Using Low-Cost Equipment to Gather High Precision Data for Flood Modeling

By Hawa Adinani, Iddy Chazua, and Hessel Winsemius

In collaboration with Deltares and with funding from GIZ, OMDTZ was able to collect accurate and precise river cross-section surveys using an innovative, low-cost technique developed between the Resilience Academy and OMDTZ.

Photo: Godfrey Kassano

The Msimbazi Basin, home to an estimated 27 percent of Dar es Salaam’s population, has historically been a fertile floodplain and the city’s main water source. Over recent years, however, the Msimbazi River has become a major source of flash floods in the city. Each year, Msimbazi flooding leads to fatalities and destroys critical infrastructure.

There have been a number of studies and projects, supported by funders such as the World Bank, to develop investment programs and detailed plans for the basin. The intent has been to catalyze investments from different actors, including the government, the private sector, and other development partners, to restore the floodplain and turn it into an asset for the city.

Better understanding the flow of the Msimbazi, the way it responds to heavy rainfall events, and where and why it floods is necessary for designing sustainable and green solutions to combat flood risk and improve living conditions. Such an understanding requires the development of a detailed hydraulic model.

Such a model needs to be based on accurate information about the amount of water the channel and surrounding floodplains can carry, and for these accurate measurements of the three-dimensional cross-section of the river are needed. To develop this model, the German International Cooperation Agency (GIZ) worked with us at OpenMap Development Tanzania (OMDTZ) and Dutch knowledge institute Deltares to collect cross-section surveys, extending an existing flood model for the Lower Msimbazi to include a much larger upstream portion of the river.

Project coverage area

From Research to Innovation to Service

It is important to mention that about one year ago, a survey of the cross-section of the river would require an international consultant to fly in with expensive equipment. (A professional-grade set of survey equipment with dual-frequency Global Navigation Satellite System (GNSS) units or acoustic profiling equipment costs tens of thousands of dollars.) Even with such expertise, a consultant would not know the environment, the possible risks of entering the Msimbazi, and the steps needed to be able to work in the neighboring wards.

Because of several clear use cases for collecting very precise geographic position data, OMDTZ teamed with Resilience Academy partner Delft University of Technology (TU Delft) to investigate if a new low-cost, dual-frequency GNSS receiver could be deployed for this end. TU Delft student Kirsten van Dongen co-developed a method with us to use the hardware, and demonstrated that it could deliver similar accuracy and precision as professional-grade survey equipment.

After this study, OMDTZ expanded the method so that locally available tools for data collection could be used in conjunction with this hardware, providing results that would be harder and more expensive to generate by hiring outside consultants. We have now deployed two units that use the real-time kinematic (RTK) mode of the new, low-cost hardware that are able to provide centimeter precise surveys.

The setup of all these tools was done by local experts from OMDTZ. At OMDTZ, our principles have always been local people, local devices, and open knowledge, and being able to use our own internal tech capabilities for high precision surveys is a tremendous achievement in working by these principles.

Deploying Human and Technical Resources

The methodology used for this project has been documented and is freely available for people who would like to further understand the process and even replicate it in other locations.

Before starting to map, the team first setup and configured the base stations. Locations were chosen based on factors including distance and a clear sky for clear satellite signals.

Base stations

Then the team trained the surveyors and made sure they were familiar with the safety measures for working in the river during data collection.

A team of 6 people (4 mappers and 2 technicians) was deployed to the field equipped with an Android smartphone with Open Data Kit (ODK) installed for data collection. We developed a hydrological data collection protocol for ODK with Deltares and carried out our first field survey with Deltares associates Bobby Russell and Hessel Winsemius, both of whom were experienced in hydrological surveys.

This first field exercise enabled us to refine our methods and safety precautions. After this, we were able to scale up and rapidly collect over 80 cross-sections of the river.

A surveyor in the field holding a locally made survey tool in RTK mode

While conducting the cross-section surveys in the river, surveyors had to actively monitor the accuracy of their GPS receivers and other devices while taking precautions to avoid deep water.

After the data was collected, it was uploaded to the server for a quality check, and surveyors redid any sections which contained errors. This was then provided to Deltares for use in their hydraulic flood model

One of the surveyed transect, visualized in 3d using DTM

Iddy Chazua, OMDTZ’s tech-lead and the man behind the setup and the analysis of this data collection, shared his experience of the fieldwork.

“Seeing the accomplishments done using low-cost differential GPS, and thinking of how young surveyors and hydrologists will eventually have access to tools they never dreamed of being able to afford in the early stages of their career, is indeed revolutionary.”

Iddy also shared his touching experience with the communities residing along the Msimbazi valley and how they use the valley in spite of knowing that it will eventually flood.

“It was a moving experience seeing how the river valley is being used by the community, but also sad how people are being affected by the same river when it floods. Seeing part of a house hanging along the river while knowing that it does not have more than weeks. But still people are living inside.”

This situation demands a robust situational analysis of the circumstances of communities residing along the valley and a multidisciplinary approach to how they can be helped to build a resilient city that considers different aspects of their needs from a ground level.

“Initial data visualization, comparing different image sources from different years, can easily show how the river changes direction and size over time, which will help in the prediction of how the river valley will evolve in the future and enable appropriate measures before more impacts can occur on the ground.”

Project Results and Data Use

The product of this assignment will be the development of a wider flood model that will assist in providing the evidentiary basis for investments in the basin. This project is complementary to the ongoing Tanzania Urban Resilience Program (TURP), implemented by the World Bank.

“Our collaboration with OMDTZ through both student work and surveying led not only to an excellent dataset, but also offered us as specialists the opportunity to convey the knowledge on hydrology and hydraulics to a local team, so that they can now collect critical river data in a sustainable way.” — Hessel Winsemius, Delft University of Technology

“The OMDTZ survey of river cross-sections and hydraulic structures along the Msimbazi River supported the creation of a comprehensive hydraulic model of the area that has led to the development of a more realistic flood hazard map of the most vulnerable areas.” — Eskedar Gebremedhin, Deltares

A screenshot of one of the flood hazard maps produced using the new survey data

Using Open Data, Open Source, and maps to solve different socio-economic challenges.