Part 3

Wildlife fauna and integration of organic matter into marine food webs

 

 

 

 

Task 10

Vertebrate food webs

 

 

 

 

 


 

 

 

 

 

 

 

 

 

 

 


Task 10:

Vertebrate Food web

 

 

 

 

 

 

 

Costa, M.J.; Catarino, F.; Salgado, J.P.; Serôdio, J.; Cabral, H. & Franco, M.A.

 

 

 

 

 

 

 

Final Report

 

 

 

 

 

 

 

 

 

 

 

 

 

Lisbon, Portugal 2000

 

 

1     Abstract

2     Introduction

3     Primary Production

3.1          Identification of the main sources of organic matter

3.1.1              Introduction

3.1.2              Methodology

3.1.3              Results

3.2          Mycrophitobenthos production

3.2.1              Introduction

3.2.2              Methodology

3.2.3              Modelling

3.2.4              Results

3.2.5              Discussion

3.2.6              References

4     Secondary production

4.1          Benthic meiofauna

4.1.1              Introduction

4.1.2              Methodology

   Study area

   Sampling

   Sediment analysis

   Meiobenthos analysis

   Data analysis

4.1.3              Results

4.1.4              Discussion

4.2          Benthic Macrofauna

4.2.1              Sampling sites

4.2.2              Methodology

Sediment analysis

4.2.3              Results

   Sediment analysis

   Species composition

   Seasonal Analysis

   Specific richness, diversity and evenness

4.3          Fish assemblages in tidal salt marsh creeks and in adjoining mudflat areas in Tagus estuary

4.3.1              Introduction

4.3.2              Methodology

   Study Area

   Data analysis

4.3.3              Results

   Abiotic conditions

   Community structure

   Seasonal variation

4.3.4              Discussion

5     Trophic analysis

5.1          Food habits of Pomatoschistus microps (Krøyer, 1838) in the Tagus salt marsh

5.1.1              Introduction

5.1.2              Methodology

5.1.3              Results

5.1.4              Discussion

5.2          Food habits of Liza ramada (Risso, 1826) in the Tagus salt marsh

5.2.1              Introduction

5.2.2              Methods

5.2.3              Results and discussion

5.3          Isotope analysis

5.3.1              Methods

5.3.2              Results

5.4          Food web

6     References

7     Publications

 

 

1           Abstract

 

In the latest projects “Comparative studies of salt marsh processes” and “effects of environmental change on European salt marshes” our team investigated the structure and functioning of the food web from the salt marsh surrounding areas of the Mira and Tagus estuaries and compared them.

In the follow up of the latest projects, EUROSAM, we pretend to study the Tagus estuary vertebrate food web within the different salt marsh compartments, such as mudflats, creeks and vegetation cover areas, to understand the importance of the direct usage of those habitats by the fish community. However studies in these areas are scarce and there is no information on the aquatic fauna present in these salt marshes. For that purpose first we identified quantitative and qualitative trends in the fish, decapod crustaceans and benthic meio and macroinvertebrates communities and we estimated the primary productivity rates of intertidal microphytobenthos of the Tagus estuary. After choosing k-species we followed-up the food web from the lowest levels, identifying and characterising the main potential sources of organic matter, to the highest ones.

 

When considering the entire intertidal area of the Tagus estuary, microphytobenthic primary production was estimated to attain 4265.1 ton C yr-1.

Detritivorous and benthic invertebrate feeders dominate the higher levels of the food web in the salt marsh tidal creeks. The major difference between the salt marsh and the adjoining area food webs is the absence or presence in lower abundance’s of crabs and several fish species, as the ell, the sea bass, the sand goby and the soles. Some of those species are potential predators of small size fish. Thus, young of the year of several fish species feeding in the salt marsh areas decrease the risk of being preyed by other fish species.

The use of the tidal creeks as feeding grounds by a high fish biomass suggest the important roll of these taxa in the organic matter transport processes between the salt marsh and the adjoining areas.

Part of the results of this task are presented in one Ph.D. Thesis (Serôdio 1999) and in two articles submitted for publication (Serôdio et al. and Salgado et al.).

 


2           Introduction

 

In the latest projects “Comparative studies of salt marsh processes” and “effects of environmental change on European salt marshes” our team investigated the structure and functioning of the food web from the salt marsh surrounding areas of the Mira and Tagus estuaries and compared them.

In the follow up of the latest projects we pretend to study the Tagus estuary vertebrate food web within the different salt marsh compartments, such as mudflats, creeks and vegetation cover areas, to understand the importance of the direct usage of those habitats by the fish community. However studies in these areas are scarce and there is no information on the aquatic fauna present in these salt marshes. For that purpose first we identified quantitative and qualitative trends in the fish, decapod crustaceans and benthic invertebrates communities and we estimated the primary productivity rates of intertidal microphytobenthos of the Tagus estuary. After choosing k-species we followed-up the food web from the lowest levels, identifying and characterising the main potential sources of organic matter, to the highest ones.

Part of the results of this task are presented in one Ph.D. Thesis (Serôdio 1999) and two article submitted for publication (Serôdio et al. submitted and Salgado et al.) included as annexes.

 

 

3           Primary Production

 

3.1           Identification of the main sources of organic matter

 

3.1.1            Introduction

 

The main goal of this study is to identify the main sources of organic matter for the aquatic consumers in salt marsh and contiguous mudflat of the upper Tagus estuary.

For that purpose we determined the C, S and N isotopic relative abundances of the main primary producers and the amount of chlorophyll a (Chl a), suspended particulate matter (SPM) and particulate organic matter (POM) in several areas of the estuary.

Stable isotopes values of POM and of the filter feeder Scrobicularia plana were analysed. POM isotopic values represent the spatial variability of both plant detritus and phytoplancton, suspended in the water column. Isotope values of S. plana are a signal of the POM that is assimilated.

 

 

3.1.2           Methodology

 

The macrophytes Spartina maritima, Arthrocnemon fruticosum and Halimione portulacoides were sampled during Spring in Hortas and Vasa sacos creeks. Water samples for Chl a, SPM and POM analysis were collected also during Spring at high tide inside both tidal creeks in the nearby subtidal areas and in the middle of the estuary and at low tide in three tributaries of the estuary (V. F. de Xira in the Tagus river, Porto alto in the Sorraia river and Ponte in the Enguias stream). These water samples were stored in dark until filtration in the laboratory was done. Water samples for POM isotope analyses were vacuum filtered until clogged onto precombusted filters. Individuals of S. plana for isotope analysis were collected from the mouth of both tidal creeks.

Samples for analysis of multiple stable isotopes were processed following the sample preparation guidelines of the Stable Isotope Laboratory, Ecosystem Center, MBL (MA, USA).

 

Tissues of macrophyte species were cleaned of mud and when present, epiphytes were removed by scraping with a razor blade. Samples were then dried to constant weight at 60 ºC. The dried tissues were ground to a fine powder with a mortar and a pestle. Plant tissue samples for 13C were checked for contamination by carbonates. Subsamples of the ground sample were acidified with several drops of 10% HCl while being observed under a dissecting microscope. If bubbling occurred the whole sample was acidified, redried at 60 ºC and stored in glass vials. Plant tissue samples for 34S were ground and rinsed in deionised water to remove seawater sulphate (resuspended in deionised water, centrifuged for 5 min, and supernatant discarded; this procedure was repeated three times), redried at 60 ºC and stored in glass vials.

Animal tissue samples for C, N and S stable isotope analysis were dissected to isolate muscle tissue and were dried at 60 ºC. The dried tissues were ground to fine powder with a mortar and a pestle. Ground animal 13C tissue samples, suspected of having carbonate contamination, were acidified with 10% HCl and redried at 60 ºC. Ground animal tissue samples for 34S rinsed in deionised water to remove seawater sulphate and redried at 60 ºC.

Each isotope analysis of a species represented a subsample from a pooled sample of several individuals. This was done to minimise the variability associated with analysis of different individual organisms and to gain enough material for S isotope analysis.

Chl a was measured using the methodology presented by Lorenzen (1967).

 

 

3.1.3           Results

 

Due to problems in the mass spectrometer that is running the samples for the stable isotope analysis it is not possible yet to present any results on these parameters. However we expect to have them at any moment. More samples for this part of the work were collected in the beginning of junne and are still being analysed. Nevertheless, here we present the results of the spatial distribution of suspended particulate matter (SPM), particulate organic matter (POM) and chlorophyll a (Chl a) in surface waters of Tagus estuary (table I).

 

Table I - Spatial distribution of suspended particulate matter (SPM), particulate organic matter (POM) and chlorophyll a (Chl a) in surface waters of Tagus estuary

 

Stations

SPM

(g.m-3)

POM

(g.m-3)

Chl a

(g.m-3)

VFX

30.0

12.0

3.2

A

33.0

13.2

1.6

Pa

90.5

21.0

1.7

H

178.8

48.7

8.7

Vs

51.0

17.7

4.4

 

 

3.2           Mycrophitobenthos production

 

3.2.1           Introduction

 

The primary production of intertidal microphytobenthos of the Tagus estuary, Portugal, was quantifyied through the formulation of a simulation model. The variables used as forcing functions in the model, in situ irradiance, productive biomass and P vs. E curve parameters a and Pm, were measured with hourly resolution. The model was used for comparing the variability in production on hourly (intraday), fortnightly (within spring-neap tidal cycles) and seasonal (month-to-month) time scales. The model allowed for the prediction of hourly production rates for the whole year and for the estimation of annual primary production of the Tagus estuary.

 

 

3.2.2           Methodology

 

Primary productivity rates of intertidal microphytobenthos of the Tagus estuary were estimated from in situ time series of measurements of photosynthetic active radiation (PAR) and temperature and from photosynthesis versus irradiance (P-I) curves measured at different times during low tide periods.

Sampling was carried out in an extensive intertidal mudflat located near Pancas salt marsh, on the south margin of the Tagus estuary, Portugal. The sampling site was representative of the intertidal areas of the estuary, being composed of fine muddy sediment, with 90.6% of particle sizes below 20 µm, and colonised by microphytobenthic communities typically dominated by diatoms (Brotas and Plante-Cuny 1998). Sediment samples were collected during low tide using plexiglas corers (1.9 cm internal diameter) and taken to the laboratory where were placed outside, in an artificial tidal system which simulated the immersion during high tide using estuarine water collected on the day of sediment sampling.

During three spring-neap tidal cycles, photosynthetically active radiation (PAR, 400-700 nm) was measured continuously at the sediment surface at the sampling site, using an underwater quantum sensor (LI-192SA, LI-COR, Nebraska, Lincoln, USA) connected to a data logger (Delta-T Logger, Delta-T Devices, Cambridge, UK), positioned ca. 10 cm above the sediment surface.

 

Primary production was determined by measuring gross oxygen photosynthetic production, using Clark type oxygen microelectrodes (5-20 µm tip 737-GC model, Diamond General) according to Revsbech and Jørgensen (1983). The microelectrode was positioned vertically within the sediment using a micromanipulator and was connected to a picoammeter (model 8100 Electrometer, Keithley Instruments, Cleveland, Ohio, USA). Photosynthesis was measured at depth intervals of 50 µm and then integrated over all depth intervals to yield the community photosynthetic rate. P-I curves were constructed by exposing the same sediment core to eight different incident irradiance levels and measuring the community photosynthetic rate under each level.

Productive biomass was estimated by measuring dark-level Chl a fluorescence, Fo, using a pulse amplitude modulation fluorometer (PAM 101 Chlorophyll Fluorometer, Heinz Walz, Effeltrich, Germany). The PAM fiberoptics was positioned perpendicularly to the sediment surface, at a fixed distance of 1 mm, and readings were considered after signal stabilisation following darkening of the sample. The relative positions of the sample surface, the fiberoptics and the oxygen microelectrode were set with the help of a micromanipulator (MM33, Diamond General, Ann Arbor, Michigan, USA) to which the corer was attached.

 

 

3.2.3           Modelling

 

The patterns of variation of PAR and productive biomass were used as forcing functions in the calculation of annual primary production rates of intertidal microphytobenthos for different tidal heights. A linear relationship between productive biomass and P-I curve parameters was used to estimate photosynthetic rate at each point in time and for different tidal heights. Fo was used to predict short-term variations in the community photosynthetic light response, by quantitatively relating Fo to P vs. E curve parameters.

Based on the in situ measurements of the irradiance level reaching the sediment surface during whole spring-neap tidal cycles (Serôdio & Catarino 1999), the model was run considering that during high tide the irradiance at the sediment surface was null.

Since the relationship between Fo and the P vs. E curve parameters allows for the direct estimation of hourly production rates from Fo and E observations, the estimation of annual production budgets was approached through the modelling of the Fo variability throughout an entire annual period. The model used for describing the hourly variation of Fo is a modification of the model formulated by Pinckney & Zingmark (1991).

The relationship between Fo and a and between Fo and Pm was quantitatively defined by calculating the slope and the y-intercept of linear regression equations fitted to paired measurements of Fo and a and of Fo and Pm.

 

 

3.2.4           Results

 

PAR reaching the surface of intertidal sediments was strongly conditioned by the periodic tidal inundation, with large and abrupt variations occurring during flooding and ebbing (Fig. 1). PAR levels decreased very rapidly to null or very low values with the incoming of the tide during most of the daytime immersion periods. Only on neap tides, when the water column during high tide is lower, some light reached the sediment surface. The variation of the time of day of tidal emersion along the Spring-neap tidal cycle resulted in a clear fortnightly variation in total daily PAR reaching the sediment surface. Maximum daily PAR values were recorded on the days when tidal inundation occurred at the beginning and at the end of the day, which coincides with the days preceding the neap tides. Minimum values were observed on the days preceding the Spring tides, in which tidal inundation occurs during the middle of the day.

 

The photosynthetic light response was also found to vary significantly, mainly on hourly time scale (Fig. 2). This variability was shown to result mainly from rapid and substantial changes in productive biomass, caused by vertical rhythmic migration of motile diatom cells (Serôdio et al., 1997).

A linear relationship was found between Fo and P vs. E curve parameters a and Pm, either when considering each fortnight period separately or when pooling the data from the three periods. Highly significant regression equations were found in all cases when a and Pm were regressed against Fo.


 

Figure 1 - Example of PAR (filled area) and temperature (line) variability at the surface of intertidal sediment during on Spring-neap tidal cycle. Horizontal bars and vertical dotted lines mark the high tide periods.


 

Figure 2 - Typical short-term variation photosynthetic light response of undisturbed microphytobenthos.

 

As a result of high variability in both incident PAR and community photosynthetic efficiency, primary production rates were found to vary considerably on hourly and fortnightly time scales, resulting in an annual pattern of daily primary production characterised by a strong fortnightly quasi-cyclic variability superimposed on an underlying seasonal trend (Fig. 3).

 

Figure 3 - Model-predicted annual variation of mean daily production rate.

 

In general, maximum hourly production rates were obtained near neap tides, during days when the low tide exposure occurred near the middle of the day, and minimum rates near Spring tides, when the opposite situation occurred. Spatially, microphytobenthic primary production was predicted to increasing with tidal height, which determines the number and extent of exposure during low tide and, hence, the annual amount of light available for photosynthesis.

When the model was run for 4 different years, the pattern of daily production was essentially the same and the estimates of annual production varied by less than 7%. For this 4-year period, annual areal production ranged from 12.47 to 13.34 mol O2   m-2 yr-1 or, converting oxygen production (mol O2) to carbon assimilation (g C) assuming a 1:12 ratio (Cammen 1991, Pinckney & Zingmark 1993a), from 149.6 to 160.1 g C m-2 yr-1. The estimation of the annual primary production of the entire intertidal area of the Tagus estuary was approached by running the model for the whole range of tidal heights (from 0.1 m to 4.1 m, at 0.1 m intervals) and by extrapolating the annual areal production values calculated for each tidal height for the total intertidal area corresponding to the same tidal height interval. For each tidal height interval, the total intertidal area was calculated from average heights of 300 m x 300 m areas computed from bathymetry charts for the Tagus estuary. From the annual areal primary production rate predicted for each tidal height interval, an annual production map for the intertidal areas of the Tagus estuary was constructed (Fig. 4). For the total 114.48 km2 of intertidal area of the Tagus estuary, mean annual primary production was estimated to reach 4265.1 ton C yr-1.

 

 

3.2.5           Discussion

 

The interference between the tidal and the day/night cycles cause the estuarine intertidal environment to be dominated by strong variability in a number of different time scales. The use of a sampling design of this type allowed for the finding that microphytobenthic primary productivity is dominated by variability on sub-seasonal (hourly and fortnightly) time scales. These results have important consequences for the design of sampling programs for the characterization of the temporal patterns of variation of microphytobenthic productivity, as the usual practice of measuring production monthly throughout the year does not provide information on the within-month variability and may yield erroneous seasonal pattern due to liasing.


 

Figure 4 - Estimated spatial distribution of microphytobenthic annual primary production for the entire intertidal area of the Tagus estuary.

 

 

The predominance of fortnightly over seasonal variability in microphytobenthic productivity predicted for the Tagus estuary is not likely to be a generalised feature of estuarine ecosystems, as it is determined by factors associated to the particular geographic location of each estuary. In general, seasonality is expected to increase with latitude, following the higher month-to-month variations in irradiance and temperature, as shown in a recent comparison of results from different estuaries (MacIntyre et al., 1996). In the Tagus estuary, the fortnightly variability in daily production is caused by the variation in the daily available irradiance along the Spring-neap tidal cycle: during Spring tides, low tide occurs early in the morning and late in the afternoon; conversely, during neap tides, low tide occurs in the middle of the day. In other estuaries, different tidal regimes and less turbid water may reduce the relative importance of the Spring-neap cycle on benthic production.

