demo/daily.synthesis.R

In jobonaf/pesco: Data fusion for air quality data

#' ---
#' title: "Demo: How to prepare daily data for PESCO"
#' author: "Giovanni Bonafe'"
#' date: "March 23rd, 2015"
#' ---

#' ## Demo: How to prepare daily data for PESCO
#' You can reproduce the following R code with 
#' ```demo(daily.synthesis)``` after loading package ```pesco```.

## load package
require(pesco)

#' With ```data()``` you load some example datasets. 
#' See ```example/prepare-datasets.R``` to prepare these datasets 
#' from ASCII and NetCDF files.

## load hourly observation
data(NO2.obs)

#' With ```str(NO2.obs)``` you can compactly display the structure 
#' of the object ```NO2.obs```.

## calculate daily maxima
NO2.obs.max <- dailyObs(NO2.obs,statistic="max",pollutant="NO2")
boxplot(data=NO2.obs.max, NO2~Time, range=0, border="orange", 
        col="orange", lty=1)

#' After calculating daily maxima and averages of ```NO2.obs```, 
#' we can remove it. Obviously, ```boxplot()``` is not required to
#' prepare the data, it is just useful to have an idea of what you did.

## calculate daily averages
NO2.obs.ave <- dailyObs(NO2.obs,statistic="mean",pollutant="NO2")
rm(NO2.obs)
boxplot(data=NO2.obs.max, NO2~Time, range=0, border="orange", 
        col="orange", lty=1)
boxplot(data=NO2.obs.ave, NO2~Time, range=0, border="olivedrab", 
        col="olivedrab", lty=1, add=T, xaxt="n", yaxt="n")

## load hourly CTM concentrations
data(PM10.ctm)

#' After calculating daily averages of ```PM10.ctm```, 
#' we can remove it. To plot ```PM10.ctm.ave``` we use 
#' ```filled.contour```. Note that ```PM10.ctm.ave``` is a list,
#' and its elements ```PM10.ctm.ave$data``` is an array with 3 dimensions.
#' Therefore we use ```PM10.ctm.ave$data[,,1]``` to select the first day.

## calculate daily averages
PM10.ctm.ave <- dailyCtm(PM10.ctm, statistic="mean")
rm(PM10.ctm)
library(fields)
filled.contour(PM10.ctm.ave$data[,,1],color.palette=tim.colors)