Details

Autor(en) / Beteiligte

• Ismail, Mohd Sabri
• Masseran, Nurulkamal

Titel

Risk assessment for extreme air pollution events using vine copula

Ist Teil von

• Stochastic environmental research and risk assessment, 2024-06, Vol.38 (6), p.2331-2358

Ort / Verlag

Berlin/Heidelberg: Springer Berlin Heidelberg

Erscheinungsjahr

2024

Link zum Volltext

Quelle

Springer Online Journals Complete

Beschreibungen/Notizen

• This study proposes an alternative risk assessment approach for evaluating extreme air pollution events through vine copula modeling. Three characteristics of unhealthy air pollution events (i.e., severity, intensity, and duration) in Klang, Malaysia, are examined. The vine copula is fitted using sequential maximum likelihood estimation and joint maximum likelihood estimation, with a subsequent comparison based on criteria such as log-likelihood, Akaike’s information criterion, and Bayesian’s information criterion. Model fitting and comparison studies demonstrate that the most well-fitted vine copula model, achieved through joint maximum likelihood estimation, comprises the Joe, Rotated Tawn type 2 (180 degrees), and Rotated BB8 (90 degrees) copulas. The positive Kendall's τ correlation coefficient (0.26) for the obtained vine copula indicates that higher values of one characteristic are likely to be associated with higher values of the other characteristics, and vice versa. Furthermore, with the upper tail dependence coefficient (0.31) surpassing the lower tail dependence coefficient (0.18), indicating stronger dependence in the upper tail of their distribution, this underscores the significance of conducting risk assessments for extreme air pollution events characterized by extreme levels of severity, intensity, and duration. A vine copula-based simulation study is conducted to delve deeper into the risk assessment, revealing that extreme air pollution events are not linked to the highest values of joint and conditional probabilities. These findings suggest that extreme values in those distinct characteristics do not consistently occur simultaneously. The return period measures also indicate that extreme air pollution events have long waiting periods. Despite the current status of extreme air pollution events in Klang being controllable, achieving effective control necessitates ongoing efforts, encompassing regulatory actions, industrial controls, robust public transportation programs, and a dedicated transition to cleaner energy sources. This task is crucial for ensuring continuous clean air quality, sustaining our environment, and avoiding negative impacts on the economy and public well-being.