Spatio-temporal Study of Pollutants in the Rio De Ondas Hydrographic Basin, West of Bahia, Brazil Using Multivariate Analysis

doi:10.21203/rs.3.rs-2239257/v1

Download PDF

Research Article

Spatio-temporal Study of Pollutants in the Rio De Ondas Hydrographic Basin, West of Bahia, Brazil Using Multivariate Analysis

https://doi.org/10.21203/rs.3.rs-2239257/v1

This work is licensed under a CC BY 4.0 License

Journal Publication

published 09 Sep, 2023

Read the published version in Environmental Monitoring and Assessment →

You are reading this latest preprint version

Water bodies are containers that receive a large load of pollutants through the release of domestic, industrial, and agricultural effluents. With this focus, this work aimed to conduct a temporal-spatial variability study in the Rio de Ondas Hydrographic Basin through multivariate statistical analysis. For this, seventeen collection sites were established in four stations along the Rio de Ondas and its tributaries between 2017 and 2018. Ionic chromatography with suppressed conductivity was used for ions determination, while ICP-OES determined metals' total concentrations. The land use and occupation assessment between 1985 and 2015 was conducted using images from Landsat 5 and 8 satellites and the descriptive and multivariate analysis of the data using version 10 of the Statistica software. The results showed that, in thirty years, there was a growth of 569% of agricultural activities in the watershed area, with significant suppression of native vegetation, favoring the transport of pollutants to rivers. Ca²⁺, PO₄²⁻, Al, Cu, and Zn concentrations showed a statistically significant difference between the seasons, with higher medians in the rainy season. Rainy season influenced the formation of three groups in the PCA, consisting of electrical conductivity, salinity, TDS, and PO₄²⁻ (group 1); temperature, Fe, SO₄²⁻ and Cl⁻ (group 2); and Ca²⁺, Mg²⁺, Na⁺, and HCO₃⁻ (group 3). The strong correlation between parameters of each group indicates anthropic influence on the watershed's water quality. However, levels are within the potability standard.

PCA

Water quality

Seasonality

Rio de Ondas

Transport of Pollutants

Agricultural Inputs.

Water resources play an essential role in the survival of living species on the planet (Bynum and Bynum, 2018; Edwards et al., 2019; Padowski, 2020; Zhiltsov et al., 2020). Through them, human beings meet their basic personal, social, and economic (agricultural and industrial) needs (Edwards et al., 2019; Padowski, 2020). However, its inappropriate use leads to significant changes in its quality, compromising the survival of species in their ecosystem and, consequently, their uses for different purposes (Khangembam and Kshetrimayum, 2019; Province, 2015; Vittori et al., 2018; Wang et al., 2019; Xu et al., 2019; Zhao et al., 2019).

Natural and anthropogenic factors can influence changes in water quality, and rivers receive a considerable load of pollutants released into the air, soil, or directly into drainage networks (Padowski, 2020; Province, 2015; Vittori et al., 2018; Zhao et al., 2019). Of these, anthropogenic activities are the primary sources of alterations, dumping urban (Castro et al., 2021; Padowski, 2020; Province, 2015), industrial (Dadi et al., 2018; Gonzalo and Camargo, 2013; Kistemann et al., 2016) and agricultural (Edwards et al., 2019; Vittori et al., 2018) effluents that can pose a risk and cause severe harmful effects to marine and human species once they reach water bodies, due to the highly toxic elements, ions and substances they carry (Dadi et al., 2018; Gonzalo and Camargo, 2013; Nargis et al., 2021; Wang et al., 2019; Xu et al., 2019; Zhao et al., 2019).

Pollutants can reach water bodies through diffuse and point source pollution (Azzellino et al., 2019; Harrison et al., 2019; Motevalli et al., 2019; O’Shea, 2002; Wang et al., 2018). The former is promoted by surface runoff over large areas and makes identification, measurement, and, consequently, controlling difficult, as there are low concentrations of polluting species in water (Carazo-Rojas et al., 2018; Libos et al., 2003; Lima et al., 2016). In turn, point source pollution leads to higher concentrations of these species because it originates from wastewater, including domestic and industrial effluents, frequently reaching water bodies at a defined location in a concentrated manner (Carazo-Rojas et al., 2018; Lima et al., 2016). These two forms of pollution can be identified through water quality parameters, such as temperature, acidity (pH), alkalinity, hardness, concentrations of chlorides, fluorides, iron, manganese, nitrogen, phosphorus, and other organic and inorganic compounds (Lima et al., 2016; Sales et al., 2014). Identifying these parameters also allows the evaluation of primary pollution sources through mathematical models, such as multivariate statistical analysis.

Multivariate analysis techniques help assess possible contamination sources based on the study of grouping and correlation of variables (Kumari and Sharma, 2019). Principal component analysis (PCA), for example, can be applied when there are many variables, promoting their grouping using data correlation (Miller and Miller, 2010). In this way, each variable's contribution is evaluated, looking for possible relationships according to the study area. In a study carried out in the Rio de Ondas Hydrographic Basin (ROHB), western Bahia, Brazil, PCA allowed the identification of two distinct groups, characterized by the influence of seasonality (dry and rainy seasons) (Rego et al., 2014). However, the study does not elucidate pollution sources besides showing a significant difference between the concentrations in the sampling periods.

ROHB supplies much of the water consumed by cities in the region and is used to produce food, mainly through irrigation, and is essential for the local population's basic needs. The scarcity of data in the scientific literature and the cultural, social, economic, and historical importance of the Rio de Ondas Watershed open space for research about its water quality and assessment of environmental contamination. In this sense, this work aimed to carry out a spatio-temporal study of the concentration of pollutants in the waters of the ROHB utilizing multivariate statistical analysis.

2.1 Characterization of the study area and sampling sites

The Rio de Ondas Hydrographic Basin (ROHB) is located at coordinates 11°55' and 12°34' of south latitude, and 46°23' and 45°0' of west longitude, with an area of 5,157 km² (Almeida et al., 2016). The region is part of the cities' territories of Luís Eduardo Magalhães, São Desidério and Barreiras.

