On the earthquakes in the Northern Baltic Shield in the spring of 1626

ORI GIN AL PA PER

On the earthquakes in the Northern Baltic Shieldin the spring of 1626

R. E. Tatevossian • P. Mantyniemi • T. N. Tatevossian

Received: 25 August 2009 / Accepted: 2 March 2010 / Published online: 27 February 2011� Springer Science+Business Media B.V. 2011

Abstract This study starts from the earthquakes of May 14, 1626 and June 22, 1626 as

given in existing parametric catalogs for the Baltic (Fennoscandian) Shield. The first shock

is located in North-western Russia, the second in Finland that belonged to Sweden at that

time. A search for previously unknown Russian sources of information is performed, and

secondary Swedish sources are replaced by primary ones. The contemporary sources are

two Russian chronicles and two Swedish manuscripts. In addition, a later reminiscence is

used. The contents of the sources are critically analyzed and augmented with background

information. A new interpretation of one Swedish manuscript is presented. The earthquake

dates are analyzed. A credible source of errors follows from the different calendars used.

Macroseismic intensity assessment is possible for two places. A new solution of one

earthquake felt in both territories is proposed. The available data are too fragmentary to

prove it beyond doubt, but the scenario is feasible in many ways. Tentative earthquake

parameters are calculated. Epicentral intensity is assessed at 6–7 (EMS), magnitude is

estimated at 4.7–5.7. The epicenter is located in Russia close to the border between the two

territories.

Keywords Historical earthquakes � Historical seismology � Macroseismic sources �Baltic (Fennoscandian) Shield

R. E. Tatevossian (&) � T. N. TatevossianInstitute of the Physics of the Earth, Russian Academy of Sciences, B. Gruzinskaya 10,Moscow 123995, Russiae-mail: [email protected]

P. MantyniemiInstitute of Seismology, Department of Geosciences and Geography, University of Helsinki,POB 68, 00014 Helsinki, Finlande-mail: [email protected]

123

Nat Hazards (2011) 57:133–150DOI 10.1007/s11069-010-9516-7

1 Introduction

The Baltic (Fennoscandian) Shield in North-eastern Europe is situated in an intraplate

domain that exhibits low seismic activity (Fig. 1). The Precambrian of the study region

consists of the Archaean Domain extending to the Kola Peninsula in the northeast. The

available seismicity records comprise two earthquakes close in space and time in 1626:

those of 14 May and 22 June. They stand out from the earthquake list in that they are

followed by an interruption of entries for almost 125 years. Also, the given magnitudes are

among the highest determined in the region. Therefore, the events are crucial for the

accurate assessment of recurrence intervals and seismic hazard in the region.

According to the New Catalog (1977, 1982), the earthquake of May 14, 1626 was of

magnitude 5.4, which is the largest in the subregion. The other magnitude is estimated at

4.7, both with an accuracy of ±1.0 unit (Fig. 1). The first shock is located in North-western

Fig. 1 A regionalseismotectonic map. Light graycircles are earthquake epicentersof magnitude ML2-4.3 recordedbetween 1998 and 2007according to the Fennoscandianearthquake catalog (Ahjos andUski 1992). Dashed red linesdelineate mapped shear zones asfollows: Bothnian Bay (BBZ),Central Lapland (CLZ),Hirvaskoski (HSZ),Kandalaksha-Puolanka (KPZ),Kuusamo (KSZ), Oulujarvi(OSZ), Quark Shear Zones (QSZ;Koistinen et al. 2001). Starsdenote epicenters given for theearthquakes of May 14, 1626(in Russia) and June 22, 1626(in present-day Finland) in theNew Catalog of Kondorskayaand Shebalin (1977, 1982). Errorbars show the accuracy oflocation. Magnitudes, epicentralintensities (MSK-64) and depthsare shown. Brackets are used inthe New Catalog to markdoubtful solutions. Figure framescorrespond to the seismic regionof Baltic Shield (region XIII) inthe New Catalog. Black lines arecurrent state borders betweenFinland and Russia

134 Nat Hazards (2011) 57:133–150

123

Russia, the second in present-day Finland that was Swedish territory at the time of the

occurrence. The earthquake of May 14, 1626 has been reconsidered several times. For

example, Stepanov and Pribylova (2003) evaluated a magnitude of 5.0, while Nikonov

(2004) suggested an increase to 6.5. In contrast, the information available for the earth-

quake of June 22, 1626 confines to entries in parametric catalogs only. Panasenko (1969),

the New Catalog and Penttila (1978) provided an identical epicenter; Ahjos and Uski

(1992) modified both coordinates by 0.2–0.3�. The respective magnitude estimates range

from 4.6 to 5.2.

The intent of this contribution is to verify the initial documentary data available for the

earthquakes in the Northern Baltic Shield in the spring of 1626. First, the sources are

reviewed, then the possibility to find previously unknown sources is carefully investigated,

and finally the scope of the information available for the earthquakes is evaluated and

discussed. Possible earthquake scenarios are presented, and tentative earthquake parame-

ters are estimated for the new solution.

2 The earthquake of May 14, 1626

2.1 Sources for the earthquake in May 1626

The starting point for the study on the earthquake of 14 May is notes by Yakovleva (1958)

referenced in the record line of the New Catalog (1977, 1982). Yakovleva found a later

manuscript sewed to the Pogodin’s collection of 1563 manuscripts. It describes events in

the Moscow State and, in particular, in Northern Dvina region in the 1500s and 1600s.

Historian M.P. Pogodin titled the manuscript as Records on Dvina events in 7061 (1553)–7134 (1626) (Zapiski Dvinskikh sobytiy v 7061 (1553)–7134 (1626)). On the other side of

the page 105 of the Pogodin’s collection, there is a record on earthquake on May 4, 1626

felt on the shores of the White Sea1:

In year 7134, May in the 4 day, earth shook in all Pomor’e, on Solovki and in Ust’-Kola, and on Dvina up to Siskoy Monastery.

