Compare the connectivity of two scenarios

We can measure the connectivity of a given habitat and barrier with habitat_connectivity(). We can also compare the connectivity, say for example if you have the same area habitat and barrier, but you want to understand what the change in connectedness is when you remove, or add some habitat, or some barrier(s). This function helps you do that, by comparing a "scenario" connectivity object against a "baseline" connectivity object (both created by habitat_connectivity() or summarise_connectivity()).

Usage

compare_connectivity(scenario, baseline, scenario_name = NULL)

Arguments

scenario

A connectivity object (from habitat_connectivity() or summarise_connectivity()). The scenario to compare against baseline. Must be a single row.

baseline

A connectivity object (from habitat_connectivity() or summarise_connectivity()). The reference the scenario is compared against. Must be a single row, and match scenario on species, interpatch_distance, and resolution.

scenario_name

Character. An optional label for the scenario, for example "Bentley Project". Appears in the scenario_name column on every row. Defaults to NULL, which gives NA.

Value

A compare_connectivity object: a tibble with four rows (baseline, scenario, change, pct_change, in the measure column) and the same metric columns as summarise_connectivity() output: scenario_name, measure, species, interpatch_distance, n_patches, effective_mesh_ha, prob_connectedness, patch_area_mean, patch_area_total_ha, and data_resolution. The change row is scenario - baseline, so a positive value means the scenario is higher than the baseline; the pct_change row expresses that same change as 100 * change / baseline, which is the readable form for metrics whose absolute deltas are very small. Metric values are held at full precision - they are not rounded - so change is exact.

Details

If you have raster or vector layers rather than connectivity objects, use habitat_connectivity_comparison(), which runs habitat_connectivity() on each scenario for you and then calls this function.

See also

habitat_connectivity_comparison(), which starts from habitat and barrier layers instead of connectivity objects, habitat_connectivity(), and summarise_connectivity().

Examples

# build `connectivity` objects cheaply (no spatial pipeline) and compare them
baseline <- summarise_connectivity(lizard_areas_connected)
# a scenario in which one connected patch (the first row) is lost
scenario <- summarise_connectivity(lizard_areas_connected[-1, ])
compare_connectivity(scenario = scenario, baseline = baseline)
#> # Connectivity comparison: baseline / scenario / change / pct_change
#> # A tibble: 4 × 10
#>   scenario_name measure    species             interpatch_distance n_patches
#>   <chr>         <chr>      <chr>                             <dbl>     <dbl>
#> 1 NA            baseline   Blue-tongued Lizard                  50     73   
#> 2 NA            scenario   Blue-tongued Lizard                  50     72   
#> 3 NA            change     Blue-tongued Lizard                  50     -1   
#> 4 NA            pct_change Blue-tongued Lizard                  50     -1.37
#>   effective_mesh_ha prob_connectedness patch_area_mean patch_area_total_ha
#>               <dbl>              <dbl>           <dbl>               <dbl>
#> 1         4.47                1.70e- 5         3600.               26.3   
#> 2         4.47                1.70e- 5         3648.               26.3   
#> 3        -0.0000113          -4.28e-11           47.6              -0.0172
#> 4        -0.000252           -2.52e- 4            1.32             -0.0654
#>   data_resolution
#>   <chr>          
#> 1 2x2            
#> 2 2x2            
#> 3 2x2            
#> 4 2x2            
#> # change = scenario - baseline (positive = scenario is higher)
#> # pct_change = 100 * change / baseline

# label the scenario so it can be told apart from others
compare_connectivity(
  scenario = scenario,
  baseline = baseline,
  scenario_name = "Bentley Project"
)
#> # Connectivity comparison: baseline / scenario / change / pct_change
#> # A tibble: 4 × 10
#>   scenario_name   measure    species             interpatch_distance n_patches
#>   <chr>           <chr>      <chr>                             <dbl>     <dbl>
#> 1 Bentley Project baseline   Blue-tongued Lizard                  50     73   
#> 2 Bentley Project scenario   Blue-tongued Lizard                  50     72   
#> 3 Bentley Project change     Blue-tongued Lizard                  50     -1   
#> 4 Bentley Project pct_change Blue-tongued Lizard                  50     -1.37
#>   effective_mesh_ha prob_connectedness patch_area_mean patch_area_total_ha
#>               <dbl>              <dbl>           <dbl>               <dbl>
#> 1         4.47                1.70e- 5         3600.               26.3   
#> 2         4.47                1.70e- 5         3648.               26.3   
#> 3        -0.0000113          -4.28e-11           47.6              -0.0172
#> 4        -0.000252           -2.52e- 4            1.32             -0.0654
#>   data_resolution
#>   <chr>          
#> 1 2x2            
#> 2 2x2            
#> 3 2x2            
#> 4 2x2            
#> # change = scenario - baseline (positive = scenario is higher)
#> # pct_change = 100 * change / baseline