Search for z~7 Ly-alpha emitters with Suprime-Cam at the Subaru Telescope
Pascale Hibon, Nobunari Kashikawa, Christopher Willott, Masanori Iye, Takatoshi Shibuya
aa r X i v : . [ a s t r o - ph . C O ] S e p Search for z ∼ α emitters with Suprime-Cam at the SubaruTelescope. P.Hibon , , N.Kashikawa , , C.Willott , M.Iye , , T.Shibuya , ABSTRACT
We report a search for z=7 Ly α emitters (LAEs) using a custom-madeNarrow-Band filter, centered at 9755˚A with the instrument Suprime-Cam in-stalled at the Subaru telescope. We observed two different fields and obtainedtwo sample of 7 Ly α emitters of which 4 are robust in each field. We arecovering the luminosity range of 9 . − . erg s − in comoving volumes of ∼ × and 4 . × M pc .From this result, we derived possible z ∼ α luminosity functions for the fullsamples and for a subsample of 4 objects in each field. We do not observe, in eachcase, any strong evolution between the z=6.5 and z ∼ α luminosity functions.Spectroscopic confirmation for these candidate samples is required to establish adefinitive measure of the luminosity function at z ∼ Subject headings: cosmology: early universe, galaxies: luminosity function, massfunction, galaxies: distances and redshifts
1. Introduction
Galaxies formed at high redshifts play a key role in understanding how and when thereionization of the universe took place. They also help constrain the physical mechanismsthat drove the formation of the first stars and galaxies in the universe. Over the last decade, Gemini Observatory, La Serena, Chile; email [email protected] School of Earth and Space Exploration, Arizona State University, Tempe, AZ 85287 Optical and Infrared Astronomy Division, National Astronomical Observatory, Mitaka, Tokyo 181-8588,Japan Department of Astronomy, School of Science, Graduate University for Advanced Studies, Mitaka, Tokyo181-8588, Japan Herzberg Institute of Astrophysics, National Research Council,5071 West Saanich Road, Victoria, BCV9E 2E7, Canada ∼
6, which corresponds to ∼ α line, allows us to probe the high-redshift Ly α luminosity function.The search for the redshifted Ly α emission at the longest possible wavelength is complicatedby the presence of OH emission lines within the terrestrial atmosphere. This strong emis-sion line is responsible for the faintness limit at which celestial objects can be detected withground-based telescopes at near-infrared (IR) wavelengths. Fortunately, there are spectralintervals with lower OH-background that allow for a fainter detection limit from the ground-based observations. This is known as the narrow-band (NB) imaging technique.This is one of the most successful methods to detect strong Ly α emission lines of galaxies,since it relies on a specific redshift interval as well as a selected low-sky background window.This filter allows us for maximum detection of light from the Ly α emitters at the centralwavelength, while minimizing the adverse influences of sky emission.Over hundred of z > α LF was derived. This LF shows an apparent deficit comparedto the z=5.7 Ly α LF of Shimasaku et al. (2006), corresponding to a possible luminosity evo-lution from z=5.7 to z=6.5 of L ∗ z =6 . ∼ . − . L ∗ z =5 . . They conclude that the reionizationof the universe is not been completed at 6.5.Ouchi et al. (2010) have now obtained the largest sample to date of 207 LAEs at z=6.6 withthe NB imaging technique. Their derived z=6.6 apparent Ly α LF indicate a decrease from5.7 at 90% confidence level with a more dominant decrease of luminosity evolution ( L ∗ ) thannumber evolution (Φ ∗ ), in agreement with Kashikawa et al. (2006). They claim thereforethat the hydrogen in the IGM is not highly neutral at z=6.6.Hu et al. (2010) have obtained 88 z ∼ ∼ α emitters. Their results on theevolution of the Ly α LF are in agreement with previous works from Malhotra & Rhoads(2004) and Kashikawa et al. (2006).Iye et al. (2006) have first confirmed spectroscopically a z=6.96 LAE. From this result,Ota et al. (2008) assumed an evolution of density from z=5.7 to z ∼
7. They found thatthe IGM is not highly neutral at z ∼ x z =6 . HI ∼ . − .
36 to x z ∼ HI ∼ . − .
64. Hibon et al. (2011) has also performed a search forz ∼ ∼ z ∼ α emitters with the custom-made N B
973 filter ( λ center =9755˚A,FWHM=200˚A). This paper first presents the data (Section 1) and the data reduction proce-dure (Section 2). We then describe the method of selection and contamination of low-redshiftinterlopers for high redshift LAEs in Section 3. We present the final sample of z ∼ α luminosity function at this redshift in Section 4.Throughout this study, we adopt the following cosmological parameters : H = 70 km.s − .M pc − ,Ω m = 0 .
3, Ω Λ = 0 .
