plot.seas.sum: Plot boxplots of normalized seasonal sums

In seas: Seasonal Analysis and Graphics, Especially for Climatology

Plot boxplots of normalized seasonal sums

Description

Plots normalized seasonal sums using boxplots.

Usage

## S3 method for class 'seas.sum'
plot(x, var, norm = "days", year.filter, ylim,
    start = 1, rep = 0, col = "lightgrey", main, ylab, ...)

Arguments

x	a seas.sum object created by seas.sum
var	name of seasonal variable in x
norm	a variable to normalize by, either "days" (to produce unit/day) or "active" (unit/day, when active); it may also be a matrix with the same dimensions as x$days
year.filter	use only these years for analysis
ylim	either a single value for c(0, ylim), or a range of c(min, max) for the y-axis limits
start	starting bin at left-hand side of plot
rep	repeat bins on right-hand side of plot
col	colour for boxplot, default is "lightgrey"
main	title for plot; if it is missing, then it will automatically be generated
ylab	y-axis label; if it is missing, then it will automatically be generated
...	ignored

Details

This function is a boxplot interpretation of a seas.sum object. This is not the same as treating var as a continuous variable and using seas.var.plot, since a seas.sum object has been smoothed. Daily extreme values are not well represented here as a result.

Warning

The appearance of the boxplots are sensitive to the width parameter specified in the seas.sum function on strongly discontinuous variables. Small bin widths capture the discontinuities better than wider bins, and changes the skew of the distribution.

For instance, the median will appear to decrease as width decreases.

Author(s)

Mike Toews

See Also

seas.sum, image.seas.sum, seas.norm

Examples

data(mscdata)
par.orig <- par(no.readonly=TRUE)
on.exit(par.orig)

dat <- mksub(mscdata, id=1108447)
dat.ss <- seas.sum(dat)

# Normalized by the number of days in each bin
plot(dat.ss)

# Normalized by the number of active days in each bin
plot(dat.ss, norm="active")

# Snow, using a different start day, and a better y-axis:
dat2.ss <- seas.sum(dat, var="snow", width="mon",
                    start.day=as.Date("2000-08-01"))
par(yaxs="i")
plot(dat2.ss, var="snow")
plot(dat2.ss, var="snow", norm="active")

seas documentation built on May 2, 2022, 5:08 p.m.