Evryscope science: exploring the potential of all-sky gigapixel-scale telescopes

1.8 MB·English

by  Nicholas M. Law

Draftversion January15,2015
PreprinttypesetusingLATEXstyleemulateapjv.5/2/11
EVRYSCOPE SCIENCE: EXPLORING THE POTENTIAL OF ALL-SKY GIGAPIXEL-SCALE TELESCOPES
Nicholas M. Law1, Octavi Fors1, Jeffrey Ratzloff1, Philip Wulfken1, Dustin Kavanaugh1, David J. Sitar2,
Zachary Pruett2, Mariah N. Birchart2, Brad N. Barlow3, Kipp Cannon4, S. Bradley Cenko5, Bart Dunlap1,
Adam Kraus6, Thomas J. Maccarone7
Draft version January 15, 2015
ABSTRACT
5
Low-cost mass-produced sensors and optics have recently made it feasible to build telescope ar-
1
rays which observe the entire accessible sky simultaneously. In this article we discuss the scientific
0
motivation for these telescopes, including exoplanets, stellar variability and extragalactic transients.
2
To provide a concrete example we detail the goals and expectations for the Evryscope, an under-

n construction780MPixtelescopewhichcovers8,660squaredegreesineachtwo-minuteexposure;each
a night, 18,400 square degrees will be continuously observed for an average of ≈6 hours. Despite its
J
small 61mm aperture, the system’s large field of view provides an ´etendue which is ∼10% of LSST.

3 The Evryscope, which places 27 separate individual telescopes into a common mount which tracks
1 theentireaccessibleskywithonlyonemovingpart,willreturn1%-precision,many-year-length,high-
cadence light curves for every accessible star brighter than ∼16th magnitude. The camera readout
]
times are short enough to provide near-continuous observing, with a 97% survey time efficiency. The
M
arraytelescopewillbecapableofdetectingtransitingexoplanetsaroundeverysolar-typestarbrighter
I than m =12, providing at least few-millimagnitude photometric precision in long-term light curves.
. V
h Itwillbecapableofsearchingfortransitinggiantplanetsaroundthebrightestandmostnearbystars,
p wheretheplanetsaremucheasiertocharacterize;itwillalsosearchforsmallplanetsnearbyM-dwarfs,
- for planetary occultations of white dwarfs, and will perform comprehensive nearby microlensing and
o
eclipse-timingsearchesforexoplanetsinaccessibletootherplanet-findingmethods. TheEvryscopewill
r
t also provide comprehensive monitoring of outbursting young stars, white dwarf activity, and stellar
s activity of all types, along with finding a large sample of very-long-periodM-dwarfeclipsing binaries.
a
When relatively rare transients events occur, such as gamma-ray bursts (GRBs), nearby supernovae,
[
or even gravitationalwave detections from the Advanced LIGO/Virgo network, the array will return

1 minute-by-minute lightcurves withoutneeding pointing towardsthe eventas it occurs. By co-adding
v images,thesystemwillreachV∼19inone-hourintegrations,enablingthemonitoringoffaintobjects.
2 Finally, by recordingall data, the Evryscope will be able to provide pre-eventimaging at two-minute
6 cadence for bright transients and variable objects, enabling the first high-cadence searches for optical
1
variability before, during and after all-sky events.
3
0
1. 1. INTRODUCTION notsensitivetotheverydiverseclassofshorter-timescale
0 Synopticskysurveysgenerallycoververylargeskyar- objects, including transiting exoplanets, young stellar
5 eas to detect rare events. Since it is usually infeasible to variability,eclipsingbinaries,microlensingplanetevents,
1 coverthousandsofsquaredegreeswithasingletelescope, gamma ray bursts, young supernovae, and other exotic
: transients, which are generally studied with individual
v they repeatedly observefew-degree-widefields, use large
small telescopes staring at single fields of view.
i aperturestoachievedeepimaging,andtiletheirobserva-
X Inthis paper we explorean approachto reachingrare,
tionsacrossthesky. Theresultingsurvey–suchasPTF
short-timescale events across the sky: using a large ar-
r (Law et al. 2009a), Pan-STARRS (Kaiser et al. 2010),
a SkyMapper(Keller et al.2007),CRTS(Djorgovski et al. rayoftelescopestoplaceapixeloneverypartofthesky
and integrating throughout the night to achieve depth.
2011),ATLAS(Tonry2011),andmanyothers–isneces-
These systems have been prohibitively expensive up to
sarilyoptimizedforeventssuchassupernovaethatoccur
now because of the extremely large number of pixels
on day-or-longer timescales. However, these surveys are
required to cover the sky with reasonable sampling, to
say nothing of the logistics of building and maintaining
1Department ofPhysics andAstronomy, UniversityofNorth the very large numbers of telescopes and storing the re-
CarolinaatChapelHill,ChapelHill,NC27599-3255, USA
2Department of Physics and Astronomy, Appalachian State sulting data. The rise of consumer digital imaging and
University,Boone,NC28608, USA decreasing storage costs offer a solution to these prob-
3DepartmentofPhysics,HighPointUniversity,833Montlieu lems. New surveys have exploited mass-produced com-
Avenue, HighPoint,NC27268, USA pactCCDcamerasandcameralensestocovereverlarger
4Canadian Institute for Theoretical Astrophysics, 60 St.
areas (e.g. Vestrand et al. 2002; Pollacco et al. 2006;
GeorgeSt.,Toronto, ON,M5S3H8
5Goddard Space Flight Center, Mail Code 661, Greenbelt, Pepper et al. 2007; Bakos et al. 2009; Mal ek et al. 2010;
MD20771 Law et al. 2013; Shappee et al. 2014). The logical end-
6Department of Astronomy, Univ. of Texas at Austin, 2515 pointofthisapproachisasurveywhichcoverstheentire
Speedway, StopC1400,Austin,TX78712
7Department of Physics, Texas Tech University, Box 40151, skywithgoodpixelsampling;wehereexplorethescience
LubbockTX79409-1051 capabilities of such a system.
2 N.M. Law et al.
TABLE 1
All-sky gigapixel-scale telescope science cases
Science case Section
Bright, known objects §4
Transitingexoplanets §4.1
Bright,nearbystars §4.1.1
HabitableplanetsaroundnearbyM-dwarfs §4.1.2
Whitedwarftransits §4.1.3
TESSplanetyieldenhancement §4.1.4
Otherexoplanet detection methods §4.2
Transitandeclipsetimingexoplanetdetection §4.2.1
Stellarpulsationtimingexoplanetdetection §4.2.2
Nearby-starmicrolensing §4.2.3
Stellarastrophysics §4.3
Mass-radiusrelation §4.3.1
Youngstars §4.3.2 Fig.1.— The Evryscope all-sky gigapixel-scale telescope cur-
White-dwarfvariabilitymonitoring §4.3.3 rently under construction at UNC Chapel Hill. The system con-
Variabilityfromaccretingcompactobjects §4.4 sistsof27individualtelescopesbasedoncommodityhardware,en-
Unexpected stellarevents §4.5 capsulatedina1.8m-diametercustom-moldeddomewhichmimics
Faint transient events §5 the sky’s hemisphere, and mounted on an off-the-shelf equatorial
Nearby,YoungSupernovae §5.1 mount.
