Domestic Wastewater for Forage Cultivation in Cerrado Soil

Fertigation of agricultural crops that are not directly used in human food, with domestic wastewater is a viable alternative for the sustainable use of water resources. The development of agricultural practices that provide high productivity with the sustainability of agroecosystems has been a great challenge. Thus, our aims were to use of domestic wastewater in the planting of Brachiaria brizantha cv Marandu, as an alternative for animal feed production in Cerrado soils, and to study the physical-chemical and microbiological impacts of the fertigation. These impacts were evaluated, respectively, by physical-chemical indicators content and diversity of nitrogen fixing bacteria (NFB) and arbuscular mycorrhizal fungi (AMF) in the DGGE profile. The NPK contents of the wastewater were used to determine the five fertigation managements (M1 to M5). M1 and M2 managements had no wastewater and M3 to M5 contained 20, 40 and 60% of NPK from the wastewater. The managements in a completely randomized design with 20 plots and 4 replicates were distributed. Soil samplings prior to fertigation and at the end of the experiment were performed. Leaf biomass productivity was determined in three different grass cuts. After fertigation, changes in physical-chemical indicators and in the viable microbial cells counts were observed. The NPK of wastewater increased the abundance of NFBs and AMFs. Leaf biomass productivity per hectare was directly proportional to NPK concentration. In addition, wastewater did not alter the nutritional composition of Marandu grass. Therefore, the fertigation with domestic wastewater showed to be a viable and promising alternative for reuse of this water in Cerrado soil for animal feed production.


Introduction
The water crisis and new discussions on the sustainable use of water should contribute to intensify studies of wastewater reuse in agriculture.Thus, the use of domestic wastewater to irrigate crops, which are not used directly in human food, can be a viable alternative (Silva et al., 2016).
Brazil has the greatest biodiversity of the planet (Dias-Filho, 2014).The Cerrado is the second largest Brazilian biome.It has grassland, savannic and forest physiognomies that dominate the Brazilian Midwest (Sano et al., 2008).This biome is a biodiversity hotspot, because of its species richness and high endemism level (Myers et al., 2000).
In the 1970s, the Cerrado was subject to large financial investments for pastures formation (Andrade et al., 2005).In this period, the replacement of fat grass by more productive grasses, mainly, Brachiaria (B.decumbens, B. humidicola and B. brizantha) was done.In addition, the new agricultural areas formation and the abandonment of degraded areas have favored the growth and adaptation of invasive plant species (Dias-Filho, 2014).Thus, the development of agricultural practices that provide high biomass yields with sustainability of agroecosystems has been a great challenge.
Our objectives were to use of domestic wastewater in the planting of Brachiaria brizantha cv Marandu, as an alternative for animal feed production in Cerrado soils, and to study the physical-chemical and microbiological impacts of the fertigation. jas.ccsenet.

Experi
The exper altitude of At this exp wastewate

Soil Pr
The soil ty Prior the e removed b The liming saturation application The dolom power (NP
All managements had liming and water from an artesian well.The water volume of each management was in the ratio between the water balance of Palmas-TO city and NPK content (Table 1).Note.NPK: nitrogen, potassium, and phosphorus.

Planting and Management of Grass
B. brizantha due to resistance to long dry periods, good adaptation in Cerrado soil, high regrowth capacity and tolerance to the leafhopper was used (Soares Filho et al., 2002).
The seeds were obtained on the market of Palmas-TO city.They had 60.3% of purity, 80% of germination rate and of 48.24% cultural value.Thus, in the planting, 1.75 kg/ha of seeds were used.Embrapa (1984) recommends 1.5 to 2 kg of viable seeds per hectare.The planting took place on June 24, 2015, with the seeds applied to the haul.
In the plantation area were performed 20 plots with nine m 2 (3 m × 3 m).The management of the plots was determined by lottery.The vertical dimension of the plots was 1.00 m, because of the maximum depth of soil samples.
The effective root depth of 43 cm was used to determine the total water blade for the grass.
In fertilization, phosphorus (P) and potassium (K) contents were applied according to Alcântara and Bufarah (1999).In this step, 627.66 kg of superphosphate ha -1 and 119.37 kg ha -1 of potassium chloride were applied.The urea was placed as a nitrogen source in a dosage of 150 kg ha -1 (Lopes et al., 2013).

