1. The distribution and abundance of the European eel, Anguilla anguilla, were investigated in the Mondego River, a strongly modified river in central Portugal. Nine freshwater sites, located below the first impassable obstacle to eel migration, were surveyed for 2 years. The study was conducted monthly during the first year and seasonally during the second year. Fish were caught by electric fishing and, physical and biotic variables potentially influencing eel length distribution and abundance were determined at each sampling session. 2. Canonical correspondence analysis (CCA) was used to explain the relationships between abiotic and biotic habitat variables, and eel length distribution. Spatial variation was strongly associated with distance from the sea, number of obstacles, river width and percentage of instream cover. Depth, river flow, water temperature and cannibalism had weaker effects on the longitudinal distribution, but contributed as explanatory variables for the model. The influence of spatial variables outweighed the importance of temporal variables as predictors in this model. 3. Length distribution changed in space from a dominance of small eels (<100 mm) closer to the sea to larger eels ( ‡250 mm) in the upper reaches of the river. The smallest length class (eels <100 mm) was strongly associated with wider stretches where instream cover, mainly composed of aquatic submerged macrophytes, was abundant (>75% area coverage). On the contrary, larger eels ( ‡250 mm) preferred deeper stretches with stronger river flow and less instream cover. 4. The effect of biotic interactions with conspecifics as well as with the non-native red swamp crayfish Procambarus clarkii, in sites with poor instream cover, modified this general pattern of distribution leading to fewer eels <100 mm at some sites closer to the sea. Therefore, despite the ubiquity and plasticity shown by this species, there were clear ontogenetic variations in habitat use such that the first two canonical axes of a CCA accounted for 58.4% of the spatial variability in size structure.

No abstract

No abstract

Type-specific reference conditions that describe sites with no or only very minor anthropogenic disturbances are a basic requirement of the European Water Framework Directive. The reference condition approach implies a previous determination of criteria acceptable for the definition of near-natural stretches. In this paper, a methodology based on selection and validation procedures is applied to a Portuguese watershed to guide the identification of reference sites. The methodology consisted of three phases: (1) a preliminary site inspection through the use of maps, available data and an extensive screening field campaign that resulted in the selection of 52 potentially undisturbed small-and medium-sized rivers stretches, ranging from 43 to 1069 m in altitude, with catchment areas between 4 and 641 km 2 and representing siliceous and calcareous riverbeds; (2) a detailed site selection based on ten pre-defined criteria involving physical/social/biological attributes that indicated catchment land use changes and alien vegetal riparian species as major human impacts; (3) a site validation procedure involving detailed investigation of benthic macroinvertebrate communities, riparian vegetation, in-stream habitat quality and chemical parameters that showed riparian wood-related problems as a major restriction to validate a stretch as a near-natural site. Only about 12% of all investigated river sites, accounting for 600 m of all 5200 m studied, could be considered as near-reference stretches or sites retaining essential natural functions. Selection and validation procedures can identify different sites as reference, which reinforces the need for applying both procedures. The results presented can help to accomplish the requirements of the EU -Water Framework Directive by selecting reference sites as the first step to establish biological reference conditions and, simultaneously, to form a basis for nature conservation strategies.

The influence of river flow on the fish community was assessed for the Tagus estuary (Portugal), based on sampling surveys carried out between 1979 and 2002. Four estuarine areas were sampled using similar fishing gear and effort in all the years considered in this study (1978-1980; 1995-1997; and 2001-2002). According to river freshwater flow values, sampling years were classified as wet (mean value of 714 m 3 s -1 , sd = 110 m 3 s -1 ) or dry (mean value of 164 m 3 s -1 , sd = 19m 3 s -1 ). Species richness varied between 22 and 39 according to the year, but no significant differences were related to river flow. The number of species per ecological guild was also similar in wet and dry years. Fish assemblage was dominated by marine occasional, estuarine resident and marine-estuarine opportunist species that represented near 90% of all fish species. The highest densities were represented by estuarine resident species. Fish density in dry and wet years differed significantly (mean density of 10.51 individuals 1,000 m -2 and 3.62 individuals 1,000 m -2 , respectively), and the major differences were registered for estuarine resident, marine-estuarine opportunist and catadromous species. These differences probably reflected the estuarine habitat availability and also differences in fish densities in some estuarine areas under different flow conditions. The multivariate ordination analyses performed outlined both seasonal and spatial variation trends in fish distribution and abundance. The estuarine longitudinal gradient and its relationship with species distribution were less evident in dry years. Relationships between species abundance and river flow were different according to species, which is probably due to different needs in the timing and magnitude of river flow.