Climatic Variations on the East-coast of the
Atlantic At the end of the Middle Ages.

In northern Europe the variations in the oceanic circulation chiefly manifested themselves in:

A: Storm floods in the North Sea and the Baltic.

B: The freezing of the Baltic and the Cattegat in certain winters.

C: The removal of herring fishery from the Baltic and Öresund to the northern parts of the Cattegat, Skagerak and the North Sea.

These variations naturally exerted an influence upon the climate. According to the hypothesis enounced in the foregoing chapter these phenomena were caused by variations in the tide generating force which will repeat themselves in future as certainly as they have occurred in the past.



In the last centuries of the Middle age Holland passed through a stage of transformation caused partly by the fact that its coast is slowly subsiding. In addition to this, inundations from the rivers carried away the soil of its peat-moors. Then came the attack on the sea-coast by storm floods. Several such floods are on record from the early Middle age but the greatest devastations set in from the latter part of the 12th century, followed in rapid succession by the great floods of the 13th and 14th century which is denoted as “The catastrophic age” of Holland. It culminated with the destruction of South Holland, “the Waard”, beginning in the 15th century, i. e. the time of the perihelion-node-apside and the absolute maximum of the tidal force.

Both before and after that time calamities of this kind have occurred more rarely (1). From the 15th century to our time only the great storm floods are on record viz. in 1570, 1717, 1825, (2). The principal disasters of the catastrophe were:

1219: (the “Marcellus”) (3)

1246: (the “Lucas”

1287: (the “Lucia”

1241: (the”Elizabeth”. (4)

Which seems to have surpassed all others in its disastrous effect.

The catastrophe occurred on the 18 Nov. 1421 and devastated towns of Dodrecht and Geertruidenberg. The storm floods between 1374 and 1421 mark the culmen of the catastrophic period for South-Holland (5). Also from England we have records of heavy storm floods in 1404 and 1412. The latter occurred the 12th October and is described in the following curious manner: “Sea flooded thrice without ebbing”, which brings back to our memory the remarkable description in the Kings Mirror of the great “havgerdingar” which were said to have occurred in the Greenland seas in the 13th century but never have been observed or heard of since.(6)

In these centuries also was achieved the last stage of the transformation of the old Friesian lake-district into the present Zuidersee. It is impossible to fix with surety the exact time of the catastrophes which mark the different stages of the incessant encroaching action of the sea upon the Ijssel, the ancient mouth of the Rhine, and the whole subject is very much debated. According to a number of old chronicles it happened so. In the time of the Romans the interior part of the present Zuidersee formed a shallow lake, Lacus Flevo, the shores of which were widened by the erosion of the rivers. Then came the attack on the sea-shores by storm floods. The Marsdiep by Helder was formed finally in the 15th century. In 1170 the isthmus between North Holland and Friesland was broken through. In 1250 the sea reached Enkhuitzen engulfing the intervening land where now lies the island of Wieringen. On the 4th of December 1287 the great catastrophe occurred which Hennig thus describes: “extraordinary heavy storm flood in the North Sea causing the loss of more than 80,000 human lives. The sea broke into the s.c. Lacus Flevo forming the present Zuidersee. Simultaneous storm floods occurred in Norfolk and Suffolk” (7).

In support of this version is adduced the fact that there existed in the 10th century (960-985) the flourishing towns Stavoren, on the shore of the “Flie”. And Medemblik. It is also noticeable that the name Zuidersee first appears in documents from the year 1340. Before that time the name “Aelmere” had substituted the ancient denomination of “Lacus Flevo”. According to a private communication from professor van Everdingen in Bilt, it seems certain that the part of the Zuidersee south of Enkhuizen was open to navigation from 1395. The final stage of the formation of the Zuidersee probably fell in the latter time of the catastrophic age (8), viz. the 14th century. On account of the innumerable sequences of storm floods which then devastated the Hollandish coast it seems difficult to judge with certainty the much debated question of the veracity (9) of the old chroniclers.

In the 14th century the coasts of Friesland and Schleswig were exposed to destruction. On this subject much has been written lately and here I give a quotation from the last publication on the subject.

  1. As I cannot enter upon the literature concerning these catastrophes which is extremely vast, I prefer to follow the latest treatise published in Sweden on this subject by Dr. A. Norlind: “The geographische Entwiklung des Rheindeltas”, (Lund and Amsterdam 1912).
  2. A. Nordlund I., c. p. 235, 236.
  3. This was followed by other floods in 1248, 1249, 1257 (the Gereonflood”) and 1267 (A. Norlind l. c. p. 249)
  4. A. Norlinds description of this storm flood fills the pages 196 to 204 in his paper: Die geographische Entwicklung des Rheindeltas.
  5. “Die letzte Hälfte des 15tem Jarhunderts war von unheilverkündenden Vorzeichen für den südhollandischen Waard sehr reich gewesen. Es wurde in den Jahren 1374 (zweimal) 1376, 1377, 1393, 1396, und 1397 überschwemmt. Diese Uberschwemmungen waren teils vom Flusse teils vom Meere gakomen. Das schwerste Gefahr drechts aber von Seitum des Maeres. Und sie kam in deutlicher und eindringlicher Gestalt was allen die nicht blind waren offenbar warden musste."A. Norlin l. c. p. 194.
  6. Est ad hoc in mari Groenladiae mirandum quod cingulorum maris nomine venit… oriuntur tres decumani fluctus qui totum mare cingunt nullibi pervii, montibus altiores praesuptisque montium jugis admodum similes; raraque prestant exempla eorum qui salvi haec pericula evaserint!. (“Speculum Regal”).
  7. In Walford's essay – on the Famines of the world (Statistical Journal Sept. 1878, London) we read the following passages: “Holland: 1283-87, a dreadful storm laid the whole country on both sides of the Zuidersee under water. To such height did the water rise that Count Florence took advantage of the circumstance to subdue the inland towns by armed vessels called “cogs” (Davis' Holland). England: 1287. Winter excessively rainy, great floods 1 Jun. Sea broke in from the Humber to Yarmouth forced by the winds. In December on Suffolk and Norfolk coasts. Plague all the year.
  8. A. Norlind L. e Dr. Norlind who has kindly superseded the preceding description of storm floods is of the opinion that the climax of the catastrophic epoch fell in the 13-14 th century.
  9. On the other hand we ought to be on guard against the predisposition common to geographers and meteorologists of the present time to believe and minimize all historic accounts of phenomena of catastrophic character which seem to them incompatible with every day experience.