Gamma-RayBursts §5.2
Gravitational waveEMcounterparts §5.3 We summarize in Section 6.
Unknownorunexpected transients §5.4
2. PERFORMANCEMETRICSANDSCIENCEREGIMES
To provide an example set of capabilities, we discuss The Evryscope concept multiplexes many small-
theEvryscope(Figure1),thearraytelescopewearecur- aperture,low-costtelescopestocoverasmuchofthevisi-
rently constructing at the University of North Carolina bleskyaspossible. Thetelescopesareallmountedintoa
at Chapel Hill. The system (Law et al. 2012b,c, 2013, rigidhemisphere. Thehemisphereactsasaproxyforthe
2014) is a low-cost 0.8 gigapixel robotic telescope that dome of the sky itself; when rotated at the sidereal rate,
images8,660square degreesin eachexposure. Contrast- all the mounted cameras track the sky simultaneously.
ing the traditional telescope (Greek; far-seeing) to this
instrument’semphasisonoverwhelminglywidefields,we 2.1. Performance metrics
havenamedourarraytelescope the Evryscope,fromthe
With a fixed field of view on the order of the size of
Greek for wide-seeing.
theentireaccessiblesky(8,000-12,000squaredegreesde-
The Evryscope is designed to open a new parameter
pendingonacceptableairmass),theremainingimportant
spaceforopticalastronomy,tradinginstantaneousdepth
variables for the instrument design are: 1) the aperture
andskysamplingforcontinuouscoverageofmuchlarger
ofthe telescopesand2)the numberofpixelsspreadover
sky areas. An Evryscopeis essentiallya 10cm-scaletele-
the field of view.
scopepointedattheentireaccessibleskysimultaneously.
A useful metric for the evaluation of all-sky surveys
As such, these systems will complement large-telescope
is how rapidly the sky can be covered to a particular
pointed surveys by enabling much shorter-cadence ob-
SNR.Thisclearlyscaleslinearlywiththe fieldofviewof
servations of much larger numbers of targets. Although
the survey, but evaluating the scaling of the SNR with
we concentrate on the Evryscope array in this paper,
the other quantities requires including the effects of an
other systems are under development, including Fly’s Evryscope-like design’s relatively large pixel scale8. We
Eye (Vida et al. 2014) and HATPI (G. Bakos, private
start with the standard imaging noise equation:
communication).
Theabilitytoproducearealtime(few-minute-cadence) S T A
movie of the sky will enable realtime searches for tran- SNR= p exp tel (1)
sientandvariablephenomenaofalltypes,thephotomet- p(Sp+SsP2)TexpAtel
ric monitoring of millions of stars simultaneously, and where S is the source flux in photons per second per
p
can provide pre-imaging of unexpected events detected squaremeterofcollectingarea;T istheexposuretime;
exp
by other surveys. These capabilities have the potential A is the telescope collecting area in square meters; S
tel s
of significantly contributing to many fields, a selection istheskybackgroundfluxpersquarearcsecond;andPis
of which we describe in this paper. We summarize the the pixel scalein pixels per arcsecond. As we areaiming
science cases addressed in Table 1. for a simple metric rather than a precise calculation, we
The paper is organized as follows. In §2 we describe simplify the expression by assuming that all sources are
the astronomical regimes that currently-feasible all-sky concentratedinsinglepixelsandneglectingtheeffectsof
telescopescanaddress;in§3weintroducetheEvryscope dark current, readout noise and quantum efficiency.
as an example of such a system and detail the projected This relation can be simplified by exploring two
performance of our prototype system, as well as provid- regimes: bright targets where the sky background can
ing overviews of the data reduction design and the per- beneglected,andfainttargetswheretheskybackground
formance of our prototype systems which test the per- dominates the noise:
formance of individual Evryscope cameras. In the fol-
lowing sections we detail the science contributions such 8 set by the currently-feasible number of pixels to distribute
a system can make, to bright, known-object variability acrossthesky;currently,roughly0.5-3gigapixelsarefeasibleatthe
surveys (§4) and relatively faint transient surveys (§5). $106 levelwithreasonablestorageandcomputationrequirements.
All-Sky Gigapixel-Scale Telescopes 3
of view effectively cancels their relatively small aper-
tures). The arrays can therefore meet or exceed the
performance of current sky surveys in the bright-source
100 photometric-monitoring regime, including transiting ex-
2g oplanets(§4.1),microlensingevents(§4.2.3),eclipsingbi-
e
d nary and stellar pulsation monitoring for exoplanet de-
2m tection(§4.2.1and§4.2.2),andgeneralstellarvariability
e / (§4.3).
du 10 In the faint-source regime the sky background due to
n
e the large pixel scale becomes by far the limiting factor,
Et
suppressingthesystemperformancebyfactorsoftensor
hundreds compared to the bright-star case; this means
that all-sky telescope arraysare likely only to effectively
1 contributewheretheabilitytoachieveextremelyhighca-
SDSSWIYNK-EOLDTCIFHTASAST-FSRNMA-WPCSSEaumSpseVrWSTASSkPyMVaIpSpTeSArubaPruT-FHHSCAT-DSEoSutPhanSETvAryRsKRceoSppleeLrSST dsekaeyrnlcyies-tooimvfeoervomebrarsniedyrivnoagbtijioemncptssoo(rfotnarensackreyb.yaTrsehuaipss)eirnsnpcolruvedaadees(a§er5xo.tu1rn;edmoptethliye-
cal observations of gamma-ray-bursts and orphan after-
Fig.2.— The ´etendues of currently-operating general-purpose glows(§5.2); andcross-matchingobservationswithhigh-
sky survey instruments (plus LSST; adapted from Tyson 2010) cadence all-sky surveys operating at other wavelengths
compared to the Evryscope, the example all-sky array telescope
(§5.4) or spectra (§5.3).
described in this paper. We caution that the large pixel scale of
thearraytelescope andothersurveysbasedonsimilartechnology Thecross-overbetweenthese tworegimesissetby the
makes this comparisononlyvalidforbrightsources whichrequire pixel scale, and pushing to covering more pixels across
high-cadence monitoring. To calculate the ´etendues for multiple- the sky would push the crossover between ”bright” and
telescope surveys, we combine the telescope’s FoVs where a sur-
”faint”sourcestomuchfainterlevels. Thepixelsampling
vey has multiplesites and/or telescopes observing different fields;
where a survey has multiple telescopes simultaneously observing issetbythenumberofdetectorsandtelescopesthatcan
thesamefieldwecombinetheirapertures. be feasibly purchasedand mounted, along with the data
storage and reduction challenges. When the pixel scale
pushes to similar levels to current sky surveys (around
BrightsourceSNR=pS T A (2)
p exp tel
1 arcsecond per pixel), an all-sky-telescope array with a
similar´etenduetothoseskysurveysbecomescompetitive
SppTexpAtel at all magnitudes and timescales. Moving to providing
BackgroundlimitedSNR= (3)
p(S P2) simultaneous arcsecond-scale sampling across the acces-
s
sible sky would require 100s-of-gigapixels instruments,
For a particular instantaneous field of view (FoV), in which will be challenging.
