Toward Global Stochastic River Flood Modeling

Abstract

Global flood models integrate flood maps of constant probability in space, ignoring the correlation between sites and thus potentially misestimating the risk posed by extreme events. Stochastic flood models alleviate this issue through the simulation of flood events with a realistic spatial structure, yet their proliferation at large scales has historically been inhibited by data quality and computer availability. In this paper, we show, for the first time, the efficacy of modeled river discharge reanalyses in the characterization of flood spatial dependence in the absence of a dense stream gauge network. While global hydrological models may show poor correspondence with absolute observed river flows, we find that the rate at which they can simulate the joint occurrence of relative flow exceedances at two given locations is broadly similar to when a gauge‐based statistical model is used. Evidenced over the United States, flood events simulated using observed gauge data from the U.S. Geological Survey versus those generated using modeled streamflows have similar (i) distributions of site‐to‐site correlation strength, (ii) relationships between event size and return period, and, importantly, (iii) loss distributions when incorporated into a continental‐scale flood risk model. Extremal dependence is generally quantified less accurately on larger rivers, in arid climates, in mountainous terrain, and for the rarest high‐magnitude events. However, local‐scale errors are shown to broadly cancel each other out when combined, producing an unbiased flood spatial dependence model. These findings suggest that building accurate stochastic flood models worldwide may no longer be a distant aspiration.

LISFLOOD-FP 8.0: the new discontinuous Galerkin shallow-water solver for multi-core CPUs and GPUs

“…For the catchment-scale flood simulation, the DG2-GPU runtime was less competitive due to widespread overland flow, involving frictional forces acting on thin water layers, which imposed an additional time step restriction in the current DG2 formulation. It is expected that this restriction could be lifted by formulating an improved DG2 friction scheme based on the finite-volume friction scheme of Xia and Liang (2018). Overland flow does not feature in the EA benchmark tests, where DG2-GPU runtimes remain competitive, being only 5-8 times slower than ACC on the same grid.…”

“…Hit rate (H), false alarm ratio (F) and critical success index (C) for the DG2 and FV1 predictions of maximum flood extent calculated against the reference solution of the ACC solver at x = 10 m. tion on the time step size to maintain stability in the discharge slope coefficients. This challenge has recently been addressed in finite-volume hydrodynamic modelling using an improved friction scheme that calculates the physically correct equilibrium state between gravitational and frictional forces (Xia and Liang, 2018). Extending this friction scheme into a discontinuous Galerkin formulation is expected to al-leviate the time step restriction and reduce DG2 solver runtimes for overland flow simulations.…”

Multisourced Flood Inventories over the Contiguous United States for Actual and Natural Conditions

“…The gauge-based flood events are regarded as "confirmed" events. For headwater sub-catchments (upstream to Strahler Order 2), a simulated flood event will be determined as a "confirmed" event, if it corresponds well to an immediate downstream "confirmed" flood event according to flood timing, because a downstream event is likely caused by its immediate upstream flood event (Wing et al, 2020). For higher order subcatchments, a simulated flood will be recorded as a "confirmed" event, if it corresponds to a "confirmed" event in its immediate upstream/downstream subcatchment with the same Strahler Order, or its lower order sub-catchment in the major upstream tributary.…”

Urbanizing the floodplain: Global changes of imperviousness in flood-prone areas

“…On the other hand, GAUD implicitly incorporates (via data fusion) multiple global urbanization datasets while the model-derived floodplain delineation is unique in terms of spatial coverage, resolution, and expicit accounting of flood risk. Therefore, we only used single datasets for mapping the floodplain and urbanization but report the results in a regional context (e.g., per country) to abate the effects of local-scale errors 23 .…”

“…On the other hand, GAUD implicitly incorporates (via data fusion) multiple global urbanization datasets while the model-derived floodplain delineation is unique in terms of spatial coverage, resolution, and expicit accounting of flood risk. Therefore, we only used single datasets for mapping the floodplain and urbanization but report the results in a regional context (e.g., per country) to abate the effects of local-scale errors 23 .As of 2015, 16.2% of the world's urban area was within the 100-year floodplain (105,657 km 2 ). The extent of urban development on the 100-year floodplain was similar for each continent with the notable exception of Asia: 12.0% and 12.7% for North and South America, 7.9% for Oceania, 12.5% for Europe, 13.1% for Africa, and 22.7% for Asia.…”