Building predictive models of soil particle-size distribution

Is it possible to build predictive models (PMs) of soil particle-size distribution (psd) in a region with complex geology and a young and unstable land-surface? The main objective of this study was to answer this question. A set of 339 soil samples from a small slope catchment in Southern Brazil was used to build PMs of psd in the surface soil layer. Multiple linear regression models were constructed using terrain attributes (elevation, slope, catchment area, convergence index, and topographic wetness index). The PMs explained more than half of the data variance. This performance is similar to (or even better than) that of the conventional soil mapping approach. For some size fractions, the PM performance can reach 70 %. Largest uncertainties were observed in geologically more complex areas. Therefore, significant improvements in the predictions can only be achieved if accurate geological data is made available. Meanwhile, PMs built on terrain attributes are efficient in predicting the particle-size distribution (psd) of soils in regions of complex geology.

Relationships between microbial activity and soil physical and chemical properties in native and reforested Araucaria angustifolia forests in the state of São Paulo, Brazil

Araucaria angustifolia (Bert.) O. Kuntze is the main component of the Mixed Ombrophilous forest and, in the State of São Paulo, it is associated with a high diversity of soil organisms, essential for the maintenance of soil quality, making the conservation of this ecosystem a major and pressing challenge. The objective of this study was to identify the physical and chemical properties that are most closely correlated with dehydrogenase enzyme activity, basal respiration and microbial biomass under native (NF) and replanted (RF) Araucaria angustifolia forests in three regions of the state of São Paulo, in winter and summer. The main differentiating factors between the areas were also determined. Each forest was represented by three true replications; at each site, from around the araucaria trees, 15 soil samples (0-20 cm) were collected to evaluate the soil physical, chemical and microbiological properties. At the same points, forest litter was sampled to assess mass and chemical properties. The following microbiological properties were evaluated: microbial biomass carbon (MBC), basal respiration (CO2-C), metabolic quotient (Q: CO2), dehydrogenase enzyme activity (DHA) as well as the physical properties (moisture, bulk density, macroporosity and total porosity), soil chemical properties [pH, organic carbon (org-C), P, Ca, K, Mg, Al, H+Al], litter dry mass, and C, N and S contents. The data were subjected to analysis of variance (TWO-WAY: ANOVA). A Canonical Discriminant Analysis (CDA) and a Canonical Correlation Analysis (CCA) were also performed. In the soil under NF, the values of K, P, soil macroporosity, and litter dry mass were higher and Q: CO2 and DHA lower, regardless of the sampling period, and DHA was lower in winter. In the RF areas, the levels of moisture, porosity and Q: CO2 were higher in both sampling periods, and DHA was higher in winter. The MBC was only higher under NF in the summer, while the litter contents of C, N and S were greater in winter. In winter, CCA showed a high correlation of DHA with CO2-C, pH and H+Al, while in the summer org-C, moisture, Mg, pH and litter C were more associated with DHA and CO2-C. The CDA indicated H+Al, available P, total porosity, litter S content, and soil moisture as the most discriminating variables between NF and RF, but moisture was the most relevant, in both seasons and CO2-C only in winter. The combined analysis of CCA and CDA showed that the contribution of the microbiological variables to a differentiation of the areas was small at both samplings, which may indicate that the period after reforestation was long enough to allow an almost complete recovery of the microbial activity.

Comparison between detailed digital and conventional soil maps of an area with complex geology

Since different pedologists will draw different soil maps of a same area, it is important to compare the differences between mapping by specialists and mapping techniques, as for example currently intensively discussed Digital Soil Mapping. Four detailed soil maps (scale 1:10.000) of a 182-ha sugarcane farm in the county of Rafard, São Paulo State, Brazil, were compared. The area has a large variation of soil formation factors. The maps were drawn independently by four soil scientists and compared with a fifth map obtained by a digital soil mapping technique. All pedologists were given the same set of information. As many field expeditions and soil pits as required by each surveyor were provided to define the mapping units (MUs). For the Digital Soil Map (DSM), spectral data were extracted from Landsat 5 Thematic Mapper (TM) imagery as well as six terrain attributes from the topographic map of the area. These data were summarized by principal component analysis to generate the map designs of groups through Fuzzy K-means clustering. Field observations were made to identify the soils in the MUs and classify them according to the Brazilian Soil Classification System (BSCS). To compare the conventional and digital (DSM) soil maps, they were crossed pairwise to generate confusion matrices that were mapped. The categorical analysis at each classification level of the BSCS showed that the agreement between the maps decreased towards the lower levels of classification and the great influence of the surveyor on both the mapping and definition of MUs in the soil map. The average correspondence between the conventional and DSM maps was similar. Therefore, the method used to obtain the DSM yielded similar results to those obtained by the conventional technique, while providing additional information about the landscape of each soil, useful for applications in future surveys of similar areas.

