As a result of the samplings in the spring, summer and autumn periods, a total of 965 individuals were examined and 17 species belonging to seven families (Baetidae, Heptageniidae, Leptophlebiidae, Ephemeridae, Ephemerellidae, Potamanthidae and Caenidae) were identified (Table 2). Due to the deformation in the individuals belonging to the Potamanthidae family, the diagnoses were left at the Gen sp. level, while the Ecdyonurus sp. taxon belonging to the Heptageniidae family could be identified up to the genus level due to the lack of character.
Among the individuals collected, the highest number of species was found in the Heptageniidae family with nine species, while the highest number of individuals were found in the B. rhodani taxon, which is a member of the Baetidae family. When the dominance (%) of the species are examined according to the seasons; while B. rhodani was the most dominant species in spring (72.55%), summer (44.17%) and autumn (61.49%) periods, followed by R. semicolorata (11.52%) in spring, E. alpicola (17.67%) in summer, E. affinis (22.36%) in autumn. When the frequency values (%) of the species are examined, in the same seasonal order; while B. rhodani with 85%, 65% and 55% ratios was the most frequently encountered taxa in the stations, followed by R. semicolorata (30%) in spring, C. macrura (35%) in summer. C. macrura and E. lateralis were detected the same rates in autumn (%20). The continuous presence of B. rhodani in all three periods indicates that the ecological tolerance of the species is high (Table 3).
Table 3
Seasonal abundance (N/m2), % dominance (D) and % frequency (F) values of Ephemeroptera specimens detected in the Eastern Mediterranean Basin
|
Spring
|
Summer
|
Autumn
|
Species
|
N/m2
|
%D
|
%F
|
N/m2
|
%D
|
%F
|
N/m2
|
%D
|
%F
|
Baetis rhodani
|
378
|
72.55
|
85.00
|
125
|
44.17
|
65.00
|
99
|
61.49
|
55.00
|
Cloeon dipterum
|
|
|
|
3
|
1.06
|
5.00
|
|
|
|
Caenis luctuosa
|
|
|
|
1
|
0.35
|
5.00
|
|
|
|
Caenis macrura
|
|
|
|
15
|
5.30
|
35.00
|
9
|
5.59
|
20.00
|
Ecdyonurus macani
|
1
|
0.19
|
5.00
|
|
|
|
|
|
|
Ecdyonurus sp.
|
|
|
|
1
|
0.35
|
5.00
|
2
|
1.24
|
5.00
|
Electrogena affinis
|
|
|
|
26
|
9.19
|
20.00
|
36
|
22.36
|
15.00
|
Electrogena lateralis
|
3
|
0.58
|
10.00
|
21
|
7.42
|
25.00
|
12
|
7.45
|
20.00
|
Epeorus alpestris
|
10
|
1.92
|
5.00
|
|
|
|
|
|
|
Epeorus alpicola
|
34
|
6.53
|
25.00
|
50
|
17.67
|
25.00
|
3
|
1.86
|
5.00
|
Epeorus assimilis
|
6
|
1.15
|
15.00
|
15
|
5.30
|
15.00
|
|
|
|
Heptagenia perflava
|
6
|
1.15
|
15.00
|
6
|
2.12
|
5.00
|
|
|
|
Rhithrogena semicolorata
|
60
|
11.52
|
30.00
|
2
|
0.71
|
5.00
|
|
|
|
Ephemera vulgata
|
1
|
0.19
|
5.00
|
|
|
|
|
|
|
Serratella ignita
|
17
|
3.26
|
20.00
|
18
|
6.36
|
15.00
|
|
|
|
Paraleptophlebia submarginata
|
3
|
0.58
|
15.00
|
|
|
|
|
|
|
Potamantidae Gen. sp.
