This project’s objectives are based on the climatic information contained in ships’ logbooks for the period 1750 to 1850. Officers on board eighteenth and nineteenth century sailing vessels maintained detailed log books of the ships’ activities and management. Included within these records were observations of the current weather. These observations were made at least three times daily and were used as an indispensable aid to navigation in a period before reliable methods of determining longitude were widely available. We are fortunate that many thousand such log books have survived. This project concentrates on those held in British, Dutch, French, Spanish and Argentinean archives. These collections contain the vast majority of extant logbook material. Most of the logs are from vessels engaged in military or official government activities, but a significant number have also survived from the ships engaged in the Spanish Postal Service and the quasi-governmental enterprises of British, Dutch and French trading companies. The use of logbooks from these sources, with their interests in different parts of the World, ensures a near-global coverage of oceanic information.

The recorded data are concerned with wind direction and wind force as these two elements more than any others contributed to the speed and direction of the vessels. Other weather elements were also recorded such as precipitation, fog, ice cover, state of sea and sky. Although non-instrumental (some temperature and air pressure records begin to appear in the nineteenth century but they are relatively few in number), the data have been shown by the small scale studies thus far undertaken to be reliable and accurate. They can be subjected to statistical analysis, used for synoptic reconstructions and for scientific interpretation and scrutiny.

The log books represent a hitherto little explored but potentially important source of oceanic climatic information for a period before the establishment of networks of instrumental observatories. The abstraction, digitisation and analysis of these data will make a significant contribution to our knowledge and understanding of global and regional climatic variation.

Objective 1

To produce and make freely available for the scientific community the World’s first daily oceanic climatological database for the period 1750 to 1850.

The database will contain climatological information for the North and South Atlantic, the Indian and the Pacific Oceans. Entries will be made chronologically day by day but will include identification by geographic location. Information on wind direction, wind force and other recorded weather elements will be included. Sufficient log books exist for a daily record of the observations for each the oceanic regions to be abstracted. Each entry could contain up to ten items of regularly recorded information with, in total, nearly one million, entries. The database will be designed to allow for later additions as other sources become available and for easy abstraction of information/data blocks specified by the user. It will be supported by a substantial body of metadata focusing on background material such as the details of the vessel in question, the recording officer and the purpose of the voyage.

Objective 2

To realise the potential of the database to provide a better knowledge of oceanic climate variability over the study period.

It is important not only to produce but to utilise the database to improve our knowledge and, ultimately, understanding of climatic variability over the oceans between 1750 and 1850. Several further objectives fall under this broad heading for this critical, pre-industrial and, as far as oceanic climates are concerned, pre-instrumental period.

Objective 2a. To prepare summary and derivative measures from the database to complement and integrate with other contemporary series.

A number of climatic series can be derived from the database. These can be further summarised at the seasonal, annual, decadal and multi-decadal time scales. The geographic coverage of the raw data allows for global as well as regional summaries. These derived series can be usefully compared with contemporary, land-based data to shed light on regional climate responses to changing conditions over the oceans. Much of this land-based data have already been gathered under the auspices of various past EC projects.

Objective 2b. To use the database to determine the character and scale of oceanic climatic change and variability at various time scales during the final stages of the pre-industrial period.

The rapidity and nature of climatic variations across oceanic areas can be objectively identified. This is important because the project’s data are drawn from the ‘pre-industrial’ period when anthropogenic climatic control is widely regarded as absent and the results will aid testing of general circulation models.

Objective 2c. To extend the NAO record by reference to derived information from the database.

Because of the abundance of data from the North Atlantic region and its importance to furthering our understanding of European climate, particular emphasis will be placed on the North Atlantic Oscillation and how it manifests itself in the abstracted data. An index will be developed to reconstruct the NAO record to 1750. Reference to contemporary land-based data will shed light on the role of the NAO in governing European climates between 1750 and 1850. Twentieth century data can be used to examine the consistency of this relationship over the multi-decadal time scale.

Objective 3

To use the information to extend and enhance existing oceanic-climate databases.

The proposers recognise that this project is relevant to other research being undertaken. It offers the possibility of supplementing and extending established data sets, in particular the world-wide Comprehensive Ocean-Atmosphere Data Set (COADS), thereby extending the period of cover backwards by a further century from its current starting point in the mid-nineteenth century.

Objective 4

To disseminate the project’s findings and to stimulate interest and awareness in this source with a view to fostering its further development and realising its scientific potential.

