Quick Jump Menu
Info Ask a Question
Advanced Search Search the Knowledgebase
Asteroid Photometry and Reporting
Views: 13272 Last Updated: 2010-07-15 19:18

Orbit of asteroid 'tzecmaun'Ron Wodaski and Steve Pastor recently attended a workshop at Mt. Lemmon in Arizona to learn about asteroid photometry and reporting. Ron is in the process of preparing this wiki article with details of how Tzec Maun telescope users can search for asteroids, perform astrometry used in orbital determinations, and report results to the Minor Planet Center. This article assumes that the Big Mak will be used for imaging.

Note: Click the thumbnail at far right to see the orbit of asteroid tzecmaun, recently discovered and named in honor of the foundation by one of our users, Erwin Schwab. We have also included an animation of the discovery frames below right. Asteroid 'tzecmaun' is located below the text with the designation number and name. (Previously known and much brighter asteroid '7992' is at upper left of the animation, magnitude approximately 15.)

Note: This article is intended for team leaders and advanced individual users. It is not intended for beginners - it assumes that the reader is already familiar with the underlying concepts involved in imaging, using Astronomy Studio software, astrometry, etc.

Gathering Data

The first step in the process is taking images with one of the research telescopes. For this tutorial, I will assume that the Big Mak is being used. It has a focal ratio of 3.8, and an aperture of 14", which makes it a good telescope for detecting faint objects. However, other telescopes having long focal lengths and large apertures can also be used.

Asteroids are most commonly found near the ecliptic, the plane in which the planets orbit. Asteroid orbits extend above and below the ecliptic, but when you are starting out, and seeking to learn the necessary skills for detecting asteroids, we recommend you stay close to the locations where the majority of the asteroids can be found. In fact, it is very useful to detect known asteroids as the first step in your learning process - you will know that you are doing it right if you can detect known asteroids.

The basic unit of work for asteroid detection is the image set. This is a series of four images, with a delay between images. Any asteroids in the field of view will move relative to the stars. Technically, you can reasonably detect asteroids with three images, but a four-image set provides redundancy that is useful for detection of dim objects, and for dealing with the occasional problem image.

The delay between image sets is based on the rate of motion of typical asteroids. The average asteroid moves at a rate of approximately 0.5 arcseconds per minute. (That is an average or typical rate; actual rates vary - e.g., 0.3, 0.6, 0.7.)  Since you would want at least a few pixels between images of an asteroid, you need to know the image scale of the telescope used for asteroid detection to determine the delay between images in a set.

The details of the Big Mak can be found TK in this article. The image scale is 1.05"/pix (1.05 arcseconds per pixel). At a rate of 0.5"/min, an asteroid is moving across one pixel every 2.1 minutes (1.05 divided by 0.5). If we want the average asteroid to move 5 pixels between images in a set, then we need to have 5 * 2.1 or 10 and a half minutes from the start of one exposure to the start of the next exposure.

Typically, asteroid searches use exposure times of 1-3 minutes. Don't forget to add in the download delay when calculating your delay time. If we take 2-minute exposures, and assume a 30-second download time for the ST-10XME on the Big Mak, then we need a delay of (10.5 - 2.5) or 8 minutes.

While a delay is a simple way to deal with the asteroid rates of motion, it is not how real asteroid searches are done. Instead of a delay, a more typical approach is to slew to another area of the sky and take an image there. Since we want at least 10.5 minutes between the start of successive exposures, and it takes 2.5 minutes for each exposure (plus whatever slewing time is required), we could also take five sets of images. Each set would be in a different field of view; we would return to the same field of view for each set.

The image at right shows an example of multiple fields of view. The fields are right next to each other, but they could also be in completely different parts of the sky! The process would work like this: slew to the center of the A1 field of view (FOV), take a 2-minute image. Slew to the center of the A2 field, take 2-minute image. Continue until the image at the center of the A5 field is taken. Then slew back to the A1 field, take a 2-minute image. This is the second image in the set for the A1 field of view.

At the end of the process, there will be five images sets. We recommend that you take four images in each set. Since this example uses five image sets, the total delay from start to start in an FOV is about 12.5 minutes. This works - it is larger than the minimum we established of 10.5 minutes.

