Home

Richard Pearson Senior Thesis - ROVOR

1. Rx D 4 Z I 3 1 where I is the raw unprocessed frame initially from the CCD R is the real light from the object of interest is the individual pixel variation D is the background noise and Z is the bias level Calibration frames are taken to understand the contribution of each term Note that different types of astronomical work require a slightly different image equation i e Equation 3 1 and thus different types of calibration frames For ROVOR an optical telescope without a spectrometer the calibration frames are of three types biases or zeroes darks and flats 3 1 1 Biases An electronic offset is applied to CCD data so the A D amplifier operates linearly at low light levels This offset is measured by a bias or zero frame which is essentially a zero second exposure image with the shutter closed The purpose of this is to simply read each pixel s biased level To remove this bias level from the raw unprocessed image one simply directly subtracts the bias frame from the raw frame Therefore Equation 3 1 now becomes Rxo D I Z 3 2 A good practice is to collect a large amount of bias frames average them together and subtract them from the raw image When these are averaged it is wise to use a 3 1 Calibration Frames 14 min max rejection This throws out the minimum and maximum values associated with each pixel
2. Remember that the D here has been bias subtracted as well If you find that subtract ing the bias from the dark actually causes more error you can just use the averaged dark frame for subtraction with the bias level still intact For my research this is 3 1 Calibration Frames 15 the process that I used We found that with the previous CCD the Apogee 47p it was best to use a collection of dark frames without the bias subtracted from those frames So instead of having I Z D we used the form I Dz where Dz was the dark that had not been bias subtracted mathematically you can see these two equations are essentially equivalent The latter however did not include an averag ing of the bias level We hoped that by leaving out the bias averaging the statistics would improve we did not run any test to determine if this statistically improved the photometry all we know is that it did not make it worse One last thought regarding dark subtraction on the new current CCD the FLI version of an Ap8 We have been running the CCD at a temperature of 30 C This provided excellent images and fine photometry However we found that the CCD could go down to 45 C By doing so a 60 second dark frame had the same counts as a bias frame thereby eliminating the need for the dark subtraction because D Z 0 This is common for cryogenically cooled optical CCDs whos
3. Once Orchestrate is open go to View gt Properties gt Client This will allow Orchestrate to connect to TheSky CCDSoft the Telescope and Filter Wheel You must find the name of the machine that is being used This is found under System Properties gt Computer Name The Full Computer Name is what you will put in the Machine Name in Orchestrate Then put in the user name and password of the machine you are using the login you used when the computer starts up In the case here we use ROVOR 02 for the Machine Name rovor and outhouse For filters instead of using names we use numbers Here are the numbers associated with each of the filters a 1 Ha Narrow b 2 Ha Wide c 3 B d 4 V e 5 R f 6 1 Filter Wheel Setup 1 Once you have downloaded and installed the FLISetup package for Windows from the Finger Lakes Instrumentation website you need to make sure that the USB connection for the filter wheel is in place When it is and the computer recognizes that it is there then continue On the Start menu go to All Programs gt Finger Lakes Instrumentation gt FLI Filter This is actually the only program that we will end up using for ROVOR so if you deem it necessary the other programs can be removed as well When you first open FLI Filter it will ask you to designate a port for the software to speak to the filter wheel Select the option to have it scan for the USB connection and if you wanted have it search
4. We were able to observe Mk501 in the B V R and J bands Due to the nature of the project we desired the highest signal we could get with the shortest exposure time We decided to look for microvariability primarily in the R band Hence the majority of our observational 51 5 1 Observational Data 52 Filter Total Nights Total Images B 49 584 V 49 492 R 61 3693 I 24 176 Table 5 1 Total observational nights and frames over the 2009 and 2010 campaign of Mk501 for each filter Note the heavy emphasis in the R band Intraday variability or IDV was searched specifically in the R band region time was devoted to R band observations see Table 5 1 Tables 5 2 and 5 3 show the entire observational campaign The processed R band data for Mk501 is shown in Figures 5 1 and 5 2 The first shows the normal differential magnitudes obtained with our reductions from April 22 to July 6 2009 The error bars used here include the error in the standard stars when converting the instrumental magnitudes into true apparent magnitudes these fall around 0 02 mag The second figure displays the R band values obtained by our collaborator Alberto Sadun from the University of Denver His error bars do not include the error in the standard adjustment and range around 0 004 mag Also instead of using just one frame per data point he averaged together about five frames to get a single data point per night His plot ranges from April 22 to Ju
5. What is Orchestrate Orchestrate is a program that automates nearly all processes of controlling the telescope Slewing moving the scope adjusting the filter wheel and controlling the camera are some of the core procedures that can be controlled by Orchestrate Orchestrate accomplishes these tasks by telling TheSky6 how to move the scope and CCDSoft how to operate the camera and filter wheel Orchestrate makes it possible to automate long observing sessions You can create a script of commands load it into Orchestrate and let it run the show Orchestrate also takes a lot of the sting out of making a TPoint model Orchestrate s layout is similar to Excel with columns for commands arguments comments status error messages mxi and so forth In the Command File Edit View Options Connections Data Macro Window Help column you insert commands like Ds ta Ble Wel alolx sls SlewToRaDec or SetResolution In the Arguments column you put modifiers necessary to complete the command For example SetResolution needs an argument to set the resolution either High Medium or Low You can enter helpful comments in the Comment column that will improve the readability of your script The Status column gives you the status of the command This column usually shows one of four things a Dion dd hmi blank cell indicating that the command has not yet been attempted Done indicating that the command successfully ran Working indicating
6. Chapter 3 Reduction and Photometry Techniques Calibrating CCD images to collect statistically viable photometry is an art of error minimization A CCD image is an array of photosensitive wells that convert incoming photons into electrons The electrons are counted by an analog to digital A D amplifier and reported as counts a really technical name However the wells within the CCD do not distinguish between photons photons arriving from a celestial object a flashlight a random cosmic ray or any other source are not differentiated on the CCD This is where the art of reduction comes into play To provide good photometry from the desired celestial object all other possible forms of light or photon counts need to be accounted for and removed A series of calibration frames are used to remove the unwanted photons Following the reduc tion photometry calculations are performed on the object of interest This section describes the various types of calibration frames the programs used for reductions and two types of CCD photometry Also Steve B Howell s book entitled Hand book of CCD Astronomy is one of the best resources and is highly recommended 12 3 1 Calibration Frames 13 it s actually a fairly nice read 9 3 1 Calibration Frames A CCD image is a collection of object light a bias level and background noise or light by pixels of varying sensitivity This can be expressed as a simple equation
7. Once TheSky6 is connected to the telescope you are ready to give the scope some orders Click on the star you want the scope to point at this brings up the Object Information dialogue box Click the slew button it is the icon in the lower right corner of the box that looks like a little green telescope with a red arrow by it This will make the scope point at the currently selected object Be sure the object listed at the top of the box is indeed the target you want to point at DO NOT SLEW THE TELESCOPE Object Information 3 x General Multimedia Utility Telescope Object 1 of 4 You will be asked to confirm the slew Click ok and the white crosshairs indicating where the scope is pointing should start moving toward the target Of course the telescope should start moving as well be sure that nothing is in the way of the scope as it slews Wait for a few seconds after the white crosshairs settle on the target so you don t accidentally take an exposure while the scope is still moving Object name GSC 6726 62 Magnitude 13 2 Equatorial RA 13h 36m 37s Dec 29 56 35 current Equatorial 2000 RA 13h 36m 07s Dec 29 53 46 Horizon Azim 171 33 29 Alt 20 04 53 Visibility Rise 19 01 Set 03 24 Transit time 23 11 Object type Star Source catalog GSC catalog Hour angle 00h 36m 37s Air mass 2 91 Screen x 538 00 420 00 menda This is the Slew Button 4 Us
8. order 1 WAIT for a couple of minutes The computer and or CCD are confused so just be a little patient 2 Try closing CCDSoft and re opening it In doing this you may have to reset the Autosave parameters again 3 Try disconnecting from the CCD Wait again to see if it responds 4 If CCDSoft will not close or disconnect from the CCD the last resort is to cut the power to the CCD After this you should be able to close CCDSoft it may automatically close itself Wait for a couple of seconds and turn the power back onto to the CCD Then start up CCDSoft reconnect to the CCD begin cooling and reconnect everything in Orchestrate In fact you may have to restart Orchestrate as well Just try not to abort any image downloads Cutting the power to the CCD isn t a very fun thing to do There have been a couple of times when everything has connected just fine CCDSoft to the camera filter wheel and focuser theSky to the telescope Orchestrate to the telescope the mount the camera and filter wheel and yet when you go to home the scope it displays an error The error it usu ally displays when it cannot find its home position is something to the effect of Telescope could not be homed Some telescopes do not support this com mand This is probably true but our telescope does support this command and it is essential to the telescope actually working after it has been parked It is most likely a mechanic
9. 52 2 DEC 39 45 36 6 The AGN model has been extended to include every galaxy in the universe It proposes that every galaxy in the universe holds a blackhole at its center To better confirm and clarify this presumption a study of AGN is necessary and Mk501 was chosen for examination Three things were hoped to be obtained from the Mk501 observations First we hoped to find small short variability in the blazar This is otherwise known as intraday variability or IDV Second day to day or rather night to night variations known as intranight variations were sought Third periodic behavior in the IDV and intranight variability was searched for All of these were to assist in confining the universal AGN model The IDV and intranight variability helps constrain the size and mass of the central blackhole The periodicity helps in better determining the physical processes occurring at the center of AGNs TeV stands for a teraelectron volt which is a trillion electron volts 1 2 Photometry 3 1 2 Photometry The nature of the Mk501 campaign required reevaluations of the nature of charged coupled device CCD calibrations Photometry calibrations must be performed on the CCD images to obtain accurate measurements This is usually completed by obtaining intensity readings from the images using computerized apertures or circles defining areas to read centered on the objects of interest However for extended objects like Mk501 t
10. CD s Disc 2 believe If you can t find it here it is 5050 C6396046 Eject TheSky6 DVD Insert the Paramount CD back in and install TheSky6 Update Driver You will notice that there are two functions that we do not use lAClient and AServer These are used if there are going to be other people at different locations using ROVOR If at some point you find that they would be helpful they re here and waiting You will also notice that there is an Optional Software button with features that include PrecisionPEC Automapper and TheSky Pocket Edition The latter is not needed at all for ROVOR We looked into using AutoMapper but found that it was more of a problem than it was worth So just use the directions on how to map and run a TPoint Model that we have put together It is actually quite simple PrecisionPEC would be useful if we were to get a Guider put on ROVOR at any time It costs somewhere around 100 but again it would not be helpful unless ROVOR had the ability to AutoGuide Next thing to do is to check for updates for each of these products To do this you have to go to the Software Bisque website www bisque com On the left side you ll see Support Center Click it Then on the subsequent screen click Free Updates on the right side Sadly you have to actually go to the site every time to get the updates and they actually come relatively frequently So check often If they require you to sign in to their site use t
11. J 198 261 266 1975 6 D Gall and for the VERITAS Collaboration Multi wavelength Observations of Markarian 501 ArXiv e prints 2009 7 J Quinn et al Detection of Gamma Rays with E gt 300 GeV from Markarian 501 Astrophys J 456 L83 1996 8 S R McNeil Ph D thesis Brigham Young University 2004 9 S B Howell Handbook of CCD Astronomy 2 ed Cambridge University Press 2006 68 BIBLIOGRAPHY 69 10 11 12 16 17 18 19 20 R Ierusalimschy L Henrique de Figueiredo and W Celes Lua 5 1 Reference Manual http www lua org manual 5 1 manual html 2008 C Pace BVRI comparison stars near selected very high energy blazars 2010 undergraduate senior thesis A U Landolt UBVRI Photometric Standard Stars Around the Celestial Equa tor Updates and Additions 137 4186 4269 2009 F Murtagh Multiresolution Software Website 2009 Accessed June 2009 A U Landolt UBVRI photometric standard stars in the magnitude range 11 5 16 0 around the celestial equator Astron J 104 340 371 1992 L Heidelberg Knigstuhl 1652 398 MRK 501 http www lsw uni heidelberg de projects extragalactic charts 1652 398 html Accessed April 2009 M Villata C M Raiteri L Lanteri G Sobrito and M Cavallone BVR photometry of comparison stars in selected blazar fields I Photometric sequences for 10 BL Lace
12. The hypothesis is that Mk501 was too small to output a decent bulge profile However to prove the worth and value of the method we imaged an extended object with a larger brighter bulge Messier 81 M81 a large nearby galaxy host to an AGN see Figure 3 6 SBP photometry was previously performed on M81 by McNeil 8 His results for M81 are presented in Table 5 4 along with further analysis regarding the SBP photometry on M81 are found in 85 1 Therefore much to our dismay we had 3 3 Photometry 31 Figure 3 6 M81 as viewed from the ROVOR site It has a highly active nucleus in the center of its grand design spiral nature to resort to performing normal differential photometry on Mk501 A description of our methods is described in this section The ideas behind differential photometry and its usefulness was discussed in the introduction of the Photometry section 83 3 To summarize there are standard stars assumed to be constant in the FOV of Mk501 By imaging Mk501 the other standard stars are imaged too Mk501 can then be compared against the unvarying standards to look for internal variations The standards in the Mk501 field are shown in Figure 3 7 To perform differential photometry one aperture size is chosen for the stars and for Mk501 To be consistent with other work previously performed as well as current col laborator assistance we chose to work with an aperture radius of 7 375 arcseconds We opted
13. The new CCD the Ap8 downloads images in 5 7 seconds Theoretically we could produce about twice as many images on any given night Realistically we obtained about 1 5 times as many images It specifically helped with remotely obtaining flat fields which are constricted by the quickly setting sun Secondly the Ap8 has a FoV of about 23 x 23 arcminutes whereas the Ap47p has a 12 x 12 arcminute FoV The larger FoV enabled more stars in the field for comparison It also allowed images of larger AGN to fit fully on the CCD frame Lastly and I think most importantly the new Ap8 can get colder As discussed in 83 1 2 a cooler CCD has less dark current and smaller pixel fluctuations The Ap8 has no problem holding its own at 45 C while the Ap47p was peaking at 30 C However when the change of CCDs occurred there were some things lost The most significant and important difference in CCDs was the large discrepancy in the gain and readout noise levels In the science that we are performing with ROVOR the larger gain and readout noise contributes very little to the photometric error As stated previously the Ap47p had 2 to 5 times better noise levels In the S N equation these noise levels are wrapped up in the osky terms Therefore if the object s signal 4 2 CCD Imperfections 46 Cobj is large enough a 2 or 5 X spy won t make that much difference For Mk501 Cov gt 50000 counts with some of the comparison stars reaching
14. about this is that there is no danger of breaking a simulated telescope The following exercises will guide you through the process of selecting commands and arguments connecting to the simulated instruments debugging and running your script For help on encoding Arguments be sure to look at pages 11 12 of the Orchestrate pdf help file Read this entire section before beginning Your script should include commands to accomplish the following tasks 1 Set the camera resolution to medium 2 Slew to the following targets in the order they are listed Sirius RA 05h 53m 33s Dec 01d 35m 21s NGC 1976 Azimuth 273d 32m 19s Altitude 24d 26m 10s M 78 RA 04h 42m 25s Dec 04d 19m 39s Betelgeuse Rigel 3 After slewing to each target Wait for 2 seconds then take a 10 second exposure 4 Save your Orchestrate document as Exercisel in c Program Files Sofware Bisque Orchestrate Scripts 5 Make the necessary connections in the Connections menu 6 Run your script lnixJ Your Orchestrate ll Command Arguments document should look 4 SetResolution Medium something like this 2 SlewToObject Sirius WaitFor 2 Takelmage 10 05h 53m 33s 01d 35m 215 A note on connections Be sure that the telescope is set up as a simulator In TheSky6 go to Telescope Simulator and be sure the name is set to NGC1976 2 10 273d32m 19s 24d 26m 10s Simulator For this WaitFor 2 practice exercise you will Takelmag
15. intranight variability was identified every night for Mk501 from April to July 2009 periodic behavior was not satisfactorily detected There seems to be evidence of a 20 day period but the high error bars and relatively low ampli tude intranight variability prevent this possibility from fully solidifying Analytical programs structure functions and Period04 were presented as a way to better identify periodic behavior in Mk501 These should be used to find periodic IDV and intranight variability in future data processing 66 67 Additionally after attempting to use the new SBP photometry technique on Mk501 we concluded that this particular blazar was just too dim and too small to produce viable data This provided useful limitations concerning the SBP pho tometry reductions and permitted follow up observations of M81 to compare with the McNeil 2004 data 8 Between two 2 minute subsequent images of M81 a differ ential magnitude of 0 005 0 001mag was found The FWHM of the residual light matched the FWHM of the image PSF within 0 35 pixels This fits in the parameters of the SBP photometry and proves that SBP photometry should be used to identify low amplitude variations such as IDV Overall the year long campaign will add a huge amount of data to the continuing observations of blazars and particularly Mk501 The data collected will be used to more fully constrain the present AGN models Collaborations with other telescope fa
