The Wide-field Infrared Survey Explorer (WISE): MissionDescription and Initial 4n-orbit Performance

Edward L. Wright', Peter R. M. Eisenhardt', Amy Mainzer', Michael E. Ressler, Roc M.Cutri3 , Thomas Jarrett', J. Davy Kirkpatrick', Deborah Padgett, Robert S. McMillan',

Michael Skrutskie5 , S. A. Stanford' Martin Cohen', Russell G. Walker s John C.Mather', David Leisawitz 9 , Thomas N. Gautier III', Ian McLean', Dominic Benford9,

Carol J. Lonsdale10 , Andrew Blain'', Bryan Mendez", William R. Irace', Valerie Duval'y wFengclivan Liu', Don Royer', Ingolf Ileinrichsen', Joan Howard 11 , Mark Shannonll

Martha Kendall ll Amy L. Walsh", Mark Larsen' Joel G. Cardon" Scott Schick",_ MarkSchwalm 17 , Mohamed Abide , Beth Fabinsky' Larry Naes 14 , Chao-^ ei Tsai'

Wright@astro.ucla.edu

ABSTRACTThe all sky surveys done by the Palomar Observatory Schmidt, the European Southern Ob-

servatory Schmidt, and the United Kingdom Schmidt, the InfraRed Astronomical Satellite andthe 2 Micron All Sky Survey have proven to be extremely useful tools for astronomy with valuethat lasts for decades. The Wide-field Infrared Survey Explorer is mapping the whole sky fol-lowing its launch on 14 December 2009. WISE began surveying the sky on 14 Jan 2010 andcompleted its first full coverage of the sky on July 17. The survey will continue to cover the sky asecond time until the cryogen is exhausted (anticipated in November 2010). WISE is achieving 5sigma point source sensitivities better than 0.08, 0.11, 1 and 6 mJy in unconfused regions on theecliptic in bands centered at wavelengths of 3.4, 4.6, 12 & 22 pm. Sensitivity improves towardthe ecliptic poles due to denser coverage and lower zodiacal background. The angular resolutionis 6.1", 6.4", 6.5" & 12.0" at 3.4, 4.6, 12 & 22 µm, and the astrometric precision for high SNRsources is better than 0.15".

1. Introduction'UCLA Astronomy, PO Box 951547, Los Angeles GA

90095-1547

2 jet Propulsion Laboratory, 4800 Oak Grove Drive,Pasadena, CA 91109

'Infrared Processing and Analy=sis Center, California In-stitute of Technology, Pasadena CA 91125

4 University of Arizona, 1629 E. University Blvd., Tuc-son, AZ 85721, USA

"Department of Astronomy, University- of Virginia,Charlottesville VA 22903

s Physics Department, University of California, Davis,CA 95616

7 Institute of Geophysics and Planetary Physics, LLNL,Livermore. CA. 94551

s donterey Institute for Research in Astronomy, 200 8thStreet, Marina CA 93933

NASA Goddard Space Flight Center, Greenbelt 1411)207,71

1 °National Radio Astronomy Observatory, Char-

The Wide-field Infrared Survey Explorer (WISE)will complete a mid-infrared survey of the entire

lottesville VA 22903"Ball Aerospace & Technologies Corporation, 1600

Commerce Street, Boulder, CO 80301' ) Space Dynamics Laboratory, 1695 North Research

Park way, North Logan, UT 84341rs California Institute of Technology, Pasadena CA 9112514 Lockheed Martin Advanced Technology Center (re-

tired') , Palo Alto, CA"Practical Technology Solutions, Inc., P.O Box 6336,

North Logan, Utah 84341

rs Space Sciences Laboratory, University of California,Berkeley CA 94720

17 L-3 Communications SSG-Tinsley, Wilmington, MA01887

sky by mid-July 2010 with much higher sensitivitythan previous infrared survey missions. The mostcomparable previous mission, the InfraRed Astro-nomical Satellite (IRAS; Neugebauer et al. 1954;Beichman et al. 1988) was launched in 1983 andmapped the whole sky in 4 bands with 62 individ-ually wired detectors. WISE is mapping the wholesky in 4 infrared bands W1..W4 centered at 3.4,4.6, 12 & 22 pm using a 40 cm telescope feeding ar-rays with a total of 4 million pixels. The increasednumber of detectors leads to a much higher sensi-tivity: WISE is achieving a sensitivity more thanone hundred times better than IRAS in the 12 pmband. While IRAS had two far-infrared bands at60 and 100 pm, WISE has two mid-infrared bandsat 3.4 & 4.6 pm. In these bands the only all-skysurvey that has been done is from the Diffuse In-fraRed Background Experiment (DIR.BE) on theCORE satellite, and DIRGE used only a singlepixel per band observing a 0.7° beam. A pointsource catalog was constructed by Smith et al.(2004) from the DIRGE data with flux limits of60 & 50 Jy at 3.5 & 4.9 pm, and WISE shouldreach flux limits that are 5 x 10 4 times lower inthese bands.

2. Implementation

WISE is a MIDEX (medium class Explorer)mission funded by NASA. The project is man-aged and operated by the Jet Propulsion Labora-tory (JPL) for the PI, Edward L. Wright. Majorcomponents of the project were built by contrac-tors, with the Space Dynamics Laboratory (SDL)of Utah State University building the instrumentand Ball Aerospace & Technologies Corporationbuilding the spacecraft. See Liu et al. (2008) formore details about the development of the WISEflight system. The data processing and analysis isbeing done by the Infrared Processing and Analy-sis Center (IPAC). Education and Public Outreachactivities for WISE are led by the UC BerkeleySpace Sciences Lab (1tlendez 2008).

The WISE flight system is 285 cm tall, 200 cmwide and 173 cm deep. It has a mass of 661 kg.It uses 301 W of power, while the solar panels canprovide over 500 W. The 40 cm diameter telescopesits inside a solid-hydrogen cooled cryostat. Thecryostat plus telescope and camera have a mass of347 kg, and they solid hydrogen has a mass of 15.7

Fig. 1.— Diagram showing the WISE flight sys-tem in survey configuration with cover off. Thespacecraft bus to the left of the biped supportswas provided by BATC, and the cryogenic instru-merit to the right of the bipods was provided bySDL.

kg at launch. Figure I shows the flight system inoperational configuration.

2.1. Mission Design

It is easiest to observe sources when the Sun isnot in the same part of the sky, since maximiz-ing the elongation from the Sun makes it mucheasier to keep optics cool, and to avoid scatteredlight from the Sun. But to scan the entire sky onemust observe the ecliptic poles, which are always90° from the Sun. Thus WISE adopted a basicstrategy of scanning great circles with a center lo-cated at the Sun, which keeps the solar elongationat 90°. These circles scan from the North Eclip-tic Pole (NEP) to the South Ecliptic Pole (SEP),crossing the ecliptic at longitudes ±90` from theSun. The entire sky is covered by this scan pat-tern in half a year as the Earth-Sun line turns by180".

Confusion noise would prevent detection of keyWISE science targets unless the beam size wereless than 50 square arc-seconds, and since thereare 5 x 10 11 square arc-seconds in the sky, WISEclearly needs to transmit large quantities of datato the ground. This led to choosing a low Earth or-bit to minimize the transmission loss. The actual

47'

1 frame 1 orbit 2 orbits Many orbits

Fig. 2. Coverage by WISE in one frame (every11 seconds), in one orbit, in two orbits, and inmany orbits. The gray levels show the depth ofcoverage with the darker areas having more cov-erage..

f

pr

Fig. 3. Cartoon showing WISE coverage on thesphere for 1 orbit, for 2 consecutive orbits, andfor 2 orbits separated by 20 days, illustrating thehighly redundant coverage at the ecliptic poles.

WISE orbit started with a mean radius of 6909kin mid-December 2009, and the radius haddecreased by 0.8 km due to atmospheric drag bymid-June 2010. In a polar orbit the average geoidheight is (Re,, + Rl,) f 2 = 6368 km, so the averagealtitude above the geoid is 540 km. This altitudewas chosen to reduce the exposure to trapped radi-ation belts in the South Atlantic Anomaly (SAA).Since looking at the Earth would both blind thedetector arrays and overwhelm the cryogenic sys-tem, a Sun synchronous orbit was chosen, with aninclination {97.51 chosen to give a precession rateof 360° in one year. The right ascension of the as-cending node was chosen so that the local time atthe equator crossings is either 6 A-M or 6 P `1I. Forthe actual WISE launch at 14:09:33 UT on 14 Dec2009 the ascending node is at 6 Pl1 local time.Since the end of the orbit normal close to the Sunis at —7.5° declination for dawn launch from theWestern Test Range (at Vandenberg AFB, CA),during June the angle between Sun and the orbitnormal can be as large as 31°, and eclipses occur

Fig. 4.-- WISE pointing and orbit during theJune solstice. Note that WISE points perpen-dicular to the Earth-Sun line and not toward thezenith.

over the South pole. Figure 4 shows the the rela-tionship between theWISE orbit, the Earth-Sunline, and WISE line-of-sight during the eclipse sea-son. COBS was in a 99° inclination orbit at 900km altitude with a June-centered eclipse season.The alternative of a 6 PM launch from the VTRgives an eclipse season centered on the Decembersolstice, as was the case for IRAS.

By synchronizing the scans and the orbit so thetelescope is pointed at the NEP when the satel-lite is closest to the North pole of the Earth, oneachieves a scan pattern that covers the entire sky,with the line of sight (LOS) never closer to theSun than 90° and never further than 31' from thezenith.

