As stated on our main page, the aim of
this research-project is to find the answer to the question: How
sure are we that there exists dark matter. To do this, we have
decided to examine the assumptions which are involved in the different
kinds of evidence, which are also stated on our main page. One of these involves the value of
and this one will be discussed here
to answer the question: How strong is the evidence for the
existence of dark matter following from the value of
?
is
defined to be the ratio of the average mass density to the critical
mass density of the universe. I will explain this in more detail in
the following section, in which I will also explain how this parameter
is related to dark matter. At that point, the following subquestions
follow in a natural way:
?
based on the mass, which we can detect?
will differ from each other, since
the first includes dark matter, while the second is purely based on
the matter we are actually able to detect.
?There are three possible scenario's for the future of the universe:

An important quantity in cosmology is the critical mass density. This is
defined to be the mass density, which is required for a flat
universe.
is closely related to
these quantities, since it is defined in the following way:

How is
related to dark matter? The
problem is, that when
is measured,
values of about 0.3 to 0.45 (Eke e.a. 1998, De Laix 1998) are
obtained, but when
is estimated based
on the mass we can detect, values of about 0.1 to 0.2 (Evrard 1987,
Huchra and Geller 1982) are obtained. The difference between these
values has to be dark matter, since that is the only matter, which is
not taken into account in the estimation, but is in the
measurements.
For this research-project, the question to be answered is: how strong is this piece of evidence for the existence of dark matter. To answer this question, the following questions automatically arise:
?
based on the mass, which we can detect?
?I've found four ways to measure
:
. This seemed the most
promising method, because it offered a way to measure
directly instead of measuring other
parameters first and calculating
from
those results and it is based on redshifts, which we can measure quite
accurately, so I've chosen to concentrate on this method.
The theory behind the method involving redshifts consists of one
formula, in which the decelleration-parameter, q0 and
are the only constants, One can find
values for these constants by fitting one's measurements with this
formula. The best way to find the assumptions which are involved in
this technique is following the path, that leads to this
formula. Since the mathematics are
quite cumbersome, I have decided to give a sketch of what the method
does and list of assumptions which are involved. For a more thorough
discussion, on will have to go through the mathematics.
The description, which follows, is not an accurate description, but it does give an idea of what this method does in a way, which is a lot easier to imagine than the real line of reasoning. Imagine, that you are looking at a sample of stars (or other objects, but I will keep calling them stars) on the sky, which form a group. These stars will have a certain redshift. This redshift can be explained by the Doppler-effect, since the stars will be moving with respect to the earth. But let's assume for the moment, that those stars don't move with respect to us, so they're moving with us, but that we still measure the same redshift and that the equation for the Doppler-shift is still valid. The only way we can accomplish this construction is to accept, that the speed of light is NOT constant!
Since we can only measure the redshift of stars in our line of sight, the speed of light only changes in the direction of the line of sight and not in the direction perpendicular to it, so we have mathematically created a new reality in which the speed of light within a group of stars has two different values in two perpendicular directions. This has important consequences.Now, imagine, that you lived in that group of stars in that mathematically created world. As mentioned above, the speed of light would have two different values in two perpendicular directions. Imagine, that you would apply the normal Doppler-equation in your situation and would measure the redshift of two stars: one in the direction in which the speed of light hasn't changed and one in the direction in which the speed of light has. Imagine you measured the same redshift for both stars. In that case, the outcome of your experiment would be, that these stars wouldn't have the same velocity.
The consequence of this is, that one of the stars will move quicker
away from you than the other. Now imagine that you do this with a
group of stars, which are more or less spherically distributed around
you and which have the same redshift. In one direction the speed of
light will differ from the other two (we're now entering the third
dimension, which is also unaffected, since there is only one direction
in which the value of the speed of light changes). The result is, that
if you would live long enough, you would see how this sphere deformed
into an ellipse, since some stars are moving out quikker than other
stars. This rate of deformation from a sphere into an ellipse can be
measured and expressed in terms of
and q0 (Alcock and Pazcynski 1979)
To get to the final result, one has to run through a lot of mathematics, during which one has to make the following assumptions:

This expression is only valid if the curvature of the universe,
determined by the value of a constant k, follows from k=1. There are
also formulae for other curvatures (Alcock and Paczynski 1979), but
I'll only use this one, since all measurements and estimations
indicate, that
<1, so k seems to be
equal to 1, so I only mention this one here. Since this is a very
complicated expression and because results are obtained by fitting
data to it, I've made some graphs, which relate the ratio
/z
to z for
different values of
0.
Before I could start plotting those graphs, I needed to make some choices:
Since measurements
using this technique result in values of ca. 0.4, I have decided to
make three graphs: one for each of the values 0.35, 0.4 and 0.45. I
have chosen to do so, since by doing this, it might be possible to
see, how sensitive this technique is: if the graphs differ a lot, the
technique will be more accurate than in the case, in which they are
quite similar.


