3.1 Descriptive analysis
The results of the descriptive statistics of the variables analyzed in the sources are listed in Table 1.
Table 1. Descriptive statistics of the analyzed variables for the four water sources.
|
Cristal (GWS)
|
Camaquã (SWS)
|
Capão do Leão (SWS) |
Rio Grande (SWS) |
Variables
|
Average
|
Sd
|
Cv (%)
|
Average
|
Sd
|
Cv (%)
|
Average
|
Sd
|
Cv (%)
|
Average
|
Sd
|
Cv (%)
|
T air (ºC)
|
22,42
|
5,84
|
26,04
|
23,08
|
7,02
|
30,43
|
14,00
|
3,67
|
26,2
|
21,58
|
6,20
|
28,73
|
T water (ºC)
|
21,16
|
2,88
|
13,63
|
20,33
|
5,14
|
25,28
|
14,63
|
2,46
|
16,81
|
21,16
|
6,40
|
30,27
|
Al total (mg L-1)
|
0,04
|
0,09
|
213,89
|
2,59
|
2,25
|
85,76
|
2,62
|
3,04
|
115,53
|
4,64
|
3,14
|
67,8
|
DAlum. (mg L-1)
|
0,00
|
0,00
|
0,00
|
0,36
|
0,08
|
50,42
|
0,17
|
0,5
|
69,60
|
0,52
|
0,77
|
56,32
|
TIron (mg L-1)
|
1,95
|
1,04
|
53,24
|
1,52
|
0,99
|
65,14
|
1,98
|
1,69
|
85,22
|
2,62
|
1,56
|
59,32
|
DIron. (mg L-1)
|
1,16
|
0,92
|
79,16
|
0,42
|
0,44
|
56,29
|
0,45
|
0,24
|
54,32
|
0,46
|
0,46
|
98,74
|
Mn total (mg L-1)
|
0,23
|
0,08
|
34,89
|
0,14
|
0,13
|
95,62
|
0,35
|
0,77
|
221,11
|
0,06
|
0,04
|
72,16
|
Alkalinity (mg L-1)
|
70,81
|
16,08
|
22,71
|
25,66
|
6,50
|
25,32
|
25,33
|
6,64
|
26,20
|
34,66
|
8,23
|
23,73
|
Hardness (mg L-1)
|
44,16
|
4,29
|
9,72
|
17,30
|
3,69
|
21,34
|
19,30
|
3,79
|
19,65
|
30,68
|
7,28
|
23,72
|
Ca total (mg L-1)
|
11,14
|
1,15
|
10,33
|
3,53
|
0,90
|
25,67
|
3,68
|
0,87
|
23,57
|
7,28
|
1,77
|
24,28
|
Mg total (mg L-1)
|
3,96
|
0,71
|
17,81
|
2,02
|
0,67
|
33,39
|
2,50
|
0,23
|
23,32
|
2,98
|
1,32
|
44,26
|
Apparent color (uH)
|
35,32
|
25,18
|
71,29
|
70,29
|
44,06
|
62,68
|
62,75
|
41,70
|
66,46
|
98,43
|
33,40
|
33,93
|
Turbidity (uT)
|
11,21
|
12,43
|
110,87
|
24,47
|
19,17
|
78,35
|
25,96
|
28,22
|
39,12
|
38,57
|
19,41
|
50,33
|
ST (mg L-1)
|
156,67
|
15,41
|
9,83
|
85,83
|
17,46
|
20,35
|
97,75
|
28,67
|
29,34
|
130,67
|
19,41
|
30,60
|
pH
|
7,09
|
0,43
|
35,74
|
7,35
|
0,23
|
3,12
|
7,16
|
0,29
|
4,06
|
7,39
|
0,40
|
5,36
|
OM (mg L-1)
|
1,15
|
0,41
|
35,74
|
4,25
|
0,96
|
22,71
|
5,25
|
2,42
|
46,03
|
8,58
|
2,61
|
30,40
|
PT (mg L-1)
|
0,18
|
0,13
|
73,97
|
0,06
|
0,08
|
31,05
|
0,15
|
0,07
|
45,03
|
0,08
|
0,04
|
48,76
|
Nitrate (mg L-1)
|
0,04
|
0,03
|
67,04
|
0,44
|
0,42
|
94,59
|
0,58
|
0,19
|
32,73
|
0,30
|
0,34
|
115,44
|
Chlorides (mg L-1)
|
15,44
|
0,87
|
5,66
|
3,97
|
0,46
|
11,68
|
7,6
|
1,02
|
13,37
|
10,96
|
3,08
|
28,08
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Legend: Sd: standard deviation; Cv: coefficient of variation in %; OM: organic matter.
