Mapping of cropland humus content of the Bryansk region using machine learning methods DOI Creative Commons

Lidiya Yuryevna Konoplina,

J. L. Meshalkina, В. П. Самсонова

et al.

Lomonosov Soil Science Journal, Journal Year: 2024, Volume and Issue: 79(№4, 2024), P. 130 - 140

Published: Jan. 1, 2024

The FAO methodology within the Global Soil Nutrient and Budget Maps (GSNmap) project was tested for first time mapping humus content with a spatial resolution of 250 meters per pixel in soils Russian Federation at regional scale, using Bryansk Region as an example. map created R software environment data from Agrochemical Service remote sensing, global databases soil maps. centroids sites which composite samples were taken by selected sampling points. set predictors available under expanded additional data, including maps soil-forming rocks. importance assessed Boruta algorithm, is usually used initial stage random forest. model caret package quantile regression forest method. modeling efficiency coefficient (MEC) 55%, determination (R2) 0.57. reflects current information that can be to monitor dynamics organic matter assess state arable region.

Language: Английский

Machine learning for predictive mapping of exceedance probabilities for potentially toxic elements in Czech farmland DOI
Jan Skála, Daniel Žížala, Robert Minařík

et al.

Journal of Environmental Management, Journal Year: 2025, Volume and Issue: 380, P. 125035 - 125035

Published: March 24, 2025

Language: Английский

Citations

0
Fine-resolution baseline maps of soil nutrients in farmland of Jiangxi Province using digital soil mapping and interpretable machine learning DOI
Bifeng Hu, Yibo Geng,

Kejian Shi

et al.

CATENA, Journal Year: 2024, Volume and Issue: 249, P. 108635 - 108635

Published: Dec. 9, 2024

Language: Английский

Citations

1
Mapping Topsoil Behavior to Compaction at National Scale from an Analysis of Field Observations DOI Creative Commons
Anne C Richer-De-Forges, Dominique Arrouays, Songchao Chen

et al.

Land, Journal Year: 2024, Volume and Issue: 13(7), P. 1014 - 1014

Published: July 8, 2024

Soil compaction is one of the most important and readily mitigated threats to soil health. Digital Mapping (DSM) has emerged as an efficient method provide broad-scale maps by combining information with environmental covariates. Until now, input DSM been mainly composed point-based quantitative measurements properties and/or type/horizon classes derived from laboratory analysis, point observations, or maps. In this study, we used field estimates map behavior at a national scale. The results previous study enabled clustering six different behaviors using in situ observations. potential responses effective land management tool for preventing future compaction. Random forest was make spatial predictions over cultivated soils mainland France (about 210,000 km2). Modeling performed 90 m resolution. us spatially identify clusters possible Most were consistent known geographic distributions some types properties. This consistency checked comparing both local-scale external sources information. best predictors available digital (clay, silt, sand, organic carbon (SOC) content, pH), indicators structural quality SOC clay covariates (T °C relief-related covariates). Predicted interpretable support recommendations mitigate compactness soil–scape Simple observational data that are usually collected surveyors, then stored databases, valuable mapping assessment inherent sensitivity

Language: Английский

Citations

0
Mapping of Cropland Humus Content of Bryansk Oblast Using Machine-Learning Methods DOI

L. Yu. Konoplina,

J. L. Meshalkina, В. П. Самсонова

et al.

Moscow University Soil Science Bulletin, Journal Year: 2024, Volume and Issue: 79(4), P. 500 - 508

Published: Dec. 1, 2024

Language: Английский

Citations

0
Mapping of cropland humus content of the Bryansk region using machine learning methods DOI Creative Commons

Lidiya Yuryevna Konoplina,

J. L. Meshalkina, В. П. Самсонова

et al.

Lomonosov Soil Science Journal, Journal Year: 2024, Volume and Issue: 79(№4, 2024), P. 130 - 140

Published: Jan. 1, 2024

The FAO methodology within the Global Soil Nutrient and Budget Maps (GSNmap) project was tested for first time mapping humus content with a spatial resolution of 250 meters per pixel in soils Russian Federation at regional scale, using Bryansk Region as an example. map created R software environment data from Agrochemical Service remote sensing, global databases soil maps. centroids sites which composite samples were taken by selected sampling points. set predictors available under expanded additional data, including maps soil-forming rocks. importance assessed Boruta algorithm, is usually used initial stage random forest. model caret package quantile regression forest method. modeling efficiency coefficient (MEC) 55%, determination (R2) 0.57. reflects current information that can be to monitor dynamics organic matter assess state arable region.

Language: Английский

Citations

0