the political border with de Gualtar 4710-057 BRAGA, PORTUGAL Spain (Galicia region). Its estuary trends NNE-SSW (5) INIAP-IPIMAR, Av. 5 de Outubro, 8700-305 OLHÃO, PORTUGAL and presents a high-mesotidal regime: the maximum corresponding author: [email protected] amplitude of the astronomical spring tide reaches 4 m,

17 J. Moreno, F. Fatela, C. Andrade, J. Cascalho, F. Moreno & T. Drago

and may be amplified by storm surge (Taborda and MATERIAL AND METHODS Dias, 1991). The dynamic tidal effects are felt up to a distance of 40 km upstream, due to the tidal regime and Minho and Coura estuaries to the smoothness and low gradient of the Minho's outlet (Alves, 1996). Sixteen sediment samples were collected along the main channels of Minho and Coura estuaries The estuary is very shallow due to widespread (Figure 1) in July 2002, using a small Van Veen grab siltation: a significant part of the bottom emerges sampler. during low water spring tide (Figure 2), when connection with the sea is made by two shallow Each sample consists of 10 cm3 of the topmost 1 channels which incise the bottom sediment down to -1 cm layer of surface sediment that was stored immersed m depth (south) and -2 m depth (north) (Alves, 1996). in alcohol and Rose Bengal solution, [1g/l] (Lutze, 1964). The widest development of intertidal environments in the left margin of the Minho occurs at its confluence Sediment has been washed through a 63 µm sieve with the Coura river - Caminha tidal marsh -, and foraminifera were concentrated by flotation in providing an adequate site to investigate the present- trichloroethylene before picking (Murray, 1979). day distribution of foraminiferal assemblages in When possible, 100 individuals were counted and relation with the vertical zonation of the tidal marsh. identified in each sample from the living (stained) This distribution will be discussed with respect to microfauna, following the Loeblich and Tappan (1988) salinity, temperature, pH and tidal regime. generic classification.

Figure 1. A - Location of study area in Portugal. B - Minho estuary and Coura tributary. Numbered solid dots indicate subtidal sampling locations in the estuary. Dark dashed area represents the tidal marsh. Sampling profiles PR - Pedras Ruivas, CP - Railway Bridge and PIC - Pinelas.

18 Living Foraminiferal Assemblages from the Minho and Coura Estuaries (Northern Portugal): A Stressful Enviroment

Figure 2. Oblique view of the Minho estuary from Sta. Tegra - Galicia (October 2003); A - low-water at spring tide; B - high-water at spring tide. At each sampling location temperature and salinity environments of the Minho and Coura estuaries vertical profiles were determined under summer (Figure 1). These samples were submitted to the same conditions, at low and high waters in spring tide, to procedure described for the subtidal estuarine samples, evaluate the maximum penetration of marine influence except foraminiferal collection which was performed inside the estuary. These data were obtained using a using a micropipette and a wet picking procedure. WTW conduktometer LF 191, in sailing upstream. Temperature, salinity and pH were measured along Caminha tidal marsh the profiles, during low water, in sediment pore-water. This water seeped and accumulated inside perforated Thirty two surface sediment samples were PVC tubes previously inserted into the sediment to a collected in April 2002 (under spring tide conditions) depth of 40 cm below the surface, following De Rijk along the cross shore profiles of Pedras Ruivas (PR), (1995). Water parameters were measured with a railway bridge (CP) and Pinelas (PIC), which extend multiparameter probe Horiba U-22 and a WTW across the salt marsh, tidal flat and channel conduktometer LF 191. A few sampling sites in the

19 J. Moreno, F. Fatela, C. Andrade, J. Cascalho, F. Moreno & T. Drago

tidal flat did not supply any interstitial water. The During low-tide a significant part of the bottom temperature and salinity of estuarine water has also estuary emerges. The salinity profile shows that brackish been controlled close to the bottom, in high and low conditions (4 to 0.5‰) advance downstream and occupy water during spring tide. the lower estuary until 4 km from the estuary mouth. Salinity values of 4.7 ‰ have been found at the mouth Altimetry of the PR profile have been obtained of the estuary just close to the bottom, whereas salinity using a Zeiss Elta R55 total station and was further of surface water at the same site does not exceed 2 ‰. connected to the national hydrographic datum This pattern is once more replicated by the distribution (Hydrographic Zero -HZ- which lies 2 m below mean of temperature, which shows a strong fluvial influence sea level) using a differential GPS. These data were along the entire estuary (Figure 3b). combined with tabulated astronomical tide prevision (Instituto Hidrográfico, 2002) to calculate relative time The Coura estuary presents well mixed waters of submersion of the sampling locations. Data on during high tide (Figure 4), throughout the 4.5 km elevation and time of successive high- and low-waters upstream from the confluence with the Minho estuary. during the year of 2002 was interpolated using a Nevertheless, polyhaline waters (30 to 16.5‰) do not simple sinusoidal fit (adequate to describe advance more than 1 km, where they are replaced by monochromatic astronomical-driven semidiurnal brackish waters (16.5 to 0.5‰). Temperature follows tides), to calculate a series of water heights every 10 the same pattern. During low tide the Coura estuary is minutes. This series was further processed to compute flooded by freshwater. the annual submersion time of a given location as a function of ground elevation. The pH values of marine water measured inside the estuary range between 8.07 and 8.35. In the Minho RESULTS estuary, pH of the brackish waters fall within the interval 8.31 (S=3.96‰) and 7.95 (S=2.13‰), but the Environmental parameters Coura estuary shows a more acidic tendency, yielding a pH of 6.35 at the confluence with the Minho during Minho and Coura estuaries high-water slack (S=28.8 ‰) and 5.81 during low- water slack (S=0.3 ‰). Preliminary measurements obtained in April 2002 showed a strong variation in salinity during each tidal Caminha tidal marsh cycle at the Minho estuary. At the confluence with the Coura estuary (2.5 km upstream from the mouth), Synoptic measurements of salinity, temperature salinity ranged between 0.3 ‰ in low-water and 28.8 and pH obtained during low-waters along the three ‰ in high-water spring tide. These results indicated profiles are presented in Table 1. the need to increase detailed measurements of temperature and salinity during July spring tides to Salinity in the Pedras Ruivas transect (PR) yielded control the extension of marine influence inside the 0.9 ‰ in the subtidal domain (PR1), contrasting with Minho and Coura estuaries. values of 15.8 ‰ (PR5) to 7.7 ‰ (PR10) in pore-water from the marsh; the temperature presented a low range This contrasting trend was confirmed along the of variability: 11.8ºC (PR1), 12.1ºC (PR8) to 10.3ºC deeper channel of the Minho estuary (Figure 3a). In (PR6); the pH was 5.81 at the subtidal (PR1) and high-tide it behaves as a partially-mixed basin. The ranged between 6.10 (PR5) and 4.03 (PR8). salt-wedge, defined by the 35‰ contour, penetrates 4 km upstream, whilst halocline cannot be followed The Railway Bridge transect (CP) exhibited a more than 10 km upstream (Figure 3a). The salinity of 4.36 ‰ at the subtidal (CP1), against 15.4 ‰ distribution of temperature shows a similar pattern, (CP2) to 7.2 ‰ (CP7); the temperature also presented a preserving the strong contrast between marine low variability: 11.4ºC (CP1) against 12.9ºC (CP4) to (<13.5ºC) and continental waters (22ºC) during 10.9ºC (CP7); the pH was 6.35 at the subtidal (CP1) summer. and ranged between 4.50 (CP2) and 7.95 (CP5).

20 Living Foraminiferal Assemblages from the Minho and Coura Estuaries (Northern Portugal): A Stressful Enviroment

Figure 3a. Profiles of temperature and salinity in the Minho estuary (10th of July 2002 - high-tide 16:12h; tidal height 3.38m)

Temperature of sediment pore-water doesn't show a significant variation in the Caminha tidal marsh. The CP transect (at the Coura mouth) shows the less acidic sediment pore-water; the more acidic values are recorded along the PIC transect, close to the inside limit of Caminha tidal marsh, at the Coura inner estuary. Considering each profile, the highest and the lowest pH values occur in the marsh domain, with exception to CP2. Salinity of sediment pore-water (measured during low tide) is significantly higher in the marsh domain than in the tidal flat and channels, with exception again to CP2.

Foraminiferal Assemblages

Figure 3b. Minho and Coura estuaries Profiles of temperature and salinity in the Minho estuary (11th of July 2002 - low-tide 11:34h; tidal height 0.72m) Most of the sampling sites are barren of The Pinelas transect (PIC) yielded a salinity of foraminifera. In the Minho estuary only 3 of the 11 0.1 ‰ in the subtidal (PIC1) against 2.9 ‰ (PIC4) to sites yielded foraminifera (Figure 5); in the Coura 1.6 ‰ (PIC5); the temperature presented once more estuary they were only found at the confluence with a low variability: 11.9ºC (PIC1) against 12.3ºC the Minho estuary (Caminha, site 2a), where the (PIC5) to 11.2ºC (PIC6); the pH yielded 4.33 in the assemblage is dominated by Miliammina fusca subtidal (PIC1) and ranged between 5.65 (PIC5) and (Brady) (91.2%). The foraminiferal assemblage of 3.65 (PIC4). Camarido (site 1a) is dominated by Bolivina spp.

21 J. Moreno, F. Fatela, C. Andrade, J. Cascalho, F. Moreno & T. Drago

Figure 4. Profiles of temperature and salinity in the Coura estuary (11th of July 2002 - high-tide 16:53h; tidal height 3.48m).

(46.7%) and Reophax nodulosus Brady (15.6%), foraminifera, H. germanica and Elphidium cf. followed by M. fusca (Brady) (4.9%), Cibicides incertum (Williamson), are present but in low lobatulus (Walker and Jacob), Gavelinopsis praegeri proportion, ranging between 4.3% and 0.5% (Figures (Heron-Allen and Earland) and Haynesina germanica 6a and b) and exhibit very thin tests. At Pinelas (Ehrenberg) (4.1% each). At the slope of Canosa island transect, calcareous foraminifera are completely absent (site 3a, 4 km upstream the Minho estuary) the (Figure 6c). assemblage is dominated by Jadammina macrescens (Brady) (41.1%). Haplophragmoides wilberti M. fusca, H. manilaensis, Psammosphaera spp. Andersen (10.1%), Trochammina inflata (Montagu) and Pseudothurammina limnetis (Scott and Medioli) (9.5%), Haplophragmoides manilaensis Andersen are the dominant species in all transects. They (7.6%) and M. fusca (6,3%) also appear. H. correspond to agglutinated foraminifera, characteristic manilaensis (50.9%) becomes the most important from marginal marine environments and related to specie at Lanhelas margin (site 5a, 7.5 km upstream very low salinity conditions (Murray, 1991; Sen Gupta, the Minho estuary), followed by M. fusca (13.9%), H. 2002). wilberti (7.4%) and Cribrostomoides jeffreysii (Williamson) (6.5%). Along the PR and CP transects five different settings may be distinguished based upon Caminha tidal marsh foraminiferal assemblages distribution (Figure 7):

A total of 13 different species have been identified 1) Channel and tidal flat: strongly dominated by. M. in the living assemblages (Figures 6a, b and c). The fusca and Psammosphaera sp., associated with H. density of living specimens ranges between 0 and 300 germanica and Elphidium sp.; extending from individuals per cm3 of sediment, the higher values infratidal to 2.7 m above HZ; area submerged more being found in the transition between low and high than 22% of time along the year. Foraminifera are marsh and the lower ones in the tidal flat. Calcareous absent from PR2 and PR3 (tidal flat).

22 Living Foraminiferal Assemblages from the Minho and Coura Estuaries (Northern Portugal): A Stressful Enviroment

Table 1. Salinity, temperature and pH of pore-water and elevation of sampling sites, relative to the Hydrographic Zero (HZ).

2) Low marsh (sites PR4, PR5 and CP3): strongly The Pinelas transect is barren of foraminifera at the dominated by M. fusca and Psammosphaera sp., channel and the assemblages from the tidal marsh associated with Saccammina sp., J. macrescens show a similar trend than the other transects. and H. germanica; placed between 2.7 m and 3.0 Nevertheless, the low foraminiferal abundance does m (above HZ), with a submersion time inferior to not allow to establish a secure approach and a clear 22%. zonation in this profile.

3) Lower high marsh (sites PR6, PR7, CP4, CP5 and DISCUSSION CP6): strongly dominated by M. fusca associated with Psammosphaera sp., P. limnetis and H. The Minho estuary provides a small wilberti; from 3.0 m to 3.5 m (above HZ), with a accommodation volume for the tidal prism and the submersion time inferior to 14%. large sand shoal that occupies most of the estuarine mouth constricts the sea water input, which is more 4) Upper high marsh (sites PR8 and CP7): strongly effective during spring tides. In addition, the Minho dominated by P. limnetis associated with H. basin drains the rainiest region of Iberian Peninsula, manilaensis and Haplophragmoides sp.; from 3.5 with an average annual precipitation of 1300 mm m to 3.86 m (above HZ), with a submersion time (Alves, 1996). All together these characteristics inferior to 1%. prevent extensive penetration of a salt wedge into the estuary, which essentially behaves as "partially mixed" 5) Highest high marsh shows a strong presence of H. (Brown et al., 1991). During ebb tide, the sea water is manilaensis, which dominates assemblages in sites completely flushed out. PR10 and CPas, associated with Haplophragmoides sp., placed above 3.86 m Living (stained) benthic foraminifera have been (above HZ), with a submersion time also inferior found in only 4 of the 16 sampling sites from the to 1%. Minho and Coura estuaries. Sediment from these sites

23 J. Moreno, F. Fatela, C. Andrade, J. Cascalho, F. Moreno & T. Drago

is mud, reflecting a low energy environment, in transects are consistent in showing higher salinity contrast with the coarser grain size (mainly sand) of values of interstitial water at the transition from low to most of the estuarine sediments (Fatela et al., 2003). high marsh (Figure 6). The lower salinity values recorded at the highest points in the Pedras Ruivas One aspect requiring further research is the high marsh can be explained by the shorter submersion relationship between the occurrence of calcareous time of the area and the increased exposure to rain foraminifera and the penetration of marine water inside showers and surface run-off from surrounding land, the estuary. These foraminifera are apparently contributing to dilute salt content in pore-water. confined within the upstream limit of the 30‰ Another cause for this lower salinity, still under isohaline, placed at Pedras Ruivas, 4 km away from the research, is the possible deflection or accommodation estuary mouth. of the fresh water output from the Coura river over the fringe marsh during rising tide. The variation of The tidal marsh is a relatively "stable" salinity in pore-water within the tidal flat (when environment, in spite of extreme daily salinity present) reflects the dynamics of fluviatile and marine variations of the estuarine waters, between fresh and water led by tidal forcing. fully marine conditions. The more regular, large scale fluctuations in the main environmental parameters that The pH inside the estuary seems also to be related characterise tidal influence in the estuarine water are with the tide cycle and the consequent balance between smoothed in the tidal marshes. For instance, the marine and freshwater, namely in the Coura salinity of sediment pore-water at Pedras Ruivas confluence. This trend follows the observations ranges in the interval 15.8 ‰ and 7.7 ‰ along the tidal recorded in different estuaries world-wide (e.g. marsh transect at low tide, while freshwater conditions Debenay et al., 2002; Cunha and Dinis, 2002). In the prevail at the tidal flat (S=0.9 ‰) and main channel pore-water of the tidal marshes the relationship (S=0.3 ‰). between pH and salinity is unclear, as we have observed the lowest values of pH (Table 1) at Pinelas The variation in salinity of pore-water along the (located in the inner Caminha tidal marsh transect) and tidal marsh transects seems to be related with site at the highest highmarsh location of the remaining elevation and consequently, submersion time. All transects. This pattern seems to reflect higher

Figure 5. Living (stained) foraminiferal assemblages from Minho estuary.

24 Living Foraminiferal Assemblages from the Minho and Coura Estuaries (Northern Portugal): A Stressful Enviroment

Figure 6a. A - Pedras Ruivas (PR) transect (April 2002) showing location of sampling sites, B - salinity and C - species abundances in the foraminiferal assemblages. concentration of fresh water input from rain and carbonate layer. This can be interpreted as a direct surface run-off from surrounding land, as a outcome of the strong influence of acidic freshwater consequence of shorter tidal induced flooding. with lower Ca contents (Debenay et al., 2002; Le Probably, litter accumulation also contributes to the Cadre et al., 2003), highly probable in the Minho and low values of the pH recorded in these sites. Coura tidal marsh where run-off drains a granitic basement. In fact, Paiva et al. (1993) reported low Ca The foraminifera present in the depositional concentrations in the grain size fraction < 2 mm of environments of the Minho and Coura estuaries sediment samples from the Minho river, collected correspond essentially to species with agglutinated between the estuary mouth and 32 km upstream. tests. Porcellaneous foraminifera are completely These authors measured Ca contents by EDXRF absent at all sites and despite the dominance of (Energy Dispersive X-Ray Fluorescence calcareous hyaline species close to the estuary mouth Spectrometry) and values ranged from 900 ppm to (sites 1a and 1b) their abundance is very low at all the 8200 ppm. These low values reflect the geochemical remaining sites. The presence of living Rotaliina in signature of systematic Ca depletion relative to the the tidal marsh transects is confined between the contents of the average siliceous rocks (Wedepohl, channel and the low marsh subenvironment, and the 1995) exposed to weathering in the Minho river tests within these environments often lack the basin.

25 J. Moreno, F. Fatela, C. Andrade, J. Cascalho, F. Moreno & T. Drago

Figure 6b. A - Railway Bridge (CP) transect (April 2002) showing location of sampling sites, B - salinity and C - species abundances in the foraminiferal assemblages.

Benthic foraminifera assemblages composed of the CONCLUSIONS agglutinated species M. fusca, Psammosphaera sp. and H. manilaensis occur frequently in many tidal marsh The low-salinity foraminifera assemblages of the environments around the world. They represent often Caminha tidal marsh reflect the semidiurnal extreme the most abundant species, and are interpreted as variations between fluviatile and marine conditions in indicators of low salinity (Murray, 1991; Sen Gupta, this estuary. The site-scale morphology, precipitation 2002; Debenay et al., 2002). and hydrodynamic-level are considered as the main factors leading to the complete flushing out of marine The foraminiferal assemblages studied in the water from the Minho and Coura estuaries during each present work are clearly dominated by M. fusca, low tide. Psammosphaera sp., H. manilaensis and P. limnetis, reflecting the semidiurnal variation between brackish All the studied transects show a consistent and freshwater conditions at the Minho and Coura distribution of foraminiferal assemblages. M. fusca estuaries. The distribution of foraminiferal species and Psammosphaera sp. are the dominant species on along the Caminha tidal marsh transects is determined the channel, tidal flat and low marsh, where the by sites altimetry and consequently by the different environmental parameters (temperature, salinity and relative submersion time, fitting the common marsh pH) show the strongest daily fluctuations. Three zones zonation (Figure 7). can be clearly defined on the high marsh:

26 Living Foraminiferal Assemblages from the Minho and Coura Estuaries (Northern Portugal): A Stressful Enviroment

Figure 6c. A - Pinelas (PIC) transect (April 2002) showing location of sampling sites, B - salinity and C - species abundances in the foraminiferal assemblages.

- IB, where M. fusca and Psammosphaera sp. are potential tool to the Late Quaternary coastal evolution dominant, associated with P. limnetis, J. studies of this region, to be developed in future works. macrescens, T. inflata and H. wilberti. ACKNOWLEDGEMENTS - IA2, where P. limnetis is the dominant species associated with H. manilaensis and This paper is a contribution of the Envichanges Haplophragmoides sp.; project, funded by FCT (PLE/12/0). We would like to thank Dr. Carlos Antunes (ICBAS, Univ. Porto) for the - IA1, where H. manilaensis and Haplophragmoides nautical facilities and to Celso Pinto for assistance sp. are the dominant species. during field work. The authors are grateful to G. Frances and to the anonymous reviewer for the The close relationship between Caminha tidal thorough revision of the manuscript that greatly marsh zonation and the altimetry of sites, represent a contributes to its improvement.

27 J. Moreno, F. Fatela, C. Andrade, J. Cascalho, F. Moreno & T. Drago

Figure 7. Zonation of tidal marsh foraminiferal assemblages.

