DTRIndicator {ClimProjDiags}

Diurnal temperature range indicator (DTR) of multidimensional arrays

Description

This function computes the diurnal temperature indicator, defined as the number of days where the diurnal temperature variation exceeds the vulnerability threshold (defined as the mean(tmax -tmin) + 5 from the reference period).

Usage

DTRIndicator(
  tmax,
  tmin,
  ref,
  by.seasons = TRUE,
  dates = NULL,
  timedim = NULL,
  calendar = NULL,
  ncores = NULL
)

Arguments

tmax

A numeric multidimensional array containing daily maximum temperature.

tmin

A numeric multidimensional array containing daily minimum temperature. This array must be the same dimensions as tmax parameter.

ref

An output list from the DTRRef function with the same dimensions as parameters tmax and tmin, except the time dimension, containing the mean diurnal temperature variation for the reference period.

by.seasons

If TRUE (by default), the DTR is computed for each season (December-January-February, March-April-May, June-July-August and September-October-November) seperately. If FALSE is specified, the montly mean DTR is computed.

dates

A vector of dates with a calendar attributes. If NULL (by default), the 'time' attributes of parameter 'tmax' and 'tmin' are considered.

timedim

An integer number indicating the position of the time dimension in the parameters tmax and tmin. If NULL (by default), the dimension called 'time' in parameter tmax and tmin is considered as time dimension.

calendar

A character indicating the calendar type.

ncores

The number of cores to be used when computing the index.

Value

A list of length 3:

• $dtr.ref, an array with the same dimensions as the input data, but with the time dimension reduce from daily to monthly or seasonal resolution depending on the selected resolution in by.season.

• $year, a vector of the corresponding years.

• $season, a vector of the seasons or months corresponding to the resolution selected in by.season.

Examples

##Exmaple with synthetic data:
tmax <- 1 : (2 * 3 * 730 * 1)
dim(tmax) <- c(lon = 2, lat = 3, time = 730, model = 1)
tmin <- (1 : (2 * 3 * 730 * 1)) - 1
dim(tmin) <- c(lon = 2, lat = 3, time = 730, model = 1)
time <- seq(as.POSIXct("1900-01-01 12:00:00", tz = "", 
                      format = "%Y-%d-%m %H:%M:%S"), 
           as.POSIXct("1901-31-12 18:00:00", tz = "", 
                      format = "%Y-%d-%m %H:%M:%S"), "day")
time <- as.POSIXct(time, tz = "CET")
metadata <- list(time = list(standard_name = 'time', long_name = 'time', 
                            calendar = 'noleap',
                            units = 'days since 1970-01-01 00:00:00', 
                            prec = 'double', 
                            dim = list(list(name ='time', unlim = FALSE))))
attr(time, "variables") <- metadata
attr(tmax, 'Variables')$dat1$time <- time
attr(tmax, 'Variables')$common[[2]]$dim[[3]]$len = length(time)
attr(tmax, 'Variables')$common[[2]]$dim[[3]]$vals <- time
attr(tmin, 'Variables')$dat1$time <- time
attr(tmin, 'Variables')$common[[2]]$dim[[3]]$len = length(time)
attr(tmin, 'Variables')$common[[2]]$dim[[3]]$vals <- time
a <- DTRRef(tmax, tmin, by.seasons = FALSE, ncores = NULL)

aa <- DTRIndicator(tmax, tmin, ref = a, by.seasons = FALSE, ncores = NULL)
str(aa)
dim(aa$indicator)