Nikonov (2004) related another document to this earthquake. It is Dvina Chronicle(Dvinskiy Letopisets), a chronicle compiled in Kholmogory. It includes the following

report (PSRL, 1977 v. 33, p. 172)2:

1627. A terrible quake. Under governing of this voevoda in Dvina occurred a terriblequake in 135, May, day 20, in All Saints week, at 5 o’clock before Monday, because of oursins anger of God shook the earth, and many people saw this quake, but some of them weresleeping. And God save the people from the quake.

Novgorod, Kholmogory, Velikiy Ustyug, Vologda and Perm’ were the main centers of

documentation in Northern Russia until the fall of Novgorod in 1478, when it was annexed

to the Moscow State. Chronicles compiled in Novgorod and Perm’ end before the 1600s.

Compilation of chronicles continued in Kholmogory, Velikiy Ustyug and Vologda (Fig. 2).

When verifying all the chronicles compiled there (PSRL, 1959, v.26; PSRL, 1977, v. 33;

PSRL, 1982, v.37), it became clear that the only information on the event of interest is

given in Dvinskiy Letopisets (Table 1). This chronicle was published for the first time in

1774 by G. F. Miller, the second time in 1791 by N. I. Novikov. Both editions have been

lost. A. A. Titov prepared the third edition in 1889 on the basis of four handwritten copies

1 Texts in original languages can be found in the Appendix.2 PSRL–Polnoye sobraniye russkikh letopisey (Complete Collection of Russian Chronicles).

Nat Hazards (2011) 57:133–150 135

123

of the chronicle. Only two of them survived until now, including the one of the Archae-

ographic Commission, so it is not possible to reconstruct his work. A. A. Titov used mostly

his own copy, to which he appended pieces from other copies, without explicit description

of his actions. Therefore, when publishing the chronicle in 1977, the previous edition was

ignored, and only handwritten copies were used.

A total of fourteen copies of Dvinskiy Letopisets are still extant. They are kept in

Moscow, Sankt-Petersburg and Helsinki. Copyists often added, skipped or altered original

information, which can be the reason for confusion later. Different copies extend their

Fig. 2 Places related to the earthquakes of 1626. The red line outlines the border between Russia andSweden defined in 1617. White circles are towns and lesser centers of population. In Russia, the dashedyellow area is Pomor’e in the narrow sense, while dashed violet is Pomor’e in the broad sense according toTatishchev (2003). The red contour corresponds to ‘‘over Dvina up to Siyskoy Monastery’’ according to theassumption made in this study. The centers of documentation in Northern Russia have been underlined;chronicles compiled in these centers report on events in the first quarter of the 1600s. Distances fromepicenter as given in the New Catalog are shown. In Sweden (present-day Finland), the dashed brown areacorresponds to the province of Pohjanmaa (Osterbotten in Swedish) as it was defined in 1634, the doublydashed area to the municipality of Ii. The inset shows the rectangular area around Lake Oulujarvi in moredetail. The cross denotes the church of Paltamo located at a distance of almost 20 km from the main centerof population, and the black hexagon the fortress of Kajaani

136 Nat Hazards (2011) 57:133–150

123

descriptions to somewhat different times, but all of them cover the year of interest. As most

of the Russian Chronicles, Dvinskiy Letopisets starts by quoting fragments of older Pan-

Russian chronicles, but with time passing it concentrates mostly on local life in Northern

Russia.

Verification of chronicles associated with Northern Russia leads to the conclusion that

there are only two documents reporting on an earthquake in Russia around 1626: Recordson Dvina events and Dvinskiy Letopisets. The dates given in the chronicles are discussed in

Sect. 4.

2.2 Assessment of the sources

Several questions arise from the information given in Records on Dvina events in 7061(1553)–7134 (1626) as reported by Yakovleva (1958):

(1) The origin of the document is not clear. Where was it compiled and what are its

sources?

(2) What does Pomor’e mean? This toponim has different definitions. In the narrow

meaning it refers to the southern shores of the White Sea from Onega to Kem’. In the

broadest sense it means the huge area from Karelia up to the Urals (Fig. 2).

According to the definition by Tatishchev (2003), Pomor’e included regions of

Arkhangel’sk, Kholomogory, Vaga, Tot’ma, Vologda, Kargopol’, Charonda and

Olonets. In which sense is it used in the cited document?

(3) In the 1600s, the whole region along the Dvina River from Vychegda up to the White

Sea was named Dvina; Siyskoy Monastery is more or less in the center of this region.

What does ‘‘up to Siyskoy Monastery’’ mean: to the north or to the south of it? The

first one looks more reasonable (it is contoured in red in Fig. 2), nevertheless this is

an assumption.

Is one of the two sources of a higher priority? Dvinskiy Letopisets is a local source of

clear origin, in contrast to the Records on Dvina events. Local sources usually contain more

detailed descriptions, but here the situation is strangely the opposite. Kola and ‘‘all

Table 1 Information on chronicles related to Northern Russia

Title Year of publication Timeinterval

Place ofcompilation

Main features(Pan-Russian/local)

Descriptiononearthquake

Cholmogorskayaletopis’

PSRL, v. 33, Leningrad,Nauka, 1977, 250 pages

852–1659 Kholmogory Pan-Russian/local

No

Dvinskiyletopisets

PSRL, v. 33, Leningrad,Nauka, 1977, 250 pages

1342–1790 Kholmogory Local Yes

Ustyuzhskayaletopis’

PSRL, v. 37, Leningrad,Nauka, 1982, 228 pages.

852–1745 VelikiyUstyug

Pan-Russian/local

No

Vologodskayaletopis’

PSRL, v. 37, Leningrad,Nauka, 1982, 228 pages.

862–1700 VelikiyUstyug

Pan-Russian/local

No

Vologodsko-Permskayaletopis’

PSRL, v. 26, Moskva-Leningrad, Izd. AN SSSR,1959, 415 pages

854–1538 Vologda Pan-Russian/local

No

Letopisets L’vaVologdina

PSRL, v. 37, Leningrad,Nauka, 1982, 228 pages

1192–1797 Vologda Local No

Nat Hazards (2011) 57:133–150 137

123

Pomor’e’’ disappeared from the place names in Dvinskiy Letopisets; when reporting Dvina,

Siyskiy Monastery has been omitted. Absence of information on Ust’-Kola possibly fol-

lows from its remoteness from Kholmogory, where the chronicle was compiled, but then

Siyskiy Monastery is only 75 km to the south from Kholmogory. For the same reason, it is

difficult to explain the lack of information on Pomor’e. Because of these contradictions and

lack of information on the origin of the manuscript Records of Dvina events, it is not

possible to assess priorities to the manuscripts.