This projects concentrates on British, Dutch, French, Spanish and Argentinean log books. Most of these are sources that have already been subject to preliminary examination. It is known that other sources exist elsewhere in Europe that for reasons of accessibility, lack of opportunity or appreciation of their scientific value, have not been exploited. An important objective is to encourage work in these areas by fellow academics.

The most significant innovatory aspect of this work is its exploration of a long-existing but relatively under-utilised source of reliable scientific information, viz. ships’ log books from the late eighteenth and early nineteenth centuries. Never before has this source been exploited in the proposed degree of detail or over such a large number of items and geographic range. Early work, making limited use of log book data for purposes of pressure field reconstruction was carried out by Lamb and Johnson (1966). Oliver and Kington (1970) drew attention to the potential of log books for climatic research but only much more recently have any steps been taken towards realising this potential, e.g. Landsberg (1985).

The scientific work is poorly advanced however and developing only in a few countries. The reasons for this are manifold but a prevailing lack of awareness of the nature or existence of ships’ log books or of their climatological contents has been a factor. In addition, the practical tasks of abstracting detailed data from thousands of items are formidable for any individual or small group. Archaic vocabulary, ambiguous terminology and the absence of any international standards for weather terms as well as the problems of calligraphy and language have also hindered the study of this internationally-based source. More generally, a concern on the part of some scientists regarding non-instrumental data may also be a factor. Yet, as the widely-acclaimed works of Lamb (1995) and Ladurie (1972) have demonstrated, such concerns are ill-founded and this project aims to extend further the scientific application of historical data.

Current research using this source has thus far been limited to case studies and specifically-focused analyses (e.g. Farrington et al, 1998 and Catchpole, 1992). These have been useful in themselves and have given a strong indication that the data source is both extensive in terms of geographical coverage and reliable in the consistency and accuracy of the climatic record. Other studies have also revealed how the data can be used to produce either detailed daily synoptic reconstructions or long series of climate records. Such relatively small-scale studies could of course continue but the point has been reached when an attempt at a more comprehensive exercise, drawing upon many log books over wide areas and spanning collectively many decades, must be undertaken to realise the potential of this unique part of the European heritage and to bring it to the attention of the wider scientific community.

The second important innovative aspect of this project is its concentration on oceanic areas. Past researches of the climates of the eighteenth and nineteenth centuries have principally focused on land-based data. Although a Danish project (Frydendahl. et al 1992) has been carried out using digitized log book data for the early eighteenth century the oceans have for the period 1750-1850 have yet to be examined in detail. Whilst instrumental data of the character that are normally associated with land-based observations, such as temperatures, rainfall and air pressure, are so scarce as to render any comprehensive review untenable, the data for wind direction and force can provide a reliable reconstruction of wind flow and air pressure fields. Similar work using more restricted sets of land-based wind data in Sweden are models in this respect showing how wind flow data can be profitably interpreted within a wider regional context (Jönsson and Holmquist, 1995). Indeed it can be argued that such marine-based wind data are more reliably recorded and more representative of synoptic-scale factors than are land-based observations subject as they are to local influences.

Whilst attention will be devoted to the wind data, other climatic variables are also available; precipitation, heavy fog and other extreme occurrences such as ice cover and icebergs were consistently recorded. These data will be included in the database. They will also contribute to the preparation of time series information to identify significant climatic events and periods. For example, although precipitation depths are not recorded, wet and dry phases can be distinguished from the data through the ‘rain days’ series. Whilst similar series have been already assembled for many land-based sites no comparable undertaking exists for oceanic areas.

This project, whilst innovative in many fundamental aspects is at the same time part of the wider field of established climatic studies where the state of the science is much more advanced. Close attention will be given to examination of the climatic record revealed in the various series. Comparisons with contemporary evidence from land-based observations will be made in the search for a more comprehensive knowledge of climate variation that has hitherto lacked a detailed oceanic climatic element. The series will also be examined for signals and teleconnections between major areas of activity. The ENSO signal is perhaps the most widely sought at present, but the North Atlantic Oscillation (NAO), if only because of its profound influence on European climate, demands attention. In addition the strength of the NAO signal is found over the sea and for the time period embraced by the project now other direct observational data are available for those areas. Finally, the project will formally extend existing databases of which COADS (Woodruff et al, 1993) is the most apposite. The log book data source provides the only current means by which such extensions backwards through time can be confidently made.