The LSST calibration hardware system design and

development

Patrick Ingraham

, Christopher W. Stubbs

, Chuck Claver

, Robert Lupton

, Constanza

Araujo

, Ming Liang

, John Andrew

, Je? Barr

, Kairn Brannon

, Michael Coughlin

, Merlin

Fisher-Levine

c,d

, William Gressler

, Jacques Sebag

, Sandrine Thomas

, Oliver Weicha

, and

Peter Yoachim

Large Synoptic Survey Telescope, 950 N Cherry Ave, Tucson, AZ 85719, USA

Department of Physics, 17 Oxford Street, Harvard University, Cambridge, MA 02128, USA

Department of Astrophysical Sciences, Princeton University, Princeton, NJ 08544, USA

Brookhaven National Laboratory, Upton, NY 11973, USA

Department of Astronomy, University of Washington, 3910 15th Ave NE, Seattle, WA 98195

ABSTRACT

The Large Synoptic Survey Telescope (LSST) is currently under construction and upon completion will per-

form precision photometry over the visible sky at a 3-day cadence. To meet the stringent relative photometry

goals, LSST will employ multiple calibration systems to measure and compensate for systematic errors. This

paper describes the design and development of these systems including: a dedicated calibration telescope and

spectrograph to measure the atmospheric transmission function, a collimated beam projector to characterize the

spatial dependence of the LSST transmission function and a ?at-?eld screen illumination system to measure the

high-frequency variations in the global system response function.

Keywords: Calibration, LSST, photometry, operations, spectrograph, atmospheric transmission

1. INTRODUCTION

The Large Synoptic Survey Telescope (LSST) Project

will perform precision photometry over the visible sky

at a 3-day cadence using an 8.4 m diameter telescope that forms an image of the sky on a 3.2 Gigapixel focal

plane array.

The telescope, camera and infrastructure are currently under construction

and are scheduled to

begin commissioning in 2019 with the 10-year survey starting in 2022. One of main deliverables for LSST data is

precision photometry of both resolved and un-resolved objects (e.g. galaxies and stars). The relative photometric

design requirements are speci?ed to be 5 mmag (0.5%) repeatability in the bvri ?lters and 7.5 mmag in the uzy

?lters, for bright unresolved point sources under a wide range of observing conditions. The scienti?c bene?ts of

such high quality measurements impact several science cases including: photometric redshift determination of

galaxies, photometric metallicity determination of stars, and high-?delity determination of supernovae redshifts;

one of the fundamental probes in exploring the nature of Dark Energy and measuring the expansion rate of the

universe.

Meeting the photometric precision requirements is a signi?cant challenge and necessitates the calibration

and correction of multiple forms of systematic error. One example of systematic error that plagues photometry

measurements is the e?ect of atmospheric transmission since it is known to evolve both temporally and spatially

over the course of ˘ 2-3 LSST pointings. Similar to other surveys, calibration measurements and corrections must

also be determined for static e?ects such as vignetting and for system properties that may evolve over longer

timescales, such as optical throughput. This paper describes multiple hardware systems that LSST is developing

to measure and compensate for numerous sources of systematic errors, particularly errors impacting photometry

measurements.

Further author information: (Send correspondence to Patrick Ingraham): E-mail: pingraham@lsst.org

Characterization of the optical properties of LSST is accomplished using two independent systems that are

located inside the main telescope rotating enclosure (dome). The ?rst system, discussed in Section

is a custom-

made Collimated Beam Projector (CBP) that projects a ?eld of sources onto user-de?ned discrete sections of

the telescope optics. This device enables the ability to characterize the low to mid frequency spatial dependence

of the telescope and instrument transmission function, monitor ?lter throughput evolution and assist in the

characterization of ghosting e?ects. The second system is a calibration (?at-?eld) screen that will be illuminated

by both a white-light and a tunable monochromatic illumination system. The calibration screen system will

produce data to measure the high-frequency variations in the global transmission function.

To compensate for the e?ects of atmospheric transmission and its temporal and spatial variability, LSST will

utilize a robotic 1.2-meter diameter auxiliary telescope dedicated to measuring absorption features caused by

earth's atmosphere that are imprinted in the observed spectra of bright stars. Section

of this paper describes

the Auxiliary Telescope, its enclosure, and the spectrograph that is being speci?cally optimized to characterize

the atmospheric transmission properties at high cadence in coordination with the main telescope.

