Docstoc

ANALYSIS-OF ECLIPSING BINARY STARS THROUGH PHOTOMETRY

Document Sample
ANALYSIS-OF ECLIPSING BINARY STARS THROUGH PHOTOMETRY Powered By Docstoc
					MS. JUNE SHIH June 6, 1002

Advisors Of Research: Dr. Anthony Shoup Prof. Gaurang Yodh

ANALYSIS OF ECLIPSING BINARY STARS THROUGH PHOTOMETRY WITH UC IRVINE OBSERVATORY

Binary star systems are two stars created at the same time, gravitationally bound to each other orbiting around their common center of mass (Figure 1). Our observations focus on an eclipsing binary system. These stars appear as a single point of light to an observer, but based on its brightness variation and spectroscopic observations we can say for certain that the single point of light is actually two stars in close orbit around one another. The variations in light intensity from eclipsing binary stars are caused by one star passing in front of the other relative to an observer. If we assume that the stars are spherical and that they have circular orbits, then we can easily calculate how the light varies as a function of time, called a light curve, for eclipsing binary stars. These photometric observations and calculations can be performed using the UC Irvine observatory computer programs. Further observations of the eclipsing binary as a spectroscopic binary, using the Doppler shift of their spectra where absorption lines shift periodically back and forth relative to each other, can give data for the velocity curves. Using both photometric and spectroscopic results, we can determine properties such as individual velocities, mass, radii, densities, temperatures of the stars, and the true orbital sizes. Because binary stars are easy to observe, their results play an important role in astronomical studies as a guide to accuracy and stellar evolution. [1]

Figure 1 - Overhead view of binary star; Orbital motion around center of mass

-1-

Binary stars play an important role in our ongoing pursuit to understand the universe and its constituents. Their existence out-numbers any other stellar structure, and probably more so since many remain undetected. Many stars have later been recategorized as part of a binary system or multiple star system. This observed frequency is beneficial to gathering data. [2] The main importance of binary stars is that they are the primary source of our knowledge of the fundamental properties of stars. The evolution of a star depends almost entirely on its mass and chemical composition [2]. Therefore in order to test stellar evolution theories and structure, we need as much accurate data as possible. This knowledge is usually successfully analyzed with binary stars since fundamental properties such as mass, radii, velocities, separation distance, period, and more can be determined. Knowing these properties give information about stars whose proximity hinders studies of them. These stars benefit from binary stars “born” from the same population. A population is a group of stars all born or created at about the same time. Since the universe goes through stages of chemical composition, stars in the same population usually have the same composition, thus having roughly the same properties as the observed binary star. Therefore binary stars can be thought of as the foundation or basis to calibrate data.

There are several types of binary star systems. Their classification is based on their observational techniques. The four main categories of binary stars are: visual, astrometric, spectroscopic, and eclipsing. Visual binary stars, or double stars as it is commonly known, are true binary systems in which both stars are resolvable in a telescopic eyepiece. It is valuable because it is the standard followed for studies of other star luminosities. Astrometric binaries are pairs that have one too faint in comparison to the other and/or too close to its brighter half to be separated in a telescope. Gravitational effects are used and its orbital motion is determined through astrometric methods. Spectroscopic binaries are irresolvable through a telescopic eyepiece. They appear as one star. By using both spectra analysis and the theory of Doppler shifting, their radial velocities can be determined, and ultimately concluding the single star point as a binary star. The last of the four types, and probably the most informative are eclipsing binaries. An eclipsing binary star, like the spectroscopic type, is irresolvable through a telescopic eyepiece. It appears as a single star in the sky. In order for a binary star to be classified as eclipsing, the plane of orbital motion must be sufficiently edge-on to the observer. A 90 angle of inclination (Figure 2), is entirely edge-on from the point of the observer. Conversely a 0 angle of inclination would resemble that of Figure 1; an overhead view.

Figure 2 - Edge –on plane of orbital motion

-2-

If the observer’s angle of inclination is too low, no eclipse will be observed and eclipsing binary type observations would conclude nothing about the stars. Ideal conditions would be entirely edge-on as shown in the figure above, but not imperative. As long as a portion of the star is clipped, the light coming from the star will vary, and that is what gives good conditions for taking light curve data. A light curve measures time-dependent flux. Because eclipsing binaries are viewed as a single star, data points represent total surface brightness from both stars. It shows occurrences at a particular point in the period, or phase, at which the primary and secondary eclipses occur. A primary eclipse (Figure 3, a) is when the brighter, hotter burning star is blocked. This displays the largest drop, or lowest minimum on the graph, in surface brightness, I. A secondary eclipse (Figure 2, b) is when the less bright star is shadowed. This displays the smaller of the two minima.