|
2
|
0.38
|
5.00
|
|
|
|
|
|
|
Total
|
521
|
100
|
-
|
283
|
100
|
-
|
161
|
100
|
-
|
Shannon-Wiener diversity (H) and Shannon-Evenness density (E) indices, which are applied respectively to calculate species diversity and population-related densities at stations, are frequently preferred in ecological studies as they give more objective results without distinguishing rare and dominant species (Özkan et al., 2020). The Shannon-Wiener index value (H) increases as the number of taxa increases and the distribution widens in a community. If this value is greater than 3, it represents the high quality water class, between 1–3 represents the moderately polluted water class, and less than 1 represents the polluted water class (Wilhm & Dorris, 1968). The Shannon Evennes (E) index, which is defined as the relative density of species in a community, takes a value between 0 and 1, and when this value approaches 1, it indicates that the diversity in the community is high (Magurran, 1988). Based on this situation, according to the results of the diversity and density values calculated at the stations (Table 4); the highest diversity was observed at the 6th (1.55) and 14th (1.33) stations, and the lowest at the 9th (0.15) and 13th (0.26) stations in the spring period. The E value, which expresses balance-equality, was calculated at the 19th (0.92) and 15th (0.84) stations the highest, and the lowest at the 12th (0.52) and 10th (0.54) stations. Although the species richness is the same, the differences in the diversity (H) values of the stations that differ in terms of abundance values or the disproportions in the H and E values of the stations with high species richness vary according to the distribution characteristics of the taxa at the stations. For example; if the 12th and 15th stations with the same species richness are examined, it can be said that the 15th station has a higher H value and therefore exhibits a more balanced distribution. As a matter of fact, it is seen that the E value, which expresses balance-equality, is calculated higher at the 15th station. On the other hand, it was observed that the E value was lower than the stations with less diversity, against the high H value at the 6th station, where the diversity was the most. This is explained by the more heterogeneous distribution of individuals in that station. This situation is also valid for the index values calculated for other seasons. In the summer period; the highest diversity value (H) was calculated at the 15th (1.46) and 6th (1.41) stations, and the lowest at the 9th (0.29) and 20th (0.31) stations. The most balanced distribution (E) was calculated at the 12th and 13th stations, which have the same value (0.83), and the lowest E values were calculated at the 3rd and 5th stations, which also had the same value (0.66). In the autumn period; the highest diversity (H) was calculated at the 12th (1.01) and 19th (0.98) stations, and the lowest at the 3rd (0.22) and 4th (0.25) stations. The most balanced distribution was observed with the same E value (1.00) at the 15th and 20th stations, and the lowest at the 3rd (0.62) and 4th (0.64) stations. Despite the low diversity at station 20, the higher E value compared to other stations is due to the fact that only 2 species are observed as a single individual in this station. Since they have only one species in all three seasons, no significant results could be obtained at stations with species richness of 1, and H and E values were calculated as 0 and 1, respectively. In addition, Ephemeroptera specimens could not be detected at the 3rd, 18th and 20th stations in the spring, at the 1st, 8th and 11th stations in the summer, and at the 7th and 18th stations in the autumn and sampling could not be carried out due to the drying up of the 1st, 2nd, 8th, 11th, 13th and 14th stations of the autumn period. The fact that the species diversity is noticeably less in the samplings of the autumn period compared to the other periods is explained by the fact that the waters dry out and the sampling cannot be carried out, and the increase in the species eliminated from the environment due to the changing water data.
