How to obtain a
Originated 2017 January 14
This tutorial includes;
– a definition of astrometry and its uses
- CCD basics
– what is needed to perform accurate astrometric measurements (to better than 1 arc
– how to obtain an Observatory Code from the Minor Planet Center (MPC) which
qualifies you to submit astrometry to that organisation. (note that an MPC Observatory Code is not needed to submit comet observations in ICQ format)
The ‘Guide to Minor Body Astrometry’ on the MPC website is a MUST READ. In addition ‘Astrometry How-to’ by Tim Spahr, Director of the MPC, located at the end of this tutorial may help you avoid many of the common astrometric pitfalls.
What why and how
What is astrometry?
Astrometry in its broadest sense is the measurement of positions, parallaxes and proper motions of an astronomical body on the sky. In this tutorial I will use a narrower definition – the measurement of the position of an asteroid in terms of its Right Ascension (RA) and Declination (Dec) known as equatorial coordinates. For nearby objects such as comets and asteroids, in particular Near Earth Objects, these coordinates can vary slightly depending on your location on the Earth. Hence the importance, when obtaining an ephemeris from the MPC for example, of specifying one’s exact position which includes latitude, longitude and height. When you get your Observatory Code then inputting that will supply these parameters.
Why is it needed?
Accurate astrometry is necessary to:
– determine and/or refine the orbit of a newly discovered comet or asteroid,
– prevention of loss and recovery of lost comets and asteroids,
How is it done?
In the days of astrophotography measurement of position was a complicated and time consuming task involving the use of a mechanical plate measuring engine, Figure 1, – a single measurement could take several hours to complete. Nowadays it requires: a CCD image, an accurate star catalogue, an astrometry software package and a PC – measurement time is reduced to minutes if not seconds. The accuracy of the position so determined will, to a large extent, only be as good as the accuracy of the reference stars used. Astrometrica allows you to select from several catalogues which it accesses via the internet, or which may be stored locally on your PC – UCAC 4 is recommended for astrometry.
Figure 1. Plate measuring machine. Credit
Tools of the trade
Here I describe my set-up (as was) to give you some ideas as to how you might like to proceed. Mine started out as a visual observatory so may not by untypical for newcomers to the world of the amateur astronomer. When I began observing I had a vague notion that I would proceed from visual observations to photographic imaging. After several failures with film and with affordable CCD cameras becoming more readily available I abandoned the former for the latter.
My very first telescope was a cheap disaster, my second was a very good, very rugged TAL 10cm (4in) reflector and my third and present instrument a 25cm (10in) Newtonian reflector. The latter coupled with a Starlight Xpress MX516 CCD camera allows me to reach approximately mag 15 with a 30 sec exposure. So I would suggest that a 25cm (10in) reflector is a good starting point but, if you can afford it, then the larger the better. Naturally the larger the telescope the more expensive it will be and you will need a larger, more costly observatory to house it.
I should mention robotic telescopes accessible over the internet at this point. Images do cost money but, compared with the cost of purchasing your own telescope, camera and observatory this may not seem too large a burden. Other plusses are access to larger telescopes in both hemispheres. Organisation such as Sierra Stars Observatory Network (SSON) and itelescope have reasonably priced starter packages - you can find tutorials on this website at http://www.britastro.org/projectalcock/Using%20the%20SSON%20robotic%20telescope.htm and http://www.britastro.org/projectalcock/iTELESCOPE.htm
A schematic of my set-up is shown in Figure 2 It includes a:
– reflecting telescope on a go-to equatorial mount controlled by Megastar via a Skysensor
– CCD camera controlled by Astroart. You will need a CCD camera with an FOV of at
least 10×10 arc mins or you may find there are too few reference stars on the image
Mine is 10×8 arc mins and I do have to check the number of stars in the FOV when
choosing targets as sometimes there are too few
– Laptop PC
software available with an internet connection is another option
– CD R/W drive to save images.
Figure 2. My set up (as it was some years ago).
Your setup may well be very different but, in summary, you will need: a telescope on a motorised mount, a means of controlling it, a CCD camera, imaging software, an accurate time source and a large amount of storage for your images. You will notice an aerial in Figure 2 – this connected my observatory laptop via a wireless router to a PC in my study which allowed me to control matters from a warm room once I had everything set up.
Starlight to image
A CCD camera consists of a matrix of picture elements, pixels, which convert the incident light into electrons which are fed to a computer and converted into an image. It is perhaps worthwhile just dwelling for a few moments on how pixels convert photons into electrons, Figure 3, as this will help to understand the meaning of, and the need for dark frames, flat fields and image calibration.