Figure 3 - Light curve; surface brightness vs. time (phase)

The light curve also maps out the period. A period is an interval of time characterized by the occurrence of a certain condition, event, or phenomenon [3]. The universal point used in astronomy for calculating a binary stars period is the point of lowest surface brightness, or scientifically referred to as the epoch. The epoch is in units of Julian Date (JD). This is the number of days that have passed since Julius Caesar declared day one. The calculation of one day is heliocentric. It is relative to the sun, neglecting effects due to changes in Earth’s position in orbit, where the sun is relatively stationary. Compared to our 365-day orbital period, commonly these binaries have periods less than a day. Overall they do not exceed 10-day periods with the exception of E Aurigae whose orbital period is 27.1 years. [2] When plotted over time, the pattern of the light curve will repeat itself, with the intervals of repetition equaling the period. Thus we use the phase of the period to communicate the time of significant occurrences. Changing the period coordinates to range from 0.0 to 1.0 from the beginning of the period to the end, respectively, we use

-3-

the fractions to describe how far along into the period the point in question is. For example using our Earth’s 24-hour day, if we set the starting time (0.0) of the phase to be 1:00AM then a full period would be back to 1:00AM the next day (1.0). Now at phase one quarter (0.25), it is equivalent to saying a quarter of the period time has passed. This would bring us to 7:00AM. So we can say 7:00AM is a phase shift of 0.25 relative to 1:00AM being the beginning of the day. Again phase shift is relative to the starting point of the period. Notice in figure 3 the period from the 0.0 to 1.0, starts and ends at the epoch. This is the standard light curve format in eclipsing binary studies In some eclipsing binary light curves, the obvious is not so obvious. For example primary and secondary eclipses are not determinable or one minimum can be mistaken as fluctuations contained in the error bar of data collection. These common types of light curve characteristics have been used to categorize eclipsing binary stars one step further. The three main eclipsing binary types are: Algol,  Lyrae, and W UMa. An Algol type (EA) binary star has a noticeably large difference in minima. This shows that one star is significantly brighter than the other. Sometimes the dimmer star’s eclipse can’t even be seen due to the drop in brightness’s comparable size to the error bars in data points. Algol stars are probably the hardest of the three types to observe and take data accurately on. In figure 4, you can see the primary eclipse at phase 0.7. Notice the large drop in total surface brightness.
phase

In an Algol type system, the two stars are further apart compared to the other eclipsing binary types. This proximity observation is drawn from the graph’s relatively constant surface brightness (flat-topped feature). This further suggests that the effects due to proximity are small. In further detail, the stars either have large periods or orbits so far apart that they each radiate freely without interference from the other eclipsing over. In opposition to the latter, if the 0 0.2 0.4 0.6 0.8 1.0 stars were closer together (described in the later eclipsing binary Figure 4 - Algol Type (EA) [4] star types), you will see a more frequently varying surface brightness due to the orbital motion intertwining more frequently.

-4-

The second type is  Lyrae (EB). You can see from figure 5 that these types of eclipsing binary stars still have a noticeable difference in surface brightness. The proximity of the two stars is much closer, meaning their orbit size is much smaller, and/or it has a smaller period. When it is like this, the time lapse between the stars position to be in front, side, Figure 5 - B Lyrae Type (EB) [5] or behind is much shorter causing the EB light curve to be continuously variable [2, pg 7] The last type of eclipsing binary, which is the category my particular binary star falls under, is the W UMa (EW). Figure 6 is an example of a W UMa type binary star light curve. Like the EB light curve, it has a constantly varying surface brightness, though more apparently so. The EW light curve has no point resembling a plateau, where as the EB light curve has no plateau, has smoother, flatter maxima regions. The most noticeable feature of an EW light curve is the similarity in minimums. The primary and secondary eclipses are almost indistinguishable. This tells us that the two stars are similar in surface brightness (magnitude).