At the north direction, its limits are the Rio de Janeiro basin; at the south, the Rio das Fêmeas basin; at the east, the Rio Grande basin, which includes 32 perennial rivers along its course, all of which belong to the São Francisco watershed. The main tributaries of the Rio de Ondas basin are: Rio de Pedras, Rio Borá and Rio Vereda das Lajes. Figure 1 shows the study area's location and sampling points.

The west of Bahia, the focus region of the present study, has been highlighted nationally for the production of agricultural commodities, such as soy, corn, and cotton, with an average annual output of 42 bags per hectare in the rainfed system (1,270 thousand hectares), and 50 bags per hectare in the irrigated system (36 thousand hectares) for soybeans, 145 bags per hectare in rainfed (245.5 thousand hectares) and 192 bags per hectare in irrigated (7 thousand hectares) for corn, and 4.05 tons per hectare in rainfed (285.3 thousand hectares) and 4.35 tons per hectare in irrigated (22.6 thousand hectares) for cotton (AIBA, 2015). The region's crop production is based on cultivation using central pivots, mainly due to the extended period of drought. In 2014, for example, the ROHB area had 190 central pivots (Almeida et al., 2016) out of a total of more than 1,000 (AIBA, 2015), which draws attention to the concern about land use and occupation’s influence on water quality.

To characterize the water quality and identify polluting sources, sample collections were carried out during the 2017/2018 harvest period, the first being at the end of 2017's dry season (D1), in October, when the soil was being prepared for planting. The second sampling period took place after the beginning of the rainy season (R1) in December 2017, the beginning of the planting period, and the third in the final stage of the rainy season (R2) in March 2018, the harvest period for some crops such as soybean and corn. Finally, the fourth sampling period was at the beginning of the dry season (D2), June 2018, the cotton harvest period (Borém and Freire, 2014; Sediyama et al., 2015). In each time (D1, R1, D2 and R2), seventeen surface water samples were collected at different points distributed along the ROHB. An additional sample was collected from a spring with preserved vegetation, identified as a background point (BP).

2.2 Sampling

Water samples were collected in glass flasks, previously washed with 5% (v/v) neutral Extran® solution and placed in a styrofoam box with ice until arrival at the laboratory, where they were filtered through a cellulose acetate 0.45 µm membrane filter. During sample collection, a Soil Control MMAD-5 multiparameter probe was employed to determine the following variables: pH, temperature (T), electrical conductivity (EC), total dissolved solids (TDS), and salinity (S).

2.3 Device and Conditions

For the anions and cations determinations, we adopted the methodologies proposed by Thermo Fisher Scientific (MICHALSKI, 2009), following the criteria and procedures of the Standard Methods for the Examination of Water and Wastewater, by the American Public Health Association (APHA), the American Water Works Association (AWWA), and the Water Environment Federation (WEF) (APHA/AWWA/WEF, 2012).

For the determination of cations (Na⁺, K⁺, Mg²⁺, and Ca²⁺) and anions (F⁻, Cl⁻, NO₃⁻, PO₄²⁻ and SO₄²⁻) by ion chromatography (IC), a Dionex, model ICS-90, with suppressed conductivity detection and automatic sampler 858 Professional Sample Processor was used. Conductivity suppressor was an Atlas AAESTM in recycle mode, 45mA. The eluent was a Na₂CO₃:NaHCO₃ 3.5:1 mM solution with isocratic flow at 1.2 mL min^− 1, approximate pressure of 1500 psi and 25 µL loop volume. The ion exchange columns and pre-columns were made of quaternary alkylammonium functionalized surface, models AS14A (4 x 250 mm, 5 µm i.d.) and AG14A (4 x 250 mm), respectively, for anions, and CS14A (4 x 250 mm, 5 µm i.d.) and CG14A (4 x 250 mm) for cations.

The ionic balance was performed by calculating the product of the charges by the concentration (µeq L^− 1) of each ion (Scopel et al., 2005). For this, we considered the major ionic species present in the samples. The results showed that, for most samples, the errors fell within the range of ± 20%. This data is presented in more detail in the supporting information file.

The quantification of the elements Al, Fe, Mg, Mn, Zn, Cu, and Si was performed by ICP/AES (5100, Agilent, Hainesport, NJ, USA). The operational parameters were as follows: nebulizer flow of 0.70 L min^− 1, nebulizer back pressure of 307.9 kPa, plasma argon flow of 11.98 L min^− 1, argon pressure of 527.9 kPa, auxiliary gas flow of 1.00 L min^− 1, and RF power of 1,199.6 W.

2.4 Land use and occupation in Rio de Ondas Hydrographic Basin

Landsat 5 and 8 satellite images were used for landscape classification and temporal analysis between 1985 and 2015, corresponding to 30 years of analysis. Orbits 220 − 68 and 220 − 69 were used as references to cover the study area, and the images were obtained from the United States Geological Survey's (USGS) EarthExplorer database. The landscape types were determined from observations of the region's thematic maps, considering the limitations of the spectral resolution of Landsat images. The reference pixels were collected in ENVI using the Region of Interest tool. The landscape types or elements determined were: natural vegetation (Cerrado), pasture, agriculture, forestry, irrigation pivot, and urban areas.

2.5 Statistical Analysis

The analytical results were organized in matrices, and descriptive statistics were used to facilitate the interpretation of the data since it describes its characteristics through mean values, standard deviations, percentages, maximums, and minimums. Thus, the Kolmogorov-Smirnov normality test was performed to apply the appropriate correlation and hypothesis tests for parametric and non-parametric variables (p < 0.05).

Kruskal-Wallis test (U test) was applied for variables with non-parametric distribution to identify the different groups. At the same time, one-way ANOVA and multiple-comparison post hoc (Bonferroni) were utilized for variables with parametric distribution to identify the different means. The least acceptable level of significance was used for representation purposes (p < 0.05) (Marques et al., 2013).

The water analysis data was submitted to principal components analysis (PCA), in which concentrations below the LOQ were substituted with a value of half the LOD to avoid loss of information (Christofaro and Leão, 2014).

This article used the components that presented an accumulated variance of around 60% of the total variance. The factor loading values classification corresponded to the absolute composition of the principal components (PC) and were marked as relevant (> 0.75), medium (> 0.50 and ˂0.75), and irrelevant (˂0.50) (Finkler et al., 2015). The Statistica software version 10 was employed on all statistical calculations.

