ERDDAP is NOAA's data server, and several satellite products this package wants are on one. It matters here for a practical reason: the same products at PO.DAAC need an Earthdata login, and ERDDAP serves them with no account at all, subsetting server-side by longitude, latitude and time.
Value
a named list, one entry per dataset, each with server,
dataset_id, variables, has_altitude, start, end, resolution,
label and reference
What is built in, and what is not
Three datasets ship. They are the ones that are global, current, and relevant to a shelf study:
MUR | SST, 0.01 deg daily | 2002-06 onward |
VIIRSCHL | chlorophyll, daily | 2020-05 onward |
VIIRSCHL2018 | chlorophyll, daily | 2012-01 to 2022-07 |
There is no global VIIRS SST here. The VIIRS SST dataset on these servers covers the US West Coast only (-128 to -115 E), which is no use on the Atlantic shelf, and nothing global turned up in its place. MUR is the SST to use instead — it is a blended analysis that takes VIIRS among its inputs, so it carries the same sensor's information at higher resolution and over a much longer record.
The two chlorophyll entries are the same instrument processed twice.
VIIRSCHL2018 is the 2018 reprocessing, which stops in July 2022;
VIIRSCHL is the current gap-filled product, which starts in May 2020. They
overlap for two years and are not the same numbers, so a series spanning both
has a seam in it.
MUR is not a model SST
MUR is a satellite analysis of the foundation temperature — the
temperature below the daily warming layer — where a model SST is the
topmost model level. On a calm sunny afternoon those differ by a degree or
more. Both arrive in a column called SST, and source_of() is what
distinguishes them.
See also
accessERDDAP(), erddap_dataset() for any other ERDDAP dataset
Examples
names(erddap_datasets())
#> [1] "MUR" "VIIRSCHL" "VIIRSCHL2018"
erddap_datasets()$MUR$dataset_id
#> [1] "jplMURSST41"