We propose a probabilistic deep learning approach for the prediction of maximum water depth hazard maps at high spatial resolutions, which assigns well-calibrated uncertainty estimates to every predicted water depth. Efficient, accurate, and trustworthy methods for urban flood management have become increasingly important due to higher rainfall intensity caused by climate change, the expansion of cities, and changes in land use. While physically based flood models can provide reliable forecasts for water depth at every location of a catchment, their high computational burden is hindering their application to large urban areas at high spatial resolution. While deep learning models have been used to address this issue, a disadvantage is that they are often perceived as “black-box” models and are overconfident about their predictions, therefore decreasing their reliability. Our deep learning model learns the underlying phenomena a priori from simulated hydrodynamic data, obviating the need for manual parameter setting for every new rainfall event at test time. The only inputs needed at the test time are a rainfall forecast and parameters of the terrain such as a digital elevation model to predict the maximum water depth with uncertainty estimates for complete rainfall events. We validate the accuracy and generalisation capabilities of our approach through experiments on a dataset consisting of catchments within Switzerland and Portugal and 18 rainfall patterns. Our method produces flood hazard maps at 1 m resolution and achieves mean absolute errors as low as 21 cm for extreme flood cases with water above 1 m. Most importantly, we demonstrate that our approach is able to provide an uncertainty estimate for every water depth within the predicted hazard map, thus increasing the model’s trustworthiness during flooding events.
Abstract. The presence of ephemeral ponds and perennial lakes in the Sudano-Sahelian region of West Africa is strongly variable in space and time. Yet, they have important ecological functions and societies are reliant on their surface waters for their lives and livelihoods. It is essential to monitor and understand the dynamics of these lakes to assess past, present, and future water resource changes. In this paper, we present an innovative approach to unravel the sediment and water balance of Lac Wégnia, a small ungauged lake in Mali near the capital of Bamako. The approach uses optical remote sensing data to identify the shoreline positions over a period of 22 years (2000–2021) and then attributes water surface heights (WSHs) to each observation using the lake bathymetry. We then present a novel methodology to identify and quantitatively analyze deposition and erosion patterns at lakeshores and in lake beds. The method therefore represents a significant advancement over previous attempts to remotely monitor lakes in the West African drylands, since it considers not only changes in water depth to explain recent declining trends in lake areas, but also changes in the storage capacity. At Lac Wégnia, we recognize silting at the tributaries to the lake, but overall, erosion processes are dominant and threaten the persistence of the lake because of progressive erosion through the natural levee at the lake outlet. This factor contributes 66 %±18 % to the decreasing WSH trend, while 34 %±18 % of the dry-season lake level changes are explained by increasing evaporation from the lake and by possibly falling groundwater tables. Due to the decreasing reservoir capacity of the lake, WSHs are declining even in the wet season in spite of positive rainfall patterns.
<p>Water resources in the African Sahel Region are under increasing pressure due to climatic changes, population growth and land degradation. Often, societies rely on surface water from lakes and rivers to sustain their lives and livelihoods. It is therefore essential to monitor and understand the dynamics of these water bodies to assess past, present, and future water resource changes.</p><p>Here we use satellite imagery and altimetry to determine water level and storage changes in small water bodies across the African Sahel. The method consists of detecting the ever-shifting edge of lakes and rivers in Landsat and Sentinel-2 optical imagery and assigning heights to shoreline points using altimetric data from ICESat-2satellite. This so-called &#8220;waterline method&#8221; assumes that the water-land boundary can be regarded as a contour line that connects points of equal elevation. We present novel extension of the waterline method which also allows to identify bathymetry changes over time from shoreline position observations. By tracking the temporal changes of surface water contour shapes, we can quantitatively analyse erosion and deposition processes. Past reservoir capacity changes and water storage variations are thus retrieved from optical remote sensing data, which are available over much longer periods of time and at higher revisiting frequenciesthan altimetry data.</p><p>The operational implementation of the method offers access to the water levels and storage variations of more than 300 water bodies in 10 Sahelian countries over the period 2000-2021. The identified spatio-temporal trends reveal fascinatingly heterogeneous patterns of drying and wetting across the Sahelian zone. Wet-season water level data reveal increasing trends over the last 20 years from West to East. Dry-season water availability then depends to a large degree on storage capacity.</p><p>Finally, we use the method for a detailed attribution analysis to identify drivers of change at Lac W&#233;gnia, a designated RAMSAR site in Mali. The lake is characterized by an alarming decrease of dry-season surface water extent over the last 20 years. We recognize silting at the tributaries to the lake, but overall, erosion processes are dominant and threaten the persistence of the lake because of continuous backward erosion at the outlet of the lake. This explains the decreasing trend in water levels even for the wet-season, in spite of positive rainfall patterns.</p>
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.