Articles | Volume 9, issue 1
https://doi.org/10.5194/soil-9-21-2023
© Author(s) 2023. This work is distributed under
the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
https://doi.org/10.5194/soil-9-21-2023
© Author(s) 2023. This work is distributed under
the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
Original research article
| 
11 Jan 2023
Original research article |
| 11 Jan 2023
| 11 Jan 2023
Shapley values reveal the drivers of soil organic carbon stock prediction
Alexandre M. J.-C. Wadoux
CORRESPONDING AUTHOR
Sydney Institute of Agriculture & School of Life and Environmental Sciences, The University of Sydney, Sydney, Australia
Nicolas P. A. Saby
INRAE, Unité de Recherche Info&Sols, Orléans, France
Manuel P. Martin
INRAE, Unité de Recherche Info&Sols, Orléans, France
Related authors
Lei Zhang, Lin Yang, Thomas W. Crowther, Constantin M. Zohner, Sebastian Doetterl, Gerard B. M. Heuvelink, Alexandre M. J.-C. Wadoux, A.-Xing Zhu, Yue Pu, Feixue Shen, Haozhi Ma, Yibiao Zou, and Chenghu Zhou
Earth Syst. Sci. Data, 17, 2605–2623, https://doi.org/10.5194/essd-17-2605-2025, https://doi.org/10.5194/essd-17-2605-2025, 2025
Short summary
Short summary
Current understandings of depth-dependent variations and controls of soil organic carbon turnover time (τ) at global, biome, and local scales remain incomplete. We used the state-of-the-art soil and root profile databases and satellite observations to generate new spatially explicit global maps of topsoil and subsoil τ, with quantified uncertainties for better user applications. The new insights from the resulting maps will facilitate efforts to model the carbon cycle and will support effective carbon management.
Mercedes Román Dobarco, Alexandre M. J-C. Wadoux, Brendan Malone, Budiman Minasny, Alex B. McBratney, and Ross Searle
Biogeosciences, 20, 1559–1586, https://doi.org/10.5194/bg-20-1559-2023, https://doi.org/10.5194/bg-20-1559-2023, 2023
Short summary
Short summary
Soil organic carbon (SOC) is of a heterogeneous nature and varies in chemistry, stabilisation mechanisms, and persistence in soil. In this study we mapped the stocks of SOC fractions with different characteristics and turnover rates (presumably PyOC >= MAOC > POC) across Australia, combining spectroscopy and digital soil mapping. The SOC stocks (0–30 cm) were estimated as 13 Pg MAOC, 2 Pg POC, and 5 Pg PyOC.
Alexandre M. J.-C. Wadoux, José Padarian, and Budiman Minasny
SOIL, 5, 107–119, https://doi.org/10.5194/soil-5-107-2019, https://doi.org/10.5194/soil-5-107-2019, 2019
Lei Zhang, Lin Yang, Thomas W. Crowther, Constantin M. Zohner, Sebastian Doetterl, Gerard B. M. Heuvelink, Alexandre M. J.-C. Wadoux, A.-Xing Zhu, Yue Pu, Feixue Shen, Haozhi Ma, Yibiao Zou, and Chenghu Zhou
Earth Syst. Sci. Data, 17, 2605–2623, https://doi.org/10.5194/essd-17-2605-2025, https://doi.org/10.5194/essd-17-2605-2025, 2025
Short summary
Short summary
Current understandings of depth-dependent variations and controls of soil organic carbon turnover time (τ) at global, biome, and local scales remain incomplete. We used the state-of-the-art soil and root profile databases and satellite observations to generate new spatially explicit global maps of topsoil and subsoil τ, with quantified uncertainties for better user applications. The new insights from the resulting maps will facilitate efforts to model the carbon cycle and will support effective carbon management.
Benjamin Loubet, Nicolas P. Saby, Maryam Gebleh, Pauline Buysse, Jean-Philippe Chenu, Céline Ratie, Claudy Jolivet, Carmen Kalalian, Florent Levavasseur, Jose-Luis Munera-Echeverri, Sebastien Lafont, Denis Loustau, Dario Papale, Giacomo Nicolini, Bruna Winck, and Dominique Arrouays
EGUsphere, https://doi.org/10.5194/egusphere-2025-592, https://doi.org/10.5194/egusphere-2025-592, 2025
Short summary
Short summary
Soil is a large pool of carbon storing globally from two to three times more carbon than the atmosphere and vegetation. We compute the soil stock evolution from 2005 to 2019 for a wheat-maize-barley-oilseed-rape crop rotation at a French crop site. The soil carbon stock decreased by around 70 ± 16 g C m-2 yr-1 over the period, leading to a total loss of around 8 % of the initial soil stock. This strong destocking is primarily explained by a decrease in the residue return to the site.
