Spring 1999
Nearby Supernova Campaign
Photometry Guide


TABLE OF CONTENTS


SUMMARY


INTRODUCTION

On the face of it, the goal of the photometric follow-up of this spring's supernova is to obtain lightcurves in standard bandpasses, such as B and V, from which peak magnitudes and stretches can be determined. However, the supernovae will be at a range of redshifts, and no CCD+filter combination can perfectly reproduce the standard bandpasses, so our observations will never be entirely true to the standard system without invocation of some of the very corrections we are trying to determine. I suggest that the real goal of the photometric and spectroscopic follow-up is to obtain a spectrum, flux versus wavelength, for all the SNe from the date of discovery through some period after maximum. From this viewpoint, photometry provides low-resolution spectra.

This may make it sound like we should just do spectroscopy; then if desired broadband flux measurements could be synthesized for the standard bandpasses. However, besides the obvious time/signal-to-noise limitations, it is very difficult to obtain accurate relative total flux measurements from spectroscopic observations. Examples of factors which make absolute spectrophotometry difficult include:

All of this is to remind you that for this program the meaning of a broadband magnitude is ambiguous, and to encourage you to think of photometry as low-resolution spectroscopy. From this perspective you can see that just having an image of a supernova is inadequate - we must be able to determine the flux on an absolute system and the effective bandpass. This perspective also points to the ambiguity of common photometric terms like "atmospheric extinction" and "color-term" in the context of observing redshifted objects with time-varying spectra.


INTERNAL CALIBRATION

Biases:

During the afternoon, and again in the morning, obtain about 11 bias frames. Make sure the dome is dark or that the instrument is absolutely light-tight; for typical CCD readout times light leaks can be a problem even though the shutter is never opened. If the bias has little or no 2-D structure this calibration will be sufficient. Most CCD's have an overscan region that will track the DC component of the bias. In the worst cases the bias will have significant 2-D structure, this structure can vary with time, and there might not be an overscan region. If you suspect this is the case, take bias frames periodically throughout the night for safety. It is not uncommon to see a little periodic time-variable 2-D structure in bias images cased by electronic pickup. Assess whether the level of this noise is significant enough to affect photometry, and if so, try to get a technical assistant to check it.

Dome Flats:

During the afternoon, obtain 5 domeflats in each filter. Exposures should be set so the brightest parts of the image (not necessarily individual pixels) are about 2/3 of the way to saturation. Make sure the telescope is close to the correct focus in each filter since this will affect how well dust shadows on the dewar window can be flattened.

Twilight Flats:

During twilight obtain 5-7 twilight flats, in each filter if possible. Exposures should be set so the brightest parts of the image (not necessarily individual pixels) are about 1/3-2/3 of the way to saturation. Make sure the telescope is close to the correct focus in each filter since this will affect how well dust shadows on the dewar window can be flattened. The twilight sky is not intrinsically "flat"; in general the flattest part of the twilight sky is about 15 degrees from the zenith in the direction opposite the sun. Adjust the telescope pointing around this canonical location to avoid the Galactic plane and bright stars (consult Norton's Star Atlas). Also make sure to dither the telescope by several arcminutes between exposures just in case there are bright stars in the field. The brightness of the twilight sky changes by a factor of 2 roughly every three minutes; this can be used to estimate the appropriate exposure time based on previous exposure times. If obtaining flats in several filters, start in the filter with the lowest flux, generally I, R, U, V, then B (blue light is scattered more, but also U response of CCD is usually poor). Of course since our most important filters are UBV, you will want to make sure to get these.

Darks:

Each CCD will have a dark current. Most of our photometry observations will be short enough that the mean dark current can be ignored. There can be hot pixels, however, and so dark images can be useful in checking whether supernova photometry is affected by any of these. Most of our runs are short (1 night) so it is difficult to obtain good quality dark frames, since a good meaurement requires several (for cosmic ray rejection) long (to build up enough signal that isn't lost due to charge-transfer inefficiency) exposures in a completely dark dome. Certainly if the night is cloudy and a given CCD has been/will be used more than once, take several 3600 second dark frames just so they are available.

FOCUSING

Since observations are needed in UBVRI, focusing can be a hassle. Changes in focus with bandpass are principly due to differences in filter glass thickness and refraction indices, so the relative focus values between filters are constant. These values are often well known, and may be posted in the observatory or on a web page. If not, determining relative focus values at the start of a run is important, and once determined should be archived (e.g. on our web page). Note that it is typical for the telescope structure to cool significantly during the start of the night, and these changes will also affect the focus. Therefore, make sure that filter offsets are accurate and not due to cooling over the course of taking focus measurements.

