Fluvial flooding hazard assessment in Northern Italy: potential and informativeness of different geomorphic classifiers

The steady increase of economic losses and social consequences caused by flood events in Europe, as a result of the combined effects of anthropization (e.g. land-use and land-cover changes) and climate change, calls for updated and efficient technologies for assessing pluvial, fluvial and coastal flood hazards and risks. In this context, the EIT-Climate KIC SaferPLACES () project aims at exploring and developing innovative and simplified modelling techniques to assess and map flood hazard and risk in urban environments under current and future climates. Concerning fluvial flooding, detailed inundation maps can be accurately obtained by means of hydrological and hydraulic numerical models, whose application, though, is often very resource intensive. For this reason, consistent and harmonized national flood hazard maps are still lacking in many countries of the world. Several studies have proved that flood-prone areas can be delineated by considering linear binary geomorphic classifiers, which are computed by analysing Digital Elevation Models, DEMs, and whose threshold values are calibrated relative to existing hydraulic flood hazard maps. One of these indices, the so-called Geomorphic Flood Index (GFI), was recently shown to be cost-effective, reliable and efficient for identifying flood-prone areas in several test sites in the United States, Africa and Europe. As part of the activities of SaferPLACES, in this study we test different geomorphic classifiers (GFI included) for the identification of flood-prone areas in a wide area in Northern Italy (c.a. 100000 km2, including Po, Adige, Brenta-Bacchiglione and Reno river basins). We refer to the recently compiled MERIT (Multi-Error-Removed Improved-Terrain) DEM, a 3sec-resolution (~90m at the equator) DEM developed by removing multiple error components from existing spaceborne DEMs. As reference maps for the calibration, we select the flood hazard maps provided by (i) the Italian Institute for Environmental Protection and Research (ISPRA), and (ii) the Joint Research Center (JRC) of the European Commission. Our study confirms the better performances of GFI compared to other geomorphic classifiers, also providing useful information regarding the sensitivity of GFI threshold values relative to different reference hazard maps; it also suggests as a promising avenue for future researches the combination of multiple geomorphic indices through data-driven approaches and artificial intelligence.

Download Full-text

Combination of geomorphic classifiers through Machine Learning-based techniques for flood hazard assessment

10.5194/egusphere-egu21-4723 ◽

2021 ◽

Author(s):

Andrea Magnini ◽

Michele Lombardi ◽

Simone Persiano ◽

Antonio Tirri ◽

Francesco Lo Conti ◽

...

Keyword(s):

Machine Learning ◽

Hazard Assessment ◽

Water Depth ◽

Flood Hazard ◽

Research Centre ◽

Hazard Mapping ◽

Hazard Maps ◽

Flood Hazard Assessment ◽

Flood Prone Areas ◽

Flood Hazard Maps

Every year flood events cause worldwide vast economic losses, as well as heavy social and environmental impacts, which have been steadily increasing for the last five decades due to the complex interaction between climate change and anthropogenic pressure (i.e.&#160;land-use and land-cover modifications). As a result, the body of literature on flood risk assessment is constantly and rapidly expanding, aiming at developing faster, computationally lighter and more efficient methods relative to the traditional and resource-intensive hydrodynamic numerical models. Recent and reliable fast-processing techniques for flood hazard assessment and mapping consider binary geomorphic classifiers retrieved from the analysis of Digital Elevation Models (DEMs). These procedures (termed herein &#8220;DEM-based methods&#8221;) produce binary maps distinguishing between floodable and non-floodable areas based on the comparison between the local value of the considered geomorphic classifier and a threshold, which in turn is calibrated against existing flood hazard maps. Previous studies have shown the reliability of DEM-based methods using a single binary classifier, they also highlighted that different classifiers are associated with different performance, depending on the geomorphological, climatic and hydrological characteristics of the study area. The present study maps flood-prone areas and predicts water depth associated with a given non-exceedance probability by combining several geomorphic classifiers and terrain features through regression trees and random forests. We focus on Northern Italy (c.a. 100000 km2, including Po, Adige, Brenta, Bacchiglione and Reno watersheds), and we consider the recently compiled MERIT (Multi-Error Removed Improved-Terrain) DEM, with 3sec-resolution (~90m at the Equator). We select the flood hazard maps provided by (i) the Italian Institute for Environmental Protection and Research (ISPRA), and (ii) the Joint Research Centre (JRC) of the European Commission as reference maps. Our findings (a) confirm the usefulness of machine learning techniques for improving univariate DEM-based flood hazard mapping, (b) enable a discussion on potential and limitations of the approach and (c) suggest promising pathways for further exploring DEM-based approaches for predicting a likely water depth distribution with flood-prone areas.&#160;