Table 4
Calculated index values of species diversity and density at stations according to seasons (S: number of species, N: individual numbers)
Stations
|
Seasons
|
Mean/Standard deviation
|
Spring
|
Summer
|
Autumn
|
S
|
N
|
H
|
E
|
S
|
N
|
H
|
E
|
S
|
N
|
H
|
E
|
H
|
E
|
1
|
1
|
2
|
0
|
1
|
-
|
-
|
-
|
-
|
-
|
-
|
-
|
-
|
0.00 ± 0.00
|
0.33 ± 0.58
|
2
|
1
|
55
|
0
|
1
|
1
|
6
|
0
|
1
|
-
|
-
|
-
|
-
|
0.00 ± 0.00
|
0.67 ± 0.58
|
3
|
-
|
-
|
-
|
-
|
3
|
13
|
0.69
|
0.66
|
2
|
18
|
0.22
|
0.62
|
0.30 ± 0.35
|
0.43 ± 0.37
|
4
|
3
|
26
|
0.74
|
0.7
|
3
|
21
|
0.78
|
0.73
|
2
|
30
|
0.25
|
0.64
|
0.59 ± 0.30
|
0.69 ± 0.05
|
5
|
2
|
7
|
0.41
|
0.75
|
3
|
9
|
0.68
|
0.66
|
1
|
2
|
0
|
1
|
0.36 ± 0.34
|
0.80 ± 0.18
|
6
|
7
|
45
|
1.55
|
0.67
|
5
|
26
|
1.41
|
0.82
|
2
|
11
|
0.59
|
0.9
|
1.18 ± 0.52
|
0.80 ± 0.12
|
7
|
1
|
9
|
0
|
1
|
1
|
3
|
0
|
1
|
-
|
-
|
-
|
-
|
0.00 ± 0.00
|
0.67 ± 0.58
|
8
|
1
|
3
|
0
|
1
|
-
|
-
|
-
|
-
|
-
|
-
|
-
|
-
|
0.00 ± 0.00
|
0.33 ± 0.58
|
9
|
2
|
56
|
0.15
|
0.58
|
2
|
12
|
0.29
|
0.67
|
1
|
2
|
0
|
1
|
0.15 ± 0.15
|
0.75 ± 0.22
|
10
|
3
|
88
|
0.48
|
0.54
|
3
|
24
|
0.84
|
0.77
|
3
|
46
|
0.74
|
0.7
|
0.69 ± 0.19
|
0.67 ± 0.12
|
11
|
2
|
17
|
0.36
|
0.72
|
-
|
-
|
-
|
-
|
-
|
-
|
-
|
-
|
0.12 ± 0.21
|
0.24 ± 0.42
|
12
|
4
|
19
|
0.73
|
0.52
|
2
|
5
|
0.5
|
0.83
|
3
|
6
|
1.01
|
0.92
|
0.75 ± 0.26
|
0.76 ± 0.21
|
13
|
2
|
28
|
0.26
|
0.65
|
2
|
5
|
0.5
|
0.83
|
-
|
-
|
-
|
-
|
0.25 ± 0.25
|
0.49 ± 0.44
|
14
|
5
|
52
|
1.33
|
0.75
|
3
|
54
|
0.89
|
0.81
|
-
|
-
|
-
|
-
|
0.74 ± 0.68
|
0.52 ± 0.45
|
15
|
4
|
16
|
1.21
|
0.84
|
6
|
26
|
1.46
|
0.71
|
2
|
11
|
0.69
|
1
|
1.12 ± 0.39
|
0.85 ± 0.15
|
16
|
3
|
14
|
0.51
|
0.56
|
4
|
32
|
1.01
|
0.68
|
1
|
7
|
0
|
1
|
0.51 ± 0.51
|
0.75 ± 0.23
|
17
|
3
|
71
|
0.53
|
0.57
|
2
|
7
|
0.41
|
0.75
|
2
|
8
|
0.38
|
0.73
|
0.44 ± 0.08
|
0.68 ± 0.10
|
18
|
-
|
-
|
-
|
-
|
1
|
3
|
0
|
1
|
-
|
-
|
-
|
-
|
0.00 ± 0.00
|
0.33 ± 0.58
|
19
|
3
|
13
|
1.01
|
0.92
|
2
|
26
|
0.36
|
0.72
|
3
|
18
|
0.98
|
0.89
|
0.78 ± 0.37
|
0.84 ± 0.11
|
20
|
-
|
-
|
-
|
-
|
2
|
11
|
0.31
|
0.68
|
2
|
2
|
0.69
|
1
|
0.33 ± 0.35
|
0.56 ± 0.51
|
The calculated Shannon-Wiener diversity index ranged between 0.00 ± 0.00-1.18 ± 0.52 on average (Table 4). While the average H value was between 1–3 at the 6th and 15th stations, the diversity values were calculated below 1 at the other stations. Although H values below 1 represent the polluted water class, this is due to the low number of individuals in these stations and the low homogeneity of distribution (E) of the individuals. Therefore, the evaluation of the water quality of the stations according to the Shannon-Wiener diversity index calculated on the Ephemeroptera community should be considered together with physicochemical measurements and biotic index values. As a matter of fact, it was determined that the 11th station according to the final average of the measured physicochemical variables, and all the other stations except the 7th and 8th stations according to the final quality classes of the calculated biotic index values were in the high, good and moderately polluted water classes (Tables 6 and 10).
Among the 20 stations, the highest similarities were calculated between stations 1st, 2nd and 8th and stations 10th and 19th (100%). It was followed by 80% similarity between stations 3rd and 9th, 6th and 15th, and 7th and 20th. The lowest similarities were determined as 18% between the 6th station and the 13th and 20th stations. On the other hand, the 18th station differs from all stations (Fig. 2 and Table 5). This is explained by the fact that the C. dipterum taxon was not encountered at any other station other than this station, and that no other species other than this one was detected at this station.
