ENVIRONMENTAL FRAGILITY BY MACHINE LEARNING ALGORITHMS

Authors

DOI:

https://doi.org/10.4215/rm2022.e21034

Abstract

The advancement of predictive models by Machine Learning Algorithms (ML) associated with environmental data enables the improvement of models of environmental fragility, which are essential tools for decision-making. This study aimed to derive a prediction of environmental fragility by testing ML associated with environmental covariates in the state of Minas Gerais. We use physical-environmental variables (soil, geology, climate, relief) with a weight of fragility for the attributes and calculation of the average to obtain a model of Potential Environmental Fragility (PEF). Subsequently, we extracted the PEF values to a 4,800-point grid, which was used to generate a new prediction by ML called PEFML. This prediction was based on testing five algorithms and a set of 105 environmental covariates. The results indicated that the best-performing PEFML prediction was the Random Forest model (R2 0.59 and RMSE 0.47), indicating a predominance of the low environmental fragility level. The PEF and PEFML models have strong correlations (0.7 Pearson); however, PEFML has stronger correlations with other environmental data. Therefore, the PEFML prediction is a robust model that captures information from covariates and has coherent spatial patterns.

Keywords: Environmental fragility model; Spatial prediction; Random Forest; Environmental planning.



Downloads

Download data is not yet available.

Author Biographies

  • Cristiano Marcelo Pereira de Souza, State University of Montes Claros, Montes Claros (MG), Brazil

    PhD in Soils and Plant Nutrition from the Federal University of Viçosa (2018); Currently, Postdoctoral Researcher in Geography at the State University of Montes Claros. He participates in the Graduate Program in Geography (master's degree) at Unimontes, teaching disciplines in the master's degree: pedology, watersheds, scientific writing, and reading landscapes. He received a scholarship from different development agencies, such as: CNPq, CAPES, FAPEMIG, FAPESB, CENTEV-UFV. He served as a collaborating researcher at UFV from 2018 to 2020 in an extension project linked to the Geography department. He has experience in the field of physical geography, pedology, geomorphology, agronomy, geology. Working mainly on the following topics: soil-landscape relationship, GIS, modeling with machine learning algorithms, landscape dynamics, geoenvironmental mapping, watershed management, and studies of basins with structural control

  • Lucas Augusto Pereira Silva, Federal University of Uberlândia, Uberlândia (MG), Brazil
    PhD student in Geography at the Federal University of Uberlândia - UFU. Works in the area of Physical Geography, with emphasis on the following topics: environmental analysis with remote sensing applications, geoprocessing and machine learning, studies in biogeography and the influence of climate change scenarios (past and future) on the dynamics of land use and land cover and vegetation domains in semi-arid regions.
  • Gustavo Vieira Veloso, Federal University of Viçosa, Viçosa (MG), Brazil

    PhD in Soil Science and Plant Nutrition at the Federal University of Viçosa. He has experience in the area of Agricultural and Soil Engineering, with emphasis on Machine Learning, Modeling, Data Analysis, Time Series, Machine Design, Geochemistry, Geostatistics, Statistics is Electronics. Working mainly on the following topics: Machine Learning, Modeling, Data Analysis, Time Series and Statistics. 

  • Marcos Esdras Leite, State University of Montes Claros, Montes Claros (MG), Brazil

    Professor at the Department of Geosciences State University of Montes Claros-UNIMONTES. Professor of the Graduate Program in Geography/ UNIMONTES and the Graduate Program in Social Development/ UNIMONTES. Graduated in Geography/Unimontes. PhD in Geography from the Federal University of Uberlândia - UFU. Has experience in teaching and research in Geography, with emphasis on Geotechnologies. Coordinator of the Geoprocessing Laboratory/UNIMONTES. He completed a doctoral technical internship at Universidade Nova de Lisboa with a FAPEMIG scholarship (2008). Editor of Cerrados Magazine (2015 to 2016). Editor of Social Development Magazine (2013 to 2015). Member and coordinator of the Chamber of Applied Social Sciences - CSA/FAPEMIG (2017 to 2020). Member of CODEMA from Montes Claros/MG (2019 to 2021). Coordinator PPGEO/Unimontes (2018 to 2021). Coordinator of PPGDS/Unimontes (Current).

  • Elpídio Inácio Fernandes Filho, Federal University of Viçosa, Viçosa (MG), Brazil

    PhD in Agronomy (Soils and Plant Nutrition) from the Federal University of Viçosa (1996). He is currently a Professor at the Federal University of Viçosa. He has experience in the area of Agronomy, with an emphasis on Pedometry. Working mainly on the following topics: Expert systems, Agricultural fitness, Geographic information systems, software development.

Downloads

Published

2023-02-15

Issue

Vol. 21 (2022)

Section

Artigos

License

Authors who publish in this journal agree to the following terms:

  1. Authors retain the copyright and grant MERCATOR the right of first publication, with the work simultaneously licensed under the Creative Commons Attribution License, which allows the sharing of the work with recognition of the authorship of the work and initial publication in this journal.
  2. Authors are authorized to sign additional contracts separately, for non-exclusive distribution of the version of the work published in this journal (e.g., publish in an institutional repository or as a book chapter), with acknowledgment of authorship and initial publication in this journal.
  3. Authors are allowed and encouraged to publish and distribute their work online (e.g., in institutional repositories or on their personal page) at any point before or during the editorial process, as this can generate productive changes as well as increase the impact and citation of the published work (see The Effect of Free Access).
  4. Authors are responsible for the content of the manuscript published in the journal.

How to Cite

ENVIRONMENTAL FRAGILITY BY MACHINE LEARNING ALGORITHMS. (2023). Mercator, 21. https://doi.org/10.4215/rm2022.e21034