Using Bayesian networks with rule extraction to infer the risk of weed infestation in a corn-crop

This paper describes the modeling of a weed infestation risk inference system that implements a collaborative inference scheme based on rules extracted from two Bayesian network classifiers. The first Bayesian classifier infers a categorical variable value for the weed-crop competitiveness using as input categorical variables for the total density of weeds and corresponding proportions of narrow and broad-leaved weeds. The inferred categorical variable values for the weed-crop competitiveness along with three other categorical variables extracted from estimated maps for the weed seed production and weed coverage are then used as input for a second Bayesian network classifier to infer categorical variables values for the risk of infestation. Weed biomass and yield loss data samples are used to learn the probability relationship among the nodes of the first and second Bayesian classifiers in a supervised fashion, respectively. For comparison purposes, two types of Bayesian network structures are considered, namely an expert-based Bayesian classifier and a naive Bayes classifier. The inference system focused on the knowledge interpretation by translating a Bayesian classifier into a set of classification rules. The results obtained for the risk inference in a corn-crop field are presented and discussed. (C) 2009 Elsevier Ltd. All rights reserved.

Monitoring Services for Industrial Applications

Sao Paulo Research Foundation (FAPESP) in Brazil

In-Line Monitoring of Particle Size during Emulsion Polymerization under Different Operational Conditions using NIR Spectroscopy

In the present work, the sensitivity of NIR spectroscopy toward the evolution of particle size was studied during emulsion homopolymerization of styrene (Sty) and emulsion copolymerization of vinyl acetate-butyl acrylate conducted in a semibatch stirred tank and a tubular pulsed sieve plate reactor, respectively. All NIR spectra were collected online with a transflectance probe immersed into the reaction medium. The spectral range used for the NIR monitoring was from 9 500 to 13 000 cm(-1), where the absorbance of the chemical components present is minimal and the changes in the NIR spectrum can be ascribed to the effects of light scattering by the polymer particles. Off-line measurements of the average diameter of the polymer particles by DLS were used as reference values for the development of the multi-variate NIR calibration models based on partial least squares. Results indicated that, in the spectral range studied, it is possible to monitor the evolution of the average size of the polymer particles during emulsion polymerization reactions. The inclusion of an additional spectral range, from 5 701 to 6 447 cm(-1), containing information on absorbances (""chemical information"") in the calibration models was also evaluated.

Biogeochemical cycles of nutrients in tropical Eucalyptus plantations Main features shown by intensive monitoring in Congo and Brazil

The sustainability of fast-growing tropical Eucalyptus plantations is of concern in a context of rising fertilizer costs, since large amounts of nutrients are removed with biomass every 6-7 years from highly weathered soils. A better understanding of the dynamics of tree requirements is required to match fertilization regimes to the availability of each nutrient in the soil. The nutrition of Eucalyptus plantations has been intensively investigated and many studies have focused on specific fluxes in the biogeochemical cycles of nutrients. However, studies dealing with complete cycles are scarce for the Tropics. The objective of this paper was to compare these cycles for Eucalyptus plantations in Congo and Brazil, with contrasting climates, soil properties, and management practices. The main features were similar in the two situations. Most nutrient fluxes were driven by crown establishment the two first years after planting and total biomass production thereafter. These forests were characterized by huge nutrient requirements: 155, 10, 52, 55 and 23 kg ha(-1) of N, P, K, Ca and Mg the first year after planting at the Brazilian study site, respectively. High growth rates the first months after planting were essential to take advantage of the large amounts of nutrients released into the soil solutions by organic matter mineralization after harvesting. This study highlighted the predominant role of biological and biochemical cycles over the geochemical cycle of nutrients in tropical Eucalyptus plantations and indicated the prime importance of carefully managing organic matter in these soils. Limited nutrient losses through deep drainage after clear-cutting in the sandy soils of the two study sites showed the remarkable efficiency of Eucalyptus trees in keeping limited nutrient pools within the ecosystem, even after major disturbances. Nutrient input-output budgets suggested that Eucalyptus plantations take advantage of soil fertility inherited from previous land uses and that long-term sustainability will require an increase in the inputs of certain nutrients. (C) 2009 Elsevier B.V. All rights reserved.