16. is needed Richard can be reached at rlpearson_the_third yahoo com 82 There is an account on gmail for ROVOR rovor byu gmail com password outhouse there is chatting capabilities on the left hand side Installation Anew 10 Insert Paramount CD Click on the Install Software button Install MKS 4000 USB Driver Install CCDSoft Software Serial Number for CCDSoft should be written on an email to Dr Moody It should just be in the front of the CD case If not here it is for Physics and Astronomy Department Brigham Young University DO68 9F6160FF If CCDSoft cannot find the SBIG Driver and if it does not ask you to download the cll files then go to CCDSoft when it is uploaded and have it open the SBIG Driver Checker If CCDSoft cannot find the SBIG Driver and if it does not ask you to download the dll files then go to C gt ProgramFiles gt SoftwareBisque gt CCDSoft gt CameraDrivers gt SBIG when it is uploaded and have it open the SBIG Driver Checker Download and then Update Now this is only if you need the SBIG dll files Install TPoint Software The Serial Number for Tpoint is written on the CD brilliant whoever did that me just in case for Physics and Astronomy Department Brigham Young University 1066 FE266053 Install Orchestrate Eject the Paramount CD Insert TheSky6 DVD it should be just one DVD there are 4 CD s if you wish to go that route Again the serial number should be written on one of TheSky6
17. optical intraday variability in a complete sample of radio selected BL Lacertae objects Astron Astrophys 305 1996 M Jang and H R Miller The Microvariability of Selected Radio Quiet and Radio Loud QSOs Astron J 114 565 574 1997 G E Romero S A Cellone and J A Combi Optical microvariability of southern AGNs Astron Astrophys Suppl Ser 135 477 486 1999 A C Gupta J H Fan J M Bai and S J Wagner Optical Intra Day Vari ability in Blazars Astron J 135 1384 1394 2008 K K Ghosh B D Ramsey A C Sadun and S Soundararajaperumal Optical Variability of Blazars 127 11 26 2000 BIBLIOGRAPHY 71 30 31 3 32 33 34 35 36 37 38 39 K K Ghosh C Kim B D Ramsey and S Soundararajaperumal Optical Microvariability of Blazars Journal of Korean Astronomical Society 34 9 15 2001 A C Gupta W G Deng U C Joshi J M Bai and M G Lee Multi color optical variability of the TeV blazar Mrk 501 in the low state New Astronomy 13 375 384 2008 A A Jensen Structure Function Analysis of Blazars AP Librae and 3c279 2008 Undergraduate senior thesis accessed March 2010 J H Simonetti J M Cordes and D S Heeschen Flicker of extragalactic radio sources at two frequencies Astrophys J 296 46 59 1985 P Lenz and M Breger Period04 User Guide
18. or brighter The reason for this is to avoid confusion the greater proportion of dimmer stars makes it easier to confuse them This is one of the drawbacks of this method Because there are only so many bright stars it would be hard to get a model with a huge number of data points As the stars march westward over the course of the night you can map the same stars over and over again because their altitude and azimuth have changed As you add points to your model a little numbered red line appears on TheSky6 This line is like the crumbs that Hansel and Gretel left behind them to find their way out of the forest This red line will help you visualize where you have mapped and where you have not You can read more about this line on pp 24 25 of the TPoint manual 3 Slewing the scope Click on the star in TheSky6 This brings up the Object Information dialogue box Along the bottom B CAR A U E right hand side of this box is an icon of a little green telescope Click on this icon You will be asked to confirm the slew Click ok to confirm The telescope should start slewing Before you can use TheSky6 to tell the scope where to point there are a few things you need to make sure of First make sure TheSky6 is connected to the telescope To do this click on Telescope Link This EEN TRE DONE should establish a link between the scope and IS CLOSED TheSky6 If it doesn t check the troubleshooting OR ONLY PARTIALLY section OPENED
19. the connections Make sure TheSky6 and CCDSoft are open Also be sure that TheSky6 has a Connect to TheSky A connection to the telescope and CCDSoft is connected to the camera and filter wheel Next click on all of the connections under the Connections menu to make the necessary connections You should now be able to run your script Orchestrate Orc3 v Camera Filter Wheel Once you have completed your script and made the proper connections now you need to try and run it There are two buttons that will execute your script one is Step and the other is Run These buttons are found just below Macro on the E E x toolbar From left to right these buttons are Run Step Stop and Abort The Stop and Abort buttons are only available while Orchestrate is running a command In other words the stop and abort buttons are grayed out until you hit either the Run or Step button The step command will only execute the currently highlighted command and the run command will run the entire script If you need to stop the script hitting the stop button will stop the script after it finished the command it is currently working on The abort button stops everything in its tracks it is like the big red button on a band saw that you hit when you want it to stop right now Now it is time for some practice You will have to know how to use Orchestrate to run a simulated telescope before you attempt to run the ROVOR telescope The nice thing
20. the standard magnitudes are used in transforming the instrumental magnitudes You may also remember that the secondary standards in the same field of view as the object have been calibrated in order to transform the object s magnitude These secondary standards normally carry even higher errors than the first extended standard fields Some of the sec ondary standards have errors up to 0 09 mag This discrepancy results from smaller telescopes performing the secondary calibrations a lack of observations and or a lack of plentiful photometric observations Regardless care must be used in calculating and tracking error throughout the photometry process Chapter 5 Results The Mk501 campaign was designed to locate and identify variations in this particu lar blazar Three types of variations were searched for intraday variability IDV intranight variation and periodic behavior For this campaign we obtained many images in one night in order to find IDV We also to monitored Mk501 for a number of months to find both the intranight and periodic variations The following chapter outlines the observations of Mk501 over the past year and describes the conclusions drawn from the observations 5 1 Observational Data ROVOR was able to observe Mk501 for over 60 nights The nights varied in quality ranging from very photometric to fairly variant seeing but because of differential photometry the data is still usable just with higher errors
21. the telescope to change sides The telescope is technically a RCOS Carbon Truss 16 inch Ritchey Chr tien Telescope 2 1 Hardware 7 Figure 2 2 A view of the RC Optical System telescope used at the ROVOR site 2 Slight mechanical of the meridian when it tracks objects through the meridian pointing limits are acquired with this feature but the software packages eliminate this drawback A charge coupled device CCD and filter wheel are attached at the back end of the telescope Two CCDs have been used the past two years First was a 1k x 1k Apogee Ap47 camera which provided a 0 7375 arcsec pixel resolution and a 12 x 12 arcminute field of view FOV The current CCD is a 1k x 1k Apogee Ap8 camera in a Finger Lakes Instrumentation FLI body and gives 1 35 arcsec pixel resolution and a 22 x 22 arcminute FOV In front of the CCD six filters are held in a FLI filter wheel B V R I wide and narrow Ha A description of the sensitivity of each filter can be found in Dr J Ward Moody s ROVOR files The meridian is the imaginary line that runs directly through due North to due South through the zenith The zenith is the imaginary point directly overhead 3For the project described here the Ap47 CCD was used in a binned mode or a binning of the pixels 2 by 2 This essentially doubled the plate scale from 0 735 to 1 475 arcsec pixel Though it limits the minimum spatial resolution it was sufficie
22. 9 0 Woe O SSEW OY IJEWIJS ULI IM Cp p 1Y UONIUY P snipes PTIJISZIEMIIS ot SUIS y p pi0 y p p 1y OS T 9 809 3481 9 809 tp a P Pp Y psp I9AI9SQO O snipe ryds y Figure A 2 This is the slide that followed the previous one It goes through the derivation of the radius and mass limitations Appendix B Troubleshooting with ROVOR e If Orchestrate chokes through the night while following an object through the meridian the problem is most likely that the object was within the meridian limits of the telescope Those limits are displayed in theSky as a blue and red bang on either side of the meridian A couple of things can be done to avoid or fix the problem 1 Plan the script to avoid any object within about 20 minutes of the merid ian on either side that 20 minutes is actual time not arcminutes This is the easiest fix just avoid the problem 2 Orchestrate can take a list of enumerated errors and disregards them if they occur in the script You can search for details in the Orchestrate help file and on the Software Bisque website www bisque com Simply there is a file with errors listed for Orchestrate to ignore Using this shouldn t hurt ROVOR it is fairly effective too e If CCDSoft freezes during an image download or if you accidentally aborted the download of an image you can attempt a number of fixes in the following 76 77
23. Communications in Asteroseis mology 146 53 136 2005 S NASA Hubble Essentials Accessed March 2010 R D Crisp Richard Crisp Homepage 2009 Accessed September 2009 R D Crisp in Residual bulk image characterization and management in CCD image sensors E Bodegom and V Nguyen eds SPIE 2009 No 1 p 72490L J R Janesick Scientific Charged Coupled Devices 1 ed SPIE The Interna tional Society for Optical Engineering 2001 H L Johnson and W W Morgan Fundamental stellar photometry for stan dards of spectral type on the revised system of the Yerkes spectral atlas As trophys J 117 313 1953 BIBLIOGRAPHY 72 40 A J Barth L C Ho and W L W Sargent The Black Hole Masses and Host Galaxies of BL Lac Objects In Active Galactic Nuclei From Central Engine to Host Galaxy S Collin F Combes amp I Shlosman ed Astronomical Society of the Pacific Conference Series 290 601 2003 41 J H Fan G Z Xie and R Bacon The central black hole masses and Doppler factors of the gamma ray loud blazars 136 13 18 1999 42 D Merritt and L Ferrarese The Mo o Relation for Supermassive Black Holes Astrophys J 547 140 145 2001 43 F M Rieger and K Mannheim On the central black hole mass in Mkn 501 Astron Astrophys 397 121 125 2003 44 F M Rieger and K Mannheim Implications of a possible 23 day peri
24. Ip putag GI das yoro ur Y dn umg TI ovyujyp 0T N vy oqur perq Poyuyy Vu paysnfpe mou oq AUO OL S IeJ9p 910UL 10 T 8 ves SdqS 0M 94 Jo 3931 o8 Nq oy USITY 6 das ou St sty Seur yoes 10 snipes sa Vu ayy 9 dogg Wo songea gold 8 9SBUIT puov s 10 Y SADIS ULIOJI9 Y y VJ 80 6 Z oS sniper yore 103 Vu vore opnyuseul oy oynduroy 9 SULI snjnuue jo gore Aq y dogg Wo nea IPMA V eore uue syus 973 oyETNITeO Q T 8 998 syunoo I uur y ZULIJqNS Aq ST Oq 2 snqnutre squs ayy agemo v sjoxid ZT 03 Jo sniper wo Arrens SOILD OLIYWUOIUOD optsut J SsyUNOD yor SASd Soseurl eN G IBVUII I SUIS 10 PIUIQUIOI e aS JUIBIJ VUIIOJOI Y UO PPA T JUQUUUUIO uoraididosa days 3 3 Photometry 22 10000 Lar h Pixel Value ADU Pixel Value ADU 5000 Radius pixels a A radial profile of a star s PSF b A 3D image of a star s PSF Figure 3 2 Images of a single star s PSF An average of many star s PSF across the whole CCD becomes the image s PSF The PSF is recorded as a FWHM of a gaussian profile particular Mk501 project by combining 60 images from May 28 2009 Step 2 Once the reference frame decision has been chosen the next step is to make the two images PSF the same The significance of this step cannot be fully impressed The PSF determines how much the incoming light is spread onto the CCD A point source a regular star crea
25. MARKARIAN 501 MONITORING CAMPAIGN USING ROVOR by Richard L Pearson III A senior thesis submitted to the faculty of Brigham Young University in partial fulfillment of the requirements for the degree of Bachelor of Science Department of Physics and Astronomy Brigham Young University April 2010 Copyright c 2010 Richard L Pearson III All Rights Reserved BRIGHAM YOUNG UNIVERSITY DEPARTMENT APPROVAL of a senior thesis submitted by Richard L Pearson III This thesis has been reviewed by the research advisor research coordinator and department chair and has been found to be satisfactory Date J Ward Moody Advisor Date Eric Hintz Research Coordinator Date Ross Spencer Chair ABSTRACT MARKARIAN 501 MONITORING CAMPAIGN USING ROVOR Richard L Pearson III Department of Physics and Astronomy Bachelor of Science The remote telescope facility of Brigham Young University ROVOR observed Mk501 a variable blazar for over 60 nights These observations took place from April 2009 to March 2010 Possible microvariability At 10 to 30min was found in three nights thereby limiting the size and mass of the central black hole within Mk501 No periodic behavior was identified but intranight variations occurred daily SBP photometry a new reduction method was first used to analyze the Mk501 data However the method requires a larger brighter object M81 was therefore observed and SBP photom
26. There have been a number of statistical tests produced for analyzing variability in AGNs However only the variability confidence parameter was used in this Mk501 campaign Nevertheless the rest are briefly described here so that when future analy sis of the data occurs these tests will be readily available The tests were constructed to look for intraday variability IDV or microvariability over an evening of obser vation Three of the most significant are the variability confidence parameter IDV amplitudes and structure functions of electrons per pixel from the readout noise Howell expresses that each form of the S N equation reveals the same conclusions so each is alright to use This was just to help in understanding of the other errors associated with the S N equations 4 1 Overview 39 Variability Confidence Parameter The variability parameter applied to AGN is described in a number of sources 25 28 This parameter which we will define as is defined as a level of confidence in the IDV Basically it finds the variation of data points throughout the night in relation to the error or the a It assumes that differential photometry is being used this means that there is a designated comparison star and a check star see 83 3 2 By definition the parameter is eae 4 4 O Astars The cam is the observational error in the differential magnitude of Mk501 or any AGN as defined in 3 3 2 The CAstars is
27. ackground of TheSky as TheSky operates the mount and telescope If for some reason the TPoint model does not seem to correct the pointing error enough SB also produces an additional program to take out periodic error fluctuations for an additional 100 00 called Precision PEC We have not found the need to purchase this software TPoint has performed excellently The commercial off the shelf software from SB continues to operate ROVOR Since working out most of the kinks in the communication and hardware interface the software has given little resistance to nightly observations Still further improvements 4This idea sprouted in March 2010 with Bret Little currently and undergraduate He was at the head of the interface construction 5If in assembling the mount and telescope the alignment was off by only a half a degree the error in the telescope pointing would equivocally be 30 arcminutes thus the object of interest would most likely be found outside of the telescope s FOV 2 2 Software 10 can be provided a GUI wrap around program for Orchestrate or a purchase of SB s Precision PEC are a couple of suggestions 2 2 2 LabView Programs LabView is a very powerful and useful graphical programming language Before the renovation of ROVOR which upgraded ROVOR to the features found in 82 1 and 82 2 students put together a main control terminal for the original mount and telescope using LabView Currently LabView
28. adjusted magnitude conversion into counts is complete the count values for each image are summed up separately CnAref and gt Cnaooj Basically this is summing up the total counts under the SBP curve Remember to add together the CrnAref for the reference image instead of the unadjusted counts For the SBP example Figures 3 4 and 3 5 the totals come out to be gt Craref 2659 37 and 3 3 Photometry 30 Y Craomj 2647 27 Step 12 Finally we are now able to find the differential magnitude between these two images To do so we use the total counts found in Step 11 from each of the images in the following equation AM 2 5log Spee 3 12 nAobj The derivation of this equation can easily be reproduced but will not be done here For your interest the SBP example that we have been following produces a value of 0 00495 this means there was a 0 00495 difference between the object and reference frames Care must be taken to understand the true meaning of the negative sign or positive sign as the case may be If it is easier to work in magnitudes without the negative sign in front of the 2 5 as in Equation 3 10 then take off the negative sign Just remember that you did that when playing around with other numbers and data 3 3 2 Normal Differential Photometry Before continuing a confession is necessary After months of trying to make the SBP photometry method work with Mk501 the errors were unacceptably large
29. al problem There is a tightening locking screw 78 on the top part of the mount that permits the telescope to rotate but not slide from position If that is too tight it will not allow the telescope to rotate at all We found this out when the temperature variations between day and night were large At night cooler the mount would contract and the telescope would not be able to find home If theSky gives an error saying the scope could not find home you must manually release the tightening screw have fun driving down to Delta When you get down there the telescope should be beeping That s its signal for not being able to move If it is not beeping however there is probably something else going on and I have no idea what that would be If for some reason Orchestrate errors on a Take Image command a number of sources of error are available 1 Orchestrate is not connected to the CCD 2 CCDSoft is not connected to the CCD 3 The arguments for the Take Image command were incorrect 4 CCCSoft froze during a download see above 5 CCDSoft tried to save the file but the disk space was full Recently we have had problems with incredibly slow internet connections When logging into the computers down at the ROVOR site through the VNC connection the screen reloads every ten minutes After further investigations and a talk with HughesNet the internet satellite provider Bret Little was able to access a