WISE has adopted the 2XIASS strategy of mov-ing the telescope at a constant inertial rate, andusing a scan mirror to freeze the sky on the ar-rays for the integration time. The field of view(FOV', of WISE is 47 1 , and by choosing to have asmall (10%) overlap between frames, the cadencebetween frames is set at 11.002 seconds per frame.For WISE 8.8 seconds of this is spent. integrat-

0.2o:

00

W1c00 0.8a

`m

0 0.6

0.4 A4

100

90

80

70

60

50

40

d 30

20

10

0

1.0............

4x '.............°.

16x

,F

0 30 60 90 120 150 180 210 240 270 300Days since 1 Jan 2010 00:00 UT

Fig. 5. VISE survey coverage vs. date. As of15 July 2010 more than 990 of the sky had beencovered to depth of 8 frames or more. The fractionof the sky covered to a depth of 4, 12 or 16 framesis also shown. The dotted portions of the curvesshow, the anticipated coverage until the expectedexhaustion of the cryogen. The small slowdowns inthe survey progress occur when the Moan crossesthe scan path. The dashed line shows a simpleprediction of the sky coverage based only on thelongitude of the Sun, starting on 14 Jan 2010.

ing and it takes about 1.06 seconds to read outthe arrays. The LOS then Jumps forward 42' in1.1 seconds. There are about 15 orbits per day,so the scan circle advances by about 4' per or-bit, giving about 12 orbits where the FOV hits agiven source. Figure 2 shows how the redundancybuilds up, while Figure 3 shows scans on the ce-lestial sphere.

The Moon crosses the scan circle twice a. month,and this would leave gaps in the sky coverage be-cause the Moon is bright enough to cause straylight problems when it is less than 15° from theline of sight. Thus WISE uses a modified scanpattern where the scan circle gets slightly aheadbefore the Moon interferes, and then drops slightlybehind to recover the region the Moon blotted out.The Moon moves 13° per day in ecliptic longitude,so with a. 15' exclusion zone WISE needs to be1.2` ahead ,just before the Moon crosses the scancircle, and then drop back to 1.2° behind. In addi-tion, the SAA affects part of 4 orbits per day, andto mitigate reductions in the sky coverage due tothe SAA, a small toggle of ±0.2 deg is added tothe longitude of the scan circle center, with thesign changing with a two orbit period. Finally,

2 5 10 20 30X Cµm1

Fig. 6.— The weighted mean VISE relative spec-tral response functions in electrons per photon.

the scan circle central longitude is biased by +5°on the side of the scan where the Sun is approach-ing the scan, which is the 6 PM side of the orbit.This means that the scans are actually 90° and95° fiom the Sun when crossing the ecliptic, al-lowing for a 5 day recovery period after a satelliteanomaly without causing a gap in sky coverage.

The WISE mission design provides at Ieast 8frames on over 99% of the sky in a. 6 month sur-vey interval after allowing for data lost to theMoon and the SAA. Heinrichsen '&- Wright (2006)describe the mission operations system in more de-tail. Figure 5 shows the survey progress vs. date.In mid-July 2010 WISE completed its first passover the sky, but will continue to survey until itscryogen runs out.

Duval et al. ('2004) and Mainzer et al. (2005)describe the WISE hardware design.

2.2. Instrument

The WISE telescope, imager, dichroic beam-splitter, detectors, and cryostat were built by theSpace Dynamics Laboratory of Utah State Uni-versity with major subcontracts from LockheedMartin Advanced Technology Center, L3 Com-munications SSG-Tinsley, and DRS Technologies.Larsen k; Schick (2005) describe the WISE sci-ence payload in more detail. The instrumentwas tested, then passed a vibration test, andwas tested again, before being delivered to BallAerospace in May 2009 for final integration withthe spacecraft. The testing included a determi-nation of the optimal focus, a measurement of thepoint spread function, and a determination of both

4

10°

0

411 € _ 4

Q10-t76C

0 -2

,

10-3

CDW

10410 20 30

X [PM]

Fig. 7.— The weighted mean WISE relative spec-tral response functions after normalizing to a peakvalue of unity, on a logarithmic scale.

the relative spectral response and the absolute sen-sitivity of the system.

The WISE short-wavelength channels employ4.2 and 5.4 µm cutoff HgCdTe arrays fabricated byTeledyne Imaging Sensors with 1024 x 1024 pix-els each 18 pin square. For the long-wavelengthchannels, the detectors are Si:As BIB arrays fromDRS Sensors & Targeting Systems with the same1024 x 1024 pixel format and pitch. The first fourand last four pixels in each row and column areused as non-illuminated reference pixels, so the ef-fective size of the arrays is 1016 x 1016 pixels. Themedian pixel scale is 2.757"f pixel with a rangeof ±0.6c among the different axes and arrays.The band 4 data is binned 2 x 2 on board, giv-ing 5.5"Ipixel. All four arrays image the samefield of view simultaneously using 3 dichroic beamsplitters.

All of the arrays use sampling up the ramp. The11 sec cadence between frames is divided into 10parts. A reset-read of the arrays occurs during thefirst 1.1 sec, followed by 8 read cycles each 1.1 secapart and a final reset cycle while the scan mirrorflays back. For the W3 and W4 arrays the 9 readsare multiplied by weights of -4, -3, -2, -1, 0, 1, 2,3 & 4 and summed to give the slope of the ramp.The WI and W2 arrays showed excess noise in thefirst sample so the weights used are 0, -7, -5, -3,-1,1,3,5& 7.

The spectral response of the system has beendetermined in three different ways: the whole sys-tem response was measured using a Fourier trans-form spectrometer (FTS), the system response

was estimated using the product of componentdata (measured with an FTS), and the responsewas computed from component design calculations. Since the sharp edges in the response func-tion cause ringing and negative values in the FTSspectrum, a. "best" combined estimate of the rel-ative spectral response was computed as follows:first a power law was fit to the ratio of the totalsystem measured response to either the compo-nent prediction or the design prediction. The pre-dictions were then adjusted to better match themeasured transmission using these fits. The fitswere weighted by the square of the response func-tion times a I /C

2 statistical weight, so the cor-rection is best in the filter passbands. After theadjustment the three methods gave consistent re-sponses. Finally the best estimate for the spectralresponse was made using a weighted mean of themeasured, predicted from components, and designvalues. In the weighted mean, zero or negativeresponses were given zero weight, the design. andcomponent predictions were given unit weight, andthe measurements were given weights of SNR2.This weighted mean relative spectral response isshown in Figures 6 and 7. The curves in the loga-rithmic plot have been normalized to a peak valueof 1. Note that the small red leak in the W3 trans-mission is based on data with a low SNR and maynot actually exist.

WISE measures the signal S given by

so( R(A)FudInv x R(A)AFada (1)

since the relative spectral response R is electronsper photon and dn-, x F„dv j (hv) .x F„ d Inv.

Specifying a central wavelength for broad bandslike the WISE filters, especially the very wideW3 band, is always ambiguous. LVe have cho-sen to specify the isophotal wavelengths for ourfilters. The isophotal wavelength and magnitudezeropoint are defined by requiring that a constantF.' give the same signal as Vega in a WISE band,and that Fa = F; ` ` (A:,,). Tokunaga & Vacca( 2005) give a clear definition of the isophotal wave-length, and values for Mauna Kea Observatoryfilter set. Cohen et al. (1992 ) give the IR spec-trum of Vega, based on a. model from Kurucz(1991, private communication to Martin Cohen).It is important to remember that WISE satu-rates on Vega, so the magnitude zeropoint is actu-

TABLE 1FLUX CORRECTIONS AND COLORS FOR POWERLAWS AND BLACKBODIES

F, f,, (W f^(T4'2) f,(i 3) fe (Ih'4) [WI-W2] [W2-W3] [W3-W4]

V3 1.0283 1.0206 1.1344 1.0142 -0.4040 -0.9624 -0.8684v' 1.0084 1.0066 1.0088 1.0013 -0.0538 -0.0748 -0.0519v l 0.9961 0.9976 0.9393 0.9934 0.2939 0.8575 0.7200v0 0.9907 0.9935 0.9169 0.9905 0.6393 1.8357 1.4458V- 1 0.9921 0.9943 0.9373 0.9926 0.9828 2.8586 2.1272v- 2 1.0000 1.0000 1.0000 1.0000 1.3246 3.9225 2.7680V- 3 1.0142 1.0107 1.1081 1.0130 1.6649 5.0223 3.37341/_ 4 1.0347 1.0265 1.2687 1.0319 2.0041 6.1524 3.9495B,(100) 17.2062 3.9096 2.6588 1.0032 10.6511 18.9307 4.6367Bv (141) 4.0882 1.9739 1.4002 0.9852 7.7894 13.0371 3.4496Bv (200) 2.0577 1.3448 1.0006 0.9833 5.4702 8.8172 2.4949B„(283) 1.3917 1.1124 0.8791 0.9865 3.8329 5.8986 1.7552B„(400) 1.1316 1.0229 0.8622 0.9903 2.6588 3.8930 1.2014B,(566) 1.0263 0.9919 0.8833 0.9935 1.8069 2.5293 0.8041B,(800) 0.9884 0.9853 0.9125 0.9958 1.1996 1.6282 0.5311B„(1131) 0.9801 0.9877 0.9386 0.9975 0.7774 1.0421 0.3463K2V 1.0038 1.0512 1.0030 1.0013 -0.0963 0.1225 -0.0201G2V 1.0049 1.0193 1.0024 1.0012 -0.0268 0.0397 -0.0217

ally based on fainter stars calibrated to the Vegasystem. Therefore the debris disk around Vega(Aumann et al. 1984) is relevant only if it has af-fected the calibration used to determine the mag-nitudes of the fainter standards. A convenient andaccurate fit to the continuum of our Vega spectrumin the 2.5 < A < 29 ,um range we use is given by