The coincidence of maximum daily production rates with minimum resuspension of benthic biomass during neap tides (considering the tidal amplitude as an indicator of the resuspension intensity) is likely to cause a marked fortnightly variability in the availability of microphytobenthic biomass for the estuarine food web. In the period between Spring and neap tides, the simultaneous increase in daily production and decrease in resuspension intensity favours the accumulation of microphytobenthic biomass on the tidal flats, but at the same time a reduction in the transfer of benthic biomass for the pelagic food web. On the opposite phase of the Spring-neap cycle, while biomass on the tidal flats is expected to decrease following to the reduction of daily production and the increase in resuspension rates, the availability of benthic microalgae for filter feeders or benthic grazers on other areas of the estuary is expected to increase. Such a process would contribute to attenuate the impact of the fortnightly variation in intertidal primary production at the ecosystem-level primary and secondary productivity.

Considering the interannual variation in the annual estimates, a maximum range of 149.6 - 160.1 g C m-2 yr-1 and a mean value of 155.8 g C m-2 yr-1 are obtained for the annual areal production rate of the studied intertidal microphytobenthos of the Tagus estuary. These values are within the range of estimates reported for other estuaries (for a recent review, see MacIntyre et al., 1996), and close to previous estimates made for this estuary. Based on measurements of community net photosynthesis, Brotas & Catarino (1995) estimated annual production as 47 and 178 g C m-2 yr-1, values calculated for sites at tidal heights of 1.4 m and 3.1 m, respectively. The use of different methodologies to measure microphytobenthic photosynthesis and the different methods used to extrapolate from hourly to annual production rates, makes difficult a direct comparison between the estimates obtained in both studies. However, when the present model is run for the tidal height of 3.1 m, and the resulting estimate is converted to net production (80% of gross production, value estimated for similar microphytobenthos communities in the Tagus estuary), a mean value of 164.9 g C m-2 yr-1 is obtained, 7.3% lower than the estimate of Brotas & Catarino (1995).

Because the model was not validated for tidal heights or locations other than the used for estimating its parameters, some caution must be used in the analysis of the results obtained concerning the estimation of annual production rate for the whole estuary. The main causes for failure in the model predictions are the variation in the fraction of microalgae that are motile and overall biomass, which are usually lower in sandier sediments, and the effects of desiccation on sites that are not inundated during neap tides (tidal height above 2.9 m). Also variations over space in the community physiological light response, associated to differences in taxonomic composition or to photoacclimation to different light regimes, may contribute for errors in the model predictions.

The annual gross primary production for the whole intertidal area of the Tagus estuary estimated by taking into consideration the representativeness of the different tidal heights and the production estimated for each tidal height interval, 4265.1 ton C yr-1, was substantially different from the value obtained by directly extrapolating the areal production rate estimated at a single tidal height for the entire intertidal area, 17858.9 ton C yr-1. This highlights the importance of the development and validation of improved production models that consider the variability of production rates with tidal height in the estimation of ecosystem-level production.

 

 

3.2.6           References

 

Brotas V, Catarino F (1995) Microphytobenthos primary production of Tagus estuary intertidal flats (Portugal). Netherlands Journal of Aquatic Ecology 19:333-339.

Brotas, V. & M. R. Plante-Cuny (1998) Spatial and temporal patterns of microphytobenthic taxa of estuarine tidal flats in the Tagus Estuary (Portugal) using pigment analysis by HPLC. Marine Ecology Progress Series 171: 43-57.

Cammen LM (1991) Annual bacterial production in relation to benthic microalgal production and sediment oxygen uptake in an intertidal sandflat and an intertidal mudflat. Marine Ecology Progress Series 71:13-25.

MacIntyre HL, Geider RJ, Miller DC (1996) Microphytobenthos: The ecological role of the “secret garden” of unvegetated, shallow-water marine habitats. I. Distribution, abundance and primary production. Estuaries 19:186-201.

Pinckney J, Zingmark RG (1991) Effects of tidal stage and sun angles on intertidal benthic microalgal productivity. Marine Ecology Progress Series 76:81-89.

Pinckney JL, Zingmark RG (1993) Modelling the annual production of intertidal benthic microalgae in estuarine ecosystems. Journal of Phycology 29:396-407.

Revsbech, NP & BB Jørgensen. 1983. Photosynthesis of benthic microflora measured with high spatial resolution by the oxygen microprofile method: Capabilities and limitations of the method. Limnology and Oceanography 28: 749-756.

Serôdio J 1999 Modelling the primary productivity of intertidal microphytobenthos. Role of migratory rhythms studied by in vivo chlorophyll fluorescence. University of Lisbon, 168 p.

Serôdio J, Catarino F 1999 Fortnightly light and temperature variability in estuarine intertidal sediments and implications for microphytobenthos primary productivity. Aquatic Ecology 33:235-241.

Serôdio J, Marques da Silva J & Catarino F 1997 Nondestructive tracing of migratory rhythms of intertidal microalgae using in vivo chlorophyll a fluorescence. Journal of Phycology 33:542-53.

Serôdio J, Marques da Silva J & Catarino F. Use of in vivo chlorophyll a fluorescence to quantify short-term variations in the productive biomass of intertidal microphytobethos. Submited for publication.

 

 

4           Secondary production

 

4.1           Benthic meiofauna

 

4.1.1           Introduction

 

The importance of meiofauna in the estuaries is very high and according to Coull (1999) it plays a role in three fields: it facilitates biomineralization, feeds various higher trophic levels and show a high sensibility to antropogenic actions, making them excellent organisms for estuarine pollution biomonitoring.

According to Heip et al. (1992) the latitude has influence on the meiobenthic communities; therefore the majority of studies have been conducted at latitudes different from the Tagus estuary latitude, mainly in Northern Europe (e.g. Merilaeinen, 1988; Escravage et al., 1989; Vanreusel, 1991; Heip et al., 1992; Steyaert & Vincx, 1996; Moodley et al., 1998a; Blome et al., 1999) and in Southern EUA (e.g. Bell, 1979; Fleeger et al., 1982; Montagna et al., 1983; Coull, 1985; Coull, 1999).

Meiofauna are badly known in Southern Europe estuaries. In Portugal their knowledge is limited to the work of Rosado & Bruxelas (1995), Rosado (1996), Adão & Marques (1999) and Austen et al. (1989). In Tagus estuary there is only a study comparing the communities in several European estuaries (Heip & Herman, 1995). In this work only four sites were sampled in a single season (Soetaert et al., 1995).

Meiofauna are characterised by high densities (between 105 e 107 ind m-2) (Coull, 1999). In temperate estuaries this group is dominated by nematodes and followed generally by copepods (Heip et al., 1990; Heip et al., 1995), although there may be significant differences in sub-dominant groups (Soetaert et al., 1995). Spatial distribution of meiofauna in sediment follows an aggregated pattern (Fleeger et al., 1990; Steyert et al., 1999) both horizontally and vertically. Although this is a complex issue since environmental gradients are sharper vertically, it is also in this direction that fauna and flora show a strong vertical zonation (Joint et al., 1982).

The processes that generate and maintain the vertical distribution pattern in different places are scarcely known and according to Steyaert et al. (1999) they are an important challenge for modern ecology research.

In temperate zones, tidal and shallow subtidal meiofauna typically shows a seasonal pattern (Smol et al., 1994). According to Coull (1999) this seasonality is directly or indirectly related to temperature pattern by means of factors induced by it. According to the same author (1985) other biological processes may also have a bear on meiofauna seasonality.

Although other groups such as microphytobenthos (Brotas, 1995, Brotas et al., 1995), macrofauna (Calvário, 1982; Costa et al., 1996; Costa et al., 1999), ichthyofauna (Costa, 1982; Costa et al., 1996; Cabral, 1998) and birds (Moreira, 1995), are relatively well studied in the Tagus estuary, the knowledge of meiofauna is practically non existent.

For a better understanding of meiobenthic community in the Tagus estuary we intend to establish their composition in terms of great taxonomic groups, their vertical distribution pattern and their variation in time.

 

 

4.1.2           Methodology

 

Study area

The Tagus estuary on the Portuguese Western coast (38°44’N, 9°08’W) is the largest Portuguese estuary and one of the largest estuaries in Europe. It covers an approximate area of 320 Km2, where 40% are tidal areas (Bettencourt, 1979) made up basically of sand and mud flats and some salt marshes. The largest area of the tidal platforms is on the Southern margin.

In the Tagus estuary the tides are semi-diurnal with amplitudes higher than in the ocean, varying from 2 to 4,6m (Costa, 1999).

The salinity of an average high tide varies from typical values of 36‰ in the coastal zone to 27‰ in the widest part of the estuary, with a quick decreasing gradient upstream with values of 0,5‰ 50km from the mouth of the river. In low tide the values at the mouth of the river are 30-33‰, maintaining the sharp gradient from the widest part of the estuary up to Vila Franca de Xira (Costa, 1999).

The temperature amplitude is also higher in the estuary than in the ocean, with a highest amplitude in the upper zones of the estuary (Bettencourt, 1979).

The sampling sites are on the Southern margin near Alcochete (Fig. 5)

 

Figure 5 - Tagus estuary map showing the sampling sites (square). The tidal areas are shown in brown.

 

Two sampling sites were chosen in the tidal zone: Hortas and Eucaliptal. The first site, Hortas, is a mud flat with a salt marsh on one side and a small water channel, Ribeira das Enguias, on the other side. The sediment is muddy and is little disturbed by man since it is a difficult place to reach, even though it is common to observe in low tide high concentrations of birds feeding in that area, what causes some disturbance to the sediment.

The other sampling site, Eucaliptal, is an area located on the other margin of Ribeira das Enguias, and is very easy to reach. The sediment is more heterogeneous and sandier. Is a preferential place for placing boats and catching clams and crabs, and so it is more disturbed by man.

 

Sampling

Six cores were taken in each sampling site, using a piston-style core with a 3.57cm inner diameter and buried 10cm deep, according to Fleeger et al. (1988), Thiestle & Fleeger (1988) and Soetaert et al. (1994). The 6 cores comprehend 2 replicates taken in 3 different spots, one replicate for determining the meiobenthic community and the other for the analysis of chlorophyll a and pheopigments analysis. Two cores seem to be the minimum accepted for analysing the meiobenthic community; 3 cores are accepted as an adequate number of cores for this kind of study (Bouwman, 1987; Steyaert com. pess.).

Each core was cut in four layers, 0-1cm, 1-3cm, 3-6cm and 6-10cm. In each site, sediment samples were also taken to determine the granulometry, water content and organic matter content in the sediment. These samples were also cut in the same layers mentioned above. Samples were taken in Summer and Autumn.

The identification of the samples includes two letters and one number: the first letter refers to the sampling site, H to Hortas and E to Eucaliptal; the second letter refers to the season, S to Summer and A to Autumn; the number indicates the layer, 1 for the first layer (0-1cm), 2 for the second (1-3cm), 3 for the third (3-6cm) and 4 for the fourth (6-10cm).

 

Sediment analysis

The samples for granulometry were dried at 60ºC for 24h. About 100g dry weight were washed in a 63µm sieve to wash out the mud. The sediment retained in the sieve was dried again at 60ºC for 24h and passed through two sieves of 2mm and 63µm mesh. The sediment retained and centrifuged in each sieve was then weighted to determine the mud content (the part smaller than 63µm), the sand content (the fraction smaller than 2mm but bigger than 63µm) and the gravel content (the fraction bigger than 2mm).

Water content was assessed by drying the samples at 60ºC for 24 hours while the total organic matter was determined after destruction in a muffle furnace (2h at 450ºC).

 

The pigments, chlorophyll a and pheopigments were extracted with 90% Acetone for 24h in dark at 4ºC, and then centrifuged. The chlorophyll a, normally used, as an indicator of the microphytobenthos biomass, and the pheopigments concentration was determined with a spectrophotometer, accordingly to the methodology presented by Lorenzen (1967). The pigment concentration was expressed in µg/g of dry sediment.

 

Meiobenthos analysis

The sediment samples for the determination of the meiobenthic community were placed in plastic bottles identified and fixed with a 4% formaldehyde solution neutralised with Borax (Pfannkuche & Thiel, 1988).

The sediment was then washed through two sieves with a 1mm and 38µm mesh and the sediment in the 38µm sieve was centrifuged in a LUDOX HS-40. Each sample was centrifuged three times for 10 min, at 2600rpm, in 400ml recipients. The meiofauna retained in the supernatant was removed after each centrifuge. Finally all meiofauna was keep in a 4% formaldehyde solution neutralised with Borax and stained wit Bengal rose.

The organisms were classified in the main taxonomic groups and counted with binocular and/or microscope. The densities of each group were calculated. The density unit most used is the number of individuals per 10cm2 of the sediment surface. This unit is used to compare equal sediment volumes; since in this work the sediment volume of each layer is different, the unit used was the number of individuals per 10cm3. When the whole core is considered and compared with the other cores the volume is the same and then the unit used was ind/10cm2, which allows the values to be compared with data of other studies.

Data analysis

The community structure was evaluated through a correspondence analysis, using the CANOCO (4.0 version) program. The groups with a low occurrence (present in less then 10 of the 48 samples) were excluded (i.e. Bivalvia, Gastropoda, Isopoda, Tardigrada and Insecta). The data referring to water, organic matter and mud content, and the chlorophyll a and pheopigments concentration were included as co-variables.

 

 

4.1.3           Results

 

The values of the water, organic matter and mud content in the sediment are shown in table I.

Since the gravel content was not relevant, table I only shows the value of the mud content, the other being the sand content. In Hortas, the mud content was always higher than 99%. In Eucaliptal, the sediment was sandier, especially in Autumn when it varied from mud in the first layers to sand in the fourth layer.

In general a decrease of the water and organic matter content decreased with depth.

 

Table I – Percentage of water (WC), organic matter (OM) and mud content in the sediment for the two sampling sites in Summer and Autumn.

 

Station

Layer

WC

OM

Mud

HS

1

49

7,1

99,7

2

44

6,7

99,8

3

47

6,9

99,8

4

46

6,9

99,6

ES

1

56

8,4

89,3

2

55

8,4

91,1

3

51

8,6

91,2

4

44

7,6

87,6

HA

1

64

8,4

99,4

2

54

7,9

99,7

3

51

7,9

99,6

4

47

8,0

99,5

EA

1

54

8,5

85,6

2

49

5,5

58,3

3

34

4,0

43,0

4

30

4,0

36,0

 

The values of chlorophyll a and pheopigments concentration in the sediment and the chlorophyll a / pheopigments ratio are shown in figure 6.

 

a                                         b                                     c                                  d

Figure 6 - Mean values of chlorophyll a and pheopigments concentration in the sediment and chlorophyll a / pheopigments ratio at the two sampling sites in Summer and Autumn. (a – Hortas in Summer; b – Eucaliptal in Summer; c - Hortas in Autumn; d – Eucaliptal in Autumn).

In general the values of chlorophyll a and pheopigments concentration reached their maximum in the first layer and decreased with depth. Higher values of chlorophyll a were observed in Autumn.

The pheopigments presented higher values than chlorophyll a. In Hortas these values were higher in Summer than in Autumn, while in Eucaliptal the values related to Autumn were higher in the first layers and lower in the deepest layers.

The chlorophyll a / pheopigments ratio was higher in the first layers and decreased with depth. This ratio increased from Summer to Autumn.

From the groups considered by Higgins & Thiel (1988) the following taxa were identified in the samples collected in Hortas and Eucaliptal: Sarcomastigophora, Ciliophora, Turbellaria, Nematoda, Rotifera, Polychaeta, Oligochaeta, Tardigrada, Ostracoda, Copepoda, Isopoda, Halacaroidea, Insecta, Gastropoda e Bivalvia. Nauplii of crustaceans were also found in great quantities and although they do not represent a taxonomic group they were also included in this study.

Taking into account all the samples observed, the meiofauna density varied from 0,3 to 438,2 individuals per cm3 of sediment. The density and diversity of taxonomic groups were greater in the first layers, decreasing with depth. The densities of the meiobenthic groups are shown in figure 7.

 

 

Figure 7 - Mean density values of the meiobenthic groups considered at the two sampling sites in Summer and Autumn.

 


As regards the vertical distribution and the dominant groups, there was a clear difference between the two seasons. In Summer more than 90% of the individuals were found in the two first layers, from which 69% (Hortas) and 76% (Eucaliptal were on the first layer (fig. 8). In Autumn the individuals had a deeper distribution, reaching 90% only in the third layers.

 

 


Figure 8 - Mean percentage of individual in each layer and respective mean density value (ind/cm3) at the two sampling sites in Summer and Autumn.

 

The dominance of the different groups is shown in figure 9. The Nematoda were always the dominant group in the first three layers. In Summer for both sites the dominant group was the Sarcomastigophora, while in Autumn the Nematoda group continued to be dominant, at both sampling sites.

Figure 9 - Relative mean percentage of the meiobenthic groups considered in each layer in Summer and Autumn at the two sampling sites.

 

In Summer the diversity was lower than in Autumn. Only the groups Nematoda, Copepoda, Sarcomastigophora and the crustacean nauplii presented high densities, while in Autumn the groups Rotifera, Turbellaria and Ciliophora presented high densities too.

 

In general at Eucaliptal the densities were higher than at Hortas, and there was an increase of the meiobenthic density from Summer to Autumn although this increase was not as clear in Eucaliptal (48%) as in Hortas (76%) (Fig. 10).

