Hasil untuk "River protective works. Regulation. Flood control"

Mingxia Lu, Ting Ni, Jiuping Xu

ABSTRACT Effective assessment of urban flood risk is essential for decision‐making under mixed and uncertain information conditions. This study proposes a multi‐indicator hesitant fuzzy assessment framework integrating hazard, exposure, and vulnerability. The framework couples the storm water management model (SWMM) with geographic information system techniques to simulate rainfall runoff processes and characterize spatial variations in flood risk. Quantitative indicators are derived from hydrological simulation and GIS‐based spatial analysis, and an urban flood risk distribution map is generated for the main urban area of Chengdu. The results indicate that high and very high flood risk zones account for 10.18% and 9.43% of the study area, respectively, and are mainly concentrated in districts with high imperviousness, dense populations, and complex transportation networks. Weight analysis shows that vulnerability is the dominant contributor to flood risk, with a weight of 0.412, followed by hazard at 0.362 and exposure at 0.225, highlighting the importance of reducing urban vulnerability and enhancing drainage capacity. The proposed framework provides a practical and transferable approach for urban flood risk assessment and offers scientific support for emergency response planning and disaster prevention strategies in cities with similar characteristics.

Céline Cattoën, Jono Conway, Nava Fedaeff et al.

Abstract Effectively forecasting and communicating flood hazards at national or continental scales is critical to reducing impacts of flooding. Yet, it remains a challenge due to the predominance of ungauged catchments in often complex and steep terrain. We present the development, communication, and evaluation of a national flood awareness system, the Aotearoa (New Zealand) Flood Awareness System, AFAS. Forecasts are produced with an uncalibrated, semi‐distributed hydrological model, driven by a high‐resolution convective‐scale atmospheric model with statistical perturbations in rainfall, soil moisture and baseflow to generate a 50‐member ensemble. We implement a relative flow and flood exceedance threshold framework to evaluate hourly forecasts across six categories from below normal to extremely high. Forecast performance is categorically assessed against observations, for a 2.5‐year reforecast, at 272 sites nationwide, up to 48 h ahead. Overall, AFAS produces skilful streamflow forecasts in catchments with complex topography, even with operational delays ingesting observations. We explore a novel approach to river forecast communication using daily videos. We suggest rethinking large‐scale streamflow forecast communication by balancing a depth with breadth approach (pointwise absolute flows versus distributed relative flows), to raise collective awareness before and during natural disasters. AFAS appears to be the first system producing public‐friendly videos to communicate streamflow forecasts in their topographical context. Future development of AFAS will benefit from a federated approach across national and regional agencies, including sharing of real‐time weather observations, forecasting tools and expertise.

Jens Reinert, Cordula Dittmer, Daniel F. Lorenz et al.

ABSTRACT Extreme heavy rainfall in Western Europe on 13–15 July 2021 caused severe flooding, notably in Germany's Rhineland‐Palatinate and North Rhine‐Westphalia. This study examines Flood Forecasting, Early Warning, and Disaster Response weaknesses during this event, focusing on the city of Stolberg. An interdisciplinary mixed‐methods approach integrated meteorological, hydrological, and social science research. Data included river gages, precipitation measurements, warnings, and 300 documents, with 30 expert interviews. Weaknesses included imprecise meteorological forecasts due to dynamic weather, leading to general warnings without specific impact guidance. Limited flood forecasting hindered local preparation and response, exacerbated by an emergency response system unprepared for the event's scale. The top‐down approach of Flood Forecasting and Early Warning conflicted with the bottom‐up processes of Disaster Response, hampering effective crisis management. The study reveals critical weaknesses and calls for improved forecasting, integrated response plans, communication protocols, and crisis channels to enhance flood resilience. Future research should explore these issues in other extreme flood events and compare international Flood Forecasting, Early Warning, and Disaster Response systems.

