Changes in soil chemical properties under different farming systems exploration in semiarid region of Paraíba

The change of Caatinga natural vegetation in the Brazilian semiarid region, to different systems of agricultural exploitation, that is, the replacement of complex and stable systems by simple and unstable systems has caused changes in soil properties that are dependent on the climate, the type of crops and the management adopted. Based on areas of Caatinga native vegetation, this study aimed to evaluate the changes in the chemical characteristics of soils under different farming systems in Paraiba backwoods. Vertissol samples were collected at 0 to 10, 10 to 20, 20 to 30 and 30 to 40 cm and chemically characterized. Selected systems were native vegetation, sparse vegetation, pasture, annual and permanent crops. Based on the statistical analysis, it was concluded that the replacement of native vegetation by agricultural farming systems in the region of watershed Riacho Val Paraiso, PB, caused changes only in pH, potassium and sodium in the soil attributes. There was a trend of soil chemical properties increasing in the areas of agricultural cultivation and with depth. In all areas of agricultural farming systems, soil fertility is suitable for most crops. 
 
   
 
 Key words: macronutrients, land use, soil depth.


INTRODUCTION
The caatinga biome, occupying an area of about 850,000 km², about 11% of the national territory, is the main existing ecosystem in the Brazilian Northeast region under semiarid climate.The population in this area corresponds to about 20 million inhabitants.This area has significant socioeconomic and ecological importance.
The complex and stable systems of the natural vegetation of the caatinga has been replaced by simple and unstable crop systems.This has caused changes in soil chemical and physical properties which are dependent on climate, crop type and management type (Santana and Souto, 2011).This change may result in decreased vegetation cover causing soil loss and, consequently, reduced soil fertility.
According to Chaves et al. (2006) the indiscriminate use of natural vegetation, intensive pasture and irrational use of agriculture are factors that have contributed to accelerate the process of environmental imbalance.The consequence of this process is related to the reduction of soil fertility and biodiversity (Travassos and Souza, 2011).Barros et al. (2013), for example, noted that carbon and nitrogen stocks in the soil under cultivation of sugar cane decreased compared to the soil under native forest.Likewise, agricultural cultivation in soils of the floodplains of the river Guamá caused a reduction in the concentration of P and K (Lopes et al., 2006).
The characterization of soil in a particular area is fundamental for sustainable models aiming to maximize output and mitigation of natural resource degradation.Among the soils found in the semiarid region are the Vertisols characterized as clayey or loamy soils with high content of expandable clay minerals (2:1) causing the appearance of "slickensides" and splitting the subsurface layers of soil in dry season, and may or may not provide the "Gilgai"-type microrelief (EMBRAPA, 1999).Generally, there are high amounts of exchangeable bases (S) and saturation (V%) and a reaction ranging from moderately acidic (on the surface of some of these soils) to strongly alkaline (at the horizon C) (EMBRAPA, 1999).It was evident in the surveys that the depth profile of the AC ranges from 60 to 130 cm.
In agricultural systems, the soil chemical properties are altered, positively or negatively, depending on the soil management adopted.In this sense, Santos and Ribeiro (2002), when studying the effects of irrigated agriculture on chemical properties of soils of the São Francisco submedium region found that the chemical properties were affected differently, depending on the adopted management.With the objective of evaluating land use, Corrêa et al. (2009) found that in relation to native vegetation, uses of cultures with short cycles, discarded areas, pastures, and fruit cultures were higher in the three analyzed layers of pH, exchangeable Ca and Mg attributes, sum bases, base saturation, and available P. Lopes et al. (2006) also observed that the pasture showed greater sustainability of fertility than the system under rice cultivation.
Generally, in environments under natural vegetation, less variation occurs in chemical and physical soil properties when compared to farm management systems; thus, the natural vegetation is a sure indicator to evaluate different types of land use, allowing the evaluation of the sustainable or unsustainable use of certain agricultural practices (Menino et al., 2012).
Based on areas of Caatinga native vegetation, this study aimed to evaluate the changes in the chemical characteristics of soils under different farming systems in Paraiba backwoods.

