Units

First, a brief word about units. Although it may not always be obvious, the underlying philosophy of Figaro is to provide tools which you may use as you wish, and not to force you to reduce your data in a way imposed by the author of the software. It would be in keeping with this for there to be no restrictions on the units that can be used for flux calibration. However, there are practical limitations.

Magnitude units, such as the AB79 system of Oke & Gunn,³ give results that have a comfortable feel for optical astronomers, and also fall into a numerically convenient range, generally being numbers between 0 and 25. Unfortunately, being logarithmic (not to mention going backwards) they are, while not actually difficult to handle, sufficiently different to linear units to make it impossible to write software that can deal with them other than as a special case.

Rather than do that, Figaro compromises. The flux calibration routines insist on the calibration being performed using linear units. When the flux calibrated spectrum has been produced in such units, it may then be converted to the AB79 scale. This means that rather than have a lot of software that operates on both linear and logarithmic data, we have one routine (`abconv') that will perform the conversion.

With that restriction, Figaro will allow you to use whatever linear units you prefer. As will become clearer soon, the units used are determined entirely by entries in the original tables used by the routine `gspike', and if you insist on `Joules per minute per square foot per Angstrom' you are quite at liberty to prepare your own table giving the flux density for your standard at the appropriate wavelengths in these units. You will even find that `splot' will label your axes properly when you plot the data!

As far as linear units go, there are objections to using `ergs per sec per square cm per Hz' and even more so to `ergs per sec per square cm per Angstrom' on the grounds that the numbers involved are ridiculously small (and in the latter case can easily go outside the floating point range, which is a serious problem!). So, mainly to be able to avoid having to type commands such as `splot high=2.5e-27 low=1.5e-27', the preferred units for Figaro files are Janskys, mJy, or micro-Janskys.

For most of the flux standards for which Figaro supplies tables, two tables are provided: one in AB magnitudes, usually a direct copy of the published data, and one in (milli- or micro-) Janskys, usually the result of a semi-automatic conversion from the former. If you particularly like ergs/s/cm**2/A, then you can either provide your own flux tables in these units and work from them directly, or you can use the `flconv' command to convert from a spectrum calibrated in Janskys into these units.