Extract the maximum available weather observation closest to a
specified target time within a given time window.
If no valid observations are found, return NA.
Usage
get_weather_obs(
data,
target_column,
timestamp_column,
target_hour,
target_minute,
time_window
)Arguments
- data
A data frame containing weather observations.
- target_column
The meteorological variable to retrieve.
- timestamp_column
The column containing timestamps of the observations.
- target_hour
An integer specifying the target hour (0-23) for retrieving observations.
- target_minute
An integer specifying the target minute (0-59) for retrieving observations.
- time_window
An integer specifying the allowable time window (in minutes) around the target time.
References
Davis, Robert E., Owen Himmel, Parker K. Sims, and Christopher M. Fuhrmann. “Mortality and Diurnal Temperature Range in Virginia.” International Journal of Biometeorology, January 30, 2025. https://doi.org/10.1007/s00484-025-02850-6
Examples
df <- data.frame(
timestamp = (c("2024-03-01 06:50:00",
"2024-03-01 06:53:00",
"2024-03-01 07:10:00",
"2024-03-01 07:30:00",
"2024-03-01 07:55:00")),
temperature = c(15, NA, 16, NA, 17)
)
# retrieve the closest temperature observation around 07:00 within a 30-minute window
get_weather_obs(df,
target_column = temperature,
timestamp_column = timestamp,
target_hour = 7,
target_minute = 0,
time_window = 30)
#> [1] 16
# Note: Use `!!rlang::sym()` if `target_column` or `timestamp_column`
# is provided as a character string.
get_weather_obs(df,
target_column = "temperature",
timestamp_column = timestamp,
target_hour = 7,
target_minute = 0,
time_window = 30)
#> [1] NA
get_weather_obs(df,
target_column = !!rlang::sym("temperature"),
timestamp_column = timestamp,
target_hour = 7,
target_minute = 0,
time_window = 30)
#> [1] 16