PCAB correlations towards PICE - Different patterns

Post Reply
taxelson
Site Admin
Posts: 19
Joined: 28 Dec 2008, 18:39
Real name: Torbjörn Axelson
Location: Björbo, Dalarna, Sweden
Contact:

PCAB correlations towards PICE - Different patterns

Post by taxelson »

Picea abies (PCAB) and Pinus cembra (PICE) – Is it possible to date samples from one of those species towards a chronology of the other?

To answer this question I selected from ITRDB Pola020 (PCAB, 1500m, 49°15'N 19°55'E by F. Schweingruber) and Pola300 (PICE 1550m 49°14'N 20°00'E by Bednarz). They are from almost the same location in the Beskidy mountains near the Slovakian border, and rather near the maximum elevation present in the actual area. I did also select aust007 (PCAB 1850m 47°08'N 11°17'E by F. Schweingruber) and aust002 (PICE ?m 46°52'N 11°01'E by Giertz), located in the Austrian Alps, not far from the Italian border. The distance between those is about 50 km. There are alps higher than 3000 meters in the actual area, so the forest is present only in the valleys.

The correlation between the Austrian collections is high (corr=0.59, Prop2yrs) and 6 of 24 samples (25%) from the Austrian PICE-collection (the shorter) have corr >=0.50 towards the PCAB-collection, and 14 of 24 (58%) have corr >=0.40. It means that many samples in this area could be expected to be datable towards a reference of the other specie. The highest value for a single sample is as high as 0.60 over 195 years.

The correlation between the Polish collections is in contrast very low. Corr=0.32. This means that not even a floating chronology which contains many samples could surely be dated towards a chronology of the other specie in this area.

The difference in elevation is not very big, and may not itself explain the difference in correlation between the species at the two sites, but perhaps the high mountains in the alpine area have an impact according to temperature and/or precipitation, which force the alpine chronologies congruating?

I may have removed samples from the original chronologies if they looked erroneous. I have also removed the oldest part of the meansample chronologies as the sample depth is only one.


Below are block correlation analysis for the two sites:

Code: Select all

[size=85]
pola020 compared to the reference pola300.
Best matches for the whole sample:  -123 0,33 (2092)      5 0,32 (1964)     98 0,26 (1871)  
The sample is currently dated to 1964 which is used in the "Aimed at" column.
Block length: 50   Table sorted by Corr coeff.

Block  -----Aimed------   -------Best    ------------Three best matches with {hitAt,Corr coeff.,SetsSampleTo}
start  --------at  year   around that    ---1stBestMatch-------    ---2ndBestMatch-------    ---3rdBestMatch-------
    0      5 0,37  1964       5 0,37         5 0,37    5 (1964)       98 0,37   98 (1871)       87 0,36   87 (1882)
   10     15 0,39  1954      15 0,39        15 0,39    5 (1964)      127 0,37  117 (1852)      157 0,36  147 (1822)
   20     25 0,39  1944      25 0,39        25 0,39    5 (1964)      167 0,35  147 (1822)      107 0,34   87 (1882)
   30     35 0,41  1934      35 0,41        67 0,43   37 (1932)       35 0,41    5 (1964)      142 0,37  112 (1857)
   40     45 0,47  1924      45 0,47        45 0,47    5 (1964)      132 0,40   92 (1877)       77 0,35   37 (1932)
   50     55 0,42  1914      55 0,42       162 0,50  112 (1857)       55 0,42    5 (1964)       48 0,34   -2 (1971)
   60     65 0,17  1904      65 0,17       172 0,44  112 (1857)       80 0,43   20 (1949)       24 0,37  -36 (2005)
   70     75 0,02  1894      77 0,11*       90 0,47   20 (1949)       13 0,44  -57 (2026)       83 0,41   13 (1956)
   80     85-0,02  1884      84 0,14*      100 0,48   20 (1949)       23 0,41  -57 (2026)      141 0,32   61 (1908)
   90     95 0,02  1874      94 0,16*      110 0,43   20 (1949)       33 0,40  -57 (2026)       99 0,35    9 (1960)
  100    105 0,29  1864     105 0,29        85 0,34  -15 (1984)        3 0,33  -97 (2066)      120 0,30   20 (1949)
  110    115 0,35  1854     115 0,35       144 0,35   34 (1935)      115 0,35    5 (1964)       64 0,30  -46 (2015)
  120    125 0,41  1844     125 0,41       125 0,41    5 (1964)       74 0,39  -46 (2015)      102 0,37  -18 (1987) 

____________________________________________________________________________


aust002 compared to the reference aust007
Best matches for the whole sample:     4 0,59 (1971)   -370 0,42 (2345)   -362 0,35 (2337)  
The sample is currently dated to 1971 which is used in the "Aimed at" column.
Block length: 50   Table sorted by Corr coeff.

