Engineering properties of peat: a data-driven approach

Marco D'Ignazio; Rasmus Sillanpää; Tim Länsivaara; Marco D'Ignazio; Rasmus Sillanpää; Tim Länsivaara

doi:10.3934/geosci.2026009

AIMS Geosciences

2026, Volume 12, Issue 1: 229-251. doi: 10.3934/geosci.2026009

Previous Article Next Article

Research article Special Issues

Engineering properties of peat: a data-driven approach

1.
Terra Research Centre, Faculty of Built Environment, Tampere University, P.O. Box 600, FI-33014, Tampere, Finland
2.
Destia Oy, Lahti, Finland

Received: 29 September 2025 Revised: 10 December 2025 Accepted: 28 January 2026 Published: 27 February 2026

Peat is a highly variable organic soil that presents major challenges for geotechnical engineering. In Finland, where peatlands cover one-third of the land area, infrastructure often intersects deep deposits that are difficult to characterise and costly to improve. Conventional practice relies on conservative assumptions or large-scale replacement, which can be expensive and carbon intensive. In this study, we compiled a harmonised database of over 250 datapoints from Finnish, Nordic, and Western European sources, focusing on compressibility, yield stress, and undrained shear strength. Regression analyses were benchmarked against Random Forest machine learning models. The results confirmed that the compression index is well predicted from water content, whereas yield stress and undrained shear strength display high variability under regression. Random Forest models provided modest improvements over conventional regression for strength-related parameters, while compressibility remained well captured by empirical correlations. Cross-parameter estimators offer additional tools where direct strength testing is unavailable. The findings underline the complementary roles of regression and machine learning in peat characterisation. Moreover, a hybrid workflow is proposed: regressions for early screening and conservative design, and machine learning for refined, site-specific assessment, supporting more sustainable infrastructure development on peatlands.
- peat,
- database,
- compressibility,
- shear strength,
- machine learning,
- correlations
Citation: Marco D'Ignazio, Rasmus Sillanpää, Tim Länsivaara. Engineering properties of peat: a data-driven approach[J]. AIMS Geosciences, 2026, 12(1): 229-251. doi: 10.3934/geosci.2026009

Related Papers:

Abstract

Peat is a highly variable organic soil that presents major challenges for geotechnical engineering. In Finland, where peatlands cover one-third of the land area, infrastructure often intersects deep deposits that are difficult to characterise and costly to improve. Conventional practice relies on conservative assumptions or large-scale replacement, which can be expensive and carbon intensive. In this study, we compiled a harmonised database of over 250 datapoints from Finnish, Nordic, and Western European sources, focusing on compressibility, yield stress, and undrained shear strength. Regression analyses were benchmarked against Random Forest machine learning models. The results confirmed that the compression index is well predicted from water content, whereas yield stress and undrained shear strength display high variability under regression. Random Forest models provided modest improvements over conventional regression for strength-related parameters, while compressibility remained well captured by empirical correlations. Cross-parameter estimators offer additional tools where direct strength testing is unavailable. The findings underline the complementary roles of regression and machine learning in peat characterisation. Moreover, a hybrid workflow is proposed: regressions for early screening and conservative design, and machine learning for refined, site-specific assessment, supporting more sustainable infrastructure development on peatlands.