MATERIALS AND METHODS
The study area is located in the watershed of Val Paraiso stream, between the cities São João do Rio do Peixe and Sousa, inserted in the Northwest portion of Paraíba State, Brazil (situated in the parallel of latitude 6°37'54" to 6°44'29" South and meridians of longitude 38° 18'21" to 38°24'12'' West).According to Koppen classification, the climate is warm tropical climate of severe drought, reaching over 35°C at times of higher temperatures.In the study area the average annual temperature is 27°C and the index average annual rainfall is 967.23 mm; however during the survey, an average temperature of 28.6°C and pluviometric index of 6.2 mm were recorded.The vegetation is basically composed of Caatinga Hiperxerófila and the predominant soil in the Val Paraiso watershed is classified as Vertisol (Fernandes Neto, 2009).
In the watershed, five farming systems exploration were identified: native vegetation, taken as reference (area covered by arboreal natural vegetation); sparse vegetation (area covered by natural vegetation typical of the caatinga in recovery; this area was deforested ten years ago, submitted by three consecutive years for agricultural cultivation, and currently is in the process of forest recovery), pasture (area covered by a sparse vegetation and planted; this area under pasture was deforested for over 40 years and yet is being used as pasture for ruminant animals); annual crops (areas of temporary crops; this area is being used for over 60 years with annual crop under constant activities of burning, disking and plowing) and permanent agriculture (crop with high vegetation cover, mainly composed of permanent crops; this area was deforested for 5 years and is under permanent cultivation agriculture).
For each farming systems, exploration were opened for profiles occurring in the same soil class.In each profile, the soil samples were collected from July to August 2012, at depths of 0 to 10, 10 to 20, 20 to 30 and 30 to 40 cm.These samples, after being air dried and passed through a sieve of 2 mm, were characterized chemically according to the methods recommended by Embrapa (1997).The chemical elements analyzed were: calcium (Ca), magnesium (Mg), sodium (Na), potassium (K), hydrogen (H), aluminum (Al) and phosphorus (P).From these data, the sum of exchangeable bases (EB), cation exchange capacity (CEC) and base saturation percentage (BS%) were calculated.
The experimental design was completely randomized in factorial scheme 4 × 5, with four replicates (four profiles) that is four depths (0 to 10; 10 to 20; 20 to 30 and 30 to 40 cm) and five sites (native vegetation, sparse vegetation, pastures, annual agriculture and permanent agriculture).
Data were analyzed using descriptive statistics by calculating the maximum, minimum, mean and coefficient of variation.Analysis of variance (ANOVA) and Tukey test at 5% probability were made for comparison of means of the results, according to Ferreira (2000).

