Influence of land use and occupation on the water quality of a microbasin in the southwestern Amazon

Water resource management in Brazil is constantly evolving, and greater knowledge about this resource allows better planning and more sustainable uses. In Brazil, the improvement of water resource management faces the difficulty of implementing the instruments of the National Water Resources Policy, such as classification of water bodies. Thus, to help 15 improve the water management instruments in the country’s northern region, the objective of the present study was to diagnose the influence of land use and occupation on the water quality of the Igarapé Nazaré microbasin. For this purpose, indirect methods of landscape analysis were applied based on the processing of remote sensing images in a GIS. For the water quality analysis, 10 collection points were selected in the watershed, with samples collected at each one in four periods (high water; HW/LW transition; low water; LW/HW transition). In the collected samples, 14 parameters were analyzed, namely: 20 temperature, pH, electrical conductivity; turbidity; water transparency and depth; dissolved oxygen; chlorophyll a, ammonia, nitrite, nitrate, total phosphorus and dissolved phosphorus; total coliforms and E. coli. The spatial analysis showed that the microbasin has about 84% anthropized territory, with emphasis on agriculture, and sources of pollution from industries, fish farming and domestic sewage. Parameter analyses showed high values of total phosphorus (0.005 27.55 mg.L-1), total coliforms (4,103 1,09,106 CFU) and E. coli (0 5.8,105 CFU), and low DO concentration (0.0 8.33 mg.L-1), below the 25 official limit established in all periods analyzed The water quality of the Igarapé Nazaré microbasin was found to suffer strong anthropic interference, requiring improvement of the sanitary infrastructure of city of Ji-Paraná, for maintenance of the watershed in class 2.


Statistical treatment
Analysis of variance (ANOVA) was used to evaluate the seasonal differences of the parameters considered in the case of parametric data, as determined by the Shapiro-Wilks test, and the Kruskal-Wallis test was applied in the case of nonparametric data. All the tests were performed with the free statistical program Past 3.25. The data were also submitted to multivariate analysis and principal component analysis (PCA), with the objective of verifying the relationship of the variables between 95 collection periods, to detect possible patterns. For the PCA, the data were arranged in a matrix, and the statistics were computed with the program XLSTAT for Excel. In all analyses, the results were considered significant with p ≤ 0.05.

Land use and occupation
The geoprocessing allowed obtaining information on the land use and occupation in the Igarapé Nazaré microbasin, as well as 100 identifying the main point sources of pollution, as depicted in Figure 2.
The main sources of effluents identified in the microbasin were associated with food processing (two sources), meat packing (five sources) and treated domestic sewage (two sources, one from a housing complex and the other from a penitentiary). All these sources released effluents rich in organic matter (Cardoso, 2015;Marçal & Silva, 2017;Lunelli, 2019), which when not treated adequately pose risks to water quality. Barbosa (2012), in analyzing the water quality of the Pirarara River in the 105 municipality of Cacoal, Rondônia, observe that the impairment of water quality was mainly due to release of domestic and industrial wastes, by identifying contaminants typical of these types of effluents at the most degraded points.
Pasture for grazing livestock (figure 3) is the main land use in the microbasin (73.3 km2), followed by native or secondary vegetation (17.4 km2) and urban area (11.3 km2). According to the Brazilian Institute of Geography and Statistics (IBGE, 2017), animal husbandry is one of the main economic activities in the state of Rondônia, which occupies sixth place in the 110 national ranking of cattle herd size (14 million head). This explains the large number of slaughterhouses (n =5) in the microbasin.
Besides the domestic and industrial point sources of pollution identified, there are also aquaculture sources (classified as water bodies, with area of 3.18 km2) in the microbasin. This is a rapidly expanding economic activity in Rondônia (Pereira et al., 2019). This activity uses large amounts of water and produces effluents rich in organic matter, similar to domestic waste, 115 posing a water quality risk if not adequately managed (Macedo & Sipaúba-Tavares, 2010;Assunção et al., 2016).
Besides the point sources, there are several diffuse sources of pollution. Their origin is usually difficult to pin down and the quantities are hard to calculate due to the many sources and causes, such as soil erosion, runoff containing particulate matter such animal manure, soot and ash from land clearance by burning, among other factors, as discussed by Sodré (2012), Of particular note is that according to the Sewage Atlas of the National Water Agency (ANA, 2017), only 18.6% of the domestic 120 sewage in Ji-Paraná is submitted to some type of treatment, and the flow of raw sewage into Igarapé Nazaré was 29.1 L/s that year.
Studies of other microbasins in the state, such as Ji-Paraná (Nunes et al., 2018) and Ouro Preto do Oeste (Prestes et al., 2018), as well as of the city of Manaus in the state of Amazonas, have shown proportions of anthropized areas similar to those found by us, with values of 62.4% (Manaus), 85.94% (Ouro Preto do Oeste) and 95.5% (Ji-Paraná). 125

