1. Briffa, K.R., Bartholin, T.S., Eckstein, D., Jones, P.D., Karlén, W., Schweingruber, F.H. and Zetterberg, P. (1990). A 1400-year tree-ring record of summer temperatures Fennoscandia. Nature 346: 434-439.

Tree-ring data have been used to reconstruct the mean summer (April-August) temperature of northern Fennoscandia for each year from AD 500 to the present. Summer temperatures have fluctuated markedly on annual, decadal and century timescales. There is little evidence for the existence of a Medieval Warm Epoch, and the Little Ice Age seems to be confined to the relatively

short period between 1570 and 1650. This challenges the popular idea that these events were the major climate excursions of the first millennium, occurring synchronously throughout Europe in all seasons. An analysis of past warming trends suggests that any summer warming induced by greenhouse gases may not be detectable in this region until after 2030.

2. Briffa, K.R., Jones, P.D., Bartholin, T.S., Eckstein, D., Schweingruber, F.H., Karlén, W., Zetterberg, P. and Eronen, M. (1992). Fennoscandian summers from AD 500: temperature changes on short and long timescales. Climate Dynamics 7: 111-119.

Quantitative estimates of 1480 years of summer temperatures in northern Fennoscandia have previously been derived from continuous tree-ring records from northern Sweden. Here we show the results of spectral analyses of these data. Only a few peaks in the spectra are consistently significant when the data are analyzed over a number of sub-periods. Relatively timestable peaks are apparent at periods of 2.1, 2.5., 3.1, 3.6, 4.8, ~32-33 and for a range between ~55-100 years. These results offer no strong evidence for solar-related forcing of summer temperatures in these regions. Our previously published reconstruction was limited in its ability to represent long-timescale temperature change becauses of the method used to standardize the original tree-ring data.

Here we employ an alternative standardization technique which enables us to capture temperature change on longer timescales. Considerable variance is now reconstructed on timescales of several centuries. In comparison with modern normals (1951-70) generally extended periods when cool conditions prevailed, prior to the start of the instrumental record, include 500-700, 790-870, 1110-1150, 1190-1360, 1570-1750 (A.D.) with the most significant cold troughs centered on about 660, 800, 1140, 1580-1620 and 1640. Predominantly warm conditions occurred in 720-790, 870-1110, and 1360-1570 with peaks of warmth around 750, 930, 990, 1060, 1090, 1160, 1410, 1430, 1760 and 1820.

7. Eronen, M., Hyvärinen, H. and Zetterberg, P. (1999). Holocene humidity changes in northern Finnish Lapland inferred from lake sediments and submerged Scots pines dated by tree rings. The Holocene 9: 569-580.

Conclusive evidence for a rise in water levels has been found in connection with lake-sediment studies undertaken partly in collaboration with the pine megafossil sampling and dendrochronological work in northern Finnish Lapland. The change in lake-level stands is shown by slow sedimentation rate in the early to mid-Holocene and an increase thereafter. These data indicate a regional rise in water levels during the latter part of the Holocene following a relatively dry period between 8000 and 4000 BP. Synchronous changes, also indicating rising water levels, have been observed in the diatom and cladoceran assemblages of the sediment cores. Subfossil Scots pines (Pinus sylvestris, L.) have been preserved in large quantities in small lakes in Lapland, because in many cases the rising water level has inundated the trunks after death. The position of the subfossil trunks and stumps often indicates that the pines have been growing on dry land at the sides of the lakes in which they are now submerged. Traces of a bark beetle (Tomicus minor, Hart.) have been detected in a few old pine logs found far outside the present distribution area of this insect.

A total of 1722 samples of pine subfossils have been collected in the forest-tundra ecotone region of Lapland to build a continuous pin ring-width chronology over 7000 years long. The long chronology is almost finished, but its two parts are still separated by a short discontinuity around 250 B.C. An absolutely dated, year-by-year chronology over 2000 years long extends from the present close to that time. The older, over 5000 years long continuous floating chronology is fixed to the timescale by several radiocarbon dates. A total of 1212 samples of pine wood have been dated and assembled within the chronologies by tree-ring cross-matching. These substantial data indicate a gradual retreat of pine tree and forest limits with some marked regional differences during the past 5000 years. According to the preliminary interpretations of the tree-ring data the variability of Holocene summer temperatures has increased towards the present time. Shifts in climatic development leading to cooler and more unstable conditions seem to have occurred in mid-Holocene time, and between 2500 and 2000 BP. The increase in humidity is probably in association with these changes in the high-frequency variability of temperature.

