A multivariate method for matching soil classification systems, with an Australian example
H. F. Teng A , R. A. Viscarra Rossel
B and R. Webster C D
A School of Environmental Ecology and Biological Engineering, Wuhan Institute of Technology, Wuhan 430205, China.
B School of Molecular and Life Sciences, Faculty of Science and Engineering, Curtin University, GPO Box U1987, Perth WA 6845, Australia.
C Rothamsted Research, Harpenden, Hertfordshire AL5 2JQ, UK.
D Corresponding author. Email: richard.webster@rothamsted.ac.uk
Soil Research 58(6) 519-527 https://doi.org/10.1071/SR19320
Submitted: 14 November 2019 Accepted: 4 May 2020 Published: 25 June 2020
Abstract
Differences between local systems of soil classification hinder the communication between pedologists from different countries. The FAO–UNESCO Soil Map of the World, as a fruit of world-wide collaboration between innumerable soil scientists, is recognised internationally. Ideally, pedologists should be able to match whole classes in their local systems to those in an international soil classification system. The Australian Soil Classification (ASC) system, created specifically for Australian soil, is widely used in Australia, and Australian pedologists wish to translate the orders they recognise into the FAO soil units when writing for readers elsewhere. We explored the feasibility of matching soil orders in the ASC to units in the FAO legend using a multivariate analysis. Twenty soil properties, variates, of 4927 profiles were estimated from their visible–near infrared reflectance (vis–NIR) spectra. We arranged the profiles in a Euclidean 20-dimensional orthogonal vector space defined by standardised variates. Class centroids were computed in that space, and the Euclidean distances between the centroids of the ASC orders and units in the FAO scheme were also computed. The shortest distance between a centroid of any ASC order and one of units in the FAO classification was treated as a best match. With only one exception the best matches were those that an experienced pedologist might expect. Second and third nearest neighbours in the vector space provided additional insight. We conclude that vis–NIR spectra represent sufficiently well the essential characters of the soil and so spectra could form the basis for the development of a universal soil classification system. In our case, we could assign with confidence the orders of the ASC to the units of the FAO scheme. A similar approach could be applied to link other national classification systems to one or other international systems of soil classification.
Additional keywords: Australia, class matching, principal components analysis, world soil map.
References
Bidwell OW, Hole FD (1964) An experiment in the numerical classification of some Kansas soils. Soil Science Society of America Journal 28, 263–268.| An experiment in the numerical classification of some Kansas soils.Crossref | GoogleScholarGoogle Scholar |
de Caritat P, Lech ME, McPherson AA (2008) Geochemical mapping “down under”: selected results from pilot projects and strategy outline for the National Geochemical Survey of Australia. Geochemistry Exploration Environment Analysis 8, 301–312.
| Geochemical mapping “down under”: selected results from pilot projects and strategy outline for the National Geochemical Survey of Australia.Crossref | GoogleScholarGoogle Scholar |
Esfandiarpour I, Salehi MH, Karimi A, Kamalai A (2013) Correlation between Soil Taxonomy and World Reference Base for Soil Resources in classifying calcareous soils: (a case study of arid and semi-arid Iran). Geoderma 197–198, 126–136.