Fa = 1.0158 x 10-16 (l - 0.0083ln(a 11 8.891 µm)2)xBa(14454 K), (2)

where BA is the Planck function, which matchesthe continuum spectrum to an average absoluteerror of 0.0450. The temperature in this adhoc fit is much higher than the effective tem-perature of Vega and the solid angle is muchlower than the real solid angle due to the in-crease with wavelength of the free-free and bound-free opacites. Absorption lines reduce the in-bandfluxes by only 0. 113, 0.57, 028 & 0.14'% in bandsI through 4. To avoid possible multiple solu-tion we use the continuum spectrum when solv-ing for A,,,. The isophotal wavelengths so de-fined are 3.3526, 4.6028, 11.5608 &- 22.0883 yinfor W I.A. The measured s ystem transmissions

were smaller than the expected values at longwavelengths, leading to an effective wavelength forWISE band 4 of 22 Mm instead of the expected23 um. Absolute measurements of Vega by NISX(Price et a1. 2004) show that a 2.7oio upward off-set from the model spectrum is needed at 21.3 pm(Cohen 2009). NN,-"e have applied this correctionto the WISE 22 ym band, giving magnitude zero-points on the Vega system in the WISE passbandsof F. = 8.180 x 10 -r5 2.415 x 10-r5 6.515 x10 -1' 5.090 x 10-1s W/cm2 ,/µm, which convertto F,° = 306.681, 170.663, 29.0448 &;, 82839 Jyin tiVLM4 using F,° _ (A? ,/c)F 0̂ . There is anoverall systematic uncertainty of ±1.5% from theVega spectrum in these flux zeropoints.

W%e have also observed in-flight a discrepancybetween red and blue calibrators in W3 and WW,the 12 and 22 pin bands. This amounts to about.-171c and 9s in the fluxes for W%3 and W'4, withthe red sources appearing too bright in W4 andtoo faint in W'3. The flux differences could beresolved by adjusting the effective wavelength ofW3 and WXA 3-51c blueward and 2-3% redward.respectively. This would change the zero rnagui-

6

tude Fv by about -8% in W3 and -}-4% in W4.But the zeropoints and isophotal wavelengths re-ported here are based on the relative spectral re-sponses derived from ground calibration withoutany such adjustment. The instrumental zeropointsthat define the conversion from counts to magni-tudes have been based on standard stars, whichare the blue calibrators. Given the discrepancy

between red and blue calibrators we estimate thatthe conversion from magnitudes to Janskys arecurrently uncertain by ±10% in W3 and W4. Up-dated values will be provided in the ExplanatorySupplement accompanying the preliminary datarelease.

This definition of the isophotal wavelength andflux zeropoint means that the color correctionterm for a source with a different spectrum thanVega vanishes by construction when Fa is a con-stant (Fv a v-z) and very nearly vanishes forRayleigh-Jeans sources with Fh a A` or Fv a v'.These spectral energy distributions bracket thevast majority of WISE sources, so the the colorcorrections are generally small. But the extremelywide W3 filter does lead to color corrections aslarge as 0.1 magnitude for a constant Fv . Table Igives the flux correction factors in the WISE bandsfor several input spectra, and the WISE colors forthese spectra. These factors multiply the signalS given by a. spectrum F„ = F;(Ais"/A)-^. Thusa spectrum with F" = coast = 29.0 Jy gives asignal that is f (W3) = 0.9169 times the signalfrom Vega, so one would need a constant Fr, of29.010.9169 = 31.7 Jy to give zero magnitude.

Outer Solar system objects like Centaurs willhave color temperatures close to 100 K and fora 100 K blackbody the flux correction factorf,(W3) = 2.6588 is quite large. Thus a 100 Kblackbody would need 29.0/2.66 = 10.9 Jy at11.56 um to give zero magnitude in tV3.

The most common stars in the WISE catalogat high galactic latitude should be G-K dwarfs, sothe flux correction factors and colors for a K2Vand a G2V have been included in Table 1 usingKuruez (1993) spectra. Since the W2 baled in-cludes the fundamental CO bandhead, the fluxcorrection factors give the ratio between the sitnal from the star to the signal from a constant F'\equal to the average over a log normal passbandwith a 9.1% FNN'HVl centered on Ar.,o.

2.2.L Optics

The optical design and assembly of the WISEtelescope and camera was done by L-3 Commu-nications SSG-Tinsley. The optics consist of anafocal 40 cm diameter telescope that produces aparallel beam of light that is fed into the scan mir-ror, which works in the parallel beam, and theninto the all-reflecting camera. There are six mir-rors including folding flats in the afocal telescopebefore the scan mirror, and six mirrors in thecamera after the scan mirror. All ofthe mirrorsare gold-coated giving a high infrared transmis-sion. The design is described in more detail bySchwalm et al. (2005).

9.2.2. Cryostat

The cryostat (Naes et al. 2008) was built by theLockheed Martin Advanced Technology Center. Ituses solid hydrogen to cool the telescope to lessthan 12 K, and the Si:As arrays to less than 7.5K. There are two tanks of solid hydrogen: a largersecondary tank that cools the telescope and optics,the short wavelength arrays, and shields aroundthe smaller primary tank that just cools the Si:Aslong wavelength arrays. It has a predicted on-orbit lifetime of 10.75+ 1 5 months. Schick & Lloyd0.

(2009) describe the cryostat support system andtest results.

2.3. Spacecraft

The WISE spacecraft was built by Ball Aerospace,with a design based on the R.S300 series of sin-gle string spacecraft. The NextSat component ofthe Orbital Express mission was the first, RS300spacecraft to be launched, and it worked flawlessly.SWRI provided the spacecraft avionics.

'SSE is a three-axis controlled spacecraft thatis commanded to follow long scans at a constantinertial rate. The spacecraft provides attitude con-trol of better than 75, jitter of less than 1.3", anddrift rate variation of less than 0.2"/sec over 9 sec-onds. Angular momentum is stored on board in 4reaction wheels, and any buildup of excess angu-lar momentum is dumped using magnetic torquerrods. Primary attitude information is providedI y two Ball CT-633 star trackers. A fiber opticinertial measurement unit, Sun sensors and mag-

netometers are used during safe and emergencymodes of the spacecraft.

7

A fixed solar panel provides over 500 watts ofpower. WISE is oriented so the solar panel isalways pointing very nearly at the Sun. Dur-ing eclipse, a 20 amp-hour lithium-ion batteryprovides flight system power. Science data andflight system telemetry are stored on a 96GB flashmemory card for later transmission to earth. Afixed high gain antenna is used to transmit datato the Tracking and Data Relay Satellite System(TDRSS). The spacecraft body steers the antennato TDRSS for downlink. Science operations haveto be stopped during data transmission, so thedata downlinks are scheduled while the satellitepasses over the poles of the Earth, where the skycoverage is highly redundant.

2.4. Mission Operations

Except for looking up instead of down, WISE isvery similar to an Earth observing satellite. WISEis in a Sun-synchronous low Earth orbit like theLandsats and many other Earth observers, andit uses the TDRSS to download a large volumeof data. Thus WISE is being operated by theEarth Sciences Operations Center at JPL whichalso operates the JASON ocean topography map-ping mission.

The preparation of a command load for WISEstarts with a set of survey planning parameters,which then are used along with a predicted orbitto generate the survey plan for a week in the fu-ture. Times when a turn to TDRSS would be com-patible with the WISE Sun and Earth pointingconstraints are chosen, and a complete sequenceof pointing commands covering half a week thatcombines observing with TDRSS downloads is cre-ated. Additional commands for momentum un-loading and annealing the W3 and NN'4 detectorarrays are inserted into the command sequence.This process is repeated twice a week to allow forthe uncertainties in the predicted orbit caused bythe fluctuating atmospheric drag due to solar ac-tivity. In the first half of 2010 the solar activityhas been anomalously low, minimizing the errorsin the predicted orbits.

2.5. Data Processing & Archiving

WISE science data processing, archiving anddistribution is performed by the Infrared Process-ing and Analysis Center (IPAC), California In-

stitute of Technology. The processing softwareand operations system is based on algorithms,pipelines and architecture used for the 2MASS,Spitzer Space Telescope and GALES projects.

Following each TDRSS downlink contact, sci-ence image data packets are sent to IPAC wherethey are uncompressed and assembled into rawFITS format images in the four WISE bands.The raw FITS image headers are populated withflight system engineering telemetry, navigationdata and survey events information that is sentasynchronously from Mission Operations at JPL.The images from each downlink transfer are thenpassed to the WISE Scan/Frame pipeline thatperforms instrumental calibration (droop correc-tion, linearization, bias subtraction, flat-fielding),source extraction, photometric and astrometriecalibration and artifact identification on the in-dividual exposure frames. The WISE Multi-frame pipeline operates on collections of calibratedsingle-frame images that cover a region on the sky.The single-frame images are coadded by register-ing them onto a common pixel grid, matchinglocal background levels, and optimally combin-ing resampled pixel values using outlier rejectionto suppress transient pixel events such as cosmicrays, noise excursions and fast moving objects(Masci & Fowler 2009). Sources are detected andcharacterized on both the resulting combined im-ages and individual frames, and spurious detec-tions of image artifacts are identified and flagged.Both Scan/Frame and Multiframe processing areaccompanied by detailed quality assessment tomonitor data accuracy relative to WISEs sciencedata requirements.

