Digital soil mapping workflow for forest resource applications: a case study in the Hearst Forest, Ontario
Abstract
Accurate soil information is critically important for forest management planning and operations but is challenging to map. Digital soil mapping (DSM) improves upon the limitations of conventional soil mapping by explicitly linking a variety of environmental data layers to spatial soil point data sets to continuously predict soil variability across a landscape. Thus far, much DSM research has focussed on the development of ultrafine-resolution soil maps within agricultural systems; however, increasing availability of light detection and ranging (LiDAR) data presents new opportunities to apply DSM to support forest resource applications at multiple scales. This project describes a DSM workflow using LiDAR-derived elevation data and machine learning models (MLMs) to predict key forest soil attributes. A case study in the Hearst Forest in northeastern Ontario, Canada, is used to illustrate the workflow. We applied multiple MLMs to the Hearst Forest to predict soil moisture regime and textural class. Both qualitative and quantitative assessment pointed to the random forest MLM producing the best maps (63% accuracy for moisture regime and 66% accuracy for textural class). Where error occurred, soils were typically misclassified to neighbouring classes. This standardized, flexible workflow is a valuable tool for practitioners that want to undertake DSM as part of forest resource management and planning.
Résumé
Des informations précises sur les sols sont absolument essentielles pour la planification et les opérations d’aménagement forestier mais elles sont difficiles à cartographier. La cartographie numérique des sols (CNS) constitue un progrès par rapport aux limites de la cartographie conventionnelle des sols en reliant une variété de couches de données environnementales à des ensembles de données spatiales ponctuelles des sols pour prédire la variabilité à travers un paysage de façon continue. Jusqu’à maintenant, beaucoup de travaux de recherche sur la CNS ont mis l’accent sur le développement de cartes des sols à très haute résolution pour des systèmes agricoles. Cependant, la disponibilité croissante de données lidar offre de nouvelles opportunités d’appliquer la CNS en support à des applications qui concernent les ressources forestières à de multiples échelles. Ce projet décrit un flux de travail de CNS qui utilise des données altimétriques dérivées du lidar et des modèles d’apprentissage automatique (MAA) pour prédire des attributs importants des sols. Une étude de cas dans la forêt de Hearst, dans le nord-est de l’Ontario, au Canada, est utilisée pour illustrer le flux de travail. Nous avons appliqué plusieurs MAA à la forêt de Hearst pour prédire le régime d’humidité et la classe de texture. Une évaluation tant qualitative que quantitative indiquait que le MAA de forêt aléatoire produisait les meilleures cartes (précision de 63 % pour le régime d’humidité et de 66 % pour la classe de texture). Lorsqu’il y avait des erreurs, les sols mal classés étaient typiquement placés dans les classes voisines. Ce flux de travail standardisé et flexible est un outil précieux pour les praticiens qui veulent entreprendre la CNS en tant que composante de la planification et de la gestion des ressources forestières. [Traduit par la Rédaction]
Get full access to this article
View all available purchase options and get full access to this article.
References
Adhikari A. and Hartemink A.E. 2016. Linking soils to ecosystem services — a global review. Geoderma, 262: 101–111.
Akumu C.E., Johnson J.A., Etheridge D., Uhlig P., Woods M., Pitt D.G., and McMurray S. 2015. GIS-fuzzy logic based approach in modeling soil texture: sing parts of the Clay Belt and Hornpayne region in Ontario Canada as a case study. Geoderma, 239–240: 13–24.
Akumu C.E., Baldwin K., and Dennis S. 2019. GIS-based modeling of forest soil moisture regime classes: using Rinker Lake in northwestern Ontario, Canada as a case study. Geoderma, 351: 25–35.
Altman N.S. 1992. An introduction to kernel and nearest neighbor nonparametric regression. Am. Stat. 46: 175–185.
Banton, E. 2010. Photo-interpretation manual for ecosites in Ontario. Forest Resources Inventory Program, Ontario Ministry of Natural Resources.
