Dataset used in this study
We used the 5 km scale historical UK weather record held by the UK Met Office’s National Climate Information Centre (Perry and Hollis, 2005). This gridded dataset covers the whole of the UK and includes daily maximum and minimum temperature and rainfall data from observation stations from 1960 to 2016.
We developed indices relating to the risk of occurrence of four extreme weather events, namely (i) heat waves, (ii) cold snaps, (iii) extreme rainfall (flood), and (iv) low rainfall (drought). We employed a threshold approach and for each grid point extracted the frequency each year that the five day rolling mean temperature or rainfall exceeded this threshold for a set number of days. We split the resulting dataset into two 27 year time periods, 1961–1988 and 1989–2016, reflecting the Met Office’s definition of long-term averages for weather data of 30 years (World Meteorological Organization, 2017), while keeping two discrete time periods of equal length.