2. MAIN TELESCOPE IN-DOME CALIBRATION HARDWARE

This section describes the equipment located inside the main telescope building that is used as part of routine

calibration activities. The characterization of the telescope transmission function is performed using a combina-

tion of measurements using the calibration screen and the collimated beam projector. Both of these systems are

mounted to the dome, as shown in Figure

. The calibration systems have undergone signi?cant redesign from

the previous publication

to minimize technical risk and to increase the operational e?ciency. The inclusion of

the CBP to the In-Dome Calibration hardware complement enabled the relaxation of the illumination unifor-

mity requirement on the calibration screen therefore lower-risk designs could be accommodated. The following

subsections describe the components of the In-Dome Calibration systems and the hardware that facilitates their

operation.

Figure 1. The location of the calibration screen (shown in green) and collimated beam projector (indicated by the red

circle) in the LSST dome.

2.1 Collimated Beam Projector (CBP)

Determination of the optical transmission function and how it evolves with time is traditionally performed using

dome or sky ?ats (uniform illumination of the entire ?eld of view) and/or star ?ats, where a ?eld of stars is

rastered around the focal plane and the change in their properties is examined. Performing observations to

create star ?ats is a time consuming endeavour and should be done in photometric conditions; arguably the most

valuable time to perform science observations. Furthermore, dome ?ats, sky ?ats, and star ?ats all measure

the integrated transmission function of the optical system. With the collimated beam projector, the equivalent

of star ?ats can be reproduced from inside the telescope dome for fractional areas of the telescope pupil at

user-de?ned positions. Moreover, the measurements can be performed with and without a ?lter in the beam to

separate ?lter transmission properties from the other optical elements.

Figure 2. The Collimated Beam Projector is a small ˘ 30 cm telescope used as a projector (modeled as a paraxial lens in

inset A), to propagate the image of a series of simulated stars (pinholes) through the telescope, camera and onto the LSST

focal plane (camera shown as a side view in inset B). By articulating the projector and telescope the entire transmission

function of the telescope can be measured. Field angles shown are ? 1

The CBP is located in the telescope dome opposite the calibration screen between the top two rows of vent

gates, as indicated by the red circle in Figure

. The optical telescope assembly used for the CBP will be a

wide-?eld ˘ 30 cm diameter telescope on an actionable mount. Located at the CBP focal plane will be a mask

that is illuminated via a wavelength tunable monochromatic source (further discussed in section

2.3

). The mask

will be held in a mask wheel that will allow observers to switch between multiple mask designs. The nominal

mask will consist of single pinhole for each CCD, including the guiders and wavefront sensing devices. The CBP

will be used to measure the low to mid- spatial frequency variations of the transmission function. In traditional

dome and sky ?ats these measurements are often highly contaminated from ghosting e?ects that can manifest

as systematic error in the photometric measurements. Because the CBP only illuminates a small portion of the

pupil at a time, ghosting contamination is avoided. Multiple prototypes of the CBP have been tested and used

to help de?ne the LSST CBP design. Delivery of the device is expected in 2017. Readers are encouraged to see

Coughlin et al

from these proceedings for details on CBP design evolution and operation.

Figure 3. The calibration screen will be illuminated from a single central optic. The use of a single optic ensures only

low-frequency illumination non-uniformity whose e?ects are mitigated through Collimated Beam Projector measurements.

Mitigation of scattered light is performed via ba?es near the central optic and from blackening the surfaces not visible

to the LSST focal plane.

2.2 Calibration Screen

The calibration screen is used for obtaining ?at-?eld calibration frames in both monochromatic and polychromatic

light. The re?ective portion of the screen is an annulus with an inner and external diameters of 4.2 and 9.3

m, respectively. A blackened area surrounding the re?ective portion is present to minimize scattered light from

angles exceeding the 3.5

LSST ?eld-of-view. During operations, the screen rests in tilted the position shown

in Figure

. The screen is of normal incidence to the telescope boresight at an elevation angle of 22

. To

facilitate maintenance of the dome vent gates and to allow servicing of equipment located on the calibration

screen structure, the dome screen may be rotated into a vertical position. The re?ective material used on

the calibration screen has yet to be selected. Due to the large wavelength range of the LSST survey and the

requirement for monochromatic ?ats taken at 1 nm increments, a smooth re?ectivity pro?le is required. Surfaces

exhibiting di?use re?ectance under consideration include Spectralon, however, another option may be purchasing

a commercial prefabricated screen, such as the Draper M1300.