Amicie A. Delahaie, Lauric Cécillon, Marija Stojanova, Samuel Abiven, Pierre Arbelet, Dominique Arrouays, François Baudin, Antonio Bispo, Line Boulonne, Claire Chenu, Jussi Heinonsalo, Claudy Jolivet, Kristiina Karhu, Manuel Martin, Lorenza Pacini, Christopher Poeplau, Céline Ratié, Pierre Roudier, Nicolas P. A. Saby, Florence Savignac, and Pierre Barré
SOIL, 10, 795–812, https://doi.org/10.5194/soil-10-795-2024, https://doi.org/10.5194/soil-10-795-2024, 2024
Short summary
Short summary
This paper compares the soil organic carbon fractions obtained from a new thermal fractionation scheme and a well-known physical fractionation scheme on an unprecedented dataset of French topsoil samples. For each fraction, we use a machine learning model to determine its environmental drivers (pedology, climate, and land cover). Our results suggest that these two fractionation schemes provide different fractions, which means they provide complementary information.
Christophe Djemiel, Samuel Dequiedt, Walid Horrigue, Arthur Bailly, Mélanie Lelièvre, Julie Tripied, Charles Guilland, Solène Perrin, Gwendoline Comment, Nicolas P. A. Saby, Claudy Jolivet, Antonio Bispo, Line Boulonne, Antoine Pierart, Patrick Wincker, Corinne Cruaud, Pierre-Alain Maron, Sébastien Terrat, and Lionel Ranjard
SOIL, 10, 251–273, https://doi.org/10.5194/soil-10-251-2024, https://doi.org/10.5194/soil-10-251-2024, 2024
Short summary
Short summary
The fungal kingdom has been diversifying for more than 800 million years by colonizing a large number of habitats on Earth. Based on a unique dataset (18S rDNA meta-barcoding), we described the spatial distribution of fungal diversity at the scale of France and the environmental drivers by tackling biogeographical patterns. We also explored the fungal network interactions across land uses and climate types.
Amicie A. Delahaie, Pierre Barré, François Baudin, Dominique Arrouays, Antonio Bispo, Line Boulonne, Claire Chenu, Claudy Jolivet, Manuel P. Martin, Céline Ratié, Nicolas P. A. Saby, Florence Savignac, and Lauric Cécillon
SOIL, 9, 209–229, https://doi.org/10.5194/soil-9-209-2023, https://doi.org/10.5194/soil-9-209-2023, 2023
Short summary
Short summary
We characterized organic matter in French soils by analysing samples from the French RMQS network using Rock-Eval thermal analysis. We found that thermal analysis is appropriate to characterize large set of samples (ca. 2000) and provides interpretation references for Rock-Eval parameter values. This shows that organic matter in managed soils is on average more oxidized and more thermally stable and that some Rock-Eval parameters are good proxies for organic matter biogeochemical stability.
Mercedes Román Dobarco, Alexandre M. J-C. Wadoux, Brendan Malone, Budiman Minasny, Alex B. McBratney, and Ross Searle
Biogeosciences, 20, 1559–1586, https://doi.org/10.5194/bg-20-1559-2023, https://doi.org/10.5194/bg-20-1559-2023, 2023
Short summary
Short summary
Soil organic carbon (SOC) is of a heterogeneous nature and varies in chemistry, stabilisation mechanisms, and persistence in soil. In this study we mapped the stocks of SOC fractions with different characteristics and turnover rates (presumably PyOC >= MAOC > POC) across Australia, combining spectroscopy and digital soil mapping. The SOC stocks (0–30 cm) were estimated as 13 Pg MAOC, 2 Pg POC, and 5 Pg PyOC.
