Mid-infrared spectroscopy coupled with process-based modelling to assess soil organic matter dynamics in a tropical maize-forage intercropping system

• Grant number: 22/01234-4
• Support Opportunities: Scholarships abroad - Research Internship - Doctorate (Direct)
• Start date: August 22, 2022
• End date: August 14, 2023
• Field of knowledge: Agronomical Sciences - Agronomy - Soil Science
• Principal Investigator: Ciro Antonio Rosolem
• Grantee: Laudelino Vieira da Mota Neto
• Supervisor: Marcelo Valadares Galdos
• Host Institution: Faculdade de Ciências Agronômicas (FCA). Universidade Estadual Paulista (UNESP). Campus de Botucatu. Botucatu , SP, Brazil
• Institution abroad: University of Leeds, England
• Associated to the scholarship: 21/05167-7 - Organic matter quality and carbon in the soil in a maize-forage intercropping system under nitrogen fertilization, BP.DD

Abstract

Land-based solutions to mitigate the global climate change crisis require the adoption of sustainable agriculture management practices. Intercropped systems - maize-forages cropped simultaneously - are key strategies to increase soil carbon (C) sequestration and agricultural resilience to extreme weather events. The complex biogeochemical C and nitrogen (N) dynamics is a challenge for soil organic matter (SOM) modeling, especially on an ecosystem basis. The Microbial Efficiency-Matrix Stabilization (MEMS) model aims to address this challenge by framing the flow of C and N molecules from the plant litter - above and belowground materials with contrasting chemistry quality - across microbial decomposition and transformation - as C:N stoichiometry - into the stabilization on mineral matrix surfaces - persistence - based on quantifiable particulate (POM) and mineral-associated organic matter (MAOM) fractions. In order to broaden our knowledge of C and N dynamics in tropical intercropped systems fertilized with N and the response to climate scenarios, we propose as a first objective to input our data (POM and MAOM) into the MEMS model and estimate SOM dynamics. This can support sustainable management decision-making and estimates of tradable C credits, especially in highly-intensive tropical agriculture. To this purpose, it is necessary to adopt swift, cost-effective and eco-friendly techniques that allow tracking down physical and chemical soil properties. Spectroscopic techniques are reliable physical methods that use hyperspectral data based on the interaction of electromagnetic waves [mid-infrared spectroscopy MIR 400-4000 cm-1 wavelength] with soil samples. Coupling soil spectra with machine learning tools [e.g. Partial Least Squares Regression (PLS); Principal Component Analyses (PCA)] has shown feasible for rapid soil properties diagnosis, such as C and N content in the bulk soil or in SOM fractions. Therefore, the second objective is to predict soil C and N with MIR spectra information and use the data to feed MEMS model. (AU)