Real-time flood warning systems as part of digital and innovative non-structural solutions have been widely used to prepare decision makers, operators, and affected population to alleviate socio-economic flooding consequences [1]. Many models have been introduced recently to provide more accurate flood forecasts with longer lead times. However, they rely highly on availability of input data which may contain missing values in measurement for one or more timesteps mainly due to wide range of reasons such as random/systematic errors and blunders. Hence, real-time early warning systems cannot be operated properly unless these missing data are properly infilled [2]. Despite data imputation techniques have been mainly employed in pre-processing step of historical data i.e., models training and validation, they have not been properly elaborated in real-time operation practically [3].

This paper aims to propose a new event-based data imputation method for infilling rainfall and water level missing data appearing in real-time operation of flood early warning systems. Event identification is first used to divide the real-time data into the wet or dry weather conditions which then are used for selecting the best strategy of infilling missing data. Imputation decision framework takes advantage of various imputation techniques including t-copula, move-median, and kriging based on external available benchmarks and temporal location of missing data. Proposed methodology is tested in real-world case study of urban drainage system in London, UK. Conventional techniques such as linear regression, kriging, nearest neighbourhood, t-copula, inverse distance, and similar calendar are first compared together and best techniques are then tested with proposed methodology in three real-time scenarios as (1) missing rainfall intensity, (2) missing water level, (3) missing both rainfall and water level. Recurrent neural network model used for flood forecasting and results are demonstrated for the next 3hr-ahead predictions.

Results show the proposed method can reduce root mean square error (RMSE) from 55% to 13%, 43% to 12%, and 97% to 17% for the above scenarios, respectively. Furthermore, using external benchmark data resources, i.e. other near rainfall/water level stations, shows very efficient when missing data appears at early steps of rainfall events where selected conventional techniques suffer from predicting rainfall pattern. Finally, when both water level and rainfall intensity were missing, the proposed imputation method can reduce RMSE from 197mm to 117mm (RMSE was originally 100 for no missing data) for 3hr-ahead predictions. Generally, this study shows the proposed imputation method can better infill the missing data, especially those in the flood event by using correlated data in other weather/gauging stations and flexibility in applying different methods.

References

[1] Piadeh, F., Behzadian, K., Alani, A.M. (2022). Multi-Step Flood Forecasting in Urban Drainage Systems Using Time-series Data Mining Techniques. Water Efficiency Conference, West Indies, Trinidad and Tobago, repository.uwl.ac.uk/id/eprint/9690 [Accessed 31/12/2022].

[2] Piadeh, F., Behzadian, K., Alani, A. (2022). A critical review of real-time modelling of flood forecasting in urban drainage systems. Journal of Hydrology, 607, 127476.

[3] Ben Aissia, M., Chebana, F., Ouarda, T. (2017). Multivariate missing data in hydrology–Review and applications. Advances in Water Resources, 110, pp.299-309.

How to cite: Piadeh, F., Behzadian, K., and Rizzuto, J. P.: Event-based Flood Data Imputation for Infilling Missing Data in Real-time Flood Warning Systems, EGU General Assembly 2023, Vienna, Austria, 23–28 Apr 2023, EGU23-4524, https://doi.org/10.5194/egusphere-egu23-4524, 2023.

Flood modelling, particularly for urban contexts, often suffers from the long computation time and detailed data requirements. Real-time forecasting requires fast, data-efficient tools. A recent break-through in rapid flood simulation has allowed us to develop www.fastflood.org, an open-source platform for near-instant fully spatial flood prediction. It uses the fast sweeping method for elevation model correction, flow accumulation, and estimation of steady-state discharge. Afterwards, a compensated partial steady state is estimated through catchment characteristics and a flood depth field is iteratively back-calculated. We will show that our method, in various real-world tests, achieves over 97% accuracy compared to traditional models in flood extent modelling, while obtaining a speed increase of over 1500 times. Together with recently published global datasets on elevation and land cover, the web-based tool allows for rapid setup and simulation of flood scenarios. Due to the speed of the method, editing of risk-reducing measures (reservoirs, barriers, channel alterations) can be carried out with direct feedback on the consequences for small to medium sized areas. Testing on flash and fluvial flood events show very good promise for early warning methods by linking the flood simulation technique with weather forecasts in real-time. Finally, by employing recent IT technologies such as web-assembly, the model can run completely on the user’s machine, allowing for a sustainable, free model.

