rangemap_enm: Species distributional ranges based on ENMs/SDMs outputs

Description Usage Arguments Details Value Examples

View source: R/rangemap_enm.R

Description

rangemap_enm generates a distributional range for a given species using a continuous raster layer produced using ecological niche modeling or species distribution modeling tools. This function binarizes the model in suitable and unsuitable areas using a user specified level of omission or a given threshold value. Optionally, representations of the species extent of occurrence (using convex hulls) and the area of occupancy according to the IUCN criteria can also be generated. Shapefiles can be saved in the working directory if it is needed.

Usage

1
2
3
4
5
6
rangemap_enm(model_output, occurrences = NULL, threshold_value = NULL,
             threshold_omission = NULL, min_polygon_area = 0,
             simplify = FALSE, simplify_level = 0, polygons = NULL,
             extent_of_occurrence = TRUE, area_of_occupancy = TRUE,
             final_projection = NULL, save_shp = FALSE, name,
             overwrite = FALSE, verbose = TRUE)

Arguments

model_output

a RasterLayer of suitability for the species of interest generated using a ENM or SDM algorithm, that will be binarized using the a user-defined threshold_value or a value calculated based on a percentage of omission (0 - 100) defined in threshold_omission. If the layer is projected, this projection must be WGS84 (EPSG:4326); if not projected, WGS84 projection will be assigned for the analysis.

occurrences

a data.frame containing geographic coordinates of species occurrences, columns must be: Species, Longitude, and Latitude. Geographic coordinates must be in decimal degrees. occurrences may not be defined, but if so, threshold_value must be defined. Default = NULL.

threshold_value

(numeric) value used for reclassifying the model_output. This value will be the lowest considered as suitable for the species and must be inside the range of values present in model_output. If defined, threshold_omission will be ignored. If occurrences is not defined, this parameter is required. Default = NULL.

threshold_omission

(numeric) percentage of occurrence records to be excluded from suitable areas considering their values of suitability in the continuous model (e.g., 0, 5, or 10). Ignored if threshold_value is provided. Default = NULL.

min_polygon_area

(numeric) minimum area of polygons that will be kept as part of the species ranges after defining suitable areas and convert raster layer to polygon. Default = 0. A value of 0 will keep all polygons.

simplify

(logical) if TRUE, polygons of suitable areas will be simplified at a tolerance defined in simplify_level. Default = FALSE.

simplify_level

(numeric) tolerance to consider when simplifying polygons created from suitable areas in model_output. Lower values will produce polygons more similar to the original geometry. Default = 0. If simplifying is needed, try numbers 0-1 first. Ignored if simplify = FALSE.

polygons

(optional) a SpatialPolygons* object to clip polygons and adjust extent of occurrence to these limits. Projection must be WGS84 (EPSG:4326). If NULL, the default, a simplified world map will be used.

extent_of_occurrence

(logical) whether to obtain the extent of occurrence of the species based on a simple convex hull polygon; default = TRUE.

area_of_occupancy

(logical) whether to obtain the area of occupancy of the species based on a simple grid of 4 km^2 resolution; default = TRUE.

final_projection

(character) string of projection arguments for resulting Spatial objects. Arguments must be as in the PROJ.4 documentation. See CRS-class for details. If NULL, the default, projection used is WGS84 (EPSG:4326).

save_shp

(logical) if TRUE, shapefiles of the species range, occurrences, extent of occurrence, and area of occupancy will be written in the working directory. Default = FALSE.

name

(character) valid if save_shp = TRUE. The name of the shapefile to be exported. A suffix will be added to name depending on the object, as follows: species extent of occurrence = "_extent_occ", area of occupancy = "_area_occ", and occurrences = "_unique_records".

overwrite

(logical) whether or not to overwrite previous results with the same name. Default = FALSE.

verbose

(logical) whether or not to print messages about the process. Default = TRUE.

Details

All resulting Spatial objects in the list of results will be projected to the final_projection. Areas are calculated in square kilometers using the Lambert Azimuthal Equal Area projection, centered on the centroid of occurrence points given as inputs or, if points are not provided, the resulting range.

Value

If occurrences and threshold_omission are defined, a sp_range object (S4) containing: (1) a data.frame with information about the species range, and Spatial objects of (2) unique occurrences, (3) species range, (4) extent of occurrence, and (5) area of occupancy.

If instead of occurrences and threshold_omission, threshold_value is provided, the result will be a sp_range object (S4) of two elements: (1) a data.frame with information about the species range, and (2) a SpatialPolygons object of the species range.

If extent_of_occurrence and/or area_of_occupancy = FALSE, the corresponding spatial objects in the resulting sp_range object will be empty, an areas will have a value of 0.

Examples

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
# parameters
sp_mod <- raster::raster(list.files(system.file("extdata", package = "rangemap"),
                                    pattern = "sp_model", full.names = TRUE))
data("occ_train", package = "rangemap")

thres <- 5
save <- TRUE
name <- "test"

enm_range <- rangemap_enm(model_output = sp_mod, occurrences = occ_train,
                          threshold_omission = thres)

summary(enm_range)

rangemap documentation built on Sept. 5, 2021, 5:17 p.m.