2. Observations and Data Reduction2.1. Observations
The data were taken with the Suprime-Cam instrument (the Subaru Prime Focus Cam-era), installed at the 8m Subaru telescope at the National Astronomy Observatory of Japan.This instrument delivers a mosaic of ten 2048 × λ = 8000 − . The total area of the survey, the two fields considered, is 2340 square arc minutes.We observed using the custom-made NB filter centered at 9755˚A with a wavelengthrange of ∆ λ =200˚A ( N B Based on observations obtained with WIRCam, a joint project of CFHT,Taiwan, Korea, Canada, France,at the Canada-France-Hawaii Telescope (CFHT) which is operated by the National Research Council (NRC)of Canada, the Institute National des Sciences de l’Univers of the Centre National de la Recherche Scientifiqueof France, and the University of Hawaii. This work is based in part on data products produced at TERAPIX,the WIRDS (WIRcam Deep Survey) consortium, and the Canadian Astronomy Data Centre. This researchwas supported by a grant from the Agence Nationale de la Recherche ANR-07-BLAN-0228
Using the Data Reduction Software developed for the Subaru Suprime-Cam instrument(Ouchi et al. 2004; Yagi et al. 2002), we performed the following steps: a bias subtraction,a flat fielding, a distortion and atmospheric dispersion correction, a sky subtraction, a badregions masking such as satellite trails and AG probe, the alignment of the individual expo-sures, and the final co-adding step resulting on a stacked image. After eliminating the lowsignal-to-noise regions at the edges of the field of view, we obtain an effective area of 1118arcmin for D33 field, and 1202 arcmin for D41 field.We realize the astrometric calibration on the individual images before the final stacking us-ing the UCAC2 catalog (Zacharias et al. 2004).We need then to adjust the WCS parameters to obtain a more precise alignment. For thispurpose we use the IRAF task msctpeak to interactively align the catalog stars in our imagesand update their headers. We finally obtain an astrometry calibration for each individualimages with a precision of rms ∼ ± . u ∗ , g ′ , r ′ , i ′ , z ′ broad band CFHT-LS data (described inthe next section) is based on the SDSS data for stars with 17 < i ′ <
21 and the Megacam-SDSS color transformation equations of Regnault et al. (2009).The precision obtained in u ∗ , g ′ , r ′ , i ′ , z ′ is between 0.03 and 0.02 mag. As the N B
973 filter is included in the z ′ -bandfilter, we calibrated our N B
973 stacked images, using the AUTO magnitude from the z ′ band SExtractor catalog. We performed the calibration on 1500 non-saturated stars within16 < z ′ <
20. Considering the photometric error on the broad-band calibration, we obtain aphotometric calibration precise at 0.1 magnitude in
N B limitmag.cl available in the SDFRED package. This task constructs a count distribution from randomphotometric apertures on the image. Then it fits a Gaussian profile into this distribution,obtains the sigma number of this profile and calculates the limiting magnitude of the image.We report in Table 1 the limiting magnitude for each band in each field.
Very deep optical imaging data of our observed fields are available through the CFHT-LS. For the purpose of this study, we made use of the T0006 release. These data productsare available from the CADC archive and take form of image stacks in the u ∗ , g ′ , r ′ , i ′ , z ′ filters and of ancillary data such as weight maps, catalogs etc. The spectral curves of thefilter u ∗ , g ′ , r ′ , i ′ , z ′ are similar to the ones of SDSS filters. We also have in hand the Near 5 –Infrared deep Imaging data from the WIRCam Deep Survey (WIRDS) (PIs: C.Willott, J-P.Kneib) (Bielby et al. in preparation).These optical data have been calibrated photometrically using the SDSS photometry andthe NIR data using 2MASS photometry (McCracken et al. 2010). Considering internal andexternal photometric error sources, the uncertainty on the optical and the NIR data pho-tometry is ∼ ∼ geomap/geotran tasks.Our complete set of data is therefore scaled at 0.2 arcsec/pixel and is covering an effectivearea of 1118 and 1202 arcminutes , for D33 and D41 fields, respectively.A summary of the observational data for both fields, D33 and D41, used in this paperis provided in Table 1. Figure 1 shows the transmission curves of the filters correspondingto the multi-band data used in this study.
