Contents:

• Why use spectral lines for pointing/focusing?
• How to do a spectral line pointing/focus
• What's possible, what's not
• How does it work?
• Extending the list of sources
• Acknowledgement

Why use spectral line fivepoints/focusing?

Because most of the many continuum sources we list are not suitable for local pointing and focusing with the heterodyne receivers. Almost all blazars are too weak, the planets are rarely in the right place if they are up, and most other bright pointing targets are concentrated towards the Galactic Plane. Whereas SCUBA receives the integrated power over a wide bandwidth of around 100 GHz, and can make use of off-source pixels to compensate for sky noise, the present heterodyne receivers have a relatively small continuum bandwidth of not more than 2 GHz, and being single-beam instruments, cannot remove sky noise.

There is a class of objects which fill this gap quite well and use the principal ability of the heterodyne receivers to detect line emission: evolved late-type stars. As a rather amorphous group such objects include Asymptotic Giant Branch stars through planetary nebula. Examples include IRC+10216, NGC7027, CRL618, RX Boo and LP And. The brighter objects of this class are nearby, relatively speaking, and therefore tend to be scattered over the sky. Many of them are variable stars, some known for centuries. They are often surrounded by a dense shell of gas and dust. The gas often gives rise to strong lines of CO and sometimes HCN and CS, while the dust means that these objects can be good continuum targets also.

The current set of objects we use for spectral line pointing appears at the end of the JCMT source catalog, but the list is not complete, since basically any small-diameter object which shows a reasonably bright line is suitable.

Spectral line fivepoints and foci confer an operational advantage in addition: the use of the technique removes the need to reconfigure the system from DAS to CBE and back.

Dual-channel pointing/focus measurements provide improved sensitivity for a given integration time. A second positive feature is that one obtains therefrom a mean pointing for the two beams, more appropriate for dual-channel observations. The dual-channel heterodyne receivers do have a slight "squint" within the limits of measurement.

How does spectral line pointing/focusing compare with that on continuum targets? Aside from fulfilling the obvious need for local pointing, the evidence is largely anecdotal at this stage. However, recent tests during very unstable daytime conditions show that in a 225-GHz tau of 0.3+ using receiver B3 it was straightforward to obtain both pointing and focus determinations in a single pass using RX Boo. Such observations are well beyond the "normal use" limits of B3, and pointing on any continuum target aside from the brightest planets would have been very difficult.

How do I do a spectral line fivepoint or focus?

A spectral line fivepoint consists of spectra taken at five positions (center, and offsets in the four cardinal directions), just as for a regular continuum fivepoint observation, except that the signal used in the fit to the data is the integrated intensity of the line in each case. A spectral line focus traditionally has 5 measurements also, but can contain up to 10 such (there is no requirement that the number of points be odd). A larger number is likely to produce a better result.

The steps involved in carrying out a spectral line fivepoint or focus are:

• Tune the receiver in single or dual channel mode to the frequency of the line (usually, but not always a CO line) and the LSR velocity of the object chosen for pointing. The latter is important for correct on-line reduction of the data, since the line needs to be centered in the spectrum.
• Configure the DAS for any of the normal bandwidths - 125 through 920 MHz. The bandwidth chosen depends to some extent on the frequency and the target, but for receivers A3 and B3 250 MHz is usually enough.
• Choose non-continuous calibration; otherwise the system will spend a disproportionate amount of time calibrating.
• Use beamswitching with a throw/rate of 60"/1 Hz, although larger switches are OK. Position-switching is not used.
• Calibrate.
• Issue the normal FIVE-point or FOCUS command, and choose an integration time; usually 10 seconds will be enough, although times from 4 through 30 seconds are typical.

In dual-channel mode the result will be the average of the separate pointing or focus determinations. The individual fits for the two channels appear on the TSS command terminal (which is echoed to the user stations).

That's it.

Current Capabilities

Although spectral line pointing was in place for some years in a rather limited way, the software was expanded considerably in September/October 2001 by Firmin Oliviera. At present:

• Both pointing and focusing in spectral line mode is operational.
• Pointing always has 5 measurements, whereas a focus observation can contain up to 10 measurements.
• Focusing can be carried out in any of the three axes (x,y, and z).
• Both single- and dual-channel observations are allowed.
• DAS bandwidths between 125 and 920 MHz may be used.

• Wideband mode; that is, single-channel 1800 MHz.

How does it work?

For each observation the software merges the DAS subbands together as necessary, and uses the given velocity range to fit a linear baseline outside the line emission region. The integrated intensity of the line at each pointing or focus offset is used as input to the fitting algorithm. In dual channel mode the results are kept separate and averaged subsequently to provide a final fit.

Extending the list of suitable sources

Any small-angular diameter source which has an integrated line intensity bright enough is fair game. The common targets in the current pointing list have all been tested at some point in the JCMT's history, and all of them are OK at the CO 2-1 and CO 3-2 lines. The brighter ones should be OK also at the higher transitions of CO.

One could also extend the list to special cases, not necessarily evolved stars. For instance MWC349 is a weak continuum target which has very weak CO lines, but enormously bright hydrogen recombination lines which can be used for spectral line pointing. Small-diameter galaxies with appreciable CO emission can be used; although the line may have a weak peak intensity, the integrated intensity can be quite sufficient for five-pointing.

In all cases not in the current JCMT catalog the source should be added to the catalog; both the velocity and the velocity range must be entered in the catalog. The latter is critical information; it is the velocity range to zero emission and is used to define the baseline fit region and the range for calculation of the integrated intensity. If the source needs to be entered "by hand", the velocity range can be included by giving the command "send control set vrange" at the ICL prompt.

Thanks!

Very many thanks to Firmin Oliviera for his passion for and interest in this project and all his efforts to make it come to pass. Thanks, too, to Jim Hoge for his enthusiastic and able assistance at the telescope during the tests.