IL NUOVOCIMENTO Vol. ?, N. ? ? Chacaltaya: towards a solution of the knee ....? 1 0 0 O. Saavedra(1) and L. Jones(2) 2 (1) Dipartimento di Fisica Generale Torino, Italy. n Istituto Nazionale di Fisica Nucleare, Torino,Italy. a (2) Department of Physics, University of Michigan, Ann Arbor, USA. J 1 3 1 v 4 Summary. — Cosmic rays physics is currently being studied with rather sophis- 5 ticated detectors running in a variety of experimental conditions and atmospheric 5 depths around the world. In this paper we describe the reasons why cosmic ray 1 physicsexperimentsathighaltitudeslikeChacaltayaaresoimportantforresolving 0 some of the open problems in cosmic physics. A discussion on the futureprospects 1 ofthehighaltitudemountainlaboratoriessuchasChacaltayaforcosmicrayphysics 0 is presented. / h PACS 96.40 – Cosmic Rays. p PACS 96.40.De – Composition, energy spectra and interactions. - PACS 96.40.Pq – ExtensiveAir Showers. o r t s a : v i X 1. – Introduction r a One of the most important problems in cosmic ray physics, not yet resolved after nearly 40 years, is the existence of the ”knee in the primary spectrum. And there are many other important problems in the field of cosmic rays that should be studied, in particular experimental investigations,and at high altitude sites. One may ask why are observations of cosmic rays at high altitude laboratories are so important? One reason is because it is possible to observe the air shower cascades produced by lower energy of primary cosmic rays than at lower altitudes, due to their attenuation in the atmospheric overburden. This is particularly relevant in the current studies of gamma ray sources and gamma bursters. Also, to study the shower cascades of higher energies at an early stage of development where showers present a minimum of fluctuations. The atmospheric depth of Chacaltaya (540 g/cm2) corresponds to the maximum development of the showers in the energy range of 10 - 500 TeV, giving a maximum detection probability. Moreover the fluctuations in the development of the showers are much lower at the observation level of Chacaltaya than at lower altitudes. In this paper a possible experiment using a variety of different techniques and carried out by a wide international collaborationis also presented and discussed. (cid:13)c Societa`ItalianadiFisica 1 2 O.SAAVEDRAand L.JONES 2. – The Chacaltaya Laboratory TheChacaltayaCosmicRayResearchLaboratory,nearLaPaz,Bolivia,islocatedat an elevation of 5220 meters above sea level corresponding to 540gr/cm2, equivalent to about7 nuclearinteractionmeanfree paths (for TeVenergyprotons)and14.1radiation ◦ ◦ lengths. Its geographic position is at 16 South Latitude. 291.8 East Longitud. and −4◦ of geomagnetic latitude corresponding to 13.1 GV cutoff rigidity. Its geographic location in the Southern Hemisphere is also important in order to observe a part of the skywhereγ-astronomyisnotaswelldevelopedasitisintheNorthernHemisphere. The Galactic Center, in particular, can be observed at low zenith angle. The Chacaltaya Laboratory is the highest continuously functioning research station on the globe, and providesa unique opportunity for researchon cosmic rayphenomena. At energies above 1014eV, the flux of primary cosmic rays is so low that direct observation by balloon-orsatellite-borneinstruments (withareasofonlyafew squaremeters)isnot feasible. For example, the integral primary cosmic ray flux of energies above 1016eV is only one particle per (m2·sr·yr). Consequently, for the most sensitive indirect studies of cosmic rays with energies of and above 1015eV (one PeV), it is necessary to deploy extensivedetectorsystemsatashighanelevationaspossible,toreducetheofatmosphere overburden. . 21. Hadronic intensities at Chacaltaya. – A experiment using emulsion chambers in combination with the extensive air showertechnique is being operated at Chacaltayaby the SYS collaboration. The collaboration is observing bundles of high energy gammas andhadronsintheairshowercores(withemulsionchambers)associatedwithairshowers detected bythe EASarray. The experimentemploys35plasticscintillators,32emulsion chamber units and a hadron calorimeter. Each emulsion chamber unit, with dimensions of50cm·50cm·15cm,consistingof15layersof1cmPbplatesandtwosheetsofX-ray films. Below each unit a scintillator detector of the same area as the emulsion chamber detects the bundle of charged particles produced in the emulsion chamber material by the hadronic component of the shower through local nuclear interactions. The gamma ”family” is connected to the hadronic component through the geometri- calposition,andthe hadroniccomponentisrelatedtothe airshowerthroughthe arrival time [1]. Each unit of the hadron calorimeter provides an output which is a measure of the energy released in the scintillator and is converted to the number of (minimum- ionizing)chargedparticles. Thedataproducedbytheairshowerarrayandbythehadron calorimeter are recordedwhen at leastone unit of the hadroncalorimeter records a par- ticle density n ≥ 103 (particles/0.25 m2). 