3. Sample3.1. Catalog generation
We generate the catalogs using the software SExtractor of Bertin & Arnouts (1996).We used the dual-image mode : the first image, the detection image, has been settled asthe combined
N B
973 image, the second image, the measurement image, corresponds to theresampled images from the optical and NIR bands.We choose to detect objects in 7 pixels above a threshold of 1.2 σ . The aperture used for thephotometry is 1 ′′ . Criterion
N B combined ) > σ .We selected objects with a 5 σ detection on the combined N B
973 images.Criterion u ∗ , g ′ , r ′ , i ′ , χ ) < σ .Due to the Gunn- Peterson trough, i.e. the nearly complete absorption of the flux short- 6 – Megacam WIRCamu* g’ r’ i’ z’NB973 J H Ks100000.20.40.60.81
Fig. 1.— Transmission curves of the filters corresponding to the complete set of data usedin this paper. All transmissions are normalized to 100% at maximum. 7 –ward of Lyman α as a result of the large neutral hydrogen column density in the IntergalacticMedium (IGM), we observe spectral discontinuity at redshifts greater than about 6. We aretherefore searching for objects which are not detectable in optical ( u ∗ , g ′ , r ′ , i ′ ) bands. Apossible method is therefore to select objects with less than a 3 σ detection in filters blue-ward of the expected Ly α emission : u ∗ , g ′ , r ′ , i ′ .The color break between the optical and N B
973 filters is high and covers a wide spectralrange. Moreover, for the CFHT-LS, the Terapix data center generated deep χ image com-bining the g ′ , r ′ and i ′ images. We consider therefore this χ image as well as the otherblue-ward optical filters and we select objects with less than a 3 σ detection in the χ image.In summary, we selected objects with less than a 3 σ detection in u ∗ , g ′ , r ′ , i ′ ,and χ bands.Criterion z ′ and N B
973 data.The
N B
973 filter used for this study is included in the broad z band filter. Although we areexpecting an excess of flux in
N B z ′ − N B > . mag . We used the same NB973 filter than Ota et al. (2008) have usedfor the search of robust z=7 LAEs in the SDF field. Their search was successful as it resultedon the spectroscopic confirmation of the only z ∼ z ′ − N B
973 of our candidates with a second color criterion : z ′ − N B > . mag . We showour candidates in the Figure 2 representing z ′ − N B
973 vs
N B
973 for both fields, and the twocolor criteria. The points answering z ′ − N B > . mag but not z ′ − N B > . mag are the points with no z ′ band detection. We used then the 3 σ limit of the z ′ band data forthese objects.Criterion N B − J < N B − H <
N B − Ks < a D33 D41 D33 D41MegaCam u ∗ g ′ r ′ i ′ z ′ N B
973 5 6.3 24.3 24.7WIRCam J H K s a σ magnitude limits in apertures 2 ′′ in diameter for MegaCam, WIRCam and Suprime-Cam.
22 22.5 23 23.5 24 24.5-101234 22 23 24 25-101234
Fig. 2.— Color-Magnitude Diagram z ′ − N B
973 vs
N B
973 for D33 field (left) and D41(right), showing the candidates(red triangles) obtained with the criteria z ′ − N B > . mag (green dashed line). We also compare them to the criterion z ′ − N B > . mag (blue dashed line). The black dots represent all the objects present in each field withSNR( N B combined ) > σ . 9 – Since the narrow band and broad band data were taken at different times, it is possiblefor transient objects to appear in one filter and not another. Transient objects that arebrighter than the narrow band detection limit will be considered as candidates if they arenot visible in the broad band images.Kulkarni & Rau (2006) used a transient rate for SNe, including Type Ia, b,c and Type II, of5 × Gpc yr − . From Cappellaro et al. (1999), we know that the transient rate of Type IaSNe is approximately a factor 3.4 lower than the transient rate including Type Ia, b, c andType II SNe. Therefore, by applying a transient rate of 5 × / i ′ − N B
973 = 0 .
33, and use the results of variability. For their largemagnitude variation (1.1-1.6 mag), Ota et al. (2008) find one object at each epoch giving P ∼ −