REFERENCES Loeblich, A.R.Jr. & Tappan, H. 1988. Foraminiferal Genera and their Classification. Van Nostrand Reinhold Company, 1, 970 pp., 2, Alves, A. 1996. Causas e Processos da Dinâmica Sedimentar na 847 pl. Evolução Actual do Litoral do Alto Minho. PhD thesis, Univ. Lutze, G.F. 1964. Zum Farben Rezenter Foraminiferen. Meyniana, 14: Minho, Braga. Unpublished. 442 pp. 43-47. Brown, J.; Colling, A.; Park, D.; Phillips, J.; Rothery, D. & Wrigh, J. Murray, J.W. 1979. British Nearshore Foraminiferids. Synopsis of the 1991. Waves, Tides and Shallow-Water Processes. The Open British Fauna (New Series), 16, Academic Press, London, 60 pp. University, 187 pp. Murray, J.W. 1991. Ecology and Palaeoecology of Benthic Cunha, P. & Dinis, J. 2002. Sedimentary Dynamics of the Mondego Foraminifera. Longman Scientific & Technical, London, 397 pp. Estuary. In: Aquatic Ecology of the Mondego River Basin, Global Paiva, P.C.; Araújo, M.F.; Dias, J.M. & Jouanneau, J.-M. 1993. Importance of Local Experience. P.A. Pardal, J.C. Marques & Distribuição Elementar em Sedimentos do Rio Minho. Memórias M.A. Graça (eds.): 43-62, Imprensa da Universidade, Coimbra. do Museu e Laboratório de Mineralogia e Geologia da Debenay, J.P.; Guiral, D. & Parra, M. 2002. Ecological Factors acting Faculdade de Ciências da Universidade do Porto, 3: 561-566. on Microfauna in Mangrove Swamps. The Case of Foraminiferal Sen Gupta, B. 2002. Foraminifera in marginal marine environments. Assemblages in French Guiana. Estuarine, Coastal and Shelf In: Modern Foraminifera. Sen Gupta, B. (ed.): Kluwer Academic Science, 55(2): 509-533. Publishers, p. 141-159. De Rijk, S. 1995. Salinity Control on the Distribution of Salt Marsh Taborda, R. & Dias, J.M.A. 1991. Análise da sobre elevação do nível Foraminifera (Great Marshes, Massachusetts). Journal of do mar de origem meteorológica durante os temporais de 1978 e Foraminiferal Research, 25(2): 156-166. 1981. Geonovas special issue 1: 89-97. Fatela, F.; Moreno, J.; Cascalho, J.; Silveira, T.; Fradique, C. & Drago, Wedepohl, K.H. 1995. The composition of the continental crust. T. 2003. Variação Extrema da Salinidade: Uma Condicionante Geochimica et Cosmochimica Acta, 59 (7): 1217-1232. da Distribuição dos Foraminíferos "Vivos" no Estuário dos Rios Minho e Coura. Ciências da Terra, (UNL), nº esp. V, CD-Rom, pp. A68-A71. Lisboa. Le Cadre, V.; Debenay, J.-P. & Lesourd, M. 2003. Low pH Effects on Ammonia beccarii Test Deformation: Implications for Using Test Deformations as a Pollution Indicator. Journal of Foraminiferal Research, 33(1): 1-9. (Received: May, 11, 2004. Accepted: July, 12, 2004)

28 Thalassas, 2005, 21 (1): 29-33 An International Journal of Marine Sciences

SEXUAL DIMORPHISM IN THE RADULA OF PISANIA PUSIO (LINNAEUS, 1758) (MOLLUSCA, GASTROPODA, BUCCINIDAE)

H. MATTHEWS-CASCON(1), H. A. PEREIRA ALENCAR(2), S. GUIMARÃES RABAY(2) & R.M.S. MOTA(3)

Keywords: Mollusca, prosobranch, radula, sexual dimorphism.

ABSTRACT females and males had five dimensions measured. For comparison of the radula size between sexes, the "t" Pisania pusio (Linnaeus, 1758) is an inhabitant of test of Student was used. For the males the average size the intertidal zone, being generally found under the of the radula was of 10.25±2.4 mm of length, for the rocks during the low tide. It presents spindle, smooth females, the average size of the radula was of 8.87±1.2 and glossy shell measuring 34 mm of length, with mm of length. The length of radular ribbon of showed purplish brown color with narrow, revolving, dark a statistically significant difference between male and brown bands, sometimes in the shape of arrowheads female (p=0.002). The size of the lateral teeth of the and rachiglossate radula type. The objective of this males and females showed statistically significant study was to investigate the occurrence of sexual difference in five dimensions measured and the size of dimorphism in the size of the radula and in the shape the rachidian teeth of males and females showed of the teeth among adults of Pisania pusio. Forty-two statistically significant difference in four dimensions adult specimens of P. pusio were collected in the measured. Pacheco Beach, Caucaia, State of Ceará, Northeast Brazil. Twenty specimens were males and twenty-two INTRODUCTION were females. The radula was removed, and then measured with a milimetric slide under compound Prosobranch mollusks may present sexual microscope. Sixty lateral and rachidian teeth of both differences in the shell, radula and in some parts of the body. The differences in the shell are generally related to size and format, with the females usually larger than the males as in the species Voluta ebraea (Matthews, 1969), Pugilina morio (Matthews-Cascon, Matthews (1)Departamento de Biologia, Centro de Ciências, Instituto de Ciências do Mar. Universidade Federal do Ceará (UFC), & Belucio, 1990) and Lacuna pallidula (Bandel, 1976; Campus do Pici, Bloco 906, Fortaleza - Ceará - Brasil, Fretter, 1984; Fretter & Graham, 1994). Dimorphism 60.455-970. e-mail: [email protected] only affecting the weight of the animals was registered (2)Departamento de Biologia, Centro de Ciências, Universidade Federal do Ceará in some species of Polinices, where the males are (3)Departamento de Estatística e Matemática Aplicada, Centro generally heavier than the females (Bernard, 1968; de Ciências, Universidade Federal do Ceará Fretter, 1984). Sexual dimorphism in shell color is

29 H. Matthews-Cascon, H. A. Pereira Alencar, S. Guimarães Rabay & R.M.S. Mota

found in Cypraea gracilis, where the females are red and the males brown (Griffiths, 1961). In Ficus subintermedia, the propodium in the males is sharp while in the females is dull (Fretter, 1984). Sexual dimorphism in radula size is found in Nassa francoliana (Maes, 1966; Fretter, 1984) where the rachidian tooth is larger in the males. In some Cypraeidae the radula of females is smaller and the shape of the rachidian tooth presents sexual dimorphism (Schilder & Schilder, 1961). The radula of some species of Drupella presents sexual differences in the lateral teeth (Arakawa, 1958a). Another species that presents sexual dimorphism in the radula is Tricolia variabilis, where the males have less marginal teeth than the females (Robertson, 1971).

The Buccinidae Pisania pusio is a slow moving gastropod found in the intertidal zone under the rocks associate with tunicates, sponges and calcareous algae. Pisania pusio presents spindle smooth and glossy shell Figure 1. Five dimensions measured in the lateral tooth of Pisania pusio: a- Lateral tooth measuring in average 34 mm of length, with purplish width; b- Lateral tooth outer cusp length; c- Lateral tooth width between outer brown color with narrow, revolving, dark brown and middle cusps; d- Length of inner cusp; e- Width of inner cusp bands, sometimes in the shape of arrowheads and rachiglossate type radula. The objective of this study was to investigate the occurrence of sexual dimorphism in the size of the radula and in the shape of the radular teeth among adults of Pisania pusio (Linnaeus, 1758).

MATERIAL AND METHODS

Forty-two adult specimens of P. pusio (20 males and 22 females) were collected in the intertidal zone in the Pacheco Beach (38º38'48"W, 03º41'24"S) Caucaia, State of Ceará, Northeast Brazil, in October of 2001. The animals had the shell length measured and were anesthetized in 30% magnesium chloride and later fixed in 70% alcohol. After the initial examination for determination of the sex, the proboscis was removed and boiled in solution of saturated potassium hydroxide (KOH), until the soft parts dissolved and remained just the radula.

The radulae of all individuals were measured on a milimetric slide, and those of six males and of six females were clarified with xylol, stained with Congo Figure 2. red diluted in absolute alcohol and had their teeth Five dimensions measured in the rachidian tooth of Pisania pusio: a- Rachidian removed under compound microscope. The teeth were tooth length; b- Rachidian tooth width; c- Rachidian tooth width between middle cusps; d- Rachidian tooth length without the cusps; placed in a microscopic slide with glycerin and e- Rachidian tooth length of all middle cusps.

30 Sexual Dimorphism in the Radula of Pisania pusio (Linnaeus, 1758) (Mollusca, Gastropoda, Buccinidae)

measured under an optic microscope with a graduated sexes in the variance was also statistically significant ocular lens. Sixty lateral and rachidian teeth of both (p<0.01), with the measures of the females being less females and males had five dimensions measured variant. (Figs.1, 2). The rachidian tooth of P. pusio males presented Four radulae (two of each sex) were observed and higher values than that of females in four of the five photographed under scanning electron microscope dimensions measured, with the differences being (SEM) model XI 30, covered with carbon filament statistically significant (Table 1). (Metalizer the Sputer Coater SCD 0,50). All the statistical analysis were developed with the program The lateral teeth of P. pusio showed statistically GraphPad Instat® Version 3.06, 32bit for Windows. significant difference among the sexes in all five The normality of the distribution of the collected data dimensions measured, when the males presenting was verified with the test KS, while the veracity of higher values (Table 2). difference in the samples was verified with Mann- Whitney test. Besides the differences in the size, P. pusio presented sexual dimorphism in the shape of the All shell measurements were made with a vernier radular teeth. The rachidian tooth is subquadrate with caliper to 0.1mm. The shell length of Pisania pusio three big cusps in the middle, but in the females the was the distance from the apex to the tip of siphonal base of the rachidian is concavewhile in the males is canal. straight. The lateral teeth has three cusps inclined toward the central part of radular row (Figs. 3,4), but RESULTS in the males the inner cusp is longer and slim than the females. The radula ribbon of P. pusio males presented in average a length of 10,25±2,38 mm and that of females The shell length in adults of Pisania pusio did not was in average 8,87±1,2 mm. The difference was showed a statistically significant difference between statistically significant (p=0.02). The difference among males and females (p=0.3385).

Table 1. Comparative measurements of the rachidian tooth of males and females in Pisania pusio from Pacheco Beach, Caucaia, State of Ceará, Northeast Brazil, collected in October of 2001. N=60.

31 H. Matthews-Cascon, H. A. Pereira Alencar, S. Guimarães Rabay & R.M.S. Mota

Table 2. Comparative measurements of the size of the lateral teeth of males and females in Pisania pusio from Pacheco Beach, Caucaia, State of Ceará, Northeast Brazil, collected in October of 2001. N=60.

DISCUSSION Sexual dimorphism in the size the radular ribbon is largely dependent on the growth of individual teeth, There have been many studies on the gastropod because the length of radular ribbon is determined by radula, where the radular characters are generally the tooth size or the thickness of the tooth bases along considered constant within the species or that the longitudinal axis by the number of rows (Fujioka, individual variation does not exceed the difference 1985a). between species (Fretter & Graham, 1994). However, many studies have demonstrated that the radular According to Fujioka, (1985b) sexual dimorphism characters are modified by many factors as seasonal in the radula of Muricidae was predominantly changes, sexual differences, age (Maes,1966, Fujioka, observed in the genera with simplified tricuspidate 1984, 1985a, 1985b, 1985c). rachidians such as Mancinella and Cronia. Pisania pusio rachidian tooth is also simplified with three big Sexual dimorphism in the radula of Muricidae cusps in the middle. have been reported in four species of Drupella (Arakawa, 1958a; Fujioka, 1982) in two species of Robertson (1971) belived that sexual dimorphism Nassa (Maes, 1966), in four species of Mancinella of the radula found in species of the genus Hiloa may (Arakawa, 1958b; Fujioka, 1985a), and in Morula be selectively advantageous for their different food musiva and Cronia margariticola (Fujioka, 1984). requirements. The biological significance of this Neverthless, in Buccinidae the only case reported was dimorphism in Pisania pusio if any, is yet unknown. It for Pisania luctuosa (Cernohorsky, 1971), where as is not possible at present to determine whether this found in the present study for Pisania pusio, the radula radular dimorphism is functional. ribbon was longer in the males than in the females. ACKNOWLEDGEMENTS Larger teeth in the radula of adults males than in females and sexual dimorphism in the shape of the We are very grateful to Dr.Paulo Cascon who rachidian tooth, found in this study for Pisania pusio provided useful insights and for revising English was also reported for some species of Mancinella, a language. For assistance with electron microscopy we muricid gastropod (Fujioka, 1985b). thank Daercio da Costa Magalhães.

32 Sexual Dimorphism in the Radula of Pisania pusio (Linnaeus, 1758) (Mollusca, Gastropoda, Buccinidae)

Figure 4. Male radula of Pisania pusio (S.E.M.). Measure bar=100µm.

Figure 3. Female radula of Pisania pusio (S.E.M.). Measure bar=100µm.

REFERENCES

Arakawa, K.Y. 1958 a. On the remarkable sexual dimorphism of the Fujioka, Y. 1985 c. Seasonal aberrant radular formation in Thais radula of Drupella. Venus 19 : 206-214. bronni (Dunker) and T. clavigera (Kuster) (Gastropoda: Arakawa, K.Y. 1958 b. Some notes on the radula of Purpura echinata Muricidae). J. Exp. Mar. Biol. Ecol., 90: 43 - 54. Lamarck. Venus, 20: 69 - 75. Griffiths, R.J. 1961. Sexual dimorphism in Cypraeidae. Proc. Malacol. Bandel, K. 1976. Spawning, development and ecology of some higher Soc. London, 34: 203-206. Neogastropoda from the Caribbean Sea of Columbia (South Maes, V.O. 1966. Sexual dimorphism in the radula of the muricid America). Veliger 19 (2) : 176-193. genus Nassa. Nautilus, 79: 73-80. Bernard, F.R. 1968. Sexual dimorphism in Polinices lewisi Matthews, H. 1969. Notas sobre a familia Volutidae no Nordeste (Naticidae). Nautilus 82: 1-3. brasileiro (Mollusca : Gastropoda). Arq. Ciên. Mar, 9: 71-75. Cernohorsky, W.O. 1971. Indo-Pacific Pisaniinae (Mollusca: Matthews-Cascon, H.; Matthews, H. & Belucio, L. 1990. Notas sobre Gastropoda) and related buccinid genera. Rec. Auckland Inst. a anatomia, sistemática e biologia de Pugilina morio Linnaeus, Mus., 8: 137 - 167. 1758 (Mollusca: Gastropoda). Arq. Ciên. Mar, 28: 3-8. Fretter, V. 1984. Prosobranchs. In: K.M.Wilbur (Ed.), The Mollusca. Robertson, R. 1971. Sexually dimorphic archaeogastropods and Reproduction vol. (1 - 45pp.). New York: Academic Press, Inc. radulae. Annu. Rep. Am. Malacol. Union., 75-78 pp. Fretter, V. & Graham, A. 1994. British prosobranch molluscs. London: Schilder, F.A. & Schilder, M. 1961. Sexual differences in cowries. The Ray Society. 820 pp. Proc. Malacol. Soc., 74: 207-209. Fujioka, Y. 1982. On the secondary sexual characters found in the dimorphic radula of Drupella (Gastropoda: Muricidae) with reference to its taxonomic revision. Venus, 40: 203 - 223. Fujioka, Y. 1984. Sexually dimorphic radulae in Cronia margariticola and Morula musiva (Gastropoda: Muricidae). Venus, 43: 315 - 330. Fujioka, Y. 1985 a. Systematic evaluation of radular characters in Thaidinae (Gastropoda: Muricidae). J. Sci. Hiroshima Univ., Ser. B, Div.1, 31: 235 - 287. Fujioka, Y. 1985 b. Sexually dimorphic radulae in four species of Mancinella (Gastropoda: Muricidae). Mukaishima Mar. Biol. Station, 238: 73 - 81. (Received: May, 24 2004. Accepted: December, 29, 2004)

33

Thalassas, 2005, 21 (1): 35-44 An International Journal of Marine Sciences

CALCAREOUS NANNOPLANKTON FROM THE GUADIANA ESTUARY AND ALGARVE CONTINENTAL SHELF (SOUTHERN PORTUGAL): AN ECOLOGICAL MODEL

J. FERREIRA (1) & M. CACHÃO (2)

Keywords: Calcareous nannoplankton; Ecology; Guadiana river; Gulf of Cadiz; Hydrology; Oceanography; Portugal.

ABSTRACT they approach shoreline while more euthrophic communities increase their density towards the coast. During September and November 2001, water samples from six stations along the Guadiana river INTRODUCTION mouth and estuary and 28 surface sediments along three transects over its continental shelf between 9-331 Coccolithophores form a major component of the m water depth were collected for calcareous extant marine nannoplankton and are one of the main nannoplankton analysis. The most abundant open ocean primary producers. They play a significant coccolithophores both in the water column and surface role in the CO2-O2 exchanges between the ocean and sediment were Gephyrocapsa ericsonii, G. oceanica, the atmosphere, seriously affecting both the biological G. muellerae and Emiliania huxleyi. The distribution of and the carbonate pumps (Sikes et al., 1994) and extant and fossil calcareous nannoplankton producing an additional feedback to climate changes abundances in the water column show distinct Summer (Westbroek et al., 1993). and Winter patterns. During Summer fossil nannoliths predominate over extant ones while during Winter the The coccoliths, which form a discontinuous opposite pattern is registered. This is interpreted as a calcareous shell around the living single cell, result of seasonal changes in productivity, constitute the most important component of deep-sea hydrographic and oceanographic conditions. On the sediments and providing highly significant other hand, distinct ecological behaviours can be information for the interpretation of global change in inferred from calcareous nannoplankton abundances the geological record (Perch-Nielson, 1989). when plotted along continental shelf transects: oligothrophic communities decline its abundance as Calcareous nannoplankton comprises present day coccolithophores and other phytoplankton groups. Mainly oceanic, their role in coastal environments is (1) Geology Centre. Edif. C6 Campo Grande, 1749-016 still poorly understood. Several works have already Lisboa, Portugal; [email protected] addressed this issue in distinct sectors of the coastal (2) Centre and Dep. Geology Fac. Scienc. Univ. Lisbon, Edif. C6 Campo Grande, 1749-016 Lisboa, Portugal; margin off Portugal (Cachão, 1993; Abrantes & Moita, [email protected] 1999; Cachão & Oliveira, 2000; Cachão et al., 2000;

35 J. Ferreira & M. Cachão

Figure 1. Study area and surface sediment sampling locations.

Moita, 2001). The area off the Guadiana estuary though has not yet been analysed in detail (Fig. 1). The present work focuses on the distribution and behaviour of the most common species of calcareous nannoplankton dwelling in neritic environments off southern Portugal. Its main purpose is to determine the importance of calcareous nannoplankton in the sedimentary, environmental and ocean dynamic processes that characterize the present day Guadiana river mouth and adjacent shelf area.

The methodology used in this work allowed to determine absolute nannolith and coccolith densities necessary for this quantitative palaeoecological study.

HYDROGRAPHY AND GEOLOGICAL FRAMEWORK

The study area is strongly influenced by Guadiana river discharges and the proximity to the Gibraltar Strait. The Guadiana river flow volume is marked by unusually strong seasonal changes, as well as changes associated with dry and wet years (Loureiro et al., 1986). The minimum measured outflow volume is approximately 6 m3s-1, while the maximum outflow 3 -1 Figure 2. during floods may be as high as 3000 m s or more Water samples location. (Ortega & Garzón, 1997).

36 Calcareous Nannoplankton from the Guadiana Estuary and Algarve Continental Shelf (Southern Portugal): An Ecological Model

The coast adjacent to the estuary has an average Water column samples were collected from the tidal amplitudes of ca. 2 m, reaching a maximum of 3.4 Guadiana estuary, both near surface and near bottom, m at spring tides. Tidal currents at the river mouth during upstream and downstream campaigns (Fig. 2). reach 0.6 ms-1 (direction 340º) during peak flood tide, All water samples were filtered directly at the location and 1.2 ms-1 (direction 140º) during peak ebb tide or in the laboratory through 0.45 µm Milipore cellulose (Instituto Hidrográfico, 1998). At high tide, saline nitrate membrane filters. A sector of approximately 25º coastal waters penetrate 40-50 km into the hinterland of each filter was afterwards permanently mounted on a of the estuary (Rocha et al., 2002). slide and cover slide with synthetic cement (Entellan).

The sediment supply to the study area is brought The surface sediments (Fig. 1) were sampled with by coastal drift, as well by the river's basin runoff. a Smith-MackIntyre sampler and then dried in laboratory. The sediment samples where weighted and MATERIALS AND METHODS filtered through 0.45 m Milipore cellulose nitrate membrane filters. Again, a sector of approximately 25º Several sets of samples were collected from the of each filter was permanently mounted on a slide and water column along the Guadiana River during the cover slide with synthetic cement (Entellan). SIRIA Project (Sept. and Nov. 2001, Portuguese Hydrographic Institute) and from three transects The identification and quantification were performed on surface sediments of the continental shelf under with a polarizing optical microscope Ortholux II Pol-BK the projects SIRIA, EMERGE, ODIANA and under x1250 magnification. The Taxonomy follows CRIDA. Perch-Nielsen (1989) and Bown (1999).

Table 1. Surface sediments absolute abundance of calcareous nannofossils x 106 liths per 1g of sediment.

37 J. Ferreira & M. Cachão

OCEANOGRAPHY

The study area is located about 200 km Northwest of the Strait of Gibraltar where the Mediterranean Outflow Water (MOW) is strongly present.

The dense Mediterranean Outflow Water forms a high-velocity bottom current that interacts with the Spanish and Portuguese continental slope sediment at depths between 400 and 1800 m (Nelson et al., 1993).

This Mediterranean water plume spreads westward and descends the continental slope, mixes intensely with the overlying North Atlantic water while still in the Gulf of Cadiz, and thereby looses much of its density anomaly (Baringer & Price, 1999).

According to Nelson et al. (1999) above 300 m water depth there is a strong southeastward inflow of North Atlantic Superficial Water (hereafter called Atlantic inflow) over the Gulf of Cadiz shelf that intensifies toward the Strait of Gibraltar. The shelf breaks at about 130 ± 20 m water depth. Below 300 m water depth there is a significant development of bottom-current deposited sediment and bedforms grown by deep MOW shear northwestward of Gibraltar along the Cadiz continental slope.

According to Baringer & Price (1999), the jet-like current of outflow water is about 100 m thick and has a maximum speed of 1 ms-1 decreasing westward. The current maximum is near the middle of the outflow layer, and there are generally thick shear layers in the interfacial boundary layer between the Mediterranean water and the North Atlantic water and in the bottom boundary layer. Within the Strait the velocity reaches 1.2 ms-1 until encounters the continental slope and begins to descent and accelerate to maximum speeds of Figure 3. -1 Difference between living and fossil nannoliths in surface sediments. 1.4 ms . Under the influence of the Coriolis force, the For sample location see Fig. 1. current is deflected to the north, and the flow makes a nearly inertial turn of 90º in less than 20 km to align itself along the local topography. Further downstream secondary circulation within the outflow that includes the flow decelerates to less than 0.6 ms-1 (more a downslope (towards the south) component closest to typically about 0.3 ms-1) with the deepest, saltiest part the bottom and upslope (towards the north) component of the outflow moving fastest. Eventually, the flow in the interface between the outflow and the water decelerates to speeds of about 0.1 to 0.3 ms-1 near Cape directly above it (approximately 100 m above the St. Vicente where the outflow starts to completely floor). The strong acceleration due to the continental separate off the bottom. Two important conclusions slope appears to be the main reason for the MOW to can be gained from this. First, there is a clear mix strongly with the North Atlantic Central Water.

