Holliday et al.'s Gish gallop: Platinum
Holliday et al.'s "comprehensive refutation" of the Younger Dryas impact hypothesis (YDIH) is a highly misleading Gish gallop.
An earlier blog post addressed the presence of impact melts and microspherules in the YDB. By themselves, they confirm a cosmic impact at Abu Hureyra near the YD onset. Due to the similarity of debris at the YDB across a wide range of sites, it's likely there were cosmic impacts across a wide area within a short timespan, indicating a singular event.
Another blog post examined the presence of nanodiamonds in the YDB. As they occur in the debris layer with microspherules at Abu Hureyra, they are also excellent impact indicators. Kinzie et al. (2014) showed they are likely present in many other YDB layers, along with apparent impact spherules, signalling a widely distributed cosmic impact, circa 10,800-10,900 BCE.
The remaining topics to be discussed are, therefore, the presence of platinum at the YDB, and the apparently synchronous timing of this evidence across ~ 50 sites on several continents. This blog post, therefore, focusses on the platinum evidence. This corresponds to section 11 of Holliday et al..
Following discovery of the Greenland platinum signal by Petaev et al. (2013), which is expected to be global given its ~20 year lifetime in the atmosphere, Moore et al. (2017) set out to find a corresponding platinum abundance in the YDB at many sites. Their main result is shown below;
The horizontal blue band is the expected position of the YDB based either on direct radiocarbon dating or the presence of proxies, specifically Clovis points. The dots joined by lines indicate platinum measurements (Moore et al., 2017). The symbols on the right show the location of microspherules and/or nanodiamonds. The correlation with a platinum signal is excellent. Platinum anomalies are also apparent at other levels at Arlington Canyon.
Moore et al. (2017) tested several other sites as shown above. Again, the agreement is excellent, albeit somewhat noisy at some sites due to measurement limitations. Since then, platinum and other PGE abundances have been detected at many more sites consistent with the YDB, including at Abu Hureyra. Given the microspherule evidence alone confirms a cosmic impact at Abu Hureyra, it seem likely that the platinum signal in Greenland and at the YDB is indicating the same event. Powell (2022) summarizes these findings.
11. Purported YDIH evidence of impact: Platinum Group Elements
Following the study of Greenland ice cores by Petaev et al. (2013a), YDIH proponents
focused on platinum (Pt) and added platinum group elements (PGEs) as YDB impact indicators
(Andronikov et al., 2016a; Mahaney et al., 2017; Moore et al., 2017; 2019; 2020; Wolbach et al.,
2018b; Pino et al., 2019; Thackeray et al., 2019; Teller et al., 2020). However, at least two
principal aspects of the Pt anomaly in Greenland ice were misinterpreted by the YDIH
proponents.
Let's see.
The first mistake is using the Pt spike alone as an indicator of ET matter.
We already have highly correlated abundances of microspherules, other melts and nanodiamonds in the YDB. We therefore expect to also find enrichment of PGE elements in the YDB layer, although we cannot know in advance their relative abundances.
As stressed
by Jaret and Harris (2021, p. 2), “To convincingly show evidence of meteoritic components, the
full suite of PGEs should match known meteorite groups.” Petaev et al. (2013a) explicitly stated
that while the Pt spike in ice requires an injection of Pt-rich matter, it does not identify the nature
of that matter because both ET and crustal Pt-rich materials exist, but they have quite different
Pt/Ir and Pt/Al ratios.
The composition of comets is relatively uncertain. So it is possible that the PGE composition might not match any known meteorite groups which are mainly thought to originate from asteroids. However, in general the basic composition of comets is thought to be similar to primitive chondritic (stony) meteorites, plus lots of added volatile ices.
