Skip to contents

Apply manual sampling bias to occurrences via a grid

Source: R/apply_manual_sampling_bias.R
apply_manual_sampling_bias.Rd

This function adds a sampling bias weight column to an sf object containing occurrences. The sampling probabilities are based on bias weights within each cell of a provided grid layer.

Usage

apply_manual_sampling_bias(occurrences_sf, bias_weights)

Arguments

occurrences_sf

An sf object with POINT geometry representing the occurrences.

bias_weights

An sf object with POLYGON geometry representing the grid with bias weights. This sf object should contain a bias_weight column and a geometry column. Higher weights indicate a higher probability of sampling. Weights must be numeric values between 0 and 1 or positive integers, which will be rescaled to values between 0 and 1.

Value

An sf object with POINT geometry that includes a bias_weight column containing the sampling probabilities based on the sampling bias.

See also

Other detection: apply_polygon_sampling_bias()

Examples

# Load packages
library(sf)
#> Linking to GEOS 3.12.1, GDAL 3.8.4, PROJ 9.4.0; sf_use_s2() is TRUE
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
library(ggplot2)

# Create polygon
plgn <- st_polygon(list(cbind(c(5, 10, 8, 2, 3, 5), c(2, 1, 7, 9, 5, 2))))

# Get occurrence points
occurrences_sf <- simulate_occurrences(plgn)
#> [using unconditional Gaussian simulation]

# Create grid with bias weights
grid <- st_make_grid(
    plgn,
    n = c(10, 10),
    square = TRUE) %>%
  st_sf()
grid$bias_weight <- runif(nrow(grid), min = 0, max = 1)

# Calculate occurrence bias
occurrence_bias <- apply_manual_sampling_bias(occurrences_sf, grid)
occurrence_bias
#> Simple feature collection with 32 features and 3 fields
#> Geometry type: POINT
#> Dimension:     XY
#> Bounding box:  xmin: 2.608597 ymin: 1.500189 xmax: 9.700949 ymax: 8.662383
#> CRS:           NA
#> First 10 features:
#>    time_point sampling_p1 bias_weight                  geometry
#> 24          1   0.5803610   0.4212345  POINT (8.62278 1.500189)
#> 11          1   0.4470324   0.9998340 POINT (9.700949 1.557037)
#> 30          1   0.9976994   0.7985516 POINT (7.259084 2.571956)
#> 4           1   0.8868568   0.6641659  POINT (7.990637 1.90122)
#> 31          1   0.8853797   0.6641659 POINT (8.206332 1.997754)
#> 13          1   0.7385367   0.8321967  POINT (8.577629 1.94152)
#> 25          1   0.9993741   0.3226733 POINT (9.315453 1.811303)
#> 2           1   0.6696007   0.3915279 POINT (6.044068 2.804665)
#> 14          1   0.9832881   0.3915279  POINT (6.70267 3.335745)
#> 23          1   0.7326365   0.2751708 POINT (7.802674 3.113804)

# Visualise where the bias is
ggplot() +
  geom_sf(data = plgn) +
  geom_sf(data = grid, alpha = 0) +
  geom_sf(data = occurrence_bias, aes(colour = bias_weight)) +
  geom_sf_text(data = grid, aes(label = round(bias_weight, 2))) +
  theme_minimal()