Development of an ANN-Based Urban Flood Alert Criteria Prediction Model and the Impact of Training Data Augmentation

Urban flooding occurs during heavy rains of short duration, so quick and accurate warnings of the danger of inundation are required. Previous research proposed methods to estimate statistics-based urban flood alert criteria based on flood damage records and rainfall data, and developed a Neuro-Fuzzy model for predicting appropriate flood alert criteria. A variety of artificial intelligence algorithms have been applied to the prediction of the urban flood alert criteria, and their usage and predictive precision have been enhanced with the recent development of artificial intelligence. Therefore, this study predicted flood alert criteria and analyzed the effect of applying the technique to augmentation training data using the Artificial Neural Network (ANN) algorithm. The predictive performance of the ANN model was RMSE 3.39-9.80 mm, and the model performance with the extension of training data was RMSE 1.08-6.88 mm, indicating that performance was improved by 29.8-82.6%.

Study of Flood Disaster Mitigation Analysis for Transportation Routes in Panakukkang District, Makassar City

This study conducted an analysis study of flood disaster mitigation for transportation routes in the Panakukkang district of Makassar City. By using ArcGis software, the results of the simulation of safe and vulnerable zone levels based on color indicators are known. There are 5 villages in Panakukkang District which are flood safe zones, with the number of evacuation sites, namely 21 buildings. Then there are 4 villages which are flood alert zones with 2 evacuation sites, 2 buildings. On the first evacuation route there are 8 reference points namely Reference Point C with the distance to the nearest evacuation site 3.22 km and a travel time of 64.3 minutes. Then the reference point A with a distance to the nearest evacuation site is 2.85 km and a travel time of 57 minutes. While the reference point F is the closest point to the nearest evacuation distance 0.71 km and the travel time is 14.2 minutes. All these reference points require travel speeds of 3 km / h on foot. On the second evacuation route there are 6 Reference Points namely reference point A with distance to the nearest evacuation point 1.94 km and travel time 38.8 minutes, reference point E with distance to nearest evacuation location 1.23 km and travel time 24.6 minutes. Then at the reference point C is the closest point to the nearest evacuation distance 0.72 km and the travel time is 14.4 minutes.

Preservation storage in a flood damage mitigation effort at the National Library of France

During a flood alert, the decision to evacuate a threatened collection of a library is an important one. If not thought out carefully, a hastily executed move can expose valuable collections to unforeseen threats. Although floods are usually slow to develop in Paris, the decision to make a preventive evacuation must be taken at the appropriate moment, considering the time needed for the relocation, the reality of the threat and the need for service continuity. In the context of its flood protection plan, the National Library of France has conceived a box model that contributes to saving time in case of a flood and prevents damage during an evacuation. Combining accessibility to documents with security requirements, this model can be implemented in different contexts.

MIDAS: A New Integrated Flood Early Warning System for the Miño River

Early warning systems have become an essential tool to mitigate the impact of river floods, whose frequency and magnitude have increased during the last few decades as a consequence of climate change. In this context, the Miño River Flood Alert System (MIDAS) early warning system has been developed for the Miño River (Galicia, NW Spain), whose flood events have historically caused severe damage in urban areas and are expected to increase in intensity in the next decades. MIDAS is integrated by a hydrologic (HEC-HMS) and a hydraulic (Iber+) model using precipitation forecast as input data. The system runs automatically and is governed by a set of Python scripts. When any hazard is detected, an alert is issued by the system, including detailed hazards maps, to help decision makers to take precise and effective mitigation measures. Statistical analysis supports the accuracy of hydrologic and hydraulic modules implemented to forecast river flow and flooded critical areas during the analyzed period of time, including some of the most extreme events registered in the Miño River. In fact, MIDAS has proven to be capable of predicting most of the alert situations occurred during the study period, showing its capability to anticipate risk situations.

