Data access - focus on 73N to 77N, -130W to -160 W from 7-10 October 2022

[dksasaki]

We need to understand which data products should be used. This issue focuses on exploring and discussing details of the different datasets we are investigating.

[dksasaki]

When accessing the following earthdata webpage (product VNP29 ) the following figure appears:

The colors correspond to the masks in the netcdf file - for instance (look in the Attributes section):

<xarray.DataArray 'SeaIceCover_Map' (number_of_lines: 100, number_of_pixels: 100)>
array([[253., 253., 253., ..., 253., 253., 253.],
       [253., 253., 253., ..., 253., 253., 253.],
       [253., 253., 253., ..., 253., 253., 253.],
       ...,
       [253., 253., 253., ..., 253., 253., 253.],
       [253., 253., 253., ..., 253., 253., 253.],
       [253., 253., 253., ..., 253., 253., 253.]], dtype=float32)
Dimensions without coordinates: number_of_lines, number_of_pixels
Attributes:
    coordinates:    latitude longitude
    long_name:      Sea Ice Cover map with masks
    mask_meanings:  200-missing, 201-no_decision, 211-night, 225-land, 237-in...
    mask_values:    [200 201 211 225 237 250 252 253 254]
    valid_range:    [  0 100]

where:

mask_meanings:

200-missing,
201-no_decision,
211-night
225-land
237-inland_water
250-cloud
252-unusable_L1B_data
253-bowtie_trim
254-no_L1B_data

I found a 1.1 version of this document..

[lauracrews]

I think we may need to use the level 1b data. We can also access those data on Earthdata.

There are a few different kinds of files. For each satellite (NPP and JPSSS) there are three types of imagery:

• Imagery resolution (I-bands) which are 375 m horizontal resolution
• Moderate resolution (M-bands) which are 750 m horizontal resolution
• Day-night band: A single band for low-light conditions which are 750 m horizontal resolution

Each imagery file has an associated geolocation file that contains lat/lon coordinates for each pixel as well as other information about the satellite.

Here's a list of the short names for each product:

NPP

• Imagery: VNP02IMG + geolocation VNP03IMG
• Moderate: VNP02MOD + geolocation VNP03MOD
• Day-night band: VNP02DNB + geolocation VNP03DNB

JPSS

• Imagery: VJ02IMG + geolocation VJ03IMG
• Moderate: VJ02MOD + geolocation VJ03MOD
• Day-night band: VJ02DNB + geolocation VJ03DNB

There are also "near-real time" NRT versions of there products - I am not sure what the difference is with those.

[lauracrews]

The next challenge is figuring out how to create a "true color" image from the many spectral bands available in the imagery resolution and moderate resolution bands. Essentially, I think this is a matter of understanding which bands roughly correspond to RGB. Thisexample from Satpy might be useful but it's unclear what kind of file is being used (maybe this actually is a hint at what kind of file we need?)