BrVQ: Difference between revisions

From Cybis Wiki
Jump to navigation Jump to search
Line 19: Line 19:
'''Historical:''' 1197-1490
'''Historical:''' 1197-1490


Block 2b is very problematic. It is dated by Bråthen to 1490, and the youngest part is obviously correctly dated, but the older part does not correlate at all to other references at its given position! Instead that older part correlates as if the whole 2b-block should be dated to 1197, i.e the samples in the oldest part are erroneously dated 293 years too late! Samples of two very different ages have been mixed together.
'''Block 2b''' is very problematic, which is revealed by a [[block analysis]]. It is dated by Bråthen to 1490, and the youngest part is obviously correctly dated, but the older part does not correlate at all to other references at its given position! Instead that older part correlates as if the whole 2b-block should be dated to 1197, i.e the samples in the oldest part are erroneously dated 293 years too late! Samples of two very different ages have been mixed together.


When carefully reading the documentation and constructing the samples positions and lengths these mistakes were revealed: Six samples from Götene and Kungslena churches which are from early 12'th century have been mixed into the chronology at a position 293 years too late. As there is only an average series available, only the early and late ends of Block 2b may be used in a new reference as these "clean ends" are based on one curve and are not created as a mean value mix of incorrectly matched chronology members, i.e  904-974 (misdated as 1197-1267) and 1419-1490. And then only 939-974 (35 yrs) and 1419-1482 (63 years) have at least a four samples depth.  
When carefully reading the documentation and constructing the samples positions and lengths these mistakes were revealed: Six samples from Götene and Kungslena churches which are from early 12'th century have been mixed into the chronology at a position 293 years too late. As there is only an average series available, only the early and late ends of Block 2b may be used in a new reference as these "clean ends" are based on one curve and are not created as a mean value mix of incorrectly matched chronology members, i.e  904-974 (misdated as 1197-1267) and 1419-1490.


On page 32-33 is a list of 12 samples which are said to be the samples in the average serie "Block2b", but that list is not correct. That is obvious when compared to the number of samples said to be behind each value in the average chronology. The correct list could almost be reconstructed from the complementary list at pp. 90-94 and the years when the number of years are altered. It is than revealed that one sample from Hedared is forgotten in the first list and that one sample from Götene church is confused i the table.  
On page 32-33 is a list of 12 samples which are said to be the samples in the average series "Block2b", but that list is not correct. That is obvious when compared to the number of samples said to be behind each value in the average chronology. The correct list could almost be reconstructed from the complementary list at pp. 90-94 and the years when the number of samples per year are altered in the chronology. It is than revealed that one sample from Hedared is forgotten in the first list and that one sample from Götene church is confused i the table.  
{| class="wikitable" border=1
{| class="wikitable" border=1
|-
|-

Revision as of 14:52, 19 June 2009

BraVSQ Oak Chronology for the region Västra Götaland by Alf Bråthen. It was originally published in the book: Alf Bråthen: Dendrokronologisk serie från västra Sverige 831-1975. Rapport RAÄ 1982:1. ISBN 91-7192-511-2.

In the book four different Chronologies are given, with annual average values and the number of measurements behind each value. There are also detailed documentation about the origins of each sample. The four chronologies are originally named: "Block 1", "Block 2", "Block 2b" and "Block 3". Unfortunately there are some severe errors in the material, which have to be corrected before the chronologies can be used. The wrong date for the Lödöse Chronology was discovered soon after the publication, and a corrected (and enlarged) chronology was published by Bråthen later [ref needed].

Still the material from the book is very valuable and useful, when missdatings are corrected and bad sequences are removed. The measurements were originally done in 0.1 mm resolution, but the average series are given in 0.01 mm resolution.

An analysis of the BraVSQ-material

For the following analysis, I have used four independent oak-chronologies: OlandQUSP, WestDK, BohusRec and HallandQU (see List of references). The documentation and result of this analysis is available at http://www2.hemsida.net/taxelson/dendro/BrathenEk.zip (WebCite). Two weighted chronologies of correct and corrected blocks are present is available at http://www2.hemsida.net/taxelson/dendro/BraVSQ.dec . In BVSQgt4 all parts with less than four samples depth is removed.

