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.

Soil functioning in a toposequence under rainforest in São Paulo, Brazil

Studies of soil-water dynamics using toposequences are essential to improve the understanding of soil-water-vegetation relationships. This study assessed the hydro-physical and morphological characteristics of soils of Atlantic Rainforest in the Parque Estadual de Carlos Botelho, state of São Paulo, Brazil. The study area of 10.24 ha (320 x 320 m) was covered by dense tropical rainforest (Atlantic Rainforest). Based on soil maps and topographic maps of the area, a representative transect of the soil in this plot was chosen and five soil trenches were opened to determine morphological properties. To evaluate the soil hydro-physical functioning, soil particle size distribution, bulk density (r), particle density (r s), soil water retention curves (SWRC), field saturated hydraulic conductivity (Ks), macroporosity (macro), and microporosity (micro) and total porosity (TP) were determined. Undisturbed samples were collected for micromorphometric image analysis, to determine pore size, shape, and connectivity. The soils in the study area were predominantly Inceptisols, and secondly Entisols and Epiaquic Haplustult. In the soil hydro-physical characterization of the selected transect, a change was observed in Ks between the surface and subsurface layers, from high/intermediate to intermediate/low permeability. This variation in soil-water dynamics was also observed in the SWRC, with higher water retention in the subsurface horizons. The soil hydro-physical behavior was influenced by the morphogenetic characteristics of the soils.

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.

Optimization of the sampling scheme for maps of physical and chemical properties estimated by kriging

The sampling scheme is essential in the investigation of the spatial variability of soil properties in Soil Science studies. The high costs of sampling schemes optimized with additional sampling points for each physical and chemical soil property, prevent their use in precision agriculture. The purpose of this study was to obtain an optimal sampling scheme for physical and chemical property sets and investigate its effect on the quality of soil sampling. Soil was sampled on a 42-ha area, with 206 geo-referenced points arranged in a regular grid spaced 50 m from each other, in a depth range of 0.00-0.20 m. In order to obtain an optimal sampling scheme for every physical and chemical property, a sample grid, a medium-scale variogram and the extended Spatial Simulated Annealing (SSA) method were used to minimize kriging variance. The optimization procedure was validated by constructing maps of relative improvement comparing the sample configuration before and after the process. A greater concentration of recommended points in specific areas (NW-SE direction) was observed, which also reflects a greater estimate variance at these locations. The addition of optimal samples, for specific regions, increased the accuracy up to 2 % for chemical and 1 % for physical properties. The use of a sample grid and medium-scale variogram, as previous information for the conception of additional sampling schemes, was very promising to determine the locations of these additional points for all physical and chemical soil properties, enhancing the accuracy of kriging estimates of the physical-chemical properties.

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.

Effects of tillage systems on physical properties of a cohesive yellow argisol in the northern state of Espírito Santo, Brazil

Tillage systems are a key element of the technology of crop production, both with a view to crop yield and from the perspective of soil conservation and sustainability of the production system. The aim of this paper was to evaluate the effects of five tillage systems on the physical properties of a cohesive Yellow Argisol. The experiment was installed in the field on January 21, 2011 and lasted 260 days, in an area previously used as pasture with Brachiaria grass without liming or fertilization, but irrigated by a low pressure spray system. The treatments, in five replications and in a randomized block design, consisted of: 1) disk plow (twice) + disk harrow + ridge-furrow tillage (raising a ridge along the planting row), 135 days after transplanting (DP + RID); 2) disk plow (twice) + disk harrow (DP no RID); 3) subsoiler (SB); 4) disk plow (twice) + disk harrow + scarification with three shanks along the plant row (DP + SPR); and 5) disk plow (twice) + disk harrow + scarification with three shanks in the total area (DP + STA). In all tillage systems, furrows were mechanically opened for the papaya plants. After the treatments, the mechanical resistance to penetration was determined, followed by soil moisture, mean weight diameter (MWD), geometric mean diameter (GMD), bulk density (BD), macroporosity (Ma), microporosity (Mi), and number of fruits per plant. There were differences in penetration resistance (PR) between treatments. The subsoiler was more effective to decrease RP to a distance of 0.35 m from the plants, perpendicular to the plant row. The scarifier resulted in a lower PR than DP or SB, even at the depth of 0.40 m, and it was more effective at greater distances perpendicular to the plant. All tillage systems induced a PR between 2.0 and 3.0 MPa at the depth with the highest concentration of papaya tree roots (0-0.25 m), improving the physical conditions to this depth. There was no statistical difference among the treatments for BD, Ma, Mi, MWD, and GMD at a depth of 0.20 m. The disk plow changed the physical properties of the soil most intensely to a depth of 0.20 m. The use of scarification, reduced tillage with a forest subsoiler, or ridge-furrow tillage did not improve the physical properties in the rhizosphere. Reduced tillage with a forest subsoiler resulted in a lower number of fruits per plant than all other treatments, which did not differ from each other.