Elisa Bruni, Bertrand Guenet, Yuanyuan Huang, Hugues Clivot, Iñigo Virto, Roberta Farina, Thomas Kätterer, Philippe Ciais, Manuel Martin, and Claire Chenu
Biogeosciences, 18, 3981–4004, https://doi.org/10.5194/bg-18-3981-2021, https://doi.org/10.5194/bg-18-3981-2021, 2021
Short summary
Short summary
Increasing soil organic carbon (SOC) stocks is beneficial for climate change mitigation and food security. One way to enhance SOC stocks is to increase carbon input to the soil. We estimate the amount of carbon input required to reach a 4 % annual increase in SOC stocks in 14 long-term agricultural experiments around Europe. We found that annual carbon input should increase by 43 % under current temperature conditions, by 54 % for a 1 °C warming scenario and by 120 % for a 5 °C warming scenario.
Claire Froger, Nicolas P. A. Saby, Claudy C. Jolivet, Line Boulonne, Giovanni Caria, Xavier Freulon, Chantal de Fouquet, Hélène Roussel, Franck Marot, and Antonio Bispo
SOIL, 7, 161–178, https://doi.org/10.5194/soil-7-161-2021, https://doi.org/10.5194/soil-7-161-2021, 2021
Short summary
Short summary
Pollution of French soils by polycyclic aromatic hydrocarbons (PAHs), known as carcinogenic pollutants, was quantified in this work using an extended data set of 2154 soils sampled across France. The map of PAH concentrations in French soils revealed strong trends in regions with heavy industries and around cities. The PAH signatures indicated the influence of PAH emissions in Europe during the industrial revolution. Health risks posed by PAHs in soils were low but need to be considered.
Alexandre M. J.-C. Wadoux, José Padarian, and Budiman Minasny
SOIL, 5, 107–119, https://doi.org/10.5194/soil-5-107-2019, https://doi.org/10.5194/soil-5-107-2019, 2019
Related subject area
Pedometrics
Quantifying spatial uncertainty to improve soil predictions in data-sparse regions
Insights into the prediction uncertainty of machine-learning-based digital soil mapping through a local attribution approach
Reference soil groups map of Ethiopia based on legacy data and machine learning-technique: EthioSoilGrids 1.0
Accuracy of regional-to-global soil maps for on-farm decision-making: are soil maps “good enough”?
How well does digital soil mapping represent soil geography? An investigation from the USA
Kerstin Rau, Katharina Eggensperger, Frank Schneider, Michael Blaschek, Philipp Hennig, and Thomas Scholten
EGUsphere, https://doi.org/10.5194/egusphere-2025-166, https://doi.org/10.5194/egusphere-2025-166, 2025
Short summary
Short summary
Uneven data collection can make it hard to predict soil properties accurately in new areas. We developed a method to show where predictions are reliable and where more data is needed. By training a model in one region and applying it to another, we found that our approach effectively recognized river patterns but was biased toward overrepresented soil types. This tool can guide smarter data collection, helping improve predictions and make better use of resources for soil management.
Jeremy Rohmer, Stephane Belbeze, and Dominique Guyonnet
SOIL, 10, 679–697, https://doi.org/10.5194/soil-10-679-2024, https://doi.org/10.5194/soil-10-679-2024, 2024
Short summary
Short summary
Machine learning (ML) models have become key ingredients for digital soil mapping. To explain why the ML model is reliable, we apply a popular method from explainable artificial intelligence to the uncertainty prediction, with an application to the mapping of hydrocarbon pollutants on urban soil. We show the benefit of a joint analysis of the influence on the best estimate and the uncertainty to improve communication with end users and support decisions regarding covariates’ characterisation.
Ashenafi Ali, Teklu Erkossa, Kiflu Gudeta, Wuletawu Abera, Ephrem Mesfin, Terefe Mekete, Mitiku Haile, Wondwosen Haile, Assefa Abegaz, Demeke Tafesse, Gebeyhu Belay, Mekonen Getahun, Sheleme Beyene, Mohamed Assen, Alemayehu Regassa, Yihenew G. Selassie, Solomon Tadesse, Dawit Abebe, Yitbarek Wolde, Nesru Hussien, Abebe Yirdaw, Addisu Mera, Tesema Admas, Feyera Wakoya, Awgachew Legesse, Nigat Tessema, Ayele Abebe, Simret Gebremariam, Yismaw Aregaw, Bizuayehu Abebaw, Damtew Bekele, Eylachew Zewdie, Steffen Schulz, Lulseged Tamene, and Eyasu Elias
SOIL, 10, 189–209, https://doi.org/10.5194/soil-10-189-2024, https://doi.org/10.5194/soil-10-189-2024, 2024
Short summary
Short summary
This paper focuses on collating legacy soil profile data and on the production of an updated national soil type map of Ethiopia, EthioSoilGrids version 1.0, using legacy data and a machine-learning approach. Given its quantitative digital representation, the map and the associated data make tremendous contributions to agricultural development planning and digital agricultural solutions, as well as improving the accuracy of global predictive soil mapping efforts.