Total Water Capacity in the Soil
The total water capacity (TWC) in the soil depends on the effective roots depth (Z).In grazing, the TWC is of 30 to 100 cm (Klar, 1991).TWC for B. brizantha cv.Marandu, in middle texture soil, with pressures from 1/5 to 15 atm was of 43.1357 cm (Cunha et al., 2010).
Water availability factor in the soil for forage is of 0.3 to 0.7 (Bernardo et al., 2008).Thus, the actual water capacity (AWC) in the soil of 25.8 mm was calculated, respectively, with 51.6 mm and 0.5 of TWC and water availability factor.

Irrigation Management of Crop
The climate of the region is humid and subhumid (C2wA"a") with a water deficit in the winter, annual evapotranspiration between 1.5 and 1.60 mm, mean annual temperature of 27.5 °C, and relative humidity of 80% (INMET, 2013).
The wastewater application was on August 12, 2015, when the grass was about 20 cm of height.In this assay, thirty fertigations were done in the morning and afternoon periods.These applications were done with a watering can (10 liter) on the leaves.
About 6000 students, teachers and visitors from the CEULP/ULBRA campus produce the domestic wastewater.
For transportation and storage of this wastewater was used a pump (2 hp), 250 meters of hose of 50 mm in diameter and a water box of 1000 liters.
The comNPK was ground and diluted in the water of an artesian well or in the wastewater, because of the small amount.

Sampling and Characterization of the Wastewater
The physical, chemical and microbiological analyses were using three samples (0.5 L) of the wastewater (Apha, 2005).The electrical conductivity and pH were determined, respectively, by instrumental and potentiometric methods.The iodometric and oxidimetric methods were used, respectively, for the biochemical oxygen demand (BOD) and the chemical oxygen demand (COD).The total solids, nitrogen, and minerals were determined in analytical balance, Kjedahl method, and spectrophotometry.Total and faecal coliform counts were determined by the most likely number (MPN) method.

Sampling and Characterization of the Soil
The soil sampling was done in two periods with a Dutch auger (Raij, 2001).The first one was carried out in April 2015, before planting the grass and applying the wastewater, in four points (A1 to A4) and in four soil depths (0-10, 10-20, 20-30 and 90-100 cm) (Figure 1).These samples were identified as A1SD1 to A1SD4, A2SD1 to A2SD4, A3SD1 to A3SD4, and A4SD1 to A4SD4.After the randomization of the experiment, the A1 point was within the M4 management.A2 was in M3 management and the A3 and A4 points were, respectively, in the M1 and M2 managements (Figure 1, Table 1).
The second sampling was performed at the end of the assay, in February 2016.Twelve samples were obtained for each fertigation management (M1 to M5), with four replicates and three soil depths (0-10, 10-20 and, 20-30 cm).
Twenty grams of soil were placed in dark plastic bags and added in a Styrofoam box containing dry ice.These samples were used for physical, chemical and microbiological analyses.The physical-chemical indicators of soil were determined according to Standard Methods (APHA, 2005).

Analysis of the Soil Microbiota
Viable microbial cells and DGGE profile were performed as described in Carvalho et al. (2018).

Sampling and Characterization of the Plant
The plant cuts were carried out on November 13, 2015, January 4, 2016 and February 23, 2016.
The first cut occurred after 140 days of sowing the grass seeds.The planting occurred in the dry season.So, the germination was observed after 30 days.The second and third cuts occurred, respectively, after 50 days and 100 days of the first cut.The cuts were performed at a height of about 10 cm from the soil.
After the plant cuts with a pruning shears, the green mass was determined in an analytical balance.After the green mass determination, the samples were placed in an oven at 65 o C, with forced ventilation for 72 hours.
After cooling to room temperature (25±5 o C), the dry matter in the air (DMA) was determined in analytical balance.
The DMAs were submitted to milling in a Willey mill to determine the nutrients (N, P, K, Ca, Mg, S, Zn, Cu, Fe, Mn, Na, Co, Mo and B), etheric extract, acid detergent fiber (ADF) and neutral detergent fiber (NDF).
After the nitroperchloric digestion of DMAs (Tedesco et al., 1985), the elements contents were determined by the Kjeldahl method (nitrogen), spectrophotometry (phosphorus), flame emission photometry (potassium and sodium), and plasma emission spectrophotometry (other minerals).
The actual dry matter (ADM) content was determined in an oven at 105 °C until the stabilization of its weight using DMA.The ash content was determined in muffle at 5500C using the MSE.
The ash content was determined in muffle at 550 o C.
All these analyses were performed according to the methodology of the Association of Official Analytical Chemists (1990) and Embrapa (1999).

Statistical Analyses
The assay was performed in a completely randomized design (CRD) with five fertigation managements and four replicates.
Statistical analyses of soil and grass samples were performed, respectively, by CRD and CRD in a 5 × 3 factorial scheme.The factorial was the five doses of wastewater (Table 1) and the three cuts.The comparison between the viable microorganisms counts was done by analysis of variance (ANOVA) and Tukey's test.The comparison between dry matter yields were performed by ANOVA and a regression analysis.These analyses were performed using the free trial version of the Minitab 17 software (2016) at the 5% level.
The clusters and main component of the soil and plant samples were performed in Sigma Plot Software (2016).
The DGGE profiles were analyzed by the unweighted pair group method with arithmetic mean (UPGMA) and Jaccard similarity index in the Bionumerics software (Version 5.1).The similar bands were those with 0.5% of significance level by the post-hoc Bonferroni test.

Wastewater Characterization
The chemical composition of the domestic wastewater had higher N, P and K levels than nutrients that shows its potential in the fertigation (Table 2).However, the management of this wastewater over the soil is important, because of high ammonia level (Table 2).This compound has a high evaporation rate in the soil.The total N, P and K contents (Table 2) were used to determine the wastewater proportion of the fertigation managements (Table 1).
The domestic wastewater composition can vary according to the collection site, climate and the economic and social situation of the population (Kong et al., 2015).
The N content of the wastewater was more than 70 mg L -1 (Table 2).The P and total nitrogen contents of crude domestic sewage were, respectively, 4 to 12 and 20 to 70 mg L -1 (Metcalf & Eddy, 2003).
The K content of the wastewater of the CEULP/ULBRA was lower than 30 mg L -1 .This value was also obtained in Pescod (1992).
The BOD and COD were lower than the values described by Metcalf and Eddy (2003), and Araújo et al. (2010) which were 110 and 800 mg L -1 for raw sewage.However, the BOD/COD ratio of our wastewater was similar to the obtained for domestic sewage by Simões et al. (2013).
The sodium content was lower than 40 mg L -1 described by Von Sperling (2006).
The basic pH of the wastewater and liming in the soil contributed to the acidity correction (Table 2).Seed germination and seedling growth depend on the acidity of the soil.
The wastewater had of total and fecal coliforms counts smaller than the amount observed in other domestic effluents (Araújo et al., 2010).However, this count was similar to the obtained in domestic sewage from a septic tank (Fonseca, 2007).

Physical-Chemical Indicators Characterization of Soil Before Fertigation
The sand, silt, and clay amount did not show differences (p < 0.05) between the soil depths (Table 3).Thus, this soil is sandy-loam with middle texture.
Only the pH, base saturation and clay not reduced its values with soil depth (Table 3).In the surface, the pH was more acid than at depths above 20 cm.This result may be due to the higher organic matter and minerals in the 0-10 cm than other depths.In addition, there was a positive linear correlation (R 2 = 0.8795) between base saturation and pH in soil depths.According to Ronquim (2010), the pH of the Cerrado soil is acid and depends on the sampling points, time and soil depth.Soils from six no-till crops also had acid pH more on the surface than other depth (Nicolodi et al., 2008).
In 0-10 cm, the sampling points had different distributions of the physical-chemical indicators, with few overlapping of them (Figure 2).These results show the heterogeneous character of the soil.The unequal distribution of nutrients in the soil has also been reported in other studies (Pavinato & Rosolem, 2008;Nicolodi et al., 2008;Ronquim, 2010).We observ to A2 than nt was more sim nts (Figure 2).

