The function calculate in which classe(s) of a texture triangle (classification system defined by 'class.sys') lies each soil sample (with texture data) in the table 'tri.data'. As a sample may lie inside a texture class, but also at the edge of 2 or more texture classes, the function does not only output one single texture class per sample. If 'PiC.type' is 'n' or 'l', it rather output a table where each column is a texture class and each row a texture sample, and yes / no information about the belonging of the sample to each texture class. Alternatively, If 'PiC.type' is 't'it will output a text string (per sample) containing all the texture classes to which that point belong. The texture data in 'tri.data' can be transformed into another particle size system prior to their classification if needed. See the options base.css.ps.lim, tri.css.ps.lim, dat.css.ps.lim, css.transf and text.transf.fun. ON DEFAULT VALUES OF TT.points.in.classes() ARGUMENTS? As TT.points.in.classes() shares its arguments with many other functions, their default value is not defined in TT.points.in.classes() source code, but rather in a dedicated list object called 'TT.par' and stored in the environment TT.env. The function TT.get() is used to retrieve the default value of the arguments defined in TT.par (see ?TT.get). For instance, to know the default value of 'class.sys', you can type TT.get("class.sys"). To set a different default value for a given argument in R, use TT.set() (see ?TT.set). For instance to change the default value of 'class.sys', type TT.set( "class.sys" = "USDA.TT" ).
TT.points.in.classes(
tri.data,
class.sys = NULL,
PiC.type = NULL,
css.names = NULL,
text.sum = NULL,
base.css.ps.lim = NULL,
tri.css.ps.lim = NULL,
dat.css.ps.lim = NULL,
css.transf = NULL,
text.transf.fun = NULL,
trsf.add.opt1 = NULL,
trsf.add.opt2 = NULL,
text.tol = NULL,
tri.sum.tst = NULL,
tri.pos.tst = NULL,
collapse = NULL,
texture2xy = FALSE,
blr.tx = NULL,
blr.clock = NULL
)Data frame. Data frame containing the CLAY, SILT and SAND 'coordinates' of the texture data points to be classified The data frame can contain more column than needed (ignored). The data frame must have column named CLAY, SILT and SAND (uppercase, the order has no importance) or named after the 'css.names' argument (alternative names). The sum of CLAY, SILT and SAND must be equal to 'text.sum' ('text.tol' determines the error tolerance).
Single text string. Text code of the texture classification system to be used for the classification of 'tri.data'. Possible values are "none" (no classification plotted), "USDA.TT" (USDA texture triangle), "HYPRES.TT" (texture triangle of the European Soil Map), "FR.AISNE.TT" (French texture triangle of the Aisne region soil survey), "FR.GEPPA.TT" (French GEPPA texture triangle), "DE.BK94.TT" (German texture triangle), "UK.SSEW.TT" (Soil Survey of England and Wales), "AU.TT" (Australian texture triangle), "BE.TT" (Belgium texture triangle), "CA.EN.TT" (Canadian texture triangle, with English class abbreviations) and "CA.FR.TT" (Canadian texture triangle, with French class abbreviations) (see the package vignette for a complete list).
Single character string. If equal to 'n', then a table of 0, 1, 2 or 3 is outputed (0 if the sample does not belong to a class, 1 if it does, 2 if it lies on an edge and 3 if it lies on a vertex). Notice that the accuracy of the classification is not garanteed for samples lying very close to an edge, or right on it. See <http://www.mail-archive.com/r-help@r-project.org/msg96180.html>
Vector of 3 character strings. Name of the columns in 'tri.data' that contains the CLAY SILT and SAND values, respectively. If NULL, default c("CLAY","SILT","SAND") value is assumed. Not to be confused with 'css.lab' that defines the labels of the CLAY SILT and SAND axes in the plot.