3.1 Change in ROHB's vegetation cover (Spatio-temporal analysis)

Changes in ROHB's vegetation cover between 1985 and 2015 are shown in Table 1.

Table 1

Quantitative classification of ROHB's land use and occupation.
Classes	Area in 1985	Area in 2015
Agriculture	7.55%	40.88%
Urban area	0.01%	0.38%
Cerrado (natural vegetation)	91.73%	52.65%
Pasture	0.68%	1.57%
Irrigation pivot	0.03%	4.08%
Silviculture	0.00%	0.43%
Total	100%	100%

It was observed that there was a 569% growth of agricultural activities in the ROHB area in thirty years, while native vegetation suffered a 57.40% suppression of its total area in the same period (Almeida et al., 2016). The expansion of agriculture in the cerrado began in the 1970s, with several factors converging on the growth of agribusiness in this region, like the availability of land, technological and financial strengthening, and the water availability associated with climatic and topographical conditions (Almeida et al., 2016; Sicsú and Lima, 2000). However, this dynamic caused significant changes in the basin’s vegetation cover.

A study on the effects of land use and change in vegetation cover on the water quality of the uMngeni River in South Africa showed a strong relationship between the reduction of natural vegetation and the nutrient concentrations increase in the analyzed samples (Namugize et al., 2018). This increase in nutrient concentration can be caused by in natura release of domestic sewage (Lima et al., 2016; Vryzas, 2018), surface runoff from chemical fertilized agricultural areas (Libos et al., 2003; Lima et al., 2016; Vryzas, 2018), or domestic industrial effluents (Lima et al., 2016; Vryzas, 2018). Likewise, the land use and occupation by agricultural activities in ROHB may be influencing the variables’ concentrations in our study.

3.2 Descriptive statistical analysis of data and seasonality

Analysis of ROHB’s surface waters was performed to determine twenty-three parameters presented individually for each sampling period in supplementary material (SM). Table 2 presents the results of the descriptive analysis of the variables studied between the dry (D1 and D2) and rainy (R1 and R2) periods.

Table 2

Summary of variables from the ROHB's hydrological periods, N = 17.
Variables	D1	R1	R2	D2	p	Maximum allowable (Brazil, 2017; CONAMA, 2005)
Variables	Average or median (min - max)	Average or median (min - max)	Average or median (min - max)	Average or median (min - max)	p	Maximum allowable (Brazil, 2017; CONAMA, 2005)
NH₄⁺ (mg L^− 1)	0.068 ^abc (< LOQ- 0.120)	0.127 ^abcd (< LOQ–0.283)	0.100 ^abcd (0.042–0.155)	0.155 ^bcd (0.050–0.450)	0.001227	1.200
Ca²⁺ (mg L^− 1)	0.544 ^abcd (0.210–1.200)	0.650 ^abcd (0.220–1.179)	0.770 ^abcd (< LOQ − 2.408)	0.417 ^abcd (0.040–1.615)	1.000000	-
Mg²⁺ (mg L^− 1)	0.025 ^abd (< LOQ − 0.100)	0.025 ^abcd (< LOQ − 0.219)	0.135 ^bcd (< LOQ − 0.439)	*0.030 ^abcd (< LOQ − 0.278)	0.000549	-
Na⁺ (mg L^− 1)	0.331 ^ad (0.030–1.540)	*0.710 ^abd (0.193–6.60)	2.776 ^cb (0.414–7.815)	*0.494 ^abd (0.191–7.155)	0.000000	200.000
K⁺ (mg L^− 1)	0.105 ^acd (< LOQ − 0.540)	0.345 ^bcd (< LOQ − 0.773)	0.323 ^abcd (< LOQ − 1.571)	0.321 ^abcd (0.173–0.761)	0.046912	-
F⁻ (mg L^− 1)	<LOQ	<LOQ	<LOQ	<LOQ	-	1.400
Cl⁻ (mg L^− 1)	0.713 ^ad (0.540–1.760)	1.271 ^cb (0.750–2.120)	1.023 ^bcd (0.620–2.130)	0.733 ^acd (0.490–1.250)	0.000015	250.000
NO₃⁻ (mg L^− 1)	*0.010 ^acd (< LOQ − 0.420)	0.514 ^bcd (< LOQ − 1.060)	0.405 ^abcd (< LOQ − 2.000)	0.422 ^abcd (0.180–1.170)	0.000745	10.000
PO₄²⁻ (mg L^− 1)	*0.005 ^abcd (< LOQ − 0.450)	*0.005 ^abcd (< LOQ − 1.720)	*0.005 ^abcd (< LOQ − 0.670)	*0.005 ^abcd (< LOQ − 0.610)	1.000000
SO₄²⁻ (mg L^− 1)	*0.025 ^ad (< LOQ − 0.530)	0.588 ^bc (< LOQ − 2.290)	0.498 ^bc (0.340–0.760)	*0.025 ^ad (< LOQ − 0.460)	0.000205	250.000
HCO₃⁻ (mg L^− 1)	1.238 ^abd (< LOQ − 5.200)	1.988 ^abd (< LOQ − 15.67)	6.309 ^cd (0.040–16.87)	3.489 ^abcd (0.040–15.89)	0.002048	-
pH	5.542 ^abc (4.820–6.970)	5.462 ^abc (4.720–6.750)	5.50 ^abc (4.640–7.000)	7.297 ^d (4.800–9.520)	0.000002	6.000–9.000
Sal. (mg L^− 1)	1.942 ^ab (0.001–7.800)	3.535 ^abcd (1.600–8.200)	5.012 ^bcd (1.100–9.600)	4.594 ^bcd (0.700–8.200)	0.004095	300.000
Cond. (µS cm^− 1)	4.594 ^abd (0.000–15.20)	7.876 ^abcd (3.100–16.30)	10.31 ^bcd (3.100–19.80)	8.818 ^acbd (1.800–16.70)	0.003939	-
TDS (mg L^− 1)	3.071 ^ab (0.001–10.40)	4.788 ^abcd (2.500–11.20)	7.053 ^bcd (2.000–12.60)	6.200 ^bcd (1.100–12.20)	0.004234	200.000
Temp. (°C)	24.85 ^a (23.60–26.70)	25.87 ^bc (24.20–27.80)	26.38 ^bc (25.00–28.30)	22.78 ^d (21.3–24.44)	0.000000	-
Al_total (mg L^− 1)	0.030 ^abcd (< LOQ − 0.049)	0.049 ^abcd (< LOQ − 0.080)	0.047 ^abcd (< LOQ − 0.116)	*0.038 ^abcd (< LOQ − 0.248)	1.000000	0.100
Cu_total (mg L^− 1)	*0.002 ^abcd (< LOQ − 0.009)	0.002 ^abcd (< LOQ − 0.004)	0.002 ^abcd (< LOQ − 0.006)	0.002 ^abcd (< LOQ − 0.003)	1.000000	0.009
Fe_total (mg L^− 1)	0.026 ^acd (< LOQ − 0.084)	0.096 ^bc (< LOQ − 0.419)	0.051 ^abcd (< LOQ − 0.128)	0.026 ^acd (< LOQ − 0.098)	0.005524	0.300
Mn_total (mg L^− 1)	<LOQ	<LOQ	<LOQ	<LOQ	-	0.100
Zn_total (mg L^− 1)	*0.003 ^abcd (< LOQ − 0.078)	0.005 ^abcd (< LOQ − 0.026)	0.006 ^abcd (< LOQ − 0.018)	0.002 ^abcd (< LOQ − 0.006)	1.000000	5.000
The means with different letters in superscript are statistically different after adjusting for the Bonferroni test (p < 0.05); *Comparisons using the Kruskal-Wallis test (Median - Non-parametric data); p = lowest p-value for the compared group; LOQ – Limit of quantification.

