Soil legacy data rescue via GlobalSoilMap and other international and national initiatives

Arrouays, Dominique; Leenaars, J.G.B.; Richer-de-Forges, Anne C.; Adhikari, Kabindra; Ballabio, Cristiano; Greve, Mogens Humlekrog; Grundy, Mike; Guerrero, Eliseo; Hempel, Jon; Hengl, Tomislav; Heuvelink, G.B.M.; Batjes, N.H.; Carvalho, E. Burigo De; Hartemink, Alfred E.; Hewitt, Alan J.; Hong, Suk-Young; Krasilnikov, Pavel; Lagacherie, Philippe; Lelyk, Glen; Libohova, Zamir; Lilly, Allan; McBratney, Alex B.; McKenzie, N. J.; Vasquez, Gustavo M.; Mulder, Vera Leatitia; Minasny, Budiman; Montanarella, Luca; Odeh, Inakwu O. A.; Padarian, José; Poggio, Laura; Roudier, Pierre; Saby, Nicolas P.A.; Savin, I. Yu.; Searle, Ross; Solbovoy, Vladimir; Thompson, James A.; Smith, Scott; Sulaeman, Yiyi; Vintila, R.; Rossel, Raphael Viscarra; Wilson, Peter; Zhang, Gan‐Lin; Swerts, M; Oorts, Katrien; Kārkliņš, A.; Liu, Feng; Navarro, Alexandro R. Ibelles; Levin, Arkadiy; Laktionova, T.M.; Dell'Acqua, Martin; Suvannang, Nopmanee; Ruam, Waew; Prasad, Jagdish; Patil, N. G.; Husnjak, Stjepan; Pásztor, László; Okx, J. P.; Hallett, Stephen R; Keay, C. A.; Farewell, Timothy S.; Lilja, Harri; Juilleret, Jérôme; Marx, Simone; Takata, Yusuke; Yagi, Kazuichi; Mansuy, Nicolas; Panagos, Panos; Liedekerke, Mark van; Skalský, Rastislav; Sobocká, Jaroslava; Kobza, Josef; Eftekhari, Kamran; Alavipanah, Seyed Kacem; Moussadek, Rachid; Badraoui, Mohamed; Silva, Mayesse Aparecida da; Paterson, Garry; Gonçalves, M. C.; Theocharopoulos, Sid; Yemefack, Martin; Tedou, Silatsa; Vrščaj, Borut; Grob, Urs; Kozák, Josef; Borůvka, Luboš; Dobos, Endre; Taboada, Miguel Ángel; Moretti, Lucas; Rodríguez, Dario

doi:10.1016/j.grj.2017.06.001

Cited by 127 publications

(67 citation statements)

References 82 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Thanks to the effort of some groups to rescue soil legacy data (Arrouays et al, 2017), and cheaper and faster methods to analyse soil samples, there is more soil data available than ever before. This data availability not only allow us to use new ML algorithms which usually require more observations but also opens the door to new ways to train those models.…”

Section: New Good Practicesmentioning

confidence: 99%

Machine learning and soil sciences: A review aided by machine learning tools

Padarian¹,

Minasny²,

McBratney³

2019

Preprint

Self Cite

View full text Add to dashboard Cite

Abstract. The application of machine learning (ML) techniques in various fields of science has increased rapidly, especially in the last ten years. The increasing availability of soil data that can be efficiently acquired remotely and proximally, and freely available open-source algorithms, have led to an accelerated adoption of ML techniques to analyse soil data. Given the large number of publications, it is an impossible task to manually review all papers on the application of ML in soil science without narrowing down a narrative of ML application in a specific research question. This paper aims to provide a comprehensive review of the application of ML techniques in soil science aided by a ML algorithm (Latent Dirichlet Allocation) to find patterns in a large collection of text corpus. The objective is to gain insight into publications of ML applications in soil science and to discuss the research gaps in this topic. We found that: a) there is an increasing usage of ML methods in soil sciences, mostly concentrated in developed countries, b) the reviewed publication can be grouped into 12 topics, namely remote sensing, soil organic carbon, water, contamination, methods (ensembles), erosion and parent material, methods (NN, SVM), spectroscopy, modelling (classes), crops, physical and modelling (continuous), c) advanced ML methods usually perform better than simpler approaches thanks to their capability to capture non-linear relationships. From these findings, we found research gaps, in particular: about the precautions that should be taken (parsimony) to avoid overfitting, and that the interpretability of the ML models is an important aspect to consider when applying advanced ML methods in order to improve our knowledge and understanding of soil. We foresee that a large number of studies will focus on the latter topic.

show abstract

Section: New Good Practicesmentioning

confidence: 99%

Machine learning and soil sciences: A review aided by machine learning tools

Padarian¹,

Minasny²,

McBratney³

2019

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…However, the availability of public SOC information is limited across large areas of Latin America and large discrepancies exist in current global SOC estimates (Tifafi et al, 2018). Thus, there is a pressing need to validate the accuracy of global SOC estimates, improve interoperability and contribute to the capacity of countries to meet the GlobalSoilMap specifications (Arrouays et al, 2017) to inform policy decisions around climate change mitigation strategies.…”

Section: Introductionmentioning

confidence: 99%

No silver bullet for digital soil mapping: country-specific soil organic carbon estimates across Latin America