Jonathan J. Maynard, Edward Yeboah, Stephen Owusu, Michaela Buenemann, Jason C. Neff, and Jeffrey E. Herrick
SOIL, 9, 277–300, https://doi.org/10.5194/soil-9-277-2023, https://doi.org/10.5194/soil-9-277-2023, 2023
Short summary
Short summary
Accurate information on soil properties is critical for identifying soil limitations and the management practices needed to improve crop yields on smallholder farms. This study evaluated the accuracy of soil map information for agronomic decision-making. Based on four publicly available soil maps in Ghana, we found that soil map data significantly overestimated crop suitability, potentially leading to ineffective agronomic investments by smallholder farmers.
David G. Rossiter, Laura Poggio, Dylan Beaudette, and Zamir Libohova
SOIL, 8, 559–586, https://doi.org/10.5194/soil-8-559-2022, https://doi.org/10.5194/soil-8-559-2022, 2022
Short summary
Short summary
Maps of soil properties made by machine learning techniques are increasingly applied in Earth surface process modelling and agronomy. Maps of the same area made by different methods appear quite different and also differ from field-based polygon soil survey maps. We explore these differences both visually and numerically, using methods that quantify the spatial patterns. Readers can apply the methods to their areas of interest in the USA with the supplied R Markdown scripts.
Cited articles
Arrouays, D., Deslais, W., and Badeau, V.: The carbon content of topsoil and
its geographical distribution in France, Soil Use Manage., 17, 7–11,
2001. a
Arrouays, D., Jolivet, C., Boulonne, L., Bodineau, G., Saby, N. P. A., and
Grolleau, E.: A new projection in France: a multi-institutional soil
quality monitoring network, Comptes Rendus de l'Académie d'Agriculture de
France (France), 2002. a
Batjes, N. H.: Total carbon and nitrogen in the soils of the world, Europ.
J. Soil Sci., 47, 151–163, 1996. a
Beucher, A., Rasmussen, C. B., Moeslund, T. B., and Greve, M. H.:
Interpretation of convolutional neural networks for acid sulfate soil
classification, Front. Environ. Sci., 9, 679, https://doi.org/10.3389/fenvs.2021.809995, 2022. a, b, c
Breiman, L.: Random forests, Mach. Learn., 45, 5–32, 2001. a
Chen, S., Arrouays, D., Angers, D. A., Chenu, C., Barré, P., Martin, M. P.,
Saby, N. P. A., and Walter, C.: National estimation of soil organic carbon
storage potential for arable soils: A data-driven approach coupled with
carbon-landscape zones, Sci. Total Environ., 666, 355–367,
2019. a
Greenwell, B.: Package “fastshap”, R package
version 0.0.5,
https://CRAN.R-project.org/package=fastshap (last access: 10 April 2022), 2020. a
Guo, L., Sun, X., Fu, P., Shi, T., Dang, L., Chen, Y., Linderman, M., Zhang, G., Zhang, Y., Jiang, Q., Zhang, H., and Zeng, C.: Mapping soil organic carbon stock by
hyperspectral and time-series multispectral remote sensing images in
low-relief agricultural areas, Geoderma, 398, 115118, https://doi.org/10.1016/j.geoderma.2021.115118, 2021. a
Hengl, T., Mendes de Jesus, J., Heuvelink, G. B. M., Ruiperez Gonzalez, M.,
Kilibarda, M., Blagotić, A., Shangguan, W., Wright, M. N., Geng, X.,
Bauer-Marschallinger, B., Guevara, M. A., Vargas, R., MacMillan, R. A., Batjes, N. H., Leenaars, J. G. B., Ribeiro, E., Wheeler, I., Mantel, S., and Kempen, B.: SoilGrids250m: Global gridded soil
