parameter_predictions_boxplots: Hydraulic parameter predictions boxplots

In jschap1/WSEw: River Geometry Package for Studying Water Surface Elevation-Width Relationships

Description

Makes box and whisker plots showing predicted hydraulic parameter error

Usage

parameter_predictions_boxplots(z.l, z.sb, z.nl, z.nlsb, k,
  lumped = FALSE, legend = TRUE, ...)

Arguments

z.l	linear model parameter predictions or prediction errors
z.sb	SB model parameter predictions or prediction errors
z.nl	NL model parameter predictions or prediction errors
z.nlsb	NLSB model parameter predictions or prediction errors
k	exposure level
lumped	flag, if true, lump the cross sections together and just compare methods
legend	flag for legend
...	other parameters as taken by boxplot()
absolute	flag for plotting absolute error

Details

If lumped = FALSE, plots the results over each cross section separately. This is the updated version, which can take either predictions or prediction error as input.

Examples

Test case
z.l <- z0.l.error[1:3,,1:2]
z.sb <- z0.sb.error[1:3,,1:2]
z.nl <- z0.nl.error[1:3,,1:2]
z.nlsb <- z0.nlsb.error[1:3,,1:2]
parameter_predictions_boxplots(z.l, z.sb, z.nl, z.nlsb, k = 12, names = bplab)
z.l[,12,]
z.sb[,12,]
z.nl[,12,]
z.nlsb[,12,]

bplab <- c("1","1","1","1","2","2","2","2","3","3","3","3")
lumpedlab <- c("L","SB","NL","NLSB")
par(mfrow = c(2,2))
par(opar)
k <- 16
z0 error
parameter_predictions_boxplots(z0.l.error, z0.sb.error, z0.nl.error, z0.nlsb.error,
                               k = k,
                               lumped = TRUE,
                               main = paste("z0 error at", 100*expo[k], "percent channel exposure"),
                               ylab = "z0 error (m)",
                               legend = FALSE,
                               notch = TRUE,
                               names = lumpedlab)
A0 error
parameter_predictions_boxplots(A0.l.error, A0.sb.error, A0.nl.error, A0.nlsb.error,
                               k = k,
                               lumped = TRUE,
                               main = paste("A0 error at", 100*expo[k], "percent channel exposure"),
                               ylab = "A0 error (sq. m)",
                               legend = FALSE,
                               notch = TRUE,
                               names = lumpedlab)

parameter_predictions_boxplots(z0.l, z0.sb, z0.nl, z0.nlsb, z0.true.ra, k = k, lumped = FALSE,
                               main = paste("z0 error at", 100*expo[k], "percent channel exposure"),
                               ylab = "average z0 error (m)", legend = FALSE, A0 = FALSE, notch = TRUE, names = bplab)
abline(h = z0.true.ra[3])

jschap1/WSEw documentation built on May 12, 2019, 10:50 a.m.