 

 

Figure 10 – Mean density (ind/10cm2) of the meiobenthic groups per core at the two sampling sites, in Summer and Autumn. The value of that mean density for the group Nematoda is also indicated in the figure together with the percentage of Nematoda in relation to the total individual number.

 

The density of the groups Nematoda, Copepoda, Polychaeta, Oligochaeta and Ostracoda and also the crustacean’s nauplii increased from Summer to Autumn. The groups Rotifera, Turbellaria and Ciliophora occurred almost exclusively in Autumn. The density of the group Sarcomastigophora decreased from Summer to Autumn. Although the density of the group Bivalvia apparently increased from Summer to Autumn, and the groups Gastropoda, Insecta, Isopoda and Tardigrada only occurred in Autumn, these groups presented very low densities which did not make it possible to establish a pattern.

The results of the correspondence analysis are shown in table II and figure 11.

 

Table II Eigen values and variance cumulative percentage.

 

Axes

1

2

3

4

Total inertia

Eigen values

0.516

0.377

0.176

0.142

1.511

Species-environmental correlation

0.774

0.687

0.345

0.373

 

Species variance cumulative percentage

34,1

59,1

70,7

80,1

 

Species-environment variance cumulative percentage

53,3

84,0

87,7

91,1

 

 

 

The two first factorial axes explain 59% of the variance related to the meiobenthic groups and 84% resulting from the relation between these groups and the environmental variables considered.

 

Figure 11 – Top: Representation of the meiobenthic groups in the first factorial plan.

Bottom: Representation of the vectors related to the environmental parameters.

 

As we can see in figures 11 and 12 the second, third and fourth layers do no show any apparent difference between the two sampling sites, and that is why they are grouped. For the first layer the items related to Eucaliptal are circumscribed to a smaller area, although they are in the same area than Hortas.

The first layers are grouped according to the two seasons. In Summer the second, third and fourth layers are all grouped, while in Autumn they are individualised. The fourth layers do not show differences between the two seasons.

The trends observed in this analysis are the following: two groups related to the first layer, one the Summer (HS1 and ES1) and another for Autumn (HA1 and EA1); a group related to the second layers in Autumn (HEA2), another group related to the third layers in Autumn (HEA3); and a last group including the second and third layers in Summer (HES2 and HES3) and all the fourth layers (HE4).

 

 

Figure 12 – Identification of the different groups resulting from the correspondence analysis. The arrows indicate the depth increase: the blue arrow in Summer and the black arrow in Autumn.

 

The groups Rotifera, Turbellaria e Ciliophora and the crustacean nauplii are associated to the first layers in Autumn. The Sarcomastigophora are associated to the first layers in Summer. The Polychaeta occurred deeper, that is why they are associated to deeper layers. Although the other groups also occurred in deeper layers they presented higher densities in the first layers and that’s why they are closer to them.

The chlorophyll a and pheopigments concentration and the water and organic matter content varied inversely to depth and presented higher values in Autumn what might have conditioned the direction of the vectors that are pointing to the fist layers, mainly the Autumn ones. The organic matter content increase from Summer to Autumn was not very significant, so the related vector is in an intermediary position. The mud content is essentially associated to the Hortas samples since in this place the values were always greater than 99%.

 

 

4.1.4           Discussion

 

Oxygen availability is one of the main factors that condition the vertical distribution of meiofauna (Coull, 1988). According to Cartaxana & Lloyd (1999) the oxygen concentrations measured in muddy sediment of the low salt marsh of the Tagus estuary reach very low values already in the first millimetre. They continue to fall in the second millimetre and keep on falling, although not so sharply until they became undetectable at 14mm depth. Bioturbation also brings oxygen to anoxic zones (Fenchel, 1996; Forster, 1996). According to Forster (1996) the great spatial and temporal heterogeneity of these oxidation states can affect the meiofauna migration and the biogeochimical processes. Therefore and since the first layers are always muddy the sediment is oxidised only very close to the surface; in the second layer if oxygen is still present its level will be very low and it will disappear altogether in the third and forth layers.

Oxygen may also be present in anoxic zones due to the bioturbation already mentioned caused by tube-dwelling animals.

The oxygen availability limits the distribution of several meiobenthic groups that can be found almost exclusively in the first layer (Copepoda, Rotifera, Turbellaria, Ostracoda, Bivalvia, Gastropoda and nauplii).

Although the groups Nematoda, Sarcomastigophora, Polychaeta, Oligochaeta and Halacaroidea often show higher densities in the first layer they also attain high values and sometimes higher in lower layers. Apparently the groups Polychaeta, Oligochaeta and Halacaroidea don’t show any vertical distribution pattern. Since they are active diggers they can live at depths greater than the 10cm used in this study; this can explain why it is not possible to find any vertical distribution pattern.

In the Sarcomastigophora group the organisms are fixed. Therefore it may be difficult to distinguish living organisms from those that were dead by the time they were caught although they were also stained. For this reason the density values of this group may be overestimated for the lower layer. According to Gooday (1988) the Sarcomastigophora group is generally found near the sediment surface where they can find nutrients and interstitial water is well aerated. Since these organisms feed on algae (Gooday, 1988) and the food availability is higher on the sediment surface, as we can see by the chlorophyll a in figure 2, their distribution tends to decrease as the depth increases although according to Moodley et al. (1998b) the foraminifers (Sarcomastigophora) can migrate through anoxic zones what suggests that some of them are facultative anaerobes. According to the same author (1998c) the sulphide concentration (often correlated with the absence of oxygen) may be important to determine their distribution since they tolerate sulphide but only for survival effects, as they do not reproduce in its presence.

The Nematoda group has shown two different vertical distribution patterns according to the season. In Summer it occurred almost exclusively in the first layer, like the other groups, while in Autumn their abundance in the second layer was also very high and sometimes even higher. Although the Nematoda density has increased between Summer and Autumn, their density in the first layer decreased and there was a vertical redistribution of the individuals which were also found in the second layer. According to Montagna et al. (1983) in muddy sediments when the temperature rises the depth of the redox potential discontinuity decreases, i. e. moves closer to the surface. Thus due to the sun exposure and the high temperatures the sediment underwent in Summer the RPD depth is lower and the tolerance limits for each organisms moved upwards and influenced their vertical distribution. According to the same author, this effect also conditions the diatom distribution what explains the same patterns seen as regards the chlorophyll a concentration.

According to De Deckere et al. (in prep.) the Nematoda migrate during a tidal cycle and can migrate into the sediment reaching 15mm deep. Thus in favourable conditions exist in deeper layers the Nematoda could also have migrated into this layers.

As regards the groups Insecta, Isopoda and Tardigrada, the small number of individuals (2, 1 and 1 respectively) didn’t allow any conclusions to be taken as regards the vertical distribution patterns.

According to Soetaert et al. (1995) Nematoda were the dominant group in the Tagus estuary followed by Copepoda, while in this work the second dominant group for both places is Sarcomastigophora in both places and in Autumn, the Turbellaria group in Hortas and Ciliophora group in Eucaliptal, although the densities of other groups were also high.

As regards seasonality this is apparent in almost all meiobenthic groups, as well as the chlorophyll a values. This shows that microphytobenthos increased between Summer and Autumn too.

As can be seen in figure 8 the greatest differences occurred between the two seasons but there were also differences between the two sampling sites. These differences are more relevant in the upper layers and all the first layers are well separated and those relating to Summer are distinct from those relating to Autumn. As already seen the second and third layers relating to Autumn are also separated due to the great number of Nematode present in them. The fourth layers are separated neither by season nor by site. So the seasonality effect gradually disappears, as the depth increases, i. e. it is evident in the first layer but it is not seen between 6 and 10cm.

This decreasing effect of seasonality can also be seen for the chlorophyll a values which in the fourth layer are all equivalent to and in accordance with the values found by Brotas & Serôdio (1995) for the same depth in the Tagus estuary.

As it has already been said, according to Montagna et al. (1983) in muddy sediments when temperature rises the RPD depth decreases and animal densities also decrease. This appears to be the factor responsible for the lower densities this author obtained in Summer. In the Tagus estuary the exposed sediment is subjected to higher temperatures in Summer than in Autumn and this has a bear on the RPD depth.

In Hortas, the water content is rather lower in Summer than in Autumn. This difference is only seen in the upper layers what indicates that it has been evaporation, probably due to the higher temperatures felt in Summer and which results in a greater evaporation of interstitial water. In Eucaliptal, the granulometry of the sediment is rather different which difficult any comparisons as regards the water content since this is basically conditioned by the granulometry of the sediment (Calvário, 1982).

According to Coull (1985) in muddy sediments the seasonal abundance is negatively related to salinity. In Summer due to sun exposure, high temperatures and water evaporation, the salinity of interstitial water can reach high values which together with the high temperatures and the lower RPD depth may be responsible for the lower density and diversity seen in Summer.

The chlorophyll a/pheopigments ratio increases between Summer and Autumn which complies with the results of Brotas et al. (1995). The fact that more and better food was available, since the number of living algae increase as compared with detritus, has also provided better conditions for meiofauna growth as they mainly feeds on detritus, diatoms and bacteria (Coull, 1988).

From Summer to Autumn, the mean density of meiofauna per core increased in Hortas and also in Eucaliptal, although not in such a clear way. This was probably due to the fact that in the latter site the samples collected in Autumn were from sediments with a higher content of sand what can change the meiofauna densities and implies lower densities of microphytobenthos (Brotas et al., 1995; Brotas & Serôdio, 1995).

This increase from Summer to Autumn is also seen in the percentage of organic matter, what again indicates that the food availability increased from Summer to Autumn.

Although spatial correlation has already been documented (e. g. between Copepoda abundance and the microphytobenthos (Blanchard, 1990; Sandulli & Pinckney, 1999; Santos et al., 1995), this work didn’t show that type of correlation, as it would be necessary to use a rather different method especially aimed at that kind of study. In addition, according to Montagna et al. (1983) the physical factors apparently influence meiofauna and diatoms in the same way. So a simple correlation between the variation of chlorophyll a values and densities of meiobenthic groups may not mean a response from Copepoda to the varying quantity of food.

As Nematoda are the dominant group of the estuarine meiofauna the seasonality of meiofauna is closely linked to the seasonality of this group and according to Li & Vincx (1993) the abundance peaks of the Nematoda dominant species may occur in Spring, Summer, Autumn and Winter. This work did not cover an annual cycle so it could not be determined when the abundance peak of this group and all the meiobenthos in the Tagus estuary will occur.

According to Bouwman (1987) when there is seasonality, the density of estuarine meiofauna increases generally in Spring after the low values seen in Winter. A second peak can occur in Autumn although the abundance peak of each group may occur in other seasons of the year (Coull, 1988). This is the case of the Sarcomastigophora group that as opposed to the trend seen in other groups decreased its density between Summer and Autumn in Eucaliptal.

In most studies the abundance peak of meiofauna occurs at the end of Winter or in the beginning of Spring (Coull, 1985; Vincx, 1989; Rutledge & Fleeger, 1993; Li et al., 1996; Santos et al., 1996; Coull, 1999). In studies made in the Ems estuary the abundance peak is seen in Summer (Bowman et al, 1984; Essink & Keidel, 1998). In the Mira estuary, a study developed by Adão & Marques (1999) shows higher values for the Nematoda group at the end of the Winter.

Although there are exceptions, such as the work performed by Vincx (1989) during three years, the work of Li & Vincx (1993) that lasted for seven years and the work of Coull (1985) that lasted for eleven years and was extended for 22 years (Coull, 1999), great part of the studies that include time variations are based in periods of about one year (e.g. Bell, 1979; Montagna et al, 1983; Bowman et al, 1984; Rutledge & Fleeger, 1993; Blome & Faubel, 1996; Santos et al., 1996; Li et al., 1996; Essink & Keidel, 1998; Adão & Marques, 1999).

According to Coull (1985) the interannual variability is much greater than the seasonal variability and so the comparison of seasonality patterns should be considered with some care.

In the study performed by Soetaert et al. (1995) the samples were collected in April 1992. The values of the total abundance of meiofauna were rather lower than those found in this work. In the former work the highest mean density was approximately 2000 ind / 10cm2 and the lowest approximately 400 ind / 10cm2, while in this work the mean values per core varied from 3587 to 6965 ind / 10cm2 (fig. 6). Although in Soetaert et al. (1995) work the samples were collected only until 5cm deep and in this work until 10 cm deep, the difference between the densities is still high, since we found more than 90% of meiofauna in the first 6 cm deep (fig. 4).

As it has already been said, interannual variability is much greater than seasonal variability (Coull, 1985); so it can not be said that in the Tagus estuary the densities are lower in Spring than in Summer and Autumn, although in the work developed by Danovaro (1996) in the North of Italy the highest abundance peak was seen in October and the lowest in April. A similar seasonality pattern could exist in the Tagus estuary.

Since the sampling sites considered by Soetaert et al. (1995) show very different conditions of granulometry and salinity, it is difficult to compare the density and diversity seen in the two works. The four sampling sites are placed along a salinity gradient and the site that shows salinity equivalent to the two sites studied in this work is sandy and it only has 36% of mud. As the work performed by Soetaert et al. (1995) is mainly aimed at Nematoda it only shows density values related to the groups Nematoda, Copepoda, Gastrotricha, Turbellaria, Ciliata e Foraminifera. Therefore an overall comparison of the results of both studies is not possible.

 

 

4.2           Benthic Macrofauna

 

4.2.1           Sampling sites

 

Sampling took place seasonally between Autumn 1998 and Summer 1999 in the two sites established: Hortas, near Ribeira das Enguias, and Vasa Sacos, near the mouth of the Sorraia River (Fig. 13).

 

 

Figure 13 – Representation of both sites sampled for the benthic macroinvertebrate communities

 

At each site sampling stations were located in:

                - mudflats which are flooded daily and have an altitude between -0,8 m and -0.1 m (mean sea level).

                - creeks mouth which is the transition area between the mudflats and the creek.

                - creeks that are also flooded daily and have an altitude between –0.1 m and 0.2 m (mean sea level).

                - the pioneer area covered by Spartina maritima has a altitude, in relation to mean sea level of 0.6, being daily flooded excepted during neap tides.

                - the creek margins covered mainly by two species of salt marsh plants: Halimione portulacoides and Arthrocnemon perenne. This area presents an altitude of 1.2 m (mean sea level) being submersed during Spring tides.

 

 

4.2.2           Methodology

 

Sediment analysis

Sediment samples for determination of the water and total organic matter content and granulometric analysis were collected using a Van Veen grab or a core.

Water content was assessed by drying the samples at 60ºC for 24 hours while the total organic matter was determined after destruction in a muffle furnace (2h at 450ºC).

For the determination of the grain size, the sediment samples were dried at 60ºC, weighted and washed to remove the fine portion (<63 mm) and dried again at the same temperature. Sediment was then sorted in a 7 sieve series (2 mm, 1 mm, 500 µm, 355 µm, 250 µm, 125 µm and 63 µm) and weighted according to grain size. The fine fraction was determined in relation to the initial sample dry weight.

 

Benthic macroinvertebrates sampling

Two different methods of sampling were used to collect benthic macroinvertebrates. In the creeks and mudflats a Van Veen grab (with an area of 0.05 m2) was used. At the salt marsh sites a core (0.12 m diameter) was used to collect sediment samples. To prevent decomposition buffered formalin stained with Bengal Rose was added to the samples.

At the laboratory, the sediment samples were washed in a sieve with 500 µm mesh size. The benthic invertebrates were than identified to the lowest level possible, counted and weighted.

Data were analysed through the evaluation of species richness S (expressed as the total number of taxa in each site), density D (expressed in number of individuals per m2) and biomass B (expressed in g of dry weight per m2) which was performed according to Ruhmor (1990). Two indices were also used: Shannon-Wiener’s diversity H’ (Shannon and Weaver, 1963) and evenness J (Pielou, 1966).

 

 

4.2.3           Results

 

Sediment analysis

As expected, for both sites, the values of total organic matter were always superior in the stations with vegetation, pioneer and creek margin areas. In most of the seasons there seems to be a gradient of the percentage of organic matter from the mudflat, with lower values, to the creeks (Fig. 14).

In opposition the percentage water was higher in the areas with shorter flooded periods. An exception was observed for the pioneer area in Vasa sacos where a series of small ponds retain the water for a longer period.

 

 

Figure 14 - Variation of the water content and total organic matter percentages in Hortas and Vasa sacos during the Autumn, Spring and Summer. (M - mudflat, Cm - creek mouth, C - creek, Sp – pioneer area, HA - creek margins)

 

The granulometric analysis revealed a high resemblance between all the stations with the predominance of fine grains, always superior to 97%.

 

 

Species composition

In the salt marsh area a total of 37 taxa of benthic invertebrate were identified (Tab. II), 31 in Hortas and 30 in Vasa sacos with annual average densities of 2707.3 and 1519.1 ind/m2, respectively. The most abundant taxon in these areas was the oligochaetes. The ostracodes were mostly present in Hortas, being the main responsible for the higher density in this area.

 

Table III - Benthic macroinvertebrates community taxa with average ind/m2) in each site.

 

Taxa

Hortas

Vasa sacos

Phylum Protozoa

 

 

   Class Sarcodina

 

 

      Order Forameniferida

 

 

            Forameniferida n.i.

1.0

 

Phylum Nematoda

 

 

            Nematoda n.i.

22.1

11.8

Phylum Annelida

 

 

   Class Polychaeta

 

 

      Order Errantia

 

 

         Family Nereidae

 

 

            Hediste diversicolor

17.3

4.8

            Nereidae n.i.

0.3

0.4

      Order Sedentaria

 

 

         Family Spionidae

 

 

            Pygospio elegans  Claparède

0.7

1.8

            Polydora sp.

