seas (version 0.6-0)

sdsm: Read and write from SDSM

Description

Reads and writes the data format used in SDSM's DAT and OUT extensions.

Usage

# reading
read.sdsm(file, start = 1961, end = 2000, calendar)
# writing
write.sdsm(dat, var, start, end, file = "")

Arguments

file

name of DAT or OUT file

dat

data.frame of variables to be written

start

starting year

end

ending year

var

name of variable to be written from dat

calendar

calender used for data; if unspecified, this is assumed to be proleptic Gregorian (normal); however, for CCCma models this should be "365_day", and for Hadley models this should be "360_day"; see year.length

Value

read.sdsm returns a data.frame of the measured variables. The variables are named V1…Vn, for n ensembles.

If a calendar is specified, this is stored as an attribute in the date data frame column.

Details

This function readings and writes climate data with the Statistical Downscaling Model, or SDSM. The model uses DAT extensions for input data, such as daily observations of mean temperature, and OUT extensions for modeled output.

References

Wilby, R.L., Dawson, C.W. and Barrow, E.M. 2002. SDSM --- a decision support tool for the assessment of regional climate change impacts, Environmental Modelling Software, 17 (2), 145--157, 10.1016/S1364-8152(01)00060-3

SDSM can be downloaded free-of-charge for Windows platforms from https://www.sdsm.org.uk/

CGCM1 and HADCM3 model data for SDSM can be downloaded from the Canadian Climate Impacts and Scenarios website: https://web.archive.org/web/20120218192015/http://www.cics.uvic.ca/scenarios/sdsm/select.cgi (archived)

See Also

read.msc, change

Examples

# NOT RUN {
# reading
fname <- system.file("extdata", "GF_2050s_precip.OUT", package="seas")
gf50 <- read.sdsm(fname)
gf50.ss <- seas.sum(gf50, var=paste("V", 1:20, sep=""), name="Grand Forks")

# analysis
image(gf50.ss, var="V1")
image(gf50.ss, var="V2")
image(gf50.ss, var="V3")

# writing
data(mscdata)
hj <- mksub(mscdata, id=2100630)
fname <- paste(tempdir(), "HJ_Obs_prcp.DAT", sep="/")
write.sdsm(hj, "precip", 1961, 2000, fname)
# }