Article
Engineering, Civil
Maria Kaiser, Stephan Guennemann, Markus Disse
Summary: This study developed a novel methodology based on tree-based classifiers to assess flood susceptibility at a regional scale using spatially distributed and catchment-related factors. The methodology was evaluated in the region of Bavaria (Germany), and all three models performed well, with the CatBoost model achieving the highest performance. It was found to be crucial to consider sample density and coverage of the study area when modeling large territories. The study also proposed an overall susceptibility score for cities based on the generated susceptibility map.
JOURNAL OF HYDROLOGY
(2022)
Article
Geosciences, Multidisciplinary
Romulus Costache, Alireza Arabameri, Ismail Elkhrachy, Omid Ghorbanzadeh, Quoc Bao Pham
Summary: This study evaluated the susceptibility to floods in the Buzau river basin in Romania using 6 machine learning models, with Random Forest showing the highest accuracy among them.
GEOMATICS NATURAL HAZARDS & RISK
(2021)
Article
Geosciences, Multidisciplinary
Manish Singh Rana, Chandan Mahanta
Summary: This study aimed to improve flash flood susceptibility modeling by incorporating ensemble approaches into bivariate and multivariate statistical models. A flash flood and geospatial database were developed, and weights were assigned to influencing factors based on correlation and weight of evidence (WOE) methods. Multiple models were built and validated, with the WOE-ANN model outperforming all machine learning models.
Article
Geography, Physical
Laura Melgar-Garcia, Francisco Martinez-Alvarez, Dieu Tien Bui, Alicia Troncoso
Summary: Floods are devastating weather-induced disasters with significant impact on socio-economic development and the environment. This research compares three convolutional neural networks for predicting flood-prone areas, achieving an impressive accuracy rate of 96.01%. The flood susceptibility maps derived from this study provide valuable insights for local authorities to enhance decision-making and risk management strategies.
INTERNATIONAL JOURNAL OF DIGITAL EARTH
(2023)
Article
Environmental Sciences
Ze Wang, Heng Lyu, Chi Zhang
Summary: This study introduces a semi-supervised graph-structured model, GAT, to overcome the problem of data scarcity in cities and enable flood susceptibility mapping. Results show that GAT performs well in flooded-nonflooded sample classification and generates a rational flood susceptibility map.
GEOCARTO INTERNATIONAL
(2023)
Article
Geosciences, Multidisciplinary
Mahya Norallahi, Hesam Seyed Kaboli
Summary: Machine learning models were used to predict urban flood hazard areas with limited data, showing good prediction performance in the absence of detailed hydraulic-hydrological data.
Article
Geosciences, Multidisciplinary
Mohamed Abdelkareem, Abbas M. Mansour
Summary: Earth observation from space plays a crucial role in characterizing, detecting, and managing natural hazards, such as flash floods. This article focuses on developing a flash flood hazard zone (FFHZ) map using satellite images and GIS analysis. The study identifies vulnerable areas in the Wadi Qena Basin and assesses the impact of floods on New Qena City. The findings highlight the need for protective measures, including dams, reinforced infrastructure, and extending flood canals.
Article
Geosciences, Multidisciplinary
Mehrnoosh Taherizadeh, Arman Niknam, Thong Nguyen-Huy, Gabor Mezosi, Reza Sarli
Summary: This study proposes a framework for mapping flood-prone areas using GIS, remote sensing data, and MCDM techniques. The hybrid MCDM model combines DEMATEL with GIS-based ANP to evaluate flood vulnerability in Golestan province, Iran. Fourteen criteria related to flood potential were considered and analyzed, and the study identified the northern and central parts of the study area as being at higher risk of flooding. The proposed GIS-DANP model provides a valuable tool for flood management and decision-making, aiding in risk reduction and minimizing casualties.
Article
Environmental Sciences
Ahmed M. Al-Areeq, Radhwan A. A. Saleh, Abdulnoor A. J. Ghanim, Mustafa Ghaleb, Nabil M. Al-Areeq, Ebrahim Al-Wajih
Summary: This study used geoprocessing and computational techniques to map flood susceptibility in the Qaa'Jahran watersheds in Dhamar, Yemen. The GWO_LM_ANN model, trained using a hybrid algorithm, outperformed other machine learning models, achieving high precision, sensitivity, specificity, F1 score, accuracy, and AUC. The findings have significant implications for disaster preparedness and response, offering targeted and efficient non-structural solutions.
GEOCARTO INTERNATIONAL
(2023)
Article
Environmental Sciences
Ahmed M. Youssef, Biswajeet Pradhan, Abhirup Dikshit, Ali M. Mahdi
Summary: Geohazard risk is high in Arab countries due to ineffective disaster preparedness measures, mismanagement, lack of public awareness, inadequate funding and lack of stakeholder support. In this study, flood susceptibility modelling was conducted in Egypt using machine learning technique (Support Vector Machine) and deep learning method (Convolutional Neural Networks). The results showed that deep learning technique (CNN) provided better prediction accuracy than machine learning technique (SVM).
