UVOT Modes

The Swift UltraViolet/Optical Telescope (UVOT) has a different way of setting exposure times and filter combinations that I will attempt to explain below.



The Swift/UVOT has preset observing modes with various filter combinations and time distributions.  These modes allow it to dynamically adjust when the planned schedule (the Pre-Planned Science Timeline or PPST) is interrupted by a gamma ray burst (GRB) or other high priority target uploaded as an automatic target (AT) which will override the observation of any targets with a lower figure of merit value for the time that the AT is visible.  Most modes divide up an observing time between requested filters according to a specific time weighting scheme.  This weighting can be done for the same filter as well.  For example, many single filter observations in the optical allow for "earth limb protection" by taking three separate snapshots with a time weighting of 1:10:1.  The short observations at the beginning and the end can be discarded if observing close to the earth limb results in a higher background that makes the exposure unusable without wasting the entire observing window.  If the background isn't too high, those exposures can still be used separately or coadded with the other exposures.  Earth limb protection is also accomplished by putting the less affected UV filters at the beginning and end of filter sequences.  For grism observations, a "slew-in-place" using a UV filter to set up the observation protects the grism exposure at the front end while a UV filter observation is usually at the end of the grism exposure sequence to protect the back end of the observation. 

If a target is observed according to the PPST, it will divide up the planned time based on the filter order and weighting and observe accordingly.  If the observation is interrupted in the middle (by a GRB or AT), the remaining filters are missed for that snapshot/orbit (and possibly future orbits as the targets likely overlap in visibility).  If an active GRB or AT overlaps with the beginning of the planned observation, the remaining time will be divided up amongst the requested filters in the requested ratios.

If a target is uploaded as an AT, the actual exposure time is not always correctly anticipated (but can be predicted on the ground using the planned timeline).  The spacecraft weights the time until the next observing constraint but may terminate the observation before that when a higher figure of merit target in the plan comes up.  To help get more filters in those cases we have some unscaled modes that do the filter sequence for a given set of time.

In order to conserve filter wheel rotations (one of the few things likely to wear out on Swift), the default mode utilizes a "filter of the day" where one of the UV filters (including u) is used for most of the day for targets where XRT is the primary instrument.  Even if XRT is the primary instrument, for SN targets I recommend utilizing a multi-filter UVOT mode to get valuable UV/optical data on the SNe at the same.  If the utility of that data is in question, I am more than willing to look at the data and determine whether it is useful or not.  We have had many deep observations in a single filter that could have yielded multiple useful colors.   This post gives some of the recommendations and sample wording to use as a UVOT mode justification.

This page is intended to suggest observing modes for Swift's UVOT.  These modes are especially recommended for red objects (ie low UV flux compared to their optical flux), but because the UV filters have lower sensitivity and the optical data can be obtained from the ground, weighting the time heavily to the UV is rarely a bad thing.

Note, I will often say things like "all 6 UVOT filters."  The UVOT actually has 7 lenticular filters, but the white filter is very broad.  It is useful for very faint objects like some gamma ray burst (GRB) afterglows that you might not detect in narrower bands, but you don't get color information and the conversion to flux is very spectrum dependent.  So I ignore it for nearby SN observations.

Default recommended observing mode for Supernovae in the PPST
6 UVOT filters heavily weighted to the UV -- 0x223f  [time weighting w2:m2:w1:u:b:v::5:8:3:1:1:1]

Default recommended observing mode for Supernovae as an uploaded AT
6 UVOT filters heavily weighted to the UV -- 0x0270 (this will do all 6 filters in that ratio for a 1000 second snapshot.  If the snapshot is longer, any extra time is dumped into the uvm2 exposure.  Since the uvm2 filter is the least sensitive and has the sharpest sensitivity cutoff (ie no red leak), getting more time and a higher S/N in the uvm2 is almost always a good thing.

uvw1+u+b+v (ie a target is too faint for the mid-UV filters) -- 0x2238 [time weighting: w1:u:b:v::6:2:1:1]

There are also special modes for if the SN is too bright (m<13) or for other filter combinations, but it is probably better to work with me or an ODS to figure out what is appropriate.