PUBLISHED VERSION Aab, A.;..; Bellido Caceres, Jose Alfredo; BenZvi, S.;..; Clay, Roger William; Cocciolo, G.;..; Cooper, Matthew John; Coutu, S.;..; Dawson, Bruce Robert; deAlmeida, R. M.;..; Grubb, Trent Daniel; Guardincerri, Y.;..; Harrison, Thomas Alan; Harton, J. L.;..; Herve, Alexander Edward; Hill, Gary Colin; Hojvat, C....; Malacari, Maximus Dominic; Maldera, S.;..; Saffi, Steven Jason; Saftoiu, A.;..; Sorokin, Jennifer Sally; Spinka, H.;..; Wahrlich, Philip Shane; ... et al.; Pierre Auger Collaboration Probing the radio emission from air showers with polarization measurements Physical Review D, 2014; 89(5):052002 ©2014 American Physical Society http://journals.aps.org/prd/abstract/10.1103/PhysRevD.89.052002 PERMISSIONS http://publish.aps.org/authors/transfer-of-copyright-agreement http://link.aps.org/doi/10.1103/PhysRevD.62.093023 “The author(s), and in the case of a Work Made For Hire, as defined in the U.S. Copyright Act, 17 U.S.C. §101, the employer named [below], shall have the following rights (the “Author Rights”): [...] 3. The right to use all or part of the Article, including the APS-prepared version without revision or modification, on the author(s)’ web home page or employer’s website and to make copies of all or part of the Article, including the APS-prepared version without revision or modification, for the author(s)’ and/or the employer’s use for educational or research purposes.” 20th May 2014 http://hdl.handle.net/2440/82998 PHYSICAL REVIEW D 89, 052002 (2014) Probing the radio emission from air showers with polarization measurements A. Aab,1 P. Abreu,2 M. Aglietta,3 M. Ahlers,4 E.J. Ahn,5 I.F.M. Albuquerque,6 I. Allekotte,7 J. Allen,8 P. Allison,9 A. Almela,10,11 J. Alvarez Castillo,12 J. Alvarez-Muñiz,13 R. Alves Batista,14 M. Ambrosio,15 A. Aminaei,16 L. Anchordoqui,17 S. Andringa,2 T. Antičić,18 C. Aramo,15 F. Arqueros,19 H. Asorey,7 P. Assis,2 J. Aublin,20 M. Ave,13 M. Avenier,21 G. Avila,22 A.M. Badescu,23 K.B. Barber,24 R. Bardenet,25 J. Bäuml,26 C. Baus,27 J.J. Beatty,28 K.H.Becker,29J.A.Bellido,24S.BenZvi,30C.Berat,21X.Bertou,7P.L.Biermann,31P.Billoir,20F.Blanco,19M.Blanco,20 C.Bleve,29H.Blümer,27,26M.Boháčová,32D.Boncioli,33C.Bonifazi,34R.Bonino,3N.Borodai,35J.Brack,36I.Brancus,37 P. Brogueira,2 W.C. Brown,38 P. Buchholz,1 A. Bueno,39 M. Buscemi,15 K.S. Caballero-Mora,13,40 B. Caccianiga,41 L. Caccianiga,20 M. Candusso,42 L. Caramete,31 R. Caruso,43 A. Castellina,3 G. Cataldi,44 L. Cazon,2 R. Cester,45 S.H. Cheng,40 A. Chiavassa,3 J.A. Chinellato,46 J. Chudoba,32 M. Cilmo,15 R.W. Clay,24 G. Cocciolo,44 R. Colalillo,15 L. Collica,41 M.R. Coluccia,44 R. Conceição,2 F. Contreras,47 M.J. Cooper,24 S. Coutu,40 C.E. Covault,48 A. Criss,40 J.Cronin,49A.Curutiu,31R.Dallier,50,51B.Daniel,46S.Dasso,52,53K.Daumiller,26B.R.Dawson,24R.M.deAlmeida,54 M.DeDomenico,43S.J.deJong,16,55G.DeLaVega,56W.J.M.deMelloJunior,46J.R.T.deMelloNeto,34I.DeMitri,44 V.deSouza,57K.D.deVries,58L.delPeral,59O.Deligny,60H.Dembinski,26N.Dhital,61C.DiGiulio,42A.DiMatteo,62 J.C.Diaz,63M.L.DíazCastro,64 P.N.Diep,65F.Diogo,2C.Dobrigkeit,46 W.Docters,58 J.C.D’Olivo,12 P.N.Dong,65,60 A. Dorofeev,36 J.C. dos Anjos,64 M.T. Dova,66 J. Ebr,32 R. Engel,26 M. Erdmann,67 C.O. Escobar,5,46 J. Espadanal,2 A. Etchegoyen,11,10 P. Facal San Luis,49 H. Falcke,16,68,55 K. Fang,49 G. Farrar,69 A.C. Fauth,46 N. Fazzini,70 A.P.Ferguson,48B.Fick,63J.M.Figueira,11,26A.Filevich,11A.Filipčič,71,72N.Foerster,1B.D.Fox,73C.E.Fracchiolla,36 E.D. Fraenkel,58 O. Fratu,23 U. Fröhlich,1 B. Fuchs,27 R. Gaior,20 R.F. Gamarra,11 S. Gambetta,74 B. García,56 S.T.GarciaRoca,13D.Garcia-Gamez,25D.Garcia-Pinto,19G.Garilli,43A.GasconBravo,39H.Gemmeke,75P.L.Ghia,20 M. Giammarchi,41 M. Giller,76 J. Gitto,56 C. Glaser,67 H. Glass,70 F. Gomez Albarracin,66 M. Gómez Berisso,7 P.F. Gómez Vitale,22 P. Gonçalves,2 J.G. Gonzalez,27 B. Gookin,36 A. Gorgi,3 P. Gorham,73 P. Gouffon,6 S. Grebe,16,55 N. Griffith,77 A.F. Grillo,33 T.D. Grubb,24 Y. Guardincerri,53 F. Guarino,15 G.P. Guedes,78 P. Hansen,66 D. Harari,7 T.A. Harrison,24 J.L. Harton,36 A. Haungs,26 T. Hebbeker,67 D. Heck,26 A.E. Herve,24 G.C. Hill,24 C. Hojvat,70 N. Hollon,49 E. Holt,26 P. Homola,1,35 J.R. Hörandel,16,55 P. Horvath,79 M. Hrabovský,79,32 D. Huber,27 T. Huege,26 A. Insolia,43 P.G. Isar,80 S. Jansen,16,55 C. Jarne,66 M. Josebachuili,11,26 K. Kadija,18 O. Kambeitz,27 K.H. Kampert,29 P. Karhan,81 P. Kasper,70 I. Katkov,27 B. Kégl,25 B. Keilhauer,26 A. Keivani,82 E. Kemp,46 R.M. Kieckhafer,63 H.O.Klages,26M.Kleifges,75J.Kleinfeller,47,26J.Knapp,83R.Krause,67N.Krohm,29O.Krömer,75D.Kruppke-Hansen,29 D. Kuempel,67 N. Kunka,75 G. La Rosa,84 D. LaHurd,48 L. Latronico,3 R. Lauer,85 M. Lauscher,67 P. Lautridou,50 S. Le Coz,21 M.S.A.B. Leão,86 D. Lebrun,21 P. Lebrun,70 M.A. Leigui de Oliveira,87 A. Letessier-Selvon,20 I. Lhenry-Yvon,60 K. Link,27 R. López,88 A. Lopez Agüera,13 K. Louedec,21 