Welcome to my webpage.  I am a Brown University student, and I spent my summer doing astrophysics research at Lawrence Berkeley National Laboratory.

This is a picture of the lab:

It has a beautiful view!
 
 

The project I was working on this summer involved...
 
 

WHAT ARE THE BLUE BLOB GALAXIES?

    There are still many things that are unknown about galaxies. One mystery which has yet to be solved is exactly how galaxies evolve and age. The larger goal of my research this summer was to contribute to the understanding of a specific type of galaxy evolution.

    There are two major types of galaxies: spiral and elliptical. Previous research on galaxies led to the formation of the morphology-density relation. The morphology density relation basically states that galaxies with spiral morphology, which are young, often forming stars, and emit a blue spectra, will be less densely clustered than elliptical galaxies, which are older, do not form stars, and emit a redder spectra. Most of the galaxies in clusters are therefore red/elliptical, while the blue/spiral galaxies are spread throughout the sky.
 
 

The galaxies I am studying are called the Blue Blob galaxies.  These are nine of the Blue Blob galaxies.  The galaxies are the blobs located at the center of these nine boxes:

The Blue Blob galaxies are unusual because they are  blue spiral galaxies which appear to be located near the center of the a galaxy cluster predominantly composed of red elliptical galaxies.  That cluster, which is known as the Coma Cluster, is one of the most dense and massive structures known in the universe.  It is 2.5 x 10^15 times the mass of the sun.  It has thousands of galaxies in it.  Only 46 of these galaxies are hypothesized to be a part of the Blue Blob.  The image shown below is an X-ray image of the gas in the Coma cluster.  This image has added color, the red representing high X-ray emission and the blue lower X-ray emission.  The X-ray emission is proportional to the square of the the density of X-ray gas.  In this image, north is up, west is towards the right, and east is left.  The center of the Blue Blob galaxy region, which is located 17 degrees west and 6 degrees north of Coma, would be just about in the yellow section of this image, on the right-hand side:
 
 

Evidence has shown that galaxies are still falling into the Coma cluster.  However, most of the galaxies found in the cluster are elliptical.  But if the galaxies outside the cluster are mostly spiral, where do the elliptical galaxies in the cluster come from?  Perhaps the Blue Blob galaxies can help to answer that question...
 
 
 
 
 
 
 
 

RAM PRESSURE STRIPPING

My mentor, Greg Aldering, has hypothesized that the Blue Blob galaxies may provide rare examples of a hypothesized phenomenon known as ram pressure stripping. Ram pressure stripping occurs when blue galaxies, which may have originally been more dispersed, are pulled into a galaxy cluster (such as Coma) by its strong gravitational force. As the blue galaxies come into the cluster, the ionized hydrogen in the cluster will cause the neutral hydrogen in the galaxies to collapse and stars will begin to form at a very rapid rate. However, this rapid increase in star formation will not last long. Soon, much of the hydrogen that was originally in the galaxy will be used up in the formation of stars. Then, the ionized hydrogen in the cluster will push the rest of the hydrogen out of the galaxy, and star formation will cease. It is possible that when this galaxy suddenly ceases star formation, it may become a red galaxy like the other galaxies in the cluster.   This is because there will no longer be young blue stars present in the galaxy.  This may explain why the galaxies in clusters are mostly red.
 

The plot shown below is based on a model of ram pressure stripping ( Fuijta and Nagashima).  It shows the trajectory with which a galaxy would fall into the Coma Cluster.  The y-axis shows distance from the center of the Coma Cluster, which is located at zero kpc.  The x-axis shows time (in millions of years) before the galaxy reaches the center of the cluster.  The arrow shows where  the Blue Blob galaxies are hypothesized to be in relation to the center of Coma.  They are about 2 million years away from reaching the center.
 
 

This next plot compares the rate of star formation with distance from the center of the cluster.  The y-axis shows the star formation rate, and the x-axis shows distance to the center of the cluster, which is located at zero.  The galaxy begins 1700 kpc away from the center of the cluster, and as it falls into the galaxy, its star formation rate increases.  Then, it suddenly drops and ceases star formation.
 