the bright source regime the time taken to cover the
wholeskytoaspecifiedSNRandsourcebrightnessthere- 3. THEEVRYSCOPE
fore scales as 1 . In the faint-source regime the As a concrete example of a currently-feasible all-sky
FoV×Atel
time taken scales with P2 . telescope array, we here detail the Evryscope, currently
FoV×Atel under constructionat UNC Chapel Hill. The Evryscope
consists of a single hemisphere that contains twenty-
2.2. Science with all-sky telescope arrays
seven 61mm-aperture telescopes, each with a rectangu-
From the above scaling relations we see that for suf- lar 28.8MPix interline CCD imaging a 380 sq. deg. FoV
ficiently bright sources the pixel scale is not the limit- using Rokinon 85mm F/1.4 lenses and including a five-
ing factor, and we recover the standard ´etendue metric: element filter wheel. The hemisphere tracks the sky on
the telescope aperture times the field of view. For faint a standard German Equatorial mount, imaging an in-
sources the pixel scale becomes the most important lim- stantaneous 10,200 sq. deg. FoV (8,660 sq. deg. when
itingfactor,withtheachievableSNRlimitedbythepixel overlapbetween cameras is taken into account). The in-
scale squared. dividualtelescopes are fixed into holes in an aluminium-
These two performance regimes set where current reinforced fibreglass dome. The interline CCDs provide
relatively-large pixel all-sky telescope arrays can most anelectronicshutter,soduringnormaloperationtheen-
effectively contribute: monitoring known bright objects tire instrument operates with only one moving part: the
wherethepixelscaleisrelativelyunimportant(oncesys- RA drive. Each camera has a five-element filter wheel
tematicsarecontrolledfor);andsearchingforrareshort- used as a dark shutter; the survey normally operates
variability-timescale objects, where the need to monitor with a single filter, althoughwe retain the option for fu-
large numbers of targets or large areas of sky rapidly turemulti-filter surveys. Alldata is storedandanalyzed
is paramount. The source-magnitude crossover between on-site, with ∼100TB/year of compressed FITS images
the two regimes depends on the exact details of the storedinto network-accessible-storage(NAS) units. The
hardware and site, but exploring the currently-feasible hardware specifications are summarized in Table 2, in-
combinations for the systems described here we find the cluding an estimate of the photometric performance of
crossovertypically occurs around 15th magnitude. the system. We plan to deploy the Evryscope at the
In the bright sourceregime, the´etendues ofcurrently- CTIO observatory in 2015.
feasibleall-skytelescopesystemsexceedallcurrentlarge- The photometric performance calculations in Table
telescope sky surveys (Figure 2; the enormous field 2 use conservative assumptions: median V-band sky
4 N.M. Law et al.
TABLE 2
The specifications of the Evryscope
Hardware
System design 27individualtelescopes; sharedequatorialmount
Telescopeapertures 61mm
Telescopemounting Fibreglassdomew. aluminiumsupports
Detectors 28.8MPixKAI29050interline-transferCCDs
7e-readoutnoise@4sreadouttime
50%QE@500nm;20,000e-capacity w. anti-blooming
Fieldofview 380sq. deg. pertelescope(23.8◦×16.0◦)
10,200sq.deg. instantaneous total
8,660sq. deg. excludingoverlapregions
Skycoveragepernight 18,400sq. deg. (2-10hourspernight)
Total detector size 780MPix
Sampling 13.3′′/pixel
Observingstrategy Trackfor2hours;resetandrepeat
Datastorage Alldatarecordedforlong-termanalysis
100TB/year(compressed)
Performance
PSF50%enclosed-energydiameter 2pixelsincentral 2/3ofFoV;2-4pixelsinouter1/3
Exposuretimes 120sstandard(plusshorterforbright-starmode)
Surveyefficiency 97%efficiencyfrom4scamerareadout
Limitingmagnitude mV=16.4(3-sigma;120sexposure)
mV=18.2(3-sigma;1hour)
mV=19.0(3-sigma,1night)
Photometricperformance 1%photometry onmV<12starsevery2minutes(inc. scintillation)
3-millimagphotometryonmV=11.5starsevery20minutes
3-millimagonmV=6starsin10mins(saturation-limitedshortexps.)
1%photometry onmV=15starseveryhour
90°
60°
30°
n
o
ati
n0°
ecli
D
-30°
-60°
-90°0° 60° 120° 180° 240° 300° 360°
Right ascension
Fig.3.— The instantaneous Evryscope sky coverage, including 0" 20" 40" 60" 80" Outside FoV
Maximum differential refraction per ratchet
the individual camera fields of view, for a mid-latitude Northern-
hemisphere site (30◦N). The SDSS DR7 photometric survey cov-
Fig.4.— The maximum atmospheric-refraction-induced change
erage(Abazajianetal.2009)isshownforscale.
in stellar position over the course of a two-hour tracking ratchet,
for a system with a field of view of 120◦ located at a site with
brightness at a good dark-sky site of mV=21.8 (SDSS 33◦N latitude. 28% of the field maintains <13′′motion; 56% less
measurements; Abazajian et al. (2003)); atmosphere + than26′′(onePSFFWHMfortheEvryscopesystem). Weusethe
telescope throughput 45%; 50% of encircled energy simple atmospheric model detailed in Meeus (1991), adjusted for
withina4-pixelaperture,anda25%light-lossduetoav- typicalobservatoryaltitudes.
erage vignetting and angular quantum efficiency effects
each night. This “ratcheting” survey setup is designed
across the field.
for both exoplanet transit searches (precision long-term
The optimal survey filter depends on the exact sci-
photometry) and co-adding of images for deep imaging
ence being targeted; for the purposes of this paper we
(transient searches). To aid in both precision photome-
assume a V-band filter will be used, although the fil-
tryandco-adding,theratchetingsurveymaintainsPSF-
ter wheels give flexibility during operations to perform
width-levelpositioningfor≈56%oftheFoVontwo-hour
multiple-filter surveys.
timescales (limited by atmospheric refraction in part of
the field; Figure 4).
