Short and long-term temporal variation of the zooplankton in a tropical estuary ( Amazon region , Brazil )

Nyctemeral and seasonal distribution of the zooplankton of the Taperaçu estuary and the effect of the hydrological parameters on the dynamics of these organisms were studied in March (rainy season) and September 2005 (dry season). Zooplankton samplings were carried out during spring tides at 2-hours intervals during a 24 h period. Samples were collected with a plankton net of 120 μm mesh size. Simultaneous temperature, salinity, pH and dissolved oxygen concentration measurements were accomplished at subsurface. The estuary presented a high seasonal variation in salinity (9.1-40.0) with regimes oscillating from mesohaline to polihaline/marine during the rainy and dry season, respectively. Temperature, pH and salinity were significantly higher during the dry season. A total of 50 taxa were identified, belonging to the following groups: Arthropoda, Sarcomastigophora, Cnidaria, Mollusca, Nematoda, Platyhelminthes, Bryozoa, Chordata, Annelida and Chaetognatha. Copepods dominated quantitatively the local zooplankton with 50% of the total organisms. Total zooplankton densities oscillated from 16,491 to 397,476 ind.m-3 during the dry and rainy season, respectively. Rainfall was the principal responsible for the hydrological parameters oscillations, mainly salinity, which influenced directly the population dynamics of the zooplankton of the Taperaçu estuary.


INTRODUCTION
Estuaries are located at the interface between the continental and marine domains (Lam-Hoai et al., 2006) and represent one of the most productive ecosystems worldwide (Day et al., 1989).This productivity is associated with peculiar characteristics, such as the constant environmental fluctuations to which these ecosystems are subjected which are resultant from the marine and freshwater influxes, retaining nutrient and other substances that stimulate its productivity (Elliott & Mclusky, 2002).
In these ecosystems, spatial and temporal variations of zooplankton are driven by environmental factors (rainfall, tidal currents, amongst others), food availability and the interaction of these factors (Li et al., 2000;Li et al., 2006).Zooplankton population size appears to be controlled by tidal exchanges (Grindley, 1984) and distribution patterns seem to be related to the strength of freshwater output.For estuarine mangrove ecosystems that are not subject to high salinity variation (i.e.estuaries with restricted freshwater discharge), tides have a major influence on the structure and density of the zooplankton communities present within the estuary (Rodriguez, 1975;Robertson et al., 1988).
The Amazon estuaries are characterized by semidiurnal macrotides and harbour the second largest continuous mangroves area on earth (Herz, 1991;Kjerfve et al., 2002).Rainfall regimen and tidal circulation in these ecosystems appears to be among the principal factors underlying its productivity.Perhaps surprisingly, detailed studies on the estuarine zooplanktonic communities from the Amazon littoral are scant and essentially focused on copepods specific composition (Wright, 1936 a,b;Calef & Grice, 1967;Krumme & Liang, 2004), copepods biomass (Magalhães et al., 2006) and on ichthyoplankton structure and dynamics (Barletta-Bergan et al., 2002).Thus, the purpose of this study was to determine the nyctemeral and seasonal variation in the zooplankton composition of Taperaçu estuary and to evaluate the effect of the hydrological parameters on the distribution patterns of these organisms.
It is located inside the Extractive Marine Reserve of Caeté-Taperaçu, which is used by riverine traditional extractive populations.Climate is equatorial hot and humid (Martorano et al., 1993) with annual average temperature and rainfall of 25.50 o C and 2,500 mm, respectively.This region shows two marked seasons: the rainy season (from December to May) and dry season (from June to November) (Souza-Filho & Paradella, 2001).

Sample collection and laboratory procedures
Zooplankton subsurface samples were collected each two hours in a fixed station (00 o 55'06.8"Sand 46 o 44'00"W) (Figure 1) during a nyctemeral cycle through horizontal plankton hauls of 3 minutes performed with a 120 µm mesh size plankton net equipped with a mechanical flowmeter to estimate the filtered water volume.Collections were carried out in the spring tides of March/2005 (dry period) and September/2005 (rainy period).Retained organisms were immediately fixed in 4% neutral formalin (Sodium tetraborate -borax).Subsurface water samples were collected and analysed in situ to determine conductivity (salinity), temperature, dissolved oxygen and pH using a multiparameter analyser.Rainfall data from the meteorological station of Tracuateua (01º05'S and 47º10'W), located 17 km far from the county of Bragança, were obtained from the 'Instituto Nacional de Meteorologia' -INMET.
Samples were identified to the lowest taxonomic level possible under an stereoscopic microscopy using specific literature (Trégouboff & Rose, 1957;Boltovskoy, 1981;Boltovskoy, 1999).To compensate for uncertainty in identification procedures, genera and more extended groups were also considered.Prior to sorting and identifying the organisms, samples were fractioned with a Folsom Plankton Sample Splitter (McEwen et al., 1957) owing to the high abundance of organisms or detritus.Frequency of occurrence was calculated and different taxa were grouped according to the criteria established by Matteucci & Colma (1982).
To evaluate the ecological patterns of zooplankton species from Taperaçu estuary, Shannon-Wiener specific diversity (H') and Pielou's equitability (J') were also calculated using the following equations: where pi is the number of individuals from species i divided by the number of individuals from all combined species (ni/N) and S corresponds to the total number of species in the sample.

