Stephen McIntyre has just lately once more fired up the seminal web site for uncovering deficiencies within the works of Mann et al, ClimateAudit.org
His newest put up ends a 25 yr thriller surrounding the well-known MBH98 paper. A Swedish engineer, Hampus Soderqvist, reversed engineered the reconstruction and deduced that:
Mann’s checklist of proxies for AD1400 and different early steps was partly incorrect (Nature hyperlink now useless – however see NOAA or right here). Mann’s AD1400 checklist included 4 collection that weren’t truly used (two French tree ring collection and two Moroccan tree ring collection), whereas it omitted 4 collection that have been truly used. This additionally utilized to his AD1450 and AD1500 steps. Mann additionally used an AD1650 step that was not reported.
Soderqvist’s discovery has an essential utility.
The well-known MBH98 reconstruction was a splice of 11 completely different stepwise reconstructions with steps starting from AD1400 to AD1820. The proxy community within the AD1400 step (after principal elements) consisted 22 collection, growing to 112 collection (after principal elements) within the AD1820 step. Mann reported a number of statistics for the person steps, however, as mentioned again and again, withheld the essential verification r2 statistic. By withholding the outcomes of the person steps, Mann made it unimaginable for anybody to hold out routine statistical exams on his well-known reconstruction.
Nevertheless, by reverse engineering of the particular content material of every community, Soderqvist was additionally in a position to calculate every step of the reconstruction – precisely matching every subset within the spliced reconstruction. Soderqvist positioned his outcomes on-line at his github web site a few days in the past and I’ve collated the outcomes and positioned them on-line right here as properly. Thus, after virtually 25 years, the outcomes of the person MBH98 steps are lastly out there.
Remarkably, Soderqvist’s discovery of the particular composition of the AD1400 (and different early networks) sheds new mild on the controversy about principal elements that animated Mann’s earliest realclimate articles – on December 4, 2004 as realclimate was unveiled. Each articles have been assaults on us (McIntyre and McKitrick) whereas our GRL submission was below overview and whereas Mann was searching for to dam publication. Soderqvist’s work exhibits that a few of Mann’s most vehement claims have been unfaithful, however, oddly, unfaithful in a method that was arguably unhelpful to the argument that he was making an attempt to make. It’s fairly bizarre.
Soderqvist is a Swedish engineer, who, as @detgodehab, found a exceptional and deadly flaw within the “signal-free” tree ring methodology utilized in PAGES2K (see X right here). Soderqvist had figured this out a few years in the past. However I used to be unaware of this till just a few days in the past when Soderqvist talked about it in feedback on a latest weblog article on MBH98 residuals.
https://climateaudit.org/2023/11/24/mbh98-new-light-on-the-real-data/
The put up is a protracted and technical one to which I can’t do correct justice, and I counsel studying the unique at Local weather Audit
