Download specific assets from a set of STAC items
Usage
rsi_download_rasters(
items,
aoi,
asset_names,
sign_function = NULL,
merge = FALSE,
gdalwarp_options = c("-r", "bilinear", "-multi", "-overwrite", "-co",
"COMPRESS=DEFLATE", "-co", "PREDICTOR=2", "-co", "NUM_THREADS=ALL_CPUS"),
gdal_config_options = c(VSI_CACHE = "TRUE", GDAL_CACHEMAX = "30%", VSI_CACHE_SIZE =
"10000000", GDAL_HTTP_MULTIPLEX = "YES", GDAL_INGESTED_BYTES_AT_OPEN = "32000",
GDAL_DISABLE_READDIR_ON_OPEN = "EMPTY_DIR", GDAL_HTTP_VERSION = "2",
GDAL_HTTP_MERGE_CONSECUTIVE_RANGES = "YES", GDAL_NUM_THREADS = "ALL_CPUS",
GDAL_HTTP_USERAGENT = "rsi (https://permian-global-research.github.io/rsi/)"),
...
)
Arguments
- items
A
StacItemCollection
object, as returned byrsi_query_api()
.- aoi
Either an sf(c) object outlining the area of interest to get imagery for, or a
bbox
image containing the bounding box of your AOI.- asset_names
The names of the assets to download. If this vector has names, then the names of the vector are assumed to be the names of assets on the STAC server, which will be renamed to the elements of the vector in the final output.
- sign_function
A function that takes the output from
query_function
and signs the item URLs, if necessary.- merge
Logical: for each asset, should data from multiple items be merged into a single downloaded file? If
TRUE
, this returns a single file for each asset, which has been merged via gdalwarp. No resampling or compositing is performed, but rather each pixel uses the last data downloaded. This is fast, but precludes per-item masking and compositing. IfFALSE
, each asset from each item is saved as a separate file.- gdalwarp_options
Options passed to
gdalwarp
through theoptions
argument ofsf::gdal_utils()
. The same set of options are used for all downloaded data and the final output images; this means that some common options (for instance,PREDICTOR=3
) may cause errors if bands are of varying data types. The default values are provided byrsi_gdalwarp_options()
.- gdal_config_options
Options passed to
gdalwarp
through theconfig_options
argument ofsf::gdal_utils()
. The default values are provided byrsi_gdal_config_options()
.- ...
Passed to
item_filter_function
.
Value
A data frame where columns correspond to distinct assets, rows correspond to distinct items, and cells contain file paths to the downloaded data.
Examples
if (FALSE) { # interactive()
aoi <- sf::st_point(c(-74.912131, 44.080410))
aoi <- sf::st_set_crs(sf::st_sfc(aoi), 4326)
aoi <- sf::st_buffer(sf::st_transform(aoi, 5070), 100)
landsat_image <- get_landsat_imagery(
aoi,
start_date = "2022-06-01",
end_date = "2022-08-30",
download_function = rsi_download_rasters
)
}