GEOCARTO INTERNATIONAL
(2022)
Article
Computer Science, Interdisciplinary Applications
Rohan Singh Wilkho, Nasir G. Gharaibeh, Shi Chang, Lei Zou
Summary: Structured databases on flash flood events lack comprehensive information, especially emerging data such as visual media. To address this issue, a flash flood information retrieval (FF-IR) system was developed, which utilizes machine learning models to automate and improve the information retrieval process. The FF-IR system outperforms direct Google searches by over 100% in terms of F2-score, offering a valuable tool for natural hazard researchers and practitioners to enhance flash flood risk assessments and mitigation planning.
ENVIRONMENTAL MODELLING & SOFTWARE
(2023)
Article
Environmental Sciences
Zhengyang Cheng, Konstantine P. Georgakakos, Cristopher R. Spencer, Randall Banks
Summary: This paper demonstrates the use of numerical hydrologic and hydraulic modeling in quantifying flash flood risks, evaluating flood mitigation measures, and building operational warning systems for small urban communities. Case studies from Tegucigalpa, Honduras, and Istanbul, Turkey, showcase the application and transition of these models in real-world scenarios. Limited data and budget constraints are discussed, and feasible flood mitigation plans are recommended based on cost-to-performance analysis.
Article
Engineering, Environmental
Badiya Salele, Yakubu Aminu Dodo, Dalhatu Aliyu Sani, Mohammed Awad Abuhussain, Barno Sayfutdinovna Abdullaeva, Adam Brysiewicz
Summary: This study simulated runoff in pervious and impervious urban areas using the SWAT model, and enhanced the model with the emotional artificial neural network (EANN). The SWAT-EANN couple model was used to assess urban flooding. The study's findings provide valuable insights for urban planning, administration, and development, aiming to prevent flooding and environmental issues.
WATER SCIENCE AND TECHNOLOGY
(2023)
Article
Computer Science, Information Systems
Til Prasad Pangali Sharma, Jiahua Zhang, Narendra Raj Khanal, Foyez Ahmed Prodhan, Lkhagvadorj Nanzad, Da Zhang, Pashupati Nepal
Summary: Basin geomorphology plays a crucial role in flood risk evaluation, particularly in flood-prone areas like Nepal. The study used SRTM data and GIS technology to assess flood risks in the East Rapti River basin, identifying potential risk sub-basins based on morphometric parameters and landcover types. The results highlighted the importance of drainage density, relief, and rainfall intensity in contributing to flash floods, emphasizing the need for flood-resilient city planning in high-risk sub-basins.
ISPRS INTERNATIONAL JOURNAL OF GEO-INFORMATION
(2021)
Article
Construction & Building Technology
Hai -Min Lyu, Zhen-Yu Yin
Summary: This study developed a framework for evaluating flood susceptibility in the Guangdong-Hong Kong-Macau Greater Bay Area (GBA) using tree-based machine learning and geographical information system techniques. Tree-based models were used to predict flood susceptibility, with the categorical boosting model performing the best. The obtained flood susceptibility maps provide suggestions for flood disaster mitigation in the GBA.
SUSTAINABLE CITIES AND SOCIETY
(2023)
Article
Environmental Sciences
F. Polong, Q. B. Pham, D. T. Anh, K. U. Rahman, M. Shahid, R. S. Alharbi
Summary: In developing countries, satellite-based precipitation products can be used to support decision-making and planning of hydrological resources due to limited availability of in situ hydro-meteorological data. This paper evaluates and compares four high-resolution satellite-based precipitation products with measurements from rain-gauge stations, and finds that ARC2 performs the best in reproducing the occurrence frequency and detecting precipitation events.
INTERNATIONAL JOURNAL OF ENVIRONMENTAL SCIENCE AND TECHNOLOGY
(2023)
Article
Environmental Sciences
Duong Tran Anh, Ahad Hasan Tanim, Daniel Prakash Kushwaha, Quoc Bao Pham, Van Hieu Bui
Summary: This study proposes a hybrid model combining deep learning LSTM with the Discrete Element Method (DEM) to predict the porosity of gravel riverbed. The simulation results show that the model is reliable and accurate, and it can effectively reduce simulation time.
INTERNATIONAL JOURNAL OF SEDIMENT RESEARCH
(2023)
Article
Engineering, Environmental
Chandini P. C. Senan, R. S. Ajin, Jean Homian Danumah, Romulus Costache, Alireza Arabameri, A. Rajaneesh, K. S. Sajinkumar, Sekhar L. Kuriakose
Summary: This study aims to identify flood-vulnerable zones in southern India using local self-governing bodies as mapping units, and evaluate the predictive accuracy of the AHP and F-AHP models. The results demonstrate that the AHP model is more efficient than the F-AHP model in demarcating flood vulnerable zones.
