acquireDirectoryStructure: Acquire structure of input directory

In geanders/futureheatwaves: Find, Characterize, and Explore Extreme Events in Climate Projections

Description

This function parses the structure of the user-provided directory of climate projection files to create a list of the climate models and ensemble members included.

Usage

acquireDirectoryStructure(dataFolder, coordinateFilenames, tasFilenames,
  timeFilenames, models_to_run, dataDirectories, threshold_ensemble,
  thresholdBoundaries)

Arguments

dataFolder	Character string with pathway to a directory with climate projection data. This directory must have a specific structure– see the futureheatwaves vignette for guidance on setting up this directory.
coordinateFilenames	Character string the with filename of each grid point location file. This filename should be identical for all ensemble member subdirectories included in the dataFolder directory. See the package vignette for an example of the required structure for this file.
tasFilenames	Character string the with filename of each climate projection file. This filename should be identical for all ensemble member subdirectories included in the dataFolder directory. See the package vignette for an example of the required structure for this file.
timeFilenames	Character string the with filename of each projection dates file. This filename should be identical for all ensemble member subdirectories included in the dataFolder directory. See the package vignette for an example of the required structure for this file.
models_to_run	A character vector with either "all" (the default), in which case the function runs through all models in the dataFolder directory, or the names of the models to run, using the names of each model's subdirectory within the data directory (e.g., c("bcc1", "ccsm")).
dataDirectories	A list object, with two elements, one for each of the two subdirectories included in the main directory. Typically, these will be separate directories of historical and projection experiments from climate models. Each element of the list should be named with the name of the subdirectory and should provide a numeric vector with the starting and ending years of the data within each of the two subdirectories (e.g., list("historical" = c(1990, 1999), "rcp85" = c(2060, 2079)) for a dataFolder with historical experiment data for 1990 to 1999 and RCP8.5 projections for 2060 to 2079).
threshold_ensemble	A character vector giving the name of the ensemble member that should be used when determining the city-specific threshold temperatures for each climate model (e.g., "r1i1p1"). This threshold is used for relative extreme event definitions. See the futureheatwaves vignette for more on extreme event definitions. If any climate model lacks that ensemble member for the specified dates for calculating the threshold, it will be excluded from the processing.
thresholdBoundaries	A numeric vector with the custom time boundaries to be used to determine the threshold variable values for the extreme event definition. The required format for this vector is c(start year, end year), with the restriction that bounds must be contained within the time boundaries of one of the two experiment subdirectories specified by the dataDirectories argument in gen_hw_set. The default value is 1990 to 1999.

Value

The function returns a list object outlining the file structure of the dataFolder directory.

Note

Projection, grid locations, and projection time files within the dataFolder directory must be comma-separated (.csv) files and must be named using the names specified by the arguments coordinateFilenames, tasFilenames, and timeFilenames. See the futureheatwaves vignette for more information about setting up the dataFolder directory.

geanders/futureheatwaves documentation built on May 17, 2019, 12:14 a.m.