“Wahrscheinlich bildeten dies Friesischen Inseln in alten Zeiten eine zusammenhängende Dünenkette nur durchflossen von der grösseren Flüssen. Der ganzen Bodengestaltung bei den freisischen Inseln und im ausgedehnten Wassergebiet ist nach Krümmel erst durch die Gezeitenströmung der charakteristische Typ aufgedrückt worden. (Perterm. Mitt. 1889 s. 129 ff). Die säkulare Senkung von der die Dargschikten unter der Marsch und die submarinen Wälder und Moore Zeugniss ablegen, so wie der, nach Browne and der Schwelle der historischen Zeit, nach Walther vor 6,000 bis 8,000 Jahren, erfolgte Durchbruch des anglo-französischen Isthmus, ferner der stärkere Flutwechsel der ja erst nach diesem Ereigniss besonders in die Erscheinung tritt, machten ihre verheerenden Einwirkungen auf die Dünenkette geltend."


The freezing in certain winters in the Baltic and Cattegat
and the growth of the Hanseatic herringsfishery
during the Middle ages.

Both of these phenomena are important when studying the cause of climatic variations in historic time. In the first place it is necessary to prove their reality by carefully sifting and comparing available data which are numerous but often contradicting as to the time and extent of freezing a.s.f. We have come to look upon climatic variations as quite irregular and incomprehensible and because of this there is now a tendency to discredit the statements of the chronicles regarding the severe winters and occasional abnormally hot summers of the 13th and 14th centuries, the shifting in the climatic seasons, the variations between drought and extreme precipitation, a.s.f. The old writers are supposed to have exaggerated, because in their time humanity was less able to protect itself against the influence of such variations and it is assumed that, as navigation was generally suspended during winter, the ice-conditions of the sea were judged from what could be seen from shore. As to the abundant herring fishery of the Öresund, it is argued that in certain years even now a tolerably abundant fishery is carried on off the coast of Scania and that the astounding records of the old Hanseatic fishery are due to an enormous exaggeration of the ancient returns of the yield.

In my opinion, the state of the Baltic during the Middle ages, when known, will furnish the key to the mystery of the climatic variations. The statements relating to this subject are so numerous and distributed over so many ancient chronicles and later summaries of chronicles that the collection and sifting of this material will require a special study.

In France the interest for historic climatology was awakened by Arago, in Sweden by Ehrenheim in his celebrated presidential address to the Academy of Sciences of Stockholm in 1824. The scientific scrutiny of the Swedish material was begun by R. Rubenson and continued by N. Ekholm,(1) who in the 5th chapter of his paper “Öm klimatets ändrigar och dess orsaker” gives the most complete summary of the subject known hitherto. The name of the chapter is: “Kilmatväxslingar under historisk tid, särskildt I nordvästra Europa.” Ekholm commences his discourse thus:

“It remains to consider the climate variations in historic time. The observation material is certainly very comprehensive but there is such a want of order and regularity in it that it seems almost impossible at present to get a clearer view of the different kinds of phenomena and the connection between them.”

1. Quarterly Journ. Of the R. Metorol. Soc. 1901 January

The material from which such deductions can be drawn consists, according to Ekholm, partly of historic records, numerous but unreliable and irregular and partly in actual observations of the weather which however do not date further back than 150 years. Our knowledge of climate variations during the Middle ages must thus be based chiefly on historic records. As their authenticity has been questioned, because in some cases the writers have taken down their accounts from hearsay or copied one another, I have tried to collect new documents and concentrate my research upon one problem, viz., the state of the Baltic. For, if it is possible to show that the records regarding the freezing of this sea and the herring fishery therein are based on actual facts, it is absolutely certain that hydro graphic changes have occurred in the circulation of the waters of our seas and we have obtained a basis for our endeavour to discover their causes. In the Icelandic, the Swedish and particularly in the Danish historic literature I have found material wherewith to compare and criticize these records. This chapter contains a collection of such statements which I have obtained partly directly , partly from information given by Swedish authorities on Mediaeval historic literature.