2.3 About the places related to the documents

Kola is mentioned only in Records on Dvina events in 7061 (1553)–7134 (1626). It is clear

that Kola is crucial for the epicenter position given in the New Catalog (Fig. 2). If this

locality appeared erroneously in the chronicle, the parameterization of the earthquake of

May 14, 1626 is influenced dramatically. Grounds for these speculations can be found out

by carefully examining the maps showing the historical and geographic background of

Kola and Pomor’e in 1626 (Fig. 3). The earthquake occurred during a flourishing time for

Kola and the entire Russian Lapland that lasted from the mid-1500s to mid-1600s. Fast

declination of the region only started in 1665, when the custom rules were toughened, after

which most of the population left Kola (Golubtsov 1911).

For the first time, Kola was drawn on a European map in 1539 (map of Olaus Magnus).

All coastal area of Russian Lapland was rather well developed. Besides the maps, there is

Fig. 3 Contemporary maps of Northern Russia and neighboring states: a Upper part: A miniature map ofNorthern Russia. The arrow points at Kola stockade town. Published in 1601 in Frankfurt by J. Keerbergen,L. Hulsius, drawing by Abraham Ortelius. Lower part: Map fragment from the Atlas � Speculum OrbisTerrae � 1593. Cartographer: Anthonio Iankinsono Anglo, Antverpen. Places mentioned in the text areunderlined. b A fragment of a map of Denmark, Sweden and Norway. � Novbus Atlas � 1635. Most ofKola is mapped rather accurately

138 Nat Hazards (2011) 57:133–150

123

direct evidence (reproduced by Golubtsov 1911) of active life in the region between 1550

and 1650. For example, the monastery in Kandalaksha was constructed in 1532 and that in

Pechenga in 1533. Kola became a permanent settlement between 1533 and 1550, when the

construction of Kol’skiy stockade town started. The construction was completed in 1582.

Kola became the regional administrative center. First voevoda (governor) of Kola inau-

gurated a market place and arranged custom services. The city walls were built in 1583

(cf. Fig. 3). A year before the earthquake, there were 500 soldiers and artillery in the Kola

garrison (Shaskol’skiy 1962). Kola remained an administrative center of the region during

the first quarter of the 1600s. That is why any information on events within the region

could enter the outer world through Kola. Afterward, the information may start to be

associated with Kola. This phenomenon is well known in historical seismology. For

example, there is a big cluster of earthquakes in Shiraz, Iran, when it was the capital of

Persia (Tatevossian et al. 2003). It should also be noted that some important localities

along the trade road such as Kandalaksha (distance 190 km from the given epicenter),

Kem’ (130 km) and Varzuga (65 km, i.e. almost six times closer to the given epicenter

than Kola) are not mentioned in the chronicles.

At the time of the earthquake, both Siyskiy and Solovki monasteries were very well

linked with Moscow. Siyskiy Monastery was founded in 1525 in the small island of Lake

Mikhailov. The monastery gradually grew more powerful, and in ca 1545 its buildings

could be found in Kholmogory, Una and Nenoks. In 1579, the monks’ property included a

large piece of land extending to ‘‘Siya and Khorobritse–6 versts [a verst ca 1 km] and

toward Kargopol’–50 versts’’ (Slavyanskaya entsiklopediya…, 2004). The monastery

became an important clerical and administrative center of Dvina land well known in

Moscow. Three wooden churches were built in 37 years, and a set of masonry buildings

was formed during the second half of the 1600s. The monastery had a town church in

Kholmogory, where the chronicle Dvinskiy Letopisets was compiled.

Solovki. The monastery was founded at the end of the 1400s. It became the clerical

center of Pomor’e very quickly and was of Pan-Russian importance in the 1500s. Active

construction works were started at that time. The monastery became a key point in the

defense system of Pomor’e, when Swedish activity grew in the region in the late 1500s and

early 1600s. A permanent garrison was formed, and it had as many as 1040 soldiers

between 1614 and 1626. Solovki was so crucial in defending Pomor’e that the tsar waved

aside taxes for 5 years and presented it large pieces of land. However, the monastery ended

up in opposition to both religious and civil authorities, when it refused to accept the clerical

reform of patriarch Nikon in 1653. Troops sent from Moscow took over the monastery in

1676, and from that time it lost its importance. Both the Solovki and Siyskiy monasteries

were closed in the 1920s. This caused destruction and loss of local documents.

3 The earthquake of June 22, 1626

3.1 Sources for the earthquake in June 1626

The parametric earthquake catalogs that include an entry for June 22, 1626 rely on the

descriptive catalog of Renqvist (1930). It gives three sources: the New Finnish Manual(1693), Mathesius (1734) and Messenius (1865). Given the long delay between the event

and these sources, there is obvious doubt whether they can be regarded as original. A closer

examination showed that the New Finnish Manual (1693) only contains a remark of a large

earthquake in the province of Pohjanmaa (Osterbotten in Swedish) in 1626. Mathesius

Nat Hazards (2011) 57:133–150 139

123

(1734) is a geography thesis focusing on the same province. It mentions damage to the

church in Paltamo due to an earthquake on the day of Eschilli (Eskil in Swedish) in 1626.

The source for this information is a description of the municipality of Paltamo by Johannes

Andreae Cajanus in 1663. Messenius (1865) stands for the second edition of Rhymechronicle of Finland. The other editions appeared in print in 1774 and 2004 (Messenius

1774, 2004). The third edition was prepared using strict scientific criteria and the original

manuscript from the 1620s. Prior to Renqvist (1930), Moberg (1894) listed seismic

occurrences in present-day Finland, relying on Thuneld (1794) in the 1626 case. Thuneld

(1794) is the third edition of description of the geography of Sweden. This source was

classified as secondary, because it gives the same information about the church sustaining

damage in an earthquake in 1626.