Statistical analysis
Basic community data (organisms abundance and number of taxa) and community indices (Shannon-Wiener's diversity and Pielou`s equitability, using all taxonomic levels) were tested for differences among samples.The assumption of data normality and homogeneity of variances were tested through Lilliefors (Conover, 1971) and Bartlett's Chi-square test (Sokal & Rohlf, 1969), respectively, using STATISTICA 6.0 package (StatSoft, 2001).When data were not normal, log (x+1) transformations were performed to achieve near normality distributions.One-way ANOVA followed by Fisher LSD post-hoc test were applied to compare hydrological parameters (salinity, dissolved oxygen and temperature), total and most representative zooplankton densities, specific diversity and equitability between different day-night, seasonal and tide periods.However, when the variances were heterogeneous, non-parametric Mann-Whitney U-test was applied (Zar, 1999).To investigate similarities among periods, tides and sampling months, hierarchical agglomerative (WPGMA) analyses of similarity were calculated using Bray-Curtis similarity index and log (x+1) transformed total density (ind.m -3 ) data using the PRIMER (Plymouth Routines Multivariate Ecological Research) statistical package, version 6.1.6.Spearman rank correlation analyses were also used to verify correlations between biotic and abiotic variables employing the computer software package, STATISTICA, version 6.0.

RESULTS
Monthly average rainfall registered between 2000 and 2005 varied from 2.20 mm (November) to 510.60 mm (March), characterizing two seasonal periods: dry period (August-December) and rainy period (January-July), showing a wide range during the annual cycle.In the sampled months, rainfall oscillated between 0.30 mm in September/2005 and 501.60 mm in March/2005, with a total annual amount of 1,820.50mm, which represented a decrease of 5.1% with relation to the average obtained in the last 5 years (Figure 2).
Taperaçu zooplankton community was represented by 38 Arthropod taxa, two taxa each of Cnidaria, Mollusca and Chordata, and one taxon each of Sarcomastigophora, Chaetognatha, Nematoda, Platyhelminthes, Bryozoa and Annelidea.Copepoda was the Subclass comprising the higher number of Arthropoda, with 25 taxa (50%) (Table 1), from which six were more representative.
Cluster analysis (77% of similarity) based on species density in each sample showed three well defined groups (Figure 7).The first group (1) included samples of both seasonal periods, which presented the highest densities of Oithona oswaldocruzi (from 7,353 ind/m -3 to 61,869 ind/m -3 ).However, this group could be divided into two subgroups: 1a, comprised of rainy period samples with the highest observed densities; and 1b, including samples obtained in the dry period.Group 2 comprised samples collected in September (dry period) showing high densities of Paracalanus quasimodo.Group 3 was constituted of samples collected in the rainy period, with unidentified copepod nauplii and copepodites presenting the highest densities.Group formation showed a clear seasonal variation concerning to zooplankton composition and abundance.
Spearman correlation coefficient revealed a significant positive correlation between temperature and Cirripedia

