Skip to contents

Download and Aggregate Monthly Climate Data (Single Station)

Source: R/dl_monthly_single.R
smn_dl_monthly_single.Rd

Downloads daily SMN data for a station using smn_dl_daily_single(), then aggregates it to monthly values for one or more variables applying a completeness rule per month. The function can return either a long table (one row per month and a column per requested variable), or reduce tables (one per variable, with YEAR, JAN..DEC, and annual summaries).

Usage

smn_dl_monthly_single(
  station,
  variables = "all",
  start_date = "1961-01-01",
  end_date = Sys.Date(),
  max_missing_frac = 0.2,
  output_format = c("reduce", "long"),
  aggregator = NULL,
  csv_file = NULL,
  daily_del_na = "no"
)

Arguments

station

Character or numeric. SMN station code (passed to smn_dl_daily_single()).

variables

Character vector of variables to aggregate, or "all". Valid names are "prec", "evap", "tmax", "tmin". Default "all".

start_date

Start date (inclusive). Coerced with as.Date(). Default "1961-01-01". Passed to smn_dl_daily_single().

end_date

End date (inclusive). Coerced with as.Date(). Default Sys.Date(). Passed to smn_dl_daily_single().

max_missing_frac

Numeric in [0, 1]. If the fraction of missing days in a month exceeds this threshold, the monthly value is set to NA. Default 0.2.

output_format

Output format. Either "long" or "reduce" (default "reduce").

aggregator

Optional single function to apply to all variables (e.g., median) or a named list mapping variable → function, e.g., list(prec = sum, tmax = mean). If NULL, defaults are used: sum for prec,evap and mean for tmax,tmin.

csv_file

Optional path to write CSV output.

  • For "long": a single CSV at csv_file.

  • For "reduce" with multiple variables: one file per variable, using paste0(file_path_sans_ext(csv_file), "_", var, ".csv").

daily_del_na

Character "yes" or "no"; forwarded to smn_dl_daily_single() to control row removal at the daily stage. Default "no". Note: the monthly completeness rule is applied after downloading daily data; this argument is only about cleaning rows at the daily function.

Value

  • If output_format = "long": a single data.frame with columns station, latitude, longitude, altitude, date, <vars...>.

  • If output_format = "reduce" and a single variable was requested: a wide data.frame.

  • If output_format = "reduce" and multiple variables were requested: a named list of wide data.frames (one per variable).

Details

Long format

If output_format = "long", the result contains one row per month with columns: station, latitude, longitude, altitude, date, <vars...>, where <vars...> are the requested subset among prec, evap, tmin, tmax. The date column is the first day of each month.

Reduce format

If output_format = "reduce", the result is a wide table (or a named list of tables when multiple variables are requested) with columns: YEAR, JAN..DEC, ACUM, PROM, MONTHS. For sum-type variables (prec, evap) ACUM is the row sum; for mean-type variables (tmax, tmin) ACUM is the row mean. PROM is always the row mean across valid months; MONTHS is the count of non-NA months.

Daily data are obtained via smn_dl_daily_single() with output_format = "full" (to carry station metadata). Missing fraction is computed against the calendar number of days per month within the requested window; absent daily rows count as missing. The date in the long table is set to the first day of each month (YYYY-MM-01).

Examples

if (FALSE) { # \dontrun{
# 1) All variables, monthly, long format (one row per month)
m_long <- smn_dl_monthly_single(
  station = "15101",
  variables = "all",
  start_date = "1984-01-01",
  end_date   = "2020-12-31",
  output_format = "long"
)

# 2) Only precipitation, reduce format (wide)
m_prec <- smn_dl_monthly_single("15101", variables = "prec", output_format = "reduce")

# 3) Two variables, strict completeness (<= 10% missing), custom aggregator
m_custom <- smn_dl_monthly_single(
  station = "15101",
  variables = c("tmax","tmin"),
  max_missing_frac = 0.10,
  aggregator = median,
  output_format = "long"
)

# 4) Control daily cleaning from here (do not drop NAs before monthly)
m_cfg <- smn_dl_monthly_single(
  station = "15101",
  variables = "all",
  daily_del_na = "no",
  output_format = "long"
)
} # }