Yuri Gelsleichter

Digital soil mapping, GIS, satellite monitoring, and precision agriculture

About me

I am Yuri Andrei Gelsleichter, Mechanical Technician (IFSC, 2004), Environmental Engineer (UNISUL, 2015), Ph.D. in Science, Technology and Innovation in Agriculture (UFRRJ, 2020), Postdoctoral Researcher (Hungarian University of Agricultural Sciences, 2023).

During this journey, I became a specialist in Digital Soil and Environment Mapping, Precision Agriculture, Analysis and Data Visualization applying the R programming language.

Currently also working with remote sensing, data mining and automation, and research in the area of soil classification and spectroradiometry, and precision agriculture. Co-founder of Data Situ.

CV em Português 🇧🇷

Projects

Open Science to Increase Reproducibility In Science (OSIRIS)

https://osiris4r.eu/

Reproducibility in science needs to be standardized with protocols and checklists

Reproducibility is essential for the quality of Research and Innovation (R&I) but is currently limited. In the project, my role is to test and contribute to computational reproducibility, seeking to develop protocols and checklists for improvement in R&I.

Soil mapping in Itatiaia National Park

doi.org/10.1016/j.geodrs.2023.e00641

The incredible thing about this project was to apply proximal remote sensing to satellite images to almost double the predictive capacity of the model

Remote Sensing products are applicable for spatial prediction of soil properties, in the process called Digital Soil Mapping (DSM).

The aim of the study was to combine Vis-SWIR soil laboratory spectra with the DSM technique to create new hyperspectral images to use as covariates in a new soil mapping method.

The advantages of the method were demonstrated throughout the spatial prediction of Total Carbon content in soils from the upper part of the Itatiaia National Park. Using information from the Vis-SWIR spectra of the upper soil horizon from 72 points in the upper part of INP, 130 hyperspectral images were generated, that is, subsurface hyperspectral image. The validation methods were 8-fold cross-validation (CV) and external validation executed using the Random Forest algorithm.

The DSM achieves a mean determination coefficient (R2) in the CV of 0.39 and Root Mean Square Error (RMSE) of 4.6, while the creation of a new model with a set of hyperspectral images gave a CV R2 of 0.60 and RMSE of 4.06. The method can be applied in dense or low vegetation areas, for agricultural or conservation purposes, for various soil properties, with all and each wavelength.

International Experiences

MATE

https://uni-mate.hu/

Postdoctoral fellowship at MATE University

2022 - current

In this experience, I am constantly deepening my research strategies and programming in R

Research on classification, mapping, and spectral characterization of soils; Lessons in R programming, soil improvement, and protection

ISRIC

https://www.isric.org/

Guest researcher

2019

Spatial modeling of land use

Development of land use scenarios for calculating carbon stocks in Argentina

MATE

https://uni-mate.hu/

Exchange at MATE University

2017 - 2018

Soil classification provides a foundation for land use planning

International collaboration for the Universal Soil Classification System by including centroids of Brazilian soils

Education

Federal Rural University of Rio de Janeiro

Doctor in Science, Technology and Innovation in Agriculture

2017 - 2020

Founded in 1910, UFRRJ has a tradition in Agronomy research and teaching

At this stage of my professional life, I gained a great deal of knowledge about Soils and Remote Sensing and Geographic Information Systems applied in R programming, where I also learned about data flow and automation

University of Southern Santa Catarina

Bachelor in Environmental and Sanitary Engineering

2009 - 2015

UNISUL is consolidated as an educational hub in Greater Florianopolis

In addition to understanding environmental management, water and waste treatment, I was awakened to Remote Sensing and Geographic Information Systems

Federal Institute of Santa Catarina

Industrial Mechanics Technician

2003 - 2004

Created in 1909 in Florianopolis, Santa Catarina, IFSC has a tradition in teaching

During the course, I gained a lot of knowledge about industrial mechanics including metrology, projects, and manufacturing processes

Recent Publications

Relevant studies:

Enhancing Soil Mapping with Hyperspectral Subsurface Images generated from soil lab Vis-SWIR spectra tested in southern Brazil; 2023; Gelsleichter, Y.A.; Costa, E.M,; Anjos, L.H.C.; Marcondes, R.A.T; doi

Past and Future Responses of Soil Water to Climate Change in Tropical and Subtropical Rainforest Systems in South America; 2023; Arévalo, S.M.M.; Delgado, R.C.; Lindemann, D.S.; Gelsleichter, Y.A.; Pereira, M.G.; Rodrigues, R.A.; Justino, F.B.; Wanderley, H.S.; Zonta, E.; Santana, R.O.; Souza; R.S.; doi

Degradation of South American biomes: What to expect for the future?; Delgado, R.C.; Santana, R.O.; Gelsleichter, Y.A.; Pereira, M.G.; doi

Mapping soil properties in a poorly-accessible area; Costa, E.M.; Pinheiro, H.S.K.; Anjos, L.H.C; Marcondes, R.A.T.; Gelsleichter, Y.A.; doi

Spatial Bayesian belief networks: a participatory approach for mapping environmental vulnerability at the Itatiaia National Park, Brazil; Costa, E.M.; Pinheiro, H.S.K.; Anjos, L.H.C; Gelsleichter, Y.A.; Marcondes, R.A.T.; doi