When taking focus series, look for additional clues in the images that may help later in differentiating between poor focus and bad seeing. For instance, on many telescopes the images suffer noticable astigmatism on either side of focus. The orientation of the elliptical image will rotate by 90 degrees on either side of focus. Also look at the crispness of the diffraction spikes on bright stars - the spikes will be doubled if the focus is poor. Do not confuse diffraction spikes with blooming caused by saturation of the CCD. Exposure series should have integration times of roughly 10 seconds at each focus setting, and I recommend that at least 7 focus settings be tried. The longer exposures do a better job of averaging over seeing variations and the range of focus settings helps average over longer-term seeing variations and helps define the focus curve. Note that a focus curve is not a parabola; once the instrumental defocus is significantly less than the atmospheric seeing the focus curve will "flatline" (except for seeing variations) for several focus settings until the defocus once again becomes significant. If the focus curve is fairly flat this indicates that the focus step size was too small and never reached values with significant defocus. Note that it is important to determine the focus well even if the seeing is poor, since if the seeing improves you do not want to have to spend time refocusing. In general the focus will depend on telescope support temperature and telescope zenith angle (due to gravitational loading). The sensitivity of various telescopes to these factors can vary widely. Again, sometimes the temperature and/or zenith-angle dependence of the focus is known and posted.

STANDARD STAR OBSERVATIONS

External calibration is obtained by observing stars with known brightnesses in standard bandpasses. The most basic photometric quantities are the flux zero-point, the atmospheric extinction as a function of airmass, and whether the flux zero-point depends on the colors of the standard stars (color term). That is photometry at its most basic, however the real story is more complicated. For example, atmospheric extinction is wavelength dependent (modifying the weighting across the filter bandpass), and can be time dependent as well since aerosols and molecules have different wavelength-dependent scattering and absorption properties and the aerosal content of the atmosphere can change during the night (e.g. as an atmospheric inversion layer sinks below the summit during the night). A color-term is a simple linear approximation to correct for the fact that the bandpass in use may be shifted or have a different shape than a standard bandpass, but the combination of filter bandpass and supernova spectrum could easily require non-linear, and even non-monotonic, corrections.

Viewed spectroscopically the SN spectrum is simply being measured over a slightly non-standard wavelength interval and with a slightly non-standard weighting across the resolution element (filter bandpass). Observations of standard stars help to constrain atmospheric extinction and non-standard bandpasses, but ultimately they are most useful in this context when observations are able provide linearly (or higher order) independent estimates of these quantities.

Thus, you should not assume that observations of a few standard stars at low airmass will provide sufficient calibration. Observations of standard stars need to be obtained over a range of airmasses (up to airmass of 3), colors, and times, and it is critical to decorrelate these parameters (e.g. red stars and blue stars must be observed at both high and low airmass and at the start and end of the night). This kind of aggressive calibration is not necessary every night, but must be done at least once for each telescope, detector, filter, SN-field combination. If the calibration situation is desparate, at least try to observe standard star fields having stars with a range of colors fitting onto one CCD field; this allows an approximate determination of the system color terms.

Since standard stars are fairly bright, they can be observed during twilight. Unless the standard-star field is crowded, the focus doesn't have to be perfect. During evening twilight, you may want to dial-in canonical focus values if they known; otherwise you will have to determine the focus before observing standard stars during the period when the telescope is cooling. It is important to plan standard star observations well in order to obtain a good range of airmasses and colors. For many standard star fields, several standard stars can fit in the field of view but of course you need to determine the telescope coordinates that will place the field in the correct location. Some such fields have been composed and are listed below. Also, by turning on standard stars and using the zoom feature in TheSky you can compose additional standard star fields). As mentioned earlier, try to decorrelate airmass, color, and time. For instance, if you plan to observe three standard star fields at the start of the night, do not observe them in order of their airmass, otherwise airmass and time will be correlated. Also, it can be useful in establishing the quality of the night to make repeat observations, for instance UBVRI and then U again. Standard star observations should also be taken during the night. These observations are basically a check on changes in atmospheric extinction since the system response is usually stable. Thus, it may be sufficient to take observations in extinction sensitive filters, i.e., U, B, and maybe V. Note that if the CCD temperature changes, the R and I response will change. Therefore, make sure that the CCD temperature is kept stable, and note any significant changes (1 degree or more).

In order to be able to determine cross-telescope calibration very precisely, every telescope must at some point obtain observations of certain fundamental reference standard star fields under photometric conditions. These are:

Other useful faint standard star fields are listed below, centered for large and small fields of view, and with a tally of the number of blue, green, yellow, and red stars:

Note that Peter Stetson has added numerous standard stars to many of these Landolt 1992 fields. These data can be obtained at CADC.

Note that many of the faint standard star sequences are too faint to be used with smaller telescopes in some bandpasses (namely, U and B at the Lick 1-m). Instead, brighter Landolt standards can be used. A useful list of these, along with finder charts, can be found on Lick Observatory standard star page.