We noticed that there is no statistically significant distinction between the hydrological periods for the parameters Ca²⁺, PO₄²⁻, Al_total, Cu_total, and Zn_total, where p > 0.05. However, for most of the variables, there were significant differences, i.e., p < 0.05, between the medians of the groups (D1, D2, R1, R2), evidencing the influence of the hydrological periods on the water quality parameters for the study area. In all samples, at the four sampling points, F⁻ and Mn_total were below the limit of quantification.

Temperature and the concentrations of Na⁺, Cl⁻, SO₄²⁻, and Fe_total were significantly different in the dry and rainy seasons, with the highest values in the rainy periods. Thus, the ions Na⁺, Cl⁻ and SO₄²⁻ may be associated with anthropogenic sources, such as domestic effluents, fertilizers, and septic tanks, since the rain causes an increase in surface runoff (Barzegar et al., 2016). Moreover, the water temperature can be affected by natural and anthropogenic factors of energy exchange, modifying the rivers' physical, chemical, and biological processes and, consequently, the survival of aquatic species (Hanafiah et al., 2018; Mustapha et al., 2013). In this study, temperatures were higher in the rainy season when river flows are higher, with an increased heat absorption capacity and less water uptake for crops irrigation (Xin and Kinouchi, 2013). The higher Fe_total concentrations in the rainy season are related to the increase of soil leaching, which, even with low levels of the element, can reach the drainage networks in this period (Rego et al., 2021; Freitas et al., 2014).

The pH at the D2 sampling period showed higher values, significantly different from the other periods. The pH in D1, R1, and R2 agrees with results presented in other works in the study area, being slightly acidic due to the natural characteristics of the region, especially the presence of native vegetation (Rego et al., 2014; Rego et al., 2017). The period D2 is affected by point sources, which may be related to the decrease in water temperature in the same season (Souza et al., 2021).

For the other parameters, the concentrations were very similar between the hydrological periods, showing, in most cases, no significant difference. However, in the rainy season, the average concentrations tend to be higher, justified by the leaching of soils in agricultural regions, with the presence of fertilizers and soil correctors. The seasonal distinction for water quality parameters was previously observed in another work by our research group in the same study area, in which the multivariate statistical analysis showed how the seasons influence the concentrations of the analytes (Rego et al., 2014).

3.3 Principal Components Analysis

PCA was performed to reduce the number of variables and facilitate the extraction of more relevant information in assessing ROHB's water quality. For this, we used the average of each variable in D1 and D2 to study the dry season and the averages in R1 and R2 for the rainy season. The Kaiser criterion was followed to select the number of principal components (PC), selecting only principal components with eigenvalues greater than 1(Kaiser, 1960; Souza et al., 2021). In addition, the accumulated variance was also considered to determine the number of factors that must be extracted, and (Hair et al., 2009) suggest the minimum level of 60% is acceptable. Table 3 shows that just by adding PC1 and PC2, the criteria are met, with eigenvalues > 1, and the added factors explain around 66% of the total variance, not requiring other components.

Table 3

Eigenvalues and variances obtained in the PCA for the variables in Table 2.
Components	Eigenvalues	Variance (%)
1	4,.94275	41.18960
2	2.95582	24.63187
3	1.86204	15.51704

Figure 2 shows the graph of the active variables' weights, pointing to 3 different groups that are correlated. Group 1 includes electrical conductivity, salinity, TDS, and PO₄²⁻. In group 2, there is temperature, Fe_total, SO₄²⁻ and Cl⁻. Furthermore, group 3 consists of Ca²⁺, Mg²⁺, Na²⁺, and HCO₃⁻.

In group 1, salinity is the total amount of solid inorganic material dissolved in natural water, and salinization of water refers to an increase in TDS and overall chemical content of the water. The electrical conductivity of water represents its ability to transmit electrical current due to the presence of dissolved substances, mainly inorganic, which dissociate into cations and anions (Berger et al., 2013; Sharifi et al., 2012). The concentration of PO₄²⁻ ions may directly influence TDS levels and conductivity due to the existing relationship with the levels of soluble compounds (Takiyama et al., 2003). Therefore, these parameters are logically positively correlated.

The highest salinity, TDS, and conductivity values were found in P17, located on the Vereda das Lajes River, which is reached by water drained from small farmers' lands who raise animals in improvised channels upstream of this site. Given this, the surface runoff coming from rainwater in fertilized soils and with animal food's waste can increase these parameters. The same was observed in a previous article by (Rego et al., 2014) in the same region.