Baveye P.C., Baveye J., and Gowdy J. 2016. Soil “ecosystem” services and natural capital: critical appraisal of research on uncertain ground. Front. Environ. Sci. 4: 41.
Behrens T., Zhu A.X., Schmidt K., and Scholten T. 2010. Multi-scale digital terrain analysis and feature selection for digital soil mapping. Geoderma, 155(3–4): 175–185.
Binkley, D., and Fisher, R. 2019. Ecology and management of forest soils. 5th ed. Wiley-Blackwell.
Blackburn, C.E., Bond, W.D., Breaks, F.W., Davies, D.W., Edwards, G.R., Poulsen, K.H., et al. 1985. Evolution of Archean volcanic-sedimentary sequences of the western Wabigoon subprovince and its margin: a review. In Evolution of Archean supracrustal sequences. Edited by L.D. Ayres, P.C. Thurson, K.D. Card, and W. Weber. Geological Association of Canada, Special Paper 28.
Blackford, C. 2020. DSM-workflow-for-forest-resource-applications (version v1.0.1). 10.5281/zenodo.3894915 [accessed 16 June 2020].
Breiman L. 2001. Random forests. Mach. Learn. 45(1): 5–32.
Brenning, A., Bangs, D., and Becker, M. 2018. RSAGA: SAGA geoprocessing and terrain analysis. R package version 1.3.0. Available from https://CRAN.R-project.org/package=RSAGA.
Brungard C.W., Boettinger J.L., Duniway M.C., Wills S.A., and Edwards T.C. 2015. Machine learning for predicting soil classes in three semi-arid landscapes. Geoderma, 239–240: 68–83.
Bui, E.N., Simon, D., Schoknecht, N., and Payne, A. 2007. Adequate prior sampling is everything: lessons from the Ord River Basin, Australia. In Digital soil mapping: an introductory perspective. Edited by P. Lagacherie, A.B. McBratney, and M. Voltz. Developments in Soil Science, Volume 36. Elsevier. pp. 193–204.
Bulmer, C., Pare, D., and Domke, G.M. 2019. A new era of digital soil mapping across forested landscapes. In Global change and forest soils. Edited by M. Busse, C.P. Giardina, D.M. Morries, and D.S. Page-Dumroese. Volume 36. Elsevier. pp. 345–371.
Canadian Society of Soil Science. 2020. Working Group – Soils of Canada. Available from https://soilsofcanada.ca/digital-soil-mapping/working-group.php [accessed 27 January 2020].
Cohen J. 1960. A coefficient of agreement for nominal scales. Educ. Psychol. Meas. 20: 37–46.
Conrad O., Bechtel B., Bock M., Dietrich H., Fischer E., Gerlitz L., et al. 2015. System for Automated Geoscientific Analyses (SAGA) v. 2.1.4. Geosci. Model Dev. 8(7): 1991–2007.
Cortes C. and Vapnik V. 1995. Support-vector networks. Mach. Learn. 20(3): 273–297.
Crins, W.J., Gray, P.A., Uhlig, P.W.C., and Wester, M.C. 2009. The ecosystems of Ontario, Part 1: ecozones and ecoregions. Ontario Ministry of Natural Resources.
Didan, K. 2015. MOD13Q1 MODIS/Terra Vegetation Indices 16-Day L3 Global 250m SIN Grid V006 [Data set. NASA EOSDIS Land Processes DAAC]. [accessed 17 December 2019].
Drever C.R. and Lertzman K.P. 2001. Light-growth responses of coastal Douglas-fir and western redcedar saplings under different regimes of soil moisture and nutrients. Can. J. For. Res. 31(12): 2124–2133.
Dyke, A.S. 2004. An outline of North American deglaciation with emphasis on central and northern Canada. In Developments in quaternary sciences. Vol. 2. pp. 373–424.
Gallant J.C. and Dowling T.I. 2003. A multiresolution index of valley bottom flatness for mapping depositional areas. Water Resour. Res. 39(12): 1347–1359.
Goetz J.N., Brenning A., Petschko H., and Leopold P. 2015. Evaluating machine learning and statistical prediction techniques for landslide susceptibility modeling. Comput. Geosci. 81: 1–11.
