Package 'rasterList'

Crop methods for a RasterList-class object.

Description

Crop methods for a RasterList-class object.

Usage

## S4 method for signature 'RasterList'
crop(x, y, check.RasterList = TRUE, ...)

Arguments

x	a valid object
y	a Spatial Object or an Extent
check.RasterList	logical value. If it is TRUE, it checks the x RasterList-class object. Default is FALSE.
...	further arguments

Value

a "cropped" RasterList-class object

Examples

precf <- system.file("map/Mekrou_precipitation.grd", package="rasterList")
prec <- stack(precf)

## Sample L-moments 
if (requireNamespace("lmom",quietly = TRUE)) {
library(lmom)

samlmom <- stack(rasterList(prec,FUN=samlmu))
## Fitting a Random Probability Distribution: it is a 'rasterList' Object
fitdist <- rasterList(samlmom,FUN=pelgam)

##### ZOOM IN 
## set a mask 
mask <-raster( extent(fitdist)/4 )


fitdist_masked <- crop ( x = fitdist,y=mask)
}

---

Is a RasterList-class object ?

Description

Is a RasterList-class object ?

Usage

is.RasterList(x)

Arguments

x	a valid object

Value

a logical variable

Examples

r <- rasterList()
is.RasterList(r)
rr <- raster()
is.RasterList(rr)

f <- system.file("external/test.grd", package="raster")
ra <- rasterList(f)
is.RasterList(rr)

---

Raster methods for a RasterList-class object.

Description

Raster methods for a RasterList-class object.

Usage

## S4 method for signature 'RasterList'
raster(x, FUN = NULL, ...)

Arguments

x	a valid RasterList-class object
FUN	if it not NULL a function is applied to all elements of the list slot in x.
...	further arguments

Value

a RasterLayer-class object

See Also

stack,RasterListApply

Examples

f <- system.file("external/test.grd", package="raster")
ur <- rasterList(raster(f),FUN=function(x,d){x+0:d},d=10)

r1 <- raster(ur)
r2 <- raster(ur,FUN=function(x){x[2]})

---

Creates a RasterList-class object

Description

The method rasterList is the constructor of a RasterList-class from a generic object.

Usage

rasterList(object = NULL, list = NULL, object.name = NA, ...)

as.RasterList(object, ...)

rasterList(object = NULL, list = NULL, object.name = NA, ...)

## S4 method for signature 'RasterLayer'
rasterList(object = NULL, list = NULL, object.name = NA, ...)

## S4 method for signature 'RasterStack'
rasterList(object = NULL, list = NULL, object.name = NA, ...)

## S4 method for signature 'RasterBrick'
rasterList(object = NULL, list = NULL, object.name = NA, ...)

## S4 method for signature 'RasterList'
rasterList(object, list = NULL, object.name = NA, FUN = NULL, ...)

Arguments

object	the object to coerce
list	a list object to assign to the raster map.
object.name	character string containing the name to assign to object.
...	further arguments for raster(generic) or FUN (RasterList-class)
FUN	function that can be used to apply to each element of the list in a RasterList-class

Details

The argument FUN is useful to create or transform RasterList-class from other Raster* classes.

Value

a RasterList-class object.

Examples

f <- system.file("external/test.grd", package="raster")
rr <- rasterList(f) 
rs <- as.RasterList(f)
# The package-provided datasets shall be only used as example datasets. 
precf <- system.file("map/precipitation.grd", package="rasterList")##
## A resampled preciptation raster map based on CHIRS dataset:
## Funk, Chris, Pete Peterson, Martin Landsfeld, Diego Pedreros, James Verdin, 
## Shraddhanand Shukla, Gregory Husak, James Rowland, Laura Harrison, 
## Andrew Hoell and Joel Michaelsen.
## "The climate hazards infrared precipitation with stations - a new environmental 
## record for monitoring extremes". Scientific Data 2, 150066. doi:10.1038/sdata.2015.66 2015. 
## http://chg.geog.ucsb.edu/data/chirps/
##
## Not run: 
if (requireNamespace("lmom",quietly = TRUE) & requireNamespace("lubridate",quietly = TRUE)) {
## Sample L-moments
 library(lmom)
 library(lubridate)
 
prec <- stack(precf)
samlmom <- stack(rasterList(prec,FUN=samlmu))
## Fitting a Random Probability Distribution: it is a 'rasterList' Object
fitdist <- rasterList(samlmom,FUN=pelgam)
}

## End(Not run)

if (requireNamespace("lmom",quietly = TRUE) & requireNamespace("lubridate",quietly = TRUE)) {
library(lmom)
library(lubridate)
precf <- system.file("map/Mekrou_precipitation.grd", package="rasterList")
prec <- stack(precf)
 # Set time
time <- as.Date(names(prec),format="X%Y.%m.%d")
year <- sprintf("X%04d",lubridate::year(time)) ##as.character(time,format="X%Y")

## Compute Annual Precipitation (sum aggregration)
yearlyprec <- stackApply(x=prec,fun=sum,indices=year)
## L-moments
samlmom <- stack(rasterList(yearlyprec,FUN=samlmu))
fitdist <- rasterList(samlmom,FUN=pelgam)

}

---

RasterList-class

Description

Class RasterList

Details

It contains RasterLayer-class with the following adjoint slots:

list:

a list of generic object whese length corresponds to the number of cells. Each list element for each cell;

name:

an identification name of the object. Default is NA.