3.1. Telescope array tracking modes
An alternate approach would be to track the system
The Evryscope will track 8,660 square degrees for two for the length of an exposure only and ratchet back on
hours at a time before moving back (Figure 3); the field each exposure. However, this approach leads to stars
of view is wide enough that stars are continuously ob- moving across the telescopes’ FoVs from one exposure
served for an average of ≈6 hours each night (more at to the next, leading to potentially large PSF changes
high declinations; Figure 5). In total 18,400 square de- with the current commercial camera lens optical quali-
grees are continuously observed for at least two hours ties, and thus has the potential to greatly increase sys-
All-Sky Gigapixel-Scale Telescopes 5
The pipeline stores the data in a compressed format
whichallowsrapidretrievaloftime-seriescut-outimages
of interesting areas of the sky, also allowing easy after-
the-fact measurements of transients detected by other
surveys. Later developments of the pipeline will include
realtimedifferentialimageanalysistodetectnewsources
incrowdedregionsofthe sky. The detaileddesignofthe
Evryscopepipeline willbe describedin a future publica-
tion.
3.3. Evryscope Performance Testing
3.3.1. AWCams: High Canadian Arctic Planet-Search
Telescopes
Fig.5.—TheEvryscopeskycoverageinone10-hournight. The
intensityofthecolourationcorrespondstothelengthofcontinuous The AWCams (Arctic Wide-field Cameras) are two
coverage(between2and10hours,instepsoftwohours)provided small telescopes designed to search for exoplanet tran-
bytheratchetingsurvey. Foramid-latitudeNorthern-hemisphere
site(30◦N). sits around bright stars (V=5-10). They are similar to
individual Evryscope cameras except that they lack the
tematic errors in the photometry. The PSF differences
tracking that enables long exposures. The cameras are
areuptofactors-of-twoinFWHMfromthecentertothe
deployed at the PEARL atmospheric science laboratory
edges of each telescope’s FoV, and vignetting can also at 80◦N in the Canadian High Arctic, where continuous
vary by up to 50% (Law et al. 2013). In a few-minute-
winterdarknessgreatlyincreasestheirsensitivitytolong-
tracking mode, this change is large enough to produce
period exoplanets. The cameras, the AWCam project,
rapid changes in the SNR of the stars and the limiting
and the attained performance are described in detail in
magnitude of the system on each part of the sky, also
Law et al. (2012c, 2013, 2014a).
leading to potential data reduction and survey perfor-
The AWCams have been operating in the Arctic for
mance issues.
three winters, including a test run in February 2012
For these reasons we have designed the Evryscope for
and full-winter operations in the 2012/13 and 2013/14
two-hourtracking,simplifyingthephotometricreduction
winters. The robustness of the hardware and enclo-
andkeepingthePSFsandlimitingmagnitudesstablefor
sure design has been validated by essentially uninter-
at least 60 measurement epochs in each tracking period.
rupted operation throughout the entire deployment pe-
The two-hour-tracking mode also guarantees that each
riod, including a total of 10 months of completely unat-
part of the sky is covered for at least two hours each
tended robotic operation. Throughout the winters the
night, simplifying time-series photometry.
cameras kept themselves (and crucially their windows)
There are two possible tracking surveys: 1) observing
clear of snow and ice, took over 60TB of images, and
newfieldcenterseachnight,slightlyshiftedinRA;and2)
consistently maintained few-millimagnitude photomet-
maintainingfieldcentersforlongerperiodsatthe costof
ric precisions over several-month timescales (Law et al.
observingatless-optimalairmasses. The firstintroduces
2014a). This performance with similar lenses and CCDs
extra systematic noise from the night-to-night changes
to Evryscope hardware demonstrates that exoplanet-
of the delivered PSF and vignetting on each star; the
detection-level photometric precisions are achievable
second introduces extra airmass-produced systematics.
with our software pipelines even in relatively-hard-to-
We will decide the optimal strategy on the basis of on-
reduce untracked data.
sky systematics measurements.
3.3.2. Tracking Single-Telescope Prototype
3.2. Evryscope data reduction
We have also built and operated a prototype track-
The Evryscope data will be stored and analyzed
ing unit-telescope system based at the Appalachian
on-site. The system will generate approximately State University Dark Sky Observatory (DSO). The in-
100TB/yearofcompressedFITSimagesstoredintolow- dividual Evryscope telescope was mounted to a Cele-
costnetwork-accessible-storage(NAS)units. Theon-site stron CPC1100 using a custom-built dovetail plate and
analysispipeline,basedonthepipelinesdevelopedforthe wedge, providing tracked exposures over the two-minute
Evryscope arctic prototype cameras (§3.3.1), astromet- Evryscopetimescaleswithatolerancesmallenough(<1
rically and photometrically calibrates images, extracts pixel) to emulate the final Evryscope performance. The
sources and then associates them with a reference cat- dark-sky conditions at DSO have allowed us to verify
alog made from previous Evryscope epochs. The cur- that the limiting magnitude and image quality calcula-
rent version of the pipeline can keep up with the in- tions described above match the actual performance of
coming Evryscope data stream using a 12-core server, the system under observatory conditions.
allowing realtime data analysis. Imaging data will be
stored on-site and then physically transferred on hard 3.4. Summary of Evryscope Performance
disk drives for further analysis every few months. On
In the targeted range of declinations (110-degrees of
shortertimescales,thesource-associatedlight-curveswill
declination) the Evryscope will generate a dataset in-
be transferred by Internet to a cluster at UNC Chapel
cluding:
Hill for detrending and transit detection.
The realtime source association tests allow quick de- 1. two-minute-cadence multi-year light curves for ev-
tection of obvious new objects which may require rapid ery star brighter than m =16 in the target range
V
follow-up, such as supernovae (§5.1) and GRBs (§5.2). of declinations.
6 N.M. Law et al.
1.0
1-year survey
3-year survey
0.8
y
babilit 0.6
o
pr
n
o
Fig.6.— We evaluated the crowding levels for the Evryscope Detecti 0.4
bydeveloping atool tosimulateimages onthebasisofUSNO-B1
photometry(Monetetal. 2003),includingtheeffectsofcrowding, 0.2
the lens PSFs, the detector sampling, and photon and detector
noise. We show above representative 10-arcminute fields insingle
exposures and in one-hour co-adds, scaled to show the faintest 0.0
0 20 40 60 80 100 120 140 160 180 200
detectable stars. Crowding does not limit photometry of at least
90% of stars above 15 degrees galactic latitude in 120s exposures Period / days
(30degrees in60mexposures). Fig. 7.— The probability of detecting at least three significant
eclipse/transiteventsinanEvryscopesurvey,for1%-leveltransits
2. millimagnitudeminute-cadencephotometryforev- andeclipsingbinarieswithhighlysignificantdetectionsineachdat-
ery star brighter than m =12 in the target range apoint. Weassume33%weatherlossesandsimulatetheobserving
V
windoweffectsfora20-degree-declinationtargetobservedfromour
of declinations.
plannedEvryscopesite. Thedetectionefficiencyforlonger-period,
low-transit-depth planets will be reduced by the number of data
3. minute-by-minute record of all events in the sky pointstobesearched.
down to m =16.5, with only 3% deadtime for im-
V
age readout. is covered during transit times. The Evryscope’s con-
tinuous coverage enables averaging over more than 100
4. m =19 in one-hour integrations; every part of the datapoints and more than an hour for each transit oc-
V
sky observed for at least 6.5 hours per night. currence, enabling much improved photometric perfor-
mance compared to surveys which cover large sky areas
TheEvryscope’s13”/pixelimagescaleissmallenough by imaging smaller areas in sequence.
toallowseparatedprecisionphotometryforatleast90%
ofstarsobservedabove15degreesgalacticlatitude(Fig- 4.1.1. Bright, nearby stars
ure 6).
