Gouldian Finch

Photo by Kym Nicolson. Licensed under CC BY 4.0.

1 Introduction

This vignette demonstrates how to analyze occurrence data for Gouldian Finch in Australia, using records from the Atlas of Living Australia (ALA).

The dataset has been prepared for you to explore, making it suitable for both study and practice with real-world ecological data. In this vignette we provide short examples of how to manipulate and visualize the dataset, but you are encouraged to develop your own creative approaches for analysis and visualization.

This is the glimpse of your data :

library(dplyr)
library(ecotourism)
data("gouldian_finch")
gouldian_finch |> glimpse()
Rows: 3,922
Columns: 14
$ obs_lat     <dbl> -38.28428, -35.54665, -35.41374, -33.76889, -33.76889, -33…
$ obs_lon     <dbl> 141.6337, 138.8717, 149.0669, 151.1119, 151.1119, 151.1119…
$ date        <date> 2015-01-25, 2021-01-03, 2016-12-13, 2014-04-24, 2014-04-2…
$ time        <chr> "15:15:00", "15:01:16", "10:30:00", "00:00:00", "00:00:00"…
$ year        <dbl> 2015, 2021, 2016, 2014, 2014, 2014, 2017, 2017, 2017, 2014…
$ month       <dbl> 1, 1, 12, 4, 4, 4, 1, 2, 2, 12, 5, 4, 4, 4, 4, 5, 4, 6, 6,…
$ day         <dbl> 25, 3, 13, 24, 24, 24, NA, 17, 9, 10, 28, 27, 26, 17, 23, …
$ hour        <int> 15, 15, 10, 0, 0, 0, 0, 15, 14, 13, 15, 6, 6, 6, 6, 6, 8, …
$ weekday     <ord> Sunday, Sunday, Tuesday, Thursday, Thursday, Thursday, Sun…
$ dayofyear   <dbl> 25, 3, 348, 114, 114, 114, 1, 48, 40, 344, 148, 118, 117, …
$ sci_name    <chr> "Chloebia gouldiae", "Chloebia gouldiae", "Chloebia gouldi…
$ record_type <chr> "HUMAN_OBSERVATION", "HUMAN_OBSERVATION", "HUMAN_OBSERVATI…
$ obs_state   <chr> "Victoria", "South Australia", "Australian Capital Territo…
$ ws_id       <chr> "948280-99999", "946770-99999", "949250-99999", "957650-99…

2 Visualization

2.1 Spatial Distribution Map

Distribution of Occurrence Gouldian Finch Sightings in Australia

library(ggplot2)
library(ggthemes)

gouldian_finch |> 
  ggplot() +
    geom_sf(data = oz_lga) +
    geom_point(aes(x = obs_lon, y = obs_lat), color = "red") +
    theme_map()

Keep in mind that the natural range of the Gouldian Finch is in northern Australia. Occasional records from the south may reflect birds in captivity, such as those observed in zoos, rather than wild populations.

Weekday Distribution of Gouldian Finch Sightings

week_order <- c("Monday", "Tuesday", "Wednesday", "Thursday", "Friday", "Saturday", "Sunday")

gouldian_finch |> 
  ggplot(aes(x = factor(weekday, levels = week_order))) +
    geom_bar() +
    labs(x = "Weekday", y = "Number of Records") +
    theme_minimal()

Monthly Distribution of Gouldian Finch Sightings

library(lubridate)

gouldian_finch |>
    dplyr::mutate(month = month(month, label = TRUE, abbr = TRUE)) |>
    ggplot(aes(x = factor(month))) +
    geom_bar() +
    labs(x = "Month", y = "Number of Records") +
    theme_minimal()

Yearly Distribution of Gouldian Finch Sightings

gouldian_finch |>
    ggplot(aes(x = factor(year))) +
    geom_bar() +
    labs(x = "Year", y = "Number of Records")+
    theme_minimal()

4 Relational visualization

We want to see if gouldian_finch occurrences are related to precipitation on the same day from the weather dataset.

Here’s a short R script that:

  1. Joins gouldian_finch with weather using ws_id and date.

  2. Counts daily occurrences.

  3. Plots precipitation vs number of gouldian_finch sightings.

library(ggbeeswarm)

# Prepare gouldian_finch occurrence counts per day
gouldian_finch_daily <- gouldian_finch |>
  group_by(ws_id, date) |>
  summarise(occurrence = n(), .groups = "drop")

# Join with weather data for precipitation
gouldian_finch_weather <- gouldian_finch_daily |>
  left_join(weather |> select(ws_id, date, prcp), 
            by = c("ws_id", "date"))

gouldian_finch_weather |>
  filter(!is.na(prcp)) |>
  mutate(rain = if_else(prcp > 5, "yes", "no")) |>
  ggplot(aes(x = rain, y = occurrence)) +
  geom_quasirandom(alpha = 0.6) +
  ylim(c(0, 15)) +
  labs(
    title = "Relationship between rainy day and gouldian finch occurrence",
    x = "Rainy",
    y = "Number of Gouldian Finch records"
  ) +
  theme_minimal()

gouldian_finch_weather <- gouldian_finch_daily |> 
  left_join(
    weather |> select(ws_id, date, temp, prcp),
    by = c("ws_id", "date")
  )


ggplot(gouldian_finch_weather, aes(temp, occurrence, color = prcp)) +
  geom_point(alpha = 0.5) +
  scale_color_viridis_c() +
  labs(
    title = "Gouldian Finch occurrence vs temperature, colored by precipitation",
    x = "Mean daily temperature (°C)",
    y = "Occurrences",
    color = "Precipitation (mm)"
  ) +
  theme_minimal()