This appears not to be clear to some, it seems to me from the discussion in the first post of this mini-series. So I'll try to clarify with an image, a version of the previous graph(s):
Here I have marked in green the most obvious rapidly successive branchings in the tree (there may be others in the lineages not considered here as well). What do we see?
1. The tightest of all of this rapidly branching sequences is the K-NOP bloc. Logically, as they are only separated by one, recently discovered, SNP. This means that these two nodes happened within a short period (here estimated as 1000 years: c. 66-65,000 BP).
2. This K-NOP "supernode" can be included within another: F-IJK-K-NOP. Nevertheless this one is maybe the less tight of all marked sets (this supernode is estimated to last some 8,000 years instead: c. 73-65,000 BP).
3. Within haplogroup E, we see two "supernodes":
3.1. E-E1: which means that E2, E1a and E1b split in a short period (est. 2000 years: c. 47-45,000 BP).
3.2. E1-E1b: which means that E1b1a and E1b1b1 and E1b1b2 a split in about the same epysode (est. 3000 years: c. 33-30,000 BP).
4. Within haplogroup I we can easily appreciate that I1, I2a and I2b split also in a rapid sequence (est. 3000 years: c. 40-37,000 BP).
5. Within haplogroup J can also see that J1, J2a and J2b branched from the parental node in a similarly brief window (est. 3000 years: c. 37-34,000 BP).
6. The other haplogroup where we can see such "supernodes" is R1:
6.1. First, R1a, R1b1a, R1b1c, R1b1b1 and R1b1b2 split in a brief epysode, estimated in 4000 years here (c. 25-21,000 BP maybe).
6.2. Second (within R1b1b2) R1b1b2a*, R1b1b2a1* and the widespread R1b1b2a1a branched out in a similarly brief period (est. 3000 years), that is coincident with the late Magdalenian timeline (c. 16-13,000 BP).
So as conclussion:
First of all, someone explains me how the new discovered SNP defining the NOP macro-haplogroup makes any difference at all in the sequence of diversification of K in a very short period, generally believed to have happened in South Asia.
Second, someone explains me how to make sense archaeologically of R1b1b2 arriving from somewhere in the SW regions of Asia to Europe within what seems to be more a Solutrean than a Gravettian context? Has any proto-Solutrean been found in West Asia (for instance)? Proto-Magdalenian maybe? I know of nothing of the kind, except maybe a late "Aurignacoid" industry of Iran someone mentioned to me once.
Third, haplogroup I looks more and more a pervivence of Aurignacian colonization and might even be used to fine tune the chronology, pushing it back 2,000 (Aurignacian sensu stricto) or even 6,000 years (Balcanic Bacho Kiro culture and other transitional groups). An expansion within West Asia (J and G) may be parallel to Aurignacian or rather a little later in time (curiously enough).
And that's all for today.