J. LozanoBahilo,39 L. Lu,29,83 A. Lucero,11 M. Ludwig,27 H. Lyberis,34 M.C. Maccarone,84 M. Malacari,24 S. Maldera,3 J. Maller,50 D. Mandat,32 P. Mantsch,70 A.G. Mariazzi,66 V. Marin,50 I.C. Mariş,20 H.R. Marquez Falcon,89 G. Marsella,44 D. Martello,44 L. Martin,50,51 H.Martinez,90O.MartínezBravo,88D.Martraire,60J.J.MasíasMeza,53H.J.Mathes,26J.Matthews,82J.A.J.Matthews,85 G.Matthiae,42 D.Maurel,26 D.Maurizio,64 E.Mayotte,91 P.O.Mazur,70 C. Medina,91 G.Medina-Tanco,12 M.Melissas,27 D. Melo,11 E. Menichetti,45 A. Menshikov,75 S. Messina,58 R. Meyhandan,73 S. Mićanović,18 M.I. Micheletti,92 L. Middendorf,67 I.A. Minaya,19 L. Miramonti,41 B. Mitrica,37 L. Molina-Bueno,39 S. Mollerach,7 M. Monasor,49 D. Monnier Ragaigne,25 F. Montanet,21 B. Morales,12 C. Morello,3 J.C. Moreno,66 M. Mostafá,36 C.A. Moura,87 M.A. Muller,46 G. Müller,67 M. Münchmeyer,20 R. Mussa,45 G. Navarra,3 J.L. Navarro,39 S. Navas,39 P. Necesal,32 L.Nellen,12A.Nelles,16,55J.Neuser,29P.T.Nhung,65M.Niechciol,1L.Niemietz,29T.Niggemann,67D.Nitz,63D.Nosek,81 L. Nožka,32 J. Oehlschläger,26 A. Olinto,49 M. Oliveira,2 M. Ortiz,19 N. Pacheco,59 D. Pakk Selmi-Dei,46 M. Palatka,32 J.Pallotta,93N.Palmieri,27G.Parente,13A.Parra,13S.Pastor,94T.Paul,95,96M.Pech,32J.Pe¸kala,35R.Pelayo,88I.M.Pepe,97 L. Perrone,44 R. Pesce,74 E. Petermann,98 S. Petrera,62 A. Petrolini,74 Y. Petrov,36 R. Piegaia,53 T. Pierog,26 P. Pieroni,53 M. Pimenta,2 V. Pirronello,43 M. Platino,11 M. Plum,67 M. Pontz,1 A. Porcelli,26 T. Preda,80 P. Privitera,49 M. Prouza,32 E.J.Quel,93S.Querchfeld,29S.Quinn,48J.Rautenberg,29O.Ravel,50D.Ravignani,11B.Revenu,50J.Ridky,32S.Riggi,84,13 M. Risse,1 P. Ristori,93 H. Rivera,41 V. Rizi,62 J. Roberts,69 W. Rodrigues de Carvalho,13 I. Rodriguez Cabo,13 G.RodriguezFernandez,42,13J.RodriguezMartino,47J.RodriguezRojo,47M.D.Rodríguez-Frías,59G.Ros,59J.Rosado,19 T. Rossler,79 M. Roth,26 B. Rouillé-d’Orfeuil,49 E. Roulet,7 A.C. Rovero,52 C. Rühle,75 S.J. Saffi,24 A. Saftoiu,37 F. Salamida,60 H. Salazar,88 F. Salesa Greus,36 G. Salina,42 F. Sánchez,11 P. Sanchez-Lucas,39 C.E. Santo,2 E. Santos,2 E.M. Santos,34 F. Sarazin,91 B. Sarkar,29 R. Sarmento,2 R. Sato,47 N. Scharf,67 V. Scherini,44 H. Schieler,26 P. Schiffer,14 A. Schmidt,75 O. Scholten,58 H. Schoorlemmer,73,16,55 P. Schovánek,32 F.G. Schröder,26,11 A. Schulz,26 J. Schulz,16 S.J.Sciutto,66M.Scuderi,43A.Segreto,84M.Settimo,20,1A.Shadkam,82R.C.Shellard,64I.Sidelnik,7G.Sigl,14O.Sima,99 A.Śmiałkowski,76R.Šmída,26G.R.Snow,98P.Sommers,40J.Sorokin,24H.Spinka,100,70R.Squartini,47Y.N.Srivastava,96 1550-7998=2014=89(5)=052002(18) 052002-1 © 2014 American Physical Society A. AAB et al. PHYSICAL REVIEW D 89, 052002 (2014) S. Stanič,72 J. Stapleton,77 J. Stasielak,35 M. Stephan,67 M. Straub,67 A. Stutz,21 F. Suarez,11 T. Suomijärvi,60 A.D. Supanitsky,52 T. Šuša,18 M.S. Sutherland,82 J. Swain,96 Z. Szadkowski,76 M. Szuba,26 A. Tapia,11 M. Tartare,21 O. Taşcău,29 N.T. Thao,65 J. Tiffenberg,53 C. Timmermans,55,16 W. Tkaczyk,76 C.J. Todero Peixoto,57 G. Toma,37 L. Tomankova,26 B. Tomé,2 A. Tonachini,45 G. Torralba Elipe,13 D. Torres Machado,50 P. Travnicek,32 D.B. Tridapalli,6 E. Trovato,43 M. Tueros,13 R. Ulrich,26 M. Unger,26 J.F. Valdés Galicia,12 I. Valiño,13 L. Valore,15 G. van Aar,16 A.M. van den Berg,58 S. van Velzen,16 A. van Vliet,14 E. Varela,88 B. Vargas Cárdenas,12 G. Varner,73 J.R. Vázquez,19 R.A. Vázquez,13 D. Veberič,72,71 V. Verzi,42 J. Vicha,32 M. Videla,56 L. Villaseñor,89 H. Wahlberg,66 P. Wahrlich,24 O. Wainberg,11,10 D. Walz,67 A.A. Watson,83 M. Weber,75 K. Weidenhaupt,67 A. Weindl,26 F. Werner,26 S.Westerhoff,101B.J.Whelan,40A.Widom,96G.Wieczorek,76L.Wiencke,91B.Wilczyńska,35H.Wilczyński,35M.Will,26 C.Williams,49T.Winchen,67B.Wundheiler,11S.Wykes,16T.Yamamoto,49T.Yapici,63P.Younk,102G.Yuan,82A.Yushkov,13 B.Zamorano,39E.Zas,13D.Zavrtanik,72,71M.Zavrtanik,71,72I.Zaw,69A.Zepeda,90J.Zhou,49Y.Zhu,75 M.ZimbresSilva,46andM.Ziolkowski1 (The Pierre Auger Collaboration)* 1Universität Siegen, Siegen, Germany 2LIP and Instituto Superior Técnico, Technical University of Lisbon, Portugal 3Osservatorio Astrofisico di Torino (INAF), Università di Torino and Sezione INFN, Torino, Italy 4University of Wisconsin, Madison, Wisconsin, USA 5Fermilab, Batavia, Illinois, USA 6Instituto de Física, Universidade de São Paulo, São Paulo, SP, Brazil 7Centro Atómico Bariloche and Instituto Balseiro (CNEA-UNCuyo-CONICET), San Carlos de Bariloche, Argentina 8New York University, New York, New York, USA 9Ohio State University, Columbus, Ohio, USA 10Universidad Tecnológica Nacional–Facultad Regional Buenos Aires, Buenos Aires, Argentina 11Instituto de Tecnologías en Detección y Astropartículas (CNEA, CONICET, UNSAM), Buenos Aires, Argentina 12Universidad Nacional Autonoma de Mexico, Mexico, Distrito Federal, Mexico 13Universidad de Santiago de Compostela, Spain 14Universität Hamburg, Hamburg, Germany 15Università di Napoli “Federico II” and Sezione INFN, Napoli, Italy 16IMAPP, Radboud University Nijmegen, Netherlands 17University of Wisconsin, Milwaukee, Wisconsin, USA 18Rudjer Bošković Institute, 10000 Zagreb, Croatia 19Universidad Complutense de Madrid, Madrid, Spain 20Laboratoire de Physique Nucléaire et de Hautes Energies (LPNHE), Universités Paris 6 et Paris 7, CNRS-IN2P3 Paris, France 21Laboratoire de Physique Subatomique et de Cosmologie (LPSC), Université Joseph Fourier Grenoble, CNRS-IN2P3 Grenoble INP, France 22Observatorio Pierre Auger and Comisión Nacional de Energía Atómica, Malargüe, Argentina 23University Politehnica of Bucharest, Romania 24University of Adelaide, Adelaide, South Australia, Australia 25Laboratoire de l’Accélérateur Linéaire (LAL), Université Paris 11, CNRS-IN2P3 Orsay, France 26Karlsruhe Institute of Technology–Campus North–Institut für Kernphysik, Karlsruhe, Germany 27Karlsruhe Institute of Technology–Campus South–Institut für Experimentelle Kernphysik (IEKP), Karlsruhe, Germany 28Ohio State University, Columbus, Ohio, USA 29Bergische Universität Wuppertal, Wuppertal, Germany 30University of Wisconsin, Madison, Wisconsi, USA 31Max-Planck-Institut für Radioastronomie, Bonn, Germany 32Institute of Physics of the Academy of Sciences of the Czech Republic, Prague, Czech Republic 33INFN, Laboratori Nazionali del Gran Sasso, Assergi (L’Aquila), Italy 34Universidade Federal do Rio de Janeiro, Instituto de Física, Rio de Janeiro, Brazil 35Institute of Nuclear Physics PAN, Krakow, Poland 36Colorado State University, Fort Collins, Colorado, USA 37‘HoriaHulubei’NationalInstituteforPhysicsandNuclearEngineering,Bucharest-Magurele,Romania 38Colorado State University, Pueblo, Colorado, USA 39Universidad de Granada and C.A.F.P.E., Granada, Spain 40Pennsylvania State University, University Park, USA 052002-2 PROBING THE RADIO EMISSION FROM AIR SHOWERS … PHYSICAL REVIEW D 89, 052002 (2014) 41Università di Milano and Sezione INFN, Milan, Italy 42Università di Roma II “Tor Vergata” and Sezione INFN, Roma, Italy 43Università di Catania and Sezione INFN, Catania, Italy 44Dipartimento di Matematica e Fisica “E. De Giorgi” dell’Università del Salento and Sezione INFN, Lecce, Italy 45Università di Torino and Sezione INFN, Torino, Italy 46Universidade Estadual de Campinas, IFGW, Campinas, Sao Paulo, Brazil 47Observatorio Pierre Auger, Malargüe, Argentina 48Case Western Reserve University, Cleveland, Ohio, USA 49Enrico Fermi Institute, University of Chicago, Chicago, Illinois, USA 50SUBATECH, École des Mines de Nantes, CNRS-IN2P3, Université de Nantes, Nantes, France 51Station de Radioastronomie de Nançay, Observatoire de Paris, CNRS/INSU Nançay, France 52Instituto de Astronomía y Física del Espacio (CONICET-UBA), Buenos Aires, Argentina 53Departamento de Física, FCEyN, Universidad de Buenos Aires y CONICET, Argentina 54Universidade Federal Fluminense, EEIMVR, Volta Redonda, Rio de Janeiro, Brazil 55Nikhef, Science Park, Amsterdam, Netherlands 56Instituto de Tecnologías en Detección y Astropartículas (CNEA, CONICET, UNSAM), and National Technological University, Faculty Mendoza (CONICET/CNEA), Mendoza, Argentina 57Universidade de São Paulo, Instituto de Física, São Carlos, São Paulo, Brazil 58Kernfysisch Versneller Instituut, University of Groningen, Groningen, Netherlands 59Universidad de Alcalá, Alcalá de Henares Spain 60Institut de Physique Nucléaire d’Orsay (IPNO), Université Paris 11, CNRS-IN2P3 Orsay, France 61Michigan Technological University, Houghton, Michigan, USA 62Università dell’Aquila and INFN, L’Aquila, Italy 63Michigan Technological University, Houghton, Michigan, USA 64Centro Brasileiro de Pesquisas Fisicas, Rio de Janeiro, Rio de Janeiro, Brazil 65Institute for Nuclear Science and Technology (INST), Hanoi, Vietnam 66IFLP, Universidad Nacional de La Plata and CONICET, La Plata, Argentina 67RWTH Aachen University, III. Physikalisches Institut A, Aachen, Germany 68ASTRON, Dwingeloo, Netherlands 69New York University, New York, New York, USA 70Fermilab, Batavia, Illinois, USA 71J. Stefan Institute, Ljubljana, Slovenia 72Laboratory for Astroparticle Physics, University of Nova Gorica, Slovenia 73University of Hawaii, Honolulu, Hawaii, USA 74Dipartimento di Fisica dell’Università and INFN, Genova, Italy 75Karlsruhe Institute of Technology–Campus North–Institut für Prozessdatenverarbeitung und Elektronik, Germany 76University of Łódź, Łódź, Poland 77Ohio State University, Columbus, Ohio, USA 78Universidade Estadual de Feira de Santana, Brazil 79Palacky University, RCPTM, Olomouc, Czech Republic 80Institute of Space Sciences, Bucharest, Romania 81Charles University, Faculty of Mathematics and Physics, Institute of Particle and Nuclear Physics, Prague, Czech Republic 82Louisiana State University, Baton Rouge, Los Angeles, USA 83School of Physics and Astronomy, University of Leeds, United Kingdom 84Istituto