9. Eronen, M., Lindholm, M. and Zetterberg, P. (1994). Extracting palaeoclimatic information from pine tree rings in Finland for historical climatic proxy data collection. In Frenzel, B. (ed.): Proceedings of the ESF workshop on 'Climatic trends and anomalies in Europe 1765-1715', Bern, September 1992. Special issue 8. Paläoklimaforschung - Palaeoclimate Research 13: 43-50.

Forest limit pines in northern Fennoscandia constitute a valuable source of palaeoclimatic data, as their ring-width variations are indicative of summer temperatures and cross-correlation techniques can be used to extend the pine chronologies back over thousands of years. Tree-ring data also add to our knowledge of the nature of climatic changes during historical times.

Historical climatic proxy data can be correlated with tree-ring records leading to improved data quality. Stable isotope measurements from the tree-rings will also provide valuable information on past climatic variations.

20. Eronen, M. & Zetterberg, P. (1996). Climatic Changes in Northern Europe Since Late Glacial Times, with special reference to dendroclimatological studies in northern Finnish Lapland. Geophysica , 32(1-2): 35-60.

The oxygen isotope variations in deep Greenland ice cores indicate large and extremely rapid temperature oscillations during the last ice age. Strongly fluctuating climatic conditions also characterized the final few thousand years of the Weichselian glaciation. Temperatures rose to levels near their present values during the Bolling-Allerod interstadial, but during the Younger Dryas episode (12,700- 11,500 years ago) the cold ice age conditions came back. The cold spell ended with an abrupt warming about 11,500 years ago and it marked the beginning of the present warm interglacial, the Holocene. The Holocene climate has also varied, but all oscillations in temperatures have been an order of magnitude smaller than those common in the North Atlantic region during the last ice age. Temperatures rose to about present-day values in early Holocene time and generally continued to rise slowly until 6000-5000 BP. During the last 5000 years climatic conditions have gradually become cooler and obviously somewhat more unstable. The variations in humidity show some differences in various regions of Fennoscandia, but since 2500 BP, wetness has enerally increased.

There are different climatic proxy data which can be used for Holocene studies, but the extraction and interpretation of signals of small short-term changes is often difficult because of unavoidable inaccuracies in data and dating. The main trends of Holocene climatic development are relatively well known over northern Europe, but the short-term variations are not known in detail. Tree rings provide a possibility to study high-frequency climatic variability, since annual and even seasonal resolution in dating can be achieved using dendrochronology. The pine tree-ring data collected from the tree-line area of northern Fennoscandia indicate changes in past summer temperatures. The absolute tree-ring curve constructed from subfossil pines (Pinus sylvestris, L.) of Finnish Lapland indicates a duration of over 2000 years, extending to 165 B.C. and, after a ca. 200-year gap, the older unbroken part of the chronology extends until about 7500 calendar years before present. Preliminary interpretations of the data suggest that the climatic variability in the study area increased around 5000 BP with a subsequent trend towards cooler and wetter climatic conditions.

22. Eronen, M. and Zetterberg, P. (1996). Expanding megafossil data on Holocene changes at the polar/alpine pine limit in northern Fennoscandia. Paläoklimaforschung - Palaeoclimate Research , 20: 127-134.

Megafossils of Scots pine collected from small lakes and peat bogs in northern Fennoscandia provide excellent material for dendrochronological research and at the same time provide information on past changes in the timberline.The material is of great value for studying Holocene climatic history. The tree-ring analyses are now rapidly yielding new megafossil dates. Once the hundreds of samples can be cross-matched and the age of the resulting chronology can be accurately determined, the result will be a large number of accurately dated megafossils. About 500 subfossil pine logs have been dated so far, and the resulting data have improved our knowledge of the Holocene history of the pine forest in the far north.

The data will expand as the dendrochronological work proceeds, since the total number of samples collected is now over 1000. While new samples are being added to the general tree-ring chronology for Finnish Lapland, which extends back in time about 7000 years, attempts are being made to fill in the gap of about 300 years in the curve shortly before the beginning of the Christian era. The present megafossil data corroborate the earlier conclusion that the time of maximum distribution of pine in the northern Fennoscandia was between 2000 and 4000 BC. During the later part of the Holocene the limit of pine has generally retreated, but there are differences in this development between different regions.