2.5

0.1

            Streblospio shrubsolii Buchanan

117.1

125.1

            Spionidae n.i.

2.1

7.0

         Family Cirratulidae

 

 

            Cirratulidae n.i.

1.8

 

         Family Capitelidae

 

 

            Capitelidae n.i.

45.8

 

            Polichaeta n.i.

9.2

0.3

   Class Oligochaeta

 

 

            Oligochaeta n.i.

972.5

835.1

Phylum Mollusca

 

 

   Class Bivalvia

 

 

      Order Eulamellibranchia

 

 

         Family Scrobicularidae

 

 

            Scrobicularia plana (Da Costa)

83.1

9.4

            Abra tennuis (Montagu)

3.3

 

   Class Gastropoda

 

 

      Order Mesogastropoda

 

 

         Family Hydrobiidae

 

 

            Peringia ulvae (Pennant)

175.1

40.2

      Order Basommetophora

 

 

         Family Ellobiidae

 

 

            Phytia myosotis (Draparnaud)

28.9

88.5

Phylum Arthropoda

 

 

   Class Arachnidae

 

 

      Order Araneae

 

 

            Araneae n.i.

 

5.2

      Order Acari

 

 

            Acari n.i.

2.2

1.7

 

 

 

 

 

 

Taxa

Hortas

Vasa sacos

   Class Ostracoda

 

 

            Ostracoda n.i.

916.8

1.1

   Class Crustacea

 

 

      Order Isopoda

 

 

         Family Gnathiidae

 

 

            Paragnathia formica (Hesse)

125.0

82.6

         Family Anthuridae

 

 

            Cyathura carinata (Kröyer)

26.0

33.8

         Family Sphaeromatidae

 

 

            Sphaeroma monodi Bocquet, Hoestlandt & Levi

62.7

149.7

      Order Amphipoda

 

 

         Family Talitridae

 

 

            Orchestia gammarela (Pallas)

 

20.6

         Family Gammaridae

 

 

            Gammaridae n.i.

1.0

0.7

         Family Corophiidae

 

 

            Corophium sp.

2.2

1.7

            Amphipoda n.i.

 

0.7

   Class Insecta

 

 

      Order Diptera

 

 

         Family Limoniidae

 

 

            Limoniidae n.i.

48.1

31.3

         Family Dolichopodidae

 

 

            Dolicopodidae n.i. (larvae)

31.9

35.4

         Family Chironomidae

 

 

            Chironomidae n.i. (larvae)

 

19.1

         Family Ceratopogonidae

 

 

            Ceratopogonidae n.i. (larvae)

0.7

 

            Diptera n.i. (Pupae)

1.8

3.2

      Order Coleoptera

 

 

            Coleoptera n.i (larvae)

4.4

3.7

      Order Homoptera

 

 

            Homoptero n.i (larvae)

 

0.1

            Insecta n.i. (larvae)

1.8

2.2

            Insecta n.i. (adulto)

 

1.5

 

The most represented groups were the polychaetes, isopods and insects. Within the polychaetes, Streblospio shrubsoli clearly dominated in both sites. In the case of isopods diferent species dominated, Paragnathia formica in Hortas and Sphaeroma monodi in Vasa sacos. The same happened with the insects, with Limoniidae and Dolichopodidae larvae being the most abundant in Hortas and Vasa Sacos respectively. Although the two species of gastropods were identified in both sites, Peringia ulvae revealed the highest density in Hortas while in Vasa sacos it was Phytia myosotis.

 

The analysis of the macroinvertebrates taxa per site (Tab. IV) showed that in the mudflats, creeks mouth and creeks habitats the communities were dominated by oligochaetes, polychaetes (mainly S. shrubsolii) and ostracodes (only in Hortas), presenting higher densities in the mudflats. The bivalve Scrobicularia plana and the gastropods P. ulvae were also abundant. All these taxa presented lower densities in Vasa sacos. However, there were some differences between the creeks habitat and the mudflats and creeks mouth habitats, not only on the abundance of the taxa refered (specialy for Hortas), but also on the higher representativity of insect larvae taxa which was much lower in these two last habitats.

In what concerns the biomass values, S. plana was the dominant taxa in the mudflats and in the creeks mouth of both areas. P. ulvae and ostracodes in Hortas and Hediste diversicolor and oligochaetes in Vasa sacos also presented high values of biomass in these habitats. Inside the creeks the relative importance of the polychaete H. diversicolor and of the isopod Cyathura carinata increased, while there was a decrease in the biomass values of S. plana, specialy in Hortas.

The pioneer vegetation areas showed, in Vasa sacos, the poorest benthic community, though most of the taxa presented considerable densities, namely P. ulvae with 118.0 ind/m2. This area, in both sites, was characterised by the increase in abundance of many taxa, especially insects and isopods. In Hortas the oligochaetes clearly dominated, but with several other taxa like P. formica, S. monodi and P. ulvae registering values above 200.0 ind/m2. Regarding the biomass, the values observed in Vasa sacos were lower than those in Hortas but P. ulvae had the highest values in both, 0.4 and 7.6 g/m2, respectively. S. plana registered a decrease in its densities, nevertheless in Hortas it was still important in terms of the biomass.


Table IV – Average density (ind/m2) and biomass (g/m2) of the important benthic macroinvertebrate community taxa in both sites.

 

 

Taxa

Hortas

Vasa sacos

Mudflat

Creek mouth

Creek

Pioneer area

Creek margin

Mudflat

Creek mouth

Creek

Pioneer area

Creek margin

D

B

D

B

D

B

D

B

D

B

D

B

D

B

D

B

D

B

D

B

H. diversicolor

8.1

4.4´10-2

3.8

7.9´10-2

4.3

6.5´10-2

25.8

0.2

44.2

0.9

7.4

9.2´10-2

11.1

8.6´10-2

5.6

4.5´10-2

 

 

 

 

S. shrubsolii

130.0

4.6´10-3

300.2

1.1´10-2

132.8

5.4´10-3

22.1

7.4´10-4

 

 

189.1

6.0´10-3

156.9

7.0´10-3

265.0

1.610-2

14.7

-

 

 

Spionidae n.i.

9.8

-

 

 

0.7

7.1´10-5

 

 

 

 

24.0

-

3.7

-

7.4

-

 

 

 

 

Oligochaeta n.i.

323.6

1.0´10-2

335.7

1.3´10-2

230.0

7.3´10-3

1876.8

5.0´10-2

2096.2

8.4´10-2

1950.2

7.6´10-2

649.7

2.2´10-2

1180.9

7.4´10-2

88.5

1.5´10-3

306.0

6.9´10-3

S. plana

216.2

14.0

185.0

0.3

3.1

1.7´10-2

11.1

7.6

 

 

30.6

14.6

8.8

3.0

7.5

0.4

 

 

 

 

P. ulvae

150.9

0.5

238.7

0.4

30.8

6.8´10-2

317.1

7.7

138.3

1.8

31.9

9.2´10-2

12.4

2.3´10-2

38.7

3.0´10-2

118.0

0.4

 

 

P. myostis

 

 

 

 

0.7

5.0´10-4

11.1

-

132.7

1.2´10-2

 

 

 

 

 

 

 

 

442.5

1.8´10-2

Ostracoda n.i.

1553.2

0.1

2980.2

0.2

50.5

2.5´10-3

 

 

 

 

1.8

-

3.7

-

 

 

 

 

 

 

P. formica

 

 

 

 

 

 

320.8

0.1

304.2

9.3´10-2

 

 

 

 

 

 

 

 

413.0

8.3´10-3

C. carinata

13.8

5.8´10-3

 

 

3.8

1.4´10-2

73.7

2.8´10-2

38.7

7.7´10-3

33.7

2.4´10-2

11.1

9.4´10-3

21.0

4.7´10-2

73.7

5.8´10-2

29.5

4.1´10-4

S. monodi

 

 

 

 

 

 

221.2

0.1

92.2

3.3´10-2

1.3

1.6´10-3

 

 

2.5

8.0´10-3

 

 

744.8

7.7´10-2

O. gammarela

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

103.2

2.0´10-2

Corophium sp.

 

 

 

 

 

 

11.1

6.6´10-3

 

 

 

 

 

 

1.3

1.9´10-4

 

 

154.9

2.8´10-3

Limoniidae n.i.

 

 

 

 

2.9

3.4´10-3

44.2

1.2´10-2

193.6

6.4´10-2

 

 

 

 

1.9

4.0´10-3

 

 

70.1

0.4

Dolicopodidae n.i.

1.84

-

 

 

44.4

1.4´10-2

47.9

5.5´10-2

62.7

1.2´10-2

1.8

3.7´10-4

18.8

6.5´10-3

19.9

9.4´10-3

66.4

8.1´10-3

 

 

 


In the creek margins, both sites presented similar benthic community structure, with a high representativity of a large number of groups namely, isopods, amphipods, oligochaetes, gastropods and insects. It was noticeable in these areas the reduced number of polychaetes, the absence of bivalve species and the high number of insects taxa. The highest values of density in the creek margins were registered for oligochaetes (2096.2 ind/m2) in Hortas and S. monodi (744.8 ind/m2) in Vasa sacos. The gastropds, represented by P. ulvae and Phytia myosotis were also abundant, and in Hortas P. ulvae had the highest value of biomass (1.8 g/m2) of this area while in Vasa sacos, Dolicopodidae was the taxon with higher biomass value.

 

Seasonal Analysis

The analysis of the seasonal variation of the density and biomass in both sites showed that the highest densities were obtain in different seasons, Winter for Vasa sacos and Summer for Hortas. However these fact resulted from the high densities of ostracodes in the first site and of oligochaets in the last one.

In terms of biomass the highest values were obtain during the Spring for both sites with 12.5 g/m2 and 17.5 g/m2 respectively for Vasa sacos and Hortas.

 

In Hortas, the areas without salt marsh vegetation presented in the Autumn and Winter the highest densities of oligochaetes while the ostracodes were mostly present during the Spring and Summer.

Regarding the area colonised by S. maritima (pioneer area), P. ulvae was the most abundant species in the Autumn (59.0 ind/m2) and Summer (1150.4 ind/m2) while during the Winter and Spring it was substituted by S. shrubsolii (88.5 ind/m2) and oligochaetes (7477.9 ind/m2), respectively. Nevertheless, in Autumn and Winter the number of taxa identified was very reduced.

In the Spring, P. formica showed high densities in both areas where salt marsh vegetation was present. In the area colonised by H. portucaloides and A. fruticosum (creek margin), only one taxon was identified during Autumn and Winter. However an increase in the number of taxa was observed in the Spring with the presence high densities of insect larvae (especially Limoniidae with 774.3 ind/m2), isopods (mainly P. formica with 1216.8 ind/m2) and gastropods (P. ulvae, 376.1 ind/m2 and P. myosotis, 531.0 ind/m2). During the Summer this area was clearly dominated by oligochetes (3849.6 ind/m2) which corresponded to about 89% of the total number of individuals captured.

In terms of biomass, in the Hortas mudflat it was S. plana the dominat species, except during the Autumn (P. ulvae). It was also in this area that during the Spring the highest value of this parameter, for all sampling areas, was registered (46.2 g/m2). In the creek mouth and creek area, in most seasons, the same species dominated, P. ulvae during the Autumn and Summer and H. diversicolor in the Winter. In the areas colonised by salt marsh plants it was P. ulvae that had the highest biomass values in the Spring and Summer, though in the pioneer area during the Spring, S. plana revealed the highest value (31.0 g/m2) of the two areas.

 

In Vasa sacos the oligochaetes were the dominant group in almost every areas without vegetation for most of the sampling period. However, during the Summer in the creek area the most abundant taxon was P. ulvae with 120.0 ind/m2, while in creek mouth it was S. shrubsolii (35.0 ind/m2)

In the pioneer area insects’ larvae and the gastropods were the most represented group during the Summer. In the remaining seasons few taxa were present in the area.

In creek margins, during the Autumn and Winter only insects’ larvae were identified. This area was dominated, in the Spring by P. formica and oligochaetes with densities of 1651.9 ind/m2 and 1062.0 ind/m2, respectively. This last taxon, along with nematodes was, during the Summer, the most abundant groups although their densities were inferior to those registered in Spring.

The areas without vegetation revealed that in most seasons where S. plana was present the highest values of biomass belong to this species. However, duringt the Autumn in the mudflats and in the Summer for the creek area, the oligochaetes and P. ulvae, respectively presented the highest biomasses. When S. plana wasn’t captured, two taxa dominated, H. diversicolor in the creek mouth ( Autumn and Winter) and C. carinata in the creek (Autumn). For the pioneer area it was clear that P. ulvae was the most important species in terms of biomass. Only during the Spring, when P. ulvae wasn’t identified, the isopod C. carinata had the highest biomass value. In the creek margin different taxa presented the highest taxa. Thus, during the Autumn ans Winter it was the Dolichopodidae larvae, while in the Spring and Summer, the species that most contributed for the biomass were the isopod S. monodi and the gastropod P. myosotis, respectively.

 

Specific richness, diversity and evenness

The analysis of the specific richness, diversity and evenness for both sites during the Autumn, Winter, Spring and Summer (Tab. V) showed that the diversities in Vasa sacos were in most of the cases lower than in Hortas. This seemed to be related with the reduced values of evenness in that site, originated by the higher dominance of one taxon, oligochaetes.

In general, in Hortas, the areas without vegetation were the ones with the highest values of diversity, mainly due to the reduced number of taxa present in the areas with vegetation. However in Spring, with the increase of the specific richness, the areas colonised by vegetation registered high diversity values (1.083 in the pioneer area aand 1.556 in the creek margin).

For the creek and the mudflat the seasonal flutuations of all the parameters were very reduced. In oposition the creek mouth presented a high seasonal variability in the diversity values mainly due to the flutuation in the number of species captured.

 

In Vasa sacos the dominance of oligochaetes from Autumn till Spring was responsible in the areas without vegetation, for the low values of evenness and diversity, between 0.186 and 0.408 in mudflats during the Autumn and 0.526 and 1.155 in the Spring for the creek. In Summer, due to the reduction in the relative importance of the oligochaetes, and thus the increase of the eveness values, the diversity reached higher values (1.292 in mudflats, 1.331 in creek mouth and 1.513 in the creek). In the areas colonised by salt marsh plants it was observed a similar pattern as described for Hortas, with a very reduced species richness during the Autumn and Winter and an increase of this parameter during the Spring which originated high diversity values.

 


 

Table V - Specific richness (S), diversity (H’) e evenness (J) of the benthic macroinvertebrate community in Hortas and Vasa sacos

 

 

 

Hortas

Vasa Sacos

 

 

Mudflat

Creek mouth

Creek

Pioneer area

Creek margin

Mudflat

Creek mouth

Creek

Pioneer area

Creek margin

 

Autumn

S

10

5

7

2

1

9

7

7

3

1

H’

1.540

1.082

1.354

0.637

0.000

0.408

0.718

0.635

0.860

0.000

J

0.669

0.672

0.696

0.918

-

0.186

0.369

0.327

0.783

-

 

Winter

S

8

7

8

3

1

6

6

7

3

2

H’

1.581

1.173

1.497

1.011

-

0.611

0.757

0.806

0.956

0.693

J

0.760

0.603

0.720

0.921

-

0.341

0.423

0.414

0.870

1.000

 

Spring

S

7

1

8

12

13

7

5

9

2

16

H’

1.552

0.000

1.359

1.083

1.556

0.820

0.730

1.155

0.693

1.945

J

0.797

-

0.636

0.436

0.606

0.421

0.454

0.526

1.000

0.702

 

Summer

S

11

12

8

4

8

9

4

7

8

6

H’

1.120

0.930

1.510

0.734

0.487

1.292

1.331

1.513

1.562

1.229

J

0.467

0.374

0.726

0.530

0.234

0.588

0.960

0.777

0.751

0.686

 

 

 


4.3           Fish assemblages in tidal salt marsh creeks and in adjoining mudflat areas in the Tagus estuary

 

4.3.1           Introduction

 

The importance of salt marshes as nursery habitats for fish has been emphasised by several authors, mainly for eastern USA estuarine systems (e.g., Cain & Dean, 1976; Weinstein & Brooks, 1985; Rozas 1995).

For European estuaries, some studies concerning the structure and dynamic of the fish assemblages in salt marsh areas were also developed, namely those conducted by Labourg et al. (1985), Drake & Arias (1991), Cattrijsse et al. (1994) and Costa et al. (1994). Despite these contributions, knowledge on fish assemblages of European salt marshes is scarce.

Salt marsh tidal creek areas are characterised by a high instability due to fluctuations of water level and the direction and strength of the flow, which induce a high variability in the abiotic conditions (Labourg et al., 1985). Thus, the use of tidal creeks by fishes is controlled by the hydroperiod. During high tide the creeks are flooded allowing the access of fishes. At low tide the water drains almost completely being the fishes forced to move to the adjoining areas.

The utilisation of these habitats may be discussed in terms of cost-benefit. The high instability of abiotic factors increases the risk of stranding and enables the occurrence of a large number of species. Fish species that tolerate such conditions can benefit of the high food availability induced by the extreme productivity of these marsh areas (Shenker and Dean 1979). This fact associated with the low predation pressure within those habitats emphasises its potential as feeding grounds and refuges for fish (Cain & Dean, 1976; Shenker & Dean, 1979; Bozeman & Dean, 1980).

The fish assemblages in salt marsh areas reflect a seasonal pattern with pulses of transient estuarine species, which regularly colonise these shallow and intertidal areas (Bozeman & Dean 1980; Rozas 1995).