Source detection and characterization in theScan/Frame and Multiframe pipelines exploitWISES simultaneous four band measurements.Sources are detected by thresholding on a com-bined four-band, matched-filter image, similarto the chi-squared detection method describedby Szala,y et al. (1999). Profile-fit photometry isperformed on all bands simultaneously, using amaximum-likelihood fitting procedure that alsooperates on multiple, independent image framesin the hlultiframe pipeline.

Astrometric calibration of single-exposure WISEimages is performed by solving for the frame cen-ter positions and rotations using the in-situ mea-sured position of 2MASS PSC astrometric refer-

10 15 20Vega magnitude

0.00

0.6

0.5E

0.4

0.3

E 0.2

E0.1

Fig. 8.----- The Wolf-Rayet wind bubble NGC 2359in the infrared using W1 as blue, W2 as cyan, W3as green, and W4 as red. Image width is 0.4°, withNorth to the left and NIvest up.

ence stars. Band-to-band rotations and offsets,plate scales, and field distortion are determinedby off-line analysis using the apparent positionsof reference stars in thousands of image frames.Full WCS information derived from the astromet-ric solutions is included in the single-frame imageheaders. The single-frame astrometric solutionsare propagated to the coadded images and ex-tracted source lists in the Mnitiframe pipeline.

Processed WISE images and extracted sourcedata and metadata will be archived and served tothe community via the web-based and machine-friendly interfaces of the NASA JIPAC InfraredScience Archive (IRSA). WISE data products willbe interoperable with other data centers and ser-vices through the IRSA infrastructure.

3. On-orbit Performance

WISE observes each patch of sky many timesover a period of a. day or more during its normalsurvey operation. These multiple frames can becombined into co-added multicolor mosaics suchas Figure 8, which shows the the Wolf-Rayet windbubble NGC 2359 (Schneps &:: Wright 1980). TheWISE project is posting an image of the week onits multimedia gallery which can be reached vialhttp:l i wise.astro.ucla.edu.

While the sensitivity of WISE is specified for8 coverages, this is a worst case. Even oil theecliptic, where the coverage is the sinallest, themost likely number of fraines covering a given

Fig. 9.--- Standard deviation of WISE magnitudesderived from the repeatability on sources seen in11 frames. The standard deviations have beenbinned into bins with width 0.2 mag, and the re-sulting histograms have been fit with a gaussianusing outlier rejection. The points and errorbarsshow the derived means and standard deviations.From left to right, the open squares show W4, thefilled squares show W3, the open circles show W2,and the filled circles show W1.

source is 11. The individual frames are ana-lyzed to provide astrometric and photometric in-formation. Analyzing the scatter among the val-ues from individual frames gives the noises shownin Figure 9. The dashed line shows a 5:1 SNRin 8 frames scaled to the 11 frame case plot-ted, giving a(m) = 0.185 mag. The curves show6(m) = 0.01 -I- [2.5/ ln(10)]n/10 -0.4, . The magnitudes where the curves cross the dashed line are17.11, 15.66, 11.40 & 7.97 mag. These give rawsensitivities of 44, 93, 800 & 5500 uJv for 5a in 8frames. There will be additional uncertainty dueto source confusion which is not included in thesenoises. Allowing for confusion gives sensitivites of0.08, 0.11, 1 & 6 mJy.

The astrometric accuracy of WISE can be es-timated by looking at the differences betweenWISE and UCAC3 (Zacharias et al. 2010) posi-tions, shown in Figure 10 for sources with SNR> 20. At lower SNRs, an additional error ofabout FtG'HXI/ (2 SNR) must be added in quadra-ture. Since this additional error is 0.15" for a PSFFWH11.I of 6 sec and SNR= 20, and the observedwidth of the distribution implies a- = 0.17" forSNR > 20, NO=ISE positions for high SNR sourcesshould be better than 0.15" for lor and 1 axis.

2000

1500

1000

500

c00 0U

1500

1000

500

0 1 1 1 1 1 1 r i

-1.0 -0.8 -0.6 -0.4 -0.2 0.0 0.2 0.4 0.6 0.8 1.0WISE-UCAC [arc-sec]

Fig. 10.- Histograms of the difference in right as-cension and declination between WISE and UCACpositions for objects with SNR in W1 > 20.

1.0

- W1----- W2

0.8 -----• W3

W ... W4U)0 0.6CL

Q^ 0.4

0.2

0.00 2 4 6 8 10 12 14 16 18

Radius ["]

Fig. 11. - Azimuthally averaged point spreadfunctions for WISE.

4

3cr

Ec

2

1n

M

0

0 2 4 6[4.6] - [ 12] in mog

Fig. 12.- Color-color diagram showing the loca-tions of interesting classes of objects. Stars andearly type galaxies have colors near zero, whilebrown dwarfs are very red in Wl-W2, spiral galax-ies are red in W2-W3, and ULIRGS tend to be redin both colors.

The image duality for WISE is good, as shownin Figure 11. These curves show the azimuthallyaveraged point spread function for WISE withFWHMs of 6.1", 6.4", 6.5" & 12.0" in WISE bandsIto4.

The observed saturation levels for WISE are0.3, 0.5, 0.7 & 10 Jy for bands 1 through 4. Satu-rated pixel values are replaced with a flag indicat-ing which of the 9 samples up the ramp was thefirst to saturate, so there is still a coarse indicationof the brightness in saturated regions.

4. Science Goals

GUISE has achieved 5a point source sensitivitiesbetter than 0.08, 0.11, 1 & E mJy at 3.4, 4.6, 12& 22 pm in regions observed in 8 or more frames,which is expected to be more than 99% of the sky.These sensitivities correspond to Vega magnitudesof 16.5, 15.5. 1L2 & 7.9. Thus WISE will go amagnitude deeper than the 2MASS K 5 data in Wlfor sources with spectra close to that of an A0 star,

Dec

RA

10

LLit> 0.1

001100 200

T

500 1000

eff

4

3

2

1

0 r 97

O

t1_^ 0 703

> _.

0 483

0 325

rd2

s^4

t

f't U

xr.'

$&i

`1

fx

0 2

0.5 1 2x

5 10 20

Fig. 13.— Brown dwarf spectra selected fromBurrows et al. (2003) smoothed to 1% resolution.Curves are labeled with the effective temperaturefrom 797 K at the top to 312 K at the bottom.

and even deeper for moderately red sources like Kstars or galaxies with old stellar populations.

The fundamental objective of WISE is to pro-vide a sensitive all-sky survey in the mid-infrared.This survey will probably provide its biggest pay-off in ways that have not yet been imagined. Butseveral fields can be identified where an all-skysurvey will provide a guaranteed payoff.

Many classes of extragalactic objects will bequite red in the WISE colors, as will brown dwarfstars which are among the Sun's closest neigh-bors. Figure 12 shows the color-color diagram con-structed from the WL.W3 bands, with the regionsoccupied by various types of objects illustrated.

4.1. Brown Dwarf Stars

Brown dwarf stars are very faint, since theyare not massive enough to fuse hydrogen into he-lium. As a result, they gradually fade and cool,and old brown dwarfs will be, very cool and faint.Both Jupiter with L = 10 -^ L„ and Cliese 229Bwith L = 10_, L,-,,have very strong emission at

Fig. 14^ Flux in the WISE bands comparedto the bolom.etric flux as a function of Teff forthe models of Burrows et al. (2003). Simple func-tional fits are shown.

4.6 pm due to a lacy of methane absorption atthis wavelength (Kirkpatrick 2005). Thus the 4.6µm band of WISE is a powerful tool for findingcool brown dwarfs. Most of the brown dwarfsfound prior to WISE have been discovered usingthe z' band of SDSS or the J band of 2MASSor UKIDSS, but fairly high temperatures are re-quired before there is substantial emission in ei-ther of these short wavelength bands. As a resultthe currently known sample of brown dwarf,,; is bi-ased toward the hotter, and thus younger, objects.WISE is able to find 10 Gyr old brown dwarfs andthus should detect a. high density of stars in thesolar neighborhood. The expected number den-sity of brown dwarfs is 1-2 times the density ofordinary stars (Reid et al. 1999; Chabrier 2002).

The expected number of brown dwarfs thatWISE will see can be computed using models forthe emitted spectra of brown dwarfs and the lumi-nosity and effective temperature of brown dwarfsas a function of mass and age (Burrow=s et al.2003). 'These model spectra are plotted asvF,,,% Fb„i in Figure 13. The spectra can alsobe used to compute the dimensionless ratios ofvF„ j Fb,.t for each WISE band, which turn out to

11

10-150.2 0.5 1 2 5 10 20 50 100 200

Wavelength [µrn]

10-10

_6 10-11a^

E 10-12

(D10-13

10-14

be well described by simple functions of the effec-tive temperature L. The functional forms plottedin Figure 14 are given by:

vF„ C X4

= 0.161 (3)

Fbot

with x = hv1(1.34kTe) for band I in blue,

vFv 2.25 min(T^/230,1)2

Fbol

15 x4(4)R 4 2x — 1

with x = by f (2.29kTe) for band 2 in green,

vF„ 15 x4= 0.55— (5)

Fbol 7T 4 ex — 1

with x = hvl(1.05kT,,) for band 3 in orange, and

vFv exp[-5.4log(T^ j170)] if T,, > 317= 0.9

Fbol ex( 20100 //T 70)2] otherwisep [— log( T,,, f 1(6)

for band 4 in red. Thus WISE band 2 sees a fluxfor Te f f > 230 K that is similar to a greybodywith a temperature 2.25 times higher than Te f f ,and with an emissivity 2.25 times higher than ablackbody. The fluxes in WISE bands 1 and 3also look like higher temperature greybodies butwith emissivities much smaller than a blackbody.The flux in WISE band 4 does riot behave like ablackbody.

