Delineation of coffee terroir zones using machine learning and support of portable X-ray fluorescence

• Grant number: 24/09330-8
• Support Opportunities: Scholarships in Brazil - Scientific Initiation
• Start date: July 01, 2024
• End date: June 30, 2025
• Field of knowledge: Agronomical Sciences - Agronomy - Soil Science
• Agreement: MCTI/MC
• Principal Investigator: Michele Duarte de Menezes
• Grantee: Bruno Bommediano
• Host Institution: Escola de Ciências Agrárias. Universidade Federal de Lavras (UFLA). Ministério da Educação (Brasil). Lavras , SP, Brazil
• Associated research grant: 21/06968-3 - From seed to cup: internet of things technology in the quality coffee production chain, AP.TEM

Abstract

The term "terroir," widely used in the world of wines and applicable to coffee, reflects the conditions affecting the plant, such as climate, soil, terrain, temperature, and humidity - but also human factors, such as agricultural production practices and post-harvest processes. When interacting with the plant's genetics, unique characteristics are generated reflecting in the flavor and aroma of coffee. Given the multivariate and complex interactions of factors influencing coffee quality in defining terroir, machine learning techniques present great potential when applied to unravel, quantify, and understand data in operational environments in agribusiness (Patrício and Rieder, 2018). Furthermore, they have the potential to convert sensor data obtained rapidly, more affordably, and in an environmentally friendly manner into useful or desirable information. This is the theoretical framework of the so-called soft sensors, a term that emerged from combining the words "software," as sensor signal evaluation models are usually implemented in computer programs, and "sensor," because these models create sensor-like information from hardware. The portable X-ray fluorescence (pXRF) soft sensor is capable of detecting the total content of various chemical elements from the periodic table (from Mg to U), as each one has a typical fluorescence energy. Data is obtained quickly (about 1 minute) through proximal detection with little or no pre-processing of samples (Silva et al., 2021), which is the most significant advantage of this sensor. In coffee farming, special attention will be given to the levels of K, Cl, and S of pXRF, as these chemical elements are essential from the perspective of beverage quality. K has long been considered the quality element in plant nutrition (Malavolta, 1980). Amorim et al. (1973) reported that in coffee plants fertilized with potassium chloride, the Cl- anion can have a negative effect on the beverage. S is related to caffeine levels. (AU)