STOCHASTIC ENVIRONMENTAL RESEARCH AND RISK ASSESSMENT
(2023)
Article
Environmental Sciences
Dipankar Ruidas, Asish Saha, Abu Reza Md Towfiqul Islam, Romulus Costache, Subodh Chandra Pal
Summary: This study proposes a bivariate logistic regression method to delineate the flash flood hazard map in the Gandheswari river basin. The method shows high predictive performance in both training and testing datasets.
ENVIRONMENTAL SCIENCE AND POLLUTION RESEARCH
(2023)
Article
Geosciences, Multidisciplinary
Muhammad Shafeeque, Yi Luo, Arfan Arshad, Sher Muhammad, Muhammad Ashraf, Quoc Bao Pham
Summary: The present study uses the SPHY model and CMIP6 climate data to quantitatively assess glacio-hydrological changes in the Upper Indus Basin. The study finds that future glacier area and freshwater availability will decrease, and the contributions of snowmelt, glacier melt, baseflow, and rain-runoff to total runoff will change, as will the contribution of the critical zone. In addition, low flows are projected to increase while high flows are likely to decrease. Warming temperature is identified as the dominant driver for changes in glacier area and total runoff.
Article
Engineering, Multidisciplinary
Pakorn Ditthakit, Sirimon Pinthong, Nureehan Salaeh, Jakkarin Weekaew, Thai Thanh Tran, Quoc Bao Pham
Summary: This study compares the applicability of machine learning methods and the GR2M model for simulating monthly runoff. The results show that machine learning algorithms, particularly SVR-rbf, outperform the GR2M model in stations with low correlation coefficients between input and output datasets.
AIN SHAMS ENGINEERING JOURNAL
(2023)
Article
Water Resources
Quoc Bao Pham, Babak Mohammadi, Roozbeh Moazenzadeh, Salim Heddam, Ramiro Pillco Zola, Adarsh Sankaran, Vivek Gupta, Ismail Elkhrachy, Khaled Mohamed Khedher, Duong Tran Anh
Summary: Lakes play a crucial role in enhancing the sustainability of the natural environment and reducing risks associated with the food chain, agriculture, ecosystem services, and recreational activities. This research focuses on simulating monthly lake water levels using hybrid data-intelligence models based on the ensemble adaptive neuro-fuzzy inference system (ANFIS) and metaheuristics algorithms, with a specific emphasis on the seasonal effect on Titicaca Lake water-level fluctuations. The ANFIS-WOA model demonstrated the best prediction performance, with a root mean square error (RMSE) of approximately 0.08 m, a mean absolute error (MAE) of approximately 0.06 m, and a coefficient of determination (R-2) of approximately 0.96, when the inputs considered were Xt-1, Xt-2, Xt-3, Xt-4, and Xt-12. Furthermore, the study found that long-term seasonal memory serves as a suitable input for lake water-level models, with the 1-year time series providing the best estimation for the water level of Titicaca Lake.
APPLIED WATER SCIENCE
(2023)
Article
Ecology
Aishwarya Sinha, Suresh Nikhil, Rajendran Shobha Ajin, Jean Homian Danumah, Sunil Saha, Romulus Costache, Ambujendran Rajaneesh, Kochappi Sathyan Sajinkumar, Kolangad Amrutha, Alfred Johny, Fahad Marzook, Pratheesh Chacko Mammen, Kamal Abdelrahman, Mohammed S. Fnais, Mohamed Abioui
Summary: This research uses geospatial tools, AHP, and fuzzy-AHP models to identify wildfire risk zones in Wayanad Wildlife Sanctuary and Kedarnath Wildlife Sanctuary. Both natural and anthropogenic factors contribute to the fire occurrences in these areas. The validation of the risk maps shows that both models have high prediction accuracy, with the F-AHP model performing slightly better. The created models can be used to implement effective policies to reduce the impact of fires in similar protected areas.
Article
Environmental Sciences
Loubna Hamdi, Nabil Defaflia, Abdelaziz Merghadi, Chamssedine Fehdi, Ali P. Yunus, Jie Dou, Quoc Bao Pham, Hazem Ghassan Abdo, Hussein Almohamad, Motrih Al-Mutiry
Summary: This study uses GPS data and PS-InSAR techniques to monitor land subsidence in the Cheria basin in Algeria. The results show significant changes in land surface, with a maximum subsidence of 500 mm over 6 years. These findings can be used to identify vulnerable areas and evaluate surface deformation for potential damage reduction in the future.
