How to compute extreme cold levels to design power plants in the climate change context?

 The expected impact of climate change on temperature extremes is an increase in both the frequency and intensity of heat waves, while cold waves are expected to become less frequent and associated with milder cold temperatures. However, cold waves cannot be ruled out, as cold temperatures similar to those experienced in the past can still occur, at least in the near future, albeit with a lower probability.

While many studies have focused on estimating hot extremes in the context of non-stationary climate change, fewer have addressed the estimation of cold extremes, which must be considered for the design of new installations. Unlike hot extremes, which will intensify over time, the coldest values that might affect existing or planned installations are expected to occur now or in the very near future.

Temperature extremes exhibit different types of non-stationarities: a seasonal cycle, the human-induced climate change trend, and interannual to decadal variability. The seasonal cycle is commonly handled by selecting the season prone to the analyzed extremes. Various methods have been proposed to account for the trend due to human-induced climate change in extreme value estimations, either by considering trends in the parameters of statistical extreme value distributions (Coles, 2001; Parey et al., 2007; Gilleland and Katz, 2016; Barbaux et al., 2025, among others) or by computing a reduced variable whose extremes can be considered stationary and then back-transformed (Parey et al., 2013, 2019; Mentachi et al., 2016). However, for cold extremes, interannual variability generally plays a more significant role.

Therefore, in this study, we propose and test an approach to infer extreme cold values representative of the current climate by combining extreme deviations from the average winter mean and variance, as observed during the coldest winters in the past, with the average conditions of current winters. The methodology will first be described then illustrated with examples.

 

References:

Coles S (2001) An introduction to statistical modelling of extreme values, Springer Series in Statistics. Springer, London

Parey S, Malek F, Laurent C, Dacunha-Castelle D (2007) Trends and climate evolution: statistical approach for very high temperatures in France. Clim Change 81:331–352. https://doi.org/10.1007/s10584-006-9116-4

Gilleland, E., & Katz, R. W. (2016). extRemes 2.0: An Extreme Value Analysis Package in R. Journal of Statistical Software, 72(8), 1–39. https://doi.org/10.18637/jss.v072.i08

Occitane Barbaux, Philippe Naveau, Nathalie Bertrand, Aurélien Ribes, Integrating non-stationarity and uncertainty in design life levels based on climatological time series, Weather and Climate Extremes, Volume 50, 2025,100807, ISSN 2212-0947, https://doi.org/10.1016/j.wace.2025.100807.

Parey S, Hoang TTH, Dacunha-Castelle D (2013) The importance of mean and variance in predicting changes in temperature extremes. J Geophys Res Atmos 118:8285–8296. https://doi.org/10.1002/jgrd.50629

Parey, S., Hoang, T.T.H. & Dacunha-Castelle, D. Future high-temperature extremes and stationarity. Nat Hazards 98, 1115–1134 (2019). https://doi.org/10.1007/s11069-018-3499-1

Mentaschi, L., Vousdoukas, M. I., Voukouvalas, E., Sartini, L., Feyen, L., Besio, G., & Alfieri, L. (2016). The transformed-stationary approach: a generic and simplified methodology for non-stationary extreme value analysis. Hydrology and Earth System Sciences, 20(9), 3527–3547. https://doi.org/10.5194/hess-20-3527-2016