Follow-up observations of transiting exoplan-
This large dataset will allow a wide range of science
ets, by either emission spectra during secondary
to be performed simultaneously; we detail potential sci-
eclipse (Madhusudhan et al. 2011; Knutson et al.
entific contributions for Evryscope-like systems in the
2007) or transmission spectroscopy (Sing et al.
following sections. To simplify the descriptions of the
2011; Snellen et al. 2010; Tinetti et al. 2007;
accessible targets, we assume that multiple systems will
Vidal-Madjar et al. 2003; Charbonneau et al. 2002)
be deployed to cover both the Northern and Southern
techniques, have revealed direct measurements of
hemispheres,andsorefertocoveringallstarsofparticu-
albedos, atmospheric composition, chemistry, and
lartypes,ratherthanlimitingtoaparticularhemisphere.
even phase curves showing features on the planetary
Weprovideaquick-referencetothesciencecasesinTable
cloud layers. These observations have been performed
1.
for only a very few planets, however, because they
require one thing above all else: photons from the host
4. BRIGHT,KNOWN-OBJECTREGIME:EXOPLANETS
star. Without a star significantly brighter than most
ANDSTELLARVARIABILITY
narrow- field transit searches can currently monitor, it
TheEvryscope’suniquecontributiontoexoplanetswill
is prohibitively expensive to reach sufficient signal-to-
be its ability to monitor extremely large sky areas at
noise on the most interesting spectral features of the
high-cadence. It will thus simultaneously cover large
planet transit even with future observatories such as
numbers of rare, widely-separated targets that would
JWST (Madhusudhan et al. 2014; Rauer et al. 2011;
otherwise require individual telescopes for each target.
Kaltenegger & Traub 2009). Current exoplanet transit
This will enable large surveys for rare transiting objects
surveys (Table 3) are limited to fields of view in the
as well as providing the datasets for other more exotic
few-hundred-square-degree range at best and cannot
exoplanet-detection methods.
effectively search for transits around a large sample of
rare bright stars.
4.1. Transiting Exoplanets
The Evryscope will have an order-of-magnitude larger
The Evryscope’s uniquely large field of view enables fieldofviewthanthe next-largestcurrentexoplanetsur-
transit searches in stellar populations that, because of veys,allowingalong-termtransitsurveycovering70,000
their rarity, have been inaccessible up to now. Com- stars brighter than m =9 (Perryman& ESA 1997) (the
V
paredto currenttransiting planet surveys(Table 3), the brightest stars will be observedin a short-exposure sup-
Evryscope has at least 15 times the instantaneous field plement to the standard Evryscope survey strategy).
of view, at least five times the number of pixels, and Given the Kepler-measured planetary population rate
similar sky sampling. Crucially, the photometric perfor- (e.g. Howard et al. 2012) and detailed observing effi-
mance on bright stars is scintillation-limited, and so the ciency,weatherwindow,geometricandfalse-positivecor-
ultimate photometric performance of the system is set rections(Figure7;Law et al.2013),weestimatethatthe
by the aperture and the length of time that each star Evryscopewillatleastdoubletheknownnumberoftran-
All-Sky Gigapixel-Scale Telescopes 7
TABLE 3
The Evryscope comparedto currenttransiting planetsurveys
Survey FoVa Aperture ′′/pixel Sites MPix/site Targets Ref.
/ sq deg. / mm
KELT 676 42 23.0 2 16 Brightstars Pepperetal.(2007)b
SuperWASP 488 111 13.7 1(8tels.) 32 Generaltransits Pollaccoetal.(2006)
HAT-South 128 180 3.7 3(6tels.) 128 Generaltransits Bakosetal.(2009,2012)c
MEarth 2.8 400 0.8 2(8telseach.) 32 M-dwarfs Irwinetal.(2009)d
CSTAR 20 145 15 1(Antarctic) 4 Brightstars Wangetal.(2014)e
TFRM-PSES 19 500 3.9 1 16 M-dwarfs Forsetal.(2013)
AWCams 504&1295 71&42 22&36 1(Polar) 32 Brightstars Lawetal.(2013)f
NGTS 96 200 1.1 1(12tels) 48 Brightstars Wheatleyetal.(2013)
Kepler 105 950 4.0 1 95 Earth-likeplanets Kochetal.(2010)g
Evryscope 8,660 61 13.3 1 780 Brightstars,MDs,WDs Lawetal.(2012c)
aMaximumsimultaneousFoVoverallsites
bMultiplesitesobservingdifferentfields
cMultiplesiteswithcontinuous observationof128sq. deg.
dMultiplesitesobservingdifferentfields
eAntarcticsitewithcontinuous observationsduringwinter
fArcticsitewithcontinuous observations duringwinter
gSpace-based telescopewithcontinuous observations
siting giant planets around stars brighter than m =9. (Frith et al. 2013) and 8,889in (L´epine & Gaidos 2011).
V
With few-millimagnitude photometric precision, the
4.1.2. Rocky planets in the habitable zones of M-dwarfs Evryscope will be sensitive to planets as small as a few
Earth radii around the small-radius mid-M-dwarfs. On
Despite being the most common stellar type in
the basis of Kepler planet statistics we estimate that
our galaxy, the transiting planetary population around
a long-term Evryscope survey would detect 5-10 small
M-dwarfs has not yet been explored in detail be-
planets around bright M-dwarfs with good sensitivity
cause of their extreme faintness compared to solar-
to targets in the relatively small-radius habitable zones
type stars. Kepler can only cover a few faint thou-
of those faint stars. In addition, the system will simul-
sand targets in this mass range (e.g. Muirhead et al.
taneously searchfor giant planets transiting a 100-times
2012; Dressing & Charbonneau 2013) and individually-
larger sample of late M-dwarfs. This capability will be
targeted surveys like MEarth are also limited to a few
highly complementary to the new generation of infrared
thousand bright M-dwarfs at most (Berta et al. 2013).
radial velocity planet surveys that will come online at
The results of a large exoplanet transit survey covering
roughly the same time as the Evryscope survey (e.g.
1,000,000 M-dwarfs (Law et al. 2012a) and radial veloc-
Quirrenbachet al. 2010; Artigau et al. 2014).
ity surveys of a few hundred of the brightest M-dwarfs
(e.g. Butler et al. 2006; Endl et al. 2006; Bonfils et al.