5. Down to i ′ ∼ .
5, we found approximately 7500 objects in D33 and 9000 objectsin D41. From this method, we obtain therefore an estimated number of variable objects of ∼ ∼ Considering Figure 9 of Hawley et al. (2002) presenting several color diagrams for L-,T- dwarfs detected in SDSS, we know that M-, L-, T- dwarfs have z ′ − J > z ′ − J >
2. Therefore, although our J-band is shallower than our NB973 limitingmagnitude, the criterion
N B − J <
We estimated the lower value on equivalent width a line emitter, with a flat continuum in f ν , would require to be selected with our criteria using the formula from Rhoads & Malhotra(2001): EW min ∼ (cid:18) f NB f BB (cid:19) ∆ λ NB = (cid:20) σ NB σBB − (cid:21) ∆ λ NB (1)with f NB and f BB the flux in N B
973 and g ′ band respectively, ∆ λ NB the width of the N B σ NB and σ BB , the flux uncertainties in N B
973 and g ′ band respectively. We obtainedtherefore an EW D min ∼ A and EW D min ∼ . A . Foreground line emitters would thenrequire an equivalent width EW D min ≥ A and EW D min ≥ . A to contaminate our Ly α selection sample.1- H α at z ∼ α emitters at z ∼ EW < A . From Equation 1, the lower limit onthe H α emitters equivalent width in our survey is 840˚A in D33 field and 714.2˚A in D41 field.H α emitters at z ∼ iii ] at z ∼ iii ]emitters at z ∼ iii ] emitters at z ∼ K s band than in J band and with typical J bandmagnitude ∼
23 (AB). We will have therefore observed these objects in our very deep NIRdata. Considering the lower limit on equivalent width that line emitters will have to con-taminate our candidate sample and the NIR magnitudes of typical z ∼ iii ] emitters,the contamination by these foreground emitters is therefore ruled out.Cardamone et al. (2009) found low redshift [O iii ] emitters with very high observed EWs,reaching values until 1500˚A. However, these emitters have very bright magnitudes in opticalbands : 18 ≤ g ′ , r ′ , i ′ , z ′ ≤
21 and blue spectra. They would be therefore easily detected bythe CFHT-LS data. These emitters are identified as Pea galaxies : luminous blue compactgalaxies. 11 –3- [O ii ] at z ∼ ii ] emitters at z ∼ z ′ − K s vs K s . z ∼ ii ] emittershave K s ǫ [20; 23], and z ′ − K s >
0. As our narrow-band filter is included in the z ′ band filter,as seen in Figure 1, we can assume that z ′ − K s >
0. This colour does not agree with ourCriterion z ′ − K s vs J − K s . Most of their [O ii ] emitters have not only a z ′ − K s > J − K s >
0. Inthe case of our sample, a deeper J band image would help us eliminate the [O ii ] emittersas a possible contaminant to verify that our candidates have J − K s <
0, as deduced fromCriterion K s band data do not cover entirely both fields. Although the contamination byz ∼ ii ] emitters is very unlikely due to the K s magnitude range, we cannot completelyrule it out. • Balmer break galaxiesBalmer break galaxies at z ∼ z ′ − J for z ∼ z ′ − J >
0. Daddi et al. (2004)shows that z ≥ z ′ − Ks > .
5. Moreover, in the case ofcontinuum objects, we can assume that z ′ − Ks ∼ N B − Ks . During the selectionof our candidate sample, we applied Criterion z ′ − J <
0. If the candidates wereBalmer break galaxies, we should have detected them in J with a brighter magnitudein J than in z ′ . • Extremely Red Objects (EROs)We add a criterion to avoid selecting extremely red objects (Cimatti et al. 2002).We constrain this contamination by applying
N B − J <
0, and also verifying
N B − H <
N B − Ks < Ks band than in NB973.These objects cannot contaminate our candidate samples. However, as our NIR data do notcover our entire fields, we cannot completely rule out this contamination. 12 – In order to estimate the number of false detections that could pass our selection criteria,we create an inverse
N B
973 combined image by multiplying this
N B
973 image by -1. Wethen applied the first criteria used for the candidate selection : SNR(
N B combined ) > σ .None detection meets this condition. We found out therefore that our candidate samples arenot contaminated by false detections. After this analysis, we can conclude that it is unlikely that our z ∼ α at ∼ iii ] at ∼ ii ]at ∼ We obtain a final sample in each observed field. The D33 final sample contains 7z ∼ N B
N B ∼ N B
N B EW ) derived from thephotometric data and report them in Table 2 and Table 3. It is interesting to observe thatthe candidates from the D41 field, deeper than the D33 field (see Table 1) are automaticallyfainter than the D33 ones. The observational conditions for both fields show a systematicdifference of at least 0.1 arcsec in seeing. EW rest = (cid:18) f NB ∆ λ z ′ − f z ′ ∆ λ NB f z ′ − f NB (cid:19) ×
11 + z (2)where f NB is the observed flux in the narrow-band combined image, f z ′ is the observedflux in the z ′ broad-band image, ∆ λ NB and ∆ λ z ′ are the width of the N B
973 filter (200˚A)and the z ′ band filter (928˚A) respectively.Following Ota et al. (2008), we assume that 77% of the NB flux comes from the Ly α
13 –line : f Lyα ∼ . f NB . We use the detection limit in z ′ band, to derive in turn lower EW limits.In the D33 field, LAE N B
973 magnitude corresponding to 94% completenessand the LAE
N B
973 magnitude to 89% completeness. In the D41 field, LAE
N B
973 magnitude corresponding to 96% completeness and the LAE
N B
973 magni-tude to ∼
89% completeness. To estimate the completeness, we added 200 artificial star-likeobjects per bin of 0.1 magnitude in blank regions of the stacked images. We then run SEx-tractor on the image with the same parameters as previously used for object detection. Thisprocedure has been repeated 20 times. The average count on 20 times of the number of artifi-cial stars retrieved in each magnitude bin provides a direct measure of the completeness limit.