But when Eisenhardt et al. (2010) surveyed10 square degrees with Spitzer to a flux limitat 4.5 pin 20 times deeper than the WISE allsky sensitivity, they found fewer very red sourcesthan predicted by combining the Burrows et al.(2003) models with either the Chabrier (2003)log normal IMF or the Reid et al. (1999} powerlaws. The models could be brought into agree-ment with the observed counts by assuming thatabout 50'X,, of the IRAC 4.5 yin flux was sup-pressed by an absorber other than methane orwater vapor. The amount of absorption neededis highly correlated with the assumed mass func-tion, but the Eisenhardt et al. (2010) observa-tion of about 10 brown dwarfs gives a model-independent prediction that WISE should findabout 10(1 41253 /1'10)/20' ' n. 400 late-T and Yclass brown dwarfs. The WISE sample of browndwarfs will be 3 magnitudes brighter than theSpitzer sample of Eisenhardt tit al. (2010), andthus much easier to follow up with the HST or theKeck telescopes.

Fig. 15.— Plot showing the spectrum of anL * galaxy at redsluft z = 0.33, a ULIRG(FSC15307 redshifted to z = 3 and scaled bya factor of 3), and observations of the z =6.42 QSO J1148+5251 with a scaled compos-ite quasar spectrum in orange, compared tothe WISE all-sky expected sensitivities in red.Other surveys are indicated with dot areas cor-responding to their sky coverage: from the left,GALEX AIS (Martin & GALEX Science Team2003), SDSS in green (York et al. 2000), DPOSSin black (Djorgovski et al. 1995), UKIDDS LAS inblue (Lawrence et al. 2007), 2MASS in magenta(Skrutskie et al. 2006), Akari in blue open circles(Matsuhara et al. 2005), and the IRAS FSC inblack (Moshir et a1. 1992).

4.2. Ultra-Luminous Infrared Galaxies

WISE will detect the most luminous galaxiesin the Universe. The Ultra-Luminous InfraredGalaxies (ULIRGs) are due to mergers that lead todust-enshrouded star formation and also to AGNactivity as gas is disturbed out of stable circu-lar orbits and falls into the central super-massiveblack holes in the merging galaxies.

In the concordance ACDM cosinology struc-ture formation proceeds in a bottom-up manner,so large objects are formed by the mergers ofsmaller objects. The peak of large mergers oc-curs at redshift:s 1 to 2 and leads to the peak inthe star formation history of the Uiiherse, andWISE has sufficient sensitivity at 22 Alin to seethe top end of the luminosity function at redshift3. As a result, WISE will detect many galaxieswith L > lo in L., ^. Quasar activity is also associ-ated with mergers, and the evolving quasar hir i-nosity function of Hopkins et al. (20071 predicts

12

TABLE 2NUMBER OF BROWN DWARFS WITH FLUXES GREATER THAN THE WISE 4.6 PM REQUIRED FLUX LLvIITOF 160 [tJY FOR EITHER THE CHABRIER (2002) LOG NORMAL IMF (LGNM) OR REID ET AL. (1999)

POWER LAWS WITH VARIOUS INDICES a FROM 0.4 TO 1.3.

IMF T, < 300 T, < 500 T, < 750 d < 1.3 pe

lgnm 6.9 221.1 1339.7 0.8820.4 2.6 76.4 593.5 0.3300.7 52 120.9 670.7 0.5341.0 10.6 196.6 920.8 0.9271.3 22.4 330.4 1310.2 1.743

that WISE will see over 1.5 x 10 5 quasars at 22pin and over 3.6 million quasars at 4.6 µm, withredshift distributions shown in Figure 16.

The abundance of ULIRGs as a function of lu-minosity and redshift will provide important in-formation about the formation mechanisms thathave led to the galaxies and clusters we see at lowredshift today.

106

W1 4.2.1. Red1Dusty AGN and QS0s

-a 105V)

z 104

103

102

1011 1 1 1 1 1 i

0 1 2 3 4 5 6z

Fig. 16. Number of QSOs and AGN brighterthan the expected sensitivity in the 4 WISE bandsover the entire sky, based on the Hopkins et al.(2007) luminosity function. The WI and W2curves lie almost on top of each other, with W3and W4 giving lower counts.

WISE will characterize and probe fully the ex-tent of the population of obscured active galacticnuclei and will provide a highly uniform set ofmid-infrared photometry for virtually all knownActive Galactic Nuclei (AGN) and Quasi-StellarObjects (QSOs). Most of what is known aboutthe properties and evolution of AGN and QSOshas been deduced from studies of objects dis-covered in ultraviolet-excess and optical emis-sion line surveys. However, infrared and radiosurvevs reveal that manv AGN in the universehave remained hidden from short wavelengthsurveys because of reddening and obscurationby dust in and around their nuclei ( Low et. al.1988; 'vVebster et al. 1995). Estimates of the frac-tion of AGN missing from optical/UV selectedsamples vary considerably, ranging from 15 toover 5010 (Richardset al. 2003: Brown et al. 2007;Glikman et al. 2004, 2007,. Such an obscuredpopulation may account for at least part of thehard X-ray background (Comastri et al. 1995),and it may contribute measurably to the far in-frared background.

The most extensive searches for red AGNhave been based on the 2'-1IASS PSC, using ei-

13

10-1 z

U W

W10-13N 0EUa^

1 0-14

10-15

0.4 1 10 100X [µm]

Fig. 17.— Fluxes from an 0.13 kin diameter nearEarth asteroid at a distance of 0.5 AU with elon-gation 90° for three albedos: 0.04, 0.145 and 0.52from bottom to top on the left side of the plot.These models assume a Lambertian surface. Thephase angle is 63°. The WISE 5u point source sen-sitivities for 8 stacked frames are shown as solidcircles.

Cher near-infrared color selection or combina-tion with mid- and far-infrared and radio mea-surements (Cutri et al. 2001; Gregg et al. 2002;Glikman et al. 2007). These studies have foundlarge numbers of predominantly low redshift AGN,adding considerably to the complete census ofAGN in the local universe. However, 2MASS isa relatively shallow survey and is biased againstthe most heavily obscured AGN at higher red-shifts. WISE will extend the 2MASS studies tohigher redshifts, probing the luminosity and num-ber density evolution of red AGN over a. significantfraction of the age of the universe.

One of the highest redshift QSOs, SDSS1148+5251, is shown in Figure 15, with data fromJiang et al. (2006) and a scaled composite spec-trum constructed from Vanden Berk et al. (2001)and Glikman et al. (2006). A similar source is de-tectable by WISE at 3.4 & 4.6 urn over the whole=sky, and also at 12 um with the better sensitiv-ity achieved at the actual ecliptic latitude of thissource (,3 = 46'--).

4.3. Asteroids

Asteroids are typically fairly dark, but do havea large range of albedos. As a result, the visualmagnitude is an imprecise indicator of the size ofall Stuart & Binzel (2004) show a range

in albedos from 0.023 to 0.63 for near Earth aster-oids, leading to a 5:1 range in estimated diametersand a 125:1 range in volumes. But the light thatis not reflected is absorbed and reradiated in thethermal infrared. Figure 17 shows fluxes from anasteroid with low, median and high albedos. Theratio of optical to infrared fluxes gives the albedo,and the infrared flux and color temperature give agood indication of the diameter of an asteroid.

WISE is sensitive enough to see 1.2 km diameter asteroids in the main belt at 2.7 AU fromthe Sun, and 0.13 kin near Earth objectsthat are 0.5 AU from the Earth and 1.1 AU fromthe Sun. These limits are for objects with knownorbits so that multiple frames can be stacked, andare for 5u. This gives 10`/c accuracy in diame-ters, which is comparable to the errors caused bythe unknown thermal inertia and rotation poles ofmost asteroids (Wright 2007). WISE will be ableto provide radiometric diameters for about 3 x 105asteroids using image stacking analysis. For anasteroid with a bolometric albedo of A = 5%, theWISE stacked frame sensitivity of 800 uJy at 12um corresponds to a,n optical flux of roughly

VIR AF;,(opt) ^ — FJR) = 2 pJy (7)vo,,t 1 — A

which is all magnitude of 23. If good op-tical data are available, all of these asteroids willhave well-determined albedos.

For new asteroids, a diameter of 2.4 km inthe main belt or 0.25 km for a Near Earth Ob-ject (NEO) at a distance of 0.5 AU gives enoughsignal-to-noise in a single frame to allow detec-tions. The NEOWISE program (Mainzer et al.2010) is searching the WISE data for moving ob-jects and sending positional information to theIAU Minor Planet Center. WISE should detectmore than 105 asteroids on single frames.

VISE provides about 10 measurements of theflux spaced over 30 hours for a typical asteroidwhich call used to derive rotation periods fortens of thousands of asteroids.

The accurate, diameters for previously knownasteroids plus view asteroids derived by the WISEtask will allow a significantly more precise knowl-edge of the size frequency distribution (SFD) ofsub-kin NEOs. Since only a few dozen sub-kmobjects currently have diameters that have beenaccurately measured by either radar or thermal

14

IR data, WISE data will significantly improve ourknowledge of the SFD of these smaller objects.Diameters derived from WISE measurements willreduce uncertainties in the SFD associated withthe traditional reliance on the optical H magni-tude. Also, since infrared discoveries of new as-teroids are relatively insensitive to albedo, the setof new discoveries made by WISE are much closerto a diameter-limited survey than optical surveys.This will remove a. significant source of uncertaintyin SFD calculations derived from optical discover-ies alone.