Rhyme chronicle of Finland contains narratives of fictional and real important persons

and political events until 1628, written in rhyme. The author, Swedish Johannes Messenius

(about 1580–1636), was most probably an eyewitness to the only natural phenomenon

mentioned in the text. His exact whereabouts at the moment of the earthquake are known:

he was held political prisoner in the fortress of Kajaani not far from Lake Oulujarvi from

1616 to 1635 (Fig. 2). Messenius was a prolific writer at a time when illiteracy was far

more common. What is missing from Renqvist (1930) is the main study of Johannes

Messenius, the monumental Scondia illustrata (the description of Scandinavia), completed

in the 1630s. It was published in the early 1700s. A translation from Latin into Finnish in

1988 was used in this study (Messenius 1988). The 10th book of Scondia illustrata is

devoted to the history of Finland and contains a few comments on the earthquake. This

work and Rhyme chronicle of Finland are not independent of each other:

Next I want to write about a strange thing [1626]

it is an earthquake

that travelled through the entire Finland

One cape only suffered damage in the municipality of Ii

it tilted so that

it sank in an awful way.

In the deep water it is seen no more

than the tops of tall trees.

An earthquake is never followed by anything good,

which has also been prophesized.

The following year, in the same municipality,

one day, the nineteenth of April,

it rains almost only blood.

This is not a good sign either.

God be merciful and lenient with us

for the sake of His Son Jesus Christ.

Messenius J, Rhyme chronicle of Finland

[1626] When dawn was breaking on the 13th of June, Finland was shaken by a

terrible earthquake. It made the ground sink over quite a considerable area in the

municipality of Ii, which made possible the creation of a lake in this very place.

Messenius J, Scondia illustrata

The description of the municipality of Paltamo (Cajanus 1663) that includes a remark of

a church suffering damage due to an earthquake in 1626 is not contemporary with the

140 Nat Hazards (2011) 57:133–150

123

earthquake. Nevertheless, it has value, because it is independent of the texts of Messenius.

The author, Johannes Andreae Cajanus, spent his entire life (1626–1703) in Paltamo, except

for school and studies elsewhere between 1636 and 1648. He served as the vicar for about

the last 40 years of his life. Cajanus gave a testimony to the consequences of the earthquake:

This church (…): but in the big earthquake that happened over most of Pohjanmaa

[Osterbotn] in 1626, on the day of Eskil, it was badly broken, so that the eastern

gable leant; and since that time the peasants have been unwilling to spend on the

church, because of which it now seems to be falling

Johannes Andreae Cajanus, The description of the municipality of Paltamo

The description of the municipality was likely prepared to commemorate the inspection

of the diocese by bishop Johannes Terserus in 1663, which may have been the first ever

visitation by a bishop in the area (Jokipii 1960). Besides the history of the parish and

related persons, the description extends to the geography of the municipality as well as

ethnography and folklore of its inhabitants. The church of interest was the second in the

history of the parish, completed in 1599. It was most likely a rectangular timber building.

At the time of the visit it still existed, but in a bad shape, and preparations of building a

new church had been commenced.

3.2 A new interpretation of the sources

The impressive consequence of the earthquake in the municipality of Ii described by

Johannes Messenius cannot be understood as a tectonic land subsidence: if an earthquake

leaves permanent deformation in the ground, the magnitude has to be large, i.e. M C 6.5

(Pantosti and Yeats 1993). Had an earthquake of this size occurred in the territory, the

consequent damage would not have been confined to a single church. However, population

tables do not show any abrupt decrease in the number of houses either in Ii or in Paltamo

around 1626; instead, habitation increased quite favorably there in the 1620s (Keranen

1984, 1986; Vahtola 1998). In Paltamo, the increase in the mid-1620s may in part be

attributed to improved bookkeeping rather than an actual change. The number of houses

became stable toward the end of the decade, and only started to decline in the early 1630s,

when severe famine hit the area. It is more credible that the description refers to a sec-

ondary earthquake effect such as a landslide or subsidence of riverbanks or lakeshores, and

the magnitude of the earthquake may have been moderate.

The new interpretation is that Johannes Messenius may have located the landslide

erroneously. It is certain that Messenius cannot have been an eyewitness to occurrences in

the municipality of Ii. As a political prisoner he could neither become familiar with local

geography nor gather information freely, had he wanted to. It is not clear if he learned much

Finnish, the language of ordinary people in the area, during the long imprisonment. The

gnomic remark of it raining blood the following year may have roiled the location of the real

earthquake effects. Messenius, despite his formal education in political science and law,

shared the superstitious notion of that time that earthquakes were omens of misfortune.

The landslide may have occurred along the shores of Lake Oulujarvi (Fig. 2). This

reasoning is based on the following: (1) It is more probable that Messenius obtained some

hearsay from across a short distance (about 10 km at the shortest) rather than from far

away (170 km). (2) Many of the banks of Lake Oulujarvi consisted of sand and were

notoriously unstable (Leiviska 1914). This characteristic feature is hidden in local place

names such as Paltamo, palta or palte being a very old word that means rolling banks.

Nat Hazards (2011) 57:133–150 141

123

A plausible sequence of events is that the ground shaking triggered an amount of soil to

roll into the water. (3) Pine trees are typical of the edges of the banks (Fig. 4); in the

event of a landslide they would fall into the water as described in Rhyme chronicle ofFinland. However, Messenius supposedly preferred suitable words to actual facts when

thinking of a rhyme; the fallen trees would be in chaotic order in the water rather than

stand upright.

There are a number of possible capes by Lake Oulujarvi, but it is not possible to

pinpoint the one of interest on the basis of Messenius’ texts alone. According to Rhymechronicle of Finland, the land subsidence was the only damage following the earthquake,

but it is known from the description of Johannes Andreae Cajanus that the church of

Paltamo, not far from the fortress of Kajaani, was partly broken (Fig. 2). An explanation

may be that both incidents occurred in the vicinity of the same cape, but this remains

speculative.

Johannes Messenius provided a valuable piece of information in Scondia illustrata: the

date of 13 June. It is mentioned, as if in passing, that the earthquake occurred at dawn. This

is a detail that may tell of a genuine eyewitness. It is credible that the time of the day is

remembered and reported afterward, especially if the reporter was awakened by the event.

