documentation and tarball 0.2.2

.Rbuildignore

@@ -1,2 +1,4 @@
 ^.*\.Rproj$
 ^\.Rproj\.user$
+tmp/
+tar/

DESCRIPTION

@@ -1,8 +1,8 @@
 Package: pesco
 Type: Package
 Title: Data fusion for air quality data
-Version: 0.2
-Date: 2015-02-26
+Version: 0.2.2
+Date: 2015-02-27
 Author: Lucia Paci, Giovanni Bonafe'
 Maintainer: G.Bonafe' <[email protected]>
 Imports: geoR, akima, fields, caTools, ncdf, timeDate, chron

NEWS

@@ -0,0 +1,5 @@
+Changes in pesco 0.2.2 [2015-03-27]
+=====================================
+ * added documentation (man/*.Rd files)
+
+

R/daily_synthesis.R

@@ -26,15 +26,17 @@ dailyObs <- function(data, statistic, pollutant,
                      Time="Time", Code="Code",
                      others=c("Name","Municipality",
                               "Lat","Lon","Elev","Type")) {
-  codes <- unique(data[,Code])
+  codes <- unique(data[,Code])  ## extract station codes
   flag=TRUE
   for (code in codes) {
     #print(code)
-    dat <- data[data[,Code]==code,]
+    dat <- data[data[,Code]==code,] ## select data of a single station
     ttt <- dat[,Time]
     sta <- dailyStat(x=dat[,pollutant],time=ttt,statistic,necess=0.75)
-    out <- data.frame(sta, names(sta),
-                      dat[rep(1,length(sta)),
+    ## build a dataframe with...
+    out <- data.frame(sta,   ## ...daily statistic,...       
+                      names(sta),  ## ...dates and...
+                      dat[rep(1,length(sta)),  ## ...station attributes
                           c(Code,others)])
     colnames(out) <- c(pollutant, Time, Code, others)
     if(flag) {

R/geo.functions.R

@@ -54,3 +54,20 @@ ll2utm.grid <- function(lat,lon,round=-2,iz=32) {
                        y=round(utm$y,round))
   return(coords)
 }
+
+## euclidean distance
+Dist <- function(x,y,xi,yi) sqrt((x-xi)^2+(y-yi)^2)
+
+## find nearest
+which.nearest <- function(xo,yo,xi,yi) {
+  out <- rep(NA,length(xo))
+  xo <- as.vector(xo)
+  yo <- as.vector(yo)
+  xi <- as.vector(xi)
+  yi <- as.vector(yi)
+  for(i in 1:length(xo)) {
+    out[i] <- which.min(Dist(xo[i],yo[i],xi,yi))[1]
+  }
+  return(out)
+}
+

R/interpolate.R

@@ -1,19 +1,3 @@
-## euclidean distance
-Dist <- function(x,y,xi,yi) sqrt((x-xi)^2+(y-yi)^2)
-
-## find nearest
-which.nearest <- function(x,y,xi,yi) {
-  out <- rep(NA,length(x))
-  x <- as.vector(x)
-  y <- as.vector(y)
-  xi <- as.vector(xi)
-  yi <- as.vector(yi)
-  for(i in 1:length(x)) {
-    out[i] <- which.min(Dist(x[i],y[i],xi,yi))[1]
-  }
-  return(out)
-}
-
 ## interpolation (to grid or to sparse points)
 Interp <- function(x,y,z,xp,yp,
                    method="linear",

R/read.R

@@ -16,7 +16,7 @@ read.ncdf.arpaer <- function(con=NULL, pollutant="pm10", lev=1,
   lat  <- get.var.ncdf(nc,varid="lat")
   Time <- as.POSIXct(get.var.ncdf(nc,varid="Times"), 
                      format="%Y-%m-%d_%H:%M:%S", tz=tz.in)
-  Time <- tz.change(time.in=Time,tz.in=tz.in,tz.out=tz.out)
+  Time <- tz.change(x=Time,tz.in=tz.in,tz.out=tz.out)
   value  <- get.var.ncdf(nc,varid=pollutant)
   if(length(dim(value))==4) {
     value <- value[,,lev,]
@@ -161,13 +161,13 @@ read.grads <- function(filectl) {
 }
 
 # Funzione R per la lettura di file in formato estra_qaria.
-read.qaria <- function(filename) {
-  hourly <- scan(filename,skip=2,what="",nlines=1)[4]=="hh"
-  dat <- read.fwf(file = filename,
+read.qaria <- function(file) {
+  hourly <- scan(file,skip=2,what="",nlines=1)[4]=="hh"
+  dat <- read.fwf(file = file,
                   na.strings=c("    -9999.0","     -999.0"),
                   widths=c(4,rep(3,2+as.integer(hourly)),11),
                   header=FALSE,skip=3)
-  header=scan(filename, skip=2, what="", n=4++as.integer(hourly))
+  header=scan(file, skip=2, what="", n=4++as.integer(hourly))
   names(dat) <- header
   strTime <- paste(dat$aaaa,dat$mm,dat$gg,sep="-")
   if(hourly) {

