Skip to contents

Summarise operational days by user-defined seasons

Source: R/cam_summarise_op_by_season.R
cam_summarise_op_by_season.Rd

Takes the output from cam_get_op_days() and sums the number of operational days per season. Seasons are defined by Julian day start cutoffs and can be 1..n segments that wrap the year.

Usage

cam_summarise_op_by_season(
  calendar_df,
  grouping = c("project", "project_id", "location", "location_id"),
  seasons = c(spring = 99, summer = 143, winter = 288),
  labels = NULL,
  date_col = "date",
  operating_col = "operating",
  na_as = FALSE,
  by_year = FALSE,
  wide = TRUE
)

Arguments

calendar_df

A tibble/data.frame with at least date (Date) and operating (logical/NA), typically the output of cam_get_op_days().

grouping

Character vector of grouping columns to keep in the summary. Defaults to c("project","project_id","location","location_id"); only columns present are used. If none are present, aggregation is global.

seasons

Season starts. Named numeric, unnamed numeric + labels, or a data frame with season and start columns. Defaults to c(spring = 99, summer = 143, winter = 288).

labels

Optional labels if seasons is an unnamed numeric vector.

date_col

Name of the date column in calendar_df. Default "date".

operating_col

Name of the logical operating column. Default "operating".

na_as

Logical: treat NA in operating as TRUE when counting? Default FALSE.

by_year

Logical: also summarise by calendar year? Default FALSE.

wide

Logical: return one column per season (TRUE) or a long table with season and operating_days (FALSE). Default TRUE.

Value

A tibble. In wide mode: grouping cols (+ year if by_year), one column per season (counts), and total_days (row sum). In long mode: grouping cols (+ year), season, operating_days, and total_days.

Details

Season definitions (seasons) can be supplied in any of these forms:

  • Named numeric vector (recommended): e.g., c(spring = 99, summer = 143, winter = 288).

  • Unnamed numeric + labels=: e.g., seasons = c(99, 143, 288), labels = c("spring","summer","winter").

  • Data frame with columns season (labels) and start (Julian day).

Rules:

  • Starts must be integers in 1..366, unique, and will be sorted.

  • Classification is circular: the last season runs to day 366, then wraps to the first start.

Grouping: By default the function looks for c("project","project_id","location", "location_id) but only use the ones actually present in calendar_df. If none are present, the result is aggregated across all records.

Operating NA handling: By default operating = NA is counted as FALSE (na_as = FALSE). Set na_as = TRUE to count unknown days as operational.

See also

cam_get_op_days

Author

Marcus Becker

Examples

if (FALSE) { # \dontrun{
cal <- cam_get_op_days(image_reports,
                       grouping = c("project_id","location","location_id"),
                       span = "operational", missing_as = FALSE)

# Default seasons, by year, wide format
sum1 <- cam_summarise_op_by_season(cal, by_year = TRUE, wide = TRUE)

# Two-season example, long format
sum2 <- cam_summarise_op_by_season(
  calendar_df = cal,
  seasons     = c(IceFree = 120, FreezeUp = 305),
  by_year     = FALSE,
  wide        = FALSE
)
} # }