Comparison between artificial neural networks and maximum likelihood classification in digital soil mapping

Soil surveys are the main source of spatial information on soils and have a range of different applications, mainly in agriculture. The continuity of this activity has however been severely compromised, mainly due to a lack of governmental funding. The purpose of this study was to evaluate the feasibility of two different classifiers (artificial neural networks and a maximum likelihood algorithm) in the prediction of soil classes in the northwest of the state of Rio de Janeiro. Terrain attributes such as elevation, slope, aspect, plan curvature and compound topographic index (CTI) and indices of clay minerals, iron oxide and Normalized Difference Vegetation Index (NDVI), derived from Landsat 7 ETM+ sensor imagery, were used as discriminating variables. The two classifiers were trained and validated for each soil class using 300 and 150 samples respectively, representing the characteristics of these classes in terms of the discriminating variables. According to the statistical tests, the accuracy of the classifier based on artificial neural networks (ANNs) was greater than of the classic Maximum Likelihood Classifier (MLC). Comparing the results with 126 points of reference showed that the resulting ANN map (73.81 %) was superior to the MLC map (57.94 %). The main errors when using the two classifiers were caused by: a) the geological heterogeneity of the area coupled with problems related to the geological map; b) the depth of lithic contact and/or rock exposure, and c) problems with the environmental correlation model used due to the polygenetic nature of the soils. This study confirms that the use of terrain attributes together with remote sensing data by an ANN approach can be a tool to facilitate soil mapping in Brazil, primarily due to the availability of low-cost remote sensing data and the ease by which terrain attributes can be obtained.

Phosphorus extracted by ion exchange resins and mehlich-1 from oxisols (latosols) treated with different phosphorus rates and sources for varied soil-source contact periods

Despite the large number of studies addressing the quantification of phosphorus (P) availability by different extraction methods, many questions remain unanswered. The aim of this paper was to compare the effectiveness of the extractors Mehlich-1, Anionic Resin (AR) and Mixed Resin (MR), to determine the availability of P under different experimental conditions. The laboratory study was arranged in randomized blocks in a [(3 x 3 x 2) + 3] x 4 factorial design, with four replications, testing the response of three soils with different texture: a very clayey Red Latosol (LV), a sandy clay loam Red Yellow Latosol (LVA), and a sandy loam Yellow Latosol (LA), to three sources (triple superphosphate, reactive phosphate rock from Gafsa-Tunisia; and natural phosphate from Araxá-Minas Gerais) at two P rates (75 and 150 mg dm-3), plus three control treatments (each soil without P application) after four contact periods (15, 30, 60, and 120 days) of the P sources with soil. The soil acidity of LV and LVA was adjusted by raising base saturation to 60 % with the application of CaCO3 and MgCO3 at a 4:1 molar ratio (LA required no correction). These samples were maintained at field moisture capacity for 30 days. After the contact periods, the samples were collected to quantify the available P concentrations by the three extractants. In general, all three indicated that the available P-content in soils was reduced after longer contact periods with the P sources. Of the three sources, this reduction was most pronounced for triple superphosphate, intermediate for reactive phosphate, while Araxá phosphate was least sensitive to the effect of time. It was observed that AR extracted lower P levels from all three soils when the sources were phosphate rocks, while MR extracted values close to Mehlich-1 in LV (clay) and LVA (medium texture) for reactive phosphate. For Araxá phosphate, much higher P values were determined by Mehlich-1 than by the resins, because of the acidity of the extractor. For triple superphosphate, both resins extracted higher P levels than Mehlich-1, due to the consumption of this extractor, particularly when used for LV and LVA.

Forms and accumulation of copper and zinc in a sandy typic hapludalf soil after long-term application of pig slurry and deep litter

Successive applications of pig slurry and pig deep litter may lead to an accumulation of copper (Cu) and zinc (Zn) fractions in the soil profile. The objective of this study was to evaluate the Cu and Zn forms and accumulation in a Sandy Typic Hapludalf soil after long-term application of pig slurry and deep litter. In March 2010, eight years after initiating an experiment in Braço do Norte, Santa Catarina (SC), Brazil, on a Sandy Typic Hapludalf soil, soil samples were collected from the 0-2.5, 2.5-5.0, 5-10 and 10-15 cm layers in treatments consisting of no manure application (control) and with applications of pig slurry and deep litter at two levels: the single and double rate of N requirement for maize and black oat succession. The soil was dried, ground in an agate mortar and analyzed for Cu and Zn contents by 0.01 mol L-1 EDTA and chemically fractionated to determine Cu and Zn. The applications of Pig deep litter and slurry at doses equivalent to 90 kg ha-1 N increased the contents of available Cu and Zn in the surface soil layer, if the double of this dose was applied in pig deep litter or double this dose in pig slurry, Cu and Zn migrated to a depth of 15 cm. Copper is accumulated mainly in the organic and residual fractions, and zinc preferentially in the fraction linked to clay minerals, especially in the surface soil layers.