Figure 3. Pixels – Light to image
Perceived wisdom is that a resolution of approximately 2 arc secs/pixel gives satisfactory results in terms of astrometric (and photometric) accuracy. What you have, or can afford, in the way of a telescope or CCD camera may govern the result. For example when I started CCD imaging I already possessed a 10in (25cm) Newtonian reflector and my pockets were only deep enough for the cheapest Starlight Xpress CCD camera. You may be able to compensate for too high a resolution by binning – whereby the output from several pixels is combined electronically.
A little maths to show how to calculate the CCD field of view (FOV) and resolution (using my equipment as an example):
– First calculate the plate scale which is (206265)/Focal length of the telescope in mm =
206265/1626 = 126.9 arc sec/mm,
– Then calculate the size of the CCD chip from pixel size (0.0098×0.0126 mm) and
number of pixels on the chip (500×290) = 4.9 x 3.6 mm,
– The FOV is therefore ((126.9×4.9)/60)×((126.9×3.6)/60) = 10.4×7.6 arc mins,
– The resolution is the plate scale divided by the number of pixels/mm = 126.9/89 = 1.4
The need to calibrate
Each pixel generates electrons which are fed to your computer in the form of electric current to form an image. The number of electrons generated by each pixel depends on the amount of incident light (photons), and the always present dark current. The output of each pixel is measured in Analogue Digital Units (ADUs) - the maximum, full-well or saturated value being usually being 65,536. To obtain a true representation of the area of sky imaged, the dark current must be subtracted and the variation in number of electrons generated in each pixel by the same quantity of incident photons allowed for. Calibrating the images - applying dark frames and flat fields will perform the necessary corrections. While this might sound complicated it is quite easy to do or rather imaging processing software will do the job for you. Note that the CCD camera should not be moved until all images and calibration frames have been obtained and that the latter should be obtained during each imaging session.
A dark frame is obtained by taking an exposure of the same duration as the image while preventing any light falling on the CCD. This can be done quite simply by fitting a light tight cap on the end of the telescope tube, Figure 4. Best practice is to take a number of dark frames, at least five, and combine them. Photometry software packages such as AIP4WIN have facilities to do this.
Figure 4. Light tight cap for obtaining dark frames.
A flat field is slightly more complicated in that you need to take an image of a uniformly illuminated surface. There are a number of ways of doing this but I have found a ‘light bin’ to give satisfactory results. ‘Bin’ because I made it from a plastic waste bin as shown in Figure 6. Images are obtained with the telescope pointed at the illuminated surface – a minimum of 3 seconds exposure is recommended and the ADU value of any pixel should not exceed around 50% of maximum. Imaging software such as Astroart will provide the required image statistics telling you whether you have under or over exposed the flat field image.
to dark frames of the same duration as your images you will also need dark
frames of the same duration as your flat fields. Some software packages such as
several calibration frames in
Light bin for obtaining flat fields. Credit
Obtaining an observatory code
This section explains how to obtain, or rather how I obtained, an observatory code including: selecting the target asteroids, the basics of imaging and the measurement of those images. The same procedure can also be applied to the projects described in the following chapter. Please note that although your set-up may be different the procedure will be essentially the same.
Why do you need one?
Observatory codes are issued by the MPC when they are satisfied with the accuracy of your astrometry. Having obtained such a code you can then submit astrometric data to them.
Choosing asteroids to image
How to obtain your observatory code is illustrated by the procedure I followed towards the end of the year 2000 to obtain that code for my home observatory. As suggested on the MPC website I selected six asteroids with numbers in the range 400 – 3000. Using Megastar I chose asteroids that were in the south east to ensure that they would be visible for a number of nights and thus avoid having to restart the exercise should I experience a period of cloudy skies (perusing your long-range weather forecast before starting this exercise may help you to avoid such a trap). It also helps to reduce time spent moving between asteroids if they are as close together as possible.
Asteroids, between magnitudes 13 and 16, chosen were:
(3259) Brownlee Discovered by J Platt at Palomar in 1984,
(403) Cyane Discovered by A Charlois at Nice in 1895,
(1040) Klumpkea Discovered
by B Jekhovsky at
(1719) Jens Discovered
by K Reinmuth at
(1468) Zomba Discovered
Discovered by A Kopff at
Choosing brighter asteroids (in CCD terms) allows the exposure times to be kept fairly short and therefore minimises tracking errors.