The illumination of the calibration screen will utilize a single optical element that protrudes from the cali-

bration screen at the center of the annulus. An example of a design under consideration is shown in Figure

This is a signi?cant design evolution from the previously presented design where the illumination sources were

located on the top-end of the telescope. In using a single re?ection element, the illumination pattern on the

screen is limited to only low-frequency illumination non-uniformity. The only higher-frequency non-uniformity's

would originate from the support structure for the re?ector, or the re?ecting screen itself. The conceptual design

in Figure

has a diverging beam originating from the light source that is re?ected and dispersed by an aspheric

optical element. Because the optical quality of this component is not of critical importance due to the calibration

screen utilizing a di?use re?ective material, the piece could be machined from aluminum then polished to reduce

?guring artifacts. This results in the central optic being easy to fabricate, light-weight, straightforward to mount,

and cost-e?ective. The calibration screen will be delivered to the site and integrated in late 2018.

Because the light source is originating from the backside of the calibration screen and our operational re-

quirements dictate that broadband dome ?at ?elds are taken daily while the dome is in the park position, the

white light source(s) must be mounted directly to the calibration screen. Due to environmental considerations

and safety concerns, the monochromatic source will not be located in the dome or on the observatory ?oor. Light

delivery systems are further discussed in section

2.3

2.3 Light Sources and Delivery Systems

The calibration screen will be illuminated by both broadband light sources and a tunable monochromatic source.

The monochromatic source is expected to be a tunable laser that covers the 320-1125 nm bandpass range in 1

nm increments. The laser will be located in an enclosed section of the camera utility room on the base enclosure

level (level 5) of the facility just outside the lower enclosure. Transporting this light from the source to the

calibration screen and CBP is a non-trivial problem.

Original plans to transport the light to the previous multi-projector system located on the top-end of the

telescope mount utilized a broadband ?ber optic nearly 80 m in length. The absorption of blue light over this

length of ?ber ( ˘ 90%) made this design challenging operationally due to the amount of time required to perform

the measurements. Various ?ber optic con?gurations were considered for the central illuminator design discussed

in Section

2.2

, but all su?ered from absorption in the blue and the need for human intervention to connect the

?ber to the dome when needed. For these reasons, it was decided to free-space propagate the laser in enclosed

tubes from the laser room to the calibration screen and the collimated beam projector. This both removes the

?ber absorption issue and ensures that calibrations can be performed without human intervention.

The monochromatic light will be propagated from the source, through a shutter and beam expander system,

re?ected vertically into the ceiling then propagated horizontally through a hole in the concrete wall of the lower

enclosure. From the lower enclosure, a powered steering mirror will then direct the light vertically through a

hole in the observing ?oor and up to the lower-enclosure and dome interface. From this point, the light must be

directed to calibration screen and the CBP (albeit not simultaneously). Because the dome azimuth repeatability

requirement subjects the laser tube placement to a 2.5 cm displacement error, a beam steering system is required

where one steering mirror is located in the lower-enclosure (as mentioned previously) and the other is on the

calibration system mounted in the dome. The beam position will be determined using two cameras, one looking

at the focus to measure pointing, the other imaging a conjugate pupil to measure beam centering.

Because the CBP observations must be performed during the day while the dome is in the park position,

and the CBP is not located directly above the laser tube originating from the lower enclosure, an optical bench

will be placed high in the dome to steer the beam into a CeramOptec PowerLightGuide fused end ?ber bundle

to transport the light to the CBP. Transmission losses in the ?ber are not of signi?cant concern since the light

required for the CBP is only a small fraction of the power required for the calibration screen.