Single numerical. Sum of the 3 particle size classes for each texture value (fixed). The real sum of the 3 particle size classes in 'tri.data' should be >= text.sum * (1-text.tol) OR <= text.sum * (1+text.tol), where 'text.tol' is an argument that can be changed. If some of the texture values don't match this requirement, an error occur (function fails) and TT.points.in.classes returns a of bad values with their actual particle size classes sum. You can 'normalise' you data table () prior to the use of TT.points.in.classes, by using the function TT.normalise.sum(), so all values match the 'text.sum' criteria. See also 'tri.sum.tst' that can be set to FALSE to avoid sum of particle size classes tests.
Vector of 4 numericals. Particle size boundaries (upper and lower) of the 3 particle size classes (CLAY, SILT and SAND, starting from the lower size of CLAY particles, 0, to the upper size of the SAND particles, 2000), in micrometers, FOR THE BASE SYSTEM. These particles size class limits are the references and all other texture values with different limits will be converted into that reference if (and only if) css.transf == TRUE (not default). If NULL, 'base.css.ps.lim' will be set to the default value of the texture classification system chosen ('class.sys'). The transformation function is set by 'text.transf.fun' and is a log-linear interpolation by default.
Vector of 4 numericals. Particle size boundaries (upper and lower) of the 3 particle size classes (CLAY, SILT and SAND, starting from the lower size of CLAY particles, 0, to the upper size of the SAND particles, 2000), in micrometers, FOR THE TEXTURE TRIANGLE. If not NULL, different from 'base.css.ps.lim', and css.transf == TRUE (not default), then the CLAY SILT and SAND coordinates of the texture triangle will be converted into the 'base.css.ps.lim' reference. If NULL, 'tri.css.ps.lim' will be set to the default value of the texture classification system chosen ('class.sys'). The transformation function is set by 'text.transf.fun' and is a log-linear interpolation by default.
Vector of 4 numericals. Particle size boundaries (upper and lower) of the 3 particle size classes (CLAY, SILT and SAND, starting from the lower size of CLAY particles, 0, to the upper size of the SAND particles, 2000), in micrometers, FOR THE TEXTURE DATA TABLE ('tri.data'). If not NULL, different from 'base.css.ps.lim', and css.transf == TRUE (not default), then the CLAY SILT and SAND coordinates of the texture data in tri.data will be converted into the 'base.css.ps.lim' reference. If NULL, 'tri.css.ps.lim' will be set to the default value of the texture classification system chosen ('class.sys'). The transformation function is set by 'text.transf.fun' and is a log-linear interpolation by default.
Single logical. Set to TRUE to transform the texture coordinates of the texture triangle ('class.sys') or the texture data ('tri.data') into the base particle size class limits. See 'base.css.ps.lim' for the base plot particle size class limits, 'tri.css.ps.lim' for the triangle particle size class limits and 'dat.css.ps.lim' for the data table particle size class limits. The transformation function is set by 'text.transf.fun' and is a log-linear interpolation by default. The default value is FALSE, so no transformation is made.
R function with the same argument names and same output as the function TT.text.transf(). 'text.transf.fun' is the function that transform the texture values from one system of particle class size limits to another. Only used if css.transf == TRUE. Default value is text.transf.fun=TT.text.transf. See also 'base.css.ps.lim', 'tri.css.ps.lim' and 'dat.css.ps.lim'.
Non pre-defined format. If the user specifies its own texture transformation function in 'text.transf.fun' (not TT.text.transf()), then he can use 'trsf.add.opt1' and 'trsf.add.opt1' as new, additional, argument for his function. So the format of 'trsf.add.opt1' depends on the function defined by the user in 'text.transf.fun'.
Non pre-defined format. If the user specifies its own texture transformation function in 'text.transf.fun' (not TT.text.transf()), then he can use 'trsf.add.opt1' and 'trsf.add.opt1' as new, additional, argument for his function. So the format of 'trsf.add.opt1' depends on the function defined by the user in 'text.transf.fun'.
