Modelling Species Distribution in the Southern Ocean

Species Distribution Modelling is a correlative approach that helps assess the relationship between species occurrence records and their surrounding environmental conditions. Studying the distribution of benthic species in the Southern Ocean constitutes an interesting approach to interpolate the little amount of information we have in this broad scale and poorly accessible area. We provide here a set of environmental descriptors suitable for the modelling work. The present vignette describes the procedure to select and download them.

Environmental descriptors list

The present dataset is composed of 58 environmental descriptors, set at a 0.1 degree grid-cell resolution, on the Southern Ocean extent (80 to 45 degrees S, -180 to 180 degrees longitude). Layers are provided as a netCDF file format.

The full metadata file, giving details on each of the data sources, is available from https://data.aad.gov.au/metadata/records/environmental_layers.

How to download the data

The first steps to download the data are to define the directory in which the data will be stored. In this example, we use a temporary folder.

library(blueant)
my_data_directory <- tempdir()

## the data source we want
data_source <- sources_sdm("Southern Ocean marine environmental data")

## put these together into a blueant configuration
config <- bb_config(my_data_directory) %>% bb_add(data_source)
config

Now we can ask blueant to fetch and unpack the data:

status <- bb_sync(config, verbose = TRUE)

The verbose = TRUE argument helps to display the progress output. Once completed, the status object provides some information about the files that were downloaded. In particular, the downloaded netCDF files can be listed with:

nc_files <- Filter(function(z) grepl("\\.nc$", z), status$files[[1]]$file)

## the first few files
head(basename(nc_files))

And then you can open your data with the raster package

library(raster)

## create a raster stack of all layers
x <- stack(nc_files)

## the layers in this object
names(x)

## select the layers you want according to the layer names
subset(x, c("depth", "slope", "POC_2005_2012_mean"))

## or subset by layer number
## e.g. plot the first four layers
plot(subset(x, 1:4))

You can also consult the blueant vignette for further information.



AustralianAntarcticDivision/blueant documentation built on Jan. 9, 2025, 8:04 a.m.