Show simple item record

resumen

Abstract
Spatially resolved estimates of change in soil organic carbon (SOC) stocks are necessary for supporting national and international policies aimed at achieving land degradation neutrality and climate change mitigation. In this work we report on the development, implementation and application of a data-driven, statistical method for mapping SOC stocks in space and time, using Argentina as a pilot. We used quantile regression forest machine learning to [ver mas...]
dc.contributor.authorHeuvelink, Gerard B.M.
dc.contributor.authorAngelini, Marcos Esteban
dc.contributor.authorPoggio, Laura
dc.contributor.authorBai, Zhanguo
dc.contributor.authorBatjes, Niels H.
dc.contributor.authorvan den Bosch, Rik
dc.contributor.authorBossio, Deborah
dc.contributor.authorEstella, Sergio
dc.contributor.authorLehmann, Johannes
dc.contributor.authorOlmedo, Guillermo Federico
dc.contributor.authorSanderman, Jonathan
dc.date.accessioned2020-10-15T11:17:38Z
dc.date.available2020-10-15T11:17:38Z
dc.date.issued2020-05-20
dc.identifier.issn1365-2389
dc.identifier.otherhttps://doi.org/10.1111/ejss.12998
dc.identifier.urihttp://hdl.handle.net/20.500.12123/8054
dc.identifier.urihttps://onlinelibrary.wiley.com/doi/full/10.1111/ejss.12998
dc.description.abstractSpatially resolved estimates of change in soil organic carbon (SOC) stocks are necessary for supporting national and international policies aimed at achieving land degradation neutrality and climate change mitigation. In this work we report on the development, implementation and application of a data-driven, statistical method for mapping SOC stocks in space and time, using Argentina as a pilot. We used quantile regression forest machine learning to predict annual SOC stock at 0–30 cm depth at 250 m resolution for Argentina between 1982 and 2017. The model was calibrated using over 5,000 SOC stock values from the 36-year time period and 35 environmental covariates. We preprocessed normalized difference vegetation index (NDVI) dynamic covariates using a temporal low-pass filter to allow the SOC stock for a given year to depend on the NDVI of the current as well as preceding years. Predictions had modest temporal variation, with an average decrease for the entire country from 2.55 to 2.48 kg C m−2 over the 36-year period (equivalent to a decline of 211 Gg C, 3.0% of the total 0–30 cm SOC stock in Argentina). The Pampa region had a larger estimated SOC stock decrease from 4.62 to 4.34 kg C m−2 (5.9%) during the same period. For the 2001–2015 period, predicted temporal variation was seven-fold larger than that obtained using the Tier 1 approach of the Intergovernmental Panel on Climate Change and United Nations Convention to Combat Desertification. Prediction uncertainties turned out to be substantial, mainly due to the limited number and poor spatial and static, whereas SOC is dynamic and SOC dynamics are of particular interest to carbon sequestration and land degradation studies. Thus, there is a clear need to extend spatial SOC mapping to space–time SOC mapping. temporal distribution of the calibration data, and the limited explanatory power of the covariates. Cross-validation confirmed that SOC stock prediction accuracy was limited, with a mean error of 0.03 kg C m−2 and a root mean squared error of 2.04 kg C m−2. In spite of the large uncertainties, this work showed that machine learning methods can be used for space–time SOC mapping and may yield valuable information to land managers and policymakers, provided that SOC observation density in space and time is sufficiently large.eng
dc.formatapplication/pdfes_AR
dc.language.isoenges_AR
dc.publisherWiley
dc.rightsinfo:eu-repo/semantics/openAccesses_AR
dc.sourceEuropean Journal of Soil Science : 1-17 (First published: 20 May 2020)es_AR
dc.subjectArgentinaes_AR
dc.subjectCarbon Stock Assessmentseng
dc.subjectEstimación de las Existencias de Carbonoes_AR
dc.subjectClimate Changeeng
dc.subjectCambio Climáticoes_AR
dc.subjectLand Degradationeng
dc.subjectDegradación de Tierrases_AR
dc.subject.otherQuantile Regression Roresteng
dc.subject.otherBosque de Regresión de Cuantileses_AR
dc.subject.otherSpace-time Mappingeng
dc.subject.otherMapeo Espacio-tiempoes_AR
dc.titleMachine learning in space and time for modelling soil organic carbon changees_AR
dc.typeinfo:ar-repo/semantics/artículoes_AR
dc.typeinfo:eu-repo/semantics/articlees_AR
dc.typeinfo:eu-repo/semantics/publishedVersiones_AR
dc.description.filFil: Heuvelink, Gerard B.M. ISRIC - World soil information; Holanda. Wageningen University. Soil Geography and Landscape Group; Holandaes_AR
dc.description.filFil: Angelici, Marcos E. Instituto Nacional de Tecnología Agropecuaria (INTA). Instituto de Suelos; Argentinaes_AR
dc.description.filFil: Poggio, Laura ISRIC - World soil information, Wageningen; Holandaes_AR
dc.description.filFil: Bai, Zhanguo ISRIC - World soil information, Wageningen, The Netherlandses_AR
dc.description.filFil: Batjes, Niels H. ISRIC - World soil information, Wageningen, The Netherlandses_AR
dc.description.filFil: an den Bosch, Rik ISRIC - World soil information, Wageningen, The Netherlandses_AR
dc.description.filFil: Bossio, Deborah The Nature Conservancy; Estados Unidoses_AR
dc.description.filFil: Estella, Sergio Vizzuality; Españaes_AR
dc.description.filFil: Lehmann, Jhoannes. Cornell University. Soil and Crop Sciences; Estados Unidoses_AR
dc.description.filFil: Olmedo, Guillermo F. Instituto Nacional de Tecnología Agropecuaria (INTA). Estación Experimental Agropecuaria Mendoza; Argentinaes_AR
dc.description.filFil: Sandermann, Jonathan. Woods Hole Research Center; Estados Unidoses_AR
dc.subtypecientifico


Files in this item

Thumbnail

This item appears in the following Collection(s)

common

Show simple item record