4.7 Article

Improving pedotransfer functions for predicting soil mineral associated organic carbon by ensemble machine learning

Related references

Note: Only part of the references are listed.
Article Geosciences, Multidisciplinary

Clay content mapping and uncertainty estimation using weighted model averaging

Dongxue Zhao et al.

Summary: This study compared the importance of proximally and remotely sensed data for predicting clay at district scale, using different models and averaging techniques. The results showed that gamma-ray data was crucial for topsoil clay prediction, while slope was important for subsoil. Random forest model was found to be the best for predicting topsoil clay, with Granger-Ramanathan averaging recommended as a protocol for district-scale clay prediction.

CATENA (2022)

Article Multidisciplinary Sciences

Mapping high resolution National Soil Information Grids of China

Feng Liu et al.

Summary: Traditionally, soil spatial information has been presented as coarse polygon maps. However, solving global and local issues requires higher quality and more detailed soil information. In this study, we integrated predictive soil mapping paradigms and high-performance computing to generate high-resolution national gridded maps of nine soil properties at multiple depths across China. The predictions achieved significantly more detailed and accurate results compared to previous soil maps, making it a significant contribution to the GlobalSoilMap.net project.

SCIENCE BULLETIN (2022)

Review Soil Science

Digital mapping of GlobalSoilMap soil properties at a broad scale: A review

Songchao Chen et al.

Summary: Digital soil mapping (DSM) is increasingly used to deliver soil information efficiently. Research shows that while the number of DSM publications continues to grow exponentially, the majority focus on applications rather than methodology, with a particular emphasis on the mandatory soil properties for global soil mapping.

GEODERMA (2022)

Article Multidisciplinary Sciences

Global stocks and capacity of mineral-associated soil organic carbon

Katerina Georgiou et al.

Summary: This study estimates the stocks and storage capacity of mineral-associated organic carbon globally by analyzing soil profiles. The results show that the current carbon stocks in non-permafrost mineral soils account for 66% to 70% of soil carbon in surface and deeper layers, respectively, but only 42% to 21% of the mineralogical capacity. Agricultural management regions and deeper soil layers exhibit the largest undersaturation of mineral-associated carbon. Soil carbon furthest from its mineralogical capacity is more effective at accruing carbon over years to decades.

NATURE COMMUNICATIONS (2022)

Article Biodiversity Conservation

Ensemble modelling, uncertainty and robust predictions of organic carbon in long-term bare-fallow soils

Roberta Farina et al.

Summary: Simulation models representing soil organic carbon dynamics were evaluated through comparison with experimental data from long-term bare-fallow plots at six sites. The study found that using a generic parameterization applicable in Central Europe proved to be adequate in describing changes in SOC, providing sufficiently reliable estimates compared to models with separate calibration strategies.

GLOBAL CHANGE BIOLOGY (2021)

Article Soil Science

Using bias correction and ensemble modelling for predictive mapping and related uncertainty: A case study in digital soil mapping

Jean-Daniel Sylvain et al.

Summary: The study combines bias correction and ensemble modeling to improve the accuracy of digital soil mapping, reduce conditional bias, and provide uncertainty assessment. The performance of ensemble modeling surpasses individual models and underdispersion in uncertainty analysis is identified. Global mapping products show low performance and important conditional bias compared to the proposed approach.

GEODERMA (2021)

Article Environmental Sciences

Soil organic carbon fractions in the Great Plains of the United States: an application of mid-infrared spectroscopy

Jonathan Sanderman et al.

Summary: Spectroscopy is a powerful tool for increasing soil data availability in understanding carbon cycling. A calibration transfer methodology was developed to analyze soil organic carbon distribution in nearly 8000 soil samples in the Great Plains ecoregion. Different fractions of soil organic carbon showed unique depth distribution profiles, with factors like clay content, soil type, and climate influencing their distribution.

