Ramani Huria 2.0 is collecting drainage information in the most flood prone areas across Dar es salaam. This information is used to develop a flood model which requires accurately collected specifications of drains such as depth, width, blockage (by either vegetation or material), connectivity, and diameter (typically for culverts).

To develop this model, Ramani Huria is working with a hydrologist in Deltares– Netherlands, Hessel Winsemius. Dr Hessel is an expert in the field of hydrology, with particular applications in global and regional flood risk modelling. This week he was in Dar es Salaam, to look closely at flood modeling with the local team. One of the major issues that he noticed was how hard the team were working to deal with data quality assurance – they had to identify errors, missing data and warnings manually on QGIS by checking the attribute table of a specific segment. These issues which were time consuming, tiresome and not convenient. Critically, there’s no facility in QGIS that checks specifically for drainage connectivity, so Hessel created a Python-based utility called Hydro-OSM to do just that!

Drain Data Check Model

To make data quality checks more efficient, Hessel designed the Hydro-OSM data quality assurance model to automatically identify  errors, warnings and missing information of drains. This method makes data cleaning easier than the previously employed method, and specifically checks attributes relevant to urban drainage. For cases of missing information or errors field teams will revisit the site and fix the error.

Hessel believes that if we are successful in developing the model it would be a stepping stone for decision makers in their effort of creating a flood resilient city. He’s also looking forward to helping the community develop flood models.

Hessel Checking the Functionality of the Model

The aim of all of this is to use OSM data to create a flood model (automated kind of procedure) to simulate how water moves from one place to another in an established network ie drain, ditch, waterway etc.

For the model to function, drains needs to be connected as much as possible. Although some drains in Dar es Salaam are not connected, Hessel is still working on this to see how unconnected drains can be included in the model.

All of this information needs a well developed or complete data model which Ramani Huria has already successfully established. More about the drainage data model that has been developed can be read on the previous blog  here.