Horizon identification using excel

Size: px

Start display at page:

Download "Horizon identification using excel"

Lester Joshua Wheeler
5 years ago
Views:

1 Symposium no. 21 Paper no Presentation: oral FITZPATRICK E.A. Horizon identification using excel Department of Plant and Soil Science, University of Aberdeen, Cruickshank Building, St Machar Drive, Aberdeen AB24 3UU, UK Abstract Soil classifiers have taken their lead from biologists but have not produced an internationally accepted system even though the WRB is supported by the IUSS. More recently Mazaheri et al. (1995a, 1995b) have tried to introduce Fuzzy logic into soil classification. They achieved a small amount of success when their technique was applied to selected soils in Australia but failed when applied to other soils. Like most classifiers before them, they applied their technique to whole soils. Since soils are extremely variable and do not show homology of their constituent horizons (FitzPatrick 1988) more attention should be given to identifying horizons and less effort in trying to produce a hierarchical system for whole soils. Keywords: horizon identification, reference points, coordinate principles, soil formulae Horizon Designation The underlying concept is that soil properties are coordinates, (Crowther, 1953) which intersect in space to create reference points (horizons) which can be thought of as forming a universe of points with varying distances one from the other. The horizons are arranged in groups based on one or more conspicuous properties. The groups are given names ending with on. Thus all the pale coloured upper horizons form a group of albons with the symbol Ab. The various individual horizons within the albons are given the symbols Aba, Abb, etc. At present it is not possible to apply this mathematically to all situations thus many of the groups are created subjectively. The system attempts to place unknown horizons within the universe of points based upon a calculated score within Excel. Thus the basic theoretical units are reference points, which have fixed values for each property created on coordinate principles. The identification is based on the concept of the closest match or Euclidian distance between the unknown and fixed reference values. It should be stressed that this system uses single fixed values and not value ranges as used in other systems. This allows a certain degree of overlap or fuzziness. These situations present many philosophical difficulties and also problems with regard to nomenclature. Should horizons be named on the basis of their polygenetic development? At this stage in the development of Pedology, it seems that an almost completely pragmatic approach should be adopted, based on the dominance of the property or properties of the horizons. These properties may be continuing to form at present or they may be inherited partially or completely from a previous phase of soil formation

2 Types of Horizons There are four types of horizon. Uniform or homogeneous horizons. Intergrade horizons Compound horizons Composite horizons Uniform or homogeneous horizons These have a single set of dominant properties as a result of progressive development but they may have been through a number of distinct phases. A Podzol horizon may have passed through a Cambisol stage. Intergrade horizons These have properties that grade between two reference points. These are not given names but are designated by the use of the group symbols. The symbols are combined and placed in round brackets with the dominant set of features placed first. The intergrade (Cha-Vra) displays properties of both chernons and vertons but is more like a chernon. Compound horizons These have a combination of properties of two or more reference points. These properties are usually contrasting and develop as a result of either contrasting seasonal processes or they have one set of properties superimposed upon an earlier set of properties. The designation of the seasonally compound horizons is based on the dominant summer process. For example, a strongly reduced horizon that freezes during the winter would be regarded as a type of cerulon and designated Crd. The designation of a compound horizon resulting from climatic change or progressive soil evolution would show the symbols of both horizons. For examle, an argillon may become calcified as a result of climatic change. These are designated in a similar manner to intergrades but the symbols are placed in curly brackets thus {Ccc-Ara}. Composite horizons These have discrete volumes of two or more horizons. These are designated by placing the reference point symbols in round brackets and separated by a plus sign. One of the best examples is found in Chernozems, which have crotovinas and designated as follows: (Cha+2C-z-) where Cha is the symbol for a chernon and 2C-z- is the symbol for moderately calcareous loess. Whole Soil Designation, Nomenclature and Classification Any divisions or boundaries created within soils, are arbitrary. The divisions do not delimit discrete entities. THIS SITUATION DEFIES CLASSIFICATION. The best that can be achieved for soils is to give each soil a designation and to have one or two higher levels of grouping using subjective methods to suit individual users. The designation of soils can be achieved by using formulae that are produced by writing in order the horizon symbols as they occur in vertical sequence (Sokolovsky, 1930; Vilensky, 1927). The thickness of each horizon in centimetres is also shown by the number following the horizon symbols. Each symbol is separated by a swing dash as shown in the following example:

