Calculate the average movement for cells in a BirdFlow model

calc_movement_vectors() calculates the average modeled movement of birds from each location over a single transition in a BirdFlow model.

Usage

calc_movement_vectors(bf, start, direction = "forward")

Arguments

bf: A BirdFlow model
start: The starting timestep for the transition to be modeled
direction: "Forward" by default. Set to "backward" to calculate vectors for a transition backwards in time.

Value

A data frame with columns:

i: The location index of the starting location. This is the row in the distribution matrix that corresponds to the location.
start: The starting timestep of the transition.
end: The ending timestep of the transition.
start_x: The x coordinate of the starting cell of the transition.
start_y: The y coordinate of the starting cell of the transition.
end_x: The weighted average destination x coordinate of all transitions from the starting cell, with weights set to the transition probability for each destination.
end_y: The weighted average destination y coordinate.
weight: This is the proportion of the population at the starting cell in the eBird S&T distribution for the starting timestep.
width: This is a range rescaling of weight and is used by both plot_movement_vectors(), and animate_movement_vectors() to set the line width of the arrows.

Details

In practice each row of the transition matrix represents a single starting location (start_x and start_y) and the values in the row represent transition probability to a number of destination cells. The probabilities are used as weights to calculate an average of both the destination x and y coordinates (end_x, and end_y). The ending coordinates of the arrow represented the expected average destination of all birds that start in the cell.

Examples

bf <- BirdFlowModels::amewoo
mv <- calc_movement_vectors(bf, 7)