Imaging the asteroids
The method I use to locate the first asteroid, frowned upon by some, is to move my telescope to the required coordinates using Megastar’s goto facility with a CCD framing eyepiece installed. I then replace (very carefully so as not to move the telescope) the eyepiece with the CCD camera. The camera and drawtube are marked so that the camera is installed in the correct orientation. A stop ring or mark on the draw tube aids focusing which may need a slight adjustment after the first images have been obtained. I focus by eye by taking short exposure time images and adjusting the focus manually, as I would if an eyepiece were installed, until the images of stars are as sharp as possible. This should suffice for the whole imaging session unless, for example, the temperature changes significantly. Figure 7 show the Starlight Xpress MX516 camera mounted on my telescope. The stop ring can be seen above the focusing knob.
Figure 7. Starlight Xpress CCD camera with stop ring in position.
To assist focusing a mask, Figure 8, can be placed over the end of the telescope similar to the placement of the light tight cap. The mask is made from cardboard with three holes, making an equilateral triangle, cut out. If the image is out of focus each star will be depicted by a cluster of three dots. The focus is adjusted until these three dots merge into one.
Figure 8. Focusing mask.
The six asteroids were each observed over a period of one hour on two separate nights. It took approximately three weeks to complete all the observations. Having aligned the telescope on the target asteroid a typical observation run proceeded as follows (note that I have amended the process I used then to reflect how I would do it now):
– Set the
PC to the correct time using a
port (If you have an internet connection handy then a clock sync program, Dimension
4 is recommended, can be downloaded).
– Enter the asteroid name and number into the FITS header (images are always saved in
– Take a number of test images to verify tracking, focus and that the pixels are not
saturated. A maximum of 2/3 of the full-well value for any pixel is recommended.
The maximum exposure time is also limited by the ability of your telescope to track
accurately, and the motion of the asteroid. The maximum exposure time to
avoid trailing due to the motion of the asteroid is given by the equation;
Exposure(secs) = 60x(image scale in arcsecs/pixel)/(motion in arcsecs/min)
– Set the Astroart software to: Continuous, 5 min delay between exposures, Autosave
image as a FITS file. This allows all the image data to be saved including the date and
time of the exposure. The MPC requires that the mid-points of the exposure times are
submitted so check that your imaging software is assigning this time rather than the
start time of the exposure for example. Your astrometry software may allow you to
input whether the time in the FITS header is start, middle or end of exposure and, from
this, calculate the mid-time. The delay between images ensures the PC clock resets to
the correct time if the software you use freezes the clock while the image is
downloaded to your PC.
– Image for a period of one hour (after the first few images were obtained I checked to
ensure that they had been saved correctly – well worth doing).
– Obtain dark frames. At this time I did not obtain flat fields, as I was not planning to
include magnitudes in my report.
A Goto feature on your telescope certainly helps as you can image one asteroid and fairly easily switch to the next, image it and so on (remember that each asteroid must be observed on two separate nights). If you have a Newtonian reflector try and stay away from the meridian so that you don’t have to reverse the telescope as it passes that point.
– From the dozen or so images obtained for an asteroid during each of the two evenings it
was observed, I selected three images taken roughly fifteen minutes apart.
– Each image was calibrated by subtracting the dark frame and applying a flat field but
don’t process the images in any other way.
– The position of the asteroid was measured using Astrometrica and the USNO-B1.0
catalogue (suggest you use the UCAC-4 catalogue). Tutorials are available on the
Astrometrica website and it is quite easy to use. Briefly:
– Set up the configuration file,
– Load dark frame and flat field masters
– Load images,
– Carry out astrometric data reduction (if you are unsure as to where the asteroid is in
your image compare it with a star chart – Astrometrica circles the stars so picking out
the asteroid should not be too difficult) . Alternatively you could use the blink facility.
Figure 9 shows the Object Verification window which appears when you click on the
asteroid in your image. Entering the asteroid designation and clicking on Accept will
add the relevant data to the MPC Report File. Figure 10 shows an image after
astrometric data reduction – the asteroid is now circled and numbered.
Figure 9. Astrometrica Object Verification screenshot.
Figure 10. Image after astrometric data reduction
Verification of results
It is wise
to check ones results prior to submission to the MPC. This can be done by
comparing your astrometry with an ephemeris from the
Figure 11. Astrometry plotted using Guide.
Submission of results to the
There is a very specific format that MUST be used to report observations to the MPC – an example report is shown below. Software such as the widely used and recommended Astrometrica will produce reports in the required format. Whatever software you use do ensure your observations are reported in plain ASCII format (plain text not, for example, HTML format). Each line in the body of the report is 80 characters long so set your emailer to automatically break lines at a slightly larger number.
