Core Ideas
Phosphogypsum application increases grain yield, particularly in acidic subsoils.
Phosphogypsum rates can be based on Ca2+ saturation in the ECEC of the subsoil.
We show a new method for phosphogypsum recommendation based on M5‐Rules.
Phosphogypsum (PG) is used in tropical and subtropical agriculture when subsoil acidity is an important yield‐limiting factor. However, the conditions that increase crop yield under PG application in continuous no‐till systems remain unclear. In addition, the methods used in Brazil to estimate the PG requirements are sometimes imprecise. Thus, to develop an accurate method for establishing PG recommendation, a dataset from experiments performed in southern Brazil and selected published papers involving PG application on various Oxisols under continuous no‐till was analyzed by computational techniques of data mining using the M5‐Rules algorithm to create regression models. Experimental areas consisted of annual crops managed under a long‐term (≥10 yr) no‐till system. These included maize (Zea mays L.), soybean [Glycine max (L.) Merr.], wheat (Triticum aestivum L.), and barley (Hordeum distichum L.). Results allowed the proposal of a new method for establishing the PG requirement to increase Ca2+ saturation to 60% in the effective cation exchange capacity (ECEC) at the 20‐ to 40‐cm soil layer when this is lower than 54% using the following equation: PG (Mg ha−1) = (0.6 × ECEC – exchangeable Ca2+ content in cmolc dm−3) × 6.4. The PG rates indicated by this method agreed with those leading to the maximum economic yields obtained in most studies conducted on continuous no‐till soils in Brazil. The proposed method might be efficiently used when subsoil acidity is an important growth‐limiting factor.
Spectral reflectance sensors may be useful in discriminating biomass and nitrogen status in plants. A field experiment was carried out on a loamy Typic Hapludox under no-tillage in Ponta Grossa, Parana State, Brazil, aiming to evaluate the efficiency of active canopy sensors (GreenSeeker 505 and Crop Circle ACS-470) to discriminate wheat response to nitrogen fertilization by determining the dry mass, nitrogen status, and grain yield. A randomized complete block design was used and four treatments were replicated 12 times. The treatments consisted of nitrogen application, as urea, at the rates of 0, 50, 100, and 150 kg ha -1 . The nitrogen rates were applied in topdressing at tillering of wheat crop. The readings from the sensors were sensitive to changes caused by nitrogen rates application and yet had close and significant correlations with the dry biomass production and nitrogen uptake by wheat plants. GreenSeeker 505 and Crop Circle ACS-470 sensors showed similar efficiency in discriminating biomass production and the nutritional status in the wheat crop related to nitrogen. Wheat grain yield was high and it did not follow the gains in the production of dry mass in the shoot.
RESUMOA metodologia de Redes Neurais Artifi ciais (RNAs) tem sido aplicada nas soluções de diversos problemas, dentre eles, nas aplicações voltadas a áreas específi cas cujo objetivo geralmente é auxiliar na tomada de decisões. Parte destas aplicações é resolvida com simuladores, por exemplo, o JavaNNS e o SNNS. Em determinadas situações, porém, é necessário buscar informações ou valores que estão em variáveis, ou ainda, implícitos nos códigos de algoritmos de treinamento destes simuladores, não sendo acessíveis diretamente ao usuário, situação em que o uso dos simuladores torna-se insufi ciente. Surge então a necessidade de desenvolver sistemas específi cos, implementando todos os mecanismos de criar e de treinar as RNAs, sendo também necessário estabelecer um meio para validar os resultados obtidos dos sistemas. Diante disso, este trabalho apresenta o desenvolvimento de um
Precision Agriculture has the goal of reducing cost which is difficult when it is related to fertilizers application. Nitrogen (N) is the nutrient absorbed in greater amounts by crops and the N fertilizers application present significant costs. The use of spectral reflectance sensors has been studied to identify the nutritional status of crops and prescribe varying N rates. This study aimed to contribute to the determination of a model to discriminating biomass and nitrogen status in wheat through two sensors, GreenSeeker and Crop Circle, using the Resilient Propagation and Backpropagation Artificial Neural Networks algorithms. As a result was detected a strong correlation to the sensor readings with the aboveground biomass production and N extraction by plants. For both algorithms it was established a satisfactory model for estimating wheat dry biomass production. The best Backpropagation and Resilient Propagation models defined showed better performance for the GreenSeeker and Crop Circle sensors, respectively.
A field experiment examined the wheat response following soybean due to urea-N fertilization at sowing (0, 20, 40, and 60 kg N ha −1) and top dressing (0, 30, 60, and 90 kg N ha −1). Normalized difference vegetation index (NDVI) measurements using a GreenSeeker active sensor were taken at different growth stages, at various sensor heights above the canopy, and at different times of day. Both the shoot dry matter yield and the grain yield were not affected by N fertilization at sowing and increased with increasing N rates at top dressing. A maximum economic yield was obtained at 55 kg N ha −1 in top dressing, causing a 33% increase in grain yield and an economic return of US$ 189.50 ha −1. Differences in NDVI were found before and after top-dressing N fertilization, indicating that GreenSeeker was efficient in monitoring wheat N nutrition. NDVI varied depending on the measurement distance, stabilizing from 0.30 to 1.20 m above the canopy. Because of the influence of incident radiation, higher NDVI values were obtained at the beginning and end of the day. The results suggest that for the wheat cultivar Quartzo following soybean under minimum soil disturbance, while there is no need to apply N at sowing, an important increase in grain yield can be obtained with top-dressing N fertilization. In addition, to improve the accuracy in developing N fertilizer recommendation models using GreenSeeker, a consistent protocol for spectral reflectance readings, mainly regarding the time of day, is required.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.