Physical properties of a Red-Yellow Latosol and productivity of a signalgrass pasture fertilized with increasing nitrogen doses

The physical properties and fertility of the soil are important factors in the formation and establishment of pasture. Changes in physical properties affect the movement of water, air, nutrients and roots, which, in turn, affect the productivity and longevity of pastures. The objective of this study was to evaluate the physical properties of the soil and the dry matter yield of a pasture with signalgrass cv. Basilisk (Brachiaria decumbens cv. Basilisk), fertilized with increasing nitrogen doses (N), on a dystrophic Red-Yellow Latosol. The experiment was conducted on the Fazenda Rio Manso of the Universidade Federal dos Vales do Jequitinhonha e Mucuri, in Couto de Magalhães de Minas, State of Minas Gerais, Brazil. To evaluate the annual forage yield, a split plot scheme in a randomized block design with four replications was used, with N doses (0, 50, 100, 150, and 200 kg/ha/year) in the plots and growing seasons (first and second) in the subplots. For soil evaluation, a split plot scheme was used with N doses (0, 25, 50, 75 and 100 kg/ha/cut) in the plots and three sampling times (prior to the experiment, at the end of the first growing season and at the end of the second growing season) in the subplots in a randomized block design with four replications. This analysis was performed separately at two soil depths (0-3 and 10-13 cm). Forage samples were analyzed for the annual dry matter yield (DMY), and soil samples were analyzed for pre-consolidation pressure (σp), initial soil bulk density (Bd), total pore volume (TPV) and void index (Vd). Higher nitrogen doses increased the dry matter yield of signalgrass pasture and the pre-consolidation pressure of the soil. The total pore volume and void index decreased, and the initial soil bulk density increased, though without promoting soil compaction.

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.

Quality of surface water related to land use: a case study in a catchment with small farms and intensive vegetable crop production in southern Brazil

Water degradation is strongly related to agricultural activity. The aim of this study was to evaluate the influence of land use and some environmental components on surface water quality in the Campestre catchment, located in Colombo, state of Parana, Brazil. Physical and chemical attributes were analyzed (total nitrogen, ammonium, nitrate, total phosphorus, electrical conductivity, pH, temperature, turbidity, total solids, biological oxygen demand, chemical oxygen demand and dissolved oxygen). Monthly samples of the river water were taken over one year at eight monitoring sites, distributed over three sub-basins. Overall, water quality was worse in the sub-basin with a higher percentage of agriculture, and was also affected by a lower percentage of native forest and permanent preservation area, and a larger drainage area. Water quality was also negatively affected by the presence of agriculture in the riparian zone. In the summer season, probably due to higher rainfall and intensive soil use, a higher concentration of total nitrogen and particulate nitrogen was observed, as well as higher electrical conductivity, pH and turbidity. All attributes, except for total phosphorus, were in compliance with Brazilian Conama Resolution Nº 357/2005 for freshwater class 1. However, it should be noted that these results referred to the base flow and did not represent a discharge condition since most of the water samples were not collected at or near the rainfall event.