The results of descriptive statistics of the air and water temperatures of the springs (Table 1) reflected the variations corresponding to the seasons in which the samples were taken. There was a high coefficient of variation of air temperature due to the common temperature range in the state, where the seasons are well defined. The water temperature showed a similar behavior, but with lower values than those recorded for the air, with the lowest amplitude for the Capão do Leão Spring. The lowest coefficients of variation of air and water Tcº were found for the groundwater source.
The results of the total Al monitoring reports showed results with large variation among the springs; Capão do Leão and Rio Grande had maximum concentrations above 10 mg L-1. The groundwater source showed no DAlum and for the others, the concentrations ranged from 0.06 mg L-1 (Camaquã Reservoir) to 2.54 mg L-1 (Rio Grande Spring).
The results for TIron ranged from 0.07 to 7.20 mg L-1 for the Rio Grande and Capão do Leão sources, respectively. As for DIron, the highest mean value was observed in the groundwater source (Table 1), related to the local geological formation. According to Hussin et al. (2016) the hydrogeochemical evolution of shallow groundwater is governed by the processes of precipitation, weathering, dissolution and ion exchange.
The values found for total Mn varied from zero, for Rio Grande, to 2.80 mg L-1, Capão do Leão, which also had the highest coefficient of variation. The maximum value found for total Mn coincides with the sampling in the rainy season, although the other variables have not changed in the same proportion, allowing to infer that there has been solubilization of manganese. The manganese cand be naturally found along with iron in both surface and ground water (Shah et al. in press), is na essential nutriente, however, both excess and deficiency can cause adverse impacts (CAWST, 2009).
Regarding turbidity and TS, the Camaquã Reservoir presented the lowest values, 1.60 uT and 57 mg L-1, respectively, indicating that the lotic environment favors the sedimentation of particles that constitute these variables.
The occurrence of color and turbidity is related to pH, which should be measured, since lower pH values are related to greater intensity of true color, which is an indicator of the concentration of organic matter. The pH range found in the studied sources varied between 6.6 and 8.2, obtained in the source that supplies Rio Grande. The lowest pH range was found in the reservoir spring, which supplies Camaquã. GWS showed the lowest pH values.
Evaluating the mean values of this parameter for the different sources (Table 1), similar results are observed, unlike the results reported by Chowdhury and Husain (2020), in which the surface source presented a better result than the groundwater source. The higher pH values could be attributed to bedrock geology ((Muhammad and Usman, 2022).
Importantly, pH determines alkalinity, which is the ability to react quantitatively with a strong acid. High alkalinity values are associated with organic matter decay (Omer, 2019). The descriptive analysis of the alkalinity results obtained a minimum value of 16 mg L-1 for Capão do Leão, and a maximum of 99 mg L-1 for the groundwater source (Cristal). Groundwater, due to its geochemical characteristics, has high alkalinity. High alkalinity registered for Cristal coincides with changes in turbidity, indicating the influence of rainfall, and characteristics of this source.
The lowest values of hardness and total Ca (12 and 2 mg L-1, respectively) were obtained in the source that supplies Camaquã and for total Mg (0.8 mg L-1), the source used by the municipality of Rio Grande.