Thomas Thaler, Christian Kuhlicke, Thomas Hartmann

Jens Reinert, Elena‐Maria Klopries, Holger Schüttrumpf

ABSTRACT The July 2021 flood event severely affected western Germany, particularly Rhineland‐Palatinate and North Rhine‐Westphalia, with rainfall intensity and water levels exceeding the 100‐year return period in many areas. This study analyzes the performance of flood monitoring and forecasting systems during this extreme event, focusing on the integration of meteorological and hydrological data within detection, monitoring, and forecasting processes. The findings reveal critical challenges in translating meteorological information into actionable flood warnings, including discrepancies between predicted and observed precipitation, underestimation of maximum discharges (exceeding the 100‐year return period by up to 730% in some areas), and the absence of comprehensive flood forecasting systems in key regions. Approximately 60% of the investigated gauging stations experienced exceedances of all defined flood warning thresholds within 6 h or less, highlighting the rapid escalation of the event and the constrained response time for emergency measures. The results of the study underline the need to redesign flood information and warning systems: especially for extreme situations warnings must not be based only on measured data and not on single forecast values but mainly on potential risks and communicated uncertainty/probability.

Ying Gong, Weili Wang, Yufeng Zhou et al.

Abstract To improve the effectiveness of flood disaster relief operations, by ensuring timely and accurate delivery of urgently needed supplies to affected areas, this study focuses on the problem of emergency material distribution during floods. With the objective of minimizing the overall delivery time of emergency materials, we propose a coordinated optimization model that integrates trucks, speedboats, and drones for effective distribution of emergency supplies in flood‐affected areas. To solve this optimization problem, we introduce an improved adaptive large neighborhood search (IALNS) algorithm, which builds on the traditional ALNS framework through refined tuning of deletion and insertion operators. Comparative analyses are conducted with a genetic algorithm, simulated annealing algorithm, and tabu search algorithm. The results reveal that the average performance gap of IALNS compared to these methods is 91.13%, 152.72%, and 16.92%, respectively. The experimental results demonstrate that the efficiency of the proposed model and algorithm in addressing the emergency supply distribution problem during flood disasters, highlighting the superior performance of IALNS. This research contributes to enhancing disaster response strategies, ultimately leading to improved outcomes for flood‐affected communities.

Li Li, Kyung Soo Jun

River flood routing computes changes in shape of a flood wave over time as it travels downstream along a river. Conventional flood routing models, especially hydrodynamic models require high quality and quantity of input data such as measured hydrologic time series, geometric data, hydraulic structures and hydrological parameters. Unlike physically based models, machine learning algorithms, which are data driven models, do not require much knowledge about underlying physical processes and can identify complex nonlinearity between inputs and outputs. Due to the higher performance, less complexity, and low computation cost, novel machine learning methods as a single application or hybrid application were introduced by researchers to achieve more accurate and efficient flood routing. This paper reviews the recent application of machine learning methods in river flood routing.

سید مسعود سلیمان پور, سمیرا زندی‌فر, امید رحمتی et al.