30. ample of his work The errors in these expanded fields still tend toward the 0 01 mag level Therefore when using absolute photometry in addition to the error in collecting the data the S N or SNR value an error in comparison is added Using the knowledge of propagating errors 4 1 3 a total error Oo can be calculated 2 2 2 Otot TsnR obj O standard 4 7 Here the first term is the S N of the object of interest from Equation 4 3 and the second term is the error associated with the transformation using the standardized fields For further information regarding these transformation errors which typically come out to 0 01 0 03 mag see the senior project completed by Cameron Pace 11 The resultant 0 falls around 0 02 0 04 mag The error value usually satisfies projects looking at long term variability However when searching for variability on the timescales of minutes to hours where the variability is on the order of 0 02 mags the high Giot values will be no good Another solution is needed 4 3 Photometric Error 49 4 3 2 Differential Photometry Error Differential photometry is a useful alternative to absolute photometry It uses the magnitudes of the stars surrounding the object of interest on a CCD frame By finding the difference between the stars and the object of interest a differential magnitude is obtained for the object of interest Also stars within an object of interest can be standardiz
31. at DSa aE 3 3 1 Surface Brightness Profile SBP Photometry 3 3 2 Normal Differential Photometry io Varas Se ee ae Error Analysis A MONET VEW e Bee bot a Sah Se a oh here a bea he ae FS ne 4 1 1 Standard Deviation do A ane AR ae A Bla ae 4 12 Signal to noise Ratio 4 4 al poder ane dk ae pa oe 4 1 3 Propagation of Error 2S e a She Sea 4 1 4 AGN Microvariability Tests ur soto it ia ir Sek 4 2 CCD Imperfections 10 da ee ee da 4 2 1 Gain and Readout Noise oboe osas ei vi viii wwe EH oO o oo un EH 13 13 14 15 17 18 19 30 CONTENTS vii 4 2 2 Ghosting and Quantum Hysteresis 46 4 3 Photometric Error 4 4 2 4 5 4 ach eee Bi ee BS AE ER 47 4 3 1 Absolute Photometry Error 4 47 4 3 2 Differential Photometry Error 49 5 Results 51 Sl Observational Data e vieras e a aa e BS 51 5 2 Data Analysis de a io eG ht e e hee RS 97 D2 IOV Dataa a A Oe eae ee 58 52 2 Intranet Variability dedo a Bo 59 5 2 3 Penodie Variability ts ace eke a bd ae bo a Be 63 524 MEL Pate a gee eh SS hae hed a E 63 6 Conclusion 66 Bibliography 68 A Blackhole Radius and Mass Derivation 73 B Troubleshooting with ROVOR 76 C ROVOR Weather Station 80 D Customized User Manuals 82 Index 103 List of Figures 2 1 2 2 3 1 3 2 3 3 3 4 3 9 3 6 3 7 4 1 5 1 5 2 5 3 5 4 5 9 5 6 5 7 A l A 2 EOS io ad Dd A Eee Gab eee Ge Si es feat a 6 ROWV OR
32. bout some vocabulary words Selecting terms for the TPoint model requires a decent understanding of statistics Rather it requires a decent understanding of a few statistics words Root Mean Squared RMS The RMS basically tells you how well the scope will point at something For example if the RMS is 40 arcseconds that means the scope will usually point to within 40 arcseconds of your intended target You will want the smallest RMS possible ideally less than 30 arcseconds Value This is a measure of how important a given term is to the TPoint model The higher the value the more that particular term is contributing to the model A term with a small value or a value that is smaller than its sigma should not be used Ideally all terms you use should have a value that is at least three times greater than the sigma Sigma This is the measure of the error of individual terms A term with a large sigma has a large standard deviation That means TPoint can t really tell if this term is helping the model at all Do not include terms with a large sigma even if they improve the RMS Population Standard Deviation PSD This is a measure of the Sigma of the whole group of terms You want keep this value as small as possible The basic idea of making a TPoint model is to statistically improve the pointing of the telescope Your task in creating a TPoint model is to minimize the RMS The terms you add will account for different mechan
33. cilities is highly recommended especially to encourage simultaneous observations of Mk501 in the X ray y ray or even radio regions of the electromagnetic spectrum These simultaneous observations will show correlations between various regions and help solidify central blackhole theories The contents of this thesis were to guide students through the operation reduction and calibration processes associated with ROVOR and AGN research This is by no means a complete collection of everything one would ever need to know concerning these topics However it was meant to simplify the learning process A significant amount of time was used in coping with the nuts and bolts of a remote telescope facility and learning how to use the reduction tools in their most effective ways This was to pass the knowledge on to others so that scientific progress can more readily move through BYU s Physics and Astronomy department Bibliography 1 F Cain How Many Galaxies in the Universe 2009 Accessed April 2010 2 E P Hubble Extragalactic nebulae Astrophys J 64 321 369 1926 3 O D A Carroll Bradley W An Introduction to Modern Astrophysics 2 ed Pearson Education 2007 4 C K Seyfert Nuclear Emission in Spiral Nebulae Astrophys J 97 28 1943 5 M Ulrich T D Kinman C R Lynds G H Rieke and R D Ekers Nonther mal continuum radiation in three elliptical galaxies Astrophys
34. d using the SlewToAzAlt command If you are using the SlewToRaDec telescope for research you will probably want the resolution arar A set to High If you do not give Orchestrate a command to SlewToAsAlt set the resolution it automatically sets it to a 1x1 bin ratio SetTrackingh ate Jog Don t feel like you have to memorize all of these commands like in some CS class There is a handy list of commands you can pull up To access the command list you must have an Orchestrate document open Then go to View Command List You can highlight a cell in the Command column then double click on one of the commands from the Command List to insert it into that cell As mentioned earlier all commands require some sort of modifier in the Arguments column for the command to run After you insert a command into the command column hit the right arrow or click in the cell in the Arguments column and insert your argument See pages 11 art O xj 12 of the Orchestrate pdf help file for a list of expected A arguments SetT emperature SetResolution SetFrameMode SetShutterStrobing WaitFor WaitUntil ImageThenSlewT o AutoM ap AppStart AppStartup There are a few things you must File Edit View Options Connections Data Macro Window Help do before Ci co fel _ 83 6 ASS l attempting to ops use your script First make sure that Orchestrate is connected to TheSky6 as well as CCDSoft Goto Connections to see
35. d also show up on the darks This did not eliminate the problem but it helped reduce its effects during photometric calculations By analyzing the dark and bias images we concluded that the RBI did not horribly 3The Ap8 is discussed in a minute Just hold on 4 3 Photometric Error 47 effect our results leaving 10 s of counts this is on the order of the gain and readout noise The ghosting on the Ap8 was quite similar to the Ap47p Therefore we concluded RBI to be an additional but not significant source of error for both CCDs As for quantum hysteresis QEH defined as the random or non linear variations associated with the signal transfer the Ap47p may experience some symptoms 38 We did not spend time on descriptively identifying the complete QEH signature for the CCDs Rather when we saw that the RBI error was sufficiently low as well as the gain and readout noise the effects of QEH were ruled insignificant as well Again it is something to take note of for future reference 4 3 Photometric Error Even if all of the instruments and optics on a telescope system are perfect errors in observation and calibration factor in These errors are caused by atmospheric effects seeing transparency etc and pure error limitations The latter is a built in error to the standardization process as described in 3 3 However differential photometry generally provides lower errors as compared to absolute photometry Furt
36. d though that it is important to know how to do it this way This is just like long division you know how to do it so you can still divide when you don t have a calculator When large pointing errors appear such as when mapping the first six points or crossing over the meridian this technique may be the only way to go The automapping feature of method two cannot handle very large errors and when it fails you will have to do it manually as described above Let s look at each of the above listed steps in detail 1 Insert a TPoint object into TheSky6 From TheSky6 click Edit Insert New Object TPoint Model then click ok You have to insert a TPoint Model object into TheSky6 This allows TheSky6 to enter the mapping data into TPoint 2 Choosing a star If you have never mapped a star before don t worry about the following stuff Just pick a bright star and get started with the next step As mentioned above you have some decisions to make after mapping the first six stars First it is a good idea to map a bunch of stars on the west side of the meridian before hopping over to the east side There are a couple of reasons for this If you flip over to the other side of the meridian too early TPoint will have trouble modeling your data points Richard and did this once and the pointing error kept getting bigger and bigger There was no hope of salvaging or even understanding the model so we had to start over after two or three hou
37. d times filename saving and temperature limitations Orchestrate is the scripting software for all ofthe SB applications In Orchestrate itself one compiles an automated script which sends commands to the telescope cam era and filter wheel There are many nuances particular to this software and many 2 2 Software 9 of them are included in Appendix D for reference But even though it is not the eas iest and most user friendly package it has helped make ROVOR a truly remote and somewhat robotic telescope To help create nightly scripts an additional graphical user interface GUI is being constructed The GUI will wrap around Orchestrate and run it in the background Through it the user will be able to more easily con struct nightly scripts to be sent to ROVOR The TPoint software is used quite rarely but holds some of the most important information regarding the operation of the telescope The sole purpose of the TPoint package is to correct telescope pointing errors One may think that if great care were taken to properly align the mount and balance the telescope then no corrections should be needed However the guiding precision necessary for untrailed images is one part in 15 million and no gears are machined that precisely TPoint is used to analyze and remove this pointing error by building a TPoint Model Details on producing a model and its use can be found in Appendix D Once the model is completed it runs in the b
38. de variations obtained using surface bright ness profile photometry The two images dates are shown as well as the number of days between the images and their associated variation Data retrieved from 8 behavior Reference to the Mk501 data will be in regards to the R band data unless otherwise noted Also the data collected on M81 is discussed here 5 2 1 IDV Data Large amounts of R band images of Mk501 were collected on several nights The high volume of images was to help look for IDV Microvariability or intraday variability in Mk501 has been seen at timescales as short as 12 minutes with a magnitude variation of 0 13mag 29 A variation of 0 05mag was found over 15 minutes as well 28 The goal of the ROVOR campaign was to further reduce the temporal resolution timescales of the IDV The significance of finding IDV helps put limits on the size of the emitting region at the nucleus of Mk501 Therefore the smaller the temporal resolution the more limited the model There have only been 3 nights that have displayed possible IDV activity Those nights are 7 May 11 May and 6 July 2009 and are shown in Figures 5 4 5 5 and 5 6 respectively Table 5 5 shows the information regarding the IDV associated with these nights As can be seen there is quite a lack of confident IDV values The confidence Details regarding the derivation of this limitation is found in Appendix A 5 2 Data Analysis 59 parameter is le
39. ds multiply the aperture radius by the plate scale of the CCD For example in the case of the Ap8 CCD that is currently down at ROVOR 3 3 Photometry 26 radii of 1 pixel 1 475 2 pixel 2 95 3 pixel 4 425 and so on Then the annulus area A is just AE 3 6 An 7T r rz 3 7 where r is the radius of the large aperture and r is the radius of the small aperture Then Equation 3 5 can be used with Equation 3 7 to find the counts area in each annulus Cr Cna 3 8 1 2 3 8 C C Cra gt 2 3 9 n r r5 3 9 The resultant values for each annulus are used to create the SBPs after converting them into magnitudes as shown next Step 6 To create the SBPs we use the values found from Step 5 to compute a magnitude Using the magnitude relation of Mna 2 5log Cya 3 10 where mpa is the magnitude area 1 calculated from Cna the values computed in Equations 3 8 and 3 9 McNeil uses the mag area 1 reference many times in his thesis so we continue to use it here for clarity 8 l sec7 He divided the annulus counts by their In actuality McNeil found a magnitude area area and by the exposure length of the image This may have been because of varying exposure lengths between the comparison images However with the research presented in this senior thesis the comparable images had the same exposure the division of the exposure time has been r
40. e 10 also want to make sure SlewToObject M78 that the Impose slew WaitFor 2 limits checkbox is Takel 10 imi akelmage unchecked Similarly SlewToRaDec D4h 42m 25s 04h 19m 39s WeaitFor 2 gt make sure CCDSoft is running a simulated camera In CCDSoft go to Camera Setup and be sure it is set on Simulator After the SetResolution command your script should be a repetition of three basic commands a slew command a Wait command and a Takelmage command If you mess up entering in the RA and Dec coordinates it is impossible to go back and edit one of the values you have to re enter the entire string of coordinates Also if you want to insert a new command line between commands click on the bottom command so it is highlighted the go to Edit Insert to insert the new command line One final note In the above screenshot you will notice that the argument for command number 17 has 04h 19m 39s as the Declination coordinates This is different from the 04d 19m 39s that was specified in the problem The SlewToRADec command expects the first Dec coordinate to be given in degrees not hours If you enter it in hours instead of degrees the telescope will not slew where you expect it to Have a practice problem like 1 reproduce the following screen One of the commands should give an error message 2 make all of the proper connections and execute your script 3 figure out what went wrong with the command that threw an error me
41. e CCDSoft to take an exposure of the star Go into CCDSoft In the Take Image tab of the Camera Control dialogue box set Bin to 2x2 In the Exposure portion of the box be sure Minutes is set to 0 and set an appropriate number ofseconds maybe 1 or2 Now click the Take Image button to take a picture M Camera Control 5 x Setup Take Image Focus Tools Autoguide Color AutoSave It really isn t that hard to take an image Just be Exposure Subframe r Bin Imager Minutes jo M On Size 2x2 C Autoguider sure you have the correct exposure time Seconds 2 000 z p Image MN El _Iske image entered in Also be sure Frame Lig a Delay s 0 00 Reduction fAutoDak Abort CCDSoft is connected to O the camera These i exposures we are taking Fiter 7 I Tonew window for mapping purposes don t have to be professional quality or Device Linked Status Temperature Shutter Filter Max 7 Imager Yes Ready 0 0 0 Closed 100 anything You will have Autoguider No to play around with the exposure time to get it right Sometimes you need a couple seconds other times a tenth of a second is enough You should definitely set the Bin to 2x2 This will lower the resolution but for these pictures you don t care about that Binning 2x2 will make the download time a lot faster This method is already time consuming and you will want to cut every cor
42. e average difference For the example SBP i e Figure 3 4 there was an 1 needed This is denoted as a Am and was adjustment of 0 039237 mag area applied to the reference image as seen in Figure 3 5 Note that it does not matter which SBP is adjusted for consistency the reference image is chosen to be adjusted to match the object image And remember the adjustment is not just for the bulge it is applied to the whole SBP including the nucleus region of the SBP These new adjusted values are denoted as MnAadj 3 3 Photometry 29 8 a 7 s 6 a a 2 5 g a a 3 a a 2 a B a Am 0 039237 mag area 0 02 04 06 08 1 12 14 Log Radius log arcsecond a Adjusted Reference Object Figure 3 5 The SBP of the reference image was adjusted by the displayed Ama As compared with Figure 3 4 the bulge light is almost perfectly in line This is again a semi log plot with the x axis a log radius used to better show small variations Step 10 Now that the SBPs have been aligned and matched we convert the adjusted mag 1 area values the MnAaqj from Step 9 back into counts area for each annulus To do this we use the inverted relation of Equation 3 10 107 4Mnacdj a CnAref 3 11 where the MpnAaqgj is the adjusted m a values the value with the additional Ama Therefore the reference frames adjusted counts area is denoted by Crare f Step 11 With the
43. e temperature gets to about 100 C the thermal noise due to heat is no longer present and dark frames are not used I strongly suggest to begin working the CCD at 45C Though it is not as cold as 100 C diminishing the dark variance would help improve the quality of the resultant data 3 1 3 Flats The flat frame or flat field is designed to find out how each individual pixel varies when exposed to light The CCD manufacturer tries and the cost per CCD is usually directly related to the quality of the CCD to make each pixel react the same to incoming photons However imperfections exist and must be handled hence the flat field The idea behind the flat field is to expose every pixel on the CCD with the same 3 1 Calibration Frames 16 number of photons or the same intensity of light If you know that each pixel is receiving the same amount of light then the variations from pixel to pixel will be a measure of their sensitivity Some pixels will be brighter and others will be dimmer By normalizing the frame each pixel will have its own weight factor in Equation 3 1 H is the pixel s weight factor If the bias dark subtracted image is then divided by the weight factors pixel to pixel the real light from the objects is found Z D YS 3 4 3 3 4 There are further calibrations completed on the desired science objects during the photometry process but those only come after the comp
44. ed and used as further extensions from the original standardized fields see previous section This allows astronomers to work on non photometric nights because variations across a single CCD frame are assumed to be constant This allows for more obtainable even if slightly worse data By using differential photometry one generally lowers the error associated with the final product Therefore it is used to find small amplitude variations in objects such as Mk501 The error is depleted because differential photometry removes the intrinsic standard error the Ostandard in 4 3 1 In other words the standard values and error are not used to re calibrate the object of interests instrumental magnitude Instead the standard stars instrumental magnitudes and S N errors are compared with the objects instrumental magnitude and S N error Hence the total error Gor is now oe sn Roo F ha 4 8 where sy r o is the S N of the object of interest and stars is the combined S N values of all the stars used for comparison against the object Typically one star is chosen as a comparison star and a second star chosen as a check star The instrument magnitudes of the object and comparison star are used for the differential photometry and their errors are used in the equation above to find a Giot for that frame The 4Tf a group of stars are used to compute a comparison magnitude the Ostar will be the average of each star