information based on machine learning, PLOS ONE, 12, e0169748, https://doi.org/10.1371/journal.pone.0169748, 2017. a
ISO 10694: Soil quality – Determination of organic and total carbon
after dry combustion (elementary analysis), Standard, International
Organization for Standardization, Geneva, CH, 1995. a
Jones, A., Montanarella, L., and Jones, R.: Soil Atlas of Europe, European
Commission, 2005. a
Keenor, S. G., Rodrigues, A. F., Mao, L., Latawiec, A. E., Harwood, A. R., and
Reid, B. J.: Capturing a soil carbon economy, Roy. Soc. Open Sci., 8,
202305, https://doi.org/10.1098/rsos.202305, 2021. a
Kempen, B., Dalsgaard, S., Kaaya, A. K., Chamuya, N.,
Ruipérez-González, M., Pekkarinen, A., and Walsh, M. G.: Mapping
topsoil organic carbon concentrations and stocks for Tanzania, Geoderma,
337, 164–180, 2019. a
Lacoste, M., Martin, M. P., Saby, N. P. A., Paroissien, J.-B., Lehmann, S.,
Richer-De-Forges, A. C., and Arrouays, D.: Carbon content and stocks in the
O horizons of French forest soils, in: GlobalSoilMap: Basis of the Global
Spatial Soil Information System, edited by: Arrouays, D., McKenzie, N.,
Hempel, J., de Forges, A. R., and McBratney, A. B., CRC Press, Boca Raton,
FL, 2014. a
Laroche, B., Richer-De-Forges, A. C., Leménager, S., Arrouays, D.,
Schnebelen, N., Eimberck, M., Toutain, B., Lehmann, S., Nguenkam, M.-E. T.,
Héliès, F., Chenu, J.-P., Parot, S., Desbourdes, S., Girot, G., Voltz, M., and Bardy, M.: Le programme inventaire gestion conservation des
sols de France: volet référentiel régional pédologique,
Étude et Gestion des Sols, 21, 25–36, 2014. a
Lemenih, M. and Itanna, F.: Soil carbon stocks and turnovers in various
vegetation types and arable lands along an elevation gradient in southern
Ethiopia, Geoderma, 123, 177–188, 2004. a
Lugato, E., Panagos, P., Bampa, F., Jones, A., and Montanarella, L.: A new
baseline of organic carbon stock in European agricultural soils using a
modelling approach, Glob. Change Biol., 20, 313–326, 2014. a
Lundberg, S. M. and Lee, S.-I.: A unified approach to interpreting model
predictions, in: Proceedings of the 31st International Conference on Neural
Information Processing Systems, edited by: von Luxburg, U., Guyon, I., Bengio, S., Wallach, H., and Fergus, R., 4768–4777, Curran Associates Inc., Red Hook,
New York, 2017. a
Lyapustin, A., Wang, Y., Korkin, S., and Huang, D.: MODIS Collection 6 MAIAC algorithm, Atmos. Meas. Tech., 11, 5741–5765, https://doi.org/10.5194/amt-11-5741-2018, 2018. a
Martin, M. P., Orton, T. G., Lacarce, E., Meersmans, J., Saby, N. P. A.,
Paroissien, J. B., Jolivet, C., Boulonne, L., and Arrouays, D.: Evaluation of
modelling approaches for predicting the spatial distribution of soil organic
carbon stocks at the national scale, Geoderma, 223, 97–107, 2014. a, b, c
Martin, M. P., Dimassi, B., Román Dobarco, M., Guenet, B., Arrouays, D.,
Angers, D. A., Blache, F., Huard, F., Soussana, J.-F., and Pellerin, S.:
Feasibility of the 4 per 1000 aspirational target for soil carbon: A case
study for France, Glob. Change Biol., 27, 2458–2477, 2021. a
Meersmans, J., Martin, M. P., Lacarce, E., De Baets, S., Jolivet, C., Boulonne,
L., Lehmann, S., Saby, N. P. A., Bispo, A., and Arrouays, D.: A high
resolution map of French soil organic carbon, Agron. Sustain.
