Models for predicting coffee yield from chemical characteristics of soil and leaves using machine learning

BACKGROUND

Coffee farming constitutes a substantial economic resource, representing a source of income for several countries due to the high consumption of coffee worldwide. Precise management of coffee crops involves collecting crop attributes (characteristics of the soil and the plant), mapping, and applying inputs according to the plants' needs. This differentiated management is precision coffee growing and it stands out for its increased yield and sustainability.

RESULTS

This research aimed to predict yield in coffee plantations by applying machine learning methodologies to soil and plant attributes. The data were obtained in a field of 54.6 ha during two consecutive seasons, applying varied fertilization rates in accordance with the recommendations of soil attribute maps. Leaf analysis maps also were monitored with the aim of establishing a correlation between input parameters and yield prediction. The machine-learning models obtained from these data predicted coffee yield efficiently. The best model demonstrated predictive fit results with a Pearson correlation of 0.86. Soil chemical attributes did not interfere with the prediction models, indicating that this analysis can be dispensed with when applying these models.

CONCLUSION

These findings have important implications for optimizing coffee management and cultivation, providing valuable insights for producers and researchers interested in maximizing yield using precision agriculture. © 2024 Society of Chemical Industry.

A CABINE DOS Can the tractor's cabin allow the noise at the operator's station?PODE ATENUAR O RUÍDO NO POSTO DO OPERADOR?

Constant exposure to high noise levels can cause health problems for agricultural machinery operators. Machines equipped with closed cabs can attenuate noise and preserve the operator from an unhealthy condition. Some studies show that in cabin tractors it is possible to renounce the use of hearing protection, however indoor equipment and air conditioning can reach high levels of noise, which need to be further investigated. The present study aimed to evaluate the noise emitted by a cabined agricultural tractor, in different points, inside and outside cabin, in the engine rotations of 800rpm, 2180rpm and 2500rpm. In each reading, 100 noise intensity values were automatically collected, over two minutes with the aid of a digital decibel meter device. The results were compared with the levels assumed by Annex 1 of the regulatory Brazilian standard NR 15, and it was found that the cabin was able to isolate the noise emitted by the engine and its external components in 15%, however, the internal noise reached 81dB(A). The cabin can attenuate external noise and offer a healthy working environment, however, in long working hours it is recommended to use personal protective equipment such as ear protectors.

PLATAFORMA ARDUINO APLICADA NO DESENVOLVIMENTO DE UM SISTEMA DE COLETA DE DADOS MICRO-AMBIENTAIS EM CASAS DE VEGETAÇÃO

A coleta de dados e o monitoramento ambiental na agricultura são atividades de alta complexidade, demandando normalmente equipamentos de alto custo. A coleta de dados deve ser realizada preferivelmente próxima ao estrato das culturas vegetais, pois desta forma obtém-se dados fidedignos do microambiente ao qual a planta está inserida. Objetivou-se desenvolver um sistema de coleta de dados, de baixo custo, para a obtenção de parâmetros relacionados à luminosidade, umidade do solo, umidade do ar e temperatura em ambiente agrícola. Utilizou-se a metodologia específica de desenvolvimento de produtos e, plataforma de prototipagem Arduino para montagem do sistema Datalogger. Foram realizados testes de funcionamento em duas casas de vegetação A1 e A2 pelo período de 24 horas. A concepção do sistema, com a utilização de suportes de sensores, permitiu a coleta de dados sem interrupção e com adequada acurácea. A plataforma Arduíno e sensores acessórios mostraram-se perfeitamente aplicáveis para a aquisição e armazenamento de dados em casas de vegetação. O protótipo de Datalogger desenvolvido apresentou redução de custo de 600 até 3000% em relação aos componentes disponíveis no mercado com funcionalidades semelhantes.