Reliability of Documentary Data

In relation to the context in which documentary evidence was generated we need to distinguish between individual and institutional sources:

Sources produced by individuals may have a very high – daily or sub-daily – resolution (cf. weather diaries) including narratives of all weather elements, which seem to be needed for describing a climatic event. At best, they also refer to societal consequences of extreme events and to measures taken by the authorities and households. In contrast, authors are subjective in the selection, to some extent also in the interpretation of events. Usually, the records include gaps and they end at the latest with the death of the observer. Original records written down within the lifetime of a chronicler are in most cases correct. Records copied from older chronicles and compilations (i.e. chronological arrangement of texts on climatic anomalies and (natural) disasters originating from different sources) are error-prone, particularly with regard to dating. Therefore, Euro-Climhist strictly distinguishes between contemporary and non-contemporary records.

Sources produced by institutions (e.g. hospitals, bishoprics, municipalities, military, ecclesiastical, or civil authorities) constitute the second category of documentary evidence. Institutional bodies were typically not primarily interested in describing climate and they often kept detailed accounts in order to document their agricultural activities in case of auditing or enquiry. The date of those activities depended on climate and weather and, therefore, constitute relevant information for historical climatology. In communities with wine production, grape harvest dates were fixed according to the grape ripening to provide a correct delivery of the tithes to the benefitted authorities. The same procedure is known for the beginning of the grain harvest. The administrative routines involved some standardization in the way records were kept, and more importantly, institutions unlike individuals often worked in the same way for centuries and in doing so they generated very long records. They are easily quantified and statistically related to instrumental measurements, unless there is a sufficient overlap between the proxy and the instrumental series (cf. data processing). Temperatures and precipitation in the pre-instrumental period may be estimated in this way, albeit just for periods of several months.

The comparison of different data types reveals whether they complete and corroborate each other. For example, the Schaffhausen chronicler Hans Stockar wrote with regard to the weather in 1522: “It was cold and rainy in April, May and June. I had to put on a fur coat for not freezing to death on Pentecost [8th June] and some people were heating their living-room”.
Hence, this individual source contains clear information about the weather in April, May and June. It is highly significant for the harvest of winter grain and of grapes, whether it had been cold or warm, dry or wet during those months. Combining Stockar’s report with the findings on grain and grape harvest dates that were deduced from institutional economic sources (e.g. Wetter, Pfister 2011; Wetter et al. 2013), gives a homogenous result: The three months described by Stockar as being very cold actually delayed the grain and grape harvest in 1522 significantly.

Euro-Climhist: Query for 1522
(access 18th August, 2015)
1522-April / very cold / Schaffhausen (SH) / S: Stockar, Chronik
1522-May / very cold / Schaffhausen (SH) / S: Stockar, Chronik
1522-June / very cold / Schaffhausen (SH) / S: Stockar, Chronik
1522-June / fog, variable / Schaffhausen (SH) / S: Stockar, Chronik
1522-July 28 / rye harvest begins (209 days after New Year): late / Swiss Mittelland / S: Wetter, Pfister 2011
1522-October 21 / grape harvest begins (294 days after New Year): late / Swiss Mittelland / S: Wetter et al. 2013

Attention must be paid to the dating style. In 1582 Pope Gregory XIII introduced the new Gregorian style named after him. It succeeded to the Julian style calendar that had been in use since Roman times. Due to the fact that a solar year does not last exactly 365 days and six hours, but a bit more than eleven minutes less, a gap had yawned between the calendar date and the actual position of the sun, making a difference of about one day per century. Pope Gregory XIII ordered to leave out the ten days between 4th and 15th October of this year (and to skip singular leap years for the future). Due to the struggle of confessions between the Catholic and the Protestant Church during that time, the improved calendar was not accepted immediately in the protestant territories (but also not in some catholic and in the orthodox areas). In the Swiss Confederation it was only stepwise introduced, at first in most catholic territories, in 1700 also in most protestant territories. For phenological series such as grape or grain harvest data, any chronological information has to converted intro Gregorian style. However, this has not been done in several scientific studies due to ignorance. In particular, dating according the feasts of Christian saints that were common until the sixteenth century has to be harmonized.