Ice Age Refuge: Where's the Science?

[Hetware]

I see the Haplogroup to Ice Age refuge mapping repeated on countless web sites and in discussion groups. It is stunningly ironic that I was caustically denounced for having proposed a far more modest scenario than what is currently in vogue. Of course personal attacks are not scientific arguments. I'm interested in the science that supports the proposed models.

From what little I know about the science of the LGM, I have the impression that the exact geographic and chronological boundaries are far from established. Does anybody know what scientific data was used to support the current models depicting the geographic structure of the hypothetical LGM refuges described on the Genographic website?

[M.O'Connor]

Maybe you should take your ideas to the scientific researcher, being the
Genographic Program.

Have you tried to contact the Geno Program or National Geographic with your ideas? Maybe they will accept your ideas?

[Hetware]

Maybe you should take your ideas to the scientific researcher, being the
Genographic Program.

Have you tried to contact the Geno Program or National Geographic with your ideas? Maybe they will accept your ideas?

It's just a matter of time before the validity of my original hunch is established. What I want to know is how they are determining that certain regions were isolated from others during the Ice Age. Are there published findings from geological studies that we can read? Did someone simply look at a map and guess which areas might have been isolated? The sense I have is that we are dealing with the latter.

[M.O'Connor]

nobody can draw a line and say..Y-R1b here and Y-R1A there. I believe the boundries are approximate.

[Rick De Lotto]

Google for ice age sea level maps and everything seems to begin to fall into place. It is fairly obvious, for example (OK, at least to me) that yDNA's founding I1B population was clustered on what is now the floor of the Adriatic. My current belief is that the main refuges were on what is now the deeply-submerged continental shelf, and that the Dinaric (I1b) and Sardinia (I1b2) designations merely reflect the highest ground in the ice free zone—Sea level rose very rapidly at the end of the last glacial period, separating related populations.

There was a lot of room, and the area was the rich, sheltered Po River delta. Recent bathymetric research points out that during the last glacial maximum (LGM) the sea level dropped 120 meters, or over 350 feet, the Adriatic was dry for about a third of its length in the north, with the coastline situated roughly at a line drawn from between Ferrara and Ravenna to Pula, on the Istrian peninsula of current Croatia, just south of the 45th parallel. The Eastern shoreline seems to have been where the barrier islands are now. The gradual retreat upland as sea level rose would have forced a migration to what is now Venice and the mountain ranges north of it. The increased melting MAY have caused the Po to be difficult to cross, as well... hard to say, as the maps I have seen show VERY little detail. Anyone have access to bathymetric maps for the area? Or even good sailing ones? That could give an idea if the gorge was too steep to cross.

It would be interesting to try to find some paleosoil maps from the Po Valley to see what things might have been like.

[EBurgess]

I think it is based on modern haplogroup clinal maps and the geological evidence of ice free areas in europe at the time of the LGM

Sort of like when scientists initially talked about the ice free corridor along the rockies in North America. They presumed that the clovis people came down this route based on some archeology. While this route was very likely used, the latest research in north america now shows evidence for a coastal route that may or may not have been at the same time. Aside from the genetic evidence, there is some archeological:

http://www.usd.edu/esci/alaska/oykc.html

So I think the LGM refuges are just a working hypothesis. Gotta start somewhere

[Hetware]

Atlantis in the Adriatic?

Google for ice age sea level maps and everything seems to begin to fall into place. It is fairly obvious, for example (OK, at least to me) that yDNA's founding I1B population was clustered on what is now the floor of the Adriatic. My current belief is that the main refuges were on what is now the deeply-submerged continental shelf, and that the Dinaric (I1b) and Sardinia (I1b2) designations merely reflect the highest ground in the ice free zone—Sea level rose very rapidly at the end of the last glacial period, separating related populations.

I just recently saw a map showing that chunk of land in the Adriatic that was exposed during the LGM. There are many complications and unresolved questions which need to be considered. Some models hold that there may have been a few very abrupt rises in sea level as the ice retreated. There are many candidate reservoirs for huge inland glacier dammed lakes which may have burst open. These events may have not only caused a "sudden" rise in sea level, they may also have caused dramatic changes in the global climate.

I also have serious reservations regarding the actual time-depth of the Y-chromosome phylogeny. It may merely be homoplasty, but the branches of the various R* haplogroups do not appear to be very far apart in terms of haplotypes.

The Younger Dryas period may also have had significant consequences on the current distribution of related branches. It seems probable there was an expansion into previously uninhabitable regions followed by a return ot those previous conditions some 13 centuries later. For the sake of discussion, let's assume the lands into which the population expanded were likewise repopulated with game animals, and there were no other human populations to compete with. If an initial population of 20 individuals was able to sustain a growth rate of 1% per annum, in 1300 years the result would be 4.4 million individuals.

Upon the abrupt return of colder conditions, that population would likely have found refuges different from the original homeland. Additionally, studies of pollen deposites in various regions show that land that is currently habitable may have been desert during some periods of the Holocene. We can formulate any number of hypotheses based on these kinds of observations.

Anyone have access to bathymetric maps for the area? Or even good sailing ones? That could give an idea if the gorge was too steep to cross.

It would be interesting to try to find some paleosoil maps from the Po Valley to see what things might have been like.

I don't know about the Po valley, but there have been some interesting submarine studies in the Black Sea showing a substantial Danube canyon. There is also evidence for an ancient shoreline in that same data. The depth of the shoreline is not as deep as the shoreline evidence claimed by Ballard. That, however, may not be as hard to explain as it seems. The relative depth discrepancy may be due to unequl isostatic depression of different points along the ancient shoreline.