38 Calcareous Nannoplankton from the Guadiana Estuary and Algarve Continental Shelf (Southern Portugal): An Ecological Model

The dispersal patterns of shelf surface sediment are Table 2. Water samples distribution during winter 2001 (x 103 liths/l). related to the southeastward currents of the Atlantic inflow water moving towards Gibraltar, the terrigenous input from river sources and the physiography of the shelf. The sediment dispersal of the outer shelf and upper slope, in contrast, is influenced by location of sea level lowstand processes and MOW currents flowing northwestward from Gibraltar (López- Galindo, 1999).

RESULTS

Surface sediments

In shelf surface sediments the calcareous nannoplankton assemblages are generally abundant and were dominated by the small placoliths of Gephyrocapsa ericsonii (6 to 403 x 106 liths/g) and Emiliania huxleyi (1.3 to 288 x 106 liths/g) followed by Gephyrocapsa oceanica (4.6 to 209 x 106 liths/g) and G. muellerae (4.5 to 148 x 106 liths/g). Other assemblage components include Syracosphaera spp. (1 to 81.3 x 106 liths/g), Coccolithus pelagicus (0.3 to 34.2 x 106 liths/g), Calcidiscus leptoporus (0.2 to 33.8 In all of the three sampled transects the extant x 106 liths/g), Umbilicospaera sibogae (0.2 to 25.4 x calcareous nannoplankton dominate over the fossil 106 liths/g), Helicosphaera carteri (0.2 to 18.5 x 106 forms although the reworked ones occur in unexpected liths/g) and Umbellosphaera tenuis (1.7 to 15 x 106 high abundance considering the surrounding liths/g). Additional minor components include geological units that outcrop along this area. In all Rhabdosphaera clavigera (1.1 to 8.5 x 106 liths/g), samples both the nannoplankton and nannofossil Braarudosphaera bigelowii (0.2 to 5.3 x 106 liths/g), abundances decline as we approach the shore. Only Pontosphaera discopora (2.6 to 3.9 x 106 liths/g) and species such as Gephyrocapsa oceanica reach their P. multipora (0.4 to 3.9 x 106 liths/g). Reworked fossil maximum density in the inner shelf. forms, Cretaceous to Pliocene in age, are an important component of the assemblage. The dominant forms are Dictyococcites productus (1.8 to 148 x 106 liths/g), Reticulofenestra minuta (1.1 to 89.4 x 106 liths/g), Watznaueria barnesae (0.16 to 40.8 x 106 liths/g), Cyclicargolithus floridanus (0.16 to 40.7 x 106 liths/g) and Discoaster spp. (0.8 to 24 x 106 liths/g). Other reworked components include Reticulofenestra haqii- minutula (0.9 to 22 x 106 liths/g), Tetralithoides simeonidesii (0.47 to 20 x 106 liths/g), Hughesius tasmaniae (1.1 to 18 x 106 liths/g), Cruciplaccolithus asymmetricus (1.1 to 12.7 x 106 liths/g) and Cribrosphaerella ehrenbergii (0.3 to 10 x 106 liths/g). The distribution of nannofossil taxa in surface sediments is presented in Table 1 and the relative proportion between living and fossil elements are Figure 4. Relative abundance between living and fossil plotted in Figure 3. coccoliths during winter in the water column.

39 J. Ferreira & M. Cachão

Water column Table 3. Water samples distribution during Summer 2001 (liths/l).

The water column calcareous nannoplankton assemblages from Guadiana's river mouth are utterly diverse. The samples were collect upstream until no nannoplankton was retrieved.

Regarding the water samples collected during the Winter campaign (Nov 30th and Dec 1st, 2001) the distribution of the maximum taxa abundance is given in Table 2 and the relative abundance between living and fossil ones is shown in Figure 4.

The calcareous nannoplankton assemblages in the water column were dominated by Gephyrocapsa ericsonii (6 to 186 x 103 liths/l) and Gephyrocapsa oceanica (12 to 141 x 103 liths/l) followed by Emiliania huxleyi (49 to 94 x 103 liths/l), Gephyrocapsa muellerae (2 to 56 x 103 liths/l) and Coccolithus pelagicus (0.5 to 19 x 103 liths/l). Other assemblage components include Syracosphaera spp. (1.5 to 6 x 103 liths/l), Calcidiscus leptoporus (4 x 103 liths/l) and Helicosphaera carteri (2 x 103 liths/l). Additional minor components include Umbilicosphaera sibogae (0.7 to 0.8 x 103 liths/l) and Pontosphaera discopora (0.7 x 103 liths/l). Again, the reworked fossil, Cretaceous to Pliocene in age, forms are dominated by Cyclicargolithus floridanus (1.8 to 19 x 103 liths/l) and Dictyococcites productus (2.1 to 19 x 103 liths/l). Other assemblage components include Reticulofenestra minuta (4.7 to 10 x 103 liths/l), R. haqii-minutula (1.5 to 9 x 103 liths/l) and Sphenolithus spp. (1.5 to 9 x 103 liths/l). Additional minor components include Watznaueria barnesae (3 to 4 x 103 acknowledged, though the large majority are reworked liths/l). forms belonging to the Upper Cretaceous. Assemblages of extant coccolithophores are The water column analysis shows that during dominated by placoliths of Emiliania huxleyi (6 to 50 Winter there is a predominance of coccoliths of extant x 103 liths/l), Gephyrocapsa oceanica (22 to 38 x coccolithophores over the fossil ones. Only 103liths/l), Gephyrocapsa muellerae (20 to 26 x 103 Gephyrocapsa oceanica, G. ericsonii and G. muellerae liths/l), Calcidiscus leptoporus (11 to 26 x 103 liths/l) increase their abundance inside the estuary before and Gephyrocapsa ericsonii (9 to 26 x 103 liths/l). The demise. Diversity in upstream surface samples within reworked forms are dominated by Prediscosphaera the estuary was surprisingly reported. cretacea (16 to 142 x 103 liths/l), Biscutum ellipticum (41 to 77 x 103 liths/l), Cribrosphaerella ehrenbergii Regarding the water samples collected during the (23 to 74 x 103 liths/l), Watznaueria barnesae (13 to 74 Summer campaign (September 18th, 2001) the x 103 liths/l), Arkhangelskiella confusa (74 x 103 maximum distribution of the coccoliths assemblages is liths/l), Biscutum magnum (13 to 49 x 103 liths/l), given in Table 3 and the relative abundance between Reticulofenestra haqii-minutula (3 to 38 x 103 liths/l), living and fossil ones is shown in Figure 5. An increase Retecapsa crenulata (5 to 37 x 103 liths/l),) and both in diversity and abundance in coccoliths was Eiffellithus eximius (7 to 26 x 103 liths/l).

40 Calcareous Nannoplankton from the Guadiana Estuary and Algarve Continental Shelf (Southern Portugal): An Ecological Model

In all samples both the nannoplankton and nannofossil abundances decline as we approach the shore which is what is expected to find in this neritic domain where conditions for calcareous nannoplankton to thrive are less favourable.

Two of the most representative taxa found on surface sediments have a distribution that can be related to their optimal environmental conditions thus allowing us to classify them either as eu- or oligothrophic species.

Such a behaviour was already hypothesized by Cachão (1993) to explain the gradual decrease towards shore on relative abundance of Calcidiscus leptoporus Figure 5. Relative abundance between living and fossil over the continental shelf when compared to other coccoliths during summer in the water column. species such as Helicosphaera carteri and Coccolithus pelagicus. This pattern was also found by Cachão & Moita (2000) and Moita (2001). In fact the regularity DISCUSSION of the drop of C. leptoporus placoliths towards the coast allowed to compute a logarithmic regression The extant coccolithophore assemblages found in relating the percentage of this species to the water all samples both in the water column and surface column depth. This logarithmic regression indicates sediments follow those found by Sierro et al (1999) in that C. leptoporus cannot tolerate neritic Western deep terraces located between 400 and 700 m water Iberia shelf conditions and thus its placoliths are post- depth, just below the upper continental slope in the mortem gradually and logarithmically diluted over the Gulf of Cadiz and those found by Kenyon (2001) and shelf by coastal currents and waves (Cachão, 1993). Pinheiro et al (2003) also in the Gulf of Cadiz. In all On the other hand, other species seem to find on neritic these cases the Recent coccolith assemblages are environments good thriving conditions (e.g. mainly composed by Emiliania huxleyi¸ Gephyrocapsa Gephyrocapsa oceanica in the present work) or even spp., Coccolithus pelagicus, Calcidiscus leptoporus, become specialized (e.g. Coccolithus pelagicus, Rhabdosphaera clavigera and Syracosphaera spp. Cachão & Moita, 2000) in such environments.

In all the three transects the densities of calcareous nannoplankton are closely related with the granulometry of surface sediments. This is evidenced by the similarity of the distribution pattern of both extant and fossil coccoliths in all three transects (Fig. 3). This co-variation can only be achieved if factors such as near bottom currents influence sedimentation of present day coccoliths deriving from neritic productivity and the passive transport of reworking forms within the oceanic water masses. According to Gonzalez et al (2003), surface samples with just 10% of silty-clay fraction occur below 50 m water depth, in which there is a significant decline in the nannoplankton abundance (see Fig. 3). At this depth 8 8 densities drop from 4 x 10 to 2 x 10 liths/g or less as Figure 6. it approaches shoreline. Gephyrocapsa oceanica versus G. ericsonii densities along the continental shelf

41 J. Ferreira & M. Cachão

Different authors define distinct assemblages with The presence of such a variability and abundance neritic affinity for different coastal regions around the of reworked calcareous nannofossils is not easily world. Okada (1983) recognized Emiliania huxleyi understood if we bear in mind that the Guadiana basin and Gephyrocapsa spp. as having neritic affinity covers mainly Paleozoic metamorphic units and along the Western Pacific Ocean, but they can also be Triassic continental sediments and subvolcanic rocks found in other regions (see Brand, 1994 in Winter et (Mapa Geologico de la Peninsula Ibérica, 1981) al., 1994). In this work only Gephyrocapsa oceanica completely void of any reworked nannoliths. On the displays a distinct preferential development on other hand, coastal drift may only bring Neogene sediments from the middle shelf off the Guadiana reworked forms from the nearby littoral cliffs cut into estuary, while E. huxleyi, G. muellerae and G. the Miocene Cacela Formation, which contain Upper ericsonii as well as Coccolithus pelagicus have their Tortonian marine calcareous nannofossils (Cachão, maximum concentrations on surface sediments 1995). Other possible source for the reworking forms retrieved from the outer shelf. In fact, Okada (1983) are Lower Cretaceous outcrops although the closest referred G. oceanica as a species that may dominate ones can only be found near 30 km to the west, at surface sediment assemblages from very shallow to Olhão (Correia, 1988). semi-confined seas. NERITIC ECOLOGICAL MODEL FOR The explanation for the inverted pattern between COCCOLITHOPHORES Summer and Winter reworked and extant assemblages inside Guadiana estuary is probably due to a The behaviour of the several calcareous combination of factors that include: nannoplankton taxa was analysed across three transects. Two of the most abundant extant species i) Seasonal change in local coccolithophore (Gephyrocapsa oceanica and Gephyrocapsa ericsonii) productivity conditions; in fact, lith abundance when plotted along the continental shelf showed a changes, being three fold higher during Winter distinct behaviour with their densities shifting in than at Summer which may be related to an different directions (Fig. 6). increase in local productivity conditions in the water column, mainly affecting G. oceanica, G. Figure 6 shows that the community mainly ericsonii, G. muellerae and C. pelagicus; composed of species dwelling in oligothrophic oceanic realms declines its abundance as it approaches the coast ii) Variations in the river outflow. While during since environmental conditions have abruptly changed wintertime the increase in rainfall and continental by input of nutrient-rich river waters and upwelling runoff washes out the calcareous nannoplankton settings (Fiúza, 1983; Jordan, 2002) and they can't (extant and reworked) from the estuary, during compete against other phytoplankton groups for summertime the river's outflow decreases and a nutrients under more turbulent conditions. This group is higher number or coccoliths enters passively inside thus characterized as an oligothrophic selected the estuary. In the present case during summertime assemblage or K-selected community. (Fig. 7). the dilution affects both extant and reworked coccoliths. However, dilution was important for However certain coccolithophores increase their Winter more pronounced upstream reduction in abundance towards the shore. Because of this marine influence as shown by both extant and behaviour certain species of calcareous nannoplankton reworked coccolith absence on stations 5 and 6 have conditions to compete for nutrients in these more (see Fig. 4); euthrophic environment. For this reason we considered these as part of an euthrophic selected assemblage or iii) Changes in intensity of oceanic currents that r-selected community. (Fig. 7). redistribute reworked forms over the entire shelf. Inside the estuary reworked forms were three to "Shore" was defined as the area where seven fold higher during Summer than during nannoplankton abundance drops to zero due to high Winter. hydrodynamic near shore conditions. In these

42 Calcareous Nannoplankton from the Guadiana Estuary and Algarve Continental Shelf (Southern Portugal): An Ecological Model

environments no silty-clay particles are allowed to Observed differences between fossil and extant settle due to wave and longitudinal coastal currents. coccoliths within the water column inside the Guadiana's estuary were related to both seasonal The trend of the euthrophic community is to changes in local coccolithophore productivity abruptly drop its abundance towards the shore (Fig. 6), conditions, changes in the intensity of oceanic currents again due to the high energy gap established at this that redistribute reworked forms over the entire shelf depth which is corroborated by the sediments and variations in the river outflow between summer granulometric properties retrieved in this domain. and wintertime. However the r-community both as coccoliths as well as coccospheres have the ability to overcome this Significant changes in coccolith densities for hydrodynamic gap and reappear in paralic domains distinct coccolithophores, although modulated by before it progressively demises as it enters into fresh bottom depositional conditions (derived from water fluvial conditions. (Fig. 7). sedimentological properties) allowed to propose two basic distinct behaviours for this group within the CONCLUSIONS neritic environment: K-selected community mainly composed of species dwelling in oligothrophic oceanic Calcareous nannofossil assemblages are generally realms that can't compete against other phytoplankton abundant and well preserved in all collected samples, groups for nutrients under more turbulent conditions with profuse reworked Cretaceous to Pliocene and therefore decline its abundance as it approaches specimens. the coast; and r-selected community composed of certain species of calcareous nannoplankton with The richness of both extant and fossil calcareous conditions to compete for nutrients in these more nannoplankton on surface sediment and water column euthrophic environments. samples from the Guadiana river mouth and adjacent neritic environments demonstrate its usefulness on This ecological model is as a tool to unveil the present day ecological studies and future behaviour of these phytoplanktonic communities in palaeoecological interpretations even in coastal continental shelf samples and thus interpret their domains. record on future palaeoenvironmental studies.

Figure 7. Ecological model for coccolithophore distribution in coastal domain.

43 J. Ferreira & M. Cachão

ACKNOWLEDGEMENTS A.M., Fernandez, E., Rey, D., & Rosón, G. (Eds.) Special Volume on the 4th Symposium on the Atlantic Iberian Continental Margin, Thalassas, 19 (2a) 50-51. The authors wish to express thanks to Carlos Instituto Hidrográfico (1998). Tidal Charts 1999, 1. Marques da Silva (Fac. Sciences Univ. Lisbon) for his Jordan, R.W. (2002). Environmental Applications of Calcareous valuable suggestions and critical comments on this Nannofossils. In Haslett, S.K. (Ed.) Quaternary Environmental Micropalaeontology, Arnold, 185-206. study, and Rui Gama (Network Stratigraphic Harvest Kenyon, N.H., Ivanov, M.K., Akhmetzhanov, A.M., Akhmanov, G.G. House, UK) for his help on some taxonomic (2001). Interdisciplinary Approaches to Geoscience on the North reviewing. Catarina Corredeira (Nuclear and Tech. East Atlantic Margin and Mid-Atlantic Ridge. IOC Technical Inst.), Anabela Oliveira and Aurora Bizarro Series, 60, UNESCO, 64 pp. López-Galindo, A., Rodero, J. & Maldonado, A. (1999). Surface facies (Hidrographic Inst.) and are also acknowledge for their and sediment dispersal patterns. Southeastern Gulf of Cadiz availability and support. Spanish margin. Marine Geology, 155: 83-97. Loureiro, J.J.M., Nunes, M.N. & Machado, M. (1986). Bacia hidrográfica do rio Guadiana in Monografias hidrológicas dos This work was funded by the project CRIDA principais cursos de água de Portugal continental. M.P.A.T., (PLE/8/00) in cooperation with project CANAL S.E.A.R.N. Direcção Geral dos Recursos e Aproveitamentos (POCTI/32724/PAL/2000). Hidráulicos, 341-407. Mapa Geologico de la Peninsula Ibérica, Baleares y Canarias (1981). Instituto geologico y minero de España. Madrid. REFERENCES Moita, M.T. (2001). Estrutura, variabilidade e dinâmica do fitoplâncton na costa de Portugal continental. PhD Thesis, Univ. Abrantes, F. & Moita, T. (1999). Water column and recent sediment Lisbon, 272 pp. data on diatoms and coccolithophorids, off Portugal, confirm Nelson, C.H., Baraza, J. & Maldonado, A. (1993). Mediterranean sediment record of upwelling events. Oceanologica Acta, 22 (3): undercurrent sandy contourites, Gulf of Cadiz, Spain. 319-336. Sedimentary Geology, 82: 103-131. Baringer, M.O. & Price, J.F. (1999). A review of the physical Nelson, C.H., Baraza, J., Maldonado, A., Rodero, J., Escutía, C. & oceanography of the Mediterranean outflow. Marine Geology, Barber, Jr., J.H. (1999). Influence of the Atlantic inflow and 155: 63-82. Mediterranean outflow currents on Late Quaternary sedimentary Brand, L.E. (1994). Physiological ecology of marine coccolithophores, facies of the Gulf of Cadiz continental margin. Marine Geology, In Winter, A. & Siesser, W. (eds) Coccolithophores: 39-49. 155:99-129. Bown, P.R. (1999). Calcareous Nannofossil Biostratigraphy. British Nelson, C.H. & Maldonado, A. (1999). The Cadiz margin study off Micropalaeontological Society Series, Kluwer Academic Spain: an introduction. Marine Geology, 155: 3-8. Publishers, Cambridge: 315 pp. Ortega, J.A. & Garzón, G. (1997). Inundaciones históricas en el rio Cachão, M. (1993). Distribuição de nanoplâncton calcário em Guadiana: sus implicationes climáticas. Cuaternario Iberico, 365-167. sedimentos superficiais da plataforma continental portuguesa Perch-Nielsen, K. (1989). Cenozoic calcareous nannofossils. In H. (Dados preliminares). Gaia, 6: 1 - 19. Boli; J.B. Saunders & K. Perch-Nielsen (Eds.), Plankton Cachão, M. (1995) - Utilização de nanofósseis calcários em Stratigraphy, Cambridge University Press, vol.1: 427-554. biostratigrafia, paleoceanografia e paleoecologia: aplicações ao Pinheiro, L.M., Ivanov, M.K., Sautkin, A., Akhmanov, G., Magalhães, Neogénico do Algarve (Portugal) e do Mediterrâneo ocidental V.H., Volkonskaya, A., Monteiro, J.H., Somoza, L., Gardner, J., (ODP 653) e à problemática de Coccolithus pelagicus. PhD Hamouni, N., Cunha, M.R. (2003). Mud volcanism in the Gulf of Thesis, Fac. Sc.Univ. Lisbon, 356 pp. Cadiz: results from the TTR-10 cruise. Marine Geology, 195: Cachão, M., Oliveira, A. & Vitorino, J. (2000). Subtropical winter 131-151. guests, offshore Portugal. Journal of Nannoplankton Research, Rocha, C., Galvão, H. & Barbosa, A. (2002). Role of transient silicon 22: 19-26. limitation in the development of cyanobacteria blooms in the Cachão, M. & Moita, M.T. (2000). Coccolithus pelagicus, a Guadiana estuary, south-western Iberia. Marine Ecology productivity proxy related to moderate fronts off Western Iberia. Progress Series, 228: 35-45. Marine Micropaleontology, 39: 131-155. Sierro, F.J., Flores, J.A. & Baraza, J. (1999). Late glacial to recent Cachão, M. & Oliveira, A. (2000). (Cocco)liths versus (cocco)spheres: paleoenvironmental changes in the Gulf of Cadiz and formation approaching the ecological performance of coccolithophores. of sandy contourite layers. Marine Geology, 155: 157-172. Journal of Nannoplankton Research, 22: 29-34. Sikes, C.S., Wierzbicki, A. & Fabry, V.J. (1994). From atomic to global Correia, F. (1988). Estudo biostratigráfico e microfácies do Cretácico scales in biomineralization. Bulletin de l'Institut carbonatado da Bacia Sedimentar meridional Portuguêsa Oceanographique (Monaco) (Sp. Issue), 14: 1-47. (Algarve). PhD thesis Fac. Sc. Univ. Lisbon, 377 pp. Westbroek, P., Brown, C.W., Bleijswijkbrownlee, J.C., Brummer, G.J., Fiúza, A. (1983). Upwelling patterns of Portugal. In Suess, E., Thiede, Conte, M., Egge, J., Farnandez, E., Jordan, R., Knappertsbusch, M., J. (Eds.) Coastal Upwelling, its Sedimentary Record. Part A. Stefels, J., Veldhuis, M., Van der Wal, P. & Young, J. (1993). A model Responses of the Sedimentary Regime to present Coast system approach to biological climate forcing: the exemple of Upwelling, Plenum Press, New York, 85-98. Emiliania huxleyi. Global and Planetary Change, 8: 27-46. Gonzalez, R., Dias, J.M.A., Lobo, F., Mendes, I. & Plaza, F. (2003). The sedimentary shelf cover between the Ria Formosa and the mouth of the Guadalquivir Estuary (northern Gulf of Cadiz, SW Iberia). In Vilas, F., Rubio, B., Diez, J.B., Francés, G., Bernabeu, (Received: January, 7, 2004. Accepted: November, 25, 2004)

44 Thalassas, 2005, 21 (1): 45-52 An International Journal of Marine Sciences

SPATIAL DISTRIBUTION AND COMPOSITION OF SUSPENDED SEDIMENTS IN RIA DE AVEIRO LAGOON

I. ABRANTES (1), F. ROCHA (2) & J. A. DIAS (3)

Keywords: Suspended sediments; Ria de Aveiro lagoon; mineralogical composition

ABSTRACT all stations, higher mean values of suspended sediments concentration and of quartz, Opal C/CT and During four tidal cycles concentration and pyrite occur during spring tides due to the intensity of mineralogical composition of suspended sediments the tidal currents. During winter neap tide the higher were studied in the Ria de Aveiro coastal lagoon. mean values of suspended sediments concentration Temporal and spatial variations were found and seem and of quartz and kaolinite observed in stations close to be mainly controlled by tidal currents magnitude to the tidal inlet suggest that wind-induced and bottom properties. The highest mean value of resuspension and horizontal advection from the suspended sediments (24 mgL-1) was found at station 3 adjacent coastal area is taking place. Semidiurnal that combines a zone relatively deeper with a bed variations are explained, especially during the summer, locally rich in fine-grained sediments. Besides, water by tidal-current velocity asymmetry (ebb dominance). and suspended matter from northern side of S.Jacinto Seasonal variability, with winter higher suspended Channel and from Laranjo Bay converge in this area. sediments concentration and higher values of quartz, The lowest suspended sediment mean value (13.2 kaolinite and calcite, is probably related with wind mgL-1) was observed in station 7, which is also a climate, biological activity and coastal wave regime. deeper zone, but with a sandy bed. In general, the mineralogical suite identified is dominated by INTRODUCTION mica/illite (mean = 45.1%) followed by quartz (mean = 19.9%) and by kaolinite (mean = 12.7%). In almost The Ria de Aveiro lagoon, located between 40º52' and 40º30' N, is the major portuguese lagunar system, being the main part of a wet system with a total surface 2 (1) E.S.E./Instituto Superior Politécnico de Viseu, of 250 km . Connected to the Atlantic Ocean by an R. Dr Maximiano Aragão, 3500 Viseu, Portugal, artificial channel (1.3 km long, 350 m wide and 20 m [email protected] deep),this lagoon has a irregular and complex (2) Centro de Minerais Industriais e Argilas, Univ. Aveiro, 3810 geometry characterized by extensive intertidal zones Aveiro, Portugal, [email protected] (3) Univ. Algarve/ FCMA, Campus de Gambelas, 8000 Faro, and four main channels: S. Jacinto, Ílhavo, Mira and Portugal, [email protected] Espinheiro (Figure 1).

