INTRODUCTION

Many of the changes observed in aquatic environments have resulted in large part from the expansion of agriculture and, therefore, of anthropic actions and are currently causing great concern regarding the availability and quality of water bodies¹. Rivers and streams are ecosystems that have been highly impacted by population growth, the discharge of high quantities of industrial and domestic effluents, construction of dams, destruction of habitats, and introduction of exotic species^2,3. These anthropic events result in imbalances in continental freshwater ecosystems and affect trophic interactions and biodiversity⁴. These effects result from increased nutrient and pollutant input originating from agricultural, industrial, and domestic sources⁵, which are the primary causes determining river quality and biodiversity decline. For this reason, much effort has been dedicated to detecting, quantifying, and mitigating these effects over the last few decades.

Natural lotic environments can be characterized by the presence, quantity, and composition of organic and inorganic agents and the diversity of their aquatic communities, and these habitats are subject to structural and seasonal variation due to internal and external factors⁶. The primary function of water resource monitoring is to evaluate whether certain organisms are present or absent in some regions of the lotic system. The quantity and diversity of species sensitive to frequent disturbances in these environments⁷, i.e., the balance between different communities in the ecosystem, are evaluated. The composition of aquatic organisms, especially invertebrates, reflects the changes in aquatic systems⁸ and is affected not only by factors specifically related to the water conditions but also by the dynamics between biotic and abiotic factors (biotope) of the ecosystem⁹.

Benthic macroinvertebrates are aquatic animals at least 0.25 mm in length that inhabit the sediment and can colonize different types of substrate, such as branches, leaves, stones, and aquatic macrophytes, during all or part of their life cycles^10,11. Their distribution and the specific communities that they form in aquatic environments are affected by the substrate composition, food availability¹², and the physicochemical characteristics of water¹³. Because the integrity of benthic macroinvertebrate communities is closely related to the presence of pollutants in the habitat, benthic macroinvertebrates can be effectively used in environmental impact studies to evaluate continental ecosystems^14,15. Therefore, these animals can be used as bioindicators of water quality, allowing for evaluating the ecological effects of different water pollution sources¹⁶. They can be used in environmental quality studies due to characteristics such as their varying abundances in different aquatic ecosystems (lentic and lotic), the extent of their locomotive ability, and their presence or absence in samples before and after environmental impact events^17,18,19,20.

Changes in the natural composition of benthic macroinvertebrate populations, regarding spatial dynamics, have been used as efficient biological tools in water pollution monitoring studies²¹. These studies gained full relevance when they described the impacts on aquatic trophic chains²². One study analyzed macroinvertebrate diversity at different points in the River Thames in the United Kingdom for one year using the Biological Monitoring Working Party (BMWP), average score per taxon (ASPT), and Shannon-Wiener diversity index to evaluate the impacts of water quality factors and physical changes to the habitats of these organisms²³. The results indicated that less altered sampling sites harbored greater diversity of organisms and that water quality was the primary limiting factor of biodiversity. In Brazil, the BMWP index has been used in studies performed in the Doce River basin in the state of Minas Gerais²⁴, in an Atlantic forest fragment also in Minas Gerais²⁵ and in the Sinos River in the state of Rio Grande do Sul²⁶, indicating that water quality biomonitoring studies in Brazil are still scarce and geographically restricted.

Many studies of continental aquatic environments have been performed in temperate and tropical climates using benthic macroinvertebrates as water quality indicators^27,28,29,30, but few studies have been conducted in the Amazon region. This lack is the cause of great concern since many of the cities in this region are closely associated with water bodies (e.g., for transport, recreation, and food). A ten-year study of bauxite waste discharge into Lake Batata in the municipality of Oriximiná, state of Pará, reported a decrease in the density of benthic macroinvertebrates in specific locations in the lake when compared to Lake Mussurá located in the same municipality, but that was not anthropized, which raised questions related to substrate particle size and its correlation to higher or lower biodiversity²⁹. A three-year study on physicochemical and biological conditions at five locations in the Mindu Stream in the city of Manaus in the state of Amazonas identified an apparent decrease in macroinvertebrate diversity in areas under anthropic pressure that was driven by the deforestation of riparian forests, illegal land occupation, and consequently high domestic waste disposal³¹. The same study also observed greater macroinvertebrate diversity in areas with less urbanization despite decreased riparian forests in those areas. A study of the water conditions at 12 points along the Maroaga River, located in Presidente Figueiredo, in the state of Amazonas classified the river as class I of the BMWP index, which means it is "clean or not significantly altered"³².

The present study aimed to evaluate the water quality conditions in a stretch of the Ouricuri River located in the municipality of Capanema in the northeastern state of Pará, using the BMWP and ASPT indices.

