A Mid Infrared Search for Variable Stars in the Milky Way Galaxy
Brian Brooks
Advisor: Dr. Robert Benjamin
Contact me at: bbroo009@odu.edu


Home -- Introduction -- Data -- Conclusions -- About Me -- Thanks



A Mid Infrared Search for Variable Stars in the Milky Way Galaxy



Image of blinking source at                                                                                                                                   
  L030.572+0.696.gif
L=030.572 B=+0.696
Alt ID: 2MASS J18443910-0146570
Classification: Possible Asymptotic Giant Branch Star
References: Robitaille et (2008)
Image of blinking source at                                                                                                                                   
  L003.253-1.468.gif
L=003.253 B=-1.468
Alt ID: IRAS 17557-2652
Classification: Maser
References: Sevenster, M.N. et (1997),AAS,122,79S
Image of blinking source at                                                                                                                                   
  L029.716+1.878.gif
L=029.716 B=1.878
Alt ID:IRAS 18362-0202
Classification: Star
References: Sevenster, M.N. et (2001)AA,366,481S
Image of blinking source at
  L284.223-0.473.gif
L=284.223 B=-0.473
Alt ID:
Classification:
References: None

Displayed above is a set of blinking gifs that I created using DS9 imaging software. Our variable sources were obtained by the two overlapped areas as depicted in Fig 1 on the introduction page. The GLIMPSE team developed a program that would find variable stars in the Milky Way Galaxy by comparing matched sources and identifying variations in the apparent magnitudes. The gifs were created to test that the program for the 10-sigma cuts were variable sources or not. The 10-sigma cut contain a total of 1060 variable stars. By hand, I created one acrmin blinking gifs for 50 of our variable sources.

Fig 2: Histogram for apparent magnitudes of variable stars (column 1 histogram of magnitudes, column 2 difference in magnitudes)

After creating the blinking gifs, we wanted to know more about the distribution of variables. Here, displayed in Fig 2 for column 1, is a histogram of the apparent magnitudes for the variable stars. The first plot is displayed for band 1 (3.6 micron wavelength) epoch 1 which corresponds to GLIMPSE survey and epoch 2 for DEEP GLIMPSE in blue. There is a peak at around 7.5 and another at 11.5 magnitudes. In the second column is the difference in magnitude for the variable stars between the two epochs. Our variability ranges from 0.5 to around 5.5 apparent magnitudes. Two important aspects from the pair plots are our variability range and the bimodal distribution of the variables in the first two bands. Bands 3 and 4 do not display a bimodal distribution due to the lack of data from a detector malfunction in the DEEP GLIMPSE project. After learning about our distributions we sought to uncover the galactic location of variable stars.

Fig 3: Galactic map with location of faint and bright variables

In hopes of determining the galactic location, we split the bimodal distribution into two arrays of data. Anything from 0 to 9.5 magnitudes was catalogued as a bright variable star and 9.5 to 20 magnitudes was identified as a faint variable star. Fig 3 is a galactic map with the longitude on the x-axis and latitude on the y-axis. Faint variables plotted in white and the bright variables displayed in red. From this plot, we noticed that the majority of the bright variable stars are concentrated closer to the galactic center of the galaxy with the fainter stars distributed in both regions.

Fig 4: Histogram of color separation

Next we went further into describing our distribution by plotting a histogram for the color separation at 3.6-8.0 microns. What we noticed was that, on average, brighter variables plotted above in Fig 4 were redder than the fainter variables in black. Information that we gathered from this plot has lead us to believe the reason for bright variables being redder is from the dust that bright stars emit, thus we are able to detect in the infrared.

Fig 5: Infrared Dark Cloud (IRDC)

Fig 6: Star Forming Region (SFR)

Fig 7: Near Star Forming Region (nearSFR)

Fig 8: Field Star

The next step was to classify the environment of our 1060 10-sigma variables. An environment could either be an IRDC (Infrared Dark Cloud), nearSFR (near Star Forming Region), SFR( Star Forming Region), or Field Star. An infrared dark cloud is a dense region of a giant molecular cloud that is cold. These regions repesent highly active star forming regions in the Milky Way. In the images above there is a set of variables in the first row and the environment they reside in directly below each of them. We were able to detect IRDC by observing and noticing if we had a green cluster of stars like in Fig 5. The green cluster comes from the dust that the stars in the region emit, allowing the stars light to interact with the dust and displaying it as green with our infrared detection. Next in Fig 6 we have a SFR. The glow is from chemical compounds called Polycyclic Aromatic Hydrocarbons (PAHs). These compounds are excited by ultraviolet radiation given off by massive stars causing the PAHs to glow in the 8.0 micron band. This glow tells us that there was recent star formation. Any star that was not in a SFR but in the 2 acrmin frame of a SFR was classifed as a near star forming region which is shown in Fig 7. Lastly if, SFR, nearSFR, or IRDC were not present then the environment was a Field Star as shown in Fig 8. This information will be used for development of a web page for our 10-sigma variables.

Fig 9: SIMBAD Classification of source findings

Fig 9 displays a list of SIMBAD classifications for our 10-sigma variables. The SIMBAD sources are separated by bright stars in red and faint stars in green. In total, we recovered 368 SIMBAD sources based on our 10-sigma variable stars. Next we subtracted any infrared sources, star, and classifications with a question mark. We are left with a total of 56 sources. This analysis of the SIMBAD findings will help us determine more of a distinction between the bright and the faint stars, thus allowing us to determine what types of sources are varying.

Useful links

The following links are very useful for looking up info on UNIX, web page making, and astrophysical data and journals.

SIMBAD (Stellar/Galactic database)

NED (Extragalactic database)

UNIX tutorial

Web page basics

NASA Astrophysics Data Service


Home -- Introduction -- Data -- Conclusions -- About Me -- Thanks