percent.days.op.threshold: Lengths of strings of TRUE values
In climdex.pcic: PCIC Implementation of Climdex Routines

Description Usage Arguments Details Value Note See Also Examples

Computes fraction of days above or below the baseline threshold for each day, and averages them using the date factor passed in.

1
2
3

percent.days.op.threshold(temp, dates, jdays, date.factor,
  threshold.outside.base, base.thresholds, base.range, op = "<",
  max.missing.days)

`temp`	Sequence of temperature values.
`dates`	Sequence of associated dates.
`jdays`	Sequence of associated days of year.
`date.factor`	Factor to aggregate data using.
`threshold.outside.base`	Sequence of thresholds to be used for data outside the base period.
`base.thresholds`	Data structure containing sets of thresholds to be used inside the base period; see climdexInput-class.
`base.range`	Date range (type PCICt) of the baseline period.
`op`	Comparison operator to use.
`max.missing.days`	Maximum number of NA values per time period.

This function computes fractions of days above or below baseline thresholds for each day, then aggregates them using date.factor. It is used to implement TN/TX 10/90p.

A vector consisting of the mean fraction of days above or below the supplied set of thresholds.

If date.factor is omitted, daily series will be returned.

climdexInput-class.

library(PCICt)

## Parse the dates into PCICt.
tmax.dates <- as.PCICt(do.call(paste, ec.1018935.tmax[,c("year",
"jday")]), format="%Y %j", cal="gregorian")
tmin.dates <- as.PCICt(do.call(paste, ec.1018935.tmin[,c("year",
"jday")]), format="%Y %j", cal="gregorian")
prec.dates <- as.PCICt(do.call(paste, ec.1018935.prec[,c("year",
"jday")]), format="%Y %j", cal="gregorian")

## Load the data in.
ci <- climdexInput.raw(ec.1018935.tmax$MAX_TEMP,
ec.1018935.tmin$MIN_TEMP, ec.1018935.prec$ONE_DAY_PRECIPITATION,
tmax.dates, tmin.dates, prec.dates, base.range=c(1971, 2000))

## Compute monthly tx90p.
tx90p <- percent.days.op.threshold(ci@data$tmax, ci@dates, ci@jdays,
                                   ci@date.factors$monthly,
                                   ci@quantiles$tmax$outbase$q90,
                                   ci@quantiles$tmax$inbase$q90,
                                   ci@base.range, ">",
                                   ci@max.missing.days['monthly']) *
         ci@namasks$monthly$tmax