In order to use the monochromatic source with the calibration screen, the dome must be rotated away from the

park position. In this con?guration the dome cooling is reduced. Maintaining cooling during the day is critical for

minimizing dome seeing e?ects. This is one of the reasons why the monochromatic ?at ?eld observations will be

performed during cloudy nights. Other reasons include: minimal scattered light, multi-hour windows to perform

the calibration and optimization of daytime telescope access. Monochromatic ?ats need only be measured 3-4

times per year, whereas broadband ?ats must be taken daily to track dust movement on the optical elements.

Because broadband ?ats will be taken daily, the dome must remain in the park position to ensure e?ective

cooling. For this reason, the broadband source(s) will be mounted on the calibration screen itself and the light

will be directed to the central illumination optic. The broadband sources are also less subject to environmental

constraints and the safety precautions to personnel are signi?cantly reduced. Light sources currently under

consideration include LEDs similar to what is used for DECam calibration

and broadband sources such as the

Horiba KiloArc and the Energetique EQ-1500.

2.3.1 Illumination Characterization Systems

Characterization of the calibration screen illumination is pertinent to ensuring no systematic error is introduced

into the photometric corrections. The monochromatic dome ?ats will be used to synthesize a ?at-?eld image

matching a spectrum of the night sky for use in accurate background subtraction. The broadband ?ats will be

used to monitor changes to the dust patterns on the optical components (particularly the ?lters). By examining

the evolution in the daily broadband ?ats, corrections can be made to the monochromatic ?ats so long as the

spectral energy distribution of the broadband ?at is known. For this reasons, a ?ber-fed spectrograph will

be used to measure the spectral energy distribution of the light re?ected from the calibration screen. The

spectrograph can also be used to measure the line width of the monochromatic source. For this purpose, two

AvaSpec-ULS2048x64 TEC spectrographs, one for red wavelengths and the other for blue wavelengths, from

Avantes have been selected to measure the spectral energy distribution to up to a resolution of 0.7 nm. Although

the spectrographs will have an illumination calibration, we will perform monitoring of the variation of ?ux levels

using photodiodes.

The National Institute Standards and Technology (NIST) has calibrated the quantum e?ciency of Hama-

matsu S2281 photodiodes to accuracies that surpass photometric standard stars by an order of magnitude.

Several studies on the use of these photodiodes for astronomical calibration discuss their advantages in detail

{

and their usage amongst the community is increasing.

The LSST calibration plan includes several of these

photodiodes throughout the calibration procedure as a basis for comparison of calibration frames. Having a

standard at this level of precision enables accurate monitoring of the transmission response and absolute trans-

mission of ?lters. It also provides a mechanism to ensure each monochromatic ?at ?eld frame has the desired

signal. This is facilitated by using a shutter located at the output of the laser, rather than relying upon the

camera shutter. The current of the photodiodes will be measured using Keithley 6517b electrometers that will

be located in cooled electronics cabinets located on the secondary mirror support assembly of the main telescope.

3. AUXILIARY TELESCOPE AND SPECTROGRAPH

Characterization of the absorption properties of the atmosphere during LSST observations will be performed by

1.2 m diameter Auxiliary Telescope located ˘ 300 m north-east of the main telescope. At the time of writing,

the excavation for the building foundation and pier have performed but neither have been poured. The telescope

will housed in a 30 foot (9.1 m) diameter circular two-storey building. The lower ?oor will contain the control

electronics and observatory support equipment

as well as four remotely operable vent gates. The Auxiliary

Telescope does not have a high-image quality requirement hence the building does not require an active air

conditioning system. However, e?orts are being made to promote e?cient passive cooling. The observing ?oor

(2nd ?oor) is made of a grating to promote air ?ow entering through the dome shutter then passing through

the building and out of vent gates on the lower ?oor. The telescope mount, mirror cell and pier has fan-

driven circulation units to assist in temperature stabilization and uniformity. The Auxiliary Telescope dome

is currently under construction by Ash Manufacturing Company and is scheduled for delivery early summer,

2017. The enclosure will be equipped with the SmartDome controller developed by Astronomical Consultants &

Equipment Inc. The dome rotation speed will also be increased by using four motors rather than the nominal

two motor system.

