D. Seth Murray
Department of Anthropology, University of North Carolina

An Introduction to the Historical Climatology of Burgundy

Original manuscript: May 10, 2000
Revised manuscript: February 17, 2001
 

Introduction

Real and imagined changes in global climate have contributed to a proliferation of studies on climate and climate history in recent years. Global climate changes, for example, have perplexed scientists, policy makers and the public alike due to the inherent ambiguities and complexities of the global ecosystem. Although the interest in the link between climate and human society as well as climate change isn’t a new phenomenon, serious scholarship of the matter is a relatively recent development, mainly because of the paucity of reliable historical climate data. The results of many climatological research projects have been unsatisfactory because of the limited time scale at which they operate. The majority only retrodict climate history until 1955, and that has the potential to effectively skew the results of a climate change project, which in turn is often the basis for important governmental and industry policy decisions. Although climate studies have usually not held much appeal outside of their own discipline, the lack of time depth is a serious flaw which is now beginning to be addressed by a variety of concerned scientists such as geographers, biologists, and historians, as well as (paleo)climatologists and meteorologists. Historical climatology is particularly important and relevant because of its broad implications on economic policy making, for studies of global climate change, as well as research on sustainability and system resilience.

Some of the more engaging climate work has directly pursued a strategy of reconstructing the historicity of climate events and changes. While Emmanuel LeRoy Ladurie was using ethnohistorical data to describe climate changes on past people’s lives, Hubert Lamb was advocating that climate and human histories are dialectical, mutually interactive systems. By combining anecdotal information with hard scientific material, these two individuals’ work did a great deal to originally dispel the notion that changes in climate over time are phenomena that researchers can elucidate. Christian Pfister has been researching climate history for the past several decades in Europe, and particularly in Switzerland, synthesizing and analyzing an enormous amount of climate information from both proximate and direct sources in an attempt to reconstruct a refined chronology of European climate history. Pfister’s efforts have largely been successful as he and his research team have compiled a comprehensive database of proxy and direct meteorological observations for the past ten centuries called EURO-CLIMHIST. Leroy Ladurie and H. Lamb’s early influential work, Pfister’s research design and conclusions, coupled with a seminar directed by Dr. Carole Crumley on the evolution of landscapes in Burgundy that was taught at the University of North Carolina in the spring of 1999, provided much of the impetus for this paper.

The goals of this paper are to provide a summary overview of methodological and technical issues or data that are relevant to the historical climatology of Burgundy. This aim is not for this research paper to serve as a conclusive statement with regards to the climate history of Burgundy, but rather for this to serve as a catalyst and guide for future researchers wishing to complement or weigh their research objectives against the scale of historical climatology. Burgundy, located in East-Central France, was chosen as a focus region for this project in light of the interdisciplinary French Project led by Dr. Carole Crumley that has been operating in this area for the past 25 years. Within the larger context of this research team, a multitude of research topics have approached the common topic of the evolution of the Burgundian landscape. Previous research has traced multiscalar changes in the relationship between humans and their surroundings into a diachronic framework that encompasses more than two millennia. Research techniques and approaches within the French Project have included archaeology, ethnography, remote sensing, ethnohistory, political economy, medieval architecture, environmental science, and geomorphology to name but a few. The fundamental strength of the French Project lies in a shared vision of the Burgundian landscape as an intermediary between natural and social processes across time.

De Vries has produced a research report that accompanies and complements this historical climatology paper, and therein he provides a more detailed and thorough treatment of both Burgundy as a region and as the research location for the French Project (also cf. Crumley & Green 1987 for more detailed information). However, Crumley’s description of the historical climatology of Burgundy remains the seminal and most comprehensive treatment of the topic. Crumley outlines the diachronic relationships between and among environmental and human systems since the Neolithic period in Burgundy, and puts forth a concise chronology of Burgundian climate and social history since 900 B.C. Additional and more detailed information on the French Project in Burgundy is also available on the World Wide Web at: [http://www.unc.edu/depts/anthro/french].

The body of this paper is divided into five distinct sections. The essay begins with a section devoted to an on-going research project, known as EURO-CLIMHIST and headed by the Swiss historical climatologist Christian Pfister, along with some of the methods and conclusions utilized by his team. The following section will then proceed to separate discussions of historical climatology in the 19^th- and 20-th centuries, as well as paleoclimatological research. All three of these sections deal with texts that are of potential use (although sometimes only marginally so) to French Project members interested in refining or expanding a historical climatology component. I conclude this paper by presenting some additional print and electronic (eg. Internet) sources for procuring climatological data for France and with specific reference to Burgundy, in order to facilitate the work of future French Project researchers.