The role of the Tagus estuary as nursery for fish, particularly the upper areas has been reported in a large number of studies (e.g., Costa, 1982; Costa & Bruxelas, 1989; Cabral, 1998; Costa & Cabral, 1999). According to these authors, the most important grounds are located in the adjoining areas of salt marshes. The most abundant fish species that use these nursery grounds are sea bass, Dicentrarchus labrax (L. 1758), and soles, Solea solea (L. 1758) and Solea senegalensis Kaup 1858. However, no studies focused the role of the salt marsh areas for the fish assemblage and the interaction between these habitats.

In the present paper a comparative analysis of the fish assemblages in two salt marsh tidal creeks and their adjoining mudflat areas was performed in order to understand the relative importance of these habitats for the Tagus fish community.

 

 

4.3.2           Methodology

 

Study Area

The Tagus estuary has an area of 320 km2, of which 113,8 km2 are intertidal. About 13 km2 of the intertidal area is covered by salt marsh vegetation (Catarino et al., 1985). The Tagus estuary is mesotidal with semi-diurnal tides and a tidal range of about 4 m.

The present study was carried out on two salt marsh areas located in the upper part of the estuary (Fig. 15). This region is characterised by an extensive area of mudflats composed of fine muddy sediments, with 90,6% of particles with less than 20 mm diameter (Brotas et al., 1995). On the upper part of the mudflats the pioneer/lower marsh areas are dominated by Spartina maritima L. while the middle marsh is mainly composed by Halimione portulacoides L. and Arthrocnemon perenne Miller, especially along the creeks.

Samples were taken in two tidal creeks (Hortas and Vasa sacos) and in the adjacent mudflat areas (Fig. 15). The creek beds are approximately 150 m long, 15 m wide at the mouth and the height between the mud bottom and the marsh surface is ca. 1.5 m.

Both creeks have no connection with other neighbouring creeks, drying almost completely during ebb tides.


 

 

Figure 15 - Location of the Tagus estuary and sampling areas.

 

Sampling methods

Two different sampling methods were used according to habitat. In the creeks, samples were taken monthly from September 1998 until August 1999, during daylight at ebb tides of similar amplitudes, using fyke and gill nets (3mm and 50mm mesh size, respectively). The gill nets were used to capture the large specimens, mainly mugilids, and therefore to avoid the clog of the fyke net.

Fishes were collected on each 30 min, from high tide until the creek was completely dry. At each collection temperature and salinity were measured using a Hidrolab multiprobe sonde.

In the mudflat areas, samples were done monthly from January 1995 until December 1996 using a 4 m beam trawl with 10 mm mesh size and one tickher chain. Trawls were towed at about 1 knot of speed and lasted 20 min. Three trawls were performed monthly per sampling area. Temperature and salinity were also measured at each sampling period and area.

All fish caught were counted, measured (total length to the nearest 1 mm) and weighed (wet weight with 0.001 g precision) in the laboratory.

 

Data analysis

Spearman rank correlation coefficient was calculated to compare the species abundance in the creeks and in the mudflats (Zar, 1996).

The analysis of the structure of the fish assemblages was based on ecological and trophic guilds (adapted from Elliott & Dewailly, 1995), indicating the species’ use of the estuary and the feeding habits, respectively. Five ecological guilds (i.e. fresh water adventitious, marine adventitious, catadromous, nursery and resident species) and four trophic guilds (i.e. detritivorous, strictly planktivorous, strictly invertebrate feeder and carnivorous other than the two previous guilds) were used. The Chi-square statistic was used to test differences in the number of species per guild (Zar, 1996).

For the most abundant species the percent of juveniles and adults individuals was determined based on length at maturity reported by Almeida (1996) for Liza Ramada (Risso, 1826), Pestana (1989) for Sardina pilchardus (Walbaum, 1792), Dinis (1986) for S. senegalensis, and Bouchereau et al. (1990) and (1993) respectively for Pomatoschistus minutus (Pallas, 1770) and Pomatoschistus microps (Krøyer, 1838).

Species richness (S), evenness (J) and Shannon-Wiener’s (H’) diversity indices were calculated for each sampling area and season (Ludwig & Reynolds 1988).

 

 

4.3.3           Results

 

Abiotic conditions

The variation of the temperature values was similar in the two creeks as well as in the two mudflat areas (Fig. 16). In both habitats, the maximum values were obtained in July and August (23.3ºC and 22.5ºC in Hortas and Vasa sacos creeks, and 23.2ºC and 24.8ºC in the respective mudflat areas). The minimum values were recorded in January and were lower in the creeks (7.8ºC and 9.5ºC) comparatively to both mudflat areas (12.0ºC).

 

Figure 16 - Water temperature and salinity in the study areas during the sample period.

 

The maximum and minimum salinity values were obtained in the same months as the extreme temperature values. The seasonal variation pattern was similar in all the areas (Fig. 16).

 

Community structure

A total of 28 fish species were identified, of which 12 occurred in the salt marsh tidal creeks and 26 in the mudflat areas (Table VI). Species composition in the two intertidal creeks was similar as well as in the two mudflat areas considered.

 


Table VI - Relative abundance, biomass and ecological (FW- fresh water adventitious, MA- marine adventitious, C- catadromous, N- nursery and R- resident) and trophic guilds (D- detritivorous, P- strictly planktivorous, IF- strictly invertebrate feeder and C- carnivorous other than the two previous guilds) of the fish species captured in the salt marsh creeks and in the mudflat areas at Hortas and Vasa sacos.

 

 

 

 

Hortas

Vasa sacos

 

 

 

Creek

Mudflat

Creek

Mudflat

Species

EG

TG

Nº ind. (%)

Biomass

(% g)

Nº ind. (%)

Biomass

(% g)

Nº ind. (%)

Biomass (% g)

Nº ind.

(%)

Biomass

(% g)

Anguilla anguilla

C

C

0.4

< 0.1

0.1

1.5

0.1

< 0.1

0.4

2.3

Atherina presbyter

R

P

0.1

< 0.1

< 0.1

< 0.1

< 0.1

< 0.1

< 0.1

< 0.1

Barbus bocagei

FW

IF

 

 

 

 

 

 

< 0.1

1.1

Chelon labrosus

N

D

 

 

< 0.1

0.7

 

 

0.1

3.6

Ciliata mustela

N

C

 

 

< 0.1

0.1

 

 

< 0.1

0.1

Conger conger

N

C

 

 

< 0.1

0.1

 

 

< 0.1

0.1

Dicentrarchus labrax

N

C

0.3

< 0.1

7.7

4.8

< 0.1

< 0.1

1.0

0.9

Diplodus bellottii

N

IF

 

 

 

 

 

 

0.1

< 0.1

Diplodus sargus

N

IF

< 0.1

< 0.1

 

 

 

 

0.1

0.2

Diplodus vulgaris

N

IF

 

 

 

 

 

 

< 0.1

< 0.1

Engraulis encrasicolus

R

P

0.1

< 0.1

0.8

0.6

< 0.1

< 0.1

2.0

0.8

Gambusia holbrookii

FW

P

0.1

< 0.1

 

 

< 0.1

< 0.1

 

 

Gobius niger

R

IF

 

 

0.1

0.1

 

 

0.6

0.3

Halobatrachus didactylus

R

C

 

 

 

 

 

 

< 0.1

< 0.1

Hippocampus hippocampus

R

P

 

 

< 0.1

< 0.1

 

 

< 0.1

< 0.1

Liza aurata

N

D

0.1

< 0.1

0.1

0.3

0.1

1.0

0.1

0.2

Liza ramada

C

D

49.9

99.6

6.0

55.3

5.6

90.5

4.9

56.6

Mugil cephalus

N

D

< 0.1

0.2

0.1

0.7

 

 

0.4

12.9

Platichthys flesus

N

IF

 

 

0.3

2.2

 

 

0.1

0.5

Pomatoschistus minutus

R

IF

 

 

9.8

1.1

 

 

24.1

2.6

Pomatoschistus microps

R

IF

46.4

0.1

58.7

3.0

49.8

6.0

57.6

2.2

Raja clavata

MA

C

 

 

< 0.1

< 0.1

 

 

< 0.1

0.5

Sardina pilchardus

N

P

1.5

< 0.1

< 0.1

< 0.1

43.8

2.3

0.1

< 0.1

Solea senegalensis

N

IF

 

 

15.0

29.0

 

 

5.5

13.4

Solea solea

N

IF

 

 

0.1

0.2

 

 

0.5

1.2

Sparus aurata

N

IF

 

 

0.1

0.1

 

 

< 0.1

0.2

Syngnathus thyple

R

P

 

 

< 0.1

< 0.1

 

 

 

 

Syngnathus sp.

R

P

1.0

< 0.1

1.0

0.1

0.4

< 0.1

2.2

0.1

 

Total

24708

1052999

13572

143873

46384

81261

7079

95503

 


The creeks fish assemblage was mainly composed by estuarine resident species and by marine species that use the estuary as nursery (Fig. 17). Fresh water species were only present in periods of intense rainfall, while marine occasional species were completely absent from these areas. In the mudflat areas the number of species per ecological guild did not differ statistically from creeks (c2= 1.664; d.f.= 4; p> 0.05). In that habitat the species that uses the area as nursery was the most representative one, comprising 54% of the species in Vasa sacos and 50% in Hortas.

 

 

Figure 17 - Distribution of the fish species per ecological guilds (FW- fresh water adventitious, MA- marine adventitious, C- catadromous, N- nursery and R- resident species) for the saltmarsh creeks and for the mudflats at Hortas and Vasa sacos.

 

Considering the trophic guilds, no statistical differences were found between the creek and the mudflat areas (c2= 2.876; d.f.= 4; p> 0.05). The planktivorous species was the predominant group in the creeks (Fig. 18), especially in Vasa sacos where it was composed by 50% of the species. Some carnivorous species were also present in these habitats. The detritivorous were the only big-size species present in the creeks. In the mudflat areas, where the proportion of species according to ecological guilds was more equable, the invertebrate feeders were the predominant group, representing 32% in Hortas and 42% in Vasa sacos.

 

 

Figure 18 – Distribution of the fish species per trophic guilds (D- detritivorous, P- strictly planktivorous, IF- strictly invertebrate feeder and C- carnivorous other than the two previous guilds) for the saltmarsh creeks and for the mudflats at Hortas and Vasa sacos.

 

No statistical differences were found in the species abundance between both creeks(r=0.937; p<0.05) and mudflats (r=0.815; p<0.05). However, the abundance in the creeks was significantly different from the mudflats (r=0.345; p>0.05).

Three species were numerically dominant in the creeks, comprising over 95% of the total number of individuals captured: S. pilchardus (43.8% in Vasa sacos), L. ramada (mainly in Hortas with 49.9%) and P. microps in both creeks (46.4% and 49.8% respectively for Hortas and Vasa sacos). This species was also the most abundant in both mudflat areas, representing more than 50% of the captures.

Although absent in the creeks S. senegalensis, in particular at Hortas (15.0%), and P. minutus, mainly in Vasa sacos (24.1%), were also abundant in the mudflats.

In terms of biomass L. ramada was the most important species in all studied areas. The values obtained for this species ranged from 91% to almost 100% in the creeks and from 55% to 57% in the mudflat areas. In these zones S. senegalensis was also an important fraction of the biomass, especially in Hortas where it represented 29.0% of the captures. In the creeks, the total biomass observed for Hortas was more than 10 times higher than the value obtained for Vasa sacos.

 

Seasonal variation

The diversity index (H’) for the creeks was always lower than in the mudflats and does not follow the same pattern for both areas (Table VII). In Hortas the higher values occurred during the Autumn when the specific richness (S) was higher and there were no dominant species. During Summer the diversity values were very reduced following the same tendency of the evenness (J). In Vasa sacos the lowest values were observed during Winter when evenness was very reduced. The highest diversity was obtained in Spring and Summer as a result of the increase in the species richness and evenness index

 

Table VII - Seasonal values of species richness (S), evenness (J) and Shannon-Wiener’s (H’) diversity indices for the salt marsh creeks and for the mudflat areas at Hortas and Vasa sacos.

 

Hortas

Vasa sacos

 

Creek

Mudflat

Creek

Mudflat

 

H’

S

J

H’

S

J

H’

S

J

H’

S

J

Autumn

0.91

11

0.38

1.00

19

0.34

0.43

6

0.24

1.17

20

0.39

Winter

0.65

5

0.40

1.02

13

0.40

0.08

5

0.05

1.09

12

0.43

Spring

0.76

7

0.39

1.46

14

0.55

0.91

9

0.41

1.75

19

0.60

Summer

0.03

7

0.02

1.59

16

0.57

0.87

5

0.54

1.26

20

0.42

 

Both mudflat areas presented a similar pattern with highest diversities during Spring and Summer following the increase of the evenness values.

The species present in the creeks showed a high seasonality (Table VIII) in the use of these habitats. Only two species were captured during all the year, P. microps and L. ramada. A great number of the species (e.g. Engraulis encrasicolus (L. 1758), D. labrax, Anguilla anguilla (L. 1758), S. pilchardus and Liza aurata (Risso 1810)) were only captured during part of the year as juveniles. The highest abundances for most of the species was observed during the Spring, while in the Winter only a few individuals were caught, mostly from resident species.

The highest abundance of P. microps in the creeks was observed during Spring, mainly in Vasa sacos, with a population composed mostly by juveniles. In the Summer there was a decrease in the abundance of this species and the presence of adults in these areas was scarce. During the Autumn/Winter the number of individuals increased again, although in this time mainly adults were present. In the mudflat areas the highest number of individuals of this species was observed during these late seasons followed by a gradual reduction of the abundances until the Summer, when the concentration of individuals start increasing again.

L. ramada was always present in the creeks with a high number of individuals. Nevertheless, the abundances in Hortas were always higher than in Vasa sacos. The exception was during Spring when similar quantitatives, mostly small juveniles, were observed for both creeks. In the Summer in Hortas, there was an increase in the number of captures of both juveniles and adults. In the Autumn and Winter there was a gradual decrease of the abundance and mostly adults were present in these areas. For both adjacent areas the peak of abundance of this species was observed during the Winter. As described for the creeks, the number of individuals in Hortas was always superior to that obtained for Vasa sacos.

In the creeks S. pilchardus was present, mostly in Vasa sacos, from the beginning of the Spring until early Autumn. The highest abundance of this species was observed during the Spring when small juveniles began to appear in these areas. In the Summer a marked decrease in the number of individuals was observed leading almost to the disappearance of the species from these areas. The presence of this species in the adjoining areas was very reduced.

In the mudflat areas P. microps was the most abundant species in most of the seasons with a maximum during Autumn. Exceptions to this dominance were observed for Hortas in the Spring when a high number of individuals of S. senegalensis and D. labrax were captured.

 

Table VIII - Number of individuals and percentage of juveniles per season for the most abundant species captured in the salt marsh creeks and in the mudflat areas at Hortas and Vasa sacos (Aut. – Autumn, Win. – Winter, Spr. - Spring, Sum. – Summer).

 

 

Hortas

Vasa sacos

 

Creek

Mudflat

Creek

Mudflat

Species

Aut.

Win.

Spr.

Sum.

Aut.

Win.

Spr.

Sum.

Aut.

Win.

Spr.

Sum.

Aut.

Win.

Spr.

Sum.

 

Dicentrarchus

labrax

 

 

80

100%

5

100%

14 100%

1

100%

748

100%

278

100%

 

 

6

100%

 

3

100%

 

40

100%

26

100%

 

Liza

ramada

1254

45%

208

38%

2291

83%

8565

87%

67

98%

594

67%

119

53%

32

81%

29

71%

11

81%

2343

99%

217

35%

74

71%

155

25%

99

46%

16

50%

 

Pomatoschistus

minutus

 

 

 

 

406

0%

735

2%

26

6%

168

 

 

 

 

 

408

1%

185

0%

191

12%

924

18%

 

Pomatoschistus

 microps

1707

40%

71

2%

9569

98%

17

57%

4222

6%

2749

1%

312

6%

687

1%

301

11%

1724

6%

20492

95%

603

88%

1688

0%

760

0%

429

5%

1199

5%

 

Sardina

pilchardus

4

100%

 

368

100%

5

100%

 

 

 

2

100%

 

 

20307

100%

20

100%

2

100%

1

100%

1

100%

2

100%

 

Solea

senegalensis

 

 

 

 

666

100%

99

100%

770

100%

497

100%

 

 

 

 

128

100%

44

100%

91

100%

129

100%

 

 

 

4.3.4           Discussion

 

The different sampling methodologies used for salt marsh creeks and adjoining mudflats areas enabled a direct comparison between both habitats. Unlike the fyke, that is almost a unselective gear, the beam-trawl underestimates the abundance of both large and fast swimming specimens, namely mugilids, and small fishes (considering the differences in mesh size comparatively to the fyke net used). However the same sampling technique could not be applied in both habitats due to their different characteristics. These constraints were considered for the selection of the analytical tools to compare the fish assemblages of these two habitats.

From the 48 species reported in recent studies conducted in the Tagus estuary (Costa et al. 1998; Cabral 1998) only 26 were present in salt marsh mudflat areas and 12 in intertidal creeks. The low number of species in the creeks suggested that only a limited number of the fish species that occurred in the upper part of the estuary used those habitats. The fish species richness in intertidal creeks reported for other European estuarine systems varied. In the North Sea, Cattrijsse et al. (1994) reported the occurrence of 13 fish species in the salt marsh creeks of the Westerschelde estuary while Laffaille et al. (1998) found 23 fish species in the creeks of the macrotidal system of Mont Saint-Michel. Higher species richness (39 species) was obtained for the Cadiz Bay (Drake & Arias, 1991). However, these creeks were subjected to a high marine influence, which surely induce an increase of fish species number.