45. ed study of these sources should help in better comprehending the usefulness of the structure function in searching for microvariability and periodicities The simple structure function equation is SF lt F t F t dt gt 4 6 where F t is the flux of the object at time t and F t dt is the flux of the object at a later time dt lt gt denotes the average of the inner quantity This function can be used for each night s data as well as the data over a large number of days Examples of structure functions are displayed in Figure 4 1 There are many important features that show up in structure functions A list of these features from Jensen 32 are as follows e Maximums in the SF are characteristic timescales A characteristic timescales is the minimum time intervale in which microvariability can be detected e If there is a periodicity in the lightcurve of an AGN then in the SF it is displayed as a minimum e A SF can show multiple maximums and therefore multiple characteristic timescales Also only one maximum can be seen in the SF if there happened to be a bright flare in the data Be cautious to fully rely on the SF to find the characteristic timescales e If calculating the SF for a series of nights the SF will display the aliasing or windowing of the data For instance if data from 10 consecutive nights are used to calculate a SF the SF will show a periodicity of about 1 day or the time separat
46. emoved Mathematically the exposure division was simply a constant magnitude adjustment mag area 2 5 log mtsdann __1_ _9 5 Jog lt tts 22 4 9 5 log exposure Hence the exposure was left area exposure area out for simplicity s sake 3 3 Photometry 27 9 9 8 4y 8 y 7 a 7 n a B 6 6 g a i a E 5 E 5 a ER E E 4 a 3 x 8 y 3 3 y B 2 5 2 L y m w 8 1 1 0 7 0 i 10 15 20 D 02 04 0 6 0 8 1 12 14 Radius arcsecond Log Radius log arcsecond m Reference Object a Reference Object a b Figure 3 4 Examples of a SBP of Mk501 Notice that for magnitudes the more negative the brighter an object Hence the y axis has been flipped to be consistent with brighter being up The second image is a semi log plot with the z axis a log radius Step 7 If you have not already done so repeat steps 3 through 6 for the second image Both images values from Step 6 will be used to perform the final differential photometry Step 8 We are now ready to create the SBPs for each image the reference and object image Both SBPs are plotted on the same plot To form a SBP the m 4 values from Step 6 are used on the y axis with the associated radii in arcseconds on the x axis Figure 3 4a shows an example of an SBP completed on Mk501 Figure 3 4b is a plot with the log radius along the z axis This second image shows the d
47. ere were two different CCDs used for the Mk501 campaign and each of those cameras have their own gain and readout noise levels The measurements for each of the CCDs are presented in Table 4 1 4 2 CCD Imperfections 43 MRK 501 S4 1749 701 a S4 17494 701 b S5 1803 784 a S5 1803 784 b 3C 371 a log SF NV 0 a 40 A PKS 2005 489 s5 id BL Lac Las PKS 2240 260 PKS 2254 074 a PKS 2254 074 b fi 2 1 0 1 2 log At days Figure 4 1 This shows examples of structure functions SFs of many blazars Note the maximums and minimums of each function The SF for Mk501 is second from the top Image obtained from Figure 5 in Heidt and Wagner 25 4 2 CCD Imperfections 44 Gain MReadnoise Filter e ADU e ADU Ap47p old B 2 86 5 18 V 1 69 3 07 1 15 2 09 Ap8 new B 3 46 11 89 V 4 52 15 54 R 3 01 10 36 I 4 50 15 48 Table 4 1 This table displays the values of the gain and readout noise as sociated with each of the CCDs used in the Mk501 campaign Additional details regarding each CCD is found in the Hardware section 2 1 The ADU is another term for counts or even photons These values were calculated by Professor Alberto Sadun a collaborator from the University of Denver Email correspondences are found in the rovor byuQgmail com address The gain is simply a scale factor used by an analog to digital A D converter The scale factor is the
48. ertaining to variable galaxies and an overview of the project at hand are found below 1 1 Active Galactic Nuclei and Markarian 501 Active galactic nuclei AGNs describe a category of galaxies that are host to an active core or nucleus This active core emits high energy radiation Seventeen years after the first discovery of an active galaxy in 1908 by Edward A Fath Carl K Seyfert reported that a small percentage of galaxies have very bright nuclei 3 4 1 1 Active Galactic Nuclei and Markarian 501 2 Since then further clarifications have been made to the Seyfert galaxies including the AGN The AGN model consists of the active core being driven by a supermassive black hole with an accretion disk swirling about it The action of the accretion disk perturbs the material enough so that the closer it gets to the blackhole the hotter it becomes It is this material that is responsible for the high energy emission in AGNs AGNs have been further divided in terms of the direction the emission jet faces with respect to the earth A blazar is a group of AGNs that has the jet facing directly toward the earth The elliptical galaxy Markarian 501 Mk501 falls into this category 5 It is one of the few known TeV blazars and is relatively nearby z 0 034 6 7 Mk501 was chosen as the target of interest for a year long campaign It is relatively bright 13mag and is accessible in the sky for most of the year RA 16 53
49. es error from each night gt y l0snr so TSNR tar 3 3 2 explains the error associated with each Mk501 image Obviously the larger the amplitude the more confident you will be that there is in fact IDV occurring The same is true for a high confidence value However caution must be taken to closely look at the differential magnitude light curves for each night Many AGN scientists not only plot the differential magnitudes of the object but also plot the comparison star s differential magnitudes the magnitude difference between the comparison star and the check star 29 31 In this way they can determine if the variability seen in the object is truly inherit in the object or not Equations 4 4 and 4 5 help enumerate and solidfy the IDV trends Structure Functions or Duty Cycles Structure functions sometimes referred to as duty cycles is another analytical way of searching for small scale variation flares and or periodicities in AGN It is out of the scope of this senior thesis to completely describe every feature derivation and equation relating to the structure function However this is a very useful tool that should be used in further analyzing active galactic lightcurves The simplified equations and conclusions shown here are found in Anders A Jensen s thesis 32 There are many additional papers that describe and use the 4 1 Overview 41 structure function on AGN data e g 25 27 28 31 33 A detail
50. es and make adjustments until the star is centered Once you think you have moved the scope enough go back into CCDSoft and take another exposure You should be able to tell from the new exposure if you guessed the right direction If you guessed wrong try moving along the other axis If you moved too far you should change the Move setting in the Motion Controls to a lower slew rate If the star seems so big that it fills the center of the crosshairs you may need to set the exposure to a lower time If the star is bright and you have a long exposure time it will just be a big blob of light To do this right you will want the star to look more like a point After a few iterations of taking exposures and making adjustments you should get the star pretty close to the center of the crosshairs As you get it close enough that you only have to make slight adjustments you should set the Motion Controls to slew at 2x After you figure out how the N S E W buttons on the motion controls correspond to the axes of the exposure it will be easier to fine tune the motion of the scope 7 When the star is centered tell TheSky6 to Map the star Click on the Map button by the slew button on the Object Information dialogue box to map the star If the star seems way off you will be prompted to confirm the map Click ok to confirm x This is easy Just click on the Map button by General Multimedia Utity Telescope the Slew button If the s
51. es remain con stant throughout the whole campaign The average difference in the standard values N i MsSi DE is found and then applied to Mk501 s instrumental magnitude ms M mmg ms The resulting value is Mk501 s true magnitude The m quantity acts as a calibrated zero point that turns mmg into a true apparent magnitude As one final note the error associated with AM is only a combination of the Poisson statistics associated with Mk501 and its nearby standards On the other hand the latter magnitude calculation introduces further error associated with the standard star calibrations In fact the AM error is about 10 times smaller than the error in the true apparent magnitude This is why the simple differential magnitude is used for IDV detection When reporting long term monitoring values the error associated with the true magnitude is not as significant and is evident in the plots within Chapter 5 Cameron Pace a graduating senior at BYU is creating standard values for more stars to be used in calibrations such as these See his senior thesis and upcoming publication s for more detail 11 Chapter 4 Error Analysis As stated in the introduction of Chapter 3 the point of photometric reductions is to minimize the error associated with the results from each CCD image To mini mize error a brief description regarding statistical mechanisms is required the art of error minimization is an understa
52. esult in a decent model First right click on the TPoint Model object in the upper left corner of TheSky6 then go Show As Icon to TPoint Model Object Open to EEE open TPoint You will see an Excel spreadsheet type layout To edit the terms that are being used in your model click on Model Fit Data This will bring up the following window m pe vimulator Pay close attention to the Sky RMS and the PSD along the top of this window They are the most important numbers you need to watch throughout the process of selecting terms As you can see in the picture below the first six terms are already selected Unless you have a really good reason to mess with these terms don t If you want to add terms other than these the obvious thing to do is click on the More Terms button like the one in the picture below Clicking this button will bring up a whole list of new terms for you to include Start by holding down the shift key and highlighting the TF TX HCES HCEC DCES DCEC DAF DNP IE and IA terms and clicking the use checkbox Then you can close this little window Make sure the checkbox next to the Fork Flexure term is not checked as in the photo above The ROVOR scope doesn t even have a fork so it makes no sense to include this term Now that you have included all of these terms we need to keep the ones that helped the model and get rid of the LK x ones that hurt it m Tems Sky RMS 90 75 PSD 108 47 To do
53. etry was per formed to prove the quality of the SBP photometry data A variation of 0 005 0 001mag was found over a period of 2 minutes thus verifying the usefulness of this reduction method Additionally tools to help with ROVOR operation are included ACKNOWLEDGMENTS I would like to especially thank all those that supported me and worked with me through the past three years specifically my wife my family Dr J Ward Moody Cameron Pace I would also like to thank the Physics and Astronomy department of Brigham Young University for their continued sup port and funding of the ROVOR project making this and other senior projects possible Contents Table of Contents List of Figures 1 Introduction 1 1 Active Galactic Nuclei and Markarian 501 12 ein E Ge eaten Ge a Loe TS eet 1 3 Remote Telescopes 2 died des Senn Eds we ROVOR Setup 2A e A awe g eh he a Sh EPR psd che Us adit gto be Sod a the Se keen we Dud SOWIE e dn A re dee aoe Beeb ae BOS ne ce BM Boe 2 2 1 Software Bisque Packages o 2 2 2 LabView Programs ni A a ae tia als 2 2 3 Weather Stallion a ts A nD ey Reduction and Photometry Techniques 3 1 Calibration Frames E a de a ie Sul ALY t Biases e ua das dee a O a eee alee D ETA O A A ee OR ENE E te ILI PSA ES EN Seeds E AA ip ay Exe 3 2 ITIRAF and Mira ad SB ahs eo Bh Oeste EA a ls HS est ts Git a 22 tes dee Dio Sh ok Sng Ae ce ip Ao nw Bice Oe a had
54. he axes of the motion controls are flipped in relation to the axes of the exposure when you take exposures on the other side of the meridian have been talking a lot about the Motion Controls Let me explain what they are and how to use them The motion controls are a little dialogue box in TheSky6 that you can use to direct the motion of the scope They are like the electronic counterpart to the hand paddle motion controls Instead of the joystick you click on the button of the direction you want to go You can read more about how to use the motion controls on page 231 of TheSky6 user guide Click on Telescope Motion Controls to bring up my the controls You can change the controls to move in Ra and Dec or in Alt Azm You will want to leave them in Alt Az In the picture on the left you can see there is a little radio button that says Move but it is grayed out When you bring up the motion controls on the computer connected to the ROVOR scope this button will not be grayed out Be sure to select Move rather than Jog You can set the rate of motion relative to the sidereal rate 16x is ok for big moves but for fine tuning use 4x or even 2x As was mentioned above for the first picture you will just have to pick a direction and go with it Hold down the button you chose and the scope will move a little bit in that direction You should hear the servos in the mount move as the scope slews Motion Controls 6 Take exposur
55. he counts in each an nulus or C The various sized apertures create a bullseye target when superimposed on the object Figure 3 3 The annulus or ring is defined as the area between two apertures The SBPs are created with the counts found in each annulus To get the counts in each one simply subtracts the counts of the inner aperture from the counts of the outer aperture As the apertures increase in radius the number of counts within the aperture will increase the resultant of the subtraction will be a positive number Note that the most inner aperture acts as its own annulus Mathematically it looks like the following Cn Cs Cy 3 5 Here i and j are integers denoting the radius of the aperture where 7 ranges from the minimum radius to one less than the maximum radius fmin lt j lt Tmax 1 and i 1 3 Step 5 1 or Cna calcu After each C has been found we can now produce the counts area lation for each annulus This simple calculation divides the counts in each annulus the C found in Step 4 by the area associated with that annulus The area of the annulus is determined by subtracting the area of the two apertures the same aper tures used to determine the annulus counts or the areas of i and j Because the apertures are round circles the simple geometric equation of the area of a circle is used Area rr The radius used in the area calculation is in arcseconds To con vert from pixels to arcsecon
56. he normal procedures absolute and normal differential photom etry did not seem sufficient Finding IDV and periodicities in Mk501 require smaller errors than normal photometry will allow Using the Ph D thesis written by Stephen McNeil 8 we further investigate outline and detail surface brightness profile SBP photometry Proof of the validity of this new process is included as photometry com piled on another AGN M81 Chapter 3 describes the process and chapter 5 outlines the results of the photometry both normal and new 1 3 Remote Telescopes To fulfill the goals of the Mk501 campaign nightly observations from a good dark site was required Over the past two years Brigham Young University BYU un dergraduates have been working to build an operable remote telescope facility The remote facility was run during the evening at a location far from city lights and BYU campus The Remote Observatory for Variable Object Research ROVOR in Delta Utah was the result As a co constructor in the ROVOR facility I have put together this senior thesis to provide most of the information one would need to operate ROVOR It includes specialized user manuals troubleshooting help as well as some software and reduction code It was in a way designed to be an all encompassing user manual Therefore 1 3 Remote Telescopes 4 Chapter 2 describes in much more detail the features of this telescope and the reason for its construction Chap
57. her expla nation of each are described below Note that the values stated for Ctot in both types of photometry come from the data processed for the Mk501 campaign using the ROVOR telescope 4 3 1 Absolute Photometry Error To briefly summarize absolute photometry consists of looking at the object of inter est then slewing to a standard star field Moving back and forth between the two fields allows for comparison between the object of interest and the standardized stars 4 3 Photometric Error 48 Using transformation equations derived from the standardized star fields a true mag nitude is collected for the object of interest Even though this is ideal the process requires photometric nights in order to observe A photometric night allows an astronomer to assume that the sky is acting the same in one area as another This stability is needed in order to compare one part of the sky to another part Though photometric nights come relatively rarely they are the most pristine In context of error the error found intrinsically within absolute photometry is based on the original standardization fields associated with specific types of filters The error claimed with the production of the standardized fields is typically around 0 01 mag see 39 for specifics Those original fields have been expanded to include more standard star fields in the optical region that expansion was completed by A U Landolt see 12 for an ex
58. his to sign in unless you want to create an account for yourself USER NAME gt rlpearson PASSWORD gt outhouse You can then go to Downloads and Download Free Updates 11 12 13 14 15 16 Find TheSky6 CCDSoft Orchestrate and TPoint Updates for Windows Download and install them from the Desktop As of March 7 2009 the versions should be a TheSky6 6 0 0 61 b CCDSoft 5 00 092 c Orchestrate 1 00 035 d TPoint 1 00 511 One more download place to check Download Hot Fixes Sometimes they put random IMPORTANT things in here Check every so often There are still more things to download Go to www rcopticalsystems com to get the firmware for the TCCII that is attached to the telescope Go to Accessories gt Telescope Command Center Then click on the link to the Software Firmware Find the Full Installation Package Download it Then go back and update the Firmware as needed Still more things to download Go to the Finger Lakes Instrumentation website http www flicamera com index html to get the driver for the filter wheel Go to Support gt Software Download the FLISetup package Now we need to make sure that a configuration file is located in CCDSoft for the CCD that is being used Currently we are using an AP47p Hopefully it is in the Configuration File in CCDSoft If it is not then we should have the driver around Also CCDSoft will need the INI file for the camera We should have all those
