Build a stress-test series by ordering whole years — rh_scenarios_from

Reorders the years of a precipitation series by their annual total to create increasing (dry to wet) or decreasing (wet to dry) sequences, useful to stress-test a system against runs of dry years. Note: partial boundary years (a series starting or ending mid-year) are included as-is, so their totals reflect only the observed days; trim the series to whole years first if that matters for the ordering.

Usage

rh_scenarios_from_years(
  data,
  order = c("increasing", "decreasing", "asis"),
  value_col = "value",
  date_col = "date"
)

Arguments

data: A data frame with a date column and a value column.
order: "increasing" (dry to wet), "decreasing" (wet to dry) or "asis" (chronological).
value_col, date_col: Column names. Defaults "value" and "date".

Value

A data frame with columns step, year, date, value (years concatenated in the requested order). Feed precip = result$value to rh_simulate().

Examples

s <- rh_scenarios_from_years(precip_pi, order = "increasing")
head(s)
#>   step year       date value
#> 1    1 1919 1919-01-01   0.0
#> 2    2 1919 1919-01-02   8.2
#> 3    3 1919 1919-01-03   0.0
#> 4    4 1919 1919-01-04   0.0
#> 5    5 1919 1919-01-05   0.0
#> 6    6 1919 1919-01-06   0.0