Benthic insects of the El Tala River (Catamarca, Argentina): longitudinal variation of their structure and the use of insects to assess water quality

The aim of this work was to determine the structure of the benthic entomofauna and its variation along the El Tala River (Catamarca, Argentina). Five sampling stations were established, considering the location of nearby housing with respect to the watercourse. The following variables were determined in situ: altitude, latitude and longitude, bedstream width, river depth, river-current speed, water and air temperatures. Benthic insects were collected with a square parcel sampler of 0.09-m2 area and 300-μm net opening and identified to the family level. Faunal density, richness, and diversity exhibited a longitudinal variation. From sampling Stations 1 (reference site) to 3, the number of orders and families decreased, whereas in sampling Station 4 those values increased and continued to do so through to Station 5 (downstream station). Station 5 showed the highest family richness (17) and the highest value for the Shannon-Wiener index (2.74) and the lowest value in Simpson’s Dominance index (D = 0.22). These values could be explained because of the self-cleansing capabilities of the river downstream. The water quality of El Tala River is Class I (very clean and non-impacted), according to the results obtained from the application of the biotic Biological-Monitoring–WorkingParty and Average-Store-per-Taxon indices.


Introduction
The concept of river continuum refers to the species distribution in the river from its headwaters to its mouth, according to the water's characteristics; the availability of light and food and the occurrence of predators, among other characteristics (Vannote et al., 1980).
Living organisms are widely used as biotic indicators for the monitoring and evaluation of water quality, with the benthic macroinvertebrates being the most frequently recommended for aquatic ecosystems (Rosenberg and Resh, 1996;Figueroa et al., 2003;Guerrero-Bolaño et al., 2003;Pavé and Marchese, 2005;Giacometti and Bersosa, 2006;Gonçalves Oliveira et al., 1997;Correa-Araneda et al., 2010). In the Catamarca province (Argentina), benthic entomofauna have been used to evaluate the water quality of rivers and mountain streams by the application of the biotic indices from the Biological-Monitoring-Working-Party (BMWP') and Average-Store-per-Taxon (ASPT') (Rodríguez Garay, 2007;Salas, 2005Salas, , 2007. The El Tala River constitutes as one of the principal hydric resources for consumption and recreation, and at the same time, serves as the water reservoir for the city of San Fernando del Valle in Catamarca ('El Jumeal' Dam). In recent years there has been a marked increase in the number of settlements along the river's banks, the majority of these are precarious in construction and lack any system of sewage disposal (Saracho et al., 2002). We expect that the results of this report will indicate whether a health risk exists, or whether the rivers own self-cleansing capabilities are still sufficient to avoid a present threat to the well-being of the city's population. The aim of this study was to determine the structure and longitudinal variation of the benthic entomofauna in the El Tala River, in order to associate those parameters with the physicochemical characteristics of the water and thereby assess the water quality of the El Tala through the use of the benthic insect community as a bioindicator.

Study sites
The El Tala is a mountain river with a permanent water regime. It is located along the eastern side of the Ambato-Manchao Mountains and drains towards the central valley of Catamarca ( Figure 1). Five sampling stations were established along the river, considering the location of nearby housing and the access to the watercourse. Station 1 was located in an uninhabited area and taken as a reference site; Station 2 was located after the first group of settlements; Station 3 was located after the second group of settlements; Station 4 was located in an area without nearby dwellings; and Station 5 was established in an area without anthropic disturbances, where the water inlet for human consumption is located (Figure 1).
The following parameters were determined in situ at each station: the altitude, latitude, and longitude were calculated by means of a GPS eTrex Legend. The width of the stream bed consisted of a measurement of the stream bed at the water line to the north and to the south of a 15 m stretch. The rivers average depth was measured with a plastic ruler every 20 cm across the river from one bank to the other, and to the north and to the south of the 15 m stretch. The river current speed was determined with a stopwatch by measuring the average time a floating object (e. g., a tennis ball) took to traverse a distance of 15 m. The marginal arbustive and arboreal vegetation were noted by direct observation. The air temperature was determined with a mercury thermometer in the shade at a height of 1 m. The water temperature was measured with a mercury thermometer. The water physicochemical parameters were measured by means of a digital Cibar Corning multimeter (pH, electrical conductivity, total dissolved solids, and dissolved oxygen).