Captain C. J. Speerschneider of the Danish Met. Inst. has collected a great deal of information regarding the ice years in the Baltic. I have been able to compare my historic data with those of Captain S. and I have found that several of the old records have been wrongly copied and suffer from confusion of geographical names. Captain S. has particularly pointed out that the statements regarding the freezing of the Skagerak in reality concern the Cattegat, as in old times no proper distinction was made between the two seas. Thus when the Icelandic records state that in one winter packs of wolves went across the ice from Norway to Denmark, or when the Diarium Minoritorum Wisbyensium relates of the year 1296: “congelatum est mare tanto rigore ut eqvitari poterat de Opslo ad Jutiam”, it probably does not mean that communication was kept upon the ice directly between Christiania (Oslo) and Skagen, but probably between some place in the coast of Bohuslän and Jutland. But even in the freezing of the North Sea or the Skagerak should not be taken literally, that of the Baltic, occurred so frequently in Mediaeval time, is a fact that cannot be doubted or explained otherwise. As the large material collected by Captain Speerschneider is soon to be published I will here only consider a few years, which may be regarded as typical for a mediaeval ice-winter.

1306- The freezing of the Baltic during the winter 1306-1307 is a well known historic fact which cannot be doubted. So severe a winter had not occurred for many years. The entire Baltic was frozen over from Livonia and the Gulf of Finland to the Sound. There the ice is said to have measured 15 ells in thickness and for 14 weeks to have formed a solid bridge between Sweden and the Danish islands. The war between Sven Estridsson and the Dukes was stopped by the cold.

In Diarium Minoritorum Visbyensium we find: “fuit hiemps maximus ita quod mare inter Olandiam et Estoniam exstitit congelatum.”

In the Icelandic records (Annales regii) and in the Skalholt Annals a. v. I have found the following description of the ice winter 1306-1307, which was felt in Icelandic seas as well as in the Baltic.

1306. Haffis fyrir Nordanland um allt sumar 15 alna harr. Frost sva mikit I pyversku landi ok I Franz sem pa er stera lagi eru Noregi ok gengv af Ravdstock til Danmerkr yfir Eyrassvnnd.”

From 1322-1323 we have the greatest number of notes regarding a mediaeval ice-winter in the Baltic. Captain Speerschneider has permitted me to quote from his collection of historic meteorological data, drawn from Danish and German records, everything that concerns this particular winter.

“We are told in the history of Olaus Magnum, that the frost was so hard that it became possible to cross the ice from the coast of Lybeck to Denamrk, and the hostelries were put up on the ice.”

Both Hennig (Katalog bemerkenswerther Wittersungsereignisse, Berlin 1904) and Mansa (Folkesygdomme og Sundhetspleiens Historie i Danmark, Köpenhamn 1873) relate that the winter began on the 30 Nov. 1322 and lasted to the half of Lent 1323 had passed. Peter Olsen says that the cold culminated in March. Hvidfeldt (Scriptores rer. Dan. II: 528) tells us that there was general traffic across the frozen Belt from Candle mass 2/2 for 6 weeks and that the Drost, Herr Laurits rode across from Taarborg with the troops. Pfaff (Über den strengen Wintern etc. Kiel 1809) speaks of the severe winter and that it is possible to ride and drive from Germany to Denmark and from Lübeck to Danzic over the Baltic, so late as 28/2 Arago (Aeuvres completes) repeats after him that one could ride from Denmark to Danztzic.

Mansa says that the entire Baltic was frozen between Denmark, Venden and Fehmern. Strelow (Chronica Guthilandorum Köbhvn, 1833) says that one could drive across de ice between Sweden and Gulland.


A. De frigoribus autem illius Germaniae seu Gothicis maris plura meminit Albertus Crantzius, diligentissimus omnium regionum scriptor. Ait enim: anno 1323 gelidissimo frigore constringebatur mare ut pedestri itinere per glaciem de littore Lubicensi in Daniam & in Prussiam mare transiretur dispositis pero loca oportuna in glacie hospiciis (Olaus Mangus' historie 1555).

B. Erat autem 1323 qvum gelidissimo frigore mare constringeretur ut pedestri per glaciem itinere de littore nostro in Daniam inqve Prussiam mare transiretur dispositis per oportuna loca in glacie hospitiis, si quid commentibus intervenisset (Albertus Krantzius Francof. 1580).

C. Via communis erat ambulantibus et eqvitantibus in multidine super glaciem per passagium mari Baltici eundo et reeundo qvasi per continuum sex septimanarum spatium circa purificationis Maria virginis similiter per passagia inter Seelandiam et proximos partes Slaviae (Appenix ad incerti annals Danarum).

D. Between Norway, England and the Flanders (from the harbour Leven in the latter country) the ships were frozen in the open sea so that the merchants went visiting one another over the ice. When the ice broke up most of the ships frozen in the south were saved but few remained of those frozen in the north. (Suhm a St. II.61).

Captain Speerschneider summarizes his impressions of the ice winter 1322-1323 in this manner.

“The winter was severe and began early. From the 1st of Feb. to the middle of March the ice lay in the Cattegat, the Belts and the western Baltic. The ice was used for communication across the great Belt and the Fehmerbelt, probably also the Sound. There must at least have been ice along the whole of the Baltic coast of Germany extending far out into the sea and also between Sweden and Gulland.”

“Also in the southern countries the winter must have been severe. Hennig says, that the Gulf of Venice was frozen over; Arago says that the winter was severe in France and Italy.” This shows that the cold culminated between 1 Febr. and 15 March “as it does now.” 1394 “I was possible to walk across from the coast of Venden to Denmark and Sweden”. (Captain Speerscheniders notes).

“one of the coldest winters on record. The sea between Sweden and Denmark was frozen over.” (Captain Speerschneiders notes).

Hennig's Katalog has the following note on the winter 1407-1408:

“Der grosse winter von 11 Nov. Bis 27 Jan; in England v. 11 Dezember bis 2 Ja. Alle Flüsse tragen die schwestern Lastwagen, über den gefrorenen Skagerak (Kattegat?) laufen die Wölfe von Norwegen nach Jutland. Seit 3 Jan. eranlasst plötslich starkes Tauwetter in England grosse Überschwemmungen hervorgerufen wurden.