4. Discussion4.1. Variance
Two sources of variance can affect a high redshift study : the Poisson variance and thecosmic variance, in other words the fluctuations in the large scale distribution of the galaxies.In order to estimate the value of this cosmic variance we used the on-line calculator fromthe model of Trenti & Stiavelli (2008). We obtain then a value of 37% and 36% of cosmicvariance in D33 and D41 fields respectively. http://casa.colorado.edu/ trenti/CosmicVariance.html Table 2: Table of the z ∼ .
96 LAE candidates for the D33 field.
Name
N B
973 Error SNR (
N B z ′ Error SNR ( z ′ ) EW a (˚A)LAE > > > > > > > > > > a In the rest-frame 14 –Fig. 3.— Thumbnail images of the D33 candidates listed in Table 2. Each window is15 ′′ × ′′ . Objects names and passbands are located above and to the left of the thumbnails,respectively. 15 –Fig. 4.— Thumbnail images of the D41 candidates listed in Table 2. Each window is15 ′′ × ′′ . Objects names and passbands are located above and to the left of the thumbnails,respectively. 16 –Considering the limited number of objects in our sample and the large comoving volumeof our survey, our results are as limited by Poisson noise – ∼
38% for the 7 objects of D33field and for the 7 objects of D41 field – than by clustering. Therefore, in the case wherethe full sample in each observed field (7 candidates in each field) is taken into account (seeParagraph 4.2.1), the error bars represent the Poisson variance and the cosmic variance. Inthe two other cases (see Paragraphs 4.2.2 and 4.2.3), our results are more limited by thePoisson noise than by clustering. The error bars are therefore representing the Poisson noiseonly.For D33 field, we obtain a total fractional error on number counts of 0.65. As we obtain7 objects in D33, we should expect between 4.5 and 9.4 objects in D41, if the field-to-fieldvariation is within 1 σ . As we found 7 objects D41 field, we can therefore confirm that thefield-to-field variation is within 1 σ . This result was expected as none of our field are knownto be located in overdense or underdense large-scale structure at this redshift. Following Ota et al. (2008), we assume that on average 77% of the narrow-band fluxcomes from the Ly α line. We therefore apply the same correction factor during the conver-sion from N B
973 to Ly α fluxes.We fit to the Ly α luminosity function of these z ∼ L ), given by Φ( L )d L = Φ ∗ (cid:18) LL ∗ (cid:19) α exp (cid:18) − LL ∗ (cid:19) d LL ∗ (3)Table 3: Table of the z ∼ .
96 LAE candidates for the D41 field.
Name
N B
973 Error SNR (
N B z ′ Error
SN R ( z ′ ) EW a (˚A)LAE > > > > > > > > > > > > a In the rest-frame 17 –in order to compare with previous high redshift works (Hibon et al. 2010; Ouchi et al. 2010,2008; Ota et al. 2008). The error bars shown in Figure 5 represent the total errors (cumu-lative Poisson errors and cosmic variance) as explained in the previous section 4.1. In thecase of Figure 6 and 7, the results are more limited by the Poisson noise than by the cosmicvariance. The error bars represent then only the cumulative Poisson errors. Considering thelow number of candidates in our sample, we choose to fit two out of three of the Schechterfunction parameters. Following Hibon et al. (2010) and Ouchi et al. (2008), we set the faintend slope of the luminosity, α , to α = − .
5, and determine Φ ∗ and L ∗ by χ minimization.We corrected the data from the completeness by number weighting.We decided to study the following different cases : • The full sample of each field is real. • The samples are contaminated by 50% . We then consider only the 4 brightest objectsof each field as real (referred later as bright samples). This is justified by the fact thatthe brightest candidates are the most robust ones and we are focusing on building thebright end of the z ∼ α luminosity function. • We chose to derive a common Ly α LF for the D33 and D41 LAEs brightest candi-dates. We have therefore a sample