Pos(Meerkat2016)037 ∗ Alkistis.Poursidou@Port.Ac.Uk Speaker

Home , Intensity mapping

PoS(MeerKAT2016)037 http://pos.sissa.it/ ∗ [email protected] Speaker. Radio detections of the redshiftedveys 21cm can line provide with a new neutral way hydrogen tocosmological (HI) probe questions. the intensity evolving mapping Intensity history sur- mapping of the (IM) Universeing measures and the the tackle underlying fluctuations fundamental matter of distribution, the and HIspectrum therefore signal without can trac- the be need used to to detect reconstructveys individual the can galaxies. matter observe power large This volumes means very thatproviding fast, excellent intensity and redshift mapping give information. sur- us We access explorewith to the the possibility large South of cosmological African performing scales such MeerKAT while radio aArray telescope, survey (SKA). We which also is propose to a use precursor cross-correlationssurvey to between and the the optical MeerKAT Square intensity galaxy mapping Kilometre surveys, inrobust order cosmological to measurements. mitigate Our forecasts important show systematic thattering effects precise measurements and and of lensing produce the signals HI can clus- becosmological made parameters in across the a near wide future. range of These redshift. can be used to constrain HI and ∗ Copyright owned by the author(s) under the terms of the Creative Commons c Attribution-NonCommercial-NoDerivatives 4.0 International License (CC BY-NC-ND 4.0). MeerKAT Science: On the Pathway to25-27 the May, SKA, 2016, Stellenbosch, South Africa Institute of Cosmology & Gravitation, UniversityBurnaby of Road, Portsmouth, Portsmouth, Dennis PO1 Sciama 3FX, Building, UnitedE-mail: Kingdom A. Pourtsidou HI Intensity Mapping with MeerKAT PoS(MeerKAT2016)037 ]. 58 9 we the . 2. [ . It . 0 2 (2.1) 1 ) z 1 z < ( 18 = z . H z 0 < + A. Pourtsidou 67 3 and . ]. Here we will . 0 ]. In Figure 0 11 ' , 11 = ]. For modelling the 7 z HI 10 b , here– in approximately 9 : the L band 0 2 3 1000 MHz). A detailed de- is the HI density, < HI ]. Note that the survey will also f Ω 11 < 4000 deg , = its value today [ CDM expansion history and keep all ) h z Λ where , sky , k A 45 (580 ( . 100 P 1 mK ≡ 2 HI 0 < 0 2 b H ) z 2 z / H b ) ¯ 1 the HI bias. The mean 21cm emission brightness + z T < ( 1 ( , and for the HI bias we use H 4 HI z . h b ) = , and ) ] is an innovative technique that uses neutral hydrogen z z z 8 , ( k , ( 7 HI 00039 , . HI Ω 6 0 P , + 5 180 , 4 is calculated using the formalism described in [ , ) = 3 N z , 00048 ( . P 2 b 0 ¯ , T 1 ' HI Ω is a 64-dish SKA pathfinder on the planned site of SKA1_MID, with 20 dishes 1 1420 MHz), and the UHF band 0 < is the matter power spectrum, and f P < The HI power spectrum can be written as http://www.ska.ac.za/science-engineering/meerkat/ M. Santos et al, “MeerKLASS: MeerKATNote Large that Area the Synoptic survey Survey", will this have volume. to operate in one band only. Recent studies have shown that we can deliver precision cosmology (e.g. Baryonic Acous- Intensity mapping [ MeerKAT As we can see from the above plots, a very good signal-to-noise ratio can be achieved over a 1 2 3 temperature is given by where take interferometric data simultaneously, andgalaxy various evolution, science clusters, galaxy cases HI can emission, be and polarization. achieved: cosmology, Hubble parameter as a function of redshift The noise power spectrum tic Oscillations measurements) using SKA Phase 1 and the IM method [ to map the large-scale structureuse HI of as the a Universe tracer in ofthey three the do underlying dimensions. not dark detect matter Intensity distribution individualmeasuring mapping but, the galaxies; surveys unlike integrated traditional instead, intensity galaxy of they surveys, the treat redshifted the 21cm line 21cm across sky the as sky and a along diffuse redshift. background HI density we use plot the HI power spectrum and measurements errors for two bins centred at already in place. A large sky intensity mapping survey with MeerKAT has been proposed wide range of scales andlogical redshifts. parameters. These measurements As can a be first used approach to we constrain assume HI a and flat cosmo- HI Intensity Mapping with MeerKAT 1. Introduction demonstrate that we can use thecosmological MeerKAT parameters array across to exploit a the wideSKA1_MID IM range technique large of and sky redshift. constrain surveys, HI and This and investigate it will what can will be be help an achieved important us usingmove step maximise the on towards their auto-correlation to scientific HI cross-correlations clustering output. measurements, withsurveys We and are optical will then expected galaxy first to surveys. yield preciseious Cross-correlations and survey-specific robust between challenges cosmological different (foregrounds, constraints. systematics)correlation. They that can alleviate are var- expected to drop out in cross 2. HI intensity mapping with MeerKAT will scan a few thousand5 degrees months on total the observation skylarge time. –we (cosmological) scales, take The and there array are(900 two will frequency operate bands available in single dish mode in order to access scription of the noise properties of such a survey can be found in [ PoS(MeerKAT2016)037 } H , . (2.2) 03 at 1 . A − 0 D , = 8 σ HI A. Pourtsidou 2 HI . Ω 01 Mpc b . 1 / , ], and constrain 3 (L band) and 0 in linear theory, . 3 is in the L-band = 8 9 . HI 0 z σ = 0 HI f Ω k = b = { δ ∆ / z z f ] 1 − 2 - c 0 p 1 , M 005 at [ -binning . ) k k z 0 , k ]. Then the only unknown in the ( ) = P 12 2 ] HI 2 b µ 2 (right). Note that 2 (UHF band) for the survey parameters ] a comprehensive analysis is performed . HI . ) 1 z 1 ]. These are more than one order of mag- 14 Ω ( ( = = ]. For example, the (marginalised) fractional / HI 14 survey using MeerKAT. The black solid line is the z ) z β 2 14 4 3 2 HI 0 0 0

