which selected Band-1 programs are ex-
tended for two full semesters after the con-
clusion of their first, with a new rollover sys-
tem, in which we allow all Band-1 programs
to begin execution before their designated
semester and continue executing through-
out the entire semester after that.
This proposal was discussed with the STAC,
the data that led to it were described, and
the STAC approved the proposal; their rec-
ommendations can be seen on the Gem-
ini website. It appears that the amount of
time required to support this will be small
enough that, at least initially, we will not
topslice anything from the TAC process in a
given semester. We will reassess after a year
of operating this way. As with the formal
rollover before it, this policy won ’t apply to
Target of Opportunity programs, for which
completion rate is not in our control, or for
Large and Long Programs, limited-term
partner programs, or programs using visit-
ing instruments.
Figure 1.
A 19-arcsecond-wide
field in globular cluster
M15, imaged in half a
minute with ‘Alopeke
at 832 nm. The stacked
raw frame (left) has
seeing of approximately
1 arcsecond and
significant elongation
due to windshake.
Point sources in the
reconstructed image
(right) have FWHM
approximately 0.15
arcsecond. These
commissioning data
cover the central quarter
of the `Alopeke field, but
the technique should
also be extensible to the
full field.
22
‘Alopeke Update
‘Alopeke (Hawaiian for “Fox”) arrived at
Gemini North in October. It is a more so-
phisticated variant of DSSI, the speckle
camera which has been visiting Gemini
since 2012. This new instrument occupies
essentially the only spot on the telescope
where it is possible to get light to it without
disturbing other instrumentation — that
is, in the small gap between the calibration
GeminiFocus
unit (GCAL) and the Instrument Support
Structure (ISS). There’s not much room in
there, but ‘Alopeke is small enough to fit.
Therefore, although it’s a visiting instru-
ment maintained and operated by a non-
Gemini team, it is able to remain on the
telescope at all times and thus offers much
greater scheduling flexibility.
‘Alopeke has the usual speckle capabilities
— two-color simultaneous speckle imaging
over a 5 arcsecond field, significantly larg-
er than was possible with DSSI, allowing
diffraction-limited imaging in the visible
— but now with a wide-field mode cover-
ing 60 arcseconds with rapid (26 Hertz full-
frame) readout. This, of course, enables fast,
two-color photometry over the larger field-
of-view and should be excellent for occulta-
tion or high-speed photometry work.
Interestingly, the early commissioning data
show that the wider field may also be ame-
nable to image reconstruction. The figure
here shows a field in the globular cluster
M15, taken in poor conditions (1 arcsecond
seeing and very windy). Individual expo-
sures were just 60 milliseconds with two
sets of 500 images in each filter. Integrating
all of the readouts produces, as expected,
a blurry image consistent with the seeing,
and with significant elongation due to wind-
shake. From that rather uninspiring input,
the team’s image reconstruction produces
a remarkably sharp image, with 0.15 arcsec-
ond point spread function. Strictly speak-
January 2018