(The movement of 5 pixels is arbitrary. After all, some asteroids will move faster than 0.5"/min; some will move slower. The reason for having a reasonable amount of movement between images in a set is that the asteroid has to move far enough to make detection easier. if the asteroid barely moves between images in the set, the detection software will not find it.)

At the conclusion of your data collection phase, you will have one or more image sets to work with.

Note: You can use the scripting features built into the Astronomy Studio software on the advanced telescopes to create image sets. Please see TK the article Creating Scripts for details.

Asteroid Detection Software

There are multiple ways to analyze your data. You can, for example, simply load the files into a program like MaxIm DL, and create an animation of the aligned images. Stars will be stationary, and any asteroids will be moving. It should be fairly easy to see the brightest (and thus typically already known) asteroids with your eye. However, if you are going to start looking for undiscovered asteroids, you need the ability to detect dim moving objects in your image sets. It is possible to do this by eye, but it is slow, tedious, and difficult.

Software exists to help with the process. In this example, you'll learn how to use Astrometrica to do the detection. You'll need to download Astrometrica, at least one star catalog, and the asteroid orbital information from the Minor Planet Center.

Downloading Astrometrica

The Astrometrica web site is located at:

  http://www.astromehttp://www.astrometrica.at/trica.at/

Click the 'Downloads' link in the menu at top left to get to the page where you can download the software. Click on the first link in the list to download the self-installing version of Astrometrica.

Note: Astrometrica is not free software; it costs $29.95 as of the time this was written. However, it has a long evaluation period (100 days). You can use Astrometrica for more than 3 months before you must pay for it.

Downloading a Star Catalog

You will need a star catalog in order to use Astrometrica. The readme materials that come with the program give you several options. For this example, the UCAC catalog is assumed. It is smaller than some of the more complete catalogs, but has a selection of stars that is well-suited to asteroid astrometry. The primary benefit: the proper motion corrections for nearby stars are included in the data. This improves the accuracy of the astrometry because it can be updated for the current epoch.

To see detailed information about star catalogs that can be used with Astrometrica, go to the Astrometrica web site and click on the 'Star Catalogs' link in the menu at upper left.

Downloading MPCORB.DAT and COMET.DAT

You will need the latest copy of the MPC's orbital database. You can find download instructions here:

  http://www.cfa.harvard.edu/iau/MPCORB.html

If you just want to download the necessary files for asteroid and comets, you can do that from this site without any login or even needing an FTP client:

  http://mpcorb.klet.org/

Astrometrica Configuration for Big Mak

Astrometrica must be configured properly in order to analyze image sets. The importance of this cannot be overstated - Astrometrica will produce garbage if you don't configure it '''exactly''' right!

You can download and modify a configuration file that is set up for the Big Mak here. (not available yet)

Click the wrench icon at the far left of the toolbar in Astrometrica to access configuration parameters.

Observing Site Parameters

The site parameters must be set correctly for the Big Mak. Click the image thumbnail at right to view the correct settings. Asteroids are close to earth, so there will be parallax between different observing stations. The latitude, longitude, and altitude of the observing site must be recorded. The MPC (Minor Planet Center) code must also be entered. The code for the Big Mak observatory is H10.

The contact details should be set up as shown in the image that the thumbnail links to. The only fields that can be changed are 'Observer' and 'Measurer'. The rest of the contact information, especially the email, must not be changed. The MPC will send email to us, and you can view the 25 most recent emails by clicking on the MPC codes on the Portal page.

CCD Parameters

The CCD parameters must be set correctly for the Big Mak. Click the image thumbnail at right to view the correct settings. The settings that are specific to the Big Mak are:

  •  Focal Length
  •  Position Angle
  •  Pixel Width
  •  Pixel Height
  •  Saturation
  •  Time in File Header: Start of Exposure

The only setting that you would change is the Color Band. We recommend that you use one of the BVR filters for asteroid images. If you do so, select the appropriate radio button, and do not check 'Clear/no filter used'. If you use the clear filter, select Red (R) because the peak response of the chip in the ST-10 is in the red, and do check 'Clear/no filter used'. Note the following detailed explanation from the Astrometrica Help file:

 The radio buttons in this section define in which color band the CCD operates. If you use  color filters, choose the color band according to the bandpass of the filters. If you work  with a unfilered CCD (or with a clear filter), choose this according to the peak sensitivity  of the CCD chip (so the sofware knows weather it should use the B, V or R magnitudes  from the reference star catalogue for comparison), and tick the 'clear/no filter used' box  (so the magnitude will be reported as unfiltered in the MPCReport file). Most CCDs have  peak response close to the R color band. A notable exception are the Sony interline CCD  chips, which can be closer to the V band.