27. Eronen, M., Zetterberg, P. and Okkonen, J. (1994). Kirveen ja vaakanävertäjän (Tomicus minor, Hart.) jälkiä Lapin subfossiilisissa männyissä (Abstract: Axe marks and traces of a bark beetle (Tomicus minor, Hart.) in subfossil pines from Finnish Lapland). Terra 106: 238-248.

Subfossil pines (Pinus sylvestris, L.) have been collected from northern Finnish Lapland and adjacent areas of Norway in order to construct a dendrochronological master curve extending back over 7000 years. The aim is to make palaeoclimatological interpretations from the variations in annual ring widths, buts some other interesting findings have also emerged from the material. We describe here some traces of prehistoric axe blows found on subfossil trunks and imprints made by the bark beetle Tomicus minor, Hart. Mining patterns typical of this insect species were found in a subfossil pine recovered from a small lake over 150 km north of the northernmost current observations of living T. minor. Dendrochronological dating of the pine concerned suggests an age of ca. 2800 B.C., but the date is a little uncertain, because it was difficult to match the tree rings with the master chronology of Finnish Lapland. Nevertheless, the available evidence suggests that the tree died thousands of years ago.

The traces of axe blows were cut at that site for several centuries around 1000 A.D. The pine logs had not been shaped any further after cutting. The upper part of the trunk was intact and even the thicker branches were still left on many of them . The most probable explanation is that the lake was a place used for hunting wild reindeer during the winter, when the site was frozen. Lake Ailigasjärvi is situated in a relative narrow passage formed between the steep slope of the fell Ailigastunturi and the broad Tana River(Tenojoki). The trees were probably felled close to the lake and pulled onto the ice to build a barrier for the reindeer herds so that the trapped reindeers could be killed. Many archaeological finds show that areas close to Lake Ailigasjärvi were continuously settled by the Saami during the Iron Age, at the times when these subfossil trees were cut.

30.Lehtonen, H., Huttunen, P. and Zetterberg, P. (1996). Influence of man on the forest fire frequency in North Karelia, Finland, as evidenced by fire scars in Scots pines. Annales Botanici Fennici 33: 257-263.

Fire scars on living Scots pines (Pinus sylvestris L.) and pine stumps dated by dendrochronology were used as evidence for fire history, the chronology of which was established for a period of 582 years, 1412-1994. A total of 36 different fire years were identified,

the average incidence of fires in the area being once every 11.2 years. The mean fire interval was 36.7 years on the upper part of the hill and 58.6 years on the lower part. Forest fires increased in the 17th century and decreased at the end of the 19th century.

39. Zetterberg, P. (1986). Dendrokronologisesti ajoitettuja rakennuksia Pohjois-Karjalassa. (Summary: Dendrochronologically dated historical buildings in North-Karelia, Finland). Joensuun yliopisto,Karjalan tutkimuslaitoksen julkaisuja (University of Joensuu,Publications of Karelian Institute) 75: 1-25.

In North-Karelia, Finland, a dendrochronological master curve of 394 years has been constructed from living Scots pine (Pinus sylvestris). This chronology has been applied in the present study for the precise dating of two 18th century wooden buildings (situated in Ilomantsi and Lieksa).

Cross dating between the master curve and ring width series from 10-11 beams of both the buildings gave the exact years of felling of the trees. These years are A.D. 1766(building in Ilomantsi) and A.D. 1786 (building in Lieksa). Addition of these data to the master curve extended the dendrochronology of North-Karelia back to the year A.D. 1375.

40. Zetterberg, P. (1987). Site chronologies from eastern Finland and Soviet Karelia. Ann. Acad. Sci. Fennicae, Serie A, Geologica 145: 49-55.

Fifteen site chronologies based on living Scots pines (Pinus sylvestris L.) are constructed from eastern Finland and Soviet Karelia. Eight chronologies from differing habitats are compared both visually and statistically. The results show that it is possible to synchronize and cross-date chronologies from poor peatland sites with chronologies from mineral soil sites.

There is also strong agreement between the Finnish chronologies and a Soviet chronology, which reflects the dendrochronological homogeneity of the region concerned.

