Closed torse closed 1 year ago
This looks good, though I'm wondering about just snow, vs snow and ice together? Or is snow meant to also imply ice? At Planet in our UDM we do snow_ice_percent
: Percent of snow and ice values in dataset. Snow_ice values represent scene content areas (non-blackfilled) that are hidden below snow and/or ice.
If yours does cover ice too then probably good to have it in the description.
I am curious on the datasets you're thinking of that use this. Just want to be sure it cuts across a few, and fits with all of them.
Good point! I double checked with our data provider and indeed the attribute also imply ice covered surfaces. I've updated the description in the README.md accordingly but left the schema.json as is. Let me know you think this should be updated as well (e.g. "title": "Snow/Ice Cover"
).
Our datasets are Sentinel-2 based L2 and L3 products over Germany computed by the MAJA and WASP processor respectively. The access to the datasets via STAC are currently beeing developed and available for preview (S2_L2A_MAJA and S2_L3A_WASP). In an technology project we've also provided experimental access to the original Sentinel-2 data as COGs that also include the S2 specific snow_ice_percentage
attribute: S2_L2A_MSI_COG.
I'm late to the discussion here, but wondering why snow_cover is being added? Is there something particular about this ground cover type (as opposed to say, land, ocean, vegetated, barren, etc.) that impacts the electro-optical observation data? Sensor saturation?
Initial minimalistic PR for eo:snow_cover field