Use este identificador para citar ou linkar para este item: http://repositorio.ufla.br/jspui/handle/1/46925
Título: Soil compaction in no-tillage system: diagnosis, monitoring and alleviation
Título(s) alternativo(s): Compactação do solo em sistema plantio direto: diagnóstico, monitoramento e mitigação
Autores: Silva, Bruno Montoani
Tormena, Cássio Antônio
Oliveira, Geraldo César de
Dias Júnior, Moacir de Souza
Moreira, Silvino Guimarães
Palavras-chave: Física do solo
Manejo do solo
Preparo ocasional
Resistência à penetração
Aprendizagem de máquina
Solos - Compactação
Soil physics
Soil management
Occasional preparation
Penetration resistance
Data do documento: 23-Ago-2021
Editor: Universidade Federal de Lavras
Citação: PEIXOTO, D. S. Soil compaction in no-tillage system: diagnosis, monitoring and alleviation. 2021. 190 p. Tese (Doutorado em Ciência do Solo) – Universidade Federal de Lavras, Lavras, 2021.
Resumo: Many challenges have arisen with the adoption of no-tillage (NT), for example: soil compaction, weed management and stratification of soil organic matter and nutrients. To overcome these challenges, rural landowners and researchers have used occasional tillage (OT), which consists of using some method of soil tillage, such as chiseling, subsoiling, plowing and harrowing in consolidated NT. In this sense, the general objective was to improve the accuracy of the diagnosing, monitoring and alleviation of soil compaction in NT. The specific objectives were: 1) to perform a global meta-analysis of the effects of OT on annual crop yield, soil physical, chemical and biological properties, soil erosion and weed control; 2) to identify the soil physical properties that are more related to the crops response, and therefore, more suitable for the diagnosis of soil compaction; 3) to test the machine learning approach in sorting the soil physical properties and its relationship with crops yield; 4) to propose a methodology for diagnosing, monitoring and management of soil compaction in NT using the penetration resistance as an indicator; 5) linking knowledge of modeling vertical pressures applied by agricultural machines, soil load-bearing capacity (pre-consolidation pressure) and soil physical quality (least limiting water range) for preventive management of soil compaction; 6) to propose the use of machine learning for the diagnosis of soil compaction in NT. An experiment was implemented in October 2015 at Farm Santa Helena, in the municipality of Nazareno, Minas Gerais, in a Typic Hapludox, clay texture. The treatments consisted of soil management to mitigate compaction, combining subsoiling and two forms of lime application, two subsoiling frequencies (two and three years), chiseling, gypsum application and a control treatment (continuous NT), totaling seven managements. The crops evaluated in the period from 2015 to 2019 were: soybeans (2015/2016 and 2017/2018), maize (2016/2017 and 2018/2019), common beans (2017) and wheat (2018). The meta-analysis showed that OT did not affect crop yield, pH, available phosphorus and microbial activity; improved soil physical properties and weed control; and reduced aggregate stability, total organic carbon and soil erosion. It was found with the experiment that the OT improved the soil physical properties and increased the crops yield subsequent, especially soybeans, promoting economic benefits. The soil physical properties most sensitive of OT and the crops response were the penetration resistance, aeration capacity, macroporosity, relative field capacity and the “S” index. The penetration resistance should be evaluated between the matric potential of -0.03 and -0.50 MPa, preferably -0.10 MPa, for the diagnosing and monitoring of soil compaction in NT, therefore, drier than the field capacity suggested in the literature. From this, a methodological proposal using penetration resistance as an indicator of soil compaction in NT areas was suggested and tested. The link of modeling vertical pressures applied by agricultural machines, soil load-bearing capacity and soil physical quality was efficient to assist in the preventive management of soil compaction in annual crop production systems. Finally, the classification approach of machine learning algorithms proved to be efficient in the diagnosis of soil compaction in NT, by combining the responses of a set of soil physical properties and crop yield, improving decision making regarding the use of OT.
URI: http://repositorio.ufla.br/jspui/handle/1/46925
Aparece nas coleções:Ciência do Solo - Doutorado (Teses)

Arquivos associados a este item:
Arquivo Descrição TamanhoFormato 
TESE_Soil compaction in no-tillage system diagnosis, monitoring and alleviation.pdf6,13 MBAdobe PDFVisualizar/Abrir


Os itens no repositório estão protegidos por copyright, com todos os direitos reservados, salvo quando é indicado o contrário.