Research Article |
Corresponding author: Sirlei Maria Hentges ( sirleihentges95@gmail.com ) Academic editor: Ana Maria Leal-Zanchet
© 2021 Sirlei Maria Hentges, Tieli Cláudia Menzel, Cristiane Maria Loebens, Samuel Elias Siveris, David Augusto Reynalte-Tataje, Milton Norberto Strieder.
This is an open access article distributed under the terms of the Creative Commons Attribution License (CC BY 4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Citation:
Hentges SM, Menzel TC, Loebens CM, Siveris SE, Reynalte-Tataje DA, Strieder MN (2021) Structure of aquatic macroinvertebrate communities in streams of a sub-basin in the Pampa Biome, Southern Brazil. Neotropical Biology and Conservation 16(2): 249-271. https://doi.org/10.3897/neotropical.16.e60579
|
The Piratinim River is located in the northwest of Rio Grande do Sul, and represents an important effluent from the Uruguay River, with streams located far from urbanized areas, in conditions similar to those of environmental integrity, but under the influence of agricultural activities. In this study, we aim at investigating the structure of the aquatic macroinvertebrate community in streams of the Piratinim river basin by observing both spatial and local scales. The sampling was carried out in six streams distributed in three regions (upper, middle and lower) of the basin, thus exploring its upstream, intermediate and downstream stretches, during spring and autumn. Macroinvertebrates were collected using aquatic dipnets and were identified at the family level; trophic-functional categories were established according to the classification adapted to the state of Paraná. The spatial and temporal variations of the groups and of the biotic attributes (density, richness, diversity and equitability) were subjected to Kruskal-Wallis non-parametric tests and a posteriori Dunn’s tests. The abiotic variables were obtained to verify possible influence on the composition of the macroinvertebrate community, evaluated through a Canonical Correspondence Analysis. We sampled 11,564 macroinvertebrate individuals from 72 taxa, and found a predominance of the collector-filter trophic group. Abundance and richness were different between streams; the highest densities were found in the streams located in the upper region of the watershed (Chuní and Itú). The highest taxon richness was found in the lower region of the watershed (Guaracapa stream), and the lowest richness was found in the two streams for the intermediate region (Santana and Ximbocu). Diversity and equitability did not vary; temporal variations were not found. Canonical correspondence analysis explained 31.7% of the data variability. The main environmental variables that influenced macroinvertebrates distribution were temperature, electrical conductivity, dissolved oxygen, altitude and extension of the riparian forest. Seasonality and the longitudinal gradient along the basin represented determining factors for the structure and distribution of the macroinvertebrate community in the tributary streams of the Piratinim River.
environmental descriptors, lotic environments, Piratinim watershed, river ecology, seasonal variations
One of the main goals of community ecology is to identify processes that influence the patterns of community structure along spatial and temporal scales. Patterns observed in communities at a particular scale are often the consequence of complex interactions among several processes occurring at multiplex scales (
Rivers and streams are complex ecosystems with patchy environmental heterogeneity at multiple spatial and temporal scales (
Benthic macroinvertebrates are an important group of organisms which are found in sediment present beneath the water column and that act as key components in any aquatic ecosystem. Studying them is important because the macrobenthic organisms are known as indicators of anthropogenic stress due to their sedentary habitat. Furthermore, they maintain various levels of interaction between the community and environment (
Some studies indicate that aquatic macroinvertebrate communities show gradual changes in species richness and composition along the longitudinal dimension as not only the stream order and composition but also habitat complexity increase, giving rise to nested patterns composition or variations in macroinvertebrate functional groups (
The present study was conducted in the Piratinim River basin, an important river in southern Brazil and the Pampa biome. The pampa biome occupies a large portion of South American territory, including part of Argentina, Uruguay, and the state of Rio Grande do Sul, Brazil (
The samplings considered some of the main tributaries of Piratinim in its upper, middle and lower sections in two seasons of the year. The objective of our study was to apply a multiscale approach to identify the main factors leading the spatial and temporal changes in the structure of aquatic macroinvertebrates of Piratinim river basin, investigating: 1) How the structure and distribution of the macroinvertebrate community varies between basin regions, between streams and in different stretches of streams and 2) The relative importance of climatic seasonality.
This study was conducted in streams on the right bank of the Piratinim River sub-basin (Middle Uruguay River), located in the Northwest region of the state of Rio Grande do Sul, Brazil. The basin is located between the geographical coordinates 28°00' to 29°05'S and 54°05' to 56°00'W, covering the geomorphological province of the Southern Plateau. Its surface area is approximately 7,596.07 km2, with approximate dimensions of 155 km in the East-West direction, and 40 km North-South (
According to the Köppen global climate classification (
The coverage area of the basin is inserted in the phytoecological region Steppe Savanna, within the Pampa biome, with natural and semi-natural remnants. In a small portion of its extension, there is an area of “Ecological Tension”, which configures a transition with the Seasonal Deciduous Forest (
In the studied region, spring is characterized by an increase in temperature and water volume in streams; autumn is characterized by a reduction in these two factors. These conditions were also observed during the study.