R/time.functions.R

@@ -21,26 +21,8 @@ Ndays.in.year <- function(year,tz="Africa/Algiers") {
   return(nd)
 }
 
-squeeze<- function(x) {
-  if(length(x)==0) {
-    out <- "-"
-  } else {
-    x <- sort(x,decreasing=F)
-    xb <- c(NA,x)[1:length(x)]
-    xa <- c(x,NA)[2:(1+length(x))]
-    ext <- (xa-x)>1 | (x-xb)>1
-    sel <- which(is.na(ext)|ext)
-    x.sel <- x[sel]
-    sela <- c(sel,NA)[2:(1+length(sel))]
-    sym <- c("-",",")[as.numeric(sela-sel==1)+1]
-    sym[is.na(sym)] <- ""    
-    out <- paste(rbind(x.sel,sym),collapse="")
-  }
-  return(out)
-}
-
-tz.change <- function(time.in,tz.in="UTC",tz.out="Africa/Algiers") {
-  time.txt <- format(time.in, format="%Y-%m-%d %H:%M:%S")
+tz.change <- function(x,tz.in="UTC",tz.out="Africa/Algiers") {
+  time.txt <- format(x, format="%Y-%m-%d %H:%M:%S")
   time.out <- as.POSIXct(time.txt, tz=tz.in)
   attributes(time.out)$tzone <- tz.out
   return(time.out)

README.md

@@ -6,5 +6,16 @@ Data fusion for air quality data
 ###### Authors
 Lucia Paci, [Giovanni Bonafè](mailto:[email protected])
 
-######Description
+###### Description
 PESCO stands for "Post-processing and Evaluation with Statistical methods of a Chemistry-transport-model Output". The package provides functions to perform data fusion for air quality data, correcting the output of a deterministic CTM with observed data, with a Trans-Gaussian Kriging approach.
+
+###### Installation
+You can install the package by downloading the tarball and (in R)
+```
+install.packages("/path/to/tarball/pesco_0.2.2.tar.gz", repos = NULL, type = "source")
+```
+Otherwise, you can install the latest version directly from GitHub
+```
+require(devtools)
+install_github("jobonaf/pesco")
+```

TODO.md

@@ -3,48 +3,48 @@ TODO
 
 #### Datasets
 
-* ~~observed PM10 (estra_qaria)~~
-* ~~observed NO2 (estra_qaria)~~
-* ~~CTM concentrations (NetCDF, one day)~~
-* ~~emissions (ASCII table)~~
-* ~~elevation (ASCII table)~~
-* ~~population (ASCII table)~~
+- [x] observed PM10 (estra_qaria)
+- [x] observed NO2 (estra_qaria)
+- [x] CTM concentrations (NetCDF, one day)
+- [x] emissions (ASCII table)
+- [x] elevation (ASCII table)
+- [x] population (ASCII table)
 
 #### Essentials
 
-* ~~read data~~
-* ~~prepare data~~
-* ~~kriging~~
-* ~~test the kriging~~
-* ~~all-inclusive function to prepare all you need to process a day~~
-* build source package
+- [x] read data
+- [x] prepare data
+- [x] kriging
+- [x] test the kriging
+- [x] all-inclusive function to prepare all you need to process a day
+- [ ] build source package
 
 #### Extras
 
-* write the output of kriging
-* cross-validation
-* functions to manage annual runs
-* functions to calculate annual indicators
+- [ ] write the output of kriging
+- [ ] cross-validation
+- [ ] functions to manage annual runs
+- [ ] functions to calculate annual indicators
 
 #### Documentation
 
 * doc of functions:
- * ~~aqstat.functions.Rd~~
- * ~~date.functions.Rd~~
- * ~~time.functions.Rd~~
- * ~~char.functions.Rd~~
- * ~~geo.functions.Rd~~
- * read.Rd
- * prepare.Rd
- * daily_synthesis.Rd
- * boxcox.Rd
- * kriging.Rd
- * interpolate.Rd
- * crossval.Rd
-* ~~doc of datasets~~
-* ~~demo for reading data~~
-* ~~demo for preparing data~~
-* ~~demo for kriging~~
-* demo for cross-validation
-* references
-* description of the theory
+ - [x] aqstat.functions.Rd
+ - [x] date.functions.Rd
+ - [x] time.functions.Rd
+ - [x] char.functions.Rd
+ - [x] geo.functions.Rd
+ - [x] read.functions.Rd
+ - [x] prepare.functions.Rd
+ - [x] daily_synthesis.Rd
+ - [x] boxcox.Rd
+ - [x] kriging.Rd
+ - [x] interpolate.Rd
+ - [ ] crossval.Rd
+- [x] doc of datasets
+- [x] demo for reading data
+- [x] demo for preparing data
+- [x] demo for kriging
+- [ ] demo for cross-validation
+- [ ] references
+- [ ] description of the theory