# Multi-Resolution Digital Elevation Model for TPU Saint-Venant Flood Simulation (Arkansas River, 2 to 8 m)

## Facts

- Provider: Google Research (simulation_research/flood)
- License: CC-BY-4.0 (https://github.com/google-research/google-research/blob/master/README.md)
- Format: TIFF (in sampled files)
- Stored at: Google Cloud Storage, gs://gresearch/flood-sim (us)
- Access: public, readable without an account
- Category: Earth Observation & Geospatial
- Source: https://github.com/google-research/google-research/tree/master/simulation_research/flood
- Page: https://scigantic.com/archive/3934ad53-87bb-536f-87d0-cf339e96da16
- Keywords: Earth Observation & Geospatial, whole genome sequencing, TIFF, Google Cloud Storage

## About

Digital elevation models (DEMs) used as the bathymetry and terrain input to Google Research's physics-based flood simulation, which solves the 2D Saint-Venant (shallow water) equations with Manning friction on Cloud TPUs (notebook flood_simulation_tpu.ipynb in google-research/simulation_research/flood). The same floodplain is provided at five grid spacings, 2m.tiff, 3m.tiff, 4m.tiff, 5m.tiff and 8m.tiff, so the solver can be run and benchmarked at increasing resolution. Measured 2026-11-16 from the GeoTIFF header of 8m.tiff: 5766 x 2683 single-band 32-bit float grid with 8 m pixels in WGS 84 / UTM zone 15N (EPSG:32615), origin at easting 492812, northing 3898087, a roughly 46 x 21 km reach of the Arkansas River valley in central Arkansas (the notebook's 'Conway' example). File sizes run from 35 MB (8 m) to 495 MB (2 m), 0.97 GB in total. The notebook drives an inflow boundary condition (default 15000 m3/s) across one edge and writes water depth h and discharges q_x, q_y every cycle. Open with rasterio or tifffile. Datasets in the google-research repository are released under CC BY 4.0; cite the google-research flood simulation project.

## Load the data

```python
import gcsfs
fs = gcsfs.GCSFileSystem(token='anon')
fs.ls('gresearch/flood-sim/', detail=False)[:10]
```

Lists the files with anonymous access. Runs on any machine, with no cloud account.

## Data structure

File types in sampled files: TIFF

Example file: `flood-sim/2m.tiff`

## Open it

- In a hosted notebook with the data mounted: https://scigantic.com/archive/3934ad53-87bb-536f-87d0-cf339e96da16
- Full record as JSON (schema card included): https://api.scigantic.com/api/archives/3934ad53-87bb-536f-87d0-cf339e96da16
- Over MCP: `get_schema_card` with id `3934ad53-87bb-536f-87d0-cf339e96da16` at https://api.scigantic.com/mcp

## Related topics

- [Earth Observation & Geospatial datasets](https://scigantic.com/topics/earth-observation-and-geospatial)
- [TIFF datasets](https://scigantic.com/topics/tiff)
- [Whole genome sequencing datasets](https://scigantic.com/topics/whole-genome-sequencing)