4.1.3. White dwarf transits
2013) suggest that giant planets are extremely rare
around M-dwarfs, although rocky planets seem to be Whitedwarfsareanattractiveexoplanettransit-search
much more common (e.g. Howardet al. 2012). The one target because their small size enables the detection of
transiting planet detected around a relatively bright M- extremely small objects – rocky planets can occult the
dwarf, GJ1214b (Charbonneau et al. 2009), has been a star, moon-sized objects give 10%-range transit signals,
subject of huge interest, with dozens of characterization andlargeasteroidsmaybedetectable(Drake et al.2010;
papers published. Agol2011). Thedetectionofatransitingplanetarounda
Thenextstepistopushtowardsthedetectionoflarge white dwarf wouldgive intriguing insights into the char-
samplesofthe apparentlycommonrockyplanetsaround acteristics of planets in such an extreme environment;
M-dwarfs, comparing their population statistics and ul- the planetary system evolution during the star’s red gi-
timately composition and mass-radius relation to those ant phase; and possibly constraints on the properties
of planets around solar type stars. Recent Kepler plane- of very small rocky bodies. However, the small white
tary population statistics suggest that the nearest tran- dwarf radius greatly reduces the geometric transit prob-
siting rocky planet in the habitable zone of an M-dwarf ability, even for hypothetical close-in planets; the small
is less than 9pc awayfrom us (Dressing & Charbonneau size also reduces the transit time to minutes. Because
2013). However, to have a chance of finding these plan- white dwarfs are very faint it has been very difficult to
ets around M-dwarfs bright enough and nearby enough obtain a reasonably large sample of white dwarfs in a
to use for characterization, a large sample of nearby, transitsurveyatall,let aloneatthe requiredminute ca-
brightM-dwarfsmust be covered. In turn, their random dences. Furthermore, the short transit lengths require a
and sparse distribution across the sky means we must veryhighobservationaldutycycleoneachtargettohave
covera verylargesky areato reacha significantnumber a reasonable chance of detecting the transits.
of targets. The Evryscope will be capable of simulta- TheEvryscope’sall-skyall-the-timesurveywillbe the
neously monitoring all bright, nearby late K stars and first to be able to cover a very large sample of rela-
M-dwarfs for transiting rocky planets. A number of re- tively bright white-dwarfs; because the white-dwarfs are
cently released all-sky nearby M-dwarf catalogs provide sosmall,evenfainttargetscanbeeffectivelysearchedfor
the starting target lists for this bright M-dwarf survey: rocky-planet transits by searching for drop-outs where
2,970 for CONCH-SHELL (Gaidos et al. 2014), 8,479 in thewhitedwarfdisappearsforafewminutes. Evenfind-
8 N.M. Law et al.
ingoneortwohighly-significantdipswouldmakeawhite detection of a comparable number of planets in single-
dwarftransitcandidateworthfollowingupwithotherfa- datapoint-per-transit detections (Voss et al. 2013). Al-
cilities. though the number of targets covered is large, the false-
From current catalogs of bright white dwarfs (see positiveconcentrationinthislow-cadencedatawillmake
§4.3.3) we estimate that the Evryscope will be able to confirmation difficult and the low cadence will produce
simultaneouslycoverhundredsofwhite dwarfswith bet- a low probability of detection for even short-period
ter than 10% photometric precision in each two-minute exoplanets. The Evryscope’s photometric precision is
exposure, and ∼ 103 white dwarfs each night. This will comparable to Gaia’s photometric precision (2mmag at
enable us to place limits on the populations of – or even m = 12; de Bruijne 2012), although the Evryscope’s
G
discover – Mercury-sized objects. As a by-product of cadence is almost 20,000 times faster. The Evryscope’s
this exoplanet transit search around white dwarfs, the higher cadence will allow it to achieve higher detection
Evryscope will also detect new eclipsing white dwarf bi- sensitivities for short period planets and confirm long-
naries and periodically variable white dwarfs (§4.3.3). period planets detected in Gaia multi-year dataset.
4.1.4. Planet yield enhancement for TESS and other 4.2. Other exoplanet detection methods
exoplanet surveys 4.2.1. Transit and eclipse timing for exoplanet detection
TheTESS mission,thefollow-ontotheKeplermission, Transit timing variations allow us to use changes in
will cover the entire sky, searching for rocky transiting eclipse times to measure the influence of other bodies in
exoplanets around more 200,000 bright, nearby stars. a system on the transiting/eclipsing body’s orbit. Mea-
With its fourcameraswith a24◦×24◦ fieldofview each, surements of the variations have been successfully ap-
the mission requires a median stare-time on each part plied to the confirmation of multiple-planet systems de-
of the sky of months. With multiple-year surveys cover- tectedbyKepler(e.g. Mazeh et al.2013). Similartiming
ing a large fraction of the entire sky simultaneously, the variations have recently produced possible detections of
Evryscope will offer a highly complementary dataset to planets orbiting eclipsing binary stars (e.g. Potter et al.
TESS. 2011; Marsh et al. 2013), including faint stellar types
such as white dwarfs that are not easily amenable to
4.1.5. TESS host characterization photometry otherplanetsurveymethodssuchasradialvelocitiesand
reflected-light direct imaging. An all-sky telescope will
TESS is currently planned for launch in 2017
recordminute-cadence lightcurves for every eclipsing bi-
(Ricker et al.2014);bythetimeTESSisoperationalthe
narybrighterthanm =16.5. Withmultiple-year,every-
Evryscopewillhavecollectedathree-yeardatasetonev- V
night coverage when the targets are up, we will auto-
ery star in the TESS survey. The three-year Evryscope
maticallyobtainhundredsofpreciseeclipse-timesforthe
dataset will enable a sensitive high-cadence search of
thousandsofshort-periodobjectsintheFoV.Compared
every TESS target for eclipsing binaries (of all peri-
to the current standard approach of selecting and moni-
ods), flare stars, exotic binaries, rotational modulation,
toringindividualinteresting targetsonlongertimescales
sunspots, and all other intrinsic photometric variability,
(Marsh et al. 2013) this massively-multiplexed eclipse-
on timescales similar to the TESS cadence.
timing survey will enable a much larger and more com-
4.1.6. Increasing the TESS planet yield prehensiveeclipse-timingsearchforexoticplanetarysys-
tems.