 

My job this summer was to carry out an investigation of whether or not the Blue Blob galaxies are an example of ram pressure stripping. In order to provide further evidence that they are, I needed to show that the Blue Blob galaxies are in fact located in the Coma cluster. My job was to reduce the data which came straight from the telescopes. Once reduced, it is possible to calculate the redshifts of the galaxies, and that redshift will tell us where the galaxies are located. In addition, the reduced data will provide the necessary information to calculate the rate of star formation in the galaxies.
 
 
 
 
 
 
 
 

WHAT ARE SPECTRA?

The data I have reduced is spectrophotometric data. This means that each image is not a direct picture of the sky, but rather a picture of light which is separated by wavelength, somewhat like the separation that occurs using a prism, or in a rainbow. A spectrograph (shown in picture above), which is used to make these spectra, works as follows:
a) light comes in from the sky and passes through a narrow slit.
b) light passes through the collimater, which makes the light waves travel parallel to each other.
c) the light then travels through a prism or grating which separates the wavelengths of light.
d) the lens focuses the separated wavelengths onto the CCD plate.
The images go into a computer file.
 
 

THE RAW IMAGE:

This is the image that I got straight from the telescope. It has been recorded with an instrument called a Charged Coupled Device (CCD). The CCD works using electrons, which are activated by photons from the sky, to form an image on a CCD chip. On the image shown here, the galaxy spectra is the straight horizontal line through the middle. As you can see it is very hard to make out the galaxy. The vertical lines are the spectrum of the light that fills the night sky. There are also cosmic rays, which look like tiny bright dots. In addition to these visible obstructions from seeing the galaxy spectra accurately, there are also many internal errors due to uncertainty in the CCD. There are many steps to reducing spectrophotometric data. I used a computer program called IRAF to do this. On this poster, I will outline some of the most important processes of the reduction.

STEP ONE:

The first step is to subtract an electronic offset that is applied to the CCD detector. This is called bias subtraction. To do this step I subtract this a bias spectrum, which was taken with the telescope shut, from the galaxy spectra. This is a bias spectrum:

STEP TWO:

Next, I do a step which is called "flat-fielding." For flat-fielding, I use spectra that are made by taking an image of a uniform light which is placed inside the telescope dome. Since the light used to make the flat is a uniform light, all the pixels on the CCD chip should be of the same brightness. However, because they are so tiny, there can be slight manufacturing discrepancies in the absorption levels of the pixels, and therefore they do not all absorb light evenly. Thus, the unevenness of the dome flat spectra will tell you how the pixels differ from each other. The spectra of the dome flats are fit to a smooth curve, and this curve is divided into the appropriate galaxies in order to correct for this uncertainty. This is what a dome flat looks like:

STEP THREE:

Cosmic ray subtraction has to be done individually for each spectra. This process involves focusing in on each cosmic ray, and then smoothing it out using the brightness value of the pixels which border it. It is very time consuming, but afterwards the image looks much cleaner. This is a cosmic ray that hit almost parallel to the CCD chip:

STEP FOUR:

When the image comes straight from the telescope, it is set up in terms of x number of pixels vs. y number of pixels. The y axis represents the length of the slit used to make the spectra. The x-axis should represent the wavelength. In order to change the x-axis from number of pixels to wavelength, it is necessary to specify which x values equal which wavelengths. To do this an image called an arc, or comparison, is used. The arc spectra were taken with lamps of known wavelength. The graph below is a plot of an arc spectra. It shows brightness vs. x-pixels. Using tables of calculated wavelengths, I was able to identify the wavelengths on this graph. I then use this wavelength calibration to transform all of the other spectra to a y vs. wavelength scale.
 