Table 5
Similarity rates among stations (Bray-Curtis)
|
1
|
2
|
3
|
4
|
5
|
6
|
7
|
8
|
9
|
10
|
11
|
12
|
13
|
14
|
15
|
16
|
17
|
18
|
19
|
20
|
1
|
1
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
2
|
1.00
|
1
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
3
|
0.50
|
0.50
|
1
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
4
|
0.40
|
0.40
|
0.29
|
1
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
5
|
0.40
|
0.40
|
0.57
|
0.75
|
1
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
6
|
0.22
|
0.22
|
0.36
|
0.50
|
0.33
|
1
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
7
|
0.67
|
0.67
|
0.40
|
0.33
|
0.33
|
0.20
|
1
|
|
|
|
|
|
|
|
|
|
|
|
|
|
8
|
1.00
|
1.00
|
0.50
|
0.40
|
0.40
|
0.22
|
0.67
|
1
|
|
|
|
|
|
|
|
|
|
|
|
|
9
|
0.67
|
0.67
|
0.80
|
0.33
|
0.33
|
0.40
|
0.50
|
0.67
|
1
|
|
|
|
|
|
|
|
|
|
|
|
10
|
0.33
|
0.33
|
0.50
|
0.44
|
0.44
|
0.46
|
0.57
|
0.33
|
0.29
|
1
|
|
|
|
|
|
|
|
|
|
|
11
|
0.67
|
0.67
|
0.40
|
0.33
|
0.33
|
0.20
|
0.50
|
0.67
|
0.50
|
0.29
|
1
|
|
|
|
|
|
|
|
|
|
12
|
0.33
|
0.33
|
0.75
|
0.22
|
0.44
|
0.46
|
0.29
|
0.33
|
0.57
|
0.40
|
0.29
|
1
|
|
|
|
|
|
|
|
|
13
|
0.50
|
0.50
|
0.33
|
0.29
|
0.29
|
0.18
|
0.40
|
0.50
|
0.40
|
0.25
|
0.40
|
0.50
|
1
|
|
|
|
|
|
|
|
14
|
0.33
|
0.33
|
0.25
|
0.67
|
0.44
|
0.77
|
0.29
|
0.33
|
0.29
|
0.60
|
0.29
|
0.40
|
0.25
|
1
|
|
|
|
|
|
|
15
|
0.25
|
0.25
|
0.60
|
0.55
|
0.55
|
0.80
|
0.22
|
0.25
|
0.44
|
0.67
|
0.22
|
0.50
|
0.20
|
0.67
|
1
|
|
|
|
|
|
16
|
0.33
|
0.33
|
0.50
|
0.22
|
0.44
|
0.31
|
0.57
|
0.33
|
0.29
|
0.60
|
0.29
|
0.40
|
0.25
|
0.20
|
0.50
|
1
|
|
|
|
|
17
|
0.33
|
0.33
|
0.50
|
0.44
|
0.44
|
0.46
|
0.29
|
0.33
|
0.57
|
0.20
|
0.29
|
0.60
|
0.50
|
0.40
|
0.50
|
0.20
|
1
|
|
|
|
18
|
0.00
|
0.00
|
0.00
|
0.00
|
0.00
|
0.00
|
0.00
|
0.00
|
0.00
|
0.00
|
0.00
|
0.00
|
0.00
|
0.00
|
0.00
|
0.00
|
0.00
|
1
|
|
|
19
|
0.33
|
0.33
|
0.50
|
0.44
|
0.44
|
0.46
|
0.57
|
0.33
|
0.29
|
1.00
|
0.29
|
0.40
|
0.25
|
0.60
|
0.67
|
0.60
|
0.20
|
0.00
|
1
|
|
20
|
0.50
|
0.50
|
0.67
|
0.29
|
0.57
|
0.18
|
0.80
|
0.50
|
0.40
|
0.75
|
0.40
|
0.50
|
0.33
|
0.25
|
0.40
|
0.75
|
0.25
|
0.00
|
0.75
|
1
|
The physicochemical variables measured seasonally and the final quality classes related to the mean values of these variables are presented in Table 6 with their color codes (WQR, 2021). When the average water quality classes in the stations are evaluated according to the seasons over the physicochemical variables (WQR, 2021); while all stations have first class water quality in all three seasons according to pH and total nitrogen, the 8th station in terms of temperature, the 1st, 7th, 8th, 9th, 10th, 11th, 12th, 13th, 18th and 19th stations in terms of electrical conductivity, the 1st, 3rd, 8th, 13th and 19th stations in terms of dissolved oxygen were determined in second class water quality. On the other hand, only the 11th station was determined to be in polluted water quality in terms of dissolved oxygen, while the variables in all other stations were determined to be of high water quality. When the final quality classes of the stations for all variables are evaluated; the 11th station was observed in polluted water quality, the 1st, 3rd, 7th, 8th, 9th, 10th, 12th, 13th, 18th and 19th stations were observed in moderate polluted water quality, and all other stations in high water quality. (Table 6). The 11th station is a station that is rich in vegetation. The reason for the pollution here is that with the increase in temperature in the summer period, the river water decreases and the organic debris such as vegetation parts mixed with the water from the surrounding reduces the amount of dissolved oxygen. As a matter of fact, the temperature reached 24.70 Co in the summer and the dissolved oxygen value of 8.6 mg/l in the spring period decreased to 2.33 mg/l in this period.