Block 1

Recent: 1615-1975. At least 4 samples depth: 1634-1974, 9 contemporary stems as best. No problems found.

Block 2

Historical: 1337-1804. At least 4 samples depth: 1380-1779, 16 contemporary stems as best. Problems:

  • The period 1554-1613 has almost zero-correlation towards other references, and has to be removed. It is obviously erroneous. The reason for this problem is not yet known.
  • Value for 1650 is missing.

Block 2b

Historical: 1197-1490

Block 2b is very problematic, which is revealed by a block analysis. It is dated by Bråthen to 1490, and the youngest part is obviously correctly dated, but the older part does not correlate at all to other references at its given position! Instead that older part correlates as if the whole 2b-block should be dated to 1197, i.e the samples in the oldest part are erroneously dated 293 years too late! Samples of two very different ages have been mixed together.

When carefully reading the documentation and constructing the samples positions and lengths these mistakes were revealed: Six samples from Götene and Kungslena churches which are from early 12'th century have been mixed into the chronology at a position 293 years too late. As there is only an average series available, only the early and late ends of Block 2b may be used in a new reference as these "clean ends" are based on one curve and are not created as a mean value mix of incorrectly matched chronology members, i.e 904-974 (misdated as 1197-1267) and 1419-1490.

On page 32-33 is a list of 12 samples which are said to be the samples in the average series "Block2b", but that list is not correct. That is obvious when compared to the number of samples said to be behind each value in the average chronology. The correct list could almost be reconstructed from the complementary list at pp. 90-94 and the years when the number of samples per year are altered in the chronology. It is than revealed that one sample from Hedared is forgotten in the first list and that one sample from Götene church is confused i the table.

Id Bråthen's id Description use Correct dating (used part)
Ba1 Bälinge 1 1436-1314 yes yes
Ba2 Bälinge S. remstycke 1482-1313 yes yes
Ba3 Bälinge N. remstycke 1506-1280 1490-1280 yes
Go1 Götene 1 1418-1213 yes 1125-920[1]
Go3 Götene 3 1396-1239 yes 1103-946
Go4 Götene 4 1301-1157 1301-1197 1008-864 (-904)
Go5 Götene 5 1383-1198[2] yes 1090-905
He1 Hedared 1 1487-1338 yes yes
He3 Hedared 3[3] 1426-1268 1426-1302[4] yes
He4 Hedared 4 1455-1327 yes yes
He5 Hedared 5 1488-1346 yes yes
Ku1 Kungslena 1 1401-1268 yes 1108-975[5]
Ku2 Kungslena 2 1393-1227 yes 1100-934

In a book 1995 Bråthen is obviously aware of the correct datings for Götene church (1125) and Kungslena church (1118-1122).[6]

Block 3

Lödöse, archeological: 780-1298 (Misdated: 831-1349). At least 4 samples depth: 867-1185. As best: 19. No other problems found.

Notes

  1. At p.102 Bråthen remarks that there are Earlywood only for the year 1418. According to the missdating this is to be interpreted as summer (june) 1125! He also remarks that Prof. Erik Lundberg, in a paper 1968, dated the church to 12' century. Lundberg was right! (The correct dating (1125) is noted in Bebyggelseregistret)
  2. See p.91 (erroneously 1488-1346 in main table)
  3. Not in list, but according to p.91, and as the youngest year corresponds with an altering of the annual number of samples.
  4. It may also possibly continue backwards in all its length. There are several possibilities which sample is the truncated one: It may be He3 or Go4 (young end) or Ku1 (old end).
  5. The logs from Kungslena church are cut after 1108 and not after 1401.
  6. Alf Bråthen: Dated wood from Gotland and the diocese of Skara, Forlaget Hikuin, 1995, p. 92, ISBN 87-87270-75-7