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.

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.

Prediction of soil orders with high spatial resolution: response of different classifiers to sampling density

The objective of this work was to evaluate sampling density on the prediction accuracy of soil orders, with high spatial resolution, in a viticultural zone of Serra Gaúcha, Southern Brazil. A digital elevation model (DEM), a cartographic base, a conventional soil map, and the Idrisi software were used. Seven predictor variables were calculated and read along with soil classes in randomly distributed points, with sampling densities of 0.5, 1, 1.5, 2, and 4 points per hectare. Data were used to train a decision tree (Gini) and three artificial neural networks: adaptive resonance theory, fuzzy ARTMap; self‑organizing map, SOM; and multi‑layer perceptron, MLP. Estimated maps were compared with the conventional soil map to calculate omission and commission errors, overall accuracy, and quantity and allocation disagreement. The decision tree was less sensitive to sampling density and had the highest accuracy and consistence. The SOM was the less sensitive and most consistent network. The MLP had a critical minimum and showed high inconsistency, whereas fuzzy ARTMap was more sensitive and less accurate. Results indicate that sampling densities used in conventional soil surveys can serve as a reference to predict soil orders in Serra Gaúcha.

Soils associated with rock outcrops in the Brazilian mountain ranges Mantiqueira and Espinhaço

Espinhaço and Mantiqueira are two mountain ranges of great importance in Brazil. In the uppermost parts of these areas, unique ecosystems occur, generally associated with rock outcrops, they are collectively called High Altitude Rocky Complexes. These environments show distinct soil and biota characteristics in relation to the surrounding biome. The soils are generally shallow, coarse textured, with high Al3+ and varying amounts of organic matter. Entisols, Inceptsols and Histosols are dominant, directly associated with the rock outcrops, and forming a complex mosaic of soils. Some of these soils are endemic, based on peculiar conditions of parent materials, topography and vegetation, and this pedodiversity is important for detecting unique and endangered soils. In these soils, organic matter is highly humified, with a great amount of soluble forms and conspicuous presence of charcoal. Spodic horizons and dark water rivers are typically associated with quartzite and quartzite outcrops, formed by illuviation of organic compounds, being less common in granitic rocks. The very low nutrient content of these soils and other environmental limitations required the development of specific physiological and morphological plant adaptations. Most high altitude environments are unstable under current climatic conditions, and anthropic interventions may be accelerating this process. Detailed soil surveys are necessary for a better understanding of the role of these soils in ecological processes and for the development of adequate conservation policies.

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.

Climate, soil and tree flora relationships in forests in the state of São Paulo, southeastern Brasil

ABSTRACT - (Climate, soil and tree flora relationships in forests in the state of São Paulo, southteastern Brasil). With the aim of verifying possible influences of abiotic features on the spatial distribution of forest tree species and families, thirteen surveys in the state of São Paulo were selected, representing different conditions (localization at the extreme coordenates and altitudes, succesional stages, surveying methods). By applying Jaccard's Index to the binary matrices of 806 synonymized specific binomina and 79 families (Cronquist's system) phenograms were constructed using the method of the unweighted pair grouping by mathematical average (UPGMA). The species formed two floristic blocks: hygrophyllous (yearly rainfall greater than 2000 mm without dry season) and mesophyllous (yearly rainfall about 1400 mm with variable dry season). The latter was divided in two other groups: the high-altitudinal (median altitudes higher than 750 m, frost average frequency greater than 3 days/year) and low-altitudinal. Both mesophyllous floristic blocks were subdivided according to soil conditions (texture, eutrophism, acid or allic dystrophism, iron content). At the family level the relations were weak, but also showed the soil nutritional status as a possible constraint to the spatial partition of families.