This class inherits the RasterLayer-class class considering each pixel of the raster is a generic object.

Author(s)

Emanuele Cordano

See Also

raster,Raster-class

Examples

showClass("RasterList")

---

RasterList: A package for Rasters Where Cells are Generic Objects

Description

The aim of this package is to develop a way to make some complex operations on each cells of a Raster Maps. Generally raster contains numeric values in each cells and in each band. Sometimes complex operation required the definition of particular object, in case such operation should be executed for each cell of a raster map, it becomes challanging. Then RasterList-package makes these operations easy to be implented with few lines of codes.

Details

The RasterList-class is a S4 class that inherits the RasterLayer-class and it an added slot called list. The list slot is a list object of so many elements how many are the cells of the inherited RasterLayer-class class, so that there is bijective corrensponce between a raster cell and a list element. The RasterList package provides three categories of important functions:

rasterList: it is the constructor of a RasterList-class object from a generic object, it also transforms a RasterList-class into another one throungh a function argument.

rasterListFun: it is a function that constructs a particular RasterList-class object in which the objects are function-type.

RasterListApply: it is a function that allows to operate among two or more RasterList-class objects defined in the same spatial extent through a generic function fun(x,y,...) where x and y are each cell/element of two RasterList-class objects given as arguments x and y of RasterListApply.

Some examples, concerning the fitting of a probability function for each cell of a stack (RasterStack-class object) of precipitation time-series or the estimation of soil water retention curve for each cell of a raster map, are shown throughout the package manual. The precipitation example dataset in raster format were extracted by the CHIRPS database:

Funk, Chris, Pete Peterson, Martin Landsfeld, Diego Pedreros, James Verdin, Shraddhanand Shukla, Gregory Husak, James Rowland, Laura Harrison, Andrew Hoell and Joel Michaelsen. "The climate hazards infrared precipitation with stations - a new environmental record for monitoring extremes". Scientific Data 2, 150066. doi:10.1038/sdata.2015.66 2015 , https://chc.ucsb.edu/data/chirps.

The package-provided datasets shall be only used as example datasets.

The development of this package has been sponosored by ACEWATER2 and "Water for Growth and Poverty Reduction in the Mekrou" projects of the Joint Research Centre of the Europan Commission (http://aquaknow.jrc.ec.europa.eu).

---

A function for operations among RasterList-class objects.

Description

A function for operations among RasterList-class objects.

Usage

RasterListApply(..., FUN = NULL)

Arguments

...	a set of arguments containg the RasterList-class objects whose lists are operated by FUN.
FUN	a function

Value

a RasterList-class object

See Also

mapply,rasterList

Examples

f <- system.file("external/test.grd", package="raster")

ra <- rasterList(f)
rb <- rasterList(f)

rm <- RasterListApply(x=ra,y=rb,z=10,FUN=function(x,y,z){x+y+z})

### Fitting a probability distribution for precipitation 
### in each cell with "lmon" package (L Moments) 

if (requireNamespace("lmom",quietly = TRUE)) {
library(lmom)


# The package-provided datasets shall be only used as example datasets. 
precf <- system.file("map/Mekrou_precipitation.grd", package="rasterList") ## 
## A resampled preciptation raster map based on CHIRS dataset:
## Funk, Chris, Pete Peterson, Martin Landsfeld, Diego Pedreros, James Verdin, 
## Shraddhanand Shukla, Gregory Husak, James Rowland, Laura Harrison, 
## Andrew Hoell and Joel Michaelsen.
## "The climate hazards infrared precipitation with stations -a new environmental 
## record for monitoring extremes". Scientific Data 2, 150066. doi:10.1038/sdata.2015.66 2015. 
## http://chg.geog.ucsb.edu/data/chirps/
##
prec <- stack(precf)
## Sample L-moments 
samlmom <- stack(rasterList(prec,FUN=samlmu))
## Fitting a Random Probability Distribution: it is a 'rasterList' Object
fitdist <- rasterList(samlmom,FUN=pelgam)

## KS TESTING

kstesting <- RasterListApply(x=rasterList(prec),y="cdfgam",para=fitdist,FUN=ks.test)

## Mapping of p-value 
pval_ks <- raster(kstesting,FUN=function(x){x$p.value})

}

---

Execution of the elements of a RasterList

Description

This fuction transmors a generic RasterList-class object into another RasterList-class object where elemets are all function-type.

