NavigationRandom image User loginSyndicate |
Summarizing Circular Data in R: Aspect AngleSubmitted by dylan on Wed, 2012-10-10 15:26.
The orientation of terrain surface (aspect) can have dramatic effects on landscape-scale variation in soil temperature and moisture. Summarizing aspect angle is complicated by the fact that sampled values are measured on a circular scale. The circular package for R can be used to summarize, model, and visualize this type of data. An example function is presented below that demonstrates several components of the circular package: special data structures, summary functions, plotting functions and their application to circular data derived from measurements collected by compass. Spread and central tendency are depicted with a combination of circular histogram and kernel density estimate. The circular mean, and relative confidence in that mean are depicted with an arrow: longer arrow lengths correspond to greater confidence in the mean.
# load library (you may have to install this library first) library(circular) # plot a vector of aspect measurements, with units of degrees, measured via compass aspect.plot <- function(p, p.bins=60, p.bw=30, p.axis=seq(0, 350, by=10), plot.title=NULL) { # remove NA p <- na.omit(p) # make a circular class object: using native degrees, template sets zero and direction c.p <- circular(p, units='degrees', template="geographics") # compute the mean m.p <- mean(c.p) # compute mean resultant length rho.p <- rho.circular(c.p) # setup custom axis a.p <- circular(p.axis, units='degrees', template="geographics") # circular histogram plot(c.p, axes=FALSE, stack=TRUE, bins=p.bins, shrink=1.45, cex=1, sep=0.06, pch=21, col=1, bg='RoyalBlue') # add circular density, bw is the smoothness parameter lines(density(c.p, bw=p.bw), col='RoyalBlue', lty=2) # add axes: note work-around for strange bug... axis.circular(at=(-a.p) + 90, labels=a.p, cex=0.6) # annotate north text(0, 1.125, 'North', cex=0.85, font=1) # annotate mean with an arrow arrows.circular(m.p, shrink=rho.p, length=0.15, col='RoyalBlue') # add title text(0, -0.25, plot.title, cex=0.85, font=2) } # generate some fake data p.narrow <- runif(n=25, min=215, max=280) p.wide <- runif(n=25, min=0, max=270) # set figure margins to 0, 2-column plot par(mar=c(0,0,0,0), mfcol=c(1,2)) # plot aspect.plot(p.narrow, p.bw=10, plot.title='Soil A') aspect.plot(p.wide, p.bw=10, plot.title='Soil B') ( categories: )
Reply |