Tree-ring characterization of Araucaria columnaris Hook and its applicability as a lead indicator in environmental monitoring

Tree-rings have frequently been used for dating of trees and to determine annual growth increments and forest dynamics, but little is known in tropical conditions about their utilization for environmental monitoring. This paper presents the results of Araucaria columnaris tree-ring characterization by wood anatomy and X-ray densitometric analysis and the determination of Pb concentration. Core samples from twelve araucaria trees were extracted from two sites exposed to air pollution due to intense traffic of vehicles and industrial activities. The tree-rings distinctly presented radial variation in early-latewood thickness and density, and characteristics of juvenile and mature wood. Anatomical and X-ray densitometric analysis were useful to delimit the tree-ring boundaries and to date the tree-rings, as well as to prove the annual formation. The lead concentration in annual araucaria tree-rings, analyzed with graphite furnace atomic absorption spectrometry, indicated the seasonal presence of the heavy metal in the environment during the 30 years studied, although the Pb did not affect tree growth. (c) 2008 Elsevier GmbH. All rights reserved.

Leaching of nutrients from a sugarcane crop growing on an Ultisol in Brazil

Leaching is disadvantageous, both for economical and environmental reasons since it may decrease the ecosystem productivity and may also contribute to the contamination of surface and ground water. The objective of this paper was to quantify the loss of nitrogen and sulfur by leaching, at the depth of 0.9 m, in an Ultisol in Sao Paulo State (Brazil) with high permeability, Cultivated with sugarcane during the agricultural cycle of crop plant. The following ions were evaluated: nitrite, nitrate, ammonium, and sulfate. Calcium, magnesium, potassium, and phosphate were also evaluated at the same depth. The sugarcane was planted and fertilized in the furrows with 120 log ha(-1) of N-urea. In order to find out the fate of N-fertilizer, four microplots with (15)N-enriched fertilizer were installed. Input and output of the considered ions at the depth of 0.9 m were quantified from the flux density of water and the concentration of the elements in the soil solution at this soil depth: tensiometers, soil water retention curve and soil solution extractors were used for this quantification. The internal drainage was 205 mm of water, with a total loss of 18 kg ha(-1) of N and 10 kg ha(-1) of S. The percentage of N in the soil solution derived from the fertilizer (%NSSDF) was 1.34, resulting in only 25 g ha(-1) of N fertilizer loss by leaching during all agricultural cycle. Under the experimental conditions of this crop plant, that is, high demand of nutrients and high incorporation of crop residues, the leached N represented 15% of applied N and S leaching were not considerable; the higher amount of leached N was native nitrogen and a minor quantity from N fertilizer; and the leached amount of Ca, Mg, K and P did not exceed the applications performed in the crop by lime and fertilization. (C) 2009 Elsevier B.V. All rights reserved.

Bayesian ratemaking procedure of crop insurance contracts with skewed distribution

Over the years, crop insurance programs became the focus of agricultural policy in the USA, Spain, Mexico, and more recently in Brazil. Given the increasing interest in insurance, accurate calculation of the premium rate is of great importance. We address the crop-yield distribution issue and its implications in pricing an insurance contract considering the dynamic structure of the data and incorporating the spatial correlation in the Hierarchical Bayesian framework. Results show that empirical (insurers) rates are higher in low risk areas and lower in high risk areas. Such methodological improvement is primarily important in situations of limited data.

Spatio-temporal modeling of agricultural yield data with an application to pricing crop insurance contracts

This article presents a statistical model of agricultural yield data based on a set of hierarchical Bayesian models that allows joint modeling of temporal and spatial autocorrelation. This method captures a comprehensive range of the various uncertainties involved in predicting crop insurance premium rates as opposed to the more traditional ad hoc, two-stage methods that are typically based on independent estimation and prediction. A panel data set of county-average yield data was analyzed for 290 counties in the State of Parana (Brazil) for the period of 1990 through 2002. Posterior predictive criteria are used to evaluate different model specifications. This article provides substantial improvements in the statistical and actuarial methods often applied to the calculation of insurance premium rates. These improvements are especially relevant to situations where data are limited.

