[AAVSO-HEN] swift era photometry (long)

[Arne Henden]

Swift is about to be launched (November 8; less than three
weeks from now).  We should probably start thinking about
observing plans to best support Swift, while continuing to
observe HETE and INTEGRAL bursts.
   Swift's UVOT is an intensified CCD, great for photon counting
and low count rates, but with a definite blue response (spectral
range 170-650nm).  The filter complement includes Johnson U,B,V
but nothing redder.  They expect to get to the afterglow within
a minute or so after detection with the gamma-ray telescope, and
then to execute a preset observing plan.  First, it takes a 100second
exposure of the field for a finding chart; this will be telemetered
to the ground within 50seconds.  So you have about 4 minutes from
first detection until the first CCD image is received on the ground.
Then it starts a preprogrammed mode of observing, still TBD from
what I understand.  They could cycle through filters, use one of
the two grisms to obtain low resolution spectra, or stay in one
filter and get a time series.  With the open filter, they expect to
be able to detect a B=24 point source in 1000 seconds, a mode they
should not expect to use very often!  However, this means they can
detect a 14th magnitude afterglow through a filter in 10 seconds or
so, not great but with excellent time resolution.  A 30cm telescope
above the earth's atmosphere is a pretty capable instrument!
   So how can amateurs best contribute?  First, the burst alerts will
be rapid, and an afterglow should be obvious within a few minutes.
This means even small field of view systems will be effective, and
with minute-scale alerts, afterglows should be pretty bright (say, 15th
magnitude).  Fast exposures are the way to go, at least for the first
half hour or so.  Second, Swift is red-blind, so we should use red
filters.  This has several advantages: less scattered light from the
moon and city lights so that you can go fainter; Rc hits the peak QE
of most amateur CCD systems; the really exotic bursts will be the
high redshift ones, where the UBV emission will be absorbed and you
will only see the afterglow in the red passbands anyway.
   Now there are several possible red filters that should be
considered.  First, the Cousins Rc is an obvious choice since most
photometrists already have one in their filter wheel.  A second filter
choice would be Cousins Ic, even further redward than Rc.  However,
the typical Ic filter using the Bessell glass prescription has a blue
turnon that emulates the Cousins Ic bandpass, but then uses the CCD
red sensitivity falloff to form the red edge of the bandpass.  A CCD
detects light out to 1100nm, and so there is a red tail to the normal
glass Ic. You see this extra response when looking at very red objects
like Mira variables.  Some filter manufacturers will sell you an
interference Ic, which has the proper red cutoff to emulate the
Cousins I bandpass.
   While either Ic bandpass is ok, I'd suggest an alternative:
the Sloan Digital Sky Survey z' filter.  This is a simple glass
filter using RG830 glass that turns on at 830nm and then uses the
CCD response to form its red edge.  While the concept is similar
to the Bessell Ic, it is redwards of Ic.  A combination of Rc and z'
would give two red bandpasses, a very useful complement to the UVOT.
Schuler sells the z' filter as his Iz filter.  Your CCD does not have
high QE in the z' bandpass, so I'd reserve use of this filter to either
early-time observing or when you have a larger aperture telescope.
   So my preferred filters would be: Rc first, z' second, an interference
Ic third, and a Bessell Ic fourth.  I would recommend most small aperture
(say, less than 40cm) telescopes observe with Rc, and larger telescopes
observe with one of the redder filters.  Note that I am highly
recommending the use of some filter.  The reason is that the UVOT will
discover almost all of the afterglows before you would have a chance,
and the best use of amateurs is to obtain calibrated photometry during
the first few hours (after which the big guns like Keck and VLT will
be observing anyway).
   Observing cadence will depend on your telescope system and how soon
after the alert that you are able to get onto the field.  As you can see
from the recent HETE alerts, most bursts will be faint ones, and your
best chance will be observing within the first hour or two.  I'd initially
suggest following the ROTSE-III mode:
      10 5-second exposures
      10 20-second exposures
       n 60-second exposures until 40mins after the burst
       n 5-minute exposures afterwards

Adjust this pattern depending on the brightness of the afterglow.
   I'll be talking with Aaron at the Fall AAVSO membership meeting
next week and we will soon present a scheme for reporting this photometry.
In the meantime, why not look into purchasing
a z' filter and start getting your teams together to prepare for
the flood of alerts that will start in a couple of months?
Arne