Relationship of water quality and land use and occupation
Article 15, § 2, of Resolution 91/2008 from the National Water Resources Council specifies that if sufficient information is not available to classify a surface freshwater body, class 2 can be adopted. Therefore, the parameters defined by Resolution 357/2005 from the National Environmental Council (CONAMA) for class 2 are used as reference values for comparison of the results found in this study. 130 The average water temperature for the four periods analyzed (Figure 4a) was 30 °C. That value is similar to those found in studies of microbasins near the Igarapé Nazaré microbasin. Trindade et al. (2019) studied five water bodies in Ji-Paraná and found values ranging from 25 °C to 30 °C in the HW/LW transition period. In turn, in the Igarapé Mangueira microbasin, also located in the municipality of Ji-Paraná, the temperatures varied between 26 °C and 32 °C in the HW and LW periods according to Sousa et al. (2019). 135 Figure 4a also shows that the maximum temperature in the HW period was recorded at P5 (37.9 °C). That result is certainly related to factors like the absence of a permanent protection area and the influence of the land use and occupation in the surrounding area, since this point receives large amounts of effluents from two meat packing plants and part of the urban area of Ji-Paraná, as we observed in the field.
The ANOVA results did not identify significant changes in temperature over the four seasonal periods (F = 2.63; p = 0.065). 140 According to Nobre et al. (2009), the mean air temperature in the Meridional Amazon region, where Rondônia is located, varies little, ranging between 24 °C and 26 °C, with annual amplitude that can reach 3 °C to 4°C. Table 2 contains the values of water depth and transparency. For P4, the transparency values are very low compared to the water depth, demonstrating the receipt of organic and inorganic particulate matter of natural or anthropic origin. P4 is the point where a small tributary flows into Igarapé Nazaré, and is near a point of discharge of effluents from a meat packing plant. The 145 effluents from slaughterhouses contain a high concentration of organic matter, which can degrade the water quality if not treated properly (Cardoso, 2015).
The ANOVA results did not show significant variations in the averages of the periods for water depth and transparency (F = 1.17 and 1.68; p = 0.34 and 0.19). Siqueira et al. (2012), studying the water quality of the Parauapebas River (Pará State), reported that periods with less rainfall tended to have greater water transparency due to the lesser leaching of organic matter. 150 However, for P4 we observed the opposite, which can be explained by the entrance of wastewater from the nearby meat packing plant, so the smaller flow in the low water period suffers a greater impact of the organic load in the effluent received.
Phosphorus is an element that occurs naturally in water, but in high amounts it is considered a pollutant, mainly in still surface water, where it provokes eutrophication, increasing the population of algae and plants, which can consume the dissolved oxygen in the water and kill fish (Lamparelli, 2004;Klein & Agne, 2012).
The results for phosphorus (Figures 4,b and c) in all periods and at practically all the sampling points were above the threshold of 0.1 mg.L-1 for class 2 lotic water bodies set by CONAMA Resolution 357/2005. Silva et al. (2019), in analyzing the total phosphorus concentration in urban springs in Ji-Paraná, observed that of the 8 points analyzed, the concentration in the water of all of them was above that value in all periods analyzed. Their results together with ours demonstrate that anthropic activities degrade the water quality of streams in urban areas even at their origin. 160 Points P1, P2 and P3, located in a relatively well preserved rural APP, were the only ones to present concentrations below 0.1 mg.L-1 in all the periods. The others (P4 to P10) all had levels greater than 0.1 mg.L-1, with highlight on P4, which in the low water period had total phosphorus concentration of 27.55 mg.L-1. The highest concentrations of total phosphorus were found in the low water period at all the points sampled.
The lowest concentrations of dissolved phosphorus were also found at points P1, P2 and P3, while the highest concentrations 165 were found at P4 (2.06 mg.L-1) and P7 (0.12 mg.L-1). These values can reflect the presence of aquatic flora, since this is the main form of phosphate assimilated by aquatic plants (Esteves, 2011).
Factors such as the presence of slaughterhouses and food processing plants (which occupy 10.66% of the Igarapé Nazaré microbasin, located around the urban perimeter), along with low flows and volumes of water in the stream and entry of varied effluents are likely reasons for the high levels of phosphorus. 170 Therefore, it is extremely important to adequately treat the effluents generated by these industrial concerns, since the organic load is high (biological oxygen demand of up to 4,200 mg.L-1 according to Aguilar, 2002). Excess phosphorus can cause serious harm to aquatic ecosystems (Thebaldi et al., 2011;Orssatto et al., 2018;Lunelli et al., 2019).
With respect to nitrogen compounds, in the case of nitrate (Figure 4f), the majority of the points had concentrations higher than the limit specified by CONAMA Resolution 357/2005, of 10 mg.L-1 in the HW and HW/LW periods. The concentrations 175 of ammonia (Figure 4d) were also above the regulatory threshold of 3.7 mg.L-1/pH < 7.5 in the low water period (varying from 0.005 to 4.77 mg.L-1). In turn, nitrite was below the value of 1.0 mg.L-1 set by the Resolution during all the sampling periods. The presence of high concentrations of these nitrogen compounds can seriously harm aquatic flora and fauna, due to eutrophication, increased water toxicity and proliferation of algae, among other negative effects, as well as posing risks to human health (Zoppas et al., 2016;Reismann et al., 2017). 180 Of particular relevance during the low water period, when the ammonia concentrations were above the limit of 3.7 mg.L-1 (CONAMA Resolution 357/2005), there were episodes of fish mortality in Igarapé Nazaré and one of its tributaries, Córrego Bonzinho, suggesting that one of the causes of the fish die-off was the high concenration of ammonia in the water column.