59. ic nights Differential photometry was used in this project in order to maximize the imaging time with ROVOR Two variations of differential photometry were attempted for this project surface brightness profile photometry and normal differential photometry 3 3 1 Surface Brightness Profile SBP Photometry Normal differential photometry works well with point source objects or stars The method has draw backs once the point sources become extended objects This project 3 3 Photometry 20 focused on one extended object namely Mk501 see Figure 3 1 an elliptical AGN The problem with extended objects and normal differential photometry is that the other stars are quite far away from the point of interest on the object The point of interest is the center region of the elliptical galaxy the black hole emission jet region Because the stars are so far away from the center of Mk501 and other extended objects in general flat fielding errors are amplified To overcome this McNeil studied the use of surface brightness profiles SBPs on extended objects to obtain differential photometry 8 Instead of using surrounding stars as a constant standard the bulge of the galaxy or extended object is used as a reference Two separate images and their respective SBP bulges are used to compare the magnitude of the central region The difference in the two SBPs should be the differential magnitude of the central region The process is described be
60. ical problems within the scope and improve the RMS You must be careful however not to increase the PSD The point is if you add a term that makes the RMS go down but increases the PSD that term did more harm than good and you should not include it In the end you want the lowest RMS possible without increasing the PSD There are two different approaches to making a TPoint model We will try to chart a middle course between these two approaches First terms can be added that will account for known problems with the scope For example the telescope tube will almost certainly flex as you slew around the sky You will want to use the TF Tube Flexure and or the TX Tube Flexure tan terms to account for the tube flexing As you become more familiar with the ROVOR scope you will notice other specific errors to account for This first approach centers around the idea of having a well understood model that accounts for well understood problems The other approach is to make a model and add terms that make it work better without necessarily understanding the specific problems the terms are accounting for This second method could be said to be the Empirical Evidence Approach The best thing to do is to use the terms that prove to be statistically significant nave a large value and then see if any Empirical terms can be added to clean it up a little After you have your data set the 20 points mapped following sequence will almost always r
61. ifference in the profiles a little better than the first regular plot It also emphasizes the profile s trends a a smooth sloping tail associated with the bright bulge b a turning over curve to where the slope is approaching zero and c a brighter central region associated 3 3 Photometry 28 with the jet emitted by the blackhole Step 9 After the SBPs have been calculated and plotted the next step is to align the profiles At the beginning of this section it was briefly explained that instead of using far away stars for calibration and differential photometry we were going to use the profile of the galaxy bulge light This bulge light is displayed as the smooth outer portion larger radius of the SBP To align the profiles simply subtract the m a values corresponding to the bulge But where does the bulge begin or more importantly where does the nucleus light end This is a question that must be decided for each extended object observed because they will vary For Mk501 we felt confident in declaring that anything past 6 pixels or about 8 arcseconds was safe Also our collaborator from the University of Denver suggested an aperture radius of 7 arcseconds to be consistent with other research We also concluded this based upon trial and error Once the bulge differences are found average these difference values from each annulus together Then adjust the entire SBP each m a value of either image with th
62. ing the large chunks of data 4 2 CCD Imperfections 42 e If a minimum or periodicity is detected at long timescale in the SF and the minimum follows a maximum it could simply be a result of a flare The point of the SF is to look for characteristic timescales and periodicities It helps to further quantify the lightcurves of the AGN As a side note there is a program that is used in the BYU astronomy department and many others called Period04 It is used primarily for periodic star studies such as 6 Scuti variables but it may also be a good program to try It extracts the individual frequencies from the multiperiodic content of time series that contain data gaps 34 Whatever method is used finding microvariability limits and periodic behavior is the core of AGN monitoring 4 2 CCD Imperfections It seems that everything constructed by humans contains some sort of imperfection Even the optics on the ground breaking Hubble Space Telescope had major issues when first put into orbit 35 The CCD s used for other scientific research are no exception Some of these CCD glitches include gain and readout noise quantum hysterisis ghosting and many others These listed will be discussed in further detail here 4 2 1 Gain and Readout Noise The gain and readout noise on a CCD are introduced in during the manufacturing process Each value varies with filter because of the variation in the intensity of incoming photons Th
63. is only used for one simple important action the dome operation A National Instruments data acquisition DAQ device is powered through a USB connection with the computer It sits on the wall of the outhouse with wires from the dome control box in the doghouse attached to its I O leads The program that currently operates the dome is entitled MainControl Additionally this program displays the weather station s text file and a screenshot of the infrared camera Though LabView has made it quite easy to compile these needed operations the program uses a large amount of the central processing unit CPU on the computer When LabView is running the time it takes for CCDSoft to download images during an observation run is increased by about 40 it takes about 8 seconds to download an image without LabView running and almost 12 seconds with it running It will be necessary in the future to create another method for displaying the current weather and perhaps an even simpler program to open and close the dome 2 2 3 Weather Station The software to run the weather station was put together by Daniel E Wilcox a BYU alum Once it is running from dos two text files are output one text file is a 2 2 Software 11 complete record of the day s weather the other file is continuously updated currently every 30 seconds The latter file is read into the LabView program as described in 2 2 2 Further details are found in Appendix C
64. ke this The only problem here is that the CH term is one of the original six terms You should not get rid of any of these six terms unless you know exactly what you are doing The TPoint manual strongly suggests using the DAF term since our scope is has a Paramount ME mount We have found this to be good advice After you throw out a term you need to go back and look at the Fit Information again Removing a term can really affect the value and sigma of the other terms Terms that had a good value sigma ratio before may not be worth a thing now Also the order in which you add or subtract terms can have an affect on the RMS You will have to carefully select the terms that do the most for your RMS and PSD Don t be afraid to play around with it a little bit Another thing you will have to worry about is to make sure you have a large enough data set to justify all of the terms you are using The larger your set the more terms it will statistically support After you have chosen the terms you want to use it is now time to consider adding statistical terms These terms also known as harmonic terms are used to correct any harmonics that may be present in the data The idea is to add terms that account for any sine or cosine trends that may be present in the data You should read up on this on page 64 of the TPoint users manual The easiest way to find harmonic terms is to i click on the More Terms button and then click A on the Suggest Ter
65. l 8 Usually there are two other larger circles defining a ring or annulus for sky subtraction However this project replaced the values in the sky annulus with another method Areas surrounding Mk501 were chosen as vacant and empty regions on the CCD these are defined as sky areas The counts in these regions were averaged together to find an average pixel value of the sky background That value was then subtracted pixel by pixel across the whole frame Therefore the sky subtraction still occurred just not with the sky annulus 3The reason for this change was again due to the fact that we were working with an extended object Because there is no definitive place where the galaxy light ends placing an annulus for background subtraction might include some of the galaxy light itself To be safe we decided to go with this procedure Performing the subtraction this way is actually quite similar to the process used to calibrate photometers see 14 Standard procedure would be to observe the star then observe the black empty sky nearby This would ocurr throughout the night Part of the reductions would then include subtracting the background sky noise counts from the star s counts Now that CCDs can image a whole area of the sky at once choosing these dark empty areas can be easily found and used for sky subtraction 3 3 Photometry 25 Step 4 After C has been retrieved from each aperture we must find t
66. les replacing the older one How to make a TPoint model This article is intended to help you understand the process of creating a TPoint model It is not intended to substitute diligent study of the TPoint manual and other resources This article will touch on collecting data points controlling the telescope and the camera using TheSky6 and CCDSoft using Orchestrate to set up an automated mapping run choosing which terms to use with the TPoint model and other subjects e What is a TPoint model A TPoint model is a model that accounts for the inherent mechanical errors of a telescope that affect the telescope s ability to point directly at its intended target With this model in place TheSky6 can give the telescope modified directions so it will point more closely to its intended target As expected a good TPoint model requires a thorough understanding of the intrinsic errors of the telescope This understanding only comes by pointing at many many stars and recording the results The TPoint model can only account for the periodic or repeatable errors of the telescope Some errors like those caused by mirror flop are very difficult to address and cannot be corrected using TPoint TPoint has many formulae or algorithms each designed to correct for a specific type of pointing error The most challenging part of creating a TPoint model is deciding which of these algorithms or terms to include The process of choosing terms is as cl
67. letion of the reduction pro cess or the division of the normalized flat field Obtaining a flat field is a fairly tricky process that still has flaws within it They are possibly the most error prone images you will ever take The best flat field frames are obtained by pointing the telescope about 5 to 10 degrees from the zenith just after sunset or just before sunrise We assume that the images we take at this part of the sky at this time of night equally illuminates the CCD giving every pixel the same amount of light It is best to obtain frames that hover around 20 000 peak counts for optimum signal to noise This typically means that flat frame exposures run from 1 to 5 seconds Therefore flat frames do not need any dark subtraction they only need to be bias subtracted And just like the other calibration frames the best idea is to collect a large number of frames then min max average them together However you will need to obtain flat frames for each filter because each filter changes the specific frequency of light that interacts with the CCD This is a crucial part of flat fielding Additionally because the ROVOR telescope is settled on an equatorial mount the CCD image flips 180 when crossing the pier This means that the flat fields will Normalizing an image involves taking the average value of the frame and dividing the frame by that average This makes the frame average nearly equal to 1 3 2 IRAF and Mira 17 lo
68. low and follows the steps outlined in the simple list form of Table 3 1 Step 1 After reductions have been completed on the object frames an important decision needs to be made The SBP photometry requires both a reference or comparison image and an object image The reference image can be the subsequent image from the object frame or it can be a combined image of 20 60 images The combining of images to create a reference frame helps improve the photon statistics however it also smears the PSF of the reference image which degrades the statistics Having a combined image reference frame also allows you to use the reference frame for multiple nights so that each individual object frame is being compared to the same reference frame through the whole campaign The benefit of using the subsequent image allows a direct image to next image comparison and therefore minute by minute variations throughout the whole night and campaign The former idea was chosen for this 21 3 3 Photometry SUIIO O QBIIBA Jo eyo Aloyeuedxo YIM AIgowogzoyd JS 103 sdoys T E AQEL o3eun 39910 o SuIs y UI Lore tod snjuue ue UT ya syunoo fv OSBUIT ODUITAJOL IY UI vere Jod SAMUUL ue uny syunoo poysnfpe f24vuo eare Jed suue ue Uy epnyuseu paysnfpe PV us Loe Iod suue ue UT YMA spnyruseur vuu LIW tod suue ue umpa suno A snpuue ue uym syuNCD o 91319d8 UL UAM SUNO 2 25 20 Go PBI opnyUseUL eIyUoIOH
69. ly 31 2009 Figure 5 3 shows Sadun s reductions of each band through July 31 2009 Obviously the full data set has not been processed There is still much more data to reduce and compile The reduction was not completed for a number of reasons First the unsuccessful attempt with the SBP photometry put things quite behind schedule Second the re analyzing of the data due to bad error calculations lengthened the time required for reductions Third reducing the M81 data also took more time than planned Therefore there is quite a bit of work available for others to pick up The results we have nevertheless give sufficient results to answer our 5 1 Observational Data 53 Date 2009 UT Date UT JD Filter N observations 22 April 2454943 B V R 20 11 11 30 April 2454951 B V R 80 40 40 06 May 2454957 R 165 07 May 2454958 R 195 09 May 2454960 B V R 24 16 121 10 May 2454961 B V R 24 16 121 11 May 2454962 R 223 14 May 2454965 R 45 15 May 2454966 R 45 16 May 2454967 B V R 15 15 90 17 May 2454968 B V R 15 15 225 18 May 2454969 B V R 15 15 225 19 May 2454970 B V R 22 15 90 28 May 2454979 B V R 15 15 180 29 May 2454980 B V R 15 15 133 23 June 2455005 B VW R I 6 6 51 6 27 June 2455009 B V R I 8 8 8 8 29 June 2455011 B V R I 8 8 8 8 30 June 2455012 B V R I 8 8 8 8 05 July 2455017 R 7 06 July 2455018 R 178 07 July 2455019 B V R 10 9 152 09 July 2455021 B V R 15 9 165 10 J
70. m 07 May 2009 The Mk501 photometry filled diamonds was calibrated using star 1 as the comparison star and star 2 as the check star see Figure 3 7 The differential magnitudes of the two stars are shown as empty squares The stellar magnitude has been arbitrarily adjusted to fit on the same plot as the Mk501 differential magnitudes 5 2 Data Analysis 61 4962 65 4962 7 4962 75 4962 8 4962 85 JD 2450000 Figure 5 5 Normal differential photometry of Mk501 in the R filter from 11 May 2009 The Mk501 photometry filled diamonds was calibrated using star 1 as the comparison star and star 2 as the check star see Figure 3 7 The differential magnitudes of the two stars are shown as empty squares The stellar magnitude has been arbitrarily adjusted to fit on the same plot as the Mk501 differential magnitudes This night represents a higher confidence value 4 1 4 but no visible IDV and calibrations are completed this campaign should span more than 330 days Analysis of the long term data shows again the relatively quiescent state of Mk501 The R band only varies 0 lmag over the first 100 days The maximum variation of 0 68mag occurred between 19 May and 28 May or 8 776 days This gives a change of about 0 08mag per day The minimum variation was 0 003mag occurring from 29 June to 30 June or 0 999 days This shows that between ever
71. mmunication and operation Other hardware was designed and created by BYU faculty with the same users in mind 2 1 Hardware 6 Figure 2 1 An outside view of the Robotic Observatory for Variable Object Research ROVOR in Delta Utah The main telescope is housed in the larger building the dog house while the computer controls are held in the smaller building the outhouse ROVOR is housed within the Lifferth Dome a facility designed and constructed by Wes H Lifferth of the Physics Astronomy Department at BYU The dome cover is completely removed before telescope operation providing three specific benefits no need to operate through a slit horizon to horizon viewing and a simple open and close The first and third benefits allow for straightforward communication and operation The horizon to horizon viewing allows the telescope to track any object in the viewable sky with relative ease making many of the projects associated with ROVOR feasible Since the telescope sits high in the building a set of mercury switches attached to the telescope mount prevent the dome from closing or opening if the telescope is in a hazardous position The telescope itself is a 16 inch RC Optical tube situated on a Software Bisque Paramount ME see Figure 2 2 This German equatorial mount allows the telescope to easily track diurnal objects which are objects that move with the general motion of the stars However the equatorial mount requires
72. ms button This brings up the screen to the left Notice that it is empty Cancel You will need to click on Start to get the search rolling TPoint will give you a whole list of suggested terms The names of the terms are given on the left of the screen and then there are two columns of numbers The first column is the RMS while the second is the PSD These figures represent what the RMS and PSD will be e if you choose to add that term to your model The terms are organized with the best RMS and PSD first You should be careful to select harmonic terms that lower the RMS and the PSD 99 of these terms are not actually useful and you won t want to include them their RMS and PSD are worse than what your current model uses If you see one or maybe two of these terms you want to include highlight them and then click Add to Model When you are finished you should have a well understood model that accounts for the various mechanical imperfections of the mount If your model just does what it does and you don t know why then you really should do your homework Your understanding of TPoint should not be limited to this article please read up on the magic of TPoint in the TPoint manual How to use Orchestrate This article is intended to help you understand how to use Orchestrate It should not be a substitute for reading the Orchestrate manual We will discuss the logic behind Orchestrate as well as some do s and don ts