Hastie, T., Tibshirani, R., and Friedman, J. 2009. The elements of statistical learning: data mining, inference, and prediction. 2nd ed. Springer-Verlag, New York. 10.1007/978-0-387-84858-7.
Hearst Forest Management. 2019. Surficial geology, site types and climate. Hearst Forest Management Inc., Hearst, Ont. Available from http://www.hearstforest.com/english/surficial.html [accessed 13 January 2020].
Heung B., Bulmer C.E., and Schmidt M.G. 2014. Predictive soil parent material mapping at a regional-scale: a random forest approach. Geoderma, 214-215: 141–154.
Heung B., Ho H.C., Zhang J., Knudby A., Bulmer C.E., and Schmidt M.G. 2016. An overview and comparison of machine-learning techniques for classification purposes in digital soil mapping. Geoderma, 265: 62–77.
Heung B., Hodúl M., and Schmidt M.G. 2017. Comparing the use of legacy soil pits and soil survey polygons as training data for mapping soil classes. Geoderma, 290: 51–68.
Hijmans, R.J. 2019. raster: geographic data analysis and modeling. R package version 3.0–7. Available from https://CRAN.R-project.org/package=raster.
Ho T.K. 1998. The random subspace method for constructing decision forests. EEE Trans. Pattern Anal. Mach. Intell. 20: 832–844.
Hounkpatin K.O.P., Schmidt K., Stumpf F., Forkuor G., Behrens T., Scholten T., et al. 2017. Predicting reference soil groups using legacy data: a data pruning and Random Forest approach for tropical environment (Dano catchment, Burkina Faso). Sci. Rep. 8: 9959.
Jenny, H. 1941. Factors of soil formation. Dover Publications.
Johnson, J.A., Uhlig, P., and Wester, M. 2015. Field guide to the substrates of Ontario. Ontario Ministry of Natural Resources, Sault Ste. Marie, Ont.
Khaledian Y. and Miller B.A. 2020. Selecting appropriate machine learning methods for digital soil mapping. Appl. Math. Modell. 81: 401–418.
Kühn J., Brenning A., Wehrhan M., Koszinski S., and Sommer M. 2009. Interpretation of electrical conductivity patterns by soil properties and geological maps for precision agriculture. Precis. Agric. 10(6): 490–507.
Kuhn, M., Wing, J., Weston, S., Williams, A., Keefer, C., Engelhardt, A., et al. 2019. caret: classification and regression training. R package version 6.0-84. Available from https://CRAN.R-project.org/package=caret.
Leniham J.M. 1993. Ecological response surfaces for North American boreal tree species and their use in forest classification. J. Veg. Sci. 4: 667–680.
Li S., MacMillan R.A., Lobb D.A., McConkey B.G., Moulin A., and Fraser W.R. 2011. LiDAR DEM error analyses and topographic depression identification in a hummocky landscape in the prairie region of Canada. Geomorphology, 129(3–4): 263–275.
Lin L.I. 1989. A concordance correlation coefficient to evaluate reproducibility. Biometrics, 45(1): 255–268.
Mackasey, W.O., Blackburn, C.E., and Trowell, N.F. 1974. A regional approach to the Wabigoon–Quetico belts and its bearing on exploration in Northwestern Ontario. Ontario Division of Mines.
Mansuy N., Thiffault E., Paré D., Bernier P., Guindon L., Philippe V., et al. 2014. Digital mapping of soil properties in Canadian managed forests at 250m of resolution using the k-nearest neighbor method. Geoderma, 235-236: 59–73.
Mansuy N., Valeria O., Laamrani A., Fenton N., Guindon L., Bergeron Y., et al. 2018. Digital mapping of paludification in soils under black spruce forests of eastern Canada. Geoderma Regional, 15: e00194.
McBratney A.B., Mendonça Santos M.L., and Minasny B. 2003. On digital soil mapping. Geoderma. 117(1–2): 3–52.