Dev., 32, 841–851, 2012. a
Minasny, B., McBratney, A. B., Malone, B. P., and Wheeler, I.: Digital mapping
of soil carbon, Adv. Agron., 118, 1–47, 2013. a
Mishra, U., Lal, R., Slater, B., Calhoun, F., Liu, D., and Van Meirvenne, M.:
Predicting soil organic carbon stock using profile depth distribution
functions and ordinary kriging, Soil Sci. Soc. Am. J., 73,
614–621, 2009. a
Mohammadifar, A., Gholami, H., Comino, J. R., and Collins, A. L.: Assessment of
the interpretability of data mining for the spatial modelling of water
erosion using game theory, Catena, 200, 105178, https://doi.org/10.1016/j.catena.2021.105178, 2021. a
Molnar, C.: Interpretable Machine Learning: A Guide for Making Black Box Models
Explainable, Lulu Press, Raleigh, 2020. a
Orchard, V. A. and Cook, F.: Relationship between soil respiration and soil
moisture, Soil Biology and Biochemistry, 15, 447–453, 1983. a
Padarian, J., McBratney, A. B., and Minasny, B.: Game theory interpretation of digital soil mapping convolutional neural networks, SOIL, 6, 389–397, https://doi.org/10.5194/soil-6-389-2020, 2020. a, b, c
Pelletier, J. D., Broxton, P. D., Hazenberg, P., Zeng, X., Troch, P. A., Niu,
G.-Y., Williams, Z., Brunke, M. A., and Gochis, D.: A gridded global data set
of soil, intact regolith, and sedimentary deposit thicknesses for regional
and global land surface modeling, J. Adv. Model. Ea.
Syst., 8, 41–65, 2016. a
Plutzar, C., Kroisleitner, C., Haberl, H., Fetzel, T., Bulgheroni, C.,
Beringer, T., Hostert, P., Kastner, T., Kuemmerle, T., Lauk, C., Levers, C., Lindner, M., Moser, D., Müller, D., Niedertscheider, M., Paracchini, M., Schaphoff, S., Verburg, P., Verkerk, P. J., and Erb, K.-H:
Changes in the spatial patterns of human appropriation of net primary
production (HANPP) in Europe 1990–2006, Reg. Environ. Change, 16,
1225–1238, 2016. a
Poggio, L., de Sousa, L. M., Batjes, N. H., Heuvelink, G. B. M., Kempen, B., Ribeiro, E., and Rossiter, D.: SoilGrids 2.0: producing soil information for the globe with quantified spatial uncertainty, SOIL, 7, 217–240, https://doi.org/10.5194/soil-7-217-2021, 2021. a, b
Probst, P., Wright, M. N., and Boulesteix, A.-L.: Hyperparameters and tuning
strategies for random forest, Wiley Interdisciplinary Reviews, Data Mining
and Knowledge Discovery, 9, e1301, https://doi.org/10.1002/widm.1301, 2019. a, b
R Core Team: R: A Language and Environment for Statistical Computing, R
Foundation for Statistical Computing, Vienna, Austria,
https://www.R-project.org/, last access: 10 April 2022. a
Rabus, B., Eineder, M., Roth, A., and Bamler, R.: The shuttle radar topography
mission – a new class of digital elevation models acquired by spaceborne
radar, ISPRS J. Photogramm., 57, 241–262,
2003. a
Rahman, N., de Neergaard, A., Magid, J., van de Ven, G. W. J., Giller, K. E.,
and Bruun, T. B.: Changes in soil organic carbon stocks after conversion from
forest to oil palm plantations in Malaysian Borneo, Environ. Res.
Lett., 13, 105001, https://doi.org/10.1088/1748-9326/aade0f, 2018. a
Reichstein, M., Bahn, M., Ciais, P., Frank, D., Mahecha, M. D., Seneviratne,
S. I., Zscheischler, J., Beer, C., Buchmann, N., Frank, D. C., Papale, D., Rammig, A., Smith, P., Thonicke, K., van der Velde, M., Vicca, S., Walz, A., and Wattenbach, M.:
Climate extremes and the carbon cycle, Nature, 500, 287–295, 2013. a
Rovira, P., Sauras-Yera, T., and Romanyà, J.: Equivalent-mass versus
fixed-depth as criteria for quantifying soil carbon sequestration: How
relevant is the difference?, Catena, 214, 106283, https://doi.org/10.1016/j.catena.2022.106283, 2022. a
Running, S. and Zhao, M.: MOD17A3HGF MODIS/Terra net primary production
gap-filled yearly L4 global 500 m SIN grid V006, NASA EOSDIS land processes
DAAC, 2019. a
Saby, N. P. A., Arrouays, D., Antoni, V., Lemercier, B., Follain, S., Walter,
C., and Schvartz, C.: Changes in soil organic carbon in a mountainous
French region, 1990–2004, Soil Use Manage., 24, 254–262, 2008. a
Saby, N. P. A., Chenu, J.-P., Szergi, T., Csorba, A., Bertuzzi, P., Toutain, B., Picaud, C., Gay, L., and Creamer, R.: French RMQS soil profile and monitoring dataset with related management practices data, Recherche Data Gouv, V1 [data set], https://doi.org/10.15454/AIQ9WS, 2020. a
Shapley, L. S.: A Value for n-Person Games, in: Contributions to the Theory of
Games, edited by: Harold William, K. and Albert William, T., Vol. 28,
Annals of Mathematics Studies, chap. 17, 31–40, Princeton University
Press, Princeton, 1953. a
Stewart, C. E., Plante, A. F., Paustian, K., Conant, R. T., and Six, J.: Soil
carbon saturation: linking concept and measurable carbon pools, Soil Sci.