Single numerical. Tolerance on the sum of the 3 particle size classes. The real sum of the 3 particle size classes in 'tri.data' should be >= text.sum * (1-text.tol) OR <= text.sum * (1+text.tol). See 'text.sum' for more details, as well as 'tri.sum.tst' (to prevent texture sum tests).
Single logical. If TRUE (the default), the sum of the 3 texture classes of each texture value in 'tri.data' will be checked in regard to 'text.sum' and 'text.tol'. If FALSE, no test is done.
Single logical. If TRUE (the default), the position of texture values in 'tri.data' are tested to check that they are not OUTSIDE the texture triangle (i.e. that some texture values may be negative).
Single character string. If PiC.type = "t" and a sample lie on the edge of 2 texture classes, then both will be outputed in a single character string, separated by 'collapse'. Example of output: [1] "C" "VF, F" "C" "C" "M"
Single logical. Set to FALSE to avoid any transformation of the texture data (trigonometric) prior to testure data classification. Setting to FALSE avoid some numerical accuracy problems when a point is on the border of a texture class.
Vector of 3 character strings. The 1st, 2nd and 3rd values must be either CLAY, SILT or SAND, and determines the particle size classes associated with the BOTTOM, LEFT and RIGHT axis, respectively. CLAY, SILT and SAND order in the vector is free, but they should all be used one time. The CLAY, SILT and SAND names must appear whatever the corresponding columns names in 'tri.data' (eventually set by 'css.names') and whatever the labels of the axis in the plot (eventually set by 'css.lab')
Vector of logicals, eventually with NA values. Direction of increasing texture values on the BOTTOM, LEFT and RIGHT axis, respectively. A value of TRUE means that the axis direction is clockwise. A value of FALSE means that the axis direction is counterclockwise. A value of NA means that the axis direction is centripetal. Possible combinations are c(T,T,T); c(F,F,F); c(F,T,NA) and c(T,NA,F), for fully clockwise, fully counterclockwise, right centripetal and left centripetal orientations, respectively.
Julien Moeys [aut, cre], Wei Shangguan [ctb], Rainer Petzold [ctb], Budiman Minasny [ctb], Bogdan Rosca [ctb], Nic Jelinski [ctb], Wiktor Zelazny [ctb], Rodolfo Marcondes Silva Souza [ctb], Jose Lucas Safanelli [ctb], Alexandre ten Caten [ctb]
require( "soiltexture" )
# Create a dummy data frame of soil textures:
my.text <- data.frame(
"CLAY" = c(05,60,15,05,25,05,25,45,65,75,13,47),
"SILT" = c(05,08,15,25,55,85,65,45,15,15,17,43),
"SAND" = c(90,32,70,70,20,10,10,10,20,10,70,10),
"OC" = c(20,14,15,05,12,15,07,21,25,30,05,28)
) #
# Display the table:
my.text
# Classify according to the HYPRES / European Soil Map classification
TT.points.in.classes(
tri.data = my.text[1:5,],
class.sys = "HYPRES.TT"
) #
# Classify according to the USDA classification
TT.points.in.classes(
tri.data = my.text[1:5,],
class.sys = "USDA.TT"
) #
# Classify according to the HYPRES / European Soil Map classification,
# returns logical values
TT.points.in.classes(
tri.data = my.text[1:5,],
class.sys = "HYPRES.TT",
PiC.type = "l"
) #
# Classify according to the HYPRES / European Soil Map classification,
# returns text
TT.points.in.classes(
tri.data = my.text[1:5,],
class.sys = "HYPRES.TT",
PiC.type = "t"
) #
# Classify according to the HYPRES / European Soil Map classification,
# returns text,
# custom class separator in case of points belonging to
# several classes.
TT.points.in.classes(
tri.data = my.text[1:5,],
class.sys = "HYPRES.TT",
PiC.type = "t",
collapse = ";"
) #
Run the code above in your browser using DataLab