1 1 1

]). In [ b

I H

] c p M K m [ ) ( k

P 1 2 3 2 1 and we have used a 2 (UHF band). Another approach is to assume that [ HI . . 15 0 1 2 HI Ω CDM expansion history, we ﬁnd ( b = = 2 Λ δ b . We can therefore employ the Fisher matrix formalism z ¯ z T ∆ HI 3 (left) and at 3 b . . 0 2 (UHF band) [ 0 ) = HI . 09 at = µ = 1 . 1 Ω - ; z 0 z 0 z = 1 , at = k z ( ) 3 (L band) and 13% at z HI . , HI ] ]. Assuming a 0 1 k Ω P ( − / c 03 at 14 = . HI p , is the redshift space distortion parameter equal to has been measured using the smooth part of the HI signal [ HI z 0 P M [ b Ω ¯ 16 k HI T [ ] the 21cm maps acquired at the Green Bank Telescope were combined with δ the linear growth rate. RSD measurements can be used to break the degeneracy β ) = 17 HI a observations (see Table 2 in [ HI b b ln α and is 3% at d HI ˆ z ) / and z · Ω ]. In [ D ( ( ˆ k 8 2 is in the UHF band. The cyan area represents the measurement errors for a total observation 2 HI / . - ln 18 σ ) 0 1 HI detection in autocorrelation with a 4000 deg 5 months. The width of the bins is = d , Ω f 1 HI = ∼ µ ≡ 17 b 3 (L band) and z 1 3 2 f . Cross correlations between HI intensity mapping and optical galaxy surveys can provide pre- Including redshift space distortions (RSD) the HI power spectrum can be written as HI 0 0 0

1 1 1

] to forecast constraints on the HI abundance and bias along redshift. Using the aforementioned