The number of fish species reported in studies conducted in USA estuarine salt marsh areas also differed substantially (e.g., Shenker & Dean, 1979; Weinstein et al., 1980; Yoklavich et al., 1991). As pointed out by Cattrijsse et al. (1994) these differences may be due to latitudinal effects, but the particularities of both estuarine fish assemblage and salt marsh abiotic and biotic conditions should be the major determinant of species richness.

Regarding the use of the estuary by the different species, the dominance of estuarine resident species and fish that use the estuary as nursery noticed for the creeks assemblages in the Tagus, was also outlined for other estuaries (e.g., Rozas, 1995; Kneib, 1997).

In terms of feeding habits, the dominant trophic group in two habitats was different. The comparison of the food availability in these two areas would be extremely useful to explain the preponderance of planktonic feeders in the creeks and of benthic feeders in the mudflat areas. The low abundance of piscivorous species was a common aspect of the creeks fish assemblages, reported by several authors (e.g., Cain & Dean, 1976; Miltner et al., 1995).

A general characteristic of tidal creeks fish assemblage is the dominance, both in number and biomass, by few species (Cain & Dean, 1976; Kneib, 1984; Rakocinski et al. 1992; Cattrijsse et al., 1994). In several studies conducted in European estuaries, P. microps (Drake and Arias 1991; Catrijsse et al. 1994) and L. ramada (Drake & Arias, 1991; Laffaille et al., 1998) has been pointed out as the most important fish species in tidal creeks, as observed in the present study.

Laffaille et al., (1998) in Mont Saint-Michel noticed a low abundance of P. microps but other Gobiidae, P. minutus and P. lozanoi (de Buen, 1923), presented high densities in tidal creeks. In the Tagus and also in other estuaries (Drake & Arias, 1991; Catrijsse et al. 1994) P. minutus was abundant in the salt marshes adjoining areas but not in tidal creeks. These findings suggest that, depending of site characteristics, a spatial segregation may be observed in alternative estuarine habitats such as salt marsh creeks.

Drake & Arias, (1991), Catrijsse et al., (1994) and Laffaille et al., (1998) found also other fish species, namely Platichthys flesus (L. 1758), S. solea, S. senegalensis and D. labrax in tidal creeks. In the Tagus, some of these species were particularly abundant in the adjoining areas, but were only occasionally caught in the creeks. Since the use of tidal creeks can be view under a cost-benefit perspective, a comparative analysis of the food availability in these two estuarine habitats could help explaining the absence of these species in tidal creeks.

Despite the similarities in variation pattern of temperature and salinity recorded in the two creeks analysed in the present study, several differences in species abundance were found in the two sites. Rakocinski et al., (1992) suggested that geomorphological characteristics of sites are probably the most important factors determining habitat selection by fishes in estuarine tidal creeks. The extension of the nearby mudflat area and the proximity to permanent subtidal areas could support the differences in the abundance of L. ramada obtained between the two creeks analysed.

Seasonal changes in the fish assemblages of tidal creeks, reflected mainly the recruitment or pulses of abundance of different species. Some recruits migrate from other estuarine areas, namely P. microps (Bouchereau et al., 1993), or from nearshore areas, namely L. ramada (Almeida, 1996) and S. pilchardus (Ré, 1984).

The results obtained for the Tagus showed that diversity and evenness decreased in periods of intensive recruitment, being the variation pattern different according to the site. Drake & Arias (1991) outlined that these indices were higher in Winter and late Summer and both were significantly correlated with density of resident species.

Most of the species described as using the upper Tagus estuary as nursery areas were absent from the creeks or were occasional. Nevertheless, those areas are apparently important for P. microps and L. ramada both as nursery areas and as feeding grounds for the adults. For S. pilchardus the importance seems reduced comparing with the high densities of larvae observed by Ré (1984) on the mouth of the estuary and on the adjacent coastal areas. However, the analysis of the importance of these habitats for the fish community should include the role of the different species as a trophic link between the highly productive salt marsh areas and the adjacent estuarine areas or the nearby coastal areas. Laffaille et al. (1998) estimated that in Mont Saint-Michel the fish community was responsible for the export of 50 tons DW of organic matter per year, with the mullets responsible for 70% of this transport. The extremely high biomass values of L. ramada in the Tagus tidal creeks associated with the foraging behaviour of this species suggest the transport of a large amount of organic matter from the tidal creeks towards adjoining estuarine areas or the nearby coastal areas.

 

 

5           Trophic analysis

 

For the study of the food web in the Tagus salt marsh two approaches were used:

- multiple stable isotope analysis (C13/ C12, S34/ S32, N15/ N14) were applied to k-species, representative of the different trophic levels, sellected after the biological characterisation of the area;

- gut content analysis of the most representative fish species present in the study salt marsh, the common gobie Pomatoschistus microps and the thin-lipped grey mullet Liza ramada.

Both methods were used in order to complement and compare between themselves.

 

 

5.1           Food habits of Pomatoschistus microps (Krøyer, 1838) in the Tagus salt marsh

 

5.1.1           Introduction

 

The common goby Pomatoschistus microps (Krøyer, 1838), is very common in estuaries, lagoons and along the shores of Europe. In the Tagus estuary, studies of these species are resumed to Costa (1988), Moreira (1991) and to Salgado (1995), this is possibly due to their low economic value, despite their great abundance and extremely important role in the estuarine food web (Costa, 1988). Their role in the diet of different organisms from Tagus estuary has been emphasised by several authors, namely decapods, Crangon crangon Linnaeus and Carcinus maenas Linnaeus, in teleosts like Anguilla anguilla Linnaeus (Costa et al., 1992), Solea senegalensis Kaup and Dicentrarchus labrax Linnaeus (Cabral, 1998) and in piscivorous birds such as Egretta garzetta Linnaeus (Moreira, 1992), Recurvirosta avosetta Linnaeus and Calidris alpina (Moreira, pers. comun.).

The aim of this work is to deepen knowledge about the trophic place of P. microps in this salt marsh food web and to understand seasonal changes in the feeding strategies of this species.

 

 

5.1.2           Methodology

 

This study was carried in two salt marsh tidal creeks located in the upper zone of the Tagus estuary. These areas are described in detail in the analysis of the fish community section.

 

Field samples were taken monthly from September 1998 until August 1999, during daylight at ebb tides of similar amplitudes, using fyke nets (3mm mesh size). Fishes were collected on each 30 min, from high tide until the creek was completely dry. At each collection temperature and salinity were measured using a Hidrolab multiprobe sonde. After capture, gobies were placed on ice to prevent post mortem digestion.

In the laboratory total length was measured to the nearest 1mm and wet weight was determined to the nearest 0.0001g. For each species, a maximum of 60 digestive tracts, randomly selected, were analysed every month per site. The food items found in the gastro-intestinal tracts were identified to species whenever possible, counted and wet weighed to the nearest 0.0001g.

 

For quantitative analysis of the diet, occurrence (Oi), numeric (Ni) and weight (Wi) frequencies were used (empty digestive tracts were eliminated) as well as the dietary coefficient (Qi) a mixed method proposed by Hureau (1970):

 

Qi = NI x Wi

 

This method describes preferential (Q >200), secondary (200> Q >20) and accidental (Q < 20) prey.

 

In order to know the variation of the feeding habits with the length, the individuals were grouped into eight size classes of total length. Class I £ 19mm; 20mm £ class II £ 24mm; 25mm £ class III £ 29mm; 30mm £ class IV £ 34mm; 35mm £ class V £ 39mm; 40mm £ class VI £ 44mm; 45mm £ class VII £ 49mm; class VIII ³ 50mm.

 

With the purpose of comparing the food ingested before and after the period of residence inside the creeks, samples were performed during the flood on the mouth of the creek. The gut content of the individuals sampled was analysed and the results were compared with the ones obtained for the fish captured during ebb tide.

 

 

5.1.3           Results

 

The gut contents of a total of 442 Pomatoschistus microps, ranging from 10 to 51mm, were analysed.

The polychaetes were the most frequent item in the diet of this goby (Table IX), occurring in 50% of the gut contents analysed. However other items such as mysids (35.29%), bivalves (33.94%), oligochaetes (28.96%) and isopods (27.60%) were also frequent.

 

 

 

 

 

 

Table IX - Occurrence (O), numerical (N) and weight (W) frequencies and dietary coefficient (Qi) of the preys present in the gut contents of P. microps.

Taxa

%O

%N

%W

Qi

Bivalves

33.94

11.30

41.30

466.76

Oligochaetes

28.96

27.38

7.33

200.69

Polychaetes

50.00

32.50

14.57

473.53

Isopods

27.60

7.86

3.88

30.53

Amphipods

9.50

0.80

4.23

3.39

Mysids

35.29

8.73

18.10

158.00

Copepods

10.41

7.52

0.02

0.16

Decapods

6.11

0.54

2.18

1.19

Insects

2.49

0.30

0.07

0.02

Fish

6.56

0.78

8.30

6.43

Others

4.30

2.28

0.02

0.04

 

According to the results of the dietary coefficient (Table I) three preys constitute the preferential items in the feeding habits of P. microps. Polychaetes, mostly the spionidae Streblospio shrubsolii, bivalves, almost exclusively siphons of Scrobicularia plana, and Oligochaetes, with a dietary coefficient of 473.53, 466.76 and 200.69, respectively. These results were mainly due to the high weight component (41.30%) in the case of the bivalves, while for the annelids there was a higher numerical contribution (32.50% and 27.38%, respectively for polychaetes and oligochaetes).

Secondary preys were mysids with a high weight component (8.73%) and isopods, mostly Sphaeroma serratum, with similar contributions from the numerical and weight components.

 

The seasonal analysis of the prey consumed by the common gobie (Table X) shows that the polychaetes were preyed with abundance during the all year, representing always between 25 and 37% of the total preys ingested. However there were some fluctuations in the consumption of other feeding groups.

The importance of the bivalves was mostly observed during the Autumn and Winter, respectively with 17.07% and 14.44% decreasing in the other two seasons.

 

Table X - Numerical frequencies per season of the preys present in the gut contents of P. microps.

Taxa

Autumn

Winter

Spring

Summer

Bivalves

17.07

14.43

7.69

5.21

Oligochaetes

13.46

47.39

20.07

13.96

Polychaetes

31.17

25.93

37.37

35.63

Isopods

8.81

9.81

6.59

4.38

Amphipods

0.48

0.46

1.20

0.63

Mysids

21.55

1.01

6.99

25.42

Copepods

1.12

0.59

16.13

0.21

Decapods

0

0

1.17

0.42

Insects

0.16

0.08

0.55

0

Fish

0

0.04

1.72

0

Others

6.17

0.25

0.52

14.17

 

In the Winter it was observed the higher consumption of oligochaetes, representing in this season 47.39% of the total number of preys, although there was a marked decrease in the abundance of this taxa during the other seasons, their numerical frequency was always superior to 13%.

The ingestion of mysids was higher during the Summer and Autumn, while the copepods were only important during the Spring with 16.13%, when the young of the year began to colonise these areas.

The relative importance of the group “others”, which joins the remaining taxa, in the Summer is due to the abundant presence of ostracods and acaridae during this season.

 

The analysis of the feeding habits of P. microps per length classes (Figure 19) shows that the copepods were the most abundant taxa consumed by the individuals of the classes I and II, representing in the class I (Lt £ 19mm) 78.70% of the total number of preys ingested. Nevertheless, the mysids were also important part of the diet of this size individuals, as in for the remaining classes. During the goby growth the importance of the copepods decreases, while the polychaetes began to increase. Thus, in the classes III and IV this taxon was the most represented one with numerical frequencies superior to 30%, followed in order of abundance by the mysids with near 19%.

 


Figure 19 - Numerical frequencies of the taxa ingested by de different length classes of P. microps. Class I £ 19mm; 20mm £ class II £ 24mm; 25mm £ class III £ 29mm; 30mm £ class IV £ 34mm; 35mm £ class V £ 39mm; 40mm £ class VI £ 44mm; 45mm £ class VII £ 49mm; class VIII ³ 50mm.

 


For the length classes V, VI and VII the oligochaetes and the polychaetes represents more than 60% of the taxa consumed. However an increase in the abundance of bivalve siphons was noticed. In the individuals with total length superior to 49mm (class VIII) the bivalves were the most important item in numerical frequency with 33.68% followed in decreasing abundance by oligochaetes and polychaetes. In this size class it was noted a higher abundance in the consumption of a larger number of taxa.

From the total number of guts analysed during the ebb tides only 8.89% were empty.

 

The analysis of the gut contents of the 46 gobies collected during flood tides showed that 69.6% were empty. Only 4 taxa were present on the remaining ones, fish (1 clean vertebral column), spionidae (2 individuals), copepods (18 individuals) and gnathiidae (18 individuals).

 

 

5.1.4           Discussion

 

The diet P. microps included bivalves, gastropods, annelids, crustaceans, insects and fishes. Such versatility was associated with their capacity to use different feeding strategies such as biting and suction (Hamerlynck & Cattrijsse, 1994).

The preferential preys of the common goby were polychaetes, bivalves (siphons) and oligochaetes. These preference for endobenthic preys were observed by other authors, distinguishing polychaetes (Salgado et al., unpublished), the isopod Corophium volutator (Magnhagen, 1986; Pihl, 1985) and bivalve siphons (Gee, 1985) as dominant preys. Polychaetes, despite being present in the results obtain by these authors, never revealed the importance assumed in the studies performed in the Tagus estuary.

Comparing with the results obtained by Salgado et al. (unpublished) for the upper estuary subtidal areas, in the present study there was a higher abundance of epibenthic and pelagic taxa, as mysids, shrimps and fishes, while the presence of amphipods (mainly the benthic Corophium volutator) and copepods (essentially harpacticoids) was higher and there were no decapods in the guts analysed in that study. These difference in the results could be associated with different prey availability in both sites and with the absence of P. minutus in the salt marsh areas, a potential competitor for the same trophic niche.

 

This species is considered an opportunistic carnivore feeding on organisms that they select on the basis of relative availability (Pihl, 1985). This fact is corroborated by the abundance of the different preys on the digestive tracks of P. microps following their abundance in the creeks and by the presence of a great number of juveniles of different species for which the presence in the area is only seasonal.

Polychaetes and oligochaetes were the only taxon abundantly consumed during the all year. The ingestion of the remaining items was highly seasonal. The bivalves were mainly consumed during the autumn and winter, mysids mostly in summer and autumn and copepods almost exclusively in the spring, following their highest natural abundance’s in the study areas. The high number of copepods ingested during that season was also due to the abundant presence of small juveniles of P. microps in the area in this period.

 

Juveniles of P. microps increase their niche width as they grew (Thorman, 1983). The main preys of the smallest size classes (I and II) were epibenthic copepods, while mysids represent a secondary prey. With the growth of this goby dimension, it was observed a decrease in the importance of this prey and an increase in the consumption of all the other preys, most of them endobenthic. Thus, in the following size classes (III and IV), while the mysids maintained their importance, polychaetes began to be the main item ingested and bivalves were mostly important as biomass.

For classes V, VI and VII, the annelids were the preferential preys, with an increasing consumption of bivalves. This prey was the main taxon ingested by the biggest individuals caught (³ 50mm). The high abundance in the consumption of a larger number of taxa suggests that the individuals of this late size class explore a more wide trophic niche.

 

The reduced number of preys and the high percentage of empty guts in the analysis of the individuals captured on the mouth of the creek during flood suggests that this goby feeds almost exclusively on the items available inside the creeks. This fact is corroborated by the presence of a clean fish vertebral column inside one of those guts.

 

The anterior position of the mysids and bivalves in the digestive track suggests that these preys are preferentially ingested in the final period of permanency in the creeks while polychaetes and oligochaetes are mainly ingested in the upper areas of the salt marsh creeks or by the time they arrive there.

 

 

5.2           Food habits of Liza ramada (Risso, 1826) in the Tagus salt marsh

 

5.2.1           Introdution

 

The thin-lipped grey mullet (Liza ramada Risso, 1826) is one of the most successful inshore teleost fish that inhabit Portuguese waters (Almeida, 1996). In the Tagus estuary is one of the most common and abundant mugilids.

Feeding directly on the energy produced by the first level of the food web, they avoid the competition with other trophic groups and have a large contribute on the optimisation of the energy transference processes (Odum, 1970).

The feeding strategies of this species in the Tagus estuary were already investigated by Almeida (1996). Nevertheless, there is no information on the importance of the salt marsh for the feeding ecology of this mugilid and on their roll on the transference of organic matter from these highly productive areas to the subtidal areas of the estuary.

 

 

5.2.2           Methods

 

Samples of L. ramada were collected in May/June in Hortas salt marsh creek, using the same methodologies applied for P. microps. Samples were done during flood and ebb tides in order to compare the food intake before and after residence in salt marshes. The samples were collected in consecutive days, with tide of the same amplitude.

In the laboratory total length was measured to the nearest 1mm, wet weight was determined to the nearest 0.0001g and the contents of the cardiac region of the stomach was removed and weighted to the nearest 0.0001g (stomach content wet weight). A 200mg sample was removed from each stomach, and suspended in 5 ml of distilled water. The samples were shaken thoroughly and a known volume pipetted onto a slide with an etched grid. A constant area was examined in all the samples and the food items found in it were, wherever possible, identified to the genus level and counted. The identification and counting were made using a stereo microscope at x 400 magnification.

Determination of the percentage of organic matter in the stomach contents was derived from loss on ignition at 480ºC after 24h.