The stations are generally distributed in untouched areas where there are reference conditions. Even if it is observed that agriculture and animal husbandry activities are carried out close to some stations, it can be said that this situation is not at a level that threatens the physicochemical properties of the stream. On the other hand, there are no settlements close to the rivers. This situation supports that the streams are also clean in terms of domestic waste pollutants. The changes in pH, electrical conductivity and total nitrogen amount in summer and autumn periods are due to the evaporation of the waters due to the increase in air temperature and, accordingly, the increase in organic matter in the environment and the decrease in dissolved oxygen. However, no change in the physicochemical variables of the waters that could cause serious problems was observed in all three seasons. Therefore, it can be said that the physicochemical variables measured for all stations are generally observed in reference ranges at a level that does not pose a major threat, and that there is no intense domestic, industrial and agricultural pollution pressure that will adversely affect the aquatic system.
According to the result of Pearson correlation analysis; a statistically significant positive correlation was observed between temperature, electrical conductivity and total nitrogen, while a significant negative correlation was observed between dissolved oxygen and temperature and total nitrogen (p < 0.01) (Table 7).
Table 7
Pearson correlation among physicochemical variables (EC: Electrical Conductivity, DO: Dissolved Oxygen, TN: Total Nitrogen)
|
Temperature (°C)
|
pH
|
EC(µS/cm)
|
DO (mg/L)
|
TN (mg/L)
|
Temperature (°C)
|
1
|
|
|
|
|
pH
|
-0.085
|
1
|
|
|
|
EC(µS/cm)
|
0.659*
|
0.087
|
1
|
|
|
DO (mg/L)
|
-0.705*
|
0.268
|
-0.250
|
1
|
|
TN (mg/L)
|
0.398*
|
-0.050
|
0.065
|
-0.432*
|
1
|
*: The correlation is significant at the p < 0.01 level. |
In the Canonical Correspondence Analysis (CCA), which was applied to evaluate the relationships between Ephemeroptera taxa and physicochemical variables, 17 taxa and five environmental variables belonging to a total of 46 samples were used for three seasons. The eigenvalue coefficients of the first two axes were calculated as 0.324 and 0.231, respectively, and Monte Carlo Permutation test was found to be significant for all axes (F = 3.09 and p = 0.002). While there was a positive correlation between total nitrogen, pH and temperature variables in the ordination graph obtained, a negative correlation was observed between these variables and dissolved oxygen (Fig. 3). As a matter of fact, the result of the applied Pearson correlation analysis also supports this relationship (Table 7). The most determining variables in the distribution of species were determined as total nitrogen, temperature and electrical conductivity.
Dissolved oxygen in aquatic systems is one of the most important variables in terms of providing information about habitat quality, pollution level and organic matter concentration. It is known that aquatic creatures living in environments where oxygen is needed, need dissolved oxygen in their energy-requiring metabolic activities and reproduction, and the amount of dissolved oxygen in the aquatic environment shows the natural cleaning capacity. It is known that while the biological oxygen demand increases with the pollution caused by various natural or anthropogenic effects in the waters, the amount of dissolved oxygen decreases. The amount of oxygen in natural waters is directly related to temperature, and there is a decrease in the amount of dissolved oxygen depending on the increase in water temperature (Lawa, 1980; Wetzel, 1983; Karpuzcu, 2007). The increase in pollution and temperature value in river systems causes a direct increase in the electrical conductivity value, which expresses the capacity to transmit electric current (Kalyoncu et al., 2005). While nitrogen compounds are found in very low amounts in clean and oxygenated waters, they are observed intensely in waters that are low in oxygen and contain pollution (Yıldırım, 2006). As a matter of fact, in the CCA graph obtained, it is seen that the dissolved oxygen is positioned in the opposite direction with the other variables in parallel with the temperature increase.