The maximum value was found in the Cristal groundwater source for the three variables (53; 13.10 and 5.4 mg L-1, respectively for hardness, Ca+2 and Mg+2). From a sanitary point of view, they do not pose a risk due to their presence in water intended for human consumption. Souza (2015) investigated the water quality and found mean calcium values of 17 mg L-1 (rainy season) and 24 mg L-1 (dry season). Tay et al. (2015) reported the chemical constituents most abundant cation and anion as Na+ and HCO3– respectively, with the order of relative abundance are: Na+>Ca+2> Mg+2>K+ and HCO3->Cl->SO4 2- respectively. Li and Wu (2019) indicates that groundwater is the most important part of the water for drinking purpose, and health risk is closely related to the drinking water quality that is determined by many indices such as F–, NO3–, Mg2+ and TDS, etc.
Another variable studied refers to the organic matter in natural waters, composed of humic substances (humic and fulvic acids of plant origin) and non-humic substances (proteins, carbohydrates, algae, amino acids, carboxylic acids, and hydrocarbons). In water sources less impacted by human activities or where there are no algal blooms, humic substances prevail in the composition of organic matter, with fulvic acids present in organic matter less susceptible to coagulation during water treatment (Libânio, 2010).
The lowest concentration of organic matter was found in the underground source of Cristal, with 0.80 mg L-1, while the maximum, 12 mg L-1, in Rio Grande. In periods of higher rainfall, in flooded sections, the water overflows and causes an increase in organic matter and color, due to plant decomposition, and a consequent drop in pH. There is no reference value in Brazilian legislation for this variable directly, although the determination of oxygen consumed in an acidic medium is relevant in WTPs, as its presence, in addition to having the potential to form by-products such as trihalomethanes, can impart odor and taste to the water, interfere with the removal of iron and manganese and contribute to the development of biofilms in the distribution network (Libânio, 2010).
Total phosphorus (TP) was monitored and showed similar minimum values, the lowest for the sources of Camaquã and Rio Grande (0.02 mg L-1), while the maximum value occurred in the groundwater source of Cristal (0.58 mg L-1). Although TP does not have a potability standard, its measurement is directly related to parameters of water intended for human consumption.
According to Libânio (2010), in groundwater percolation and storage in soil interstices in areas with less interference by human action, phosphate levels may be higher than in surface water, corroborating our findings.
N2, as nitrate, presented values between zero, in the underground source and in the reservoir, to 1.48 mg L-1, also for the Camaquã Reservoir. These indicate the absence of significant contamination, as well as the results found in this study for this variable.
The results of the descriptive analysis of chlorides showed a minimum value, 3.24 mg L-1, for the reservoir that supplies Camaquã, and a maximum of 16.90 mg L-1 for the groundwater source. Wang et al. (2023) reported the Cl− concentration varied from 0.00 to 33.42 mg L-1, with an average of 1.40 mg L-1, and was relatively lower than the standard for drinking purposes. The association of chloride with TP indicated the release of domestic effluents.
3.2 Multivariate analysis
The result obtained by applying the Kruskal-Wallis test did not show statistically significant variation between years for any of the sources. Between the seasons, for the Cristal groundwater source, the water and air temperature variables showed a significant variation. For the source of Capão do Leão, there was no significant variation. The influence of precipitation and temperature on the spatial distribution of water yield in the Yangtze River Basin was significantly higher than that of other factors (Yang et al. 2023).
As for the reservoir source that supplies Camaquã, only water and air temperature, manganese and calcium showed significant variations. As to the source that supplies water to the municipality of Rio Grande, manganese, water and air temperature showed a significant variation. In He et al. (2023) the results showed significant spatial and temporal variation in the dissolved trace elements, with the highest concentrations found in the low reaches and during the wet season.
In Miyittah et al. (2020), the paired sample t-test confirmed that the surface water quality of the Aby Lagoon System varied significantly between the wet and dry season (p < 0.05), except for phosphate loads, which may have been largely influenced by year-round municipal waste discharges, unlike the results found in this study, for the source that is a reservoir (Camaquã).