One of the factors causing the high concentration of nutrients in the microbasin is likely the presence of many fish farming nitrite is rapidly consumed by processes of nitrification, denitrification and anammox, reducing the concentration in the water.
In contrast, the ammonia and nitrate concentrations were distinct among the four periods (Figures 5c and 5e). For ammonia, there was a significant increase in the LW period. According to Esteves (2011), high ammonia concentrations indicate recent contamination by effluents because of the rapid decomposition of ammonia to nitrite and then to nitrate in aquatic environments. For nitrate, the highest concentrations were found in the HW and HW/LW, indicating a longer time interval for 195 contamination.
With respect to the dissolved oxygen levels (Figure 5a), only at P2 was the concentration higher than the limit set by CONAMA The sampling points located in the urban area of Ji-Paraná (P4 to P8), in the periods HW, HW/LW and LW, had values below the regulatory limit, with the exception of P7 in the period HW/LW (5.33 mg.L-1). In the LW/HW period, there were higher 205 DO concentrations, which can be explained by the strong rainfall on the collection day, contributing to the aeration of the water.
As can be observed in Figure 5a, in periods HW/LW, LW and LW/HW, the water at P4 was anoxic, indicating low water quality of the tributary of Igarapé Nazaré. This result can be correlated with the turbidity and electrical conductivity results found for the same point (Figures 5 b and c), which had high values in the same periods. According to Esteves (2011), situations 210 like this can indicate a high load of organic matter in a water body, increasing the consumption of oxygen and making the water improper for drinking and certain other uses.
The low DO values are directly associated with the uses of the basin. Zuffo et al. (2013), who analyzed the main water bodies in Rondônia, associated the low DO observed with stock breeding in the region (the main use of the Igarapé Nazaré microbasin, occupying 70% its area), which was responsible for the high presence of organic matter. 215 Besides this, the release of organic matter in household wastewater in the urban area directly affected the DO, which was consumed by decomposition of that matter. That fact was also observed by Sousa et al. (2019)  High chlorophyll a concentration in water bodies can indicate risks of eutrophication and also the presence of cyanobacteria, which can generate contamination by cyanotoxins, degrading the water quality (Marino, 2017;Sousa et al. 2018).
Because the water from Igarapé Nazaré is used to supply industrial establishments, problems of algal blooms can compromise the water quality for these uses. Mariano and Nascimento (2018), studying water treatment for use by a beverage maker, found that the treatment costs increased due to the poor water quality, generating economic losses. 240 The presence of bacteria of the coliform group was also investigated ( Figures 5 d and e). Of this group, Escherichia coli are present in the feces of warm-blooded animals, so their presence in water bodies is an indicator of pollution by untreated sewage (Brasil, 2013).
The densities of total and fecal coliforms found indicated the strong presence of this group in the samples collected in the Igarapé Nazaré microbasin, as can be seen in Figures 5d and 5e. 245 In periods HW and LW, the presence of E. coli was not found at points P2 and P3, and P1 and P3, respectively. The water at all the other points had values above the limits stipulated in CONAMA Resolution 357/2005, of 1,000 coliforms per 100 mL of water. With respect to total coliforms, no point presented values greater than that threshold.
The high presence of E. coli, as discussed by Silveira et al. (2018), indicates the presence of fecal contamination. According to Machado et al. (2019), this parameter should be considered in defining the classification of water bodies for various uses.The 250 ANOVA results did not show differences in the average values between the periods for the biological parameters chlorophyll a, E. coli and total coliforms (F = 0.52; 0.41; 0.53; p = 0.67; 0.74; 0.67). Table 3 Figure 6 shows the projection of the samples and the variables on the axes. 260 Axis 1 explained 38.86% of the variance of the data and axis 2 explained 13.01%, for a total of 51.87%. The constitution of the axes presented a distribution of points without connection to the sampling periods, indicating a strong anthropogenic influence in the microbasin. Points P1, P2 and P3 were related to high concentrations of DO as a result of proximity to industrial establishments and the urban area of the municipality. In all the periods analyzed, those points were concentrated in the axis of greater water transparency. The other seven points were distributed on the axes with higher concentrations of nutrients, 265 turbidity, electrical conductivity and E. coli. In particular, P4, for all the periods, was fixed on the axis with highest values of total phosphorus and coliforms, reflecting the high level of pollution.
These results demonstrate that the points with greatest degradation are directly associated with the land use in the urban area of the microbasin and industrial activities. These results are similar to those reported by Finkler et al. (2015), who found that the most significant parameters in the water quality variation in watersheds in the municipality of Caxias do Sul were related 275 to anthropic activities and lack of treatment of domestic and industrial effluents, directly causing degradation of the water quality.