BIOGEOCHEMISTRY (2021)

Article Geosciences, Multidisciplinary

Digital mapping of the soil thickness of loess deposits over a calcareous bedrock in central France

Songchao Chen et al.

Summary: The study tested the prediction of loess thickness (TOL) by coupling a large soil dataset and environmental covariates. Using recursive feature elimination for variable selection and quantile regression forests for spatial modeling, the models showed good performance with reasonable prediction uncertainty. Airborne gamma-ray radiometric data and Sentinel 2 products were identified as important variables for the prediction of TOL.

CATENA (2021)

Article Geosciences, Multidisciplinary

Different climate sensitivity of particulate and mineral-associated soil organic matter

Emanuele Lugato et al.

Summary: Soil carbon sequestration is important in mitigating climate change, but the vulnerability of soil carbon to climate change needs to be accurately assessed. By separating soil carbon into different fractions, we can better understand its vulnerability and develop effective sequestration strategies. Different carbon fractions in soils have varying sensitivity to climate change, and management practices need to be implemented to enhance soil carbon stocks and resilience.

NATURE GEOSCIENCE (2021)

Article Environmental Sciences

Using Diffuse Reflectance Spectroscopy as a High Throughput Method for Quantifying Soil C and N and Their Distribution in Particulate and Mineral-Associated Organic Matter Fractions

Paulina B. Ramirez et al.

Summary: Large-scale quantification of soil organic carbon and nitrogen stocks, particularly in different carbon pools, is necessary for developing land management strategies. Diffuse reflectance mid-infrared spectroscopy coupled with partial least square analysis has shown promise for predicting soil properties accurately, but further calibration may be needed for specific soil types or land uses.

FRONTIERS IN ENVIRONMENTAL SCIENCE (2021)

Article Ecology

Predicting into unknown space? Estimating the area of applicability of spatial prediction models

Hanna Meyer et al.

Summary: Machine learning algorithms are popular for spatial mapping due to their ability to fit complex relationships, but their use is limited to data similar to the training set. The study proposes a method to assess the area where a prediction model can be reliably applied, using a Dissimilarity Index (DI) to define the Area of Applicability (AOA) and map estimated performance. Simulation studies show comparable prediction errors within the AOA to cross-validation errors, emphasizing the importance of considering the relationship between DI and cross-validation performance.

METHODS IN ECOLOGY AND EVOLUTION (2021)

Article Soil Science

SoilGrids 2.0: producing soil information for the globe with quantified spatial uncertainty

Laura Poggio et al.

Summary: SoilGrids generates global maps of soil properties using advanced technology and machine learning methods, incorporating soil observations and environmental factors to predict and analyze various soil properties worldwide. The research shows a good predictive performance for global soil properties, but also reveals the need for more soil observation data, especially in high-latitude regions.
Article Ecology

Simulating measurable ecosystem carbon and nitrogen dynamics with the mechanistically defined MEMS 2.0 model

Yao Zhang et al.

Summary: The MEMS 2.0 model is a comprehensive ecosystem model that simulates plant growth, soil water and temperature, plant and SOM decomposition, and nitrogen mineralization and immobilization. It accurately captures the distribution of soil carbon and nitrogen, as well as seasonal variations in gross primary production and ecosystem respiration. Further development will aim to represent forest and agricultural systems and incorporate new understanding of SOM decomposition.

BIOGEOSCIENCES (2021)

Article Soil Science

Soil carbon simulation confounded by different pool initialisation

Juhwan Lee et al.

NUTRIENT CYCLING IN AGROECOSYSTEMS (2020)

Article Soil Science

Model averaging for mapping topsoil organic carbon in France

Songchao Chen et al.

GEODERMA (2020)

Article Green & Sustainable Science & Technology

The role of soil carbon in natural climate solutions

D. A. Bossio et al.

NATURE SUSTAINABILITY (2020)

Article Soil Science

Increasing organic stocks in agricultural soils: Knowledge gaps and potential innovations

Claire Chenu et al.