The earthquake location is quite insufficient (‘‘the entire Finland’’). However, these kinds

of areas of perceptibility are not uncommon in old earthquake reports (e.g. Musson 1996).

According to the other key person, Johannes Andreae Cajanus, the earthquake occurred on

the day of Eskil. The feast of St. Eskil was originally fixed to 11 June, but later moved to

Fig. 4 A view of the banks ofLake Oulujarvi. Photo: HermanRenfors, 1902. Collections of theMuseum of Kainuu, Kajaani,Finland

142 Nat Hazards (2011) 57:133–150

123

12 June except in his local diocese in southern Sweden. Cajanus reported most of the

province of Pohjanmaa (Fig. 2) as the area of perceptibility.

4 An analysis of the different dates of earthquakes

A total of four dates can be found in the sources: May 4, 1626 and May 20, 1627 in the

Russian chronicles and the day of Eskil (12 June) 1626 and June 13, 1626 in the Swedish

manuscripts. The dawn of 13 June provided by Johannes Messenius in Scondia illustratawas previously ignored by seismologists. These dates are based on contemporary calen-

dars: the Julian calendar in Sweden and the Byzantine calendar in Russia. A later adoption

of the Gregorian calendar brought a leap of ten days, thus the dates as 14 May and 22 June

in existing parametric catalogs.

Both Russian sources, also the local one, mention only one earthquake, but the given

dates differ by ca 1 year. There are not enough reasons to tell which date is correct

(if either) on the basis of the authoritativeness of the sources—they are of the same order

(cf. Guidoboni and Ebel 2009). The Russian sources mention an earthquake in the same

area, south of the White Sea, which is of low seismic activity. This may support the notion

that both chronicles report the same seismic event, i.e. there is an error of dates in one or

both of the documents. There are at least two possible reasons for it:

(1) A confusion of dates may follow from the calendar used. According to the Byzantine

calendar, the first day of the year was 1 March until the end of the 1400s, then 1

September. At the time of interest, only the September-dating should have been in

use, but in reality the use of both conventions overlapped very long. It is believed that

the first chronicle definitely compiled following September-dating was the Nikonian

chronicle from the 1650s. The 1-year difference can be explained by this: May 1627

in March-dating is May 1626 in September-dating. It is possible that March-dating

was used in Pomor’e in the north where people were conservative and refused to

accept the church reformation (including the calendar reform) started by patriarch

Nikon.

(2) There were a total of fourteen copies of Dvinskiy Letopisets and, when preparing it for

publication in 1977, the editors tried to compromise over them. It is easy to make a

mistake between years 1626/1627 (134/135) when dealing with such an amount of

copies.

A confusion of 1 year will also affect dates. In the old times, the dates (in modern form

of day and month) were derivatives (less important) of dates pinned to certain religious

events. For example, in the information ‘‘in 135, May, day 20, in All Saints week, at 5o’clock before Monday’’ the feast of All Saints week is more important, and 20 May is the

consequence of the feast day. It is a movable feast, related to the Orthodox Easter date

(8 weeks after Easter). In 1626, the feast of All Saints fell on 3 June, in 1627 on 20 May.

Therefore, another date can be added to the two explicit Russian dates, a combination of

information from both Russian documents: June 3, 1626 (corresponding to June 13, 1626

on the Gregorian calendar). It follows from the assumptions that year 1626 in DvinskiyLetopisets is erroneously reported as 1627 and that the reference to the feast of All Saints

week is crucial to dating the event.

The two June dates in the Swedish manuscripts are very close together, so a natural

conclusion is to relate them to the same earthquake. Nevertheless, there may be limitations

in the accuracy of timing. Johannes Andreae Cajanus wrote the description only decades

Nat Hazards (2011) 57:133–150 143

123

after the event. The reminiscence ‘the day of Eskil’ may in practice mean that the earth-

quake occurrence was close to it, but not necessarily identical. The testimony of Johannes

Messenius shows that there was a difference of one day at least. The original manuscript of

Scondia illustrata proves that the given date was not distorted in the later editions used in

this study; however, the reliability of Johannes Messenius is difficult to assess. Scondiaillustrata was completed later than Rhyme Chronicle of Finland, which may also be a

source for a lapse of memory.

Two main scenarios can be discerned on the basis of the available information on

dating: (1) an earthquake in Russia on May 4 or June 3, 1626 or May 20, 1627, and another

one in Sweden on June 12(13), 1626; (2) only one earthquake occurred in the spring of

1626. In favor of the scenario of one seismic event is the coincidence of origin time

reported in Dvinskiy Letopisets (5 o’clock) and in Swedish manuscript (at dawn); but it is

based on assumptions of errors in dating. An evident source of errors is the calendar used,

and mistakes in writing and copying are made. Errors in dating can be suspected with a

certain probability, but the available information is not detailed enough to prove them

exclusively. It is well recognized in historical seismology that an earthquake, especially if

felt in different territories, can split into two or more separate events in documentation.

According to the first scenario, the uncertainty of dating of an earthquake of June

12(13), 1626 could be of the order of days only. It is difficult to evaluate the uncertainty of

dating for a separate Russian event. If it is formally assigned to be 1 year, it means that the

earthquake occurred at any moment during this time interval. However, this is not exactly

the case: there is rather a choice of three fixed (discrete) options in 1 year. In such a case,

the uncertainty of timing has only a formal character. According to the second scenario, the

earthquake occurred either in May or in June, so the uncertainty of dating is of the order of

a month.

5 Discussion

Disregarding the data points of Kola and municipality of Ii as discussed in Sects. 2 and 3,

the remaining localities concentrate in the vicinity of the White Sea and one point is

located by Lake Oulujarvi in present-day Finland at a distance of about 400 km (Fig. 2).

Besides point locations, wider place-names are mentioned in the sources. Johannes

Andreae Cajanus gave the area of perceptibility as ‘‘most of the province of Pohjanmaa’’.

Combining this with the toponim Pomor’e in the broadest sense (Fig. 2) leaves a much

smaller gap between the key areas. The absence of observations between the two territories

can be explained by the lack of population and documentation centers there (cf. Fig. 3b and

Sect. 2.3). The Russian–Swedish border in the region of interest at that time was not

controlled, so fishermen and hunters could move freely from one territory to the other, but

permanent habitation hardly existed. Throughout centuries frequent conflicts in the border

area also rendered permanent settlement slow and difficult. Thus, the one-earthquake

scenario is feasible on the basis of the information available on places.