of 8 bright objects in a total area of 2320arcmin ,corresponding to the sum of the D33 and D41 effective areas.The best-fit Schechter LF parameters for each case are summarized in Table 4. From the z ∼ α LF derived from the full sample of the D41 field, seen in Figure 5as the blue line and the candidates corresponding represented as filled squares, we observe apossible but weak evolution in luminosity from the z ∼ α LF from Ouchi et al. (2010)(dot-long dashed line in Figure 5), but no possible evolution from the z ∼ α LF fromKashikawa et al. (2011) (dot-short dashed line in Figure 5). The z ∼ α LF from thecomplete sample of the D33 field, shown as the red line and the filled triangles in Figure 5shows a weaker evolution in luminosity from the z ∼ α LF. However, looking at theerror bars from D33 and D41 field candidates, both weak evolutions in luminosity, becomevery discussable. From these possible z ∼ α LFs, if all the candidates are real for oneof the observed field, we cannot therefore conclude about a possible evolution in luminosityfrom z ∼ ∼
7. 18 –Fig. 5.— Best-fit Schechter function for the cumulative z ∼ α luminosity function de-rived from the D33 full sample in red and from the D41 full sample in blue. Our D33candidates are represented as triangles and the D41 ones as filled squares, the z=6.96 LAEfrom Iye et al. (2006) is the empty square and Ota et al. (2008) candidates are pentagons.Also represented here is the z=6.5 Ly α LF from Kashikawa et al. (2011) (dot-short dashedline), the z=6.5 Ly α LF from Ouchi et al. (2010) (dot-long dashed line) and the z=5.7 Ly α LF from Kashikawa et al. (2011) (dotted line). The cumulative Poisson errors and the cosmicvariance are taken in account in the vertical error bars. 19 –Table 4: Best-fit Schechter LF parameters for α = − . L ∗ (erg s − )) log(Φ ∗ (Mpc − ))6.96 (1) . +0 . − . − . +0 . − . (2) . +0 . − . − . +0 . − . (3) . +0 . − . − . +0 . − . (4) . +0 . − . − . +0 . − . (5) . +0 . − . − . +0 . − . (6) . +0 . − . − . +0 . − . (7) . +0 . − . − . +0 . − . (8) . +0 . − . − . +0 . − . References. (1, 2) derived from our D33 and D41 full samples resp.; (3, 4) derived from thebrightest candidates in each field; (5) derived from a common sample between D33 andD41 fields; (6)Ota et al. (2008); (7) Ouchi et al. (2010); (8) Ouchi et al. (2008)
We derived Ly α luminosity functions for the two different z ∼ ∼ ∼ By fitting a unique Schechter LF for both samples (as seen in Figure 7), we assumethat the brightest objects from both samples are real z ∼ ∼ α LF from Ouchi et al. (2010) but notthe z ∼ α LF from Kashikawa et al. (2011) . Depending on the z ∼ α LF weconsidered, the best-fit Schechter derived LF, seen as the red line on Figure 7, could agreewith the observed evolution between z ∼ ∼ ∼ ∼
7. 20 –Fig. 6.— Best-fit Schechter function for the cumulative z ∼ α luminosity function derivedfrom the D33 bright sample in red and from the D4 1 bright sample in blue . Our D33candidates are represented as triangles and the D41 ones as filled squares, the z=6.96 LAEfrom Iye et al. (2006) is the empty square. Also represented here is the z=6.5 Ly α LF fromKashikawa et al. (2011) (dot-short-dashed line), the z=6.5 Ly α LF from Ouchi et al. (2010)(dot-long-dashed line) and the z=5.7 Ly α LF from Kashikawa et al. (2011) (dotted line).The vertical error bars represent the cumulative Poisson errors, as the bright samples aremore limited by the Poisson noise than by clustering. 21 –Fig. 7.— Best-fit Schechter function for the cumulative z ∼ α luminosity function derivedfrom a common sample composed by the D33 bright sample in blue and from the D41 brightsample in red. Our D33 and D41 candidates are represented as triangles, the z=6.96 LAEfrom Iye et al. (2006) is the empty square. Also represented here is the z=6.5 Ly α LF fromKashikawa et al. (2011) (dot-short-dashed line), the z=6.5 Ly α LF from Ouchi et al. (2010)(dot-long-dashed line) and the z=5.7 Ly α LF from Kashikawa et al. (2011) (dotted line).The vertical error bars represent the cumulative Poisson errors, as the common sample ismore limited by the Poisson noise than by clustering. 22 –