The Auxiliary Telescope, previously known as the Calypso Telescope (shown in Figure

), was located on

Kitt Peak and has been brought to the National Optical Astronomical Observatory (NOAO) facility in Tucson

to undergo a signi?cant refurbishment before being transported and re-commissioned in Chile. Astronomical

Consultants & Equipment Inc. was awarded the contract. The refurbishment work includes replacements of all

drive motors, controllers and electronics. Maintaining compatibility of components with the main telescope is a

common theme throughout hardware selection for all components. Wherever possible, Kollmorgen motors and

Copley Motor controllers are being utilized and interfacing will utilize National Instrument (NI) Compact RIO

devices. The mirror cell will be re?tted with new bellows systems. The secondary mirror support system will be

re-worked to include a new commercial o?-the-shelf 6-Axis Hexapod. The most signi?cant change to the original

telescope design is the the installation of Heidenhain Tape encoders to the azimuth axis, and Heidenhain Ring

Encoders to the elevation axis and instrument rotators. During operation at Kitt Peak, the telescope su?ered

from pointing problems. This was due to multiple reasons such as temperature induced expansion and contraction

of azimuth drive surface that was not accounted by the rotary encoder and/or control software. Furthermore,

the instrument rotators used a combination of friction drives and rotary encoders that were subject to slipping.

Preserving that system would make satisfying the LSST pointing requirements challenging. In order to install

the azimuth tape encoder, a new surface to support the tape is being machined and installed above the drive

Figure 4. The Auxiliary Telescope while located on Kitt Peak. The telescope is now undergoing refurbishment and will

be ready for observations in June 2018.

surface and below the telescope fork. Fabrication of this new encoder disk is now underway. Encoder rings are

being installed on the elevation axis, and the rotary encoders are being replaced with ring or tape encoders for

the instrument rotators.

The refurbishment being done in Tucson is expected to be completed in April, 2017. The telescope will

then remain in Tucson for a period of ˘ 4 months to be used as a testbed for the Telescope and Site software

team to perform software testing and demonstration of the Observatory Control Software,

Telescope Control

System,

and Communications Middleware.

During this time, the secondary and tertiary mirrors will have

their coatings removed and will then be hard-coated with high-re?ectivity metallic coatings. Upon the installation

of the dome and the completion of the Auxiliary Telescope building, the entire telescope will be shipped to Chile

for integration, test, and ?nal acceptance. After veri?cation of the telescope performance, the telescope will be

ready for use in summer 2018, when the spectrograph will be commissioned.

The spectrograph being designed to perform characterization of the atmospheric absorption pro?le has under-

gone signi?cant evolution from what was previously presented.

The conceptual observing plan for the Auxiliary

Telescope had it operating independent of the LSST position, slewing about the sky to a ?xed table of targets

spanning large ranges of airmass. The observing plan has been expanded to support multiple observing strate-

gies for characterizing the atmospheric transmission as a function of time and position that can be optimized for

observing conditions and/or LSST ?lter. The most demanding strategy is where the Auxiliary Telescope follows

the pointing of the LSST telescope as closely as possible. This puts increased time pressure to measure the sky

spectrum since the LSST telescope changes pointing every ˘ 40 seconds and has a fast slew speed.

To assist in

the determination and observing scheduling, a greatly simpli?ed version of the LSST Scheduler

is envisioned.

This reduced Scheduler will utilize the same telemetry used to determine the LSST pointing and will use the

predicted future pointings of LSST in determining the pointings for the Auxiliary Telescope.

To increase the observing cadence of the Auxiliary Telescope, the spectrograph now utilizes a slitless design.

This has multiple bene?ts including: removing the acquisition sequence required to position the target in the

slit, relaxing the pointing requirement and the removal of di?erential slit losses that are particularly problematic

at higher airmasses. Calculations of water vapour absorption at red wavelengths (800-980 nm) as a function

of spectral resolution demonstrates that the resolution requirement on the spectrograph may be reduced to

R=150 for wavelengths longer than 800nm. This also results in a lesser exposure time to obtain the required

signal-to-noise ratio.

1600

1800

2000

2200

2400

2600

2800

3000

3200

Detector Pixel

500

1000

1500

2000

Intensity [ photons / pixel ]

4000

6000

8000

10000

wavelength [ ]

A0V star, m

=12

20s exposure

Seeing = 1''

Figure 5. Left: The simulated spectrum as measured on the detector when using the Ronchi Grating. Right: The

priliminary optical design starting at the instrument ?ange for a Rochi Grating (R ˘ 100) and a Grism (R ˘ 300). A

variety of dispersers will be held in a disperser wheel assembly that will enable selection of the appropriate instrument

setup depending on observing conditions and strategy.

