Mt Almagre and the Divergence Problem

1
Mt Almagre and the Divergence Problem
PP51C-0665
Stephen McIntyre and Peter Holzmann www.climateaud
it.org

• Methods
• Sampled 64 cores (36 trees), of which 38 cores
  (20 trees) at or near Graybill site with other
  cores at lower sites
• Measurement and cross-dating by U of Guelph (J.
  Franks). Thus far 38 cores (23 trees) cross-dated
• WinDendro used for measurement (see example
  below)
• Measurement data and meta-data online within 3
  months of sampling (www.climateaudit.org/data/colo
  rado). ITRDB archive to follow. Meta-data
  includes 5-digit co-ordinates, altitude, aspect
  strip bark, heart rot, compression.
• Comprehensive photo gallery with pictures of all
  trees (picasaweb.google.com/Almagre.Bristlecones.2
  007) and panoramas (www.gigapan.org keyword
  Almagre).
• Analysis is in progress as measurement data
  received only recently and more cross-dating
  appears possible.
• Average ring width is only 0.41 mm
• Strip Bark
• Strip bark forms said to be problematic (North et
  al, 2006)
• Major differences between individual cores in
  strip bark trees
• Not obvious how to model error structure
  statistically
• Lloyd (pers. comm.) confirmed strip bark at
  similar foxtail core.
• (see UWL132 at right)
• Chronologies

• Mt Almagre, Colorado
• Updating the Mt Almagre bristlecone pine
  chronology offered an opportunity to test several
  hypotheses
• It was a classic Lamarche and Graybill site that
  had not been updated since 1983
• it is a treeline site. Indeed it was the highest
  (3,600 m) chronology at ITRDB going back to
  AD1000
• It was a bristlecone site, used as an individual
  proxy in Crowley and Lowery 2000 and as a
  component of Manns PC1 in multiple studies
• It was very close to Colorado Springs (and
  Boulder)
• No location map but believed to be near a Forest
  Service road and accessible by 4-wheel drive
• It was a southern chronology in Cook et al
  (2004) terms and could test the hypothesis that
  divergence was limited to northern sites.

• Discussion
• Almagre is a bristlecone treeline site that has
  been used as a temperature sensitive site in
  reconstructions
• on geographic grounds, it should be more
  temperature limited than California bristlecone
  sites it is about 90 miles north and 125 meters
  higher than the Sheep Mountain CA bristlecone
  site (about 120 miles north and 80 meters higher
  than the Upper Wright CA foxtail site)
• Under the hypothesis (Mann et al 1998, 1999,
  2003, 2007) that bristlecones are teleconnected
  to global climate fields, Almagre ring widths
  would be expected to increase in the warm 1990s
  and 2000s.
• Instead, Almagre has the divergence problem
  characteristic of northern sites ring widths
  have declined in the 1990s and 2000s from high
  levels in the late 19th and mid-20th century.
• The site has an extremely arid appearance and it
  appears likely that growth is moisture limited
  even at treeline
• Growth suppression in the 1840s is very
  distinctive. Woodhouse et al (BAMS 2002)
  reported a mega-drought from 1845-56 in eastern
  Colorado and the Colorado Front Range and the
  growth suppression at Almagre may be related.
  Growth suppression in the 1840s is characteristic
  of other Colorado Front Range bristlecone sites
  and can even be perceived in California
  bristlecone chronologies.
• Developing a proper statistical model for strip
  bark presents formidable difficulties since the
  growth pulses appear to exceed climatic effects
  and combine extreme non-normality and
  autocorrelation. Under the circumstances, it was
  encouraging that we were able to obtain similar
  results to Graybill from a different sample.
  However, the effect of cross-dating selection has
  not yet been evaluated.
• The existence of unarchived Graybill measurement
  data at the University of Arizona was identified.
  Archiving of this data is encouraged.

Modern portion of Almagre Tree 30 showing narrow
recent ring widths
Panorama of Graybill site on Mt Almagre
Identifying the Graybill Site Individual trees
were identified from online photos and tags. We
located 16 tagged trees of which 8 have been
sampled. We reconciled the tags to the ITRDB
archive (thanks to C. Hallman). Only 3 of the 8
sampled trees had been archived. It appears that
Graybill sampled 42 trees, of which only 21 are
archived, although additional measurement data
exists in Tucson.
Two cores from Upper Wright foxtailLloyds field
notes confirmed strip bark.
Two cores from Almagre strip bark tree 31Black
middle of strip red towards edge
Further Information Contact Stephen McIntyre
www.climateaudit.org Email stephen.mcintyre_at_utor
onto.ca
Acknowledgements U.S. Forest Service Jenny
Franks of the University of Guelph Leslie
Holzmann, Nola McIntyre, Leslie Thomas, Tori
Bommarito, Rebecca Lee.
Combined
chronology without power transformation

Combined chronology with power
transformation of 0.375