Temperature, Fe_total, and SO₄²⁻ and Cl⁻ ions, which are in group 2, may be related to the use of agricultural inputs in the study area. Fe has been used for soil stabilization and improvement of nutrient status (Friesl et al., 2004; Warren et al., 2006) while SO₄²⁻ and Cl⁻ ions are used in agriculture for soil correction (CaSO₄) and fertilization (KCl), which, when dissociated, become essential nutrients for plants (Brasil et al., 2020; Paula et al., 2020).

The correlation of Ca²⁺, Mg²⁺, and HCO₃⁻ in group 3 may be related to the dissolution of predominant materials in the soils in some points of the studied area, constituted of carbonate rocks, mainly calcite (CaCO₃) and dolomite (CaCO₃ and MgCO₃) of the Bambuí Group (Amorim and De Lima, 2007; CPRM, 2019, 2004). However, Ca²⁺ and Mg²⁺ can also come from limestone leaching from agricultural areas since they are essential secondary macronutrients in agriculture (Sediyama et al., 2015).

The highest Ca²⁺, Mg²⁺, and HCO₃⁻ concentrations were found in P1R, near the main river's source where there is the presence of native vegetation but also close to many agricultural areas with water channels suppressed by agriculture (Damasceno, 2011). P10D and P10R also showed high values of Ca²⁺, Mg²⁺, and HCO₃⁻, which may be associated with the contribution of the Bambuí karst (Amorim and De Lima, 2007; Rego et al., 2021).

Na⁺, also present in group 3, cannot be related to the study area’s geology, which does not contain halite (NaCl). P8R (4.191 mg L^− 1) and P17R (4.482 mg L^− 1) had higher concentrations of Na⁺, traced to the water flow from regions with agricultural activity. Therefore, the high concentrations of Na⁺ at these points are explained by the flow of domestic and agricultural effluents (Rego et al., 2014).

Figure 3 represents a scatter plot for the points analyzed in PC1 and PC2, showing the separation between the rainy and dry seasons, except for P17D, P10D, P2R, and P12R.

The points of the rainy season are in the same quadrants as the variables presented in Table 2, indicating a strong correlation, mainly due to the leaching of residues in the soil of areas close to the sampling points. These results agree with the descriptive statistical analysis shown in the previous topic.

The application of multivariate analysis provided helpful information on the levels of metals, ions, and physicochemical parameters in ROHB. The data's descriptive statistical analysis showed that the median concentrations of Ca²⁺, PO₄²⁻, Al, Cu, and Zn were higher in the rainy season, which may be related to the more significant soil leaching in this period and with the release of effluents containing pollutants. In addition, the study showed that the region underwent a significant change in land use and occupation: agriculture and livestock grew by 569% in thirty years, while native vegetation was suppressed by 57.40%.

In the PCA, two principal components (PC1 and PC2) with a total variance of 66% separated the parameters into three groups. Group 1 shows a direct correlation between electrical conductivity, salinity, TDS, and PO₄²⁻explained by how the levels of ions dissolved in water increase conductivity and salinity. Group 2, formed by temperature, Fe, SO₄²⁻ and Cl⁻, may be related to the presence of agricultural inputs. Group 3, with Ca²⁺, Mg²⁺, Na²⁺, and HCO₃⁻, can be explained by the dissolution of predominant minerals in soils formed from carbonate rocks from the Bambuí Group (P10, P11, and P17) together with the use of agricultural correctives. These results are coherent with the anthropogenic activities present in the study area. However, for most variables, the concentrations in the samples are still below the safe limits allowed for human consumption.

Acknowledgments

This study was supported by the Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES, Brazil), Fundação de Amparo à Pesquisa do Estado da Bahia (FAPESB, Brazil) and Financiadora de Estudos e Projetos (FINEP, Brazil).

Author contribution Enoc Lima do Rego: Conceptualization; Data curation; Writing - review & editing; Investigation; Methodology. Joelma Ferreira Portela: Data curation; Investigation; Methodology. Camila de Lima Ribeiro: Data curation; Writing - review & editing. João Pedro Rudrigues de Souza: Writing - review & editing Myller de Sousa Tonhá: Writing - review & editing; Validation; Methodology. José Domingos Santos da Silva: Writing - review & editing. Jurandir Rodrigues de Souza: Supervision; Writing - review & editing. All authors reviewed the manuscript.

Competing interests

The authors declare no competing interests.

Funding

Competing interests

The authors declare that they have no competing interests.

Availability of data and materials

The datasets analyzed during the current study are available from the corresponding author on reasonable request.

