Landscape Metrics for Categorical Map Patterns 🗺️ in R
spatialze_lsm()
with directions
argumentcalculate_lsm
(all metrics: more than 5 times faster with 70% less memory
allocation for augusta_nlcd
; larger increases were found for smaller data)
and window_lsm
(a single metric: more than 6 times faster for augusta_nlcd
;
larger increases were found for smaller data)calculate_lsm
extras_df
object that lists which extras are needed by
each metrictibble::tibble()
with tibble::new_tibble(list())
in most functions.
This change is partially responsible for improvements of the window_lsm
speedraster_to_points
with get_points
in several places.
The get_points
function is based on the column and row numbers multiplied by
the resolution, not actual coordinates.table
with (faster) tabulate
in lsm_p_core
prepare_extras
,
get_area_patches
, get_class_patches
, get_complexity
, get_enn_patch
,
get_points
, and get_perimeter_patch
window_lsm
behaviour for situations with NAs values and non-square windowsterra
and sf
instead of raster
and sp
as underlying frameworksshow_*
functions to avoid ggplot2
warningshow_correlation
points_as_mat()
helper functionextract_lsm
returned an no-needed warning messageget_patches
returns a unique patch id for all classeswindow_lsm()
get_boundaries()
for matrix input and return_raster = TRUE
rel_mut_inf
to list_lsm()
ggplot2
versionlsm_l_ai
if class with only one cell existslsm_c_lsi
, lsm_c_nlsi
, lsm_l_lsi
not using cell surfaceslsm_l_relmutinf
to calculate relative mutual information