45 I. Abrantes, F. Rocha & J. A. Dias

136.7 Mm3 for maximum spring tide and 34.9 Mm3 for minimum neap tide (Dias et al., 2000). Consequently, Ria de Aveiro hydrodynamics is essentially dominated by tidal forcing.

The tides at the mouth are predominantly semidiurnal and are present in the entire lagoon (Dias et al., 1999; Dias, 2001).Tidal currents are strongly dependent on the local geometry. Increasing in narrow and deepest channels (beginning of S. Jacinto and Espinheiro channels), where can be higher than 1 ms-1, never exceeds 0.4 ms-1 in the shallow sections of the channels. Maximum current speed are found three hours after high and low tide, being the ebb tide currents slightly higher than the flood ones (Dias et al., 1998).

Figure 1. Satellite image of Ria de Aveiro (NASA, 2004). Sediments transport model developed by Dias (2001) point out to residence times less than 2 days, at The average depth in the lagoon is about 1 m, with central sector of the lagoon, indicating high water exception for the navigation channels where dredging renovation, due to a strong marine influence. operations maintain a depth of about 7 m (Dias et al., 2000). Investigations about suspended sediments concentrations in Ria de Aveiro have been done in the The Ria de Aveiro lagoon is geologically very north western channels and show semidiurnal young and its origin is related to the southward variations (Silva, 1994; Pereira, 1995). A two- transport of sediments by alongshore currents. From centuries X to XVIII this transport of sediments along the Portuguese west coast originated a long spit that isolated the estuary of Vouga river from the coastal ocean (Martins, 1947; Girão, 1951; Abecassis, 1955).

In the Ria de Aveiro channels, the surface sediments are a combination of medium to fine sands with a variable content of finer particles (silt and clay), which increases with the distance from the lagoon mouth. The inner zone of intertidal flats integrates, from top to the bottom, sand, mixed and lutitic flats that are mainly composed, respectively, by medium to fine sand, clay silty and clay sandy sediments (Rocha et al., 2000).

This coastal lagoon corresponds to a mesotidal estuary (Davies, 1964) with a tidal range of 3.2 m for maximum spring tide and of 0.6 m for minimum neap tide (Dias et al., 2000). Two rivers Vouga and Antuã constitute the major sources of freshwater. The total mean freshwater input during a tidal cycle is about 1.8 Figure 2. Ria the Aveiro with sampling stations and plots of suspended sediment mean 3 Mm (Moreira et al., 1993), while the tidal prism is concentration (in the legend, between parentheses is indicated the station number).

46 Spatial Distribution and Composition of Suspended Sediments in Ria de Aveiro Lagoon

dimensional depth-integrated transport model for For the semi-quantitative determination of clay cohesive suspended sediments has also been applied to and non-clay minerals, the relative content of each simulate tidal evolution of suspended sediments in the identified mineral was estimated on the basis of its lagoon by Lopes et al. (2000) and indicated that characteristic peak area corrected by the corresponding sediments concentrations show spatial, semidiurnal reflective power (Rocha, 1993), as recommended by and fortnightly variability. The study of suspended Barahona (1974), Schultz (1964), Thorez (1976), sediments mineralogy has been done by Gomes (1987) Mellinger (1979) and Pevear & Mumpton (1989). but just on rivers discharging in Ria de Aveiro. Therefore, there was no previous research about RESULTS AND DISCUSSION suspended sediments mineralogy in Ria de Aveiro channels. Concentration

The purpose of this work is to contribute to the During the study period, the concentration of characterization of suspended sediments distribution suspended sediments showed a complex spatial and and composition in the central area of the Ria de temporal pattern, depending on the tidal phase and Aveiro lagoon and to infer the sedimentary exchanges amplitude and on the season (Tables 1 and 2, Figure 2). between ocean and the lagoon. The differences in suspended sediments MATERIALS AND METHODS concentration within the lagoon seem to be largely determined by the magnitude of the currents and by the Suspended sediments were collected in seven sites bottom properties. In fact, the highest mean value of (Figure 2) located in the four main channels (1 - mouth suspended sediments (24 mgL-1) was found at station 3 of lagoon; 2 and 3 - S. Jacinto Channel; 4 - Mira which is located in a zone relatively deeper of the S. Channel; 5 and 7 - Espinheiro Channel and 6 - Ílhavo Jacinto Channel where the bed is locally rich in fine- Channel). grained sediments. Water and suspended matter from northern side of S. Jacinto Channel and from Laranjo Sampling was performed during one summer tidal Bay converge also in this area. The lowest suspended cycle (September 2001) and one winter tidal cycle sediment mean value (13.2 mgL-1) was observed in (February 2002), at approximately 2-hour intervals, station 7, which is also a deeper zone, but with a sandy including high tide and low tide measurements. The bed. The proximity of this station to the Vouga river surveys cover a range of tidal heights at the lagoon mouth seems to reflect the weak contribution of this mouth from 1.2 m to 2.9 m in summer and 1.1 m to 2.2 river in suspended sediments, comparatively to the m in winter. Water samples were collected from the apparent supply from bottom and shore estuary surface, middle-depth (when depth is higher than 3 m) erosion, during the study period. and 1 m above the bottom at each site using a Van Dorn® bottle. In general terms, the higher concentrations of suspended sediments were found close to the bottom To determine the concentration of suspended denoting resuspension and deposition processes matter, between 1 and 3 L aliquots were filtered with (Tables 1 and 2, Figure 3). the classic vacuum system using 0.45 µm Millipore® (47 mm diameter) pre-weighted filters. The filters The higher values were observed during the ebb were dried at 40 ºC for 24 h and reweighed. (Tables 1 and 2) and could be explained mainly by the asymmetry in the current velocity over the tidal cycle. Mineralogical studies were based on X-ray diffraction (XRD) determinations, using a Phillips PW Fluctuations on suspended sediments concentration 3040/60 diffractometer. All samples were analysed in are a function of current velocity, but advection of the range from 2º to 40º 2 θ, at 1° 2 θ /min, with Cu- sediments resuspended in remote areas and the time Ka radiation. The XRD reflections were evaluated taken for the particles to disperse throughout the water with the Phillips X'Pert 1.2 and Profit softwares. column after they have been eroded (scour lag) may

47 I. Abrantes, F. Rocha & J. A. Dias

Table 1. Flood and ebb averaged suspended sediment concentration (mgL-1) 1 m from the bottom for the summer and winter surveys. Minimum and maximum values observed in each situation are indicated between right parentheses.

Table 2. Flood and ebb averaged suspended sediment concentration (mgL-1) near surface for the summer and winter surveys. Minimum and maximum values observed in each situation are indicated between right parentheses.

also account. On the other hand, resuspension may not coastal wave regime than with rivers input, which were be related with current velocity but with wind-induced below the mean flow during the surveys. turbulence, which presumably was out of phase with tidal currents. The relative importance of each process is In the winter neap tide winds blow essentially from difficult to evaluate due to the complexity of the north with intensities generally higher than 3 ms-1. It is processes and to the lack of observations and of possible that wind-induced turbulence eroded the hydrological measurements concurrent with the water bottom mud deposits in the large shallow areas, as well sampling campaigns. as the finer sediments in the intertidal zones. In this case, resuspend particles are subsequently advected Nevertheless, every time currents intensity through the estuary. increases, what happens generally during spring tides, remobilization and transport of sediments become The seasonal variations of SSC can also be more efficient, therefore higher suspended sediments influenced by biological processes. The examination concentration were observed (Figure 4). under binocular microscope of the suspended sediments retained in the filters showed that biogenic Seasonal variability, with winter higher SSC in material (predominantly plant fragments, copepods almost all stations (Tables 1 and 2), was probably more and diatoms) and aggregates are more abundant in the related with wind climate, biological activity and bottom samples during the summer than in the winter.

48 Spatial Distribution and Composition of Suspended Sediments in Ria de Aveiro Lagoon

Figure 3. Figure 4. Suspended sediments mean concentration in surface and near the bottom. Suspended sediment mean concentration (S S T- Summer spring tide; S N T- Summer neap tide; W S T- Winter spring tide; W NT- Winter neap tide).

If flocculation owing to organic processes is surrounding region of the Aveiro lagoon outcrop important, fine-grained sediments are more easily Paleozoic and Proterozoic gneisses, migmatites, kept in suspension in the winter situation than in the granitoids, micaschists and schists, northwards, as well summer, because flocculation is less important in the as Cretaceous sandstones and shales, southwards. first case. Actually, the identified mineralogical suites are clearly similar to the fine fractions (<63µm) mineralogy In winter highly coastal wave energetic conditions identified in the Vouga watershed soils (Pereira, 1989) were observed. Measurements from the wave buoy and in sediments and metasediments from Aveiro located in Leixões (northern of Ria de Aveiro) region (Delgado et al., 1992; Rocha, 1993; Rocha et indicated that significant wave heights and periods al., 2000; Chaminé et al., 2001, 2002). exceeded 1.5 m (2.5-3.6 m in neap tide) and 7 s, respectively. These conditions probably induced high The central sector of the lagoon, being the most resuspension of the coastal particles, especially during influenced by the sea waters influx-reflux and neap tides. Consequently, more sediment could be characterized by higher hydrodynamism, shows the supplied to the lagoon system by the flood tidal vanishing of the mineralogical signatures of the detrital currents. inputs coming from north, east (from the Vouga watershed soils) and south Mineralogical Composition Actually, the mineralogical suites are similar, The mineralogical suite identified in the suspended along the tide cycles, at any sampling stations. sediments is, for all sampling sites, dominated by Nevertheless, the results (Figure 5) show some discrete mica/illite (mean = 45.1%) followed by quartz (mean variations, sometimes with seasonal character, of the = 19.9%) and by kaolinite (mean = 12.7%). K- relative abundance of each one of the identified feldspars, plagioclases, calcite and opal C/CT are minerals. common as accessory minerals. Less common and always in very small quantities, pyrite, anhydrite, The seasonal analysis of the suspended matter dolomite, siderite, amphiboles, zeolites, goethite and mineralogical composition (Figure 5) point out to a gibbsite were also identified. discrete increase of quartz and kaolinite during winter. Nevertheless, this trend is not shown for quartz from The mineralogical composition of the studied stations 6 (ebb tides) and 7 (always). samples shows the "signature" of the mineralogical composition of the soils and rocks outcropping in the During summer, mica/illite and chlorite show, for hydrographic basins being drained to the lagoon. In the the majority of the stations, a relative increase during

49 I. Abrantes, F. Rocha & J. A. Dias

Figure 5. Box-Plot of contents of some minerals present in suspended sediments (f- flood; e- ebb; SD- standard desviation).

50 Spatial Distribution and Composition of Suspended Sediments in Ria de Aveiro Lagoon

flood tides; however, relative abundance of quartz and CONCLUSIONS kaolinite increase during ebb fluxes, as a consequence of tidal currents velocities effects on resuspension, The sedimentary dynamics of the Aveiro lagoon is transport and deposition of particles. This behaviour essentially dominated by tidal forcing. As in other becomes more complex during winter, some stations estuarine systems, the suspended sediments showing a trend exactly the opposite of the one showed concentration fluctuated with tidal amplitude and during summer. phase and show seasonal variability.

Among all the identified minerals, calcite is the The mineralogical composition of the suspended one showing a distribution more homogeneous along sediments is a function of the fluvial input, bottom the sampling periods, being its higher concentration composition and oceanic contribution. During the values related to flood tides. Exceptions were stations study period suspended sediment composition show 3 and 5, during spring tides and station 2 during winter temporal and spatial variability. Calcite plays a role as neap tide. proxy to the ocean biogenic contribution.

The increase of quartz and kaolinite identified Exchanges of mica/illite, quartz and kaolinite during winter seems to be explained by the input of between the lagoon system and the coastal ocean oceanic particles into the lagoon, particularly during occurred during the study period. spring tides. In fact, these two minerals occur abundantly in the bottom sediments of the adjacent AKNOWLEGEMENTS continental shelf (Abrantes et al. 2000) as well as, northwards (up to Douro and Minho), in the The authors would like to thank the Departments suspended sediments of the NW Iberian Margin of Biology and Physics of Aveiro University (for using (Oliveira, 2001; Oliveira et al., 2002). However, it their boats and equipment), Prof. João Dias (from must be taken into account that this increase is Physics Department of Aveiro University), Mr. Rui sometimes related to the ebb fluxes, therefore being Marques (from Biology Department of Aveiro someway dependent of erosion of the bottom University) and the colleagues of the Geosciences sediments. Department who contributed for the fieldwork. The first author also thanks the Ministério da Ciência e do Along the tidal cycle, the compositional variability Ensino Superior for the PhD Grant. shown by the studied samples from station 1 (located close to mouth) point out to suspended sediments REFERENCES exchange between the lagoon and the ocean, with Abecassis, C. (1955) - The history of a tidal lagoon and its seasonal character. improvement (The case of Aveiro, Portugal). Proceedings of Fifth Conference on Coastal Engineering, pp. 329-363. Mica/illite denotes (Figure 5) an output trend Abrantes I., Rocha F., Vidinha J. & Dias J. A. (2000) - Fracções finas (from the lagoon to the shelf) during the winter spring dos sedimentos da plataforma e vertente continental superior entre Espinho e Aveiro-resultados preliminares. 3º Simp. Marg. tides, whereas quartz and kaolinite show an opposite Cont. Ibérica Atlântica, Faro, 405-406. tendency, being transported from the ocean to the Barahona, E. (1974) - Arcillas de ladrillería de la provincia de lagoon during this period. On the other hand, during Granada: evaluación de algunos ensayos de materias primas. Ph.D. Thesis, Granada Univ., Spain, 398pp. summer tides, mica/illite shows an increase along the Chaminé, H.; Rocha, F.; Gomes, C. & Moço, L.P. (2001) - Overview floods, being quartz and kaolinite exported to the based on clay mineralogy and organic metamorphism data of shelf. midlle Palaeozoic black shales from Albergaria-a-Velha - Coimbra region (Porto - Tomar shear zone, NW Portugal). Bol. Soc. Esp. Mineralogia, 24-A, pp. 53-54. Calcite plays a role as Proxy to the ocean biogenic Chaminé, H.; Rocha, F.; Moço, L.P.; Fernandes, J.P.; Flores, D.; contribution, being related to the flood phases, and Gomes, C.; Lemos de Sousa, M.J.; Gama Pereira, L.C. & showing (Figure 4) its higher concentration values Fonseca, P.E. (2002) - Middle and upper Palaeozoic basins from Estarreja - Coimbra - Tomar region (Porto-Tomar-Ferreira do close to the mouth, decreasing inside the lagoon with Alentejo shear zone, W Portugal): a clay mineralogy, organic the increase of distance from the mouth. metamorphism, palynology and tectonostratigraphy review.

51 I. Abrantes, F. Rocha & J. A. Dias

Strati'2002 - 3eme Congrès Français de Stratigraphie, C. Gaillard Mellinger, R.M. (1979) - Quantitative X-ray diffraction analysis of & P. Hantzperque (eds.), Docum. Labor. Géol. Lyon, 156, pp. 68- clay minerals. An evaluation. Saskatchewan Research Council, 69. Canada, SRC Report G-79: 1-46 Davies, J. H. (1964) - A morphogenetic approach to world shorelines. Moreira M. H., Queiroga H., Machado M. M., Cunha M. R.(1993). Zeit. Geomorphol., 8: pp 27-42. Environmental gradients in a southern estuarine system: Ria de Delgado, H.; Rocha, F. & Gomes, C. (1992) - Evolution of the Aveiro Aveiro, Portugal, implication for soft bottom macrofauna lagoon during the last 500 years based on clay mineralogy. Min. colonization. Neth. J. Aquat. Ecol., 27 (2-4): 465-482. et Petr. Acta, Vol. XXXV-A: 105-110. NASA (2004) - Earth Science Applications Directorate. Adress: Dias, J.M. (2001) - Contribution to the study of the Ria de Aveiro http://zulu.nasa.gov/mrsid/mrsi.pl. hydrodynamics. Portugal: University of Aveiro, Ph.D. thesis, Oliveira A. T. C.(2001) - Dinâmica da matéria particulada em 288p. suspensão na plataforma continental minhota e sua relação com Dias, J.M.A.; Pilkey, O.H. & Heilweil, V. M. (1984) - Detrital mica: a cobertura sedimentar. Ph.D. thesis, Univ. Algarve, 278pp. environmental significance in North Portugal continental shelf Oliveira, A.; Rocha, F.; Rodrigues, A.; Jouanneau, J.; Dias, J.A.; sediments. Comunicações dos Serviços Geológicos de Portugal, Weber, O. & Gomes, C. (2002) - Clay minerals of the 70 (1), pp. 93-101. sedimentary cover from the Northwestern Iberian shelf. Progress Dias, J.M.; Lopes, J.F. & Dekeyser, I. (1998) - An exploratory study of in Oceanography, Elsevier, 52(2-4), pp. 233-247. the dynamics of Ria de Aveiro, Portugal. In: KIM, H.; LEE, S.H. Pereira, V. (1989) - Contribuição para o conhecimento dos and LEE, S.J. (eds), Hydrodynamics - Theory and Applications. Cambissolos húmicos do Médio Vouga. Geociências, Rev. Univ. Seoul, Uiam Publishers, pp. 619-624. Aveiro, Vo. 4, Fasc. 1, 75-86. Dias, J.M.; Lopes, J.F. & Dekeyser, I. (1999) - Hydrological Pevear, D.R. & Mumpton, F.A. (1989) - Quantitative mineral analysis characterization of the Ria de Aveiro, Portugal in early Summer. of clays. CMS Workshop Lectures, 1. The Clay minerals Society, Oceanologica Acta, 22, 473-495. Colorado (USA). Dias, J.M.; Lopes, J.F. & Dekeyser, I. (2000) - Tidal propagation in the Rocha, F. (1993) - Argilas aplicadas a estudos litoestratigráficos e Aveiro lagoon, Portugal. Physics and Chemistry of the Earth (B), paleoambientais na Bacia Sedimentar de Aveiro. Ph.D. thesis, 25(4), 369-374. Aveiro University, Aveiro, 399 pp. Dias, J.M.; Lopes, J.F. & Dekeyser, I. (2003) - A numerical model Rocha, F.; Bernardes, C. & Delgado, H. (2000) - Caracterização system application to the study of the transport properties in Ria textural e mineralógica dos sedimentos da laguna de Aveiro, de Aveiro lagoon. Ocean Dynamics, 53, 220-231. Portugal. Abstr. "3º Simp. Margem Contin. Atlant. Ibérica" (Univ. Girão, A.A. (1951) - Evolução morfológica da região do Baixo Vouga. Algarve, Faro): 17-18. Boletim do Centro de Estudos Geográficos de Coimbra, 2/3, pp. Schultz L. G. (1964) - Quantitative interpretation of mineralogical 75-85. composition from X-ray and chemical data for Pierre shale. U.S. Gomes, C.S.F. (1987) - Minerais da carga sólida em suspensão de rios Geol. Surv. Prof. Paper, 391-C: 1-31. e ribeiros que afluem à Ria de Aveiro. Geociências, Rev. Univ. Aveiro, vol. 2, fasc. 1-2, pp. 41-45 Martins, A.F. (1947) - A configuração do litoral português no último quartel do século XIV. Biblos, XXII(I), pp. 163-197. (Received: March, 11, 2004. Accepted: December, 16, 2004)

52 Thalassas, 2005, 21 (1): 53-58 An International Journal of Marine Sciences

PALAEOENVIRONMENTAL ANALYSIS OF DOURO ESTUARY BASED ON MINERALOGICAL PARAMETERS

F. ROCHA(1), T. DRAGO(2), F. NAUGHTON(2) & T. SILVEIRA(3)

Keywords: Palaeoenvironmental analysis; Douro estuary; Mineralogical parameters

ABSTRACT the sedimentology, carbonates, organic matter, mineralogy, foraminifera, nannoplankton and pollen Sedimentary studies in the Portuguese continental assemblages have permitted to recognise the shelf have been mainly developed after the eighties, geological evolution of the area, including the shift of involving different interdisciplinary approaches, fluvial to marine facies. The aim of this work is among them sedimentology, geochemistry and therefore, to present the mineralogical evolution mineralogy. Most of those studies have been focused determined in the sedimentary record between 19.5 m on the northwestern Portuguese shelf, particularly in and 44 m depth. Results are discussed in order to sedimentary deposits extending at the adjacent shelf characterise the sediment distribution patterns and to northwards of Douro river, but they have been recognise and evaluate some temporal changes. essentially based on the study of superficial sediments or in recently deposited sediments. In this context, INTRODUCTION studies of stratigraphic succession of lagoonal and estuarine paleoenvironments may give helpful The terrigenous sedimentary particles contribute, information. A study on the stratigraphic succession of for their characteristics, to the identification of the sedimentary record of Douro estuary is being alluvium deposits sources and of paleogeographic carried out aiming at the recognition and distinction of terms, assuming that they are not the result of the environmental changes, which have occurred successive reworking cycles, nor remarkably modified during the Late Quaternary. Some results obtained on by diagenesis.