By observing asteroids at the time of their lo-cal afternoon and comparing this to their thermalfluxes in their local morning, WISE would enablemeasurement of the temperature differentials thatgive rise to the Yarkovsky effect, a phenomenonthat is often the dominant uncertainty in the longterm orbit propagation for small asteroids Measur-ing the Yarkovsky effect via the thermal infraredprovides rather different information for a differentpopulation than the radar method (Chesley et al.2003). The radar technique uses precision obser-vations of an asteroids trajectory over a decadeto look for small changes to the ephemeris causedby the Yarkovsky force. In order to obtain suffi-cient astrometric precision, only NEOs which arevery nearby can be observed. Radar data allowone to compute the tangential acceleration a pro-duced by the Yarkovsky effect. However, usinginfrared to look for temperature differentials be-tween an asteroids local morning and evening re-quires only two epochs of observation separated by3-8 months. The infrared technique is not limitedto NEOs and will also work for Main Belt asteroidssuch as the Baptistina family. The infrared dataallow one to compute the diameter which givesthe mass m, and the morning us. evening differ-ence gives the tangential force F. Thus combiningIR and radar data allows measurements of all thequantities in F = a. For asteroids in a family,the change in semi-major axis divided by the age ofthe family gives an estimate for the acceleration a,so the IR. data can be used to verify the time sincethe collision that created the family. The hypoth-esis (Bottke et al. 2007) that the K r/T impactorwas part of the Baptistina family depe=nds on theestimated 160 Xlyr age of this family, so measuringthe Yarkovsky force and diameter for a sample ofBaptistina family members will provide a valuable

test of this idea.. The infrared Yarkovsky detectiontechnique is improved by knowledge of an objectslightcurve. As mentioned above, the WISE ob-serving cadence is excellent for measuring asteroidlightcurves.

4.4. Other Science Goals

An all-sky survey in the thermal infrared can beused to address many scientific questions, not allof which require surveying the whole sky. But bysurveying the whole sky, WISE will enable theseprojects among many others:

4.4.1. Comet Trails and Zodiacal Banns

WISE will study the nearest planetary debrisdisk: our own zodiacal cloud and asteroid system.The interplanetary dust cloud is a complex distri-bution of dust grains from numerous sources evolv-ing under the influence of planetary perturbations,collisional evolution and solar radiation sources.The zodiacal dust cloud was long thought to comefrom comets (Whipple 1955, 1967, 1976) but cur-rent estimates of cometary dust production are anorder of magnitude below that needed to main-tain the cloud. The IRAS dust bands show thatasteroids are a substantial contributor to the in-terplanetary dust, cloud. There are at least sevendust band pairs girdling the ecliptic. Comet debristrails were also discovered by IRAAS (Sykes et al.1986, 1990) which provided more than 100 obser-vations of the trails associated with 8 short periodcomets. WISE scans through the ecliptic 30 timesper day which will permit frequent observationsof dust bands and cornet trails with an angularresolution many times better than IRAS. This al-lows a detailed assessment of the population andevolution of these structures in the zodiacal cloud.

4.4.2. Young Stars and Debris Disks

WISE will provide a. robust statistical databasefor studying star formation, the evolution of cir-curnstellar disks, and the dissipation of zodiacalclouds around stars of all masses, extending eveninto the brown dwarf mass regime. Detecting thesmall IR excess from the later stages of disk dissi-pation requires the ability to accurately predictthe long wavelength photosphcric emission andthe ability to accurately measure the total longwavelength emission. The simultaneous 4 color

15

10-10

6 10-11a>Cn

E 10-12

N 10-13

L21> 10-14

10-15

0.2 0.5 1 2 5 10 20 50 100 200Wavelength [grn]

Fig. 18.— Figure showing an MO T Tauri star ata distance of 2.5 kpc (TTS) and an A2 star with aplanetary debris disk at 300 pe (PDD) comparedto the VISE expected sensitivities. The unlabeledpoints for other surveys are the same as in Figure15.

photometry provided by WISE will enable the re-quired predictions, and the sensitivity of WISEwill enable detection of optically thick disk emis-sion for solar type stars out to 1 kpc distance anddown to the hydrogen burning limit in the Tau-rus and Ophiucus star forming regions. Becauseyoung stars are variable, the simultaneous four-color coverage of WISE provides a precise meansof characterizing thermal excess from circumstel-lar disks or zodiacal material. (A possible secondcoverage of the object by WISE 6 months after thefirst will even provide clues as to the nature of thevariability, e.g. extinction vs. accretion.)

Because WISE is an all-sky survey it will re-veal isolated young stars well outside obvious star-forming regions. For example, two of the nearestclusters (of any age) were discovered only withinthe last decade though their YSO-like properties:the TW Hydrae Association (d = 55 pc, by in-frared emission) and the Eta Chamaeleonis Clus-ter (d = 96 pe, by X-ray emission). Both haveabout a dozen known members and are youngerthan 10 Myr. Surprisingly neither is associatedwith a molecular cloud. Thy' Hya covers morethan 700 sq. degrees. A complete survey would bean impossible task for missions like Spitzer but isthe natural focus of WISE. Such intermediate-agevery-nearby clusters provide an ideal laboratoryfor studying the late stages of planetary accretion.

WISE will find young stars which are still ac-

creting molecular cloud material over a substan-tial fraction of the Galaxy: WISE will measurestars with optically-thick circumstellar disk emis-sion down to the hydrogen-burning limit (0.08Mra) out to 500 parsecs (e.g. the Orion molecu-lar cloud). Solar-type "Class I" and T-Tauri starswith optically-thick disks will be found to substan-tial distances (Figure 18 shows the MO T Tauristar Sz 82 scaled to a distance of 2.5 kpc), thanksto their strong excesses at 12 and 22 µm. Ex-panding upon Spitzer's survey of selected regionsof a few molecular clouds, WISE will provide a fullinventory of low-luminosity young stellar objectsin many dozens of star forming clouds extendingover the entire sky. WISE will also discover hun-dreds of the very youngest stars: the elusive "Class0" sources. These objects, which are in the earlystages of gravitational collapse, are rare becausethey remain in this phase for less than 100,000 yr.They are deeply embedded in dust, so they arevisible only at wavelengths longer than 20 µm, al-though they may be seen in scattered light in themid-Ili,.

WISE will survey planetary debris disks aroundthousands of stars in the solar neighborhood. Dur-ing the early stages of planet formation micron-sized grains accrete into successively larger parti-cles. The resulting reduction in grain surface areamakes the circumstellar disk optically thin, andthe infrared excess substantially decreases. Thedisk does not clear uniformly due to both the vari-ation in accretion/dynamical timescale with dis-tance from the star, and the shepherding effects offorming planets (Kuchner & Holman 2003). Thisnon uniformity is evident in the distribution of in-frared excess as a. function of wavelength throughthe mid-infrared. WISE will exploit these signa-tures to reveal the systematics of and refine thetimescales for disk clearing and solar system for-mation.

Although the planetesimal/planet formationphase is short. lived, dust excess signatures canpersist into the main-sequence phase as demon-strated in stars like Vega by IRAS (Am ann et al.1980. These excesses arise from small grains withPoy nting-Robertson lifetimes of tens of thousandsof years. Their existence argues that they mustbe actively replenished by the erosion of planetes-imals and thus they are signatures of successfulplanetary formation. Indeed, high resolution 20

16

ym imaging of infrared-excess objects like HR4796reveal dust rings which may betray the existence ofinvisible planets. Figure 18 also shows the knowndebris disk system ^ Lep scaled to a distance of300 pc. With the short wavelength WISE bandspinning down the photospheric flux, the excess at22 yin is easily detectable.

The sensitivity to see the photospheric flux andenough wavelength coverage to predict the pho-tometric flux are, once again, the key to reveal-ing excesses from eroding planetesimals. In addi-tion to studying the statistical distribution of suchsmall IR excesses, of great interest to the TPFmission, WISE will provide JWST with hundredsof targets for high-resolution imaging akin to theground-based and HST images of HR4796.

4.4.3. Interstellar Dust

WISE detects most components of interstellardust and produces very good maps of interstel-lar dust in the Galaxy. The 3.4 and 12 µm fil-ters include prominent PAH emission features, the4.6 pia filter measures the continuum emissionfrom very small grains, and the 22 pm filter seesboth stochastic emission from small grains and theWien tail of thermal emission from large grains.The WISE survey will map the high latitude dustto a sensitivity 60 times better than IRAS at 22ym and 200 times better at 12 µm. For typicalhigh latitude dust showing PAH emission the 3.4ym map will provide a signal to noise ratio compa-rable in sensitivity to the 12 pm map. The WISEsurvey thus extends the mapping of high latitudedust to cover the entire sky, filling in the holes leftby IRAS in the dimmest regions of sky, and ex-tending the Spitzer legacy projects GLIMPSE andMIPSGAL to the entire sky. The 3.4 pm map willprovide the first high sensitivity map of extendedemission of the entire sky at this wavelength.

These extended emission maps will allow studyof the composition and detailed structure of in-terstellar dust. Unique to WISE, a comparison ofthe 3.3 µm PAH feature with other PAH emissionfeatures in diffuse clouds can yield information onPAH composition. Results of the study of dust inthe Galaxy can also be compared to dust emissiondetected by WISE in nearby galaxies.

