Soil fertility prediction based on inputs and outputs modeling

• Grant number: 17/07786-0
• Support Opportunities: Research Grants - Innovative Research in Small Business - PIPE
• Start date: April 01, 2018
• End date: May 31, 2019
• Field of knowledge: Agronomical Sciences - Agronomy - Soil Science
• Agreement: FINEP - PIPE/PAPPE Grant
• Principal Investigator: Leonardo Afonso Angeli Menegatti
• Grantee: Leonardo Afonso Angeli Menegatti
• Company: InCeres Desenvolvimento de Software e Processamento de Dados S/A
• CNAE: Tratamento de dados, provedores de serviços de aplicação e serviços de hospedagem na internet
• Principal investigators: Sergio Silva Ferreira ; WILSON RODRIGO PATRICIO FERREIRA
• Principal investigators: Michel Rudan Isaias Vargas ; Renan Monteiro Pinto Neto
• Associated researchers: José Paulo Molin ; Rafael Otto
• Associated scholarship(s): 18/17283-9 - Soil fertility prediction based on inputs and outputs modeling, BP.TT; 18/10777-6 - Soil fertility prediction based on inputs and outputs modeling, BP.TT; 18/06261-4 - Soil fertility prediction based on inputs and outputs modeling, BP.TT; 18/06808-3 - Soil fertility prediction based on inputs and outputs modeling, BP.TT

Abstract

Soil nutrient Dynamics are being studied along the years and a great amount of knowledge has being accumulated. At the same time, weather information from agricultura areas across Brazil are being collected in a way never expected before. The soil chemical analysis is the most used tool to access soil fertility levels and prescribe recommendations for fertilizations and amendments. The objective of this practical research is to build an application (software) to use cumulated information to predict soil fertility an nutrient availability for a given area from a soil grid analysis as starting point a input and output modeling. the methodology is based on weather information an machine learning to improve the models. the straight benefit is to increase the useful lifetime of a grid soil analysis allowing it use for better recommendations. The nutrients that will be focus on this research are Calcium, Magnesium, Phosphorus an Potassium. The dynamic models will take into account the inputs from fertilizations and amendments. Nutrient exportation, leaching and runoff will be used to estimate the losses an the final nutrient balance. With the Soil Fertility prediction it will be possible to integrate with the recommendations system for inputs prescription given a yield goal. The application will prescribe recommendations for soil correction and soil fertilization base on the modeling prediction. (AU)