1 The Impact of NRC Recommendations on Climate Reconstructions U11B-05 AGU December 2006 Meeting Stephen McIntyre Toronto, Canada www.climateaudit.org/pdf/agu06.ppt
Jan 18, 2016
1
The Impact of NRC Recommendations on Climate Reconstructions
U11B-05
AGU December 2006 MeetingStephen McIntyreToronto, Canada
www.climateaudit.org/pdf/agu06.ppt
2
NRC Panel and M&M:
M&M Issue NRC PanelMBH Principal Components Method severely biased
Agreed
Bristlecones are flawed proxies Should be “avoided”
MBH failed verification test said to have been used
Agreed
Could not claim “warmest year” or “warmest decade” based on data and methods
Disallowed confidence claims prior to 1600
3
Key Recommendations and Findings
Precipitation-sensitive proxies e.g. “moisture-sensitive trees and isotopes in tropical ice cores” should not be used without “climatologic justification” of any apparent proxy-temperature correlation;
Strip-bark bristlecone and foxtails should be “avoided”. Many reconstructions are based on the same datasets
and are not independent. Some are not robust with respect to removal of individual proxies.
4
“Plausible”...
NRC panel used 4 reconstructions as comfort for comparing modern-medieval levels, but did verify that cited reconstructions met proxy quality standards. Medieval-modern relationship reverses with trivial and justifiable variations in proxy selection.
5
Most MBH “proxies” are indistinguishable from high-frequency noise. The pattern comes from bristlecones.
-3-2-1012
3
-3-2-10
123
1000 1200 1400 1600 1800 2000
-3-2
-10123
1000 1200 1400 1600 1800 2000
Left – 6 MBH99 proxies. Right: top – contribution of proxy “classes” (proxy type * continent) to MBH reconstruction, bristlecones in red. Bottom – same with each proxy class scaled.
6
Bristlecones and foxtails occupy cold arid niche
Mooney: “bristlecone pine is responding in gross terms primarily to a moisture gradient in the White Mts”. Its “main competitor” is sagebrush.
Lloyd and Graumlich: “previous winter precipitation is the most important factor governing growth variation”
7
Underwater medieval trees in the Sierra Nevadas ...
Major long-term changes in local hydrology; Lakes re-formed only in the Little Ice Age Graumlich: late 20th century “less remarkable for warmth than for high winter precipitation totals”
8
Bristlecone/foxtail chronologies are inconsistent with ecological temperature estimates...
Miller et al 2006: Medieval forests on summits were “typical of forests currently 300-500m lower”; medieval annual temperatures were “3.2 deg C warmer than at present”.
Left: Medieval tree above treeline; right- bristlecone and foxtail ring width chronologies
9
Spaghetti in the Urals
• Briffa’s Yamal chronology inconsistent with other chronologies;
• Inconsistent with ecological estimates that medieval temperatures were 2.5-3.5 deg warmer;
Top: Spaghetti graph showing Briffa 1995 (black), Briffa 2000 (Yamal – red); Polar Urals update (Esper 2002 – blue) Bottom- Treeline at Polar Urals (Shiyatov 1995)
10
One site can affect a reconstruction …
Blue – Briffa 2000 reconstruction using Yamal chronology; yellow – impact using Polar Urals update. Impact on D'Arrigo et al 2006 will be similar since 5 of 6 series overlap.
800 1000 1200 1400 1600 1800 2000
-0.5
0.0
0.5
1.0
1.5
2.0
2.5With Polar Urals UpdateWith Briffa Version
11
δO18 Decreases at Mount Logan and Law Dome
No rational basis for attributing Mt Logan and Law Dome to circulation changes, while citing Guliya and tropical cores as evidence of “unusual” temperature increases
Panel did not discuss Law Dome when stating that no Antarctic sites show medieval warming
12
Spaghetti Guliya
Three different versions of Thompson’s 1992 Guliya δO18 series used in 2006 studies. Reconciliation and quality control is a prerequisite for statistical analysis. Results from all samples need to be archived.
13
Monsoon Proxies used in Temperature Composites
Top: Three monsoon proxies used in temperature composites; right – journal titles and picture of Dulan juniper
14
Arbitrary selection of ocean sediment proxies
G. Bulloides pattern is not characteristic of other high-resolution sediment proxies;
Many high-resolution ocean proxies show distinct MWP, even in Antarctica;
Dating precision is insufficient to prove or disprove synchronization
15
Seeming modern-medieval differential depends on a very few problematic proxies:
1. Bristlecone/foxtail ring width chronologies
2. Selection of Yamal, Siberia ring width chronology in preference to updated Polar Urals;
3. Use of precipitation or even monsoon wind speed proxies in temperature reconstructions – Thompson’s Himalaya ice core isotopes; Arabian Sea cold water G Bulloides; Dulan junipers
16
A Different Spaghetti Recipe
Variations of standard reconstructions using Polar Urals update instead of Yamal and Sargasso Sea SST instead of G Bulloides wind speed proxy and Yakutia instead of problematic bristlecones/foxtails.
17
Conclusions:
There are very few proxies that actually contribute to seeming modern-medieval differential;
These proxies are used over and over; These proxies, especially bristlecones,
do not meet NRC panel recommendations or reconcile with ecological information;
None of the “comfort” studies meets NRC panel recommendations for proxy quality