Need for judicious selection of runoff inputs in a global flood model

Bibliographic Data

ID15545442
AuthorsJayesh Parmar (0000-0001-6992-3549, Indian Institute of Technology Bombay, corresponding author), Mohit Prakash Mohanty (0000-0001-7783-2073, Indian Institute of Technology Roorkee), Subhankar Karmakar (0000-0002-1132-1403, Indian Institute of Technology Bombay)
Year2025
Volume20
Issue2
Pages024032-024032
Publication date2025-01-24
Peer ReviewedYes
Open AccessYes
TypeARTICLE
VenueEnvironmental Research Letters (JOURNAL)
Journal identifiersISSN: 1748-9326 • E-ISSN: 1748-9326
PublisherIOP Publishing (PUBLISHER • GB)
DOI10.1088/1748-9326/adaa89
OpenAlexW4406802952
LanguageEN
References cited65

Numerous flood hazard assessment and risk management studies depend on hydrodynamic flood models, which require detailed inputs. However, these models face challenges when assessing flood hazards and risks at national scales due to the unavailability of input data and high computational demands. Recent advancements in global flood models (GFMs) have emerged as promising solutions. These widely adopted GFMs, capable of producing flood characteristics, require runoff input typically derived from land surface models (LSMs) or global hydrological models (GHMs), which are prone to inherit cascading uncertainties. Moreover, the utilization of a single runoff input into a GFM can produce biased and misinterpreted flood hazards due to underestimation or overestimation of GFM outputs. To highlight these implications, the present study examines GFM simulations forced with eight state-of-the-art model runoff datasets, including LSMs, GHMs, and reanalysis observations, uncovering unsafe inter-model flood depth variation (IMDV). Focusing on the flood-prone Mahanadi River Basin (MRB) of India, the study observes that IMDV surpasses the self-help range of humans (0.2 m) for 65% of the MRB region, and exceeds human and vehicle safety thresholds (2 m) for 15% of the region, based on four past flood events from the Dartmouth Flood Observatory. These regions exhibiting high IMDV overlap with densely populated areas, potentially affecting 1.66–3.65 million people. Thus, the injudicious use of runoff in GFM for flood disaster planning can lead to inaccurate flood hazard and risk assessments, significantly affecting populous regions. An alternative approach is recommended, advocating for the use of multiple simulations incorporating diverse runoff datasets. This approach would generate conservative and optimistic flood scenarios, leveraging each model’s strengths. Such comprehensive hazard scenarios would enhance flood management and decision-making for policymakers by addressing the uncertainty and providing possible impacts through risk assessments

Flood myth · Geography · Geotechnical engineering · Hydrology (agriculture · Model selection · Selection (genetic algorithm · Surface runoff · Computer Science · Environmental Science · Flood Risk Assessment and Management · Hydrology and Drought Analysis · Hydrology and Watershed Management Studies · Artificial Intelligence · Ecology · Geology

Tools

Open DOIOpen Access