Review and Discussion of the Literature and Climate data

The difficulties in either initiating a historical climatology research project or incorporating feature elements of historical climatology into a larger extant research scheme are numerous. First, the nature and availability of climate data fluctuates depending on the geographical location of the research. Certain localities, in Europe for example, have a larger written record than other places, and as such, ethnohistorical documentation can theoretically provide a wider range of climate-related information to researchers. Of course, this breadth of information sources can sometimes complicate the historical climatological reconstruction of a place by virtue of the multiplicity of data types and forms that must be factored into an analysis. A relatively lengthy European climate chronology in turn compels the researcher to pursue and acquire data from a number of different sources. However, this seemingly daunting task has been successfully met at least once recently through the work of a Swiss research team led by Christian Pfister.

1) The EURO-CLIMHIST research project:

EURO-CLIMHIST is a collection of climate data which at first encompassed 1000-1525 AD and that now includes data until the present. The data were obtained from over 30,000 proxy documentary sources from Western European countries such as France, England, Germany, Benélux, and Austria, with a majority of the data originating from northwestern Switzerland. Christian Pfister originally conceived this database (now available in digital format, although still not in a relational database) as a method for chronologically integrating proxy climate data and its description into quantitative estimates of monthly precipitation and temperature averages. Proxy data here includes direct or indirect climatic measurements, and consists of natural (e.g. geophysical, dendrochronological, and biological) as well as anthropogenic data (e.g. instrumental observations, descriptive personal accounts).

The incipient efforts towards constructing a pan-European dataset constituted of multi-proxy sources originated in a conference held in 1990. This conference was an interdisciplinary meeting of researchers engaged in various aspects of paleoclimatological work. The time period covered during this conference (1675-1715) is known as the "late Maunder Minimum," a rather brief but intense period of climatological phenomena. The collaborations emanated from this conference ultimately led to the creation and development of the EURO-CLIMHIST database under the direction of Dr. Christian Pfister at the University of Bern, Switzerland. The conference proceedings are folded into a volume that addresses both the evaluation and appropriateness of paleoclimatic reconstruction methodologies. This volume includes a section that outlines regional and local reconstruction of climatic conditions from 1675-175 for Eastern and Western European localities. Of particular interest to French Project researchers are the chapters by C. Pfister on Paris and Switzerland. The last focus of this edited volume is upon the efficacy and accuracy of mapping paleoclimatic change and events based upon a scheme of multi-proxy sources.

EURO-CLIMHIST has a developed system for verifying and evaluating data reliability; essentially that a record was made in both temporal and spatial proximity to the phenomena observed (i.e., it is preferable that a person have written in their diary about the conditions during a snow storm in their own village rather than those in a neighboring village). The database takes qualitative observations and assigns them a numerical value or code, thereby allowing for statistical analyses. It is thus critical that all EURO-CLIMHIST participants observe the same standards in data entry lest inconsistencies dilute the validity of constructive data. However, the description for processing and manipulating the EURO-CLIMHIST data found in Frenzel’s edited volume is no longer current as Pfister’s team has since upgraded the software package for its database and is attempting to make it directly available to the public before the fall of 2000.

One of the most useful outcomes of the EURO-CLIMHIST project to date has been the elucidation and inventorying of severe climatic anomalies, particularly those that have occurred during the 800 years. This result has allowed Pfister to trace the outlines for a "macro-history" of European climate changes and weather patterns, as well as giving a detailed local climate history for Bern, Switzerland (where a significant portion of the data retrieved for the EURO-CLIMHIST project was originally produced). This research project has placed an emphasis on regional level (here Europe) climatological analyses that create an attractive discursive and analytical position for meditating between issues of local and global climate histories.

2) 19^th-century climate information:

Harington’s edited volume, Year without a Summer (1992), is comprised of papers compiled for a 1986 conference hosted by the Canadian Museum of Nature. The purpose of the conference was to unite specialists from a variety of disciplines, including historians, dendrochronologists, volcanologists and geographers, to discuss the global climatological phenomena associated with the massive 1815 eruption of the Tambora volcano in Indonesia. The atmospheric disturbances caused by this eruption clearly effected weather and climate patterns over all continents. The title of this volume was specifically chosen in reference to the generally cold regional conditions that resulted in Western Europe, eastern North America and China.

