From laghas to lifelines: nature-based solutions platform
Turning flood water into water security.
An open-science platform for Garissa's Nature-Based Solutions Opportunity Scan, prepared for the World Bank GFDRR, University of Turku and ESA scale-up process. It maps where lagha and Tana River floods hit, screens sites where flood water could be captured with nature-based solutions, compares ECMWF, GFS and ICON forecasts, and includes a trilingual community manual.
The problem
Garissa suffers both flash floods and drought: water arrives as a hazard and is gone before the dry season.
What we built
- 1
Derived seasonal river channels (laghas) from SRTM elevation data and classified them by flash-flood hazard.
- 2
Measured flood exposure of surveyed schools and health facilities using UNOSAT flood extents.
- 3
Screened candidate sites for sand dams, hafirs, water pans, check dams and spreading bunds.
- 4
Published the method, data catalogue and code openly, with a story notebook and a grounded AI assistant.
Highlights
- 35+ map layers with an El Niño runoff simulator
- Sortable register of candidate sites with CSV and GeoJSON download
- 16-day forecasts from three global models for nine locations
- Community manual in English, Af-Soomaali and Kiswahili
- 12-indicator results framework with baselines
Built with
- Python
- QGIS
- Google Earth Engine
- Jupyter
- GeoJSON
- Gemini
Honest status: Figures are screening-level GIS estimates from SRTM 30 m and UNOSAT data, not engineering studies. All candidate sites must be validated with communities, WRA, NEMA and County Water before design.
Built by our founder for the County Government of Garissa, before TovuTech was incorporated.
Want something like this?
We design, build, host and secure systems like this end to end, and train your team to run them.
More success stories