AIBA. (2015). Yearbook of the Western Region of Bahia Crop 2013/2014. Retrieved July 12, 2022 from https://aiba.org.br
Almeida, R.D.S., Latuf, M.D.O., Santos, P.S. (2016). Deforestation analysis in the Ondas River Basin in the period of 1984 to 2014, Western of Bahia. Cad. Prudentino Geogr. 1, 41–63.
Amorim, V., De Lima, O.A.L. (2007). Using IP-resistivity soundings it is possible to define the structural geometry and the lithological variability within the Urucuia aquifer in the Rio das Femeas basin, Bahia. Rev. Bras. Geofis., 25, 117–129. https://doi.org/10.1590/S0102-261X2007000200002
APHA/AWWA/WEF. (2012). Standard Methods for the Examination of Water and Wastewater. Stand. Methods. Retrieved January 30, 2022 from https://www.standardmethods.org/
Azzellino, A., Colombo, L., Lombi, S., Marchesi, V., Piana, A., Andrea, M., Alberti, L. (2019). Groundwater diffuse pollution in functional urban areas: The need to define anthropogenic diffuse pollution background levels. Sci. Total Environ., 656, 1207–1222. https://doi.org/10.1016/j.scitotenv.2018.11.416
Barzegar, R., Asghari Moghaddam, A., Tziritis, E. (2016). Assessing the hydrogeochemistry and water quality of the Aji-Chay River, northwest of Iran. Environ. Earth Sci., 75(1486), 1–15. https://doi.org/10.1007/s12665-016-6302-1
Berger, J.S., Hermes, E., Zenatti, D.C., Gonçalves, M.P., Lins, M.A., Wulf, V.S. (2013). Relationship between solids and electrical conductivity in tubular biodigester used in the cassava effluent treatment. Sci. Agrar. Parana. 12, 377–383. https://doi.org/10.18188/1983-1471/sap.v12nsupp377-383
Borém, A., Freire, E.C. (2014). Cotton: from planting to harvesting, first ed. UFV, Viçosa, MG.
Brasil, E.C., Lima, E. do V., Cravo, M. da S. (2020). Use of gypsum in agriculture, in: Raij, B. Van (Ed.), General Aspects Related to the Use of Fertilizers and Correctives. Embrapa, Belém, PA., pp. 133–145.
Brazil. (2017). Consolidation Ordinance No. 5, of September 28. (2017). Ministério da Saúde. Retrieved March 3, 2022 from https://bvsms.saude.gov.br/bvs/saudelegis/gm/2017/prc0005_03_10_2017.html
Bynum, H., Bynum, B. (2018). Exhibition Water and our planet. Lancet 391, 1567. https://doi.org/10.1016/S0140-6736(18)30895-X
Carazo-Rojas, E., Pérez-Rojas, G., Pérez-Villanueva, M., Chinchilla-Soto, C., Chin-Pampillo, J.S., Aguilar-Mora, P., Alpízar-Marín, M., Masís-Mora, M., Rodríguez-Rodríguez, C.E., Vryzas, Z. (2018). Pesticide monitoring and ecotoxicological risk assessment in surface water bodies and sediments of a tropical agro-ecosystem. Environ. Pollut., 241, 800–809. https://doi.org/10.1016/j.envpol.2018.06.020
Castro, M.F., Almeida, C.A., Bazán, C., Vidal, J., Delfini, C.D., Villegas, L.B. (2021). Impact of anthropogenic activities on an urban river through a comprehensive analysis of water and sediments. Environ. Sci. Pollut. Res., 28, 37754–37767. https://doi.org/10.1007/s11356-021-13349-z
Christofaro, C., Leão, M.M.D. (2014). Treatment of censored data in environmental studies. Quim. Nova, 37, 104–110. https://doi.org/10.1590/S0100-40422014000100019
CONAMA. (2005). Resolução CONAMA N° 357, De 17 De Março De 2005* (Rectified), Conselho Nacional do Meio Ambiente. Retrieved September 3, 2022 from https://www.icmbio.gov.br/cepsul/images/stories/legislacao/Resolucao/2005/res_conama_ 357_2005_classificacao_corpos_agua_rtfcda_altrd_res_393_2007_397_2008_410_2009_430_2011.pdf
CPRM. (2019). Integrated Hydrological and Hydrogeological Studies in the Urucuia Aquifer Region. Cia. Pesqui. Recur. Minerais. Retrieved September 21, 2022 from http://www.cprm.gov.br/publique/Hidrologia/Pesquisa-e-Inovacao/Estudos-Hidrologicos-e-Hidrogeologicos-Integrados-na-Regiao-do-Aquifero-Urucuia---SAU-5208.html
CPRM. (2004). Geological chart of Brazil to the millionth - SF 24. Cia. Pesqui. Recur. Minerais. Retrieved September 21, 2022 from https://dados.gov.br/dataset/sf_24_vitoria
Dadi, D., Mengistie, E., Terefe, G., Getahun, T., Haddis, A., Birke, W., Beyene, A., Luis, P., Bruggen, B. V. D. (2018). Assessment of the effluent quality of wet coffee processing wastewater and its influence on downstream water quality. Ecohydrol. Hydrobiol., 18, 201–211. https://doi.org/10.1016/j.ecohyd.2017.10.007
Damasceno, O.S. (2011). Fluvial geomorfology of the hidrographic basin of Ondas river. Paths Geogr., Retrieved March 3, 2022 from https://seer.ufu.br/index.php/caminhosdegeografia/article/view/16198/9111
Edwards, E.C., Nehra, A. (2019). Importance of Freshwater for Irrigation, in: Goldestein, M.I, Dellasala, A.D. (Eds.), Encyclopedia of the World’s Biomes. Elsevier Inc, Washington, pp. 22–28. https://doi.org/10.1016/B978-0-12-409548-9.11913-X
Finkler, N. R., Peresin, D., Cocconi, J., Bortolin, T.A., Rech, A., Schneider, V.E. (2015). Surface water quality data by principal component analysis. Rev. Ambient. e Agua, 10, 782–792. https://doi.org/10.4136/ambi-agua.1468
Freitas, P.L. De, Polidoro, J.C., Santos, H.G., Prado, R.B., Calderano, S.B., Greogoris, G., Manzatto, C.V., Dowich, I., Bernardi, A.C. de C. (2014). Identification and physical chemical characterization of sandy and loamy texture latosols from western region of Bahia state, Brazil. Cad. Geociências, 11, 83–93.
Friesl, W., Horak, O., Wenzel, W.W. (2004). Immobilization of heavy metals in soils by the application of bauxite residues: Pot experiments under field conditions. J. Plant Nutr. Soil Sci., 167, 54–59. https://doi.org/10.1002/jpln.200320941
Gonzalo, C., Camargo, J.A. (2013). The impact of an industrial effluent on the water quality, submersed macrophytes and benthic macroinvertebrates in a dammed river of Central Spain. Chemosphere, 93, 1117–1124. https://doi.org/10.1016/j.chemosphere.2013.06.032
Hair, J.F., Black, W.C., J.Babin, B., Anderson, R.E. (2009). Multivariate Data Analysis: A global perspective, seventh ed. Pearson Education, New Jersey.
Hanafiah, M.M., Yussof, M.K.M., Hasan, M., Abdulhasan, M.J., Toriman, M.E. (2018). Water quality assessment of Tekala River, Selangor, Malaysia. Appl. Ecol. Environ. Res., 16, 5157–5174. d https://doi.org/10.15666/aeer/1604_51575174
Harrison, S., Mcaree, C., Mulville, W., Sullivan, T. (2019). The problem of agricultural ‘ diffuse ’ pollution: Getting to the point. Sci. Total Environ., 677, 700–717. https://doi.org/10.1016/j.scitotenv.2019.04.169
Kaiser, H.F., 1960. The application of electronic computers to factor analysis. Educational and Psychological Measurement. 20, 141–151. https://doi.org/10.1177%2F001316446002000116
Khangembam, S., Kshetrimayum, K.S., 2019. Evaluation of hydrogeochemical controlling factors and water quality index of water resources of the Barak valley of Assam, Northeast India. Groundw. Sustain. Dev., 8, 541–553. https://doi.org/10.1016/j.gsd.2019.02.001
Kistemann, T., Schmidt, A., Flemming, H. (2016). Post-industrial river water quality — Fit for bathing again ? Int. J. Hyg. Environ. Health, 219, 629–642. https://doi.org/10.1016/j.ijheh.2016.07.007
Kumari, R., Sharma, R.C. (2019). Assessment of water quality index and multivariate analysis of high altitude sacred Lake Prashar, Himachal Pradesh, India. Int. J. Environ. Sci. Technol., 16, 6125–6134. https://doi.org/10.1007/s13762-018-2007-1
Libos, M., Filho, O.C.R., Zeilhofer, P. (2003). Modeling of Non-Point Pollution in the Cuiabá River Basin Based on Geoprocessing. Rev. Bras. Recur. Hídricos, 8, 115–135. https://doi.org/10.21168/rbrh.v8n4.p115-135