Hubert Lamb’s chapter details his previous research that reproduced diachronic reconstructions of global atmospheric circulation since 1750. In the case of the year 1816, Lamb mobilized a wealth of archival information as proxy data to demonstrate that the cold weather in western Europe could be attributed to a prevailing concentration of low pressure systems situated over Europe for an unusually long period of time. However, while Lamb’s brief description of atmospheric phenomena is illuminating, he makes no attempt to explicitly frame his conclusions about European climatic conditions in 1816 in relationship to the Tambora eruption, which was after all his starting thesis and the purpose of this edited volume.

In the same tome, John Kington has a chapter of immediate relevance to this research report and to the climate history of Burgundy. Here Kington scrutinizes the ethnohistorical documents produced by the German meteorologist Heinrich Branes in 1816. Branes goal was to collect weather data for all of Europe. But as there was no network for systematic recording of observations extant in 1816, Branes instead had to rely on earlier climatological recordings made by the Société Royale de Médecine and the Societas Meteorologica Palatina during the 1780s. This chapter presents a map with the locations of the 1780 observations stations scattered throughout Europe from Lisbon, to Reykavik, to Moscow. Interestingly enough, the densest concentration of weather stations on this map appears situated in and around the Burgundy region (361). This map and the original from which it is drawn may be useful starting points for identifying climatological data sources of potential use to the French Project. Kington was able to broadly compare the effects of the Tambora eruption on regional European climate in 1816 with similar effects of volcanic activity in Japan and Iceland in 1783. Thus Kington’s line of inquiry opens the door for further comparison of historical climate data with other known eruption events. Finally, this article lists the contact information for the agency responsible for the Société Royale de Médecine and the Societas Meteorologica Palatina data:
 

The last chapter from Harinton’s edited volume that I will present here is by K. P. Briffa & P. D. Jones. Using multiple long-term climatic records from across Europe, these authors demonstrate that the unusually extreme weather of 1816 was seasonally confined to the late spring and the summer. Furthermore, the authors postulate that the climatic conditions may be considered particularly anomalous when placed in the context of the broader climatic norms for the early 1800s. Perhaps the most useful item of interest to French Project members is the comprehensive list of forty-six weather stations in France that have continuously recorded weather information from 1810-present. In relationship to Burgundy, there are four different stations that have extensive records for surface temperature and could be considered as regional proxy references: Paris and Châlons in France, or Geneva and Basel in Switzerland. A word of caution is warranted here for those readers intending to consult these records, as it is quite conceivable that some stations haven’t maintained the same standardized methods for collecting or recording weather observations, and as such, the provenience of all data must be carefully documented before being utilized. The authors also present a table that lists the names of all the sites that have continuous records of precipitation since 1810, with the stations in Paris or Nancy being the closest in relationship to Burgundy. Finally, Briffa & Jones briefly discuss the nature of the dendrochronological data recovered for Europe during this period. While the authors argue that tree-ring records also indicate that 1816 was a year of unusually limited growth, their dendrochronological analysis stems from a limited sample number from a small number of sites, that in turn represent a poor range of geographical variability in Europe.

3) 20^th-century climate information:

Philippe Lamarque’s research is particularly relevant for future French Project members wishing to test the accuracy of historical climate data. Using temperature and precipitation data from several dozen weather stations in the lower Rhône valley, Lamarque offers a methodology for testing and validating series of historical climate data in reference to baseline climate data so as to determine their spatial consistency and patterning. Using data from Météo-France’s database (cf. discussion in this paper, section 5), Lamarque’s study described statistical means for detecting changes in the mean of climate series values, as well as a method for determining when "inhomogeneities" [sic] occur. From the results of this author’s study, we can conclude that reconstituted climatological data may be useful measures for interpolating climatic trends (for a similar test run at a larger regional scale, cf. Pfister & Lauterberg).

The systematic recording of climatic observations expanded during the 20^th-century concurrently with the investigations into the relationship between climates and human social activities. Geographers have been engaging questions surrounding climate change and effect in part because climate necessitates an integration of time and space variables. However, geographers examine climate through either a physical approach or a social approach. That is, geographers most frequently study climate as a network of physical processes operating within a climate system or as a variable that both directly and indirectly impacts human life at local, regional and global scales. Both physical and social perspectives are useful devices, but as social scientists, it is imperative that some shape of the two be synthesized and incorporated into a research design such as that employed in the French Project.