di Astrofisica Spaziale e Fisica Cosmica di Palermo (INAF), Palermo, Italy 85University of New Mexico, Albuquerque, New Mexico, USA 86Faculdade Independente do Nordeste, Vitória da Conquista, Brazil 87Universidade Federal do ABC, Santo André, Sao Paulo, Brazil 88Benemérita Universidad Autónoma de Puebla, Mexico 89Universidad Michoacana de San Nicolas de Hidalgo, Morelia, Michoacan, Mexico 90Centro de Investigación y de Estudios Avanzados del IPN (CINVESTAV), México, Distrito Federal, Mexico 91Colorado School of Mines, Golden, Colorado, USA 92Instituto de Física de Rosario (IFIR)–CONICET/U.N.R. and Facultad de Ciencias Bioquímicas y Farmacéuticas U.N.R., Rosario, Argentina 93Centro de Investigaciones en Láseres y Aplicaciones, CITEDEF and CONICET, Argentina 94Institut de Física Corpuscular, CSIC-Universitat de Vale`ncia, Valencia, Spain 052002-3 A. AAB et al. PHYSICAL REVIEW D 89, 052002 (2014) 95University of Wisconsin, Milwaukee, Wisconsin, USA 96Northeastern University, Boston, Massachusetts, USA 97Universidade Federal da Bahia, Salvador, British Airways, Brazil 98University of Nebraska, Lincoln, Nebraska, USA 99University of Bucharest, Physics Department, Romania 100Argonne National Laboratory, Argonne, Illinois, USA 101University of Wisconsin, Madison, Wisconsin, USA 102Los Alamos National Laboratory, Los Alamos, New Mexico, USA (Received 19 September 2013; published 14 March 2014) Theemissionofradiowavesfromairshowershasbeenattributedtotheso-calledgeomagneticemission process.Atfrequenciesaround50MHzthisprocessleadstocoherentradiationwhichcanbeobservedwith rathersimplesetups.Thedirectionoftheelectricfieldinducedbythisemissionprocessdependsonlyon thelocalmagneticfieldvectorandontheincomingdirectionoftheairshower.Wereportonmeasurements oftheelectricfieldvectorwhere,inadditiontothisgeomagneticcomponent,anothercomponenthasbeen observedthatcannotbedescribedbythegeomagneticemissionprocess.Thedataprovidestrongevidence thattheotherelectricfieldcomponentispolarizedradiallywithrespecttotheshoweraxis,inagreement withpredictionsmadebyAskaryanwhodescribedradioemissionfromparticleshowersduetoanegative charge excess in the front of the shower. Our results are compared to calculations which include the radiation mechanism induced by this charge-excess process. DOI: 10.1103/PhysRevD.89.052002 PACSnumbers:96.50.sb, 96.60.Tf,07.57.Kp I. INTRODUCTION Earth while the charge excess in the shower front is to a largeextentduetotheknock-outoffastelectronsfromthe Whenhigh-energycosmicrayspenetratetheatmosphere ambient air molecules by high-energy photons in the of the Earth, they induce an air shower. The detailed shower. The magnitude of the induced electric current as registration of this avalanche of secondary particles is an wellastheinducedchargeexcessisroughlyproportionalto essentialtooltoinferpropertiesoftheprimarycosmicray, thenumberofparticlesintheshowerandthuschangingin such as its energy, its incoming direction, and its compo- time. The latter results in the emission of coherent radio sition. Radio detection of air showers started in the 1960s waves at sufficiently large wavelengths [12,14]. The and the achievements at that time have been presented in shower front, where both the induced transverse current reviewsbyAllan[1]andFegan[2].Inthelastdecade,there and the charge excess reside, moves through the air with has been renewed interest through the publications of the nearly the velocity of light. Because air has a refractive LOPES[3]andCODALEMA[4]Collaborations.Wehave index which differs from unity, Cherenkov-like time deployed and are still extending the Auger Engineering compression occurs [15,16], which affects both the radi- Radio Array (AERA) [5–9] as an additional tool at the ation induced by the transverse current as well as by the Pierre Auger Observatory to study air showers with an charge excess. The polarization of the emitted radiation energy larger than 1017 eV. In combination with the data differs for current-induced and charge-induced radiation, retrievedfromthesurface-basedparticledetectors[10]and but its direction for each of these individual components thefluorescencedetectors[11]ofthisobservatory,thedata does not depend on the Cherenkov-like time compression fromradiodetectorscanprovideadditionalinformationon causedbytherefractiveindexofair.Forthisreasonwewill the development of air showers. distinguish in this paper only geomagnetic (current- An important step in the interpretation of the data induced) and charge-excess (charge-induced) radiation. obtainedwithradio-detectionmethodsistheunderstanding Thecontributionofthegeomagneticemissionmechanism, of the emission mechanisms. In the early studies of radio described as a time-changing transverse current by Kahn emission from air showers, it was conjectured that two and Lerche [12], has been observed and described in emission mechanisms play an important role: the geo- severalpapers;seee.g.Refs.