Log‐ratio transformation is the key to determining soil organic carbon fractions with near‐infrared spectroscopy

Jaconi, Angélica; Poeplau, Christopher; López, Luciano Espino; Wesemael, Bas van; Don, Axel

doi:10.1111/ejss.12761

Cited by 21 publications

(12 citation statements)

References 36 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…POC:TOC ratios were determined using near-infrared spectroscopy (FT-NIRS; MPA, Bruker Optik GmbH, Ettlingen, Germany). The basic concept for this approach has recently been proven by Jaconi et al (2019) who also used soil samples from the German Agricultural Soil Inventory, but focused on topsoil (0-10 cm) only. Here, we present POC:TOC ratios for all depth increments of the German Agricultural Soil Inventory.…”

Section: Sampling Design and Data Acquisitionmentioning

confidence: 99%

“…Here, we present POC:TOC ratios for all depth increments of the German Agricultural Soil Inventory. Our predictions of POC:TOC ratios are based on the same 105 topsoil samples used by Jaconi et al (2019), plus additional subsoil samples from 27 of these 105 sites. The 27 subsoil samples were selected to cover the complete range of clay content and C:N ratios of subsoils in order to get a representative subset of the subsoil samples ( Fig.…”

Section: Sampling Design and Data Acquisitionmentioning

confidence: 99%

“…POC was defined as the light C fraction with a density of < 1.6 g cm − 3 while MOC was defined as the remaining, heavier C fraction. We followed exactly the fractionation protocol, which was previously described by Jaconi et al (2019) for topsoil. However, because the POC content of subsoil was about one order of magnitude lower than in the topsoil, up to 120 g soil instead of 10 g for topsoils were fractionated.…”

Section: Sampling Design and Data Acquisitionmentioning

confidence: 99%

“…TOC recovery for subsoil samples ranged from 80 to 120%. Spectral pretreatments and models were implemented as proposed by Jaconi et al (2019). Additional tree-based algorithms (ranger, cubist) and modelling approaches (log-ratio transformation of POC and MOC content vs. log-ratio transformation of POC:TOC and POC:MOC, separate prediction of POC and MOC contents vs. separate prediction of POC:TOC and MOC:TOC ratios) were also tested, as detailed in the Supplementary Material.…”

Section: Sampling Design and Data Acquisitionmentioning

confidence: 99%

“…Model performance was judged based on leave-one-site-out validation, i.e. the data was partitioned by sites and the values of each site were predicted based on models that were trained only based on the samples from the other sites (Jaconi et al, 2019). This validation procedure is more robust than partitioning the dataset just once into training and testing (Hastie et al, 2009).…”

Section: Sampling Design and Data Acquisitionmentioning

confidence: 99%

See 4 more Smart Citations

Origin of carbon in agricultural soil profiles deduced from depth gradients of C:N ratios, carbon fractions, δ13C and δ15N values

2020

View full text Add to dashboard Cite

Aims Agricultural soils in Germany store 2.54 Pg of organic carbon (C). However, information about how and when this C entered the soils is limited. This study illustrates how depth profiles of organic matter can shed light on different entry paths of organic C. Methods Machine learning was used to explain total organic C (TOC), C:N, particulate organic C (POC), δ13C and δ15N values down to 100 cm depth based on pedology, geology, climate and management-related variables from the German Agricultural Soil Inventory. We estimated TOC turnover rates based on the relationship between the proportion of maize (only C4 plant) in crop rotations and soil δ13C values. Results In the upper 30 cm of cropland, fresh photosynthates added on average 0.2 to 0.8 Mg C ha− 1 year− 1. Organic fertiliser was another source of topsoil C, especially in grassland. Sandy sites in north-west Germany contained historic C from past heathland and peatland. One third of German agricultural land was found to be on colluvial and alluvial deposits, in which allochthonous C from upstream and upslope areas evidently increased the TOC content of subsoils. In and below hardpans, TOC content and C:N and POC:TOC ratios were low, indicating restricted root-derived C input. Conclusions Our data indicate that ongoing management in German agricultural soils mainly affects topsoil C, while C storage in subsoils reveals significant legacies from allochthonous, buried or translocated C inputs. Specific attention should be focused on the sustainable loosening of hardpans that could result in a slow, but significant increase in subsoil C stocks.

show abstract

Section: Sampling Design and Data Acquisitionmentioning

confidence: 99%

Section: Sampling Design and Data Acquisitionmentioning

confidence: 99%

Section: Sampling Design and Data Acquisitionmentioning

confidence: 99%

Section: Sampling Design and Data Acquisitionmentioning

confidence: 99%

Section: Sampling Design and Data Acquisitionmentioning

confidence: 99%

See 3 more Smart Citations

Origin of carbon in agricultural soil profiles deduced from depth gradients of C:N ratios, carbon fractions, δ13C and δ15N values