DISCUSSION
According to the classification of estuaries (Mouny & Dauvin, 2002;Li et al., 2006 and references therein), Taperaçu estuary can be classified as a system with restricted freshwater influx characterized by a horizontal salinity and temperature gradient throughout the year (unpublished data).However, salinity differences are more accentuated between dry and rainy seasons, varying strongly as a consequence of the high rainfall observed mainly during the rainy period (monthly average values from 202.00 to 510.60 mm).The influence of a semidiurnal macrotide with heights varying from 4 to 6 m ( DHN, 2007) to which this estuary is subjected allows the incursion of coastal waters controlled by the river influx and regulate together with rainfall, environmental parameters variation and the structure of plankton communities.
During the field work, the rainfall data obtained from Taracuateua's Metereological Station (40 km from Taperaçu estuary) showed two seasonal periods: the rainy season (from January to July) and the dry season (from August to December).The observed periods are typical for Pará State (Northern Brazil), although they usually occur in most places from December  1 to May and from June to November, respectively, with a reported delay for the beginning of the rainy season in the Bragantine zone (Moraes et al., 2005).Salinity and temperature are indicated as crucial factors controlling spatial and vertical distribution of zooplankton (Schumann & Pearce, 1997;Kibirige & Perissinotto, 2003;Telesh, 2004) and their importance in determining the composition and distribution of these organisms are well documented (Beaugrand et al., 2000;Li et al., 2006).In the present study, salinity showed a strongly seasonal variation, which was related mainly to the observed pluviometric indices that were five hundred times higher during the rainy season.Variation in salinity as a consequence of rainfall regimen has been reported for other Northeastern tropical Brazilian estuaries (Lacerda et al., 2004), but it is more distinct for the estuaries in Northern Brazil, where it may vary from 0 to 39 (Magalhães et al., 2006;Diele & Simith, 2006).In Taperaçu estuary, pluviometric indices and consequently salinity variations influenced the zooplanktonic community structure, which was dominated by copepods and other estuarine, coastal, and oceanic zooplankton species.
Pseudodiaptomus richardi, Amphipoda, Calanopia americana, Centropages furcatus and Trematoda occurred exclusively during the dry period, showing the importance of salinity for the occurrence and population density of the referred taxa in Taperaçu estuary.On the other hand, brachyuran megalopa, fish eggs and larvae were registered only during the rainy period.The occurrence of brachyuran megalopa only in the rainy period agreed  well with data obtained by Diele (2000), who reported that the reproductive period of Ucides cordatus Linnaeus, 1763 (main Brachyura species of Amazon mangroves) in the Bragantine Peninsula is seasonal, occurring only in the rainy season (March to April).
Zooplankton total densities showed a markedly seasonal variation, with the highest values observed during the rainy season.The pluviometric regimen influenced not only hydrological parameters, but also the number of identified species which presented the highest values with the increase in salinity.At Vitoria Bay estuarine system (Southeastern Brazil), Sterza & Fernandes (2006) registered the highest zooplankton densities and diversity at the stations where the highest salinities were observed.In the present study, the contribution of coastal species increases diversity by the recruitment of coastal adjacent areas.This behavior was also described by other authors for tropical estuaries (French Guiana: Lam-Hoai et al., 2006) with similar characteristics as those recorded in Taperaçu estuary.
Diel vertical migration of the zooplankton organisms was not observed due to absence of stratification in the studied station sector as a consequence of both strong local hydrodynamic (e.g.equinoctial spring tide/ September and March, strong winds/September and high precipitation rates/March) and low local depth (up to 6.0 m).Different results can be found in environments of lower energy level, such as lakes or stratified estuary (Cardoso, 2000;Perticarrari et al., 2004).
Cluster analyses separate samples of March and September, confirming the seasonal variation of zooplankton species composition and abundance (Figure 7).Oithona oswaldocruzi was observed in 100% of the collected samples, showing the estuarine character of this species, just as it was observed for other estuarine systems (Ara, 2004).
Paracalanus quasimodo shows a euhaline-marine behavior, preferring areas with high salinities (Lopes et al., 1998), which explains its high abundances during the dry period in the studied station in Taperaçu estuary.
Spearman correlation coefficient showed a significant positive correlation between temperature and Cirripedia nauplii.Thus, it suggests that high densities of these organisms could be partly attributed to the highest water temperatures observed in the dry period.Also, the increased salinity during this period appears to corroborate the increase in nauplii densities, although the obtained data disagree with previous studies performed in the Caeté estuary (Bragantine region), which showed the dominance of settled Fistulobalanus citerosum Henry, 1973 (February and March) on artificial substrates during the rainy period (Marques- Silva, 2002).Probably, these observed differences could be attributed to environmental peculiarities or the breading period of a different species in Taperaçu estuary.
Euterpina acutifrons showed a significant positive correlation with temperature, salinity, dissolved oxygen and pH.This species is well represented in estuarine eutrophic ecosystems (Boltovskoy, 1981) and has been reported as one of the most abundant copepod species in South Atlantic coastal zones, occurring mainly during the dry period (Pereira et al., 2005;Sterza & Fernandes, 2006).
Oikopleura dioica, a typical marine species also found in estuarine polyhaline zones (Mouny & Dauvin, 2002), showed a significant positive correlation with salinity, dissolved oxygen and pH, presenting the highest densities during the dry period.
Although Silva et al. (2003) reported that larvae abundance of principal zooplankton taxa, mainly Decapoda larvae, was higher during the dry season in a Northeastern tropical Brazilian estuary, the present work showed a significant negative correlation between brachyuran zoea density and salinity and temperature.The obtained results, as explained before for megalopa larvae, were directly related to the breading period of Ucides cordatus which is reported to occur in this area during the rainy season (Diele, 2000).
A significant negative correlation was observed between Pseudodiaptomus marshi density and pH.The highest abundances recorded during the rainy period agreed well with data reported in Caeté estuary (Bragantine zone) by Magalhães (2003), who described the preference of this species for less saline estuarine zones.Pseudodiaptomus acutus density presented a highly significant negative correlation with pH and salinity, respectively.These results do not agree with those reported for Brazilian subtropical estuaries (Tundisi & Matsumura-Tundisi, 1968;Montú & Cordeiro, 1998), and neither with those recorded by Magalhães et al. (2006) in the Caeté estuary (Northern Brazil), which did not observe a clear distribution pattern for this copepod species.

CONCLUSION
The present results indicate that rainfall is the principal factor responsible for the oscillations of hydrological parameters oscillations, mainly salinity, which strongly influenced the zooplankton community temporal variation in the Taperaçu estuary.

Figure 1 .
Figure 1.Location of the study area and sampled station (black arrow) in Taperaçu estuary (Bragança, Pará, Brazil) (Modified of MADAM Project).