The physical feature determining the area of perceptibility is the propagation of seismic

waves. In a shield region, the attenuation is quite slow, which would account for the long

distances between data points. The largest earthquake in Kaliningrad on September 21,

2004 was of magnitude around Mw5.2, but was felt up to distances in excess of 600 km

(Gregersen et al. 2007). However, the attenuation may have preferred directions of azi-

muth. Because the sparse population concentrated along waterways, local site conditions

could also increase the sensation of ground shaking. For instance, Siyskyi Monastery is

144 Nat Hazards (2011) 57:133–150

123

situated in a marshy island in Lake Mikhailov (Fig. 5), where the ground conditions are

exceptionally bad. The fortress of Kajaani was built on an island in a river, and the church

of Paltamo was located on the highly unstable sandy banks of Lake Oulujarvi.

A few pieces of information are available for intensity assessment. According to the

chronicle Dvinskiy Letopisets, some people were not awakened by the earthquake in Dvina

south of the White Sea. On the European Macroseismic Scale (EMS; Grunthal 1998), many

people (20–50%) are awakened at intensity level 5. This may be taken as the upper limit

for intensity in Dvina, but the actual value can have been lower. In Records on Dvinaevents in 7061 (1553)–7134 (1626), the absence of details makes it impossible to differ-

entiate the intensity; only ‘‘felt’ can be assigned to the places mentioned. In the Lake

Oulujarvi area, the damaged church and landslide give clues to intensity assessment. The

smallest intensity that harms the built-up environment is 5 on the EMS, but the definition is

based on a statistical approach to building and damage types. Wooden structures on

average would not be ranked in the lowest vulnerability classes affected at this intensity

level. Neither the structural details nor condition of the church are known, but the site can

be classified as bad. Similarly, a landslide may result from unstable slopes or other bad

conditions. For example, the railway embankment near the town of Svetlogorsk slid as a

consequence of the Kaliningrad earthquakes of September 21, 2004. The soil was very wet

because of heavy rains. Intensity in Svetlogorsk town was assessed at 5–6 (MSK-64) and

up to 6 close to the town (Aptikaev et al. 2005). The actual ground motion at the site by

Lake Oulujarvi may not have needed be particularly strong, and the corresponding

intensity may have been around 4–5 (EMS).

The new solution is that the available documents refer to the same event. Tentative

earthquake parameters can be obtained with the help of different relationships derived for

the Baltic Shield. Wahlstrom and Ahjos (1984) derived a relationship for macroseismic

Fig. 5 Siyskiy Monastery. Map approximately contemporary with the earthquake

Nat Hazards (2011) 57:133–150 145

123

magnitude that is calculated on the basis of the radius of perceptibility and epicentral

intensity. Using a minimum radius of 400 km and a maximum value of 500 km and

epicentral intensities ranging from 6 to 7 (extrapolated from observed), macroseismic

magnitudes between 4.7 and 5.1 are obtained. In the Introduction to the New Catalog

(1977, 1982), N.V. Shebalin demonstrated that for the most of the data attenuation coef-

ficient m = 3.5 fits quite well, but for shield areas it can be as low as 3.0. Macroseismic

field equation with this value and epicentral intensity and radii of perceptibility mentioned

above gives a range of magnitudes from 4.8 to 5.7. In summary of Sect. 4 and the

discussion above, the proposed solution is as follows: date June (± 1 month) 3, 1626, time

3–5 a.m. local time, latitude 64.5�N ± 100 km, longitude 31.3�E ± 50 km, I0 = 6–7

(EMS), M = 4.7–5.7. Figure 6 shows a possible area of perceptibility for intensity I = 3

fitted to the available information and the intensity assessments above.

The epicenter is located in the Russian territory in the border area. It is obvious that it

cannot be pinpointed accurately. The available intensity data points (4–5 in Paltamo and 4

Fig. 6 A possible area of perceptibility for I = 3, epicenter area and tentative parameters for the scenarioof one earthquake in the spring of 1626. Error bars of 100 km in latitude and 50 km in longitude are plotted.Intensities are given on the European Macroseismic Scale (Grunthal 1998). Dashed pink marks areas whereground shaking was felt according to the contemporary documents

146 Nat Hazards (2011) 57:133–150

123

in Siyskiy monastery) are located west and east-southeast of the proposed epicenter. They

constrain the longitude in that it cannot be moved substantially without having to change

the intensities. There are even less constrains on the latitude. The proposed epicenter is

located in an area almost devoid of seismicity according to recent high-quality seismic

recordings (Fig. 1). If the epicenter is moved northward, it coincides with the pronounced

earthquake activity bounded by the Kandalaksha-Puolanka and Kuusamo shear zones.

However, it is also known that earthquakes have occurred as surprises in places with very

little observed seismicity; thus the site cannot be argued convincingly on the basis of the

modern seismicity pattern either.

The new solution implies that seismic hazard in the region has to be reassessed. The

magnitude of the earthquake is of relevance when evaluating the seismic potential of the

region whether the epicenter coincides with an enhanced seismicity trend or not. In

comparison with the scenario of two different earthquakes, the intensities in the Russian–

Finnish border area are higher and may reach damaging levels; the radius of isoseismal 6 is

equal to 45 km using the available attenuation relationship. The investigated documents

also serve as reminders of site effects in the event of ground shaking. The pattern of

damaging ground shaking may be irregular and extend outside the epicenter area.

Acknowledgments This research was partly supported by RFBR grants 08-05-00598 and 10-05-00126.Travel grants based on contractual bilateral cooperation between the Russian Academy of Sciences and theAcademy of Finland are gratefully acknowledged.

Appendix 1: The sources in the original languages

Yakovleva (1958):

B keno 7134-uo uoly vabz d 4 leym npzckacz pevkz do dcev Govopbb, ya Cokodrax b dEcnm-Roke b ya Ldbye go Cbqcroq voyacnspm.

