timePlot(mydata, pollutant = "nox", group = FALSE, stack
= FALSE, normalise = NULL, avg.time = "default",
data.thresh = 0, statistic = "mean", percentile = NA,
date.pad = FALSE, type = "default", cols = "brewer1",
plot.type = "l", key = TRUE, log = FALSE, smooth = FALSE,
ci = TRUE, ref.x = NULL, ref.y = NULL, key.columns = 1,
name.pol = pollutant, date.breaks = 7, auto.text = TRUE,
...)date field and at least one variable to plot.pollutant = c("nox", "co") should be used.FALSE, which means they are plotted in
separate panels with their own scaled. If TRUE
then they are plotted oTRUE the time series will be
stacked by year. This option can be useful if there are
several years worth of data making it difficult to see
much detail when plotted on a single plot.normalise
can take two values, either "mean" or a string
representing a date in UK format e.g. "1/1/1998" (in the
format dd/mm/YYYY). If avg.time. A value
of zero means that all available data will be used in a
particular period regardless if of the number of values
available. Conversely, a value of statistic = "percentile" and when aggregating the
data with avg.time. More than one percentile level
is allowed for type = "default" e.g.
percentile = c(50, 95). Not usedate.pad = TRUE the time gaps between the chunks
are shown properly, rather thantype 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", "cols = "black"lattice plot type, which is a
line (plot.type = "l") by default. Another useful
option is plot.type = "h", which draws vertical
lines.TRUE.FALSE. If TRUE a well-formatted
log10 scale is used. This can be useful for plotting data
for several different pollutants that exist on very
different scales. IFALSE.ci
determines whether the 95 shown.columns to be less than the number of
pollutants.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.cutData and lattice:xyplot. For example,
timePlot passes the option hemisphere =
"southern" on to cutData to provide southern
(rather ttimePlot
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 <- timePlot(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 summary. See openair.generics
for further details.timePlot is the basic time series plotting
function in openair. Its purpose is to make it
quick and easy to plot time series for pollutants and
other variables. The other purpose is to plot potentially
many variables together in as compact a way as possible.
The function is flexible enough to plot more than one
variable at once. If more than one variable is chosen
plots it can either show all variables on the same plot
(with different line types) on the same scale, or
(if group = FALSE) each variable in its own panels
with its own scale.
The general preference is not to plot two variables on
the same graph with two different y-scales. It can be
misleading to do so and difficult with more than two
variables. If there is in interest in plotting several
variables together that have very different scales, then
it can be useful to normalise the data first, which can
be down be setting the normalise option.
The user has fine control over the choice of colours,
line width and line types used. This is useful for
example, to emphasise a particular variable with a
specific line type/colour/width.
timePlot works very well with
selectByDate, which is used for selecting
particular date ranges quickly and easily. See examples
below.
By default plots are shown with a colour key at the
bottom and in teh case of multiple pollutants or sites,
strips on teh left of each plot. Sometimes this may be
overkill and the user can opt to remove the key and/or
the strip by setting key and/or strip to
FALSE. One reason to do this is to maximise the
plotting area and therefore the information shown.MannKendall, smoothTrend,
linearRelation, selectByDate
and timeAverage for details on selecting
averaging times and other statistics in a flexible way# basic use, single pollutant
timePlot(mydata, pollutant = "nox")
# two pollutants in separate panels
timePlot(mydata, pollutant = c("nox", "no2"))
# two pollutants in the same panel with the same scale
timePlot(mydata, pollutant = c("nox", "no2"), group = TRUE)
# alternative by normalising concentrations and plotting on the same
scale
timePlot(mydata, pollutant = c("nox", "co", "pm10", "so2"), group = TRUE, avg.time =
"year", normalise = "1/1/1998", lwd = 3, lty = 1)
# examples of selecting by date
# plot for nox in 1999
timePlot(selectByDate(mydata, year = 1999), pollutant = "nox")
# select specific date range for two pollutants
timePlot(selectByDate(mydata, start = "6/8/2003", end = "13/8/2003"),
pollutant = c("no2", "o3"))
# choose different line styles etc
timePlot(mydata, pollutant = c("nox", "no2"), lty = 1)
# choose different line styles etc
timePlot(selectByDate(mydata, year = 2004, month = 6), pollutant =
c("nox", "no2"), lwd = c(1, 2), col = "black")
# different averaging times
#daily mean O3
timePlot(mydata, pollutant = "o3", avg.time = "day")
# daily mean O3 ensuring each day has data capture of at least 75\%
timePlot(mydata, pollutant = "o3", avg.time = "day", data.thresh = 75)
# 2-week average of O3 concentrations
timePlot(mydata, pollutant = "o3", avg.time = "2 week")Run the code above in your browser using DataLab