Alves et al. (2018) reported through the PCA, it was possible to identify that the most important parameters in contribution to water quality variations are total coliforms in summer-autum, water level, water temperature, and electrical conductivity in winter and color and turbidity in winter. But, the authors also reported that the PA ccording to the results found, the seasonal influence practically did not occur for the different types of sources in this study, which may be related to the period in which the samplings were carried out.
Spearman’s correlation was applied considering that variables did not follow a normal distribution (Tables 2 to 5).
Table 2. Correlation between the variables of the GWS of Cristal, RS.
Parameters
|
Season
|
Year
|
T. air
|
T. water
|
Al Total
|
TIron
|
Mn Total
|
Ca Total
|
Color
|
Hardness
|
MO
|
Turbidity
|
DIron
|
Chlorides
|
Season
|
1,000
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Year
|
-
|
1,000
|
|
|
|
|
|
|
|
|
|
|
|
|
T. air
|
-0,828
|
-
|
1,000
|
|
|
|
|
|
|
|
|
|
|
|
T. water
|
-0,852
|
-
|
0,872
|
1,000
|
|
|
|
|
|
|
|
|
|
|
Al Total
|
-
|
-
|
-
|
-
|
1,000
|
|
|
|
|
|
|
|
|
|
TIron
|
-
|
-
|
-
|
-
|
0,561
|
1,000
|
|
|
|
|
|
|
|
|
Mn Total
|
-
|
-
|
-
|
-
|
0,506
|
0,870
|
1,000
|
|
|
|
|
|
|
|
Ca Total
|
-
|
-
|
-
|
-
|
-
|
0,553
|
0,588
|
1,000
|
|
|
|
|
|
|
Color
|
-
|
0,591
|
-
|
-
|
-
|
-
|
0,536
|
0,561
|
1,000
|
|
|
|
|
|
Hardness
|
-
|
-
|
-
|
-
|
-
|
-
|
0,587
|
0,729
|
0,692
|
1,000
|
|
|
|
|
MO
|
-
|
-
|
-
|
-
|
0,509
|
0,667
|
0,512
|
-
|
-
|
-
|
1,000
|
|
|
|
Turbidity
|
-0,531
|
-
|
0,501
|
0,526
|
-
|
-
|
0,518
|
-
|
0,760
|
-
|
-
|
1,000
|
|
|
DIron
|
-
|
-
|
-
|
-
|
-
|
-
|
-
|
0,640
|
0,542
|
-
|
-
|
-
|
1,000
|
|
Chlorides
|
-0,520
|
-
|
0,638
|
0,768
|
0,548
|
-
|
-
|
-0,595
|
-
|
-
|
-
|
-
|
-0,583
|
1,000
|
For GWS (Table 2), only TP was not correlated with any other variable. It was possible to observe an increase only in color over the years. Air and water temperature had highly positive correlations with each other. Regarding the season, there was a negative correlation with air and water temperatures, turbidity, and chlorides, allowing to infer that the values decreased in samplings carried out in cold seasons. These variables were positively correlated with water temperature, indicating a greater availability of these ions due to the increase in temperature.
Color and turbidity were correlated with total Mn, indicating that this metal contributed to the high values found, which commonly occurs by the presence of DIron (Libânio 2010). Turbidity was positively correlated with air and water temperatures, showing an increase in samples taken in the summer season and with color, demonstrating fragility in well protection. Deforestation activities in Brazil, including removal of riparian vegetation also affect water quality of water bodies and contribute to deterioration (Val et al. 2019).
Organic matter showed a correlation with total Al, TIron, and total Mn, which is related to the complexation capacity of these metals. Hardness, which indicates the concentration of multivalent cations in water and is caused by the predominant combination of total Ca and total Mg ions, and to a lesser extent with manganese, showed a positive correlation between these variables, which may occur due to their similar geochemical behavior. It showed a positive correlation with color, probably due to Mn.
