geoglows/tdxhydro-postprocessing

rapid_preprocess

Louis R. Rosas, Dr Riley Hales, Josh Ogden

Inspired by previous work including

• RAPIDpy https://github.com/BYU-Hydroinformatics/RAPIDpy
• ArcGIS RAPID Toolbox https://github.com/Esri/python-toolbox-for-rapid
• RAPID https://rapid-hub.org/

Overview

Provides the master function 'PreprocessforRAPID', which:

Reads in provided stream networks and catchments

Sorts the modeling network by Strahler stream order, ascending

Fixes stream segments of 0 length for different cases as follows:

1. Feature is coastal w/ no upstream or downstream
   • Delete the stream and its basin
2. Feature is bridging a 3-river confluence (Has downstream and upstreams)
   • Artificially create a basin with 0 area, and force a length on the point of 1 meter
3. Feature is costal w/ upstreams but no downstream
   • Force a length on the point of 1 meter
4. Feature doesn't match any previous case
   • Raise an error for now

Creates three new networks:

1. A visulation network, which has the top order 1 streams dissolved with their downstream order 2 segment (for smaller file sizes)
2. A modeling network, which is similar to 1) but only preserves the geometry of the order 2 segment and the longest order 1 segment
3. A modified basins network. Any streams that were merged will have their corresponding catchements also dissolved

Calculates the muskingum parameters for the stream network and adds this information to the modeling network

Creates the following six files:

• comid_lat_lon_z.csv
• riv_bas_id.csv
• k.csv
• kfac.csv
• x.csv
• rapid_connect.csv

Creates weight tables for each of the given input ERA netCDF datasets

All out puts are saved to given directory

tdxhydro-postprocessing's Issues

Open to collaborate?

@rileyhales, I was quite excited to discover your v2 release of the GEOGloWS Streamflow project's datasets and tools. It's quite impressive that you and your team has built such an strong foundation for a next gen global river flow simulation system. I'm also impressed that you posted all your data to AWS Open Data Registry: https://registry.opendata.aws/geoglows-v2/. Nice work!

I've been tracking the development of global datasets and tools for similar purposes for the last 5 years or so with the goal of expanding Model My Watershed capabilities from CONUS to the globe.

Our team has finally gotten the funding to get started on this work, and we're very interested in building from and potentially contributing to your work. We're thinking of starting by directly using your modified TDX-Hydro Streams and Catchments that you're maintaining on AWS at http://geoglows-v2.s3-website-us-west-2.amazonaws.com/. Our idea is to develop Rapid Watershed Delineation capabilites to operate over your modified TDX-Hydro datasets.

Might you be interested in a conversation about potentially collaborating? If it doesn't make sense right now, no worries; you have admirers who will use what you share! If it does make sense to explore a potential collaboration, a low-effort and low-effort easy start might be to connect via zoom for an information exchange.