I H

] c p M K m [ ) ( k

P Ω

3 2 ( = 13 while where time of cise and robust cosmological measurements, aslike they systematics have and the foreground advantage contaminants of that are mitigatingother relevant major for [ issues one type of survey but not for the nitude better than the currentlydamped available Lyman- constraints from galaxy surveys, intensity mapping, and with the HI bias and various cosmological parameters. Forecasts for constraints on we have considered here. 3. Cross-correlations with optical galaxy surveys between predicted power spectrum the mean temperature error on HI Intensity Mapping with MeerKAT cosmological parameters fixed to the Planck 2015 cosmology [ MeerKAT IM survey parameters, weδ find and constraints are derived for multiple redshifts across the L and UHF bands. z using an IM survey with MeerKAT can be found in [ Figure 1: HI power spectrum is the prefactor [ PoS(MeerKAT2016)037 7 . 0 1 - (3.1) 0 = 7 (see 7 . 1 . z 0 0 ). = 1 = 7. The cyan z . − z 0 A. Pourtsidou 06 at . at = 0 ] 8 z CDM expansion σ 17 MeerKAT survey. 8 with a statistical f Λ 01 Mpc ) = . . 2 r 0 0 ] 1 HI − = ∼ b c a correlation coefficient k p z HI r overlap at ∆ ]. M [ 2 Ω 000 deg sky overlap with MeerKAT, 1, k ( 11 , . 2 / 0 ) r –with = z HI r b ∆ , 7. Assuming a we plot the cross-correlation power HI . ) HI b 0 z 2 , Ω HI k ( = ( Ω δ z 2 - rP HI detection in cross-correlation assuming the 0 g 1 b 3 4 0 0 . By doing so we can detect HI clustering and

1 1

5 g I b H ] c p M K m [ ) ( P k b 3 3 ¯ T 2. Modifying our approach as before by including . Figure 2: MeerKAT IM survey (UHF band)troscopic survey and with a 500 deg Euclid-like spec- area represents the measurement errorstion for time a of total 500 observa- hrs ( 1 ) = z , = k and LSST ( z g 4 , HI 4 (UHF band). In Figure . P known we can constrain the growth of structure and expansion 1 14 at . g 0 ] to constrain the quantity < b ] a comprehensive analysis is performed and constraints are derived z 19 14 < ) = and r 7 b . ¯ HI T b overlap with Euclid we find a (marginalised) 4% error on HI 2 Ω 16%. The cross correlation power spectrum can be written as [ ( ], which is assumed to have a (conservative) 500 deg / ∼ ) 20 measured from the galaxy survey, we find r g HI b b HI the galaxy bias. Here we will consider a forthcoming spectroscopic survey with Euclid-like Ω g ( b https://www.darkenergysurvey.org/ https://www.lsst.org/ We have shown that using the δ 4 5 ] for details). Another possibility is to combine MeerKAT and SKA1-MID IM surveys with 14 The redshift overlap is 0 with with the observing time reduced accordingly with respect to the fullspectrum 4 and measurement errors for a bin centredand at RSDs and considering history of the Universe. In [ for multiple redshifts across the LMID and and a UHF 7000deg bands. Note[ that assuming an IM survey withphotometric SKA1- galaxy surveys like DES weak gravitational lensing of the 21cm emission in cross-correlation [ 4. Conclusions specifications [ HI Intensity Mapping with MeerKAT the WiggleZ galaxy survey [ history and MeerKAT pathfinder we can utiliseintensity the mapping method toHI constrain and cosmological parameters across a wide rangewe of can redshift. explore andentific maximise This output the of sci- way futureveys large with sky Phase sur- thermore, 1 we of have proposed thecross-correlations to SKA. of exploit the Fur- 21cm intensity maps with galaxies,the in advantage order of toeffects. reducing have We systematic believe thatarray the can MeerKAT be atcosmology the with the forefront next of generationradio 21cm of telescopes, and weauto- and consider cross-correlation the approaches highly complementary and synergistic. accounting for the possible stochasticityfractional in error the galaxy and HI tracers– at PoS(MeerKAT2016)037 ]. , , (1997) ]. ]. A. Pourtsidou 79 , ]. ]. ]. 1610.04189 ,[ ]. 1407.6366 ]. 1001.4811 Mon.Not.Roy.Astron.Soc. ,[ ,[ , (2012) A129 Phys. Rev. Lett. Mon. Not. Roy. Astron. Soc. 0709.3672 , , ,[ 540 0801.1677 A ground-based 21cm Baryon 0802.1710 21 cm observation of LSS at z 1 ,[ Determination of z 0.8 neutral 21 cm Intensity Mapping ,[ How accurately can 21 cm Cosmology with a SKA HI intensity 1209.0343 HI intensity mapping : a single dish 0910.5007 (2015) 3745 ,[ ,[ (2017) 4251–4260 (2010) 103527 Prospects for clustering and lensing 447 Baryon Acoustic Oscillation Intensity ]. 470 (2008) 091303 D81 Theoretical and observational constraints on the (2008) 161301 ]. (2008) 023529 Astron.Astrophys. 100 , Late-time cosmology with 21cm intensity mapping 100 4 D78 (2010) 164–173 HI and cosmological constraints from intensity mapping, Phys. Rev. Planck 2015 results. XIII. Cosmological parameters Near term measurements with 21 cm intensity mapping: (2013) 1239–1256 , 1405.1452 721 ,[ 434 ]. ]. 1502.01589 ]. Phys.Rev. Phys.Rev.Lett. ,[ , , Phys.Rev.Lett. . , (2015) 21 Mon. Not. Roy. Astron. Soc. , Astrophys.J. Mon. Not. Roy. Astron. Soc. , 803 , Precise Measurement of the Cosmological Power Spectrum With a Dedicated (2016) A13 1304.3712 1509.03286 ,[ ,[ 594 1501.03989 — I acknowledge support by a Dennis Sciama Fellowship at the University of , ]. Measuring cosmological parameters with galaxy surveys . astro-ph/9706198 Astrophys.J. , ,[ Mon. Not. Roy. Astron. Soc collaboration, P. A. R. Ade et al., , (2016) 863–870 (2013) L46–L50 LANCK 1108.1474 hydrogen fluctuations using the 21 cm intensity mapping auto-correlation mapping survey 459 approach Astron. Astrophys. optical, and CMB surveys 434 experiments measurements with forthcoming intensity mapping and optical surveys 3806–3809 neutral hydrogen fraction and BAO at z<2 0902.3091 HI intensity power spectrum Mapping as a Test of Dark Energy 21cm Survey After Reionization tomography constrain cosmology? acoustic oscillation survey Instrument sensitivity and foreground subtraction [ [8] E. Switzer, K. Masui, K. Bandura, L. M. Calin, T. C. Chang et al., [9] P. Bull, P. G. Ferreira, P. Patel and M. G. Santos, [7] R. Battye, I. Browne, C. Dickinson, G. Heron, B. Maffei et al., [2] A. Loeb and S. Wyithe, [3] Y. Mao, M. Tegmark, M. McQuinn, M. Zaldarriaga and O. Zahn, [4] J. B. Peterson, R. Aleksan, R. Ansari, K. Bandura, D. Bond[5] et al., H.-J. Seo, S. Dodelson, J. Marriner, D. Mcginnis, A. Stebbins et al., [6] R. Ansari, J. Campagne, P. Colom, J. L. Goff, C. Magneville et al., [1] T.-C. Chang, U.-L. Pen, J. B. Peterson and P. McDonald, [10] M. G. Santos, P. Bull, D. Alonso, S. Camera, P. G. Ferreira et al., [13] M. Tegmark, [14] A. Pourtsidou, D. Bacon and R. Crittenden, [16] K. W. Masui, P. McDonald and U.-L. Pen, [11] A. Pourtsidou, D. Bacon, R. Crittenden and R. B. Metcalf, [12]P [15] H. Padmanabhan, T. R. Choudhury and A. Refregier, HI Intensity Mapping with MeerKAT Acknowledgments: Portsmouth. I would like toand Mario thank Santos David for Bacon, fruitful Robert collaborations Crittenden, and useful Roy discussions. Maartens, Ben Metcalf, References PoS(MeerKAT2016)037 . , A. Pourtsidou ]. 1606.00180 , 1208.0331 ,[ Measurement of 21 cm brightness (2013) L20 763 5 ]. Astrophys.J. , 1210.2130 ,[ . Cosmology and Fundamental Physics with the Euclid Satellite The WiggleZ Dark Energy Survey: Final data release and cosmological results (2012) 103518 Erasing the Milky Way: new cleaning technique applied to GBT intensity mapping D86 1510.05453 , data Phys. Rev. fluctuations at z 0.8 in cross-correlation [18] L. Wolz et al., [19] D. Parkinson et al., [20] L. Amendola et al., HI Intensity Mapping with MeerKAT [17] K. Masui, E. Switzer, N. Banavar, K. Bandura, C. Blake et al.,