Sampling was carried out in biweekly collections, during spring and autumn: from September to November 2017, and from May to June 2018, respectively. Collections were performed at 18 sampling points, distributed in six streams (Figure
During field work, a 30–45 min sampling effort was applied to collect macroinvertebrates at each site. In order to determine the streams’ width, a measuring tape was used. A multi-parameter meter (YSI Professional Plus) was used to measure abiotic parameters, such as water temperature (°C), pH, dissolved oxygen (mg.l-1) and electrical conductivity (µS.cm-1).
Macroinvertebrates were sampled using an aquatic dipnet (250 µm mesh), from downstream to upstream, in an extension of approximately 30 m. As recommended by
In the laboratory, samples were analyzed using a stereomicroscope, at 45× magnification. Macroinvertebrates were identified, whenever possible, to the Family taxonomic level, using dichotomous and/or pictorial keys by
The trophic-functional groups were determined by the classifications of
To assess the structure of the macroinvertebrate community, data were transformed into log10 (x + 1), where “x” represents organism abundance in each family. We established a Frequency of Occurrence of at least 5% for the inclusion of taxa in the analysis, in order to exclude the influence of rare families.
We calculated the Shannon-Wiener diversity index (H’) and equitability (E), as well as taxa richness (S) and organism density (ind.m-2) (
The abiotic factors used to test the relation with the biotic data included variables of temporal nature, such as: seasons (spring and autumn), included as 2 dichotomous binary variables and the days of sampling (Julian dates); limnological variables (water temperature, dissolved oxygen, electrical conductivity and pH) and spatial variables (streams width, altitude of sampling points and riparian forests width). Variables were measured through georeferenced satellite images in Google Earth Pro browser.
To verify the influence of environmental variables on macroinvertebrates distribution, a Canonical Correspondence Analysis (CCA) (
A total of 11,564 macroinvertebrate individuals were sampled in the six streams of the Piratinim watershed. Of these, 6,089 specimens were sampled during spring, and 5,475 during autumn. Among the sampled macroinvertebrates, there are aquatic and semi-aquatic species, distributed in 72 taxa from the phyla Platyhelminthes, Nematomorpha, Anellida, Mollusca and Arthropoda (see Suppl. material
The most abundant taxa were Hydropsychidae (17.08%), Philopotamidae (15.46%) and Simuliidae (11.46%). Baetidae and Chironomidae were the most frequent families, occurring in all locations during the two sample seasons. Leptophlebiidae did not occur in only one location, Hydropsychidae in two, and Simuliidae in three. Eleven taxa were sampled exclusively at one point and one season, in low abundance.
Regarding the trophic-functional groups, the predator category had the highest taxonomic richness (N = 33), while shredders were represented only by 10 families. Considering abundance, the collectors-filters group (45.7%) predominated in the watershed, followed by gathering-collectors (26.6%), predators (17.3%), scrapers (8.5%) and shredders (1.9%) (see Suppl. material
When comparing all streams, the highest densities were found in the streams located in the highest region of the watershed: Chuní (3,312 ind.m-2) and Itú (2,450 ind.m-2) (p < 0.05). In the Guaracapa stream, located in the lower region of the watershed, the highest average taxon richness (24) was found. The lowest taxon richness was found in the two streams located in the intermediate region of the watershed, Santana and Ximbocu, with an average of 19 taxa (p < 0.05). Diversity and equitability showed no statistical difference between streams (Figure
Biotic attributes: density (A), richness (B), Shannon-Wiener diversity index (C) and equitability (D) of macroinvertebrate communities in streams of the Piratinim River watershed, southern Brazil. Average values represented by equal letters do not differ statistically. ±SD = standard deviation; ±SE = standard error.
Regarding trophic-functional groups, only the gathering-collectors group showed statistical difference when comparing streams (Figure
Temporally, there was no statistical difference in community attributes between the streams of the Piratinim River watershed (Figure
Biotic attributes: density (A), richness (B), Shannon-Wiener diversity index (C) and equitability (D) of the macroinvertebrate communities per season (spring and autumn) sampled in the Piratinim River watershed, southern Brazil. Average values represented by equal letters do not differ statistically. ±SD = standard deviation; ±SE = standard error.
