Is the Universe Younger than We Thought?

In article , (Steve
Willner) writes:

In article ,
"Phillip Helbig (undress to reply)"
writes:
At this level of precision, it's probably not enough to simply
parameterize this, but rather one needs some model of the mass
distribution near the beams.

That's exactly right (at least to the extent I understood Shajib's
talk). In particular, one has to take into account the statistical
distribution of mass all along and near the light path and also (as
others wrote) the mass distribution of the lensing galaxy
itself.

These effects, i.e. that the mass in the universe is at least partially
distributed clumpily (apart from the gravitational lens itself, which
is, essentially by definition, a big clump), also influence the
luminosity distance, which of course can be used to determine not just
the Hubble constant but also the other cosmological parameters.
However, it's not as big a worry, for several reasons:

As far as the Hubble constant goes, the distances are, cosmologically
speaking, relatively small, whereas the effects of such small-scale
inhomogeneities increase with redshift.

Whether at low redshift for the Hubble constant or at high redshift for
the other parameters, usually several objects, over a range of
redshifts, are used. This has two advantages. One is that these
density fluctuations might (for similar redshifts) average out in some
sense. The other is that the degeneracy is broken because several
redshifts are involved. (If the inhomogeneity is an additional
parameter which can also affect the distance as calculated from
redshift, with just one object at one redshift one can't tell what
effect it has, but since the dependence on redshift is different for the
inhomogeneities, the Hubble constant, and the other parameters, then
some of the degeneracy is broken.)

At the level of precision required today, simply describing the effect
of small-scale inhomogeneities with one parameter is not good enough.
It does allow one to get an idea of the possible size of the effect,
though. To improve, there are two approaches. One is to try to measure
the mass along the line of sight, e.g. by weak lensing. Another is to
have some model of structure formation and calculate what it must be, at
least in a statistical sense.

There is a huge literature on this topic, though it is usually not
mentioned in more-popular presentations.

I even wrote a couple of papers myself on this topic:

http://www.astro.multivax.de:8000/he...ons/info/etas=
nia.html

http://www.astro.multivax.de:8000/he...ons/info/etas=
nia2.html

In article , I wrote:
The upshot is that the discrepancy between the local and the CMB
measurements of H_0 is between 4 and 5.7 sigma, depending on how
conservative one wants to be about assumptions.

We had another colloquium on the subject yesterday. Video at
https://www.youtube.com/watch?v=K1496gv8KCo

The points I took away a 1. both the local ("direct") measurements
and the distant ("indirect") measurements are made by two
_independent_ methods, which agree in each case. That is, the two
direct methods (SNe, lensing) agree with each other, and the two
indirect methods (CMB, something complicated) agree with each other,
but the direct and indirect measurements disagree.

2. contrary to what I wrote earlier, even a non-physical change of
dark energy with time (say an abrupt increase at some fine-tuned
epoch) cannot fix the disagreement.

3. while there have been several suggestion for new physics to fix
the problem, none of them so far seems to work without disagreeing
with other data.

What fun!

--
Help keep our newsgroup healthy; please don't feed the trolls.
Steve Willner Phone 617-495-7123
Cambridge, MA 02138 USA

On 19/11/02 9:50 AM, Steve Willner wrote:
...
... 1. both the local ("direct") measurements
and the distant ("indirect") measurements are made by two
_independent_ methods, which agree in each case. That is, the two
direct methods (SNe, lensing) agree with each other, and the two
indirect methods (CMB, something complicated) agree with each other,
but the direct and indirect measurements disagree.

2. contrary to what I wrote earlier, even a non-physical change of
dark energy with time (say an abrupt increase at some fine-tuned
epoch) cannot fix the disagreement.

Indeed someone asks this question at http://youtu.be/K1496gv8KCo?t=3785
(at about z=10^(10) in the video, I believe..) and the answer given is
that it cannot be an abrupt change, "it must be smooth". The presenter's
answer seems to invoke (partly) other observations that rule it out. (So
change in dark energy might fix it but create new disagreements, which
would bring it in category 3, below.. Or would the discrepancy already
be in matching the data actually discussed here?)

3. while there have been several suggestion for new physics to fix
the problem, none of them so far seems to work without disagreeing
with other data.

What fun!

Yes! So why are only 20 people attending?!

--
Jos

On 02 Nov 2019, (Steve Willner) wrote:
3. while there have been several suggestion for new physics to fix
the problem, none of them so far seems to work without disagreeing
with other data. ... What fun!

How about a migrating space-time curvature? Can that connect the two
places? It would have the effect of making distant places dimmer.
Try it in a static universe also, its effects simulate FRW.

[[Mod. note -- I think this is on the edge of our newsgroup ban on
"excessively speculative" submissions, but clearly *something* odd
is going on.

I wonder if it could be "just" non-uniformity in the Hubble flow
in the region of the "local" measurements?
-- jt]]

In article ,
Jos Bergervoet writes:
Yes! So why are only 20 people attending?!

Attendance was far higher than that. The video shows only one side
of the main floor of the room, and the other side is far more popular
(perhaps because it has a better view of the screen). There's a
balcony as well, and quite a few people leave at the end of the talk
and before the question period. I didn't count, but I think the
attendance was close to 100. Anyway it was about the normal number
for a colloquium here.

