Coupled Hydrological And Public Health Risks From Urban Flooding
Integrated Remote Sensing, Machine Learning, And Hydrodynamic–Ecological Modelling
https://doi.org/10.1016/j.jhydrol.2026.135999 <-- shared paper
—
^^ recent overview video created about the research
—
https://doi.org/10.1016/j.wroa.2025.100396 <-- share (earlier) paper
—
H/T @RAHUL DEOPA | Research Scholar (IIT Roorkee)
“… [U]rban floods are not merely hydraulic events; they also transport sewage, pathogens, and other contaminants across streets and communities, leading to significant public health risks…
How do we quantify microbial contamination in near real time during a flood event, when emergency conditions make field sampling unsafe, sparse, or even impossible?...
[The authors] explored whether Earth observation data, combined with machine learning, could bridge this critical monitoring gap. By combining Landsat-derived water surface temperature, machine learning, a coupled MIKE+ Flood–ECO Lab hydrodynamic–ecological model, and Quantitative Microbial Risk Assessment (QMRA), [they] estimated microbial concentrations (𝘌. 𝘤𝘰𝘭𝘪), simulated their fate and transport during floods, and quantified the associated human health risks.
The takeaway: predicting flood risk isn’t just about where the water goes; it’s about what it’s carrying and who it puts in harm’s way. Earth observation and machine learning can help close that gap when it matters most, during the emergency, not weeks after…”
--
“HIGHLIGHTS:
• Remote sensing–based approach developed for monitoring river water temperature.
• Machine learning techniques applied to link river temperature with E. coli concentrations.
• Coupled hydrodynamic–water quality modeling employed to capture spatio-temporal variation of E. coli in floodwaters.
• Quantitative Microbial Risk Assessment (QMRA) conducted to evaluate health risks from E. coli in contaminated floodwaters.
ABSTRACT: Urban flooding poses concurrent threats to infrastructure, human safety, and public health, particularly in densely populated megacities around the globe, where floodwater often acts as a mobile pathway for microbial contamination. This study develops an integrated remote sensing–machine learning–hydrodynamic modelling framework to quantify coupled flood and microbial health risks. Water surface temperature retrieved from Landsat thermal infrared imagery was used as a predictor in a Support Vector Regression model to estimate spatio-temporal variations in E. coli concentration. These predicted microbial loads were subsequently incorporated into a 3-way coupled MIKE + Flood–ECO Lab framework to simulate flood inundation, contaminant transport, and exposure pathways under flood conditions. Hydro-meteorological forcings, including rainfall, streamflow, and urban drainage flow, together with spatial inputs such as topography and land use–land cover, were used to drive the hydrodynamic simulations. The predicted E. coli concentrations were used to parameterize microbial loading, and infection risks for adults and children exposed to contaminated floodwater were quantified using the β-Poisson dose–response model. Experimenting with the proposed framework over Delhi (India)- a city highly prone to flooding and waterborne health hazards, reveals that ∼ 63% of floodplains faced “high” to “very high” flood hazards, with E. coli levels consistently exceeding safe bathing thresholds (500 MPN/100 mL). Estimated annual infection risks surpassed the USEPA benchmark, with children being the most vulnerable. The study provides actionable recommendations for holistic flood management and emphasizes the urgent need to integrate microbial water quality into flood risk assessment and public health planning, particularly in rapidly urbanizing, climate-sensitive regions…”
#publichealth #risk #hazard #watersecurity #Floodrisk #Humanhealthrisk #Urbanflooding #Hydrodynamicmodelling #waterquality #model #modeling #SupportVectorRegression #flood #flooding #urban #city #sewage #pathogens #contaminant #disease #streets #community #quantification #remotesensing #GIS #spatial #mapping #earthobservation #spatialanalysis #water #hydrology #climatechange #extremeweather #spatiotemporal #AI #machineleraning #fateandtransport #hydrodynamic #microbial #rainfall #drainage #streamflow #topography #hydrogeomorphology #Delhi #India #floodplain