Introduction
My work and interest so far has been focused on the special
type of high redshift galaxies called 
"Lyman-alpha emitters". These are
galaxies found through their strong emission in the Lyman-alpha line
(the strong resonance line in the most abundant element in the
Universe, hydrogen). Typically, these galaxies are found
through narrow-band imaging, where a field is imaged in one filter
allowing a very narrow wavelength range of light to pass
through - a so-called narrow-band filter - and one or multiple broader
filters. An example from our COSMOS project (see below) is shown here to the right. A galaxy
with a strong Lyman-alpha line at the redshift corresponding to the
narrow-band filter wavelength range will then be selected as having
excess emission in the narrow-band filter image compared to the
emission in the broader filter images. In the case to the right, this
means redshift z ~
2.25. This way, galaxies can be easily selected between redshifts ~2 -
8, corresponding to the time when the Universe was between ~0.5 - 3
Gyrs old. With these galaxies we are looking at the child-hood and
teen-age of the Universe! Finding the galaxies is merely a start, next we need to understand the galaxies we have found and work to put these results in context with other observational and theoretical results, so that we in the end, hopefully, can understand how galaxies were formed and evolved into what they are today. This is were the real challenge lies. To add to this knowledge, I am / have been involved in several projects described below.
GOODS-S project
In December 2002, my PhD supervisors Johan Fynbo and Palle
Møller started this project, aimed at characterising Lyman-alpha
emitters with the vaste amount of data that is public in surveys such
as GOODS-S. During
my PhD (2004 - 2007) I then continued with the analysis
of the data. The results were published in Nilsson et al. (2006,
2007a).
The main results from the project were:- We found a so-called Lyman-alpha blob - a large
luminous cloud emitting only in Lyman-alpha - without a detected
counterpart galaxy. This implied that the main power source of the blob
was something never seen before, cold accretion. This means that the
Lyman-alpha emission comes from converting thermal energy in a
collapsing hydrogen cloud into light. The images below show the
thumb-nail images of the blob. The narrow-band image is the top
right-most. No counter-part was found in the other images.
- We found two filamentary structures in the image. A Kolmogorov-Smirnov test revealed the likelihood of this detection to be roughly 4 sigma, i.e. a real detection. Interestingly, the blob appears to be quite centred on the filament.
- We performed an extensive SED fitting for the properties of the (stacked) sample. The results are that the average emitter has very low metallicity, is not very massive (1-10 x 108 solar masses), have little dust (AV = 0.1 - 0.4), have SFR of a few solar masses per year and are quite old (0.4 - 1.0 Gyrs).
- One object in the sample appeared to be a massive starburst galaxy due to its extremely red SED.
- We did a comparison with the colours of Lyman Break
galaxies and found that the colours agreed, but that Lyman-alpha
emitters appeared to be drawn from a fainter sub-sample of galaxies.
This project was initiated in March 2007, with the aim to gather a large sample of Lyman-alpha
emitters
at low(er) redshift, in this case z
~ 2.25. Previous surveys at these redshifts were small, and even though
narrow-band observations in the very blue wavelength regime are
challenging, the rewards are great in that a larger
fraction of the restframe SED can be observed, to a greater
signal-to-noise, with ground-based instruments. In Nilsson et al. (2009) we
published the first results, including a candidate list of 170 robust
Lyman-alpha emitting galaxies in an 0.2 deg2 field centred
on the COSMOS survey
field. Again, the choice of field was based on exploiting public
multi-wavelength data to an as large extent as possible. Analysis of
this sample is on-going, but the early results indicate that some
evolution in the sample has occurred between z >~ 3 and z ~ 2.25. Especially, the AGN
fraction goes up, the distribution of equivalent widths gets narrower
and the average galaxy appears to have more red colours. The latter is
illustrated in the plot here, with the data-points representing various
stacked populations and the lines representing best fit SEDs at
redshift z ~ 3. For details,
see Nilsson et al. (2009).
Further results are to be expected in the next few years.ELVIS project
ELVIS
- Emission-Line galaxies with VISTA
Survey - is a long-term project started already in February 2005.
The goal with this project is to find emission-line galaxies at
different redshifts, e.g. Lyman-alpha emitters at z = 8.8, [OII]-emitters at z = 2.2 and H-alpha emitters at z = 0.8. I have been involved in
the early phases of this project, including writing funding and
observing applications, designing the narrow-band filters to be used
with VISTA and dealing with the company that created the filters. VISTA
is currently delayed in its start of operations and we hope that ELVIS
can start in the winter season of 2009 - 2010. Theoretical predictions
of what may be found with ELVIS were published in Nilsson et al. (2007b).
Further information is also given on the ELVIS
homepage.