Program Parameters

The program parameters must be set correctly for the Big Mak. Click the image thumbnail at right to view the correct settings. Program settings are very important to get right - even a single incorrect setting could completely mess up your detection process. In some cases, a setting may lengthen the time to process your images considerably. For example, if the detection limit is set too low, processing may take half an hour or more!

The settings shown in the image linked to the thumbnail are for one-minute images taken with the Big Mak. If you can longer or shorter images, you may need to tweak the values in the Object Detection section. We strongly recommend you read and get familiar with the Help section that covers object detection paramaters - the meaning of these parameters is not obvious and even small changes can have big effects. The measure of how well the parameters are set is simple, however: if you detect asteroids effectively, the settings are good!

A brief description of the object detection parameters follows. Please refer to the Astrometrica help file for detailed information.

Aperture radius - This is measured in pixels. It defines the area around a detection that is used for calculations (e.g., the centroid of the detection, the brightness - flux - of the detection, etc.). Generally, an aperture that is 2x or 3x the FWHM of your stars (in pixels) is a good starting point.

Detection limit - This is a very sensitive parameters; a value that is too small will greatly lengthen the processing time, and cause false detections. The value here is the minimum signal to noise ratio that constitutes a valid detection. The value shown (3.0) is extremely aggressive; values in the 4.0 to 5.0 range are more typical.

Minimum FWHM - This is the minimum FWHM that will occur for real objects. You can use MaxIm DL or CCDSoft to measure the FWHM of the stars in your image to determine a good setting. This may change from night to night depending on the local seeing conditions. It may even change from hour to hour, so don't assume that all images in a set share the same value for FWHM. This parameter is set in pixels; the image scale of the Big Mak is 1.05"/pixel.

PSF-fit RMS - Determines how much the FWHM of the PSF (point spread function) can be varied for the purposes of detection. This is necessary because faint objects such as asteroids will have a lower signal to noise ratio, and thus a higher RMS error.

Search radius - Defines the boundary for fixed objects. For a value of 1.2 pixels, any object that moves 1.2 pixels or less will be considered a fixed object. Values from 1 to 3 pixels are valid.

Environment Parameters

You must tell Astrometrica where to find various items. Set the location of the following folders (directories):

CCD Images - This is the folder where your calibrated CCD image sets are located.

MPCORB.DAT file - This is the folder where MPCORB.DAT and COMET.DAT are stored.

Star Catalog - This is the folder where your star catalog is stored. This can be tricky; read the documentation for the catalog you are using to make sure you specify the correct folder! For example, for ucac 2, you do not specify the 'u2' subfolder; you specify the folder that contains the u2 folder!

Output - This is where the MPCReport.txt file will be placed.

Using Astrometrica

When you run Astrometrica, it will take a while to load the MPCORB.DAT file. This file contains orbital data for known asteroids. A progress bar indicates how much of the file has been read so far.

Open Image

The first task is to open an image set for processing. Click the thumbnail at left to see what the Open dialog looks like. If you have more than one set of images in your CCD Images folder, select only the images that belong to a specific set. Typically, this will be four images.

The image at left shows four images open. The images are shown with inverted brightness; it is often easier to spot dim objects this way. To invert brightness, use the Images | Invert Display menu item.

Process Images

If your configuration is correct, processing the open images is very easy. Use the Astrometry | Moving Object Detection menu item to start processing. Our images have coordinate information saved in the file; you can click OK to bypass manually entering coordinates. You will see a progress bar while the images are processed. Depending on your configuration settings, processing may take from 10 seconds to 30 minutes! Be patient.

When processing is complete, the detected objects will be shown in the images. Blue indicates objects identified as stars but not in a catalog. Green indicates reference stars whose position and brightness are known from the catalog. The full list of what color is used for what type of object is found on the Environment tab of the program settings.