Total Al was correlated with TIron, and Mn, which may be related to the type of soil, since there was no interference from rainfall. Ion exchange reactions facilitated by weathering of silicate minerals are the main factors controlling groundwater chemistry (Sunkari et al. 2020).
According to Taloor et al. (2023) based on rock water interaction, there is a considerable variation in the seasons, which indicates the role of weathering and dissolution of rock minerals. The authors also reported the multivariate statistical analysis reveals that the lithogenic factors, such as rock-water interactions and weathering of carbonate-bearing rocks, are predominantly controlling groundwater chemistry.
Table 3 lists the correlation between the variables for the surface source used to supply the municipality of Capão do Leão. Only the TS variable showed no correlation with any other variable in Capão do Leão. There was a positive correlation between the year and TIron,, alkalinity, organic matter, and turbidity, demonstrating a temporal increase in these variables. Air temperature was correlated with water temperature and with DIron and DAlum, related to the influence of temperature on the solubilization of these metals.
Total Al and TIron, were positively and strongly correlated with each other, and both with organic matter, since this variable associates with metal ions, forming complexes. There was also a relationship between these metals with color and turbidity, associating their origin with collection periods with greater rainfall.
There was a positive correlation between color and total P, organic matter, turbidity and DIron, and a negative correlation with pH, demonstrating a predominance of acids in the composition of organic matter. There was a correlation of organic matter with DIron, turbidity, nitrate, and chlorides, as well as total P was strongly associated with organic matter, turbidity, color, and chlorides, which can be related to carrying allochthonous matter from erosive processes into the water course.
Hardness, nitrate, and total Mg were positively correlated, as well as alkalinity with total Mg and nitrate. This, in turn, showed a moderate correlation with chloride, demonstrating the influence of anthropogenic sources, as found by Souza and Gastaldini (2014).
Table 3. Correlation between the variables for the SWS of Capão do Leão, RS.
Parameters
|
Year
|
T. air
|
T. water
|
Al Total
|
TIron
|
Color
|
Hardness
|
P Total
|
Mg Total
|
MO
|
pH
|
Turbidity
|
DIron
|
Nitrate
|
Chlorides
|
DAlum
|
Year
|
1,000
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
T. air
|
-
|
1,000
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
T. water
|
-
|
0,528
|
1,000
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Al Total
|
-
|
-
|
-
|
1,000
|
|
|
|
|
|
|
|
|
|
|
|
|
TIron
|
0,666
|
-
|
-
|
0,784
|
1,000
|
|
|
|
|
|
|
|
|
|
|
|
Color
|
-
|
-
|
-
|
0,818
|
0,801
|
1,000
|
|
|
|
|
|
|
|
|
|
|
Hardness
|
-
|
-
|
-
|
-
|
-
|
-
|
1,000
|
|
|
|
|
|
|
|
|
|
P Total
|
-
|
-
|
-
|
-
|
0,658
|
0,725
|
-
|
1,000
|
|
|
|
|
|
|
|
|
Mg Total
|
-
|
-
|
-
|
-
|
-
|
-
|
0,734
|
-
|
1,000
|
|
|
|
|
|
|
|
OM
|
0,509
|
-
|
-
|
0,758
|
0,906
|
0,907
|
-
|
0,751
|
-
|
1,000
|
|
|
|
|
|
|
pH
|
-
|
-
|
-0,612
|
-0,516
|
-
|
-0,548
|
-
|
-
|
-
|
-
|
1,000
|
|
|
|
|
|
Turbidity
|
0,772
|
-
|
-
|
0,725
|
0,921
|
0,677
|
-
|
0,578
|
-
|
0,780
|
-
|
1,000
|
|
|
|
|
DIron
|
-
|
0,570
|
-
|
0,667
|
-
|
0,627
|
-
|
-
|
-
|
0,597
|
-0,551
|
-
|
1,000
|
|
|
|
Nitrate
|
-
|
-
|
-
|
-
|
-
|
-
|
0,517
|
-
|
0,653
|
0,615
|
-
|
-
|
-
|
1,000
|
|
|
Chlorides
|
-
|
-
|
-
|
-
|
-
|
-
|
-
|
0,567
|
-
|
0,743
|
-
|
-
|
-
|
0,750
|
1,000
|
|
DAlum
|
-
|
0,826
|
0,786
|
-
|
-
|
-
|
-0,686
|
-
|
-0,592
|
-
|
-
|
0,500
|
-
|
-
|
-
|
1,000
|
Table 4. Correlation between the variables for the SWS Camaquã.