Lima, R.N. de S., Ribeiro, C.B. de M., Barbosa, C.C.F., Rotunno Filho, O.C. (2016). Study of point and diffuse pollution in the Funil reservoir hydropower plant contribution basin using spatially distributed Geographic Information System modeling. Eng. Sanit. e Ambient., 21, 139–150. https://doi.org/10.1590/S1413-41520201600100127676
Marques, R.C., Bernardi, J.V.E., Dórea, J.G., Brandão, K.G., Bueno, L., Leão, R.S., Malm, O. (2013). Fish consumption during pregnancy, mercury transfer, and birth weight along the Madeira river basin in Amazonia. Int. J. Environ. Res. Public Health, 10, 2150–2163. https://doi.org/10.3390/ijerph10062150
Michalski, Rajmund. (2009). Applications of Ion Chromatography for the Determination of Inorganic Cations. Critical Reviews in Analytical Chemistry – Crit. Rev. Anal Chem., 39, 230–250. https://doi.org/10.1080/10408340903032453
Miller, J.N., Miller, J.C. (2010). Statistics and Chemometrics for Analytical Chemistry, sixth ed. Pearson Education Limited, England.
Motevalli, A., Amir, S., Hashemi, H., Berndtsson, R., Pradhan, B., Gholami, V. (2019). Inverse method using boosted regression tree and k-nearest neighbor to quantify effects of point and non-point source nitrate pollution in groundwater. J. Clean. Prod., 228, 1248–1263. https://doi.org/10.1016/j.jclepro.2019.04.293
Mustapha, A., Aris, A.Z., Juahir, H., Ramli, M.F., Kura, N.U. (2013). River water quality assessment using environmentric techniques: Case study of Jakara River Basin. Environ. Sci. Pollut. Res., 20, 5630–5644. https://doi.org/10.1007/s11356-013-1542-z
Namugize, J.N., Jewitt, G., Graham, M. (2018). Effects of land use and land cover changes on water quality in the uMngeni river catchment, South Africa. Phys. Chem. Earth, 105, 247–264. https://doi.org/10.1016/j.pce.2018.03.013
Nargis, A., Habib, A., Harun-Or-Rashid, Harun, H.B., Islam Sarker, M.S., Jin, R., Liu, G., Liu, W., Al-Razee, A.N.M., Chen, K., Cai, M. (2021). Status of multielement in water of the river Buriganga, Bangladesh: Aquatic chemistry of metal ions in polluted river water. Emerg. Contam., 7, 99–115. https://doi.org/10.1016/j.emcon.2021.03.001
O’Shea, L. (2002). An economic approach to reducing water pollution: point and diffuse sources. Sci. Total Environ., 282–283, 49–63. https://doi.org/10.1016/s0048-9697(01)00938-x
Padowski, J.C. (2020). Freshwater: The Importance of Freshwater for Domestic Use, in: Goldestein, M.I, Dellasala, A.D. (Eds.), Encyclopedia of the World’s Biomes. Elsevier Inc, Washington, pp. 12–20. https://doi.org/10.1016/B978-0-12-409548-9.11912-8
Paula, R.H.R., Anjos, D.N. Dos, Freitas, P.H.G., Ribeiro, J.S. (2020). Effect of potassium chloride salinity on the emergence and initial growth of soybean seedlings. Pesquis. Agro, 3, 110–117. https://doi.org/10.33912/pagro.v3i1.664
Province, S. (2015). Factors Influencing Water Quality of Kwae-om Canal, Samut. Procedia - Soc. Behav. Sci., 197, 916–921. https://doi.org/10.1016/j.sbspro.2015.07.273
Rego, E.L., Boaventura, G.R., Leite, O.D., Araújo, D.F., Souza, A.S., Peres, L.G.M., Lima, A.S.C., da Silva, J.D.S., de Souza, J.R. (2021). Geochemical baseline of trace and major elements in sediments in the Rio de Ondas Basin (Bahia, Brazil). Environ. Earth Sci., 80, 107. https://doi.org/10.1007/s12665-020-09359-8
Rego, E.L. do, Boaventura, G.R., Bueno, M.A., Souza, A.S., Leite, O.D. (2014). Seasonality influence evaluation of metal concentrations in Ondas River Watershed – BA (Print). Br J Anal Chem., 12, 488–494.
Rego, E.L. Do, Gomes, D.B., Da Silva Júnior, J., Herok, D.L., Porto, E.D. de O., Leite, O.D. (2017). Evaluation of physic-chemical parameters of water quality on agricultural fields of western Bahia. Orbital - Electron. J. Chem., 9, 105–110. https://doi.org/10.17807/orbital.v9i2.880
Sales, R.J. de M., Araújo, J.A.F. de, Santos, S.H. (2014). Application of the streeter-phelps equations in rivers and receive point an nonpoint sources of pollutant, to evaluate the behaviour of the dissolved oxygen concentration. Periódico Eletrônico Fórum Ambient. da Alta Paul., 10, 112–123. https://doi.org/10.17271/1980082710122014908
Scopel, R.M., Teixeira, E.C., Binotto, R.B. (2005). Hydrochemical characterization of groundwaters in projected hydroelectric power plant areas – Taquari-Antas Basin/RS, Brazil. Quim. Nova, 28, 383–392. https://doi.org/10.1590/S0100-40422005000300004
Sediyama, T., Silva, F., Borém, A. (2015). Soybean: from planting to harvesting, 1^a. ed. UFV, Viçosa, MG.
Sharifi, Z., Akbar, A., Sinegani, S. (2012). Arsenic and Other Irrigation Water Quality Indicators of Groundwater in an Agricultural Area of Qorveh Plain, Kurdistan, Iran. Am-Euras. J. Agric. & Environ. Sci., 12, 548–555.
Sicsú, A.B., Lima, J.P.R. (2000). Fronteiras Agrícolas No Brasil: a Lógica De Sua Ocupação Recente. Nov. Econ., 10, 109–138.
Souza, A.T. de, Carneiro, L.A.T.X., da Silva Junior, O.P., de Carvalho, S.L., Américo-Pinheiro, J.H.P. (2021). Assessment of water quality using principal component analysis: a case study of the Marrecas stream basin in Brazil. Environ. Technol., 42, 4286–4295. https://doi.org/10.1080/09593330.2020.1754922
Takiyama, Luís Roberto, Silva, A. de Q., Costa, W.J.P., Nascimento, H. dos S. (2003). Water Quality of the Ressacas of the Basins of the Igarapé da Fortaleza and the Curiaú River, in: Takiyama, L.R., Silva, A.Q. (Eds.), Diagnosis of Surffalls in the State of Amapá: Basins of Igarapé Fortaleza and Rio Curiaú. CPAQ/IEPA e DGEO/SEMA, Macapá-AP, pp. 81–104.
Vittori, L. Di, Mazzoni, L., Battino, M., Mezzetti, B. (2018). Scientia Horticulturae Pre-harvest factors influencing the quality of berries. Scientia Horticulturae, 233, 310–322. https://doi.org/10.1016/j.scienta.2018.01.058
Vryzas, Z. (2018). Pesticide fate in soil-sediment-water environment in relation to contamination preventing actions. Curr. Opin. Environ. Sci. Heal., 4, 5–9. https://doi.org/10.1016/j.coesh.2018.03.001
Wang, J., Yang, C., He, L., Dao, G., Du, J., Han, Y. (2019). Meteorological factors and water quality changes of Plateau Lake Dianchi in China (1990–2015) and their joint influences on cyanobacterial blooms. Sci. Total Environ., 665, 406–418. https://doi.org/10.1016/j.scitotenv.2019.02.010
Wang, J., Zhao, J., Lei, X., Wang, H. (2018). New approach for point pollution source identification in rivers based on the backward probability method. Environ. Pollut., 241, 759–774. https://doi.org/10.1016/j.envpol.2018.05.093
Warren, J.G., Phillips, S.B., Mullins, G.L., Zelazny, L.W. (2006). Impact of Alum-Treated Poultry Litter Applications on Fescue Production and Soil Phosphorus Fractions. Soil Sci. Soc. Am. J., 70, 1957–1966. https://doi.org/10.2136/sssaj2006.0004
Xin, Z., Kinouchi, T. (2013). Analysis of stream temperature and heat budget in an urban river under strong anthropogenic influences. J. Hydrol., 489, 16–25. https://doi.org/10.1016/j.jhydrol.2013.02.048
Xu, X., Jiang, Y., Liu, M., Huang, Q., Huang, G. (2019). Modeling and assessing agro-hydrological processes and irrigation water saving in the middle Heihe River basin. Agric. Water Manag., 211, 152–164. https://doi.org/10.1016/j.agwat.2018.09.033
Zhao, C.S., Yang, Y., Yang, S.T., Xiang, H., Wang, F., Chen, X., Zhang, H.M., Yu, Q. (2019). Impact of spatial variations in water quality and hydrological factors on the food-web structure in urban aquatic environments. Water Research, 153, 121–133. https://doi.org/10.1016/j.watres.2019.01.015
Zhiltsov, S.S., Zonn, I.S., Semenov, A. V., Grishin, O.E., Markova, E.A. (2020). Role of water resources in the modern world, in: I., Z., S., Z., A., K., A., S. (Eds.), Handbook of Environmental Chemistry. Springer, Cham, pp. 13–29. https://doi.org/10.1007/698_2020_598