Figure 6 - W Uma Type (EW)

Figures 4, 5, and 6 all display data of binary stars not in the plane edge-on to observer. Their angles of inclinations are <90. If the binary star was edge-on, then a total eclipse would be observed and the minima on the corresponding light curve would be flat, rather than the curved-pointed shape in previous figures.

-5-

At this point the light curve has only proven beneficial in determining the period and the occurrences of the primary and secondary eclipses. The photometric light curves alone cannot provide dimensions of binary stars or their orbits, the purpose for their studies. This is due to the relative orbital semi-major axis a. Determination of the full parameters of an eclipsing binary requires both a light curve (photometry) and a radialvelocity curve (spectroscopy). Light curves alone can provide orbital inclination i, relative quantities, ratio’s of luminosities and perhaps mass [2, 13] The semi-major axes of the absolute orbits of the stars about a common center of mass, is the most crucial information in solidifying calculations. Unfortunately, only in a few cases is it possible to measure the semi-major axes a1 and a2 of the absolute orbits separately. Most cases, only relative orbit and its semi-major axis, a = a1 + a2, can be determined. The component star masses M1 and M2, we derive from two equations. [2, 2] First, the absolute and relative orbits are coupled by a = a1 + a2 and the moment equation.

a1 M 1  a2 M 2

(1)

Second, Kepler’s Third Law states: The square of the period of the components is proportional to the cube of their semi-major axis for all systems. Their ratio is constant.
p2  r3

(2)

Period is p, and r is the semi-major axis (normally noted as a, but changed as not to be confused with the acceleration a). Kepler’s Third Law is derived from basic motion equations. Newton’s Second Law:
F  ma

(3)

The acceleration of an object moving in a circular orbit (which is most common for stellar structures, and used for simplicity in calculations) where r again is the semi-major axis and v is velocity: v2 (4) a r The gravitational force equation:
F G ( M 1  M 2 )m r2

(5)

Where M1 + M2 is the binary star system.

-6-

Combining equations (3), (4), and (5) yields:

mv 2 G( M 1  M 2 )m  r r2
Canceling a factor of r, m, and rearranging the equation, we get a velocity equation.
v2  G(M 1  M 2 ) r

(6)

(7)

Since the velocities can only be determined from radial-velocity curves, v needs to be substituted. Velocity is just the distance traveled divided by the time to travel that distance. The distance traveled, again for simplicity will be approximated to a circle, is just the circumference, 2(r). The time to travel the circumference is just the period, p. 

(

2r 2 G( M 1  M 2 ) )  p r

(8)

Rearranging the equation we get Kepler’s Third Law [6][7].

4 2 p [ ] r 3 G(M 1  M 2 )
2

(9)

To actually calculate the masses of the binary stars, let’s assume we know the inclination i and the relative orbital semi-major axis a from extended light curve information. The total mass, M, of the binary system is:

4 2 a 3 M  M1  M 2  G p2

(10)

On the other hand we could have the extended light curve information to include the mass ratio.

q

M2 M1

(11)

-7-

This is where Kepler’s Third Law’s real power resides. Using the approximate orbital parameters, semi-major axis and period, we combine equation (10) and (11) to give the estimated individual star masses.
M1  1 M 1 q M2  q M 1 q

Using equation (1) and the semi-major axis definition, the semi-major axes are [2, 58, 174]:
a1  q a 1 q a2  1 a 1 q

Photometry is the most accurate means of obtaining flux measurements. It uses Charge Coupled Device’s (CCD) to acquire images and measures pixel values. In astronomy however conditions are not always ideal. Outside influences such as change in weather and atmosphere, effect results. For example if data was only taken on the variable star and clouds rolled in, the drop in brightness could be mistaken for approach into eclipse. Differential photometry takes out these effects. It takes the difference in brightness between reference objects. If clouds rolled in, the drop in brightness would be observed on all objects, and the differential data would remain constant. The reference objects are two additional stars measured within the same frame of image. Both stars must be non-variable and have magnitudes no less than 2-3 orders. Typically the brighter of the two is used as the comparison star. The comparison star is directly used with the variable binary star to calculate the light curve. The check star is to verify the comparison star is constant. The light curve is differential photometry in action with two graphs (figure 7). The main graph plots difference in magnitude between the variable star and comparison star (figure 7a). This is the graph referred to when discussing photometric light curves.