The utilization of clay as a stratigraphic marker constitutes one of the first essays of its application in (1) Centro de Minerais Industriais e Argilas (FCT), the sedimentary basins and continues, actually, to be an Univ. Aveiro, Campus de Santiago, 3810 Aveiro, Portugal, important research area. +351.234370200 (2) INIAP, IPIMAR - Inst. de Invest. das Pescas e do Mar - Av. The clay associations' mineral analysis provides 5 de Outubro 8700-305 Olhão, Portugal, +351.281326951, (3) Laboratório Marítimo da Guia - FCUL, Estrada do Guincho, important information on paleoclimatic studies 2750-642 Cascais, Portugal, + 351.214869211 essentially concerning regional data. Paleoclimatic

53 F. Rocha, T. Drago, F. Naughton & T. Silveira

A study on the stratigraphic succession of the sedimentary record of Douro estuary is being carried out aiming at the recognition and distinction of the environmental changes, which have occurred during the Late Quaternary. Some results obtained on the sedimentology, carbonates, organic matter, mineralogy, foraminifera, nannoplankton and pollen assemblages have permitted to recognise the geological evolution of the area, including the shift of fluvial to marine facies (Drago et al., 2002).

The aim of this work is therefore, to present the mineralogical evolution determined in the sedimentary record between 19.5 m and 44 m depth. Results are Figure 1. Core location discussed in order to characterise the sediment distribution patterns and to recognise and evaluate some temporal changes. signal doesn't depend significantly of biotic factors and frequently survives the post-depositional modifications. MATERIALS AND METHODS The composition of detrital mineral associations is particularly useful in the way that they record the The studied core was sampled by rotary drilling in modifications that occur in source area, in what the sand barrier of the Douro estuary (Figure 1) from - concerns the erosion conditions. 20 to -40 m depth. After being described, core was sub- sampled for multidisciplinary investigations, including Clay and non-clay minerals assemblages of grain-size, chemical and mineralogical analysis. Nine superficial sediments have been used in the recognition organic sediment samples were dated by 14C AMS. of sources since the sixties. Changes in sources or in current activity often determine modifications of the Textural analyses have been performed by means mineral assemblages. Therefore, these assemblages are of the traditional sieving method. Gravel free useful indicators of detrital sources. On the other hand, sediments follow the Flemming (2000) classification these mineral associations, supplied from land to sea, (considering only the < 63µm and > 63µm fractions often reflect the combined control of terrigenous percentages). For sediments with variable gravel supply and transportation agents and can be used to percentage, the Hydrographic Institute classification characterize recent estuarine environments. was adopted (Moita, 1986).

Sedimentary studies in the Portuguese continental For mineralogical analysis of both fine (<63 µm, shelf have been mainly developed after the work by 90 samples) and clay (<2 µm, 43 selected samples) Dias (1987), involving different interdisciplinary fractions, each sample was dried, washed with distilled approaches, among them sedimentology, geochemistry water and wet sieved to separate the clay/silt size and mineralogy. Most of those studies have been fraction. Sediment fractions <63 m were dried in an focused on the northwestern Portuguese shelf, oven at 60ºC and gently disaggregated with a porcelain particularly in sedimentary deposits extending at the mortar. X- Ray Diffraction (XRD) was used following adjacent shelf northwards of Douro river, but they have the methodology proposed by Rocha (1993). For the been essentially based on the study of superficial semiquantitative determination of minerals by XRD, in sediments or in recently deposited sediments (Araújo random-oriented powders (<63 m) and in oriented et al., 2002; Drago et al.,1999; Oliveira et al., 2002). aggregates (<2 m), criteria recommended by Barahona In this context, studies of stratigraphic succession of (1974), Schultz (1964), Thorez (1976), Mellinger lagoonal and estuarine paleoenvironments may give (1979) and Pevear and Mumpton (1989) were helpful information (as shown by Freitas et al., 2002). followed.

54 Palaeoenvironmental Analysis of Douro Estuary based on Mineralogical Parameters

Figure 2. Defined litostratigraphic units (SED1, SED2, SED3, SED4) with dating information (Drago et al., 2002). Vertical distribution of the fine fraction mineralogical composition (a) and vertical evolution of DF/DG and C/D ratio (b).

Two mineralogical indexes were computed Results from the mineralogical studies allowed the (according to Vidinha et al., 1998 a, b, 2000): characterization, along the studied core, of the vertical phyllosilicates/(quartz+feldspars) - DF/DG and evolution of the mineral composition of the fine carbonates/detrital minerals - C/D for each sample. fractions (Figure 2a) and clay fractions (Figure 3) of Phy/Qz+Feld ratio is related with the transport of the sediments as well as of the mineralogical ratios detrital materials from the hydrographic basins of the phyllosilicates/ (quartz+feldspars) - DF/DG, and nearby continental region (the higher values indicating carbonates/detrital minerals - C/D (Figure 2b). the lower hydrodynamism of the transportation agent). C/D ratio expresses the dichotomy between the detrital This characterization allowed to assess a transport and the biogenic component, the higher zonography of the core's sedimentary column, similar values indicating the lower contribution of detrital to the informal units defined by Drago et al. (2002) - materials and the higher contribution of biogenic SED1, SED2, SED3 and SED4, as well as to attempt a materials. morphoclimatic reconstruction.

RESULTS AND DISCUSSION Therefore, it was possible to distinguish, from base to top, the following units (upper and/or lower limits Sedimentary, geochemistry and palaeoecological reported to sea-level): studies of this corer reveal the existence of four sedimentary units referred by (DRAGO et al., 2002 ) SED 1 is the lower unit (13730-10310 BP) and as SED1, SED2, SED3 and SED4 (Figure 2a). consists of slightly muddy sand with some gravely

55 F. Rocha, T. Drago, F. Naughton & T. Silveira

Figure 3. Defined litostratigraphic units (SED1, SED2, SED3, SED4) with dating information (Drago et al., 2002) and vertical distribution of the clay fraction mineralogical composition. sand and muddy sand layers at the bottom; the vertical characterized by a clear increase of quartz (and, much evolution of the studied mineralogical parameters more discrete of the phyllosilicates) and decrease of allowed its subdivision into two sub-units, SED 1a and feldspars, as well as, in the clay fractions, by a 1b. SED 1a, from the core's base until -37.66 m, decrease of kaolinite, illite becoming clearly the consisting of slightly muddy sand with gravely sand predominant clay mineral whereas chlorite becomes and muddy sand layer, is characterized by a the third clay mineral, Smectite being always very mineralogical suite composed essentially by relatively discrete; these mineralogical changes point out to a coarse-grained siliciclastic minerals (quartz and morphoclimatic evolution towards milder conditions, feldspars), with a very discrete presence of although still continental ones. phyllossilicates and carbonates; clay minerals' suites are composed by kaolinite (Kt) and illite (Ill), with Mineralogical unit SED 2 begins a little earlier (- some smectite (Sm) and chlorite (Chl), these two clay 34.07 m) than sedimentological SED2 unit (-32.33 m). minerals alternating as the third more abundant clay It consists mainly of thin interbedded layers of sandy mineral of this sub-unit; these mineralogical mud and muddy sand, followed by a slightly sandy associations point out to a continental environment mud section and muddy sand sections; like SED 1, it with temperate to subtropical climatic conditions, with can be subdivided into two sub-units. SED 2a, until - milder periods (concerning both temperature and 31,16 m, consisting mainly of muddy sand and sandy pluviosity), favorable to heavy detrital supply (intense mud thin layers, is characterized by a higher relative fluvial input). SED 1b, overlying SED 1a until -34.07 increase of quartz caused by the decrease of m, consisting mainly of slightly muddy sand, is phyllosilicates and feldspars, whereas in the clay

56 Palaeoenvironmental Analysis of Douro Estuary based on Mineralogical Parameters

fractions, kaolinite decreases again in relation to illite CONCLUSIONS (in general, very degraded), and smectite becomes the third clay mineral, chlorite being rare and even absent The results obtained so far allowed to perform the in some samples; climatic conditions should be mineralogical characterization of the fine and clay temperate (but with colder periods), relatively more fractions of the sediments from the studied core as well wet and, from the morphologic point of view, more as the morphoclimatic reconstruction of the Douro plane, presenting also the first periods of marine Estuary. influence. SED 2b, until -20.66 m, consisting of slightly sandy mud, sandy mud and muddy sand is Climatic conditions should have changed from marked by a gradual increase of feldspars and, in the temperate/subtropical, pluvious, to milder ones, more clay fractions, of kaolinite relatively to illite (now temperated, whereas, from the morphologic point of more ordered), smectite decreasing relatively to view, the evolution should have been similar, mainly chlorite; therefore, climate should have changed to less continental, changing from more to less hydrodynamic wet conditions, less favourable to hydrolysis fluvial conditions, with a few periods showing some development in the source-areas, implying that the marine influence. Evidences of hotter period followed deposition environment would be more marine. This by a cold one were found and it may correspond to the upper half of SED2 although showing no significant Bölling-Alleröd interstadial and the Younger Dryas textural contrast with the lower part of the same unit, stadial. The transition to the Holocene is well presents compositional proxies indicating an represented by the climatic amelioration (circa alternation of marine/continental facies until 6050BP. 10310BP) and the appearance of the marine influence The terrestrial facies may represent regressive (9834BP). episodes, which are more probably due to large fluvial sediment input than to eustatic sea level fall. A discrete SED 3 probably represents an ancient coarser increase in carbonates indicates a first marine barrier, established in relation to the deceleration of sea incursion episode between 9834 and 9756BP. The level rise, coupled with favourable climate conditions. gradual increase of the C/D (carbonate/detrital) ratio in This unit may have constituted the foundation for the the fine fraction points out to an increase in marine development of the present-day sandy barrier (SED 4) influence noticed after 9230BP and becoming more after 1580 BP. In the sheltered back-barrier area pronounced between 6050 BP and 5750 BP, when the alternating sandy and muddy layers have been transgressive maximum occurred. deposited in this time interval and the presence of sparse foraminifera and coccoliths (only at -1,69m), SED 3 is a gravel unit, deposited after 5750 BP. together with other paleoenvironmental indicators Overlying SED 2b until -19.4 m is characterized by suggest a brackish environment with low salinity, that notorious increase of quartz (up to 100%), whereas in rapidly acquired fluvial and/or subaereal the clay fractions only a discrete decrease of chlorite is characteristics towards the top. detected; morphoclimatic conditions changed, becoming more continental and wet favourable to On top of the studied column we found heavy detrital supply (very intense fluvial input). evidences of less wet conditions, still temperate, but with some hotter and wet periods, season- SED 4 is the upper unit of the studied sedimentary contrasted, within an environment becoming again sequence; it is composed by sand with few interbedded continental. gravely sand. It shows an increase of feldspars and carbonates and a decrease of phyllossilicates; ACKNOWLEDGEMENTS concerning the clay fractions, kaolinite decreases relatively to illite whereas smectite increases in This work is being funded by FCT under the relation to chlorite; climate should be season- research project PLE/12/00 "Late Quaternary contrasting with periods less wet, temperated, ENVIronmental CHANGES from Estuarine and alternating with others more hot and wet, erosive, Continental Shelf Sedimentary Record" (ENVI- clearly continental. CHANGES).

57 F. Rocha, T. Drago, F. Naughton & T. Silveira

REFERENCES Rocha, F. (1993) - Argilas Aplicadas a Estudos Litoestratigráficos e Paleoambientais na Bacia Sedimentar de Aveiro. PhD Thesis, Univ. Aveiro, 399pp. (unpublished). Araújo, M.F.; Jouanneau, J.-M.; Valério, P.; Barbosa, T.; Gouveia, A.; Rocha, F.; Drago, T.; Naughton, F. & Silveira, T. (2003) - Ensaio de Weber, O.; Oliveira, A.; Rodrigues, A. & Dias, J.M.A. (2002). reconstituição paleoambiental do estuário do Douro, com base Geochemical Tracers of Northern Portuguese Estuarine em parâmetros mineralógicos. Thalassas, 19(2b): 175-176. Sediments on the Shelf. Progress in Oceanography, 52 (2-4): Salomons, W. & Förstner, U. (1984) - Metals in the Hydrocycle, 277-297. Springer-Verlag, Berlin, Heidelberg. Barahona, E. (1974) - Arcillas de ladrillería de la provincia de Schultz, L. G. (1964) - Quantitative interpretation of mineralogical Granada: evaluación de algunos ensayos de materias primas. composition from X-ray and chemical data for the Pierre Shale. Ph.D. Thesis, Granada Univ., Spain, 398pp. United States Geological Survey Professional Paper 391-C: 1-31. Dias (1987) - Dinâmica sedimentar e evolução recente da plataforma Taylor, S.R. & McLennan, S.M. (1985) - The Continental Crust: its continental portuguesa setentrional. PhD thesis, Univ. Lisboa Composition and Evolution. Blackwell Scientific Publications, (unpublished), 500p Oxford. Drago, T., Araújo, M.F.D., Valério, P., Weber, O., Jouanneau & J.-M. Thorez, J. (1976) - Practical identification of clay minerals. Editions (1998). Geomorphological control of fine sedimentation on the G. Lelotte, Belgique, 99 pp. northern portuguese shelf. Boletim Inst. Esp. Oceanogr.,n° 15 Vidinha, J.M.; Rocha, F.; Andrade, C. & Gomes, C. (1998a) - (1-4): 111-122. Mineralogical characterization of the fine fraction of the beach Drago T., Naughton F., Moreno J., Rocha F., Cachão, M., Sanchez and dune sediments situated between Espinho and Torreira Gõni M.F., Oliveira A., Cascalho J., Fatela F., Freitas C. & (Portugal). Geostatistical approach. Cuaternario y Andrade C. (2002)- Geological Record of Environmental Geomorfología, 12 (3-4): 49-56. Changes in the Douro Estuary (NW Portugal): Since the Late Vidinha, J.M.; Rocha, F.; Andrade, C. & Gomes, C. (1998b) - Glacial. Proc. Littoral'2002 - 6th EUROCOAST Int. Conf., III: Aplicação de índices mineralógicos na diferenciação de 341-346. ambientes litorais de deposição - Espinho / S. Jacinto, Portugal. Freitas M.C., Andrade C. & Cruces A. (2002) - The geological record 1º Simpósio Interdisciplinar Sobre Processos Estuarinos - of environmental changes in southwestern Portuguese coastal SIPRES, Faro, pp.81-84. lagoons since the Lateglacial. Quat. Int. 93-94: 161-170. Vidinha, J.M.; Rocha, F.; Andrade; C. Gomes, C. & Freitas, M.C. Flemming, B.W., 2000. A revised textural classification of gravel-free (2000) - The behaviour of mineralogical ratios in beach and dune muddy sediments on the basis of ternary diagrams. Cont. Shelf sediments. 3º Simp. Margem Contin. Atlant. Ibérica; Univ. Res., 20: 1125-1137. Algarve, Faro, pp. 407-408. Mellinger, R. M. (1979) - Quantitative X-ray diffraction analysis of clay minerals. An evaluation. Saskatchewan Research Council, Canada, SRC Report G-79: 1-46. Oliveira, A.; Rocha, F.; Rodrigues, A.; Jouanneau, J.; Dias, A.; Weber, O. & Gomes, C. (2002) - Clay minerals of the sedimentary cover from the Northwestern Iberian Shelf. Progress in Oceanography 52(2-4): 233-247. Pevear, D.R. & Mumpton, D.R. (1989) - Quantitative mineral analysis of clays. CMS Workshop Lectures, 1. The Clay minerals Society, Colorado (USA). (Received: February, 25, 2004. Accepted: December, 27, 2004)

58 Thalassas, 2005, 21 (1): 59-65 An International Journal of Marine Sciences

FECUNDITY AND SIZE AT SEXUAL MATURITY OF THE FIDDLER CRAB UCA VOCANS (LINNAEUS,1758) (BRACHYURA: OCYPODIDAE)

C. LITULO(1)

Keywords: Mangrove forests, fiddler crabs, fecundity, sexual maturity, Uca vocans, East Africa, Mozambique

ABSTRACT INTRODUCTION

Fecundity and size at sexual maturity of the fiddler Fiddler crabs (genus Uca) are common intertidal crab Uca vocans were examined in a population from inhabitants of marshes, salt flats and mangroves Inhaca Island, southern Mozambique. Crabs were swamps along tropical, neotropical and temperate randomly sampled in December 2002 and January coastlines (Christy & Salmon, 1984; Mouton & Felder, 2003 at low-tide periods. Captured crabs were 1995; Costa & Negreiros-Fransozo, 2003). These crabs measured for carapace width (CW), dissected for the built complex burrows in substratum and display determination of development stage of gonads and egg particular behavior associated with burrow utilization counting. Fecundity in Uca vocans varied from 2000 (Costa & Negreiros-Fransozo, 2003). Burrowing and (CW = 12) to 20100 (CW = 25.3) eggs and increased foraging activities of fiddler crabs promote with female size. The onset at sexual maturity is bioturbation of estuarine intertidal flats (Colpo & achieved at 23.27 and 23.88 mm CW in females and Negreiros-Fransozo, 2004). They also remove large males respectively. The fecundity of fiddler crabs is amounts of sediment and change substrate compared with respect to latitudinal gradient. characteristics, increasing water and organic matter contents, and changing the nutrient dynamics, which affects microbiotic growth and simulates vegetal production (Genoni, 1985; Colpo & Negreiros- Fransozo, 2004).

Fiddler crabs are usually characterized by a high density compared to other brachyurans found in mangrove forests (Macia et al., 2001; Skov & Hartoll, 2001; Skov et al., 2002; Hartnoll et al., 2002). According to Macia et al. (2001), the activity patterns (1) Departamento de Ciências Biológicas, Faculdade de of fiddler crabs are largely influenced by the tidal Ciências, Universidade Eduardo Mondlane, Caixa Postal 257, Tel: + 258-1-492177, Fax: +258-1-492176, periodicity, which is in turn moderated by moon phase. Maputo, Mozambique. E-mail: [email protected] These crabs usually display several reproductive traits,

59 C. Litulo

during which females extrude several egg masses a ecology and mating system of U. vocans was given by year. Consequently, the modal classes change over Salmon (1984), from a single study carried out in a time as a result of both reproduction and quick juvenile population From Cape Fergunson, Australia. In South recruitment (Thurman, 1985). Africa, Uca vocans breeds seasonally with two peaks of spawning (October-November and April-May) According Christy & Salmon (1984) and Christy (Emmerson, 1994). Nothing is known about the (1987), fiddler crabs can be separated into two groups reproductive biology of this species in Mozambique. by the morphology of their carapace front. Wide-front crabs include Central, South and North American In this study, I characterize the reproductive species, while narrow-fronts are found in the Indo- biology of Uca vocans (Linnaeus, 1758) population Pacific (Crane, 1975). The differences between wide from Inhaca, southern Mozambique, focusing on and narrow fronts are a consequence of variations of fecundity and size at sexual maturity. The fecundity of ecological pressures that promote alternatives, but with fiddler crabs is also discussed with regard to similar adaptive strategies (Christy & Salmon, 1984). geographical distribution

Fecundity and size at sexual maturity are the key MATERIALS AND METHODS parameters that should reflect the life-time investment in reproduction (Ramirez Llodra, 2002; Lopez Greco The Saco da Inhaca (26°07'S, 32°56'E) is an & Rodriguez, 2004). Temperature, salinity, food enclosed bay fringed located at Inhaca Island, southern availability, rainfall, photoperiod and lunar cycles are Mozambique. The climate at this location is a mixture the main factors that can determine the periodicity and of tropical and subtropical, partly influenced by the extension of the breeding season, fecundity as well as South-Eastern trade wind, and a northerly monsoon, the size at the onset of sexual maturity of a species but also occasionally by strong and cold South-West (Colpo & Negreiros-Fransozo, 2003; Costa & winds or cyclones from the North-Eastern. The winter Negreiros-Fransozo, 2003; Litulo, 2004). (April-September) is usually cold and dry, while summer (October-March) is warm and rainy Six species of fiddler crabs occur in East Africa: (Guerreiro et al., 1996). Guerreiro et al. (1996) and Uca annulipes (H. Milne Edwards), U. gaimardi (H. Omodei Zorini et al. (2004) give detailed descriptions Milne Edwards), U. inversa (Hoffman), U. vocans regarding the ecology, fauna and flora of the Island. (Linnaeus), U. tetragonon (Herbst) and U. urvillei (H. Milne Edwards) (Hartnoll et al., 2002; Litulo, 2004). Samples were taken from December 2002 to From these, Uca annulipes has been studied in great January 2003 during low tides. Specimens were detail with respect to behaviour (Backwell & collected by hand during the day-time by two people Passmore, 1986; Burford et al., 2001), ecology over a period of approximately 1h over the same area (Ólafsson & Ndaro, 1997; Macia et al., 2001; Skov & of about 500 m2. At each sampling occasion, twenty Hartnoll, 2001; Skov et al., 2002; Hartnoll et al., 2002) 0.25 m2 squares were set out in the area from which ten and reproduction (Emmerson, 1994, 1999; Litulo, were randomly chosen for sampling. Each randomised 2004) and very little is know about the other species. square was excavated with a corer to a depth of 30 cm. Emmerson (1994, 1999) gave the first account on the All collected crabs were bagged, labelled and stored in reproductive biology of fiddler crabs from South 70 % ethanol until further analysis. In the laboratory, Africa and, with particular reference to Mozambique, crabs were identified, sexed and checked for the this has been accomplished recently by Litulo (2004) presence of eggs on female pleopods. The carapace for Uca annulipes. width (CW) was measured using a vernier caliper (±0.05 mm accuracy). Uca vocans (Linnaeus, 1758) occurs throughout the tropical western and the Indo-Pacific region All crabs (both males and females) were dissected (Crane, 1975). It is most abundant on exposed muddy for the analysis of gonad development stages (after flats seaward of the mangrove fringes (Salmon, 1984). Yamaguchi 2001, Negreiros-Fransozo et al., 2002). The only detailed description on the reproductive For males, immature gonads (stage I): testes and vas

60 Fecundity and Size at Sexual Maturity of the Fiddler Crab Uca vocans (Linnaeus, 1758) (Brachyura: Ocypodidae)

Table 1. Comparative data of fecundity of fiddler crabs (genus Uca) from various geographical locations. (min = minimum; max = maximum; CW = carapace width).

deferens cannot be distinguished; initial maturing In order to estimate fecundity, 30 ovigerous (stage II): gonads are filamentous and translucid and females with eggs in the stage I were selected for egg are visible under magnification; advanced maturing counting. Pleopods were removed from females, (stage III): testes are fully developed and the vas placed in petri dishes filled with seawater, and eggs deferens differentiates gradually into highly detached by the gradually addition of a solution of convoluted testes. Likewise, female gonads were sodium hypochlorite. Bare pleopods were then classified as follows: immature (stage I) = ovary thin discarded by gently stirring in a beaker filled with 200 and translucent, undifferentiated and ovaries are light ml seawater. Three 1.5 ml sub-samples were taken orange; Initial maturing (stage II) = colour ranging using a pipette, with eggs counted under a dissecting from opaque white to yellow and vitellogenesis is microscope. The average value obtained was then taking place; advanced maturing (stage III) = ovary extrapolated for the whole suspension to estimate the fills almost whole thorax cavity and are dark brown. number of eggs (Diaz et al., 1983; Ramirez Llodra, 2002). For egg staging, 10-15 eggs were removed from females'pleopods and classified into three The mean size of individuals of both sexes (males developmental stages (modified from Rabalais, 1991). and females) was compared by the Student t-test Stage I: eggs laid few days, filled with orange yolk, no (Underwood, 1997). For fecundity analysis, data were signs of segmentation and 100% of volume occupied analysed using the power function (Y = aXb) of egg by yolk. Stage II: incubation at its halfway period, eggs number (EN) versus carapace width (CW). Student t- have brown colour tending to grey, the compound eyes test was employed to verify whether the morphometric of larvae are visible and 60% of the volume is relationship was positively or negatively allometric occupied by yolk. Stage III: the larvae are a few day (Underwood, 1997). Considering the percentages of from hatching and are totally formed, eggs are dark mature and immature individuals in each size class, the tending to black and little yolk is depleted. minimum size at first maturation was determined on

61 C. Litulo

DISCUSSION

Fecundity is an important parameter measured in crustacea for the estimation of the reproductive potential and future stock size of a given species or population (Hattori & Pinheiro, 2001). Moreover, fecundity is directly related to life-history traits such as egg size, age at maturity, life span and reproductive effort (Ramirez Llodra, 2002). These life-history traits may have complex interactions determining specific trade-offs between them.

The fecundity of U. vocans of the present study was found to be similar to that described for other brachyuran studied previously (see Hines, 1982 for review), with an increase in number of eggs as the Figure 1. crabs grow larger. The coefficient of determination Uca vocans (Linnaeus, 1758). Scatterplot for the relationship between egg was high for the relationship between egg number and number (EN) and carapace width (CW). size, suggesting that carapace width is a good predictor of fecundity, since explained up to 90% of the total the basis of the size at which 50% of the individuals in variance. However, there were some variations in the the population actively entered the gonad maturation. number of eggs per batch among females of similar Both females and males were considered mature from size (see Fig. 1). Such variations are often related to stage 2. seasonal variation in food availability, multiple spawns, incomplete fertilization, predation and RESULTS parasitism.

A total of 430 crabs were analysed from which 250 A positive allometry between egg number and were females and 180 males. The carapace width (CW) female size was found in the present population, an of males ranged from 5.0 to 34.7 mm. For females it increase in fecundity with an increase of female size. ranged from 4.7 to 32.9 mm. The average size of males According to Ramirez Llodra (2002), the relationship (mean ± standard deviation: 25.9 ± 4.05) was between female size and fecundity is a major significantly larger (t = 7.85, p < 0.001) than that of characteristic of reproduction in many crustaceans, and females (24.2 ± 2.93). is related to morphological and physiological constraints in energy allocation and gonad maturation. The number of eggs ranged from 2000 (CW = 12.0 In many brachyurans, carapace width is the factor that mm) to 20100 (CW = 25.3 mm) eggs with an average has the great influence on the variability of the number of 11045.37 ± 565.7 eggs respecticely. Egg number of eggs, resulting in slopes for the relationship between (EN) was positively correlated with female size (CW) egg number and size which are higher than 3. and the resulting scatterplot shows a curvilinear trend (Fig. 1), expressed by the function EN = The number of eggs and size of ovigerous females 2.0013CW3.18526 (r2 = 0.9816, N = 30, p < 0.001). The of fiddler crabs exhibit a latitudinal pattern, with the power function had a slope of 3.185 (b > 3, t = 10.35, smaller crabs inhabiting low latitudes and the larger p < 0.05), demonstrating that egg number (EN) crabs high latitude (Table 1). Several authors have increases with female size (CW). agreed that several brachyurans show an increase in their size towards higher latitudes (e.g. Emmerson, The estimated size at which 50% of the males 1994, 1999; Lardies & Castilla, 2001). This is because reach sexual maturity was 23.88 mm CW (Fig. 2A) in higher latitudes, abiotic (e.g. temperature, rainfall, whereas females matured at 23.27 mm CW (Fig. 2B). photoperiod and salinity) and biotic (e.g. food

62 Fecundity and Size at Sexual Maturity of the Fiddler Crab Uca vocans (Linnaeus, 1758) (Brachyura: Ocypodidae)

more than one female, since larger male ocypodid crabs may have greater chances of obtaining females for copulation and win more intra-specific fights (Christy & Salmon, 1984; Christy, 1987)

Size at the onset of sexual maturity is a crucial variable to be taken into account while investigating the reproductive ecology of a given organism. For crustaceans there are not always outer characteristics such as colour and size that will inform in a rapid, direct and unequivocal manner about the exact moment when an individual reaches sexual maturity (Mantelatto & Fransozo, 1996). In brachyuran crabs, this is easier in females due to unambiguous signals of breeding competency, such as the presence of eggs attached to pleopods (Flores & Paula, 2002). However, estimates based on the smallest egg-bearing female are dependent on the sample size and do not indicate the average size at which females in a given population reach maturity (Lopez Greco & Rodriguez, 2004; Ituarte et al., 2004). In this way, more precise estimates in the determination of the size at which males and females reach maturity should be based on comparative studies using a physiological (macroscopic and histological) and morphological analysis.

The present results indicated that females mature earlier than males. Many authors have reached the same results (e.g. Mantelatto & Fransozo, 1996; Flores & Paula, 2002). These differences occur as a result of small variations in morphology (size) and as a function of population origin, food availability and abiotic factors involved in the maturation processes (Sastry, 1983; Ramirez Llodra, 2002). Ramirez Figure 2. Llodra (2002) states that the relationship between Uca vocans (Linnaeus, 1758). Logistic plots fitted for the estimates of sexual maturity in males (A) and females (B). fecundity and size at sexual maturity depends on the life-history strategies of a species. Generally, earlier availability, nutrient concentration) are generally maturing species have a shorter generation because of constant, favouring gonad development and larval the shorter generation time needed to reach first release. reproduction and, the cost may be observed in a reduction in future fecundity. In contrast, species with Sexual dimorphism was evidenced in the present delayed maturity live longer, allowing them to grow study, with males reaching larger sizes than females. larger and therefore have a higher fecundity (Ramirez Lopez Greco et al. (2000) suggests that females may Llodra, 2002). Moreover, the longer lifespan may also have reduced somatic growth compared to males gives the possibility of undergoing a higher number because they concentrate their energetic budget for of lifetime fecundity and spawnings, which are gonad development. Moreover, males may reach larger characteristic in tropical species (e.g. Emmerson, sizes for successful competition for copulation with 1994).

63 C. Litulo

This study reports the first account on the Genoni, G.P. (1985). Food limitation in salt marsh fiddler crabs Uca reproductive biology of Uca vocans in East Africa. rapax (Smith) (Decapoda, Ocypodidae). Journal of experimental Marine Biology and Ecology, 87: 97-110. Further studies on reproductive cycle, embryonic Goshima, S., Koga, T. & Murai, M. (1996). Mate acceptance and development, morphological maturity, moulting and guarding by male fiddler crabs Uca tetragonon (Herbst). Journal growth are needed in to answer several questions of Experimental Marine Biology and Ecology, 196: 131-143. Guerreiro, J., Freitas, S., Pereira, P., Paula, J. & Macia, A. (1996). which remain obscure regarding the breeding biology Sediment macrobenthos of mangrove flats at Inhaca Island, of this crab. Mozambique. Cahiers de Biologie Marine, 37: 209-327. Hartnoll, R.G., Cannicci, S., Emmerson, W.D., Fratini, S., Macia, A., ACKNOWLEDGEMENTS Mgaya, Y.D., Ruwa, R.K., Shunula, J.P., Skov, M.W. & Vannini, M. (2002). Geographic trends in mangrove crab abundance in East Africa. Wetlands Ecology and Management, 10: 203-213. Thanks to the Marine Biological Station of Inhaca Hattori, G.Y. & Pinheiro, M.A.A. (2001). Fecundity and embryology Island for logistical support. Many thanks to my ever of Pachycheles milinifer (Dana, 1852) (Anomura, Porcellanidae) at Praia Grande, Ubatuba, SP, Brazil. Nauplius, 9:97-109. energetic field assistant Sergio Mapanga for tireless Hines, A.H. (1982). Allometric constraints and variable of reproductive help. effort in brachyuran crabs. Marine Biology, 69: 309-320. Ituarte, R.B., Spivak, E.D. & Luppi, T.A. (2004). Female reproductive REFERENCES cycle of the Southwestern Atlantic estuarine crab Chasmagnathus granulatus (Brachyura: Grapsoidea: Varunidae). Scientia Marina, Backwell, P.R.Y & Passmore, N.Y. (1986). Time constraints and 68: 127-137. multiple choice criteria in the sampling behaviour and matting Lardies, M.A. & Castilla, J.C. (2001). Latitudinal variation in the choice of the fiddler crab Uca annulipes. Behavioural Ecology reproductive biology of the commensal crab Pinnaxodes chilensis and Sociobiology, 38: 407-416. (Decapoda: Pinnotheridae) along the Chilean coast. Marine. Burford, F.L.R., McGregor, P.K. & Oliveira, R.F. (2001). Intersexual Biology, 139: 1125-1133. differences in the mudballs of Uca annulipes (Decapoda: Litulo, C. (2004). Reproductive aspects of a tropical population of the Ocypodidae). Journal of the Marine Biological Association of the fiddler crab Uca annulipes (H. Milne Edwards, 1837) United Kingdom, 81: 353-354. (Brachyura: Ocypodidae) at Costa do Sol Mangrove, Maputo bay, Christy, J.H. & Salmon, M. (1984). Ecology and evolution of mating southern Mozambique. Hydrobiologia 525: 167-173. systems of fiddler crabs (genus Uca). Biological Reviews, 59: Lopez Greco, L.S., Hernandez, J.E., Bolanos, J., Rodriguez, E.M. & 483-509. Hernandez, G. (2000). Population Features of Microphrys Christy, J.H. (1987). Female choice and the breeding behavior of the bicornutos Latreille, 1825 (Brachyura, Majidae) from Isla fiddler crab Uca beebei. Journal of Crustacean Biology, 7: 642- Margarita, Venezuela. Hydrobiologia, 439: 151-159. 635. Lopez Greco, L.S. & Rodriguez, E.M. (2004). Reproductive Colpo, K.D. & Negreiros-Fransozo, M.L. (2003). Reproductive output performance in Cyrtograpsus angulatus and Cyrtograpus of Uca vocator (Herbst, 1804) (Brachyura, Ocypodidae) from altimanus (Brachyura, Varunidae) from Jabalï Island, Argentina. three subtropical mangroves in Brazil. Crustaceana, 76: 1-11. Journal of Crustacean Biology, 24: 213-216. Colpo, K.D. & Negreiros-Fransozo, M.L. (2004). Comparison of the Macia, A., Quincardete, I. & Paula, J. (2001). A comparison of population structure of the fiddler crab Uca vocator (Herbst, alternative methods for estimating population density of the 1804) from three subtropical mangrove forests. Scientia Marina, fiddler crab Uca annulipes at Saco Mangrove, Inhaca Island 68: 139-146. (Mozambique). Hydrobiologia, 449: 213-219. Costa, T..M. & Negreiros-Fransozo, M.L. (2003). Population biology Mantelatto,F.L.M. & Fransozo, A. (1996). Size at sexual maturity in of Uca thayeri Rathbun, 1900 (Brachyura, Ocypodidae) in a Callinectes ornatus (Brachyura, Portunidae) from the Ubatuba subtropical South American mangrove area: results from transect region (SP), Brazil. Nauplius, 4: 29-38. and catch-per-unit-effort techniques. Crustaceana, 75: 1201- Mouton, E.D. & Felder, D.L. (1995). Reproduction of the fiddler crabs 1218. Uca longisignalis and Uca spinicarpa in a Gulf of Mexico salt Crane, J. (1975). Fiddler crabs of the world. Ocypodidae: genus Uca. marsh. Estuaries, 18: 469-491. Princeton University Press, Princeton, New Jersey. Negreiros-Fransoso, M.L., Fransozo, A. & Bertini, G. (2002). Diaz, H., Conde, J.E. & Bevilacqua, M. (1983). A volumetric method Reproductive cycle and recruitment period of Ocypode quadrata for estimating fecundity in decapoda. Marine Ecology Progress (Decapoda, ocypodidae) at a sandy beach in southwestern Brazil. Series, 10: 203-206. Journal of Crustacean Biology, 22: 157-161. Emmerson, W.D. (1994). Seasonal breeding cycles and sex ratios of Ólafsson, E. & Ndaro, S.G.M. (1997). Impact of the mangrove crabs eight species of crabs from Mgazana, a mangrove estuary in Uca annulipes and Dotilla fenestrata on meiobenthos. Marine Transkei, southern Africa. Journal of Crustacean Biology, 14: Ecology Progress Series, 158: 225-231. 568-578. Omodei Zorini, L., Contini, C., Jiddawi, N., Ochiewo, J., Shunnula, Emmerson, W.D. (1999). Comparative fecundity and reproduction in J.& Cannicci, S. (2004). Participatory appraisal for potential seven crab species from Mgazana, a warm temperate African community-based mangrove management in East Africa. mangrove swamp. Crustacean Issues, 12: 499-512. Wetlands Ecology and. Mangement, 12: 87-102. Flores, A.V.V. & Paula, J. (2002). Sexual maturity, larval release and Rabalais, N.N. & Cameron, J.N.(1983). Abbreviated development of reproductive output of two brachyuran crabs from a rocky Uca subcylindrica (Stimpson, 1859) (Crustacea, Decapoda, intertidal area in central Portugal. Invertebrate Reproduction and Ocypodidae) in the laboratory. Journal of Crustacean biology, 3: Development, 41: 21-34. 519-541.

64 Fecundity and Size at Sexual Maturity of the Fiddler Crab Uca vocans (Linnaeus, 1758) (Brachyura: Ocypodidae)

Rabalais, N.N. (1991). Egg production in crabs with abbreviated development. In: Werner,A. & Kuris, A. (eds), Crustacean egg production, University of California, Santa Barbara, A.A. Balkema, Santa Barbara, pp. 217-234. Ramirez Llodra, E. (2002). Fecundity and life-history strategies in marine invertebrates. Advances in Marine Biology, 43: 87-170. Rodriguez, A., Drake, P. & Arias, A.M. (1997). Reproductive periods and larval abundance patterns of the crabs Panopeus africanus and Uca tangeri in a shallow inlet (SW Spain). Marine Ecology Progress Series, 149: 133-142. Salmon, M. (1984). The courtship, aggression and mating system of a primitive fiddler crab (Uca vocans: Ocypodidae). Transactions of the Zoological Society of London, 37: 1-50. Satry, A.N. (1983). Ecological aspects of reproduction. In: Vernberg, F.B & Vernberg, W.B (eds), The biology of Crustacea, Vol. 8. Environmental adaptations, Academic Press, New York, pp. 179- 270. Skov, M.W. & Hartnoll, R.G. (2001). Comparative suitability of binocular, burrow counting and excavation for the quantification of the mangrove fiddler crab Uca annulipes (H. Milne Edwards). Hydrobiologia, 449: 201-212. Skov, M.W., Vannini, M., Shunula, J.P., Hartnoll, R.G. & Cannicci, S. (2002). Quantifying the density of mangrove crabs: Ocypodidae and Grapsidae. Marine Biology, 141: 725-732. Thurman, C.L. (1985). Reproductive biology and population structure of the fiddler crab Uca subcylindrica (Stimpson). Biological Bulletin, 215-229. Underwood, A.J. (1997). Experiments in ecology: their logical design and interpretation using analysis of variance. Cambridge University Press, Cambridge. Yamaguchi, T. (2001). The breeding period of the fiddler crab Uca lactea (Decapoda, Brachyura, Ocypodidae) in Japan. Crustaceana, 74: 285-293.

(Received: April, 26, 2004. Accepted: October, 22, 2004)

65

Thalassas, 2005, 21 (1): 67-76 An International Journal of Marine Sciences

MINERALOGICAL AND GEOCHEMICAL CHARACTERISATION OF SURFICIAL SEDIMENTS FROM THE SOUTHWESTERN IBERIAN CONTINENTAL SHELF

A. MACHADO(1) (3); F. ROCHA(1); C. GOMES(1); J.A. DIAS(3); M.F. ARAÚJO(2) & A.GOUVEIA(2)

Keywords: Clay minerals; REE; Surficial sediments; Southwestern Iberian Continental shelf; Guadiana mouth; Tinto-Odiel mouth; Guadalquivir mouth.

ABSTRACT rivers are compared, the first two show enrichments in I and C, being the last one enriched in K, S and I-S. The main goal of the research being carried out, Clay mineral assemblages found, either in river whose results are herein disclosed, is to contribute to estuaries, or in the continental shelf reflect the the better understanding of sediment transport and composition of clay mineral assemblages deposition processes taking place in the Southwestern corresponding to river inputs. As a matter of fact, most Iberian Continental Shelf (Gulf of Cadiz), based upon of the identified clay minerals are inherited or detrital, mineralogical (silt and clay fractions composition) and and only a very small part would be authigenic or geochemical (rare earth elements nature and diagenetic. Nature, concentration and distribution of concentration) data. clay minerals on the continental shelf, indicates that the riverine clay minerals assemblage, consisting of In the fine fraction (<38 µm) of the shelf sediments illite, kaolinite and chlorite, changes to a marine clay the major non-clay minerals, mica, chlorite, quartz, minerals suite, which besides these clay minerals, feldspars and calcite, were identified. In the clay fraction contains smectite and illite-smectite mixed-layers too. (<2 µm) of the same shelf sediments, illite (I), kaolinite (K), smectite (S), illite-smectite interstratifications (I-S) REE distribution patterns show lower REE and chlorite (C), were identified. contents in sediments of the Guadalquivir transept, fact that could be related with the higher contents of When clay minerals distribution patterns in the biogenic particles found in these sediments. Moderate transepts of Guadiana, Tinto-Odiel and Guadalquivir positive correlations between REE, mineralogical composition - Phyllosilicates (Phyl), Quartz (Q), (1) MIA, Centro de Investigação "Minerais Industriais e Chlorite (C) and illite (I) - and coarse fraction content, Argilas", Universidade de Aveiro, PT 3810-193 Aveiro, Portugal, [email protected], [email protected], suggest that rock fragments or gravel are responsible [email protected] for hosting REE. Clay content control on REE (2) ITN, Instituto Tecnológico e Nuclear, Química, E. N. 10, concentration is probably less important than the rock 2686-953 Sacavém, Portugal, [email protected] fragments content. The highest REE concentrations are (3) CIACOMAR, Centro de Investigação das Ciências Costeiras e do Mar, Universidade do Algarve, 8000 Faro, found in shallow sea sediments, mainly from Guadiana Portugal, [email protected] and Tinto-Odiel transepts.