[3,17–20].Studiesonpossible magnetic emission mechanism as proposed, amongst contributions of other emission mechanisms from air others, by Kahn and Lerche [12] and the charge-excess showers have also been reported [21–25]. An observation mechanismasproposedbyAskaryan[13].Essentialforthe ofthecharge-excesseffectinairshowershasbeenreported geomagnetic effect is the induction of a transverse electric by the CODALEMA Collaboration [26]. currentintheshowerfrontbythegeomagneticfieldofthe We present the analysis of two data sets obtained with two different setups consisting of radio-detection stations *[email protected] (RDSs)deployedatthePierreAugerObservatory.Thefirst 052002-4 PROBING THE RADIO EMISSION FROM AIR SHOWERS … PHYSICAL REVIEW D 89, 052002 (2014) datasetwasobtainedwithaprototypesetup[27]forAERA; the present study only the SD was used to determine the theotheronewithAERAitself[5–7]duringitscommission- parameters of the air showers. ingphasewhileitconsistedofonly24stations.Inaddition,we willcomparethesedatasetswithresultsfromdifferenttypes B. Radio-detection systems ofcalculationsoutlinedinRefs.[28–34]. The prototype for AERA used in the present study Thispaperisorganizedasfollows.WediscussinSec.II the experimental equipment used to collect our data. In consisted of four RDSs. Each RDS had a dual-polarized Sec.IIIwepresentthedataanalysistechniquesandthecuts logarithmicperiodicdipoleantenna(LPDA)optimizedfor that we applied on the data, while in Sec. IV we compare receiving radio signals in a frequency band centered at ourdatawithcalculations.InSec.Vwediscusstheresults 56 MHz and with a bandwidth of about 50 MHz. These andwepresentourconclusions.Forclarity,Secs.IIIandIV antennaswerealignedsuchthatonepolarizationdirection containonlythosefiguresthatarebasedontheanalysisfor was pointing along the geomagnetic north-south (NS) thedataobtainedwithAERAduringthecommissioningof direction with an accuracy of 0.6°, while the other itsfirst24stations;theresultsoftheprototypeareshownin polarization direction was pointing to the east-west AppendixC.Wementionthatanalysesofpartsofthedata (EW) direction. For each polarization direction, NS and have been presented elsewhere [19,35–37]. EW,weusedanalogelectronicstoamplifythesignalsand to suppress strong lines in the HF band below 25 MHz II. DETECTION SYSTEMS and in the FM-broadcast band above 90 MHz. A 12-bit digitizer running at a sampling frequency of 200 MHz A. Baseline detector system was used for the analog-to-digital conversion of the The detection system used at the Pierre Auger signals. This electronic system was completed with a Observatoryconsistsoftwobaselinedetectors:thesurface GPS system, a trigger system, and a data-acquisition detector(SD)andthefluorescencedetector(FD),described system. The trigger for the station readout was made in detail in Refs. [10] and [11], respectively. The SD is an usingascintillatordetectorconnectedtothesamedigitizer array consisting of 1660 water-Cherenkov detectors as was used for the digitization of the radio signals. The arranged in equilateral triangles with sides of 1.5 km. data from all RDSs were stored on disks and afterwards An infill for the SD, called AMIGA [38], has been deployed with 750 m spacing between the stations. A compared with those from the SD. To ensure that the schematicdiagramoftheobservatoryisshowninFig.1.In events, which have been registered with the RDSs, were indeedinducedbyairshowers,acoincidencebetweenthe datafromAERAprototypeandfromtheSDintimeandin location was required [27]. An additional SD station was deployed near the AERA prototype setup to reduce the energy threshold of the SD; see the left panel of Fig. 2. Further details of the AERA prototype stations can be found in Ref. [27]. AERAis sited at the AMIGAinfill[38] ofthe observa- toryandconsistspresentlyof124stations[9].Thedeploy- ment of AERA began in 2010 and physics data-taking started in March 2011. In the data-taking period presented in this work, AERA consisted of 24 RDSs arranged on a triangular grid with a station-to-station spacing of 144 m; seetherightpanelofFig.2.Forthepresentdiscussion,we will denote this stage as AERA24. The characteristics of AERA24areverysimilartoitsprototype.Acomparisonof theirfeaturesis presentedin TableI; seealsoRef.[39]for further details. One of the main differences between AERA24 and its prototype is that the AERA stations can trigger on the signals received from the antennas FIG.1(coloronline). ThedetectorsystemsofthePierreAuger whereastheprototypeusedonlyanexternaltriggercreated Observatory;theblackdotsdenotethe1660detectorstationsof by a particle detector. In addition to these event triggers, the surface detector (SD), while the buildings containing the both systems recorded events which were triggered every telescopes of the fluorescence detector (FD) are located at 10susingthetimeinformationfromtheGPSdeviceofthe the edge of the array. The prototype of AERAwas located near theBalloonLaunchingStation(BLS)oftheobservatory;AERA RDS. These events are, therefore, called minimum bias itself is located in front of the telescope buildings at Coihueco. events. 