Rapid population growth, changes in consumption habits and industrial developments from the last half of the 20th century to this time have caused significant water problems both on a global and regional scale [1–4].

While traditional water quality monitoring programs have focused on physicochemical monitoring, the limnologists have recently been increasingly reliant on biological assessments that provide important information about freshwater ecosystem conditions [5–18].

Biomonitoring studies by using bioindicators allow us to have information about the effects (chronic) that the aquatic habitat has been exposed to in the past [19, 20]. The most important advantage of using bioindicator groups to characterize the general condition of water is that they do well in reflecting the historical past of the stress sources in a region. Benthic macroinvertebrates form one of the indicator groups mostly used in determining the water quality [16].

Biotic indices are the focus of biological monitoring studies based on benthic macroinvertebrates. One of the difficulties in presenting biological observations is summarizing data and presenting it in specific ways. Therefore, various indices are used in Europe for water quality assessment studies using benthic invertebrates [21]. The Saprobi index in Germany, Biological Monitoring Working Party (BMWP) and Average Score Per Taxon (ASPT) in England, Belgian Biotic Index (BBI) in Belgium all seem to give the most reliable results specific to their geographic regions. The Yeşilırmak BMWP biotic index (Y-BMWP) has been recently developed for use only in the Yeşilırmak river basin in Turkey [22].

This study aims to assess the water quality of Sangı Stream. In addition, we aim to compare the results of five biotic indices and three diversity indices correlation with physicochemical characteristics of Sangı Stream (West Anatolia, Turkey).

MATERIALS AND METHODS

Study Area

Izmir, an area of historical importance, is located on the Aegean coast in West Anatolia of Turkey. Sangı Stream has been chosen as a study area because it is an important drinking water source for Izmir. The stream is one of the important water sources of the Tahtalı Dam basin in the Menderes district of Izmir. Sangı Stream is also used as a water source for irrigating the surrounding agricultural fields.

This study has been carried out on Sangı Stream in the Tahtali Dam Basin in Western Turkey. The length of the Sangı Stream is 25 km2. Sangı Stream flows from the Sandı Mountain (700 m) to the west of Tahtali Dam (Fig. 1).

Fig. 1.
figure 1

Map of the sampling stations in Sangı Stream.

Full size image

The sampling stations have been chosen according to the criteria for selecting operational monitoring sites given in WFD Annex V 1.3.2. [23]. The research has been conducted from February 2018 to March 2019 at six monitoring stations that have included the upstream (stations 1, 2, 3) and downstream (stations 4, 5, 6) parts of the stream. The sampling has been carried out on monthly in over a year at Sangı Stream. The characteristics of the sampling stations are presented in Table 1.

Table 1.   Key characteristics of sampling stations in Sangı Stream
Full size table

The sampling has been carried out from each station by using a kick-net with the classic 50 × 30 cm size and 250 µm mesh size according to the guidelines in scientific literature [24]. The collected samples have been fixed in ethyl alcohol (70%) and formaldehyde (4%) throughout the field study. The samples have been categorized and diagnosed to the lowest possible taxonomic level under a stereomicroscope.

Physicochemical Parameters

Simultaneously with macroinvertebrate sampling, water samples have been taken and analysed for the following parameters, PO4-P, NH4-N, NO2-N, NO3-N, Cl– and BOD5, using spectrophotometer and the proper kits. All analyses have been done by following the standard methods [25]. Water temperature (°C), pH, electrical conductivity (EC) and dissolved oxygen (DO) have been measured in the field using portable equipment.

Data Analysis

Saprobi Index (SI), Average Score Per Taxon (ASPT), Belgian Biotic Index (BBI), Family Biotic Index (FBI), Biological Monitoring Working Party (BMWP), Shannon–Wiener (SWDI), Simpson’s (SDI), Margalef (MDI) and Evenness (E1) indices have been applied on benthic macroinvertebrate data set by using ASTERICS Software Program [24]. The Bray–Curtis similarity index has been used to determine the similarities between the sampling stations based on macroinvertebrates [26]. The UPGMA algorithm has been used to illustrate similarity-based clustering relationships between sampling stations [27]. Pearson’s based correlation analysis has been performed by using SPSS version 20.0. PCA has been applied to transform the content from large data tables into a smaller data set that can be more easily visualized and analysed [28–31].

RESULTS AND DISCUSSION

The results of the analysed physicochemical variables of the water in six sampling stations located along the stream are presented in Table 2.

Table 2.   Summarized statistics of physicochemical parameters in sampling stations
Full size table

The mean value of T°C has varied from 11.9°C in station 1 in the source area of the stream, up to 16.5°C in stations 4 and 5. On the other hand, pH has shown small alteration, with minimum value of 6.80 in station 4 and maximum value of 7.11 in station 1, meaning that in all stations the stream water is alkaline [32, 33].

