| size.dist.trellis {heR.IndoorAir} | R Documentation |
Function to plot, using the lattice package,
a Trellis of particle size distributions for
number, mass, volume, surface area, emissions etc., in the
standard format: pdf = dX / dlog(Dp) where X = N, M, V, ...
[pdf = estimated probability density function.]
In this case, the pdf is
un-normalized so that one reads the physical quantity along the
vertical axis instead of the probability density, as is the case
for the usual pdf encountered in statistics.
size.dist.trellis(formula, data,
super=TRUE, fit.lnorm=FALSE,
quant="X", prefix="Data.Set",
grid=TRUE, exact10=TRUE, rect=FALSE,
lwd = 2, lty = 1:length(data), col = "black",
fit.col1="white", fit.col2="black", fit.lty="dashed", fit.lwd=2,
type = "s", fill.col="black",
pch = 1, cex = 1,
scale.cex=1.5, strip.style = 1,
strip.bg = "NA", strip.fg = "black",
strip.cex = 2,
panel = function(...) panel.size.dist(...,
lwd = lwd, lty = lty, col = col, pch = pch, cex = cex,
fit.col1=fit.col1, fit.col2=fit.col2,
fit.lty=fit.lty, fit.lwd=fit.lwd,
fill.col=fill.col, rect=rect, grid=grid,
atx=atx, ylim=ylim, aty=aty,
fit.lnorm=fit.lnorm),
strip = function(..., bg) strip.default(...,
bg = strip.bg, style = strip.style, strip.names = T),
as.table = TRUE, main, main.cex=1.4, xlab, ylab, lab.cex=1.4,
...)
The plots are designed so that the area under each bin is the proportion of "mass" that that bin contributes to the overall "mass". The function assumes that there is one extra row in each data frame containing the very top bin limit. The corresponding top bin value for number, mass, etc is taken to be a dummy value and discarded. The number of limits must always be one greater than the number of bin values.
With super=TRUE the data frames will be overlaid, otherwise
each data frame with be associated with a new conditioning variable
with levels equal to their name in the list. In this way the data
frames will be plotted in separate panels instead of laid over
each other (superimposed).
Neil E. Klepeis
nklepeis@uclink.berkeley.edu
http://socrates.berkeley.edu/~nklepeis