Effects of Management Practices and Land Use on Biological and Enzymatic Attributes of an Agricultural Area

A series of anthropogenic approaches, including burning practices and soil disturbances as soil cover removal, plowing and harrowing were experimentally undertaken to mimic land conversion for agricultural production in northern Amazonia. These manipulations led to changes in soil biological and biochemical properties. To reduce knowledge gaps concerning land conversion in the Amazon, the study objective was to evaluate the influence of land use and management practices on the biological attributes and enzymatic activity of the soil in Tepequem, a settlement in north of the Amazon, Brazil. Tepequem was chosen for being highly representative in terms of land use and management patterns in the region. Microbial biomass carbon (MBC), soil basal respiration (SBR), metabolic quotient (qCO2) and enzymatic activity were analyzed. Land use changes resulted in alterations to soil quality. The spontaneous plants found on degraded pasture ensured system diversification, protection and organic contribution, facilitating resumption of ecological balancing of the soil. Good soil quality in managed pasture was attributed to the maintenance of soil cover, provided by grasses, and the absence of soil rotation. Burning, soil disturbances and lack of cover negatively influenced the biological and enzymatic activity in sites that were preparation, deforested and burnt. Chemical attributes are significant factors influencing soil quality and health at subsistance plantation. MBC, qMIC and qCO2, acid phosphatase, β-glucosidase and urease were the most sensitive parameters of differentiation of sites in preparation from those of native vegetation and pastures.


Introduction
Between 1970 and 1990, a government-sponsored settlement program in Amazonia resulted in extensive alterations of regional landscapes (Barbosa, 1993).In northern Roraima, this resulted in dozens of agricultural settlements in forest areas with inadequate land policy and significant increase in deforestation (Barni et al., 2012).The lack of technical assistance and unavailability of rural extension support to the settlers have contributed heavily to land degradation, since the great majority of settlers still use a low-tech slash-and-burn system, with ashes used for soil fertilization (Melo et al., 2006).
During the process of conversion of natural ecosystems into agricultural areas, soil loosed its vegetation cover following land clearing, burning (Fearnside et al., 2013).It is subjected to constant soil tillage (Zuber & Villamil, 2016), reduced organic matter (Raiesi & Baheshti, 2015), changes physical and chemical properties, and strong  (Santos et al., 2013), Ultisol and Alfisol according to a Soil Taxonomy (Soil Survey Staff, 2014).
The study area was chosen for being highly representative of the study region in general, in terms of land use and management patterns adopted, as well as of native vegetation.The study sites consisted of (i) two areas of pastures, (ii) two areas in preparation for conventional tillage (turned soil and a newly burnt area), (iii) one subsistence farming area (cassava, banana and papaya), and (iv) native forest cover area, which served as a control (Table 1).
Soil samples were collected in two different periods: dry period (April 2015) and wet period (July 2015).In each study area, a block of approximately 100 × 100 m (10,000 m 2 ), subdivided into 4 plots of 50 × 50 m (2,500 m 2 ) was set-up, to provide repetitions of each treatment.In each plot, three sub-samples observations were collected.Holes 0.50 × 0.50 × 0.50 m in extent were excavated, and samples taken at 0-5 and 0.05-0.15m, with 6 samples per plot, totaling 24 samples per study area.For the biological and enzymatic analyzes, samples were stored at 4 °C in plastic bags in a freezer in the UFRR soil management laboratory.).With paths indicating intense trampling and laminar erosion.In the rainy season, the pasture had a high populations of spontaneous plants, with a low presence of cattle in the area, which is regionally characteristic of a degraded pasture.
Managed pasture P3 Hapludult (dystrofic yellow argisol) soil under pasture of Brachiaria brizantha since 1992, without liming and fertilization.In the dry season, grass height was between 0.20 to 0.50 m, very uniform, with few spontaneous plants.Cattle grazing was in rotation and animals were spent 90 days per plot.
In the wet season, ungrazable areas developed due to intense trampling effects.