43. Zetterberg, P. (1987). Museoesineiden dendrokronologinen ajoitus; esimerkkinä Lieksan huhmar. (Dendrochronological dating of wooden museum specimens). Suomen Museo 94: 109-114.

Until recently dating of wooden museum specimens has been possible only roughly by reference to their style and carving technique if no historical documents are available. Dendrochronology provides a precise method for this problem. Here dating of a wooden mortar from Lieksa museum, eastern Finland, is presented. First step was to analyze the tree species (Scotch pine Pinus sylvestris L.) from a small sample which was taken by increment corer. After that the ring widths were measured from the sample under the microscope with and accuracy of 1/100 mm.

The ring width series was then carefully examined and compared with several formerly constructed absolute references series of pine growth. Longest such pine chronology in eastern Finland extends to the year A.D. 1375. Tree ring series from the mortar was successfully correlated with all reference series, to give a date of A.D. 1911 for the outermost ring preserved. In this case some of the surface wood on the mortar had decayed away, thus the outermost ring had not been the last to form before the tree was felled. Because of this lack of some (approx. 3-4) rings the mortar was made no earlier than A.D. 1915.

44. Zetterberg, P. (1988). Dendrochronological dating of the timber of the medieval stone church of Lempäälä in Satakunta, Southern Finland. Fennoscandia Archaeologica 5: 122-126.

The tree-ring dating of the timber structures of the medieval stone church of Lempäälä indicates that seven of the eight sample timbers of pine (Pinus sylvestris L.) were felled in the period between the summers of 1503 and 1504. The dating is in agreement with previous views regarding the church and provides additional details. One of the sample timbers was dated to renovations and repairs carried out in the early 19th century. This partly decayed timber was felled in the period 1810-1825. The tree-rings of the timbers provide data on the preceding centuries as far back as the beginning of the 13th century.

The 14th and 15th centuries display several changes of growth that can be seen uniformly in all of the samples. These were probably linked to weather conditions of the period. The tree-rings also permit datings to within a year of outside damages to the logs. The material presented in this connection has been linked with corresponding samples from nearby areas to form a uniform and absolute tree-ring calendar for pine extending back in time nearly 800 years. Old pine material from Southern Finland from as far back as the 13th and 14th centuries can now be dendrochronologically dated to within a year.

45. Zetterberg, P. (1988). Dendrochronology and archaeology: dating of a wooden causeway in Renko, southern Finland. Fennoscandia Archaeologica 5: 92-104.

The dendrochronological dating is described from the point of view of what the archaeologist needs to know about the method. Main results of this study are: a) the wooden causeway in Renko was built after the year 1828 and the whole construction is of the same age, b) the applicability of the pine

tree-ring chronology for Eastern Finland in the southern parts of the country is good, c) a new 450-year pine tree-ring chronology for Southwestern Finland is established for dating wooden material in the southern part of the country.

50. Zetterberg, P. (1990). Dendrochronological dating of a wooden causeway in Finland. Norwegian Archaeological Review 23: 54-58.

A mean ring-width chronology for Scots pine in eastern Finland has been compiled based on living trees and dendrochronologically dated historical timber constructions. As an application of this master chronology, a young archeological site near the town of Hämeenlinna, southern Finland, was dated. A previous assumption of the age of this site was that it could even be medieval.

The construction was not older that 160 years, the road having been built after 1828, the year of growth of the outermost tree rings of the samples. A new pine chronology for southern Finland extending to the year AD 1539 is presented and it is now applicable for precise dating purposes in this area.

56. Zetterberg, P. (1992). Muistomerkki Laukaan historian alkutaipaleelta, Hartikan hirsikirkon jäännökset 1590-luvulta (The remains of the wooden church of Hartikka of the 1590s). Keski-Suomi, Keski-Suomen museon julkaisuja 19: 24-31.

The first church in the parish of Laukaa is assumed to have been built in 1593 around the time when the parish itself was founded. In 1685 a new church was built and the old building was completely deserted. It was still standing in 1750. At present, only a blockwork corner of six course logs of the old nave part exists. It is protected by a later roof structure, and is the oldest existing remnant of a wooden building in Central Finland.