Quantification of aluminium in soil of the Solimões Formation, Acre State, Brazil

The variety of soils in the State of Acre is wide and their chemical profiles are still not fully understood. The nature of the material of origin of these soils is indicated by the high aluminium (Al) content, commonly associated with high calcium (Ca) and magnesium (Mg) contents. The study objective was to use different methods to quantify Al in soils from toposequences formed from material of a sedimentary nature originating from the Solimões Formation, in Acre, Brazil. Trenches were opened at three distinct points in the landscape: shoulder, backslope and footslope positions. Soil samples were collected for physical, chemical, mineralogical analyses. The Al content was quantified using different methods. High Al contents were found in most of these horizons, associated with high Ca and Mg levels, representing the predominant cations in the sum of exchangeable bases. The mineralogy indicates that the soils are still in a low weathering phase, with the presence of significant quantities of 2:1 minerals. Similar Al contents were determined by the methods of NaOH titration, xylenol orange spectrometry and inductively coupled plasma optical emission spectrometry. However, no consistent data were obtained by the pyrocatechol violet method. Extraction with KCl overestimated the exchangeable Al content due to its ability to extract the non-exchangeable Al present in the smectite interlayers. It was observed that high Al contents are related to the instability of the hydroxyl-Al smectite interlayers.

Sugarcane trash management assessed by the interaction of yield with soil properties

Currently, sugarcane plays an important global role, particularly with a view to alternative energy sources. Thus, in a sugarcane field of the mill Vale do Paraná S/A Álcool e Açúcar, Rubineia, São Paulo State, managed under two green cane harvest systems (cane trash left on and cane trash removed from the soil), Pearson and spatial correlations between the sugarcane yield (variety RB855035 in the third cut) and soil physical and chemical properties were studied to identify the property best correlated with stalk yield and the best harvest method. For this purpose, two geostatistical grids (121 sampling points on 1.30 ha) were installed on a eutrophic Red Argisol (homogeneous slope of 0.065 m m-1), in 2011, to determine the properties: stalk yield and sugarcane plant population, and soil resistance to penetration, gravimetric moisture, bulk density, and carbon stock, in the layers 0-0.20 and 0.20-0.40 m. The data were analyzed by descriptive, linear correlation and geostatistical analysis. In both treatments, the property stand density was best correlated with sugarcane yield (r = 0.725 in the trash mulching treatment - TM and r = 0.769 in the trash removal treatment - TR). However, in relation to the soil properties, bulk density (0-0.20 m) was best correlated (r = 0.305 in TM, r = 0.211 in TR). Similarly, from the spatial point of view, stand density was the property that best explained the sugarcane yield. However, in the TM treatment the density (0.20-0.40 m) was the only soil property spatially correlated with stalk yield. The carbon stock in the soil of the TM was 11.5 % higher than in the TR treatment. Results of the TM treatment were best, also with regard to soil management and conservation.

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.

Compared performance of penetrometers and effect of soil water content on penetration resistance measurements

Modern agriculture techniques have a great impact on crops and soil quality, especially by the increased machinery traffic and weight. Several devices have been developed for determining soil properties in the field, aimed at managing compacted areas. Penetrometry is a widely used technique; however, there are several types of penetrometers, which have different action modes that can affect the soil resistance measurement. The objective of this study was to compare the functionality of two penetrometry methods (manual and automated mode) in the field identification of compacted, highly mechanized sugarcane areas, considering the influence of soil water volumetric content (θ) on soil penetration resistance (PR). Three sugarcane fields on a Rhodic Eutrudrox were chosen, under a sequence of harvest systems: one manual harvest (1ManH), one mechanized harvest (1MH) and three mechanized harvests (3MH). The different degrees of mechanization were associated to cumulative compaction processes. An electronic penetrometer was used on PR measurements, so that the rod was introduced into the soil by hand (Manual) and by an electromechanical motor (Auto). The θ was measured in the field with a soil moisture sensor. Results showed an effect of θ on PR measurements and that regression models must be used to correct data before comparing harvesting systems. The rod introduction modes resulted in different mean PR values, where the "Manual" overestimated PR compared to the "Auto" mode at low θ.