It is the extremely high Pt/Ir ratio at the Pt anomaly in the Greenland ice
that “rules out mantle or chondritic sources of the Pt anomaly (Fig. 2). A further discrimination
between Pt-rich crustal materials like Sudbury Footwall ore… and fractionated extraterrestrial
sources such as Ir-poor iron meteorites… is difficult because of the comparable magnitude of
the Pt/Ir fractionation in these materials. Circumstantial evidence hints at an extraterrestrial
source of Pt, such as very high, superchondritic Pt/Al ratios at the Pt anomaly and its timing,
which is clearly different from other major events recorded in the GISP2 ice core, including
well-understood sulfate spikes caused by volcanic activity and the ammonium and nitrate spikes
associated with biomass destruction” (Petaev et al., 2013a, p 12918).
Indeed, this all points towards an ET source, as expected. Note that Petaev et al. (2013) also state that "both terrestrial and extraterrestrial high-Pt sources have substantially lower Pt/Ir ratios than those at the top of the Pt peak, implying either Pt-Ir fractionation during atmospheric processing of the Pt-rich materials or multiple injections of materials with different Pt/Ir ratios not sampled so far". It is very notable that Holliday et al. ignore this inconvenient statement, especially when Petaev is a co-author of Holliday et al..
In other words, the PGE composition detected is not a sure test for the origin of the (ET) matter because i) impact processes (and presumably many other environmental processes) can modify the composition detected, and ii) the impactor might be a highly unusual ET object. Nevertheless, the observed platinum abundance is consistent with an ET impact.
The later investigation of
YDB sediments at Hall’s Cave and Friedkin sites by Sun et al. (2021, p 70) showed that highly
siderophile element “analysis including Os isotope measurement is needed to provide a clear
picture of the source of the geochemical signatures as either being extraterrestrial or mantle derived
material.”
Sun et al. (2021) did not "show" this. It is simply their assumption or hypothesis. As already stated, we cannot know the PGE composition of a specific comet in advance, although it is generally thought they are likely to be similar to primitive chondritic meteorites plus added volatile ices. Petaev et al. (2013) have already indicated that impact processes can modify that apparent composition making identification of the source uncertain. That is, we already see the GSIP2 platinum signal is very enigmatic. So, we can expect the YDB PGE signal elsewhere to be enigmatic too. In particular, if atmospheric fractionation resulted in platinum enrichment relative to iridium in the GISP2 ice core, as suggested by Petaev et al. (2013), then it is also possible it resulted in enrichment of platinum relative to osmium. Therefore, analysis of osmium in sediment layers at YDB sites might not be very helpful since it might be a very weak signal relative to platinum and quite unlike that expected for normal chondritic impactors. This agrees with the observation of Wu et al. (2013) who failed to find a clear osmium signal at several YDB sites.
It is worth noting the normal boiling points of the six PGEs (in Celsius, from Wikipedia): Palladium: 2963 ; Rhodium: 3695 ; Platinum: 3825 ; Ruthenium: 4150 ; Iridium: 4130 ; Osmium: 5012. If the impact fireball was sufficiently hot to vaporise a lot of the Platinum, but insufficient to vaporise so much Iridium and Osmium, then we would expect to observe fractionation leading to depletion of Iridium and especially Osmium in atmospheric particles (aerosols) relative to platinum. This might then explain the relative PGE abundances seen by Petaev et al., Wu et al. and Sun et al..
Another aspect is the magnitude and duration of the Pt anomaly with the maximum Pt
concentration of 82.2 ppt and Pt/Ir ratio of 1265 in an ice layer of 12.5 cm in thickness
precipitated over ~ 3.5-year period. As the injection event has likely lasted much less than 3.5
years, the peak concentrations of Pt and Ir at the anomaly could be even higher due to the
dilution effect of Pt-free ice accumulated before and after the injection in the sample analyzed,
but the Pt/Ir and Pt/Al ratios should remain the same.
Agreed.
The situation is different for YDB
sediments where a much longer accumulation time (hundreds of years) expected for a ~1 cm thick
layer of sediments.
Agreed.
The longer accumulation time allows for any PGE spike in the sediment
(corresponding to the ice core spike) to be diluted and the Pt/Ir altered by minerals from various
sources deposited in the sediments.