73. n online service that records the amount of information downloaded per hour For the record we have a limited number of megabytes available to us each day When that limit is exceeded a very low downloading cap is placed on us by HughesNet After looking at the amount downloaded per hour we 79 noticed there was something downloading 30MB hour Then the daily limit would be exceeded and the cap would be placed thereby slowing down the VNC connection like crazy There is a virus on the computer that needs to be removed We replaced the computer and installed a virus scanner Hopefully you don t have to go through those same issues Appendix C ROVOR Weather Station The actual program itself is quite a mystery and the places the text files are output are murder mysteries themselves Therefore here is the secret passageway to the text and exe files Weather WeatherStationl sensor watcherl debug To help cross all of this a shortcut has been placed on the desktop of the computer to 6 easily access the exe file The LabView program has the specific text file location embedded within it also If this becomes a bit more tedious than it needs to be contact Daniel Wilcox at the following email address dwilcox byu gmail com Hope fully the weather station lives for a long time without needing repairs or upgrades We have recently discovered that the exe file needs to be compiled for each new computer
74. nding of photometry statistics and error sources This chapter describes an overview of some basic statistics commonly used in astron omy and specifically photometry Additionally some sources of the error inherit in CCDs are discussed Also statistical measurements regarding photometric error are presented 4 1 Overview The need to minimize and report error in scientific experiments is fundamental in presenting quality data sets In astronomical photometry there are a number of com monly reported statistical measurements These include the error per observation or standard deviation the error of the mean or standard deviation of the mean the signal to noise ratio S N or SNR and many others This section presents some 34 4 1 Overview 35 details of a couple of those statistical measurements not full derivations or com prehensive explanations Additional information can be found in general statistic books 18 19 as well as specific engineering and scientist statistic references 20 21 Only discussions concerning standard deviations SNR and AGN microvariability tests are presented here 4 1 1 Standard Deviation The standard deviation per observation Cobs is a measure of the average deviation of a set of values from the sets mean In the case of astronomical photometry Cobs represents the average deviation of a set of magnitudes of a particular object from the mean magnitude value of the set The equati
75. ner you can When you click the Take Image button you should hear a soft click as the camera shutter opens As the exposure is being taken the Status bar will slowly fill with blue indicating the elapsed time of the exposure At the end of the exposure you will hear another click as the camera shutter closes In other situations you will need to pay close attention to when the shutter opens and closes At the end of the exposure the CCD will start to download the image to the computer Depending on how you set the Bin this downloading process can take anywhere from fifteen seconds to a minute and a half The scary thing about downloading is that the computer screen freezes while the image is being downloaded You can t move the mouse or get any sort of response out of the computer Don t worry This is part of the process Just be patient and in a little while the computer will return to normal and you will have your picture You can use the following trick to save time by starting taking exposures before it is even dark yet Since TheSky6 knows the coordinates of all the stars you can point the telescope and start taking images before you can even see the stars with your naked eyes For example in April May Leo is just above the sun and has many bright stars to use for your TPoint model After the sun has gone down but it is still not quite dark enough to see the stars try using TheSky6 to point the scope at Regulus and take a short 10 sec
76. no decisive results were obtained It may be argued that a rough period of 20 days is present from the R band data in Figures 5 1 and 5 2 This would be consistent with a high energy periodicity of 23 days found in Mk501 while in an active state 44 But the difference between a quiescent and active state seems to be large so a supposed relationship between these periodicities may be faulty Addi tional reductions are required to portray the whole campaign more solid periodicity measurements should arise 5 2 4 M81 Data The brighter larger M81 galaxy was used to confirm the usefulness of the SBP pho tometry As described in 3 3 2 Mk501 was just too small and too dim to provide adequate results with SBP photometry McNeil was able to produce reliable variabil ity values as shown in Table 5 4 8 He found variations of 0 057 and 0 038mag in the B band both with a 0 011 error A 0 135 0 011mag variation was found in the R band as well Therefore he showed SBP photometry was sufficiently accurate on M81 We performed the SBP photometry on a couple of M81 images from 2 March 2010 to make certain this process actually works After going through the SBP calculations we conclude that the SBP photometry method does work A differential magnitude of 0 005mag was found with an associated error of 0 001mag McNeil found larger 5 2 Data Analysis 64 variations between images but the two images used for the SBP photometry were sepa
77. nt for the Mk501 campaign presented here 2 2 Software 8 2 2 Software As stated previously the hardware and software were selected to make operating ROVOR relatively simple Also many of the hardware pieces were supplied with off the shelf software making communication between the telescope the computer and the programs rather smooth 2 2 1 Software Bisque Packages Much of the hardware uses programs released by Software Bisque SB The mount CCD and filter wheel operate by the following programs TheSky CCDSoft Orchestrate and TPoint Though each program can be run individually and manu ally when ROVOR is used robotically these programs communicate with one another almost seamlessly A brief discussion regarding the functions of each program and future improvements regarding software operation is given here More information and specialized ROVOR operation manuals are included in Appendix D TheSky displays a virtual nighttime sky set at the telescope s location Within this program the telescope is parked i e placed at the position where the telescope rests when not in use homed slewed to the default position where the telescope encoders are reset and directed throughout the night CCDSoft controls the CCD and filter wheel It is responsible for maintaining the CCD temperature changing filters adjusting the telescope focus and taking images Appendix D includes further specifics regarding initial setup downloa
78. object For quick estimations if S gt N then Equation 4 2 turns into 2 VS This is the simple error estimation of Poisson statistics But again do not use the actual counts divided by the sky value for your SNR value 2In Howell 9 he expresses the S N ratio as Con Covj 0 Nsky Naark a eee 2 The new terms are defined as follows Nszy is the number of photons per pixel from the sky contribution Naark is the number of electrons per pixel from the dark subtraction and Nyeadout is the number 4 1 Overview 38 4 1 3 Propagation of Error The propagation of error is an important concept in statistics It is a more accurate and precise way of stating statistical errors For example if there were two numbers each with an associated error a simple less accurate way of calculating the resultant error would be to add the error values together For better accuracy one would add the error values in quadrature This means to take each error square the value add the squared value to the other squared errors then square root the result An example of a series of errors added in quadrature is the noise value in the SNR equation Equation 4 2 Each term in the denominator is a squared error value that are added to another the first term is actually the square of a square root Je x This concept is important in finalizing photometric errors and is described more fully in 4 3 4 1 4 AGN Microvariability Tests
79. odicity for binary black hole models in Mkn 501 Astron Astrophys 359 948 952 2000 Appendix A Blackhole Radius and Mass Derivation 85 2 1 describes a procedure that uses IDV data to place limits on a blackhole s emitting region and mass The derivation of this limitation is included here This is a useful calculation for checking the nature of the IDV If there are already known values for the radius and mass of the blackhole the calculated radius and mass should fit within those values However this does not mean that if a true IDV is found and does not fit within previously defined ranges that the IDV is actually faulty Care must be used in determining and stating the nature of the IDV and subsequent radius and mass values 73 74 IDV constraints Searching for quicker magnitude variations a smaller At Figure A 1 This is a slide from a presentation I gave in April 2010 It shows a very basic example of what an IDV should look like in the data Obviously there is nothing that is going to be that smooth but the idea of changing from one magnitude to another is easiest illustrated as such 75 2003 UONIPS puz WU SISO 8 Md OUD saisAydoysy UJSpoy 0 uonanponuj uy Wal pajdepe uogeanap pue aing uoy T 4 IEL z2 T 4 3IyYM sS DG Te zt T A 9139 12 gt MH Y z a OG Ss IV i 2 a W QHUspal 124314 0 onp 10108 AMAL pI e SUIPNOUT T o y We p p 1
80. off the shelf program that was used in this project see chapter 2 Though not as powerful as IRAF Mira was chosen to be consistent with the off the shelf idea This overall idea was to run a remote telescope facility with off the shelf hardware and software to allow for easy operation and smooth data reductions Mira 3 3 Photometry 18 runs on Windows computers and allows for personalized scripting It is also much more user friendly than the command line IRAF and therefore easier for a beginning astronomy student to reduce ROVOR data Two advantages of Mira that the ROVOR group found is its easy reduction methods and simple program scripting specific instructions for both can be found in the Mira help folder Many of the procedures that were used for this particular project were completed using specialized Mira scripts The source code for each of the procedures is found on the ROVOR computers along with a description of the use of the program A lot of time was taken to create scripts that could be easily used and altered in the future 3 3 Photometry Photometry is used on CCD images to analyze the photon count provided by the image There are two ways to do photometry absolute and differential Absolute photometry consists of imaging two fields throughout much of a very photometric night a night that is perfectly clear with no high cirrus clouds or highly variable seeing The object of interest is imaged then followed b
81. ok different when the telescope is on the west of the pier as compared to on the east side Therefore both west and east flats are necessary We found that performing both morning and evening flats with west flats taken at night and east flats taken in the morning works fairly well This does mean that careful planning is needed when creating an automated script to ensure proper beginning and ending times Though a bit tedious this provides quality images for photometry 3 2 IRAF and Mira There are two astronomical programs that are used at BYU IRAF and Mira Each has their strengths and weaknesses Some have even found it useful to do the re ductions in Mira and the photometry in IRAF Regardless how they are used both programs are powerful and helpful The Image Reduction and Analysis Facility or IRAF is a more cryptic but pow erful tool than Mira It only runs on Linux machines there is a Linux computer room in the astronomy department It has been tested and used by many astronomers so much so that in most scientific papers the phrase Standard IRAF procedures were used is quite common Instructions on how to operate IRAF are not included here but is left to the Observational Astronomy courses Rather the program is presented here to acknowledge its existence and powerful analysis features However none of the reduction or photometry for this project was completed in IRAF for reasons explained below Mira is another
82. olves that image with a gaussian PSF of a specified input width this varies the images own PSF The script calls this command remeasures the image PSF then performs the convolution again if needed The user specifies the PSF error but a difference in PSF s of about 0 00002 is typical and quite good The convolving takes about 30 loops if the initial width is a good guess This script makes life a lot easier and provides sufficient results Step 3 Following the matching of the PSFs photometry measurements can begin without worrying about the PSF matching errors The photometry measurements will be used to build up the SBPs for each image the reference and object frame To do this we find the total counts within a user defined aperture C The aperture is simply a circle centered about the maximum point of light on an object The C within that region are then summed and reported To create SBPs apertures of varying radii are compiled together McNeil went from 1 to 10 pixels in increments of 0 5 pixels for this project we went to 12 pixels with increments of 1 0 pixels We chose to do this because of ROVOR s larger plate scale 1 375 and 1 475 arcsec pixel as compared 3 3 Photometry 24 Figure 3 3 An example of the collection of apertures superimposed on top of Mk501 The annulus that these apertures make are used to create SBPs See 83 3 1 for more details to the plate scale used by McNeil 0 3 arcsec pixe
83. on described for Cobs is given as Cons Eis Let 4 1 In this equation N number of images lt m gt average magnitude of the N images and m magnitude in a single image This is useful in determining the general deviation in a set For the Mk501 campaign the Cobs was only used to determine the deviation from the average on a given night as well as the SNR see 84 1 2 This helped determine if the night included microvariations Other than that the Cobs was not used during this project though it helpful to know what it is and what to do with it for analyzing other papers and results You will find that some report the equation above with an N 1 in the de nominator instead of the N The reason for this is based on statistical theories and personal preference Those ideas however are not described here For the Mk501 campaign Equation 4 1 was sufficient for its use 4 1 Overview 36 4 1 2 Signal to noise Ratio The SNR is a significant piece of an image s observational error It takes into account the sky noise CCD readout error CCD photon to electron conversion the gain and the Poisson error in the count measurement The readout noise and gain errors are discussed in 4 2 1 In summation the gain accounts for the number of electrons that are produced with one incoming photon It would be ideal if the relationship was 1 1 but it is not and the SNR tries to account for that mismatch Here is the SNR eq
84. on the frame including the varying jet in AGNs 5 2 Data Analysis 65 Count Difference 190 5 170 4 gt 150 4 130 4 110 4 90 4 70 4 50 4 30 4 10 5 o gt gt 4 gt 10 0 2 4 6 8 10 Radii pixels Figure 5 7 After the SBP photometry was completed on image 12 and 13 of M81 from 2 March 2010 the counts in each annulus C for each image is subtracted Then the resulting difference was plotted with the radius of the annuli on the x axis creating this plot The FWHM of this Gaussian profile 1 6 pixels equates with the FWHM of the PSF of the two images 1 95 pixels Therefore this plot establishes the credibility of SBP photometry on objects that are as bright and as large as M81 Chapter 6 Conclusion The year long campaign of Mk501 was generally successful Possible IDV values of 0 019 0 009 0 01 0 01 and 0 01 0 01mag were found on rough timescales of 8 39 and 20 minutes respectively The first IDV was found on 7 May 2009 with the last two coming on 6 July 2009 The calculated limitations on the radius of the emitting region and the mass of the associated blackhole lie within the accepted values for Mk501 This would lead one to think that the IDV is intrinsic However the size of the error associated with the IDV values and the low point to the IDV as a product of noise Additional reductions and careful examination will reveal the true nature of the IDV Though
85. ond exposure Be very very sure you know what you are doing Because the CCD is much more sensitive than your eye you can usually get a picture of Regulus or any other bright star this way You can use this technique to map data points before it is even dark enough to observe yet This way you can spend more time working instead of standing around with your teeth in your mouth waiting for it to get dark 5 Based on the exposure make slight adjustments to the scope In CCDSOoft hit the 6 key to get the crosshairs to appear Use the Motion Controls to center the image in the crosshairs This will take several iterations After you have taken an exposure you KADE now need to make slight corrections so your target star will be in the center of the image Hit the 6 key to get the crosshairs to appear Without the crosshairs it is very hard to get the image centered So lets say the star you just took a picture of is slightly above the center of the image Now you need to decide what corrections to make to get the star in the center It will not be obvious what direction you need to move the scope to get the star centered You will just have to pick a direction slew the scope that way a little bit take another exposure and see what the results are The first few stars of the night will be a challenge because you don t know yet how the directions on the Motion Controls correspond to the directions on the exposure In our experience t
86. ose to an art as TPoint will ever get As much time can easily be spent mulling over which terms to include as can be spent collecting the data points There are many ways to simplify the process of including terms which will be discussed later e What are data points A TPoint model is made up of several data points These points are used to map out and account for the pointing errors of the telescope For example suppose you point the telescope at Sirius which is located at 06h 45m 30 79s 16 43 32 452 Because of errors in the worm gears the telescope not having an exact polar alignment flexure of the telescope tube and many other things the telescope will not point exactly at Sirius It will point somewhere close to Sirius say 06h 45m 33 5s and 16 43 36 54 When you tell TPoint to Map this data point TPoint will note the RA and dec of the target object the RA and dec of where the telescope actually pointed and the angular separation known as the pointing error of the two With a large collection 100 200 points of this data in hand TPoint can account for many of the errors which caused the scope to miss its target and pointing will greatly improve e How often does a new TPoint model need to be made Anytime changes are made to the telescope such as when the polar alignment is adjusted or a new camera or filter wheel is installed a new model must be created You might be able to get away with doing a Short Ma
87. ovember 2455143 B V R 10 11 87 17 March 2010 2455272 R 15 18 March 2010 2455273 B V R 8 5 5 Table 5 3 Observational record of Mk501 throughout the 2009 and 2010 season Observational dates displayed are all 2009 except where noted N refers to the number of frames taken in each filter 5 1 Observational Data 55 T T 4980 4990 5000 5010 5020 JD 2450000 T 4970 PEK SIX RK RX ETS e O gt ma YY SEX a LO eR O 29 SALAA SCA A za dy OZ o wo a fa w a lt 409 o st a st N m w a N LO co gt 0 N N N oa oO e oOo Oo oO oO oO m is o ae apnyubey Figure 5 1 R band normal differential photometry from April 22 to July 6 2009 These values are true apparent magnitudes of Mk501 adjusted from the six standard stars found in its FOV 15 The error bars run 0 02mag 5 1 Observational Data 56 13 22 13 24 13 26 13 28 f if Magnitude z w 13 32 13 34 13 36 4940 4960 4980 5000 5020 5040 JD 2450000 Figure 5 2 R band normal differential photometry from April 22 to July 31 2009 These values were compiled by Alberto Sadun from the University of Denver a collaborator Each data point represents the differential magnitude of Mk501 from a five image combined frame Because these are not true apparent magnitudes the error bars run a