Soil profile internal drainage for a central pivot fertigated coffee crop

Coffee cultivation via central-pivot fertigation can lead to fertilizer losses by soil profile internal drainage when water application is excessive and soils have low water retention and cation adsorption capacities. This study analyses the deep water losses from the top 1 m sandy soil layer of east Bahia, Brazil, cultivated with coffee at a high technology level (central-pivot fertigation), using above normal N fertilizer rates. The deep drainage (Q) estimation is made through the application of a climatologic water balance (CWB) program having as input direct measures of irrigation and rainfall, climatological data from weather stations, and measured soil water retention characteristics. The aim of the study is to contribute to the understanding of the hydric regime of coffee crops managed by central-pivot irrigation, analyzing three scenarios (Sc): i) rainfall only, ii) rainfall and irrigation full year, and iii) rainfall and irrigation dry season only. Annual Q values for the 2008/2009 agricultural year were: Sc i = 811.5 mm; Sc ii = 1010.5 mm; and Sc iii = 873.1 mm, so that the irrigation interruption in the wet season reduced Q by 15.7%, without the appearance of water deficit periods. Results show that the use of the CWB program is a convenient tool for the evaluation of Q under the cited conditions.

Availability of heavy metals in contaminated soil evidenced by chemical extractants

Heavy metals have been accumulating in Brazilian soils, due to natural processes, such as atmospheric deposition, or human industrial activities. For certain heavy metals, when in high concentrations in the soil, there is no specific extractant to determine the availability of these elements in the soil. The objective of the present study was to evaluate the availability of Cd, Cu, Fe, Mn, Pb and Zn for rice and soybeans, using different chemical extractants. In this study we used seven soil samples with different levels of contamination, in completely randomized experimental design with four replications. We determined the available concentrations of Cd, Cu, Fe, Mn, Pb and Zn extracted by Mehlich-1, HCl 0.1 mol L-1, DTPA, and organic acid extractants and the contents in rice and soybeans, which extracts were analyzed by ICP-OES. It was observed that Mehlich-1, HCl 0.1 mol L-1 and DTPA extractants were effective to assess the availability of Cd, Cu, Pb and Zn for rice and soybeans. However, the same was not observed for the organic acid extractant.

Effect of cattle manure on sunflower production and water use in two types of soil

The aim of the present study was to evaluate water consumption, use efficiency and yield components of sunflower variety Embrapa 122 V/2000 cultivated in two types of soil (Fluvissol and Haplic Luvisol) subjected to increasing doses of cattle manure. The experiment was carried out in a greenhouse at Universidade Estadual da Paraíba. The experimental design was completely randomized in a factorial scheme. The irrigation was performed every other day, replacing the water absorbed by the plants. The water consumption and the use efficiency were evaluated, being the use efficiency determined by the ratio of the total dry mass of sunflower and the amount of water used to produce it in each treatment. Plants were harvested at 95 days after sowing when the following parameters were evaluated: number of seeds per plant, weight of seeds per plant, weight of 1000 seeds and the outer diameter of the capitulum (head). The results showed that the sunflower was positively affected by cattle manure application, increasing the production components and the water use efficiency, regardless of the type of soil. Excepting for the 1000 seeds weight and the water use efficiency, the type of soil affected significantly the water use, the number and weight of seeds per plant. The plants cultivated in Haplic Luvisol had a better performance.

Sugarcane productivity correlated with physical-chemical attributes to create soil management zone

The socioeconomic importance of sugar cane in Brazil is unquestionable because it is the raw material for the production of ethanol and sugar. The accurate spatial intervention in the management of the crop, resulting zones of soil management, increases productivity as well as its agricultural yields. The spatial and Person's correlations between sugarcane attributes and physico-chemical attributes of a Typic Tropustalf were studied in the growing season of 2009, in Suzanápolis, State of São Paulo, Brazil (20°28'10'' S lat.; 50°49'20'' W long.), in order to obtain the one that best correlates with agricultural productivity. Thus, the geostatistical grid with 120 sampling points was installed to soil and data collection in a plot of 14.6 ha with second crop sugarcane. Due to their substantial and excellent linear and spatial correlations with the productivity of the sugarcane, the population of plants and the organic matter content of the soil, by evidencing substantial correlations, linear and spatial, with the productivity of sugarcane, were indicators of management zones strongly attached to such productivity.

Heavy metal toxicity in rice and soybean plants cultivated in contaminated soil

Heavy metals can accumulate in soil and cause phytotoxicity in plants with some specific symptoms. The present study evaluated the specific symptoms on rice and soybeans plants caused by excess of heavy metals in soil. Rice and soybean were grown in pots containing soil with different levels of heavy metals. A completely randomized design was used, with four replications, using two crop species and seven sample soils with different contamination levels. Rice and soybean exhibited different responses to the high concentrations of heavy metals in the soil. Rice plants accumulated higher Cu, Mn, Pb and Zn concentrations and were more sensitive to high concentrations of these elements in the soil, absorbing them more easily compared to the soybean plants. However, high available Zn concentrations in the soil caused phytotoxicity symptoms in rice and soybean, mainly chlorosis and inhibited plant growth. Further, high Zn concentrations in the soil reduced the Fe concentration in the shoots of soybean and rice plants to levels considered deficient.