No competing interests reported.

Supportinginformation.docx

Download PDF

Journal Publication

published 09 Sep, 2023

Read the published version in Environmental Monitoring and Assessment →

Editorial decision: Major revision
31 Jul, 2023
Reviewers agreed at journal
12 Jun, 2023
Reviewers agreed at journal
09 Apr, 2023
Reviews received at journal
30 Nov, 2022
Reviewers agreed at journal
24 Nov, 2022
Reviewers agreed at journal
21 Nov, 2022
Reviewers invited by journal
21 Nov, 2022
Editor assigned by journal
16 Nov, 2022
Submission checks completed at journal
16 Nov, 2022
First submitted to journal
04 Nov, 2022

You are reading this latest preprint version

Spatio-temporal Study of Pollutants in the Rio De Ondas Hydrographic Basin, West of Bahia, Brazil Using Multivariate Analysis

Status:

Journal Publication

Version 1

Abstract

Figures

1. Introduction

2. Materials And Methods

2.1 Characterization of the study area and sampling sites

2.2 Sampling

2.3 Device and Conditions

2.4 Land use and occupation in Rio de Ondas Hydrographic Basin

2.5 Statistical Analysis

3. Results And Discussion

3.1 Change in ROHB's vegetation cover (Spatio-temporal analysis)

3.2 Descriptive statistical analysis of data and seasonality

3.3 Principal Components Analysis

4. Conclusion

Declarations

References

Additional Declarations

Supplementary Files

Status:

Journal Publication

Version 1