Issue Summary
I am experiencing systematic data access failures when requesting VNP46A3 (monthly) and VNP46A4 (annual) nighttime lights products for large countries requiring multiple tiles. This issue affects years 2013-2025, while 2012 data access works correctly. The pattern suggests a change in manifest structure or tile registration between 2012 and subsequent years.
Observed Pattern
Working Countries (Single/Few Tiles):
Belgium, Netherlands, Denmark, etc.: Successful data access for all years 2012-2025
Both VNP46A3 and VNP46A4 products retrieve correctly
Failing Countries (Multi-Tile):
Australia, Brazil, Canada: 2012 works, 2013-2025 fail systematically
Consistent error: "Tile 'hXXvYY' for date 'YYYY-MM-DD' could not be found in the manifest"
Same tiles fail consistently across multiple years (e.g., h15v11 for Brazil)
Technical Details
Using valid Earthdata credentials (verified working)
Data confirmed to exist in your archives at https://ladsweb.modaps.eosdis.nasa.gov/archive/allData/5000/
Error occurs during manifest query phase, not data download
Affects multiple data access tools, not just BlackMarblePy
Example Error Pattern
[WARNING] [download]: Tile 'h15v11' for date '2018-03-01' could not be found in the manifest.
[WARNING] [download]: Tile 'h15v11' for date '2018-04-01' could not be found in the manifest.
Questions for NASA
Manifest Structure: Did the tile registration or manifest query system change between 2012 and 2013 for VNP46A3/A4 products?
Multi-Tile Access: Is there a different API endpoint or query method required for accessing multi-tile data post-2012?
Collection Versioning: Are there different collection versions that should be specified for different year ranges?
Recommended Approach: What is the proper method for programmatically accessing complete coverage data for large countries spanning multiple tiles?
System Information
Product: VNP46A3 (monthly composites), VNP46A4 (annual composites)
Collection: 5000
Years Affected: 2013-2025
Geographic Scope: Countries requiring multiple VIIRS tiles
Access Method: LAADS API via authenticated requests
I would appreciate guidance on the correct approach for accessing multi-tile VNP46A3/A4 data for the 2013-2025 period, or clarification if there are known limitations in the manifest system for large geographic queries.
Thank you for your assistance in resolving this data access issue.
VNP46A3/A4 Manifest Lookup Issues for Multi-Tile Countries (2013-2025) - Systematic Data Access Problem
-
- User Services
- Posts: 52
- Joined: Fri Dec 13, 2024 2:40 pm America/New_York
- Has endorsed: 8 times
- Endorsed: 1 time
Re: VNP46A3/A4 Manifest Lookup Issues for Multi-Tile Countries (2013-2025) - Systematic Data Access Problem
Try using https://ladsweb.modaps.eosdis.nasa.gov/archive/allData/5200/ instead. See:
https://ladsweb.modaps.eosdis.nasa.gov/alerts-and-issues/168722
and
https://ladsweb.modaps.eosdis.nasa.gov/alerts-and-issues/168723
https://ladsweb.modaps.eosdis.nasa.gov/alerts-and-issues/168722
and
https://ladsweb.modaps.eosdis.nasa.gov/alerts-and-issues/168723
Regards,
LAADS User Services
To receive news from LAADS DAAC direct to your inbox, email laadsdaac-join@lists.nasa.gov with “subscribe” in the subject line.
LAADS User Services
To receive news from LAADS DAAC direct to your inbox, email laadsdaac-join@lists.nasa.gov with “subscribe” in the subject line.
Re: VNP46A3/A4 Manifest Lookup Issues for Multi-Tile Countries (2013-2025) - Systematic Data Access Problem
Thank-you for your response.
5200 does have some improvement, though Canada and Russia still fail with a similar failure noted previously.
We are using Blackmarble Py and have success in most instances hence are challenged why the outcome is ok in most cases. The USA may have problems in some years but we are still determining if this is the case with further testing.
Are there issues with data from these large Northern countries? Could there be an issue on your end? Has the data file format / structure changed perhaps for these countries ?
Any help is much appreciated. Thanks.
5200 does have some improvement, though Canada and Russia still fail with a similar failure noted previously.
We are using Blackmarble Py and have success in most instances hence are challenged why the outcome is ok in most cases. The USA may have problems in some years but we are still determining if this is the case with further testing.
Are there issues with data from these large Northern countries? Could there be an issue on your end? Has the data file format / structure changed perhaps for these countries ?
Any help is much appreciated. Thanks.