Tourism data

The ecotourism package contains complementary tourism datasets at quarterly tourism-region and monthly SA2 resolutions. This vignette introduces the monthly stopover reason and activity datasets, their code-name lookup tables, and the geographic metadata used to map them.

1 Data overview

tourism_reason contains monthly domestic overnight-trip estimates classified by the reason for the stopover.

data("tourism_reason")
glimpse(tourism_reason)

Rows: 196,044
Columns: 6
$ year         <int> 2014, 2014, 2014, 2014, 2014, 2014, 2014, 2014, 2014, 201…
$ month        <ord> January, January, January, January, January, January, Jan…
$ sa2_code_tra <int> 11009, 11009, 11014, 11014, 11015, 11015, 11016, 11016, 1…
$ stop_reason  <int> 1, 2, 1, 2, 1, 2, 1, 3, 1, 2, 6, 8, 1, 1, 2, 3, 1, 1, 2, …
$ trips        <dbl> 4.490612, 14.718867, 1.340082, 6.860708, 6.368905, 12.523…
$ n_sample     <int> 2, 3, 1, 2, 3, 3, 12, 1, 10, 8, 2, 2, 2, 3, 2, 1, 1, 7, 1…

tourism_activity contains the corresponding estimates classified by activity.

data("tourism_activity")
glimpse(tourism_activity)

Rows: 320,188
Columns: 6
$ year          <int> 2014, 2014, 2014, 2014, 2014, 2014, 2014, 2014, 2014, 20…
$ month         <ord> January, January, January, January, January, January, Ja…
$ sa2_code_tra  <int> 11009, 11009, 11009, 11009, 11014, 11014, 11015, 11015, …
$ stop_activity <int> 1, 3, 5, 6, 6, 7, 1, 2, 6, 1, 2, 3, 6, 1, 2, 3, 5, 6, 7,…
$ trips         <dbl> 15.6321730, 2.1298693, 2.1298693, 19.2094791, 6.8607083,…
$ n_sample      <int> 2, 2, 2, 5, 2, 1, 2, 2, 5, 10, 8, 2, 11, 13, 7, 1, 5, 19…

Both datasets cover 2014–2024. Trip estimates are expressed in thousands, and n_sample gives the survey sample supporting each estimate. Rows with zero trips or zero n_sample have been removed.

2 Add descriptive category names

The compact integer category fields can be decoded with tourism_reason_name and tourism_activity_name.

data("tourism_reason_name")
data("tourism_activity_name")

tourism_reason_name

# A tibble: 7 × 2
  stop_reason name                          
        <int> <chr>                         
1           1 Holiday                       
2           2 Visiting friends and relatives
3           3 Business                      
4           6 Other reason                  
5           7 No other reason               
6           8 In transit                    
7           9 Not stated/not asked

tourism_activity_name

# A tibble: 8 × 2
  stop_activity name                                  
          <int> <chr>                                 
1             1 Outdoor / nature                      
2             2 Active outdoor / sports               
3             3 Arts / heritage                       
4             5 Local attractions / tourist activities
5             6 Social activities                     
6             7 None of these                         
7             8 Not asked                             
8             9 Other activities

Join the lookup table appropriate to the detailed dataset:

reason_named <- tourism_reason |>
  left_join(tourism_reason_name, by = "stop_reason")

activity_named <- tourism_activity |>
  left_join(tourism_activity_name, by = "stop_activity")

reason_named |>
  count(year, name, wt = trips, name = "trips_thousands") |>
  arrange(year, desc(trips_thousands)) |>
  head()

# A tibble: 6 × 3
   year name                           trips_thousands
  <int> <chr>                                    <dbl>
1  2014 Holiday                                 43487.
2  2014 Visiting friends and relatives          31788.
3  2014 Business                                18879.
4  2014 Other reason                             5243.
5  2014 In transit                               1061.
6  2015 Holiday                                 44444.