Parametric and nonparametric statistical modelling of crop yield: implications for pricing crop insurance contracts

This article considers alternative methods to calculate the fair premium rate of crop insurance contracts based on county yields. The premium rate was calculated using parametric and nonparametric approaches to estimate the conditional agricultural yield density. These methods were applied to a data set of county yield provided by the Statistical and Geography Brazilian Institute (IBGE), for the period of 1990 through 2002, for soybean, corn and wheat, in the State of Paran. In this article, we propose methodological alternatives to pricing crop insurance contracts resulting in more accurate premium rates in a situation of limited data.

Crop rotation in center-pivot for phytonematode control: density variation, pathogenicity and crop loss estimation

Crop rotation in center-pivot for phytonematode control: density variation, pathogenicity and crop loss estimation A field study conducted over three consecutive years, on a farm using crop rotation system under center-pivot and infested with the nematodes Pratylenchus brachyurus, P. zeae, Meloidogyne incognita, Paratrichodorus minor, Helicotylenchus dihystera, Mesocriconema ornata and M. onoense, demonstrated that intensive crop systems provide conditions for the maintenance of high densities of polyphagous phytonematodes. Of the crops established on the farm (cotton, maize, soybean and cowpea), cotton and soybean suffered the most severe crop losses, caused respectively by M. incognita and P. brachyurus. Since maize is a good host for both nematodes, but tolerant of M. incognita, its exclusion from cropping system would be favorable to the performance of cotton, soybean and cowpea. Results from experiments carried out in controlled conditions confirmed the pathogenicity of P. brachyurus on cotton. Additional management with genetic resistance was useful in fields infested with M. incognita, although the soybean performance was affected by low resistance of the cultivars used for P. brachyurus. In conclusion, crop rotation must be carefully planned in areas infested with polyphagous nematodes, specifically in the case of occurrence of two or more major pathogenic nematodes.

Peach brown rot incidence related to pathogen infection at different stages of fruit development in an organic peach production system

Brown rot, caused by Monilinia fructicola, is the most widespread disease for organic peach production systems in Brazil. The objective of this study was to determine the favorable periods for latent infection by M. fructicola in organic systems. The field experiment was carried out during 2006, 2007 and 2008 using the cultivar Aurora. After thinning fruits were bagged using white paraffin bags, and the treatments were performed by removing the bags and exposing the fruit for four days to the natural infection during each of seven fruit stages from pit hardening to harvest. Throughout the entire growing season, the conidial density and the weather variables were measured and related to the disease incidence using multiple regression analyses. At the fourth day after harvest in each season, the cumulative disease incidence was assessed, and it ranged from 40 to 98%. The incidence of brown rot on fruit that were exposed during the embryo growing stage was lower than that of unbagged fruit throughout the entire season in 2006 and 2008. The relative humidity and the conidia density were significantly correlated to disease incidence. Based on our results, M. fructicola can infect peaches during any stage of fruit development, and control of the disease must be revised to account for organic peach production systems. (C) 2011 Elsevier Ltd. All rights reserved.

Water use and crop coefficient for hybrid DKB 390

Water use and crop coefficient for hybrid DKB 390. This work aims to characterize the water use of maize hybrid DKB 390 under suitable conditions of irrigation for both sufficient and below-optimal situations of nitrogen supply. Crop coefficient values for different stages are also presented as a result, in order to provide the basis for crop water budget and management throughout the cycle. A field experiment was carried Out during the main season, in which biomass, soil moisture, leaf area, climate data and light transmittance were evaluated. These have allowed deriving water balance, use and efficiency. The mentioned genotype requires around 600 nun for high yield targets, being less efficient when led under below-optimal nitrogen fertilization.