(a) Variable and

Comparison Star

(b) Comparison and Check Star
Figure 7 - Example of light curve displaying differential photometry plots

-8-

The second, less seen, graph (figure 7b) is the plot of difference in magnitude between the comparison star and check star. The only purpose of this graph is to validate to the observer that the main light curves data is accurate. Because the two stars are nonvariable, this check plot should fit a straight line. Differential photometry corrects for conditions being less than ideal. What about error in mechanical instrumentation? There is no such thing as a perfect camera, and that includes CCD’s. Several calibration methods are used. First are flats. Flats are images that account for variations in pixel to pixel in the CCD instrument. The image is taken against a uniform light source. The uniformity of this light source is what the accuracy of flat fielding relies on. Ideally every pixel would record the same value. Flats give us these variations in individual pixel ranges of error and also dust particles on the CCD lens window. A dust particle’s obstruction could coincidently be where the star would appear and lower the value of the pixel, appearing as if the star itself were dimmer. A series of frames are taken and the median is determined for each pixel. Flats are divided into the final image taken. Figure 8 is a Figure 8 – Flat Image median valued flat used in my photometry calibration. Notice the spherical patterns in contrast to the white. These are dust particles. Also note the large contrast in light and dark. The light source is coming from the right hand side and is actually uniform, not drastically brighter as it seems in figure 8. CCD Auto’s default is to use the lowest pixel value as the lower limit and highest pixel value as the upper limit to display the contrast more clearly. Darks are images that account for variations in pixel to pixel due to thermal noise. The dark is taken with the shutters on the CCD closed. Ideally it would be completely black (pixel values zero) since no light should be entering the CCD. But darks will actually have variations in pixel to pixel because of the internal thermal noise experienced with digital imaging instruments. Electrons are released in conjunction when met by a photon. Pixel values are directly related to these electron releases. Thermal noise resembles photon interactions, thereby releasing electrons giving pixel value, when no light was collected. Darks are exposed the same length of time as the final image and subtracted so a star’s brightness

Figure 9 – Dark Image

-9-

is not overestimated. Notice the white specs in figure 9. At quick glance, this dark can be mistaken for a distant saturated image of the night sky. The more time spent into acquiring good data, the more parameters that can be determined. My research was conducted over a quarter session of 10 weeks. Of those 10 weeks, 2 ½ weeks were spent training to use the observatory. Between sharing the observatory with other UC Irvine members and catching a clear night in a city located less than 10 miles from the beach, good data was hard to find. Some initial conditions were set for choosing a binary star to study. 1) The period had to be relatively small, in the range of hours less than nighttime, and not equally divisible into our 24-hr period. The reason for the latter condition is so day-to-day the same phase isn’t observed. 2) The collective brightness had to be high; 5-8 in magnitude. This allows for smaller exposure times when taking images, visibility, and lower error. 3) Availability of comparison and check star. 4) Lastly, the Right Ascension (RA), had to be reasonable compared to the window of observable night. For example a star that is just about to set when it gets dark enough to take data, does not give a good window of data acquisition. RA is measured in Local Sidereal Time (LST), which is a measure in the local hour angle of a catalog equinox [8] (position in hour angle directly overhead). These conditions narrowed the choices down to less than a handful. Due to the short span of my research, it was more beneficial for me as a student, to take data on an eclipsing binary already known. My data was to support cataloged properties. Like all growing things, a binary star’s properties will change. Astronomy catalogs are updated periodically to reflect these changes, therefore ongoing research on recorded binary stars are beneficial.

AW UMa (figure 10) fit the initial conditions beautifully (other cataloging names are HD 99946 and Guide Star Catalog (GSC) 19840113). Located in the constellation W UMa at RA 11:30:04 (east-west coordinate system) and DEC +29:57:52 (another hour angle measurement in north-south coordinates) it has a period of 0.4387259 days (10 hrs) and an epoch 2451536.5923 JD [11]. Inclination i is 79.1. The mass ratio of this system is 0.07 < q < 0.08 [12]. The magnitudes of each star are roughly 6.84 and 7.13 [9]. Two stars with magnitudes between 7-9 were within range also.

