analyze_weather: Analysis of winter conditions based on meteorological...

In bniebuhr/winteranalyzer: WInTer conditions and CHaracterization for reindeer husbandry

Description

The function analyze_weather calculates some proxies for weather conditions based on time series of snow depth, precipitation, and minimum, maximum, and average temperature for a given year. The functions n_days, n_events, and duration_events calculate the number of days, number of events, and duration of events of type 3 and 4 based on the output of the function analyze_weather.

Usage

analyze_weather(
  date,
  snow_depth,
  precipitation,
  temp_min,
  temp_max,
  temp_avg,
  start = c("first_permanent_snow", "first_date", "other date")[1],
  plot_first_snow = FALSE,
  set_start = NULL,
  ev3_temp_min_thr = -2,
  ev3_temp_max_thr = +2,
  ev3_prec_thr = 3,
  ev4_temp_avg_day = +1,
  ev4_temp_avg_next_day = -1,
  ev4_prec_thr = 3,
  evX_temp_max_thr_day = +1,
  evX_temp_min_thr_next_day = -2,
  evX_number_days_after = 3,
  evX_prec_thr = 3,
  ...
)

n_days(weather_analyzed, event = 3, first_half = TRUE)

n_events(weather_analyzed, event = 3, first_half = TRUE)

duration_event(weather_analyzed, event = 3, first_half = TRUE)

Arguments

date	date object. vector. Vector of dates in a time series of a given winter.
snow_depth	numeric. vector. Daily snow depth, in mm.
precipitation	numeric. vector. Precipitation accumulated each day, in mm.
temp_min	numeric. vector. Minimum daily temperature, in Celsius.
temp_max	numeric. vector. Minimum daily temperature, in Celsius.
temp_avg	numeric. vector. Average daily temperature, in Celsius.
start	character. Whether the analysis should be done from the "first_permanent_snow", from the "first_date" of the time series, or from "other_date". In the first case, it uses the function find_date_permanent_snow to search for when the snow comes and stays. In the last case, an initial date must be set by the parameter set_start.
plot_first_snow	logical. Whether or not to plot the time series of snow and the estimated start for analyzing the time series
set_start	POSIX. Starting date of the analysis, in case start = "other_date".
...	other parameters for the functions find_date_permanent_snow, identify_winter_event3, or identify_winter_event4.
weather_analyzed	list. output list from the function analyze_weather.
event	numeric. Which type of event (3 or 4).
first_half	logical. Whether to get events from the whole winter (FALSE) or only from the first half of the winter (TRUE), when the effects of these events may be worse.

Value

A list with three elements:

weather_indices

a data.frame with the original data plus time series of some indices calculated: cumumative precipitation, the ratio between snow and cumulative precipitation, the ratio between cumulative precipitation and snow, the difference between cumulative precipitation and snow, and when events 3 and 4 happened.

events3

a list with information about events 3. See the output from the function identify_winter_event3.

events4

a list with information about events 4. See the output from the function identify_winter_event4.

References

Lundqvist, H., Norell, L., Danel, O. 2007. Multivariate characterisation of environmental conditions for reindeer husbandry in Sweden. Rangifer, 27(1): 5-23.

Examples

library(tidyverse)
data(weather_mittadalen)

# select one year
weather <- weather_mittadalen %>%
  dplyr::filter(year == 2015)

# analyze weather
weather_condition <-
  with(weather, analyze_weather(date, snow_depth, prec, temp_min, temp_max,
                                temp_avg, start = "first_permanent_snow",
                                plot_first_snow = T))
weather_condition

n_days(weather_condition, event = 4)
n_events(weather_condition, event = 3)
n_events(weather_condition, event = 3, first_half = F)

# plot
weather_condition$weather_indices %>%
  ggplot(aes(date)) +
  geom_line(aes(y = snow_depth, color = "snow depth")) +
  geom_line(aes(y = cumulative_precitation, color = "cum prec")) +
  geom_line(aes(y = 100*prec_snow_ratio, color = "100*cum prec/snow depth")) +
  geom_hline(yintercept = 100, linetype = 2) +
  ylim(0, 250) +
  labs(x = "Date", y = "Amount (mm)", color = "")

# plot with events
weather_condition %>%
  plot_weather(term = c("snow_depth", "cum", "prec", "temp_max")) +
  geom_vline(xintercept = weather_condition$events3$event_dates, linetype = 2)

weather_condition %>%
  plot_weather(term = c("snow_depth", "cum", "prec", "temp_max"),
               add_events = c("events3", "eventsX"), first_half = F)


# for several years
n_weather <- weather_mittadalen %>%
  tidyr::nest(data = c(date, snow_depth, prec, temp_min, temp_max, temp_avg)) %>%
  dplyr::mutate(
    winter_conditions = purrr::map(data, ~ with(., analyze_weather(date, snow_depth, prec, temp_min, temp_max, temp_avg))),
    n_days3 = purrr::map(winter_conditions, ~ n_days(., event = 3)),
    n_days4 = purrr::map(winter_conditions, ~ n_days(., event = 4)),
    n_events3 = purrr::map(winter_conditions, ~ n_events(., event = 3)),
    n_events4 = purrr::map(winter_conditions, ~ n_events(., event = 4)),
    duration3 = purrr::map(winter_conditions, ~ duration_event(., event = 3)),
    duration4 = purrr::map(winter_conditions, ~ duration_event(., event = 4))
  ) %>%
  tidyr::unnest(n_days3:n_events4)

n_weather %>%
  dplyr::select(year, n_days3:n_events4) %>%
  tidyr::pivot_longer(cols = n_days3:n_events4, names_to = "what", values_to = "val") %>%
  ggplot(aes(x = year, y = val, color = what)) +
  geom_line()

bniebuhr/winteranalyzer documentation built on Dec. 31, 2020, 8:56 p.m.