In order to test photometric transformations, we would also like observations of the following spectrophotometric standard stars taken under photometric conditions:


SUPERNOVA OBSERVATIONS

Photometric observations of our supernovae should be fairly straight-forward, although the observing pace will be busy when the SNe are bright and/or on large-aperture telescopes. We expect to have finder charts and suggested exposure times available on the web, and we hope to be able to provide suggested nightly schedules customized for each telescope. If a nightly schedule is not available then you will need to plan the night's observations during the afternoon. The fast pace of the observations will make it difficult to "wing-it". Also during the afternoon, print-out the finder charts (or go prepared with them) and the latest version of the follow-up status since you never know when the Internet will bulk.

Start by moving to the nominal coordinates of the SN; most computer controlled telescopes will point to within a small fraction of the size of the CCD field. Use your experience, but generally it will not be efficient to try to confirm the pointing in the acquisition camera. Acquire a guide star using the available offset guider - make sure the guider probe does not vignette the CCD field of view. At this stage, offset the telescope in one of the cardinal directions by roughly 1/4 of a field width and reacquire the guide star. Choose a different direction for each different SN field. The goal here is to make it possible to construct sky flats; more complicated dithering is better but not mandatory.

Cycle through observations in each filter using the suggested exposure times. After the first filter observation has been read-out and the next exposure started, confirm that you have the correct field using the finder chart. Note that for short exposures we are not concerned with cosmic-ray rejection, so it is not necessary to take multiple exposures unless inspection of the images shows that the SN is affected by a cosmic-ray hit or cosmetic defect of the CCD. If the exposure times are longer than about 4 minutes, split the observation into two separate exposures. If it is efficient to do so, run through the first set of exposures in all filters and then run through the second exposures. This can help determine the photometric quality of the night. Along the way, make sure to monitor the focus. Also monitor the seeing, and if the seeing is significantly different than assumed when the suggested exposure times were calculated you will need to calculate new exposure times and a revised schedule using the tools in WhatsUp.

In the case of unconfirmed candidates, you may be requested to compare the brightness of a candidate from one night to the next, or to determine the candidate's color, so post-max SNe can be screened out. Measuring a change in brightness requires measuring the candidate and a bright, but unsaturated, reference star in images from two different nights. Since this involves relative photometry, observations taken on non-photometric nights can still be used. Determining the color is more complicated. The zero-points in the two filters must be known and the observations must be obtained in nearly photometric conditions.

At most observatories, important observational parameters are stored in the image headers. For quick reference it can also be useful to record some of these - especially those that are changed often or which must be closely monitored - in a log. Make sure the object name, right ascension, declination, exposure time, universal time, filter, and some running number are recorded. It is usually useful to also record the hour angle, airmass, telescope focus, telescope (truss) temperature, CCD temperature, relative humidity, and atmospheric pressure. For telescopes used often, it can also be useful to record the guider coordinates of a good guide star so it can be reacquired on subsequent runs.

If you are able to use WhatsUp, use the Candidates tool to check-out observations you plan to make and to check-in whether the observations were completed successfully. This is vital to avoid redundant observations. WhatsUp also provides an exposure time calculator and other tools to help plan and monitor observations.

DATA ARCHIVING

Since many of our runs are only a single night, plan carefully how to archive the data so that you don't have to stay up all the next morning writing tapes or FTP'ing data. At some observatories the data can be automatically written to tape when the detector is read-out. In this case beware of power glitches or other errors that could cause the tape to be rewound, and subsequently overwritten, by accident. Always make two copies of the data, preferable on tape, and store them separately.

It is wise to check that the data you think is on tape is really there. A good approach is to use the log of tape files to construct a list of images to delete and then delete only those images; if there are science images left over that means they didn't make it on the tape! Label all tapes with the date, telescope, detector, observer name, file format (e.g. FITS), archive format (e.g. tar), and number of images. Especially note if the data for a night/run spans more than one tape, or if there are multiple tars on one tape.

For instructions on how to FTP the data to LBL, click on FTP Instructions.

CAUTIONS

Linearity:

As the number electrons per pixel approaches the CCD full-well depth, the CCD response will become non-linear. Do not assume that the A/D convertor saturation and CCD full-well staturation are the same. For some CCD's the A/D convertor saturates before the CCD full-well saturates, well sometimes the opposite is true. If the CCD full-well saturates before the A/D convertor saturates, it is important to know where non-linearity begins to occur. At some observatories this information has been determined and recorded in the instrument documentation. Alternatively, a linearity curve can be constructed by taking domeflats with exposure time chosen to fill the CCD wells at a range of levels up to A/D saturation. Such data are most useful if a reference image is taken after each data image. This allows a check on variations due to unstable dome lamps or changing dome illumination from light leaking into the dome. For example, take 20 exposures ranging from 5 seconds to 100 seconds, with 20-second exposures interspersed.