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.

Spatial estimation of foliar phosphorus in different species of the genus Coffea based on soil properties

Information underlying analyses of coffee fertilization systems should consider both the soil and the nutritional status of plants. This study investigated the spatial relationship between phosphorus (P) levels in coffee plant tissues and soil chemical and physical properties. The study was performed using two arabica and one canephora coffee variety. Sampling grids were established in the areas, and the points georeferenced. The assessed properties of the soil were levels of available phosphorus (P-Mehlich), remaining phosphorus (P-rem) and particle size, and of the plant tissue, phosphorus levels (foliar P). The data were subjected to descriptive statistical analysis, correlation analysis, cluster analysis, and probability tests. Geostatistical and trend analyses were only performed for pairs of variables with significant linear correlation. The spatial variability for foliar P content was high for the variety Catuai and medium for the other evaluated plants. Unlike P-Mehlich, the variability in P-rem of the soil indicated the nutritional status of this nutrient in the plant.

Artificial neural networks applied for soil class prediction in mountainous landscape of the Serra do Mar¹

Soil information is needed for managing the agricultural environment. The aim of this study was to apply artificial neural networks (ANNs) for the prediction of soil classes using orbital remote sensing products, terrain attributes derived from a digital elevation model and local geology information as data sources. This approach to digital soil mapping was evaluated in an area with a high degree of lithologic diversity in the Serra do Mar. The neural network simulator used in this study was JavaNNS and the backpropagation learning algorithm. For soil class prediction, different combinations of the selected discriminant variables were tested: elevation, declivity, aspect, curvature, curvature plan, curvature profile, topographic index, solar radiation, LS topographic factor, local geology information, and clay mineral indices, iron oxides and the normalized difference vegetation index (NDVI) derived from an image of a Landsat-7 Enhanced Thematic Mapper Plus (ETM+) sensor. With the tested sets, best results were obtained when all discriminant variables were associated with geological information (overall accuracy 93.2 - 95.6 %, Kappa index 0.924 - 0.951, for set 13). Excluding the variable profile curvature (set 12), overall accuracy ranged from 93.9 to 95.4 % and the Kappa index from 0.932 to 0.948. The maps based on the neural network classifier were consistent and similar to conventional soil maps drawn for the study area, although with more spatial details. The results show the potential of ANNs for soil class prediction in mountainous areas with lithological diversity.

Index of soil physical quality of hardsetting soils on the brazilian coast

Many soils have a hard-setting behavior, also known as cohesive or "coesos". In such soils, the penetration resistance increases markedly when dry and decreases considerably when moist, creating serious limitations for plant emergence and growth. To evaluate the level of structure degradation in hard-setting soils with different texture classes and to create an index for assessing soil hardness levels in hard-setting soils, six soil representative profiles were selected in the field in various regions of Brazil. The following indices were tested: S, which measures soil physical quality, and H , which analyzes the degree of hardness and the effective stress in the soil during drying. Both indices were calculated using previously described functions based on data from the water-retention curves for the soils. The hard-setting values identified in different soils of the Brazilian Coastal Tablelands have distinct compaction (hardness) levels and can be satisfactorily measured by the H index. The S index was adequate for evaluating the structural characteristics of the hard-setting soils, classifying them as suitable or poor for cultivation, but only when the moisture level of the soil was near the inflection point. The H index showed that increases in density in hard-setting soils result from increases in effective stress and not from the soil texture. Values for Bd > 1.48 kg dm-3 classify the soil as hard-setting, and the structural organization is considered "poor".

Estimation of field capacity from ring infiltrometer-drainage data

Field capacity (FC) is a parameter widely used in applied soil science. However, its in situ method of determination may be difficult to apply, generally because of the need of large supplies of water at the test sites. Ottoni Filho et al. (2014) proposed a standardized procedure for field determination of FC and showed that such in situ FC can be estimated by a linear pedotransfer function (PTF) based on volumetric soil water content at the matric potential of -6 kPa [θ(6)] for the same soils used in the present study. The objective of this study was to use soil moisture data below a double ring infiltrometer measured 48 h after the end of the infiltration test in order to develop PTFs for standard in situ FC. We found that such ring FC data were an average of 0.03 m³ m- 3 greater than standard FC values. The linear PTF that was developed for the ring FC data based only on θ(6) was nearly as accurate as the equivalent PTF reported by Ottoni Filho et al. (2014), which was developed for the standard FC data. The root mean squared residues of FC determined from both PTFs were about 0.02 m³ m- 3. The proposed method has the advantage of estimating the soil in situ FC using the water applied in the infiltration test.

