Function to create boxplots of water-quality data that include censored values.
is the dataset with columns that begin with P followed by a number indicating concentration data and columns that begin with R followed by numbers that match those of the concentration data indicating qualification codes. See example data sets for more information about the data format, IllRivValleyCty and qwMoRivOmaha.
is a character vector with the beginning of the column headers for remarks code (default is R), and beginning of column headers for concentration data (default is P for parameter).
is a label for the plot title indicating the site where the water-quality samples were collected.
arguments to be passed to plot method.
This function determines the columns within the data set that have concentration data, based on them having column headings that start with P (or a user-specificed indicater in the second element of the qwcols argument) followed by a number. The function determines the associated remark, or qualification columns, based on them having column headings that start with R (or a user-specificed indicater in the first element of the qwcols argument) followed by numbers that match the associated concentration data. Then it determines which values are censored, indicated by a less than symbol in the R columns, performs regression on order statistics, ros, using the NADA package, and estimates values for the censored concentrations for constituents with less than 90-percent censoring. The water-quality concentrations are then depicted by boxplots.
The regression on order statistics function in R package
NADA (Lee, 2012),
ros, is an implementation
of a regression on order statistics designed for
multiply-censored analytical-chemistry data (Helsel,
2005). The method assumes data contains zero to many
left-censored (less-than) values. For highly censored
data, ros may produce a warning message. Such as,
1 2 3
Warning messages: 1: In ros(my.list$obs, my.list$cen) : Input > 80% censored -- Results are tenuous.
The boxplot will still be generated, but the user should consider the warning message when interpreting the plots. See Oblinger Childress and others (1999) for information on the remark codes used by the U.S. Geological Survey.
Karen R. Ryberg
Helsel, D.R., 2005, Nondetects and data analysis: New York, John Wiley and Sons.
Lee, Lopaka, 2012, Nondetects and data analysis for environmental data: R package version 1.5-4, http://CRAN.R-project.org/package=NADA.
Oblinger Childress, C.J., Foreman, W.T., Connor, B.F., and Maloney, T.J., 1999, New reporting procedures based on long-term method detection levels and some considerations for interpretations of water-quality data provided by the U.S. Geological Survey: U.S. Geogolical Survey Open-File Report 99–193, 19 p. (Also available at http://water.usgs.gov/owq/OFR_99-193/index.html.)
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
data(swData) # summary of water-quality concentrations apply(IllRivValleyCty[, grep("P[[:digit:]]", dimnames(IllRivValleyCty)[])], 2, summary) # simple boxplot of water-quality concentrations rosBoxPlot(IllRivValleyCty) # same boxplot function with many additional plotting arguments rosBoxPlot(IllRivValleyCty, site="05586100 Illinois River at Valley City, IL", log="y", yaxt="n", ylim=c(0.0000001, 1), qwcols=c("R", "P"), ylab=c("Concentration, micrograms per liter"), col="skyblue1", cex.axis=0.7, cex.sub=0.8, par(tcl=0.5, las=1, yaxs="i", mgp=c(3, 0.5, 0), mar=c(5, 5, 2, 2),cex.main=0.9)) axis(2, at=c(0.0000001, 0.000001, 0.00001, 0.0001, 0.001, 0.01, 0.1, 1), labels=c("0.0000001", "0.000001","0.00001", "0.0001", "0.001", "0.01", "0.1", "1"), cex.axis=0.7)
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.