Anna Choi and Tze Leung Lai’s contribution to the Discussion of ‘The First Discussion Meeting on Statistical aspects of climate change’
本文评论了两篇关于气候变化统计方法的论文，一篇用广义加性模型评估挪威水灾风险，另一篇用极值理论分析美国和格陵兰的极端温度，对斯坦福大学新成立的可持续发展学院的环境健康课程有参考价值。
Authors and Papers under Discussion: ‘Assessing present and future risk of water damage using building attributes, meterology and topograph’ Heinrich-Mertsching C, Wahl JC, Ordonez A, Stien M, Elvsborg J, Haug O, and Thorarinsdottir TL. ‘The importance of context in extreme value analysis with application to extreme temperatures in the USA and Greenland’ Clarkson D, Eastoe E, and Leeson A. Although the two papers under discussion are based on different methods and involve different illustrative applications, they are of great appeal to us who are involved in developing courses on environmental health in Stanford’s new Doerr School of Sustainability and Climate Change. In fact, September 29, 2022 marks the historic opening of the School led by Dean Arun Majumdar. The paper ‘Assessing present and future risk of water damage using building attributes, meteorology and topography’ by Heinrich-Mertsching et al. from Norway establishes a ‘nationwide, building-specific risk score for water damages associated with pluvial flooding in Norway’, and the basic methodology is to ‘fit a generalised additive model that relates the number of water damages to a wide range of explanatory variables’ (or covariates). Combining the model with an ensemble of climate projections enables projection of the (spatially varying) impacts of climate change on the ‘risk of pluvial flooding towards the middle and end of the 21st century.’ The paper ‘The importance of context in extreme value analysis with application to extreme temperatures in the USA and Greenland’ by Clarkson et al. of Lancaster University in the UK uses extreme value theory to analyse this problem and applies a random effects Peaks over Threshold approach to the case studies in Sections 3 and 4 (for air temperatures in three western US states and ice surface temperatures in Greenland, respectively). Data science, as manifested in these 2 papers of different approaches and motivations, is very useful to develop for students in the Doerr School to learn.