Comparing the artificial neural network with parcial least squares for prediction of soil organic carbon and pH at different moisture content levels using visible and near-infrared spectroscopy

Visible and near infrared (vis-NIR) spectroscopy is widely used to detect soil properties. The objective of this study is to evaluate the combined effect of moisture content (MC) and the modeling algorithm on prediction of soil organic carbon (SOC) and pH. Partial least squares (PLS) and the Artificial neural network (ANN) for modeling of SOC and pH at different MC levels were compared in terms of efficiency in prediction of regression. A total of 270 soil samples were used. Before spectral measurement, dry soil samples were weighed to determine the amount of water to be added by weight to achieve the specified gravimetric MC levels of 5, 10, 15, 20, and 25 %. A fiber-optic vis-NIR spectrophotometer (350-2500 nm) was used to measure spectra of soil samples in the diffuse reflectance mode. Spectra preprocessing and PLS regression were carried using Unscrambler® software. Statistica® software was used for ANN modeling. The best prediction result for SOC was obtained using the ANN (RMSEP = 0.82 % and RPD = 4.23) for soil samples with 25 % MC. The best prediction results for pH were obtained with PLS for dry soil samples (RMSEP = 0.65 % and RPD = 1.68) and soil samples with 10 % MC (RMSEP = 0.61 % and RPD = 1.71). Whereas the ANN showed better performance for SOC prediction at all MC levels, PLS showed better predictive accuracy of pH at all MC levels except for 25 % MC. Therefore, based on the data set used in the current study, the ANN is recommended for the analyses of SOC at all MC levels, whereas PLS is recommended for the analysis of pH at MC levels below 20 %.

USE OF SCALED SEMIVARIOGRAMS IN THE PLANNING SAMPLE OF SOIL PHYSICAL PROPERTIES IN SOUTHERN AMAZONAS, BRAZIL

There is a great lack of information from soil surveys in the southern part of the State of Amazonas, Brazil. The use of tools such as geostatistics may improve environmental planning, use and management. In this study, we aimed to use scaled semivariograms in sample design of soil physical properties of some environments in Amazonas. We selected five areas located in the south of the state of Amazonas, Brazil, with varied soil uses, such as forest, archaeological dark earth (ADE), pasture, sugarcane cropping, and agroforestry. Regular mesh grids were set up in these areas with 64 sample points spaced at 10 m from each other. At these points, we determined the particle size composition, soil resistance to penetration, moisture, soil bulk density and particle density, macroporosity, microporosity, total porosity, and aggregate stability in water at a depth of 0.00-0.20 m. Descriptive and geostatistical analyses were performed. The sample density requirements were lower in the pasture area but higher in the forest. We concluded that managed-environments had differences in their soil physical properties compared to the natural forest; notably, the soil in the ADE environment is physically improved in relation to the others. The physical properties evaluated showed a structure of spatial dependence with a slight variability of the forest compared to the others. The use of the range parameter of the semivariogram analysis proved to be effective in determining an ideal sample density.

USE OF SCALED SEMIVARIOGRAMS IN THE PLANNING SAMPLE OF SOIL CHEMICAL PROPERTIES IN SOUTHERN AMAZONAS, BRAZIL

The lack of information concerning the variability of soil properties has been a major concern of researchers in the Amazon region. Thus, the aim of this study was to evaluate the spatial variability of soil chemical properties and determine minimal sampling density to characterize the variability of these properties in five environments located in the south of the State of Amazonas, Brazil. The five environments were archaeological dark earth (ADE), forest, pasture land, agroforestry operation, and sugarcane crop. Regular 70 × 70 m mesh grids were set up in these areas, with 64 sample points spaced at 10 m distance. Soil samples were collected at the 0.0-0.1 m depth. The chemical properties of pH in water, OM, P, K, Ca, Mg, H+Al, SB, CEC, and V were determined at these points. Data were analyzed by descriptive and geostatistical analyses. A large part of the data analyzed showed spatial dependence. Chemical properties were best fitted to the spherical model in almost all the environments evaluated, except for the sugarcane field with a better fit to the exponential model. ADE and sugarcane areas had greater heterogeneity of soil chemical properties, showing a greater range and higher sampling density; however, forest and agroforestry areas had less variability of chemical properties.