2020

View full text Add to dashboard Cite

show abstract

Predicting wetland soil properties using machine learning, geophysics, and soil measurement data

Driba,

Emmanuel,

Doro

2024

J Soils Sediments

View full text Add to dashboard Cite

Purpose Machine learning models can improve the prediction of spatial variation of wetland soil properties, such as soil moisture content (SMC) and soil organic matter (SOM). Their performance, however, relies on the quantity of data used to train the model, limiting their use with insufficient data. In this study, we assessed the use of synthetic data constrained by limited field data for training an eXtreme Gradient Boosting (XGBoost) algorithm used to predict the distribution of soil properties based on geophysical measurements constrained by soil samples. Materials and methods A spatial distribution of soil apparent electrical conductivity (ECa) and laboratory measurements of SOM and SMC from twenty-two core samples were acquired at the St. Michael restored wetland near Defiance, Ohio. The correlations between ECa, SOM, and SMC were explored for predicting the spatial distribution of SOM and SMC. We used a Beta Variational AutoEncoder (β-VAE) approach to synthetically generate over 70,000 training data from the original twenty-two data from soil cores. The training data samples were taken from the latent space. The XGBoost algorithm was then trained on the β-VAE generated data and used to predict the spatial distribution of SOM and SMC at the site. We also validated the accuracy of the XGBoost predictions using an original holdout model validation technique. Results and discussions The generated synthetic data using the β-VAE include both soil attributes and ECa, which are larger and more diverse than the original training set with an absolute mean reconstructed error for SMC and SOM ranging from 0.018 to 0.022 and 0.026 to 0.041, respectively. This indicates that the β-VAE successfully generated a realistic synthetic dataset and overcame the technical barrier of using limited datasets. In addition, using generated data to expand the original training data helps the XGBoost model make more accurate predictions compared to training on the original data. The XGBoost prediction performance yielded average Lin’s concordance correlation coefficient (LCCC) values of 0.82 and 0.85 for SOM and SMC and a ratio of performance to deviation (RPD) values of 1.92 and 2.22 respectively, indicating a good performance. Conclusions This study validated the use of β-VAE to successfully generate synthetic wetland soil datasets with attributes of the original field data that can be effectively used to train the machine learning XGBoost model. The proposed framework offers an efficient solution for mapping the spatial variability of soil property in data-scarce wetland soil environments.

show abstract

An indicator for organic matter dynamics in temperate agricultural soils

Wesemael

Chartin

Wiesmeier

et al. 2019

Agriculture, Ecosystems & Environment

View full text Add to dashboard Cite

The heterogeneity of soil organic matter (SOM) and the small changes in soil organic carbon (SOC) compared to large total SOC stocks hinder a robust estimation of SOC turnover, in particular for more stable SOC. We developed a simple fractionation protocol for agricultural topsoils and tested it extensively on a range of soils in southern Belgium, including farmed soils, soils from long-term field trials, and paired sites after recent conversion to conservation farming. Our simple fractionation involves shaking the soil, wet sieving over 20 μm and analysing the SOC concentration in the soil as well as in the fine fraction (< 20 μm). Eight biological indicators measured in an earlier study across the same monitoring network for the 0-10 cm topsoil were analysed in a conditional inference forest model in order to investigate the factors influencing the SOC fractions. Soil microbial biomass N explained the largest proportion of variation in both fractions. The fine fraction was also associated with factors explaining the regional trend in SOC distribution such as farmyard manure input, precipitation, land use and flow length. The variation in SOC content between treatments both in long-term trials and in farmers' fields converted to conservation management was mainly attributed to changes within the coarse fraction. Thus, this fraction proves to be sensitive to management changes, although care should be taken to sample deep enough to represent the former plough layer inherited from the conventional tillage practice. Furthermore, the ratio between the coarse and the fine fraction showed a linear relationship (r² = 0.66) with the relative changes in SOC concentration over the last ten years. These fractions derived from a simple analytical approach are thus useful as an indicator for changes in SOC concentration. In analogy to biological indicators such as the soil microbial biomass C, the relationship between the fractions and relative changes in SOC concentration are likely to depend on climate conditions. Our methodology provides an indicator for use in routine analysis of agricultural topsoils, which is capable of predicting the effects of management practices on SOC concentrations in the short to mid-term (5-10 years).

show abstract

Log‐ratio transformation is the key to determining soil organic carbon fractions with near‐infrared spectroscopy

Cited by 21 publications

References 36 publications

Origin of carbon in agricultural soil profiles deduced from depth gradients of C:N ratios, carbon fractions, δ13C and δ15N values

Origin of carbon in agricultural soil profiles deduced from depth gradients of C:N ratios, carbon fractions, δ13C and δ15N values

Predicting wetland soil properties using machine learning, geophysics, and soil measurement data

An indicator for organic matter dynamics in temperate agricultural soils

Contact Info

Product

Resources

About