Find Moving Objects

Astrometrica looks for moving objects in the image set. This may include known and unknown asteroids and comets. Each detected moving object will be displayed in an animation (click the thumbnail at right to see a full-size version of the 'Veryify Object' dialog). You will see an animation of the frames at upper left. The identified object is shown with a red outline. By default, the object is moving against a stationary background. You can change the Zoom, Center, and Freq (speed) options to help you decide if the object is real. A graph of the object PSF (point spread function) and the background noise are shown at right of the animation. If the object is a known asteroid or comet, it's designation will be shown at bottom left. If there is more than one possible match, click the button between the two text boxes to pick from the available objects. If there is no match, enter your own designation in the text box to the right of the button. You should use a unique designation for each object, and it should be something that you can identify later. For example, if your name were Keith Edwards, you might use something like KED001 for your first unknown moving object. (Keep track of what you use, so that you don't reuse your custom designation!) You '''MUST''' use exactly six characters or the MPC report will not be correctly formatted!

If the object appears to be real, click the Accept button. Otherwise, click the Reject button. If there was more than one moving object found, you will go automatically to the next moving object.

The thumbnail at right will take you to an example image that shows a false find. Look at the graph at top right of the image - there is a single hot pixel. Note that the signal to noise ratio (SNR) is 6.2. Compare that to the true find in the previous example - the SNR there was a healthy 34.9. (If you get a lot of hot pixel 'finds', try a higher detection limit setting on the Program tab of the Program Settings dialog (File | Settings, or click on the wrench icon). When you are done reviewing possible moving objects, those that you accept will be displayed with the automatic or manual designation. See this image for an example.

It's possible that there will be known asteroid visible in the image that were not automatically detected. The thumbnail at right takes you to an image that illustrates this. Asteroid H9008 can just be made out inside the red square that is part of the overlay of known objects. That asteroid is magnitude 20.3, which is very dime and about at the detection limit for the Big Mak.

You can double-click on the square box to see the Object Verification dialog for that image. This is useful for examining the graph of the PSF and the noise in the background to help you determine if the asteroid is real. You can also use this technique to help you verify visually identified candidates when in Blink mode. See this image for an example.

To start blink mode, use the Tools | Blink Images menu item.

Tip: To manually add a moving object that you find visually, double-click on the object, and then enter a six-character designation as described above. Click the Accept button to add the object to an individual image. If you have four images in a set, then you will need to double-click on the object in all four images, enter the designation, and click Accept. Make sure you use the exact same designation in all images!

Reporting Astrometry to MPC

For each asteroid, known or unknown, that you Accept or create manually, Astrometrica will create entries in a report file. The location of the report file is set on the Environment tab of the Program Settings dialog. The name of the file is MPCReport.txt.

A typical MPCreport.txt file looks like this:

 COD G96
 OBS A. Block
 MEA A. Block
 TEL 0.6-m f/7.8 reflector + CCD
 ACK MPCReport file updated 2009.04.01 22:44:12
 AC2 aaaaaa@as.aaaaaaaa.edu
 NET UCAC-2
     adam37    C2009 03 17.34319 13 03 56.06 -04 01 57.5          18.9 V      G96
     adam37    C2009 03 17.34690 13 03 55.88 -04 01 56.7          19.1 V      G96
     adam37    C2009 03 17.35058 13 03 55.72 -04 01 55.8          18.9 V      G96
     adam37    C2009 03 17.35440 13 03 55.54 -04 01 55.0          19.0 V      G96
     adam38    C2009 03 17.34319 13 02 55.55 -04 00 17.3          19.2 V      G96
     adam38    C2009 03 17.34690 13 02 55.40 -04 00 15.5          19.2 V      G96
     adam38    C2009 03 17.35058 13 02 55.26 -04 00 13.9          19.1 V      G96
     adam38    C2009 03 17.35440 13 02 55.11 -04 00 12.0          19.2 V      G96
 ----- end -----


Note: The email address has been deliberately altered to avoid Spam for its owner.

Copy the contents of the text file, paste them into an email message, and send it to mpc@cfa.harvard.edu.

Please see the following link for detailed information on submitting reports to the MPC:

 http://www.cfa.harvard.edu/iau/info/TechInfo.html

Rss Comments
  • There are no comments for this article.
Info Add Comment
Nickname: Your Email: Subject: Comment:
Enter the code below: