createIntamapObject {intamap}R Documentation

Create an object for interpolation within the intamap package

Description

This is a help function for creating an object (see intamap-package to be used for interpolation within the intamap-package

Usage

createIntamapObject(observations, obsChar, formulaString,
            predictionLocations=100, targetCRS, boundaries, boundaryLines,
            intCRS, params=list(), boundFile, lineFile, class="idw",
            outputWhat, blockWhat = "none",...)

Arguments

observations a SpatialPointsDataFrame, SpatialPixelsDataFrame, SpatialGridDataFrame, SpatialLinesDataFrame or SpatialPolygonsDataFrame with observations. Note that there are only few methods that can actually handle interpolation of observations with a support
obsChar list with observation characteristics, used by some interpolation methods
formulaString formula that defines the dependent variable as a linear model of independent variables; suppose the dependent variable has name z, for ordinary and simple kriging use the formula z~1; for universal kriging, suppose z is linearly dependent on x and y, use the formula z~x+y. The formulaString defaults to "value~1" if value is a part of the data set. If not, the first column of the data set is used.
predictionLocations either a Spatial* object with prediction locations or an integer with the requested number of prediction locations. If boundaries are supported, the sampled prediction locations will be sampled within the boundaries
targetCRS the wanted projection for the interpolated map
boundaries SpatialPolygonsDataFrame with the boundaries of regions in the prediction region
boundaryLines SpatialPointsDataFrame with the boundaries between pairs of regions discretized as points. Will be read from file if lineFile is given or will be created from boundaries if not.
intCRS a particular projection requested for the interpolation
params parameters for the interpolation, given as exceptions to the default parameters set in the function getIntamapParams
boundFile Filename where boundaries can be found, e.g. a shapefile
lineFile Filename where paired points on boundaries can be found
class setting the class(es) of the object, see intamap-package
outputWhat List defining the requested type of output. Parameters:
mean = TRUE
Usual kriging prediction
variance = TRUE
Usual kriging error
quantile
The estimated quantile for a certain threshold
excprob
Exceedance probability for a certain threshold
cumdistr
The cumulative distribution for a certain value
MOK
Assumed unbiased prediction using the MOK method for the threshold given. See unbiasedKrige
IWQSEL
Assumed unbiased prediction using the IWQSEL method for the threshold given. See unbiasedKrige
...
Additional prediction types that do not exist in the default parameter set, particularly parameters necessary for new methods within the intamap-package.
The list defaults to list (mean = TRUE) for objects of class IDW and list(mean=TRUE, variance = TRUE) for all other objects.
blockWhat List defining particular output for block predictions. These include:
blockMax
logical; whether to predict maximum within block, if block predictions
blockMin
logical; whether to predict mimimum within block, if block predictions
fat
Prediction of area within block above a threshold (fat = threshold
blockMaxVar
logical; whether to predict the variance of the prediction of max within the block, similarly it is possible to set blockMinVar = TRUE and fatVar = threshold
...
  • Either: other elements that can be used by particular interpolation methods. These are added to the object as named elements.
  • Or: elements that have been created in earlier calls to one of the functions in the intamap-package, and that are not supposed to change in the second call. By adding these elements to the object in createIntamapObject, they can be reused without having to re-estimate them. Typical examples are the elements created from a call to preProcess

Details

This function is a help function for creating an object (see intamap-package) for interpolation within the intamap-package. The function uses some default values if certain elements are not included.

If createIntamapObject is called without predictionLocations, or if a number is given, the function will sample a set of predictionLocations. These will be sampled from a regular grid.

targetCRS and intCRS are not mandatory variables, but are recommended if the user wants predictions of a certain projection. intCRS is not necessary if the targetCRS is given and has a projection (is not lat-long). It is recommended to include the argument intCRS if all projected elements are lat-long, as many of the interpolation methods do not work optimal with lat-long data.

The ...-argument can be used for arguments necessary for new methods not being a part of the intamap-package. It is also a method for reusing previously calculated elements that can be assumed to be unchanged for the second interpolation.

Value

An object with observations, prediction locations, parameters and possible additional elements for automatic interpolation. The object will have class equal to the value of argument class, and methods in the intamap-package will dispatch on the object according to this class.

Author(s)

Jon Olav Skoien

References

http://www.intamap.org/

See Also

intamap-package and getIntamapParams

Examples

library(intamap)

# set up data:
data(meuse)
coordinates(meuse) = ~x+y
meuse$value = log(meuse$zinc)
data(meuse.grid)
gridded(meuse.grid) = ~x+y
proj4string(meuse) = CRS("+init=epsg:28992")
proj4string(meuse.grid) = CRS("+init=epsg:28992")

# set up intamap object:
idwObject = createIntamapObject(
        observations = meuse,
        predictionLocations = meuse.grid,
        targetCRS = "+init=epsg:3035",
        class = "idw"
)

[Package intamap version 1.3-3 Index]