generate detection range models Create logistic curve for detection efficiency with depth extrapolation

Generates a logistic detection efficiency model that varies with both distance and depth. The function interpolates d50 and d95 parameters linearly between minimum and maximum depths, creating a 3D detection efficiency surface.

Usage

create_logistic_curve_depth(
  min_depth,
  max_depth,
  d50_min_depth,
  d95_min_depth,
  d50_max_depth,
  d95_max_depth,
  dist_step = 5,
  depth_step = 1,
  color_option = "magma",
  plot = TRUE,
  return_model = TRUE,
  return_object = TRUE
)

Arguments

min_depth

Numeric. Minimum depth in meters for the model.

max_depth

Numeric. Maximum depth in meters for the model.

d50_min_depth

Numeric. Distance (m) at which detection efficiency = 50% at minimum depth.

d95_min_depth

Numeric. Distance (m) at which detection efficiency = 95% at minimum depth.

d50_max_depth

Numeric. Distance (m) at which detection efficiency = 50% at maximum depth.

d95_max_depth

Numeric. Distance (m) at which detection efficiency = 95% at maximum depth.

dist_step

Numeric. Distance step size for grid predictions. Default is 5.

depth_step

Numeric. Depth step size for grid predictions. Default is 1.

color_option

Character. Viridis color palette option for plots. Default is "magma".

plot

Logical. Whether to display visualization plots. Default is TRUE.

return_model

Logical. Whether to fit and return a logistic regression model. Default is TRUE.

return_object

Logical. Whether to return the complete function object. Default is TRUE.

Value

If return_object = TRUE, returns a list containing:

func

The logistic function for calculating detection efficiency

predict_grid

Function for generating prediction grids

predictions

Data frame with distance, depth, and DE predictions

params

List of input parameters

log_model

Fitted logistic regression model (if return_model = TRUE)

line_plot

ggplot2 line plot object (if plot = TRUE)

heatmap_plot

ggplot2 heatmap object (if plot = TRUE)

combined_plot

Combined plot using patchwork (if plot = TRUE)

If return_object = FALSE, returns invisibly or displays plots only.

Details

The function creates a logistic detection efficiency model where:

• d50 and d95 parameters vary linearly with depth

• Detection efficiency = 1 / (1 + exp(a + b * distance))

• b = 2.944 / (d95 - d50)

• a = -b * d50

The model assumes that detection efficiency decreases with distance and that the detection range changes with depth in a predictable manner.

See also

generate_random_points, generate_spaced_points

Examples

if (FALSE) {
# Create a basic detection efficiency model
de_model <- create_logistic_curve_depth(
  min_depth = 2,
  max_depth = 30,
  d50_min_depth = 50,
  d95_min_depth = 20,
  d50_max_depth = 150,
  d95_max_depth = 60
)

# Use the model to predict detection efficiency
efficiency <- de_model$func(dist_m = 100, depth_m = 15)

# Generate custom predictions
custom_grid <- de_model$predict_grid(
  dist_range = seq(0, 200, by = 10),
  depth_range = seq(5, 25, by = 2)
)

# Create model without plots
de_model_quiet <- create_logistic_curve_depth(
  min_depth = 2, max_depth = 30,
  d50_min_depth = 50, d95_min_depth = 20,
  d50_max_depth = 150, d95_max_depth = 60,
  plot = FALSE
)
}