The lat/lon layers are not reported in GDS2 L2P granules produced
by the NOAA Advanced Clear-Sky Processor for Ocean (ACSPO) SST system,
for new-generation geostationary sensors, including ABI onboard GOES-R
(GOES-16/17, aka "G16/17") and AHI onboard Himawari-8 ("H08") satellites.

Examples L2P data are found in PO.DAAC https://doi.org/10.5067/GHG16-2PO27
NCEI https://doi.org/10.25921/ayf6-c438 and NOAA CoastWatch
https://coastwatch.noaa.gov/cw/satellite-data-products/sea-surface-temperature/acspo-abi.html

Due to special L1b pre-processing on the ground called "geo-rectification",
lat/lon don't change for the same (i,j) geostationary pixel from image
to image, and can be easily calculated offline. Omitting these fixed
layers in GDS2 data files decreases the filesize by a factor of 2 to 3.

This folder contains the geo_nav.py script which can be invoked as: 
geo_nav.py [path to GDS2 granule]
It calculates lat/lon's for each pixel in the GDS2 granule and
adds them to the input granule as two extra layers, "lat" and "lon".
If the "lat" and "lon" layers already exist in the input
GDS2 file, the script will print an error message and exit.

Precalculated lat/lon's for G16/17 are also provided in files:
G16_089_5_W.nc -> G16 Cal/Val position at 89.5deg West (Nov 2016 - Nov 2017)
G16_072_0_W.nc -> G16 operational position at 75.0 deg West (Dec 2017 - on)
G17_089_5_W.nc -> G17 Cal/Val position at 89.5 deg West (Mar 2018 - Oct 2018)
G17_137_0_W.nc -> G17 perational position at 137.0 deg West (Nov 2018 - on)

In case of questions, email nesdis.star.acspo@noaa.gov