Download Full-text

Delineating flood prone areas using a statistical approach

10.7287/peerj.preprints.1937v2 ◽

2016 ◽

Author(s):

Ivan Marchesini ◽

Mauro Rossi ◽

Paola Salvati ◽

Marco Donnini ◽

Simone Sterlacchini ◽

...

Keyword(s):

Statistical Approach ◽

Learning Algorithm ◽

Flood Hazard ◽

Environmental Data ◽

Training Data ◽

Hazard Maps ◽

Elevation Data ◽

Terrain Elevation ◽

Flood Prone Areas ◽

Flood Hazard Maps

Floods are frequent and widespread in Italy and pose a severe risk for the population. Local administrations commonly use flow propagation models to delineate the flood prone areas. These modeling approaches require a detail geo-environmental data knowledge, intensive calculation and long computational times. Conversely, statistical methods can be used to asses flood hazard over large areas, or to extend the flood hazard zonation to the portion of the river networks where hydraulic models have still not been applied or can be applied with difficulties. In this paper, we describe a statistical approach to prepare flood hazard maps for the whole of Italy. The proposed method is based on a multivariate machine learning algorithm calibrated using in input flood hazard maps delineated by the local authorities and terrain elevation data. The preliminary results obtained in several major Italian catchments indicate good performances of the statistical algorithm in matching the training data. Results are promising giving the possibility to obtain reliable delineations of flood prone areas obtained in the rest of the Italian territory.

Download Full-text

Delineating flood prone areas using a statistical approach

10.7287/peerj.preprints.1937v1 ◽

2016 ◽

Author(s):

Ivan Marchesini ◽

Mauro Rossi ◽

Paola Salvati ◽

Marco Donnini ◽

Simone Sterlacchini ◽

...

Keyword(s):

Statistical Approach ◽

Learning Algorithm ◽

Flood Hazard ◽

Environmental Data ◽

Training Data ◽

Hazard Maps ◽

Elevation Data ◽

Terrain Elevation ◽

Flood Prone Areas ◽

Flood Hazard Maps

Download Full-text

Assessment of Flash Flood Hazard Maps Using Different Threshold Values and Indices Methods

Forum geografic ◽

10.5775/fg.2016.059.s ◽

2016 ◽

Vol XV (Suppl. 2) ◽

pp. 49-54 ◽

Cited By ~ 1

Author(s):

Marius Mătreaţă ◽

Simona Mătreaţă ◽

Romulus-Dumitru Costache ◽

Andreea Mihalcea ◽

Andreea Violeta Manolache

Keyword(s):

Flash Flood ◽

Flood Hazard ◽

Threshold Values ◽

Hazard Maps ◽

Flood Hazard Maps

Download Full-text

Delineating flood prone areas using a statistical approach

10.7287/peerj.preprints.1937 ◽

2016 ◽

Author(s):

Ivan Marchesini ◽

Mauro Rossi ◽

Paola Salvati ◽

Marco Donnini ◽

Simone Sterlacchini ◽

...

Keyword(s):

Statistical Approach ◽

Learning Algorithm ◽

Flood Hazard ◽

Environmental Data ◽

Training Data ◽

Hazard Maps ◽

Elevation Data ◽

Terrain Elevation ◽

Flood Prone Areas ◽

Flood Hazard Maps

Download Full-text

An integrated hydrological and hydraulic modelling approach for the flood risk assessment over Po river basin

10.5194/nhess-2019-356 ◽

2019 ◽

Cited By ~ 2

Author(s):

Rita Nogherotto ◽

Adriano Fantini ◽

Francesca Raffaele ◽

Fabio Di Sante ◽

Francesco Dottori ◽

...

