Species presence maps
The function detectionMaps can generate maps of observed species richness (number of different species recorded at stations) and maps showing the number of observations by species. It uses the record table produced by recordTable and the camera trap station table as input. Note that the examples are not particularly pretty because of the low number of records used in the sample data set.
Number of observed species
We first create a map of the number of observed species.
Mapstest1 <- detectionMaps(CTtable = camtraps,
recordTable = recordTableSample,
Xcol = "utm_x",
Ycol = "utm_y",
stationCol = "Station",
speciesCol = "Species",
printLabels = TRUE,
richnessPlot = TRUE, # by setting this argument TRUE
speciesPlots = FALSE,
addLegend = TRUE
)
Number of records by species
Maps of the number of independent detections of the observed species can be generated just as easily. Normally, maps for all species will be created at once. Here, to avoid cluttering the vignette, we look at one species only. This is achieved via the argument speciesToShow. Arguments richnessPlot and speciesPlots are changed compared to the observed species richness plot above. It is also possible to set both arguments to TRUE or FALSE.
# subset to 1 species
recordTableSample_PBE <- recordTableSample[recordTableSample$Species == "PBE",]
Mapstest2 <- detectionMaps(CTtable = camtraps,
recordTable = recordTableSample_PBE,
Xcol = "utm_x",
Ycol = "utm_y",
stationCol = "Station",
speciesCol = "Species",
speciesToShow = "PBE", # added
printLabels = TRUE,
richnessPlot = FALSE, # changed
speciesPlots = TRUE, # changed
addLegend = TRUE
)
The number of independent observations depends on the argument minDeltaTime in the recordTable function.
Shapefile export
Function detectionMaps comes with 4 arguments that allow for and control creation of ESRI shapefile for use in GIS software: writeShapefile, shapefileName, shapefileDirectory and shapefileProjection. The resulting shapefile will show stations as point features (as the map above), with coordinates, total species number and number of observations per species in the attribute table. The shapefile attribute table is identical to the resulting data.frame of the detectionMaps function.
The following example demonstrates the creation of a shapefile using detectionMaps. Please note that for demonstration the shapefile is saved to a temporary directory, which makes no sense in real data and must be changed by the user. The argument shapefileProjection must be a valid argument to the function CRS from the package sp, i.e., a PROJ.4 string. These can be found under http://spatialreference.org/. There, find the projection your coordinates are in, click on “Proj4” and copy the text. In this example, it is a projected coordinate system, UTM 50N with the WGS84 ellipsoid. Its PROJ.4 string is "+proj=utm +zone=50 +ellps=WGS84 +datum=WGS84 +units=m +no_defs". Because it is so widespread, here’s the PROJ4 string for standard Lat/Long coordinates using the WGS84 ellipsoid (a standard used by most GPS devices): "+proj=longlat +ellps=WGS84 +datum=WGS84 +no_defs".
# writing shapefiles requires packages rgdal and sp
library(rgdal)
library(sp) # define shapefile name
shapefileName <- "recordShapefileTest"
shapefileProjection <- "+proj=utm +zone=50 +ellps=WGS84 +datum=WGS84 +units=m +no_defs"
# run detectionMaps with shapefile creation
Mapstest3 <- detectionMaps(CTtable = camtraps,
recordTable = recordTableSample,
Xcol = "utm_x",
Ycol = "utm_y",
stationCol = "Station",
speciesCol = "Species",
richnessPlot = FALSE, # no richness plot
speciesPlots = FALSE, # no species plots
writeShapefile = TRUE, # but shaepfile creation
shapefileName = shapefileName,
shapefileDirectory = tempdir(), # change this in your scripts!