Carbon and Nitrogen Dynamics in an Oxisol as Affected by Liming and Crop Residues under No-Till

The short-term effects of surface lime application and black oat (Avena strigosa Schreb.) residues, with or without N fertilization, were evaluated in a long-term no-till (NT) system on a sandy clay loam, a kaolinitic, thermic Typic Hapludox from the state of Parana, Brazil. The main plot treatments were: control and dolomitic lime applied on soil surface at 8 Mg ha(-1). Three treatments with crop residues were evaluated on the subplots: (i) fallow, (ii) black oat residues, and (iii) black oat residues aft er N fertilization at 180 kg ha(-1). Black oat dry biomass was not affected by the treatments during 3 yr. Surface liming increased soil pH, microbial biomass, microbial activity, and bacterial/fungal ratio at the soil surface (0-5 cm), resulting in increased amino acid turnover, water-soluble humic substances formation, and N mineralization and nitrification. While the application of black oat did increase the soil pH, overall it had much less effect on soil biological processes and C and N pools than did lime. We concluded that black oat cannot replace the need for lime to optimize crop production in these tropical NT systems. In the long term, however, black oat should aid in the amelioration of acidity and replenishment of soil organic C pools and should help reduce erosion. Overall, this study suggests that overapplication of inorganic fertilizer N may occur in some tropical NT systems. Further experiments are required in NT systems to investigate the use of slow-release N fertilizers in combination with lime and black oat as a mechanism to reduce acidification and promote sustainability.

Sodicity and salinity in a Brazilian Oxisol cultivated with sugarcane irrigated with wastewater

Changes in soil sodicity-salinity parameters are one of the most characteristic alterations after treated sewage effluent (TSE) irrigation in agro-systems. Considering the importance of these parameters for agricultural management, as well as the economical value of sugarcane for Brazil, the present study aimed at evaluating effects on soil sodicity and salinity under tropical conditions over 16 months of TSE irrigation in a sugarcane plantation at Lins, Sao Paulo State, Brazil. Soil samplings were carried out in February 2005 (before planting), December 2005 (after 8 months of TSE irrigation) and September 2006 (after 16 months of TSE irrigation) following a complete block design with four treatments and four replicates. Treatments consisted of. (i) control, without TSE irrigation; (ii) T100, T150 and T200, with TSE irrigation supplying 100% (0% surplus, total of 2524 mm), 150% (50% surplus, total of 3832 mm) and 200% (100% surplus, total of 5092 mm) of crop water demand, respectively. Compared to initial soil conditions, at the end of the experiment increases of exchangeable sodium (from 2.4 to 5.9 mmol(c) kg(-1)), exchangeable sodium percentage (ESP) (from 8 to 18%), soluble Na (from 1.4 to 4.7 mmol L(-1)) and sodium adsorption ratio (SAR) of soil solution (from 3.6 to 12.6 (mmol were found in the soil profile (0-100 cm) as an average for the irrigated plots due to high SAR of TSE. Associated with the increments were mostly significant increases in clay dispersion rates at depths 0-10, 10-20 and 20-40 cm. Electrical conductivity (EC) of soil solution increased during the TSE irrigation period whereas at the end of the experiment, after short term discontinuation of irrigation and harvest, EC in the topsoil (0-10 and 10-20 cm) decreased compared to the previous samplings. Moreover, despite increasing sodicity over time mainly insignificant differences within the different irrigated treatments were found in December 2005 and September 2006. This suggests that independent of varying irrigation amounts the increasing soil sodicity over time were rather caused by the continuous use of TSE than by its quantity applied. Moreover, also plant productivity showed no significant differences within the TSE irrigated plots. The study indicates that monitoring as well as remediation of soil after TSE irrigation is required for a sustainable TSE use in order to maintain agricultural quality parameters. (C) 2008 Elsevier B.V. All rights reserved.

Evapotranspiration and crop coefficients of seedless watermelon

The research was developed in Fortaleza, CE (latitude 3 degrees 43`S, longitude of 38 degrees 32`W and altitude of 19.5 m). It aimed to determine the Evapotranspiration and the crop coefficients for the different stadiums of seedless watermelon (Citrullus lanatus, Schrad), using the water balance method in drainage lysimeter. The experimental unit consisted of four located drainage lysimeter in an area of 600 m(2). The culture Evapotranspiration (ETc) was determined by the method of the water balance, in a volume of soil control with depth of 0.45 m, whose humidity was monitored by tensiometers. The reference Evapotranspiration (ETo) was estimated by the method of the class A pan. The crop coefficients were obtained from the rate between ETc and ETo in the different stages. The ETc presented growing values from the vegetative stage to the fruiting stage, dropping straight away in the mature stage. The maximum value of kc was observed in the fruiting stage.