67 A. Machado, F. Rocha, C. Gomes, J.A. Dias, M.F. Araújo & A. Gouveia

INTRODUCTION sedimentary processes and products. Dabrio and Polo (1987), Rodríguez Vidal (1987) and Flor (1990), have Clay minerals are the main constituents of most focused the coastal dynamics and Holocene evolution. marine sediments. The composition of clay mineral Pérez et al., (1991), Cabrera et al., (1992), Nelson and assemblages in these sediments is related to the Lamothe (1993), and Elbaz-Poulichet and Leblanc weathering regimes on adjacent land masses and/or to (1996), have focused heavy metals pollution. Elbaz- the intensity and direction of the involved sedimentary Poulicheta and Dupuy (1999), Fernández Caliani et al., transport processes. In most of the marine coastal (1997), Galán et al., (1996), Gutierrez-Mas et al., regions of the world, the detrital clay associations (1997), Machado et al., (2000 a, b, 2001 a, b, 2002, reflect the combined influences of both the 2003), have focused either mineralogical or petrographic nature and climate prevailing in the geochemical aspects of the sediments. adjacent continent, as well as of the hydrodynamic regime of the offshore waters (Weaver, 1989; As matter of fact, no detailed mineralogical or Chamley, 1989). Chemical composition of shelf geochemical studies, unless those published by the sediments is very much dependent upon factors, such authors of the present work, have been performed so as: climate, vegetation cover, geomorphology, far, both in Guadiana estuary and in its adjacent pedogenetic processes and soil nature, as well as river continental shelf. hydrodynamics, physico-chemical processes, and anthropogenic activity. In the sector of the continental margin where the Gulf of Cadiz is located, both the shelf and the A large number of investigations carried out in the coastline are oriented NNW-SSE and stepped E-W. Gulf of Cadiz have been reported, especially in the The Gulf of Cadiz receives terrigenous sediment from Spanish rivers, Tinto-Odiel and Guadalquivir, and in the three main rivers: Guadalquivir, characterised by a the adjacent shelf. Borrego et al., (1993, 1995), and drainage basin whose area is estimated at 57.390 km2, López-Galindo et al., (1999), have focused the and by a mean water discharge estimated at 164 m3/s;

Figure 1. General map showing Guadiana, Tinto-Odiel and Guadalquivir transepts, the adjacent shelf, and the location of the sampling sites or stations.

68 Mineralogical and Geochemical Characterisation of Surficial Sediments from the Southwestern Iberian Continental Shelf

Table 1. Lithlogical, geochemical and mineralogical analytical data of the surficial sediments from the three studied transepts

Guadiana, characterised by a drainage basin whose The main goal of the present study was the area is estimated at 67.500 km2, and by a mean water investigation of both the mineralogical (clay minerals, discharge estimated at 78.8 m3/s; Tinto-Odiel, in particular) and geochemical (REE, in particular) characterised by a drainage basin whose area is distribution patterns, on sediments from the estimated at 3.352km2, and by a mean water discharge continental shelf of the Gulf of Cadiz. Sediments estimated at 0.7m3/s. Tinto-Odiel is characterised yet sampling took place along three transepts defined off by a considerable environmental impact due to the fact Guadiana, Tinto-Odiel and Guadalquivir estuaries, in that several sulphide mines are located in its drainage the southwestern Iberian Continental shelf. basin (Palanques et al., 1995). MATERIALS AND METHODS The Guadiana and Tinto-Odiel rivers discharge into the south-western segment of the Gulf of Cadiz The sector of the continental shelf under study is after flowing through the Iberian Pyrite Belt and drain located at the Southwest of the Iberian Peninsula, different geologic formations of Paleozoic age between the mouths of rivers Guadiana and (Schermerhorn, 1971) consisting predominantly of Guadalquivir, being limited by the coastline and the slates and volcanic-sedimentary formations in which 200m bathimetric line. Sediments sampling was large deposits of polymetallic massive sulphide are carried out in February 2001 (rainy period), on board interbedded. On the other hand, the Paleozoic rocks of of the research vessel N.R.P Andrómeda from the Sierra Morena are drained by the Guadalquivir river. Instituto Hidrográfico, Portugal, using a Smith

69 A. Machado, F. Rocha, C. Gomes, J.A. Dias, M.F. Araújo & A. Gouveia

McIntyre sampler. For the present study, 20 samples using the nuclear reactor existent at the ITN were selected from the total number of samples Laboratory, in Sacavém, Portugal. collected in the transepts referred to: 10 off Guadiana estuary, 5 off Tinto-Odiel estuary and 7 off RESULTS Guadalquivir estuary (Figure 1). Table 1, contains the lithological, geochemical and After being washed with distilled water, samples mineralogical data corresponding to both fine fraction were wet sieved through a 400 mesh, ASTM sieve. and clay fraction of the sediments from the three Sediments fraction <38 µm was dried in an oven at studied transepts. 60ºC and gently disaggregated with a porcelain mortar. The < 2 µm fraction was separated by a standard Fine fraction (<38 µm) sedimentation classification method, and well-oriented aggregates were prepared by sedimentation of clay The studied sediments consist mostly of fine to suspension onto glass slides. The composition of both medium grain size sand, sandy mud and mud, silt and clay fractions were analysed by X-ray containing, sometimes, shell fragments. diffraction using a Philips powder diffractometer equipped with an automatic slit, using nickel-filtered The fine fraction (<38 µm), besides the clay CuK -radiation emitted at 20mA and 40kV, a facility minerals, presents a variable mineralogical that was available at the Department of Geosciences, composition. Phyllosilicates (Phyl) mostly mica/illite, University of Aveiro, Portugal. Scans were run calcite (Cal), quartz (Qz), plagioclase (Plag), and K- between 2º- 40º 2θ for the identification of non-clay feldspar (FK) are the main minerals, while dolomite minerals in the silt fraction and between 2º-20º 2θ for (Dol), siderite (Sid), anhydrite (Anhy), opal C/CT the identification of clay minerals in the clay fraction. (Opal), anatase (Anat), pyrite (Py) and amphibole The identification of clay minerals present in the (Amph) are the accessory minerals. samples, illite, kaolinite, smectite, illite-smectite and chlorite, was carried out following the routine From the mineral distribution on the Guadiana procedures, which involved standard treatments of transept it may be seen that the sediments are largely solvation with glycerol and heating at 300 and 500ºC. enriched in phyllosilicates (34%) relatively to quartz The content of each clay mineral in a particular (29%) and feldspars (Plag+Fk>18%). The remaining sample, and expressed in weight %, was determined minerals represent < 8%. In terms of mineralogy, using the weighting factors introduced by Schultz shown in Table 2, Tinto-Odiel and Guadalquivir (1964), Biscaye (1965), Barahona (1974), Thorez transepts are very similar. These transepts are largely (1976) and Rocha (1993). Being the XRD-method enriched in calcite (36%-43%) relatively to semiquantitative with an accuracy of 5-10%, decimals phyllosilicates (35%-38%) and quartz (15%-8%) or have been rounded off to full numbers. feldspars (6%-3%). On the northern continental shelf of the Gulf of Cadiz, sediments are siliciclastic, On the other hand, eight Rare Earth Elements REE containing 25% of bioclastic carbonates. In terms of (La, Ce, Nd, Sm, Eu, Tb, Yb and Lu) were quantified sediment composition, a facies zonation almost as well, with an accuracy better than 5 % (Gouveia & parallel to the shoreline and isobaths could be defined Prudêncio, 2000), by Neutron Activation Analysis, (Segado et al., 1984; Gutiérrez-Mas, 1992). Table 2. Mean values (in weight %) of non-clay minerals identified in the sediment fine fraction (<38 µm), in the three studied transepts

70 Mineralogical and Geochemical Characterisation of Surficial Sediments from the Southwestern Iberian Continental Shelf

Clay fraction (<2 µm) In general, the studied shelf surficial sediments show REE concentrations, which are anomalous in The clay mineral assemblages identified in the clay regard to NASC concentration values. In the cases of fraction (<2 µm) of the shelf sediments are simple and Guadiana and Tinto-Odiel transepts, the analysed monotonous. They consist of illite (I), kaolinite (K) sediments contain REE concentrations above the and smectite (S), while chlorite (C) and illite-smectite average background values, whereas the samples (I-S) appear as accessory minerals. belonging to the Guadalquivir transept contain concentrations below the average background values Illite is the major clay mineral, being characterised for all the analysed REE. by the following parameters: content 45% - 70%; mean concentration 60%, standard deviation 8%. Kaolinite DISCUSSION is also fairly common: content 18%-28%, mean concentration 22%, standard deviation 3%. Smectite is Clay mineralogy: evidence for sediment source and common too: content 5% - 25%, mean concentration transport 12%, standard deviation 5%. Illite-smectite concentration is low throughout the shelf surficial Clay minerals distribution patterns for the <2 µm sediments: content 2% - 6%, mean concentration 4%, fraction and kaolinite/illite (K/I) ratio corresponding to standard deviation 1%. Finally, chlorite is very scarce: the analysed surficial sediments are shown in Figure 2. content 1% - 6%, mean concentration 2%, standard The shelf clay mineral assemblages are simple and deviation 1%. monotonous, and consist of illite, kaolinite and smectite, while illite-smectite and chlorite occur as Rare earth elements accessory minerals.

The REE (La, Ce, Nd, Sm, Eu, Tb, Yb, Lu) Clay minerals assemblage in the sediments of concentration values determined in the shelf surficial Guadiana estuary consists of illite (mean sediments, after being normalised relatively to REE concentration, 67%), kaolinite (mean concentration, concentration values exhibited by an average 23%), smectite (mean concentration, 3%), chlorite sediment/sedimentary rock, such as the North (mean concentration, 6%) and illite-smectite (mean American Shales Composite (NASC, Haskin et al., concentration, 1%) (Machado et al., 2003). 1968; Taylor and Mc Lennan, 1988), are displayed in Table 3. Since REE are not easily fractionated during As a rule, illite content increases seaward, being is sedimentation, sedimentary REE patterns may provide by far the most abundant clay mineral in the shelf information about the sediment provenance. sediments. It represents around 50% of the Guadalquivir transept, but its content goes up to 60% in the Guadiana and Tinto-Odiel transepts. As can be Table 3. REE contents (mean values, in mg/kg) determined in the studied observed in Figure 2a), the Guadalquivir transept is sediments, and in the NASC characterised by lower illite contents in the inner- (Haskin et al., 1968; Taylor and Mc Lennan, 1988) middle shelf (down to the depth of 100m). Otherwise in the outer shelf the illite content increases appreciably. This fact suggests that most of the illite existent in the outer shelf is transported into the Guadiana and Tinto-Odiel estuaries by the North Atlantic Surface Water.

Kaolinite is present in significant quantities (over 25%) in the Guadalquivir transept, however its content decreases gradually seaward, in opposition to illite. The Tinto-Odiel transept shows lower kaolinite contents relatively to the Guadiana transept. Kaolinite

71 A. Machado, F. Rocha, C. Gomes, J.A. Dias, M.F. Araújo & A. Gouveia

Figure 2. Clay mineral distribution patterns corresponding to the <2 µm fraction of sediments: a) Illite (%); b) Kaolinite (%); c) Smectite (%); d) Illlite-Smectite (%); e) Chlorite (%); f) K/I ratio. existent in the sediments supplied by the Guadalquivir 1996). However, in the Guadalquivir transept smectite estuary was originally formed from the rocks and soils is present in significant quantities variable within the lying on the river hydrographic basin. In general, range 6-25%. In regard to the Illite-smectite mixed- kaolinite content decreases seaward. layers they behave similarly to kaolinite.

Comparively with the sediment mineralogy Finally, chlorite is only present in minor amounts, identified in Guadiana and Guadalquivir estuaries, the the highest concentrations (<6%) being present near Tinto-Odiel estuary reveals the generalized lack of the mouth of the Guadiana estuary. Chlorite smectites. Such is fairly consistent with the concentration values in the other transepts are very mineralogical composition of the sediments delivered low. by the Tinto-Odiel rivers into the estuary since their acidic waters are devoid of suspended smectite, due to Kaolinite/illite ratio decreases seaward, in chemical dissolution induced by the acidic mine Guadalquivir transept, indicating an effective transport drainage (Fernández Caliani et al., 1996; Galán et al., and dispersion of illite provided by the river flow. That

72 Mineralogical and Geochemical Characterisation of Surficial Sediments from the Southwestern Iberian Continental Shelf

fact indicates that high kaolinite contents are which REE contents decrease gradually with depth, transported by rivers as suspended particulate material, however at the depth of 50m the contents are still high. since kaolinite settles down preferentially in the These sediments contain also high contents of heavy estuarine sediments (Weaver, 1989). Otherwise, this metals (Machado et al., 2004), therefore they contribute ratio that appears to increase seaward in the Guadiana largely to the heavy metals enrichment identified in transept, may be due to the fact of this transept being sediments of the Gulf of Cadiz. comparatively shorter than the others, and positioned closer to the river mouth (≈60m).

Figure 3, shows the variation of clay minerals (I, K, S, I-S, C) contents, in the three studied transepts.

Figure 4. Distribution patterns of ΣREE contents (in ppm).

Comparing the three studied transepts, REE distribution patterns show lower REEs contents in the Figure 3. sediments of the Guadalquivir transept, fact that could Distribution patterns of clay minerals contents. be related with the higher contents of biogenic particles found in these sediments. Moderate positive In comparative terms, clay minerals distribution in correlations between REE concentration, the three studied transepts, shows great similarity: mineralogical composition (Phyl, Qz, C, I), and sediments from Guadiana and Tinto-Odiel transepts sediment coarse fraction content, suggest that rock are enriched in I and C, whereas sediments of the fragments is a privileged component for hosting the Guadalquivir transept are enriched in K, S and I-S. REE. Clay content control on REE content is probably Clay minerals in the continental shelf reflect the less important than rock fragments content. composition of the inputs provided by the rivers. In the sediments from the Guadalquivir transept, they are Figure 5 shows the "REE sediment- Shale essentially of authigenic origin (S and I-S), whereas in normalized plot" for the studied surficial sediments the sediments of the Guadiana and Tinto-Odiel (<38 µm fractions) in each transept: Guadiana, Tinto- transepts, they are essentially of detrital origin (I, K, Odiel and Guadalquivir estuaries, using the ratios and C). Chlorite content decreases from the Guadiana between the concentrations of the individual REE in transept to the Guadalquivir transept, reflecting the the sediment and the concentration of the same geological settings of the respective hydrographic elements in the reference material, North American basins, dominated by metamorphic rocks in the case of Shale Composite (Haskin et al., 1968; Taylor and Mc Guadiana river, and by sedimentary rocks in the case Lennan, 1988). of Guadalquivir river. In the Guadiana and Guadalquivir transepts, the Rare Earth Elements: distribution and behaviour REE distribution show a pattern sub-parallel to the coastline, whereas in the Tinto-Odiel transept, the REE Relatively high REE contents were determined in distribution show a convex pattern (Figure 5). sediments from the Tinto-Odiel transept that is positioned close to the coastline (just 30m bathymetric REE distribution patterns in the surficial sediments line), that decrease rapidly with depth (Fig.4). The same are characerised by a clear enrichment in light REE happens with sediments from the Guadiana transept in (LREE) from the outer shelf to the inner shelf. A

73 A. Machado, F. Rocha, C. Gomes, J.A. Dias, M.F. Araújo & A. Gouveia

Figure 5. REE - Shale Normalized" concentrations in the surficial sediments of the three studied transepts (Guadiana, Tinto-Odiel and Guadalquivir). negative Ce anomaly is apparent in several samples, CONCLUSIONS mainly from the Tinto-Odiel transept. It is assumed that the dissolution of phosphogypsum in the Tinto Clay mineral assemblages identified in the river has favoured the formation of a negative Ce surficial sediments of the Sothwestern Iberian anomaly in the REE distribution patterns. The formed Continental Shelf consist mainly of illite, kaolinite,

SO4 complexes would have played a dominant role in smectite, being illite-smectite mixed-layers and controlling the REE distribution in Odiel river (Elbaz- chlorite accessory minerals. The nature, concentration Poulichet and Dupuy, 1999). and distribution of clay minerals indicate that the detrital supply with origin on the adjacent continental The highest REE contents were found in shallow areas is the most important process being involved, seated sediments, mainly of Guadiana and Tinto-Odiel being also important the transport southeastward by transepts. the North Atlantic Surface Water.

74 Mineralogical and Geochemical Characterisation of Surficial Sediments from the Southwestern Iberian Continental Shelf

The higher energetic environment of Guadiana Elbaz-Poulichet, F. and Leblanc, M. (1996) - Transfer de métaux d'une river, comparatively to Guadalquivir river, appears to province minière á l'océan par des fleuves acides (Rio Tinto, Espagne), C.R. Acad.Sci. Paris, 322,1047-1052. be responsible for the higher contribution of sediments Elbaz-Poulichet, F.; Morley, N.H.; Beckers, J.M. & Nomerange, P. from terrigenous sources and for the higher REE (2001). Metal fluxes through the Strait of Gibraltar: the influence contents. The lower REE concentrations determined in of the Tinto and Odiel rivers (SW Spain). Marine Chemistry, 73: 193-213. the Guadalquivir transept, reflect most probably the Fernández Caliani, J.C., Requena A. and Galán, E. (1996) - Clay increase of the biogenic carbonate component in the mineralogy and heavy metal content of suspended sediments in na sediments, fact that promotes a dilution effect on REE extremely polluted fluvial environment: the Tinto river (SW Spain). concentrations. This work may suggest an influence of Preliminary report, In: M: Ortga-Huertas, A. López-Galindo and I. Palomo (Eds.), Advances in Clay Minerals, 218-220. the Tinto-Odiel and Guadiana rivers as REE source for Flor, G. (1990) - Tipología de dunas eólicas. Procesos de erosión- the Gulf of Cadiz. sedimentación costera y evolución litoral de la provincia de Huelva (Golfo de Cádiz Occidental, Sur de España). Estudios Geol. 46, 99-109. ACKNOWLEDGMENTS Galán, E., Fernández Caliani, J.C. and Requena, A. (1996) - Provenance and evolution of clay minerals in the Tinto river. SW The present work was carried out within the Spain, Proc. 14th Conf. Clay Min. Petrol. Banská Stiavnica, framework of the following projects: SIRIA (Situação Slovakia. Galán, E., Requena, A. and Fernández Caliani, J.C. (1996) - Provence de Referência na Região Costeira Algarvia and evolution of suspended clay minerals in the rio Tinto (Tinto Influenciável pela Barragem de Alqueva), "Programa river), SW Spain. Geologica Carpathica - Series Clays. 33-38. de Ambiente e Defesa" (Ministério da Defesa Gouveia M. A., & Prudêncio, M. I. (2000). New data on sixteen reference material obtained by INAA. Journal of Radioanalytical Nacional, and Fundação das Universidades and Nuclear Chemistry, 245: 105-108. Portuguesas), EMERGE (Estudo Multidisciplinar do Gutiérrez-Mas, J.M. (1992) - Estudo de los sedimentos recientes de la Estuário do Rio Guadiana), "Programa ODIANA" and plataforma continental y Bahia de Cádiz. PhD thesis, Univ. CRIDA (Consequences of River Discharge Cádiz, Spain. Gutierrez-Mas, J.M.; Lopez-Galindo, A. and Lopez-Aguayo, F. (1997) Modifications on Coastal Zone and Continental Shelf) - Clay minerals in recent sediments of the continental shelf and (PLE/8/00). The authors thank the financial support Bay of Cádiz (SW Spain). Clay Minerals, 32, 507-515. provided. Haskin, L.A., Helmke, P.A., Paster, T.P.& Allen, R.O. (1971)- Activation analysis. In: Geochemistry and Cosmochemistry, p. 201-218, A.O. Brunfelt and E. Steinnes, eds., REFERENCES Universitetsforlaget, Oslo. López-Galindo, A.; Rodero, J. and Maldonado, A. (1999) - Surface Barahona, E. (1974) - Arcillas de ladrillería de la provincia de facies and sediment dispersal patterns: southeastern Gulf of Granada: evaluación de algunos ensayos de materias primas. Cadiz, Spanish continental margin. Marine Geology. 155, 83-98. Ph.D. Thesis, Granada Univ., Spain, 398pp. Machado, A.; Rocha, F.; Araújo, M.F.; Vitali, F.; Gomes, C. & Dias, Biscaye, P.E., (1965) - Mineralogy and sedimentation of recent deep J.A. (2004) - Geochemical characterisation of surficial sediments sea clay in the Atlantic and adjacent seas and oceans. Bull. Geol. from the southwestern Iberian continental shelf. Ciencias Soc. Am. 76, 803-832. Marinas. (in press). Borrego, J., Morales, J.A. and Pendón, J.G. (1993) - Holocene filling of Machado, A.; Rocha, F. & Gomes, C., (2003) - Clay minerals na estuarine lagoon along the mesotidal coast of Huelva: the Piedras distribution in recent estuarine sediments: the case of the River mouth, southwestern Spain. J. Coastal Res. 9. 242-254. Guadiana Estuary, Southwestern Iberian Peninsula. Actas da Borrego, J. Morales, J.A. and Pendón, J.G. (1995) - Holocene estuarine Euroclay 2003. 10 th Conference of the European Clay Groups facies along the mesotidal coast of Huelva, sout-western Spain, Association, Italy. In: W.A. Flemming and A. Batholomá (Eds.). Tidal signatures in Machado, A.; Rocha, F.; Dias, A. & Gomes, C. (2000). Distribution moden and ancient sediments. Int. Ass. Sediment. Spec. Publ. 24, patterns of clay minerals in the surficial sediments of the pp.151-170. continental shelf off Guadiana estuary (Algarve, Portugal) - Cabrera, F., Conde, B. and Flores, V. (1992) - Heavy metals in the preliminary study. Proceedings of the 1st Latin American Clay surface sediments of the tidal river Tinto (SW Spain). Fres. Conference, Ass. Port. Argilas, 2: 64-68. Environ. Bull. 1, 400-405. Machado, A.; Rocha, F.; Dias, A. & Gomes C. (2000). Aplicação de Chamley, H. (1989) - Clay Sedimentology. Springer-Verlag, 623 pp. Parâmetros Mineralógicos na Caracterização de Sedimentos Dabrio and Polo, C.J. and Polo, M.D. (1987) - Holocene sea-level Superficiais da Plataforma Continental Adjacente ao Estuário do changes, coastal dynamics and human impacts in southern Guadiana. Actas do 3º Simpósio sobre a Margem Ibérica Iberian Peninsula. In: C. Zazo (Ed.) Late Quaternary Sea-level Atlântica, Univ. Algarve, 409-410. Changes in Spain. Pp.227-247. Machado, A.; Rocha, F.; Dias, A. & Gomes, C. (2001) - Application of Elbaz-Poulicheta, F, and Dupuy, C. (1999) - Behaviour of rare statistical analysis to clay minerals data corresponding to recent elements at the freshwater - seawater interface of two acid mine sediments from the continental shelf adjacent off Guadiana rivers: the Tinto and Odiel (Andalucia, Spain). Applied estuary. Nuevas tendencias en el estudio de las arcillas. Actas da Geochemistry, 14, 1063-1072. XVI Reun. Cient.Soc.Esp.Arcillas, 92-94.