The fullness index, the frequency of occurrence, the numerical frequency of each food item and the vacuity index were estimated according to Hureau (1970).

 

 

5.2.3           Results and discussion

 

The analysis of the stomach contents of L. ramada (table XI) showed the presence of high quantities of sediment, being identified 18 taxa. The most important taxa ingested by this mugilid were the diatoms, while nematoda, foraminifera and copepoda were only taken occasionally. The most abundant food items were Gyrosigma sp. + Pleurosigma sp. and Navicula sp., representing together more than 80% of the total number of individuals identified. Almeida et al. (1993) obtained for Alcochete, in the Tagus estuary, similar results.

 

Table XI. Food items identified in the stomach contents of L. ramada.

Coscinodiscaceae

          Ciclotella sp.

          Coscinodiscus sp.

          Melosira sp.

Actinodiscaceae

          Actinoptychus sp.

Diatomaceae

          Fragillaria sp.

Achnanthaceae

          Achnanthes sp.

          Cocconeis sp.

Naviculaceae

          Diploneis sp.

          Gyrosigma sp.

          Pleurosigma sp.

          Navicula sp.

          Stauroneis sp.

Nitzschiaceae

          Nitzschia sp.

Surirellaceae

          Surirella sp.

Desmidiaceae

          Closterium sp.

Foraminifera

Nematoda

Copepoda

 

None of the stomachs analysed was empty.

The analysis of fullness index of the individuals sampled at the moment that the water arrives into the creeks (table XII) showed that at their arrived into the mouth of the creek their stomach content represents near 0.77% of their weight. However, when the same analysis was made to the individuals caught 30 min after high tide, their stomach contents represents 11.40% of their weight. In the following periods of 30 minutes this value decreases gradually, until 4.85% two hours after high tide.

 

 

Table XII. Fullness index and % of organic matter of L. ramada by the time the flood arrives in to the creeks and in intervals of 30 min. after high tide.

 

Time

 

Nº ind.

Fullness index

Standard deviation

% organic matter

Standard deviation

Flood

45

0.77

0.45

9.02

3.10

30 min

13

11.40

3.83

10.63

0.87

60 min

13

9.72

2.87

11.64

1.28

90 min

16

6.22

4.06

11.57

1.47

120 min

4

4.85

2.93

11.90

1.61

 

In this area L. ramada feeds on the extensive surface of the mudflat areas that becomes accessible to the mullet only with the rising time that they follow, as observed by Almeida et al. (1993). Nevertheless, the period of highest ingestion begins by the time they arrive into the creeks until 1/2 hour after high tide. After this period of intense grazing activity they stop feeding and return to deepest waters. This fact is confirmed by the highest densities of L. ramada obtained for the first 1/2 hour of ebb tide, diminishing the risk of stranding inside the creek.

A possible reason for that feeding behaviour could be associated with the higher densities of microphytobenthos inside the creeks (pers. obs.). The comparison between the percentages of organic matter of the stomach contents of the individuals that feed outside with those that feed inside the creeks, show higher values for the latest ones confirming the presence of higher food availability inside the creeks.

 

The total biomass of L. ramada present in the creek during one tide, of the same month, was 246 503.69 g. Thus, it is possible to estimate in more than 25 Kg per tide the amount of material transported by this species from the salt marsh towards other estuarine areas.

However, these results need to be confirmed with more samples in order to exclude the hipothesys that the individuals caught during the flood tides were not actively feeders during that tide.

 

 

5.3           Isotope analysis

 

Based on the first part of the work, biological characterization of the area, and on the literature available k-species were selected (Tab. XIII) as representative of the different levels of the food web in the Tagus estuary. Multiple stable isotope analysis (C13/ C12, S34/ S32, N15/ N14) of those taxa were used for the identification of the different pathways of the Tagus salt marsh food web.

 

Table XIII - List of the species representative of the different levels of the food web in the Tagus salt marsh

Flora

 

Microphytobenthos

 

Macrophytes

Spartina maritima

Arthrocnemon fruticuosum

Halimione portulacoides

 

Fauna

 

Polichaeta

Hediste diversicolor

Streblospio shrubsolii

Oligochaeta

Oligochaeta n.i.

Gastropoda

Peringia ulvae

Bivalvia

Scrobicularia plana

Isopoda

Cyathura carinata

Sphaeroma serratum

Amphipoda

Orchestia gammarela

Copepoda

Copepoda n.i.

Misidacea

Neomysis integer

Decapoda

Palaemonetes varians

Crangon crangon

Pisces

Liza ramada

Pomatoschistus microps

Sardina pilchardus

 

 

5.3.1           Methods

 

Samples for stable isotope analysis were collected during Spring, due to the higher availability of the taxa sellected, in two salt marsh tidal creeks, Hortas and Vasa sacos. The preparation of the samples for isotope analysis was similar as described for the identification of the main sources of organic matter. For the fish species Pomatoschistus microps and Liza ramada different size classes were used for comparison with the results obtained in the gut contents analysis. Two classes for P. microps (juveniles and adults) and three for L. ramada (juveniles < 50 mm, juveniles > 50 mm and adults ).

 

 

5.3.2           Results

 

Due to problems in the mass spectrometer only a small part of the results are available, not allowing any kind of analysis. The 13C/12C isotopic values available are represented in table XIV. We are expecting at any time the rest of the C results, as well as those of N and S.

 

Table XIV - Isotopic values (‰) of 13C/12C available for both sites Hortas and Vasa sacos.

Taxa

Hortas

Vasa sacos

Scrobicularia plana (bivalvae)

-19.17

-15.86

Oligochaeta

-

-15.20

Capitelidae (polychaeta)

-16.69

-

Hediste diversicolor (polychaeta)

-16.31

-

Sphaeroma monodi (isopoda)

-

-18.93

Orchestia gammarela (amphipoda)

-17.17

-

Peringia ulvae (gastropoda)

-16.31

-16.16

Palaemonetes varians (decapoda)

-17.06

-17.46

Pomatoschistus microps adult (pisces)

-16.04

-15.96

Liza ramada small juvenile (pisces)

-17.95

-17.90

Liza ramada big juvenile (pisces)

-14.05

-14.21

Liza ramada adult (pisces)

-

-16.35

 

 

5.4           Food web

 

The food web of the upper Tagus estuary (including subtidal and lower intertidal areas) and of the salt marsh/mud flat areas (medium and upper intertidal areas) is represented respectively in figures 20 and 21. In those diagrams are present the most abundant taxa present in the areas representative of the different levels of the food web.

In the first areas the plant biomass, mainly from riverine inputs, microalgae and salt marsh species, is decomposed and enters the detritus pathway. Microbial fungi and bacteria are primary consumers wich are preyed by small crustaceans, as the amphipod Corophium volutator. Bacteria, fungi and detritus were considered in the same group.

As deposit feeders there is a high abundance of oligochaetes and polychaetes (mainly spionidae and Hedistes diversicolor). According to Silva (1993) the rag worm H. diversicolor consumes mostly inorganic and decomposed organic matter, although the ingestion of microalgae and some benthic invertebrates also occurs. This species seems to adopt an omnivorous diet with detritivorous predominance, emerging from his tube to feed on the surface

The gastropod Peringia ulvae is a deposit feeder that also grazes on the benthic microalgae.

A large population of the bivalve peppery furrow shell (Scrobicularia plana), are present in these areas. Being a detritus feeder this species suck the detritus from the mud surface using the inhalant siphon and vacuum-cleaning the substratum.

 

The green crab Carcinus maenas is an omnivorous species that feeds on a large spectrum of death and living preys. According to Martins (1995) the main items ingested by this species are fish (especially gobies), shrimps, small crustaceans and polychaetes, as H. diversicolor.

The brown shrimp Crangon crangon, the most abundant natantia in the area, has as preferential food items polychaetes and small crustaceans, but they also prey on bivalves and small fishes (Pomatoschistus spp.). Being a generalist, the ingestion of the different preys is dependent on their abundance (Martins, 1995).

 


 



The estuarine resident fish are represented by 4 species. The common goby Pomatoschistus microps and the sand goby P. minutus feed mainly on polychaetes, oligochaetes, bivalve siphons and small crustaceans, such as copepods, mysids, amphipods, isopods and shrimps. Canibalism was also observed for both species (Salgado, 1995).

The pipe-fish Syngnathus sp. and the european anchovy Engraulis encrasicolus are planktivorous species, ingesting preferentially mysids.

 

Two catadromous migratory species occur in these areas. The common eel Anguilla anguilla according to Costa et al. (1992) ingests green crabs Carcinus maenas and fishes as preferential and secondary food sources. However, shrimps and amphipods are also important items on this species diet.

The thin-lipped grey mullet Liza ramada, is one of the most common species in the area, being the most abundant in terms of biomass. This species is a detritus feeder grazing on the extensive mudflat areas where they ingest high quantities of benthic microalgae such as Bacilliarophycea, Cyanophycea, Zygophycea and a large amount of detritus (Almeida, 1996).

 

These upper areas are used as nursery by mainly three fish species, two soles Solea solea (common sole) and Solea senegalensis (Senegal sole) and the sea bass Dicentrarchus labrax. The diet of the soles is mainly composed of polychaetes (especially spionidae and H. diversicolor) and amphipods (Corophium volutator) (Cabral, 1998). According to that author the main difference on both species diet is the higher amount of bivalves consumed by S. senegalensis.

The diet of the young sea bass is mainly based on the ingestion of crustaceans, such as decapods (mainly shrimps), mysids and amphipods (Corophium volutator).

 

The Tagus estuary has very important wintering populations of several shore bird species, mainly herons, flamingos, gulls and ducks, which use the intertidal areas as feeding grounds. It is also the most important site for wintering waders in the country.

 

According to Moreira (1995) heron’s prey on small fish and decapod crustaceans such as the brown shrimp and the green crab. The gulls in the area feed mostly on bivalves (Scrobicularia plana) and polychaetes (mainly H. diversicolor), eating also crustaceans.

The diet of the waders inclued oligochaetes, polychaetes, the gastropode P. ulvae and the bivalve S. plana, which is ingested as siphons or as the whole individual. Moreira (1995) found that this last item represents a significant percentage of biomass removed by the bird population.

 

Comparativelly, in the salt marsh/mud flat food web (figure 21) there is a higher contribution of the salt marsh plants for the total plant biomass available for the first consumers.

The abundance of the amphipod C. volutator and the shrimp C. crangon are very reduced in the salt marsh. However, their niche is occupied by the isopod Sphaeroma monodi and by the shrimp Palaemonetes varians.

Nevertheless, the major difference between both food webs is the absence or presence in lower abundances of several fish species, as the ell, the sea bass, the sand goby and the soles. Most of those species are potencial predators of small fish. Thus, young of the year of several fish species feeding in the salt marsh areas decrease the risk of being preyed by other fish species. The reduced presence of the green crab in this area is also a factor contributing to that lower predation risk.

 

What concerns the birds, the presence of flamingos and ducks is restricted to the salt marsh and the close mud flat areas (Moreira, 1995). The ducks eat salt marsh vegetation, while the flamingos use the intertidal areas adjacent to the salt marsh to feed on small invertebrates.

 

 

6           References

 

Adão, H. & Marques, J.C. 1999. Ecologia da meiofauna dos sedimentos dos povoamentos de Zostera nolti, no estuário do rio Mira (Costa Sudoeste de Portugal). Seminário “A Zona Costeira do Alentejo”.

Almeida, P.R. 1996. Biologia e ecologia de Liza ramada (Risso, 1826) e Chelon labrosus (Risso, 1826) (Pisces: Mugilidae) no estuário do Mira (Portugal). Inter- relações com o ecossistema estuarino. Ph.D. Thesis. Faculdade de Ciências da Universidade de Lisboa, Lisboa.

Almeida, P. R; Moreira, F. M.; Costa, J. L.; Assis, C. A. & Costa, M.J. 1993. The feeding strategies of  Liza ramada (Risso, 1826) in fresh and brackish water in the River Tagus, Portugal. Journal of Fish Biology, 42: 95-107.

Austen, M.C.; Warwick, R.M. & Rosado, M.C. 1989. Meiobenthic and macrobenthic community structure along a putative pollution gradient in southern Portugal. Marine Pollution Bulletin, 20 (8): 398-405.

Bell, S.S. 1979. Short- and long-term variation in a high marsh meiofauna community. Estuarine, Coastal and Marine Science, 9 (3), 331-350Bettencourt, A. 1979. Estuário do Tejo. Preliminair document (in Portuguese). Hidroprojecto, Drena, Lisboa.

Blanchard, G.F. 1990. Overlapping microscale dispersion patterns of meiofauna and microphytobenthos. Marine Ecology Progress Series, 68: 101-111.

Blome, D. & Faubel, A. 1996. Eulitoral nematodes from the Elbe estuary: Species composition, distribution, and population dynamics. Archiv fur Hydrobiologie. Supplementband. Untersuchungen des Elbe-Aestuars. Stuttgart, 110 (2-3): 107-157.

Blome, D.; Schleier, U. & Bernem, H.V. 1999. Analysis of the small-scale spatial patterns of free-living marine nematodes from tidal flats in the East Frisian Wadden Sea. Marine Biology, 133, 717-726.

Bouchereau, J.-L.;. Quignard, J.-P; Joyeux, J.-C. & Tomasini, J.-A. 1993. Structure du stock des géniteurs de la population de Pomatoschistus microps (Krøyer, 1838) (Gobiidae), dans la lagune de Mauguio, France. Cybium 17(1): 3-15.

Bouchereau, J.-L., Quignard, J.-P; Tomasini, J.-A.; Joyeux, J.-C.; & Capape, C. 1990. Cycle sexuel, condition, fécondité et ponte de Pomatoschistus minutus (Pallas, 1770) (Gobiidae) du Golfe du Lion, France. Cybium 14(3): 251-267

Bouwman, L.A. 1987. Meiofauna. in Baker, J.M. & Wolff, W.J. (ed.). Biological Surveys of Estuaries and Coasts. Cambridge University Press, 141-156.

Bouwman, L.A.; Romeijn, K. & Admiraal, W. 1984. On the ecology of meiofauna in an organically polluted estuarine mudflat. Estuarine, Coastal and Shelf Science, 19 (6), 633-653.

Bozeman, E.L. Jr. & Dean, J.M. 1980. The abundance of estuarine larval and juvenile fish in a South Carolina intertidal creek. Estuaries 3(2): 89-97.

Brotas, V. 1995. Distribuição espacial e temporal do microfitobentos no estuário do Tejo. Pigmentos fotossintéticos, povoamentos e produção. Ph.D. Thesis. Faculdade de Ciências da Universidade de Lisboa, Lisboa.

Brotas, V.; Cabrita, T.; Portugal, A.; Serôdio, J & Catarino, F. 1995. Spatio-temporal distribution of the microphytobenthic biomass in intertidal flats of Tagus estuary (Portugal). Hidrobiologia 300/301: 93-104.

Brotas, V & Serôdio, J. 1995. A mathematical model for the vertical distribution of Chlorophyll a in estuarine intertidal sediments. Netherlands Journal of Aquatic Ecology, 29 (3-4), 315-321.

Cabral, H.N. 1998. Utilização do estuário do Tejo como área de viveiro pelos linguados, Solea solea (Linnaeus, 1758) e S. senegalensis Kaup, 1858, e robalo, Dicentrarchus labrax (Linnaeus, 1758). Ph.D. Thesis. Faculdade de Ciências da Universidade de Lisboa, Lisboa.

Cain, R.L. & Dean, J. M. 1976. Annual occurrence, abundance and diversity  of fish in a South Carolina intertidal creek. Marine Biology 36: 369-379.

Calvário, J. 1982. Estudo ambiental do estuário do Tejo. Povoamentos bentónicos intertidais (substratos móveis). CNA/Tejo nº19, Technical repport (in Portuguese), Lisboa.

Cartaxana, P. & Lloyd, D. 1999. N2, N2O and O2 profiles in a Tagus estuary salt marsh. Estuarine, Coastal and Shelf Science, 48, 751-756.

Catarino, F.; Tenhunen, J. D.; Brotas, V. & Llange, O. 1985. Application of CO2-porometer methods to assessment of components of photosynthetic production in estuarine ecosystems. Marine Biology 89: 37-43.

Cattrijsse, A.; Makwaia, E.S.; Dankwa, H.R.; Hamerlynck, O. & Hemminga, M.A. 1994. Nekton communities of an intertidal creek of a European estuarine brackish marsh. Marine Ecology Progress Series 109: 195-208.

Costa, M.J. 1982. Contribution à l’étude de l’écologie des poissons de l’estuaire du Tage (Portugal). Ph.D. Thesis. Université Paris VII, Paris.

Costa, M.J. 1988. Écologie alimentaire des poissons de l’estuaire du Tage. Cybium, 12(4): 301-320.

Costa, M.J. 1999. O estuário do Tejo. Edições cotovia. Lisboa.

Costa, M.J.; Almeida, P.R.; Cabral, H.N.; Costa, J.L.; Lopes, M.T.; Pereira, C.D.; Correia, M.J.; Teixeira, C.; Salgado, J.P. & Martins, M.J. 1996. Monitorização ambiental da construção da ponte Vasco da Gama, bentos/macroinvertebrados aquáticos e ictiofauna. Technical repport (in Portuguese). Instituto de Oceanografia. Faculdade de Ciências de Lisboa, Lisboa

Costa, J. L.; Assis, C. A.; Almeida, P. R; Moreira, F. M. & Costa, M.J. 1992. On the food of european eel, Anguilla anguilla (L.), in the upper zone of the Tagus estuary, Portugal. Journal of Fish Biology, 41: 841-850.

