Calculate reference precipitation for a reference period/climate normal

The reference rainfall corresponds to the (usually) 95th or 99th percentile of rainfall within the reference period.

get_reference_precipitation(
  precipitation,
  date = NULL,
  reference_period = NULL,
  percentile = 95L,
  wet_day_threshold = 1
)

Arguments

precipitation	vector with rainfall values
date	date for each rain value. Must be the same lentgth of rainfall and must be the same type as the reference period. I.e if date is numeric (e.g. year) then the reference period must be given as numeric as well. If date is a date or a date-time ("date" or "POSIX.ct") then the reference period must be also a date or a date-time
reference_period	vector of length 2 with the beginning and end of the reference period
percentile	percentile for the reference period. Defaults to the 95th percentile
wet_day_threshold	Numeric. Amount of precipitation at which the day is considered a wet day. Defaults to 1.

Value

the 95th percentile for the reference period.

See also

Other rainfall functions: precipitation_above_reference()

Examples

library(dplyr)
#> 
#> Attaching package: ‘dplyr’
#> The following objects are masked from ‘package:stats’:
#> 
#>     filter, lag
#> The following objects are masked from ‘package:base’:
#> 
#>     intersect, setdiff, setequal, union
# Simulate one measurement of rain per day for 10 years
rainfall <- rlnorm(10*365)
years <- rep(2001:2010, each = 365)
rain_data <- tibble(rainfall = rainfall, year = years)
# We chose a climate normal
climate_normal <- c(2001, 2005)

rain_data %>%
# calculate reference rainfall
  mutate(ref = get_reference_precipitation(rainfall,
                                           year,
                                           climate_normal,
                                           percentile = 95L)) %>%
# calculate prorportion of rainfall above reference
  group_by(year) %>%
  summarise(prop_above = precipitation_above_reference(rainfall, ref))
#> # A tibble: 10 × 2
#>     year prop_above
#>    <int>      <dbl>
#>  1  2001      0.224
#>  2  2002      0.149
#>  3  2003      0.217
#>  4  2004      0.166
#>  5  2005      0.164
#>  6  2006      0.170
#>  7  2007      0.166
#>  8  2008      0.217
#>  9  2009      0.235
#> 10  2010      0.219