shapefileProjection = shapefileProjection
)
# check for the files that were created
list.files(tempdir(), pattern = shapefileName)## [1] "recordShapefileTest.dbf" "recordShapefileTest.prj"
## [3] "recordShapefileTest.shp" "recordShapefileTest.shx"# load it as shapefile
shapefileTest <- readOGR(dsn = tempdir(),
layer = shapefileName)## OGR data source with driver: ESRI Shapefile
## Source: "C:\Users\Juergen\AppData\Local\Temp\RtmpYd7kdS", layer: "recordShapefileTest"
## with 3 features
## It has 9 fields# we have a look at the attribute table
shapefileTest@data## Station utm_y utm_x EGY MNE PBE TRA VTA n_species
## 1 StationA 604000 526000 0 0 4 0 2 2
## 2 StationB 606000 523000 0 2 8 0 2 3
## 3 StationC 607050 525000 6 0 6 8 1 4# the output of detectionMaps is used as shapefile attribute table. Therefore, they are identical:
all(shapefileTest@data == Mapstest3)## [1] TRUEMaking and using a SpatialPointsDataFrame
The data frame that detectionMaps returns can be turned into a SpatialPointsDataFrame as defined in the package sp. SpatialPointsDataFrames can be used for spatial analyses and queries in R, e.g. for extracting data from raster layers such as elevation maps.
This is how it works. We use the output of detectionMaps created above and make a SpatialPointsDataFrame. Afterwards, a sample raster is created, plotted together with the points and the values are extracted.
detections_spdf <- SpatialPointsDataFrame(coords = Mapstest3[,c("utm_x", "utm_y")],
data = Mapstest3,
proj4string = CRS(shapefileProjection))
str(detections_spdf)## Formal class 'SpatialPointsDataFrame' [package "sp"] with 5 slots
## ..@ data :'data.frame': 3 obs. of 9 variables:
## .. ..$ Station : chr [1:3] "StationA" "StationB" "StationC"
## .. ..$ utm_y : num [1:3] 604000 606000 607050
## .. ..$ utm_x : num [1:3] 526000 523000 525000
## .. ..$ EGY : int [1:3] 0 0 6
## .. ..$ MNE : int [1:3] 0 2 0
## .. ..$ PBE : int [1:3] 4 8 6
## .. ..$ TRA : int [1:3] 0 0 8
## .. ..$ VTA : int [1:3] 2 2 1
## .. ..$ n_species: num [1:3] 2 3 4
## ..@ coords.nrs : num(0)
## ..@ coords : num [1:3, 1:2] 526000 523000 525000 604000 606000 ...
## .. ..- attr(*, "dimnames")=List of 2
## .. .. ..$ : NULL
## .. .. ..$ : chr [1:2] "utm_x" "utm_y"
## ..@ bbox : num [1:2, 1:2] 523000 604000 526000 607050
## .. ..- attr(*, "dimnames")=List of 2
## .. .. ..$ : chr [1:2] "utm_x" "utm_y"
## .. .. ..$ : chr [1:2] "min" "max"
## ..@ proj4string:Formal class 'CRS' [package "sp"] with 1 slot
## .. .. ..@ projargs: chr "+proj=utm +zone=50 +ellps=WGS84 +datum=WGS84 +units=m +no_defs +towgs84=0,0,0"# now we create a sample raster and extract data from it (if the raster package is available)
if("raster" %in% installed.packages()){
library(raster)
raster_test <- raster(x = extend(extent(detections_spdf), y = 500), nrows = 10, ncols = 10)
values(raster_test) <- rpois(n = 100, lambda = seq(1, 100)) # fill raster with random numbers
# plot raster
plot(raster_test,
main = "some raster with camera trap stations",
ylab = "UTM N", # needs to be adjusted if data are not in UTM coordinate system
xlab = "UTM E") # needs to be adjusted if data are not in UTM coordinate system
# add points to plot
points(detections_spdf, pch = 16)
# add point labels
text(x = coordinates(detections_spdf)[,1],
y = coordinates(detections_spdf)[,2],
labels = detections_spdf$Station,
pos = 1)
# extracting raster values. See ?extract for more information
detections_spdf$raster_value <- extract(x = raster_test, y = detections_spdf)
# checking the attribute table
detections_spdf@data
}
## Station utm_y utm_x EGY MNE PBE TRA VTA n_species raster_value
## 1 StationA 604000 526000 0 0 4 0 2 2 88
## 2 StationB 606000 523000 0 2 8 0 2 3 30
## 3 StationC 607050 525000 6 0 6 8 1 4 12The same procedure also works with the camera trap station information table instead of the detectionMaps output.