Final Considerations
The water quality in the Igarapé Nazaré microbasin was strongly influenced by anthropic activities, mainly in the urban area of the municipality of Ji-Paraná, with values above the thresholds specified in CONAMA Resolution 357/2005 for class 2 280 water bodies for nutrients and coliforms. The sampling points located in the rural area, especially those near a well-preserved APP, presented the best water quality.
Of the parameters analyzed, only the concentration of nitrite was within the class 2 limit at all the points analyzed in all four periods. Turbidity also was consistently below the threshold with the exception of P4. The levels of pH and chlorophyll a also were only outside of class 2 at one point in all the periods (2.5% of the samples). Ammonia was above the limit only in period 285 LW at 40% of the points sampled. Nitrate was above the limit at 100% of the points in the period HW/LW and 66% of them during HW. For total phosphorus and DO, 70% of the samples were above the threshold. And for total coliforms, 100% of the samples were above the limit for class 2. For the Igarapé Nazaré microbasin to remain in class 2, it will be necessary to prepare an action plan to improve the sanitary infrastructure of the municipality of Ji-Paraná, along with better monitoring of the industrial effluents discharged into the water 290 bodies, associated with better oversight of rural effluents, mainly from fish farming.

P2
Pasture area without permanent preservation area (APP).

P3
Transition between rural and urban area, with partially preserved APP.

P4
Tributary on the right bank of Igarapé Nazaré, APP under restoration, near a slaughterhouse, located in an urban residential zone.

P5
Urban area with preserved APP.

P6
Tributary on the right bank of Igarapé Nazaré, absence of APP, near slaughterhouses.

P7
Transition between rural and urban, partially preserved APP, near a slaughterhouse.

P8
Transition between rural and urban, partially preserved APP and presence of a source of raw sewage.

P9
Flooded area, preserved APP, near the mouth of Igarapé Nazaré.

P10
Tributary on the right bank of Igarapé Nazaré, absence of APP, near a slaughterhouse.