MATERIALS AND METHODS

Samples were taken from a stretch of the Ouricuri River where it runs along the urban perimeter of the municipality of Capanema, located approximately 165 km from Belém, the state capital of Pará (Figure 1). Five sampling sites were selected along the river: P1 (01°11'30.2"S, 47°09'36.1"W), located upriver from the city of Capanema, P2 (01°11'16.7"S, 47°10'16.0"W), P3 (01°11'22.1"S, 47°10'47.4"W), and P4 (01°11'23.8"S, 47°11'18.4"W) within the city grid, and P5 (01°11'22.8"S, 47°11'32.7"W) downriver from the city (Figure 2).

Datum: WGS84; Geographic Coordinate System Latitude/Longitude; Images: Landsat/DigitalGlobe/Google Earth Software; Map production date: February/2014.

Figure 1 - Location of benthic substrate sampling sites along the Ouricuri River in the municipality of Capanema, Pará, Brazil 

P: Sampling sites.

Figure 2 - Substrate sampling sites along the Ouricuri River in Capanema, Pará, Brazil 

Two substrate samples were taken at each sampling site during the drought season (October 2013). The samples were placed in 2-liter containers, labeled and transported to the laboratory, washed in running water, and filtered through a 300-μm mesh sieve, and the benthic macrofauna was preserved in 70% alcohol for subsequent analysis. According to the State of Rio de Janeiro Aquatic Macroinvertebrate Identification Manual (Manual de Identificação de Macroinvertebrados Aquáticos do Estado do Rio de Janeiro), sediment macrozoobenthos were isolated using a stereoscopic microscope, quantified, and identified to the family level³³. Intact and deep profile sediment samples were obtained using a polyvinyl polychloride tubular core sampler 7.5 cm in diameter and 80 cm long.

Physicochemical parameters of the water, such as temperature (°C), dissolved oxygen (mg/L), turbidity (NTU), total phosphorus (P), and organic matter (g/Kg), were evaluated at all sampling sites. Temperature, dissolved oxygen, and turbidity were evaluated using 1-liter water samples. Total phosphorus and organic matter samples (500 g) were sent to the Brazilian Agricultural Research Corporation (Empresa Brasileira de Pesquisa Agropecuária - Embrapa), Western Amazon.

Benthic macrofauna data were analyzed using the BMWP, ASPT, and Shannon-Wiener diversity index. The BMWP index classifies the degree of resilience of benthic macroinvertebrate families using a scale from 1 to 10. High BMWP scores were attributed to highly sensitive benthic taxa and low scores to benthic taxa with a high tolerance to organic pollution, enabling their use as a tool to diagnose water body contamination by organic material³⁴. The score for a given sampling site was obtained by adding the individual scores for all families present and the total score corresponded to a water quality class varying from good to very critical (Table 1).

The ASPT index was calculated by dividing the BMWP score by the number of families identified at a given sampling site. High ASPT scores indicate good water quality and represent a relatively high number of taxa present³⁵ (Table 2).

Biodiversity was evaluated by calculating the Shannon-Wiener index (H') [H'=-∑(pi)(log₂pi)], which supplies information on the stability of the benthic community³⁶, using BioEstat v5.0 software. The Shannon-Wiener index evaluates species abundance in a sample considering both evenness and richness of the species present. As pollution increases, the whole benthic community is exposed to intense stress, which consequently destabilizes it. The most sensitive organisms disappear, whereas the most tolerant, due to the absence of competition for food and space, spread rapidly throughout the aquatic system resulting in decreased biodiversity and thus a decreased H'.

RESULTS

The temperature did not vary significantly between sampling sites varying within approximately 1 °C.

However, there was significant variation in dissolved oxygen, turbidity, total phosphorus, and organic matter among the five sampling sites. As the water body approximated the urban area, dissolved oxygen decreased and other parameters increased (Figure 3).

Figure 3 - Variation of physicochemical parameters in the Ouricuri River, in Capanema, Pará, Brazil, across five sampling sites 

A total of 1039 individuals belonging to 14 families were collected and identified and constituted the benthic macrofauna at the five sampling sites visited in the Ouricuri River (Table 3). Following identification, the BMWP score was determined for each taxon. Of the 14 invertebrate families identified, 13 (92.8%) were found at P1 and P2, and only five (35.7%) were found at P3, P4, and P5. These results indicate an apparent decrease in biodiversity with increasing proximity to the city center. Similarly, the BMWP scores also decreased along the river (P1: 36; P2: 32; P3: 16; P4: 19; P5: 13).