To establish the time of building, a study was carried out by the Laboratory of Dendrochronology of the Centre for Karelian Studies. Bored samples were taken of the three best-preserved logs. The results showed that the timber for the logs was felled in the winters of 1589/1590 and 1590/1591 (two logs), which means that they could have been used for building in 1591 at the earliest. This supports the previous dating of the church to 1593. The samples also reveal the varying rates of tree growth from the 14th to the 16th centuries.

57. Zetterberg, P. (1994). Dendrokronologiset tutkimukset Turun linnassa (Abstract: Dendrochronological studies at the Medieval Castle of Turku). In: Drake, K., (Ed.) Tutkimuksia Turun linnassa - Åbo slott -studier 1, Turun maakuntamuseo, raportteja, 1: 39-48.

Dendrochronological analysis were made from 28 wood samples from the Medieval Castle of Turku. A major part of the samples were collected in connection with earlier investigations and renovations. 11 samples were obtained from wooden structures in various parts of the castle. Eleven samples were of pine and their rings could be dated precisely to cover time period 1114-1539. In most cases the outermmost annual rings had been carved or rotted away and the felling date could be estimated only by an accuracy of five to 20 years. The approximated felling dates vary from 1320's to 1540's. 12 samples were of spruce and five of oak.

The oak samples were collected for tree species identification only. From eight of the spruce samples the precise felling date could be defined to winter season 1428/1429.

Some preliminary conclusions from the climatic conditions during the medieval times are drawn from the ring-width material. During years 1150-1350 periods 1200-1235 and 1300-1325 seem to be cool when compared with the rest. For any final conclusions this material is too limited.

The sampling technique and requirements and restrictions of dendrochronological dating method are also briefly discussed.

72. Zetterberg, P. and Eronen, M. (1994). Subfossiilisten mäntyjen dendrokronologiset tutkimukset Lapin metsänrajavyöhykkeessä (Dendrochronological studies on subfossil forest-limit pines in Lapland). Acta Universitatis Ouluensis, Series A 251: 97-105.

Lapin metsänrajalta ja sen tuntumasta on koottu yli 1000 näytekiekkoa järvien pohjalla ja soissa säilyneistä vanhoista männyistä. Näytekiekoista on tehty dendrokronologisia mittauksia, joiden avulla saadaan selville puiden paksuuskasvun vaihtelut. Tutkimuksen tässä vaiheessa on jo mitattu noin 800 näytettä.

Puiden vuosilustosarjat voidaan ristiinajoittaa eli kytkeä toisiinsa, jos puut ovat eläneet samanaikaisesti vähintään useiden vuosikymmenien ajan. Eri-ikäisistä puista voidaan näin rakentaa useiden tuhansien vuosien pituinen lustokalenteri, jos monenikäistä puumateriaalia on saatavilla. Alustavia iänmäärityksiä kalenterin eri osille saadaan radiohiiliajoituksista, joita on ollut tämänkin tutkimuksen tukena.

Siitepölytutkimukset ja radiohiiliajoitukset (14C) osoittivat männyn levinneen Pohjois-Lappiin yli 7000 vuotta sitten, joten näin pitkän puulustokalenterin laatiminen sinne on mahdollista. Kalenterista onkin jo suurin osa valmiina, mutta kaikkia ‘leijuvia’

osakronologioita ei ole vielä saatu liitetyksi toisiinsa, sillä ensimmäisen esikristillisenvuosituhannen kohdalla on usean sadan vuoden pituinen aukko.

Männyn paksuuskasvu on Lapissa hyvin selvästi riippuvaista kesälämpötiloista. Sen vuoksi puulustokalenterista voidaan tehdä ennen muuta kesälämpötiloja koskevia paleoklimatologisia tulkintoja. Tässä työssä ei olla vielä kovin pitkällä, koska tutkimuksessa on luonnollisesti keskitytty ensin kronologian rakentamiseen. Näyttää kuitenkin siltä, että holoseenin alkupuolella ilmasto oli pohjoisessa melko vakaa verrattuna holoseenin jälkipuoliskoon, jolloin ilmaston vaihtelevuus näyttää lisääntyneen aiheuttaen vähittäistä jäähtymistä,

Paleoklimatologisen tulkinnan pohjaksi ollaan tutkimassa myös elävistä puista kairattuja vuosilustonäytteitä, joiden paksuusvaihteluja verrataan säähavaintosarjoihin. Lisäksi vanhojen mäntyjen lustoista on alettu mitata hiilen isotooppien (13C ja 12C) suhteita, jotka antavat myös tietoa ilmaston ja ympäristön muutoksista.