How to cite: van den Bout, B., Jetten, V., van Westen, C., and Lombardo, L.: Fastflood.org, an open-source super-fast flood model in the browser, EGU General Assembly 2023, Vienna, Austria, 23–28 Apr 2023, EGU23-3611, https://doi.org/10.5194/egusphere-egu23-3611, 2023.

The prediction, warning, and impact of heavy precipitation events are highly dependent on the available data basis and regional factors. Especially in small catchments, explicit warning is often hampered by the lack of runoff data. The effects of urban flash floods triggered by heavy rain events can also be difficult to predict in catchments of small streams. This greatly increases the risk of unexpected damage in these areas. Prediction systems for small catchments (up to about 200 km²), so far mostly rely only on precipitation forecasting and simple soil water balancing. With this research, a methodology improved by artificial intelligence for the prediction of flood events in small catchments is presented.

The CatRaRe catalog with heavy rainfall events of the last 20 years from the German Weather Service is used as a basis for the investigations of heavy rainfall events in small catchments. However, stationary area parameters and runoff data of streams within the catchment are further missing for a precise area-based prediction. The latter were modeled by a recurrent neural network (RNN) in the form of runoff ratio values. Thus, with additional consideration of the CatRaRe catalog, a step prediction system of selected, small catchments in North Rhine-Westphalia (NRW) is created.

Based on the Digital Elevation Model of NRW (DEM 50), catchment areas are determined and assigned to the events of the CatRaRe catalog. For each selected catchment, the maximum discharge values of a gauging station, within a catchment and given time window after the precipitation event, are investigated. As an additional factor, the ratio between the maximum discharge as well as the mean discharge after a corresponding precipitation event is determined. In catchments without a gauging station and therefore without a runoff time series, an RNN is used to fill data gaps. Recurrent networks using the LSTM (long term short memory) method have already been successfully used to simulate time series, since LSTM networks can model temporal and spatial variability well^1,2.

Further input variables for the RNN are the primary soil type and the size and topography of the respective catchment. Additional information about the pre-rainfall index and the magnitude of individual events is also incorporated as differently (area-dependent) weighted measures. Thus, in the case of a precipitation event, it can be calculated whether critical runoff values or runoff ratio values were observed during similarly intense events in the past. This information is regionalized by means of the RNN and can be transferred to non-gauged catchments.

As the data basis grows, events that have already occurred - in particular the July 2021 event - will be evaluated accordingly in further analyses and the sensitivity of the respective influencing variables for the forecast will be adjusted.

¹Kratzert, F., Klotz, D., Brenner, C., Schulz, K., and Herrnegger, M.: Rainfall–runoff modelling using Long Short-Term Memory (LSTM) networks, Hydrol. Earth Syst. Sci., 22, 6005–6022, https://doi.org/10.5194/hess-22-6005-2018, 2018

²Hu, C.; Wu, Q.; Li, H.; Jian, S.; Li, N.; Lou, Z. Deep Learning with a Long Short-Term Memory Networks Approach for Rainfall-Runoff Simulation. Water 2018, 10, 1543, https://doi.org/10.3390/w10111543

How to cite: Hotzel, A. and Mudersbach, C.: Improved event-based flood warning system for small catchments using artificial intelligence and the CatRaRe catalog, EGU General Assembly 2023, Vienna, Austria, 23–28 Apr 2023, EGU23-1231, https://doi.org/10.5194/egusphere-egu23-1231, 2023.

Predicting flash floods in the arid region of the Arabia Peninsula poses unique challenges to researchers and practitioners due to the generally limited data records and field observations. The rapid onset of these events hinders mitigation measures and limits timely decisions, resulting in fatalities and property losses. To improve our predictive capability, we deployed a flash flood forecasting system that integrates numerical weather forecasts from the Weather Research and Forecasting (WRF) model with a distributed hydrological model, the Coupled Routing and Excess STorage (CREST), and a 2D hydrodynamic model (HEC-RAS). The atmospheric component runs at cloud-resolving scales (1.6 km) to incorporate local features and strong convection. The hydrological and hydrodynamic models run at variable spatiotemporal resolution: the rainfall-runoff generation runs at 500 meter-by-hourly, routing at 30 meter-by-hourly, while the floodplain dynamics are computed at 30 meter-by-hourly. The significant differences in computational demands dictate the domain differences: CREST runs over large natural basins while HEC-RAS runs over small urban sub-basins associated with dense infrastructures and exposure.

