This document covers designing a small number of masks (<6) to cover a field, where you
already have a good idea of your high priority objects. The danger here is that if you
only select your highest priority objects you won't get many objects per mask. Remember to check how many
secondary objects you're losing by only selecting the primary objects:
Another important consideration is the location of the mask centers, as these directly affect the number
of objects that can be selected:
|| In this case the gap between 2 primary objects is slightly too small to select
another object. Instead, you could pick all three secondary objects. The choice is yours.
||The mask centers here have been well chosen. We may have lost the right most primary
object, but we won't lose any of the others.
||The mask centers here have been poorly chosen, in an effort to keep the right most
primary object we have lost 2 other objects due to overlap in y.
The range of magnitudes of the objects in any given mask is also important, remember that an increase of 1
magnitude corresponds to a factor of 2.512 in flux. If you have one object on a mask that is a magnitude
fainter than all the others you'd have to observe that mask for twice as long for the sake of a single
object, this isn't a very good use of a MOS instrument. If you have overlapping masks you can always try to include
a faint object on two masks. Alternatively, if there are no other objects to select, you may as well leave
a faint object on the mask and hope that it has a bright emission line, just make sure the observers know
they don't have to achieve the same signal to noise for that object as they do for the others.
The best masks are produced when you are aware of how all of the variables (mask centers, magnitude ranges
and object selection) will affect the final design. Personally, I spend a couple of hours playing with the
variables in maskdesign before I actually try to design the masks.