These values and 4B
). Note.* The values in the table represent the average of four replicates.These values were compared by analysis of variance (Anova) followed by the Tukey test at 5% probability (p < 0.05).
The organic matter contents were higher in the fertigation management than in the management without wastewater (Table 4).Thus, infiltration of the effluent into the deeper layers may have occurred.The organic matter is an important factor for soil fertility (Alderson et al., 2015;Malafaia et al., 2016).However, there was no significant difference (p < 0.05) in phosphorus and sodium concentrations between managements with or without wastewater (Table 4).These results are important, because it shows a low risk of salinization and eutrophication of the soil by the domestic wastewater.
The soil depth and the microbial group influenced the counts of viable microorganisms before fertigation (Table 5).Regardless of depth, the bacterial community was larger than of filamentous fungi.This prevalence of bacterial community in relation to fungi was also observed in the soil under native vegetation in the southern region of Brazil (Rech et al., 2013).According to these authors, the soil microorganism count varies with the techniques, soil depth and culture medium.However, we observed no difference (p < 0.05) in the counts of viable bacteria between the sampling points and depths of 0 to 10 cm and 10 to 20 cm (Table 5).Therefore, the technique and/or culture medium influenced the microbial cells counts more than depth.
The bacterial and fungal cell counts prior to fertigation are similar to the results obtained in other studies with soil samples (Rech et al., 2013).
Only the bacterial community was observed at all depths (Table 5).Therefore, their great dispersion in the soil is due to present aerobic, anaerobic, and nitrogen fixation species (Dunbar et al., 2002, Vázquez et al., 2000, Zehr et al., 2003).In addition, the microbial activity of six clayey soils was higher in the surface than in the depth of 60 cm showing that the bacteria are present in several soil depths (Vale Júnior, 2011).
Actinomycete colonies were not observed only at depths of 90 to 100 cm (Table 5).The absence of microbial colonies in this soil depth may be due to the nutrient and oxygen limitation.Fungal colonies were observed in 20 to 30 cm and 90 to 100 cm.Furthermore, the communities of these microorganisms are higher in the rhizosphere than other soil depths (Smith et al., 2010, Neeraj, 2011, Silva et al., 2002, 2011).
Viable microbial cells counts after fertigation were higher than before planting of B. brizantha (Tables 5 and 6).Simões et al. (2013) observed a linear increase in microbial respiration with the wastewater dose applied in the planting of castor bean.
Regardless of fertigation, there was a significant increase (p < 0.05) in the viable microbial cells counts after planting of B. brizantha (Tables 5 and 6).Thus, the wastewater did not change the viable microbial cell count (Table 5).However, the fungal colonies in the depth of 20-30 cm were observed after fertigation that may be due to the presence of roots and availability of water and nutrients (Table 6).The fungi form mycorrhizal interactions with the roots (Pozo & Azcon-Aguilar, 1997;Oehl et al., 2011).