The current spectrograph design is meant to have high throughput, minimal internal re?ection, and support

multiple observing setups to accommodate ?exible observing strategies. Throughput measurements indicate

that adequate signal to noise can be achieved in 20 seconds on 12th magnitude star but the high majority of

observations will use stars in a magnitude range of 5 < m

< 8. The instrument will also have an imaging

mode with a small ( ˘ 1x1 arcminute) ?eld of view which may be used for alignment and telescope collimation

purposes. Because of the Auxiliary Telescope's slow (f/18) beam, it is possible to put dispersive elements into

the converging beam with minimal increase in wavefront aberration. This type of design minimizes internal

re?ections and maximizes throughput. A successful demonstration of such a system was performed by inserting

a Ronchi grating into the ?lter wheel of the 0.9m SMARTS Telescope at Cerro Tololo

The ?lter and disperser combination will be implemented using a dual-wheel design. The ?lter selection will

include the LSST ?lter bandpasses plus long and short pass broadband ?lters that may be used to measure

and/or remove e?ects from disperser selection such as order contamination. The nominal observing mode will

utilize either an Amici Prism or a Ronchi grating to provide simultaneous wavelength coverage from 350-1050 nm.

Each system is optimized to have maximum spectral resolution in the 900-980 nm water feature. Using higher

dispersion will also be an option via a Grism, such as the one shown in Figure

. After a 2-month commissioning

period, the Auxiliary Telescope will begin characterizing the night sky in e?orts to measure the time evolution

and variation in spatial structure over Cerro Pach?on. These observations combined with the anticipated nightly

LSST observing schedule will be used to optimize the Auxiliary Telescope Observing strategy to provide the

highest ?delity photometric correction for LSST data.

3.1 Auxiliary Telescope Calibration

The Auxiliary Telescope will have its own array of calibration instruments albeit with reduced functionality from

the calibration equipment for the main telescope. A small calibration screen will be used to obtain dome ?ats.

The calibration screen will be front-illuminated using a Horiba Tunable Kiloarc that enables narrowband ?at

?elds over the entire 350-1050 nm wavelength range. The monochrometer will also contain a ?at mirror that will

allow broadband illumination. A single NIST-calibration photodiode system will monitor any changes in light

source intensity. A AvaSpec-ULS2048x64 TEC spectrograph from Avantes will be used to monitor the spectral

energy distribution of the source at a spectral resolution of ˘ 2 nm.

Mondrik et al in prep

4. CONCLUSION

The calibration hardware for LSST is being ?nalized and procurements are underway. The Auxiliary Telescope

refurbishment has commenced and will result in the telescope being ready for observations in summer, 2018.

The spectrograph will then be commissioned to ensure full operational support for commissioning activities and

observations. The introduction of the Collimated Beam Projector to the hardware resulted in the relaxation of

the calibration screen requirements and has enabled the construction of a lower technical risk, higher throughput

design. A targeted design optimization study is now ongoing with delivery expected in late 2018. All systems

are fully expected to remain on schedule, within budget and meet or exceed their operational requirements.

5. ACKNOWLEDGMENTS

This material is based upon work supported in part by the National Science Foundation through Cooperative

Agreement Award No. AST-1227061 under Governing Cooperative Agreement 1258333 managed by the Asso-

ciation of Universities for Research in Astronomy (AURA), and the Department of Energy under Contract No.

DEAC02-76SF00515 with the SLAC National Accelerator Laboratory. Additional LSST funding comes from

private donations, grants to universities, and in-kind support from LSSTC Institutional Members.

REFERENCES

[1] Kahn, S., \LSST ?nal design overview," Proc. SPIE 9906 (2016).

[2] Kurita, N., \Large synoptic survey telescope camera design and construction," Proc. SPIE 9912 (2016).

[3] Gressler, W., \LSST telescope and site status," Proc. SPIE 9906 (2016).

[4] Sebag, J., Gressler, W. J., Liang, M., Axelrod, T., Claver, C., and Andrew, J., \Calibration systems for

LSST," in [Ground-based and Airborne Telescopes V], Proc. SPIE 9145 , 91453L (July 2014).