TEL 0.3-m f/5.9 reflector + CCD
00941 C2008 12 07.70371 07 28 46.59 +17 25 54.9 15.8V XXX
00941 C2008 12 08.66792 07 48 38.62 +17 43 33.8 15.4V XXX
00941 C2008 12 08.67245 07 48 44.10 +17 43 36.8 15.4V XXX
00941 C2008 12 08.67695 07 48 49.64 +17 43 41.5 15.2V XXX
00941 C2008 12 08.68144 07 48 55.27 +17 43 45.3 15.2V XXX
The lines above the actual observations start with a code describing the information included in that particular line:
OBS – the observer,
TEL – description of the telescope used,
AC2 – email address to which the MPC will respond,
of observational data includes: asteroid number, C for CCD observation, date,
RA, Dec, magnitude, band (V in this case) and XXX as you do not yet have an
observatory code. Magnitude should only be included if you are confident that
this is accurate to at least ± 0.1 mag.
The first time observations are reported additional information, which may be
submitted using the COM prefix on each line, is required by the
– Postal address
– Observatory name and site
– Observatory position: longitude, latitude, height above sea level and source of
co-ordinates. These can be obtained using a
accurate co-ordinates take the average values over a long period of time (The MPC now
recommend using Google Earth),
– Details of telescope set-up.
A few days after submitting my measurements to the MPC I received a short email to the effect ‘Your site is now observatory code 940’. I also received a slap on the wrist for submitting the observations in the wrong format – but that only goes to show that the MPC staff are quite helpful as they didn’t reject my submission but translated it to the correct, ASCII or plain text, format.
An Observatory Code is specific to an observatory so if you use another telescope at a different site you will need to ensure that it has an observatory code. When using robotic telescopes check if this is so and how observations should be submitted to the MPC i.e. who is the observer and who is the measurer. As an example the Sierra Stars Observatory Network allow me to submit observations, made using their observatories, as the named observer and measurer.
Do keep a record of your observing sessions, any problems you experienced, how you overcame them and how you processed your images. It really isn’t too difficult – all you need are clear nights to obtain the images and the time and patience to process them. Astrometrica has been and will be mentioned many times in this book. In the view of many amateur and professional astronomers there is no better astrometric software.
Astrometry how-to by Tim Spahr, MPC Director
We all too frequently receive error-filled submissions of astrometry here at the Minor Planet Center (MPC), and I've decided to write a little "how-to" describing what common pitfalls can be avoided, and how to improve, in general, what folks submit to the MPC.
First, please read the Guide to Minor Body Astrometry at: http://cfa-www.harvard.edu/iau/info/Astrometry.html
That document and this ‘how-to’ overlap significantly so please read both carefully.
Minor Planet Center
Former observer with the Catalina Sky Survey (
Most common problems
False detections due to hot pixels, bad pixels, poor flat-fielding and internal reflections.
These are the most common mistakes made in data submitted to the MPC. It isn’t just amateurs who make this mistake – professional observers still report false detections of this type every week. The key is to avoid lining up these objects and "creating" a real object. By far the easiest thing to do is simply dither the telescope between successive images. This will eliminate 99% of all false detections immediately. At
CSS, we had this problem and we eventually settled on simply moving the telescope slightly before each exposure. An example pattern is as follows: image 1 was at the expected coordinates, image 2 was 30 arcsecs north of this location, image 3 was 30 arcsec south of the expected coordinates, and image 4 was 30 arcsec south and west of the expected coordinates. This simple procedure makes it nearly impossible for anything fake to line up with linear motion within a few arcsecs tolerance.
The second problem with false detections is using a bare minimum of images. I would never use two images for discovery purposes unless I had very small pixels (0.5" /pixel or so) and I had a good point-spread function for both detections. Three images is, in my opinion, the bare minimum for consideration however I do not favor this method. Simply use more than 3 images, and dither the telescope for best results.
Another very common mistake is absolute bare-minimum time intervals between images. It is best to have at least 30 minutes worth of coverage on each and every object. Note that
CSS, Lincoln Near Earth Asteroid Research (LINEAR) and Spacewatch all have t > 30 min. LINEAR averages around 70 minutes between images on each object. This gives a more robust Vaisala orbit, better linking probabilities on the subsequent nights, and lastly, prevents bad links by the observer and the MPC. These short-interval links are often misidentified, or worse, spurious objects! The MPC now requires t > 20 minutes for designations except in extreme circumstances. Keep in mind this is t > 20 min on each night. In addition, almost all false detections won't show linear motion over a 20 or 30 minute interval, but can show nearly linear motion on shorter timescales.