By the TESS launchthe Evryscope’sdataset will con-
tain at least one transit event from almost every giant 4.2.2. Stellar pulsation timing for exoplanet detection
planet transiting TESS targets, in orbital periods up to
Stellar pulsations can also serve as accurate clocks for
severalmonths (andgiventhe number ofstars surveyed,
discovering planets (see Schuh 2010). Confirmation of
alargenumberofwell-sampledsingle-transiteventsfrom
the pulse timing method’s ability to find unseen com-
much-longer-periodplanets). Thisofferstheopportunity
panions has been provided in several cases (e.g., Vinko
of using the Evryscope dataset to confirm long-period
1993; Barlow et al. 2011), althoughnone of the detected
planets that may only transit once during the TESS 60-
objectswere planets. A few substellar andplanetaryde-
day stare period. The Evryscope could thus greatly im-
tections have been reported, but without radial velocity
prove the TESS long-period planet yield, especially be-
confirmation(Silvotti et al. 2007; Mullally et al. 2008a).
causewecanusetheTESSsingle-transitdatatopickout
The pulse timing technique is most sensitive to plan-
shorter lengths of Evryscope data to search for transits,
ets when (i) the pulsations have relatively short peri-
decreasing the required significance of individual detec-
ods, from minutes to several hours; (ii) multiple high–
tions in Evryscope data.
amplitude, independent modes are present and well–
separated in frequency space; and (iii) the pulsation pe-
4.1.7. Gaia
riods are adequately stable, preferably to 1 part in 108
Gaia typically revisits each field 70 times over its 5 or better. Objects with pulsation characteristics best
year mission. Given this revisit configuration and other meeting these criteria include the hot subdwarfs, white
constraints, Gaia is expected to astrometrically detect dwarfs, δ Scutis, and roAp stars, for example. The
∼2,000 Jupiter-size planets within than 200pc, most or- Evryscope’s cadence is well–suited to monitoring these
biting aroundbrightGKdwarfsstarswithperiods rang- type of pulsations, although for the shortest-period hot
ing between 1.5 to 9 years (Sozzetti 2011; de Bruijne subdwarf and white dwarf pulsators, they might appear
2012). In addition, it has been suggested that the pho- in the super-Nyquist regime. While other planet detec-
tometry obtained during these revisits could allow the tion methods quickly lose their utility at larger separa-
All-Sky Gigapixel-Scale Telescopes 9
tion distances and longer orbital periods, the pulse tim- areaofthesky. Long-periodsystemshaveotherwisebeen
ingmethodremainsrelativelyrobustinthisregime,asit largely neglected by previous variability surveys, which
dependsprimarilyonthehoststar’soveralldisplacement do not have a significantly high long-term duty cycle to
from the barycentre (and not a perfectly edge–on or- identify the occasional eclipses of long-period systems.
bital alignment or large radial velocity). The Evryscope The Evryscope will achieve a more complete inventory
should provide pulse timings for thousands of pulsators of such systems over the entire sky, making an unprece-
that are sensitive to planetary–sized objects. dented contribution to the mass-radius relationship for
very low mass stars (e.g. Law et al. 2012a; Zhou et al.
4.2.3. Nearby-star microlensing 2014).
Typical galactic microlensing events occur on week- 4.3.2. Young stars
timescales, but exoplanets orbiting the lens star (or
even isolated planets) can be detected as much shorter Stellarvariabilityisubiquitousamongyoungstars(e.g.
timescale bumps in the light curves. Most microlens- Skrutskie et al. 1996; Carpenter et al. 2002). Newly-
ing surveys (for example, OGLE; Udalski et al. 2008) formed stars were first identified as a class from their
have been performed with larger telescopes observing variability, a feature which is still recognized in their
relatively small fields towards the galactic plane, where name (T Tauri stars; Joy 1945; Herbig 1962). This
there is a large population of background stars for lens- variability is driven by stochastic brightness variations
ing. However,occasionalspectaculareventsaroundrela- from the accretion of circumstellar material (as for T
tively nearby stars (e.g. Gaudi et al. 2008) have demon- Tauri itself) as well as quasi-periodic rotational mod-
strated that a sufficiently large-area survey has the op- ulation from spots (as in BY Draconis stars). This
portunitytodetectmuchcloserevents–anddetectplan- variability was crucial in compiling early catalogs of
ets smaller than Earth in half-AU orbits (Gaudi et al. young stars (e.g., Kenyon & Hartmann 1995), but over
2008). The key to successful microlensing planet de- the past 15 years, it has been largely supplanted by
tection is continuous monitoring and rapid follow-up. wide-field space-based surveys in the mid-infrared (e.g.,
The Evryscope’s few-minute temporal resolution, high Evans et al. 2009). However,these surveys are only sen-
photometric precision and all-sky coverage mean that sitive to stars which host protoplanetary disks or en-
planetary signatures will be directly visible in the light velopes; the disk-free population has remained largely
curves (this has recently been demonstrated in smaller unidentified. Even youth indicators like X-Ray and UV
fields by Shvartzvald et al. 2013). A survey with the emission (e.g., Wichmann et al. 1996; Scelsi et al. 2007;
Evryscope’s sky coverage is expected to detect several Findeisen & Hillenbrand 2010; Shkolnik et al. 2011) re-
near-fieldmicrolensingeventseachyear,alongwithmany mainswampedbycontaminationfromfieldintermediate-
more conventional distant events towards the galactic age stars, spectroscopic binaries, and chance alignments
plane (Gaudi et al. 2008; Han 2008). with background extragalactic sources. The Evryscope
willdirectly identify disk-freeyoungstarsbasedontheir
4.3. Stellar Astrophysics spot-driven variability (e.g. Cody & Hillenbrand 2014),
which can achieve photometric amplitudes of σ ∼ 0.1
With two-minute-cadence monitoring of every star
mag in the optical for stars younger than 100 Myr
brighter than ∼ m =16.5, all-sky array telescopes will
V (Herbst et al. 2002; Cody et al. 2013).
enablethediscoveryandcharacterizationofawiderange
of stellar variability. 4.3.3. White-dwarf variability monitoring
The per-exposure detection limit of our program is
4.3.1. Mass-radius relation from eclipsing binaries
roughly equal to that of the Edinburgh-Cape (EC) Blue
The Evryscope will provide a complete full-sky in- Object Survey (Stobie et al. 1997). Zones 1 and 2 of
ventory of eclipsing binary systems with orbital peri- the EC survey cover 3,290 sq deg of the southern sky
ods of P . 60 days, greatly expanding the number of and include 229 white dwarf stars (Kilkenny et al. 1997;
systems which are amenable to measurements of stellar O’Donoghue et al. 2013). Scaling from this surface den-
radii and dynamical masses. These measurements are sity, we expect to monitor more than 600 white dwarfs
crucial for the study of the stellar mass-radius relation, with every exposure, and more than 1,000 each night.
which is currently uncertain at the 10% level for K-M These data will be sensitive to pulsations, rotation, and
stars (e.g. Lo´pez-Morales 2007; Boyajian et al. 2012) various binary phenomena.