|
Parameters
|
Season
|
Year
|
T. air
|
T. water
|
Al Total
|
TIron
|
Mn Total
|
Alkalinity
|
Ca Total
|
Color
|
P Total
|
MO
|
pH
|
Turbidity
|
DIron
|
Nitrate
|
DAlum
|
Season
|
1,000
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Year
|
-
|
1,000
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
T. air
|
-0,799
|
-
|
1,000
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
T. water
|
-0,877
|
-
|
0,851
|
1,000
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Al Total
|
-
|
-
|
-
|
-
|
1,000
|
|
|
|
|
|
|
|
|
|
|
|
|
TIron
|
-
|
-
|
-
|
-
|
0,902
|
1,000
|
|
|
|
|
|
|
|
|
|
|
|
Mn Total
|
-0,869
|
-
|
0,551
|
0,765
|
-
|
-
|
1,000
|
|
|
|
|
|
|
|
|
|
|
Alkalinity
|
-
|
-
|
-
|
-
|
-
|
-
|
-
|
1,000
|
|
|
|
|
|
|
|
|
|
Ca Total
|
-0,653
|
-
|
-
|
-
|
-0,515
|
-
|
0,813
|
0,608
|
1,000
|
|
|
|
|
|
|
|
|
Color
|
-
|
-
|
-0,625
|
-
|
0,853
|
0,765
|
-
|
-
|
0,517
|
1,000
|
|
|
|
|
|
|
|
P Total
|
-
|
-0,525
|
-
|
-
|
0,716
|
0,730
|
-
|
-
|
-
|
0,706
|
1,000
|
|
|
|
|
|
|
OM
|
-
|
-
|
-
|
-
|
0,661
|
0,514
|
-
|
-
|
-
|
-
|
0,649
|
1,000
|
|
|
|
|
|
pH
|
-
|
-
|
-
|
-
|
-
|
-
|
-
|
0,549
|
-
|
-
|
-
|
-
|
1,000
|
|
|
|
|
Turbidity
|
-
|
-
|
-0,547
|
-
|
0,888
|
0,790
|
-
|
-
|
-0,599
|
0,849
|
0,765
|
0,578
|
|
1,000
|
|
|
|
DIron
|
-
|
-0,638
|
-
|
-
|
-
|
0,515
|
0,503
|
-
|
-
|
-
|
-
|
-
|
-0,610
|
-
|
1,000
|
|
|
Nitrate
|
-
|
-
|
-0,530
|
-
|
0,518
|
-
|
-
|
-
|
-
|
0,525
|
-
|
-
|
-
|
0,800
|
-
|
1,000
|
|
DAlum
|
-
|
-
|
-
|
-
|
0,721
|
0,745
|
-
|
-0,590
|
-
|
0,523
|
0,626
|
0,623
|
-0,814
|
0,636
|
0,709
|
-
|
1,000
|
Table 5. Correlation between the variables for the SWS Rio Grande.