The effect of intrinsic soil properties on soil quality assessments

The assessment of soil quality is based on indicators and indices derived from soil properties. However, intrinsic soil properties may interfere with other soil properties that vary under different land uses and are used to calculate the indices. The aim of this study was to assess the extent to which intrinsic soil properties (clay and iron oxide contents) explain variable soil properties (sum of bases, potential acidity, organic carbon, total porosity, and bulk density) under different land uses (native forest, no-tillage and conventional agriculture) on small family farms in Southern Brazil. The results showed that the five properties evaluated can be included in soil quality assessments and are not influenced by the clay and iron oxide contents. It was concluded that for little weathered 1:1 and 2:1 phyllosilicate rich-soils, if the difference between the maximum and the minimum clay content under the different land uses is less than about 200 g kg-1 and the iron oxide content less than about 15 g kg-1, the physico-chemical soil properties in the surface layer are determined mostly by the land use.

Water retention and availability in soils of the State of Santa Catarina-Brazil: effect of textural classes, soil classes and lithology

The retention and availability of water in the soil vary according to the soil characteristics and determine plant growth. Thus, the aim of this study was to evaluate water retention and availability in the soils of the State of Santa Catarina, Brazil, according to the textural class, soil class and lithology. The surface and subsurface horizons of 44 profiles were sampled in different regions of the State and different cover crops to determine field capacity, permanent wilting point, available water content, particle size, and organic matter content. Water retention and availability between the horizons were compared in a mixed model, considering the textural classes, the soil classes and lithology as fixed factors and profiles as random factors. It may be concluded that water retention is greater in silty or clayey soils and that the organic matter content is higher, especially in Humic Cambisols, Nitisols and Ferralsol developed from igneous or sedimentary rocks. Water availability is greater in loam-textured soils, with high organic matter content, especially in soils of humic character. It is lower in the sandy texture class, especially in Arenosols formed from recent alluvial deposits or in gravelly soils derived from granite. The greater water availability in the surface horizons, with more organic matter than in the subsurface layers, illustrates the importance of organic matter for water retention and availability.

Soil quality indicators in a rhodic kandiudult under different uses in northern Parana, Brazil

Sustainable use of soil, maintaining or improving its quality, is one of the goals of diversification in farmlands. From this point of view, bioindicators associated with C, N and P cycling can be used in assessments of land-use effects on soil quality. The aim of this study was to investigate chemical, microbiological and biochemical properties of soil associated with C, N and P under different land uses in a farm property with diversified activity in northern Parana, Brazil. Seven areas under different land uses were assessed: fragment of native Atlantic Forest; growing of peach-palm (Bactrys gasipaes); sugarcane ratoon (Saccharum officinarum) recently harvested, under renewal; growing of coffee (Coffea arabica) intercropped with tree species; recent reforestation (1 year) with native tree species, previously under annual crops; annual crops under no-tillage, rye (Cecale cereale); secondary forest, regenerated after abandonment (for 20 years) of an avocado (Persea americana) orchard. The soil under coffee, recent reforestation and secondary forest showed higher concentrations of organic carbon, but microbial biomass and enzyme activities were higher in soils under native forest and secondary forest, which also showed the lowest metabolic coefficient, followed by the peach-palm area. The lowest content of water-dispersible clay was found in the soil under native forest, differing from soils under sugarcane and secondary forest. Soil cover and soil use affected total organic C contents and soil enzyme and microbial activities, such that more intensive agricultural uses had deeper impacts on the indicators assessed. Calculation of the mean soil quality index showed that the secondary forest was closest to the fragment of native forest, followed by the peach-palm area, coffee-growing area, annual crop area, the area of recent reforestation and the sugarcane ratoon area.

