I am plotting hovmoller
plot for three rasters but the function plots the latitude on the x axis and the time on the y axis. normally (much easier to read) latitude is on the Y axis.
Reproducible example:
library("rasterVis")
r <- raster(nrows=10, ncols=10)
r <- setValues(r, 1:ncell(r))
r1 <- raster(nrows=10, ncols=10)
r1 <- setValues(r1, 1:ncell(r))
r2 <- raster(nrows=10, ncols=10)
r2 <- setValues(r2, 1:ncell(r))
St=stack(r,r1,r2)
idx <- seq(as.Date('2008-01-15'), as.Date('2008-1-17'), 'day')
SISmm <- setZ(St, idx)
hovmoller(SISmm, contour=FALSE, panel=panel.levelplot.raster,
yscale.components=yscale.raster.subticks,
interpolate=TRUE, par.settings=RdBuTheme)
This produces:
So I want the time to be on the X axis and latitude on the Y axis!
That layout is not implemented in the rasterVis::hovmoller
method. Only the common arrangement (time on the y-axis and
latitude/longitude on the x-axis) is available at the moment.
You can try a workaround using the most important parts of the
function code.
For your example, you have to create a RasterLayer
with the latitude
values. This RasterLayer
is used by zonal
to aggregate your
RasterStack
:
library('raster')
dirLayer <- init(SISmm, v='y')
z <- zonal(SISmm, dirLayer, FUN='mean', digits=2)
The result is a matrix which is converted to a data.frame
to be
displayed with lattice::levelplot
.
## Time on the x-axis, latitude on the y-axis
dat <- expand.grid(y=z[,1], x=idx)
dat$z <- as.vector(z[,-1], mode='numeric')
levelplot(z ~ x*y, data=dat,
xlab='Time', ylab='Latitude',
panel=panel.levelplot.raster,
interpolate=TRUE,
par.settings=RdBuTheme())