Dvinskiy Letopisets (PSRL, v. 33, p. 172):1627 u. O cnpaiyov npyce. Gpb cev doedole ,sk ya Ldbye cnpaiyoq npyc. B 135 uoly

vabz 20 leym d yelek. Bcex cdznsx, r goyelekmybry d 5 xacy yoxb upex palb yaibx

uyedov ,o;bbv gonpzcecz pevkz, b vyopb on k.leq bcnpzceybe pevkb dbleia, a bybbbk.lbe d no dpevz cgakb. B on nouo npyca k.leq ,ou govbkodak.

Rhyme chronicle of Finland. Messenius J (2004) Suomen riimikronikka (Rhyme

Chronicle of Finland). Edited and translated into Finnish by H. Lonnroth and M. Linna.

Publication no. 913 of Finnish Literature Society SKS, Helsinki (in Swedish):

Har nast will iagh wnderligh tingh

Skrifua, Som ahr then iordbafningh,

hon gonom hela finland gick,

En wdd ellenast skada fick

J Soknen ijo, och then wanck,

Sa han hijsligen neder sanck,

whr diupa wapnet [watnet] synes ap

Ey mehr an the hoga tras tap,

Efter iordbafningh aldrigh gatt

Kommer, Sa ahr och om then spat:

J Samma Sokn ahret thar nast,

En dagh regnar thar blod mast,

Som war then nyttonde april,

Ey haller gatt thet bada will,

Nat Hazards (2011) 57:133–150 147

123

Gudh ware oss nadigh, och hull,

For Sin Sons Jesu Christi Skull!

Scondia illustrata. Messenius J (1988) Suomen, Liivinmaan ja Kuurinmaan vaiheita

seka tuntemattoman tekijan Suomen kronikka (Episodes of life in Finland, Livonia and

Kurland and the Chronicle of Finland by an unkown author). Translated from Latin by

M. Linna, J. Lagerstedt and E. Palmen, Finnish Literature Society SKS, Helsinki (in

Finnish):

[1626] Kesakuun 13. paivan valjetessa Suomea vavahdutti kauhea maanjaristys. Sen

johdosta Iin pitajassa maa aleni varsin huomattavalla alueella, mika teki mahdolliseksi

jarven syntymisen juuri talle paikalle.

Johannes Andreae Cajanus: Cajanus JA (1663) Anno 1663 d. I Febr. …holts den

aldraforsta Visitation uti Paldamo … (Description of the municipality of Paltamo), Tid-ningar utgifne af ett Sallskap i Abo 1777 no. 16-18 (in Swedish):

Denne kyrckan (…); men i den stora jordbafningen, som ofwer storsta delen af

Osterbotn skedde 1626, pa Eskilli dag, fick hon en stor skada, at ostra gafwelen blef

alldeles lutande; och ifran den tiden hafwa Bonderna inte welat kosta pa kyrckan, hwarfore

kyckan nu synes falla omkull.

References

Manuscripts

Scondia illustrata, 10th book, manuscript by Johannes Messenius, National Library of Sweden, StockholmNew Finnish Manual (1693) Uusi suomalainen Manuale, sisallans pitawa monda caunista P. Raamatun

kirjaa, ynna evangeliumein ja epistolain, nijn myos taydellisen rucous-kirjan ja … wirsi-kirjan … Nijnetta tasa loytan 25 kirja eli cappaletta, Keyser, Stockholm. National Library of Sweden, Stockholm

Printed literature

Ahjos T, Uski M (1992) Earthquakes in northern Europe in 1375–1989. Tectonophysics 207:1–23Aptikaev FF, Nikonov AA, Alyoshin AS, Assinovskaya BA, Pogrebchenko VV, Erteleva OO (2005)

Kaliningrad earthquake of September 21, 2004, damage, International Conference on EarthquakeEngineering ‘‘EE-21C’’ to mark 40 years of IZIIS–Skopje. 27 August–1 September 2005. Skopje–Ohrid,CD, Topic 1, pp 1–18

Cajanus JA (1663) Anno 1663 d. I Febr. …holts den aldraforsta Visitation uti Paldamo … (Description ofthe municipality of Paltamo), Tidningar utgifne af ett Sallskap i Abo 1777 no. 16–18 (in Swedish)

Golubtsov NA (1911) K istorii goroda Koly Arkhangel’skoy gubernii. (On history of Kola Town inArkhangel’sk province)//Izvestiya Arkhangel’skogo obshchestva izucheniya Russkogo severa, 1, 7–16;No 5, 392–401 (in Russian)

Gregersen S, Wiejacz P, Debski W, Domanski B, Assinovskaya B, Guterch B, Mantyniemi P, Nikulin VG,Pacesa A, Puura V, Aronov AG, Aronova TI, Grunthal G, Husebye ES, Sliaupa S (2007) Theexceptional earthquakes in Kaliningrad district, Russia on September 21, 2004. Phys Earth Planet Inter64:63–74

Grunthal G (Ed.) (1998) European Macroseismic Scale 1998, Cahiers du Centre Europeen de Geodyna-mique et de Seismologie 15, Luxembourg

Guidoboni E, Ebel JE (2009) Earthquakes and tsunamis in the past: A guide to techniques in historicalseismology, Cambridge University Press, United Kingdom

Jokipii M (1960) Suomen kreivi- ja vapaaherrakunnat II (Earldoms and baronies in Finland II), Historicalinvestigations XLVIII: 2, The Finnish Historical Society, Helsinki, 352 pp. (in Finnish)

148 Nat Hazards (2011) 57:133–150

123

Keranen J (1984) Kainuun asuttaminen (The settlement of Kainuu), Diss. Studia historica Jyvaskylaensia 28,University of Jyvaskyla, Jyvaskyla, 282 pp. (in Finnish)

Keranen J (1986) Uudisraivauksen ja rajasotien kausi (The period of land settlement and border conflicts),In: The history of Kainuu 1 by Huurre, M., and J Keranen, Publisher Kainuun maakuntaliitto, Kajaani,203–696 (in Finnish)

Kholmogorskaya letopis’. Dvinskiy letopisets (Kholmogor Chronicle. Dvinskiy Chronicle (1977) In: Pol-noye sobraniye russkikh letopisey, 33, Leningrad, Nauka, 250 pp. (in Russian)