4 4 . 4 . Galactic Sh-UCtUre

Roughly half of the sky is dominated by theMilky Way's 400 billion stars, many of which arevariable in the infrared. With deep simultane-ous near- and mid-infrared observations, uniquein an all-sky survey, the 4-color WISE will ad-dress global issues of Galactic stellar structure andpopulations: the disk's warp and flare; the natureof the bar, Could's Belt, ring, and bulge; a de-tailed census of evolved stars; and the pursuit ofhalo star-streams from globular clusters and un-merged Galactic dwarf satellites. WISE comple-ments Spitzer's Legacy Project, GLIMPSE, whichsurveyed < 300 square degrees of the inner planemostly constrained to JbI < 1'. WISE will coverall the Galactic plane, and secure critical off-planestarcounts to probe the vertical population struc-ture in the arms and spurs, and scale heights ofthe central bar, molecular ring and bulge (diskcontamination diminishes rapidly off the plane).Although the stellar counts are confusion limitednear the Galactic plane with the WISE beam sizeand high sensitivity, the color-magnitude diagramsare robust to confusion effects; and thus delin-eation of differing stellar populations is still viablein the Galactic plane.

Off plane, unconstrained by confusion, WISEoffers unparalleled infrared survey depth, uniquelycapable of all-sky detection of even normal halostars. The "Spaghetti Survey" (Morrison et al.2000) of the halo uses optical photometry andspectroscopy to identify distant G-/K-giants andkinematic substructure caused'by the destructionof accreted satellites. WISE will aid the opticalquest for extra-tidal K-giants, detecting such starsto 70 kpc at 3.4 ym and 45 kpc at 4.6 µm, farbeyond the depth of 2MASS at 2.15 yin. Com-paring the optical spectra and WISE data on thequasi-continuum (3.4 gym) and CO fundamental(4.6 am) in these cool stars to synthetic spec-tra will enable the recognition of those stars thatbelong to remnant stellar swarms from partly di-gested satellites.

4.4.5. lNearby Galaxi'c4'

The angular resolution and sensitivity of WISEallow detailed study of the internal structure ofgalaxies in the local Universe, whose propertiesprovide our basic understanding of high-redshift

17

Fig. 19. Four nearby spiral and irregular galaxies that are well resolved by WISE.

galaxies. For the — 5000 galaxies within 20 Mpcof the Sun, WISE spatial resolution is < 500 par-secs (< 1 kpc at 22 pm). Of this set, nearly400 will be highly resolved, delineating globularclusters, giant molecular clouds and other discretesites of star formation. The short mid-IR bandsof WISE are an effective probe of the underlyingstellar population, and are nearly immune to ex-tinction. The long bands are sensitive to hot dustassociated with star formation, thus bridging thetime span between evolving generations of stars.WISE data will complement the Spitzer InfraredNearby Galaxies Survey Legacy project (SINGS,Kemricutt et al. 2004), expanding the SINGS sam-ple of 75 galaxies to many thousands of galax-ies, covering the widest range in morphologicaltype. It will also supplement the IRAC 3.6/4.5,am Spitzer Survey of Stellar Structure in Galax-ies (S4G, Sheth et al. 2005): WISE will providethe crucial mid-IR tracers of star formation activ-ity for this complete sample of galaxies within 30llpc.

4 . 4 . 6. Galaxy Clusters

WISE will provide a census of star formation invery large samples of normal galaxies and clustersout to cosmological distances. WISE will detectover 100 million galaxies in at least: one band, andmany million galaxies in all four of its bands. Theshort wavelength Wl-NN'2 color will be a usefulredshift indicator for galaxies out to z = 0.5, andthe flux measurements will provide an estimatefor the total stellar mass of galaxies at z < 0.5by measuring the rest frames 1.6-2.2 pin emission( Gavazzi et al. 1996). WISE will thus provide aglobal history of star formation in massive fieldand cluster galaxies over the, past 5 billion years.

Cluster galaxies generally contain stellar pop-

ulations that are old relative to those in the sur-rounding field galaxies, so the contrast of galaxyclusters relative to the field increases with redshiftwhen viewed in the IR. By virtue of covering theentire sky, WISE will be able to detect all mas-sive (M > 10"AI(:;) galaxy clusters up to z — 0.5,enabling the most complete census of the stellarmass in dense environments when used in conjunc-tion with large-area optical imaging surveys suchas SDSS, PanStarrs, DES, and LSST. When usedwith the higher spatial resolution optical imagingfrom these surveys, which can help to overcomethe confusion limit in the WISE images, WISE willalso be able to identify even higher redshift clus-ters (0.5 < z < 1) over thousands of square degreeareas at the poles where the larger than averagenumber of WISE scans will lower the noise in theWl data.

5. Data Products

WISE will produce an image atlas in its 4 col-ors that will be a stack of all the multiple framescovering each part of the sky. A catalog of sourcesextracted from these images will also be produced.These data products are modeled on the 2MASSimage atlas and the 2MASS point source catalog.WISE is not planning to produce a separate ex-tended source catalog.

In addition, the NEOWISE project is support-ing the release of the individual WISE frames anda database of all the sources extracted from the in-dividual frame s. This dataset will be ver y valuablefor retrospective studies of asteroids discovered inthe future.

There are two data releases scheduled forWISE. The first release will be a preliminary cat-alog based on the first 50 (;c of the data, to bereleased 6 months after the last data is taken for

18

WISE. Depending on the cryogen lifetime, thisrelease could cover about 7 5%, of the sky. Thepreliminary catalog will be complete to a depth of20u. A final data release covering the entire skyto a depth of 5a will be released 17 months afterthe last data are taken.

6. Prospects for Followup Observations

The low background from space makes theWISE survey very sensitive, and allows it to coverthe largest possible solid angle while still reachingfairly low flux levels. The faintness of the sourcesfound by WISE makes followup with ground-basedtelescopes quite difficult in the thermal infrared.However, the time required for a detection scaleslike BID' for background B and telescope diam-eter D, assuming diffraction-limited optics, so itwill be possible to confirm WISE sources withground-based 10 meter class telescopes, since the(10/0.4)4 = 390,625 diameter factor almost com-pensates for the factor of 2.5 x 10 6 in the back-grounds. Since WISE only integrates for 1.2 min-utes on a typical source, 15 minutes on a 10 me-ter class telescope can surpass the WISE sensi-tivity on point sources and give much better an-gular resolution. Even though ground-based fol-lowup in the thermal infrared is possible, themost useful ground-based fallowups will proba-bly be at complementary wavelengths. Exam-ples include ground-based J band measurementsof WISE brown dwarfs, ground-based optical as-trometry of WISE asteroids, ground-based opti-cal spectroscopy of WISE ULIRGs and AGN, andphotometric redshifts of galaxies based on opticaland infrared colors.

The Jaynes Webb Space Telescope will also ex-perience the low background intensity seen byWISE, will cover the WISE passbands, and will bemuch larger than WISE. Thus JWST will be themost capable platform for studying WISE sourcesin the thermal infrared ; since it will have both alarge diameter and low background. WISE data at3.4 and 4.6 lim will be used to improve the JWSTguide star catalog by providing unproved count

rate predictions and more recent astrometry.

7. Conclusion

WISE has exploited the low infrared back-ground in space and the power of large format

infrared arrays to map the entire sky with unprece-dented sensitivity and angular resolution between3.4 and 22 ym. The WISE catalog and image at-las will be extremely valuable for a large numberof astrophysical investigations ranging from aster-oid albedos in the solar system through the stellarpopulation of the solar neighborhood out to themost luminous star forming galaxies in the visibleUniverse.

This publication makes use of data productsfrom the Wide-field Infrared Survey Explorer,which is a. joint project of the University of Cali-fornia, Los Angeles, and the Jet Propulsion Labo-ratory f California Institute of Technology, fundedby the National Aeronautics and Space Adminis-tration.

REFERENCES

Aumann, H. H., Beichman, C. A., Gillett, F. C.,de Jong, T., Houck, J. R., Low, F. J., Neuge-bauer, G., Walker, R. G., & ^Vesselius, P. R.1984, ApJ, 278, L23

Beichman, C. A., Neugebauer, G., Habing, H. J.,Clegg, P. E., &; Chester, T. J., eds. 1988, In-frared astronomical satellite (IRAS) catalogsand atlases. Volume 1: Explanatory supple-ment, Vol. 1

Bottke, W. F., Vokrouhlick-y, D., &a Nesvorny, D.2007, Nature, 449, 48

Brown, M. J. L et al. 2006, ApJ, 638, 88

Burrows, A., Sudarsky, D., &: Lunine, J. I. 2003,ApJ, 596, 587

Chabrier, G. 2002, ApJ, 567, 304

2003, PASP, 115, 763

Chesley, S. R., Castro, S. J., Vokrouhlicky, D.,Capek, D., Giorgini, J. D., Nolan, M. C., itilar-got, J.-L., Hine. A. A., Benner, L. A. M., &rChamberlin, A. B. 2003, Science, 302, 1739

Cohen, M. 2009, AJ, 137, 3449

Cohen, 1I., Walker, R. G., Barlow, M. J., & Dea-con, J. R. 1992, AJ, 104, 1650

Comastri, A., Setti, G., Zarnorani, G., & Hasinger,G. 1995, A& A, 296, 1

19

Cutri, R. M., Nelson, B. O., Kirkpatrick, J. D.,Huchra, J. P., & Smith, P. S. 2001, in Bulletinof the American Astronomical Society, Vol. 33,Bulletin of the American Astronomical Society,829

