Evaluation and joint projection of temperature and precipitation extremes across Canada based on hierarchical Bayesian modelling and large ensembles of regional climate simulations

Singh, Harsimrenjit; Najafi, Mohammad Reza; Cannon, Alex J.

doi:10.1016/j.wace.2022.100443

Cited by 11 publications

(3 citation statements)

References 73 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The alternative solution is to try to relax the assumption from trying to find the best trainer among all set of trainers to find the best class of a given data element using a given set of trainers. Instead of designing a competition between the trainers to find the winning trainer, let them cooperate to find the best class for a given element from the set of tata [104]. As many trainers can be added as much as a better performance [111].…”

Section: Methodsmentioning

confidence: 99%

Ensemble learning based on relative accuracy approach and diversity teams

B. Rokaya,

D. Alsufiani

2024

Bulletin EEI

View full text Add to dashboard Cite

Ensemble learning, which involves combining the opinions of multiple experts to arrive at a better result, has been used for centuries. In this work, a review of the major voting methods in ensemble learning is explored. This work will focus on a new method for combining the results of individual learners. The method depends on the relative accuracy and diversity of teams. Instead of trying to assign weight to each different trainer, the concept of diversity teams is presented. Each team will vote as one player; however, the individual accuracies of each learner still be implemented. The concept of relaxing parameters that deal with each team as one player is presented. Our experiments demonstrate that traditional ensemble voting methods outperform individual learners. There is a limit to the superiority of the ensemble learner that any ensemble learner cannot go beyond. The proposed voting method gives the same results as the traditional ensemble voting methods, however, a different diversity of the proposed method from the traditional voting method or for different values of the relaxing parameter can be achieved.

show abstract

Section: Methodsmentioning

confidence: 99%

Ensemble learning based on relative accuracy approach and diversity teams

B. Rokaya,

D. Alsufiani

2024

Bulletin EEI

View full text Add to dashboard Cite

show abstract

“…Simulations using neural network are helpful for a wide range of engineering problems [10]. Another group of researchers have used simulation to analyze drought [11][12][13]. The simulation of this phenomenon is a suitable tool to investigate the complex and interdependent nature of this problem and is used in different other complex interdependent systems for example in resilience assessment [14], disaster management [15], resource management [16] and regional disaster planning [17].…”

Section: Introductionmentioning

confidence: 99%

Investigating lake drought prevention using a DRL-based method

Ghayoomi

Partohaghighi

2023

Preprint

View full text Add to dashboard Cite

Drought and decrease in the level of lakes in recent years due to global warming and excessive use of water resources feeding lakes is of great importance and this research has provided a structure to investigate this issue. First, the information required for simulating lake drought is provided with strong references and necessary assumptions. Entity-Component-System (ECS) structure has been used for simulation which can consider assumptions flexibly in simulation. Three major users (i.e., Industry, agriculture, and Domestic users) consume water from ground water and surface water (i.e., streams, rivers and lakes). Lake Mead has been considered for simulation, and the information necessary to investigate its drought has also been provided. The results are presented in the form of scenario-based design and optimal strategy selection. For optimal strategy selection a deep reinforcement algorithm is developed to select the best set of strategies among all possible projects. These results can provide a better view of how to plan to prevent lake drought.

show abstract

“…The theory of copula, introduced to the hydrologic community by De Michele and Salvadori (2003), is commonly applied for the multivariate analysis of flood events as it can represent a wide range of dependence structures between hydroclimatic variables (Singh et al, 2021(Singh et al, , 2022. It is a flexible approach for the frequency analysis of compound events that allows for characterizing the individual drivers with the most appropriate distribution functions.…”

mentioning

confidence: 99%

Multivariate Analysis of Compound Flood Hazard Across Canada's Atlantic, Pacific and Great Lakes Coastal Areas

Pirani

Najafi

2022

Earth's Future

Self Cite

View full text Add to dashboard Cite

Flooding, as the most common natural hazard in the world (Kundzewicz et al., 2014), has affected more than two billion people and caused approximately USD 656 billion of damage between 1998 and 2017 (AghaKouchak et al., 2020;Wallemacq, 2018). From 1980 to 2019, flood events have accounted for 41% of all the 17,300 weather-related events, 28% of 890,000 lives lost, 27% of USD 4,000 billion economic losses, and 10% of USD 1,300 billion insured losses worldwide (Golnaraghi et al., 2020). The frequency of flood events has been increasing from 1960 to 2013, globally (Tanoue et al., 2016). Similarly, the magnitude of flooding shows increases in some regions around the world (Do et al., 2020). Around 0.8-1.1 million people experience flooding and its devastating socioeconomic consequences each year (Muis et al., 2016), especially the coastal communities. The population of the low-lying coastal areas was approximately 625 million in 2000, which is anticipated to reach 949 million by the 2030s and 1.4 billion by 2060s (Neumann et al., 2015), indicating larger exposure to different Abstract Compound flooding, caused by the simultaneous or successive occurrence of two or more flood mechanisms, is mainly associated with extreme precipitation, river overflows, and storm tides across coastal areas. The interdependencies between these components can increase the risks of flood impacts, threatening coastal communities and infrastructure systems. This study quantifies the corresponding multivariate hazard over Canada's coastal areas by characterizing the dependencies between multiple drivers of flooding based on the C-vine copula statistical approach. The joint return periods of compound flooding considering the AND, OR, and Kendall scenarios are estimated and the corresponding failure probabilities are assessed. Further, the compound hazard ratio (CHR) index is applied to quantify possible under-or overestimations of the flood hazards when individual drivers are assessed independently. Analyses are performed at 41 locations across the Atlantic, Pacific, and the Great Lakes coasts, and the uncertainties are quantified based on the Bayes theorem.Results show that at approximately 50% of locations (mostly at the Great Lakes), the flood hazard associated with the AND scenario increases considerably when the dependencies are characterized compared to the (unrealistic) independence scenario, indicating the potential for compound flooding in these regions. Besides, at more than half of the studied locations, the CHR index exceeds one highlighting the interrelationships between drivers of flooding. The results of this study provide a deeper understanding of the flood mechanisms and their interdependencies across Canada's coasts, which support the development of resilient structures and improved coastal flood management. Plain Language SummaryApproximately half of the global population lives within 200 km of coastlines. The communities and infrastructure systems in the coastal environments are at risk of flooding caused by one or multipl...

show abstract

Evaluation and joint projection of temperature and precipitation extremes across Canada based on hierarchical Bayesian modelling and large ensembles of regional climate simulations

Cited by 11 publications

References 73 publications

Ensemble learning based on relative accuracy approach and diversity teams

Ensemble learning based on relative accuracy approach and diversity teams

Investigating lake drought prevention using a DRL-based method

Multivariate Analysis of Compound Flood Hazard Across Canada's Atlantic, Pacific and Great Lakes Coastal Areas

Contact Info

Product

Resources

About