Plots in preperation P4
Hapludult (dystrofic yellow argisol) soil, with bare soil surface.Area prepared for corn planting with plowing, tilling and acidity correction.Two years earlier it had been used as cattle pasture.In the wet season 90 days after the first sampling, it remained uncovered, because sown corn seeds had not germinated.

Subsistance plantation P6
Typic Hapludalf (eutrophic red nitosol) soil under subsistence farming of banana (Musa spp.), cassava (Manihot esculenta Crantz), maize (Zea mays L.) and papaya (Carica papaya L.).Presence of weeds between plants, however, with uneven cover, presence of coal indicating the use of fire for the preparation of the area.90 days after the second data collection the area had little plant cover.
The pH (H 2 O) was measured in a 1:2.5 (soil:water) solution, Ca 2+ and Mg 2+ , K + and available P were extracted by the double-acid solution (HCl 0.05 M + H 2 SO 4 0.0125 M), K + was determined by flame photometry and P by colorimetrically.Cation exchange capacity (CEC) was calculated from the results sorptive complex analysis, all determined as described by EMBRAPA (2009).Analyzes for organic soil carbon (SOC) was performed via dry methods in a Vario El III elemental analyzer (Elementar Analysensysteme GmbH, Germany).
Soil microbial biomass carbon (MBC) was determined using the chloroform-fumigation-extraction method (Vance et al., 1987), and soil basal respiration (SBR) according to Jenkinson and Powlson (1976).The metabolic quotient (qCO 2 ) was the ratio of SBR per unit over the MBC per unit time, and the microbial quotient (qMIC) from the MBC/SOC ratio.To determine enzymatic activity, samples of the first 0.15 m of soil, sieved through a 4 mm mesh, and with vegetal fragments eliminated were used.β-glycosidase and acid phosphatase enzymes were determined using 1 g of soil with their respective extractors (p-Nitrophenyl-b-D-glucopyranoside and p-Nitrophenyl phosphate), following Tabatabai (1994).Urease activity was assessed as described by Kandeler and Gerber (1988), using the rate of urea hydrolysis from a 0.5 g soil sample, and using calorimetry at 660 nm to quantify the ammonia released during the incubation period of the soil with urea, without toluene.
Analysis of variance (F test) was used to detect the existence of a difference between land uses (treatments).Tukey test was performed at 1 and 5% probability to compared mean values of indicators measured for each of the land uses.Data were submitted to principal component analysis (PCA), to determine the relative importance of the analysed characteristics (Singh, 1981).Subsequently, and using INFOGEN software, version 2013 (Balzarini & Di Rienzo, 2013), the discarding variables method was used to eliminate those variables with smallest contribution to variability.

Results and Discussion
Study area soils had high to medium acidity, low fertility, but were high in P and K, especially in the areas affected by the fires (Table 2).
Conversion of forest (P1) to the systems represented by the P2, P3, P4 and P5 sites promoted changes in the chemical properties of soils.The high acidity at P1, P2, P3, P4 and P5 reflect the advanced degree of soil weathering in these areas.However, the highest pH, observed in the initial depth of soil from the P5 site, is derived from the release of bases in the ash resulting from burning in the area.Similar results having been obtained by Melo et al. (2006) with soils from agricultural plots of Apiaú, Roraima.The much lower acidity observed at P4 in the 0.0-0.05m layer is related to addition to the soil of dolomitic limestone as part of a soil management regimen.The average pH of the P6 site is a consequence of the high base contents, inherent to the soil class there (Typic Hapludalf), a fact already recorded by other studies (Melo et al., 2006;Melo & Schaefer, 2009).High surface layer acidity at P3 and P2 is a consequence of soil impoverishment, both by weathering and because of soil exhaustion due to its use as pasture, without replenishment with chemical or organic fertilizers.
The potential acidity observed in the study areas is like that reported by Melo et al. (2006) in the evaluation of Ultisol under forest and indigenous cultivation areas in the Yanomami Indigenous Land, on the mid-Catrimani river, Roraima.This indicates it is the result of natural process, but accelerated by the incorrect uses of the soils, especially without applying a chemical corrective for the natural acidity.
Levels of exchangeable cations (Ca 2+ , Mg 2+ and K + ) and available P found on the surface, especially at the P5 site, are indicative of the contribution that occurs when ashes are used to the increase nutrient availability.By accelerating the organic matter mineralization processes, the burning of vegetation temporarily increases the exchangeable elements content, briefly providing more favorable conditions of soil fertility in the superficial layer of the soil.However, as reported by Pomianoski et al. (2006), this decreases to very low values in a few years.Note.P1 = Native forest; P2 = degraded pasture; P3 = managed pasture P4 = area prepared for maize cultivation; P5 = area under preparation, recently deforested and burned; P6 = subsistence planting.Values followed by the same letters showed no statistical difference (Tukey test, p > 0.01). jas.ccsenet.

