geospatial data for automated assignment by GPS location

[emiliom]

Summary of Discussion between Emma, Keshav and Emilio, 4/27/2018

Codes preceding the category are as follows: <desirability (D)>-<accessibility (A)>, where 1 is the most desirable / accessible

• (1D-1A) climate
  • precipitation and temperature
  • PRISM for lower 48, other dataset for Puerto Rico (probably good approach, b/c PR climate is so different!)
  • mean annual averages vs more refined estimates/indices?
  • data for the actual day when samples were collected? can we hunt for nearby stations that would give us data for the actual day. Keshav will do the leg work. Start with CUAHSI HydroClient
• (2D-?TBD) soil type and subtypes
  • I'll follow up with USDA/USFS guy
  • characteristic properties
• (3D-?TBD) land cover/use
  • I'll bring a couple of options, including the national veg map and a remote-sensing based map. Should include PR.
• (4D-?TBD) topography
  • elevation, aspect, slope
  • I'll follow up
  • https://nationalmap.gov/elevation.html, https://www.landfire.gov/slope.php (7.5 min resolution? If so, it's too coarse), https://edna.usgs.gov/ (old product?).
  • https://nationalmap.gov/3DEP/3dep_prodserv.html: "To download 3DEP products visit: https://viewer.nationalmap.gov/basic/"
• (5D-?TBD) atmospheric composition
  • CO2 and CH4 column abundance (GOSAT data, CH4 only?). But may not be as useful
  • AmeriFlux flux? Promising. But let's focus first on the other 4 mapped data types!

---

Some general comments, notes

• See the FedData R package for promising avenues and links. Dylan B is a contributor.
  • Note: for rasters it accesses tiled subsets from AWS, eg, https://prd-tnm.s3.amazonaws.com/StagedProducts/Elevation/1/ArcGrid/
• Look into Google Earth Engine (with Python API), specially for elevation derivatives
  • UW Geohackweek 2016, Google Earth Engine (with Python API)
  • https://developers.google.com/earth-engine/python_install
  • GEE tutorial from UW Geohackweek 2017. The Python API was not covered, but there are very relevant processing steps and datasets demoed in the tutorial
• Puerto Rico. Follow up with Miguel

Climate

• See our PRISM work, including information at https://github.com/BiG-CZ/BiG-CZ-Portal/blob/master/ClimatologyGriddedData.md

Soil types

• See my fleshed out notes and tests at soil data access (SDA) services #5
• See also the FedData R package
• See https://github.com/selimnairb/EcohydroLib/tree/master/ecohydrolib/ssurgo

Land cover/use

• MRLC NLCD 2011
• See the TNM WCS service at https://viewer.nationalmap.gov/services/. GetCapabilities request: https://www.mrlc.gov/arcgis/services/LandCover/USGS_EROS_LandCover_NLCD/MapServer/WCSServer?request=GetCapabilities&service=WCS
• The U.S. National Vegetation Classification
• See the WCS ingest capabilities, including NCLD specifically, at EcohydroLib. In particular: ecohydrolib/usgs/nlcdwcs.py, ecohydrolib/nlcd/daacquery.py, ecohydrolib/wcslib.py

Elevation and derivatives (slope & aspect)

• The National Map - Elevation Point Query Service. This is what I used in the demos at the CZIMEA Aug 2017 Portland meeting.
• Roland's email for additional resources (edited a bit):

  WCSs for NLCD and elevation currently exist, although the latter is characterized as a prototype. Those and other National Map services are listed here. There are a number of avenues for accessing 3DEP content (point clouds, interpolated rasters), including through the National Map viewer. There's also point query service--not sure if the content is different for this endpoint than via WCS tho.

  I will check to see about plans for web services around elevation derivatives. My understanding is that 3DEP is not supporting these because the 3DEP program is focusing its energy on building infrastructure for distributing CONUS-wide lidar in the coming years