75. Zetterberg, P. and Eronen, M. (1995). Dendrochronology of forest limit Scots pine (Pinus sylvestris L.) and postglacial climatic changes in northern Fennoscandia. In Heikkinen, O., Obrebska-Starklowa, B. and Tuhkanen, S., (Eds.): Environmental aspects of the timberline in Finland and in the Polish Carpathians . Uniwersytet Jagiellonski, Krakow 1995, pp. 57-65.

Samples from 1265 subfossil pines have been collected from 38 sites in the timberline zone of Finnish Lapland and North Norway. Most of the subfossil trees were well preserved in the bottom sediment of small lakes and ponds. Tree-ring measurements were made on two to four radii per trunk. 858 trees have now been assigned to a number of site chronologies by crossmatching their ring series, and the common master chronology for Finnish Lapland extends almost 7000 years back in time.

The absolutely dated part extends to the year 165 B.C. Subfossil pine material has been used to study Holocene climatic history and tree-line changes. Several variables, such as (1) long-term mean ring-width, (2) year-to-year variability in ring-widths, (3) data on death and regeneration periods, (4) age class gradation and (5) the size of the pine population at the sampling site, have been studied in this material. One case study of the Atlantic/Subboreal climatic shift is quoted here.

79. Zetterberg, P., Eronen, M. and Briffa, K.R. (1994). Evidence on climatic variability and prehistoric human activities between 165 B.C. and 1400 A.D. derived from subfossil Scots pines (Pinus sylvestris L.) found in a lake in Utsjoki, northernmost Finland. Bulletin of the Geological Society of Finland 66: 107-124.

Samples from 1265 subfossil pines have been collected from small lakes and peat deposits in the forest-limit zone of northern Fennoscandia in order to study past variations of climate. Many of the subfossils have been dated by dendrochronology and the chronology constructed from the measured ring-width data extends as a continuous master curve from the present back until 165 B.C. and after a short gap until about 7000 years before the present time. This material has greatly increased the number of dated pine megafossils in northern Finland which had previously been restricted only to radiocarbon-dated samples. In addition to the year-by-year information provided by tree-ring width data, the temporal distribution of pine megafossils found in the vicinity of the forest-limit zone also provides information on past climatic changes.

The 102 pine subfossils collected from Lake Ailigas, in Utsjoki, form part of the above material. They provide information about past variations in pine growth caused, to a large degree, by changing climate at this one site, but they also give glimpses of the local activities of Prehistoric Man. The data from 90 of these trees have been successfully dated using dendrochronological techniques and the results show that all of them grew during the time period beginning 3000 years before present, and that 79 pines lived during the time span 165 B.C. to A.D. 1952. In several lakes in the forest-limit zone, some subfossil trees are much older than those in Lake Ailigas. The relatively young ages of the subfossils at this site indicates that the lake has been in existence probably only during the past 3000 years, forming when climate turned more humid than in earlier times.

The present continuous master curve is about 600 years longer than the earlier published pine chronology for northern Sweden, though this has recently been extended to A.D. 1.

In the present study, the life spans of individual dated pines are considered in conjunction with the variability in the measured tree-growth curve. These provide detailed information on growth conditions and variations in temperature during the period 165 B.C. - A.D. 1400. The correlation of the present data with the previously published results from northern Sweden display great general similarity, but also some significant differences. We propose that the combined data give a more reliable indication of regional climatic variations and that the differences are due to local factors at the two locations. The similarity between these two independent series supports general conclusions made about changing summer climate over Fennoscandia after A.D. 500 and strongly suggests that our conclusions drawn here for the period 165 B.C. - A.D. 500 are on a firm basis. The present data thus enable us to infer the record of summer climate variability in northern Fennoscandia about 600 years back in time.