Study area other (Figu
The mean According contribute system and of decomp supplemen also prese physical p variation i attributed t Note.P1 = P5 = area the same l 0.01).

Another fa activity of chambers especially
Site P5 sh did not sho order of 4 between th of 51% an org a SOC levels ure 2).The use of fire at site P5 resulted in losses of CO 2 preventing conversion of soil C into MBC, causing a decrease in microbiological activity at this study site.Pomianoski et al. (2006) when evaluating effects of fire as a management tool in an agroforestry system at bracatinga found similar results.

SOC contents g to
The process of repopulation of site P2 by spontaneous plants intensified during the wet season and made the system more diversified, which was a major stimulus to microorganisms, which used them as a source of energy and carbon for their activities.Plant species diversity and their forms of dispersion directly influence the number and frequency of soil biota interactions.This, in turn, directly and indirectly influences the biological, physical and chemical attributes of the soil, especially via natural vegetation regeneration (Medeiros et al., 2017).
The wet season decline of MBC at site P3 and P6 was due to the reduction in vegetation cover.The remaining of the cattle grazing the P3 site altered soil conditions, since the area was trampled and lacked covering vegetation.
Grazing for prolonged periods may result in vegetation degradation, since this may respond differently to the intensity and frequency of grazing.At site P6, this result may also have been influenced by the withdrawal of subsistence production species and the practice of weeding, common in this type of management, which subjecting the area to stress, promoting losses of C.These results are similar to those of Silva et al. (2012), who compared soil biomass carbon of pasture and subsistence agriculture areas in the Paraíba valley.These authors reported a reduction of up to 63% in SMB-C in both areas in the wet season.
Study site microbial quotients (qMIC) in different seasons also indicated stress reduction occurred at site P2 area between the dry and rainy seasons since, with a 23% increase in qMIC, it remained statistically similar to site P1.qMIC values also revealed the impact suffered at sites P3 and P6, as a function of wet season management practices, with reduction of 49% and 44%, respectively.
Sites P1, P2 and P5 showered statistical difference (p < 0.01) between the studied layers.At site P2 MBC of soil was highest in the 0.0-0.05m layer, while at P1 and P5 was highest in the 0.05-0.15m layer (Figure 3).ation; y the boa et not a β-glicosidase and urease showed a significant interaction effect between the areas and climatic seasons, whereas the acid phosphatase showed significant differences only between the studied areas, and not for seasons.In the wet season β-glucosidase activity at site P2 was 163% higher than at P1.Only sites P2 and P3 showed statistical difference between seasons (p < 0.01).In the dry season β-glucosidase activity at sites P2 and P3 exceeded that at P1 by 86% and 153%, respectively.Other sites showed no difference (p > 0.01) (Table 4).In the wet season β-glucosidase activity at site P2 was 163% higher than at P1.Only sites P2 and P3 showed statistical difference between seasons (p < 0.01).Higher β-glucosidase activity at P2 and P3 is related to the SOC, MBC and abundance of pasture system roots.
The abundance of roots made available by the grazing system to the soil either from the presence of grasses (P3), or spontaneous plants (P2), favored microbiological activity.Kotroczó et al. (2014) also found this influence occurring during an evaluation of how soil cover (litter and roots) impact activities of the enzymes β-glycosidase and acid phosphatase.They found that, as a result of greater microbial concentrations, enzymatic activity tends to be more intense in the plant rhizosphere.The wet season increase of β-glucosidase activity at site P2 can be explained by the higher incidence of plants.Another positive influence on enzymatic activity at sites P2 and P3 was the absence of soil movement, which favored the protection of aggregates.Positive correlations have been observed between the enzymatic activity and soil aggregates as, according to Raiesi and Beheshti (2015); Zuber and Villamil (2016), these provide protective casings for microorganisms, resulting in greater enzymatic activity.
Site P2 showed extensive termite activity in the first centimeters of the soil, which potentiated increases in soil enzyme activity.Termites contribute to the physical and chemical improvement of the soil, promoting better conditions for organic compound mineralization (Lal, 1988).In contrast to sites P1, P2 and P3, land use and management at sites P4, P5 and P6 exposed the soil, due to a reduction and/or total loss of vegetation cover, so reducing soil protection, and inhibiting the biochemical activity responsible for nutrient cycling.Similar results have been reported by Udawatta et al. (2009), when comparing cultivated areas in the conventional system and agroforestry systems.------mg P nitrofenol kg -1 soil h -1 --------------mg NH 4 N kg -1 soil h Note.P1 = native forest; P2 = degraded pasture; P3 = managed pasture; P4 = area prepared for maize cultivation; P5 = area under preparation, recently deforested and burned; P6 = subsistence planting.Values followed by the same letters showed no statistical difference (Tukey test, p > 0.01).
