There is a high correlation between sward height and pasture sward structure. Therefore, in tropical grasslands, taking sward height into account has been a much better strategy in rotational stocking management than considering pre‐defined days of growth. Similarly, sward height could be used to determine the moment when tropical grasses present the best ensilability parameters. This study aimed to identify the sward height at which Panicum maximum cv. Mombaça (Guinea grass) provides the highest fermentability coefficient (FC) and to define the combination of additives that best improves the chemical composition of silage. Two trials were carried out in Selvíria, MS, Brazil, from 2015 to 2016. The first year was used to identify the highest FC, and the second year was used to identify the best combination of eight additives (citrus pulp [CIP], homofermentative and heterofermentative LAB, their combinations and control). Statistical analyses were performed using SAS (p < .05), and one contrast was defined as silage with CIP vs. silage without CIP. The height of 130 cm resulted in the highest FC (31.01). Silages inoculated with CIP had better quality than silages without CIP, due to the high crude protein (8.3 vs. 7.3% DM), DM recovery (98.6 vs. 93.3% DM), low pH (3.92 vs. 4.91) and NH3‐N values (2.49 vs. 14.73% total N). Sward height is a consistent parameter for determining the time of ensiling Guinea grass, and the inclusion of CIP is necessary to raise the silage quality.

In Brazil, there is a lack of combined soil-plant data attempting to explain the influence of specific climate, soil conditions, and crop management on heavy metal uptake and accumulation by plants. As a consequence, soil-plant relationships to be used in risk assessments or for derivation of soil screening values are not available. Our objective in this study was to develop empirical soil-plant models for Cd, Cu, Pb, Ni, and Zn, in order to derive appropriate soil screening values representative of humid tropical regions such as the state of São Paulo (SP), Brazil. Soil and plant samples from 25 vegetable species in the production areas of SP were collected. The concentrations of metals found in these soil samples were relatively low. Therefore, data from temperate regions were included in our study. The soil-plant relations derived had a good performance for SP conditions for 8 out of 10 combinations of metal and vegetable species. The bioconcentration factor (BCF) values for Cd, Cu, Ni, Pb, and Zn in lettuce and for Cd, Cu, Pb, and Zn in carrot were determined under three exposure scenarios at pH 5 and 6. The application of soil-plant models and the BCFs proposed in this study can be an important tool to derive national soil quality criteria. However, this methodological approach includes data assessed under different climatic conditions and soil types and need to be carefully considered.

The anthropogenic input of potentially toxic elements (PTEs) from industry, agrochemicals, etc., into the environment are of great concern. Models derived from pedotransfer functions can provide estimates of the levels of PTEs based on soil attributes. Based on the importance of these models in studies in contaminated areas, we assessed the concentrations of the reactive contents of Ba, Cu, Cr, Ni, Pb, and Zn in soils cultivated with vegetable crops in the state of São Paulo, Brazil. We also evaluated the influence of chemical and physical soil attributes on their reactivity and availability. The reactive contents of PTEs represent the fraction of PTEs easily sorbed at the adsorptions sites of organic matter, iron hydroxides, or clay. This fraction can supply information about the PTE content that is more or less readily released into the soil solution. The reactive and available fraction was extracted with 0.43 M HNO and 0.01 M CaCl, respectively. The proportion of reactivity of metal pools decreased in the order of Ba>Zn > Cu > Pb > Ni > Cr. The empirical models were able to predict the relationship between the reactive fractions, the pseudototal content, and the soil attributes. The available concentrations of Cr, Cu, Ni, and Pb in the soils were lower than the limit of quantification, while 3% of the Ba content and 1% of the Zn content were available in the soil solution in relation to their pseudototal content, suggesting low mobility of these elements in the soil.