Keyword(s):

Ad Hoc ◽

Flood Hazard ◽

Hydraulic Modeling ◽

Return Periods ◽

Hazard Maps ◽

Joint Research ◽

Po River ◽

Essential Information ◽

Flood Prone Areas ◽

Flood Hazard Maps

Abstract. Identification of flood prone areas is instrumental for a large number of applications, ranging from engineering to climate change studies, and provides essential information for planning effective emergency responses. In this work we describe an integrated hydrological and hydraulic modeling approach for the assessment of flood-prone areas in Italy and we present the first results obtained over the Po river (Northern Italy) at a resolution of 90 m. River discharges are obtained through the hydrological model CHyM driven by GRIPHO, a newly-developed high resolution hourly precipitation dataset. Runoff data is then used to obtain Synthetic Design Hydrographs (SDHs) for different return periods along the river network. Flood hydrographs are subsequently processed by a parallelized version of the CA2D hydraulic model to calculate the flow over an ad hoc re-shaped HydroSHEDS digital elevation model which includes information about the channel geometry. Modeled hydrographs and SDHs are compared with those obtained from observed data for a choice of gauging stations, showing an overall good performance of the CHyM model. The flood hazard maps for return periods of 50, 100, 500 are validated by comparison with the official flood hazard maps produced by the River Po Authority (Adbpo) and with the Joint Research Centre's (JRC) pan-European maps. The results show a good agreement with the available official national flood maps for high return periods. For lower return periods the results and less satisfactory but overall the application suggests strong potential of the proposed approach for future applications.

Download Full-text

Assessing fluvial flooding hazard with a DEM-based Hierarchical Filling-&-Spilling Algorithm: a case study in Northern Italy

10.5194/egusphere-egu21-2894 ◽

2021 ◽

Author(s):

Simone Persiano ◽

Francesca Carisi ◽

Huimin Wang ◽

Valerio Luzzi ◽

Paolo Mazzoli ◽

...

Keyword(s):

Water Depth ◽

Overland Flow ◽

Flood Hazard ◽

Numerical Models ◽

Northern Italy ◽

Point Sources ◽

Economic Losses ◽

Flooding Hazard ◽

Pluvial Flooding

The steady increase of economic losses and social consequences caused by flood events in Europe is triggering the development of updated and efficient technologies for assessing flood hazard over large areas, where detailed hydrologic-hydrodynamic numerical models are resource intensive and therefore scarcely suitable. In this context, the EIT-Climate KIC SaferPLACES (https://saferplaces.co) project aims at exploring and developing innovative and simplified modelling techniques to assess and map pluvial, fluvial and coastal flood hazard and risk under current and future climates, mainly based on LiDAR (Light Detection And Ranging) high-resolution DEMs (Digital Elevation Models) raster-based analysis. Within the SaferPLACES activities, a fast-processing Hierarchical Filling-&-Spilling Algorithm (HFSA), named Safer_RAIN (see Samela et al., 2020; https://www.mdpi.com/2073-4441/12/6/1514/htm), has been recently developed for mapping pluvial flooding in large urban areas by accounting for spatially distributed rainfall inputs and infiltration processes. Although it does not incorporate any detailed description of the dynamics of overland flow and water-depth routing, previous applications have shown Safer_RAIN to be an effective tool for a rapid and consistent identification of pluvial-hazard hotspots under different rainfall and land-use scenarios.Although Safer_RAIN has been conceived for pluvial flooding hazard assessment, its structure suggests its suitability for delineating flooded areas and computing water depth in the aftermath of fluvial inundation (i.e. once the dynamic components of the inundation process become negligible) in predominantly flat floodplains. To this aim, a given flood volume can be assigned to the pixels coinciding with the fluvial flooding point-sources (e.g. simulated levee breach or overtopping) as the input to Safer_RAIN, which is then used for flooding the downstream floodplain portion according to a HSFA approach. We present a first test of the fluvial-application of Safer_RAIN for the case study of the Pisciatello river (Northern Italy, floodplain area of approximately 1300 hectares). Results for different flood scenarios obtained with Safer_RAIN at 1m resolution are compared with the corresponding flooding scenarios simulated with the fully two-dimensional numerical model HEC-RAS at 1m and 5m resolutions. The outcomes of both models are compared in terms of flooded area extent and water depth distribution, highlighting potential and limitations of Safer_RAIN for identifying fluvial flooding hazard.