Block  -----Aimed------   -------Best    ------------Three best matches with {hitAt,Corr coeff.,SetsSampleTo}
start  --------at  year   around that    ---1stBestMatch-------    ---2ndBestMatch-------    ---3rdBestMatch-------
    0      4 0,58  1971       4 0,58         4 0,58    4 (1971)       54 0,41   54 (1921)       62 0,34   62 (1913)
   10     14 0,67  1961      14 0,67        14 0,67    4 (1971)       72 0,42   62 (1913)       76 0,37   66 (1909)
   20     24 0,65  1951      24 0,65        24 0,65    4 (1971)        4 0,50  -16 (1991)       86 0,47   66 (1909)
   30     34 0,63  1941      34 0,63        34 0,63    4 (1971)       38 0,43    8 (1967)       68 0,41   38 (1937)
   40     44 0,72  1931      44 0,72        44 0,72    4 (1971)      119 0,43   79 (1896)       48 0,43    8 (1967)
   50     54 0,68  1921      54 0,68        54 0,68    4 (1971)       58 0,43    8 (1967)      129 0,38   79 (1896)
   60     64 0,57  1911      64 0,57        64 0,57    4 (1971)       52 0,33   -8 (1983)       72 0,29   12 (1963)
   70     74 0,56  1901      74 0,56        74 0,56    4 (1971)       16 0,41  -54 (2029)      102 0,33   32 (1943)
   80     84 0,63  1891      84 0,63        84 0,63    4 (1971)        9 0,55  -71 (2046)       26 0,41  -54 (2029)
   90     94 0,59  1881      94 0,59        94 0,59    4 (1971)       19 0,46  -71 (2046)       36 0,35  -54 (2029)
  100    104 0,57  1871     104 0,57       104 0,57    4 (1971)       46 0,44  -54 (2029)        7 0,43  -93 (2068)
  110    114 0,59  1861     114 0,59       114 0,59    4 (1971)       39 0,49  -71 (2046)        9 0,48 -101 (2076)
  120    124 0,64  1851     124 0,64       124 0,64    4 (1971)       27 0,41  -93 (2068)       19 0,37 -101 (2076)
  130    134 0,63  1841     134 0,63       134 0,63    4 (1971)        1 0,42 -129 (2104)       57 0,31  -73 (2048)
  140    144 0,59  1831     144 0,59       144 0,59    4 (1971)        0 0,39 -140 (2115)       11 0,36 -129 (2104)
  150    154 0,57  1821                     13 0,32 -137 (2112)       24 0,31 -126 (2101)      138 0,30  -12 (1987)
  160    164 0,55  1811                     50 0,41 -110 (2085)        3 0,34 -157 (2132)      144 0,30  -16 (1991) 

________________________________
Cybis CDendro, Algorithm: Proportion of last two years growth (2,0,T)
Correlations between available references in aust002.fil and
aust007 dated to 1975 with corr >= 0,16 and with overlap >= 171
Results sorted according to decreasing correlation coefficient values.
                  T-  Over
         Corr   Test   lap
all...   0,58   9,98   197   based on 24 members
109451   0,60  10,46   195  (1969)   
109431   0,57   9,61   195  (1969)   
109221   0,57   9,64   197  (1971)   
109521   0,52   8,48   197  (1971)   
109411   0,52   8,43   195  (1969)   
109441   0,50   8,02   195  (1969)   
109261   0,49   7,79   197  (1971)   
109321   0,47   7,35   197  (1971)   
109181   0,46   7,14   197  (1971)   
109131   0,45   6,96   197  (1971)   
109241   0,42   6,47   197  (1971)   
109281   0,41   6,34   197  (1971)   
109541   0,41   6,26   197  (1971)   
109011   0,40   6,13   197  (1971)   
109481   0,39   5,76   182  (1956)   
109151   0,35   4,87   171  (1971)   
109471   0,34   5,10   195  (1969)   
109121   0,33   4,89   197  (1971)   
109531   0,33   4,81   197  (1971)   
109041   0,31   4,62   197  (1971)   
109461   0,27   3,95   195  (1969)   
109081   0,27   3,96   197  (1971)   
109301   0,19   2,70   197  (1971)   
109101   0,16   2,26   197  (1971)   


[/size]
Last edited by taxelson on 14 Jan 2009, 17:07, edited 1 time in total.
Reason: +authors & lang. correction
Post Reply

Who is online

Users browsing this forum: No registered users and 4 guests