RESULTS AND DISCUSSION
Soil pH was significantly affected (p < 0.01) by the different environments studied likewise, there were significant difference (p < 0.05) in the amounts of electrical conductivity (EC) and potassium.In relation to depth, there was a significant effect (p < 0.01) on the pH results and on the sodium (Na) and potassium (K) amount; there was a significant effect (p < 0.05) on the hydrogen (H) amount.However, the interaction of areas of farming systems exploration x depths showed a significant effect (p < 0.05) on the H amount (Table 1).The pH is the mechanism used either to identify the soil acidity, or the hydrogen ions concentration in soil solution.Regardless of the depth of soil samples, in accordance with the minimum (6.30) and maximum  2009) and Lopes and Guilherme (2004), that the soil of the region under study as a whole, presents reaction ranging from moderately acidic to highly alkaline.These pH levels are due mainly to the high levels of exchangeable bases found in the soil of the study area.Among the factors that may have contributed to these high levels, stands out the characteristics of the soil, Vertisol, and regional climate.This soil is young, little weathered, clayey with expandable clay minerals (2:1) and located in semiarid region.Therefore, due to the origin of the soil, low permeability soil, hindering drainage and low rainfall of the region, hindering the leaching of chemical elements, the soil still has naturally high levels of exchangeable bases.
Native vegetation showed minimum and maximum pH values of 6.70 and 8.08; sparse vegetation of 6.30 and 8.15; pasture of 6.70 and 7.80; annual crops of 8.86 and 6.60 and permanent agriculture of 7.60 and 8.17, respectively (Table 2) as can be seen in Figure 1.
According to the results presented in Table 3, the highest average pH values were found in areas under agricultural cultivation.This is probably due to the residual effects of anthropogenic interventions more pronounced along the cycles of crops disagreeing de (Melo et al., 2010).These authors observed no significant differences in the pH of the soil due to different ways of using the watershed Riacho do Tronco in Boa Vista, Paraíba State.
In general, the soil pH increased with depth in all profiles (Table 2) which probably is related to elevation concentrations of calcium carbonates and bicarbonates ions.
The Coefficient of Variation (CV) should be used as a parameter to validate the mean values since, according to Vanni (1998), CV above 35% shows that the average has little meaning and values greater than 65% reflect data very heterogeneous nullifying the trustworthiness of average.According to the CV classification proposed by Warrick and Nielsen (1980), the pH of the soil for all farming systems exploitation at all depths were low (CV <12%), corroborating Cavalcante et al. (2007), Souza et al. (2008) and Neves Neto et al. (2013).This can be attributed to the fact that this variable may be measured on a small scale of values and be a logarithmic function (Neves Neto et al., 2013).
According to Santana et al. (2007), electrical conductivity expresses the salts amount present in the soil solution.Thus the greater the amount of salts presents in the solution, the greater the value of the electrical conductivity.Although there was no difference between the averages of EC in different farming systems and in the depths of the profiles, there was a trend towards higher values in areas under agricultural cultivation (Table 3).According to the classification of Warrick and Nielsen (1980), the coefficients of variation were classified as low (CV < 12%) for all treatments and depths.
Data calcium (Ca) ranged from 10.70 (annual agricultural area) to 29.65 kg cmol c kg -1 (in the permanent agriculture area) and magnesium (Mg) from 4.93 (pasture area) to 17.34 cmol c kg -1 (in the sparse vegetation area) (Table 4) showed that all contents of these elements were classified as high (Lopes and Guilherme, 2004) corroborating Melo et al. (2010) and Chaves et al. (2006).Considering that levels 2 to 3 cmol c kg -1 Ca and around 4 cmol c kg -1 Mg (Raij, 1981) are adequate to crop development, it can be stated that the study areas do not exist deficiencies of these elements.
Although the results of Ca and Mg soil have not shown significant differences in farming systems exploration and different depths, disagreeing with Vasconcelos et al. (2010) and Lima et al. (2011), there was a trend of increasing values in areas of cultivation as well as increase and decrease in the levels of Ca and Mg, respectively, relative to depth.
Data sodium (Na) ranged from 1.54 (in the permanent agriculture area) to 7.62 cmol c kg -1 (in the sparse vegetation area) not presenting significant differences in studied treatments.However, the Na + increased with depth, presenting significant differences (Table 4).
According to Santos and Ribeiro (2002), the increase of Na with depth may be related to the addition of this element by irrigation water, as may have occurred a displacement of said element of colloids soil by Ca 2+ , Mg 2+ and K + , from fertilizer applied, due to its lower    energy retention.Furthermore, for environments that are not irrigated, the increase in sodium in the lower horizons can be related to more restricted drainage in the soil profile, but also because the lowest position in the landscape.The fact sodicity increase in subsurface horizons is worrying, since the plant roots reach those horizons being impaired by the presence of the element sodium.However, the Na content in the samples of the present research is not harmful, once the exchangeable sodium percentage in the complex is below 8%, that samples classified as normal in relation to sodicity.Data potassium (K) ranged from 0.18 (in the annual crops area) to 0.88 cmol c kg -1 (in the permanent agriculture area) that is from the average level (0.16 to 0.30 cmol c kg -1 ) to very high (> 0.60 cmol c kg -1 ).However, in all areas and depths analyzed, the average values ranged from high to very high (Table 4) corroborating Chaves et al. (2007).