Nutrient accumulation and biomass production of alfafa after soil amendment with silicates

Studies on the use of silicate correctives in agriculture show that they have great potential to improve soil chemical characteristics, however, little information is available on the reactivity rates of their particle-size fractions. This study investigated whether the reactivity rates obtained experimentally could be considered in the calculation of ECC (effective calcium carbonate) for soil liming, promoting adequate development of alfalfa plants. Six treatments were evaluated in the experiment, consisting of two slag types applied in two rates. The experimental ECC was used to calculate one of the rates and the ECC determined in the laboratory was used to calculate the other. Rates of limestone and wollastonite were based on the ECC determined in laboratory. The rates of each soil acidity corretive were calculated to increase the base saturation to 80%. The treatments were applied to a Rhodic Hapludox and an Alfisol Ferrudalfs. The methods for ECC determination established for lime can be applied to steel slag. The application of slag corrected soil acidity with consequent accumulation of Ca, P, and Si in alfalfa, favoring DM production.

Soil chemical properties affected by cover crops under no-tillage system

The use of cover crops in no-tillage systems (NTS) can significantly improve the soil's fertility. Thus, a study was performed to evaluate changes in chemical properties of soil caused by cover crops in a no-tillage system. The field experiment consisted of the following crop rotation: cover crops/rice/cover crops/rice. The experimental design was in randomized blocks with three replications. Treatments consisted of four cover crops (Brachiaria brizantha(Hochst. ex A. Rich.) Stapf. cv. Marandu, Brachiaria ruziziensis R. Germ. and C.M. Evrard, Panicum maximum Jacq. cv. Colonião, and Pennisetum glaucum(L.) R. Br. cv. BN-2) and fallow (control treatment). Soil samples were collected at the beginning of the summer crop in Oct 2007, Oct 2008 and Oct 2009 at 0-5 cm soil depth. The use of cover crops provided for a significant increase in the level of nutrients, soil organic matter, cation exchange capacity, and base saturation in the soil. Soil fertility improved from the first to second year with the growing of cover crops. The soil under cover crops P. glaucum, B. ruziziensis, and B. brizantha showed higher fertility than the area under fallow.

Absorption of primary macronutrients and soybean growth at different compactation densities and moisture levels in a silt loam soil

ABSTRACT Soybean cultivation is increasing rapidly in the region of Alto Vale do Itajaí, State of Santa Catarina, where there is a predominance of silt soils. The objective of this work was to evaluate the content of primary macronutrients in shoots and shoot and root vegetative growth of soybean (Glicine max L. Merrill) grown in a silt-loam soil under different compactation densities and moisture levels. A randomized block design in a 4x4 factorial arrangement was used, with four compactation densities: 1.00; 1.20; 1.40 and 1.60 Mg m-3, and four soil moisture levels: 0.130; 0.160; 0.190 and 0.220 kg kg-1 and four replications. Each pot consisted of the overlapping of three 150-mm PVC rings, where soil was maintained in the higher and lower part of the pot with a density of 1.00 Mg m-3 and in the intermediate ring, the compactation densities were increased. Values of soil density higher than 120 Mg m-3 negatively affected N, P and K uptake by soybean plants, as well as the plant mass of the shoots and roots. The higher levels of soil moisture reduced the compaction effect and promoted better absorption of P and K.

Sample surveys via electronic mail: a comprehensive perspective

Although postal questionnaires, personal interviewing, and telephone interviewing are the main methods of survey-based research, there is an increasing use of e-mail as a data collection medium. However, little, if any, published Western research in general and that of Turkish in particular have investigated e-mail survey technique from pure survey research perspective. Attempting to develop a framework to assess e-mail as a data collection mean, the purpose of this study is to explore e-mail-based questionnaire technique from complementary angles. To this goal, sample representativeness, data quality, response rates, and advantages and disadvantages of e-mail surveying are discussed.