Djorgovski, S. G., Gal, R. R., Odewahn, S. C.,de CarvaIho, R. R., Brunner, R., Longo, G.,& Scaramella, R. 1998, in Wide Field Surveysin Cosmology, ed. S. Colombi, Y. Mellier,B. Raban, 89

Duval, V. G., Irace, W. R., Mainzer, A. K.,& Wright, E. L. 2004, in Presented at theSociety of Photo-Optical Instrumentation En-gineers (SPIE) Conference, Vol. 5487, Opti-cal, Infrared, and Millimeter Space Telescopes.Edited by Mather, John C. Proceedings of theSPIE, Volume 5487, pp. 101-111 (2004)., ed.J. C. Mather, 101-111

Eisenhardt, P. R. M. et al. 2010, AJ, 139, 2455

Gavazzi, G., Pierini, D., & Boselli, A. 1996, A&A,312, 397

Glikman, E., Gregg, M.D., Lacy, M., Helfand,D.J., Becker, R.H. & White, R.L. 2004, ApJ,607,60

Glikman, E., Helfand, D. J., & White, R. L. 2006,ApJ, 640, 579

Glikman, E., Helfand, D.J., White, R.L. & Becker,R.H. 2007, ApJ, 667, 673

Gregg, M.D, Lacy, M., White, R.L., Glikrnan, E.,Helfand, D., Becker, R.H. & Brotherton, M.S.2002, ApJ, 564, 133

Heinrichsen, L & Wright, E. L. 2006, in Pre-sented at the Society of Photo-Optical Instru-mentation Engineers (SPIE) Conference, Vol.6270, Observatory Operations: Strategies, Pro-cesses, and Systems. Edited by Silva, David R.;Doxsey, Rodger E.. Proceedings of the SPIE,Volume 6270, pp. 62701C {2006).

Hopkins, P. F., Richards, G. T., & Hernquist, L.2007, ApJ, 654, 731

Jiang, L. et. al. 2006, AJ, 132, 2127

Kennicutt, R., Armus, L., Calzetti, D., Dale, D.,Draine, B., Engelbracht, C., Gordon, K., Helou,G., Hollenbach, D., Leitherer, C., Malhotra, S.,Regan, M., Rieke, G., Rieke, M., & Thornley,M. 2004, in Spitzer Proposal ID #159

Kirkpatrick, J. D. 2005, ARA&A, 43, 195

Kuchner, M. J. & Holman, M. J. 2003, ApJ, 588,1110

Kurucz, R. 1993, ATLAS9 Stellar AtmospherePrograms and 2 kmjs grid. Kurucz CD-ROMNo. 13. Cambridge, Mass.: Smithsonian As-trophysical Observatory, 1993., 13

Larsen, M. F. & Schick, S. 2005, in Presentedat the Society of Photo-Optical Instrumenta-tion Engineers (SPIE) Conference, Vol. 5904,Cryogenic Optical Systems and InstrumentsXI. Edited by Heaney, James B.; Burriesci,Lawrence G. Proceedings of the SPIE, Volume5904, pp. 166-177 (2005)., ed. J. B. HeaneyL. G. Burriesci, 166-177

Lawrence, A., Warren, S. J., Almaini, O., Edge,A. C., Hambly, N. C., Jameson, R. F., Lu-cas, P., Casali, M., Adamson, A., Dye, S.,Emerson, J. P., Foucaud, S., Hewett, P., Hirst,P., Hodgkin, S. T., Irwin, M. J., Lodieu, N.,McMahon, R. G., Simpson, C., Smail, L, Mort-lock, D., & Folger, M. 2007, MNRAS, 379, 1599

Lin, F., Cutri, R., Greanias, G., Duval, V., Eisen-hardt, P., Elwell, J., Heinrichsen, L, Howard,J., Irace, W., Mainzer, A., Razzaghi, A., Royer,D., & Wright, E. L. 2008, in Presented atthe Society of Photo-Optical InstrumentationEngineers (SPIE) Conference, Vol. 7017, So-ciety of Photo-Optical Instrumentation Engi-neers (SPIE) Conference Series

Low, F. J., Cutri, R. M., Huchra, J. P., & Klein-mann, S. G. 1988, ApJ, 327, L41

Mainzer, A. K., Eisenhardt, P., Wright, E. L., Liu,F.-C., Irace, W., Heinrichsen, I., Cutri, R., &Duval, V. 2005, in Presented at the Societyof Photo-Optical Instrumentation Engineers(SPIE) Conference, Vol. 5899, UV,lOptical/IRSpace Telescopes: Innovative Technologies andConcepts 11. Edited by MacEwen, Howard A.Proceedings of the SPIE, Volume 5899, pp. 262-273 (1 2005)., ed. H. A. MacEwen, 262--273

20

Mainzer, A. et al. 2010, in preparation.

Martin, C. &, GALEX Science Team. 2003, inBulletin of the American Astronomical Society,Vol. 35, Bulletin of the American AstronomicalSociety, 1363

Masci, F. J. &r Fowler, I W. 2009, in Astronomi-cal Society of the Pacific Conference Series, Vol.411, Astronomical Society of the Pacific Confer-ence Series, ed. D. A. Bohlender, D. Durand,P. Dowler, 67

Matsuhara, H., Shibai, H., Onaka, T., &-- Usui, F2005, Advances in Space Research, 36, 1091

M6ndez, B. J. H. 2008, in Astronomical Societyof the Pacific Conference Series, Vol. 389, EPOand a Changing World: Creating Linkages andExpanding Partnerships, ed. C. Garmany k-M. G. Gibbs Moody, I W., 385

Morrison, H. L., Mateo, M., Olszewski, E. W.,Harding, P., Dohin-Palmer, R. C., Freeman,K. C., Norris, J. E., & Morita, M. 2000, AJ,119,2254

Moshir, M., Kopman, G., & Conrow, T. A. O.1992, IRAS Faint Source Survey, Explanatorysupplement. version 2 (Pasadena: Infrared Pro-cessing and Analysis Center, California Insti-tute of Technology, 1992, edited by Moshir, M.;Kopinan, G.; Conrow, T.)

Naes, L., Lloyd, B., k- Schick, S. 2008, in Ameri-can Institute of Physics Conference Series, Vol.985, Advances in Cryogenic Engineering, ed.J. G. N,_kTchsend, I Barclay, S. Breon,J. Deniko,M. Dipirro, J. P. Kelley, P. Kittel, A. Klebaner,A. Zeller, M. Zagarola, S. van Sciver, A. Rowe,J. Pfotenhauer, T. Peterson, & J. Lock, 815822

Neugebaner, G. et al. 1984, ApJ, 278, LI

Price, S. D., Paxson, C., Engelke, C_ k- Murdock,T. L. 2004, AJ, 128 1 889

Reid, I. N., Kirkpatrick, J. D., Liebert, J., Bur-rows, A., Gizis, J. E., Burgasser, A., Dahn,C. C., 'Monet, D., Cutri, P,., Beichnian, C. A.,&: Skrutskie, M. 1999, ApJ, 521, 613

Richards, G. T. et al. 2003, AJ, 126, 1131

Schick, S. &;, Lloyd, B. 2009, in Presented atthe Society of Photo-Optical InstrumentationEngineers (SPIE) Conference, Vol. 7439, So-ciety of Photo-Optical Instrumentation Engi-neers, (SPIE) Conference Series

Schneps, M. H. & Wright, N. 1980, S&T, 59, 195

Schwalm, M., Barry, M., Perron, G., Sam-path, D., LaMalva, F., Guregian, J., &Crowther, B. 2005, in Presented at the So-ciety of Photo-Optical Instrumentation Engi-neers (SPIE) Conference, Vol. 5904, CryogenicOptical Systems and Instruments XL Edited byHeaney, James B.; Burriesci, Lawrence G. Pro-ceedings of the SPIE, Volume 5904, pp. 178-184(2005)., ed. J. B. Heaney &, L. G. Burriesci,178-184

Sheth, K. et al. 2008, in Spitzer Proposal ID#60007

Skrutskic, M. F. et al. AJ, 131, 1163

Smith, B. J., Price, S. D., &- Baker, R. 1. 2004,ApJS, 154, 673

Stuart, I S. &-- Binzel, R. P. 2004, learns, 170, 295

Sykes, M. V., Lebofisky, L. A., Hunten, D. XL,Low, F. 1986, Science, 232, 1115

Sykes, M. V., Lien, D. J., &- walker, R. G. 1990,learns, 86, 236

Szalay, A. S., Connolly , A. J., &-, Szokoly, G. P.1999, A.1, 117, 68

Tokunaga, A. T. &7 Vacca, W. D. 2005, PASP, 117,1459

Vanden Berk, D. E. et al.2001, AJ, 122, 549

NNTebster, R. L., Francis, P. J., Peterson, B. A.,Drinkwater, M. J., & Masci, F. J. 1995, Nature,31-5, 469

Whipple, F. L. 1955, ApJ, 121, 750

. 1967, SAO Special Report, 239, 1

Whipple, F. L. 1976, in Lecture Notes in Physics,Berlin Sprinuer N/erlag, Vol. 48, InterplanetaryDust and Zodiacal Light,, ed. H. ElsaesserH. Fechtig, 403 415

21

NN'right, E. L. 2007, in AASIDivision for Plan-etary Sciences Meeting Abstracts, Vol. 39,AAS f Division for Planetary Sciences MeetingAbstracts, 483

York, D. G. et al. 2000, AJ, 120, 1579

Zacharias, N. et al. 2010, AJ, 139, 2184

This 2-column preprint was prepared with the AAS LaT)Xmacros v5.2.

22