McKenzie N.J. and Ryan P.J. 1999. Spatial prediction of soil properties using environmental correlation. Geoderma, 89(1–2): 67–94.
Meinshausen N. 2006. Quantile regression forests. J. Mach. Learn. Res. 7: 983–999.
Minasny B. and McBratney A.B. 2006. A conditioned Latin hypercube method for sampling in the presence of ancillary information. Comput. Geosci. 32(9): 1378–1388.
Minasny B. and McBratney A.B. 2007. Incorporating taxonomic distance into spatial prediction and digital mapping of soil classes. Geoderma, 142(3–4): 285–293.
Minasny B. and McBratney A.B. 2016. Digital soil mapping: a brief history and some lessons. Geoderma, 264: 301–311.
Minasny B., McBratney A.B., Mendonça-Santos M.L., Odeh I.O.A., and Guyon B. 2006. Prediction and digital mapping of soil carbon storage in the Lower Namoi Valley. Soil Res. 44(3): 233–244.
Minasny, B., McBratney, A.B., Malone, B.P., and Wheeler, I. 2013. Digital mapping of soil carbon. In Advances in agronomy. Volume 118. Edited by D. Spark. Elsevier. pp. 1–47. 10.1016/B978-0-12-405942-9.00001-3.
Nigh G.D. 2006. Impact of climate, moisture regime, and nutrient regime on the productivity of Douglas-fir in coastal British Columbia, Canada. Clim. Change, 76(3–4): 321–337.
Nijland W., Coops N.C., Macdonald S.E., Nielsen S.E., Bater C.W., White B., et al. 2015. Remote sensing proxies of productivity and moisture predict forest stand type and recovery rate following experimental harvest. For. Ecol. Manage. 357: 239–247.
Odgers, N.P., McBratney, A.B., Minasny, B., Sun, W., and Clifford, D. 2014. DSMART: an algorithm to spatially disaggregate soil units. In GlobalSoilMap: basis of the global spatial soil information system. Edited by D. Arrouays, D. McKenzie, J. Hempel, A.R. de Forges, and A.B. McBratney. Taylor & Francis. pp. 261–266.
R Core Team. 2018. R: a language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. Available from https://www.R-project.org/.
Reuter H.I., Lado L.R., Hengl T., and Montranarella L. 2008. Continental-scale digital soil mapping using European soil profile data: soil pH. Hamburger Beiträge zur Physischen Geographie und Landschaftsökologie, 19: 91–102.
Riley S.J., DeGloria S.D., and Elliot R. 1999. A terrain ruggedness index that quantifies topographic heterogeneity. Int. J. Sci. 5: 23–27.
Rossiter D.G., Zeng R., and Zhang G.L. 2017. Accounting for taxonomic distance in accuracy assessment of soil class predictions. Geoderma, 292: 118–127.
Scull P., Franklin J., Chadwick O.A., and McArthur D. 2003. Predictive soil mapping: a review. Prog. Phys. Geogr. 27(2): 171–197.
Sharififar A., Sarmadian F., Malone B.P., and Minasny B. 2019. Addressing the issue of digital mapping of soil classes with imbalanced class observations. Geoderma, 350: 84–92.
Söderström M., Sohlenius G., Rodhe L., and Piikki K. 2016. Adaptation of regional digital soil mapping for precision agriculture. Precis. Agric. 17(5): 588–607.
Stumpf F., Schmidt K., Goebes P., Behrens T., Schönbrodt-Stitt S., Wadoux A., et al. 2017. Uncertainty-guided sampling to improve digital soil maps. Catena, 153: 30–38.
Taghizadeh-Mehrjardi R., Nabiollahi K., Minasny B., and Triantafilis J. 2015. Comparing data mining classifiers to predict spatial distribution of USDA-family soil groups in Baneh region. Iran. Geoderma, 253–254: 67–77.
Taghizadeh-Mehrjardi R., Schmidt K., Eftekhari K., Behrens T., Jamshidi M., Davatgar N., et al. 2020. Synthetic resampling strategies and machine learning for digital soil mapping in Iran. Eur. J. Soil Sci. 71: 352–368.