88. over 100000 counts in the photometry aperture Now if there comes a time where the S N ratio decreases the increased gain and readout noise levels will become quite significant 4 2 2 Ghosting and Quantum Hysteresis Ghosting and quantum hysteresis are less discussed imperfections of CCDs The reason I have chosen to discuss them here is that we have found that both the Ap47p and Ap8 exhibit these properties Ghosting is charge that lingers on the CCD If an image of a fairly bright star is recorded the next exposure will show a remnant of that star Its like looking at the bright sun for a second and then as you blink the image of the sun reappears in your vision Just as the ghost sun in your eyes is a little detrimental to your sight for a while ghost stars increase photometric error of images Extra photons are being added to the image that should have been cleared with the transfer of the previous image Richard D Crisp has spent much time reviewing the effect of ghosting 36 37 In fact the proper scientific term for ghosting is residual bulk image or RBI Crisp wrote an article for SPI IS amp T Electronic Imaging regarding how to manage the residual bulk image 37 For ROVOR s cameras RBI was analyzed specifically on the Ap47p To try to account for the ghosting effects we began imaging a bias and a dark about every 20 minutes throughout an observing night This helped account for the extra counts in the images because they woul
89. pping Run as described on pages 31 33 of the TPoint manual With a short mapping run some of the terms are adjusted but not all of them If this doesn t work you will have to start over and make an entirely new model This isn t that big of a deal if you are using the automated method described below e Collecting Data Points The process of collecting data points is referred to as mapping There are two very different ways of mapping these data points The first method requires more direct input from the observer the second method is almost entirely automated You should have some practice with and a good understanding of the first method before you attempt the second Before doing a mapping run it is important to decide where in the sky the points will be taken First TPoint will not work with less than 6 points Once you have mapped six points TPoint can start improving the scope s pointing Because of this you won t be able to use the second method listed above to map these first six points Also these first few data points will exhibit large errors You may even have to have someone stand on a ladder and look through the finder scope to give you directions as you use the motion controls in TheSky6 to center the scope on the desired star Another thing that will help while mapping these first few stars is to map a group of stars that are bright and close together If the stars are bright enough you can even map them before i
90. rated by many days The first set of images the B and R bands were 70 days apart the third set was 1040 days ROVOR s two 2 March 2010 images were only separated by 2 minutes This shows the exciting power of the SBP photometry To further check the validity of the procedure we performed the same final check that McNeil used 8 After creating the SBPs for each frame and converting the magnitudes back into counts the profiles are subtracted from one another These values are then plotted versus the increasing radii The resulting profile should be close to a Gaussian function with the same FWHM spread as the actual PSF of the image in fact when proceeding with Images 12 and 13 from the 02 March 2010 ROVOR data the final plot is a Gaussian and has a FWHM of 1 6 pixels see Figure 5 7 The FWHM of the images was 1 95 pixels creating a difference in FWHM values of 0 35 pixels McNeil found a difference in FWHM for the M81 images of 0 5 pixels and claimed the FWHM of the Gaussian profile is similar in size to the PSF s of the image 8 Therefore the small variation of 0 005 0 001mag is most likely intrinsically connected with M81 2McNeil goes into further details about why this should be but the simple answer is that the varying light or remaining light should be spread out the same amount as the rest of the frame 8 A CCD s PSF is created by seeing and optics effects Every point source image should have the same PSF
91. re to pump up the analog signal enough so that the A D converter can resolve it into a range of values that can be handled by a computer 24 It basically determines how sensitive the AD converter is to the flow of electrons being produced by the photons Ideally the ratio between electrons and photons would be 1 in other words the A D converter would be good enough to resolve 1 electron for every 1 photon The readout noise arises in a CCD when the counts in a pixel move through the circuity in order to be digitally recorded This noise or error accounts for the 4 2 CCD Imperfections 45 amplification scale in the gain the actual movement through the circuit and the speed of the movement Steve B Howell states that la mazing as it seems the readout speed and thus the rate at which currents flow through the on chip amplifier can cause thermal swings in the amplifier temperature which can affect the resulting read noise level 9 He continues to explain that a CCD with a slower readout time generally produces lower noise levels This is easily seen in Table 4 1 The noise levels are 2 to 5 times larger on the new CCD than the old one Let us briefly examine these pros and cons The purpose for changing out the old CCD with the new CCD was three fold First the old CCD the Ap47p took about 20 30 seconds to download each image in 1 x 1 binning When we started using 2 x 2 binning the download time moved to 12 15 seconds
92. rs of work Map at least 20 30 points on the west side before starting to map the east side Also looking at stars on the opposite side of the meridian requires the scope and counterweight shaft to switch places This takes a couple of minutes to do The more times you flip from one side of the meridian to the other the less time you will have for mapping If you drove down to the ROVOR site to do the mapping run and you are using this method every minute will be precious and you won t want to waste time Another decision you will have to make is how far down in altitude you will go The mirror of the ROVOR scope is fixed so it won t roll out and shatter on the floor In other words the telescope itself can be safely pointed at objects that are right on the horizon In practice you will never observe objects that are close to the horizon Suppose you want to observe an object that is currently at 30 altitude If your TPoint model included stars that were at about that same altitude the scope will have a hard time tracking the object you are observing It is a good idea to map stars that are at a lower altitude as part of your model If you want to observe at 30 altitude it would be a good idea to map stars down to 20 altitude Mapping stars close to the horizon is difficult because refraction starts causing problems As was mentioned earlier this method works best with bright stars The TPoint manual suggests using stars of sixth magnitude
93. rtae objects Astron Astrophys Suppl Ser 130 305 310 1998 M Fiorucci and G Tosti VRI photometry of stars in the fields of 12 BL Lac ertae objects Astron Astrophys Suppl Ser 116 403 407 1996 J R Taylor An Introduction to Error Analysis 2 ed University Science Books 1997 L J Walker Helen M Statistical Inference Holt Rinehart and Winston 1953 M R H Walpole Ronald E Probability and Statistics for Engineers and Sci entists 2 ed Macmillan Publishing Co 1978 BIBLIOGRAPHY 70 21 22 23 24 25 26 27 28 29 B P Roe Probability and Statistics in Experimental Physics 2 ed Springer Verbig 2001 L Mortara and A Fowler Evaluations of Charge Coupled Device CCD Performance for Astronomical Use In Society of Photo Optical Instrumentation Engineers SPIE Conference Series Society of Photo Optical Instrumentation Engineers SPIE Conference Series 290 28 1981 J H Simonetti Measuring the Signal to Noise Ratio S N of the CCD Image of a Star or Nebula http www phys vt edu jhs phys3154 snr20040108 pdf Accessed March 2010 M V Newberry Increasing Precision and Accuracy in Photometric Mea surements In Precision CCD Photometry E R Craine D L Crawford amp R A Tucker ed Astronomical Society of the Pacific Conference Series 189 74 1999 J Heidt and S J Wagner Statistics of
94. s S N value If the comparison magnitude is just from one star then Ostars is the S N value of that one star 4 3 Photometric Error 50 Otot Usually comes out between 0 005 and 0 01 mag more than 10 times better than the absolute photometry Then the instrumental magnitudes of the comparison and check stars are sub tracted to find their differential magnitudes again their errors are used in the equa tion above to find their own Co These values are typically lower than the object s Cto and range from 0 003 to 0 008 the stars are usually brighter than the object and therefore have more counts By plotting both of the differential magnitudes variations present in both are assumed to not be variations within the object How ever if there is a feature in the object s differential magnitudes that is not present in the comparison star s magnitudes that feature is taken to be a real variation in the object For blazars these variations happen so fast and so faintly that the smaller error in the differential photometry is essential for finding the smaller trends Nevertheless a differential magnitude cannot be reported as a true magnitude Why Because it simply is not the differential magnitude is a difference in instru mental magnitudes A true magnitude is produced when the instrumental magnitude is converted into one using standard stars with standard values Absolute photom etry does just that But again the Cia climbs once
95. ss than half of the 99 confidence value of 2 576 Visually the IDV can be seen in the 6 May and 6 July data We were not able to visually find IDV in the 11 May data and the value is not high either but we had reason to believe from previously processed data that 11 May was more likely to have IDV Instead 11 May is a good example of the typical nightly data The scarcity of IDV in the Mk501 data highly suggests the quiescent state of Mk501 When Mk501 flares IDV is highly active and has been consistently found This ROVOR campaign hoped to find plentiful IDV values to help restrict AGN models Sadly as can be seen in Table 5 5 the Am is generally on the order of its associated error and the values are low These suggest the proposed IDV values are an artifact of the nightly seeing variations and not intrinsically from Mk501 However if these values are real a new lower limit has been temporally identified with the At 8 59min on 6 May This places limits on the size of the emission region and extrapolates to the mass of the black hole region 1 55x10 m and 5 2x10 Mo In fact if the proposed IDV is inherent within Mk501 the designated R and M values fall within the range of currently accepted values of 0 01 3 x 10 Me 40 43 Another intriguing note relates to the number of observations for each of these IDV nights Each night included over 180 images Therefore it seems that the IDV is most likely to be present in nights
96. ssage fix the problem and execute the command again Index AGN Microvariability tests 38 Background AGN monitoring 1 Photometry 3 Remote Telescopes 3 Calibration Frames 13 Biases 13 Darks 14 Flats 15 CCDs 42 Ap47p and Ap8 comparison 45 Gain 42 Ghosting 46 Quantum hysteresis 46 Readout Noise 42 Residual bulk image 46 CCDSoft 8 download time 10 Data 51 Duty cylces 40 Error per Observation 35 Intraday variability IDV 38 IRAF 17 LabView dome operation 10 Mira 17 MRMaterials 22 Orchestrate 8 Photometric Error 47 Absolute photometry 47 Differential photometry 49 Photometry 18 Background subtraction 24 Differential Photometry 30 aperture size 31 SBP Photometry 19 Sky subtraction 24 Point spread function PSF 22 Precision PEC 9 Propagation of error 38 ROVOR Dome 10 Hardware 5 Software 8 LabView 10 Software Bisque 8 Weather Station 10 Signal to noise SNR 36 Standard deviation 35 Structure functions 40 TheSky 8 TPoint 9 103
97. t 0 004mag 5 2 Data Analysis 57 125 T T T T 9 w an tr e o n m Magnitude oe 145 o hd o ae gt S oe gt B bana 15 4940 4960 4980 5000 5020 5040 5060 JD 2450000 Figure 5 3 Normal differential photometry from Alberto Sadun for the B V R and I bands The R filter is shown exclusively in Figure 5 2 but it is shown here for reference and completeness objective questions and make reasonable conclusions As was mentioned in 83 3 2 our reductions of the Mk501 data turned from SBP photometry to normal differential photometry But to prove the SBP photometry worked on brighter and bigger objects we imaged another AGN M81 which was previously studied with SBP photometry by McNeil 8 Table 5 4 outlines his obser vations and findings Discussion concerning the new ROVOR M81 data and McNeil s data are described in the following section 5 2 Data Analysis Multiple types of variations were searched for by the Mk501 campaign as discussed in the introduction of this chapter Therefore this section will move through each type to discuss our findings again they are IDV intranight variations and periodic 5 2 Data Analysis 58 Filter Image 1 Date Image 2 Date Day Separation Variability Am B 11 Mar 2002 20 May 2002 70 0 057 0 011 R 11 Mar 2002 20 May 2002 70 0 135 0 011 B 23 Mar 1992 27 Jan 1995 1040 0 038 0 011 Table 5 4 Differential magnitu
98. t is fully dark The bright stars in Leo or Orion are good examples of stars to use for these first six data points These first six stars must be on the same side and ideally the west side of the meridian when you map them If you have your heart set on mapping Orion and some of Orion s stars are east of the meridian and some are west TPoint will have a meltdown Splitting up the first six stars like this is disastrous and you will have to start over After you get these first six down there are a few decisions you have to make You will have to decide how many points to map how far down in altitude you want to go and which method to use Before we discuss these decisions in any more detail let s get a good look at the first method of mapping o Method 1 Quick start just follow the items in italics to start mapping stars right away The first method is just as described above pointing the telescope at a star and recording the error This is accomplished in 7 or so steps Insert a TPoint object into TheSky6 Choose a star Use TheSky6 to tell the scope to slew to the selected star Use CCDSoft to take an exposure of the star Based on the exposure make slight adjustments to the scope Take exposures and make adjustments until the star is centered When the star is centered tell TheSky6 to map the star DLD Or OS This method requires a lot of user input is very time consuming and may be a bit frustrating Keep in min
99. tar seems way off you will be asked to confirm the map Just confirm it It will take some time to master this process ee but it is something you have to know how to Equatorial RA 12h 59m 31s Dec 20 29 10 current do As you map more and more stars the Equatorial 2000 RA 12h 59m 03s Dec 20 26 13 Horizon Azim 180 07 36 Alt 30 03 33 pointing will improve and you won t have to Visibility Rise 17 43 Set 03 28 Transit time 22 34 make so many adjustments to get the star Object t St A rotaci n celos centered Once you have thirty or more stars Pour angh Pon oom a the TPoint model will have enough data to Air mass Screen x y 543 00 346 00 support more terms gt A Bol 4 This is the Map button e Selecting Terms Add terms that will improve the RMS pointing of the model Now that you have a statistically significant sample you can begin adding terms for TPoint to account for There are six basic terms that TPoint starts out with You can read up on what these terms deal with starting on page 42 of the TPoint manual will not explain them here except to say they account for non perpendicularities in the telescope mount and the polar alignment being off As was mentioned at the beginning of this article TPoint can account for many other things These other terms are what really help TPoint adjust the pointing of the ROVOR scope Before we talk any more about selecting terms we need to talk a
100. telescope Le se bee sock Yes he bis a ee ak 7 MOL nde ea e Ne Gt Seed de he eS eS OF 19 COD PSE Sd as A as aA A oi egg 22 Mk501 Apertures dea aw hh AS RO Se RSG 24 SBP Example e y a el a A a de e oe OS Aa 27 SBP Adjustment Example o 29 MSDS A A e a ios iil eg 31 MESOL Standard Stars ale a IA AA A oS 32 Structure Function Examples lt A Y 43 Mk501 R band photometry oaoa aa aa a e e 55 Mk501 R band photometry from Dr Sadun 56 Mk501 B V R and I band photometry 57 7 May 2009 Mk501 IDV 2 a a ane koh a a 60 11 May 2009 Mk501 IDV taa ros 61 06 July 2009 Mk501 IDV 0 LS a a ta 62 M81 SBP photometry evaluation 65 IDV Radius and Mass Limiting Equations 74 IDV Radius and Mass Limits Derivation 75 vill Chapter 1 Introduction Billions of galaxies have been found in the known universe 1 Though each galaxy is generally unique many similarities exist Edwin Hubble originally distinguished between Regular and Irregular galaxies He divided the Regular galaxy group into ellipticals and spirals 2 Further classification studies have helped to further define these galactic groups and has facilitated the process of trying to understand the nature of galaxies 3 Additional projects such as the one presented here help add a considerable amount of data regarding galaxy variability Background information p
101. ter 3 goes through the reduction and photometry techniques used for ROVOR s data Chapter 4 walks through some definitions of statistical equations used with ROVOR and commonly found in astronomical literature The chapter regarding the results chapter 5 portray the excellent data of Mk501 and M81 obtained from the ROVOR site The 70 plus nights used over the past year for scientific research and over 60 of those nights included imaging Mk501 has established ROVOR as a fully operational remote telescope facility Chapter 2 ROVOR Setup The Robotic Observatory for Variable Object Research ROVOR is a remote tele scope facility owned and operated by Brigham Young University BYU see Fig ure 2 1 It is stationed on a cow pasture 12 miles northwest of Delta UT The dark skies and coarse but acceptable seeing about 3 arcseconds provide good quality observations ROVOR had been under construction constant maintenance or man ual control until its first fully robotic run on April 1 2009 Over 70 nights have been used for scientific research since then Uncomplicated hardware and straightforward software operate the telescope and make the operation easy Specifics of the ROVOR setup are described here 2 1 Hardware Since ROVOR was to be operated and maintained by primarily undergraduate stu dents the hardware was chosen to facilitate this use Most of the hardware came with off the shelf software providing relatively fluid co
