Search for a command to run...
Direct and indirect impacts of floods on drinking water services threaten to increase risks of disease outbreaks and may lead to development setbacks in low-resource settings. Especially in complex-emergency settings where data collection remains challenging and infectious disease burdens high, urban flood vulnerabilities are still poorly understood. Our study contributes to addressing this gap by combining various data sources to assess vulnerabilities of the water supply system in the face of extreme flooding in the town of Uvira, located in South Kivu in the Democratic Republic of Congo. Uvira is a town of an estimated 280,000 inhabitants (in 2020), illustrative of a complex emergency setting with limited access to basic drinking water and sanitation and a high cholera disease burden. The town experiences distinct wet and dry periods and is situated on a hilly terrain along the shore of Lake Tanganyika with five rivers flowing through it. In April 2020 the city experienced a catastrophic flood event which affected around 80 000 people and destroyed critical water infrastructure. In this study we used three complementary approaches to study flood events and related drinking water service vulnerability in Uvira: (1) we mapped the extent of three flood events, including the April 2020 event, using high-resolution optical images and open-access optical and synthetic aperture radar (SAR) data from Sentinel-1 and Sentinel-2; (2) we overlaid maps of the water supply infrastructure to identify system exposures to flooding; (3) we carried out a survey-based rapid assessment of 148 households 12 weeks after the April 2020 flood focused on drinking water-related practices. Preliminary results from flood mapping and household survey analysis suggest that households were exposed to flooding in nine out of fourteen districts, mostly in the vicinity of rivers. Critical points of the piped drinking water system affected by the flood included the main water intake for the water supply network, located on the Mulongwe river, which was destroyed and led to a 6-week disruption of the entire drinking water supply service. Around half of the survey participants reported having changed their drinking water source after the April 2020 flood. Despite regular interruptions of water services, storage capacities within households were modest at the time of the survey (median =22L per person). Results from flood extend mapping leveraging open access satellite image data from Sentinel-1 and 2 as well as high resolution optical data before and after extreme flood events, will complement these findings by highlighting neighbourhoods and water collection points which were most severely exposed to the 2020 flood event as well as to two smaller flood events in December 2020 and April 2024 in Uvira. As such, our results demonstrate the feasibility of organising remote research in complex-emergency settings by leveraging electronic data collection tools and satellite data to gain insights into flood vulnerabilities of drinking water services in resource-limited settings. Our results may be used to inform measures for strengthening the resilience of drinking water services in low-resource, data-scarce urban communities in a global context of increasing exposure to extreme flooding.