Terribile F., Coppola A., Langella G., Martina M., and Basile A. 2011. Potential and limitations of using soil mapping information to understand landscape hydrology. Hydrol. Earth Syst. Sci. 15(12): 3895–3933.
Thurston, P.C., Osmani, I.A., and Stone, D. 1991. Northwestern Superior Province: review and terrane analysis. In Geology of Ontario, Ontario Geological Survey, Special Volume 4, Part 1. pp. 81–144.
Vaysse K. and Lagacherie P. 2017. Using quantile regression forest to estimate uncertainty of digital soil mapping products. Geoderma, 291: 55–64.
Webster R. and Burrough P.A. 1972a. Computer-based soil mapping of small areas from sample data — I: multivariate classification and ordination. J. Soil Sci. 23(2): 210–221.
Webster R. and Burrough P.A. 1972b. Computer-based soil mapping of small areas from sample data — II: classification smoothing. J. Soil Sci. 23(2): 222–234.
Webster K.L., Creed I.F., Beall F.D., and Bourbonnière R.A. 2008. Sensitivity of catchment‐aggregated estimates of soil carbon dioxide efflux to topography under different climatic conditions. J. Geophys. Res. 113(G3).
Witten, I.H., Frank, E., and Hall, M.A. 2005. Data mining: practical machine learning tools and techniques. 2nd ed. Elsevier.
Yamazaki D., Ikeshima D., Tawatari R., Yamaguchi T., O'Loughlin F., Neal J.C., et al. 2017. A high-accuracy map of global terrain elevations. Geophys. Res. Lett. 44: 5844–5853.
Yamazaki D., Ikeshima D., Sosa J., Bates P.D., Allen G.H., and Pavelsky T.M. 2019. MERIT Hydro: a high-resolution global hydrography map based on latest topography dataset. Water Resour. Res. 55(6): 5053–5073.
Yang L., Jiao Y., Fahmy S., Zhu A.X., Hann S., Burt J.E., and Feng Q. 2011. Updating conventional soil maps through digital soil mapping. Soil Sci. Soc. Am. J. 75(3): 1044–1053.
Zhu A.X. 1997. Measuring uncertainty in class assignment for natural resource maps using fuzzy logic. Photogramm. Eng. Remote Sens. 63: 1195–1202.
Supplementary Material
Supplementary data (cjfr-2020-0066suppla.docx)
- Download
- 495.39 KB
Information & Authors
Information
Published In
Canadian Journal of Forest Research
Volume 51 • Number 1 • January 2021
Pages: 59 - 77
History
Received: 13 February 2020
Accepted: 30 June 2020
Accepted manuscript online: 22 July 2020
Version of record online: 22 July 2020
Copyright
© 2021.
Key Words
Mots-clés
Authors
Metrics & Citations
Metrics
Other Metrics
Citations
Cite As
Christopher Blackford, Brandon Heung, Ken Baldwin, Robert L. Fleming, Paul W. Hazlett, Dave M. Morris, Peter W.C. Uhlig, and Kara L. Webster. 2021. Digital soil mapping workflow for forest resource applications: a case study in the Hearst Forest, Ontario. Canadian Journal of Forest Research.
51(1): 59-77. https://doi.org/10.1139/cjfr-2020-0066
Export Citations
If you have the appropriate software installed, you can download article citation data to the citation manager of your choice. Simply select your manager software from the list below and click Download.
Cited by
1. A quantitative approach to defining soil nutrient regimes within ecosystem classifications for Northwestern Ontario
2. A Critical Review for Real-Time Continuous Soil Monitoring: Advantages, Challenges, and Perspectives
3. Provincial-scale digital soil mapping using a random forest approach for British Columbia
4. Mapping organic layer thickness and fuel load of the boreal forest in Alberta, Canada
5. Incorporating spatial uncertainty maps into soil sampling improves digital soil mapping classification accuracy in Ontario, Canada
View Options
Get Access
Login options
Check if you access through your login credentials or your institution to get full access on this article.