Depth (cm)
The potassium levels were significantly different in both studied areas with lowest values of depths occurrence in areas with native vegetation and layers below 10 cm depth (Table 4), corroborating with Pereira et al. (2009), Leite et al. (2012) and Lira et al. (2012).
The highest values in the area of permanent agriculture and topsoil are probably associated with a continuous supply of plant residues favoring the retention of this nutrient, the absence of tillage, which favors' the accumulation of nutrients in the sampled depth (Barreto et al., 2008) beyond the addition of potassium fertilizers, because it is a cultivated area corroborating according to Matias et al. (2009).
Calcium presented itself as the main contributor to the total exchangeable bases (EB), followed by magnesium, sodium and potassium, reflecting thus the nature of the source material.Aluminum not detected in soil samples (Table 4) with low hydrogen contents the values of the total exchangeable bases of these samples represent practically the values of cation exchange capacity (CEC), which ranged from 30.01 to 39.26 cmol c kg -1 classified as high.Due to equivalence of high values of sum of bases to the cation exchange capacity, percentage saturation of exchangeable bases corresponded to 100% in all systems of farming operation, showing high nutritional potential for plants.
Data phosphorus (P) ranged from 32.30 (in the native vegetation area) to 58.60 mg dm -3 (in the pasture area) shows that all contents of these elements were classified as high (Lopes and Guilherme, 2004).According Falleiro et al. (2003) and Leite et al. (2012) the high levels of P in the soil may be due to the residual effect of previous fertilizations, maintenance of plant residues on the soil surface, which favors the cycling of phosphorus of no soil disturbance, which promotes the formation phosphorus sites (Sa, 2004) and the very origin of the soil.
Although the results of soil P have not shown significant differences in agricultural farming systems and different depths, disagreeing Leite et al. (2012), there was a trend of increasing values in the areas of crop and decrease the levels of P in relation to depth corroborating Leite et al. (2012).The highest levels apparently in the topsoil, is related to the fact that P move by diffusion in the soil, which results in low mobility profile, contributing to its accumulation in this layer (Zalamena, 2008).According the classification of CV proposed by Warrick and Nielsen (1980), it was observed that the levels of exchangeable bases and P in all environments and at all depths, showed low variability (CV <12%) reflecting homogeneous data with high reliability.

Conclusion
The replacement of native vegetation by agricultural farming systems in the region of watershed Riacho Val Paraiso, PB, caused changes only in pH, potassium and sodium in the soil attributes.The highest concentrations of carbonates and bicarbonates of calcium in the deeper horizons of the soil increased the pH in these horizons.Soil leaching, although low, due to the semiarid climate, causes increased concentration of cations, especially potassium and sodium, in the deeper soil horizons.There was a trend of soil chemical properties increasing in the areas of agricultural cultivation and with depth.
The Vertisol is a young, little weathered and very clayey soil, with low soil permeability because it is located in semiarid region with low rainfall; thus, in all areas of agricultural farming systems, soil fertility is suitable for

Figure 1 .
Figure 1.pH values for each area: native vegetation, sparse vegetation, pasture, annual crops and permanent crops at four depths.

Table 1 .
Summary of variance analysis for the chemical soil properties.

Table 3 .
Mean values of soil pH and electrical conductivity (E.C.) depending on the areas of farming systems exploration and depth.Means followed by same letters in the lines do not differ by Tukey test to 5% probability. a

Table 4 .
Mean values of calcium (Ca), magnesium (Mg), sodium (Na), potassium (K), the sum of exchangeable bases (EB) and phosphorus (P) depending on the areas of farming systems exploration and depth.