Article
Environmental Sciences
Mohammed Achite, Nehal Elshaboury, Muhammad Jehanzaib, Dinesh Kumar Vishwakarma, Quoc Bao Pham, Duong Tran Anh, Eslam Mohammed Abdelkader, Ahmed Elbeltagi
Summary: Drought negatively impacts water resources, land and soil degradation, desertification, agricultural productivity, and food security. The standardized precipitation index (SPI) is crucial for predicting meteorological droughts and managing water resources. Five machine learning models, including support vector machine (SVM), were used to model SPI at different timescales. The SVM model was found to be the most effective for predicting SPI, and satisfactory results were achieved when applying it to sub-basin 2. The suggested model outperformed other models in estimating drought and can be helpful for predicting drought on different timescales and managing water resources.
Article
Green & Sustainable Science & Technology
Phong Nguyen Thanh, Thinh Le Van, Tuan Tran Minh, Tuyen Huynh Ngoc, Worapong Lohpaisankrit, Quoc Bao Pham, Alexandre S. Gagnon, Proloy Deb, Nhat Truong Pham, Duong Tran Anh, Vuong Nguyen Dinh
Summary: In Southeast Vietnam, frequent droughts have caused significant damage and hindered socio-economic development. Water scarcity has particularly impacted the industrial and agricultural sectors. This study examined water balance and resilience in the La Nga-Luy River basin under two scenarios: business-as-usual and sustainable development approach. The results identified areas experiencing abnormal dryness and moderate droughts, as well as regions with severe and extreme droughts. The study also demonstrated the possibility of meeting irrigation water demand under different drought conditions and highlighted the importance of increased water use efficiency.
Article
Environmental Studies
Md. Uzzal Mia, Tahmida Naher Chowdhury, Rabin Chakrabortty, Subodh Chandra Pal, Mohammad Khalid Al-Sadoon, Romulus Costache, Abu Reza Md. Towfiqul Islam
Summary: We developed a novel iterative classifier optimizer (ICO) with ensemble algorithms to build computational models for flood susceptibility mapping in the Padma River basin, Bangladesh. The models include environmental, topographical, hydrological, and tectonic circumstances, and the most influencing variables for floods are rainfall, elevation, and distance from the river. The DLNN-ICO ensemble model has the optimal predictive performance and might be a viable technique for precisely predicting and visualizing flood events.
Article
Environmental Studies
Sheela Bhuvanendran Bhagya, Anita Saji Sumi, Sankaran Balaji, Jean Homian Danumah, Romulus Costache, Ambujendran Rajaneesh, Ajayakumar Gokul, Chandini Padmanabhapanicker Chandrasenan, Renata Pacheco Quevedo, Alfred Johny, Kochappi Sathyan Sajinkumar, Sunil Saha, Rajendran Shobha Ajin, Pratheesh Chacko Mammen, Kamal Abdelrahman, Mohammed S. Fnais, Mohamed Abioui
Summary: The study aims to assess the landslide susceptibility of high-range local self-governments (LSGs) in Kottayam district using the analytical hierarchy process (AHP) and fuzzy-AHP (F-AHP) models, and compare the performance of existing landslide susceptible maps. The identification of landslide-susceptible areas and factors will help decision-makers in identifying critical infrastructure at risk and alternate routes for emergency evacuation to safer terrain.
Article
Environmental Sciences
Siham Acharki, Pierre-Louis Frison, Bijeesh Kozhikkodan Veettil, Quoc Bao Pham, Sudhir Kumar Singh, Mina Amharref, Abdes Samed Bernoussi
Summary: Crop type identification is crucial for sustainable agriculture policy development and environmental evaluations. This study examined the effectiveness of different satellite sensors and classification algorithms in identifying land cover and crop types in a Mediterranean irrigated area. The findings revealed that the Support Vector Machine algorithm performed well in extracting crop type information from high-resolution imagery.
ENVIRONMENTAL MONITORING AND ASSESSMENT
(2023)
Article
Environmental Sciences
Asit Kumar Jaydhar, Subodh Chandra Pal, Asish Saha, Abu Reza Md. Towfiqul Islam, Dipankar Ruidas, Romulus Costache
Summary: The scarcity of water supplies in coastal areas has a significant impact on sustainable development, with coastal groundwater quality playing the most important role. Heavy metal contamination leading to groundwater pollution is a major global health and environmental concern. This study found that a significant portion of the area has high human health hazard index (HHHI) and poor water quality. Heavy metal concentrations, such as Fe, As, TDS, Mg2+, Na, and Cl-, were found to be particularly high in the western part of the district. The study also highlighted the vulnerability of the region due to high concentrations of TDS, Cl-, and Na+ in the groundwater.
ENVIRONMENTAL SCIENCE AND POLLUTION RESEARCH
(2023)