Several strategies have been used to improve soil fertility using integrated crop-livestock (ICL) systems, and the harvest of the accompanying crop used for silage may have several benefits to the soil depending on the quantity and quality of the remaining litter, providing dry matter (DM) for no-tillage systems and nutrient mineralization for the subsequent crop. The objective of this study is to evaluate the rate of decomposition of macro-and micronutrients of the DM of litter produced in ICL systems at different harvest heights in one year. The study was developed at the Experimental Farm of UNESP in Selvíria, Mato Grosso do Sul, Brazil, in the dry season. The study included five replications in plots of 200 m 2 and was arranged in a completely randomized block design. The following treatments were analyzed: control sample-degraded pasture of signal grass; Marandu grass-pasture renewal from signal grass to Marandu grass; succession 45-planting of sorghum (forage sorghum cv. Volumax) and crop harvest at the height of 45 cm for silage followed by planting of Marandu grass; simultaneous seeding of Marandu grass and sorghum, and crop harvesting for silage at the height of 15 cm (Marandu + sorghum 15) and 45 cm (Marandu + sorghum 45) from the soil surface. After crop harvesting, proportional amounts of green mass remaining from each unit were collected, transferred to nylon bags, and placed in direct contact with the soil of the respective experimental unit. Each bag was opened every 30 days after closure for up to 270 days. In each bag, the DM, and macro-and micronutrients were analyzed, and the percentage of remaining material, daily decomposition rate, and half-life were calculated. The litter of the Marandu + sorghum 45 treatment contained the highest DM, decomposition rate, and nutrient content. The crop succession provided the best condition of the litter, with the highest DM on the soil surface, which improved soil conditions and made the soil less susceptible to degradation. ResumoTem sido frequente a adoção de estratégias que visem melhorias na fertilidade do solo em sistemas de integração lavoura-pecuária (ILP) e a altura de colheita para ensilagem da cultura acompanhante, pode direcionar diversos benefícios ao solo pela quantidade e qualidade da liteira remanescente, proporcionando massa seca (MS) para o plantio direto e mineralização de nutrientes para a cultura subsequente. Objetivou-se com esse trabalho estudar a dinâmica de decomposição, ao longo de um ano, da MS, macro e micronutrientes de liteiras oriundas de sistemas de ILP com diferentes alturas de colheita. O experimento foi desenvolvido em campo na Fazenda Experimental da UNESP em Selvíria, MS, na condição de sequeiro. O delineamento adotado foi em blocos completos casualizados, com cinco repetições em parcelas de 200 m 2 . Os tratamentos corresponderam a Testemunha -pastagem degradada de capim-decumbens; Marandu -renovação da pastagem de capim-decumbens para capimmarandu; Solteiro 45 -capim-marandu semeado em sucessão ao sorgo (Sorgo...

The occurrence and accumulation of heavy metals or so-called potentially toxic elements (PTEs) in soils and plants have driven long-standing concerns about the adverse effects such metals have on the environment and human health. Furthermore, contaminated food products are known to be a leading source of exposure to heavy metals for the general population. It is crucial to accurately assess the concentrations of metals in crops and the bioavailable contents of these elements in the soil. The state of São Paulo is the largest consumer market of horticultural products in Brazil with production focused essentially on urban and industrial areas, which greatly increases the degree of exposure to contaminants. The objective of the authors in this study was to evaluate the soil-plant relationships between concentrations of Cd, Cu, Ni, Pb and Zn in vegetable and garden soils in the state of São Paulo, Brazil. To accomplish this, 200 soil (0-20 cm) and plant samples were collected from 25 species in the production areas. With the exception of Cd, there was positive correlation between pseudototals (USEPA 3051a) and bioavailable contents (extracted with DTPA) of heavy metals. However, the Cd and Pb contents in plants were not significantly correlated with any of the variables studied. All random forest and tree models proved to be good predictors of results generated from a regression model and provided useful information including covariates that were important for specifically forecasting Zn concentration in plants.

Two experiments were carried out in the experimental field of the Universidade Estadual Paulista-UNESP in Ilha Solteira, São Paulo state, Brazil, in a greenhouse from April to July 2015. This study aimed at evaluating the best sowing season and response to nitrogen doses for the cultivars of yellow oat São Carlos and black IAPAR 61. The experiments were conducted in randomized blocks designs in a factorial scheme with three replicates. The sowing seasons were April 23, May 08, and May 5 and the nitrogen doses were 0; 12.5; 25; 35.5 and 50 kg ha-1 cycle. Harvests at 30 and 60 days were conducted in order to estimate of the production of dry weight (DW), crude protein (CP), neutral detergent fiber (NDF) and acid detergent fiber (ADF). For a productivity of DW, there was interaction between sowing season and oat cultivars and significant differences for CP in the second harvesting. For NDF, a significant difference was observed between harvesting. The most suitable time for sowing of both yellow oats and black oats is early May. Dry weight yield and the CP content of yellow oats increased linearly with increasing nitrogen rates while for black oats a maximum DM yield were obtained with the application of 43.5 kg ha-1 of N.

I thank the farmers of the "green belt of the São Paulo state", which opened its properties for collection of soil and plant samples used in this research and also to the colleagues Mariele Piovesan Ruiz and Carlos Formoso that contributed with information that assisted in the choice the collection areas and facilitated the contact with the farmers in selected areas.

Lettuce (Lactuca sativa) is the main leafy vegetable produced in Brazil. Since its production is widespread all over the country, lettuce traceability and quality assurance is hampered. In this study, we propose a new method to identify the geographical origin of Brazilian lettuce. The method uses a powerful data mining technique called support vector machines (SVM) applied to elemental composition and soil properties of samples analyzed. We investigated lettuce produced in São Paulo and Pernambuco, two states in the southeastern and northeastern regions in Brazil, respectively. We investigated efficiency of the SVM model by comparing its results with those achieved by traditional linear discriminant analysis (LDA). The SVM models outperformed the LDA models in the two scenarios investigated, achieving an average of 98 % prediction accuracy to discriminate lettuce from both states. A feature evaluation formula, called F-score, was used to measure the discriminative power of the variables analyzed. The soil exchangeable cation capacity, soil contents of low crystalized Al and Zn content in lettuce samples were the most relevant components for differentiation. Our results reinforce the potential of data mining and machine learning techniques to support traceability strategies and authentication of leafy vegetables.