Costa, M.J. & Bruxelas, A. 1989. The structure of fish communities in the Tagus estuary (Portugal) and its role as a nursery for commercial fish species. Scientia Marina 53(2-3): 561-566.

Costa, M.J. & Cabral, H.N. 1999. Changes in the Tagus nursery function for commercial fish species: some perspectives for management. Aquatic Ecology 33: 287-292.

Costa, M.J.; Costa, J.L.; Almeida, P. R & Assis, C. A. 1994. Do eel grass beds and salt marsh borders act as preferential nurseries and spawning grounds for fish? - An example of the Mira estuary in Portugal. Ecological Engineering 3: 187-195.

Costa, M.J.; Gordo, L.; Almeida, P.R.; Costa, J.L.; Silva, G.; Caçador, I. & Melo, R.A. 1999. Monitorização da zona de intervenção do Parque EXPO’98. Techniocal repport (in Portuguese), Instituto de Oceanografia. Faculdade de Ciências de Lisboa, Lisboa.

Costa, M.J.; Pereira, C.D.; Jorge, F.; Silva, G.; Salgado, J.P.; Costa, J.L.; Gordo, L.S. & Almeida, P.R. 1998. Environmental impact monitoring of the construction of the Vasco da Gama Bridge. Macrozoobenthic invertebrates and fishes. Technical report (in Portuguese). Instituto de Oceanografia. Faculdade de Ciências de Lisboa, Lisboa.

Coull, B.C. 1985. Long-term variability of estuarine meiobenthos: an 11-year study. Marine Ecology Progress Series, 24, 205-218.

Coull, B.C. 1988. Ecology of the marine meiofauna. in Higgins, R. P. & Thiel, H. (ed.). Introduction to the study of meiofauna. Smithsonian Institution, 18-37.

Coull, B.C. 1999. Role of meiofauna in estuarine soft-bottom habitats. Australian Journal of Ecology, 24, 327-343.

Danovaro, R. 1996. Detritus-Bacteria-Meiofauna interactions in seagrass bed (Posidonia oceanica) of the NW Mediterranean. Marine Biology, 127, 1-13.

Dinis, M.T. 1986. Quatre soleidae de l’estuaire du Tage. Reproduction et croissance. Essai d’élevage de Solea senegalensis Kaup. Ph.D. Thesis. Université de Bretagne Occidentale, Brest.

Drake, P. & Arias, A.M. 1991. Composition and seasonal flutuations of the ichthyoplankton community in a shallow tidal channel of Cadiz Bay (S. W. Spain). Journal of Fish Biology 39: 245-263.

Elliott, M. & Dewailly, F. 1995. The structure and components of european estuarine fish assemblage. Netherlands Journal of Aquatic Ecology 29: 397-417.

Escravage, V.; García, M.E. & Castel, J. 1989. The distribution of meiofauna and its contribution to detritic pathways in tidal flats (Arcachon Bay, France). Topics in Marine Biology. Proceedings of the 22nd European Marine Biology Symposium, Inst. de Ciencias del Mar, Barcelona (Spain), Scientia Marina, 53 (2-3), 551-559.

Essink, K. & Keidel, H. 1998. Changes in estuarine nematode communities following a decrease of organic pollution. Aquatic Ecology, 32, 195-202.

Fenchel, T. 1996. Worm burrows and oxic microniches in marine sediments. 1. Spatial and temporal scales. Marine Biology, 127, 289-295

Fleeger, J.W.; Palmer, M.A. & Moser, E.B. 1990. On the scale of aggregation of meiobenthic copepods on a tidal mudflat. P.S.Z.N. I: Marine Ecology, 11 (3), 227-237.

Fleeger, J.W.; Thistle, D. & Thiel, H. 1988. Sampling equipment. in Higgins, R.P. & Thiel, H. (ed.). Introduction to the study of meiofauna. Smithsonian Institution, 115-125.

Fleeger, J.W.; Whipple, S.A. & Cook, L.L. 1982. Field manipulations of tidal flushing, light exposure and natant macrofauna in a Louisiana salt marsh: effects on the meiofauna. Journal of Experimental Marine Biology and Ecology, 56 (1), 87-100.

Forster, S. 1996. Spatial and temporal distribution of oxidation events occurring below the sediment-water interface. P.S.Z.N. I: Marine Ecology, 17 (1-3), 309-319.

Gee, J. M.; Warwick, R. M.; Davey, J. T. & George, C, L. 1985. Field experiments on the role of epibenthic predators in determining prey densities in an estuarine mudflat. Estuarine, Coastal and Shelf Science, 21:429-448.

Gooday, A.J. 1988. Sarcomastigophora. in Higgins, R. P. & Thiel, H. (ed.). Introduction to the study of meiofauna. Smithsonian Institution, 126-133.

Hamerlynck, O & Cattrijsse, A 1994. The food of Pomatoschistus minutus (Pisces: Gobiidae) in the Belgian coastal waters, and a comparison with the food of its potential competitor P. Lozanoi. Journal of Fish Biology, 44:753-771.

Heip, C. & Herman, P.M.J. 1995. Major biological processes in European tidal estuaries: A synthesis of the JEEP-92 project. Hydrobiologia, 311 (1-3), 1-7.

Heip, C.; Goosen, N.K.; Herman, P.M.J.; Kromkamp, J.; Middelburg, J.J. & Soetaert, K. 1995. Production and consumption of biological particles in temperate tidal estuaries. Oceanography and Marine Biology: an Annual Review, 33, 1-149.

Heip, C.; Huys, R. & Alkemade, R. 1992. Community structure and functional roles of meiofauna in the North Sea. Netherlands Journal of Aquatic Ecology, 26 (1), 31-41.

Heip, C.; Huys, R.;, Vincx, M.; Vanreusel, A.; Smol, N.; Herman, R. & Herman, P.M.J. 1990. Composition, distribution, biomass and production of North Sea Meiofauna. Netherlands Journal of Sea Research, 26 (2-4), 333-342.

Higgins, R.P. & Thiel, H. 1988. Introduction to the study of meiofauna. Smithsonian Institution, Washinton.

Hureau, J. 1970. Biologie comparée de quelques poisson antarctiques (Nototheneiidae). Bulletin de l’Institute Oceanographique (Monaco), 68, 1-244.

Joint, I.R.; Gee, J.M. & Warwick, R.M. 1982. Determination of fine-scale vertical distribution of microbes and meiofauna in an intertidal sediment. Marine Biology, 72 (2), 157-164.

Kneib, R.T. 1984. Patterns in the utilization of the intertidal salt marsh by larvae and juveniles of Fundulus heteroclitus (Linnaeus) and Fundulus luciae (Baird). Journal of Experimental Marine Biology and Ecology 83: 41-51.

Kneib, R.T. 1997. Early life stages of resident nekton in intertidal marshes. Estuaries 20(1): 214-230.

Labourg, P.J.; Clus, C. & Lasserre, G. 1985. Résultats préliminaires sur la distribution des juvéniles de poissons dans un marais maritime du Bassin d’Arcachon. Oceanologica Acta 8(3): 331-341.

Laffaille, P.; Brosse, S.; Feunteun, E.; Baisez, A. & Lefeuvre, J.-C. 1998. Role of fish communities in particulate organic matter fluxes between salt marshes and coastal marine waters in the Mont Saint-Michel Bay. Hydrobiologia 373/374: 121-133.

Li, J. & Vincx, M. 1993. The temporal variation of intertidal nematodes in the Westerschelde. I: The importance of an estuarine gradient. Netherlands Journal of Aquatic Ecology, 27, (2-4), 319-326.

Li, J., Vincx, M. & Herman, P.M.J. 1996. A model of nematode dynamics in the Westerschelde Estuary. Geochimica et Cosmochimica Acta, 60 (6), 1019-1040.

Lorenzen, C.J. 1967. Determination of chlorophyll and phaeopigments: spectrophotometric equations. Limnology and Oceanography, 12, 343-346.

Ludwig, J.A. &. Reynolds, J.F. 1988. Statistical Ecology. A Primer on Methods and Computing. John Wiley & Sons, New York.

Magnhagen, C. 1986. Activity differences influencing food selection in the marine fish Pomatoschistus microps. Can. J. Fish. Aquat. Sci., 43: 223-227.

Martins, M. J. 1995. Estratégias alimentares de crustáceos decápodes no estuário do Rio Mira: Cracinus maenas (L.), Crangon crangon (L.), Palaemon longirostris Milne-Edwards, Palaemon serratus Pennant e Penaeus kerathurus Forskal. Relatório de estágio. Faculdade de Ciências de Lisboa: 41p.

Merilaeinen, J.J. 1988. Meiobenthos in relation to macrobenthic communities in a low saline, partly acidified estuary, Bothnian Bay, Finland. Annales Zoologici Fennici, 25 (4), 277-292.

Miltner, R.J.;. Ross, S.W & Posey, M.H. 1995. Influence of food and predation on the depth distribution of juvenile spot (Leiostomus xanthurus) in tidal nurseries. Canadian Journal of Fisheries and Aquatic Sciences 52: 971-982.

Montagna, P.A.; Coull, B.C., Herring, T.L. & Dudley, B.W. 1983. The relationship between abundance of meiofauna and their suspected microbial food (diatoms and bacteria). Estuarine, Coastal and Shelf Science, 17 (4), 381-394.

Moodley, L.; Heip, C. & Middelburg, J.J. 1998a. Benthic activity in sediments of the northwestern Adriatic Sea: sediment oxygen consumption, macro- and meiofauna dynamics. Journal of Sea Research, 40, 263-280.

Moodley, L.; Schaub, B.E.M.; van der Zwaan, G. J. & Herman, P.M.J. 1998b. Tolerance of benthic foraminifera (Protista: Sarcodina) to hydrogen sulphide. Marine Ecology Progress Series, 169, 77-86.

Moodley, L.; van der Zwaan, G.J.; Rutten, G.M.W., Boom, R.C.E. & Kempers, A.J. 1998c. Subsurface activity of benthic foraminifera in relation to porewater oxygen content: laboratory experiments. Marine Micropaleontology, 34, 91-106.

Moreira, F. 1991. Alguns aspectos de taxonomia, crescimento e reprodução de Pomatoschistus minutus (Pallas, 1770) e Pomatoschistus microps (Kroyer, 1838) (Pisces: Gobiidae) na parte superior do estuário do Tejo. Relatório de estágio. Faculdade de Ciências de Lisboa: 116p.

Moreira, F. 1992. Aves piscivoras em ecossistemas estuarinos: dieta da Garça branca pequena (Egretta garzetta) e Garça real (Ardea cinerea) num banco de vasa do estuário do Tejo. Airo, 3(1): 9-12.

Moreira, F. 1995. A utilização das zonas entre-marés do estuário do Tejo por aves aquáticas e suas implicações para os fluxos de energia na teia trófica estuarina. Ph.D. Thesis, Faculdade de Ciências da Universidade de Lisboa, Lisboa.

Odum, W. E. 1970. Utilization of the direct grazing and plant detritus food chains by the striped mullet Mugil cephalus. In Marine Food Chains (Steele, J. H., ed.), pp. 222-240. Edinburgh: Oliver & Boyd.

Pestana, G. 1989. Manancial Ibero-Atlantico de sardinha (Sardina pilchardus Walb.) sua avaliação e medidas de gestão. Dissertação apresentada para provas de acesso à categoria de Investigador Auxiliar. INIP, Portugal.

Pfannkuche, O. & Thiel, H. 1988. Sample processing. in Higgins, R. P. & Thiel, H. (ed.). Introduction to the study of meiofauna. Smithsonian Institution, 134-145.

Pihl, L. 1985. Food selection and consumption of mobile epibenthic fauna in shallow marine areas. Marine Ecology - Progress Series, 22: 169-179.

Pielou, E. C., 1966. The measurement of diversity in different types of biological collections. J. Theor. Biol., 13:131-144.

Rakocinski, C.F.; Baltz, D.M. & Fleeger, J. W. 1992. Correspondence between environmental gradients and the community structure of marsh-edge fishes in a Louisiana estuary. Marine Ecology Progress Series 80: 135-148.

Ré, P. 1984. Ictioplâncton da região central da costa Portuguesa e do estuário do Tejo. Ecologia da postura e fase planctónica de Sardina pilchardus (Walbaum, 1792) e de Engraulis encrasicolus (Linné, 1758). Ph.D. Thesis. Faculdade de Ciências da Universidade de Lisboa, Lisboa.

Rosado, M.C. 1996. Caracterização ecológica das comunidades meiobentónicas intertidais do estuário do Sado. Ph.D. Thesis, Faculdade de Ciências da Universidade de Lisboa, Lisboa.

Rosado, M.C. & Bruxelas, A. 1995. Meiofauna do estuário do Sado. Estudos de Biologia e Conservação da Natureza, I. C. N., 17.

Rozas, L.P. 1995. Hydroperiod and its influence on nekton use of the salt marsh: A pulsing ecosystem. Estuaries 18(4): 579-590.

Ruhmor, H., 1990. Soft bottom macrofauna: collection and treatment of samples. Tech. Mar. Environ. Sci., nº8, ICES. Copenhagen .

Rutledge, P.A. & Fleeger, J.W. 1993. Abundance and seasonality of meiofauna including Harpacticoid Copepod species, associated with stems of the salt-marsh cord grass Spartina alterniflora. Estuaries, 16 (4), 760-768.

Salgado, J. P. 1995. Pomatoschistus minutus (Pallas, 1970) e Pomatoschistus microps (Kroyer, 1838). Estratégia alimentar e relações interespecíficas. Relatório de estágio. Faculdade de Ciências de Lisboa: 40p.

Sandulli, R. & Pinckney, J. 1999. Patch size and spatial patterns of meiobenthic copepods and benthic microalgae in sandy sediments: a microscale approach. Journal of Sea Research, 41, 179-187.

Santos, P.J P.; Castel, J. & Souza-Santos, L.P. 1995. Microphytobenthic patches and their influence on meiofaunal distribution. Cahiers de Biologie Marine, 36 (2), 133-139.

Santos, P.J.P.; Castel, J. & Souza-Santos, L.P. 1996. Seasonal variability of meiofaunal abundance in the oligo-mesohaline area of the Gironde Estuary, France. Estuarine, Coastal and Shelf Science, 43 (5), 549-563.

Shannon, C. E. & Weaver, W., 1963. The mathematical theory of communication. University of Illinois Press, Urbana Illinois.

Shenker, J.M. & Dean, J.M. 1979. The utilization of an intertidal salt marsh creek by larval and juvenile fishes: Abundance, diversity and temporal variation. Estuaries 2(3): 154-163.

Silva, P. C. 1993. Estudo de uma população de Nereis diversicolor O. F. Muller, 1776 (annelida, polychaeta) no estuário do Mira. Relatório de estágio. Faculdade de Ciências de Lisboa: 81p.

Smol, N.; Willems, K.A.; Govaere, J.C.R. & Sandee, A.J.J. 1994. Composition, distribution and biomass of the meiobenthos in the Oostersschelde estuary (SW Netherlands). Hydrobiologia, 282-283, 197-217.

Soetaert, K.; Vincx, M.; Wittoeck, J. & Tulkens, M. 1995. Meiobenthic distribution and nematode community structure in five European estuaries. Hydrobiologia, 311 (1-3), 185-206.

Soetaert, K.; Vincx, M.; Wittoeck, J.; Tulkens, M. & Van Gansbeke, D. 1994. Spatial patterns of Westerschelde meiobenthos. Estuarine, Coastal and Shelf Science, 39 (4), 367-388,

Steyaert, M.; Garner, N.; Van Gansbeke, D. & Vincx, M. 1999. Nematode communities from the North Sea: environmental controls on species diversity and vertical distribution within the sediment. Journal of Marine Biological Association of the United Kingdom, 79, 253-264

Steyaert, M. & Vincx, M. 1996. The vertical distribution of meiobenthos in coastal sediments (Belgium). Bulletin de la Société Royale de Sciences de Liège, 65 (1), 155-157.

Thistle, D. & Fleeger, J. W. 1988. Sampling strategies. in Higgins, R. P. & Thiel, H. (ed.). Introduction to the study of meiofauna. Smithsonian Institution, 126-133.

Thorman, S. 1983. Food and habitat resource partitioning between three estuarine fish species on the Swedish West coast. Estuarine, Coastal and Shelf Science, 17: 681-692.

Vanreusel, A. 1991. Ecology of free-living marine nematodes in the Voordelta (southern bight of the North Sea). II. Habitat preferences of the dominant species. Nematologica, 37, 343-359.

Vincx, M. 1989. Seasonal fluctuations and productions of nematode communities in the Belgian coastal zone of the North Sea. Verhandelingen van het symposium “Invertebraten van Belgie”, 57-66.

Weinstein, M.P. & Brooks, H.A. 1985. Comparative ecology of nekton residing in a tidal creek and adjacent seagrass meadow: community composition and structure. Marine Ecology Progress Series 15: 15-27.

Weinstein, M.P.; Weiss, S.L. & Walters, M.F. 1980. Multiple determinants of community structure in shallow marsh habitats, Cape Fear River estuary, North Carolina, USA. Marine Biology 58: 227-243.

Yoklavich, M.M.; Cailliet, G.M.; Barry, J.P; Ambrose, D.A. & Antrim, B.S. 1991. Temporal and spatial patterns in abundance and diversity of fish assemblages in Elkhorn Slough, California. Estuaries 14(4): 465-480.

Zar, J.H. 1996. Biostatistical analysis. 3rd ed., Prentice Hall, New Jersey.