Scarcely a week goes by without further predictions, reflections and expressions of concern regarding climate change and its attendant difficulties and challenges. Recent years have seen huge advances in the field of climatology and climatic modelling, yet it remains essentially an empirical science with an unquenchable thirst for data.
These data range widely across the discipline, and present day observations from the various national networks form an indispensable source of information with which to establish the current situation and most recent trends. Most of these networks began their activities in the latter half of the Nineteenth Century, prior to which observations were made by a number of scientists, clerics and gentlemen academics, and while they corresponded to a remarkable degree, their efforts were otherwise uncoordinated. For this reason, our reliable instrumental database for past climates in Europe goes back little further than two centuries, and for yet shorter periods in time for other parts of the world. This is a pitifully brief period in relation to the age of the Earth and its gradual evolution.
In order to predict future weather with greater confidence, scientists need to know how climate has changed in the past, with present changes gauged against the so-called ‘natural’ level of variation; that is, before anthropogenic greenhouse gases began to exercise some control. If our instrumental record, however, goes back little more than two centuries, how, it might be asked, can any knowledge be secured of climates in earlier centuries and millennia?
Fortunately, scientists have at their disposal a large and growing volume of such evidence. Much of this is based on natural proxy records. This notable form of evidence is derived from a range of sources that include tree rings, ice cores, deep sea sediment records and preserved pollen. Each has its own set of advantages and disadvantages, but all are based on an environmental fingerprint left by changes in temperatures and precipitation. After all, there is no inherent constancy or fixed average in the world’s climate. Change has always, and will always, continue to occur.
Deep sea cores provide information on such changes dating back for many millennia, but offer only a very coarse scale of temporal resolution, rarely allowing researchers to see those changes at anything less than the decadal scale. Tree rings, on the other hand, provide an annual picture, but can extend the view back little more than to the conclusion of the last ice age some 10,000 years ago. All such sources, however, suffer the curse of calibration – to what extent can a change in tree ring width and composition be expressed as a temperature or precipitation change? In the case of ice cores, to take another example, how do the changes in oxygen isotope levels express the temperature changes that control them? Much time and effort has been directed, successfully, to answering these questions and providing an important insight into long term climate variation.
Ships’ logbooks in climate change studies
In recent years, a new source has been added to these natural proxies, which is essentially different both in form and in the challenges it offers in terms of comprehending the climate record. This source is formed by the written records of our societies. Assuredly, many of these documents provide no such insight, but a significant number do. These include estate papers, agricultural returns, diaries, state papers, even old newspapers and broadsheets, all of which may contain reference to weather, floods, droughts and storms (Lamb, 1982). Much has been done with these items, found in remarkable abundance across European archives (Brázdil et al, 2010).
While this source lacks the seemingly reliable and numerically irrefutable cache of instrumental observations, with such reliability more often based on assumption than objective assessment, what it offers is an insight into the weather of past times, written first hand by those that experienced its delights, horrors and pitiless imposition on their lives. One subgroup of this kaleidoscope of items, however, deserves particular attention – the logbooks of ships that navigated the world’s oceans in the Age of Sail.
Logbooks of the Royal Navy survive in the greatest number, indeed in their tens of thousands. These form the basis of the following review, but much of what is true of Royal Navy logbooks is true of their contemporaries, both in the UK and elsewhere in the archives of former imperial maritime powers such as the Netherlands, France and Spain, whose arteries of communication were sustained by a constant flow of trading and naval vessels. Various authorities, the Admiralty in the case of the Royal Navy, obliged their captains to keep a daily logbook in which were recorded the activities on board, ships sighted and the weather conditions.
As can be seen in Figure 1, for the most part these consisted of three elements:
• Wind direction – recorded on a 32 point compass • The force of the wind – using a conventional, but now archaic, vocabulary • The general state of the weather – including rain, snow, fog, mist, cloud and sea state
Such observations were, until the mid Nineteenth Century, non instrumental and made on the basis of many years of experience at sea. It should not be forgotten that most Royal Navy officers achieved rank only after entering the Navy as ‘young gentlemen’ at the tender age of ten or eleven and being then tutored in the traditions of the Senior Service. In many ways, and remarkably so, the procedures and skills they adopted and employed differ little from those used today by observers in the Voluntary Observing Ships (VOS) that report daily to the forecasting services around the world. There remains, however, an important distinction. While the latter observations are made solely for the purposes of providing data for weather forecasting, the former, those of the mariners of that now bygone age, were made for a purpose no less exacting – navigation.
In an age before longitude could be determined using the marine chronometer or by astronomical methods – both of which were not commonplace until close to the end of the Eighteenth Century – the only means by which a vessel’s movement eastwards or westwards (the surrogate of longitude) could be determined, was by careful observation of the force and direction of the wind and its influence on the movement and direction of the vessel. It was in many ways a matter of life and death to make such observations as reliably as possible – to do otherwise was to court disaster.
Benefits and advantages of logbook data
While not perhaps considered to be instrumental observations in the sense of present day thermometer, barometer or anemometer readings, logbook data are of no less scientific value. They provide a unique insight into the important question of atmospheric circulations and the organisation of high and low pressure systems reflected, as they are, in the direction and strength of airflows and winds.
Were only a handful of such documents to have survived, they would serve little purpose, but, and fortunately, there are more than 100,000 Royal Navy logbooks for the period before the mid Nineteenth Century. Given, additionally, the near global extent of the British Empire, the spatial coverage of the extant logbooks is impressively large.
The Indian and the North and South Atlantic Oceans were negotiated frequently, but regrettably given its importance for the global climate regime, the Pacific Ocean formed no part of the regular lines of imperial communication. The coverage provided by this source is summarised in Figure 2, which is taken from the Climatological Database for the World’s Oceans:1750 to 1850 (CLIWOC) project2 (Wheeler et al, 2006).