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.

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.

Chemical and physical-hydric characterisation of a red latosol after five years of management during the summer between-crop season

Agricultural production systems that include the production of mulch for no-tillage farming and structural improvement of the soil can be considered key measures for agricultural activity in the Cerrado region without causing environmental degradation. In this respect, our work aimed to evaluate the chemical and physical-hydric properties of a dystrophic Red Latosol (Oxisol) in the municipality of Rio Verde, Goias, Brazil, under different soil management systems in the between-crop season of soybean cultivation five years after first planting. The following conditions were evaluated: Brachiaria brizantha cv. Marandu as a cover crop during the between-crop season; Second crop of maize intercropped with Brachiaria ruziziensis; Second crop of grain alone in a no-tillage system; Fallow soil after the soybean harvest; and Forest (natural vegetation) located in an adjacent area. Soil samples up to a depth of 40 cm were taken and used in the assessment of chemical properties and soil structure diagnostics. The results demonstrated that the conversion of native vegetation areas into agricultural fields altered the chemical and physical-hydric properties of the soil at all the depths evaluated, especially up to 10 cm, due to the activity of root systems in the soil structure. Cultivation of B. brizantha as a cover crop during the summer between-crop season increased soil water availability, which is important for agricultural activities in the region under study.

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.

Transfer of cadmium and barium from soil to crops grown in tropical soils

Phytotoxicity and transfer of potentially toxic elements, such as cadmium (Cd) or barium (Ba), depend on the availability of these elements in soils and on the plant species exposed to them. With this study, we aimed to evaluate the effect of Cd and Ba application rates on yields of pea (Pisum sativum L.), sorghum (Sorghum bicolor L.), soybean (Glycine max L.), and maize (Zea mays L.) grown under greenhouse conditions in an Oxisol and an Entisol with contrasting physical and chemical properties, and to correlate the amount taken up by plants with extractants commonly used in routine soil analysis, along with transfer coefficients (Bioconcentration Factor and Transfer Factor) in different parts of the plants. Plants were harvested at flowering stage and measured for yield and Cd or Ba concentrations in leaves, stems, and roots. The amount of Cd accumulated in the plants was satisfactorily evaluated by both DTPA and Mehlich-3 (M-3). Mehlich-3 did not relate to Ba accumulated in plants, suggesting it should not be used to predict Ba availability. The transfer coefficients were specific to soils and plants and are therefore not recommended for direct use in risk assessment models without taking soil properties and group of plants into account.

SOIL CHEMICAL PROPERTIES AND GROWTH OF SUNFLOWER (HELIANTHUS ANNUUS L.) AS AFFECTED BY THE APPLICATION OF ORGANIC FERTILIZERS AND INOCULATION WITH ARBUSCULAR MYCORRHIZAL FUNGI

The use of organic fertilizers and the inoculation of mycorrhizal fungi in the cultivation of oil crops is essential to reduce production costs and minimize negative impacts on natural resources. A field experiment was conducted in an Argissolo Amarelo (Ultisol) with the aim of evaluating the effects of fertilizer application and inoculation of arbuscular mycorrhizal fungi on the growth attributes of sunflower (Helianthus annuus L.) and on soil chemical properties. The experiment was conducted at the Federal University of Rio Grande do Norte, Brazil, using a randomized block design with three replicates in a 4 × 2 factorial arrangement consisting of four treatments in regard to application of organic fertilizer (liquid biofertilizer, cow urine, mineral fertilizer, and unfertilized control) and two treatments in regard to arbuscular mycorrhizal fungi (with and without mycorrhizal fungi). The results showed that the physiological attributes of relative growth rate and leaf weight ratio were positively influenced by fertilization, compared to the control treatment, likely brought about by the supply of nutrients from the fertilizers applied. The growth and productivity attributes were positively affected by mycorrhization.