Impacts of land leveling on lowland soil physical properties

The practice of land leveling alters the soil surface to create a uniform slope to improve land conditions for the application of all agricultural practices. The aims of this study were to evaluate the impacts of land leveling through the magnitudes, variances and spatial distributions of selected soil physical properties of a lowland area in the State of Rio Grande do Sul, Brazil; the relationships between the magnitude of cuts and/or fills and soil physical properties after the leveling process; and evaluation of the effect of leveling on the spatial distribution of the top of the B horizon in relation to the soil surface. In the 0-0.20 m layer, a 100-point geo-referenced grid covering two taxonomic soil classes was used in assessment of the following soil properties: soil particle density (Pd) and bulk density (Bd); total porosity (Tp), macroporosity (Macro) and microporosity (Micro); available water capacity (AWC); sand, silt, clay, and dispersed clay in water (Disp clay) contents; electrical conductivity (EC); and weighted average diameter of aggregates (WAD). Soil depth to the top of the B horizon was also measured before leveling. The overall effect of leveling on selected soil physical properties was evaluated by paired "t" tests. The effect on the variability of each property was evaluated through the homogeneity of variance test. The thematic maps constructed by kriging or by the inverse of the square of the distances were visually analyzed to evaluate the effect of leveling on the spatial distribution of the properties and of the top of the B horizon in relation to the soil surface. Linear regression models were fitted with the aim of evaluating the relationship between soil properties and the magnitude of cuts and fills. Leveling altered the mean value of several soil properties and the agronomic effect was negative. The mean values of Bd and Disp clay increased and Tp, Macro and Micro, WAD, AWC and EC decreased. Spatial distributions of all soil physical properties changed as a result of leveling and its effect on all soil physical properties occurred in the whole area and not specifically in the cutting or filling areas. In future designs of leveling, we recommend overlaying a cut/fill map on the map of soil depth to the top of the B horizon in order to minimize areas with shallow surface soil after leveling.

Perennial grasses for recovery of the aggregation capacity of a reconstructed soil in a coal mining area in southern Brazil

The construction of a soil after surface coal mining involves heavy machinery traffic during the topographic regeneration of the area, resulting in compaction of the relocated soil layers. This leads to problems with water infiltration and redistribution along the new profile, causing water erosion and consequently hampering the revegetation of the reconstructed soil. The planting of species useful in the process of soil decompaction is a promising strategy for the recovery of the soil structural quality. This study investigated the influence of different perennial grasses on the recovery of reconstructed soil aggregation in a coal mining area of the Companhia Riograndense de Mineração, located in Candiota-RS, which were planted in September/October 2007. The treatments consisted of planting: T1- Cynodon dactylon cv vaquero; T2 - Urochloa brizantha; T3 - Panicum maximun; T4 - Urochloa humidicola; T5 - Hemarthria altissima; T6 - Cynodon dactylon cv tifton 85. Bare reconstructed soil, adjacent to the experimental area, was used as control treatment (T7) and natural soil adjacent to the mining area covered with native vegetation was used as reference area (T8). Disturbed and undisturbed soil samples were collected in October/2009 (layers 0.00-0.05 and 0.10-0.15 m) to determine the percentage of macro- and microaggregates, mean weight diameter (MWD) of aggregates, organic matter content, bulk density, and macro- and microporosity. The lower values of macroaggregates and MWD in the surface than in the subsurface layer of the reconstructed soil resulted from the high degree of compaction caused by the traffic of heavy machinery on the clay material. After 24 months, all experimental grass treatments showed improvements in soil aggregation compared to the bare reconstructed soil (control), mainly in the 0.00-0.05 m layer, particularly in the two Urochloa treatments (T2 and T4) and Hemarthria altissima (T5). However, the great differences between the treatments with grasses and natural soil (reference) indicate that the recovery of the pre-mining soil structure could take decades.

Arbuscular mycorrhizal fungal communities and soil aggregation as affected by cultivation of various crops during the sugarcane fallow period

Management systems involving crop rotation, ground cover species and reduced soil tillage can improve the soil physical and biological properties and reduce degradation. The primary purpose of this study was to assess the effect of various crops grown during the sugarcane fallow period on the production of glomalin and arbuscular mycorrhizal fungi in two Latosols, as well as their influence on soil aggregation. The experiment was conducted on an eutroferric Red Latosol with high-clay texture (680 g clay kg-1) and an acric Red Latosol with clayey texture (440 g kg-1 clay) in Jaboticabal (São Paulo State, Brazil). A randomized block design involving five blocks and four crops [soybean (S), soybean/fallow/soybean (SFS), soybean/millet/soybean (SMS) and soybean/sunn hemp/soybean (SHS)] was used to this end. Soil samples for analysis were collected in June 2011. No significant differences in total glomalin production were detected between the soils after the different crops. However, total external mycelium length was greater in the soils under SMS and SHS. Also, there were differences in easily extractable glomalin, total glomalin and aggregate stability, which were all greater in the eutroferric Red Latosol than in the acric Red Latosol. None of the cover crops planted in the fallow period of sugarcane improved aggregate stability in either Latosol.