| Correlation between Soil Taxonomy and World Reference Base for Soil Resources in classifying calcareous soils: (a case study of arid and semi-arid Iran).Crossref | GoogleScholarGoogle Scholar |
FAO (1995, 2003) ‘The digitized soil map of the world and derived soil properties.’ (version 3.5) FAO Land and Water Digital Media Series 1. (FAO: Rome)
FAO/IIASA/ISRIC/ISSCAS/JRC (2012) ‘Harmonized world soil database’ (version 1.2) (FAO: Rome)
FAO/ISRIC/ISSS (1998) ‘World reference base for soil resources.’ World Soil Resources Reports No 84. (FAO: Rome)
FAO/UNESCO (1971–1981) ‘Soil map of the world.’ Legend and 9 volumes. (UNESCO: Paris)
FAO/UNESCO (1974) ‘Legend of the soil map of the world’. (FAO: Rome)
FAO/UNESCO/ISRIC (1990) ‘Revised legend of the soil map of the world’. World Soil Resources Report, (FAO: Rome)
Isbell R (2002) ‘The Australian soil classification.’ revised edn. (CSIRO Publishing: Collingwood, Australia)
IUSS Working Group WRB (2006) ‘World reference base for soil resources 2006.’ World Soil Resources Reports No 103. (FAO: Rome)
Krasilnikov P, Ibáñez Martín J-J, Arnold RW, Shoba S (Eds) (2009) ‘A handbook of soil terminology, correlation and classification.’ (Earthscan: London)
Láng V, Fuchs M, Waltner I, Michéli E (2013) Soil taxonomic distance, a tool for correlation: as exemplified by the Hungarian Brown Forest Soils and related WRB Reference Soil Groups. Geoderma 192, 269–276.
| Soil taxonomic distance, a tool for correlation: as exemplified by the Hungarian Brown Forest Soils and related WRB Reference Soil Groups.Crossref | GoogleScholarGoogle Scholar |
McDonald R, Isbell RF, Speight JG, Walker J, Hopkins MS (1990) ‘Australian soil and land survey handbook: field handbook.’ 2nd edn. (Inkata Press: Melbourne)
Morand DT (2013) The World Reference Base for Soils (WRB) and Soil Taxonomy: an appraisal of their application to the soils of the Northern Rivers of New South Wales. Soil Research 51, 167–181.
| The World Reference Base for Soils (WRB) and Soil Taxonomy: an appraisal of their application to the soils of the Northern Rivers of New South Wales.Crossref | GoogleScholarGoogle Scholar |
Rayner JH (1966) Classification of soils by numerical methods. Journal of Soil Science 17, 79–92.
| Classification of soils by numerical methods.Crossref | GoogleScholarGoogle Scholar |
Rossiter DG, Zeng R, Zhang GL (2017) Accounting for taxonomic distance in accuracy assessment of soil class predictions. Geoderma 292, 118–127.
| Accounting for taxonomic distance in accuracy assessment of soil class predictions.Crossref | GoogleScholarGoogle Scholar |
Sneath PHA, Sokal RR (1973) ‘Numerical taxonomy: the principles and practice of numerical classification.’ (W.H. Freeman: San Francisco)
van Huyssteen CW, Michéli E, Fuchs M, Waltner I (2014) Taxonomic distance between South African diagnostic horizons and the World Reference Base diagnostics. Catena 113, 276–280.
| Taxonomic distance between South African diagnostic horizons and the World Reference Base diagnostics.Crossref | GoogleScholarGoogle Scholar |
Viscarra Rossel RA, Chen C (2011) Digitally mapping the information content of visible–near infrared spectra of surficial Australian soils. Remote Sensing of Environment 115, 1443–1455.
| Digitally mapping the information content of visible–near infrared spectra of surficial Australian soils.Crossref | GoogleScholarGoogle Scholar |
Viscarra Rossel RA, Webster R (2011) Discrimination of Australian soil horizons and classes from their visible–near infrared spectra. European Journal of Soil Science 62, 637–647.
| Discrimination of Australian soil horizons and classes from their visible–near infrared spectra.Crossref | GoogleScholarGoogle Scholar |
Viscarra Rossel RA, Webster R (2012) Predicting soil properties from the Australian soil visible–near infrared spectroscopic database. European Journal of Soil Science 63, 848–860.
| Predicting soil properties from the Australian soil visible–near infrared spectroscopic database.Crossref | GoogleScholarGoogle Scholar |
Zádorová T, Penížek V (2011) Problems in correlation of Czech national soil classification and World Reference Base 2006. Geoderma 167–168, 54–60.
| Problems in correlation of Czech national soil classification and World Reference Base 2006.Crossref | GoogleScholarGoogle Scholar |