052002-5 A. AAB et al. PHYSICAL REVIEW D 89, 052002 (2014) -4.5 +16.5 -5.0 +16.0 m] m] k -5.5 k +15.5 y [ y [ -6.0 +15.0 -6.5 +14.5 -27.5 -27.0 -26.5 -26.0 -25.5 -27.0 -26.5 -26.0 -25.5 -25.0 x [km] x [km] FIG. 2. Anaerial viewof theradio-detectionsystems (open triangles) in theSD.Stations oftheSD aredenoted byfilled markers, wheretheSDstationsnearesttotheradio-detectionsystemsaredenotedwithfilledsquares,fortheprototypesetup(leftpanel)andfor AERA24 (right panel). The coordinates are measured with respect to the center of the SD. III. EVENT SELECTION AND DATA ANALYSIS field vector on Earth at the location of the Pierre Auger Observatory was 24 μT and its direction was pointing to The data from the SD and the radio-detection systems werecollectedandanalyzedindependentlyfromeachother ðθb;ϕbÞ¼ð54.4°;87.3°Þ [40]. The contribution to the emitted radio signal caused by as will be described in Secs. III B and III C, respectively. the charge-excess effect (denoted as E⃗ A) is not influenced Usingtiminginformationfrombothdetectionsystems(see by the geomagnetic field B⃗ . The induced electric field Sec. III D) an off-line analysis was performed combining vectorofthiseffectisradialwithrespecttotheshoweraxis. the data from both detectors. AsexplainedinSec.IIBthedual-polarizedantennaswere directed in the NS and EW directions. Therefore, the A. Conventions relative amplitudes of the registered electric field in each For the present analysis we use a spherical coordinate of the two arms of an RDS depend on the position of the system with the polar angle θ and the azimuth angle ϕ, RDS with respect to the shower axis. The geomagnetic- where we define θ ¼0° as the zenith direction and where emission mechanism induces an electric field E⃗ G which is ϕ¼0° denotes the geographic east direction; ϕ increases pointingalongthedirectionof(−v⃗ ×B⃗ )wherev⃗ isavector while moving in the counter-clockwise direction. We in the direction of the shower. Thus, for this emission determine the incoming direction (θa and ϕa) of the air mechanism, the relative contribution of the registered shower by analysis of the SD data. For the relevant period signals in each of the two arms does not change as a of data taking, which started in May 2010 and ended in function of the position of the RDS. For this reason, it is June2011,thestrengthanddirectionofthelocalmagnetic convenient to introduce a rotated coordinate system ðξ;ηÞ TABLE I. Comparison of characteristic features of AERA24 during this data-taking period and its prototype. AERA24 AERA prototype Antenna type LPDA LPDA Number of polarization directions 2 2 −3 dB antenna bandwidth 29–83 32–84 MHz gain LNA 20 22 dB −3 dB pass filter bandwidth 30–78 25–70 MHz Gain main amplifier 19 31 dB Sampling rate digitizer 200 200 MHz Digitizer conversion 12 12 bits Trigger EW polarization particle RDS station-to-station spacing 144 216 m SD infill spacing 750 866 m 052002-6 PROBING THE RADIO EMISSION FROM AIR SHOWERS … PHYSICAL REVIEW D 89, 052002 (2014) rightpanelofFig.2.Theestimatederrorsontheincoming direction and on the determination of the core position depend on the energy of the registered air showers. These errors are smaller at higher energies. Typical directional errors in our studies range from 0.5° at 1 EeV to 1° at 0.1 EeV. The uncertainty in the determination of the core positions also reduces for higher energies and lies around 60 m at 0.1 EeVand around 20 m at 1.0 EeV. C. Event selection and data analysis for the RDSs FIG. 3. Direction of the incoming shower, denoted as vˆ, with The data from the RDSs were used to determine the respecttothepositionofRDSwhichissymbolicallyindicatedby polarization of the electric field induced by air showers. anantenna.Thedirectionofthemagneticfieldvectorisdenoted byBˆ andthedirectionξisdefinedbytheprojectionofthevector Here we take advantage of the fact that the LPDAs were vˆ ×Bˆ ontothegroundplane.Thedirectionηisperpendiculartoξ designed as dual-polarized antennas. The data for each of thesetwopolarizationdirectionswerestoredastimetraces and is also in the ground plane. The angle between the shower axisandthegeomagneticfielddirectionisdenotedasα,whileψ with2000samples(thuswithalengthof10 μs).Weusethe istheazimuthalanglebetweentheξaxisandthedirectionofthe Hilbert transformation, which is a standard technique for RDS measured at the core position. bandpass-limited signals [42], to calculate the envelope of the time trace. An example of sucha trace is shown in the upperpanelofFig.4,whichwasrecordedforanairshower in the ground plane such