Koistinen T, Stephens MB, Bogatchev V, Nordgulen Ø, Wennerstrom M, Korhonen J (2001) Geologicalmap of the Fennoscandian Shield, 1:2000,000, Geol. Surv. Finland, Espoo; Geol. Surv. Norway,Trondheim; Geol. Surv. Sweden, Uppsala; Ministry of Natural Resources of Russia, Moscow

Leiviska I (1914) Oulujarvesta (On Lake Oulujarvi). Terra 26:19–35 (in Finnish)Mathesius PN (1734) Dissertatio geographica de Ostrobotnia, Upsaliæ, literis Wernerianis. Alternativt

namn: Gronwall, Andreæ, 1671–1758Messenius J (1774) Berattelse om nagra gamla och markvardiga Finlands handlingar, hwilken innehaller en

kronika om inbyggarens harkomst, bedrifter, gudtjenst, konungar, regenter och biskopar / med swenskarim beskrefwen. Publisher Johan Christoph, Frenckell Abo (in Swedish)

Messenius J (1865) Joh. Messenii Rimkronika om Finland och dess Inbyggare (The Rhyme Chronicle ofFinland and its inhabitants by Joh. Messenius). Publisher G.W. Enlunds, Helsingfors (in Swedish)

Messenius J (1988) Suomen, Liivinmaan ja Kuurinmaan vaiheita seka tuntemattoman tekijan Suomenkronikka (Episodes of life in Finland, Livonia and Kurland and the Chronicle of Finland by an unkownauthor) (trans from Latin: Linna M, Lagerstedt J, Palmen E). Finnish Literature Society SKS, Helsinki(in Finnish)

Messenius J (2004) Suomen riimikronikka (Rhyme Chronicle of Finland), Edited and translated into Finnishby Lonnroth, H., and M. Linna, Publication no. 913 of Finnish Literature Society SKS, Helsinki(in Swedish and Finnish)

Moberg KA (1894) Uppgifter om jordskalfven i Finland fore ar 1882 (Information about earthquakes inFinland before 1882). Fennia 9(5), 26 pp (in Swedish)

Musson RMW (1996) Determination of parameters for historical British earthquakes. Ann Geophys39:1041–1047

New Catalog of Strong Earthquakes in the Territory of the Soviet Union from Ancient Times till 1975(1977) Kondorskaya NV, Shebalin NV (Eds), Nauka, Moscow, 535 pp. English edition published in1982, Boulder, USA, 608 pp

Nikonov AA (2004) Istoricheskiye zemletryaseniya (Historical earthquakes). In: Sharov NV (ed) Deepstructure and seismicity of the Karelian region and its margins. Karelian Research Center, Petrozav-odsk, pp 192–213 (in Russian)

Panasenko GD (1969) Seysmicheskiye osobennosti severo-vostoka Baltiyskogo shchita (Seismic peculiar-ities of the Northeastern Baltic Shield). Nauka, Leningrad, p 184 (in Russian)

Pantosti D, Yeats RS (1993) Paleoseismology of great earthquakes of the late Holocene. Ann Geophys36:237–257

Penttila E (1978) Earthquakes in Finland 1610–1976, Inst. Seismology, Univ. Helsinki, Report S-1, 15 ppRenqvist H (1930) Finlands jordskalv (Earthquakes in Finland). Fennia, 54, 113 pp. (in Swedish)Shaskol’skiy IP (1962) O vozniknovenii goroda Koly (On nascence of Kola town)//Istoricheskiye zapiski,

vol. 71 (in Russian)Slavyanskaya entsiklopediya. XVII vek (2004) (Slavic enciclopedia. XVII century), compiled by V.V.

Boguslavskiy), OLMA-PRESS, Moscow, 784p. (in Russian)Stepanov VV, Pribylova NE (2003) Analiz otsenok zemletryaseniya 1626 g. Kol’skogo poluostrova

(Analysis of evaluations of Kola peninsula earthquake in 1626), In: Stroyeniye, zhivaya tektonika idislokatsii platform i ikh gorno-skladchatykh obramleniy. Materialy mezdunarodnoy konferentsii,Moskva, pp 241–244 (in Russian)

Tatevossian RE, Arefiev SS, Mokrushina NG, Petrossian AE (2003) Istoricheskaya seysmichnost’ Zagrosa,Iran (Historical seismicity of Zagros, Iran). Vychislitel’naya seismologiya 34:254–285 (in Russian)

Tatishchev VN (2003) Istoriya Rossiyskaya (Russian history), Moskva, Izd-vo ACT, 3 volumes (in Russian)Thuneld E (1794) Geographie ofver Konungariket Sverige samt derunder horande lander (Geography of the

kingdom of Sweden and countries belonging to it), 3rd edn, Stockholm (in Swedish)Ustyuzhskiye i Vologodskiye letopisi XVI-XVIII vv. (Ustyug and Vologda Chronicle) (1982), In: Polnoye

sobraniye russkikh letopisey, 37, Leningrad, Nauka, 228 pp. (in Russian)Vahtola J (1998) Puhdasoppisuuden aika (1596–1721) (The era of orthodoxy (1596–1721)). In: Elo K,

Satokangas R, Vahtola J (eds) Iin seurakunnan historia (History of the parish of Ii). Gummerus,Jyvaskyla, pp 50–95 (in Finnish)

Nat Hazards (2011) 57:133–150 149

123

Vologodsko-Permskaya letopis’ (Vologoda-Perm’ Chronicle) (1959), In: Polnoye sobraniye russkikh le-topisey, 26 Moskva-Leningrad, Izdaniye AN SSSR, 415 pp. (in Russian)

Wahlstrom R, Ahjos T (1984) Magnitude determination of earthquakes in the Baltic Shield. Ann Geophys2:553–558

Yakovleva OA (1958) Letpisnoye izvestiye o zemletryasenii na severe Mosckovskogo gosudarstva 1626 g.(Chronicle information about the earthquake in the North of the State of Moscow in 1626). Izv., ANSSSR, ser. Geofizika, n. 3, p 423 (in Russian)

150 Nat Hazards (2011) 57:133–150

123