SNODAS download product documentation is not up to date

[smalers]

NOAA provides this documentation describing the SNODAS data files:

• National Operational Hydrologic Remote Sensing Center SNOw Data Assimilation System (SNODAS) Products at NSIDC
• Snow Data Assimilation System (SNODAS) Data Products at NSIDC, Version 1 (G02158) - links to the above

The documentation is from 2003 and I don't see a more recent version. It is generally accurate. However, the appendix B information is not quite accurate. Current .txt files have information similar to the following. Note that the values for Benchmark x-axis coordinate (ulxmap in the hdr file) and Benchmark y-axis coordinate (ulymap in the hdr file) in the following are different from the above documentation.

I have updated the SNODAS Tools Python code for version 2.1.0 to scrape the the values out of the txt file for the hdr file needed for processing. I don't know when the values changed or how often they change but assuming that they do not change will cause a data shift (a few hundred meters horizontally?). I have reached out to SNDAS support. I may try to figure that out because some results may have shifted. I'll need to reprocess some data.

Format version: NOHRSC GIS/RS raster file v1.1
Data source: RUC2, NESDIS, etc.
Created by module: sm_products
Created by module comment: Not applicable
Created year: 2023
Created month: 4
Created day: 26
Created hour: 10
Created minute: 29
Created second: 10
Last modified by module: sm_products
Last modified by module comment: Not applicable
Last modified year: 2023
Last modified month: 4
Last modified day: 26
Last modified hour: 10
Last modified minute: 29
Last modified second: 10
Satellite data: no
Satellite name: Not applicable
Satellite channel: 0
Satellite data calibrated: no
Description: Modeled snow water equivalent, total of snow layers
Thematic: no
Theme file: Not applicable
Data units: Meters / 1000.000000
Product code: 396487018
Attribute table: Not applicable
Data file pathname: us_ssmv11034tS__T0001TTNATS2023042605HP001.dat
Data type: integer
Data bytes per pixel: 2
Data intercept: 0.00000000000000
Data slope: 1.00000000000000
Minimum data value: 0.00000000000000
Maximum data value: 32767.0000000000
No data value: -9999.00000000000
Number of columns: 6935
Number of rows: 3351
Geographically corrected: yes
Projected: no
Projection file: Not applicable
Horizontal datum: WGS84
Horizontal precision: 0.00000000000000
Elevation above datum: no
Vertical datum: Not applicable
Vertical precision: 0.00000000000000
Benchmark column: 0
Benchmark row: 0
Benchmark x-axis coordinate: -124.729166666662
Benchmark y-axis coordinate: 52.8708333333312
X-axis resolution: 0.00833333333333300
Y-axis resolution: 0.00833333333333300
X-axis offset: 0.00416666666666650
Y-axis offset: 0.00416666666666650
Minimum x-axis coordinate: -124.733333333328
Maximum x-axis coordinate: -66.9416666666640

[smalers]

If I understand the issue you describe, there are two things to be aware of regarding the coordinates in the .txt files:

1. The benchmark y-axis coordinate does vary slightly over time.
2. The minimum x and y -axis coordinates also vary slightly between dates. This is due to minor precision issues with how these coordinates are expressed in the header files. So instead we recommend that users use the coordinates listed in this help article: How do I convert SNODAS binary files to GeoTIFF or NetCDF?

The Special Report National Operational Hydrologic Remote Sensing Center SNOw Data Assimilation System (SNODAS) Products at NSIDC was written in 2003 when we initially started archiving SNODAS data produced by NOHRSC, so it is a little outdated. The user guide is the most recent documentation Snow Data Assimilation System (SNODAS) Data Products at NSIDC, Version 1 (G02158) and we keep it updated with any changes/new information.

When users are trying to convert the binary files to geotiff or netcdf I actually point them to the help article: How do I convert SNODAS binary files to GeoTIFF or NetCDF? which has more up-to-date information. This includes the note about the spatial shift in the data for files prior to and after 01 October 2013. It is very small (~5% of a 1km grid cell) but I just wanted to mention it in case you hadn't come across it.

I hope this helps, but if I have misunderstood your question or if you have any further questions then please let me know.