and directly carries through to uncertainties in models In certain ranges of temperature, white dwarf stars
of stellar evolution (Chabrier et al. 2007; Morales et al. experience non-radial g-mode pulsations that result in
2010; Feiden & Chaboyer 2013) and determinations of photometricvariationshavingperiodsbetween∼1.5min
the radii of transiting extrasolar planets (Fortney et al. and 30 min with amplitudes ranging from 0.1% to
2007; Charbonneau et al. 2007; Swift et al. 2012). Pre- 10%(Winget & Kepler2008;Fontaine & Brassard2008;
vious results have shown that stellar radii could be bi- Althaus et al. 2010). Though this variation will not be
ased by stellar activity (Lo´pez-Morales 2007) and rota- directly visible in the Evryscope light curves of most of
tion (Kraus et al. 2011), which argues that long-period these stars, for many of them, it will announce itself by
>
eclipsingbinarysystems(∼10daysperiod,whicharenot excess scatter in the light curves and will provide can-
tidally lockedandcanrotateatthe samevelocityassin- didates for follow up time-series photometry and astero-
gle field stars) will be crucial for determining the true seismic analysis. Among the hydrogen-atmosphere pul-
mass-radiusrelation. Asmallsetoflong-periodK-Msys- sators (the ZZ Cetis) the 2-min integration times will
temswereidentifiedbyKepler(Prˇsa et al.2011),thanks mean Evryscope is relatively insensitive to the hotter
to its 100% duty cycle, but it could only survey a small pulsators, which tend to have periods from 100-200 s
10 N.M. Law et al.
and amplitudes ∼1%, but we will generally be able to TheresultsoncataclysmicvariablesfromKeplerhigh-
detect the cooler pulsators because of their longer peri- light what can be done with the Evryscope. Scaringi
ods and larger amplitudes. At m =15.5, a typical 600 (2014) has shown, for example, that with long, well-
V
s oscillation with an amplitude of 1.5% will be detected sampledKeplerobservations,itispossibletomakestud-
in one season of observing, while at the single-exposure ies of cataclysmic variables’ power spectra that can be
detection limit, signals at the same period greater than compared very well to those made for X-ray binaries.
3.5% will be detected. We note as an example that in To study things like non-linear variability properties
one season of observing, Evryscope data will constrain (Scaringi et al. 2012), time series with lengths of many
the phase of the dominant mode of the cool ZZ Ceti days and good cadence are needed. Kepler has pro-
BPM 31594(m =15) to . 3 s, sufficient, over time, to videdtheseforahandfulofsources,whiletheEvryscope
V
place constraints on cooling rates and orbiting planets shouldbeabletoprovidesuchlightcurvesformanymore
(Kepler et al. 2005; Mullally et al. 2008b). sources. Additionally, Kepler has provided such light
TheEvryscopewillalsobesensitivetorotationinsome curvesforahighlybiasedsampleofcataclysmicvariables
whitedwarfstars. Mostwhitedwarfspresumablyrotate, –theobjectswhichareknowntobebrightCVsaheadof
but it is often difficult to detect the rate of rotation, time. Transientswithlowdutycyclesarenotincludedin
which is important for understanding angular momen- the sample, so comparisons of their behaviour with that
tumlossontheAGB.White dwarfswithmagneticfields of the persistently bright objects cannot be made.
canshowspotsontheirsurfacesthatresultinphotomet- X-ray binaries are another class of object which show
ric variability as they rotate (Brinkworth et al. 2013). anecdotal evidence for minute-scale (and faster) optical
The detected periods range from 725 s (Barstow et al. variability, but which can often be hard to study. X-
1995) to years. These stars will not only be useful as ray binary outbursts typically evolve on timescales of
probes of rotation but will also be candidates for spec- weeks, meaning that it is difficult to obtain sufficient
troscopicandpolarimetricfollowuptoconfirmandstudy target-of-opportunitytime tosamplethemwell,butitis
their magnetic fields, and stable rotators can be used as also not possible to study their bright phases with clas-
probes of motion resulting from an orbiting companion sically scheduled planned observations. It is clear that
(Lawrie et al. 2013). dramaticmid-IRvariabilitycanbeseeninX-raybinaries
Whitedwarfswithbinarycompanionsproducevarious on fast timescales (Gandhi et al. 2011), and that strong
typesofphotometricoscillation. Thesecondarycanshow variability on sub-second timescales can be seen as well
periodic variation resulting from reflection effect and el- (Kanbach et al.2001). Havingconstantopticalmonitor-
lipsoidal variations, and, of course, the system may be ing to compare with intensive or all-sky monitoring ob-
eclipsing, which can, among other things, provide im- servationsinthe X-rayswillprovideavaluable resource.
portantconstraintsonthe white dwarfmass-radiusrela- Additionally, Type I X-ray bursts should be optically
tionship. Forsomebinaries,suchasthebright(m =12) detectable from many accreting neutron stars (see e.g.
V
white dwarf + hot subdwarf CD−30◦11223 (Geier et al. Pedersen et al. 1982), meaning that the Evryscope will
2013),Evryscopedatashoulddeterminethephaseofthe provide better monitoring of the rates of Type I bursts
ellipsoidal modulation to ∼3.5 s every observing season. formanymoresourcesthanX-raymonitorscanprovide.
The changeof period ofthis system resulting fromgrav-
itational wave radiation is 6 × 10−13ss−1. Given the 4.4.2. Spin-up of magnetic white dwarfs
above phase precision, Evryscope data could detect this
Inasubsetofaccretingwhite dwarfsystems,the mag-
change in approximately a decade.
netic field of the white dwarf is strong enough to chan-
nel the accretion flow down the magnetic pole. When
4.4. Variability from accreting compact objects
these systems have magnetic and rotation axes for the
In recent years it has started to become clear that white dwarf which are different, the emission varies pe-
accretion onto compact objects is a relatively univer- riodically on the spin period of the white dwarf due to
sal process, with global similarities in the accretion pro- a “lighthouse effect”. The typical spin periods of inter-
cess in disks around supermassive black holes, stellar mediate polars are a few hundred seconds, well matched
mass black holes, neutron stars, and white dwarfs (see to the Evryscope’s cadence, so that with high cadence
e.g. McHardy et al. 2006; van der Klis 1994; Scaringi coverage, it should be possible to measure their spin
2014). The new combinationofhightime resolutionand period evolution. We estimate that the rotation peri-
high duty cycle of observation has opened new parame- ods of 10-15 known sources will be observable with the
ter space for studies of cataclysmic variables, in partic- Evryscope, and periodic emission may also be visible
ular. The Evryscope will offer three major advantages from a similarly-sized group of currently unknown ob-
over Kepler – the ability to observe stars which are not, jects. Testing whether the spin-up of magnetic white
a priori, recognized as interesting; coverage of a much dwarfs agrees with theoretical models will give a rela-
wider part of the sky (which is important for observ- tively easy way to test the general theory of accretion
ing rare objects); and potentially longer time baselines. torquing that is often applied to explain how millisec-
LikeLSST, itwill be able to detect outburstsfromcata- ond pulsars form in binaries with neutron stars (e.g.
clysmic variables and X-ray binaries, but it may also be Smarr & Blandford 1976).
able to detect a possible hidden population of outbursts
from shortorbitalsystems, which should have especially 4.5. Unexpected stellar events
short outbursts (Knevitt et al. 2014).
Monitoring very large numbers of stars will nearly
4.4.1. Aperiodic variability from accretion flows inevitably lead to discovering examples of very rare
and/or unknown stellar variability. Long-term variable