that the ξ direction is the projection of the vector (−vˆ ×Bˆ) onto the shower plane with parameters θa ¼ð30.0(cid:1)0.5Þ°, ϕa ¼ð219(cid:1)2Þ° and E¼ð0.19(cid:1)0.02Þ EeV near the AERA24 site. andηisorthonormaltoξ;seeFig.3.Theanglebetweenthe Wetookseveralmeasurestoensuregooddataqualityfor incomingdirectionoftheshowerandthegeomagneticfield the received signals in the RDSs. Despite of the bandpass vector is denoted as α. filters (see Table I) a few narrow-band transmitters con- Forourpolarizationanalysis,weconsideratotalelectric taminatedtheregisteredsignals.Theeffectofthesuppres- field as the vectorial sum of the geomagnetic and of the sionofthefrequencyregionsoutsidethepassbandandthe charge-excess emission processes, remainingcontributionsfromthenarrow-bandtransmitters E⃗ ðtÞ¼E⃗ GðtÞþE⃗ AðtÞ; (1) within the passband are displayed in the middle panel of Fig. 4, which shows the Fourier transform of the time trace shown in the upper panel. These narrow-band trans- where t describes the time dependence of the radiation mitters were removed by applying two different digital received at the location of an RDS. methods. The first method operates in the time domain and involves a linear predictor algorithm based on the B. Data preprocessing and event selection for the SD time-delayed forecasted behavior of 128 consecutive The incoming directions and core positions of air time samples [43]. The second method involves a showers were determined from the recorded SD data. A Fourier- and inverse-Fourier-transform algorithm, where detailed description of the trigger conditions for the SD in the frequency domain the power of the narrow-band array with its grid spacing of 1500 m is presented in transmitters was set to zero (see, e.g., Ref. [3]). Ref. [41]. However, as mentioned before, additional SD To determine the total electric field vector, we used the stationswereinstalledintheneighborhoodoftheRDSsas simulated antenna gain pattern [44] and the incoming infills of the standard SD cell. Because of these additional direction of the air showers as determined with the SD surface detectors, we used slightly different constraints as analysis. This technique is described in detail in compared to the cuts used for the analysis of events Refs. [45,46]. In the analysis of the radio signals we used registered by the regular SD array only. These additional theanalyticsignal,whichisacomplexrepresentationofthe constraints are a limit on the zenith angle (θ <40° for the electric field vector E⃗j, where j runs over the sample prototype and θ <55° for AERA24). Furthermore, in the number in the time trace. This complex vector was caseofeventsrecordedneartheprototype,theanalysiswas constructed from the electric field itself using the Hilbert basedononlythoseeventswheretheinfillSDstationnear transformation H: this setup yielded the largest signal strength (i.e., highest particleflux)andwherethereconstructedenergybytheSD E⃗j ¼E⃗ jþiHðE⃗ jÞ: (2) was larger than 0.20 EeV. For the AERA24 events, we required that the distance from the reconstructed shower InthelowerpanelofFig.4theanalyticsignaloftheelectric axis to the infill station was less than 2500 m or that the fieldforthisparticulareventisdisplayedafterremovingthe eventcontainedatleastoneoftheSDstationsshowninthe narrow-band transmitters from the signal. 052002-7 A. AAB et al. PHYSICAL REVIEW D 89, 052002 (2014) The data obtained with the minimum bias triggers were polarization directions) caused by the rising and setting of analyzed to check the gain for the different polarization strong sources in the galactic plane; see e.g. Ref. [27]. In directions. For this comparison we used the nearly daily the present analysis only data from those RDSs were variation of the signal strength in each RDS (for the two selected where the difference between the relative gain for their two polarization directions was less than 5%. Furthermore, it is well known [47–49] that thunderstorm conditions may cause a substantial change of the radio signalstrengthfromair showersas comparedto thesignal strength obtained under fair-weather conditions. The atmosphericmonitoringsystemsoftheobservatory,located at the BLS and at AERA (see Fig. 1) register the vertical electric field strength at a height of about 4 m. Characteristic changes in the static vertical electric field strength are indicative for thunderstorm conditions and air-shower events collected during such conditions have been ignored in the present analysis. D. Coincidences between the SD and the RDSs ThedatastreamsfortheSDandRDSswerecheckedfor coincidencesintimeandinlocation.FortheSDeventswe used the reconstructed time at which the shower hits the ground at the core position. For the timing of an event registeredbyoneormoreoftheRDSsweusedtheearliest time stamp obtained from the triggered stations. We required that the relative difference in the