Check

AW UMa

Comparison

Figure 10 – Image taken with V-filter of AW UMa, comparison, and check star

- 10 -

There have been many studies on this particular binary system yet the physical characteristics are not clearly determined. Records through years have shown variations in period and light. Figure 11 is a light curve of data taken by me on one day. Starting at phase 0.97 (2452408.77014 JD) and ending phase 0.01 (2452408.82203 JD), it is just coming out of primary eclipse. This data is ideal. The plot points have little scatter or error. Figure 11b has fluctuations less than 0.05 magnitude. The program EBPlot used to construct this light curve needed specific parameters so the axes would be accurate. Initially the epoch parameter inputted was not from an updated source. Notice the orbital phase is not in sync with the data. The primary eclipse should hit at phase 0.0.
(a)

(b)

Figure 11 - Light curve of data taken on May 13, 2002; (a) V-C variable subtract comparison; (b) K-C check subtract comparison;

As explained earlier, differential photometry is most ideal for accurate measurements because the differences in brightness between two stars are recorded. In figures 11(a) and 12(a), the axes label “V-C” refers to the difference in magnitude between the variable and comparison star. In figures 11(b) and 12(b), the axes label “K-C” refers to the difference in magnitude between the comparison and check star. Since the brighter the star the smaller the magnitude number, the axes range is negative. The portion of this light curve ran 0.05189 of a JD. It covers a little over one tenth of the total period (figure 12a region A). Using my light curve, the period would roughly be 11.2 hrs or 0.467 of a day. The published period is 0.4387259 or 10.52 hrs. Considering the minimal number of hours of data collected and challenges in photometric methods, which will be discussed in more detail, the results are not too bad.

- 11 -

(a)

A

B
(b)

Figure 12 - Complete Light Curve; (a) variable; (b) check and comparison star

Figure 12 is the light curve of my research, totaling 644 data points. Figure 12a is the light curve and figure 12b is the differential check star plot. Pieces are missing but the key features of a W UMa eclipsing binary light curve can be depicted. The epoch is used to calculate phase, so the primary eclipse is the minimum located phase 0.0. It begins to rise and hits maximum at about phase 0.25. Though the next piece is missing, we can figure through max-min patterns that the graph drops and that the second eclipse must hit at phase 0.5. At just after phase 0.5 where the points continue, the rise out of secondary eclipse is observed. As expected at about phase 0.75, the plot hits another maximum and starts its decent back into the primary eclipse (phase 1.0), completing one period. The occurrences of extrema at quarter phase marks, leads to belief that AW UMa has a circular orbit. Characteristic of the W UMa type light curve, the primary and secondary eclipses only differ slightly, about 0.05 magnitude of brightness. The light curve represents useable data. The differential check star plot validates the data taken was consistent. The comparison and check star showed stability, varying only up to 0.1 magnitude. In comparison to stars that are on the order of 7-8 magnitudes, that is 1% error.

- 12 -

A couple of pieces of the light curve seem out of place. Region A of the light curve appears shifted higher relative to surrounding points. The shift is also seen in the check star plot. The rest of the light curve seems to follow a fitted line, so this shift is probably not due to the binary star. The conclusion that the comparison star that was used to plot the light curve seemed to have gotten dimmer comes from the light curve being shifted up and check star plot shifted down. On the light curve, with the binary star remaining at its expected magnitude, shifting up equals larger difference between binary star and comparison star. Therefore the comparison star is dimmer. On the check star plot with the check star remaining constant, shifting down equals a smaller difference between the comparison star and check star. Therefore the comparison star is also concluded to have gotten dimmer. The check star is not suspected to have changed due to the fact that both the light curve and check star plot were affected. If the check star were the assumed then only the check star plot would be affected since the light curve uses only the comparison star for plotting. Comparison and check stars are non-variable stars so why would they seem to appear dimmer or brighter? One definite reason is the scratch in the V-filter, which is the filter used in this research, on the CCD imaging camera. It is clearly seen in figure 8 in the top right corner resembling claw marks. The filter allows only light in the V band to pass through. With scratches on the filter, unwanted light passes causing objects to be brighter, if captured in that region, or lower in comparison to objects in that region. Secondly, stars drift across the sky as night progresses. Tracking is not 100% accurate with the observatory’s software, so the stars positions are never the same from the beginning of data acquisition to the end. Flats (figure 8) have a slight range in pixel value across the frame, so as a star drifts across this frame, the pixel values being divided in may no longer be the same, thus slightly affecting the final pixel value. Compare figures 8 and 10 and you’ll see that the check star lies in the area these two factors are in play. The large difference in differential magnitude on the check star plot (figure 12b) between phase 0.6 and 0.7 is an example of these two limiting factors. The scatter from phase 0.1 to 0.3 labeled region B is also candidate for further analysis. Rarely are raw data plots perfectly in sync with a fitted line through the graph. A light curve is no exception. But these data points tell a story. The light curve shows inconsistent, largely fluctuating data, but the check star plot fits a straight line through phase 0.1 to 0.3, as it should. Excusing both the comparison and check star as the problem, the differential magnitudes on the light curve are shifted lower With the comparison star staying constant, means the binary star was underestimated in magnitude. This could have been due to photometric errors in overexposure. If the maximum pixel value was reached for the binary star and the CCD continued exposing the image, it would continue exposing everything else in the frame (i.e. comparison star), which wouldn’t have reached max yet. When the image was downloaded, all pixel values would read true except the binary star. The pattern on the light curve is consistent with this error analysis. Observe how all the lowered value points follow the light curve’s pattern out of eclipse into a maximum.