Advances in Flood Early Warning: Ensemble Forecast, Information Dissemination and Decision-Support Systems

Floods are usually highly destructive, which may cause enormous losses to lives and property. It is, therefore, important and necessary to develop effective flood early warning systems and disseminate the information to the public through various information sources, to prevent or at least mitigate the flood damages. For flood early warning, novel methods can be developed by taking advantage of the state-of-the-art techniques (e.g., ensemble forecast, numerical weather prediction, and service-oriented architecture) and data sources (e.g., social media), and such developments can offer new insights for modeling flood disasters, including facilitating more accurate forecasts, more efficient communication, and more timely evacuation. The present Special Issue aims to collect the latest methodological developments and applications in the field of flood early warning. More specifically, we collected a number of contributions dealing with: (1) an urban flash flood alert tool for megacities; (2) a copula-based bivariate flood risk assessment; and (3) an analytic hierarchy process approach to flash flood impact assessment.

Solar-powered flood early warning system with short message service (SMS) notifications

<span>Flood is one of the most common hazards in Malaysia. Flood effects can be local, or very large, affecting the neighborhood or community and entire river basins. This flood develops slowly; sometimes over a period of days while sometimes develop quickly in just few minutes. With the real time flood information, it will allow public safety organizations and other emergency managers to effectively plan their resource deployment within the limited time of alert. Hence, this project aims to design the solar powered flood alert warning system by using solar energy as the power supply. This system will send message using GSM to the residents to notify them about the flood occurred. In this project, three LEDs were used to indicate the height of the water levels which are safe, alert and danger conditions. Each of the height have different water level that indicates the level of safety for each condition. </span>

Flood forecasting and alerts in West Africa − experiences from co-developing a pre-operational system at regional scale

&lt;p&gt;Flooding is a rapidly growing concern in West Africa. Several floods have occurred in recent years with severe consequences including loss of lives and damaged infrastructure. Flooding is also projected to increase with climate change. Access to operational forecasts is a critical component in addressing these challenges. This study presents results from our joint efforts to co-design, co-adapt, and co-operate a short- and medium-term operational hydrological forecasting and alert pilot system for West Africa, within the FANFAR project (www.fanfar.eu).&lt;/p&gt;&lt;p&gt;The system has been co-developed through a cycle of workshops, training sessions, and expert exchanges involving representatives from hydrological services, emergency management agencies, river basin organisations, and expert agencies in 17 countries in West and Central Africa. Multi-criteria decision analysis was employed to clarify and prioritize system objectives and configurations. We found that the most highly prioritized objectives were: high accuracy, clear flood risk information, reliable access, and timely production and distribution of the information. Our agile development approach also provided ample opportunities to focus development efforts on the most highly prioritized components, and incorporate stakeholder feedback in the development process.&lt;/p&gt;&lt;p&gt;The system is built on an ICT cloud platform that employs a daily forecasting chain including meteorological reanalysis and forecasting, data assimilation of gauge observations and satellite altimetry, hydrological initialisation and forecasting, flood alert derivation, and distribution through e-mail, SMS, web visualisation and API. The system is designed to enable multiple configurations and integration of several information sources (e.g. different hydrological models, observations, flood hazard thresholds etc.). We present the system configurations, stakeholder-driven adaptations, challenges, and current forecast performance. To our knowledge, the FANFAR system constitutes a significant advancement toward the vision of achieving efficient flood management in West Africa.&lt;/p&gt;

Improvement of Urban Flood Alert Criteria Prediction Model based on Neuro-Fuzzy Initial Function and Training Data

In Korean metropolitan areas, the high density of residential and commercial buildings, highly impervious surfaces, and steep slopes contribute to floods that can occur within a short duration of heavy rainfall. To prepare for this, advance warning measures based on accurate flood alert criteria are needed. In our previous study, we demonstrated the applications of a Neuro-Fuzzy model that considersthe characteristics of the basin to predict flood alert criteria in areas with no damage. However, as the number of learning materials are low, at 27, the evaluation and verification of the model has not been sufficiently accomplished, and its application is limited. Therefore, in this study, we propose an improved model based on the initializing function of the Neuro-Fuzzy algorithm, the construction of training data, and preprocessing. Compared to the existing model, the improved model reduced the average error by 48.1%~65.4% and the RMSE by 50.7%~60.1%. The new model, when applied to actual floods, showed an improvement of 0.7%~19.1% in accuracy.