According to Stavenow, Chronologia vetusta 1298-1473 contains this statement: "Anno MCC-CCXVIII. Hiemps erat nimis aspera et mare salsum fuit congelatum ita qvod homines transierunt intra Alemanniam et Daniam videlicet Gezör et Rotstock.”

According to Stavenow the following statement is contained in an anonymous record called “Remarkable occurrences in Sweden 1220-1552”. “In 1423 the winter was so severe that the Baltic was covered with ice so that it was possible to travel in sleighs along the route of vessels from Dantzic to Lübeck and from Pomerania to Denmark.”

From Hennig's Katalog: “Ausseordentlich strenger winter bis 20 Marz. Die Ostsee friert vollständig zu so dass mann zu Fuss über das Eis von Lübeck und Stralsund nach Norwegen (Dänemark?) von Reval nach Schweden gehen kann. Die Donau und andere Flüsse von 13 Jan. bis 11 Märtz derart zugefroren dass sie die schwersten Wagen tragen.“

From the 16th century also there are records of the Baltic being frozen in such a degree that the ice could be used for communication between Germany and Denmark. But this happened more and more seldom and the winter of 1636 is considered to be the last in which the Baltic was frozen between Scania and Bornholm. In the above quotations I have only made use of such notices which, as far as I can find, have never been published or compared before. Hennig's Katalog contains several notices to the same effect, f. inst. regarding the ice winter 1318-1319, like this: “Strenger winter, alle grösse Flüsse auch der Po zugefroren n 20 Tage lang derart dass sie Wagen trugen”, etc. I do not however intend to enumerate and examine all notices on ice winters which occurred at that time. The subject is of great interest and deserves a special study, which will be undertaken sooner or later. I only wish to prove the reality of the phenomenon and to obtain a general view of the time in which it occurred. To sum up, I find that the freezing of the Baltic occurred most frequently during the last centuries of the Middle ages. The period of cold winters appears to have commenced about 1200 and to have culminated in the 14th century. The year 1322-1323 may be taken as typical.

After heavy floods of the Rhine, the Rhone a. o. rivers in June and later months 1322 the cold set in at the end of November. As a rule these cold winters were followed in spring by devastating inundations of the great rivers. No less than 55 of such occurrences are mentioned by the chronicles of the 13th century. Some other years, like 1304 and 1328, however, had exceedingly mild winters and hot and dry summers. Such were the years 1387, 1393, a. o. On the whole this century was noted for its extreme climatic variations on the eastern shore of the Atlantic. As I have already shown in this paper, corresponding phenomena occurred in Iceland and on the coast of Greenland. Later on I will show the influence which these variations exercised on the harvest yield and the economic conditions of Scandinavia.

The freezing of the Baltic in severe winters appears to have been most frequent in the last centuries of the Middle ages. Then gradually it occurred more rarely. In the last 3 centuries the Baltic ahs never frozen to. The intervals between the ice-winters gradually lengthened till at least they ceased altogether. Yet the recurrence of the old severe winters has been felt up to our time, although mitigated, in the freezing of the water along the shores of the Baltic, the crowing of ice in Öresund, driftice appearing in the Cattegat, etc. Whether the freezing of the Baltic occurred before, f. inst. during the Viking-age or the time of 800-900 a.C. it is not possible to tell, but it seems improbable that so could be the case as the Sagas never mention it. Nor do the Icelandic records speak of ice having hindered navigation in the western Atlantic.

It should be observed that the culmination of the Baltic ice winters occurred simultaneously with the period of great storm floods and the absolute maximum in the tide generating force of the sun and moon at the beginning of the 14th century. Besides this there is another phenomenon, viz. the annual immigration of herrings into the Baltic and Öresund, the s. c. Hanseatic herring fishery, which culminated 1100-1500 and afterwards declined.

Summing up all that the chronicles tell us of the ice winters of that time we find:

That there was a period 6-7 centuries ago when the Baltic, the Sounds and the Cattegat were frozen over and covered by a solid sheet of ice which could be frequented by pedestrians and carriages in certain winters. This happened most frequently in the 13th, 14th, and 15th centuries but ceased in the 16th century. For the last 250 years the Baltic has not been frozen over.

It must not be concluded, however, that the average winter climate for the last centuries has been less severe than during the Middle ages. It is possible and even probable that it is so; but this “climatic improvement” which would mean a transition from a continental to a more maritime climate for the countries in the Baltic must be proved in other ways to become a recognized meteorological fact. For the freezing of the Baltic depends not, as will be shown, on meteorologic but hydrographical causes, though the phenomenon may very well have influenced the climate.

The waters of the Baltic as well as of the Cattegat and Skagerak are stratified. The surface layer in the Middle and South Baltic has a comparatively even and uniform salinity varying from 7‰ x to 9‰. The freezing point is about -0.35ºC and the maximum of density more than 1º above 0. Every particle of water, which is cooled by contact with the atmosphere in winter, sinks down to the lower border of the water layer while other particles, warmer and less heavy take its place at the surface and are cooled in their turn. In this manner the whole bulk of water is cooled by convection down to 1º or 2º above 0 during authumn and winter. Because of the thickness of the layer (40-50 m in the Baltic proper) and the rapid interchange of water in the horizontal direction in the western Baltic and the Belts the surface layer has not time to cool to its freezing point in winter. In Febr. and March the cooling reaches its maximum a little above +1º in sever and little above +2º in mild winters.

Because of the thickness of the surface layer it is impossible at present that the Belts or the southern Baltic should be covered by a continuous ice sheet. For this to happen (except of course near the shore and in shallow bights) it is necessary that the surface layer gets thinner and the bottom layer is swelled by a stronger influx through the Belts and Sound, thus raising its level. Under such conditions the Baltic would freeze as easily as would a shallow lake or one of the fjords of the Cattegat, where the thin layer of fresh water is superposed over a deep layer of salt and warm oceanic water. Whereas these fjords, f. inst. the Gullmar and the inner Christiana fjord, freeze every winter the Baltic has kept open now for 3 centuries even in very cold winters.

Knowing, as we do, that 600 years ago the Baltic froze frequently so that the ice could be used for communication between Denmark and Germany and even for so vast a distance as that between Sweden, Gothland and Esthland, we are forced to conclude that the oceanic current of salt water entering through the Belts and the Sund must have been more powerful at that time and able to raise the level of the salt water which filled the depths of the Baltic basins to a greater height than now. Under such conditions the surface of the Baltic would freeze in cold winters. A deepening by some meters of the channels in Öresund would do the same and the explanation nearest at hand would be that the Öresund and the Baltic were deeper during mediaeval time and gave wider access to the undercurrent.

I have carefully examined this alternative by taking the opinion of the geologists who have made a special study of the level of the Baltic, (Munthe, Sernander and de Geer) and by examining carefully the oldest sea charts existent of Öresund which are kept of the Pilot Office of the Royal Danish Marine. There is none of these older than the 17th century and the sounding show on the whole the same depths as our modern charts between Scania and Sjaelland.

The changes in the Baltic from the mediaeval to modern time cannot be attributed to geologic causes. They must be caused by the altered hydrographical conditions and the only possible explanation is that the salt undercurrent was more intense during the Middle ages than now, so that the bottom layer of the Baltic then attained a higher level and the surface layer of the Baltic then attained a higher level and the surface layer was thinner, which caused the latter to freeze in cold winters. There is no other explanation possible.

What then was the state of the Baltic at that time when it happened in cold winters that its surface became frozen over? The Swedish research have furnished an answer to this question. Fig. 15 gives the position of the isohalines in the western Baltic as found by F.L. Ekman in August 1877. We see that the mixed salt water, which is found in the Cattegat and the Belts, immediately upon entering the Baltic flows along its bottom to the big basin east of Bornholm. Nowadays the undercurrent cannot fill this basin to a higher level than some 40-50 meters from the surface north and east of Bornholm and 60-70 meters below the surface in the big basin east of Gothland.

Fig. 15

If the undercurrent increased as to raise the level of the bottom layer some 9-10 meters in the neighbourhood of Bornholm, the Baltic would freeze as it did in the13th and 14th century. The following table contains the results of 3 soundings at the station Southeast of Bornholm where the water is 95 meters deep and the surface layer of uniform salinity is 50 meters in thickness.

15 august 19045 november 1904 15 march 1905
55º21' Lat. N.     15º39' Long E55º20' Lat. N.     15º30' Long E.55º23' N.     15º30' Long E.
Depth Temperature Salinity Temperature Salinity Temperature Salinity
0 m 15º.o9 C 7.29%v 10º.30 Cv 7.29% 2º.2 C 7.39%
10 m 14º.61 C 7.25% 10º.23 C 7.27% 2º.12 C 7.39%
20 m 14º.60 C 7.25% 10º.23 C 7.28% 2º.08 C 7.43%
30 m 13º.40 C 7.25% 10º.23 C 7.29% 2º.15 C 7.56%
40 m 4º.70 C 7.32% 10º.23 C 7.29% 2º.20 C 7.67%v
50 m 3º.52 Cv 10.21% 7º.02 C 9.42% 3º.40 C 9.15%
- - - - - - -
75 m 3º.64 C 16.04% 4º.25 C 16.06% 5º.33 C 15.61%
95 m 3º.52 C 17.02% 3º.70 C 16.58% 4º.59 C 16.03%

It is easy to calculate from the tables that the surface layer has given off some 30.000 calories to the atmosphere from August to November. From November to the middle of March the surface of the sea had lost 367,500 calories per square meter. The temperature of the surface layer had in the meantime sunk from an average of 9º.71 C to 2º.36 C or with 7º.35 degrees. Had the surface layer been some 35.7 meter thick instead of 50 meter and given off the same amount of heat to the atmosphere, then its temperature would have sunk to the freezing point = -0.5º C and the open Baltic east of Bornholm would have been covered by ice in March 1905. The same calculation can be made for the Baltic north of Bornholm wherefrom we possess soundings from Nov. 1914 and March 1905. The Swedish station S5 is situated in the strait between Scania and Bornhom.

6 november 190415 march 1905
55º 26' Lat. N.       14º 46' Long. E. 55º 26' Lat. N.       14º 46' Long. E.
Depth Temperature Salinity Temperature Salinity
0 m 10º.6 C 7.54 ‰ 2º.30 C 7.50 ‰
10 m 10º.59 C 7.54 ‰ 2º.31 C 7.56 ‰
20 m 10º 59 C 7.52 ‰ 2º.26 C 7.65 ‰v
30 m 10º.63 C 7.67 ‰ 2º.50 C 7.80 ‰
40 m 11º.81 C 11.56 ‰ 1º.91 C 8º.04 ‰
- - - - -
60 m 11º.25 C 14.52 ‰ 1º.75 C 13.14 ‰
- - - - -
68 m 8º.87 C 15.61 ‰ 1º.71 C 13.87 ‰

Here the surface layer, which measures 40 m, had emitted some 340,000 calories per square meter from Nov. 6th 1904 to March 15th 1905. The temperature had sunk from 10.6º C to 2.16º C. If the surface layer had been 10-12 meter less thick the temperature would have sunk somewhat under 0º by the same heat emanation. The surface layer, however, is cooled farther in some winters. In 1896 its temperature sank to 1.3ºC. If in such a winter the bottom water had risen 8-9 meters nearer to the surface a continuous sheet of ice would have covered the middle part of the Baltic proper. If the influx of salt water into the Baltic were so strong that its level reached up to 30 instead of 40-45 meters from the surface of the strait of Bornholm the ice would have been safe for communication between Scania and Bornholm in the winter 1896-1897. But in the mild winter 1904-1905 the sea would have been open and free from ice.

In this way we may form an opinion as to how much the Baltic has altered in the course of the last 5 to 6 centuries. Fig. 15 is the outcome of such a calculation and shows a section of the western and Southern Baltic reconstructed after this method. A larger supply of salt water makes the Baltic physically resemble de fjords of the Cattegat, which as we know, freeze in cold winters; and the herring shoals from the North Sea would then follow with the salt current to the gates of the Baltic as they did every authumn in the Middle ages.

It would thus require no very considerable change in the hydrographical conditions to bring about the freezing of the Baltic in severe winters, or to make the herring return to its ancient haunts in the Sound at the coast of Scania. Such conditions recur approximately even in our days in certain years and at certain intervals. It all depends upon the intensity of the water supply from the ocean through the undercurrent. In other papers I have shown that this undercurrents varies under the influence of the tides. It possesses semidiurnal, diurnal, monthly, annual, multiannual and secular periods according to the variation in the tide generating force.

The daily pulsation we discovered in 1907 and succeeding years in the Great Belt, the monthly in the Gullmarfjord. The annual period has been traced in many ways by its effects: higher water level in authumn, maximum of the influx through the Great Belt and Öresund in December, the seasonal migrations of the herrings which now generally stops at the point where the Cattegat channel shoals out of Gothenburg, but in certain authumns sends contingents down to Öresund and the Belt in such a number as to suggest to certain authors the idea that the great Hanseatic fishery of the Middle ages was no other than the herring fishery which is carried on nowadays in Öresund.

Finally there is the well known fact, that in certain winters the drift ice appears in the Sund, the Baltic and the Cattegat in such quantities as to block the Sund and the Cattegat harbours for a couple of months. Such winters appear to occur periodically. In 1809-1814 there occurred such winters, the drift ice getting crammed into the strait between Scania and Bornholm.

Regarding this subject I must refer to former publications (1). The circulation of waters through the Great Belt, which is the chief channel by which the ocean-water finds its way into the Baltic, can be represented by the diagram. Diagram in Fig. 16, next page.

This figure represents a longitude section through the Great Belt, with 3 stations; one at Revsnaes in the northern entrance of the Belt, another at Langeland at its southern egress, and the third at Korsör, in the middle. The salt undercurrent enters the Baltic nearest to the bottom of the sound and introduces water with a salinity of 30‰ and 32‰, the limits of which are shown by the isohalines. The fresher surface stratum is the Baltic water which streams out from the Baltic, its lower limit being represented by the isohaline for 16 and 20‰ salt.

Between the surface current as the undercurrent there exists a limiting stratum, in which to two other water strata are mixed. We see that the ingoing and outgoing kinds of water rest on each other like two wedges, turned with their sharp ends towards each other. (Fig. 16, 17).

1. O Petersson, Strömstudier vid Östersjöns portar. Häfte III af Svenska Hydrografisk Biologiska Kommissionens Skrifter.

Fig. 16

Fig. 17

The pressure of the salt water masses from the Ocean drives in the lower wedge, and the pressure of all the river water that is collected in the Baltic presses out the out the upper water wedge. The result is an outward flowing surfacecurrent (the Baltic stream) and an ingoing undercurrent. It was first believed that these currents were continuous and that the one was dependent of the other, so that a great out current at the surface necessitated a more rapid inflow of salt water below, in order to preserve the balance in the exchange of water between the Baltic and the Ocean. A.W. Cronander, however, made some observations at the lightships in the Cattegat, which showed that both currents were not always active at the same time. Then the opinion was adopted that it would be the result of the wind and the barometrical pressure over the North Sea if the lower wedge was pressed inwards more powerfully than the upper wedge was pressed outwards; because occasions are imaginable when the under wedge is pressed inwards with such force that its salt water masses dam up the surface current, so that the latter absolutely cannot make its way out. The reverse would be the case if the water pressure, or the wind- and barometrical pressure, were strongest over the Baltic; the outgoing water wedge in the Great Belt then being able to fill the whole of the Belt all the way to the bottom and prevent the entrance of the undercurrent. In a word, the exchange of water would, it was thought, depend on temporary atmospherical causes, such as wind, rainfall, atmospheric pressure, etc. This is the opinion generally adopted at present.

But in July 1907, the Swedish Hydrographical Commission made observations at the named stations in the Great Belt and, with the help of modern apparatus, studied the movements of the limiting stratum lying between the surface current and the undercurrent. It was found that its movement was not always the same, but that it was pulsatory or periodic and it was also discovered that the movement proceeded from the lower wedge, which was pressed inward more powerfully every 12th hour of the day and regressed once in between, i.e., also in intervals of 12 hours. The limiting surface between the currents, consequently, did not lay exactly straight, or more correctly speaking, at a level slope from north to south, but went in waves as shown by the schematic illustration in Fig. 18:

Fig. 18

This was, of course, clearly a tidal phenomenon and a very unexpected one for, as is well known, the tide ceases to be felt at the embouchure of the Skagerak, no ebb and flood being noticeable at Lindesnäs. Inside Lindesnäs the tide is slightly noticeable, amounting on the Bohuslän coast to about 3-4 decimeters while in the Belt itself, at Korsör, it is about 1 ½ Danish foot at the surface. We found, however, that the tide waves in the limiting stratum at Korsör 18-20 meters below the surface, were about 10 times as high or about 3 meters. This showed that the ebb and flood of the North Sea, which seems almost to disappear in the Skagerak and Cattegat, still exist in the lower limiting stratum, although it is lessened towards the surface by the upper stratum of light Baltic water, which acts as moderator on the waves.

If this moderating surface stratum did not exist, i.e. if the Baltic was a lake (as during the Ancylus period) and the Great Belt formed the innermost bay of the North Sea and the Cattegat, with homogeneous water of oceanic origin from the bottom to surface, the alternation of the tide water along the shores of the Belt would probably be considerable. The low shores of the Danish Islands would then, perhaps, have the appearance of a marshland defended from the wave by enormous dams, as at Walcheren and on the Friesian coast of the North Sea, or else of a “drowned land”, such as exists off the west coast of Schleswig. It was, therefore, surprising to feel the pulse of the sea beat so strongly in the depths of the Great Belt. But the tidal wave which rushes in through the Belt at a depth of 18-20 meters, does not cross the exceedingly uneven bottom of the strait undisturbed and so enter the Baltic. Each tide wave drives like a cascade of salt water over the thresholds of the Belts and sinks down the deep channel of the Baltic south of the strait, and then continues its inward course along the bottom of this sea, in the form of a so called “solitary wave”.

The exchange of water between the Atlantic Ocean and the North Sea thus obtains a pulsating character, in consequence of the influence exerted by the tide water on the undercurrent. In consequence of the cascades of salt water, the deep depressions of the Baltic are filled one after the other, and when one basin is filled it pours its superfluous water in another cascade into de next, so that in proportion as the salt water enters the Baltic, the surface water is driven out. Such is the influence of the daily and semi daily tide water on the circulation in the Baltic sounds.

Besides these tidal waves in the boundary between the upper current and the bottom current we have the great waves of longer period described in the foregoing which come in from the North Sea moving slowly through the whole of the Skagerak and the Cattegat like great cascades of water which carry with them the shoals of herring from the eastern regions of the North Sea into the Cattegat, which acts like a funnel or a fish trap in the narrow innermost end of which the immigrating herrings are crowded together. There are several such tracts where the shoals of herring are crowded into close masses – “herring mountains”, as they are called by Swedish fishermen – in consequence of the small breadth of the channel, or the irregular contour of the bottom, which acts like a filter that admits the water current but stops the progress of the herring. In such places the herring fishery is carried on with seines (“snörpvadar”). In places where the channel is wider, such as around Anholt and off Varberg, fishing is best carried on with driftnets as the shoals of herring are more widely spread there. The last and the narrowest passage that the water current has to pass before it enters the Baltic is formed by the Sund and the Belts.

Here is that the greatest sea fishery of the Middle ages –the so called Hanseatic herring fishery- was pursued, the centre of which was near Skanör and Falsterbo, and was the richest during the 13th, 14th and 15th century, since which period it has diminished to the humble proportion it now has. What could have been the cause of this decline in one of the principal fisheries of Scandinavia within historic times? It cannot have been any geological alteration diminishing the exchange of water with the Baltic as was the case after the close of the Littorina period, for during the last 2,000 years the bottom of the Sund has not risen more than ¼ meter at most.

Relief chart of Skagerak and Cattegat

The reason does not lie in a geological, but in a hydrographical alteration in the intensity of the circulation of the water, which has been caused by cosmic influence. It can be shown that the tide producing powers of the moon (and the sun) were considerably greater during the centuries mentioned above than they are now, in consequence of the relative position of the orbit of the moon in respect to the earth and the sun at the period of the winter solstice.

At certain occasions during that time of the year the sun, earth and moon approached each other more nearly than usual. Such constellations return at intervals of about 1,800 years and are distinguished by an increased intensity in the circulation of the sea, and the ebb and flood, by great variations in the climate and in solar activity. The increased intensity in the circulation of the water drove the waters of the North Sea into the Cattegat, the Sund and the Baltic. During the Middle ages, as now, this did not occur in one even continuous stream but cascade-like, by means of great under water waves that pressed the salt water into the Baltic much more powerfully than now. These pressings-in were strongest in the autumn, just as now, and with the water masses or “water mountains” that the tide producing power of the moon drove into the Baltic went the great herring shoals, “the mountains of herring” of which Saxo and other chroniclers from the 13th and 14th centuries have astonishing tales to tell. This hydrographical circumstance gave Skankör and Flasterbo a period of extraordinary prosperity during the last centuries of the Middle ages.

The powerful in-current during these centuries filled also the deep channels and basins of the Baltic with salt water to a greater height than now occurs. In consequence, the fresh surface stratum was shallower and during cold winters could be cooled down below the freezing point so that, during certain winters, as for example in 1306, 1321, etc. the Baltic became one great sheet of ice, making is possible to cross the ice from Skåne and Denmark to Germany and from Gothland to Estland in Russia. I have shown that it is possible to calculate what the proportion must be between the surface water and the bottom stratum for the Baltic to freeze again as during the Middle ages. In the fig. 15 the limits of such a proportion are shown by dotted lines. If, in consequence of an increased intensity of the undercurrent, the limit of the salinity 9‰ should be altered so as to lie on a level with that shown by the dotted line, i.e. about 8-10 met. higher at Bornholm and 15-20 met. higher at Gothland, the southern Baltic would freeze as now takes place in the Gulf of Bothnia when there is a cold winter.

The same thing would result if the Sund were made a couple of meter deeper; for example, by dredging the submarine channel called the "Flintrännan” to a greater depth. We should then once more have the rich herring fisheries of the Middle ages at their old places but we should also suffer from the cold winters of that era, with the Baltic Sea completely frozen over, and Skåne, in the south of Sweden, would have the cold continental winter climate it had in the days of Tycho Brahe (middle of the 16th century).

The great immigration of herring that nowadays takes place in the autumn and winter does not, as a rule, extend farther than the tract south of Gothenburg, between Nidingen and Tistlarne, where the submarine channel of the Cattegat becomes narrower. The advance guard of the herring shoals, which makes its appearance as early as August and September, swims higher up into the upper water strata and spreads over the broad expanse of the southern Cattegat where the fish are caught by means of driftnets; the herrings are finally once more squeezed into crowded shoals at the northern mouthpiece of the Sund and the Belts. In certain years, when the sea impulses are stronger, there is a fairly rich herring fishery here which can be experienced even for some distance into the Baltic. On these occasions there is repeated in our days, although on a diminished scale, the great herring fishery of Öresund that existed in the Middle ages, see the relief chart.

The great period of the greatest possible tide power and water circulation in the sea and the most intense solar activity will not return before the lapse of more than 1,000 years. The last time it occurred was at the close of the Middle ages, and the time before that, at the close of the Bronze Age, about 600-400 B.C. But within these 2,000-year periods there are shorter lunar-periods of the second, third and fourth rank with a length of 80-90 years, 18 years, 9 years, 4 years and 2 years, all of which influence the water circulation and the fish life of the Baltic, Cattegat and Skagerak. In the annual migrations of the herrings we have a very sensitive indicator of the influence exerted by these lunar periods on the movements in the sea. When the tide producing power of the moon increases the movements in the boundary stratum at the sea become stronger and the current and the under water waves carry the herring shoals farther into the Cattegat and gather them in more crowded multitudes at the now-existing principal fishery places, viz., the coast bank south of Gothenburg between Tistlarne and Nidingen.

If the lunar power is weaker the herring shoals remain farther out, on the western and eastern sides of the submarine channel of the Cattegat, all of which acts on the results of the herring fishery. An example of this is given by the following diagram. The undulating line represents the declination of the moon, and the years when this has been greatest are marked at the upper curve, and the years when it has been least are shown at the lower arc of the curve. Between every maximum of declination there lies a period of 18.6 years. When the moon attains its highest declination, i.e. when it comes highest and lowest in the sky it exercises its greatest tide producing power on our seas. I have examined all the information that exists respecting the Swedish herring fishery for 150 years, during the herring periods of the 18th and 19th-20th century, and have marked the favourable fishing years with max., and the unfavourable with min.

We find that, without exception, the good fishery years coincide with the greatest lunar declination and tide-raising power, and the bad fishing years with the lower declination (1).

When the tide raising power of the moon last was greatest it was able to carry the herring shoals through the whole of the narrows of the submarine channel of the Skagerak and the Cattegat all the way into the Sund and the Baltic. That was 600-900 years ago.

Fig. 19

When the tide producing power diminished most, the herring migration stops at the first narrows of the submarine channel of the Cattegat, south of Gothenburg, or even farther to the north, so that, at last, the herring has to be sought for off the Skaw (Skagen) in Denmark, as was the case this winter (1913-14). When the tide generating power is not sufficient to carry the North Sea herring farther in, an advance guard of the herring shoals is usually found during December and January in the German North Sea bays and at the mouths of the Elbe and Weser. That which is lost by the Cattegat is gained by the North Sea fishery, just as happened this winter.

It is remarkable that the first growth of the herring fishery in the North Sea known in history began in the 16th century, the period when the great fishery in the Sund came to an end.

(1) In No. 7, 8, 9 of the fishery journal “Der Fischerbote”. (Hamburg, 1911). I have published a detailed description of the vicissitudes of the Swedish herring fishery during the last 150 years.
(2) Ekholm. Quaterly Journ. R. Meteor. Soc. 1901 January

This may seem to present but cheerless prospects for Swedish fishermen and, it is true of course, that the herring fishery is a difficult and dangerous pursuit. But of late years fishing vessels and fishery methods have developed so that it is now possible to look for the herring shoals at other places and to catch them by other means than a century ago, when the Swedish fishermen were contended with fishing in the fjords and sounds with standing nets (“sättgarn”) and land seines (“landvadar”). It is to be hoped too that in the same degree that we are able to discover and study the causes of the visit of the herring to the Swedish coast and its occasional absences from our waters, it will be possible to calculate in advance the prospects of the autumn and the winter fishing seasons.

Part 4: continues in: VII. The Climate of Northern Europe at the Close of the Middle age

Back to Climate Change Page         Back to English Version

You are visitor number:
since December 2008

See the weather in Argentina

See here where our readers come from

Locations of visitors to this page

Tell us your opinion!

(Please paste the link or name of this article!)