References

[1]	UNEP, Global Peatlands Assessment: The State of the World's Peatlands—Evidence for Action toward the Conservation, Restoration, and Sustainable Management of Peatlands, United Nations Environment Programme. 2022.
[2]	Forsman J, Korkiala-Tanttu L, Piispanen P (2018) Mass stabilization as a ground improvement method for soft peaty. Peat InTech, 107-139. http://dx.doi.org/10.5772/intechopen.74144
[3]	Mesri G, Ajlouni M (2007) Engineering Properties of Fibrous Peats. J Geotech Geoenviron Eng 133: 850-866.https://doi.org/10.1061/(ASCE)1090-0241(2007)133: 7(850)
[4]	Kiikkerä S (2020) Turpeen pohjanvahvistuksen ympäristövaikutusten vertailu. BSc thesis, Aalto University, Helsinki, Finland. In Finnish.
[5]	O'Kelly B (2017) Measurement, interpretation and recommended use of laboratory strength properties of fibrous peat. Geotech Res 4: 136-171.https://doi.org/10.1680/jgere.17.00006
[6]	ASTM, D 4427-13: Standard classiﬁcation of peat samples by laboratory testing, West Conshohocken, PA, USA. 2013.
[7]	Zwanenburg C, Konstantinou M, Grünewald E (2025) Geotechnical characterization and strain-rate dependent response of peat from the Netherlands. AIMS Geosci 4: 923-945.https://doi.org/10.3934/geosci.2025040
[8]	Lechowicz Z (1994) An evaluation of the increase in shear strength of organic soils, Advances in Understanding and Modelling The Mechanical Behaviour of Peat: Proceedings of the International Workshop, 16-18 June 1993, Delft, Netherlands.
[9]	Von Post L (1922) Swedish geological peat survey with the results obtained so far. Sven Mosskult Tidskr 36: 1-27.
[10]	SFS EN-ISO 14688-1, Geotechnical Investigation and Testing. Identification and Classification of Soil. Part 1: Identification and Description. 2017. Available from: https://www.iso.org/standard/66345.html.
[11]	Ronkainen N (2012) Suomen maalajien ominaisuuksia. Suomen ympäristö 2: 2012. In Finnish.
[12]	Boylan N, Long M (2014) Evaluation of peat strength for stability assessments. Proc Inst Civil Eng Geotech Eng 167: 421-430.https://doi.org/10.1680/geng.12.00043
[13]	Landva AO, La Rochelle P (1983) Compressibility and shear characteristics of Radforth peats, Testing of peats and organic soils, 820: 157-191.https://doi.org/10.1520/STP37341S
[14]	Landva AO (1980) Vane testing in peat. Can Geotech J 17: 1-19.
[15]	Landva A (1986) In-situ testing of peat, Use of In Situ Tests in Geotechnical Engineering, ASCE, 191-205.
[16]	Hobbs NB (1986) Mire morphology and the properties and behaviour of some British and foreign peats. Q J Eng Geol 19: 7-80.
[17]	Long M, Boylan N (2013) Predictions of settlement in peat soils. Q J Eng Geol Hydrogeol 46: 303-322.
[18]	Liikennevirasto (2018) Penkereiden stabiliteetin laskentaohje. Liikennevirasto publication, Helsinki. In Finnish.
[19]	Zwanenburg C, Jardine RJ (2015) Laboratory, in situ and full-scale load tests to assess flood embankment stability on peat. Géotechnique 65: 309-326. https://doi.org/10.1680/geot.14.P.257
[20]	Sillanpää R (2023) Turpeen painumaominaisuuksien ja vedenläpäisevyyden arviointi vesipitoisuuden ja maatuneisuuden avulla. MSc thesis, Tampere University, Finland. In Finnish.
[21]	Sainio S (2022) Turpeen suljettu leikkauslujuus stabiliteettilaskelmissa. MSc thesis, Tampere University, Finland. In Finnish.
[22]	Li Y, Rahardjo H, Satyanaga A, et al. (2022) Soil database development with the application of machine learning methods in soil properties prediction. Eng Geol 306: 106769.https://doi.org/10.1016/j.enggeo.2022.106769 doi: 10.1016/j.enggeo.2022.106769
[23]	Zhang P, Yin ZY, Jin YF (2022) Machine learning-based modelling of soil properties for geotechnical design: review, tool development and comparison. Arch Computat Methods Eng 29: 1229-1245.https://doi.org/10.1007/s11831-021-09615-5
[24]	Baghbani A, Choudhury T, Costa S, et al. (2022) Application of artificial intelligence in geotechnical engineering: A state-of-the-art review. Earth Sci Rev 228: 103991.https://doi.org/10.1016/j.earscirev.2022.103991 doi: 10.1016/j.earscirev.2022.103991
[25]	Liu Y, Ni J, Zhang F, et al. (2026) Bayesian-informed DNN and ensemble learning for predicting soil water characteristic curves from easily measurable parameters. Transp Geotech 56: 101791.https://doi.org/10.1016/j.trgeo.2025.101791 doi: 10.1016/j.trgeo.2025.101791
[26]	Lundberg SM, Lee SI (2017) A unified approach to interpreting model predictions. Adv Neural Inf Process Syst 30.
[27]	Muschalik M, Baniecki H, Fumagalli F, et al. (2024) shapiq: Shapley interactions for machine learning. Adv Neural Inf Process Syst 37: 130324-130357.
[28]	D'Ignazio M (2016) Undrained shear strength of Finnish clays for stability analyses of embankments. PhD thesis, Tampere University of Technology, Finland.
[29]	D'Ignazio M, Phoon KK, Tan SA, et al. (2016) Correlations for undrained shear strength of finnish soft clays. Can Geotech J 53: 1628-1645.https://doi.org/10.1139/cgj-2016-0037 doi: 10.1139/cgj-2016-0037
[30]	D'Ignazio M, Länsivaara T (2024) Undrained shear strength of Finnish soft clays: A database perspective, Databases for Data-Centric Geotechnics, CRC Press, 135-151.
[31]	Helenelund KV (1951) Om konsolidering och sättning av belastade marklager. Soil and Water Technical Researches 6. In Swedish.
[32]	Ching J, Phoon KK (2014) Transformations and correlations among some clay parameters—The global database. Can Geotech J 51: 663-685.https://doi.org/10.1139/cgj-2013-0262
[33]	D'Ignazio M, Lunne T, Andersen Knut H, et al. (2019) Estimation of preconsolidation stress of clays from piezocone by means of high-quality calibration data. AIMS Geosci 5: 104-116.https://doi.org/10.3934/geosci.2019.2.104 doi: 10.3934/geosci.2019.2.104
[34]	Bao X, Li J, Shen J, et al. (2025) Comprehensive multivariate joint distribution model for marine soft soil based on the vine copula. Comput Geotech 177: 106814.https://doi.org/10.1016/j.compgeo.2024.106814 doi: 10.1016/j.compgeo.2024.106814

Reader Comments

Your name:*

Email:*
© 2026 the Author(s), licensee AIMS Press. This is an open access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0)