Auto-grouping and charting

library(DivInsight)
library(dplyr)
#> 
#> Attaching package: 'dplyr'
#> The following objects are masked from 'package:stats':
#> 
#>     filter, lag
#> The following objects are masked from 'package:base':
#> 
#>     intersect, setdiff, setequal, union
data("Colombia")

To gain a more detailed insight of the diversity in a large area the group_radius argument can be used in the clusterise_sites() function to automatically group sites together by specifying a radius in metres.

A label is assigned to each site denoting which group it belongs to. This label can be seen in the site_group column in the cluster dataframe.

# subset the dataframe by province name 
Colombia_Meta <- subset(Colombia, stateProvince == "Meta")

# cluster occurrence data by date and generate centred coordinates for each site
clusterised_Meta_20km_groups <- clusterise_sites(
  dataframe = Colombia_Meta,
  cluster_min_length = 30,
  
  group_radius = 20000
)

# view the information for each site/cluster
print(clusterised_Meta_20km_groups[[2]])
#>    longitude latitude site_group       date
#> 1  -73.46702 4.058576          1 2000-10-01
#> 2  -73.46702 4.058576          1 2000-10-06
#> 3  -73.47267 4.073333          1 2015-10-13
#> 4  -73.47267 4.073333          1 2015-11-19
#> 5  -73.47267 4.073333          1 2016-02-15
#> 6  -73.47267 4.073333          1 2016-02-24
#> 7  -73.47267 4.073333          1 2016-05-24
#> 8  -73.47267 4.073333          1 2016-06-13
#> 9  -73.47267 4.073333          1 2016-08-03
#> 10 -73.36824 4.083131          1 2022-11-09
#> 11 -73.38409 4.080567          1 2022-11-11
#> 12 -73.37865 4.091503          1 2022-11-14
#> 13 -71.71407 4.322192          4 2012-05-01
#> 14 -72.96847 4.295741          5 2013-02-10
#> 15 -72.96847 4.295741          5 2013-02-11
#> 16 -72.96847 4.295741          5 2013-02-12
#> 17 -72.96847 4.295741          5 2013-02-13
#> 18 -72.97906 4.287842          5 2013-02-15
#> 19 -72.97054 4.288258          5 2013-02-16
#> 20 -72.97906 4.287842          5 2013-02-17
#> 21 -72.97466 4.285456          5 2013-02-18
#> 22 -72.96518 4.273713          5 2013-02-21
#> 23 -72.96660 4.272737          5 2013-02-22
#> 24 -72.96733 4.273370          5 2013-02-23
#> 25 -72.97047 4.275363          5 2013-02-24
#> 26 -72.96566 4.458016          6 2013-03-03
#> 27 -72.96566 4.458016          6 2013-03-04
#> 28 -72.96566 4.458016          6 2013-03-05
#> 29 -72.96566 4.458016          6 2013-03-06
#> 30 -72.96038 4.401621          6 2013-03-08
#> 31 -72.96413 4.403583          6 2013-03-09
#> 32 -72.96038 4.401621          6 2013-03-10
#> 33 -72.96038 4.401621          6 2013-03-11
#> 34 -72.97800 4.445970          6 2013-03-13
#> 35 -72.97800 4.445970          6 2013-03-14
#> 36 -72.97800 4.445970          6 2013-03-15
#> 37 -72.97800 4.445970          6 2013-03-16
#> 38 -71.81433 4.530222          8 2015-09-16
#> 39 -71.81433 4.530222          8 2015-10-13
#> 40 -71.81433 4.530222          8 2015-11-19
#> 41 -71.81433 4.530222          8 2016-02-15
#> 42 -73.89669 3.798917          9 2015-11-19
#> 43 -73.90230 3.799197          9 2016-02-20
#> 44 -73.58081 4.077551         10 2016-03-21
#> 45 -73.61314 4.094075         10 2016-04-12
#> 46 -72.02722 4.171914         11 2017-05-22
#> 47 -73.40181 3.531730         12 2017-08-28
#> 48 -73.40181 3.531730         12 2019-04-08
#> 49 -71.43081 3.814218         13 2021-06-26
#> 50 -71.39631 3.821191         13 2021-06-27
#> 51 -71.52242 3.857269         13 2021-06-29
#> 52 -73.58722 3.884813         14 2021-07-07

Base R commands can be used to see how many sites there are in each group and help us decide which group(s) to examine further.

# use table() and sort() to see how many sites there are in each group
clusterised_Meta_20km_groups[[2]]$site_group %>% table %>% sort %>% print
#> .
#>  4 11 14  9 10 12 13  8  1  5  6 
#>  1  1  1  2  2  2  3  4 12 12 12

Here we choose to examine sites 1, 5, 6, and 8 further by using filter_groups_by_number() to store each group’s data then plot using plot_sites_scatter_H().

# store the data for each group
Colombia_Meta_1 <- filter_groups_by_number(clusterised_Meta_20km_groups, 1)
Colombia_Meta_5 <- filter_groups_by_number(clusterised_Meta_20km_groups, 5)
Colombia_Meta_6 <- filter_groups_by_number(clusterised_Meta_20km_groups, 6)
Colombia_Meta_8 <- filter_groups_by_number(clusterised_Meta_20km_groups, 8)

# plot the data in a scatter plot
plot_sites_scatter_H(Colombia_Meta_1, main = "Shannons H at Meta #1") 
#> Warning: Use of `stats_dataframe$date` is discouraged.
#> ℹ Use `date` instead.
#> Warning: Use of `stats_dataframe$H` is discouraged.
#> ℹ Use `H` instead.
plot_sites_scatter_H(Colombia_Meta_5, main = "Shannons H at Meta #5") 
plot_sites_scatter_H(Colombia_Meta_6, main = "Shannons H at Meta #6") 
plot_sites_scatter_H(Colombia_Meta_8, main = "Shannons H at Meta #8")

The charts show varying patterns of diversity over available time frames but it is important to know where these changes are taking place.

map_start() can be used to create a new map object and map_add() can be used to add coordinate information.

# create a new map using group 1's coordinates
Colombia_Meta_map <- 
  
  map_start(
    
    clusterised_object = Colombia_Meta_1,
    site_name = "Meta #1",
    colour = "green"
  )

# add group 5's coordinates to the map object 
Colombia_Meta_map <-
  
  map_add(
    
    existing_map = Colombia_Meta_map,
    
    clusterised_object = Colombia_Meta_5,
    site_name = "Meta #5",
    colour = "purple"
  )

# add group 6's coordinates to the map object
Colombia_Meta_map <-
  
  map_add(
    
    existing_map = Colombia_Meta_map,
    
    clusterised_object = Colombia_Meta_6,
    site_name = "Meta #6",
    colour = "blue"
  )

# add group 8's coordinates to the map object
Colombia_Meta_map <-
  
  map_add(
    
    existing_map = Colombia_Meta_map,
    
    clusterised_object = Colombia_Meta_8,
    site_name = "Meta #8",
    colour = "red"
  )

Once the map has been created it can be viewed.

# view the map
Colombia_Meta_map