- 13 -

Figure 13 - AW UMa light curve against theoritical light curve (dashed line)

Figure 13 is my AW UMa light curve fitted with a theoretical light curve adjusted from error analysis. To better understand how my data was acquired and the equipments and software programs used, let me briefly describe the UC Irvine observatory and the telescope it houses. Located on the outskirts of campus, it sits in an open field relatively far away from any light pollution from the city and school. The observatory houses a large telescope that is controlled by a computer that can be used onsite at the observatory, or remotely through a Linux terminal. The observatory is 20 feet in diameter and 20 feet high. The dome is motorized and computer controlled. It consists of two parts. The dome slit which slides up and down to open and close respectively. The dome also can rotate around 360 degrees, which allows the telescope to track the sky view as the night progresses. Within the observatory is a computer controlled Ritchi-Critchen reflecting telescope with a 24-inch primary mirror and a 8.5 inch secondary mirror (figure 14).

Starlight

“Primary Mirror”

“Secondary Mirror”

“Focal Plane

FIGURE 14 – How UC Irvine’s telescope acquires images

- 14 -

Starlight is "collected" with the primary mirror, reflects to the secondary mirror, which is then reflected again to the focal plane. The telescope focuses with stepper motors that bring the secondary mirror closer to the primary as needed [10]. Attached to the telescope are several different instruments for increasing accuracy, viewing purposes, photometric, and spectrographic data collection. These instruments are connected to the telescope at all times and can be accessed or switched to use at any time for convenience. They are positioned reminiscent of north, south, east and west on a map. A rotatable mirror is located behind the primary mirror and can be positioned to direct the light from the primary mirror to its desired instrument for analysis of that light. There are 3 CCD cameras. The ST-9E is the main imaging CCD. It is used to collect the images needed for my photometric analysis of binary stars. The ST-8E is used to record spectra from the spectrograph for spectrographic analysis such as blue-red shifts and radial velocity determination [10] (table 1)

CHARACTERISTICS GAIN READNOISE FULL-WELL CAPACITY PIXEL SIZE PIXEL SCALE SIZE OF DETECTOR FIELD OF VIEW (FOV)

ST-9E 2.8 e-/adu 13 e- (Tccd –10C) 180,000 e- (unbinned) 20 x 20 microns (unbinned) 0.81” / pixel (unbinned) 512 x512 pixels 6.9’ x 6.9’ (unbinned)

ST-8E 2.68 e-/adu 18.6 e- (Tccd –10C) 16383 adu (unbinned) 32500 adu (2x2 binning) 9 x 9 micron (unbinned) 0.728” / pixel (2 x 2 binning) 1534 x 1020 pixels 9.3’ x 6.2’ (2 x 2 binning as imager)

ST-6 6.7 e-/adu 30 e4000,000 e65535 adu 23 x 27 micron 0.98” x 1.1” 375 x 242 pixels 6.1’ x 4.6’

Table 1 – Specifications of instruments attached to primary telescope;

The ST-5C is the finder is a Celestron 5" scope used as a digital finder scope used in conjunction with the main telescope for additional accuracy and guiding system. The last instrument attached is the eyepiece. When the light from the primary mirror is directed towards the eyepiece, viewing it manually displays an image that would otherwise be captured on the CCD. As mentioned before, all components of the telescope is computer controlled. From opening the observatory to activating all instruments and programs. The control computer can be used at the observatory itself, or remotely through a hyperlink connection using a Linux computer. The control computer is a 450 MHz Pentium II system running Red Hat Linux. It has a memory 256MB and two fast hard drives of 3 and 20 GB [10].