timing between the SD events and the events registered by the radio detector is smaller than 10 μs. The distribution of the relative time difference of the coincident events registered with AREA is shown in Fig. 5. The shift of about 8 μs betweentheSDandRDStimingiscausedbythedifferent trigger definitions used for each of the two different detection systems: SD and RDS. This figure clearly dis- plays the prompt coincidence peak and some random events. The events selected for further analysis are within the indicated window in this figure. Our analysis is based on 37 air-shower events, 17 registered with AERA24, the other 20 registered with the prototype. We note here that each event can produce 6 4 s nt e v e FIG. 4 (color online). Anexampleof aradio signal invarious 2 stages of the analysis. Upper panel: the square root of the quadratic sum for the signal envelopes of both polarization directions. Middle panel: the power distribution of this signal 0 inthefrequencydomain.Lowerpanel:theanalyticsignalforthe -10 -5 0 5 10 electric field E⃗ [see Eq. (2)] reconstructed from the two time t - t [µs] RDS SD tracesandfromtheincomingdirectionoftheshower.Thesignal was cleaned from narrow-band transmissions using the linear FIG. 5 (color online). Difference between the reconstructed predictor algorithm. The signal (noise) region used for this arrivaltimeoftheair-showereventsrecordedbytheSDandthe algorithmisdenotedby1(2)andhasawidthof125ns(1600ns). RDSs of AERA24. 052002-8 PROBING THE RADIO EMISSION FROM AIR SHOWERS … PHYSICAL REVIEW D 89, 052002 (2014) interest with a width of 500 ns in the recorded RDS time traces.Inthisregion,indicatedinthebottompanelofFig.4 as region 1, we expect radio pulses from recorded air- showerevents.Becauseoftimejitterthepreciselocationof theradiosignalitselfwasdeterminedusingasmallsliding windowwithawidthof125ns,i.e.25timesamples.Asa first step, we removed the narrow-band noise using the methodbasedonalinearprediction,describedinSec.IIB. Then, within the region of interest, the total amplitude in theslidingwindowwascomputedbyaveragingthesquared sum of the three electric field components and taking the square root of this summed quadratic strength. The maxi- mum strength obtained by the sliding window was then chosen to be the signal S. Thus we have (cid:1) (cid:2) 1 X25 1=2 S¼25 E⃗jþk·E⃗(cid:2)jþk ; (3) j¼1 where the left edge of this sliding window has sample identifier k. For every measured trace, k was chosen such that S reaches a maximum value. The noise level N was determinedfromregion2showninthebottompanelFig.4. This region has a width of 1600 ns (320 samples), (cid:1) (cid:2) 1 Xm1 1=2 N ¼320 E⃗m·E⃗(cid:2)m ; (4) m¼m 0 where m0 and m1 ¼m0þ319 are, respectively, the start andstopsamplesofthisnoiseregion.FortheAERA24data showninthebottompanelofFig.4,thevalueofm0 ¼520 (1600 ns). FIG.6. Scatterplotofshowerparametersforcoincidentevents Toidentifythetwomechanismsofradioemissionunder used in the analysis; the filled circles (open triangles) are data discussion, we take advantage of the different polarization for AERA24 (prototype). Upper panel: the shower energy E directions that are expected in each case (see Sec. III A), reconstructed from the SD information versus the space angle α between the magnetic field vector and the shower axis. whereweintroducedthecoordinatesystemðξ;ηÞdepicted Lower panel: the reconstructed energy E versus the distance d in Fig. 3. In the rotated coordinate system the resulting betweentheshoweraxisandtheSDstationlocatedclosesttothe strengthoftheelectricfieldinthegroundplaneisgivenas radio-detection systems (see Fig. 2). Eξ and Eη. We define the quantity R as P RðψÞ≡2 25 ReðE E(cid:2) Þ severaldatapointsinouranalysis.Thedistancedbetween P j¼1 jþk;ξ jþk;η (5) the shower axis and the SD station closest to the center of 25 ðjE j2þjE j2Þ; j¼1 jþk;ξ jþk;η the radio setup, the angle α between the magnetic field vector and the shower direction, and the shower energy E where we use the observation angle ψ, which is the are relevant parameters for the RDS triggers discussed in azimuthal angle at the shower core between the position Sec. II B. The upper (lower) panel of Fig. 6 displays a scatter plot of these coincident events in the ðE;αÞ and oftheRDSandthedirectionoftheξaxis.SinceEη hasno component in the case of pure geomagnetic emission, it is ðE;dÞ parameter space. clear from Eq. (5) that any measured value of R≠0 indicates a component different from geomagnetic emis- E. Deviation from geomagnetic polarization as a sion.ThemeasuredvalueofRincursabiasinthepresence function of the observation angle of noise, which was taken care of using the procedure ForeachshowerandeachRDS,weusedtheSDtimeand explained in Appendix A. To use signals with sufficient the ðx;yÞ coordinates of each RDS to define a region of quality the following signal-to-noise cut was used, 052002-9