defStormsDataset: Creating a 'stormsDataset' object

In StormR: Analyzing the Behaviour of Wind Generated by Tropical Storms and Cyclones

Creating a stormsDataset object

Description

The defStormsDataset() function creates a stormsDataset object from either a NetCDF or a CSV file. This is an essential first step before other stormR functions can be used.

Usage

defStormsDataset(
  filename = system.file("extdata", "test_dataset.nc", package = "StormR"),
  sep = NULL,
  fields = c(names = "name", seasons = "season", isoTime = "iso_time", lon = "usa_lon",
    lat = "usa_lat", msw = "usa_wind", basin = "basin", rmw = "usa_rmw", pressure =
    "usa_pres", poci = "usa_poci"),
  basin = NULL,
  seasons = c(1980, as.numeric(format(Sys.time(), "%Y"))),
  unitConversion = c(msw = "knt2ms", rmw = "nm2km", pressure = "mb2pa", poci = "mb2pa"),
  notNamed = "NOT_NAMED",
  verbose = 1
)

Arguments

filename	character. Name of the NetCDF (.nc)/CSV (.csv) file. Default is the test_dataset.nc file located in the inst/extdata repository of the directory (accessible by system.file("extdata", "test_dataset.nc", package = "StormR")). This test dataset is extracted from the IBTrACS.SP.v04r00.nc file and provides all the tropical cyclones that occurred around Vanuatu from 2015 to 2016 and around New Caledonia from 2020 to 2021.
sep	character. The field separator character if filename is a CSV file. Default value is set to NULL which will set the separator to ",".
fields	named character vector. This argument allows to specify the corresponding variable names (NetCDF) or headers (CSV) in the input file for each field in the output stormsDataset. By default, the corresponding variable names are set up to import data from a NetCDF file from the IBTrACS database (Knapp et al., 2010). Corresponding variable names for following fields have to (mandatory fields) or can be (recommended or optional fields) provided: "names", names of the storms (mandatory), "seasons", years of observations (mandatory), "isoTime", date and time of observations (mandatory), "lon", longitude of the observations (mandatory), "lat", latitude of the observations (mandatory), "msw", maximum sustained wind speed (mandatory), "basin", name of the area where the storm originated (recommended), "rmw", radius of maximum winds: distance between the centre of the storm and its band of strongest winds (recommended), "pressure", central pressure (recommended), "poci", pressure of the last closed isobar (recommended)
basin	character. If the basin field is provided, then storm track data will only be extracted for the named basin. By default basin=NULL, meaning that all storms irrespective of the basin they originated in are extracted. Seven basins can be used to filter the data set: "NA", for North Atlantic basin, "SA", for South Atlantic basin, "EP", for Eastern North Pacific basin, "WP", for Western North Pacific basin, "SP", for South Pacific basin, "SI", for South India basin, or "NI", for North India basin.
seasons	numeric vector. Seasons of occurrence of the storms (e.g., c(2020,2022)). In the Southern Hemisphere, the cyclone season extends across two consecutive years. Therefore, to capture the 2021 to 2022 cyclone season both years should be specified, with cyclones assigned for the year that originated in. By default all storms occurring since 1980 are extracted.
unitConversion	named character vector. StormR functions use the metric system (international system of units), therefore msw has to be provided in m.s^{-1}, rmw in km, pressure and poci in Pa. By default unitConversion=c(msw = "knt2ms", rmw = "nm2km", pressure = "mb2pa", poci = "mb2pa") to meet the requirements when importing a NetCDF file from the IBTrACS database. This argument is mandatory even if no conversion is needed. If no conversion is needed then use "None" in the corresponding fields. The following unit conversions are implemented: For msw, "knt2ms", to convert knot to meter per second (default setting), "kmh2ms", to convert kilometre per hour to meter per second, "⁠mph2ms"⁠, to convert miles per hour to meter per second, or "None", if no conversion is needed. For rmw, "nm_to_ms"to convert nautical miles to kilometre (default setting), or "None"if no conversion is needed. For pressure and poci, "⁠mb2pa"⁠, to convert millibar to Pascal (default setting), "b2pa", to convert bar to Pascal, "atm2pa", to convert atmosphere to Pascal, "psi2pa", to convert psi to Pascal, or "None", if no conversion is needed.
notNamed	character. Constant name for not named storms to remove in the database. Default value is "NOT_NAMED" (IBTrACS database)
verbose	numeric. Whether the function should display (⁠= 1⁠) or not (⁠= 0⁠) information about the processes.

Value

The defStormsDataset() function returns a stormsDataset object.

References

Knapp, K. R., Kruk, M. C., Levinson, D. H., Diamond, H. J., & Neumann, C. J. (2010). The International Best Track Archive for Climate Stewardship (IBTrACS). Bulletin of the American Meteorological Society, 91(3), Article 3. https://doi.org/10.1175/2009bams2755.1

Examples

# Creating a `stormsDataset` object with storms between 2010 and 2015
# in the South Pacific using the NetCDF provided with the package
SP_2015_2020_nc <- defStormsDataset(seasons = c(2010, 2015))
str(SP_2015_2020_nc)

# Creating a `stormsDataset` object with storms between 2010 and 2015
# in the South Pacific using the CSV provided with the package
fileName <- system.file("extdata", "test_dataset.csv", package = "StormR")
SP_2015_2020_csv <- defStormsDataset(seasons = c(2010, 2021))
str(SP_2015_2020_csv)

StormR documentation built on Sept. 12, 2024, 6:52 a.m.