STEP FIVE:

Once the image has been cleaned up, the background, which is full of light from the sky, has to be subtracted. This is the original image after all of the previous steps have been performed and the background has been extracted.  You can now clearly see emission lines (the bright sections) along it.  Those lines represent different gases coming from the galaxy. The curvature of the lines shows that the galaxy is spiral.  Because spiral galaxies rotate, one side of the galaxy moves towards the observer while the other side moves away.  The rotation causes this tilt in the emission lines.
The bright emission line you can see in this spectrum is Hydrogen Alpha.  The presence of this line shows that the galaxy is in fact forming stars!

STEP SIX:

After all of these subtractions are performed, the spectra are ready to be extracted. The extraction process changes the spectra from a two-dimensional slice of the sky to a one dimensional view of just the light from the galaxy itself. This is an extracted spectra from the Blue Blob galaxy spectra which we began with.  Here you can see wavelength (angstroms) vs. brightness.  The very bright line near the middle is an emission line of hydrogen alpha, which means that this galaxy is probably forming stars!
 

STEP SEVEN (THE RESULTS):

After making some final preparations, the spectra are finally suitable for determining redshift. The determined redshifts tell me the distance of the Blue Blob galaxies. This distance can be compared with that of the Coma Cluster, which has a redshift of 7090 km/sec, in order to finally see where the Blue Blob lies in relation to Coma.

This table shows the redshifts I have calculated, along with some other information about the Blue Blob galaxies.  The magnitude represents the overall level of brightness in the galaxy.  The larger the magnitude is, the fainter the galaxy.   The color shows how blue or red these galaxies are.  The smaller the number is, the bluer the galaxy.  The concentration index represents the concentration of the light within the galaxy.  The Blue Blob galaxies generally have higher concentration indices than typical spiral galaxies.  This might provide further support for the hypothesis that they are undergoing ram pressure stripping, because it could mean that the gas towards the outside of the galaxy has been stripped off.  The remaining gas is concentrated in the center of the galaxies.

The redshifts I have calculated are geocentric, which means that they have not yet been corrected for the movement of the earth within our own galaxy.
 

This next plot shows a view in the plane of the sky of galaxies which are in --- or falling into --- the Coma cluster. The blue squares are Blue Blob galaxies which we now know are falling into Coma. The red squares are Blue Blob galaxies whose spectra have been reduced, except for the final redshift measurement. The open squares are other galaxies which are associated with the Coma cluster, but do not have properties like the Blue Blob galaxies.
 

This plot shows the velocity of the galaxies associated with the Coma cluster as a function of projected radius from the Coma cluster. The symbols have the same meaning as in the previous plot. The galaxies within these bands are the ones which are falling into Coma. The projected radius is a two-dimensional radius which is projected from three-dimensional space, and therefore shows the closest possible distance these galaxies may be from Coma. (They could be further out, but sufficient data to distinguish that distance is beyond the reach of current techniques.) However, it may be interesting for the reader to compare the radii from this plot to the star formation rate versus radius plot discussed in ``Ram pressure stripping.''

Further analysis of these spectra will lead to information about what sorts of gases the galaxy is emitting. If ionized hydrogen is found in the spectra, that will be evidence that these galaxies are in fact forming stars. The rate of star formation can then be calculated. If this rate fits what is known about ram pressure stripping, it is possible that this research has contributed one of the best examples of this rare phenomenon!
 
 
 
 
 
 
 

For more information about my research, please check out these links:

My introductory paper on the Blue Blob galaxies

My final paper on the Blue Blob galaxies

Greg Aldering's webpage
 
 
 

References:

Aldering, G. & Odewahn S.C., 1995 AJ, 2009, 110.

Fujita Y. & Masahiro N., 1999 ApJ, 619, 516.

Gunn J.E. & Gott J.R.  1972, ApJ, 176, 1.

More Blue Galaxies in the Coma Cluster, Aldering, G. & Odewahn, S.C. (1996), ASP Conf. Ser. 88: Clusters, Lensing, and the Future of the Universe, ed. V. Trimble., 172.