The effectiveness of the operational national scale flash flood forecasting system is evaluated in this study for the extreme precipitation event that hit Jeddah on 24 November 2022. The event was the heaviest ever recorded in the area, causing widespread flash floods across Jeddah's urban and rural areas.

The atmospheric component forecast is compared to the NASA satellite precipitation product (IMERG Late) and radar-rainfall estimates that were bias-adjusted based on in situ gauge observations. Since no hydrological observations were available to the authors for this event, discharge obtained from the gauge-adjusted radar-rainfall data, which represents the benchmark precipitation, was used as a reference to assess the skill of the WRF-based flood forecasts. Finally, the effectiveness of the warning system was compared to reported localized flood incidents at the street or neighborhood level by the public ('crowd source').

The results of this study reveal an excellent temporal and spatial agreement between the forecasted precipitation from WRF and the bias-adjusted radar-rainfall estimates. The same conclusions cannot be drawn for the IMERG Late data. The satellite product seems to overestimate precipitation in most cases, which is consistent with the findings of several prior satellite validation studies. Comparing the flood quantiles for the Nov. 24th flood event indicates that the WRF-driven flood peak discharge properties agree with the radar-based ones. The differences between the flood characteristics (hydrographs peak, timing, and flood volume) when using WRF-forecasted versus radar-based benchmark precipitation were also minimal. The simulated flood inundation could capture the broad patterns of inundated areas at the city level: a high degree of agreement was reached, and more than 95% of the reported incidents across the city districts fell within the forecasted high or extreme warnings provided by the operational system on Nov. 23rd, at 12.30; therefore, more than 12 hours ahead. The importance of the study comes from the fact that it provides an effective solution and a state-of-the-art methodology to forecast such types of extreme rainfall events, which can cause major flash floods in the urban areas of Saudi Arabia.

How to cite: Sofia, G., Yang, Q., Shen, X., Mitu, M. F., Patlakas, P., Chaniotis, I., Kallos, A., Alomari, M. A., Alzahrani, S. S., Christidis, Z., and Anagnostou, E.: The operational flash-flood forecasting system for the Kingdom of Saudi Arabia: A case study of the 24th November 2022  flash flood in Jeddah, EGU General Assembly 2023, Vienna, Austria, 23–28 Apr 2023, EGU23-7434, https://doi.org/10.5194/egusphere-egu23-7434, 2023.

The accurate prediction of runoff features such as water level and flow is valuable for planning and operation of urban drainage systems (UDS), especially for appropriately acting as flood control mechanisms during extreme rainfall events which are constantly impacted by climate change variables [1]. In addition, cost-effective design, and operation of flood control measures such as smart UDS require highly accurate rainfall predictions across the catchment area, i.e., intensity and duration [2]. Furthermore, sufficient lead time is needed to activate the control mechanisms on the UDS without affecting the accuracy of the predictions. It seems that the emerging use of satellite precipitation products (SPPs) is promising for obtaining predictions with longer lead times [3]. Hence, more exploration of potential runoff predictions by using SPPs is worth investigating to achieve a more accurate and longer lead time.

This study employs a type of SPPs i.e., global precipitation measurement-integrated multi-satellite retrieval product (GPM-IMERG) to predict rainfall-runoff duration, peak and volume, as well as changes in flow over the course of the event at 30-minute intervals. In order to train and validate the machine learning model, the data from GPM-IMERG V06 was merged with ground data from the catchment precipitation gauge and flow sensor. The methodology is demonstrated by its application to the rainfall-runoff modelling of a real-world small urban sub-catchment area and its performance is evaluated by comparing it with the runoff predictions from physically based simulation models [4].

Results show that while using SPPs solely can provide accurate predictions, significant improvement can be obtained when this data is integrated with ground monitoring data. The model output can be utilised for better design, planning and management of UDS technologies as flood control tools and consequently real-time operation of UDS in urban flooding.

[1] Ferrans, P., Torres, M., Temprano, J., Sánchez, J., (2022). Sustainable Urban Drainage System (SUDS) modelling supporting decision-making: A systematic quantitative review. Science of The Total Environment. 806(2), 150447.

[2] Guptha, G., Swain, S., Al-Ansari, N., Taloor, A., Dayal, D. (2022). Assessing the role of SuDS in resilience enhancement of urban drainage system: A case study of Gurugram City, India. Urban Climate, 41, 101075.

[3] Piadeh, F., Behzadian, K., Alani, A. (2022). A critical review of real-time modelling of flood forecasting in urban drainage systems. Journal of Hydrology, 607, 127476.

[4] Broekhuizen, I., Leonhardt, G., Marsalek, J., & Viklander, M. (2020). Event selection and two-stage approach for calibrating models of green urban drainage systems. Hydrology and Earth System Sciences, 24(2), 869–885.

How to cite: Girotto, C., Piadeh, F., Behzadian, K., Zolgharni, M., Campos, L., and Chen, A.: Role of Satellite Precipitation Products in Real-Time Predictions of Urban Rainfall-Runoff by Using Machine Learning Modelling, EGU General Assembly 2023, Vienna, Austria, 23–28 Apr 2023, EGU23-10211, https://doi.org/10.5194/egusphere-egu23-10211, 2023.

Due to climate change, heavy rainfall events are occurring more frequently and at a higher rate. At the same time, the sewer system is often not designed to handle such a large amount of precipitation, which often leads to flooding of certain critical areas. However, modern sensor technology for monitoring the filling and flow rates in the sewer system, as well as for monitoring and forecasting rainfall probabilities, including quantity information, form a good basis for developing modern multifunctional early warning systems, which can serve as an aid for decisions on action. Urban water drainage systems represent complex networks with nonlinear dynamics and different types of interactions. This yields an involved modelling problem for which different off-line simulation approaches are available. Nevertheless, these approaches cannot be used for real-time simulations, i.e., running in parallel to weather now- and forecasts and enabling the monitoring and automatic control of urban water drainage systems. Alternative approaches, used commonly for automation purposes, involve parameterized linear delay systems, which can be used in real-time but lack the necessary level of detail, which is required for adequate flood risk prognostics. Given this setup, an approach for the effective modelling of detailed water drainage systems for real-time applications implemented with the open-source Storm Water Management Model (SWMM) software is addressed and exemplified for a part of the water drainage system of the city of Flensburg in northern Germany. Additionally, a freely available early-warning system prototype is introduced and used to combine weather forecast information on a 2-h prediction horizon with the developed model and available measurements. This prototype is subsequently used for data assimilation using the ensemble Kalman filter (EnKF) for the considered area in Flensburg. The project presented here is part of the NEPTUN project. NEPTUN is financed by Interreg Deutschland-Danmark with means from the European Regional Development Fund. 

How to cite: Høegh Ravn, N., Baumann, H., and Schaum, A.: Urban Flood Forecast using an Open Source Coupled 1D1D SWMM Model for Real-Time Applications, EGU General Assembly 2023, Vienna, Austria, 23–28 Apr 2023, EGU23-15800, https://doi.org/10.5194/egusphere-egu23-15800, 2023.

 Overflow forecasting in early warning systems is acknowledged as an essential task for devastating urban flood risk management. Although many machine learning models have been developed recently to forecast water levels in urban drainage systems (UDS), they usually require big and accurate data resources [1]. Alternatively, ensemble data mining models are becoming more popular, in which time-series numerical data are turned into the categorised features that classify wet weather conditions as two classes of overflow and non-overflow conditions [2]. However, the concept of time-series ensemble modelling i.e., blending different data mining techniques for predictions with different timesteps is still new [3]. Furthermore, the application of more advanced models, particularly multi-blending in these types of ensemble modelling requires more investigation. This study aims to introduce a novel multi-stacking model that integrates different decision tree frameworks by developing various base weak learner data mining techniques and associated base model performance indicators in the process of time-series blending of pre-trained stacked ensemble models. The performance of this new approach is compared by several previously developed ensemble models [2] through confusion matrix performance criteria, including hit rate, overestimation, and underestimation. This method is demonstrated by its application to a real case study of UDS located in the northwest of London for performance assessment up to 5hr ahead (i.e., 20 timesteps with 15-min intervals). In total, 140 base-models and 20 stacked models were developed that are stored in the data warehouse to use as real-time early-warning flood overflow forecasting for this case study. These developed models were used through introduced decision three framework that specified stacking blending methodology. Results show that while base models and stacked models suffer from high miss rate, especially for forecasting more than 3hrs ahead (more than 50%), the proposed multi-stacking model could perfectly maintain the miss rate (i.e., sum of over- and under-estimations) of up to 4hr-ahead predictions less than 10%, but this rate dropped to nearly 30% for 5hr-ahead predictions. However, the rate of overflow forecasting remained acceptably near 80% whereas it is recorded to less than 58% for other benchmark models. Using different decision frameworks for determining importance of each stacked model in blending mode of multi-stacking method shows could reduce errors in forecasting rate and take advantage of each model in real-time early warning urban flood forecasting.

References

 [1] Piadeh, F., Behzadian, K., Alani, A. (2022). A critical review of real-time modelling of flood forecasting in urban drainage systems. Journal of Hydrology, 607, 127476.

[2] Granata, F., Di Nunno, F., de Marinis, G. (2022). Stacked machine learning algorithms and bidirectional long short-term memory networks for multi-step ahead streamflow forecasting: A comparative study, Journal of Hydrology, 613(A), 128431.

[3] Piadeh, F., Behzadian, K., Alani, A.M. (2022). Multi-Step Flood Forecasting in Urban Drainage Systems Using Time-series Data Mining Techniques. Water Efficiency Conference, West Indies, Trinidad and Tobago. repository.uwl.ac.uk/id/eprint/9690 [Accessed 31/12/2022].

How to cite: Piadeh, F., Behzadian, K., Chen, A. S., Campos, L. C., and Rizzuto, J. P.: Real-time flood overflow forecasting in Urban Drainage Systems by using time-series multi-stacking of data mining techniques, EGU General Assembly 2023, Vienna, Austria, 23–28 Apr 2023, EGU23-8574, https://doi.org/10.5194/egusphere-egu23-8574, 2023.

Natural disasters are known to cause widespread and severe damages all over the world annually. Flood events are responsible for economic and human life losses[1]. One of the most important indicators of the damage level in a flood crisis is the number of casualties. This index is evaluated annually in all countries based on natural disasters. Studies indicate that the death rate caused by floods correlates with countries' development over time [2]. In the present study, quantitative values of three sustainability indicators were extracted in the Czech Republic, Iran, and the United States between 1990 and 2020. These indicators are the Human Development Index (DPI), Gross Domestic Product(GDP), and Climate-Change Impacts (CCI), representing the Social, Economic, and Environmental aspects of sustainable development, respectively.  Then, the mathematical relationships between the development indicators and the number of human losses caused by disasters were evaluated using statistical distributions based on time series. In the final step, using Artificial Intelligence (AI) methods, including Artificial Neural Network (ANN), Adaptive Neuro-Fuzzy Inference System (ANFIS), and Random Tree (RT), a prediction of the number of potential fatalities per natural disaster was obtained. The outcomes showed that each country's deaths caused by natural disasters could depend on different parameters and impact coefficients. In addition, the ANFIS algorithm, with more than 98% accuracy, has the most efficiency in determining the severity of the event. With the help of this AI system, it is possible to evaluate society's behavior and its resilience against floods from a holistic viewpoint[3].

How to cite: Naghedi, S. R., Huang, X., and Gheibi, M.: A smart dashboard for forecasting disaster casualties: An investigation from sustainable development dimensions, EGU General Assembly 2023, Vienna, Austria, 23–28 Apr 2023, EGU23-17237, https://doi.org/10.5194/egusphere-egu23-17237, 2023.

One of the most effective parameters in hydrologic modeling and water resource management issues such as flood warning and real time control of urban drainage systems is short term rainfall forecasting accuracy [1]. In recent decades, developing real-time urban flood forecasting models has been investigated through various studies and different machine and deep learning models have been applied as a prediction model [2]. For instance, in [3], the role of short-term rainfall forecasts in predictive real-time optimal operation (PRTOP), for five adaptive PRTOP models, have been compared.

In this study, based on the result of the [1], one of the superior rainfall forecasting models, the machine learning-based particle swarm optimization (PSO)-support vector regression (SVR) rainfall forecasting approach, is used to develop a 15-minute ahead forecast model of rainfall depth. The PSO-SVR model is linked then to the harmony search (HS)-storm water management model (SWMM) as an optimization-simulation model in a predictive real-time operation model to minimize the flood volume, objective function = min (flood volume), at the control point of a portion of an urban drainage system in Tehran, Iran. Subsequently, the effect of integrating forecasting model with the simulation-optimization model (HS-SWMM) has been examined.

The application of the proposed real-time operation approach through optimizing the operation of the system for eight selected rainfall events, each of them has been selected from different classes, reveals its outperformance over a reactive real-time operation model (RTOP) by decreasing the flood volume at the control point up to 7.5%. 

Keywords: Urban Drainage Systems, Short-term Rainfall Forecasting, Real-time Operation, Machine Learning.

[1] Adaryani, F. R., Mousavi, S. J., & Jafari, F. (2022). Short-term rainfall forecasting using machine learning-based approaches of PSO-SVR, LSTM and CNN. Journal of Hydrology, 614, 128463.

