Risk prediction for weed infestation using classification rules

This paper proposes a fuzzy classification system for the risk of infestation by weeds in agricultural zones considering the variability of weeds. The inputs of the system are features of the infestation extracted from estimated maps by kriging for the weed seed production and weed coverage, and from the competitiveness, inferred from narrow and broad-leaved weeds. Furthermore, a Bayesian network classifier is used to extract rules from data which are compared to the fuzzy rule set obtained on the base of specialist knowledge. Results for the risk inference in a maize crop field are presented and evaluated by the estimated yield loss. © 2009 IEEE.

Variabilidade temporal e mapeamento dos dados climáticos de Botucatu-SP

O trabalho teve objetivo estudar a variabilidade temporal da temperatura do ar, precipitação pluviométrica e umidade relativa do ar na cidade de Botucatu-SP, Brasil, utilizando técnicas geoestatísticas. Os dados de precipitação pluviométrica, temperatura do ar e umidade relativa do ar utilizados no presente estudo são provenientes da Estação Meteorológica da Fazenda Lageado, da Faculdade de Ciências Agronômicas-UNESP. As observações foram realizadas no período de 1988 a 2007, referem-se ao total mensal de precipitação pluvial expressa em altura de lâmina d'água (mm), médias mensais de temperatura em ºC e umidade relativa em %. Os dados foram avaliados por meio da estatística clássica e geoestatística. As variáveis climáticas tiveram sua dependência verificada por variogramas, apresentando dependência temporal maior que 76%. A série temporal de umidade relativa do ar foi a que apresentou maior alcance (8,67 meses) e, conseqüentemente, maior estabilidade climática. O conhecimento da distribuição temporal das variáveis climáticas é importante para o estudo e realização do zoneamento agroclimático, bem como para o dimensionamento do sistema de irrigação das culturas.

Métodos de interpolação para estimar o ph em solo sob dois manejos de café Arábica

There are many methods used to estimate values in places no sampled for construction of contours maps. The aim of this study was to use the methods of interpolation kriging, inverse of the square of the distance and polynomial in the representation of the spatial variability of the pH of the soil in the organic and conventional management in the culture of the coffee plantation. For that, irregular meshes were built for soil sampling in the depth of 0-0,10 meters, totaling 40 points sampling in each area. For gauging of the interpolation methods they were solitary 10% of the total of points, for each area. Initially, the data were appraised through the classic statistics (descriptive and exploratory) and spatial analysis. The method inverse square of the distance and kriging has low error in estimating dados. The method of kriging presented low variation around the average in different managements.

Delimitação do Embasamento da Bacia de Taubat

The aim of this study is to define the basement map of the Taubaté Basin applying geostatistics to seismic, gravimetric and deep wells data. The study consisted of the interpretation of eleven seismic sections obtained in the central and northeastern portions of the basin. The altitude of the basement and the distribution of faults was determined based on these sections. New information was obtained from 79 wells located mainly in the regions of São José dos Campos and Jacareí. The method of kriging with an external drift was applied, using seismic and well data as the main variables and the gravimetric map as the secondary variable. The basement contour map obtained has a strong correlation with the main faults. It was possible to obtain a better resolution in the region of São José dos Campos and in the northeast area, where the vast majority of wells are located.

Variabilidade dos atributos de um latossolo vermelho sob plantio direto no cerrado brasileiro e produtividade da soja

The spatial variability of several soil attributes (bulk density, penetration resistance, water content, organic matter content and pH) as well as soybean yield have been assessed during the 2007/08 growing season, in Selviria (MS) in a Hapludox (Typic Acrustox), under no tillage. The objectives were to assess the spatial variability of soil and plant parameters at the small plot scale and to select the best soil attribute explaining most the variability of agricultural productivity. Soil and plant were sampled on a grid with 121 points within a plot of 25,600 m 2 in area and slope of 0.025 mm -1 slope. Medium and low coefficients of variation were obtained for most of the studied soil attributes as expected, due to the homogenizing effect of the no-till system on the soil physical environment. From the standpoint of linear regression and spatial pattern of variability, productivity of soybeans could be explained according to the hydrogen potential (pH). Results are discussed taken into account that the soybean crop in no-tillage is widely used in crop-livestock integration on the national scene.

Espacialização das estimativas de contaminação de águas subterrâneas por Tebuthiuron

Improper use of pesticides can lead to significant environmental impacts such as, contamination of environmental compartments, being the aquatic compartments the most vulnerable. In this context, the spatialization of pesticides concentrations estimative in groundwater provides important insights for decision making in managing and monitoring pesticides use. This study aimed to spatialize estimatives of groundwater contamination by Tebuthiuron, from different irrigation depths in the Rio Pardo basin, Pardinho-SP, Brazil. The simulations were performed using the ARAquá computer system, considering 0mm, 200 mm and 400 mm annual irrigation depths. Geostatistical techniques were used to obtain the spatial distribution of the simulated estimative. Tebuthiuron maps estimating concentration in groundwater were obtained by Kriging interpolation method, and indicated the areas with high potential for groundwater contamination. Considering all the simulations, it was concluded that there was no risk of groundwater contamination by Tebuthiuron in the study area.

