legend.only = T（光栅包）时的图例属性

Question

When plotting only the legend (of a raster object - a colorbar):

require(raster)
r = raster()
r[] = 1
plot(r, legend=F)
plot(r, zlim=c(-10,10), legend.only=T)

how can I control the legend axis label size, tick length, and other legend properties? I know I can call par(...) before the last plot() call, but is there a cleaner way?

Answer 1

You can pass axis.args and legend.args as arguments to the legend only function call, as for image.plot in the fields package.

For example, to specify tick positions and labels, and to reduce tick label size, the following should do the trick. It will also accept arguments such as legend.width and legend.shrink .

require(raster)
data(volcano)
r <- raster(volcano)
plot(r, col=topo.colors(100), legend=FALSE, axes=FALSE)
r.range <- c(minValue(r), maxValue(r))
plot(r, legend.only=TRUE, col=topo.colors(100),
     legend.width=1, legend.shrink=0.75,
     axis.args=list(at=seq(r.range[1], r.range[2], 25),
                    labels=seq(r.range[1], r.range[2], 25), 
                    cex.axis=0.6),
     legend.args=list(text='Elevation (m)', side=4, font=2, line=2.5, cex=0.8))

Answer 2

It's also possible to work with the 'smallplot' argument in when 'legend.only=TRUE'. Small works from the bottom/left corner of the plot area smallplot=c(min % from left, max % from left, min % from bottom, max % from bottom).

# load data & plot
require(raster); data(volcano); r <- raster(volcano)
plot(r, col=topo.colors(100), legend=FALSE, axes=FALSE)
r.range <- c(minValue(r), maxValue(r))

plot(r, legend.only=TRUE, col=topo.colors(100), legend.width=1, legend.shrink=0.75,
    smallplot=c(0,.09, .3,.75)); par(mar = par("mar"))

plot(r, legend.only=TRUE, col=topo.colors(100), legend.width=1, legend.shrink=0.75,
    smallplot=c(0.3,0.5, 0.2,0.7)); par(mar = par("mar"))

plot(r, legend.only=TRUE, col=topo.colors(100), legend.width=1, legend.shrink=0.75,
    smallplot=c(0.85,0.9, 0.7,0.9)); par(mar = par("mar"))

plot(r, legend.only=TRUE, col=topo.colors(100), legend.width=1, legend.shrink=0.75,
    smallplot=c(0.7,0.90, 0.05,0.2)); par(mar = par("mar"))