Usage

rasterListFun(object)

Arguments

object

an object to be coerced to RasterList-class

Value

This function works with RasterList-class objects in which all elements of object@list slot are functions. It returns a "global" function that works at "raster" scale. The returned function will have the following usage signature: fun(xval,...) where one xval (if its lengths is different from 1) element is the applied to each element and ... are further common arguments.

Examples

library(sp)
library(rasterList)
library(soilwater)
set.seed(1234)
data(meuse.grid)
data(meuse)
coordinates(meuse.grid) <- ~x+y
coordinates(meuse) <- ~x+y
gridded(meuse.grid) <- TRUE


soilmap <- stack(meuse.grid)[['soil']]
elevmap <- rasterize(x=meuse,y=soilmap,field="elev",fun=mean)
soilparcsv <- system.file("external/soil_data.csv",package="soilwater")
soilpar <- read.table(soilparcsv,stringsAsFactors=FALSE,header=TRUE,sep=",")
## From help(meuse,help_type="html")
##soil type according to the 1:50 000 soil map of the Netherlands. 
## 1 = Rd10A (Calcareous weakly-developed meadow soils, light sandy clay); 
## 2 = Rd90C/VII (Non-calcareous weakly-developed meadow soils, heavy sandy clay to light clay); 
## 3 = Bkd26/VII (Red Brick soil, fine-sandy, silty light clay)
soiltype_id <- c(1,2,3)
soiltype_name <- c("sandy clay","sandy clay","silty clay loam")

meuse.soilrasterlist <- rasterList(soilmap,FUN=function(i,soiltype_name,soilpar){
			
		o <- NULL
		if (!is.na(i)) {
			ii <- which(soilpar$type==soiltype_name[i])	
		    o <- soilpar[ii,]				
			type <- o[["type"]]
			o <- o[names(o)!="type"]
			o <- o[names(o)!="Ks_m_per_hour"]
			names(o)[names(o)=="Ks_m_per_sec"] <- "ks"
			names(o)[names(o)=="swc"] <- "theta_sat"
			names(o)[names(o)=="rwc"] <- "theta_res"
			attr(o,"type") <- type
			## add noise
			noise <- rnorm(length(o))
			o <- o*(1+0.005*noise)
				
			o["m"] <- 1-1/o["n"]
				
				
		} else {
				
			o <- soilpar[which(soilpar$type==soiltype_name[1]),]
			type <- o[["type"]]
			o <- o[names(o)!="type"]
			o <- o[names(o)!="Ks_m_per_hour"]
			names(o)[names(o)=="Ks_m_per_sec"] <- "ks"
			names(o)[names(o)=="swc"] <- "theta_sat"
			names(o)[names(o)=="rwc"] <- "theta_res"
			o[] <- NA
		}
			
		return(o)
},soiltype_name=soiltype_name,soilpar=soilpar)


meuse.swclist <- rasterList(meuse.soilrasterlist,FUN=function(x) {
			
			o <- NA
##			swc       rwc   alpha       n         m           ks
##			9 0.4295507 0.1093227 3.39387 1.39617 0.2837546 2.018317e-07
		
			
			o <- function(psi,...,func="swc"){
				
				args <- c(list(psi=psi,...),as.list(x))
				oo <- do.call(args=args,what=get(func))
				return(oo)
				
			}
			
			
			
			
			
			
			
			return(o)
			
		})


### RasterList with soil water retenction curves (One for each cell!) 

swcfunr <- rasterListFun(meuse.swclist)

## RasterLayer of soil water content assuming a uniformly distrrubted pressure head 
psi <- -0.9
soil_water_content <- raster(swcfunr(psi))
plot(soil_water_content)

## RasterLayer of soil water content from  a generic map of soil water pressure head
psi <- 0.2-(elevmap-(5))
psi[] <- -0.9+0.1*rnorm(ncell(psi[])) ## Alternatively to the values of the previous line!
soil_water_content <- raster(swcfunr(psi))
plot(soil_water_content)

## END

---

Creates a RasterStack-class object from a RasterList-class

Description

The method transforms a RasterList-class into a RasterStack-class in case of the list elements are numeric vectors.

Usage

## S4 method for signature 'RasterList'
stack(x, ...)

Arguments

x	a rasterList-class object
...	further arguments for rasterList

Value

a RasterStack-class object

See Also

rasterList

Examples

f <- system.file("external/test.grd", package="raster")

## Creates a simple generic RasterList

rl <- rasterList(f) 

list <- as.list(as.vector(rl))
list <- lapply(X=list,FUN=function (x) {c(x,x+10,x+15)}) 

rl <- rasterList(rl,list=list,object.name="test")

ss <- stack(rl)

il <- 8331
list[[il]] <- numeric(0)
rla <- rasterList(rl,list=list,object.name="test2")
sa <- stack(rla)

---