Microbial Diversity by DGGE Profile Before of Fertigation
The amplification of the nif H gene in soil depth of 90 to 100 cm was not observed (Figure 3).Although, viable bacterial cells had been obtained in this depth.
The DGGE band profile showed a large variety of NFB, with the higher number of bands in 0 to 10 cm and 10-20 cm than other depths (Figure 3).This result can be due to the high amount of BFN cells in the rhizosphere (Silva et al., 2011).Furthermore, Da Silva (2012) showed great diversity of nif H gene in Cerrado soil by DGGE.This same author identified a predominance of this nif H gene in Actinomycete.
Similar to the observed in the soil physicochemical analyses, two main groups were observed in the UPGMA dendogram (Figure 3).The A1 point has 69% of similarity with point A2, A3 and A4 points has 89% of similarity.Therefore, the distance between sampling points influenced the bacterial diversity and shows once again the influences in the BFN Figure 3.
In    Fertigation managements (M3 to M5) had higher leaf dry mass production per hectare (ha) than the managements without wastewater (Figure 7).
In the first cut, the dry mass was the average of each fertigation management (Figure 7).In the other cuts, these values were the mean of both cuts, as there was no significant difference between them (p < 0.05) (Figure 7).
The dry mass productivity of B. brizantha was directly proportional to the wastewaterNPK (Figure 7).Furthermore, wastewaterNPK had higher leaf dry mass productivity than comNPK (Figure 8).Therefore, comNPK is less available for plant than the wastewaterNPK.
Linear equations are important for leaf dry mass productivity studies using NPK wastewater concentrations.The positive linear effect on leaf dry mass with nitrogen dose was also observed in the planting of Marandu grass (Alexandino et al., 2003) and Brachiaria decumbens cv.Basilisk (Maranhão et al., 2010).
The dry mass productivity of the second cut was similar to the third cut (Figure 7).This result shows the regrowth of the Marandu grass with the fertigation that can contribute to reduce the frequency of planting of this grass.The leaf dry mass increases with reduced cut-off interval and regrowth decreases with the cuts (Maranhão et al., 2010;Alexandino et al., 2003).
The iron, cobalt, and etheric extract content of the leaf biomass did not differ significantly (p < 0.05) between the second and third cuts (Table 7).
In the last cut, we observed a significant reduction (p < 0.05) in nitrogen, crude protein, potassium, zinc, copper, and manganese content (Table 7).However, there is a significant increase (p < 0.05) in phosphorus, calcium, magnesium, sulfur and, ADF content (Table 7).Therefore, the cuts had an influence on the nutritional composition of Marandu grass.
The calcium and NDF levels were different between without management (M1) and managements with NPK (M2 to M5) (Table 7).These managements had higher phosphorus, magnesium, and NDF content than the M1 management.Therefore, the domestic wastewater did not alter the nutritional composition of Marandu grass (Table 7).
The crude protein content, in the fertigation managements, was similar to the obtained by Serafim (2010).This author shows a variation of 11.31 to 13.81% in the crude protein content with the use of swine wastewater in the Marandu grass growth.
The etheric extract levels did not exceed the limit of 6% of the diet of ruminants (Souza et al., 2009).These levels are lower than those reported by Mari (2003).This author obtained values etheric extract in Marandu grass from 2.7 to 1.9% of dry mass.
The phosphorus concentration in leaf biomass was of 1.03 to 1.51 kg -1 (Table 7).This concentration was similar to the fodder plants (Malavolta, 1987).
The aging of the grass influenced significantly (p < 0.05) in the Cu concentration (Table 7).We observed a decrease in Cu concentration with time.Concentration of this nutrient was of 3.00 to 9.00 mg kg -1 .According to Malavolta (1987), the Cu level in grass is of 6.0 mg kg -1 .
The Fe level was high in leaf biomass, because the levels of this element in grass are of 180 to 250 mg kg -1 (Malavolta, 1987).However, the iron requirement for ruminants is of 30 to 100 mg kg -1 (McDowell, 1999).Thus, our forage would meet this requirement.
The ash content did not differ with the cuts (p < 0.05).However, the managements with wastewaterNPK had lower ash content than other treatments (Table 7).The ash content obtained in this study is similar to the observed in Marandu grass cultivated at different times of the year in the state of Piauí/Brazil (Rodrigues Júnior et al., 2015).

Conclusions
In this study, we had the following conclusions:  Fertigation is a viable alternative for the use of domestic wastewater in agriculture in the Cerrado soil, which may represent a reduction in the wastewater amount discarded in the water bodies.

Figure
Figure Cerrado.T

Figure
Figure 2. P Figure 4

Figure 3
Figure Note.CP: Crude protein, EE: Etheric extract, ADF: Acid detergent fiber, NDF: Neutral detergent fiber.The upper and lower case letters in the columns indicate, respectively, the statistical comparison between cuts and treatments.Thus, the same letters in the same column indicate that there was no significant difference by the Tukey test at 5% significance (see annexes).Statistical analyses were carried out using the free trial version of the Minitab 17 software (2016) available at http://www.onthehub.com/minitab.


Fertigation management with up to 60% NPK from wastewater also validates this alternative;  Changes in soil physical and chemical properties after fertigation depend of the time, the type of crop, and the characteristics of the soil and effluent used;  The use of fertilizers from domestic wastewater contributes positively with the abundance of NFB and AMF in the soil;  NPK of wastewater is more available to the plant than commercial NPK;  Fertigation with domestic wastewater has a low risk of salinization and eutrophication of the soil.

Table 1 .
Fertigation managements with or without domestic wastewater that were used in the planting of Brachiaria brizantha cv Marandu in Cerrado soil

Table 2 .
Physical and chemical indicators and coliform counts of wastewater

Table 4A .
Physical-chemical indicators of soil samples after the fertirrigation of Brachiaria brizantha cv Marandu in Cerrado soil Note. * The values in the table represent the average of four replicates.These values were compared by analysis of variance (Anova) followed by the Tukey test at 5% probability (p < 0.05).Table4B.Physical-chemical indicators of soil samples after the fertirrigation of Brachiaria brizantha cv Marandu in cerrado soil

Table 5 .
Viable microbial cells counts from the soil before planting of Brachiaria brizantha cv Marandu and wastewater utilization

Table 6 .
Viable microbial cells counts after the planting of Brachiaria brizantha cv Marandu