We obtain two different best-fit Schechter functions for the z ∼ ∼ ∼ ∼ ∼ ∼
7, as seen in Figure 6.By producing z ∼ ∼ ∼ α emitters which has lead to thefirst spectroscopically confirmed z ∼ N B
973 (∆ λ = 200˚ A , λ c = 9755˚ A ) and reaches a 50% completenessof N B σ ). IOK-1 has a flux of 2 × − erg s − cm − . We therefore observea wider but shallower area. This strategy is justified by our willing to constrain only thebright-end of the z ∼ α LF.The main prediction of Le Delliou et al. (2006) is a likely moderate decline of the brightend of the LF of LAEs from z=6.5 to z ∼ α LF derived from LAEs evolves as the rest-frame UVLFs obtainedfrom LBGs. From Yoshida et al. (2006), they estimated therefore a possible z ∼ α LF with a pure luminosity evolution of L ∗ z =7 = 0 . L ∗ z =5 . , with L ∗ z =5 . = 1 . . erg s − (Shimasaku et al. 2006). This inferred z ∼ α LF does not agree with our photometriccandidates sample. It confirms their idea that, if LAEs are strongly related to LBGs, theneutral hydrogen fraction in the IGM is possibly higher at z ∼ ∼ α lines of high redshift LAEs.If none of our object is a real z ∼ ∼ α LF, which helps for constraining better these models. 23 –
5. Conclusions
We observed 0.64 square degree of the WIRDS/CFHT-LS fields with the
N B
973 filter,corresponding to targeting the Ly α line at z ∼
7. After applying our selection criteria andverifying that our selection was not contaminated by low-redshift emitters, we obtained asample of seven z ∼ ∼ α Luminosity Functions. We did not find a significant evolutioneither in luminosity nor in density from z=6.5 to z ∼ ∼ α Luminosity Function. Theexact evolution of the LF beyond redshift 6.5 remains therefore a matter to debate. More-over, a single galaxy at z ∼ α LF. We need to increase the number of independent fields as well as the numberof LAEs at z=7. Once the bright end of z ∼ α LF is determined and possible evolutionfrom z=6.5 is derived, it will become easier to assess whether the 1.06 µ m and the 1.19 µ mNB filters could reveal z ∼ REFERENCES
Bertin, E. & Arnouts, S. 1996, A&AS, 117, 393Cappellaro, E., Evans, R., & Turatto, M. 1999, A&A, 351, 459Cardamone, C., Schawinski, K., Sarzi, M., Bamford, S. P., Bennert, N., Urry, C. M., Lintott,C., Keel, W. C., Parejko, J., Nichol, R. C., Thomas, D., Andreescu, D., Murray, P.,Raddick, M. J., Slosar, A., Szalay, A., & Vandenberg, J. 2009, MNRAS, 399, 1191Cimatti, A., Daddi, E., Mignoli, M., Pozzetti, L., Renzini, A., Zamorani, G., Broadhurst,T., Fontana, A., Saracco, P., Poli, F., Cristiani, S., D’Odorico, S., Giallongo, E.,Gilmozzi, R., & Menci, N. 2002, A&A, 381, L68Daddi, E., Cimatti, A., Renzini, A., Fontana, A., Mignoli, M., Pozzetti, L., Tozzi, P., &Zamorani, G. 2004, ApJ, 617, 746Delorme, P., Albert, L., Forveille, T., Artigau, E., Delfosse, X., Reyl´e, C., Willott, C. J.,Bertin, E., Wilkins, S. M., Allard, F., & Arzoumanian, D. 2010, A&A, 518, A39+Hawley, S. L., Covey, K. R., Knapp, G. R., Golimowski, D. A., Fan, X., Anderson, S. F.,Gunn, J. E., Harris, H. C., Ivezi´c, ˇZ., Long, G. M., Lupton, R. H., McGehee, P. M., 24 –Narayanan, V., Peng, E., Schlegel, D., Schneider, D. P., Spahn, E. Y., Strauss, M. A.,Szkody, P., Tsvetanov, Z., Walkowicz, L. M., Brinkmann, J., Harvanek, M., Hennessy,G. S., Kleinman, S. J., Krzesinski, J., Long, D., Neilsen, E. H., Newman, P. R., Nitta,A., Snedden, S. A., & York, D. G. 2002, AJ, 123, 3409Hayashi, M., Kodama, T., Koyama, Y., Tadaki, K.