|
Parameters
|
Season
|
Year
|
T. air
|
T. water
|
Al Total
|
TIron
|
Mn Total
|
Alkalinity
|
P Total
|
pH
|
Sólidos Totais
|
Turbidity
|
DIron
|
Nitrate
|
DAlum
|
|
Season
|
1,000
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Year
|
-
|
1,000
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
T. air
|
-0,874
|
-
|
1,000
|
|
|
|
|
|
|
|
|
|
|
|
|
|
T. water
|
-0,888
|
-
|
0,883
|
1,000
|
|
|
|
|
|
|
|
|
|
|
|
|
Al Total
|
-
|
-
|
-
|
-
|
1,000
|
|
|
|
|
|
|
|
|
|
|
|
TIron
|
-0,605
|
-
|
-
|
0,582
|
0,891
|
1,000
|
|
|
|
|
|
|
|
|
|
|
Mn Total
|
-0,869
|
-
|
0,728
|
0,850
|
0,529
|
0,725
|
1,000
|
|
|
|
|
|
|
|
|
|
Alkalinity
|
-0,752
|
-
|
-
|
0,627
|
-
|
0,616
|
0,721
|
1,000
|
|
|
|
|
|
|
|
|
P Total
|
-
|
0,721
|
-
|
-
|
0,644
|
0,525
|
-
|
0,675
|
1,000
|
|
|
|
|
|
|
|
pH
|
-
|
-
|
-0,567
|
-
|
-
|
-
|
-
|
-
|
-
|
1,000
|
|
|
|
|
|
|
Sólidos Totais
|
-
|
-
|
-
|
-
|
0,778
|
0,680
|
-
|
-
|
0,566
|
-
|
1,000
|
|
|
|
|
|
Turbidity
|
-
|
0,610
|
-
|
-
|
0,865
|
0,671
|
-
|
-
|
0,870
|
-
|
0,772
|
1,000
|
|
|
|
|
DIron
|
-
|
-
|
-
|
-
|
-
|
-
|
-
|
-
|
-
|
-
|
-
|
-
|
1,000
|
|
|
|
Nitrate
|
-0,637
|
-
|
0,564
|
-
|
-
|
0,557
|
-
|
-
|
-
|
-
|
-
|
-
|
0,633
|
1,000
|
|
|
DAlum
|
-
|
-
|
-
|
-
|
-
|
-
|
-
|
-
|
-
|
-
|
-
|
-
|
0,814
|
-
|
1,000
|
|
Table 4 presents the correlation between the variables for the Camaquã Reservoir source. In this source, all variables were correlated with each other. There was no positive correlation between the variables and seasons or years. The season influenced the air and water temperatures, as well as the reduction of total Mn and total Ca concentrations in samples taken in the winter. Temperatures were correlated with total Mn, indicating the influence of temperature on total Mn and its distribution in the reservoir.
There was a positive correlation between alkalinity, total Ca, and pH, since this salt can originate from the dissolution of rocks. Total P was correlated with total Al and DAlum, TIron, color, organic matter, and turbidity, making it possible to relate these variables to agricultural activity, a potential source of total P due to the use of fertilizers. Valentini et al. (2021) evaluated the water quality of Mirim Lagoon, of RS, using a principal component analysis, and reported that phosphorus, turbidity, and TS were grouped along the same principal component and that these are associated with pollution of agricultural origin from activities developed around the water source.
Total Al was associated with TIron, and both showed a correlation with color, organic matter, turbidity, and DAlum, which in turn was correlated with DIron, organic matter, color, and turbidity. There was also a correlation between nitrate and color, turbidity, and TS; between turbidity and color, and organic matter; between DIron and TIron, and total Mn. Such associations of variables make it possible to relate the ease of erosion typical of the type of soil in the region of the reservoir, increased by the lack of riparian forest, adding allochthonous material to the reservoir. Furthermore, the pH was negatively correlated with DAlum and DIron, and positively with nitrate, showing that the higher the pH, the lower the concentrations of these metals and the higher the nitrate.
The solubility of many of these elements (e.g. Iron and Mn) is strongly dependent on pH and redox conditions, and possibly complexation with organic carbon, which may explain the generally higher concentrations in surface waters (organic complexation and lower pH) (Flem et al. 2018).
For the surface water source of Rio Grande, the results of the correlation analysis are listed in Table 5.
For the Rio Grande water source, the season showed a negative correlation with air and water temperature, TIron and total Mn, in addition to alkalinity and nitrate, indicating that, in samples collected in the winter, the concentrations of these substances reduced. Still, DIron correlated with nitrate and DAlum. The year variable was positively correlated with total P and turbidity, showing a tendency for temporal increase in their concentrations.