Agreed. And as already stated, impact processes might have enhanced the platinum abundance relative to other PGE abundances in aerosols, making detection of iridium and osmium very difficult. Thus, if osmium has been fractionated like iridium in the GISP2 ice core, it might not a be a useful indicator of ET impact at all. Nevertheless, Moore et al. (2017) are able to detect the expected very low platinum concentration spikes in the YDB at many locations.
For example, Os isotopes and PGE data of Sun et al. (2020;
2021) for sediments below, above and within the YDB layer from the Hall’s Cave and the Debra
L. Friedkin site do show several Pt spikes, with one sample (BMC16_11.D – Sun et al., 2021)
having very high Pt/Ir ratio of 1937 and very low Pt/Lu ratio of ~0.0007 due to the dominance of
terrestrial silicate matter in sediments. Based on the Pt/Ir ratio alone, the nature of this Pt-rich
and Ir-poor material cannot be resolved. It is the dominance of silicate matter in sediments that
rules out usage of the Pt anomaly alone or even with Ir as a proxy of ET matter in them.
Sweatman (2021, p 2) describes PGEs (“especially platinum itself”) as “the most robust impact
proxies“ but clearly this is not the case here unless a comprehensive analysis of PGEs and
siderophile elements is performed. For example, a volcanic source of PGE anomalies at Hall’s
Cave and the Friedkin site was deduced based on a wider examination of 187Os/188Os isotopic
ratios as well as abundances of Os, Ir, Ru, Pt, Pd, and Re (Sun et al., 2020, 2021).
Sun et al. (2020,2021) is focussed on osmium, rather than platinum. And as already stated, osmium, if it is fractionated like iridium in the GISP2 ice core, might not be a useful guide to the origin of the platinum abundance.
In fact, Sun et al.'s data suggests that there could be separate osmium-rich events (which they conclude are volcanic in origin) and platinum-rich events (which could be signalling an ET event) including a platinum-rich event at the YDB consistent with the YDIH. On the other hand, it is also possible that the platinum and osmium signals they find could actually be coincident, and their apparent separation in time is caused by taking different samples from different laterally-separated sediment columns with respect to an undulating sediment horizon without compensating for this effect. Certainly, there is nothing in Sun et al. (2020) which refutes the YDIH.
Sweatman (2021, p 8) falsely claimed that Holliday et al. (2014) stated that in the
Greenland ice “the platinum anomaly is around 30 years too late.” No such statement was made
by Holliday et al. (2014). Perhaps Sweatman misread or misunderstood the Holliday et al. (2014)
reiteration of the Petaev et al. (2013a) finding that the Pt anomaly “precedes an ammonium and
nitrate spike in the core by ~30 years.” In this context, “precedes” would imply “too early.”
This is wordplay. By "too late" Sweatman (2021) meant to old.
However, the Pt anomaly appears in the ice core about 1 meter above the YDB, which
corresponds to about 20 years (Section 5.1).
This is misleading. Here, Holliday et al. use the definition of the YD onset as determined by some paleo-climatologists to define the YDB, which is contentious. See an earlier blog post about this issue. However, since Holliday et al. deny the impact took place, they cannot also define the YDB, since according to their position there is no YDB layer. So their position on the YDB is internally inconsistent.
In fact, this definition for the YD onset is based on a larger than normal jump in dD in the NGRIP ice core only. It ignores all other kinds of climate signal in all other ice cores. YDIH proponents, on the other hand, define the YDB according to the location of the YD impact debris - like nanodiamonds, microspherules and platinum. In this case, of course, there is no discrepancy in the position of the YDB relative to the platinum anomaly, because they are the same thing.
The ammonium and nitrate spike appears in the ice
core even higher corresponding to about 50 years later than the YDB.
Once again, this is misleading for two reasons. First, because of how they define the YDB - see above. And second because of how they define the position of the "ammonium and nitrate spike". In fact there is no "ammonium and nitrate spike". The "ammonium ion" spike plotted in Petaev et al (2013) is actually an artefact and misrepresents the data on which it based from Mayewski et al. (1993), as described in Sweatman (2021). The correct data is shown in Sweatman (2021), also provided below, which Holliday et al. ignore.