75 A. Machado, F. Rocha, C. Gomes, J.A. Dias, M.F. Araújo & A. Gouveia

Machado, A.; Rocha, F.; Dias, J. A. & Gomes, C. (2001). Schermerhorn, L.J.G. (1971) - An outline stratigraphy of the Iberian Mineralogical characterisation of fine sediments of the Pyrite Belt. Bol. Geol. Min. 82, 239-268. continental shelf off Guadiana estuary. Actas do V Congresso do Schultz, L. G. (1964) - Quantitative interpretation of mineralogical Quaternário de Países de Línguas Ibéricas - V Reunião do composition from X-ray and chemical data for the Pierre Shale. Quaternário Ibérico, Lisboa, 214-217. United States Geological Survey Professional Paper 391-C,1-31. Machado, A.; Rocha, F.; Gomes, C. & Dias, J.A. (2002). Clay minerals Segado, M., Gutierrez, J.M., Hidalgo F., Martinez, J.M. and Cepero, F. identified in the suspended particulate matter of the Guadiana (1984) - Estudio de los sedimentos recientes de la plataforma estuary, during a tidal cycle. La investigación de arcillas en continental gaditana entre Chipiona y Cabo Roche. Bol. Geol. geología, agricultura, médio ambiente y ciência de materiales Min. T.XCV-IV, 310-314. (Actas da XVII Reun. Cient. Soc. Esp. Arcillas), 177-180. Taylor, S.R. & McLennan, S.M. (1988) - The significance of the rare Nelson, C.H. and Lamothe, P.J. (1993) - Heavy metals anomalies in earths. In: Geochemistry and Cosmochemistry. Handbook on the the Tinto-Odiel river and estuary system, Spain. Estuaries, 16, Physics and Chemistry of Rare Earths, 11, pp. 485-578, K.A. 496-511. Gschneider, Jr. And L. Eyring, eds. Elsevier Science Publishers Palanques, A., I. Diaz, J. and Farran, M., (1995) - Contamination of B.V.. heavy metals in the suspended and surface sediment of the Gulf Thorez, J. (1976) - Practical identification of clay minerals. 99 pp. of Cadiz (Spain): the role of sources, currents, pathways and (Editions G. Lelotte, Belgique). sinks. Oceanologica Acta, 18, 4, 469-477. Weaver, C.E. (1989) - Clays, Muds and Shales. Developments in Pérez, M., Usero, J., Gracia, I. and Cabrera, F. (1991) - Trace metals in Sedimentology 44. Elsevier, Amsterdam. sediments from the "Ria de Huelva". Toxicol. Environ. Chem. 31- 32, 275-283. Rocha, F. (1993) - Argilas aplicadas a estudos litoestratigráficos e paleoambientais na bacia sedimentar de Aveiro. Phd Thesis, Univ. Aveiro, 399pp (unpublished). Rodríguez Vidal, J. (1987) - Recent geomorphic evolution in the Ayamonte-Mazagón sector of the South Atlantic coast (Huelva, Spain). Trab. Neóg. Cuatern. 10, 259-264. (Received: February, 3, 2004. Accepted: December, 27, 2004)

76 Thalassas, 2005, 21 (1): 77-84 An International Journal of Marine Sciences

AGE, GROWTH AND REPRODUCTION OF THE AXILLARY SEABREAM, PAGELLUS ACARNE (RISSO, 1827), FROM THE SOUTH COAST OF PORTUGAL

R. COELHO*, L. BENTES, C. CORREIA, J.M.S. GONÇALVES, P.G. LINO, P. MONTEIRO, J. RIBEIRO & K. ERZINI

Keywords: Pagellus acarne, Sparidae, age, growth, reproduction, maturity.

ABSTRACT INTRODUCTION

Axillary seabream, Pagellus acarne, caught by Sea breams (Sparidae) are a dominant component longlines in the Algarve (Southern Portugal), were of the Algarve continental shelf demersal fish sampled between August 1995 and August 1996. Age community (Gomes et al., 2001). More than 20 species was studied by counting growth increments on otoliths are found in Algarve (south Portuguese) waters and and the estimated von Bertalanffy parameters were many are among the more valuable of the commercial -1 Linf = 28.82 cm, K = 0.29 year and t0 = -1.47 year for and recreational fish species. One of these important -1 males and Linf = 32.30 cm, K = 0.18 year and t0 = -2.56 species is the axillary seabream, Pagellus acarne year for females. Evidence of the annual periodicity of (Risso 1827). This demersal fish species inhabits the deposition of increments was found by marginal various types of sea bottom, especially seagrass beds increment analyses. Macroscopic analysis of the and sand down to 500 m depth, but is more common gonads and the gonad somatic index showed that between 40 and 100 m. It has a wide geographical reproduction occurred over an extensive period of distribution along the European and African coasts, time, from May to November. Lengths at first maturity from Denmark to Senegal, and around the Madeira, were 18.10 and 17.60 cm for males and females, Azores, Canary and Cape Verde Islands. It also occurs respectively. This species was characterized as being a in the Mediterranean and in the Black Sea (Bauchot & protandric hermaphrodite. Hureau, 1986). The biology of this species has been studied in the Mediterranean (Andaloro, 1982; Stergiou et al., 1997), the Atlantic coast of Morocco and Western Sahara, (Mennes, 1985; Lamrini, 1986) and the Canary Islands (Pajuelo & Lorenzo, 2000). In Portugal, research has focused mainly on longline (Ezini et al., 1998) and gillnet (Santos et al., 1995) Universidade do Algarve, CCMAR/FCMA, selectivity. Campus de Gambelas, 8000-117 Faro, Portugal. Phone: +351 289 800100 ext.7242. Fax: +351 289 818353. Internet: www.ualg.pt/fcma/cfrg Total landings of sea breams in the Algarve have * Corresponding author e-mail: [email protected] declined in recent years from a combined total of

77 R. Coelho, L. Bentes, C. Correia, J.M.S. Gonçalves, P.G. Lino, P. Monteiro, J. Ribeiro & K. Erzini

Figure 1. Mean marginal increment in otoliths of P. acarne by month. The error bars represent ± standard error (SE). approximately 4000 t per year in 1987 to less than MATERIAL AND METHODS 2000 t per year in 2000. This decrease in landings has been very accentuated in the case of P. acarne. Biological sampling took place on the south coast Regarding the entire Portuguese coast, landings of P. of Portugal (Algarve), between August 1995 and acarne have decreased from 1949 t in 1988 to 1254 t August 1996. A total of 370 fish were captured, mostly in 2000, representing a reduction of 35.6 %. In the from commercial longliners operating along the Algarve, the reduction in recent years was even greater Portuguese South coast. The entire sample was (40.1%), with landings decreasing from 1071 t in 1987 brought to the laboratory where total length (TL, cm), to 641 t in 2000 (DGPA, 2000). total weight (Wt, g) and eviscerated weight (We, g) were measured. The date and location of capture were These trends underline the urgency and necessity recorded, along with other data such as hook size and for research on this species. In order to provide useful bait used. Saggitta otoliths were removed, cleaned in information for decision making, and to allow better distilled water and air dried. Gonads were removed, management, it is important to carry out the basic weighed and classified. Morphometric relations were research which provides the parameters necessary for established between TL and Wt. stock assessment. In this study, we focused on aspects of the population biology of P. acarne from the To estimate age, whole otoliths were immersed in Algarve, studying age, growth and maturity in fish glycerol and observed with a compound microscope exploited mainly by longlines. with amplifications between 10 and 40x, with a black

Table 1. Growth parameters of P. acarne estimated by non-linear regression from otolith readings, for all fish, males and females. All estimates are followed in brackets by the lower to upper limits of the 95 % confidence interval.

78 Age, Growth and Reproduction of the Axillary Seabream, Pagellus acarne (Risso, 1827), from the South Coast of Portugal

Figure 2. Monthly variation of relative frequency for maturity stages of pre, during and post spawning for females of P. acarne. background and under reflected white light. All (R - Rn) otoliths were read independently by between three and MI = (R - R ) six experienced researchers, and age was only assigned n n-1 for otoliths where at least three of the researchers were in agreement. Researchers did not have access to The von Bertalanffy growth function was fitted to information on size, sex or date of capture while they the individual length-at-age data of all aged specimens were counting growth increments. To quantify the by the non-linear least square method (NLIN efficiency of age estimations between readers, the procedure in SAS (1988)). Growth parameters were percent of concordant readings was calculated, compared between sexes using the Hotelling's T2 test comparing the percent of readings discordant in 0, +/- (Bernard, 1981). 1, +/- 2, +/- n years. The reproductive cycle and size at first maturity Marginal increment analysis was used to validate of the axillary seabream was studied by macroscopic the periodicity of the formation of the growth analysis of the gonads. Laboratory studies were increments. According to Beamish & McFarlane carried out and the sex and the maturity stage of (1983), this method is a simple and direct way to specimens determined either visually directly by eye validate age estimations, given that the variation of the or by compound microscope. The maturation distance of the opaque band to the border of the otolith classification stages were adopted from a number of throughout the year gives an indication of the time of others used for synchronous and total spawners the year when the bands are deposited. For this (Lagler, 1978). analysis, 10 otoliths were selected for each month. These otoliths were observed with a compound The gonad somatic index (GSI) (Htun-Han, 1978) microscope and measurements taken with an ocular was used to identify the spawning period: micrometer. The total length of the otolith (Rt) was measured along the dorso-ventral axis, as was the GSI = 100 x (gonad weight (g) / We (g)) radius (R - the distance from the focus to the margin) and the distance from the focus to the last ring (Rn) The proportion of mature individuals by size class and the penultimate ring (Rn-1). The monthly mean was used to fit maturity ogives and to estimate the size marginal increment (MI) was then calculated by: at first maturity (TL where 50% of the individuals are

79 R. Coelho, L. Bentes, C. Correia, J.M.S. Gonçalves, P.G. Lino, P. Monteiro, J. Ribeiro & K. Erzini

mature). The logistic curve was fitted by non-linear Wt = 0.012 x TL3.048 regression using the NLIN procedure in the SAS (1988) system: (r2 = 0.98, n = 370, Range: 12.4 to 36.5cm TL) The otoliths from the 370 fish were used for age 1 determination. In general, the otoliths were easy to Pi = -b(L -L ) read and only 5 specimens could not be aged. The 1 + e i 50 percentage of concordance between the different readers was 80.6%. Percentage of discordance was where Pi is the proportion of mature individuals in size 17.5%, 1.4%, 0.2%, 0.2% and 0.1% for +/- 1, +/- 2, +/- class Li, b is the slope and L50 is the size where 50% of 3, +/- 4 and +/- 5 years, respectively. the individuals are mature (Pi = 0.5). Overall, the best represented age classes were from Maturity ogives were fitted to males and females 1 to 8 (90.8% of the sample). Most of the male separately and compared with the Hotelling's T2 test specimens were aged between 2 to 5 years (65.9%), and (Bernard, 1981). only one male was estimated to be older than 8 years. This large specimen with 30cm TL was estimated to be RESULTS 13 years. Most of the females were between 3 and 7 years (71.1%). The youngest hermaphodite was 2 years Of the 370 specimens collected for population old and most of these specimens were between 3 and 6 dynamics studies, 82 (22.2%) were males, 159 (43.0%) years (72.6%). All immature specimens were were females, 15 (4.1%) were immature and 106 determined to be either age 0 or 1. (28.6%) were hermaphrodites. The sex of 8 individuals (2.2%) could not be determined. The mean total length Through the marginal increment analysis, it was of males (mean = 23.9 cm, sd = 2.9, range: 15.9 to 30 possible to validate age estimations. It was observed cm) was significantly different from the mean total that one pair of bands (one opaque and one translucent) length of females (mean = 25.8 cm, sd = 3.0, range: was formed each year. The opaque band started to form 16.7 to 36.5 cm) (t student test: P < 0.001). in July, corresponding to a sharp decrease in the marginal increment. During the rest of the year there A significant relationship between TL and Wt was was a progressive increase in the marginal increment found (ANOVA: P < 0.01): (Figure 1).

Table 2. Summary of length-at-age data for P. acarne using otoliths. All values refer to the total length in cm. SD refers to the standard deviation.

80 Age, Growth and Reproduction of the Axillary Seabream, Pagellus acarne (Risso, 1827), from the South Coast of Portugal

Table 3. Bio-geographic comparison of biological parameters for P. acarne. K, Linf and t0 refer to the von Bertalanffy growth parameters and M50 refers to length at first maturity. LFA refers to length frequency analyses. Values with an asterisk (*) refer to measurement of fork length while all the others refer to total length.

The estimated von Bertalanffy parameters, along Lengths at first maturity were 18.1 and 17.6 cm TL with the respective confidence intervals are presented for males and females respectively. Although these in Table 1. Significant differences between males and values are relatively similar, the maturity ogives females were found in these parameters (Hotelling T2 presented statistically significant differences test: P < 0.05). The observed and predicted lengths-at- (Hotelling T2 test: P < 0.05) (Figure 4). age for all fish, males and females are summarized in Table 2. DISCUSSION

Hermaphrodite fish accounted for 30.5% of the The observation and interpretation of the otoliths mature specimens. Hermaphrodite males were was very consistent, based on the independent and dominant in the smaller size classes while concordant readings of at least 3 different observers, hermaphrodite females occurred only in size classes of revealing the adequacy of these structures for 18 cm and greater. All fish in size classes greater than estimating age of the axillary seabream. Previous 31 cm were females. studies that have used otoliths to age P. acarne include Andaloro (1982) and Pajuelo & Lorenzo (2000). The axillary seabream has an extensive spawning season, from spring to autumn. Most spawning The marginal increment analyses in the otoliths females were found from May to October. Post- showed that a pair of bands (one opaque and one spawning stages occurred mostly from December to translucent) was deposited with an annual periodicity, April (Figure 2). The analysis of the GSI indicated a with the opaque band starting to be deposited in the similar spawning period as that found by analysis of summer, in July. The relatively large standard error in maturity stages. In general, the highest values started this month might be due to the fact that part of the in May and were observed until September in males individuals already started to deposit the opaque band and November in females (Figure 3). Again, the resting and have a small marginal increment (close to 0) while period was observed from December to April. others are still depositing the translucent band and

81 R. Coelho, L. Bentes, C. Correia, J.M.S. Gonçalves, P.G. Lino, P. Monteiro, J. Ribeiro & K. Erzini

Figure 3. Gonad somatic index (GSI) by month for males (A) and females (B) of P. acarne. The error bars represent ± 1 standard deviation. have large marginal increments (close to 1). This Formosa coastal lagoon, an important nursery area in analysis validated the use of otoliths for assessing age the Algarve, proved effective in the capture of in this species. Pajuelo & Lorenzo (2000) had already juveniles of other sparid species, but not the axillary demonstrated this annual deposition pattern for this seabream. This could be due to the fact that juveniles species in the Canary Islands. According to these of this species are not found in these coastal lagoons authors, the opaque band was deposited mostly during during their early life stages but, instead, live in the the summer, when water temperature is higher and adjacent coastal areas. Since they are not yet recruited food is more abundant. to the fishery they are not captured with the traditional commercial fishing gears that operate in these areas. The von Bertalanffy growth curve fitted to otolith The estimated growth parameters are similar to the based age-length data adequately explains the growth ones reported by other authors for the Mediterranean, pattern for the observed size range of the axillary the Western Sahara and the Canary Islands (Table 3). seabream. The fitted growth model probably does not The only author whose values are divergent from this properly describe juvenile growth in this species tendency is Andaloro (1982) who obtained higher because of the lack of individuals smaller than 12.4 cm values of K and lower values of Linf for species from in the sample. Beach seines used inside the Ria the eastern Mediterranean.

82 Age, Growth and Reproduction of the Axillary Seabream, Pagellus acarne (Risso, 1827), from the South Coast of Portugal

Figure 4. Maturity ogives for males and females of P. acarne, with the respective 1cm total length class observed values. The fact that males appear mostly in the lower reproductive period seems to occur mostly during the length classes and females in the higher classes, with winter, from October to March with a peak in significant differences detected in the mean lengths-at- December and January (Table 3). age of males and females, indicates that this is probably a protandric hermaphroditic species. This In Portugal, present legislation stipulates a minimum type of hermaphroditism may also be evidenced by the legal landing size of 18 cm for the axillary seabream. fact that females have a lower growth rate and a larger Given the estimated lengths at first maturity of 18.1 cm asymptotic maximum length. Therefore, in the earliest for males and 17.6 cm for females, this regulation may years of life, when the growth rate is higher, the not be appropriate and the minimum landing size should individuals are mainly males while, on the other hand, be increased as a precautionary measure. in the older age classes when the growth rate is lower most of the individuals are females. This type of ACKNOWLEDGMENTS hermaphroditism had already been described for this species (Lamrini, 1986; Arculeo et al., 2000; Pajuelo This work was funded in part by the European & Lorenzo, 2000). Even though we cannot specify the Union (DG XIV Ref. 98/082). length and age of sexual inversion, this probably occurs between 20 and 24 cm TL (between 3 to 6 years REFERENCES of age), since it is during this period that male and female proportions are identical. In the lower length Andaloro, F. (1982). Résumé des paramètres biologiques sur Pagellus classes, the sex ratio is in favor of the males and above acarne de la mer Tyrrhénienne méridionale et de la mer Ionienne septentrionale. FAO Fish. Pap., 266: 89-92. 24 cm of total length this trend is reversed and the sex Arculeo, M., Brusle-Sicard, S., Potoschi, A. & Riggio, S. (2000). ratio favors females. These values are similar to the Investigations on gonadal maturation in Pagellus acarne (Pisces, ones found by Andaloro (1982), who considered that Sparidae) in the strait of Messina (Sicily). Ital. J. Zool., 67: 333-337. Bauchot, M.L. & Hureau, J.C. (1986). Sparidae. In: Whitehead, P.J.P., sexual inversion occurs between ages 2 to 6. Bauchot, M.L., Hureau, J.C., Nielsen, J. & Tortonese, E., Eds., Fishes of the North-eastern Atlantic and the Mediterranean, This study recorded a long spawning period, Volume II, UNESCO, Paris, pp. 883-907. Beamish, R.J. and McFarlane, G.A. (1983). Validation of age between May and November, with a peak in the determination estimates: the forgotten requirement. In: Prince, summer months, while the resting period is mainly E.D. & Pulos, L.M. Eds., Proceedings of the international during the winter months. A reproductive season workshop on age determination of oceanic pelagic fishes: tunas, occurring mostly during the summer period had billfishes and sharks, NOAA Technical Report NMFS 8, U.S. Department of Commerce, Miami, pp. 29-33. already been described for the Mediterranean and the Bernard, D. (1981). Multivariate analysis as a mean of comparing Western Sahara, while in the Canary Islands the growth in fish. Can. J. Fish. Aquat. Sci., 38: 233-236.

83 R. Coelho, L. Bentes, C. Correia, J.M.S. Gonçalves, P.G. Lino, P. Monteiro, J. Ribeiro & K. Erzini

DGPA, 2000. Recursos da pesca. Série Estatística 1999. Direcção- Pajuelo, J.G. & Lorenzo, J.M. (2000). Reproduction, age, growth and Geral das Pescas e Aquicultura, Lisboa, 170 pp. mortality of axillary seabream, Pagellus acarne (Sparidae), from Erzini, K., Gonçalves, J.M.S., Bentes, L., Lino, P.G. & Ribeiro, J. the Canarian archipelago. J. Appl. Ichthyol. 16: 41-47. (1998). Species and size selectivity in a "red" sea bream longline Santos, M.N., Monteiro, C.C. & Erzini, K. (1995). Aspects of the métier in the Algarve (southern Portugal). Aquat. Living Resour., biology and gillnet selectivity of the axillary seabream (Pagellus 11:1-11. acarne, Risso) and common pandora (Pagellus erythrinus, Gomes, M.C., Serrão, E. & Borges, M.D.F. (2001). Spatial patterns of Linnaeus) from the Algarve (south Portugal). Fish. Res., 23: 223- groundfish assemblages on the continental shelf of Portugal. 236. ICES J. Mar. Sci., 58: 633-647. SAS, 1988. SAS User's guide. SAS Institute, North Carolina, 494 pp. Htun-Han, M. (1978). The reproductive biology of the dab Limanda Stergiou, K.I., Christou, E.D., Georgopoulous, D., Zenetos, A., limanda (L.) in the North Sea: gonosomatic index, hepatosomatic Souvermezoglou, C. (1997). The Hellenic seas: physics, index and condition factor. J. Fish Biol., 13: 369-378. chemistry, biology and fisheries. In: Ansell, A.D., Gibson, R.N. & Lagler, K. (1978). Capture, sampling and examination of fishes. In: Barnes, M. Eds., Oceanography and marine biology: an annual Benegal, T. Ed., Methods for Assessment of Fish Production in review, UCL Press, pp. 415-538. Fresh Waters, Blackwell Scientific Publications, pp. 7-47. Lamrini, A. (1986). Sexualité de Pagellus acarne (Risso, 1826) (Teleosteen Sparidae) de la cote Atlantique Meridionale du Maroc (21º - 26º). Cybium, 10: 3-14. Mennes, F. (1985). Multispecies assessment of fish stocks off the Western Sahara region with emphasis on the family Sparidae. Naga, 3: 5-10. (Received: September, 2, 2004. Accepted: January, 28, 2004)

84