To such manifold advantages can be added others, and in contrast to the natural proxies cited previously, logbook evidence requires little in the way of calibration, although, as noted later, they are not without their idiosyncrasies that demand attention. In that sense the records are not by proxy, but the real conditions recorded faithfully at the time of their occurrence. Moreover, self-evidently the observations cover the oceans. These embrace nearly three-quarters of the planet’s surface, yet lack anything remotely similar to the data provision available for land surfaces. In contrast to the natural proxies, these observations have a temporal scale of resolution at the daily, often hourly, level. On the other hand, the logbook record cannot extend back beyond the mid Seventeenth Century when they began to be kept, but nevertheless encompasses some important climatic episodes.
These include the closing decades of the Little Ice Age, the major volcanic eruptions of Lakigigar (1783-4) and Tambora (1815), the period of solar quiescence known at the Dalton Minimum (circa 1810 – 1830) and the final decades of ‘natural’ climate change preceding the warmings of the last century. The data, finally, can be fixed in time. The logbooks are meticulously written and always include date and day, and location markers of latitude and longitude are also given daily.
Adapting logbook data for research
All these benefits are, however, not to deny the existence of difficulties in the use of logbooks. Most significant among these issues is that of the wind force descriptions. The Beaufort Force conventions of a 13-point scale, each of which represents a wind speed change, used universally today, did not come into use in the Royal Navy until the 1830s (Wheeler and Wilkinson, 2004). Before that time the vocabulary of wind force terms was much wider than the current 13 points, and not subject to any regulation or written guidance. Nevertheless, it is clear from the logbooks that the vocabulary was commonly used and widely understood (Wheeler and Wilkinson, 2005).
In no small measure this is the result of the oral traditions and skills handed down by senior to junior officers with the passing years. As noted earlier, all Royal Navy officers gained senior rank only by proven service, acquiring their skills on board. The problem remains, however, of determining the true force of winds, described in terms that have long since passed out of use.
Before the mid Eighteenth Century, for example, the term ‘breeze’ that today covers most of the lower half of the Beaufort Scale was not in common use; instead almost all winds were termed as gales. This is a term that today betokens a violence that requires Gale Warnings, which send yachtsmen scuttling back to the safety of port. Yet in past times gales could vary from near calm to extreme violence, their strength being noted in logbooks by various adjectival qualifiers that included soft, easy, light, gentle and stiff, as well as the present day near, equating to Force 7, and severe, representing Force 9.
More than 70 of these terms were divined from the logbooks, but one of the achievements of the CLIWOC project was to provide a dictionary of such terms that expressed them in Beaufort Scale equivalents, thereby providing a measure of force that can be used in scientific studies (Garcia-Herrera et al, 2003).
The other required adjustment is to the recorded wind directions. These were estimated on the basis of magnetic north, but to be of use in climatology the bearings need to be in respect of true north. The difference between the two is known as variation and it differs over space and time. Happily, a number of data bases exist that can be used to apply the necessary corrections.
The general descriptions of the weather, fortunately, require no such changes and were prepared in a form of English understood today. While not instrumental in character, climatologists have at their disposal three important and reliable measures of past climate: wind force, wind direction and the weather’s general state.
An example of what can be gained from the use of such information can be found in the 2010 Wheeler et al study, in which a daily weather record for the English Channel was reconstructed from the mid Eighteenth Century back to 1685. Such a notably detailed and reliable record can be gathered from no other source. It reveals the variations in circulation that characterise the closing years of the Little Ice Age and the recovery from that period, arguably the coldest in North West Europe since the retreat of the last ice caps more than 10,000 years ago. The frequent invasions of cold Siberian air from the east, or of more temperate and humid air from the Atlantic can all be seen and plotted in unique detail and do much to enhance the temperature record provided by the equally long Central England Temperature (CET) series that goes back to 1659 (Manley, 1974; Jones and Hulme, 1997).
Yet more interesting results can be harnessed if such series are blended with more recent observations that form part of the various re-analysis data sets that are now available. By this means, in 2013 Barriopedro et al examined air circulation changes in the North East Atlantic region for the past 300 years, an exercise that adds once again to our understanding of how climate has behaved in the past. It has also been possible to link these maritime observations with land based data for a more comprehensive spatial picture, and in 2009 Küttel et al amply demonstrated the benefits of this approach.
Conclusions and future directions
Despite the various successes mentioned in this article, the potential of Naval logbooks remains largely unrealised. The CLIWOC and subsequent projects and activities have used only some 6,000 logbooks, yet the UK National Archives in Kew and the National Maritime Museum in Greenwich hold more than 100,000 logbooks from before the mid Nineteenth Century, with others in foreign collections. So large are the number of items and so demanding of time is the extraction of the data, which with no OCR system available for this task can only be performed manually, that projects need to be focused on specific time periods and regions.
A good example of such an enterprise is the ARCdoc project3, which focuses on the Arctic region between 1750 and 1850, using logbooks from whaling vessels and from the archives of the Hudson’s Bay Company. There are relatively few such documents in the UK, and although the collection amounts to little more than 300, they take the researcher to a region of unique climatological importance where, not only can the circulation patterns be reconstructed as they have been for elsewhere, but observations are also available on the extent and character of the polar sea ice cover.
This project has yet to be completed, but the ice front reconstructions it is offering provide scientists with their first insight into the region’s behaviour at this time. Given the anxieties surrounding polar ice retreat and its global, as well as regional consequences, such insights assume particular significance, and one wonders what the salty old sea dogs who so assiduously kept their logbooks would make of our use of their reflections.
Published: 27th May 2014 in AWE International