Collection of benthic entomofauna
A Surber sampler (area, 0.09 m 2 ; mesh size, 300 µm), was used to collect 3 sub-samples per station. The sampling was conducted in August 2006 (during the low-water season).
The samples were fixed in situ in 96% (v/v) aqueous ethanol and processed in a laboratory under a stereomicroscope (PZO Warszawa). The insects were identified at the family level through the use of specialised literature (Fernández and Domínguez, 2001).

Community parameters and diversity indices
The following community parameters were determined: abundance, density (individuals.m -2 ), and faunal richness or diversityie., the number of families per order of insects collected.

Water-quality evaluation through biotic indices
The BMWP' index was used after the adjustment in the northwestern region of Argentina (Domínguez and Fernández, 1998). This index gives a score to each macroinvertebrate family on the basis of their tolerance to different levels of organic contamination (higher scores indicate lower tolerances). The final index is obtained by calculating the values assigned to each family (Table 1). The result is then compared to the reference values to determine the class of the water quality. The ASPT' index was also calculated by dividing the value obtained for the BMWP' index by the number of taxa found (Klemm et al., 1990).

Statistical treatment
Univariate and bivariate descriptive statistical techniques were used. In addition, inferential and bivariate statistical techniques were applied by calculating the Pearson Correlation Analysis by means of the SPSS v.1.10 software. Multivariate techniques such as the Principal Component Analysis (PCA) were also applied through the use of the PC-ORD v.1.4 statistical program (Magurran, 1989).

Results
At the 5 sampling stations, the Chironomidae (Diptera) showed the highest density, while the rest of the orders and families exhibited considerably lower densities. The highest diversity corresponded to Tricoptera, with 5 families ( Table 1).
The community parameters determined for the benthic entomofauna showed variations throughout the 5 sampling stations. The highest and lowest densities were registered at Stations 2 and 3, respectively. The taxonomic composition fluctuated between 5 orders and 14 families at Station 3, to 7 orders and 17 families at Station 5. The Shannon-Wiener (log2) diversity index ranged from 1.46 to 2.74 bits, with the highest and lowest values being recorded at Station 5 and Station 2, respectively. Simpson's index ranged from 0.22 at Station 5 to 0.59 at Station 2. (Table 2).
The BMWP' index ranged from 78 points at Station 2 to 94 points at Station 1, while the ASPT' index fluctuated from 6 points at Station 2 to 6.75 points at Station 3. In all of the sampling stations, the values obtained for both indices corresponded to Class 1: very clean water with no anthropic impact (see Table 3).
The possible relationship between the water physicochemical variables (Table 4) and the community attributes of the benthic insects was analysed using the Pearson Correlation Analysis and its results are set forth in the discussion. Sorensen's Similarity index indicated that Stations 1 and 5 were the most similar (76%), with 12 families in common; while Stations 2 and 3 were the most dissimilar (41%), with 10 families in common (as shown in Figure 2).
A cluster analysis between the water physicochemical variables and the sampling sites indicated that axes I and II explained 99.99% of the variability (Table 5). Axis I (97.84% of the variability) clearly separated Station 1 from the rest of the stations, where the variables total dissolved solids and the BMWP'index value determined the clustering (as shown in Figure 3).
The results of the PCA conducted between the insect families and sampling sites indicated that axes I and II explained 98.02% of the variability (Table 6). Axis       (Figure 4).