Most of the subfossil pines preserved in the lake grew very close to the place where they were found. Reaction wood indicating tilting of the trunk, probably caused by rising water levels, was found in several trees. These trees grew on the lake shore where the soil was soft or eroding. The large number of relatively young trees indicates that storms may have occasionally felled pines. This cannot be proven (by demonstrating a common death year) however, because the outermost tree rings in the subfossil material are frequently decayed. A small number of the pines found in Lake Ailigas were cut by Man. The pines, identified by axe marks may be remnants of barriers built by the Prehistoric Saami for hunting of wild reindeer. The dates of these indicate that traps were constructed in the mid 9th century and at the end of the 11th and start of the 12th centuries, i.e. late Iron Age times

81. Zetterberg, P., Eronen, M. and Briffa, K.R. (1995). Subfossil forest limit pines from northern Finland: evidence on climatic variability, growth variations and prehistoric human activities betveen 165 B.C. and A.D. 1400. In Ohta, S., Fuji, T., Okada, N. Hughes, M.K. and Eckstein, D. (eds.) Tree Rings, from the past to the future, Proceedings of the International Workshop on Asian and Pacific Dendrochronology. Forestry and Forest Products Research Institute Scientific Meeting Report , 1:134-144.

Samples from 1265 subfossil pines have been collected from small lakes and peat deposits in the forest-limit zone of northern Fennoscandia in order to study past variations of climate. Major part of the subfossils have been dated by dendrochronology and the continuous tree-ring master curve extends from the present back until 165 B.C. and after a short gap until about 7000 years before the present.

A total of 102 pine subfossils collected from Lake Ailigas, a small lake in northernmost Finland, provide information about past variations in pine growth caused by changing climate at this one site, but they also give glimpses of the local activities of prehistoric Man. The dating of 90 of these trees show that they grew during three millennia before present, and that 79 pines lived during the time span 165 B.C. to A.D. 1952. The relatively young ages of the subfossils at this site indicate that the lake was formed when humidity of climate increased and probably it has been in existence only during the past 3000 years.

The life spans of dated pines are considered in conjunction with the variability in the tree-growth curve.

These provide detailed information about variations in growing conditions and temperature during the period 165 B.C. to A.D. 1400. The correlation of the present data with the data from northern Sweden display great general similarity, but also some significant differences. The similarity between these two independent series supports conclusions made about changing summer climate over Fennoscandia after A.D. 500 and strongly suggests that our conclusions drawn here for the period 165 B.C. to A.D. 500 are on a firm basis. The present data thus enable us to infer the record of summer climate variability in northern Fennoscandia about 600 years further back in time.

Reaction wood indicating tilting of the trunk, probably caused by rising water levels, was found in several trees. The common occurrence of relatively young trees in the data indicates that storms may have occasionally felled pines. A small number of the pines found in Lake Ailigas were cut by Man. The pines, showing evidence of axe marks may be remnants of barriers built by the prehistoric Saami-people for hunting of wild reindeer. The dates of these indicate that traps were constructed in the mid 9th century and at the end of the 11th and start of the 12th centuries, i.e. late Saami Iron Age times.

83. Zetterberg, P., Eronen, M. and Lindholm, M. (1996). The mid-Holocene climatic change around 3800 B.C.: tree-ring evidence from northern Fennoscandia. Paläoklimaforschung - Palaeoclimate Research 20: 135-146.

In order to construct a continuous pine (Pinus sylvestris, L.) chronology for northern Fennoscandia, samples from over 1000 subfossil pines have been collected from 31 sites, mainly small lakes in the polar/alpine timberline zone of Finnish Lapland and North Norway. Complete discs were sawn from the trunks to allow tree-ring measurements to be made on three or four radii per tree. More than 500 trees have now been connected to dated chronologies, and the absolutely dated part of the Finnish Lapland master chronology extends back to the year 115 B.C. Before that there is still a major gap of several hundred years around and after the Subboreal/Subatlantic (around 500 B.C.) transition, but existing floating chronologies fixed to a time scale by means of a number of radiocarbon dates cover most of the last 7000 years.

Abundant material is available for the mid-Holocene time. About 140 samples collected from 13 sites in Enontekiö and Utsjoki in Finland and Troms close to the Finnish boundary in Norway fall into the time interval 4500-3000 B.C. A number of variables such as the size of the pine population at the sites, the long-term mean ring- width, year-to-year variability in ring-widths, death and regeneration periods and the age of trees at death have been studied on this material, and a preliminary temperature reconstruction has also been made for this period. The chronologies show on average relatively small variations in ring-width before 3800 B.C., but increased variability afterwards. This suggests greater variability in summer temperatures, and thus a change in the climatic pattern after 3800 B.C. The mid-Holocene climatic change at this time was thus largely the result of a shift towards less stable conditions.