The low β-glucosidase activity found at the P5 and P4 sites may also be associated with exposure of the soil to high temperatures, which occurred in these agroecosystems: by fire at the first, and at the second by solar irradiation, resulting from a total absence of the cover and soil disturbance.According to Boerner et al. (2008), such thermal stress can impair enzymatic kinetics.Accordingly, biodegradability of soil organic matter can be reduced (Kotroczó et al., 2004;Zuber & Villamil, 2016).
β-glycosidase action includes converting cellulose to glucose (Mendes et al., 2012).Low β-glucosidase activity at the P1 site in both periods may be associated with the quantity and quality of the plant matter residues that were being returned to the soil.These are more complex in the native areas than in the agricultural areas, resulting in a consequent reduction in β-glucosidase activity.Similar effects were reported by Acosta Martinez et al. (2007), in tropical forests and dry and wet seasons.
There was no statistical between-site differences for urease activity.Though differences between seasons occurred, with a reduction in levels present at sites P1, P2 and P3 from the dry season to the wet.These observations differs from the reports of other authors who have compared activity urease between agricultural areas and those under native vegetation.For example, Raise and Bahesti (2015) reported a 28% reduction of urease activity in cultivated areas compared to native vegetation areas.
Highest acid phosphatase activity occurred at sites P1, P2 and P3, and no statistical differences existed between these areas (p > 0.01).Acid phosphatase activity at sites P4, P5 and P6 was 33% lower than at site P1 (Figure 5).
Acid phosphatase values found at sites P1, P2 and P3 confirm the positive relationship of this enzyme has with the quantity and quality of SOC.Phosphatase shows increased activity in more preserved environments, where C availability is higher, and temperature, acidity and pH more favorable, as shown by Silva et al. (2009), in an evaluation of the biological attributes of forest plantation areas.At sites P4, P5 and P6, in addition to the impact of soil management on cover and protection, one of the factors that may have contributed negatively to the activity of this enzyme, especially at P5 was the high levels of P available in the first 0.05 m of the soil (Table 1).According to Mendes et al. (2012), in P-deficient environments there is a compensation in the soil and plant system that stimulates the production of phosphatase.In consequence, high levels of inorganic P can inhibit phosphatase action.
PCA provided a clearer separation between sites P4, P5 and P6 and P1, P2 and P3.This underscored that the different forms of soil management and use and the soil chemical attributes influenced the biological and enzymatic attributes of the soil.Within-and between-site correlations found between the dry and wet seasons permitted the evaluation of which chemical, biological and enzymatic attributes contributed most to the improvement or diminution of soil quality.In CP1, the chemical attributes SOC and pH associated with the biological and biochemical attributes MBC, qMIC, acid phosphatase and β-glycosidase were reduced during the dry season in P1, P2 and P3 soils.During this period, site P3 was not being grazed, so there was little pressure on the soil.In the wet season of the attributes measured, SOC, MBC, qMIC, acid phosphatase, β-glycosidase and urease appeared most closely -6.00 -3.00 0.00 3.00 6.00 CP 1 (42.3%)linked with the quality of these environments.During this season, of all disturbed sites, P2 soil characteristic most resembled those of the control area of native vegetation (P1).These results are related to the availability, quantity and quality of organic material in various areas.
The quantity and quality of leaf litter, the higher complexity of plant structure, and the greater biological and microbiological diversity exerted a strong positive influence on both the biological and biochemical factors measured at site P1.Forested areas are also associated with higher levels of humidity, temperature and nutrient cycling, and this combination influences the microbiome, and chemical attributes as pH and SOC, as well as biological (MBC and qMIC), and biochemical attributes (β-glycosidase, acid phosphatase and urease) in both seasons.
Grasses and spontaneous plants, respectively, provided more readily-available plant material, mainly via roots, furnishing greater volumes of easy decomposed substrate at sites P2 and P3.This may well have had a positive influence on MBC and enzyme activity.Soil quality is derived from the balance between SOC and biological-plus-biochemical properties (Chaer et al., 2009).
Sites P4 and P5 showed positive correlations for C/N, H+Al, P, qCO 2 and negative correlations for SOC, dry period pH, MBC, qMIC, acid phosphatase and β-glycosidase, indicating a soil system imbalance as a consequence of human management, that had resulted in inefficient carbon use by the microbial community, thus reducing the quality of these soils.
CP2 showed a positive relation at site P6 between Ca+Mg, K and pH, reflecting the importance of chemical attributes for soil quality.However, probably due to pH values and chemical conditions of the soil class of this area, the management form there did not favor biological and biochemical activity.This is in line with the results of Acosta-Martinez et al. (2007) in tropical watershed soils in Costa Rica.