Masatoshi Yoshino’s edited volume is a collection of papers written by geographers. While it is not focused on historical climatology per se, this tome is a cogent piece of work on the relationship between humans and climate, and is a helpful way for the uninitiated researcher to gain a better understanding of how geographers measure, model and predict climate variability. The most significant value of this volume is perhaps that it frames much of the current research in climatology in relation to the institutional initiatives being carried out by such organizations as the United Nations Environment Programme (UNEP), the World Climate Programme (WCP), and the International Geographical Union’s (IUG). Yoshino’s volume elucidates the context and audience for whom climate-related data is most often being produced. This work is also beneficial as it explicitly frames the relevance and importance of climatological studies in reference to the human communities that must respond to changes in climate. Through a series of case studies from around the globe, of which Doguédroit et al.’s chapter is the only one dealing directly with France, Yoshino’s edited volume deals with the pertinence of climate to issues such as land use, dynamic agricultural responses, and resource management strategies.

4) Information on Paleoclimatology:

Paleoclimatology is the study of past climate conditions and their underlying processes prior to the use of instruments in the recording of weather observations. Paleoclimatologists employ a battery of methods for climate reconstruct: dendrochronology, palynology, lake and ocean sediments, ice core data, and coral reef formations. However, the central goal of most paleoclimatological research is to anticipate and simulate future climate changes, rather than to retrodict and interpret the importance of past climatic conditions for human populations. The USA’s National Oceanic and Atmospheric Adminstration actively pursues an agenda of paleoclimatological research and data collection. Their Paleoclimatology program maintains a database of global climate scientific data that relates to climate system variability and predictability. NOAA has made much of their database available to the public on the World Wide Web, and it is easily accessible and searchable by both geographical location and data type. Additional information is available from the NOAA website at: [http://www.ngdc.noaa.gov/paleo/paleo.html].

For those researchers searching for a contextualization of human-environment interactions in Europe within a much longer time scale, I suggest taking a look at the work of geologist and palynologist Josette Renault-Miskovsky’s (only published in French). This author, who specializes in paleoclimatology, offers in the first half of her book a wide survey and detailed description of the paleobotanical and paleofaunal data available to researchers investigating the distant past. In the second part of this work, Renault-Miskovsky examines the broad sequence of environmental conditions and their implications for the chronology of human organization, with special attention paid to sites in Mediterranean France. This theoretical emphasis reveals, in my estimation, the author’s under-privileging of human agency in their interactions with natural environment, and I would even goes as far to say that the author discounts the dialectical relationship between humans and their landscape(s). In spite of this fact, there are useful data that may be gleaned from this volume, particularly the Upper- to Late-Paleolithic tables that superimpose geological, biological and human chronologies. However, this volume is really only useful as a general introductory overview and is not a good source of proxy information in terms of the historical climatology of Burgundy.

5) Additional sources for procuring historical climatological data for France and Burgundy:

If none of the aforementioned texts and references offer a future French Project team member useful information, there are alternative avenues to pursue data acquisition. The main data center in France for climatological information is the national weather service. This office can be contacted at:
 

There are a number of data sets that are potentially of great use and relevance to researchers working in Burgundy. These data sets are available in both digital format and hard copy, and the data can be requested for specific weather stations. The closest and probably most useful weather stations for French Project researchers to consult are located in St-Yan (a French airforce base south of Digoin), Nevers, Macon or Dijon (Crumley, personal communication). The Dijon meterological observation site is housed at the city museum and can be contacted directly:

Musée de Dijon
Jardin de l'Arquebuse
21000 Dijon
Phone: (33) 03 80 76 82 76
[http://www.meteo.fr/comprendre/sites_tourismes/dijon.html]

In the following two paragraphs, I briefly summarize the description of data sets available from the National Meteorological office that may be of most immediate utility to researchers in Burgundy. Data set #0250 represents the surface climatological data for individual weather stations in France from 1949-present, and is available in digital format. Elements recorded in this set include precipitation, air temperature (minima & maxima), wind, air pressure, humidity, cloud cover, visibility, evaporation, evapotranspiration and soil temperatures. This data is also statistically arranged into daily, weekly, monthly and annual presentations of climatological means, extremes, and frequencies. Data set #0249 is a collection of climate observations from 1951-1980 similar in nature to those in the preceding data set. These data, however, are presented as a 30-year period of climatological normals that have become the standard reference for meteorologists, as recommended in the World Meteorological Organization (WMO) guidelines, and could potentially provide a baseline for comparison with proxy climate data retrieved elsewhere.