[5] Coughlin, M., Abbott, T. M. C., Brannon, K., Claver, C., Doherty, P., Fisher-Levine, M., Ingraham, P.,

Lupton, R., Mondrik, N., and Stubbs, C., \A collimated beam projector for precise telescope calibration,"

Proc. SPIE 9910 (2016).

[6] Rheault, J.-P., DePoy, D. L., Marshall, J. L., Prochaska, T., Allen, R., Wise, J., Martin, E., and Williams,

P., \Spectrophotometric calibration system for DECam," Proc. SPIE 8446 , 84466M{84466M{9 (2012).

[7] Stubbs, C. W. and Tonry, J. L., \Toward 1% Photometry: End-to-End Calibration of Astronomical Tele-

scopes and Detectors," The Astrophysical Journal 646 , 1436{1444 (Aug. 2006).

[8] Stubbs, C. W., Slater, S. K., Brown, Y. J., Sherman, D., Smith, R. C., Tonry, J. L., Suntze?, N. B.,

Saha, A., Masiero, J., and Rodney, S., \Preliminary Results from Detector-Based Throughput Calibration

of the CTIO Mosaic Imager and Blanco Telescope Using a Tunable Laser," in [The Future of Photometric,

Spectrophotometric and Polarimetric Standardization ], Sterken, C., ed., Astronomical Society of the Paci?c

Conference Series 364 , 373 (Apr. 2007).

[9] Stubbs, C. W., Doherty, P., Cramer, C., Narayan, G., Brown, Y. J., Lykke, K. R., Woodward, J. T., and

Tonry, J. L., \Precise Throughput Determination of the PanSTARRS Telescope and the Gigapixel Imager

Using a Calibrated Silicon Photodiode and a Tunable Laser: Initial Results," The Astrophysical Journal

Supplement 191 , 376{388 (Dec. 2010).

[10] Regnault, N., Guyonnet, A., Schahman?eche, K., Le Guillou, L., Antilogus, P., Astier, P., Barrelet, E.,

Betoule, M., Bongard, S., Cuillandre, J.-C., Juramy, C., Pain, R., Rocci, P.-F., Tisserand, P., and Villa, F.,

\The DICE calibration project Design, characterization, and ?rst results," Astronomy & Astrophysics 581 ,

A45 (Sept. 2015).

[11] Barr, J., Gressler, W., Sebag, J., Seriche, J., and Serrano, E., \LSST summit facility construction progress

report, reacting to design re?nements and ?eld conditions," Proc. SPIE 9906 (2016).

[12] Daly, P., Schumacher, G., Delgado, F., and Mills, D., \LSST OCS status and plans," Proc. SPIE 9913

(2016).

[13] Daly, P. and Schumacher, G., \Is the work?ow model a suitable candidate for an observatory supervisory

control infrastructure," Proc. SPIE 9913 (2016).

[14] Lotz, P., Dubois-Felsmann, G. P., Lim, K.-T., Johnson, T., Chandrasekharan, S., Mills, D., Daly, P.,

Schumacher, G., Delgado, F., Pietrowicz, S., Selvy, B., Sebag, J., Marshall, S., Sundararaman, H., Contaxis,

C., Bovill, R., and Jenness, T., \LSST control software component design," Proc. SPIE 9913 (2016).

[15] Mills, D., Schumacher, G., and Lotz, P., \LSST communications middleware implementation," Proc. SPIE

9906 (2016).

[16] Gresslera, W., Thomas, S. J., Gessner, C., Warner, M., Barr, J., Lotz, P. J., Schumacher, G., Wiecha,

O., Angeli, G., Andrew, J., Claver, C., Schoening, B., Sebag, J., Krabbendam, V., Neill, D., Hileman, E.,

Muller, G., Araujo, C., Martinez, A. O., Aguado, M. P., Marchen, L. G., de Argando, I. R., Romero, F. M.,

Rodrguez, R., Gonzlez, J. C., and Venturinie, M., \Large synoptic survey telescope mount ?nal design,"

Proc. SPIE 9906 (2016).

[17] Delgado, F., \The LSST scheduler construction plan," Proc. SPIE 9906 (2016).