-I., & Tanaka, I. 2011, MNRAS, 415, 2670Hibon, P., Cuby, J., Willis, J., Cl´ement, B., Lidman, C., Arnouts, S., Kneib, J., Willott,C. J., Marmo, C., & McCracken, H. 2010, A&A, 515, A97+Hibon, P., Malhotra, S., Rhoads, J., & Willott, C. J. 2011, ApJHu, E. M., Cowie, L. L., Barger, A. J., Capak, P., Kakazu, Y., & Trouille, L. 2010, ApJ,725, 394Hu, E. M., Cowie, L. L., McMahon, R. G., Capak, P., Iwamuro, F., Kneib, J., Maihara, T.,& Motohara, K. 2002, ApJ, 568, L75Iye, M., Ota, K., Kashikawa, N., Furusawa, H., Hashimoto, T., Hattori, T., Matsuda, Y.,Morokuma, T., Ouchi, M., & Shimasaku, K. 2006, Nature, 443, 186Kashikawa, N., Shimasaku, K., Malkan, M. A., Doi, M., Matsuda, Y., Ouchi, M., Taniguchi,Y., Ly, C., Nagao, T., Iye, M., Motohara, K., Murayama, T., Murozono, K., Nariai,K., Ohta, K., Okamura, S., Sasaki, T., Shioya, Y., & Umemura, M. 2006, ApJ, 648, 7Kashikawa, N., Shimasaku, K., Matsuda, Y., Egami, E., Jiang, L., Nagao, T., Ouchi, M.,Malkan, M. A., Hattori, T., Ota, K., Taniguchi, Y., Okamura, S., Ly, C., Iye, M.,Furusawa, H., Shioya, Y., Shibuya, T., Ishizaki, Y., & Toshikawa, J. 2011, ApJ, 734,119Kulkarni, S. R. & Rau, A. 2006, ApJ, 644, L63Le Delliou, M., Lacey, C. G., Baugh, C. M., & Morris, S. L. 2006, MNRAS, 365, 712Malhotra, S. & Rhoads, J. E. 2004, ApJ, 617, L5McCracken, H. J., Capak, P., Salvato, M., Aussel, H., Thompson, D., Daddi, E., Sanders,D. B., Kneib, J., Willott, C. J., Mancini, C., Renzini, A., Cook, R., Le F`evre, O.,Ilbert, O., Kartaltepe, J., Koekemoer, A. M., Mellier, Y., Murayama, T., Scoville,N. Z., Shioya, Y., & Tanaguchi, Y. 2010, ApJ, 708, 202Ota, K., Iye, M., Kashikawa, N., Shimasaku, K., Kobayashi, M., Totani, T., Nagashima, M.,Morokuma, T., Furusawa, H., Hattori, T., Matsuda, Y., Hashimoto, T., & Ouchi, M.2008, ApJ, 677, 12 25 –Ouchi, M., Shimasaku, K., Akiyama, M., Simpson, C., Saito, T., Ueda, Y., Furusawa,H., Sekiguchi, K., Yamada, T., Kodama, T., Kashikawa, N., Okamura, S., Iye, M.,Takata, T., Yoshida, M., & Yoshida, M. 2008, ApJS, 176, 301Ouchi, M., Shimasaku, K., Furusawa, H., SAITO, T., Yoshida, M., Akiyama, M., Ono, Y.,Yamada, T., Ota, K., Kashikawa, N., Iye, M., Kodama, T., Okamura, S., Simpson,C., & Yoshida, M. 2010, ArXiv e-printsOuchi, M., Shimasaku, K., Okamura, S., Furusawa, H., Kashikawa, N., Ota, K., Doi, M.,Hamabe, M., Kimura, M., Komiyama, Y., Miyazaki, M., Miyazaki, S., Nakata, F.,Sekiguchi, M., Yagi, M., & Yasuda, N. 2004, ApJ, 611, 660Regnault, N., Conley, A., Guy, J., Sullivan, M., Cuillandre, J., Astier, P., Balland, C., Basa,S., Carlberg, R. G., Fouchez, D., Hardin, D., Hook, I. M., Howell, D. A., Pain, R.,Perrett, K., & Pritchet, C. J. 2009, A&A, 506, 999Rhoads, J. E. & Malhotra, S. 2001, ApJ, 563, L5Salzer, J. J., Jangren, A., Gronwall, C., Werk, J. K., Chomiuk, L. B., Caperton, K. A.,Melbourne, J., & McKinstry, K. 2005, AJ, 130, 2584Shimasaku, K., Kashikawa, N., Doi, M., Ly, C., Malkan, M. A., Matsuda, Y., Ouchi, M.,Hayashino, T., Iye, M., Motohara, K., Murayama, T., Nagao, T., Ohta, K., Okamura,S., Sasaki, T., Shioya, Y., & Taniguchi, Y. 2006, PASJ, 58, 313Spergel, D. N., Bean, R., Dor´e, O., Nolta, M. R., Bennett, C. L., Dunkley, J., Hinshaw,G., Jarosik, N., Komatsu, E., Page, L., Peiris, H. V., Verde, L., Halpern, M., Hill,R. S., Kogut, A., Limon, M., Meyer, S. S., Odegard, N., Tucker, G. S., Weiland, J. L.,Wollack, E., & Wright, E. L. 2007, ApJS, 170, 377Trenti, M. & Stiavelli, M. 2008, ApJ, 676, 767Xia, L., Malhotra, S., Rhoads, J., Pirzkal, N., Zheng, Z., Meurer, G., Straughn, A., Grogin,N., & Floyd, D. 2010, ArXiv e-printsYagi, M., Kashikawa, N., Sekiguchi, M., Doi, M., Yasuda, N., Shimasaku, K., & Okamura,S. 2002, AJ, 123, 66Yoshida, M., Shimasaku, K., Kashikawa, N., Ouchi, M., Okamura, S., Ajiki, M., Akiyama,M., Ando, H., Aoki, K., Doi, M., Furusawa, H., Hayashino, T., Iwamuro, F., Iye, M.,Karoji, H., Kobayashi, N., Kodaira, K., Kodama, T., Komiyama, Y., Malkan, M. A.,Matsuda, Y., Miyazaki, S., Mizumoto, Y., Morokuma, T., Motohara, K., Murayama, 26 –T., Nagao, T., Nariai, K., Ohta, K., Sasaki, T., Sato, Y., Sekiguchi, K., Shioya, Y.,Tamura, H., Taniguchi, Y., Umemura, M., Yamada, T., & Yasuda, N. 2006, ApJ,653, 988Zacharias, N., Urban, S. E., Zacharias, M. I., Wycoff, G. L., Hall, D. M., Monet, D. G., &Rafferty, T. J. 2004, AJ, 127, 3043