Obviously bad astrometry
It is surprising that this is the second most common problem, and perhaps the most annoying. We often get observations of objects that are so clearly wrong, a simple cursory examination would show this. For example, we still receive reports every few days of objects that are reversing direction of motion between 3 measurements, or worse, two good positions and one that is off by 5 arcmins or more.
Also in this category are bad links by the observer which happen on a weekly basis. For example, an observer will go to the expected coordinates for an object on the MPC’s Near Earth Object Confirmation Page (NEOCP), find a bright object, measure it, and send it in as the NEOCP object only to find later this was a routine numbered asteroid. This is precisely why the NEOCP, and the MPC Ephemeris Service, provide you with the speed and direction of the object in question. I'm sure you'd be surprised how many routine MBAs moving 30arcsecs/hour are turned in where the observer thought this was an NEOCP object that was supposed to be moving 300arcsecs/hour. We also receive bad links where observers simply went to the MPChecker, found the object closest to their object, and pasted this designation in the observation string. This causes me no end of grief, because in most cases, it is simply easier to paste a new observer-assigned temporary designation on each object submitted. In this fashion all "new" objects are identified by our automatic software, and will receive new provisional designations which are e-mailed automatically. This also reduces e-mail traffic back-and-forth between parties (one thing that definitely helps me, since I get a few hundred e-mails per day).
Of all things, this should never be a problem, but it occurs frequently. It is imperative that the observer check the clock each and every time observations are taken. Badly timed images are not only an amateur problem – every single currently active professional survey has had some sort of timing problem, from bad local – UT time corrections, bad computer clock time, mistiming the exposure start and end, and miscalculating the midpoint of the exposure. Yes, these were all done in one form or another by professional surveys. Please do be careful here as there's no excuse for this error and trust me, I'm talking from experience. At one time
CSShad a minus 12 second error on all images due to improper coding of the start time in the FITS header and this is more or less my fault for not checking it!
Junk astrometric solutions
Another frequent problem is astrometry that is clearly the right object but is also clearly incorrect. This arises in some cases for horrible or non-converged fits on astrometric solutions. If your RMS on your solution is extremely small and uses only 3 or 4 stars, it is probably wrong. Likewise, if it is over about 0.7 arcsecs, you've also probably got a problem – keep a close eye on those solutions!
Over-observing bright objects (Not an astrometric problem, but something of note)
A sizeable number of objects need no astrometry whatsoever, and yet we still will receive literally thousands of observations each month of these objects. Bright numbered asteroids, unless they are occultation targets, radar targets or mission targets, really don't need astrometry, so don't go out of your way to target them. You may, and should, measure them if they just happen to appear in your frames but targeted astrometric observations of routine numbered MBAs by amateur astronomers are almost certainly a waste of time. Likewise, many non-numbered NEOs are absolutely hammered by amateurs for no apparent reason. A good rule of thumb is only to observe objects that you and your system can actually help. So if the current ephemeris uncertainty is only 0.3 arcsecs, there is no possible way you can dramatically improve the orbit if your astrometry is only good to 0.5 arcsecs. Also, given the sky coverage and sheer number of professional surveys in action, it is very likely that your hard work will simply be obviated the next night by a survey.
Please cease and desist from sending single positions of any object on a given night, unless this object is spectacularly important, and then only do so with gratuitous comments regarding the accuracy of the measure, and why no other measures were obtained. For example, this might be acceptable for NEOs and comets from skilled observers but single, isolated, positions for MBAs are very likely to be deleted.
Notes regarding professional programs
At this point, it is probably important for amateurs to know a thing or two about the professional programs. LINEAR takes five images of each field, total spacing about 1 hour.
CSS, Siding Spring Survey ( SSS), and Lowell Observatory Near Earth Object Search (LONEOS) all take four images of each field, with intervals varying from 20 – 60 minutes from first to last image. Spacewatch and Near Earth Asteroid Tracking (NEAT) take only 3 images, spaced by 20 – 60 minutes, but they both have small ( ~1 arsec) pixels. Each and every one of the aforementioned programs submits all objects as new objects meaning, each object observed on a given night has its own, observer-assigned unique temporary designation. These observations pass flawlessly through our automatic processing code and the remaining one-night objects that might be NEOs are left for further examination. Given that a good deal of MPC effort has been put forth to process the bulk of the data this way, other observers should consider operating in a like manner.