rgeedim Demo

This vignette shows how to extract a Google Earth Engine asset by name for an arbitrary extent and visualize it in R.

First, we load {rgeedim} and call gd_initialize(). This is a wrapper function around geedim.Initialize() that must be run before using the Python Google Earth Engine API.

library(rgeedim)
gd_initialize()

Perhaps the simplest way to specify the target extent is using the xmin/xmax/ymin/ymax arguments to gd_bbox(). This function returns a Python object equivalent to GeoJSON, which is interchangeably represented as a simple list object in R using {reticulate}.

Determine Target Region

r <- gd_bbox(
  xmin = -121,
  xmax = -120.5,
  ymin = 38.5,
  ymax = 39
)

As is standard for GeoJSON, coordinates of the bounding box are expressed in WGS84 decimal degrees ("OGC:CRS84"). Note that longitude, latitude (X, Y) coordinate pair order is implied.

Access by ID

We can find IDs of assets of interest using the Google Earth Engine data catalog: https://developers.google.com/earth-engine/datasets/catalog

To obtain an R object reference to the asset we pass the "id" to gd_image_from_id(). For example here we use Global SRTM Topographic Diversity:

x <- gd_image_from_id('CSP/ERGo/1_0/Global/SRTM_topoDiversity')

gd_image_from_id() will return geedim.mask.MaskedImage and gd_collection_from_name() will return geedim.collection.MaskedCollection objects.

Now, we pass the image result to gd_download(). We can specify output filename and target area as region arguments. See gd_bbox() for examples of making a region argument from bounding coordinates or a {terra} SpatExtent object.

Other options that can be passed to the BaseImage.download() method include scale which allows warping of the result to a target resolution. Try modifying this example to use scale=90 (~native SRTM resolution):

img <- gd_download(x, filename = 'image.tif',
                   region = r, scale = 900,
                   overwrite = TRUE, silent = FALSE
                  )

gd_download() (invisibly) returns the filename on successful download, which helps to “pipe” into functions that might read the result.

So we can use the {terra} rast() function to read the GeoTIFF gd_download() result.

library(terra)
f <- rast(img)
par(mar=c(1,1,1,1))
plot(f[[1]])

# inspect object
f