[Rick De Lotto]

I think (OK, hope) that a lot of these issues are caused by the limited genetic sample, and that All Things Will Be Made Clear when we have more data. I keep having to remind myself how shallow the information actually is right now, how swiftly things are moving, and how hard it is to keep wishful thinking out of the research process.

In many ways this is beginning to look too much like work. <g>

[Hetware]

I think it is based on modern haplogroup clinal maps and the geological evidence of ice free areas in europe at the time of the LGM

That's where my questions begin. Where's the model depicting the probable ice-free areas of Europe? Furthermore, what portions of the ice-free regions were habitable. There has been some research suggesting large areas of desert where there are now forests (or farmland and cities covering potential forests). Additionally, some regions that were once desert are now habitable.

Sort of like when scientists initially talked about the ice free corridor along the rockies in North America. They presumed that the clovis people came down this route based on some archeology. While this route was very likely used, the latest research in north america now shows evidence for a coastal route that may or may not have been at the same time.

This is the kind of thing that makes me want to find a brick wall and slam my head against it! I have more patience the Antediluvian Merovingian mythos of Atlantean Angels than I do with people who mindlessly repeat the latest model dreamt up in some ivory tower as if it were handed down from On High. I'm not suggesting that you are in the latter category. What I'm suggesting is that many parts of the world where the action really took place regarding human populations is now in Davey Jones's Locker.

Aside from the genetic evidence, there is some archeological:

http://www.usd.edu/esci/alaska/oykc.html

Thanks. That looks interesting.

So I think the LGM refuges are just a working hypothesis. Gotta start somewhere

I just found this in a PPT produced by Underhill in 2002:
Caveats: Genetics of Living Populations & History

• Provides only proxy data for actual pre-historic events
• No a priori reason for a 1: 1 correlation between the gene pool and other non-genetic evidence
• Different population histories can generate same genetic landscape
• Earlier demographic episodes may be hidden or replaced by more recent events
• Y ancestry may not always reflect the ancestry of the rest of the genome

[EBurgess]

Caveats: Genetics of Living Populations & History

Provides only proxy data for actual pre-historic events
No a priori reason for a 1: 1 correlation between the gene pool and other non-genetic evidence
Different population histories can generate same genetic landscape
Earlier demographic episodes may be hidden or replaced by more recent events
Y ancestry may not always reflect the ancestry of the rest of the genome

I think (OK, hope) that a lot of these issues are caused by the limited genetic sample, and that All Things Will Be Made Clear when we have more data. I keep having to remind myself how shallow the information actually is right now, how swiftly things are moving, and how hard it is to keep wishful thinking out of the research process.

This is really the meat of the issue. We have very old DNA evidence 20kya, at best 5kya for SNPs. Then we have STRs which frankly are only applicable when combined with other evidence like Surnames, paper trail etc...What does it tell me that I am r1B1c when so are millions of others!

There is a whole middle ground... say the last 2000 years where we don't currently have good ground to stand on.

Thank goodness for people like Victor who have the brains and the interest to look into populations and try to see patterns. I must say that while he may find clusters of STRs in his E3B data, he may not find a corresponding SNP. That is not to say the realtionship isn't there, but then you had better have a significant number of markers and regional demarcation to make your case.

I remain cautious but optimistic. I think we will find groups which show more recent SNIPS, within the last 1000 years or even more recent BUT...Not every group will have a downstream marker. Given that the non-recombining y-DNA is finite in size and that these events are rare. I think the picture that will emerge is that we will find some haplogroups like R1b1C break down even further while others remain undifferentiated. The undifferentiated at the distal end of the branches will be the rule and not the exception.

Finding somebody who is R* may be rare, but finding somebody that is R1B1C* may in fact be common!

The mantra : we need more data!

[Hetware]

Whoops! wrong thread

http://www.google.com/search?hl=en&q=Indo-Greek+Kingdom&btnG=Google+Search

[cacio]

Regarding the existence or absence of refuges, one thing that puzzled me is that I have seldom seen any paper about Italy (despite the fact that Italian scientists seem to be quite active in the field).
As an Italian, I was wondering if there are papers talking about a migration out of a possible refuge in Italy. The papers I have seen talk about the Franco Cantabrian refuge (for R1b, I, mtdna H1 and H3), the Balkan (for some I) or the Ukrainian, but none seems to have uncovered any particular migration out of Italy. (Except of course the nice fiction of the seven daughters of Eve).

Through google I happened across a paper that seems to say that few plants crossed the Alps (ie the plants that recolonized northern Europe do not come from Italy). Were paleolithic Italians equally lazy? And in any case, Italy should then show some special haplogroup distribution, which it doesn't really seem to.

[Mikey]

Cacio,

I too have marvelled at the idea that early humans didn't wait out the cold ice age in sunny Italy. I've been told the flora and fauna there was perhaps not rich enough, and that most of the land (being mountainous or mountain valleys) was also very cold.

But there is some notion that HG I bled across the Adriatic if not during the LGM, very shortly after.

Or that HG I1b2 was present in extreme NW Italy during the LGM.

[cacio]

I think one reason the article was mentioning about the plants is simply
that Italy is surrounded by the Alps, which are obviously difficult to cross for
plants and most animals. Less so for humans though.

I guess part of the story is simply that Italy has been subject to more
invasions and migrations, so its composition is more of a mix, and it
is more difficult to get clean looking results. However, it may simply
be that Italy was inhabited by a mix similar to that of southern France and the Balkans (as per Hetware's theory), but it is more difficult to isolate these now in Italy (except may be in Sardinia, which was relatively isolated).