ACURA final poster - Sites At Penn State

Mentors: Ann Schmiedekamp & Carl Schmiedekamp (Penn State University, Abington, PA)
Sue Ann Heatherly (National Radio Astronomy Observatory, Green Bank, WV)
Radio astronomy can be defined as a portion of
astronomy that deals with the radio emission of
objects in the sky. These objects emit photons which
can be recorded by the use of current technology.
Experimenters focused heavily on the occurrence of
absorption lines. This is shown when cold hydrogen
gas absorbs the 21cm of radiation that was detected
by the telescope. This causes a jump in energy level.
Our Results
HI Survey
This data was taken November 20th at right ascension 18:03:00 and
declination -22:30:00. There was one absorption line found when the
cloud’s velocity was measured at 7.21 (km/s) by the BON database
while we measured at that same cloud’s velocity at 9.21(km/s). This
absorption line was one of the best found throughout the data and
only repeated once.
This absorption line was found on November 19th at right
ascension 19:00 and declination 4:1.8:00. this translates to
galactic latitude 37.7 and galactic longitude 0.06.
36.40 (km/s)
The purpose of this project is to characterize
absorption lines in the 21 cm radiation of hydrogen
gas. We will attempt to identify absorption lines and
distinguish them from areas where hydrogen gas is
not present. This objective was motivated by the
random appearance of empty lines throughout our
data. If we are successful in classifying the pattern
then perhaps we can move closer to an explanation
and an identification of this event.
33.69 (km/s)
This absorption line was found on November 21th at right
ascension 00:23:00 and declination 65:42:00. This translates
to galactic latitude 119.90 and galactic longitude 2.00.
This cluster of absorption lines was found on November
Right ascension 18:43:00 and declination 05:23:00. In total, there
were three absorption lines one following the other. This data
showed the most absorption lines during one scan of the sky.
In order to achieve our results our team traveled to
the National Radio Astronomy Observatory located in
Green Bank, West Virginia. Data was taken at this
facility for a total of three days. The essential tool
used to accomplish the recording of photon emission
from the gas clouds is a forty-foot radio telescope.
Constructed in the early 60’s this telescope obtained
its name from its width which reaches a 40 foot
diameter. The process of the experiment consisted on
directing the forty-foot in the proper direction of the
galactic equator, calibrating our data on a chartrecorder in order to achieve proper measurements,
and recording the information on Data Studio. This
technique was applied simultaneously and repeated
for each piece of information.
8.24 (km/s)
Our team was successful in discovering several of the
absorption lines throughout the entire range of data
collected. These absorption lines seemed to appear more
often in right ascension 18:00. Throughout our analysis we
also observed slight differences in the velocities of the
absorption line in the cloud. Velocities tended to be slightly
. larger than those reported by the HI Survey. Those
intensities were always smaller when compared to the
survey data. Lastly, experimenters observed right
ascensions where absorption lines occurred multiple times
in one data.
This absorption line was found on November 19th at right ascension
18:23:00 and declination 13:09:00. In this example, the data taken by
the BON database displays a absorption line at velocity 8.24 (km/s).
However the experimenters did not find an absorption line when taking
data in this part of the sky. In
The authors thank Ming Chan in assisting with data
acquisition and Sue Ann Heatherly with expert technical
assistance in the collection of data and the operation of
the telescope. The authors also gratefully acknowledge the
Pennsylvania NASA Space Grant Consortium for generous
funding of this project.

similar documents