Corrigendum to “Millions of digitized historical sea-level pressure observations rediscovered”

We have revised the dataset associated with the paper “Millions of digitized historical sea-level pressure observations rediscovered” by E. Hawkins et al. (Geoscience Data Journal, 10, 385, doi: 10.1002/gdj3.163, 2023). The dataset includes more than 5 million observations of sea level pressure every 3 hours from April 1919 to December 1960 over the UK & Ireland which were contained in the Daily Weather Reports (DWRs) published by the Met Office.

A dataset user brought a small footnote to our attention which stated that in the original DWR documents for April 1919 to February 1930, the column giving the pressure change over the previous 3 hours was in units of half-millibars rather than whole millibars as we had previously assumed. This means that all pressure observations during this period derived using the ‘Change in last 3 hours’ column required small revisions – around 10% of the total dataset.

The ‘change over last 3 hours’ column was first introduced in the DWRs in May 1911 when the units of both pressure observations and the change in 3 hours were in/Hg using two decimal places. From May 1914 onwards, the pressure units were changed to mb, with half-millibars used for the change in pressure. After February 1930, the change in pressure was given in tenths of mb, and this was correctly used. The pressure observations from the DWRs for January 1911 to March 1919 remain unrescued.

The discussion of Figure 1 should read:

Figure 1 shows an example DWR page from 5th April 1919, showing the stations from which eight sea-level pressure observations per day can be derived. Each station has a listing for 01Z, 07Z, 13Z and 18Z, with a pressure observation converted to sea-level (given to a precision of 0.1 mb) and a change in pressure over the previous 3 hr in units of half-millibars. This allows the pressures for 22Z, 04Z, 10Z and 15Z to be calculated, but with a small uncertainty as the change is only given with a precision of 0.5mb. Note that the rows are not always complete, highlighting missing data, especially for 01Z, and therefore also for 22Z the day before.

The dataset revision means there are small visual differences in updated versions of Figures 6, 9 & 10, but these are not shown here. A revised version of Figure 7 is shown, and the discussion around Figures 7 and 8 should now read:

For example, the missing observation at Eskdalemuir in southern Scotland at 15Z on 23rd November is 956 mb, with other missing observations in Ireland from Malin Head at 972 mb and Blacksod Point at 984 mb. Recovering such individual missing observations may be worthwhile if analysing case studies of particular severe storms.

Note one almost certainly erroneous observation in the middle panel of the top row of Figure 7. The 991 mb observation for Birmingham (south-east of the lowest pressure values) at 15Z on 16th November 1928 has no correction listed in the DWRs and is correctly transcribed from the original DWR sheets. The 18Z observation is 975 mb, and this is indicated to be minus 16 half-mb from 3 hr earlier (Figure 8), resulting in a 983 mb observation for 15Z (Figure 7). It seems highly likely that the handwritten ‘−16’ should be ‘+16’, and that the 15Z observation was actually 967 mb, rather than 983 mb; this would fit the other available observations of the synoptic situation. There will be other examples such as this in the dataset, but they would likely be rejected in a reanalysis assimilation. This is an example of issue (4) listed above and suggests that the data at times derived from both a transcribed observation and a change in pressure will contain more errors. The 960 mb at Inchkeith at 15Z on 23rd November also looks too high but is similarly transcribed correctly with no correction reported.

The revised dataset reduces some of the differences with existing observations for Valentia (original Figure 10) but does not remove all the differences identified between 1922–1929. We also note that one sub-series (London at Kew) in Cornes et al. (Geoscience Data Journal, doi: 10.1002/gdj3.226, 2023) has also been revised in v1.1 of that dataset because of this issue.

We gratefully acknowledge Richard Meats for bringing this issue to our attention. We apologize for the error, but the conclusions are unchanged.