When comparing seasons, predators occurred in higher densities during autumn (p<0.05), whereas the gathering collectors had a higher density during spring. The other groups did not vary between seasons (Figure
Abundance of trophic-functional groups of macroinvertebrate communities by season (spring and autumn), collected in the hydrographic basin of the Piratinim River, southern Brazil. Average values represented by equal letters do not differ statistically. ±SD = standard deviation; ±SE = standard error.
The forward-selection procedure of CCA method indicated only six variables to be included in the analysis: altitude, stream width, riparian forest, temperature, electrical conductivity and dissolved oxygen. The physical-chemical parameters of the water and the environmental characteristics of the streams (except altitude) are presented in the Suppl. material
According to the Canonical Correspondence Analysis (CCA), the first two axes were significant (p < 0.05) and together explained 31.7% of the data variation. Axis 1 explained 18.8% and axis 2 explained 12.9% of the total variability. In addition, a significant correlation was observed between abiotic and biotic variables (p < 0.05; Table
Result of the Canonical Correspondence Analysis (CCA) applied to the matrix of abiotic and macroinvertebrate data sampled in different streams of the Piratinim River basin, southern Brazil, during spring 2017 and autumn 2018 seasons. Monte Carlo test for the significance of the ordination first axis p<0.05 (n = 999 permutations). Values in bold were statistically significant (p<0.05).
Total variance: 2,28 | ||
---|---|---|
Environment characteristics | CCA1 | CCA2 |
Electric conductivity | -0.48 | 0.42 |
Dissolved oxygen | 0.59 | 0.40 |
Altitude | -0.31 | -0.48 |
Riparian forest | 0.21 | 0.40 |
Width | -0.27 | -0.21 |
Temperature | -0.76 | -0.44 |
% of explanation | 18.8 | 12.9 |
Pearson’s Species-Environment Correlation | 0.91 | 0.88 |
CCA1 was mainly influenced by temperature, which clearly separated samples made during different seasons in the plot, where autumn samples were grouped in the right side and spring samples in the left side. Electrical conductivity increased during the spring, and it is also related to the left side of the plot. Dissolved oxygen had the highest levels recorded during autumn, in the right side.
CCA 2 presented a spatial segregation between the sampling points located upstream and downstream of the basin, influenced by altitude and riparian forest. The points of the Chuní and Itú streams, located at the bottom of the plot, are characterized by a higher altitude and larger extension of riparian forest. The points of the Ivaí and Guaracapa streams, located at the top of the plot, are characterized by a lower altitude and reduced riparian forest (Figure
A,B Result of the Canonical Correspondence Analysis (CCA) applied to the matrix of abiotic data and families of macroinvertebrates sampled during the seasons of spring 2017 and autumn 2018, in the Piratinim River basin, southern Brazil. Streams: CH = Chuní; GU = Guaracapa; IT = Itú; IV = Ivaí; S = Santana and XI = Ximbocu.
The right side of CCA1 plot was characterized by the presence of the families Lutrochidae (r = 0.49), Ceratopogonidae (r = 0.57), Gerridae (r = 0.61), Veliidae (r = 0.53), Corydalidae (r = 0.60), Libellulidae (r = 0.55) and Philopotamidae (r = 0.59). The left side, related to the increase in temperature and conductivity, mainly showed representatives of the families Chironomidae (r = -0.48), Simuliidae (r = -0.59), Baetidae (r = -0.66) and Hydropsychidae (r = -0.48). In CCA2, the lower side of the plot was mainly characterized by representatives of Hydrobiosidae (r = -0.62) and Baetidae (r = -0.41) (Figure
Studies in neotropical aquatic environments are mostly focused on areas affected by urbanism, mainly because these studies seek to investigate the themes of environmental degradation and public health through disease vectors, or because those environments are often located near the research institution (
In this study we covered a drainage basin with considerable variation in geomorphological conditions and landscape structure (
We found a highly diverse macroinvertebrate fauna in the tributaries of the Piratinim River. Other studies carried out in southern Brazil indicated a lower taxa representativeness (
Regarding the most abundant taxa, Hydropsychidae and Philopotamidae were the most representative (32.54% of the total sampled individuals). These taxa are grouped in the order Trichoptera, which is considered one of the most diverse insect Orders in lotic environments, with families widely distributed in temperate and tropical latitudes (
In the sampled tributaries, some insect families, such as Baetidae (Ephemeroptera) and Chironomidae (Diptera), were present in all collection locations. A similar result was obtained by
Usually, Ephemeroptera are used as bioindicators of good water quality; however, Baetidae has a wide range of pollution tolerance, which can occur in areas under intermediate impact. Likewise, Chironomidae is an indicator of disrupted environments; however, some genera have higher tolerance to organic contamination than others (
Density varied significantly among the investigated tributaries, and the streams in the upper region of the basin had the highest average density of individuals per m2. The sampling points from this region, even though containing the most considerable portion of riparian vegetation among the investigated streams, show a physiognomy of the surroundings conditioned to disturbances in agricultural and livestock practices close to the banks, with direct access by cattle. In part, such results may be associated with the level of disturbance around these water courses, which favor the predominance of tolerant or resistant taxa (
For the richness attribute, the highest average was recorded in the Guaracapa stream, located in the lower region of the basin. This stream has a more closed riparian forest, which is difficult to access, when compared to other tributaries analyzed in this study. In streams in the middle region of the basin, the surrounding vegetation is present, but it is more conditioned to disturbances, and in these places the lowest average values of richness were found. Therefore, in the evaluated streams, the extent of riparian vegetation, solely, does not represent an indicator of environmental integrity. It is necessary to assess how the surrounding environment interacts with these areas, enabling sources of non-specific impacts on water courses.