Download Full-text

Using image-based deep learning to identify river defences from elevation data for large-scale flood modelling

10.5194/egusphere-egu2020-8522 ◽

2020 ◽

Author(s):

Catharine Brown ◽

Helen Smith ◽

Simon Waller ◽

Lizzie Weller ◽

David Wood

Keyword(s):

Deep Learning ◽

Large Scale ◽

Flood Hazard ◽

The United States ◽

Hazard Maps ◽

Learning Techniques ◽

Geographical Regions ◽

River Flood ◽

Flood Maps ◽

Flood Hazard Maps

National-scale flood hazard maps are an essential tool for the re/insurance industry to assess property risk and financial impacts of flooding. The creation of worst-case scenario river flood maps, assuming defence failure, and additional separate datasets indicating areas protected by defences enables the industry to best assess risk. However, there is a global shortage of information on defence locations and maintenance. For example, in the United States it is estimated that there are around 160,000 kilometres (100,000 miles) of defence levees, but the location of many of these is not mapped in large-scale defence datasets. We present a new approach to large-scale defence identification using deep learning techniques.In the generation of flood hazard maps, the elevation depicted in the Digital Elevation Model (DEM) used in the hydraulic modelling is fundamental to determining the routing of water flow across the terrain and thus determining where flooding occurs. The full or partial representation of raised river defences in DEMs affects this routing and subsequently causes difficulty when developing both undefended and defended flood maps. To generate undefended river flood maps these raised defences need to be entirely removed, which requires knowledge of their locations. Without comprehensive defence datasets, an alternative method to identify river defences on a large-scale is required.The use of deep learning techniques to recognise objects in images is fast developing. DEMs and other related datasets can be represented in a similar raster format to images. JBA has developed a successful methodology which involves training a U-Net Convolutional Neural Network, originally designed for image segmentation, to identify raised river defences in DEMs. Using this defence dataset, we have been able to generate true river undefended flood maps for a selection of countries including Italy, Germany, Austria and the US. We present details of the methodology developed, the model training and the challenges faced when applying the model to different geographical regions.

Download Full-text

Design Floods Considering the Epistemic Uncertainty

Water ◽

10.3390/w13111601 ◽

2021 ◽

Vol 13 (11) ◽

pp. 1601

Author(s):

Radu Drobot ◽

Aurelian Florentin Draghia ◽

Daniel Ciuiu ◽

Romică Trandafir

Keyword(s):

Flood Hazard ◽

Epistemic Uncertainty ◽

Statistical Distributions ◽

Hazard Maps ◽

Design Flood ◽

Flood Hydrograph ◽

Recorded Data ◽

Different Shapes ◽

Significant Spread ◽

Flood Hazard Maps

The Design Flood (DF) concept is an essential tool in designing hydraulic works, defining reservoir operation programs, and identifying reliable flood hazard maps. The purpose of this paper is to present a methodology for deriving a Design Flood hydrograph considering the epistemic uncertainty. Several appropriately identified statistical distributions allow for the acceptable approximation of the frequent values of maximum discharges or flood volumes, and display a significant spread for their medium/low Probabilities of Exceedance (PE). The referred scattering, as a consequence of epistemic uncertainty, defines an area of uncertainty for both recorded data and extrapolated values. In considering the upper and lower values of the uncertainty intervals as limits for maximum discharges and flood volumes, and by further combining them compatibly, a set of DFs as completely defined hydrographs with different shapes result for each PE. The herein proposed procedure defines both uni-modal and multi-modal DFs. Subsequently, such DFs help water managers in examining and establishing tailored approaches for a variety of input hydrographs, which might be typically generated in river basins.

Download Full-text