Overlay specified SDM predictions that meet the percent overlap threshold requirement onto base geometry
Arguments
- base.geom
object of class
sfc; base geometry- sdm
object of class
sf; original SDM predictions- sdm.idx
names or indices of column(s) with data to be overlaid
- overlap.perc
numeric; percent overlap threshold, i.e. percentage of each base geometry polygon must overlap with SDM prediction polygons for overlaid density value to be calculated and not set as NA
Value
Object of class sf with the geometry of base.geom and
the data in the sdm.idx columns of sdm overlaid onto that
geometry. Note that this means all columns of sdm not in
sdm.idx will not be in the returned object.
Because the data are considered spatially intensive, the agr
attribute will be set as 'constant' for all columns in the returned object.
Additionally, the output will match the class of sdm, with regards
to the classes tbl_df, tbl, and data.frame. This means that, in addition to
being an sf object, if sdm is a tibble then the output will
also be a tibble, while if sdm is just a data frame then the output
will not be a tibble.
Details
See the eSDM GUI manual for specifics about the overlay process.
This process is equivalent to areal interpolation (Goodchild and Lam 1980),
where base.geom is the target, sdm is the source, and the data
specified by sdm.idx are spatially intensive.
Note that overlay_sdm removes rows in sdm that have NA values
in the first column specified in sdm.idx (i.e. sdm.idx[1]),
before the overlay.
Thus, for valid overlay results, all columns of sdm specified in
sdm.idx must either have NA values in the same rows
or contain only NAs.
References
Goodchild, M.F. & Lam, N.S.-N. (1980) Areal interpolation: a variant of the traditional spatial problem. Geo-Processing, 1, 297-312.
Examples
pol1.geom <- sf::st_sfc(
sf::st_polygon(list(rbind(c(1,1), c(3,1), c(3,3), c(1,3), c(1,1)))),
crs = sf::st_crs(4326)
)
pol2.geom <- sf::st_sfc(
sf::st_polygon(list(rbind(c(0,0), c(2,0), c(2,2), c(0,2), c(0,0)))),
crs = sf::st_crs(4326)
)
pol2.sf <- sf::st_sf(data.frame(Dens = 0.5), geometry = pol2.geom,
crs = sf::st_crs(4326))
overlay_sdm(pol1.geom, pol2.sf, 1, 25)
#> Simple feature collection with 1 feature and 1 field
#> Attribute-geometry relationship: constant (1)
#> Geometry type: POLYGON
#> Dimension: XY
#> Bounding box: xmin: 1 ymin: 1 xmax: 3 ymax: 3
#> Geodetic CRS: WGS 84
#> Dens geometry
#> 1 0.5 POLYGON ((1 1, 3 1, 3 3, 1 ...
# Output 'Dens' value is NA because of higher overlap.perc value
overlay_sdm(pol1.geom, pol2.sf, 1, 50)
#> Simple feature collection with 1 feature and 1 field
#> Attribute-geometry relationship: constant (1)
#> Geometry type: POLYGON
#> Dimension: XY
#> Bounding box: xmin: 1 ymin: 1 xmax: 3 ymax: 3
#> Geodetic CRS: WGS 84
#> Dens geometry
#> 1 NA POLYGON ((1 1, 3 1, 3 3, 1 ...
if (FALSE) {
# These examples take longer to run
overlay_sdm(sf::st_geometry(preds.1), preds.2, 1, 50)
overlay_sdm(sf::st_geometry(preds.2), preds.1, "Density", 50)
}