Datasetpublic

E4DRR/rim2d-simulations

RIM2D East Africa Flood Simulations Event-based 2D hydraulic inundation simulations for the GHACOF flood storyline workshop (May 2026), produced under the ICPAC E4DRR / CRAF'D project. Cases Each <case>/ directory is self-contained and can be re-run with RIM2D: <case>/ ├── input/ │ ├── dem.nc Copernicus GLO-30 DEM → reference grid │ ├── buildings.nc Overture Maps building fraction raster │ ├── channel_mask.nc channel mask… See the full description on the dataset page: https://huggingface.co/datasets/E4DRR/rim2d-simulations.

sourceHugging Facecc-by-4.0updated 4mo agoView on Hugging Face

0likes2kdownloads

Dataset Card

RIM2D East Africa Flood Simulations

Event-based 2D hydraulic inundation simulations for the GHACOF flood storyline workshop (May 2026), produced under the ICPAC E4DRR / CRAF'D project.

Cases

Each <case>/ directory is self-contained and can be re-run with RIM2D:

<case>/
├── input/
│   ├── dem.nc                  Copernicus GLO-30 DEM → reference grid
│   ├── buildings.nc            Overture Maps building fraction raster
│   ├── channel_mask.nc         channel mask (WorldCover-derived)
│   ├── iwd.nc                  initial water depth
│   ├── roughness.nc            Manning's n (from WorldCover class)
│   ├── pervious_surface.nc     GHSL-derived pervious fraction
│   ├── sealed_surface.nc       GHSL-derived sealed fraction
│   ├── sewershed.nc            sewershed extent (currently uniform)
│   ├── outflowlocs.txt         outflow point list
│   ├── fluvial_boundary_none.txt
│   └── rain/imerg_t*.nc        per-30-min IMERG V07 rainfall frames
├── output/
│   ├── case_v1_wd_<t>.nc       water depth at sim second <t>
│   ├── case_v1_vel_<t>.nc      flow velocity at sim second <t>
│   ├── case_v1_wd_max.nc       max-depth envelope
│   ├── case_v1_vel_max.nc      max-velocity envelope
│   ├── case_v1_wd_max_t.nc     time of max depth (s since start)
│   └── case_v1_mask.nc
└── simulation_v1.def           RIM2D model definition (flex format)

This dataset holds only the science inputs/outputs + model config. Viewer assets, curated storyline narratives, and stakeholder validations are co-production artefacts and live in a separate repo (<https://github.com/HillaryKoros/DevOps-hazard-modeling>).

Cases included (initial release)

Case slug	Country	Event date	Sim duration	Domain
`nairobi_2026-03-06`	Kenya	06 Mar 2026	24 h	Nairobi metro
`nile_2024_abu_hamad`	Sudan	25–30 Aug 2024	6 d	Abu Hamad
`ssd_2025-10-05`	South Sudan	05 Oct 2025	24 h	Upper Nile (Renk area)
`eth_2021-08-16`	Ethiopia	16–18 Aug 2021	3 d	Addis Akaki river
`uga_2019-05-26`	Uganda	26–28 May 2019	3 d	Central Uganda (Kampala)
`eri_2019-08-10`	Eritrea	10–15 Aug 2019	5 d	Asmara plateau highlands
`som_2023-11-19`	Somalia	19–21 Nov 2023	3 d	Lower Shabelle
`dji_2019-11-21`	Djibouti	21–23 Nov 2019	3 d	Djibouti City
`tza_2024-04-10`	Tanzania	10–17 Apr 2024	7 d	Dar es Salaam metro
`rwa_2023-05-01`	Rwanda	01–06 May 2023	5 d	Western + Northern provinces
`bdi_2024-04-01`	Burundi	01–22 Apr 2024	21 d	Bujumbura + central Burundi
`tza_2024-04-10`	Tanzania	10–17 Apr 2024	7 d	Dar es Salaam

Preview animations

Each case ships with a preview.gif showing the water-depth evolution.

Case	Preview
Nairobi (KEN)	[image]
Sudan Nile (SDN)	[image]
South Sudan (SSD)	[image]
Addis Akaki (ETH)	[image]
Uganda Central (UGA)	[image]
Eritrea Highlands (ERI)	[image]
Somalia Lower Shabelle (SOM)	[image]
Djibouti City (DJI)	[image]
Dar es Salaam (TZA)	[image]
Rwanda W/N (RWA)	[image]
Burundi (BDI)	[image]

All cases share a common pipeline: GEE (DEM/MERIT/WorldCover/GHSL) + Overture (buildings/roads) + TDX-Hydro (rivers) + IMERG V07 (rainfall) → prepare_rim2d_case.py → RIM2D GPU.

Reproduce a case

bash

cd <case>
# rain frames live at ../<case>/input/rain/ relative to v1/
RIM2D simulation_v1.def --def flex

Source code for the pipeline: <https://github.com/icpac-igad/DevOps-hazard-modeling/tree/main/hazard-model-api>

Storyline schema

Each storyline.json is structured for workshop co-production:

—event — title / date / country / trigger / description
—hydromet — auto-extracted rainfall + RIM2D statistics
—forecast_signal — counterfactual EPS analysis (TODO)
—impact — population, displacement, fatalities, sectors (TODO from EM-DAT/IFRC)
—mobility — IDP camps in inundated zone (TODO from IOM DTM / IDMC)
—narrative — setup / peak / response / lesson (curated)

License

CC BY 4.0 — please cite as:

ICPAC E4DRR / GHACOF (2026). RIM2D East Africa Flood Simulations dataset. HuggingFace Datasets: E4DRR/rim2d-simulations.

Contact

ICPAC E4DRR project — Hillary Koros

—hillary.koros@igad.int
—koroshillary12@gmail.com