It is known that very clean and undisturbed environments of the family Heptageniidae contain a lot of indicator species (Bauernfeind et al., 2002). For this reason, this family represents an important group as a bioindicator of clean water in studies to determine water quality (Kazancı et al., 2014). In our study, this family was determined as the family with the highest rate in terms of species diversity. It has been reported that this family is observed in high diversity in some studies conducted in Turkey (Türkmen, 2013; Salur et al., 2016). It has been reported that Ecdiyonurus larvae belonging to this family spread on stony and rocky soils in fast-flowing parts of rivers and have low tolerance to organic pollution (Bauernfeind & Soldan, 2012). On the other hand, it has been reported that R. semicolorata is eurytherm in terms of temperature and is distributed in high or medium flowing regions in rivers and it has been recorded in many places in studies conducted in Turkey. (Kazancı, 2001a; Türkmen, 2013; Küçüker, 2019). As a matter of fact, in the CCA graph obtained, it is seen that the taxa clustered in the same direction as the dissolved oxygen variable expressing clean conditions, and in the opposite direction with the temperature and total nitrogen variables expressing eutrophic conditions, consist of individuals belonging to the families Heptageniidae (E. assimilis, H. perflava, E. macani, E. alpestris, Ecdyonurus sp., R. semicolorata) and Leptophlebiidae (P. submarginata). Although R. semicolorata is known to be more tolerant than other Heptageniidae members, it was determined as an indicator species of clean conditions in our study. On the other hand, it has been reported that individuals belonging to the Electrogena genus are distributed in the oligosaprobic and beta-mesosaprobic regions of the rivers (Bauernfeind, 1995). The fact that the species belonging to this genus (E. affinis, E. lateralis) are positioned in the same direction as temperature, pH and total nitrogen variables in our study supports their high tolerance to pollution. It has been reported that the species belonging to the Leptophlebiidae family are generally distributed in the stony areas of the hypocrenon and rhithron regions of the rivers (Buffagni et al., 2009). It has been stated that Paraleptophlebia genus, which is a member of this family, is generally distributed in oligosaprobic regions, and it can be found in betamezosaprobic, xenosaprobic and even alpha-mesosaprobic regions, although less likely (Kazancı, 2001b).
It has been reported that individuals belonging to the Caenidae family also show distribution in alpha-mesosaprobic and beta-mesosaprobic zones (Mouthon, 1996). Caenis members of this family are known for their high tolerance to organic pollution and their adaptation to various stream types within the order Ephemeroptera. Some species can even live in brackish water environments and are generally found in sandy, loamy or gravelly soils, but they can also be distributed in stagnant waters where the current is slow (Timm, 1997; Menetrey et al., 2008). In the CCA graph obtained, it is seen that the C. macrura and C. luctuosa taxa determined belonging to this family cluster in the same position with the temperature, pH and total nitrogen variables.
It has been reported that the C. dipterum species is eurytherm in terms of temperature and is distributed in the stagnant parts of rivers (Buffagni et al., 2009). In the CCA graph obtained, it was determined that this species is located close to the temperature variable, while it is located in the opposite regions with the dissolved oxygen variable. B. rhodani, which is represented by the highest number of individuals at the stations for the spring, summer and autumn periods and has the highest incidence, is a species known for its high tolerance to organic pollution within the order Ephemeroptera. It has been reported that this species is eurytherm and prefers regions with high or medium currents (Sladeck, 1973; Buffagni et al., 2009). At the same time, it is stated that there are records in many regions from west to east of Turkey (Küçüker, 2019). As a matter of fact, in the CCA graph obtained, it was seen that this species was positioned very close to the center and clustered in a positive relationship with the determining environmental variables such as temperature, total nitrogen and electrical conductivity.
Another family within the Ephemeroptera order, known for its high tolerance to organic pollution, is known as Potamantidae (Allan, 1995). It has been reported that the larvae of the genus Potamanthus are found in the potamon regions of the rivers, in the medium and slow flowing parts (Maiorana, 1979). In our study, this family was found only in the 11th station belonging to the spring period. Due to the deformation occurring in the detected individuals, the diagnosis procedure is Potamantidae Gen. sp. level and the individuals are thought to belong to the genus Potamanthus. In the CCA graph obtained, it was determined that this species also clustered in a positive relationship with the determining environmental variables such as temperature, total nitrogen and electrical conductivity.