Guevara

Olmedo

Stell

et al. 2018

SOIL

View full text Add to dashboard Cite

Abstract. Country-specific soil organic carbon (SOC) estimates are the baseline for the Global SOC Map of the Global Soil Partnership (GSOCmap-GSP). This endeavor is key to explaining the uncertainty of global SOC estimates but requires harmonizing heterogeneous datasets and building country-specific capacities for digital soil mapping (DSM). We identified country-specific predictors for SOC and tested the performance of five predictive algorithms for mapping SOC across Latin America. The algorithms included support vector machines (SVMs), random forest (RF), kernel-weighted nearest neighbors (KK), partial least squares regression (PL), and regression kriging based on stepwise multiple linear models (RK). Country-specific training data and SOC predictors (5 × 5 km pixel resolution) were obtained from ISRIC -World Soil Information. Temperature, soil type, vegetation indices, and topographic constraints were the best predictors for SOC, but country-specific predictors and their respective weights varied across Latin America. We compared a large diversity of country-specific datasets and models, and were able to explain SOC variability in a range between ∼ 1 and ∼ 60 %, with no universal predictive algorithm among countries. A regional (n = 11 268 SOC estimates) ensemble of these five algorithms was able to explain ∼ 39 % of SOC variability from repeated 5-fold cross-validation. We report a combined SOC stock of 77.8 ± 43.6 Pg (uncertainty represented by the full conditional response of independent model residuals) across Latin America. SOC stocks were higher in tropical forests (30 ± 16.5 Pg) and croplands (13 ± 8.1 Pg). Country-specific and regional ensembles revealed spatial discrepancies across geopolitical borders, higher elevations, and coastal plains, but provided similar regional stocks (77.8 ± 42.2 and 76.8 ± 45.1 Pg, respectively). These results are conservative compared to global estimates (e.g., SoilGrids250m 185.8 Pg, the Harmonized World Soil Database 138.4 Pg, or the GSOCmap-GSP 99.7 Pg). Countries with large area (i.e., Brazil, Bolivia, Mexico, Peru) and large spatial SOC heterogeneity had lower SOC stocks per unit area and larger uncertainty in their predictions. We highlight that expert opinion is needed to set boundary prediction limits to avoid unrealistically high modeling estimates. For maximizing explained variance while minimizing prediction bias, the selection of predictive algorithms for SOC mapping should consider density of available data and variability of country-specific environmental gradients. This study highlights the large degree of spatial uncertainty in SOC estimates across Latin America. We provide a framework for improving country-specific mapping efforts and reducing current discrepancy of global, regional, and country-specific SOC estimates.

show abstract

“…Likely, M1 gives more plausible projections as it is based on measured SOC sequestration rates (Mg C yr −1 ) per bioclimatic zone and land cover class (i.e., cropland and grassland), whereas M2 makes inferences with respect to annual C increases vis à vis ‘present’ SOC stocks at a given location. Such stocks, however, are computed from best available (and often only available) data for organic carbon, bulk density, and proportion of coarse fragments (>2 mm) that were sampled and analysed between 1950 and 2015 (Arrouays et al, ; Batjes, ; Minasny, Malone, et al, ; Sanderman et al, ). Further, an implicit assumption of M2 is that possible C gains will be greatest where present SOC stocks are largest, which is counter‐intuitive (see Levèvre, Fatma, Viridiana, & Wiese, ; UNEP, ).…”

Section: Resultsmentioning

confidence: 99%

Technologically achievable soil organic carbon sequestration in world croplands and grasslands

Batjes

2018

Land Degrad Dev

View full text Add to dashboard Cite

Reported potentials for sequestration of carbon in soils of agricultural lands are overly optimistic because they assume that all degraded cropland and grassland can be subjected to best management practices. Two approaches for estimating this potential are presented. Method 1 (M1) considers literature‐derived best estimates for annual soil organic carbon (SOC) gains (Mg C ha−1) by bioclimatic zone; Method 2 (M2) assumes an annual C increase of 3 to 5 promille with respect to present SOC mass (similar to the French ‘4 pour mille’ initiative). Four management scenarios are considered, capturing the varying level of plausibility of meeting the full technological potential. According to M1, achievable gains range from 0.05–0.12 Pg C yr−1 to 0.14–0.37 Pg C yr−1, with a technological potential of 0.32–0.86 Pg C yr−1. For M2, these are 0.07–0.12 Pg C yr−1, 0.21–0.35 Pg C yr−1, and 0.60–1.01 Pg C yr−1. Consideration of the technological potential only and use of a proportional annual increase in SOC (M2), rather than using best estimates for soil carbon gains by bioclimatic zone (M1), will provide too ‘bright a picture’ in the context of rehabilitating degraded lands and mitigating/adapting to climate change. Further, M2 assumes that possible C gains will be greatest where present SOC stocks are highest, which is counter‐intuitive. Although all measures aimed at increasing SOC content should be encouraged due to the creation of win‐win situations, it is important to create a realistic picture of the amount of SOC gains that are feasible based on bioclimatic and management implementation constraints.

show abstract

Soil legacy data rescue via GlobalSoilMap and other international and national initiatives

Cited by 127 publications

References 82 publications

Machine learning and soil sciences: A review aided by machine learning tools

Machine learning and soil sciences: A review aided by machine learning tools

No silver bullet for digital soil mapping: country-specific soil organic carbon estimates across Latin America

Technologically achievable soil organic carbon sequestration in world croplands and grasslands

Contact Info

Product

Resources

About