aqp: texture – R documentation

Become an expert in R — Interactive courses, Cheat Sheets, certificates and more!

texture

Textural conversions

Description

These functions consist of several conversions between sand, silt and clay to texture class and visa versa, textural modifiers to rock fragments, and grain size composition to the family particle size class.

Usage

texcl_to_ssc(texcl = NULL, clay = NULL, sample = FALSE)

ssc_to_texcl(sand = NULL, clay = NULL, as.is = FALSE, droplevels = TRUE)

texmod_to_fragvoltot(texmod = NULL, lieutex = NULL)

texture_to_taxpartsize(
  texcl = NULL,
  clay = NULL,
  sand = NULL,
  fragvoltot = NULL
)

Arguments

`texcl`	vector of texture classes than conform to the USDA code conventions (e.g. c\|C, sil\|SIL, sl\|SL, cos\|COS)
`clay`	vector of clay percentages#'
`sample`	logical: should ssc be random sampled from the lookup table? (default: FALSE)
`sand`	vector of sand percentages
`as.is`	logical: should character vectors be converted to factors? (default: TRUE)
`droplevels`	logical: indicating whether to drop unused levels in factors. This is useful when the results have a large number of unused classes, which can waste space in tables and figures.
`texmod`	vector of textural modifiers that conform to the USDA code conventions (e.g. gr\|GR, grv\|GRV)
`lieutex`	vector of in lieu of texture terms that conform to the USDA code conventions (e.g. gr\|GR, pg\|PG), only used when fragments or artifacts are > 90 percent by volume (default: NULL))
`fragvoltot`	vector of rock fragment percentages

Details

These functions are intended to estimate missing values or check existing values. The ssc_to_texcl() function uses the same logic as the particle size estimator calculation in NASIS to classify sand and clay into texture class. The results are stored in soiltexture and used by texcl_to_ssc() as a lookup table to convert texture class to sand, silt and clay. The function texcl_to_ssc() replicates the functionality described by Levi (2017).

When sample = TRUE, the results can be used to estimate within-class, marginal distributions of sand, silt, and clay fractions. It is recommended that at least 10 samples be drawn for reasonable estimates.

Unlike the other functions, texture_to_taxpartsize() is intended to be computed on weighted averages within the family particle size control section. Also recall from the criteria that carbonate clay should be subtracted from clay content and added to silt content. Similarly, if the percent of very fine sand is known it should be subtracted from the sand, and added to the silt content. Unlike the other functions, texture_to_taxpartsize() is intended to be computed on weighted averages within the family particle size control section. Also recall from the criteria that carbonate clay should be subtracted from clay content and added to silt content. Similarly, if the percent of very fine sand is known it should be subtracted from the sand, and added to the silt content.

Value

texcl_to_ssc: A data.frame containing columns "sand","silt", "clay"

ssc_to_texcl: A character vector containing texture class

texmod_to_fragvoltot: A numeric vector containing total fragment volume

texture_to_taxpartsize: a character vector containing "taxpartsize" classes

Author(s)

Stephen Roecker

References

Matthew R. Levi, Modified Centroid for Estimating Sand, Silt, and Clay from Soil Texture Class, Soil Science Society of America Journal, 2017, 81(3):578-588, ISSN 1435-0661, doi: 10.2136/sssaj2016.09.0301.

Examples

# example of ssc_to_texcl()
tex <- expand.grid(sand = 0:100, clay = 0:100)
tex <- subset(tex, (sand + clay) < 101)
tex$texcl <- ssc_to_texcl(sand = tex$sand, clay = tex$clay)
head(tex)

# example of texcl_to_ssc(texcl)
texcl <- c("cos", "s", "fs", "vfs", "lcos", "ls",
          "lfs", "lvfs", "cosl", "sl", "fsl", "vfsl", "l",
          "sil", "si", "scl", "cl", "sicl", "sc", "sic", "c"
          )
test <- texcl_to_ssc(texcl)
head(test <- cbind(texcl, test), 10)


# example of texcl_to_ssc(texcl, clay)
data(soiltexture)
st <- soiltexture$values
idx <- sample(1:length(st$texcl), 10)
st <- st[idx, ]
ssc <- texcl_to_ssc(texcl = st$texcl)
head(cbind(texcl = st$texcl, clay = ssc$clay))


# example of texmod_to_fragvoltol()
frags <- c("gr", "grv", "grx", "pgr", "pgrv", "pgrx")
head(texmod_to_fragvoltot(frags))


# example of texture_to_taxpartsize()
tex <- data.frame(texcl = c("c", "cl", "l", "ls", "s"),
                  clay  = c(55, 33, 18, 6, 3),
                  sand  = c(20, 33, 42, 82, 93),
                  fragvoltot = c(35, 15, 34, 60, 91))
tex$fpsc <- texture_to_taxpartsize(texcl = tex$texcl,
                                   clay = tex$clay,
                                   sand = tex$sand,
                                   fragvoltot = tex$fragvoltot)
head(tex)


# example of texture_to_taxpartsize() with carbonate clay and very fine sand
carbclay <- rnorm(5, 2, 3)
vfs <- rnorm(5, 10, 3)
st$fpsc <- texture_to_taxpartsize(texcl = tex$texcl,
                                  clay = tex$clay - carbclay,
                                  sand = tex$sand - vfs,
                                  fragvoltot = tex$fragvoltot)
head(tex)


# example of sample = TRUE
texcl <- rep(c("cl", "sil", "sl"), 10)
ssc1 <- cbind(texcl, texcl_to_ssc(texcl = texcl, sample = FALSE))
ssc2 <- cbind(texcl, texcl_to_ssc(texcl = texcl, sample = TRUE))
ssc1$sample <- FALSE
ssc2$sample <- TRUE
ssc  <- rbind(ssc1, ssc2)
aggregate(clay ~ sample + texcl, data = ssc, summary)

aqp

Algorithms for Quantitative Pedology

v1.29

GPL (>= 3)

Authors

Dylan Beaudette [aut, cre], Pierre Roudier [aut, ctb], Andrew Brown [aut, ctb]

Initial release

2021-04-05