84. Zetterberg, P., Eronen, M. and Lindholm, M. (1996). Construction of a 7500-year tree-ring record for Scots pine in northern Fennoscandia and its application to growth variation and palaeoclimatic studies. In Spiecker, H., Mielikäinen, K.., Köhl, M. and Skovsgaard, J. (eds.) ' Growth Trends in European Forests: Studies from 12 countries ', pp. 7-18. Springer-Verlag, Berlin Heidelberg.

Precisely dated subfossil trees and long chronologies of tree growth provide a unique calendrical record for studies on growth variations in the past and climate history. A total of 1465 subfossil Scots pines (Pinus sylvestris L.) has been sampled from 42 sites in northern Fennoscandia. The sampling sites are mostly small lakes in the timberline zone located between 68o30' N and 70o00' N and at elevations ranging from 75 m a.s.l. to 515 m a.s.l. A total of 1023 subfossil trees have been dated by dendrochronology. From the dated trees we have constructed a chronology for each growing site and also a common chronology for the entire area. In this study, subfossil pines are used as 'growth reference' of forest site productivity in the past. The abundance of subfossil pines makes it possible to explore the long-term stand history of pine at each site over thousands of years. The unbroken tree-ring record extends from the present back until 165 BC and after a 150-200-year gap until about 7500 years before the present time. The continuous part of the master curve is 664 years longer than the earlier published pine chronology for northern Sweden. This 7500-year pine master chronology can be used for dating subfossil pines from a wide area in northern Fennoscandia.

The subfossil pine material has been used for studies on the Holocene climatic history, growth variations, year-to-year variability in ring-widths, periods of germination and mortality, population size and age structure at the sampling sites and tree-line changes. These combined data are valid indicators of diverse environmental changes in the past. Detailed register of the growth variations is available for two periods in mid- to late-Holocene times. The data for the period 4500-3000 BC suggests increased variability in tree growth after 3800 BC. The mid-Holocene climatic change at this time was thus largely the result of a shift towards less stable growing conditions. For the late Holocene period from 165 BC to the present the dating accuracy of variations in tree growth and climate is one year. The mean annual growth has temporarily varied significantly, being highly dependent on growing conditions at each site, but when the entire material is considered, in the long term the mean ring width has remained rather constant, being approximately 0.6 mm. The available data from the Holocene climatic optimum and late Holocene do not differ from each other in this respect.

85. Zetterberg, P. and Hiekkanen, M. (1991). Dendrochronological studies on the age and construction phases of the medieval stone church of Sipoo (Sibbo). Finskt Museum 97: 87-98.

Archaeological studies in recent years suggest that the medieval stone churches of the province of Uusimaa in Southern Finland are older that hitherto assumed. In 1990 a number of samples for dendrochronological analysis were taken from the church of Sipoo in connection with repairs and renovations. These samples were analysed at the Laboratory of Dendrochronology of the University of Joensuu. They indicated that the construction of the nave and the sacristy had begun in the late 1440s, and the main stage of building was in the mid-1450s. The porch, used as a bell-tower, was probably built in 1453-1454.

The present results suggest a re-evaluation of the time of construction of the medieval churches of Uusimaa, which may have been the result of planning on the diocesan level in the first half of the 15th century. The sampling process applied in this connection also showed that even careful preliminary investigations and selection of sampling locations does not necessarily ensure successful analysis and that extra samples are often required. In the church of Sipoo, for example, the timber structures in the attic of the sacristy may prove to be interesting in this respect.

89. Zetterberg, P., Saksa, A. and Uino, P. (1995). The early history of the fortress of Käkisalmi, Russian Karelia, as evidenced by new dendrochronological dating results. Fennoscandia Archaeologica , 12: 215-220.

Dendrochronological analyses were made of seven timber samples collected at archaeological excavations at the Fortress of Käkisalmi. The samples were taken from a layer containing timber 1.5-2.5 below the present surface. All samples were of pine, and their annual rings could be dated precisely to cover the time period AD 1184-1373. In most cases the outermost rings had been carved off or decomposed,

and thus the felling dates of this material vary between 1332 and ca. 1420. Only some sample had the bark present and its felling date could be dated exactly to the winter season (September-April) 1363/1364. All dating results presented here do not agree with earlier datings from this timber layer, but our material is still too limited for final conclusions.