Data set #0251 presents pluviometric data for France displayed by individual recording stations for 1860-present. This data is available in digital format and includes daily observations recorded in situ both manually (pre-20^th-century) and with automatic instrumentation. These data are arranged into daily, weekly, monthly and annual presentations of pluviometric means, extremes, and frequencies. Data sets #0254 (for years 1802-1921) and #0255 (1921-present) are only available as hard copies from individual weather stations. Climate elements in these sets include precipitation, air temperature (minima & maxima), wind, air pressure, humidity, evaporation, soil temperatures and cloud cover. There are also lists of proxy data sets, such as weather diaries, proceedings from scientific meetings and newspaper weather records available on the World Wide Web [http://www.wmo.ch/web/wcp/wcdmp/infoclim/category/cat_J.html].

All of the aforementioned data sets are available free-of-charge to researchers and teachers following Resolution 40 of the World Meteorological Organization in June 1995. According to this document, the international exchange of meteorological information and data between parties whose nations are members of the WMO is to be freely provided to individuals and institutions, as long as their purpose and goal are non-commercial in nature. The National French Meteorological office honors the WMO resolution and will provide climate data for free if asked, but they will not cover shipping and handling costs. There are several notable exceptions to the free acquisition of climate data from French Meteorological office: 1) If the research is studying climate history in order to provide context to a project with legal or commercial obligations (for example, a preliminary site survey prior to construction). And 2) if the research is conducted for the purpose of direct or indirect material gains for an individual or a group (such as cost reduction, real estate risk assessment, or productivity amelioration).

Another excellent location to begin research into historical climatology is at the Laboratoire de Météorologie Dynamique (LMD) of the Centre Nationale de Recherches Scientifiques. This office houses a research program which consists in the study of the mechanisms, the evolution and the prediction of meteorological and climatic phenomena in its largest sense. More information is directly available from the LMD on the World Wide Web at: [http://www.lmd.jussieu.fr/en/Welcome.html].

Future Directions and Conclusions

There are varieties of proxy data types that may be of use to the French Project member who wishes to incorporate some combination of the historical climatological issues raised in this paper. These include archival documentary records, cadastral profiles, demographic and epidemiological data, dendrochronological records, hydrological data such as riverine discharge and lake sediment cores, and ethnohistorical or ethnographic information. I must reiterate here that caution must be used when manipulating ethnohistorical and archival materials as these should, if at all possible, be independently tested against other forms of data. Unlike some researchers, I would not go so far to say that these historical data sets should be discounted if they cannot be separately verified with contemporary quantitative measurements. I do believe though that if possible, data should be scrutinized for its reliability in topics unrelated to climate observations. The misgivings about unreliable data stems from the difficulties in interpreting historical documents’ dating systems as well as the subjectivity of the author’s impressions (of cold, for example), which are undeniably valid concerns and should be taken into account.

A research project that accumulates findings from a variety of data sources will probably need to consider data management protocols as part of their overall research design. As with true with many undertakings, data management and retrieval can often play as much of an important role in the success (or failure) of a project as the ability to procure data in the first place. Therefore in my estimation, it is imperative that research plans look ahead to the data analysis process and decide from the outset how data can best be managed and where it will be stored. For qualitative data recovered from ethnohistoric documents for example, I suggest using the NUD*IST software package which was originally designed for ethnographic data management. NUD*IST is a non-numerical, unstructured database designed to store and retrieve text, establish a code (or index) for that text data, organize these codes prior to data analysis and create models for patterns in the data. Statistical or quantitative climate data records should be stored in a relational database, such as Microsoft Access, or in a Geographic Information System (GIS) database if there are immediate plans to graphically or cartographically display the research results. GIS packages are also critical for any sort of spatial analysis of climatological phenomena. The GIS software package currently in use by the French Project is GRASS, although other programs such as ESRI’s ArcView or ArcInfo are just as, if not more effective and powerful analytical tools.

Last updated 3/5/2007.