The inset plot shows the timing of the GISP2 platinum signal versus the oxygen isotope signal in the GISP2 ice core. Underneath is the GISP2 ammonium ion signal in Mayewski et al. (1993), which is misleadingly plotted at a single spike at its centre by Petaev et al. (2013). All this data is from the same ice core, and therefore cannot be misaligned. Clearly, all these signals are coincident within the resolution of the data. This was emphasized by Sweatman (2021).
However, higher resolution data for the ammonium anomaly is given in Wolbach et al (2018), shown above. This shows the platinum spike and sudden drop in temperature (signalled by the oxygen isotope trace) coincide with a steadily increasing ammonium signal in the GISP2 ice core. The ammonium signal is very noisy, indicating ammonium is brought to Greenland by severe storms from elsewhere, i.e. likely indicating extensive wildfires at the onset of the YD, with ammonium-enriched dust brought to Greenland's ice by stormy weather over hundreds of years. Holliday et al. have ignored this data too.
This suggests three
independent events with three different causes.
This is nonsense. The beginning of the YD ammonium ion 'bump' coincides with the platinum signal in the GISP2 ice core, as shown above. How the YD onset is defined is a matter of contention and open to interpretation. So a coincident event remains a possibility.
Sweatman (2021, p 20) also falsely claims that Holliday et al. (2014) misrepresented
Petaev et al.’s (2013b) conclusions. Holliday et al. (2014, p 522) states, “In response to Boslough
(2013) [commentary of Petaev et al. (2013a)], Petaev et al. (2013b) accept arguments against the
Pt- depositing event being the cause of the YD cooling.” The first sentence of Petaev et al.
(2013b) is, “Besides providing additional arguments against the Pt depositing event as a cause of
the Younger Dryas cooling, Boslough’s letter raises an important question about the scale of this
event.”
Clearly, with Petaev a co-author of Holliday et al., Sweatman's (2021) interpretation of Petaev's comments was incorrect. But this was not clear at the time.
Sweatman (2021) and Powell (2022) highlight the work of Moore et al. (2017).
Sweatman (2021, p 5) states that those investigators “reported the discovery of a widespread
platinum anomaly at the base of the YD black mat in several locations in North America.” As
discussed (Section 5.4), the timing and context of Pt deposition is far from clear. The work
reported by Moore et al. (2017), like most other YDIH studies, does not meet the dating criteria
set out by Kennett et al. (2008a) (Table 4).
Maybe not, but we must deal with the evidence as it is found rather than as we would like it to be. The dating evidence is dealt with separately. As stated before, no YDB site is clearly inconsistent with the timing of the YDI.
The Pt zone is assumed to be the YDB because it has
Pt (circular reasoning) and is stratigraphically about where the YDB should be.
This is false. The YDB at each location is defined according to whatever YDB proxies are present - these are either other impact indicators, like nanodiamonds and microspherules, or the upper level of Clovis artefacts etc. As shown above, the platinum is found exactly where it is expected. The agreement is superb.
The accurate and
precise dating required to identify the YDB does not allow for dates that are “about” right,
however.
This is misleading and dealt with separately.
Further, out of 11 sites with Pt reported by Moore et al. (2017), only four have black
mats (and Arlington Canyon has multiple zones dating to the YDB over a section 5 meters thick)
(Table 4).
This is misleading. No YDIH proponent has claimed the YD black mat is perfectly continuous. Likewise, the YDB is not expected to be continuous. Both will be sensitive to local conditions.
At the archaeological site of Abu Hureyra, Sweatman (2021, p 5) note that Moore et al.
(2020) “analysed debris from the burned layer… at Abu Hureyra, Syria, dated to 12,825 ± 55 cal
BP, finding elevated platinum at 6.2 ppb.” There are other (younger) “burned layers” at Abu
Hureyra (Section 4.2) and other sites in the region. Were they sampled, too?
Are Holliday et al. suggesting that every burned layer at every site in the region must be sampled in order to make any conclusions about the provenance of platinum in YDB at Abu Hureyra? How many sites must be investigated? How far away? Their demands are unreasonable. Microspherule evidence from the YDB "burned layer" layer investigated at Abu Hureyra already confirms a cosmic impact. See an earlier blog post on this issue. A platinum abundance in the same burned layer clearly supports this finding. Again, Holliday et al. provide no refutation arguments.