مقدمه امروزه خشکسالی با توجه به کاهش میزان نزولات جوی و افزایش نیاز به محصولات کشاورزی به یک تهدید طبیعی برای امنیت غذایی تبدیل شده است. خشکسالی می‌تواند بخش قابل‌توجهی از کل جمعیت را تحت تأثیر قرار دهد، به‌ویژه ساکنان مناطق خشک و نیمه‌خشک بیشتر در معرض تهدید قرار دارند. هنگامی‌که سامانه‌های آب زیرزمینی تحت تأثیر خشکسالی قرار می‌گیرند، ابتدا تغذیه آب زیرزمینی و بعداً تراز آب زیرزمینی و میزان تخلیه آن کاهش می‌یابد. ویژگی‌های خشکسالی از جمله مدت، شدت و فراوانی در سفره آب زیرزمینی نسبت به سایر انواع خشکسالی‌ها تغییر می‌کند.   مواد و روش‌‌ها با توجه به اهمیت ارزیابی و پایش خشکسالی، وضعیت خشکسالی 12 محدوده مطالعاتی حوزه آبخیز درانجیر در سال آبی 1382-1381 تا 1398-1397 بررسی شد. این حوزه آبخیز در استان‌های یزد و کرمان با مساحت و ارتفاع متوسط به‌‌ترتیب 50736.44 و 1857.90 کیلومتر مربع و متر واقع شده است. مساحت ارتفاعات و دشت در این حوزه آبخیز به‌‌ترتیب 58.99 و 40.99 درصد است. آمار تراز آب چاه‌های مشاهده‌ای در سطح حوضه با مراجعه به سازمان‌های ذی‌ربط اعم از سازمان تحقیقات منابع آب ایران (تماب)، سازمان آب منطقه‌ای استان‌‌های یزد و کرمان تهیه شد. سپس، با استفاده از روش میان‌یابی وزن فاصله معکوس مقادیر متوسط ماهانه تراز آب زیرزمینی از نقشه‌های مورد نظر استخراج شد. بر اساس رابطه‌ شاخص GRI، در محیط برنامه‌ MATLAB، مقادیر خشکسالی آب زیرزمینی محاسبه شد.   نتایج و بحث بر اساس نتایج محاسبه‌ شاخص GRI مشخص شد که شدیدترین خشکسالی‌ها در محدوده‌های کویر درانجیر، بردسیر و قریه‌‌العرب وقوع یافته است که به‌‌ترتیب کمبودی برابر با 81.38، 77.75 و 75.66 دارند. همچنین، طولانی‌ترین خشکسالی‌ها در محدوده‌‌ قریه‌‌العرب وقوع یافته است که تداومی برابر با 121 ماه دارد. لذا، محدوده‌ قریه‌‌العرب، با دارا بودن شدت و تداوم بالای خشکسالی، از شرایط بحرانی‌تر خشکسالی طی دوره‌ آماری مذکور نسبت به دیگر محدوده‌های مطالعاتی برخوردار بوده است. مطابق نتایج فراوانی طبقات شاخص GRI، برای محدوده‌های مطالعاتی حوزه آبخیز درانجیر مشخص شد که بعد از خشکسالی نرمال در تمام محدوده‌ها، خشکسالی ملایم، از بالاترین فراوانی برخوردار بوده است. نتیجه‌‌گیری کمبود و نوسانات شدید نزولات جوی، رخداد پدیده خشکسالی به‌ویژه در مناطق دارای اقلیم خشک و نیمه‌خشک را اجتناب‌ناپذیر می‌کند. از این‌رو، داشتن درک روشنی از اثر و شدت پدیده خشکسالی بر بوم‌سازگان، می‌تواند کمک شایانی به مدیران در برنامه‌ریزی و تصمیم‌گیری بهتر مدیریت منابع حوضه و استفاده بهینه از شرایط برحسب پتانسیل موجود داشته باشد. پایش تغییرات وضعیت سفره‌های آب زیرزمینی به‌‌دلیل خشکسالی و بهره‌برداری بیش از حد سایر حوزه‌های آبخیز کشور و تهیه اطلس نهایی آن توصیه می‌شود. همچنین، لازم و ضروری است که با استفاده از مدل‌های متناسب در این زمینه وضعیت خشکسالی با نتایج و یافته‌های حاصله مقایسه شود. در نهایت، برای مقابله با این پدیده طبیعی توصیه می‌شود برنامه‌های کوتاه و بلندمدت در قالب راهکارهای توسعه سطح زیر کشت محصولات گلخانه‌ای، استفاده از ارقام مقاوم به خشکی و با راندمان مصرف آب بالا، توسعه سامانه‌های تحت‌فشار در آبیاری به‌جای سامانه‌های سنتی در باغ‌‌ها و مزارع و اجرای پروژه‌های مؤثر در افزایش ذخیره منابع آب آبیاری زیرزمینی و سطحی نظیر بندهای سنگ سیمانی، خاکی، پخش ‌‌سیلاب در منطقه مورد مطالعه توصیه می‌شود.

Kim van denHoven, Carla J. Grashof‐Bokdam, Pieter A. Slim et al.

Abstract Coastal flood managers seek to anticipate future flood risk and as a result consider the adaptation of flood defences. Instead of crest heightening, dikes can be adapted to include hydrodynamic reducing vegetated foreshores to form a nature‐based hybrid flood defence, for instance; at managed realignments. In this study we investigated the potential of vegetated revetments as a natural continuous connection between the realigned dike and restored foreshore. We applied the historic grass sod transplantation technique with the aim to improve our understanding of the strength of a transplanted sod revetment. In Living Lab Hedwige‐Prosperpolder, dikes were available for in‐situ experiments during managed realignment preparations. We transplanted grass sods and studied erosion resistance after one growth season. Our results show transplanted sod vegetation continued to grow and started to attach to the clay layer. While erosion occurred under extreme wave impact and overflow, the sod pulling method revealed individual sod strength. In conclusion, sod transplantation is a good technique to source local material for green realigned dike revetments. A vegetated dike revetment can hereby create a natural continuous connection between the realigned dike and foreshore, which benefits flood protection as well as flora and fauna.