Due to the low abundance values of the other detected species, they were located independently of the physicochemical variables measured or no significant results could be obtained. It has been determined that the results obtained are in parallel with the literature information.
According to the ANOSIM results (as a result of the Doornik-Hansen test, the variances between species-season groups are not normally distributed (p = 0.00 < 0.05)); while there was a significant difference between the spring period among with summer (p = 0.00) and autumn (p = 0.02) periods, there was no significant difference between the summer and autumn periods (p = 0.55). According to the result of SIMPER analysis; the seasonal groups with the highest difference in terms of species distribution were determined as spring-summer (71.52%), spring-autumn (67.19%) and summer-autumn (62.65%), respectively. The taxon that contributed the most to the distinctions in all seasonal groups was determined as B. rhodani with the values of 47.35, 50.77 and 38.49, respectively (Table 8). The distinctions among seasons according to the distribution of species are presented in Fig. 4.
Table 8
Results of SIMPER and ANOSIM
Groups
|
SIMPER
|
One-Way ANOSIM
|
Average Dissimilarity (%)
|
Discriminating species
|
Contribution (%)
|
p value
|
Spring-Summer
|
71.52
|
Baetis rhodani
|
47.35
|
0.00*
|
Rhithrogena semicolorata
|
10.43
|
Epeorus alpicola
|
10.38
|
Electrogena affinis
|
5.76
|
Electrogena lateralis
|
5.25
|
Spring-Autumn
|
67.19
|
Baetis rhodani
|
50.77
|
0.00*
|
Rhithrogena semicolorata
|
11.45
|
Electrogena affinis
|
6.78
|
Epeorus alpicola
|
6.20
|
Electrogena lateralis
|
5.17
|
Summer-Autumn
|
62.65
|
Baetis rhodani
|
38.49
|
0.55
|
Electrogena affinis
|
15.36
|
Electrogena lateralis
|
11.26
|
Epeorus alpicola
|
10.19
|
Caenis macrura
|
5.50
|
*: p < 0.05 significance level. |
In the MANOVA analysis, in which the differences between seasons were evaluated in terms of physicochemical variables (Doornik-Hansen test showed that the variances were normally distributed (p = 7.07 > 0.05)); The covariance equality between the season groups was evaluated using the Box's test and it was seen that the variance-covariance equality could not be achieved (p = 0.04 < 0.05). For this reason, an evaluation was made on Pillai's trace statistics (Akbulut, 2010). According to the results of Pillai's trace statistics, it was determined that the seasons made a significant difference on the linear combination of environmental variables (Pillai Trace (λ) = 0.94, F(7.09), p = 0,00 < 0.05, η2 = 0.47). The variables causing this difference were determined as temperature, dissolved oxygen and total nitrogen. The electrical conductivity variable did not make a significant difference between the seasonal groups. The homogeneity of the variances was evaluated with the Levene statistical test in order to understand in which season combinations the variables causing the significant difference made a difference. Accordingly, while the homogeneity of the variances of the temperature, pH and dissolved oxygen variables was met (p > 0.05), it was determined that the variances of the total nitrogen variable were not homogeneously distributed. For this reason, evaluations were made on the Tukey HSD test (Kayri, 2009), which is one of the multiple comparison tests used in case of homogeneous distribution of variances, and the Games Howell test (Games, 1971), which is used in cases where variances are not homogeneously distributed. According to the test results; while there was no significant difference between spring and autumn seasons in terms of temperature variable (p > 0.05), it was found to be significant in other seasons combinations (p < 0.05). While there was a significant difference between summer and autumn seasons in terms of pH variable (p < 0.05), no significant difference was found in other seasons combinations (p > 0.05). Dissolved oxygen and total nitrogen variables differed significantly among all season groups ( p < 0.05) (Table 9).
Table 9
MANOVA analysis between seasonal groups and physicochemical parameters
|
MANOVA
|
Levene's Test (p)
|
Tukey HSD/Games Howell (p)
|
|
Seasons
|
Mean
|
Std. Deviation
|
Sum of Squares
|
df
|
Mean Square
|
F
|
Sig.
