This function simplifies the output of ac, dc and acdc, by processing information from the 'archive' elements of an acdc_archive-class object that hold the results of calls to the workhorse routines. This is especially useful if the algorithm(s) have been applied chunk-wise, in which case the results for each chunk are returned in a list. The function aggregates information across chunks to generate a continuous time series of results and a map of the expected proportion of time steps spent in each grid cell.

acdc_simplify(
  archive,
  type = c("acs", "dc"),
  mask = NULL,
  normalise = TRUE,
  keep_chunks = FALSE
)

Arguments

archive

An acdc_archive-class object returned by ac, dc or acdc.

type

A character that defines whether the function should be implemented for the outputs of a call to an AC* algorithm (ac or acdc), in which case type = "acs", or the DC algorithm (dc), in which case type = "dc".

mask

(optional) A spatial mask (e.g., the argument passed to bathy in ac, dc or acdc) to mask areas (e.g., land) from the overall map. If implemented, cells in masked areas are assigned NAs rather than a score of 0.

normalise

A logical input that defines whether or not to normalise the overall map so that cell scores sum to one. If normalise = FALSE, the overall map represents the expected number of time steps spent in each grid cell; if normalise = TRUE, the overall map represents the expected proportion of time steps spent in each grid cell.

keep_chunks

A logical variable that defines whether or not to retain all chunk-specific information.

Value

The function returns an object of class acdc_record-class.

Details

If the ac, dc or acdc function was implemented step-wise, this function simply extracts the necessary information and re-packages it into an acdc_record-class object. For a chunk-wise implementation, the function (a) computes the map of where the individual could have spent more or less time by aggregating the chunk-specific maps) and (b) simplifies chunk-specific records into a single contiguous time series, with re-defined time stamps from the start to the end of the time series (for AC* algorithm(s)) to return an acdc_record-class object.

See also

The AC, DC and ACDC algorithms are implemented by ac, dc and acdc. After simplification, acdc_plot_trace, acdc_plot_record and acdc_animate_record can be implemented to visualise time-specific results.

Author

Edward Lavender