Shutter Function:

Due to mechanical latency, true exposure times and requested exposure times can differ by a constant offset. At better observatories the value of such an offset will be known. Try to chose exposure times (especially for standard stars) at least 100x longer than this latency time. If this is impractial, then make sure this information is propagated through when the data are reduced. If the latency time is not know, you might consider obtaining a linearity curve during the afternoon, concentrating on short exposure times (less than 5 seconds). The linearity curve must be accurate enough to detect a latancy larger than 1/100 of the shortest standard star exposures you plan to use.

Besides a constant time offset across the field, the exposure time can vary across the CCD. For instance a leaf shutter will have slightly longer exposure times at the center of the CCD than at the edges. A way to test for this is to take a very short exposure with very bright dome lights (to get good signal) and then a longer exposure with normal domelights. After subtracting the bias level, divide these two images, smooth if necessary, and look for radial or ramp stucture. With very high S/N data sometimes the leaves of a leave shutter can be detected. If a differentional shutter term is detected, this must be taken into account when selecting standard stars, dome-flat, and twilight-flat exposures.

Amplifier Cross-talk:

It is becoming more common with large-format CCD's to use more than one amplifier to speed up the readout time. It is not unusual for cross-talk to occur between amplifiers. One problem that can occur in this case is that when a bright object is clocked through one amplifier it perturbs the signal reported by the other amplifiers. If these electronic "ghost images" are severe it might be nessary to use only one amplifier. This should be the exception rather than the rule. Again, ask the local experts or consult local documentation if you suspect cross-talk problems.

APPENDIX I - Quick Photometry

There will be many occasions when you need to measure the brightness and/or size of an image. These include focusing, measuring the seeing, obtaining rough photometric zeropoints, obtaining absolute or differential SN photometry. Here is a brief presentation of how to do quick photometry in IRAF.

It is assumed that you have IRAF installed and have a working IRAF set-up for your account (i.e. that mkiraf has been run to create login.cl and loginuser.cl, and a uparm directory. % is used to represent the system prompt, cl> is the IRAF prompt, and : is the IMEXAMINE graphics window prompt.

If you are not already in IRAF, start it as follows:

Set the display buffer to accomodate your image. For a 2048x2048 image do:

Then display the image and start-up the image examination tool. If you will be comparing two more more images, they can be loaded into the other available display buffers:

You will now notice a flashing circular cursor on the screen. Center a random star within the cursor and press r. You will see a graphics window come up showing the radial profile of the star and a radial Gaussian fit to the profile. Image parameters, including the FWHM are printed at the bottom of the graphics screen. Note the value of the FWHM for later use.

At this stage move the cursor back onto the image and press :. You will see the colon appear in the graphics window. The cursor will also try to move to the graphics window; if it is not in the graphics window, move it there. You now need to set-up several important parameters, many of which will depend on the FWHM. If you are comparing two images, determine the FWHM for both and use the larger value in setting the following parameters:

Now measure the brightness of several reference stars and the SN itself using the a key. This will report the background- subtracted flux of the stars and SN within an aperture of radius 2*FWHM. Use :? to get help on the available commands in imexamine, and use q to quit.

If you are comparing two images, first select the 2nd image buffer in ximtool to bring up the second image (displayed as in the example above) and conduct the measurements using the a key in exactly the same way. Using the measured fluxes, determine whether the ratio of fluxes of the SN compared to the ratio of the fluxes of the reference stars. This will enable you to determine whether the SN has brightened or faded (provided you are comparing measurements taken with similar filters).

APPENDIX II - Imaging Zeropoints


                   True Zero-points
               U     B     V     R     I
------------------------------------------
WIYN         23.67 25.92 26.14 26.19 25.52
KPNO 2.1m    22.04 24.34 24.92 25.08 24.42
CTIO 1.5m    22.04 23.79 24.22 24.32 23.66
CTIO 0.9m    21.00 22.74 23.17 23.28 22.62
Mt. Laguna   20.85 23.33 23.71 ----- 23.28
YALO 1.0m    19.44 22.82 23.30 23.34 22.77

                Estimated Zero-points
               U     B     V     R     I
------------------------------------------
JKT  1.0m    21.00 22.74 23.17 23.28 22.62
Lick 1.0m    21.00 22.74 23.17 23.28 22.62
MDM  1.3m    21.50 23.24 23.67 23.78 23.12
MDM  2.4m    22.04 24.34 24.92 25.08 24.42

Here the zeropoint magnitude is defined as
m = -2.5*log10(N photoelectrons/sec) + m_zero
at an airmass, X = 1.

APPENDIX III - Checklist

Under development