Variability of soil fertility properties in areas planted to sugarcane in the State of Goias, Brazil

Soil sampling should provide an accurate representation of a given area so that recommendations for amendments of soil acidity, fertilization and soil conservation may be drafted to increase yield and improve the use of inputs. The aim of this study was to evaluate the variability of soil fertility properties of Oxisols in areas planted to sugarcane in the State of Goias, Brazil. Two areas of approximately 8,100 m² each were selected, representing two fields of the Goiasa sugarcane mill in Goiatuba. The sugarcane crop had a row spacing of 1.5 m and subsamples were taken from 49 points in the row and 49 between the row with a Dutch auger at depths of 0.0-0.2 and 0.2-0.4 m, for a total of 196 subsamples for each area. The samples were individually subjected to chemical analyses of soil fertility (pH in CaCl2, potential acidity, organic matter, P, K, Ca and Mg) and particle size analysis. The number of subsamples required to compose a sample within the acceptable ranges of error of 5, 10, 20 and 40 % of each property were computed from the coefficients of variation and the Student t-value for 95 % confidence. The soil properties under analysis exhibited different variabilities: high (P and K), medium (potential acidity, Ca and Mg) and low (pH, organic matter and clay content). Most of the properties analyzed showed an error of less than 20 % for a group of 20 subsamples, except for P and K, which were capable of showing an error greater than 40 % around the mean. The extreme variability in phosphorus, particularly at the depth of 0.2-0.4 m, attributed to banded application of high rates of P fertilizers at planting, places limitations on assessment of its availability due to the high number of subsamples required for a composite sample.

Soil uses during the sugarcane fallow period: influence on soil chemical and physical properties and on sugarcane productivity

The planting of diversified crops during the sugarcane fallow period can improve the chemical and physical properties and increase the production potential of the soil for the next sugarcane cycle. The primary purpose of this study was to assess the influence of various soil uses during the sugarcane fallow period on soil chemical and physical properties and productivity after the first sugarcane harvest. The experiment was conducted in two areas located in Jaboticabal, São Paulo State, Brazil (21º 14' 05'' S, 48º 17' 09'' W) with two different soil types, namely: an eutroferric Red Latosol (RLe) with high-clay texture (clay content = 680 g kg-1) and an acric Red Latosol (RLa) with clayey texture (clay content = 440 g kg-1). A randomized block design with five replications and four treatments (crop sequences) was used. The crop sequences during the sugarcane fallow period were soybean/millet/soybean, soybean/sunn hemp/soybean, soybean/fallow/soybean, and soybean. Soil use was found not to affect chemical properties and sugarcane productivity of RLe or RLa. The soybean/millet/soybean sequence improved aggregation in the acric Latosol.

Evaluation of statistical and geostatistical models of digital soil properties mapping in tropical mountain regions

Soil properties have an enormous impact on economic and environmental aspects of agricultural production. Quantitative relationships between soil properties and the factors that influence their variability are the basis of digital soil mapping. The predictive models of soil properties evaluated in this work are statistical (multiple linear regression-MLR) and geostatistical (ordinary kriging and co-kriging). The study was conducted in the municipality of Bom Jardim, RJ, using a soil database with 208 sampling points. Predictive models were evaluated for sand, silt and clay fractions, pH in water and organic carbon at six depths according to the specifications of the consortium of digital soil mapping at the global level (GlobalSoilMap). Continuous covariates and categorical predictors were used and their contributions to the model assessed. Only the environmental covariates elevation, aspect, stream power index (SPI), soil wetness index (SWI), normalized difference vegetation index (NDVI), and b3/b2 band ratio were significantly correlated with soil properties. The predictive models had a mean coefficient of determination of 0.21. Best results were obtained with the geostatistical predictive models, where the highest coefficient of determination 0.43 was associated with sand properties between 60 to 100 cm deep. The use of a sparse data set of soil properties for digital mapping can explain only part of the spatial variation of these properties. The results may be related to the sampling density and the quantity and quality of the environmental covariates and predictive models used.