The increase in EC in drinking water indicates that the water is contaminated, or that sea water is mixed into the water. According to the One-Way Anova test, EC has varied significantly between stations (p < 0.05). The highest mean value of EC in this study has been recorded in station 4 (369.2 μS/m) in the downstream area.

Dissolved oxygen is a vital variable for a healthy aquatic life [34, 35]. The mean value of DO concentration has varied from 13.4 mg/L (station 1) to 9.65 mg/L (station 4). BOD5 is the amount of dissolved molecular oxygen used by microorganisms during the 5-day incubation period to oxidize the structure of organic substances in water at 20°C. The mean values of BOD5 in sampling stations has ranged from 1.16 mg/L in station 1 to 4.75 mg/L in station 4 [32, 33]. According to the One-Way Anova test, BOD5 has varied significantly between stations (p < 0.05).

Elements that limit efficiency in aquatic environments are mostly PO4-P, NH4-N, NO2-N, NO3-N [36]. The NO3-N range has been from 0.78 mg/L in station 1 to 2.55 mg/L in station 4. The increase of NO3- in station 4 and 5 has shown these two stations to be rich with nutrients caused by discharges of agricultural and domestic wastes in the vicinity of Sangı Stream. The minimum mean value of PO4-P has been 0.04 mg/L, registered in stations 1 and 2, and the maximum has been 3.19 mg/L in station 4. The minimum mean value of NO2-N has been 0.08 mg/L in station 1, and the maximum has been 2.66 mg/L in station 4. The minimum mean value of NH4-N has been 0.05 mg/L in station 1, and the maximum has been 3.72 mg/L in station 4. According to the One-Way Anova test, PO4-P, NH4-N, NO2-N, NO3-N have varied significantly between stations (p < 0.05).

In this study, the subdividing of the sampling points into two groups has caused the PCA analysis to show a clear spatial and temporal division (Fig. 2). According to PCA analysis, total variance has been explained as 93.9% in axis 1 and 2 where eigenvalues have been greater than 1. PCA 1 has positive loading on temperature, ToC, EC, BOI5, NH4-N, NO2-N, NO3-N, CI– and PO4-P and negative loading on DO and pH. Positive loadings on NH4-N, NO2-N, NO3-N and PO4-P have been related to organic pollution. Thus, the resulting components of PCA analysis represent pollution from agricultural and domestic wastewaters.

Fig. 2.
figure 2

Biplots for PCA analysis of water quality in Sangı Stream.

Full size image

In this study, a total of 5.363 benthic macroinvertebrate samples have been collected, belonging to nine groups: Crustacea, Oligochaeta, Gastropoda, Epheme-roptera, Plecoptera, Trichoptera, Odonata, Coleoptera and Diptera. Among all the taxonomic groups the following have been dominant: Ephemeroptera in station 1 (50.7%); Diptera in stations 2, 5 and 6 (14.1, 15.9 and 15.3%); and Crustacea in stations 3 and 4 (11.8, 11.5%) (Table 3).

Table 3.   Distribution of taxonomic groups (%)
Full size table

EPT-Taxa (%) has been one of the metrics with the best response to the physicochemical variables of freshwater. It is indicated that EPT taxa are sensitive to anthropogenic factor while Oligochaeta taxa are tolerant to anthropogenic factor in freshwater ecosystems [37]. In this study, the highest EPT-Taxa values have been obtained in station 1 (75%) during all seasons while the lowest EPT-Taxa values have been obtained in station 2 (57.1%) in autumn (Table 4).

Table 4.   Distribution of seasonal EPT-Taxa values (%)
Full size table

Indicator organisms that define the condition of the environment are used in biological monitoring studies. Bacteria, protozoans, algae, benthic macroinvertebrates, macrophytes and fish can be used as indicator organisms [38]. Benthic macroinvertebrates are the most advantageous group among all these groups. Habitat demands of benthic macroinvertebrates are more limited than other groups. Their mobility is lower. They are not displaced when they are affected by adverse conditions and they can be tracked more easily in the event of changes in community components and the loss of susceptible species. It is easier to diagnose, gather and hide than some other groups. It is found at any time of the year and its life cycles take longer than some other groups [39–45].

Cluster analysis dendrogram (UPGMA method) shows the similarities of the sampling stations (Fig. 3). According to the Bray–Curtis similarity index, the stations 1 and 2 (94%) have been the most similar to each other. The second most similar stations to each other have been determined to be stations 3 and 4 (93%). All of these close similarities and dissimilarities are likely related to the ecological and physical characteristics of the sampling stations. If we compare the bottom structure and physical characteristic of these stations, the matching sampling points are fairly similar to each other while stations 5 and 6 are different from the others. The bottom structure of stations 5 and 6 has a sandy-muddy sediment type because of their low incline and flow. These stations are close to the Tahtali Dam.