Conclusions
Land use changes in the settlement area of Tepequém in the north of the Amazon, resulted in alterations to soil quality.
The spontaneous plants found on degraded pasture ensured system diversification, protection and organic contribution, facilitating resumption of ecological balancing of the soil.
Soil quality at managed pasture is attributed to the maintenance of soil cover, provided by grasses, and the absence of soil rotation.
Burning practices, soil disturbances and lack of cover negatively influenced the biological and enzymatic quality at sites in preparation and deforestation and burning.
Chemical attributes are significant factors influencing soil quality and health at subsistence plantation.
MBC, qMIC and qCO 2 , acid phosphatase, β-glucosidase and urease were the most sensitive methods by which to differentiate sites in preparation from the native vegetation and pastures.
Figure = Native forest under prepara letter, upperca

Figure
Figure 3. S

Table 1 .
Characteristics of land uses in the Tepequém settlement during the dry season (April/2015) and during the rainy season (July/2015) Degraded pasture P2Hapludult (dystrofic yellow argisol) soil under grass pasture (Brachiaria brizantha A. Rich), and submitted to extensive and continuous cattle grazing without fertility management, and without rotation.In the dry season, there was a dry covering, interspersed with spontaneous plants (jurubeba: Solanum paniculatum L.; lobeira: Solanum lycocarpum A. St.-Hil.; vassourinha: Scoparia dulcis L.; malícia: Mimosa pudica L.

Table 2 .
Chemical properties of human-altered soils from the Tepequém Settlement-Amajari-Roraima-State, northern Brazil

Table 3 .
Values for microbial biomass carbon (MBC), microbial quotient (qMIC) in two seasons (dry and wet) from soils from human-altered areas of the Tepequém Settlement, depth 0.0-0.15m

Table 4 .
Enzymatic activity in soils from human-altered areas of Tepequém-Amajari settlement-RR.Samples taken from depths of 0.0 to 0.15 m in dry and wet seasons