Assessment of pedotransfer functions for estimating soil water retention curves for the amazon region

Knowledge of the soil water retention curve (SWRC) is essential for understanding and modeling hydraulic processes in the soil. However, direct determination of the SWRC is time consuming and costly. In addition, it requires a large number of samples, due to the high spatial and temporal variability of soil hydraulic properties. An alternative is the use of models, called pedotransfer functions (PTFs), which estimate the SWRC from easy-to-measure properties. The aim of this paper was to test the accuracy of 16 point or parametric PTFs reported in the literature on different soils from the south and southeast of the State of Pará, Brazil. The PTFs tested were proposed by Pidgeon (1972), Lal (1979), Aina & Periaswamy (1985), Arruda et al. (1987), Dijkerman (1988), Vereecken et al. (1989), Batjes (1996), van den Berg et al. (1997), Tomasella et al. (2000), Hodnett & Tomasella (2002), Oliveira et al. (2002), and Barros (2010). We used a database that includes soil texture (sand, silt, and clay), bulk density, soil organic carbon, soil pH, cation exchange capacity, and the SWRC. Most of the PTFs tested did not show good performance in estimating the SWRC. The parametric PTFs, however, performed better than the point PTFs in assessing the SWRC in the tested region. Among the parametric PTFs, those proposed by Tomasella et al. (2000) achieved the best accuracy in estimating the empirical parameters of the van Genuchten (1980) model, especially when tested in the top soil layer.

Multivariate analysis applied to the study of the relationship between soil and plant properties in a peach orchard

In the State of Rio Grande do Sul, the municipality of Pelotas is responsible for 90 % of peach production due to its suitable climate and soil conditions. However, there is the need for new studies that aim at improved fruit quality and increased yield. The aim of this study was to evaluate the relationship that exists between soil physical properties and properties in the peach plant in the years 2010 and 2011 by the technique of multivariate canonical correlation. The experiment was conducted in a peach orchard located in the municipality of Morro Redondo, RS, Brazil, where an experimental grid of 101 plants was established. In a trench dug beside each one of the 101 plants, soil samples were collected to determine silt, clay, and sand contents, soil density, total porosity, macroporosity, microporosity, and volumetric water content in the 0.00-0.10 and 0.10-0.20 m layers, as well as the depth of the A horizon. In each plant and in each year, the following properties were assessed: trunk diameter, fruit size and number of fruits per plant, average weight of the fruit per plant, fruit pulp firmness, Brix content, and yield from the orchard. Exploratory analysis of the data was undertaken by descriptive statistics, and the relationships between the physical properties of the soil and of the plant were assessed by canonical correlation analysis. The results showed that the clay and microporosity variables were those that exhibited the highest coefficients of canonical cross-loading with the plant properties in the soil layers assessed, and that the variable of mean weight of the fruit per plant was that which had the highest coefficients of canonical loading within the plant group for the two years assessed.

Three-dimensional interpolation of soil data: fertility and pedomorphological features in southern Brazil

The graphical representation of spatial soil properties in a digital environment is complex because it requires a conversion of data collected in a discrete form onto a continuous surface. The objective of this study was to apply three-dimension techniques of interpolation and visualization on soil texture and fertility properties and establish relationships with pedogenetic factors and processes in a slope area. The GRASS Geographic Information System was used to generate three-dimensional models and ParaView software to visualize soil volumes. Samples of the A, AB, BA, and B horizons were collected in a regular 122-point grid in an area of 13 ha, in Pinhais, PR, in southern Brazil. Geoprocessing and graphic computing techniques were effective in identifying and delimiting soil volumes of distinct ranges of fertility properties confined within the soil matrix. Both three-dimensional interpolation and the visualization tool facilitated interpretation in a continuous space (volumes) of the cause-effect relationships between soil texture and fertility properties and pedological factors and processes, such as higher clay contents following the drainage lines of the area. The flattest part with more weathered soils (Oxisols) had the highest pH values and lower Al3+ concentrations. These techniques of data interpolation and visualization have great potential for use in diverse areas of soil science, such as identification of soil volumes occurring side-by-side but that exhibit different physical, chemical, and mineralogical conditions for plant root growth, and monitoring of plumes of organic and inorganic pollutants in soils and sediments, among other applications. The methodological details for interpolation and a three-dimensional view of soil data are presented here.