/z
versus
z).


The upper-left graph shows the result for q0=-0.5, the
upper-right one for q0=-1.0 and the lower-left graph for
q0=-1.5, all for
equal to
0.35, 0.4 and 0.45. Since the curves are quite similar for
q0=-0.5 and q0=-1.0, it will be hard to get an
accurate value for
by fitting your
data on this graph. But the only thing, that is relevant for this
research project, is the difference between the values for
following from measurement by, for
example, this method and from estimation by, for example, the method
involving Mass-to-Light-ratios described below. Since estimates give a
value of about 0.1 to 0.2, I have made another graph showing this
relationship for
equal to 0.4, 0.2
and 0.1 respectively for q0=-1.0. This is the lower-right
graph. As you can see, these graphs differ enough from each other to
decide which value of
fits best to
your data, so we can't deny this difference by pointing out, that it
is just the result of inaccuracies in the measurements.
I've made these graphs using Mathematica and by entering the
following commands (the result is the upper-right graph):
Although there are a lot of assumptions needed to arrive at the
final equation, I still think, that this is a very good method, since
the variables which have to be measured can be measured with a high
degree of accuracy and this method offers a way to find a value for
, which does not depend on one single
measurements. but a value, which gets better by doing more
measurements, since the measurements can be chosen to be totally
independent by spreading them out over a wide range of different
redshifts. The only problem is, that the value one finds for
De Laix (1998) has made a thorough study of the errors in this
method and has come up with
=
0.3+/-0.18, while Eke e.a. (1998), who have used a method involving
cluster evolution, have found
=
0.45+/-0.25. I think that the latter value is somewhat too high since
this was found by means of cluster evolution and I think that the
average density of the universe is somewhat lower than that of a
cluster.
based on the mass, which we can detect?One way to estimate the value of
is by means of the Mass-Luminosity-Relationship of groups of
galaxies. This is based on the simple relation (A.E.Evrard, 1987):

If this equation is used, the following questions can be raised:
One way in which the Mass-Luminosity-Relationship is calibrated is
by using the virial theorem (Huchra and Geller, 1982). This is done by
measuring the luminosity and determining the mass by means of the
virial theorem and bringing those data together. For the purpose of
this research-project, this isn't the best way to do it, I think,
since, if there is dark matter, is will be counted as visible mass,
since the use of the virial theorem doesn't take into account the
difference between detectable matter and dark matter. On the other
hand, even with this possible piece of dark matter included, the value
for
found this way is still lower
than the value found by means of the redshift-method described above,
so one could also regard it as a very strong piece of evidence for the
existence of dark matter.
Another way to calibrate the Mass-Luminosity-Relationship is based on the Initial Mass Function. This is explained by Douwe, but he concludes, that this isn't a useful way to calibrate a Mass-Luminosity-Relationship.
According to direct measurements
has a value between 0.3+/-0.18 (De Laix 1998) and 0.45+/-0.25 (Eke
e.a. 1998), of which the latter may be somewhat too high. Values for
between 0.2 and 0.48 are within on
standard deviation of both measurements, so a value of
more or less directly.
According to the estimations based on the detectable mass
has a value of about 0.1 to 0.2 (Evrard
1987, Huchra and Geller 1982). Since this involves dynamics, there
will probably some dark matter be taken as visible matter, so the
value we are looking for, will, I think, be about 0.05 to 0.1.
The conclusion, that follows from these measurements is, that 50% to 90% of all the mass in the universe is dark matter.
or dig further into the calibration of the
Mass-Luminosity-Relationship. At the end of their article, Huchra and
Geller refer to some other articles, which deal with other methods to
calibrate the Mass-Luminosity-Relationship. This might be a good
starting point.