metaclipR.Regridding: Directed metadata graph construction for Transformations of...

In Predictia/metaclipR: A climate4R extension for METACLIP provenance description

Description

Build a directed metadata graph describing a regridding Transformation on a climate4R grid

Usage

metaclipR.Regridding(graph, package = "transformeR", version = "1.3.2",
  fun = "interpGrid", arg.list = NULL)

Arguments

graph	A previous metaclipR data structure from which the current step follows
package	package
version	version
fun	function name. Unused (set to "interpGrid")
arg.list	Argument list. See details

Details

This function takes as reference the semantics defined in the Data Source and Transformation ontology defined in the Metaclip Framework (http://metaclip.predictia.es/).

Argument list

The following list of arguments is required to define an aggregation:

• new.coordinates

• method

• bilin.method

The different arguments are explained in the the help page of interpGrid.

Author(s)

D. San Martín, J. Bedia

References

Climate4R page at University of Cantabria metaclip.predictia.es

See Also

Other transformation: metaclip.graph.Command, metaclipR.Aggregation, metaclipR.AnomalyCalculation, metaclipR.Climatology, metaclipR.Dataset, metaclipR.Ensemble, metaclipR.etccdi

Examples

require(transformeR)
require(igraph)
pkg <- "transformeR"
v <- "1.1.1"
# Assume a given hindcast DatasetSubset: 
data("CFS_Iberia_hus850")
DS <- subsetGrid(CFS_Iberia_hus850, members = 1:3, years = 1989:1991)
graph <- metaclipR.DatasetSubset(package = pkg,
                                 version = v,
                                 arg.list = list(members = 1:3,
                                                 years = 1989:1991),
                                 fun = "subsetGrid",
                                 output = "DS")
# Data are regridded to the regular EOBS 0.25 grid
data("EOBS_Iberia_tas")
ref.grid <- getGrid(EOBS_Iberia_tas)
# We apply the fast 'akima' interpolator
out <- interpGrid(DS,
                  new.coordinates = ref.grid,
                  method = "bilinear",
                  bilin.method = "akima")
plotClimatology(climatology(out), backdrop.theme = "coastline")
# This is how metadata is encoded:
# 1.) We identify the origin node from which the first transformation hangs:
graph$parentnodename
# 2.) Argument list
arg.list <- list("new.coordinates" = ref.grid,
                 "method" = "bilinear",
                 "bilin.method" = "akima")
# 3.) metaclipR.Regridding is called:
graph <- metaclipR.Regridding(package = pkg,
                              version = v,
                              graph = graph,
                              fun = "interpGrid",
                              arg.list = arg.list)
# This is the graph structure containing the metadata:
plot(graph$graph)

Predictia/metaclipR documentation built on May 14, 2019, 6:07 a.m.