3 Lta1~Mla20~Cbb30~B-l- CAMBISOL (Lta = litter, Mla = mullon, Cbb= cambon, B-l- = basic loam In order to be able to show the various texture patterns in soil formulae without invoking any process it is suggested that > and < symbols should be placed between horizon symbols and used singly or in pairs as follows: < the horizon above has slightly less clay or coarser texture than the horizon below << the horizon above has much less clay or much coarser texture that the horizon below > the horizon above has slightly more clay or finer texture than the horizon below >> the horizon above has much more clay or much finer texture than the horizon below The use of these symbols is shown in the formula below, and illustrates the situation in which the podzon Pzd has a finer texture than the ison In. Lta2~Fma3~Hfa2~Sbb7~Aba5<Pzd40>In20+:A-s- PODZOL (Lta = litter, Fma = fermenton, Hfa = humifon, Aba = albon, Sbb = sablon, Pzd = podzon, Inb = ison, A-s- = acid sand) Each formula characterizes a group of soils but there will be some variations such as the thickness of the horizons. Each distinctive member of each group is given a name, usually a local place name that is also given to the series (mapping unit) in which this class is dominant. The groups are arranged in subclasses and classes, both of which, particularly the latter are established on a subjective basis to suit the requirements of the operator. The subclasses are established on the basis of two, three or four prominent horizons. They are not given names but designated by writing the symbols of the horizons and placing them in square brackets. Four of the subclasses of Podzols are: [Aba~Pza] [Aba~Pzc] [Aba~Pzd] [Aba~Pzb~Isb] The classes are given names and are characterized by the presences of one prominent horizon or by a unique combination or relationship between the horizons. When the class is based on the presence of a prominent horizon it often derives its name from the horizon. The Chernozems are created because of the presence of a prominent chernon. On the other hand the Phaeozems are created on the basis of the unique combination of a phaeon and an argillon. This system allows the creation of various classes and hierarchies to suit various users. A development of the above might be the concept of soil materials with different types of arrangements. There are those situations where a given material is laterally continuous and homogeneous and fits the classical horizon concept. Then there is the composite situation where there is a mixture of different materials as in Chernozems. The compound horizons are an intimate mixture of different properties and may also be laterally contiguous. Similarly the biologically produced horizons may be laterally contiguous if large surface structures such as termitaria are excluded. The approach taken here is to blend the two concepts are far as possible and to have horizons that have some type of lateral continuity even although there may be more than one type of material present and to define and name a horizon on it easily visible and