Michael V. Martello, Andrew J. Whittle, Hannah R. Lyons‐Galante

Abstract Estimates of flood‐related damages and costs often rely on asset‐specific depth‐damage curves that characterize the fragility of a given asset. To date, there are very few depth‐damage curves that are potentially applicable to rail rapid transit infrastructure, and no studies attempt to construct these relationships specifically these asset classes. Given the lack of empirical performance data or asset‐specific reliability tests, we solicited expert engineering judgment to characterize the fragility of transit assets to saltwater flood exposure. We validate the resulting synthetic depth‐damage relationships via a benchmarking approach and demonstrate consistency with previously published depth‐damage curves for similar asset classes. The solicitation framework presented can easily be extended to other infrastructure assets and systems, potentially serving as a key step toward a more rigorous quantification of the potential risks posed to infrastructure by natural hazards and climate change.

فرهاد میثاقی, پریسا عسگری, مریم نوری

مقدمهوجود آب برای کشاورزی، از اهمیت بالایی برخوردار است و با وجود بحران آبی که همه ‌ساله هم به‌صورت کمی ‌و هم به‌صورت کیفی شدید‌تر می‌شود، باید به‌طور جدی به این مساله توجه شود. منابع آب شامل آب‌های سطحی و زیرزمینی می‌شود که از نظر کیفی، آب‌های سطحی در معرض خطر بیشتری هستند، بنابراین، برای حفظ آن­‌ها باید منابع آلودگی را شناخته و راهکارهای مناسب برای پیشگیری یا رفع این آلودگی‌ها ارائه شود. مواد و روش‌ها در این پژوهش، چرخه انتقال فسفات در حوزه آبخیز زنجان‌رود با استفاده از مدل SWAT شبیه‌سازی شده است. برای واسنجی و اعتبارسنجی از نرم‌افزار SWAT-CUP و مقادیر اندازه‌گیری شده شدت جریان متوسط ماهانه در ایستگاه آب‌سنجی سرچم بین سال‌های 1996 تا 2013، استفاده شده و برای تحلیل حساسیت نیز 26 پارامتر حساس، انتخاب شده است. سه گزینه برای شیوه آبیاری، سه گزینه برای میزان کود مصرفی و دو گزینه تلفیقی تعریف‌ شده است. به‌منظور تحلیل عدم قطعیت از شاخص‌های p-factor و r-factor و برای تحلیل کیفیت نتایج مدل، از دو شاخص ضریب تعیین (R2) و ضریب نش-ساتکلیف (NS)، استفاده شده است.  نتایج و بحثدر مرحله واسنجی رواناب ماهانه در خروجی حوضه، ضرایب r-factor ،p-factor ،R2 ، NS، به‌ترتیب 0.27، 0.11، 0.83 و 0.53 و در مرحله اعتبارسنجی به‌ترتیب 0.60، 0.18، 0.73 و 0.53 به‌­دست آمده است. نتایج نشان داد که با افزایش سطح آبیاری تحت فشار، میزان آلودگی فسفات در خروجی حوضه تغییر چشم‌گیری ندارد. در رابطه با میزان کود، کاهش 50 درصدی مصرف کودهای فسفاته، مقدار فسفات ورودی به رودخانه زنجان‌رود را به میزان 19.2درصد کاهش داده است. از طرفی، افزایش 50 درصدی مصرف کودها، فسفات ورودی را به میزان 17.7 درصد افزایش داده است.  نتیجه‌گیرینتایج حاکی از عملکرد مناسب مدل SWAT و توانایی آن در شبیه‌سازی مذکور بود. همچنین، با تغییر شیوه آبیاری سطحی به تحت فشار و افزایش راندمان آبیاری، تغییر چشمگیری در میانگین میزان فسفات خروجی از سطح حوضه ایجاد نمی‌شود. از طرفی با کاهش مقادیر کود‌دهی و جلوگیری از کود‌دهی بی‌رویه به‌وسیله کشاورزان، به میزان زیادی می‌توان مانع آلودگی منابع آب‌های سطحی و زیرزمینی شد.