Discussion
Studies conducted on watercourses in additional regions of the country (Miserendino, 1995;Mangeaud and Brewer, 1994;Scheibler, 2007;Pavé and Marchese, 2005) -and on those in other countries of the Neotropical Region (Gonçalves Oliveira et al., 1997;Baptista et al., 2001;Bueno et al., 2003) have indicated a relationship between the physicochemical variables of the water and the relative distribution of specific benthic entomofauna throughout a longitudinal gradient of those parameters.
In the sampling stations at the El Tala River, the water temperature was negatively and significantly correlated with the density of Perlidae and Ceratopogonidae and was likewise negatively and highly significantly related to the abundance of Psephenidae. This could be attributed to the preference of those families for cold water (Bachmann, 1995;Archangelsky, 2001).
The pH fluctuations were within the levels established for water-quality guidelines (i.e., 6.5-8.5) for any of the different uses of the resource (Agosba- OSN-Sihn, 1994). This parameter had a negative and significant correlation with the Shannon-Wiener diversity index and a negative and highly significant one with the richness in orders.
The dissolved-oxygen concentration was presumably not a limiting condition for the development of the benthic entomofauna since that parameter was maintained above the levels indicated by the water-quality guidelines (i. e., 4-5 mg.L -1 ; Agosba-OSN-Sihn, 1994).
The electrical conductivity showed a highly significant negative correlation with the presence of Perlidae, Psephenidae, Corydalidae and Tipulidae, and a significant negative one with the occurrence of Ceratopogonidae.
The total dissolved solids showed a significant negative correlation with the presence of Perlidae, Corydalidae, and Tipulidae, and a highly significant negative one with that of Ceratopogonidae and Psephenidae. The accumulation of suspended organic matter could act as a limiting condition for the development of these families; probably because of the concomitant reduction in the densities of the benthic algae (Marin, 2003).
Current velocity is a limiting condition for the colonization of benthic entomofauna (Boltovskoy et al., 1995). This variable was highly negatively correlated with the presence of Hydroptilidae.
With respect to the structure of the benthic entomofauna, the Chironomidae (Diptera) was dominant in the El Tala River. The chironomid dominance here agrees with the observations of other authors elsewhere (Argañaraz, 1988;Fernández et al., 1995Scheibler, 2007;Salas, 2007). The Simuliidae larval density in the sampling stations was low, with a maximum at station 3 and a minimum at station 1. The Tipulidae were collected only at Station 1, probably because of the levels of total dissolved solids. Ceratopogonidae was present at stations 1 and 2, but the absence of this family at the rest of the stations may have resulted from the higher values of water temperature, electrical conductivity, and total dissolved solids; with all three of those parameters acting as limiting conditions. Empididae was collected at all stations except Station 1, while Stratiomyidae was recorded only at Station 4. Although those two families make up part of the typical benthic entomofauna, both usually have low abundances overall .
The order of the Ephemeroptera was represented by the Baetidae, Leptohyphidae, and Leptophlebiidae families. Baetidae has the highest density at Station 1, which in turn, has an increased water temperature and higher levels of dissolved oxygen. Leptohyphiidae and Leptophlebiidae had the highest densities recorded at stations 5 and 1, respectively. The absence of the Ephemeroptera species at Station 3 could not be correlated with any of the different environmental variables analysed in this station.
Larvae and adults of Elmidae (Coleoptera) were collected in the present study, which was the most abundant family of the order, exhibiting the highest density at Station 5. The Elmidae larvae are collector-gatherers, whereas the adults are scrapers, mainly feeding here on the organic detritus (suspended material) which was abundant in all the sampling stations. The Psephenidae larvae had the highest density at Station 1 and the lowest at Station 4, in agreement with the low values of the physicochemical variables with which those larvae are inversely associated. Staphylinidae and Hydraenidae were poorly represented and were only collected at Station 3, which is consistent with data reported from other rivers of the northwestern region of Argentina .
Trichoptera exhibited the highest level of diversity among the orders. The highest density of Hydrobiosidae occurred at Station 5, probably as a result of a greater availability of food (other benthic organisms) .Hydropsychidae showed the highest density at Station 5 and the lowest at Station 1. Hydroptilidae was recorded at the highest density at Station 2 and the lowest at Station 1, in agreement with the lowest value of the current velocity, variables with which those larvae are inversely associated. The higher density of Leptoceridae was recorded at Station 4, while Glossosomatidae had the highest density at Stations 3 and 5.
Plecoptera was represented by Perlidae. A low larval density of this family was probably attributable to increasing values of water temperature, conductivity, and the concentration of total dissolved solids.
Megaloptera was represented by Corydalidae, which was recorded only at Station 1. This family was furthermore inversely correlated with electrical conductivity and the level of total dissolved solids.
Larvae of Libellulidae (Odonata) were poorly represented and only collected at stations 3 and 5.
Hemiptera was poorly represented in the El Tala River. Veelidae was recorded at Station 2, while Naucoridae and Gerridae were collected only at Station 5 -and in all instances at low densities.
The water quality of the segment studied in the El Tala River as evaluated according to biotic indices (the BMWP' and the ASPT') corresponded to Class I (very clean water with no anthropic impact). The variations observed in the index value at each sampling station cannot be attributed to environmental perturbations since the physicochemical parameters measured in the water were within the reference values for water for diverse uses (Agosba- OSN-Sihn, 1994). The high values of the biotic indices and the Shannon-Wiener diversity index, together with the good physicochemical quality of the water, allow the provisional inference that the section of the river studied manifested a good ecological status. Nevertheless, bacteriological and chemical analyses of the water (e. g., for organic matter and nitrogenated compounds) should be conducted in order to correlate those findings with the results obtained here by the measurement of biotic indices.