Soc. Am. J., 72, 379–392, 2008. a
Thierion, V., Ghaith, A., Billecocq, P., Gaudé, C., Mesona, L., Laurent,
B., Bertrand, M., and Bigot, S.: D’OSO à la cartographie de
végétation par télédétection multi-temporelle–Exemples
d’utilisation des images Sentinel-2, in: Colloque de Bilan et de
Prospective du PNTS, 2018. a
Trabucco, A., Zomer, R. J., Bossio, D. A., van Straaten, O., and Verchot,
L. V.: Climate change mitigation through afforestation/reforestation: a
global analysis of hydrologic impacts with four case studies, Agr.
Ecosyst. Environ., 126, 81–97, 2008. a
Tuanmu, M.-N. and Jetz, W.: A global, remote sensing-based characterization of
terrestrial habitat heterogeneity for biodiversity and ecosystem modelling,
Global Ecol. Biogeogr., 24, 1329–1339, 2015. a
Van Wesemael, B., Paustian, K., Meersmans, J., Goidts, E., Barancikova, G., and
Easter, M.: Agricultural management explains historic changes in regional
soil carbon stocks, P. Natl. Acad. Sci. USA, 107,
14926–14930, 2010. a
Vos, C., Don, A., Hobley, E. U., Prietz, R., Heidkamp, A., and Freibauer, A.:
Factors controlling the variation in organic carbon stocks in agricultural
soils of Germany, Europ. J. Soil Sci., 70, 550–564, 2019. a
Wadoux, A. M. J.-C. and Molnar, C.: Beyond prediction: methods for interpreting
complex models of soil variation, Geoderma, 422, 115953, https://doi.org/10.1016/j.geoderma.2022.115953, 2022. a, b
Wadoux, A. M. J.-C., Samuel-Rosa, A., Poggio, L., and Mulder, V. L.: A note on
knowledge discovery and machine learning in digital soil mapping, Europ.
J. Soil Sci., 71, 133–136, 2020. a
Wadoux, A. M. J.-C., Heuvelink, G. B. M., Lark, R. M., Lagacherie, P., Bouma,
J., Mulder, V. L., Libohova, Z., Yang, L., and McBratney, A. B.: Ten
challenges for the future of pedometrics, Geoderma, 401, 115155, https://doi.org/10.1016/j.geoderma.2021.115155,
2021a. a
Wadoux, A. M. J.-C., Román-Dobarco, M., and McBratney, A. B.: Perspectives
on data-driven soil research, Europ. J. Soil Sci., 72,
1675–1689, 2021b. a
Wang, B., Gray, J. M., Waters, C. M., Anwar, M. R., Orgill, S. E., Cowie,
A. L., Feng, P., and Li Liu, D.: Modelling and mapping soil organic carbon
stocks under future climate change in south-eastern Australia, Geoderma,
405, 115442, https://doi.org/10.1016/j.geoderma.2021.115442, 2022. a
Wright, M. N. and Ziegler, A.: ranger: A Fast Implementation of Random
Forests for High Dimensional Data in C
Zomer, R. J., Trabucco, A., Bossio, D. A., and Verchot, L. V.: Climate change
mitigation: A spatial analysis of global land suitability for clean
development mechanism afforestation and reforestation, Agr.
Ecosyst. Environ., 126, 67–80, 2008. a
Short summary
We introduce Shapley values for machine learning model interpretation and reveal the local and global controlling factors of soil organic carbon (SOC) stocks. The method enables spatial analysis of the important variables. Vegetation and topography determine much of the SOC stock variation in mainland France. We conclude that SOC stock variation is complex and should be interpreted at multiple levels.
We introduce Shapley values for machine learning model interpretation and reveal the local and...