Fig. 3.
figure 3

The similarities between sampling stations.

Full size image

The biotic indices indicate that the Sangı Stream is slightly polluted by agricultural activities. Table 5 summarizes the biological quality scores and quality classes in Sangı Stream. Water quality classes have varied from moderate (according to BMWP 5th and 6th) to high class. According to the BMWP (original), the highest score values have been obtained in stations 1 and 2 while the lowest scores have been obtained in stations 5 and 6. According to ASPT (original), the highest score values have been obtained in station 1 while the lowest scores have been obtained in station 6. According to the SI, the water quality class is oligosaprobic/betamesosaprobic at sites 1, 2 and 3 (Class I–II) while the water quality class is betamesosaprobic at sites 4, 5 and 6.

Table 5.   Average score values and water quality classes of all indices
Full size table

The average species diversity of the stations has been determined by using SDI, SWDI and MDI in Sangı Stream. According to SDI, the highest diversity values have been obtained in stations 1 and 2 while the lowest scores have been obtained in stations 5 and 6. According to SWDI, the highest diversity values have been obtained in station 2 while the lowest scores have been obtained in station 4. According to MDI, the highest diversity values have been obtained in station 2 while the lowest scores have been obtained in station 6. Species richness is a simple number of species, while species evenness determines how equal the abundance of species is. [46]. The low evenness value indicates that there is no balance in distribution of the species and the community is dominated by a taxon or some taxa [47]. According to [48], if the SWDI value is higher than “3”, it indicates unpolluted water; if the diversity value is between 1 and 3, it indicates moderate pollution, and a value smaller than “1” indicates heavy pollution. According to [49], the diversity value ranges from 0 (low density) to 1 at SDI.

Comparative analyses of biotic indices in stream systems in Turkey have been made on a regional basis in recent years [8, 14, 17, 18, 50–55].

In this study, the random sample cases (10% select case) have been made on the basis of biotic indices and physicochemical parameters to verify data sets and to determine that the data has been transferred without errors in the SPSS version 20.0. Table 6 indicates the correlations of biotic and diversity indices.

Table 6.   Pearson’s based correlation assesment between biotic and diversity indices in Sangı Stream
Full size table

As a result of the correlation analysis, SI has a significant positive correlation with FBI (r = 0.942, p < 0.01). BMWP (original) has a significant positive correlation with ASPT (original) (r = 0.848, p < 0.01), while BMWP (original) has a significant negative correlation with FBI (r = –0.795, p < 0.01). On the other hand, BMWP (Original) has a significant negative correlation with SDI (r = –0.841, p < 0.01). Note that the increase in index values of SI, FBI, BMWP and ASPT shows good ecological quality, while the increase in BBI shows bad ecological quality.

Table 7 summarizes the Pearson correlation analyses between the biotic and diversity indices and the physicochemical parameters. A significant correlation has been found between T°C and SI (r = –0.751, p < 0.05), between pH and SI (r = 0.826, p < 0.05), and between DO and SI (r = –0.751, p < 0.05). According to correlation between physicochemical parameters and indices, a significant correlation value has been determined between NO2-N, BOI5, T°C, pH and DO and BMWP (original) and SWDI. All physicochemical parameters except T°C and pH have shown a significantly correlation with BBI. According to the analysis results, the benthic macroinvertebrates are sensitive to changes in temperature and oxygen in water, and as temperature increases and oxygen decreases, sensitive organisms are replaced by tolerant organisms [56].

Table 7.   Pearson’s based correlation assesment between biotic indices and species diversity indices and physicochemical parameters
Full size table

CONCLUSIONS

According to the results, it can be predicted that the pollution factors from intense agricultural activities and urbanization may cause serious pollution pressure on Sangı Stream in the future. This study shows that the BMWP, ASPT, FBI and SI appear to be more appropriate than the BBI. The BBI implemented in various streams studies in Turkey has shown reliable results reflecting the quality of the streams. However, BBI has not shown a significant correlation with other indexes used in Sangı Stream. The fact that there are very few similar studies makes it difficult to determine their availability in Turkey. More studies are needed on this subject, and they should be carried out in streams in different regions.

Intermittent monitoring studies should be carried out in the Tahtalı Dam basin to prevent the negative effect from waste. This study shows that there is a great need for development of a reference biotic index for Turkey to assess surface waters in Turkey with high accuracy. The development of a Turkish biotic index will also enable it to be used in Mediterranean and Aegean countries with similar climatic and geographical characteristics. This index will be an important criterion for determining the water quality of freshwater ecosystems on a global scale. Therefore, this study is very important in terms of data generation.