Intensity-duration-frequency (IDF) curves can serve as useful tools in risk assessment of extreme environmental events. Thus, this study proposes an IDF approach for evaluating the risk of expected occurrences of extreme air pollution as measured by an air pollution index (API). Hourly data of Klang city in Malaysia from 1997 to 2016 are analyzed. For each year, a block maxima size is determined based on four different monsoon seasons. Generalized extreme value (GEV) distribution is used as a model to represent the probabilistic behavior of maximum intensity of the API, which is derived from each block. Based on the GEV model, the IDF curves are developed to estimate the extreme pollution intensities that correspond to various duration hours and return periods. Considering the IDF curves, we found that for any duration hour, the magnitude of pollution intensity tends to be high in parallel with increasing return periods. In fact, a high-intensity pollution event that poses a high risk of affecting the environment is less frequent than low-intensity pollution. In conclusion, the IDF curves provide a good basis for decision makers to assess the expected risk of extreme pollution events in the future.

中文翻译：

---

强度-持续时间-频率方法用于空气污染事件的风险评估。

强度-持续时间-频率（IDF）曲线可以用作对极端环境事件进行风险评估的有用工具。因此，本研究提出了一种IDF方法，用于评估以空气污染指数（API）衡量的极端空气污染的预期发生风险。分析了1997年至2016年马来西亚巴生市的每小时数据。每年，根据四个不同的季风季节确定最大区块大小。通用极值（GEV）分布用作模型来表示API的最大强度的概率行为，该概率行为是从每个块派生的。基于GEV模型，开发了IDF曲线以估计与各种工时和返还时间相对应的极端污染强度。考虑到IDF曲线，我们发现对于任何持续时间，随着返回期的增加，污染强度的幅度往往较高。实际上，对环境造成高风险的高强度污染事件要比低强度污染事件发生的频率低。总之，IDF曲线为决策者评估未来极端污染事件的预期风险提供了良好的基础。