Contents lists available at ScienceDirect Geoderma journal homepage: www.elsevier.com/locate/geoderma Allocating soil profile descriptions to a novel comprehensive soil classification system Farzin Shahbazi a, ⁎ ,1 , Jingyi Huang b , Alex B. McBratney b , Philip Hughes b a Soil Science Department, Faculty of Agriculture, University of Tabriz, Iran b Sydney Institute of Agriculture & School of Life and Environmental Sciences, The University of Sydney, Eveleigh, NSW 2015, Australia ARTICLE INFO Handling editor: M. Vepraskas Keywords: Soil taxonomy Principal component analysis Diagnostic horizons Diagnostic properties Allocation Nearest-neighbour distance ABSTRACT Previous work has been put into the creation of a comprehensive soil classification system (CSCS) using a harmonised dataset of 23 soil properties at 18 depth intervals. The classification consists of selected soil taxa from the US Soil Taxonomy, World Reference Base for Soil Resources, the Australian Soil Classification, and the New Zealand Soil Classification. In this paper, the CSCS was tested for allocation using data for from 44 soil profiles collected in Iran. A distance-based algorithm was used to allocate and name the soil profiles according to the CSCS. It was found that 36 Iranian soil profiles are close to the existing taxa of the CSCS in the taxonomic space. Three Iranian profiles with distances between 25 and 30 taxonomic units to the closest CSCS taxa were added to the CSCS and assigned with new systematic names. Allocating the remaining 5 Iranian taxa would require regenerating the nomenclature system. The CSCS has shown advantages for allocating soil profiles from other regions of the world other than the USA, Australia and New Zealand. It also enables cross-referencing with existing soil classification systems. In the future, the CSCS can be further improved by adding taxa from other global or regional soil classification systems. 1. Introduction The last few decades have seen an escalation in the importance and scope of soil science. The cataloguing and taxonomical identification of soil has become an increasing priority for a large number of nations. This is in part due to the recognition that soil security is as pertinent a priority as issues such as water security (McBratney et al., 2014). Un- fortunately, communication between these nations has not been man- aged the same way the science has, creating a series of soil classification systems that are sometime niche, sometimes with similar terms that mean different things (for example, the Salic horizon of the USDA Soil Taxonomy and World Reference Base of Soil Resources), sometimes duplicated. Identifying truly unique soil types in a different system is therefore a challenge, as even though there have been great strides towards globalizing the study of soil, non-standard descriptions and methodologies (e.g. duplex soils in the Australian Soil Classification) still percolate through smaller and regional soil classification systems. This can be exacerbated by geopolitical issues which obfuscate com- munication and ergo understanding between government agencies of differing nations. 1.1. Novel classification development Methods have been recently developed to look at these systems from a numerical perspective. Based on theories developed from the 'fifties to the turn of the century (Hole and Hironaka, 1960; Bidwell and Hole, 1964) viewing soil profile descriptions (SPD's), primarily from the Australian Soil Classification or the US based Soil Taxonomy (ST), in terms of properties and depths (Hughes et al., 2017a). Spline functions have been used to transform these soil descriptions which are typically recorded by horizon with associated horizon depths into standardized depth increments (Bishop et al., 1999). Of all the properties collected and splined in this fashion, 23 specific soil properties have been se- lected because of pedological importance and availability. These properties stacked according to depth into a single vector 414 integers long (excluding the pedon or taxonomic id) have been converted via scaled principal component analysis into a specific taxonomic space. The formula established from this process can be used to project other similarly collected and processed data into the same environment for taxonomic comparison. With SPD's or taxonomic descriptions converted in this way, other numeric techniques can be used for comparison (Hughes et al., 2018a). Of these techniques, convex hull analysis and https://doi.org/10.1016/j.geoderma.2018.05.017 Received 27 February 2018; Received in revised form 3 May 2018; Accepted 13 May 2018 ⁎ Corresponding author at: Soil Science Department, Faculty of Agriculture, University of Tabriz, 5166616471, Iran. 1 Alternative address: Sydney Institute of Agriculture & School of Life and Environmental Sciences, The University of Sydney, Eveleigh, NSW 2015, Australia. E-mail addresses: shahbazi@tabrizu.ac.ir, farzin.shahbazi@sydney.edu.au (F. Shahbazi). Geoderma 329 (2018) 54–60 0016-7061/ © 2018 Elsevier B.V. All rights reserved. T