Prediction of soil penetration resistance with the use of geostatistics and fuzzy logic in the context of soil landscape relationship

• Grant number: 16/00452-7
• Support Opportunities: Scholarships in Brazil - Scientific Initiation
• Start date: May 01, 2016
• End date: December 31, 2016
• Field of knowledge: Agronomical Sciences - Agronomy - Soil Science
• Principal Investigator: José Marques Júnior
• Grantee: Frederico Luiz Siansi
• Host Institution: Faculdade de Ciências Agrárias e Veterinárias (FCAV). Universidade Estadual Paulista (UNESP). Campus de Jaboticabal. Jaboticabal , SP, Brazil

Abstract

The spatial variability of soil penetration resistance (PR) is closely related to the potential of soil compaction and the identification of spatial areas is essential to the adoption of the best management practices to a productivity with minimum environmental impact. The aim of this scientific initiation proposal is to evaluate the potential of fuzzy logic and geostatistics to identify areas susceptible to soil compaction in the context of soil landscape relationship. The present proposal of this scientific initiation scholarship is an extension of final report ranged from January to December of 2011 and related to the process 2010/14808-1, so, the previous analyzes will help in the development of this project. The studied area of 165.4 ha in neighborhoods near to Guariba (SP) was sampled with 349 points along three identified compartments according to the landscape model of geomorphic surfaces (GS) to analyze the descriptive statistics and geostatistics of soil density, total pore volume (TPV), microporosity, macroporosity, soil penetration resistance and magnetic susceptibility (MS) of soil in the 0.00-0.20m layer. The statistics and geostatistics relationship of these attributes will be evaluated for the assembly of a rule base in order to develop mathematical models that predict the soil penetration resistance values. Scatter plots will be constructed and analyzed the geostatistics of soil penetration resistance values generated in penetrometer and those predicted by the fuzzy logic. Based on soil penetration resistance mapping constructed by the bench penetrograph values (mapping proposal 1), will be proposed mapping methods by the points that predict the soil penetration resistance by the fuzzy logic in the values of points collected in the studied area (mapping proposal 2) and with values of points collected and the interpolated by the cokriging method after construction of variograms for locations not measured in the studied area (mapping proposal 3).