3 Add tourism-region metadata

The detailed datasets share sa2_code_tra with tourism_region. The packaged region table stores this identifier as character, so harmonise the type before joining.

data("tourism_region")

region_lookup <- tourism_region |>
  mutate(sa2_code_tra = as.integer(sa2_code_tra)) |>
  distinct(sa2_code_tra, state, region, sa2, lon, lat, ws_id)

reason_with_region <- reason_named |>
  left_join(region_lookup, by = "sa2_code_tra")

reason_with_region |>
  select(year, month, state, region, sa2, name, trips, n_sample) |>
  head()

# A tibble: 6 × 8
   year month   state           region          sa2         name  trips n_sample
  <int> <ord>   <chr>           <chr>           <chr>       <chr> <dbl>    <int>
1  2014 January New South Wales Capital Country Queanbeyan  Holi…  4.49        2
2  2014 January New South Wales Capital Country Queanbeyan  Visi… 14.7         3
3  2014 January New South Wales Snowy Mountains Cooma       Holi…  1.34        1
4  2014 January New South Wales Snowy Mountains Cooma       Visi…  6.86        2
5  2014 January New South Wales Snowy Mountains Cooma Surr… Holi…  6.37        3
6  2014 January New South Wales Snowy Mountains Cooma Surr… Visi… 12.5         3

The same region_lookup can be joined to tourism_activity. The ws_id column can subsequently be joined to weather_stations when weather context is needed.

4 SA2 polygons and representative points

oz_sa2 provides simplified 2021 SA2 polygons for every TRA code represented by the detailed tourism data. Its lat and lon columns provide one representative interior point per SA2 when point geometry is more convenient.

data("oz_sa2")
glimpse(oz_sa2)

Rows: 2,350
Columns: 4
$ sa2_code_tra <int> 11007, 11008, 11009, 11010, 11012, 11013, 11014, 11015, 1…
$ lat          <dbl> -35.51037, -35.37553, -35.35231, -35.36064, -35.37693, -3…
$ lon          <dbl> 149.7535, 149.2337, 149.2253, 149.2515, 149.2015, 149.068…
$ geometry     <MULTIPOLYGON [°]> MULTIPOLYGON (((149.7607 -3..., MULTIPOLYGON…

Join an SA2 summary directly to the polygon table for mapping:

reason_by_sa2 <- tourism_reason |>
  summarise(
    trips_thousands = sum(trips),
    .by = sa2_code_tra
  )

reason_map <- oz_sa2 |>
  left_join(reason_by_sa2, by = "sa2_code_tra")

reason_map |>
  select(sa2_code_tra, lat, lon, trips_thousands) |>
  head()

Simple feature collection with 6 features and 4 fields
Geometry type: MULTIPOLYGON
Dimension:     XY
Bounding box:  xmin: 148.3952 ymin: -37.26275 xmax: 150.2157 ymax: -34.99963
Geodetic CRS:  GDA2020
# A tibble: 6 × 5
  sa2_code_tra   lat   lon trips_thousands                              geometry
         <int> <dbl> <dbl>           <dbl>                    <MULTIPOLYGON [°]>
1        11007 -35.5  150.          748.   (((149.7607 -35.08347, 149.7577 -35.…
2        11008 -35.4  149.           19.2  (((149.2192 -35.3601, 149.2148 -35.3…
3        11009 -35.4  149.         1120.   (((149.2315 -35.34224, 149.2179 -35.…
4        11010 -35.4  149.            7.63 (((149.2316 -35.34221, 149.2359 -35.…
5        11012 -35.4  149.           56.6  (((149.2065 -35.34586, 149.2045 -35.…
6        11013 -36.9  149.          229.   (((148.422 -36.88885, 148.5236 -36.9…

5 Recover omitted zero rows

Use tidyr::complete() to restore the full monthly panels; combinations absent from the packaged data are assigned zero trips and n_sample.

all_sa2 <- union(tourism_reason$sa2_code_tra, tourism_activity$sa2_code_tra)
months <- factor(month.name, levels = month.name, ordered = TRUE)

tourism_reason_full <- tourism_reason |>
  tidyr::complete(year = 2014:2024, month = months, sa2_code_tra = all_sa2,
                  stop_reason = tourism_reason_name$stop_reason,
                  fill = list(trips = 0, n_sample = 0))

tourism_activity_full <- tourism_activity |>
  tidyr::complete(year = 2014:2024, month = months, sa2_code_tra = all_sa2,
                  stop_activity = tourism_activity_name$stop_activity,
                  fill = list(trips = 0, n_sample = 0))