Concerning the trophic-functional groups, besides the fact that predators showed a high taxonomic richness, the abundance was relatively low when compared with gathering-collectors and filtering-collectors. These two groups were predominant in the total sample, representing more than 70% of all sampled individuals. This is mainly due to the fact that the most abundant families (Hydropsychidae, Philopotamidae and Simuliidae) are mostly collector-filters. However, it is important to highlight that some families or groups may have a generalist habit, adapting to food resources availability (
According to
The analysis showed that seasonality represented the most determining factor regulating the distribution of macroinvertebrate taxa (CCA1). During the spring, the electrical conductivity parameter showed the highest values, most clearly associated with streams in the upper region of the basin. Among the families of which the spatial distribution was associated with such condition, are the dipterans Chironomidae and Simuliidae. During autumn, we recorded the highest levels of dissolved oxygen, being more associated with the mouth of the basin. For this season, several families were correlated, such as the hemiptera Veliidae and Gerridae.
Electrical conductivity can be considered one of the factors that significantly influence fauna distribution (
On the axis corresponding to CCA2, segregation of the upper and lower regions of the basin were once again observed. The streams located in the upper basin, of higher altitude and extension of riparian forest, were correlated with the distribution of the families Hydrobiidae and Baetidae. However, the streams located in the lower basin, of lower altitude, exhibited a reduction in the average extension of the riparian forest, in relation to the other streams.
In fact, among the environmental variables evaluated, the width of the riparian forest, given its presence and preservation, was the one that demonstrated crucial influence in determining the distribution of the macroinvertebrate community of the streams analyzed.
Degradation and removal of riparian vegetation is considered one of the factors that may lead to an increase in water temperature of streams (
However, despite the variations and the influence of riparian forest in taxa distribution, it is important to highlight that the average stream width in the upper basin was slightly higher than in the tributaries from the low region, considering that such variables may be proportional. According to
When high values of richness and diversity are verified in a study, it is assumed that the composition and structure of a community are similar to those found in areas of environmental integrity. The results found for the Piratinim River streams revealed a richness of aquatic macroinvertebrates similar or higher to those found in other studies conducted in southern Brazil, evidently influenced by seasonal and spatial aspects. However, these studies also demonstrate the influence of non-specific sources of impact affecting these organisms. The Piratinim River drainage basin is located in an agricultural region, where the reduction of natural vegetation coverage, especially native grasslands, contributes to landscape homogenization, also affecting aquatic biodiversity.
In this sense, fauna surveys and the investigation of spatial distribution patterns and community structure of aquatic macroinvertebrates contribute to fill the current knowledge gap concerning these taxa in the Pampa biome, and to expand the patterns already described for the south of Brazil. Therefore, awareness-raising and management activities are necessary to promote the maintenance of environmental quality along the evaluated water courses.
The authors are thankful to the Federal University of Fronteira Sul – Cerro Largo campus, for logistical support during the performance of this study.
Table S1
Data type: occurrence
Explanation note: Geographic coordinates and altitude of the streams at 18 sampling points in the Piratinim river watershed, southern Brazil, during spring 2017 and autumn 2018.
Table S2
Data type: species data
Explanation note: Taxonomic composition (diversity, absolute frequency and trophic-functional groups *) of aquatic macroinvertebrates collected in 18 sampling points of six streams in the hydrographic basin of the Piratinim River, southern Brazil, in the Upstream (U), Intermediate (I) and Downstream (D) sections, during the spring (P) of 2017 and autumn (O) of 2018. * Gtf: Trophic-functional group (P = predator, S = scraper, Sh = shredders, FC = filtering collector, GC = gathering collector).
Table S3
Data type: species data
Explanation note: Abiotic factors of the streams at 18 sampling points in the Piratinim river watershed, southern Brazil, during spring 2017 and autumn 2018.