Tanzania is the most flood-affected country in East Africa, with towns and cities undergoing massive spatial expansion with a growth in informal settlements. These areas are typically characterized by limited access to adequate municipal services and infrastructure such as water supply, sanitation, solid waste, electricity and drainage. Together, these shortcomings contribute to a deterioration of environmental conditions and the presence of major health and safety risks which are particularly acute at times of flooding. In Dar es Salaam, floods in 2009, 2011, and in early and late 2014, caused loss of life and injury, while also severely impacting diverse sectors including transportation, energy and social infrastructure such as schools and health facilities. Zanzibar City is a relatively flat, low lying urban area and an important cultural and historic heritage, tourism function and the central business and administration hub for Zanzibar. However, it is a rapidly growing city that is prone to extreme rainfall flooding, which closed schools and other facilities in 2017 and 2020.
In response to these issues, the World Bank has formed a partnership with the UK Foreign, Commonwealth and Development Office, and the establishment of the Tanzania Urban Resilience Program (TURP). This is a 5-year trust fund with resources from the UK's International Climate Fund for Bank executed activities to support climate resilience dialogue and training for Tanzanian cities and includes provision for recipient executed risk mitigation investments. The main objective of the TURP is to support national and local governments in Tanzania to strengthen the management of climate risk in cities. Through the Program, Tanzania will further address challenges associated with the generation of climate risk data and information for urban management, the application of urban and land use planning tools for risk reduction, and the identification and construction of climate smart infrastructure.
TURP is currently preparing project financing for transformative flood control investments such as for Dar es Salaam's Msimbazi River, as well as establishing the Resilience Academy to foster knowledge and information on urban resilience and initiate tools to aid decision-makers and stakeholders in managing risk. One such initiative is this Flood Risk Dashboard that was developed by Deltares of the Netherlands, Upande of Kenya and OPM of Tanzania.
The dashboard and analysis that it contains provides the Government and people of Dar es Salaam and Zanzibar with the evidence-based data, information and analysis tools that will help significantly reduce the negative impacts of flooding. The analysis contained in the dashboard includes the quantification of flood risks from all sources of flooding, including rainfall, rivers and any coastal or tidal influence on the flooding. It includes the analysis and quantification of the hazard, exposure and vulnerability of all significant human, social, environmental and material assets. The hazard relating to the location, severity and likelihood of flooding occurring at a particular location. The exposure relating to the human, social, environmental or material assets, such as homes, businesses, services and infrastructure that might be in harms way. Whilst vulnerability of those human, social, environmental or material assets relates to how severely they may be affected or their ability to recover from flooding.
The dashboard is intended to allow users to access, visualize, interrogate and analyse the hazard and risk data for Dar es Salaam and Zanzibar City. This use may pertain to future city planning, disaster response planning and infrastructure planning amongst other use cases.
The analysis that the dashboard contains is based on a chain of data analysis and modelling to simulate the flooding characteristics and the resulting risk. The primary data used in the analysis was the rainfall data, supplied by the Tanzania Meterological Agency (TMA) and Trans-African Hydro-Meteorological Observatory (TAHMO). Whilst the coastal tide and surge data was provided by the Global Tide and Surge Model (GTSM). The geospatial data used to set up the analysis, such as the terrain and city mapping was sourced from the Resilience Academcy Geonode. Socio-economic data somes from the 2012 census and other sources which have been extrapolated to 2020, the base year, and 2050, the future time horizon which will include projections of moderate and extreme climate change.
The TMA and TAHMO rainfall data was analysed to create deterministic return periods which characterise the severity and likelihood of rainstorms for Dar es Salaam and Zanzibar City. These range from the regular 1 in 2 years flood events to the very extreme but infrequent 1 in 1,000 years flood events, which represent a 50 percent and 0.1 percent chance of occurring in any given year. From this statistical analysis a probabilistic event set was assembled, representing a minimum of 10,000 years of storm events for Dar es Salaam and Zanzibar City, which would provide a characterisation of flood hazard with all feasible combinations of different events occuring. These two data sets, the deterministic and probabilistic event sets were coupled with a wflow-sbm rainfall-runoff model and Delft 3D FM hydraulic models to create what are called ‘hazard’ data sets. These data sets are gridded hazard maps of flood water depth, area, duration and velocities.
Spatial data sets of the cities’ communities, buildings, infrastructure will be used to create ‘Exposure’ maps. That is maps of assets and people exposed to flooding. Vulnerability maps, derived from data sets (e.g. building and road taxonomy, but also indicators of social vulnerability), define the conditions determined by physical, social, economic and environmental factors, which increase the susceptibility of a community to the impact of hazards. So for example informal settlements are more vulnerable than concrete buildings. The hazard, exposure and vulnerability data sets are combined in the Delft-FIAT risk model, which calculates the risk for given location by relating flood depths, likelihood and the impact.
Typical indicators of risk include: 'Probable Maximum Loss' (PML), corresponding to likely loss that could occur and 'Average Annual Loss' (AAL), corresponding to the annualized expected loss. These metrics are powerful indicators for long term planning for infrastructure and housing development as well as insurance estimations, moreover they provide a basis for estimating cost-benefit analysis of flood management measures. Other indicators are equally important for disaster risk reduction and response planning, Such as 'number of residents affected by 10 centimetres deep flood water' and 'number of residents affected by 1 meter deep flood water', 'number of schools affected', 'number of hospitals affected' or even more qualitative analysis is possible, such as 'which roads or BRT stops are closed by flood waters' and 'which roads or BRT stops are unaffected by flood waters'. The risk information can be retrieved from the dashboard at the City, Municipality, Ward and Sub-Ward levels.