Alluvial soil quality in agroforestry systems and native forest of the Brazilian semiarid region

The agroforestry systems consist of techniques appropriate to local conditions, taking into account the coexistence with the semiarid in Brazil. The objective of the research was to evaluate the soil properties in agroforestry management units (SAFs), and native forest, in Bueno community, city of Irauçuba, CE. Treatments were: (1) Agroforestry management unit SAF 1 in the elevated portion of the landscape (interfluve); (2) Agroforestry management unit SAF 2 in the portion of slope (colluvium), and (3) Native forest. Soil samples were collected: 5 composite samples, derived from 15 sub-samples in areas of study in the layers 0.00-0.10; 0.10-0.20; 0.20-0.30 and 0.30-0.40 m. It was a completely randomized design, with 5 repetitions, treatments (SAF 1, SAF 2 and native forest) and the plots being repetitions. The analysis of variance was applied and means submitted to Tukey test at 5%. There were significant differences for properties analyzed, except to CEes, indicating low concentration of soluble salts, without risks of salinity (0.25 a 0.34 dS m -1 ). The exchangeable sodium (9.51 a 29.88 mg dm -3 ), with normal values of PST (0.66 a 1.35%), in the SAF 2 differed from the others. The values of exchangeable sodium considered high are not characterized with restriction according to the normal EST. The units of SAF 1 and SAF 2 contributed to the maintenance of soil quality in top condition to MN, featuring high levels of calcium, magnesium and potassium, with restrictions to the exchangeable sodium SAF 1 and SAF 2 to a lesser extent to the native forest.


INTRODUCTION
Agroforestry is an integrated approach to land use that is characterized by deliberate maintenance of trees and other woody perennials in fields and pastures.This system is one of the best known traditional practices for livelihood, suitable land management and sustainable development (Kittur and Bargali 2013;Parihaar et al., 2014Parihaar et al., , 2015)).These indigenous agroforestry systems not only support the livelihood through production of food, fodder and fuel wood, but also mitigate the impact of climate change through carbon sequestration ( al., 2011;Singh et al., 2008;Bargali et al., 2009).These include strategies such as crop rotation, consortia, crops associations with animals, which are aimed at inputs obtainment and soil plant cover (Altieri, 1999).
Agricultural production intensification tendency to meet the global demand for food, the pressure on nonrenewable natural resources and the need to produce with sustainability emphasizes the importance of understanding agricultural production factors, and especially understand the soil quality study (Gonzaga et al., 2013).
The Brazilian semiarid region is one of the most populous in the world and with rainfalls above the average of others, with spatial and temporal variability, favoring high evaporation rates.In this geographical space, shallow soils that are little weathered and with good chemical characteristics and physical limitations to agricultural cultivation dominate.
In this region, human intervention in the agricultural environment has occurred in order to use natural resources for food obtainment and, with population expansion, in the pursuit of increased productivity and production.Soil intensive preparation for agricultural crops, deforestation and fires have caused several impacts on the agro-ecosystem, such as flora and fauna, water availability and agricultural soils quality reduction, changing the regional climate.Human action in a disorderly way, without observing land agricultural potential, change the physical, chemical and biological attributes, accelerating desertification.
This predominant agricultural production model in the Brazilian semiarid region does not favor its agroecosystem genetic and social heritage conservation and protection.This fact contributes to unveil threatening life form degradation scenarios on the planet, thus accelerating the desertification process.
Given the addressed problems, the semiarid region needs an interaction culture that considers the drought phenomenon, the available natural resources and the people who inhabit this plural and diverse geopolitical space.Thus, the development of technologies and research adapted to semiarid geoenvironmental conditions are necessary to achieve an agriculture based on rational water use and natural resources alternative sources.
Sustainable agriculture in the semiarid region can be achieved by means of agricultural production systems design using technologies and management practices that conserve or improve agro-ecosystem physical basis and sustaining capacity (Franco, 2000).Agroforestry systems are recognized by farmers and nongovernmental organizations as a semi-arid interaction technology.These redesigns are integral parts of a larger system focusing on the family unit, with external inputs independence and agricultural crops diversity, ensuring food security sovereignty.Therefore, it is a recognized technology by farmers as a sustainable production method with the interaction and local features of the semiarid region.
Thus, it is essential to describe, assess the impacts and scientifically validate environmental and social improvement of this successful experience of interaction with the semi-arid, and thus build in a participatory manner a sustainable agricultural production alternative for family farmers.
In this context, the research aimed to evaluate soil chemical and physical properties in agroforestry management units (SAFs), with comparative reference to native forest in Bueno community, municipality of Irauçuba, CE.

MATERIALS AND METHODS
The research was conducted in Bueno community, municipality of Irauçuba, CE, and consisting of 37 families.Latitude: 03° 36 '24 27' S and longitude: 39° 51 '27 59 "W.Main activities are eminently of agricultural nature, where small livestock, beekeeping, rainfed crops of short cycle species such as beans, corn, sesame and sorghum, as well as production yards stand out.From experiences in the SAFs, they expanded their production systems with yards production around their homes.
The areas were defined and agroforestry management units (SAF) were implanted in July 2007.These followed local features criteria regarding landscape and degradation level, besides having a nearby water source for soil moisture use through groundwater upwelling (dam with valley).Therefore, two areas implanted with agroforestry management units (SAFs) were used, conducted with the same implantation criteria, with each one in different locations in the landscape (SAF 1), interfluve (higher land area) and (SAF 2) colluvium (lowland area), totaling 0.5 ha each SAF under study.The soil was classified as Eutrophic Fluvic Neosol (Santos et al., 2013).
The study consisted of the following treatments: (1) Agroforestry management unit, as defined in the previous SAFs design section, being located in the landscape higher portion (interfluve); (2) Agroforestry management unit, located on the lowland portion (colluvium) and 3) native forest as reference (Figure 1A to E).
Differences that distinguish SAF from units 1 and 2 were the position in the landscape, which influences the systems water dynamics.SAF 1 is constituted of good water drainage, with the environment remaining unsaturated, that is, soil pores partially filled with water and air throughout the year.In SAF 2, during the rainy season, is poor soil water drainage happens, providing a saturated environment, that is, all the pore space is filled with water.After the rainy season, soil moisture is higher compared to SAF 1, thus having more stored water in the soil, favoring managed agricultural crops development, a fact favored due to the superficial water table.
Area 3 refers to native forest (NF) and is considered as a reference, comprising a length of 3 ha, without human action for about thirty years.It is located on the opposite portion of SAFs 1 and 2 agroforestry management units, with a distance of approximately 600 m.This vegetation consists of hyperxerophilic caatinga, with predominance of the following main species: thrush, Caesalpinia pyramidalis, Mimosa hostilis, quince, Brazilian-walnut, Combretum leprosum, beach mororó and termite nests in the trees.The soil surface is covered with plant litter in different decomposition stages.
To perform the laboratory analyzes, samples with deformed structure were collected, with five composite samples, originated from 15 sub-samples in each study area in the layers of 0.00 to 0.10; 0.10-0.20;0.20-0.30and 0.30-0.40m depth, using a Dutch Table 1.Particle size distribution, textural classification and particles density in agroforestry management units (SAF 1 and SAF 1) and native forest, in the layers from 0.00 to 0.10, 0.10 to 0.20, 0.20 to 0.30 and 0.30 to 0.40 m. type auger, put up in labeled plastic bags and taken to the Soil Analysis Laboratory.Samples were air dried, buffered and passed in sieves with mesh opening of 2.00 mm, thus obtaining the thin airdried soil (TADS), which were subjected to physical, chemical and soil analysis.

Sand
For granulometric analysis, the pipette method was used, using the sodium hexametaphosphate chemical dispersant and distilled water in 20 g (TADS), with slow mechanical agitation in a shaker (Wagner 50 rpm) for 16 h.Sand (2.00 to 0.05 mm) was measured by sieving, clay (<0.002 mm) by sediment and silt (0.05 to 0.002 mm) by the difference between sand and clay fractions.
pH was obtained in water; Ca, Mg and P content through the Mehlich-1 extractor, and quantified by atomic absorption spectrophotometry (Ca and Mg) and flame photometry (P).Electrolytic conductivity (EC) and Na + were determined at 25°C in aqueous extract, at a 1:5 ratio; nitrogen (N) was digested in digester block with sulfuric acid, hydrogen peroxide and digesting mixture at 250°C for 30 min, and at 350°C for 2 h, and quantified by titration with NaOH 0.025 mol L -1 after distillation; exchangeable aluminum (Al 3+ ) by extraction with KCl 1 mol L -1 and titrated with NaOH 0.025 mol L -1 ; potential acidity (H + Al) was measured by titration with NaOH 0.025 mol L -1 after extraction with calcium acetate 0.5 mol L -1 , at 7.0 pH; and soil organic matter (SOM), by titration with ferrous ammonium sulfate 0.005 M after heated in uniform plate with potassium dichromate 0.02 M. From the analyzes, the following indexes were obtained: sum of bases (SB); effective cation exchange capacity (t); cation exchange capacity at pH 7.0 (CEC); bases saturation (V%); exchangeable aluminum saturation (m%) and exchangeable sodium percentage (ESP), according to Donagema et al. (2011).
The design was completely randomized with five replications, with treatments (agroforestry systems and native forest) and plots considered as repetitions.The results were submitted to analysis of variance, and treatment means were submitted to Tukey's test at 5% probability, using the ASSISTAT 7.6 statistical program (Silva et al., 2002).

RESULTS AND DISCUSSION
Table 1 shows particle size distribution results (granulometry), their textural classification and particle density.The studied systems and their respective layers showed no textural range, and it was classified as sandy loam, since studied units had soil class (Fluvic Neosols).It is worth noting that granulometry is a soil physical characteristic of difficult modification, since it is inherent of the source material, not being modified by soil management practices and agricultural crops.
Silt fraction values (Table 1) were high, what is an indicative of young and little weathered soils, and according to the classification and standards adopted by the National Center for Soil Research -CNPS, EMBRAPA, 90% of the municipality area consists of shallow soils characterized by crystalline rocks, corroborating with Silva et al. (2014) and Rebouças et al. (2014).In SAF 2, there was clay decrease and sand increase, what can be explained by agroforestry management units topography (SAF 1 and 2), and SAF 1 is at the highest point in the landscape, thus being ), what was characterized as mineral soil, conditioned by the source material and its mineralogical composition and silt/clay ratio (1.13 to 4.43).
In Table 2, SAFs 1 and 2 agroforestry management units and native forest soil chemical attributes mean squares are shown.Significant differences were found at 5% probability level for chemical attributes, except for phosphorus and electrical conductivity in the saturation extract.
Significant differences were found in the analyzed attributes, except for CE es , indicating soluble salts low concentration, without salinity potential risks (0.25 to 0.34 dS m -1 ) for SAF 1 and 2 agroforestry management units and native forest (Table 2).This fact can be explained by exchangeable sodium increase (from 9.51 to 29.88 mg dm -3 ), and PST normal values (0.66 to For studied systems, SAF2 pH average values differed from SAF1 and native forest (Table 3).It is noteworthy that pH values have no restrictions for crop development, although potential acidity differed between studied units, with an average value higher than others for SAF1 (11.03 cmol c dm -3 ), with restrictions also to the high exchangeable calcium content, probably coming from the source material (crystalline rocks).
Calcium, magnesium, potassium and soil organic matter chemical attributes at SAF 1 and SAF 2 agroforestry management units were higher than native forest conditions.These results can be attributed to the constant organic compounds input in several decomposition stages derived from implanted plant species diversity in combination with caatinga (local biome) during seven years from its implementation.This indicates that plants management with crop residues thinning, pruning and raking shows SAFs potential in soil organic matter increase.These results corroborate with those found by Xavier et al. (2004), and is considered a good indicator of changes in soil organic matter as a function of soil management and agricultural crops.
It was found that nitrogen (N) and soil organic matter (SOM) values, followed the same tendency, with statistical difference between studied units, being higher in the SAF 1, followed by SAF 2 and NF, where N needs SOM mineralization According to Frazão et al. (2008), N levels are strongly correlated with those of organic matter.
SAFs agroforestry systems can be a viable alternative for agriculture development in the Brazilian semiarid to the extent that arable species introduction contributes to soil properties maintenance and improvement (Table 2).Other assessed attributes qualitatively contributed through observation and experience at SAF 1 and SAF 2 study units, such as bee pasture, thermal comfort, humidity control, residue inputs at several decomposition stages, food availability and diversity for permanent food and nutritional security.Studies evaluating agroforestry systems should take into account local conditions, such as weather pattern, sampling time, soil type and landscape position.Adopted methods for soil attributes assessment are dynamic and can be modified in time and space (Marchiori Junior et al., 2000).For phosphorus (P) content, there was SAF 2 statistical difference compared to the others, but with lower values.

Depth
units, although sodium had masked the sum of base values, since sodium content were high in all units under study, SAF 1, SAF 2 and native forest (Table 3).Mean values for soil chemical attributes in layers from 0.00 to 0.10; 0.10 to 0.20; 0.20 to 0.30 and 0.30 to 0.40 m are shown in Table 4. Significant differences were verified in the 0.30-0.40layer, with higher value for CE es , indicating soluble salts low concentration, without salinity potential risks (from 0.226 to 0.413 dS m -1 ).This fact may have occurred due to increased exchangeable sodium in the 0.10-0.20 and 0.20-0.30layers (20.65 to 21.67) mg dm -3 ), differing from the other layers, with the same tendency for PST (0.97 to 1.25%).They were not characterized with restrictions.
The high sodium content in all layers (Table 4), were not characterized as solodic, with high CEC levels, statistically differing in the layers from 0.00 to 0.10 and from 0.10 to 0.20, with (15.10 to 11.02 cmol c dm -3 ). Calcium differed statistically on 0.00 to 0.10 and 0.30 to 0.30 layers, (6.51 and 3.97 cmol c dm -3 ), with magnesium having the same tendency (2.01 and 1.24 cmol c dm -3 ), and potassium differing in all layers.These are considered high levels.
The pH varied in the 0.30-0.40layer, with lower value (Table 4) without restrictions.Potential acidity has not differed between layers, and these were considered high values.
It was found that nitrogen (N) and soil organic matter (SOM) values in the layers follow the same tendency, except for soil matter at a depth of 0.30-0.40,with lower value than the others, with decrease in depth.The others, however, had lower values, which are considered low.
Soil chemical attributes interactions in agroforestry management units SAF 1, SAF 2 and native forest at layers from 0.00 to 0.10, 0.10 to 0.20, 0.20 to 0.30 and 0.30 to 0.40 m are shown in (Table 5).
The pH was statistically different in the native forest, which had lower value compared to SAF 1 and SAF 2, at the 0.00 to 0.10 layer (Table 5).Potential acidity and exchangeable aluminum differed between studied units, with higher average value (9.43 cmol c dm -3 ) in the other studied units, besides not differing between layers.Assessing CE es , SAF 1 differed statistically from other units under study in the layers of 0.20-0.30and 0.30-0.40.Significant differences in pH value for natural forests replaced by plantations were also reported by Bargali et al (1993) andJoshi et al (1997).
SAF 1 PST differed from other studied units in all layers, with higher value.(PST = 0.57%), with normal values in the studied units and layers, being influenced by calcium, magnesium and potassium high levels.Because of these exchangeable cations high levels, it was not constituted as solodic.
Exchangeable sodium results were high, differing between studied units, with SAF 1 having higher value than others and the same tendency in layers, with restrictions on agricultural crops.CEC has not differed between studied units nor between layers, but the values were considered high, thereby giving cultivation good chemical characteristics.Calcium, magnesium and potassium chemical attributes were high and have not differed between studied units, except between layers with variations.
Organic matter in the native forest had lower value and differed from SAF 1 and SAF 2 agroforestry units, differing in 0.20-0.30and 0.30-0.40subsurface layers.SAF 1 Nitrogen (N) showed higher value and differed from SAF 2 and native forest, with no statistical differences between layers.
The sum of bases (SB), effective CEC (t) and base saturation have not differed between studied units, with statistical variation between layers.It is noteworthy that Table 5.Chemical attributes interaction in Agroforestry Management Units (SAF 1 and SAF 2) and Native Forest in layers from 0.00 to 0.10, 0.10 to 0.20, 0.20 to 0.30 and 0.30 to 0 40 m for pH, CE, PST, N, MOS, P, Al 3+ , (H+Al), t, T, V, m, K + , Na + , ca (V˃50%) was higher than 50% in studied units, and is considered eutrophic (Table 5).

Conclusions
SAF 1 and SAF 2 agroforestry management units contributed to soil quality maintenance in higher conditions than native forest.SAF 1 and SAF 2 agroforestry management units and native forest showed favorable chemical attributes to soil fertility concerning calcium, magnesium and potassium content, with restrictions regarding SAF 1 and SAF 2 exchangeable sodium levels, and in a lower extent to native forest.SAF 1 and SAF 2 agroforestry management units provided higher soil organic matter intakes compared to Forest.
Arora et *Corresponding author.E-mail: alessandoln@yahoocom.br Author(s) agree that this article remain permanently open access under the terms of the Creative Commons Attribution License 4.0 International License

Table 2 .
Mean squares for soil chemical attributes in agroforestry management units (SAF 1 and SAF 2) and native forest.

Table 3 .
Soil chemical attributes means in agroforestry management units (SAF 1 and SAF 2) and native forest.