Sania Maulida, Myrna Asnawati Safitri, Endra Wijaya

The floods that occurred in several areas of the Special Capital Region of Jakarta Province are influenced by natural and unnatural factors, such as problem in spatial planning and its regulation in legal documents. This study discusses about synchronization problem between several regulations at national and regional level, namely Presidential Regulation Number 60 of 2020 and Regional Regulation of the Special Capital Region of Jakarta Province Number 1 of 2014. This study uses normative legal research method with statute and conceptual approach. The conclusion of this study is that there is still no regulation synchronization at the level of national and regional related to the floods control management. At the national level, Presidential Regulation Number 60 of 2020 already regulated in detail about floods control management through the spatial planning general policy, strategy of implementation, spatial system and structure, and direction of zoning regulation for Jakarta, Bogor, Depok, Tangerang, Bekasi, Puncak, and Cianjur regions. Different from that national level regulation, at the regional level, Regional Regulation of the Special Capital Region of Jakarta Province Number 1 of 2014, related to the floods control management, only focus on how to normalize the river as a part of drainage system in every district in Jakarta Province.

Mitchel Provan, Sean Ferguson, Enda Murphy

Abstract A numerical hydrodynamic model was used to simulate the generation and evolution of storm surges in Atlantic Canada in response to synoptic‐scale surface wind and atmospheric pressure fields. The modelling was conducted as part of a broader initiative to support community‐scale inundation modelling and coastal flood risk assessment for communities located in the Acadian Peninsula region of New Brunswick. The 44 largest storm surge events on record at a tide gauge proximate to the region of interest were simulated using the numerical model. Initially, a comparison between simulated storm surges and peak non‐tidal residuals from tide gauge records showed relatively poor agreement, producing an R2 value of 0.403. Model skill was improved by incorporating the influence of sea ice cover on air‐sea momentum transfer in the hydrodynamic model, and improved correlation with measured residuals was obtained by adding estimates of wave set‐up to the predicted storm surges, ultimately resulting in an R2 value of 0.803. The results of the simulations provided a basis for identifying distinct regional factors affecting storm surges and water level residuals and demonstrated conditions where wave set‐up and sea ice cover play an important role in contributing to extreme high water levels.

Sonam Futi Sherpa, Manoochehr Shirzaei

Abstract Extreme precipitation and flooding often lead to human and economic losses. However, the high‐resolution nationwide flooding exposure data are scarce. Availability of all‐weather space‐borne SAR satellite data can potentially improve the ability to generate high‐resolution flood map extent and exposure globally. In Iran, flooding is a major concern, given the socioeconomic vulnerabilities and increased likelihood of climate extremes. Iran experienced extreme flooding during January to March 2019, attributed to significant precipitation during October 2018 to March 2019, which is well above the long‐term averages for 1999–2019. Using Pettitt and Mann Kendall tests, nationwide precipitation records were identified by significant decreasing and increasing trends in north and south, respectively. Utilizing 673 Sentinel‐1 SAR intensity images, we applied a fast‐marching algorithm for image segmentation in combination with a Bayesian framework to obtain high‐resolution probabilistic flood maps. We found, 22, 9, and 15 states in January, February, and March, respectively, experienced flooding that covered >15% of their area with high flooded area percent in the northwestern and southeastern region. We estimated that >15, >11.32, and >11.33 million people were exposed to floods in January, February, and March, respectively. Our datasets inform flooding models and management efforts under increasing climate extremes and changing land use and cover.

Ockert Jacobus Gericke

Abstract Internationally, the occurrence and frequency of floods, along with the uncertainty involved in the estimation thereof, contribute to the practitioners' dilemma to make a single, justifiable decision when various design flood estimation methods are used. This article presents the further development of a Design Flood Estimation Tool (DFET) using Microsoft Visual Basic for Applications to assess the performance of event‐based design flood estimation methods in 48 gauged catchments in South Africa. The improved DFET proved to be an easy‐to‐use software tool for the rapid estimation and assessment of at‐site design floods in both gauged and ungauged catchments. In using a ranking‐based selection procedure, the Soil Conservation Service, Alternative Rational and Catchment Parameter methods provided the best estimates of the at‐site probabilistic flood peaks, while the Standard Design Flood method proved to be the least appropriate. Since the accuracy and uncertainty associated with each design flood method's key input parameters are unknown when applied in ungauged catchments, the incorporation of an ensemble event approach as part of the DFET calculation routines, is recommended. This will ensure that the key input parameters from an expected range of values are used to achieve probability neutrality between input rainfall and estimated runoff.

Rashed Uz Zzaman, Sara Nowreen, Maruf Billah et al.