|
Partial Eta Squared
|
Spring-Summer
|
Spring-Autumn
|
Summer-Autumn
|
Temperature
|
Spring
|
15,46
|
4,21
|
381,14
|
2
|
190,57
|
13,08
|
0,00*
|
0,38
|
0,34
|
0,00*
|
0,48
|
0,01*
|
Summer
|
21,96
|
3,86
|
Autumn
|
17,13
|
3,08
|
Total
|
18,30
|
4,73
|
pH
|
Spring
|
8,02
|
0,18
|
0,75
|
2
|
0,37
|
5,86
|
0,01*
|
0,21
|
0,20
|
0,08
|
0,41
|
0,01*
|
Summer
|
7,82
|
0,30
|
Autumn
|
8,14
|
0,27
|
Total
|
7,98
|
0,28
|
Electrical Conductivity
|
Spring
|
369,18
|
140,43
|
1884,55
|
2
|
942,27
|
0,07
|
0,93
|
0,00
|
0,20
|
0,94
|
1,00
|
0,96
|
Summer
|
382,94
|
105,39
|
Autumn
|
370,50
|
89,02
|
Total
|
374,61
|
113,75
|
Dissolved Oxygen
|
Spring
|
9,03
|
0,53
|
22,14
|
2
|
11,07
|
30,43
|
0,00*
|
0,59
|
0,14
|
0,00*
|
0,09
|
0,00*
|
Summer
|
7,44
|
0,76
|
Autumn
|
8,54
|
0,42
|
Total
|
8,31
|
0,92
|
Total Nitrogen
|
Spring
|
0,27
|
0,26
|
6,99
|
2
|
3,50
|
21,20
|
0,00*
|
0,50
|
0,02*
|
0,00*
|
0,03*
|
0,01*
|
Summer
|
1,17
|
0,52
|
Autumn
|
0,64
|
0,39
|
Total
|
0,70
|
0,56
|
*: p < 0.05 significance level. |
In the index calculation applied to determine the ecological quality ratios of water quality, Amphipoda (10), Bivalvia (2), Chironomidae (35), Coleoptera (16), Diptera (13), Ephemeroptera (17), Gastropoda (4), Heteroptera (8) Hirudinea (1), Odonata (5), Oligochaeta (2), Plecoptera (6), Simuliidae (14), Trichoptera (18), Turbellaria (1) were used individuals belonging to the orders. Species numbers of the orders are indicated in parentheses. The index and class limit values of the calculated ecological quality ratios are given in Table 10 with color codes.
According to the calculated biotic index values, in the spring period; the 1st, 7th, 8th, 12th, 13th and 20th stations are moderate, 2nd, 3rd, 5th, 10th, 18th and 19th stations are good, all other stations are in high, in the summer period; the 9th, 13th, 18th and 20th stations are moderate, the 7th station is poor, the 8th and 11th stations are bad, all other stations are in high, and in the autumn period; the 3rd and 9th stations are moderate, the 4th, 6th, 10th and 16th stations are good, the 5th and 20th stations are poor, the 7th and 18th stations are bad, all other stations are in high water classes were detected. Considering the final values of the averages; the 1st, 9th, 11th, 13th, 18th and 20th stations are moderate, 7th and 8th stations are poor were detected water quality classes, while all other stations were determined to be in good or high water quality class. In addition, while individuals could not be detected at the 1st station of the summer period, sampling could not be made due to the drying up of the 1st, 2nd, 8th, 11th, 13th and 14th stations of the autumn period (Table 10).
As a result; while the 7th and 8th stations are observed to be polluted or under the influence (poor and bad water quality class), it can be easily said that all other stations are in reference feature (high, good and moderate water quality classes) away from the pollution pressure to a large extent. This is due to the fact that the determined stations are located away from anthropogenic pressure to a large extent, and therefore, domestic or agricultural wastes do not mix into the water. On the other hand, the fact that the 11th station is in the polluted water class in terms of physicochemical parameters and the index value calculated in the summer period with poor water quality shows parallelism. It is thought that the reason for this situation is the increase in temperatures and the decrease in the amount of oxygen in the organic debris mixed into the water, and the decrease in the number of Heptageniidae individuals distributed here. As a matter of fact, in stations with bad or poor water quality classes, taxa known for their high tolerance to pollution such as B. rhodani, C. macrura and E. lateralis were observed in high abundance while, taxa known as clean water indicator such as E. assimilis were observed in very low abundances. On the other hand, Ephemeroptera individuals were not found in most of these stations. This is an expected situation for the members of this order, which is generally known to contain high levels of indicator species in clean waters (Demir, 2005; Jandry et al., 2014). It has been observed that the most observed species among the stations with reference characteristics are composed of Heptageniidae members that prefer xenosaprobic and oligosaprobic river zones, which are generally very clean or slightly polluted (Kazancı, 2001a; Bauernfeind et al., 2002).