Delft-FEWS applications around the globe


Community Water Watch (CWW) – Dar es Salaam, Tanzania

Community Water Watch (CWW) – Dar es Salaam, Tanzania

Dar Es Salaam experiences frequent flooding. Informal settlements are particularly vulnerable to floods. These settlements are unplanned, have to cope with very limited infrastructure, and solid waste frequently clogs the undersized drainage system, causing more severe flooding and direct and indirect consequences. As a result, people are driven further into poverty.

Community Water Watch (CWW) is a community-designed and -operated flood early warning service for Dar es Salaam, Tanzania. The service uses open online media communication channels, innovative affordable weather stations, and hydrological and hydrodynamic models built upon locally-collected map data. It is built entirely on local, affordable data sources. The service is a dashboard that provides predictions of riverine and/or ward-scale flooding and shares situational awareness information through online media sources. The service is implemented with partners on the ground in order to realise a community co-design. The openness of the service and the inclusive approach to its design closely follow the Principles for Digital Development (see International Financing Institutions (IFIs) such as the Department of International Development (DFID) and the World Bank endorse these principles.

Left: Flooding in Dar es Salaam city centre (NotYetUhuru, May 2019, CC BY-SA 4.0) Right: Drainage channel in Manzese ward – Dar Es Salaam, next to a school, clogged by solid waste.

The service is piloted on the Msimbazi Valley and the surrounding wards, which are the most flood prone areas in Dar es Salaam. It is offered as a scalable Software as a Service (SaaS) so that it can be expanded for use in multiple organisations with relatively low effort. The project was financed by the Netherlands Enterprise Agency (RVO, Netherlands) under the Partners for Water funding programme and carried out by a consortium consisting of Deltares, TU-Delft, FloodTags, and the Tanzania Red Cross Society (TRCS).

The CWW service is used by local communities and stakeholders who need real-time information to support relief activities. The main users of the CWW dashboard are currently the Tanzania Red Cross Society (TRCS) and the Dar es Salaam Rapid Transport (DART) agency. These types of agencies do not typically have hydrologists or other water specialists on staff. Therefore, the CWW user interface was designed to be very easy to use and displays Early Warning service outputs tailored to their needs. It provides exactly the information they need to make a better decision. The co-design process included several workshop and training activities with all relevant stakeholders. 

First workshop (left) and training activities (right) with all relevant stakeholders


In 2019, the “short rains” season, which normally lasts from October through December, continued through January 2020 and caused many flood events. During this season, the consortium kept in contact with DART and TRCS to discuss the use of the system and get feedback on how to make the system more user friendly. Furthermore, the World Bank monitored the dashboard to understand if its capabilities could add value to future loans/projects. 

“The dashboard is helping us to plan for the 70:30 busses distribution (70% of busses kept West of Kigogo bridge, and 30% East) in the system.”
- Eng Fanuel Kalugendo, DART, 17 October 2019

The CWW Dashboard, displayed in Tanzania Red Cross Society main conference room, on December 17th 2019. The dashboard was used on this day to understand and anticipate heavy flooding that occurred throughout the city. Left: visual of flood reports; Right: visual of hydrological and impact forecast.


The CWW dashboard, the web front-end of the flood early warning service, incorporates information from different components:

  1. Hydro-meteorological network installed by the Trans-African Hydro-Meteorological Observatory (TAHMO)
  2. Delft-FEWS, running operationally at Deltares to provide a forecast every hour for the following 8 hours on river depth, flooded area, number of buildings affected per neighborhood (ward), flooded area of the bus depot, and flooded bus stops. It takes water approximately 8 hours to flow from the very upstream parts of the Msimbazi River into the city of Dar es Salaam. A lead time of 8 hours was chosen because it is enough time for stakeholders to act on forecasted floods while avoiding many false alarms and misses which could result in loss of confidence in the early warnings. The Delft-FEWS system includes:
    • Data feeds: observed hydro-meteorological data from TAHMO and forecasted meteorological precipitation from the NOAA Global Forecast System (GFS) 
    • hydrological wflow model
    • 1D2D urban flood model in SOBEK 
    • Delft-FIAT impact assessment model
  3. Real-time online media feed providing information during events. 
  4. “FloodTags Assistant,” an automated assistant for the Telegram Direct Messaging App, which enables people to (1) analyse and monitor the contents of disaster management related group chats and (2) start one-on-one chats with people on the ground to gather detailed flood information. The information from this assistant is fed into the CWW Dashboard, so it can be analysed and filtered.

The CWW service is unique in that it combines real-time and forecast data, enabling preparedness, with situational awareness. The service is co-designed to provide tailored information on flood early warning and situational awareness to users lacking or with limited ICT and technical capacity to maintain a forecasting system themselves.

In this pilot, we demonstrated the need for and value of involving stakeholders (providers of data as well as users of information) from the very start in order to connect information to actual decisions. It is key to think through use cases, and how operational procedures can benefit from the information.

We aim to continue developing the dashboard, make it more robust, keep it operational for 2 more years, and provide transitional support so that more organisations can incorporate the service into their standard operating procedures. After this period, it is foreseen that a  government agency will take over hosting/maintenance of the system. 

The long-term goal from the project’s consortium is to establish this service over many cities in the world in close collaboration with early warning and relief agencies.

For more information, please contact Patricia Trambauer ( or Hessel Winsemius (