Abstract Flood havoc during 2019 in the Sangu River basin caused widespread damage to residents, crops, roads, and communications in parts of hills in Bangladesh. Developing flood hazard maps can play an essential step in risks management. For this purpose, this study assessed 12 hydro‐geomorphological factors, namely, topographic wetness index, elevation, slope, extreme rainfall, land‐use and land‐cover, soil type, lithology, curvature, drainage density, aspect, height above the nearest drainage, and distance from streams. Maps prepared by individual application of the Analytical Hierarchy Process (AHP) and Analytical Network Process (ANP) exhibit validation scores ranging from 0.77 to 0.79. It is found that the ANP‐based model under 1‐day maximum rainfall denotes a reliable hazard map presenting comparable accuracy to the field results. The hazard map under 100‐year return periods shows that a total of 0.71 million population living downstream is prone to “very high” flood because of its lowland morphology, mild slope, and high drainage density. Alarmingly, 39% of roads, 43% of farming lands, and 25% of education buildings are observed to lie in the highest flood‐prone area. Details on subdistrict level exposures have the potential to serve the decision‐makers and planners in site selection for flood management strategies and setting priorities for remedial measures.

Miguel Leal, Eusébio Reis, Susana Pereira et al.

Abstract This study focuses on the physical vulnerability of buildings to flash floods using an indicator‐based methodology. A physical vulnerability index (PhVI) that combines intrinsic vulnerability (IV) of buildings and flash flood intensity (FFI) is proposed. IV evaluates the propensity to suffer damage, resulting from indicators related to building properties. FFI estimates the potential to cause damage, resulting from indicators related to flow parameters. PhVI was applied to a critical section of a small drainage basin in Portugal where flash floods are frequent. Evaluating IV and the intensity of natural hazards is essential in physical vulnerability assessments. This study addresses two problems found in the literature: the lack of flash flood‐dedicated physical vulnerability assessments and the difficulties in assembling building properties and the intensity of natural hazards in a vulnerability index defined from indicator‐based methodologies. PhVI is a useful tool where damage records are rare or non‐existent, allowing the prioritisation of resources and application of local protection measures. This index can be adapted to other study areas and natural hazards, although more research is needed to improve the knowledge on the indicators and weights of IV and FFI.

Shuang Song, Shuai Wang, Bojie Fu et al.

Abstract Collective memory plays a controlling role in adaptation to potential flood risks, by learning from past disasters. However, with little quantitative empirical data, previous socio‐hydrological models have conceptualized the decaying process of flood memory in an oversimple approach. Here, based on survey data of 683 respondents on Ningxia Floodplain, we confirmed that flood memory decays overtime via two channels: oral communication (communicative memory) and physical recording of information (cultural memory). Using the Universal Decay Model (UDM) proposed by previous researchers provides better fitting of results to the decay of flooding memory (adjusted R2 coefficient are 0.97, 0.90, 0.95 when data of all, rural or urban respondents used, respectively) compared with the original exponential model (adjusted R2 coefficient are 0.91, 0.74, 0.59, corresponding). Then, significantly reduced losses for the same flood sequence predicted by integrating the UDM into a socio‐hydrological model by 16% and the differences between different clusters (urban and rural respondents) can even reach 22.81%. These differences suggest that previous socio‐hydrological models have been too simplistic in their conceptualizations of decaying processes associated with collective memory, which may have limited deeper insights into flood risk management.

Mohammad Ahmadlou, A'kif Al‐Fugara, Abdel Rahman Al‐Shabeeb et al.

Abstract Floods are one of the most destructive natural disasters causing financial damages and casualties every year worldwide. Recently, the combination of data‐driven techniques with remote sensing (RS) and geographical information systems (GIS) has been widely used by researchers for flood susceptibility mapping. This study presents a novel hybrid model combining the multilayer perceptron (MLP) and autoencoder models to produce the susceptibility maps for two study areas located in Iran and India. For two cases, nine, and twelve factors were considered as the predictor variables for flood susceptibility mapping, respectively. The prediction capability of the proposed hybrid model was compared with that of the traditional MLP model through the area under the receiver operating characteristic (AUROC) criterion. The AUROC curve for the MLP and autoencoder‐MLP models were, respectively, 75 and 90, 74 and 93% in the training phase and 60 and 91, 81 and 97% in the testing phase, for Iran and India cases, respectively. The results suggested that the hybrid autoencoder‐MLP model outperformed the MLP model and, therefore, can be used as a powerful model in other studies for flood susceptibility mapping.

Marcela Rojas