The colloquium list for the fall is at
https://www.cfa.harvard.edu/colloquia
if you want to see what other topics have been covered.

To the question in another message, I don't see why some local
perturbation -- presumably abnormally low matter density around our
location -- wouldn't solve the problem in principle, but if this were
a viable explanation, I expect the speaker would have mentioned it.
It's not as though no one has thought about the problem. The
difficulty is probably the magnitude of the effect. I don't work in
this area, though, so my opinion is not worth much.

--
Help keep our newsgroup healthy; please don't feed the trolls.
Steve Willner Phone 617-495-7123
Cambridge, MA 02138 USA

[[Mod. note -- I apologise for the delay in posting this article,
which was submitted on Fri, 8 Nov 2019 21:15:25 +0000.
-- jt]]

In article , Steve Willner
writes:

To the question in another message, I don't see why some local
perturbation -- presumably abnormally low matter density around our
location -- wouldn't solve the problem in principle, but if this were
a viable explanation, I expect the speaker would have mentioned it.
It's not as though no one has thought about the problem. The
difficulty is probably the magnitude of the effect. I don't work in
this area, though, so my opinion is not worth much.

I'm sure that someone must have looked at it, but is the measured Hubble
constant the same in all directions on the sky? (I remember Sandage
saying that even Hubble had found that it was, but I mean today, with
much better data, where small effects are noticeable.) If it is, then
such a density variation could be an explanation (assuming that it would
otherwise work) only if we "just happened" to be sitting at the centre
of such a local bubble.

Of course, some of us remember when the debate was not between 67 and
72, but between 50 and 100, with occasional suggestions of 42 (really)
or even 30. And both the "high camp" and "low camp" claimed
uncertainties of about 10 per cent. That wasn't a debate over whether
one used "local" or "large-scale" methods to measure it, but rather the
deference depended on who was doing the measuring. Nevertheless, it is
conceivable that there is some unknown systematic uncertainty* in one of
the measurements.

---
* For some, "unknown systematic uncertainty" is a tautology. Others,
however, include systematic uncertainties as part of the uncertainty
budget. (Some people use "error" instead of "uncertainty". The latter
is, I think, more correct, though in this case perhaps some unknown
ERROR is the culprit.

On 11/2/19 3:50:25 AM, Steve Willner wrote:
In article , I wrote:
The upshot is that the discrepancy between the local and the CMB
measurements of H_0 is between 4 and 5.7 sigma, depending on how
conservative one wants to be about assumptions.


3. while there have been several suggestion for new physics to fix
the problem, none of them so far seems to work without disagreeing
with other data.

What fun!


In the question and answer period
One person asked if the triple point of hydrogen may provide
insight to the problem of discrepancy between the local
and the CMB measurements of H_0.
The triple point of hydrogen is at 13.81 K 7.042 kPa.
Silvia Galli didn't provide an answer
other than many things are possible.

The questioning person's name was not given.
Can anyone provide some insight
into what the triple point of hydrogen
has anything to do with discrepancy between
the local and the CMB measurements of H_0?

Richard Saam

On 11/2/19 3:50:25 AM, Steve Willner wrote:
In article , I wrote:
The upshot is that the discrepancy between the local and the CMB
measurements of H_0 is between 4 and 5.7 sigma, depending on how
conservative one wants to be about assumptions.


3. while there have been several suggestion for new physics to fix
the problem, none of them so far seems to work without disagreeing
with other data.

What fun!


The Ho data is tightening:

**
Testing Low-Redshift Cosmic Acceleration with Large-Scale Structure
https://arxiv.org/abs/2001.11044
Seshadri Nadathur, Will J. Percival,
Florian Beutler, and Hans A. Winther
Phys. Rev. Lett. 124, 221301 - Published 2 June 2020
we measure the Hubble constant to be
Ho = 72.3 +/- 1.9 km/sec Mpc from BAO + voids
at z2

and

Ho = 69.0 +/- 1.2 km/sec Mpc from BAO
when adding Lyman alpha at BAO at z=2.34
**

Richard D Saam

In article , "Richard D.
Saam" writes:

The Ho data is tightening:

**
Testing Low-Redshift Cosmic Acceleration with Large-Scale Structure
https://arxiv.org/abs/2001.11044
Seshadri Nadathur, Will J. Percival,
Florian Beutler, and Hans A. Winther
Phys. Rev. Lett. 124, 221301 - Published 2 June 2020
we measure the Hubble constant to be
Ho = 72.3 +/- 1.9 km/sec Mpc from BAO + voids
at z2

and

Ho = 69.0 +/- 1.2 km/sec Mpc from BAO
when adding Lyman alpha at BAO at z=2.34
**

I guess it depends on what you mean by "tightening". If one
measurement is X with uncertainty A, and another Z with uncertainty
C, and they are 5 sigma apart, then someone measures, say, Y with
uncertainty B, which is between the other two and compatible with
both within 3 sigma, that doesn't mean that Y is correct. Of course,
if someone does measure that, they will probably publish it, while
someone measuring something, say, 5 sigma below the lowest measurement,
or above the highest, might be less likely to do so.

It could be that Y is close to the true value, but perhaps all are
wrong, or X is closer, or Z. The problem can be resolved only if
one understands why the measurements differ by more than a reasonable
amount.