- 15 -

UCIrob is the main program for controlling the dome and scope. UCIrob can open and initialize all components of the observatory. Features include: instrument selector, commands for scope moves, finder scope commands, and turning on/off of lights, CCD instruments, and much more. On the bottom of the program window displays status of elements and features of the scope/dome, such as the current position of the scope in RA, DEC, altitude (Alt), and azimuth (Az) of the scope. Scope features stats such as whether the dome is open/closed, if features of the scope are on/off, and if certain CCD's are linked, just to name a few. Current time displays in Universal Time (UT), LST, and military help for diverse tracking of the night sky. CCD Auto is the program used to control the imaging (photometry) and spectrograph (spectroscopy) CCD's. Photometry is the measurement of light. Photometry is the most accurate form of obtaining flux measurements, and is executed through the marriage of these two key photometric components.

Photometry is a very powerful tool in observing and determining parameters otherwise immeasurable. In conjunction with other observational methods, the fundamentals and life of an eclipsing binary star can be determined. Analysis of light curves and radial curves tell a story. This story provides information to surrounding stellar life, past and future stellar evolution. Binary stars are continuously studied in both new discoveries and progress of known ones such as AW UMa.

- 16 -

ACKNOWLEDGEMENTS I would like to thank Dr. Anthony Shoup for his patience and extensive availability to train, assist and answer questions. Also to Prof. Gaurang Yodh for giving me the opportunity to learn and research using the UC Irvine observatory. I’ve learned so much. More so than any astronomy book could have taught.

REFERENCES [1] Bruton, Dan. Eclipsing Binary Stars <www.physics.sfasu.edu/astro/binstar.html> [2] Kallrath, Josef and Milone, Eugene F. (1999) “Eclipsing Binary Stars: Modeling and Analysis”. New York: Springer [3] Lexico LLC. (2002) <www.dictionary.com> [4] “The Light Curve of Algol” <www.macalester.edu/astronomy/courses/phys40/mark/ltcurve.htm> [5] Bruton, Dan., Linenschmidt, Robb., and Schmude Jr., Richard W. (1996) “Watching Beta Lyrae Evolve” Internet Science Journal. <www.physics.sfasu.edu/astro/betalyra/> (23 January 1996) [6] <http://www.astro.washington.edu/astro211/webwork5/derive.html> (9 November 2000) [7] Cowles, Dennis Joseph. (2002) “Kepler’s Third Law”. Skylights Online Newsletter. 17:5 <http://www.craigmont.org/kepler.htm> (30 May 2002) [8] Fisher, Rick. (1996) Astronomical Times. <http://sadira.gb.nrao.edu/~rfisher/Ephemerides/times.html#LST> (30 June 1996) [9] Centre De Donnees Astronomiques De Strasbourg (2001) <http://simbad.harvard.edu/sim-fid.pl> (March 2001) [10] Smecker-Hane, Tammy. (2002) “The UC Irvine Observatory”. Dept. of Physics & Astronomy, UC Irvine <www.physics.uci.edu/~observat/> (1 March 2002) [11] Pribulla, Vanko, Parimucha, and Chochol. “New Photometric Minima and Updated Ephemerides of Selected Eclipsing Binaries”. Konkoly Observatory. (2001) http://www.konkoly.hu/cgi-bin/IBVS?5056 (7 April 2001) [12] Jeong, J.H., Lee, Y.S., and Yim, J.R. (1997) Light Curve Solution of The Contact Binary AW UMa. Space Sci, 14(2), 225-232

- 17 -

- 18 -


				
DOCUMENT INFO
Shared By:
Stats:
views:34
posted:11/27/2009
language:English
pages:18
Description: ANALYSIS-OF ECLIPSING BINARY STARS THROUGH PHOTOMETRY