SOIL & TILLAGE RESEARCH (2019)

Article Geosciences, Multidisciplinary

Continental-scale soil carbon composition and vulnerability modulated by regional environmental controls

R. A. Viscarra Rossel et al.

NATURE GEOSCIENCE (2019)

Article Geosciences, Multidisciplinary

Soil carbon storage informed by particulate and mineral-associated organic matter

M. Francesca Cotrufo et al.

NATURE GEOSCIENCE (2019)

Article Soil Science

Modelling and mapping soil organic carbon stocks in Brazil

Lucas Carvalho Gomes et al.

GEODERMA (2019)

Review Soil Science

LUCAS Soil, the largest expandable soil dataset for Europe: a review

A. Orgiazzi et al.

EUROPEAN JOURNAL OF SOIL SCIENCE (2018)

Article Geosciences, Multidisciplinary

Soil organic carbon stocks and their determining factors in the Dano catchment (Southwest Burkina Faso)

Ozias K. L. Hounkpatin et al.

CATENA (2018)

Article Environmental Sciences

Fine resolution map of top- and subsoil carbon sequestration potential in France

Songchao Chen et al.

SCIENCE OF THE TOTAL ENVIRONMENT (2018)

Article Meteorology & Atmospheric Sciences

WorldClim 2: new 1-km spatial resolution climate surfaces for global land areas

Stephen E. Fick et al.

INTERNATIONAL JOURNAL OF CLIMATOLOGY (2017)

Article Multidisciplinary Sciences

Hydrologic regulation of plant rooting depth

Ying Fan et al.

PROCEEDINGS OF THE NATIONAL ACADEMY OF SCIENCES OF THE UNITED STATES OF AMERICA (2017)

Editorial Material Environmental Sciences

Aligning agriculture and climate policy

A. Chabbi et al.

NATURE CLIMATE CHANGE (2017)

Review Soil Science

Linking soils to ecosystem services - A global review

Kabindra Adhikari et al.

GEODERMA (2016)

Article Environmental Sciences

The new assessment of soil loss by water erosion in Europe

Panos Panagos et al.

ENVIRONMENTAL SCIENCE & POLICY (2015)

Article Soil Science

Machine learning for predicting soil classes in three semi-arid landscapes

Colby W. Brungard et al.

GEODERMA (2015)

Article Multidisciplinary Sciences

The contentious nature of soil organic matter

Johannes Lehmann et al.

NATURE (2015)

Article Environmental Sciences

The EMEP MSC-W chemical transport model - technical description

D. Simpson et al.

ATMOSPHERIC CHEMISTRY AND PHYSICS (2012)

Article Multidisciplinary Sciences

Persistence of soil organic matter as an ecosystem property

Michael W. I. Schmidt et al.

NATURE (2011)

Article Agriculture, Multidisciplinary

Climate change mitigation: A spatial analysis of global land suitability for clean development mechanism afforestation and reforestation

Robert J. Zorner et al.

AGRICULTURE ECOSYSTEMS & ENVIRONMENT (2008)

Review Agronomy

Organo-mineral associations in temperate soils:: Integrating biology, mineralogy, and organic matter chemistry

Ingrid Koegel-Knabner et al.

JOURNAL OF PLANT NUTRITION AND SOIL SCIENCE (2008)

Article Environmental Sciences

Soil carbon saturation: concept, evidence and evaluation

Catherine E. Stewart et al.

BIOGEOCHEMISTRY (2007)

Article Soil Science

Quantifying soil organic carbon fractions by infrared-spectroscopy

M. Zimmermann et al.

SOIL BIOLOGY & BIOCHEMISTRY (2007)

Article Computer Science, Artificial Intelligence

Gene selection for cancer classification using support vector machines

I Guyon et al.

MACHINE LEARNING (2002)