Submerged macrophytes height estimation by echosounder data sample

Traditional methods of submerged aquatic vegetation (SAV) survey last long and then, they are high cost. Optical remote sensing is an alternative, but it has some limitations in the aquatic environment. The use of echosounder techniques is efficient to detect submerged targets. Therefore, the aim of this study is to evaluate different kinds of interpolation approach applied on SAV sample data collected by echosounder. This study case was performed in a region of Uberaba River - Brazil. The interpolation methods evaluated in this work follow: Nearest Neighbor, Weighted Average, Triangular Irregular Network (TIN) and ordinary kriging. Better results were carried out with kriging interpolation. Thus, it is recommend the use of geostatistics for spatial inference of SAV from sample data surveyed with echosounder techniques. © 2012 IEEE.

Variabilidade espacial dos atributos físicos em um Argissolo Vermelho sob floresta

This study aim to evaluate the spatial variability of soil physical attributes in of Alfissol forest in Manicoré region, AM. The mapping of an agroforestry growing area of 70×70m was made using a regular grid sampling of 10 × 10m. In each grid, soil samples were collected at 0.0-0.2 and 0,40-0,60 m depth, with a total of 128 sample points. Physical analyses were made (texture, soil bulk and particles density, macro and microporosity, total porosity and aggregates stability in water). With the exception of DMG, DMP and class aggregates <1.00 mm had all the physical attributes spatial dependence structure ranging from moderate to weak. Values were above the range established in the mesh (12.00 to 45.56 m), enabling to make a basis for future studies in forest area. The physical attributes presented in kriging maps different spatial behavior, however there are relationships between these attributes shown by geostatistics proving that this is effective tool for studies in forest area.

Produção de matéria seca da braquiária de acordo com os atributos químicos de um Latossolo em Selvíria, Mato Grosso do Sul

Rustic forages as the signal-grass are predominant in areas of Cerrado in extensive livestock systems that favor soil degradation. However, with time, not even these forages can have a good development in those areas. The objective of this study was to analyze the variability of plant and soil attributes; to define the linear and spatial correlations between signal-grass yield and the chemical attributes of the studied soil, and to evaluate, among the chemical attributes of the soil, which one that best explain the variability in this forage yield. The experiment was conducted in an area that had been under pasture for more than 30 years, belonging to UNESP - Ilha Solteira Campus, located in Selvíria - MS. A geostatistical grid was installed in an oxisol, for soil and plant data collection, with 121 sampling stations, consisted of eleven transections with 160 m width in the direction of the Cartesian axes. The dry mass yield of signal-grass was low, presenting high variation. The attributes MSr, N, PB, MO1, MO2, pHa1, pHa2, pHk1 and pHk2 did not vary at random. They presented data variability from low to high and followed clearly defined spatial patterns, ranging between 17.7-162.9 m. There was significant linear correlation at 1% between MSr and N leaf, and between MSr and PB. The cross semivariograms MSr=f(N) and MSr=f(PB) confirmed that the dry matter can be estimated from data of nitrogen leaf and crude protein content of this forage.

Suscetibilidade magnética do solo e estimação da capacidade de suporte à aplicação de vinhaça

The objective of this work was to evaluate the magnetic susceptibility efficiency for estimating the support capacity of areas for vinasse application. Two hundred forty-one soil samples were collected from a 380-ha area, on which soil chemical properties, clay content, and magnetic susceptibility were determined. Vinasse requirement was calculated for each sample. Data were subjected to descriptive statistical analysis, and regression models were developed between magnetic susceptibility and the other evaluated attributes. The analysis of data spatial dependence was performed using geostatistics. Kriging maps and cross variograms were built in order to investigate the spatial correlation between soil magnetic susceptibility and studied attributes. Based on the map of vinasse requirement, on the soil classes, and on the kriging map, calculations were done for average vinasse dose and average soil support capacity, weighted by the area. Magnetic susceptibility has significant linear spatial correlation with recommended vinasse doses and soil support capacity for the application of this effluent, and it can be used as a pedotransfer function for indirect quantification of soil support capacity.

Inferência espacial para mapeamento de macrófitas submersas: estudo de caso

Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq)

Métodos para extração de informações a partir de imagens multiespectrais de escalas grandes

Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES)

Classificação fuzzy de vertentes por krigagem e TPS com agregação de regiões via diagrama de Voronoi

Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq)

Inferência espacial de cianobactérias a partir da fluorescência in vivo do pigmento ficocianina

Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES)

Métodos geoestatísticos na caracterização espacial de óxidos de ferro em diferentes pedoformas

Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES)