mgcv.polarPlot(polar,
pollutant = "nox", type = "default",
resolution = "normal", limits = NA,
exclude.missing = TRUE, uncertainty = FALSE,
cols = "default", min.bin = 1, upper = NA,
ws.int = 5, angle.scale = 45, units = "(m/s)",
force.positive = TRUE, k = 100, normalise = FALSE,
main = "", key.header = "", key.footer = pollutant,
key.position = "right", key = NULL,
auto.text = TRUE, ...)ws,
wd and a pollutant. Can also contain date if plots by
time period are required.pollutant =
"nox". There can also be more than one pollutant specified
e.g. pollutant = c("nox", "no2"). The main use type determines how the data are split
i.e. conditioned, and then plotted. The default is will produce a
single plot using the entire data. Type can be one of the built-in
types as detailed in cutData e.g. "season"TRUE (the
default) removes points from the plot that are too far from the
original data. The smoothing routines will produce predictions at
points where no data exist i.e. they predict. By removing the poTRUE three plots are produced on the same scale
showing the predicted surface together with the estimated lower and
upper uncertainties at the 95% confidence intervacolours() to see the full list). An examws.int to a
lower value when the wind speeds are low to get a sensible grid
spacing.angle.scale
to another value (between 0 and 360 degrees) ws.int interval to reduce chart
clutter.TRUE. Sometimes if
smoothing data with streep gradients it is possible for predicted
values to be negative. force.positive = TRUE ensures that
predictions remain postive. This is useful for several rFALSE. Typically, value of around 100 (the default)
seems to be suitable and will resolve more features in the plot.TRUE concentrations are normalised by
dividing by their mean value. This is done after fitting the
smooth surface. This option is particularly useful if one is
interested in the patterns of concentrations for several pollkey.header =
"header", key.footer = "footer1" adds addition text
above and below the scale key. These arguments are passed to
drawOpenKey "top", "right",
"bottom" and "left".drawOpenKey. See
drawOpenKey for further details.TRUE (default) or FALSE. If
TRUE titles and axis labels will automatically try and format
pollutant names and units properly e.g. by subscripting the `2' in
NO2.polarPlot also
returns an object of class ``openair''. The object includes three main
components: call, the command used to generate the plot;
data, the data frame of summarised information used to make the
plot; and plot, the plot itself. If retained, e.g. using
output <- polarPlot(mydata, "nox"), this output can be used to
recover the data, reproduce or rework the original plot or undertake
further analysis.
An openair output can be manipulated using a number of generic
operations, including print, plot and
summarise. See openair.generics for further
details.
polarPlot surface data can also be extracted directly using the
$data operator, e.g. object$data for output <-
polarPlot(mydata, "nox"). This returns a data frame with four set
columns: cond, conditioning based on type; u and
v, the translational vectors based on ws and wd;
and the local pollutant estimate.type = "year" is chosen. The
function may fail if there are insufficient data to smooth the
surface.mgcv package that uses Generalized Additive
Models. While methods do exist to find an optimum level of
smoothness, they are not necessarily useful. The principal aim of
polarPlot is as a graphical analysis rather than for
quantitative purposes. In this respect the smoothing aims to strike a
balance between interesting (real) features and overly noisy data. The
defaults used in polarPlot are based on the analysis of data
from many different sources. More advanced users may wish to modify
the code and adopt other smoothing approaches.
These plots often show interesting features at higher wind speeds (see
references below). For these conditions there can be very few
measurements and therefore greater uncertainty in the calculation of
the surface. There are several ways in which this issue can be
tackled. First, it is possible to avoid smoothing altogether and use
polarFreq. Second, the effect of setting a minimum
number of measurements in each wind speed-direction bin can be
examined through min.bin. It is possible that a single point at
high wind speed conditions can strongly affect the surface
prediction. Therefore, setting min.bin = 3, for example, will
remove all wind speed-direction bins with fewer than 3 measurements
before fitting the surface. Third, consider setting
uncertainty = TRUE. This option will show the predicted surface
together with upper and lower 95% confidence intervals, which take
account of the frequency of measurements.
Variants on polarPlot include polarAnnulus and
polarFreq.polarAnnulus, polarFreq# load example data from package
data(mydata)
# basic plot
polarPlot(mydata, pollutant = "nox")
# polarPlots by year on same scale
polarPlot(mydata, pollutant = "so2", type = "year", main = "polarPlot of so2")
# set minimum number of bins to be used to see if pattern remains similar
polarPlot(mydata, pollutant = "nox", min.bin = 3)
# plot by day of the week
polarPlot(mydata, pollutant = "pm10", type = "weekday")
# show the 95\% confidence intervals in the surface fitting
polarPlot(mydata, pollutant = "so2", uncertainty = TRUE)Run the code above in your browser using DataLab