102. tes a gaussian profile on the CCD see Figure 3 2 By having the PSFs matched variable seeing effects and defective flat fielding effects are greatly diminished However if the PSFs are incorrectly matched false variations will be reported in the photometry The Mira program actually reports the FWHM of the PSF and that is what is used to find the PSF of the image itself IRAF can also provide a PSF template directions on how to do that can be found in the form of Dr Hintz or Physics 529 or a little bit of 329 The PSF matching was done using Mira s FWHM scripting commands The script references the use of another program only used for this one purpose It is called 3 3 Photometry 23 MRMaterials the MR standing for multi resolution 13 Only one command is used m_simu The program itself is quite large and the commands run through the dos system Every new computer that the program runs on needs a special license file I have communicated with Dr Murtagh the creator of the software through the ROVOR email address rovor byu gmail com to obtain new license files for new computers Continue to do this in the future if the need arises The application as well as licence file must be in the working folder with the images that are being processed The easiest thing to do is to copy the licence and exe file into each new working folder and then run the PSF fitting script Simply the im_simu command takes an image and then conv
103. that the command is currently being executed and error which obviously tells you that an error occurred while attempting to execute that command After the Status column comes the Error Msg column This column gives a brief description of the error that occurred it also gives you the error code Looking up these error codes on the Internet can help you diagnose the reason why the command did not execute How do I use Orchestrate As mentioned above Orchestrate runs a series of commands known as a script to make the telescope and camera do what you want them to do There are a few commands you will use over and over again The most commonly used commands include SlewToRaDec SlewToAzAlt TakeImage SetResolution WaitFor and AutoMap Most of your scripts will consist of the repetition of some sort of slew command a WaitFor command and then an Image command Different projects call for different scripts When using Orchestrate to automate the process of gathering data points for a TPoint model your first command will usually be SetResolution with an argument of Medium The Medium argument sets the bin rate to 2x2 High to 1x1 and Low to 3x3 The AutoMap command can work with images that were taken at a bin rate of 2x2 setting the resolution to medium will save you loads of time while gathering data points for a TPoint model Also gathering data points for a TPoint model is best accomplishe
104. that the weather station is going to be run This requires Visual Basic or the Visual Basic C program The Express Edition of these programs can be accessed on the Microsoft website Once these programs are installed follow the instructions found with the weather station to produce the application file Once you do the text files should be produced and a small black pop up screen should output that the computer is talking with the weather station software When the black screen is not there the weather station is not running 80 81 Still the need to more easily dissect and present the current weather weather trends and past weather is a necessity No steps have been taken in that regard however The physics department operates a weather station that sits atop the Eyring Science Center It may be worthwhile to meet with the professor overseeing the station to discuss possible ways to analyze and provide an simple weather display of the features previously described Appendix D Customized User Manuals This includes customized user manuals for setting up running and operating the ROVOR telescope These were compiled by Cameron Pace the TPoint model and Or chestrate instructions and Richard Pearson the installation instructions There are so many intricate pieces to the ROVOR facility that not all of them were included in these manuals but hopefully they provide some quality assistance If additional help and assistance
105. the observational error in the differential magnitude between the comparison and check stars If gt 2 576 then the difference is real at a 99 confidence level The 2 576 falls out of the statistical percentage value of a normal standard deviation distribution 18 This confidence parameter was not used extensively in the current Mk501 project due to the omittance of a check star Instead a stellar average assumed to be constant was used to find Mk501 s differential magnitude However recent calculations may necessitate the additional usage of single comparison and check stars More details are described in the Differential Photometry section 83 3 2 as well as the Data section 85 1 IDV Amplitudes The measurement of IDV amplitudes coincides with the confidence parameter in the previous section once is determined to be greater than 2 576 then the amplitude of the IDV is calculated This parameter measures basically the percentage of the 4 1 Overview 40 night that is above 20 The equation is 25 27 100 lt AM gt A Y Mmar Mmin 202 4 5 where A is the IDV amplitude lt AM gt is the average differential magnitude of that particular night Maz is the maximum differential magnitude with Mmin as the minimum magnitude The o represents the averaged error over the course of the night 28 If we were to use this equation for the Mk501 project the would be the average of the fram
106. the serial ports too just for fun then click Scan It should find a port that it labels flifil1 Click OK Then you will see a screen with a list of six slots from O to 5 In the bottom corner is a message that reads its current position It will say Unknown Click Home Wait for the message to show that it has actually found what filter it is at You are then ready to change the filters PROBLEMS If you have done all the necessary things above and FLI Filter still won t connect or if CCDSoft won t connect then there are a couple of things to make sure of CCDSoft Troubles The main problem here is that there are a couple of files missing that CCDSoft needs In Program Files gt Software Bisque gt CCDSoft gt Filter Wheel Plug Ins CCDSoft needs the Finger Lakes Filter Wheel Plugin dll 1 0 0 6 That same file needs to be in CCDSoft gt Filter Wheels gt Finger Lakes One last very important thing in C gt Program Files gt Common Files gt System There is a dll file there that CCDSoft needs It is called libfli dll If you right click on it you will see that this was created around September 25 2006 This one needs to be updated So back in C gt Program Files gt Finger Lakes Instrumentation you will find the same file libfli dll but if you right click it you will see that it has been update and has a new date of around July 24 2008 or something newer Copy that one and paste it into the Common Fi
107. this click Correction arcsecs on the Fit Y 1H Index error HA 2 00 i IV ID Index error Dec 180 07 Information 7 NP Non perp HA Dec 076 button This brings IV CH Non perp Dec OTA 2 14 j V ME Polar axis elevation 180 07 up the Fit IZ MA Polar axis East West 1 50 i Information J Fork flexure pola dialogue box J Tube flexure 90 6 Thi di b None Equatorial ra oe Sa a pas ea just has a bunch 4 Mica of numbers in it Fit Fit Information but they are some y Graph Telescope Simulator very important Scatter Diagram y numbers You will r User defined graph want to scroll Cie swe gt zF ances down partway till Cope Pm x ca yugale d information like the one on the next page Be careful there is a very similar list of information at the very top of this dialogue box We see three columns of numbers the second two columns are the most important to us The first lists the value and the second lists the sigma As you can see the DAF term has a sigma that is large relative to its value This means we should get rid of the DAF term This brings our RMS from 71 47 to 71 10 and our PSD from 101 07 to 96 61 This is not a very big change in RMS but it did bring down the PSD which is the real goal In the above example you will also notice that the CH term has a larger sigma than a value This is a very bad situation and you would normally want to get rid of any term li
108. throughout all the frames This then helps reduce the error in each pixel and eliminates random cosmic ray hits 3 1 2 Darks A dark frame is a measure of the random noise level surrounding the CCD It accounts for things such as hot pixels CCD temperature gradients and thermal stability The dark frame is taken with the shutter closed but with the same exposure times as the object frames For example if you are exposing for 90 seconds on a particular object then you will want to collect a series of 90 second darks Because the darks collect random fluctuations it is usually a good idea to take the average again usually with a min max rejection of many frames to use in a dark correction For ROVOR we have found it good practice to take a large number of darks at the beginning and end of an observing night along with a few throughout the night Darks must be obtained at the same temperature as object frames as well since the thermal fluctuation counts are proportional to the temperature of the CCD Note that the dark frame itself is a combination of a bias level and the noise Therefore the dark frames need to have their bias subtracted out as well The bias subtraction needs to happen before the darks are averaged together After the bias has been subtracted from the darks and the darks have been averaged together the dark counts can be removed from the bias subtracted raw frame simply by subtraction Rx 1 Z D 3 3
109. to use a sky subtraction over the whole frame instead of a sky annulus see 5To obtain this radius we used r 5 pixels binned 2 x 2 on the Ap47p and r 5 463 pixels binned 1 x 1 on the new Ap8 3 3 Photometry 32 o z o e o a s4 gt i a r eo a E 1 A 6 2 a I a e e 2 3 fa 5 i e gt amp P gt Figure 3 7 Mk501 and its associated standard stars The numbering system was derived from the finder chart 15 Step 3 in 83 3 1 for further details Once the aperture counts for Mk501 Cy x and the stars C were obtained they were converted into an instrumental magnitude m 2 5log C 3 13 Then the instrumental magnitudes for all the standard stars are averaged together Mm Dim where N is the total number of standard stars used Finally ms is subtracted by mmg or the instrumental magnitude of Mk501 This difference becomes the differential magnitude AM of Mk501 in that image The AM with its smaller associated error is used in analyzing nightly fluctuations also known as intra day variability or IDV For long term monitoring mmx is converted into a true magnitude Mk501 hovers around a magnitude of 13 To convert the mmg we move back to the standard star s instrumental magnitude Each star s magnitude is then compared to the standard 3 3 Photometry 33 values msi produced by Villata 16 and Fiorucci 17 The ms valu
110. too It is a file Folder called Apogee It can go into the Configuration File in CCDSoft as well MIRA Installation 1 We are using this program given to us by Michael Newberry one of Dr Moody s friends He is very helpful and very open in suggestions We have the 7 edition but there is now a newer version If you can convince Dr Moody to get the newer version then all of this is for naught But until he does this is the thing to use to upload the program 2 There were a number of issues with the first couple of 7 editions Finally Mr Newberry sent us the final fixed version of the 7 edition in the rovor byu gmail com email 3 You can either search MIRA on the email or go to December 28 2009 and look for the email from support mirametrics com There is a pdf file that you need to download in order to upload the mira exe file from the Mira FTP site Also included in that email is a password to open the pdf It is also given below Cu wiwuipo44 4 Input the password and open the pdf 5 Under the Downloading MIRA Pro Ultimate Edition section go to the FTP site download and install the Mira program You will need to rid the computer of any other editions on the computer The given passcode to install is given in the pdf or below 66375 c43 aahf3 677 ffgh ccxk 94kku 6 Once it s downloaded and installed that should be it though you might need the code above to install it Orchestrate Setup 1
111. uation in all its glory 22 23 5 de 2 N Cobj 00 O Osky G l sky l p This equation assumes that the sky counts have already been removed from the 4 2 object Section 3 3 goes through the sky removal process Therefore Cobj true object counts G gain o pixels in object aperture p pixels used for sky count calculation and Osky standard deviation in the sky count calculation As can be seen in this equation S is the signal from the object the numerator on the right hand side and N is the noise in the object signal the denominator on the right hand side To explain further the o is the standard deviation see 4 1 1 of the counts in the sky aperture As noted in 3 there are two ways to calculate the sky counts using larger apertures for a sky annulus around the object or selective dark areas around the CCD frame The latter option was chosen for this project An average counts pixel value was obtained for each area then each area value was averaged This became the sky background value for the particular CCD image Additionally the standard deviation was calculated for each area and averaged to become the Osky for each frame Both the subtracted sky value and sky standard deviation are placed into the header as values in SUBBACK and SUBBACKE More recent reductions put these values into SKYCNTS and SKYERR 4 1 Overview 37 Lastly the o and p are the total n
112. uly 2455022 B V R 15 9 134 17 July 2455029 B V R I 10 10 10 10 23 July 2455035 B V R 12 12 105 28 July 2455040 B V R I 8 8 8 8 29 July 2455041 B V R I 10 10 10 10 30 July 2455042 59 31 July 2455043 R 74 Table 5 2 Observational record of Mk501 throughout the 2009 and 2010 season Observational dates displayed are all 2009 except where noted N refers to the number of frames taken in each filter 5 1 Observational Data Date 2009 UT Date UT JD Filter N observations 01 August 2455044 B V 12 11 03 August 2455046 B V R 4 3 167 04 August 2455047 B V R 6 7 19 09 August 2455052 B V R I 22 18 24 8 10 August 2455053 B V R I 12 12 12 12 11 August 2455054 B V 9 9 119 17 August 2455060 B V R I 10 9 10 10 18 August 2455061 B V R I 8 8 8 8 20 August 2455063 B V R I 8 8 8 8 21 August 2455064 B V R I 8 8 8 8 22 August 2455065 B V R I 8 8 8 8 26 August 2455069 B V R I 8 8 8 8 27 August 2455070 B V R I 8 8 8 8 28 August 2455071 B V R I 6 5 6 6 29 August 2455072 R 68 01 September 2455075 B V R 24 24 30 12 September 2455086 B V R I 4 3 4 4 17 September 2455091 B V R 8 7 16 21 September 2455095 B V R I 4 4 4 4 22 September 2455096 B V R I 6 6 6 6 23 September 2455097 B V R I 4 4 4 4 24 September 2455098 B V R I 6 6 6 6 25 September 2455099 B V R I 6 6 6 6 26 September 2455100 B V R I 6 6 6 6 04 November 2455139 B V R 4 8 15 05 November 2455140 B V R 10 10 42 06 November 2455141 R 7 07 November 2455142 R 85 08 N
113. umber of pixels used in determining the object s counts and the sky s counts respectively For the circular aperture around the object the o is just the area of the aperture The p is the total number of pixels used in all of the selected dark areas for Mk501 we used 6 boxes with 100 pixels in each box Therefore p 600 To convert the SNR value from error in counts to error in magnitude dsyr a relatively simple relation is used Recall that conversion to a magnitude is OSNR 2 5 log C 4 3 If the C value represents the ratio of N or 2 1 which is the ratio between the noisy signal and the deviation in the noise then the ogyr can be found This is a relatively simple calculation to perform once S and N have already been found There are simplifications that can be performed to make the transition simpler but they lead to further uncertainty in the true error for example the simplest simplification is that asyr 1 S N 23 24 A simple calculation in a Mira script or Excel sheet takes care of the processing headache so a more accurate calculation is easy One final additional note Before I found this equation I thought that an SNR calculation was simply the ratio of the total signal Co in the SNR equation to the sky background value the header value of SUBBACK I could not have been more incorrect The SNR value is the ratio of Co to the uncertainty in the sky background relative to the
114. with a large number of observations This would mean that after further reductions 11 more nights with N gt 50 are most likely to show IDV 5 2 2 Intranight Variability Intranight variability simply refers to how Mk501 varies from night to night For this analysis long term monitoring was necessary Figures 5 1 5 2 and 5 3 show Mk501 s magnitude from April to July 2009 spanning 100 days When reductions 5 2 Data Analysis 60 Date UT N obs Am At min R 10 m M 10 Mo 07 May 2009 187 0 019 0 009 8 59 0 996 1 55 0 52 11 May 2009 206 1 094 06 July 2009 184 0 01 0 01 38 87 1 244 6 99 2 37 06 July 2009 184 0 01 0 01 19 53 1 244 3 01 1 19 Table 5 5 Display of the possible IDV nights with the number of frames N obs the magnitude variation Am the timescale of magnitude variation At the confidence parameter the radius limit of the emitting region R and the mass limit of the central black hole M Notice that none of the days include a sufficiently large to be fairly confident in the Am or At ay ATEN q ea 4 Lew tan Hip T TT gs 1111 pa 097 Hy iiig 95 7 r T 1 4958 46 4958 48 4958 5 4958 52 4958 54 4958 56 4958 58 4958 6 4958 62 4958 64 4958 66 JD 2450000 Figure 5 4 Normal differential photometry of Mk501 in the R filter fro
115. y night significant variations occurred Even though IDV may be mostly absent in a quiescent state intranight variability is abundant However the relatively calm Mk501 leads the intranight variability to search specifically for low amplitude periodic behavior 5 2 Data Analysis 62 Magnitude S 0 9 5018 65 5018 7 5018 75 5018 8 5018 85 5018 9 5018 95 JD 2450000 Figure 5 6 Normal differential photometry of Mk501 in the R filter from 06 July 2009 The Mk501 photometry filled diamonds was calibrated using star 1 as the comparison star and star 2 as the check star see Figure 3 7 The differential magnitudes of the two stars are shown as empty squares The stellar magnitude has been arbitrarily adjusted to fit on the same plot as the Mk501 differential magnitudes Note the significant variation at the beginning of the night This is an artifact of the flat fielding as it is present in both the Mk501 and stellar differential magnitudes 5 2 Data Analysis 63 5 2 3 Periodic Variability After establishing the prominent presence of intranight variability in Mk501 s quiet state searching for periodic behavior in that variability is the next step As pre viously discussed in 84 1 4 no analytical computations were performed to identify periodic behavior Visual analysis of the long term plots was the basis of the periodic searches However
116. y imaging a standard star field These fields have been observed and standardized The standardization pro vides a constant base where objects of interest can be compared The standard field stars are taken to be constant so that the object of interest can then be compared The scripting language that Mira uses is called the Lua language It is described from its online reference manual as an extension programming language designed to support general procedural programming with data description facilities It also offers good support for object oriented pro gramming functional programming and data driven programming Lua is intended to be used as a powerful light weight scripting language for any program that needs one Lua is implemented as a library written in clean C that is in the common subset of ANSI C and C 10 3 3 Photometry 19 Figure 3 1 Markarian 501 Mk501 as viewed from the ROVOR site This elliptical galaxy is classified as a blazar with a redshift of z 0 334 against the constant to look for variation For more information regarding standard photometry techniques see 11 12 and reference therein As an extension of those standard fields other stars are standardized in other fields Differential photometry uses these secondary standards as comparison stars Having these secondary stars in the field of view FOV around an object of interest permits observations to occur on not so perfectly photometr

Richard Pearson Senior Thesis - ROVOR

Contents

Download Pdf Manuals

Related Search

Related Contents