Innovation in fruit farming research: compositional nutrient diagnosis (CND), boundary line and yield prediction, with an emphasis on grapevine cultivation

• Grant number: 23/10434-0
• Support Opportunities: Scholarships abroad - Research
• Start date: December 05, 2023
• End date: June 05, 2024
• Field of knowledge: Agronomical Sciences - Agronomy - Soil Science
• Principal Investigator: Danilo Eduardo Rozane
• Grantee: Danilo Eduardo Rozane
• Host Investigator: Moreno Toselli
• Host Institution: Faculdade de Ciências Agrárias. Universidade Estadual Paulista (UNESP). Campus do Vale do Ribeira. Registro , SP, Brazil
• Institution abroad: Università di Bologna, Italy

Abstract

High yields of a quality product can only be achieved when plants are nutritionally balanced. Several factors can compromise the efficiency of crops in reaching their maximum genetic production potential, highlighting low soil fertility, especially in tropical areas, and adequate crop and agricultural input management. The project will consist of three large studies: Study 1 will establish nutritional guidelines for compositional nutrient diagnosis (CND) and develop a computational program; using the Boundary Line method; Study 2 will determine the interaction between soil analyses and the multinutrient variables obtained in Study 1, based on plant tissue contents, in order to establish fertilization recommendations as a function of soil nutrient concentrations; and in Study 3, yield predictions will be simulated and a computer program developed. The "CND-Uva" and "Predição-Uva" computer programs will be registered at INPI (National Institute of Industrial Property), which will allow producers, professionals and other interested parties to access a tool that makes the mathematical calculations required to assess nutritional status and predict grapevine production. This will be achieved with a database of nutritional (soil and plant) and climate variables that will be obtained from grapevine growers in the municipality of Pilar do Sul, São Paulo (SP) state, Brazil. The three studies will use robust data composition analysis and machine learning techniques. The project will develop prediction models and simulate the soil and leaf nutrient needs and production for Brazil and Italy, assessing the universality of guidelines/parameters, in order to promote the exchange of techniques and parameters, appropriately adapted. (AU)