Tolerant species were more abundant at P2, P3, and P4, and the presence of pollution-sensitive organisms and indicators of low pollution levels was only observed at P1 (Table 3). The BMWP scores classified the water quality in P1 as questionable, in P2, P3, and P4 as critical, and in P5 as very critical (Table 4). The ASPT scores varied from 3.1 to 4.3 and characterized the five sampling sites as impacted environments varying from moderate (P1 and P5) to severe (P2, P3, and P4) (Table 5). Similar to the determination by the BMWP score, the environmental quality of P1 was identified as being the least anthropized.

Shannon-Wiener index values varied from 0.27 to 0.49, with the highest for P1, P2, and P3 (Table 5). This index is based on the proportional abundance of a species in a sample and accounts for species evenness (equitability) and richness. Homogeneity (E; Pielou's equitability index) was calculated as the ratio between diversity observed and maximum diversity ranging from 0 to 1, where 1 indicates a situation in which all species in a given sample were equally abundant. The site with the highest equitability value was P3.

DISCUSSION

Although P1 was distant from the urban area, it exhibited similar taxonomic richness to P3 and P4, which were located within the urban area and strongly impacted. This result may be due to a combination of external and internal factors, namely the decrease in the riparian forest followed by shoreline erosion and low substrate organic matter content limiting the number of taxa present. A study conducted in Lake Batata in the municipality of Oriximiná reported a decrease in macroinvertebrate diversity in a water body exhibiting a high quantity of clay-dominated sediment originating from bauxite extraction, which altered the local sediment that was originally sandy³⁷. The increased turbidity downriver from P2 may be related to the decreased diversity observed in P3, P4, and P5.

A previous study evaluated the water and bottom sediment quality of a lake located in the municipality of Castelo in the state of Espírito Santo and showed that the instance of substrate phosphorus content greater than 40 mg/dm³ was high and could be related to eutrophication³⁸. The presence of a considerable quantity of macrophytes observed at P5 may be related to high substrate phosphorus contents, which would directly or indirectly aid their proliferation. Phosphorus is the main eutrophication agent in continental aquatic environments, and the decrease or disappearance of some specialist species and their replacement with more tolerant generalist groups is commonly observed in impacted aquatic environments³⁹.

The higher biodiversity observed at P2 was likely related to its less anthropized conditions, driven by higher organic matter and phosphorus levels at P3 and the remaining downriver sites. These increased values promoted increased metabolic activity in decomposers with a consequent decrease in dissolved oxygen and a reduction of the least tolerant taxa. This was confirmed by the presence of the Oligochaeta species, a bioindicator of anthropized environments, observed in higher quantities in P3.

The biological indices estimated for the sampling sites classified the Ouricuri River as questionable, critical, or very critical and moderately polluted to severely polluted. These biological index scores were lower than those observed in previous studies of continental Brazilian water bodies^24,30 in which a reserved area was compared to an urbanized area in the state of Minas Gerais and benthic invertebrate diversity BMWP scores ranged from 103 to 186. The change in the composition of the aquatic macroinvertebrate fauna in the Ouricuri River directly resulted from factors such as hydrological and substrate composition changes and organic material and sediment input. Specifically, a clear increase in organic material and sediment input, accompanied by a decrease in dissolved oxygen, was observed at P2.

The Shannon-Wiener index values for the five sampling sites were equal to or lower than those observed in a study of a stream in the micro basin of the Cambará River in the municipality of Cruz Alta in the state of Rio Grande do Sul⁴⁰. Based on the BMWP scores, the stream was characterized as having questionable water quality (H' = 0.46). Therefore, it was suggested that the five sampling sites analyzed had low diversity, and P4 and P5 were the most worrisome due to their low scores.

In the present study, indicator taxa of low water quality, such as Chironomidae, Oligochaeta, Planorbidae, and Planariidae, were present at sites considered altered and impacted. These benthic taxa are known to be highly resistant to anthropogenic changes, such as domestic or industrial waste discharges or the absence of riparian forests⁴¹. Dipterans belonging to the family Chironomidae were the most frequent taxa overall and were potentially responsible for a large part of the quantitative changes observed in the macrofaunal community. This is in accordance with the results of previous impact studies in which organic material discharge has been associated with urban expansion^42,43,44,45. The family Chironomidae is described as the most abundant organism in aquatic environments subjected to anthropic disturbances, more frequent in areas directly affected by water quality changes due to the presence of organic pollutants originating from domestic waste discharge^46,47.

CONCLUSION

The present study results showed evidence of environmental impact caused by anthropic actions for five sampling sites along the Ouricuri River. The presence of houses in several locations adjacent to the river was the main contamination source.