[2] Piadeh, F., Behzadian, K., & Alani, A. (2022). A critical review of real-time modelling of flood forecasting in urban drainage systems. Journal of Hydrology, 127476.

[3] Jafari, F., Mousavi, S. J., & Kim, J. H. (2020). Investigation of rainfall forecast system characteristics in real-time optimal operation of urban drainage systems. Water Resources Management, 34(5), 1773-1787.

How to cite: rezaie Adaryani, F., Mousavi, S. J., and Jafari, F.: PSO-SVR Rainfall Forecast-Assisted Real-time Optimal Operation of Urban Drainage Systems, EGU General Assembly 2023, Vienna, Austria, 23–28 Apr 2023, EGU23-4251, https://doi.org/10.5194/egusphere-egu23-4251, 2023.

The release of dangerous chemicals in flood crises is a common and recurring phenomenon. Because due to the breakdown of the infrastructure in the year, the possibility of release of dangerous pollutants is expected more than before [1]. The present research has chosen a chemical factory with Glycerin tanks as a case study in the city of Liberec (Czech Republic). As acute exposure to these compounds leads to effects such as headaches, dizziness, bloating, nausea, vomiting, thirst, and diarrhea [2], the control of this pollution can significantly reduce the health risk impacts especially during floods. In the first step of this study, all the physics coordinates of the reservoirs were examined and evaluated in different conditions. In the next step, according to the experimental calculations, the volume of Glycerin release from the reservoirs and in the flood flow was evaluated. Risk analysis was done using the HAZard & OPerability study (HAZOP) technique. Risk Number in the HAZOP method is computed based on two factors containing Risk Possibility (RP) and Risk Intensity (RI) [3]. Both the values are determined according to prediction of Adaptive Neuro Fuzzy Inference System (ANFIS) [4] based on previous studies in different countries. The results demonstrated that the integrated HAZOP-ANFIS model has different performance in various flood flow conditions.

RP value is predicted abased on three parameters include; rainfall value, mass of contaminant, and flood flow. Likewise, the RI is estimated in ANFIS method according to self-assimilative factor and continuity of floods [4]. The computations demonstrated that ANFIS technique has more than 0.9 correlation coefficient for prediction of both flood risk factors (RP and RI). Likewise, the sensitivity analysis of the prediction system is examined as per all the declared physical features which effects on RP and RI. Also, it should be mentioned that in the Glycerin content is in the range of 45%- 65% in the case study. Numerical analysis illustrated the performance of designed framework has more efficiency in the higher concentrations of the contamination. The suggested structure can be used as an early qualitative framework for acute effects of hazardous material emissions.

Keywords: HAZOP risk assessment; ANFIS; Flood; Glycerin emissions; Chemical plant

References:

[1] Yard, E.E., Murphy, M.W., Schneeberger, C., Narayanan, J., Hoo, E., Freiman, A., Lewis, L.S. and Hill, V.R., 2014. Microbial and chemical contamination during and after flooding in the Ohio River—Kentucky, 2011. Journal of Environmental Science and Health, Part A, 49(11), pp.1236-1243.

[2] Dunjó, J., Fthenakis, V., Vílchez, J.A. and Arnaldos, J., 2010. Hazard and operability (HAZOP) analysis. A literature review. Journal of hazardous materials, 173(1-3), pp.19-32.

[3] Zabihi, O., Siamaki, M., Gheibi, M., Akrami, M. and Hajiaghaei-Keshteli, M., 2023. A smart sustainable system for flood damage management with the application of artificial intelligence and multi-criteria decision-making computations. International Journal of Disaster Risk Reduction, 84, p.103470.

[4] Akbarian, H., Gheibi, M., Hajiaghaei-Keshteli, M. and Rahmani, M., 2022. A hybrid novel framework for flood disaster risk control in developing countries based on smart prediction systems and prioritized scenarios. Journal of environmental management, 312, p.114939.

How to cite: Moezzi, R., Taghavian, H., Gheibi, M., Koci, J., and Cyrus, J.: Integrated HAZard & OPerability study (HAZOP) and Adaptive Neuro Fuzzy Inference System (ANFIS) as an early alarm framework for Glycerin emission control of a chemical plant during floods: A case study of Liberec city, Czech republic, EGU General Assembly 2023, Vienna, Austria, 23–28 Apr 2023, EGU23-17416, https://doi.org/10.5194/egusphere-egu23-17416, 2023.