Since 2015, W. M. Keck Observatory has been considering the possibility of conducting nighttime operations without any staff on the summit of Maunakea. A combination of methods has been used to assess the risk of this change in operations from different perspectives. System experts were surveyed to determine potential gaps in functionality that could create risk when operating or troubleshooting systems remotely. A hazard and risk analysis of use cases that describe nightly operations was conducted to identify risks to people, observatory equipment, and science quality and quantity that arise from the absence of people on the summit during the night. Risks were also identified by mining the night time fault reporting data from 2010-2016 to determine instances where hands on presence has been required on the summit to address issues. In the current state, these known issues would result in lost time and potential risk to equipment. The risk responses developed to address these risks have identified requirements on existing systems and for new capabilities to support unattended nighttime operations at WMKO.
Remote operation of observatories has been a topic of interest for many years. This paper discusses a general approach to determining what it will take to transition from on-site summit nighttime operation to remote nighttime operation of a facility. It is informed by involvement in projects at Canada-France-Hawaii Telescope, Gemini Observatory, and W. M. Keck Observatory. While these projects had differences, they all shared the goals of upgrading an operating observatory that is on sky every night to improve efficiency of operations without negative impact on science. The approach combines project management (PMI) and systems engineering (INCOSE) methodologies and tools to develop an understanding of the impact on operations, determine scope and requirements for new capabilities as well as additional functionality for existing systems, identify and manage risks, and how to incrementally move toward remote operation by integrating changes into current operations along the way.
In 2015, W. M. Keck Observatory conducted a study of the feasibility of conducting nighttime operations on Maunakea without any staff on the mountain. The study was motivated by the possibility of long term operational costs savings as well as other expected benefits. The goals of the study were to understand the technical feasibility and risk as well as to provide labor and cost estimates for implementation. The results of the study would be used to inform a decision about whether or not to fund and initiate a formal project aimed at the development of this new unattended nighttime operating capability. In this paper we will describe the study process as well as a brief summary of the results including the identified viable design alternative, the risk analysis, and the scope of work. We will also share the decisions made as a result of the study and current status of related follow-on activity.
As part of the image quality (IQ) assessment and improvement initiative being carried out at the 3.6m Canada
France Hawaii Telescope (CFHT) on Mauna Kea, Hawaii, our objective in the work reported here is to obtain
a systematic assay of thermal sources within the dome and in the summit environment around the observatory,
and therefore mitigate their contributions to convective instability leading to 'dome seeing'. Toward this, we
undertook a nighttime overflight to capture thermal images with a calibrated infrared camera of the outer
structures of CFHT and the neighboring observatories on the summit ridge, as well as of a significant area
of the surrounding terrain. The same thermal camera was then used to image heat sources within the dome.
Using a convective heat transfer model, all these measured surface temperatures were converted to heat fluxes,
and thus used to build a thermal assay of the dome. In addition, using button type temperature loggers, we
simultaneously recorded the nighttime dome skin temperatures of CFHT and two other observatories over a
weeklong period to evaluate nighttime supercooling of the dome skin due to radiation to the cold night sky. As a
complementary goal we compared the efficacy of different paints and coatings used in observatories to minimize
this effect. Though similar studies have been carried out at other observatories, the results are rarely available
in published literature. Therefore, here we explain our methodologies, along with a detailed discussion of our
results and inferences to serve as a useful resource to the larger observing community.
In 2007, the Canada-France-Hawaii Telescope (CFHT) undertook a project to enable the remote control of the
observatory at the summit of Mauna Kea from a control room in the Headquarters building in Waimea. Instead of
having two people operating the telescope and performing the observations from the summit, this project will allow one
operator to remotely control the observatory and perform observations for the night. It is not possible to have one person
operate from the summit, as our Two Person Rule requires at least two people for work at the summit for safety reasons.
This paper will describe how systems engineering concepts have shaped the design of the project structure and
With the advent of Queue observing at the Canada-France-Hawaii Telescope (CFHT), much emphasis has been placed
on minimizing the overheads in the observing process. Ensuring telescope focus is a necessary overhead, but taking the
focus sequences required to keep focus during the night adds significant time. In order to nearly eliminate this overhead
without sacrificing good telescope focus, the necessary focus position for each instrument has been modeled as a
function of telescope temperature and position on the sky. The correct focus position is calculated instead of measured,
so focus updates are practically instantaneous. The model coefficients are updated with new data regularly. Automatic
focus using calculated focus positions has been implemented for MegaCam, WIRCam and ESPaDOnS.
We present the first results of the SuperNova Direct Illumination Calibration Experiment (SNDICE), installed
in January 2008 at the Canada France Hawaii Telescope. SNDICE is designed for the absolute calibration of
the instrumental response of a telescope in general, and for the control of systematic errors in the SuperNova
Legacy Survey (SNLS) on Megacam in particular. Since photometric calibration will a critical ingredient for
the cosmological results of future experiments involving instruments with large focal planes (like SNAP, LSST
and DUNE), SNDICE functions also as a real-size demonstrator for such a system of instrumental calibration.
SNDICE includes a calibrated source of 24 LEDs, chosen for their stability, spectral coverage, and their power,
sufficient for a flux of at least 100 electron/s/pixel on the camera. It includes also Cooled Large Area Photodiode
modules (CLAPs), which give a redundant measurement of the flux near the camera focal plane. Before installing
SNDICE on CFHT, we completed a full calibration of both subsystems, including a spectral relative calibration
and a 3D mapping of the beam emitted by each LED. At CFHT, SNDICE can be operated both to obtain a
complete one-shot absolute calibration of telescope transmission in all wavelengths for all filters with several
incident angles, and to monitor variations on different time scales.