Abstract
Weather typing, based on surface pressure charts, has been one of the principal means of analysis in synoptic climatology. Here, we use an automated scheme to derive weather types ( WTs ) and also calculate Lamb weather types ( LWTs ) for the Falkland Islands. The WTs are based on sea‐level pressure data estimated using two reanalysis products: one that extends from 1948 to 2014 and another that just uses station pressure data as input and extends back to 1871. The WTs can be used to derive counts of gale days and these will be compared with storminess estimates based on the rate of change of daily‐average pressure measurements at the principal observational site (near the capital, Port Stanley) on the islands. A particular emphasis of the paper is the reliability of the results taking into account that we are using reanalysis datasets from a very data‐sparse region of the world. More gale days are estimated during the period from about 1880 to the mid‐1910s and since the 1980s. Fewer gale days are evident during other periods, particularly from the mid‐1910s to 1947. As these changes are not evident in the storminess measure derived from the sub‐daily pressure series for the Port Stanley region, the results in terms of gale‐day counts are very suggestive of being due to differences in the quality of the reanalysis during the different periods. The reanalysis appears better the higher the number of gale days estimated. The opening of the Panama Canal in 1914 dramatically reduced the number of ships, and hence observations, rounding Cape Horn. The paper also relates seasonal counts of the LWTs and WTs to recently developed long series of temperature and precipitation for the Port Stanley region.