
Extract site x species information from long format data
Source:R/fb_format_site_species.R
fb_format_site_species.Rd
Convert a flat data.frame
with species occurrence/abundance at site level
into a proper data.frame
object that can then be used by other functions.
The final output contains sites in rows and species in columns.
Arguments
- data
a
data.frame
in a long format (see example).- site
a
character
of length 1. Name of the column with site labels.- species
a
character
of length 1. Name of the column with species names.- value
a
character
of length 1. Name of the column with species occurrence/abundance.- na_to_zero
a logical value. If
TRUE
(default)NA
are replaced by0
.
Value
A data.frame
with sites in rows and species in columns. The first
column is named "site"
and contains the name of the sites.
Examples
filename <- system.file(
"extdata", "woodiv_raw_data.csv",
package = "funbiogeo"
)
all_data <- read.csv(filename)
head(all_data)
#> site country longitude latitude species count family genus
#> 1 26351755 Portugal 2635000 1755000 JPHO 1 Cupressaceae Juniperus
#> 2 26351755 Portugal 2635000 1755000 PPIR 1 Pinaceae Pinus
#> 3 26351765 Portugal 2635000 1765000 JPHO 1 Cupressaceae Juniperus
#> 4 26351955 Portugal 2635000 1955000 JPHO 1 Cupressaceae Juniperus
#> 5 26351955 Portugal 2635000 1955000 PPIR 1 Pinaceae Pinus
#> 6 26351965 Portugal 2635000 1965000 JPHO 1 Cupressaceae Juniperus
#> binomial endemism cultivated plant_height seed_mass sla
#> 1 Juniperus phoenicea 0 0 4.88150 79.86000 4.365246
#> 2 Pinus pinaster 0 0 19.75384 55.83434 3.357539
#> 3 Juniperus phoenicea 0 0 4.88150 79.86000 4.365246
#> 4 Juniperus phoenicea 0 0 4.88150 79.86000 4.365246
#> 5 Pinus pinaster 0 0 19.75384 55.83434 3.357539
#> 6 Juniperus phoenicea 0 0 4.88150 79.86000 4.365246
#> wood_density
#> 1 0.6487500
#> 2 0.4430277
#> 3 0.6487500
#> 4 0.6487500
#> 5 0.4430277
#> 6 0.6487500
site_species <- fb_format_site_species(all_data, "site", "species", "count")
site_species[1:3, 1:4]
#> site JPHO PPIR PPIA
#> 1 26351755 1 1 0
#> 2 26351765 1 0 0
#> 3 26351955 1 1 0