The Usselo soil
(Section 5.6) in northwest Europe was sampled for Pt. Sweatman (2021, p 5) refers to “elevated
levels of platinum, other PGEs and REEs (rare-Earth elements)” citing the work of Andronikov
et al. (2016a) but fails to mention that none of the samples in that study are directly linked to
YDB dating, and some of the Pt spikes are above or below the soil. Further, the Pt spike
reported in Andronikov et al. (2016a) is not remarkable and that study concluded that their
evidence for an ET impact is equivocal (Table 4).
Nevertheless, it is not inconsistent with production by the YDIH.
Finally, the link between elevated Pt and the beginning of the YD/GS-1 is far from clear,
contrary to statements by Moore et al. (2017), Sweatman (2021) and Powell (2022). Cheng et al.
(2020) (published over three months before Sweatman’s paper) present “speleothem oxygenisotope
data that, in concert with other proxy records, allow us to quantify the timing of the YD
onset and termination at an unprecedented subcentennial temporal precision” (abstract). Their
work includes identification of the YDB and examination of the Pt record in the Greenland Ice
Sheet. Their observations and conclusions on this particular issue (p 23414) is worth quoting in
full because they are directly germane to both the onset of the YD/GS-1 and the significance of
Pt in the YDIH debate:
“an ∼20-y-long Pt-anomaly [highlighted by Sweatman 2021, p 4,5,14,17] was
identified in the Greenland Ice Sheet Project (GISP2) ice core…, which was attributed to
injections of Pt-rich dust from the [YDB impact] event and subsequent deposition at a
depth of 1,712.375 to 1,712.000 m, or at ∼12,820 B.P., based on synchronization to the
GICC05 chronology…. A closer look, however, found that the immediate hydroclimatic
impact, if any, was likely minor as inferred from GISP2 δ18O record (corresponding to a
<1‰ drop…). In the same ice core, the Pt-anomaly occurred at the middle of a gradual
increase in Ca2+ (dust proxy) from ∼1,714.00 to 1,709.90 m (∼12,870 to 12,765 B.P. on
GICC05 chronology) without disrupting the course… Provided that the GISP2 and
NGRIP records were synchronized precisely…, the Pt-anomaly did not disrupt NGRIP
and AM [Asian Monsoon] δ18O records either… Additionally, there is no clear evidence
that the YD-onset excursion has been interrupted substantially around the time of the Ptanomaly,
either in the South American Monsoon or in tropical records… These
observations are thus inconsistent with the hypothesis that the extraterrestrial event
triggered the YD unless the extraterrestrial event did not leave any imprints in the
Greenland ice core, which would be also inconceivable. Moreover, the YD as a
millennial-scale perturbation during the last deglaciation has a previous analog: a YDlike
event occurred at ∼245,000 B.P. during glacial termination-III (the third to the last
deglaciation)... Based on this paleoanalog and the preponderance of geochronological
data, we contend that the YD Impact Hypothesis remains untenable and offers a less
parsimonious explanation for the global timing and structure of the YD event, and the
data presented here provide a precise timing framework for further research in the area
[emphasis added].”
The work of Cheng et al. (2020) has already been dealt with in an earlier blog post. Essentially, their conclusions are unsupportable because they misunderstand the signals expected by the YD impact, i.e. speleothem data does not have the fine resolution required to detect a cosmic impact winter.
Summary
Once again, Holliday et al. have not provided any refutation arguments against the platinum abundance at the YDB in many locations as evidence in favour of the YDIH. Indeed, to make their Gish gallop, they resort to many blatantly false, misleading or inconsistent statements, wordplay, and a methodology that ignores the usual scientific dogma of Occam's razor. They make much of the apparently unusual osmium abundance at the YDB, but this might be explained by fractionation by the impact fireball, since osmium has a much higher normal boiling point than platinum.
Comments
Post a Comment