2408 events were selected during 4.6 years b of data taking, during which the emulsion chambers were simultaneously active. The shower size interval for these data is N =5·106−107 e The receintly published results [2] can be seen in figure 1 where the differential energy spectrum of hadrons in air showersis comparedwith the expectations from simulations. These simulations used different models to simulate multiple particle production; the UA5 algorithm (modified for hadron-nucleus collisions), VENUS [3], QGSJET [4] and HDPM [5]. Atmospheric diffusion of cosmic rays is described by the code in ref. [6] for the UA5 model, and by CORSIKA 5.20 [7] for the rest. It is apparent that the average number of hadrons for shower size N =5·106−107 is lower than that predicted by all e thesesimulations. Thistendencyisconsistentnotonlywiththerelationshipbetweenthe γ families and the accompanied air showers discussed by Kawasumi et al. [1], but also CHACALTAYA:TOWARDSASOLUTIONOFTHEKNEE....? 3 withthe conclusionsofthe KASCADE experimentatsealevel [8], wherethey havealso reportedahadronicfluxaccompanyingairshowerslowerthanthe expectationsbasedon severalmodels. In figure 2 the measuredmuontriggerrate as a function ofthe hadronic rate is compared with the predictions from severalsimulations for multiple particle pro- duction. In a more recent paper by the KASCADE group [9], this result is confirmed. These measurements: γ families and hadrons with the SYS experiment at Chacaltaya, and hadrons at the KASCADE array at sea level, are quite independent, as they are observing with different detector systems. However they display similar problems in describing shower development through the atmosphere. If we take into account these results, (i.e. lower hadronic intensities than expected in the KASCADE experiment [8] and at Chacaltaya [2], as well as lower intensities of γ-families [1]), it would seem that at primary energies in the range 1015-1016 eV a larger dissipative mechanism occurs in nuclear interactions than current models predict. In fact, if we consider cosmic phe- nomenology, we see the “knee” problem seen in the energy spectrum of primary cosmic rays in the same energy range of the emulsion chamber families (e, γ). Fig. 2. – Comparison between simulated and Fig. 1. – Differential energy spectrum of measured integral muon trigger and hadronic hadronsinairshowers(solidline)andexpecta- rates at the KASCADE calorimeter. The ex- tions from MC simulations with various inter- perimental hadronic rate is much lower than action models (See text). The EAS size range is Ne =5·106−107. the expectations by various simulation codes. . 22. Hadronic interaction at colliders and cosmic rays. – The primary cosmic ray spectrum is derived from flux measurements of EAS for which energies are measured by the total number of particles in the shower. It is obvious that such measurements at energies higher than the “knee” region can be seriously affected if the characteristics of the hadronic interaction change radically. On the other hand, due to the lack of knowledge of the high energy interactions from particle accelerators in this energy range, it is hard to make comparisons between experimental results and simulations. In fact, as has been pointed out by Jones [10] all 4 O.SAAVEDRAand L.JONES Fig.3.–Multiplicitydistributions(up)andenergydistributions(down)vs. rapidityofsecondary particles inhighenergy p−p¯collisionsforenergy range 103 -107 GeVintherest system of one oftheprimaryparticles. Verticalbandsshowtheangularacceptance oftwocolliderexperiments. these comparisonsrequirethe knowledgeof hadronicinteractionsat the highest energies and over a wide angular range. The highest energy reached currently is at Fermilab collider,studiedwiththeCDFandD0detectors,andcorrespondsto∼2000TeVproton collision with a stationary proton. The LHC collider currently under construction at CERN will extend this up to about ∼1017 eV. (equivalent cosmic ray primary energy). These collider facilities will thus covermostof the EAS energy regionof currentinterest in the contex of the ”knee” and the primary composition problems. Unfortunately, this is not the case for the angular range (or rapidity) coverage. In the upper part of figure 3 [11], the multiplicity distribution of secondary particles for different primary energies is shown, together with the rapidity range covered by the CDF experiment and UA5 experiments, which had the largest rapidity range achieved thus far among high energy collider detectors. It can be seen that most of the secondary particles are detected by both detectors. But in the lower part of figure 3, where the energy distribution of secondaryparticlesisdisplayed,itisclearthatonlyasmallfractionoftheprimaryenergy is measured by these detectors, and this fraction decreases with primary energy. This effetc is even more important in the coordinate system where one (target) nucleon is at rest,i.e. thecosmicraycase. Sincetheairshowerdevelopmentisdominatedbythefinal state energy flow, i.e. by this missing forward region, the most important information for EAS simulation is not provided by these collider experiments. 