Samuel Silverman, now retired from the Physics Department, Boston college, has provided NSSDC with several data sets of pre-computer-era auroral observation records and is in the process ofproviding additional such data. These data collections go back two millennia, and their organizing and archiving result, at least in part, from NASA Office of Space Science (OSS) funding based on peer review and determination of the value of the data.
The Sun is known to change on a wide range of time scales, at least as large as the multihundred-years as evidenced by the Maunder minimum, etc. Society has increasingly sparse relevant data available as it goes back further in time. With modern data and analyses space physicists have been coming to understand relations between solar output and auroral occurrence, so old auroral data have the potential for helping the understanding of solar variability on relatively long time scales.
The first-received of these data sets consists of about 1,900 manuscript pages of mid-1800s records and is the main focus of the present article. The other data sets are ASCII digital files summarizing old records from a few centuries BCE to about 1950. The data sets to date are as follows:
1. Daily Auroral Reports, southeastern Canada and Northeastern U.S., 1848-1853.
2. Miscellaneous catalogs; Asian and Classical Greco-Roman, up to 1500. This set includes files of geographic coordinates of the sites.
3. Observations from Greenland, Iceland, and the Faroe Islands. This set includes a separate file for 1920-1923 for Iceland.
4. Other preliminary world files for eight different eras starting with before 1838 and ending with 1917-1951.
The details of the digital data sets have not yet been fully expolored by NSSDC. Those data and additional data sets to be delivered will be discussed in a future article. Readers with an interest in these data sets may contact H. Kent Hills at NSSDC.
The hard copy data set, listed first above, contains auroral observatioins from southeastern Canada and northeastern United States. For each night during 1848-1853, log-book entries were made at military posts and volunteer locations, recording local weather conditions and whether an aurora was visible. When an auroral display (especially a bright one) was visible, there is often a description of it, and many of these descriptions give considerable detail. In some cases a standard sky chart is included, with markings to indicate the details. Currently, NSSDC can supply paper copies of these pages. Depending on requester interest, NSSDC will consider making digital scanned copies. Readers may contact H. Knet Hills at NSSDC for further information.
The fuller description of this data set is as follows:
Daily Auroral Reports, Southeastern Canada and Northeastern U.S., 1848-1853.
This hard copy data set consists of approximately 1,900 legal-size (8 1/2 x 14 inches) pages of auroral observations systematically recorded at several stations in the northeastern U.S. and adjacent southeastern Canada under the direction of Joseph Henry, first secretary of the Smithsonian Institution, and John Henry Lefroy, then a captain in the Royal Artillery of Great Britain and director of the Toronto Magnetic Observatory in the 1840s and early 1850s.
Lefroy initially received permission to place registers in the guard rooms of military posts, to be kept by the noncommissioned officers at Quebec, Kingston, London, and Montreal, and this was later extended to other parts of Canada. Henry instituted a network of voluntary observers of meteorological parameters. The Smithsonian provided their volunteers with instructions for the auroral observation that were similar to those provided to the Canadian military observers by the directors of the Toronto Observatory. Lefroy then used the Smithsonian instructions to extend and improve his own.
For each night from January 1848 through April 1853, log-book entries were made for each of the locations that made observations, recording local weather and sky conditions and whether aurora were visible. When an auroral display (especially a bright one) was visible, there is often a description of it, and many of these descriptions give considerable detail. In some cases a standard sky chart is included, with markings to indicate the details.
Apparently, only the Canadian military observations were duly recorded every night. The principal reporting stations are Montreal, Kingston, London, Toronto, and Quebec. When the Canadian entries show aruoras present, then the manuscript often inculdes (in different handwriting) also reports from other locations in neighboring regions of the United States. But when the Canadian entries show no aurora present, then there usually is no U.S. entry. It is unknown if this is due to a decision by the compiler of the manuscript or to the less systematic observations of the volunteers, who may have made no record if there were no auroras visible. The Canadian entries cover each day of the more than five-year time span with the exception of four missing months from April through July of 1850.
Despite original intentions of the two organizers, publication of the combined observations never happened, although the dates and places of the auroras were published in the general catalog [J. Lovering, "On the Periodicity of the Aurora Borealis," Mem. Am. Acad. Arts Sci., 10, 9-351, 1866-1871]. The pages of this data set are reproductions from the original unpublished manuscript in the U.S. National Archives. Background information on these observations has been provided (S. M. Silverman, "Joseph Henry and John Henry Lefroy -- A Common 19th Century Vision of Auroral Research," EOS, TRANS. AM. GEOPHYS. U., 70, no. 15, pp. 227-240, April 11, 1989).