4 measurable properties. For example, horizon Dra-duron is easily recognised in the field as being hard and will not slake in dilute HCl, therefore the cementation is presumably silica, thus it is regarded as a duron. However thin sections show in many durons that there are numerous clay coatings and deposits of calcite secondary to the clay coatings, and finally the silica coatings. This is clearly a compound horizon but that can only be decided on the basis of detailed study. Thus it is simply regarded as a duron but in detail it is an argillon-calcon-duron {Ara-Ccb-Dra}. The system attempts to place unknown horizons within the universe of horizons based upon a calculated score. This is achieved by creating spreadsheets in Excel with the reference horizons arranged horizontally starting in column D row C. the properties arranged vertically starting in column A row 4. A reference value is given to each of the properties of each horizon. These are scaled values with most being from 0 to 9 but for toxicity the value is increased to 100 in order to get a clear separation of polluted from nonpolluted soils. The original values for unknowns are placed in the column B, then the system automatically calculates the scaled values for each property using the VLOOKUP function in the cells in column C. The score for each property of each horizon is determined using the formula. This formula subtracts the values in column C from those in the rest of the spreadsheet. These results are squared and put into an array. All the properties should be used but in some cases some may not have been determined then cell is left empty. An IF statement in the formula: =IF($C4=,0,(IF($D4=,0,($C4-D4))^2)) places a zero in the array for empty cells. Thus as many or as few properties can be used which makes the system as general or as definitive as required and introduces a high degree of flexibility. The formula also places zeros in the array if property codes are missing and adds to the flexibility. The values for each horizon are summed to give the scores for that horizon. The lowest value is the closest match to the unknown. The data for the unknown can be entered cell by cell or they can be copied from a data set. This is quicker and would be the method operated by an institution. In this method one or more spreadsheets are set up to contain the original field and laboratory data. A column is selected, copied and pasted into column B. In a separate table below; The Reference names, Calculated scores, WRB equivalent names and Horizon types, are arranged vertically and linked to the data in the rows above. If these columns are selected and ordered in ascending order according to the SCORE column then the lowest score appears at the top and is the closest match. In practise the macros Ctrl+u, Ctrl+m and Ctrl+l will order the columns. Because each soil has many properties, getting a perfect match is virtually impossible so that scores under about 300 are considered to be very good. For polluted soils the score will be very high. To compare the identified unknown with a known reference horizon of the same name click on the name and symbol to display the descriptions of the horizons and an illustration of a profile with a similar horizon. This also shows the formula for the soil. Soil Formulae Filter All of the formulae for a given project can be placed in a single spread sheet and then filtered for any single horizon or combination of horizons. Thus as many different hierarchies as require can be created. Generally it might be desirable to create

5 hierarchies based on conventional lines having classes such as Podzols and Chernozems, but it may be necessary to group together all soils that contain other horizons such as fragons (fragipans). Advantages of This System This system has a number of distinct advantages over other traditional systems. There are: (1)It is extremely precise because it uses all available properties. (2)In spite of being very complicated to produce it is nevertheless very simple to operate since little personal memory if needed. (3)Shows the position of the horizon in multidimensional space, namely its. (4)relationship to other horizons and closeness of the unknown to the reference horizon. (5)Variable number of properties can be used for an identification. (6)Multiple hierarchies can be produced to suit individual users. (7)More properties and/or reference points can easily be added without having to rework the whole system. (8)Easily adapted to suit local needs. (9)The system can easily be searched to find, for example, all horizons with granular structure and high ph as opposed to moderate or low ph. (10) Because it is produced on CD-ROM there is almost endless space for adding material, especially illustrations. A CD-ROM giving a description of all the horizon groups and the detailed methodology is available from the author. Acknowledgements I would like to thank the many colleagues that made helpful suggestions especially M. Aitkenhead and F. Terribile. References Crowther, E.M The sceptical soil scientist. J. Soil Sci. 4: FitzPatrick, E.A Soil Horizon Designation and Classification. Technical Paper 17. International Soil Reference and Information Centre (ISRIC), Wageningen, The Netherlands. p Mazaheri, S.A., A.J. Kopi, A.B. McBratney and B. Constable. 1995a. Australian Soil Identification Spreadsheet (ASIS): A Program for Allocating Soil Profiles to Australian Great Groups (GSG). Version 1.1. p. 20. Mazaheri, S.A., A.J. Kopi, A.B. McBratney and B. Constable. 1995b. A fuzzy allocation scheme for the Australian Great Soil Group Classification. European Journal of Soil Science 46: Sokolovsky, A.N The nomenclature of the genetic horizons of the soil, pp In Proc. of the 2 nd International Society of Soil Science. Vol. 5. Leningrad, Moscow. Vilensky, D.G Concerning the principles of genetic soils classification. Contribution to the study of the soils of Ukrania, pp In 1 st Iternational Congress of Soil Science. Vol

Traffic accidents and the road network in SAS/GIS

Traffic accidents and the road network in SAS/GIS Frank Poppe SWOV Institute for Road Safety Research, the Netherlands Introduction The first figure shows a screen snapshot of SAS/GIS with part of the