With regard to metals, there was a positive correlation of total Al and TIron with total Mn, total P, TS, and turbidity, associating their increase in the water body with surface runoff. Still, total solids were correlated with turbidity, which was not observed in the other sources..
Alkalinity showed a positive correlation with TIron, total Mn, and total P, which in turn was correlated with turbidity and TS, is commonly related to erosion processes that contribute to the increment of soil particles in water bodies. Calazans et al. (2018) analyzed a Brazilian river basin and reported that the parameters thermotolerant coliforms, total manganese, and total phosphorus were considered the most relevant and total chromium, total cadmium, selenium, total dissolved copper, and total boron the less relevant for the characterization of water quality. The autores also reported the discharge of domestic sewage and industrial wastewater are the main sources of pollution responsible for the surface water quality deterioration in the basin.
In order to understand a large number of variables studied, a cluster analysis was carried out, which groups the variables according to their similarities within the class and dissimilarities between the different classes, showing high homogeneity within the cluster and high heterogeneity among the clusters. It was applied considering the data of the variables used in the previous steps, already standardized, and separated by water source, as illustrated in Figure 2.
Two groups of variables were obtained for the Cristal groundwater source (Figure 2). Group 1 included air and water temperature, total Al, total Mg, OM pH, total solids, nitrate, and chlorides. Group 2 comprised the variables TIron, total Mn, total Ca, color, hardness, TP, turbidity, DIron, and alkalinity showing the relationship between metal concentration and changes in color and turbidity due to their solubilization when oxidized, as well as, due to the characteristics of alluvial deposits, where higher concentrations of TIron and total Mn occur. Still, it can be inferred that in the hardness and alkalinity of this water source, multivalent cations prevail, such as total Ca, TIron, and total Mn.
For the Arroio Duro Reservoir (Figure 2), air and water temperature, total Mn, alkalinity, total Ca, hardness, total Mg, pH, and DIron belonged to group 1, in which minerals predominated. In group 2, the allocated variables were total Al, TIron, color, TP, OM, TS, turbidity, nitrate, and DAlum, usually associated with anthropogenic alterations, such as the existing agricultural activity around the reservoir, which has areas without riparian vegetation, carrying substances into its interior, in addition to propensity to erosion of the soil in the region.
In the Capão do Leão water source (Figure 2), air and water temperature, total Mn, alkalinity, total Ca, total Mg, hardness, and pH belonged to the first group, prevailing mineral elements that make up alkalinity and hardness. Group 2 was formed by total Al, TIron, color, TP, OM, TS, turbidity, DIron, nitrate, chlorides, and DAlum; and in this group, the variables associated with the degradation of water quality by human activities predominate, such as the runoff of fertilizers used in agricultural activity, surface runoff, which also indicate possible contamination by sewage.
The grouping of variables in the Rio Grande water source (Figure 2) was distributed as follows: in group 1, air and water temperature, total Al, TIron, total Mn, alkalinity, total Ca, hardness, TP, total Mg, OM, turbidity, and TS, predominating in this group minerals, total metals, and one of the main indicator parameters of environmental degradation, the TP. In group 2, color, pH, DIron, DAlum, nitrate, and chlorides.
The surface sources presented similar groups regarding their compositions. In group 1, minerals and metals present in hardness and alkalinity predominated, while in group 2, variables indicating some type of environmental degradation in the water body prevailed, mainly associated with soil management with erosive potential and that end up carrying particles and nutrients into water bodies. The release of contaminants from diffuse anthropogenic sources as well as different natural conditions (surface water quality distance to shore, leaky conditions between aquifers, etc.) may also directly or indirectly affect the hydrogeochemical composition of the groundwater (Islam et al. 2019). Yan et al (2023) reported the surface water is affected by the weathering and dissolution of both silicate and carbonate rocks, while GW is mainly affected by the weathering and dissolution of silicate.