3. – Future prospects As is well known the energy region in the cosmic ray spectrum at primary energies of 1015−1016 eV presents the enigmatic “knee”. The origin of this change of slope of CHACALTAYA:TOWARDSASOLUTIONOFTHEKNEE....? 5 the primary cosmic rays spectrum is still unresolved, even after a great deal of efforts has been dedicated with increasingly sophisticated equipments working in a variety of atmosphericdepthsinthearoundtheworld. Thequestionabouttheoriginofthe“knee” is still open. To know the reason of the existence of the knee is a major challenge faced in the very near future. Is it due to a change in production and/or acceleration mech- anism at the source, propagation through interstellar space, or perhaps a change in the nature of the high energy interactions (as suggested by some emulsion experiments at high altitude laboratories)? Thecurrentthinkingisthatperhapsnotjustonereasonbuttwo(ormore)coincident mechanisms are operating in this energy region to produce the ”knee”. What should be the next steps for investigation of cosmic rays through this energy range? One could expect that some answers to the problems of nuclear interactions and multiple particle productionwillbegivenbythecurrentacceleratorsorfromtheLHCexperiments. How- ever, as it is shown in figure 3, we lose information in the very forward directions with anincreaseincollider energy,unless differentdetector architecturesspecifically sensitive to small-angle (high rapidity) final states are utilized. In order to fully exploit the great potential of high altitude laboratories such as Cha- caltaya, the simultaneous measurement of different observables should be undertaken. For example, consider an array built by 225 scintillation counters arranged on a 7 m grid covering an area of ∼ 104 m2. Since the atmospheric conditions at Chacaltaya for Cherenkov measurements are optimal, a wide aperture Cherenkov array could complete this EAS detector. In the central part of the array an hadronic calorimeter should be installed. This calorimeter should meet the following conditions: a) enough thickness to realiably measure the energy of hadrons of up to 100 TeV, b) large enough area to have enough statistics, for primaries of at least 1016 eV, for example 100 m2, c) carpet detector on the top (streamer tubes or RPC counters) to measure the fine structure of the electromagnetic component of the associate shower, d) a tracking system in the top layers to measure the arrival direction of surviving hadrons, and e) fine-grained track- ing detectors in between layers, such as a silicon array, scintillating fibers or emulsion chambers. TheinstallationshouldalsohavemuondetectorsdistributedaroundtheEAS array and under the hadron calorimeter. . 31. Direct measurement of survival proton spectrum up to 100 TeV. – Up to the presentthe energyregionupto somehundreds ofTeV has beeninvestigatedby balloon- andsatellite-borneinstruments,andabovetheseenergiesonlybyground-basedairshower experimentswhicharerelativelyinsensitivetotheprimarynuclearcomposition. Thepro- posal below suggests how a direct measurement of the primary proton spectrum could be achieved. The surviving protons arriving at Chacaltaya suffer an attenuation given by N = N · o e−540/λ(E) where λ(E) is the nuclear interaction mean free path. The energy of such events is measured by the calorimeter and they are easily recognized by the lack of ac- companying particles detected either in the carpet or in the EAS array. In addition, during moonless nights no accompanying Cherenkov light must be detected with the Cherenkov array. This confirms that no absorbed (lower energy) shower in the upper atmosphere is associatedwith the surviving proton. Evenif this measurement is limited to lower proton energies and to a small fraction of events, the information that we can obtain is of crucial importance for backgroundrejection. 6 O.SAAVEDRAand L.JONES SinceonlyprotonscanreachtheChacaltayalevelwithoutinteractingintheatmosphere, the final result could be the measurement of the direct primary spectrum of protons provided, all the uncertainties, both in the inelastic proton air nuclei cross section and diffractionscatteringonairnucleibeminimized. Withtheproposedareaofthecalorime- ter,therewouldbe∼80events/yearforE ≥100TeV. Themeasurementinthisenergy p regionisofcrucialimportanceforthecalibrationoftheprotoncontentinprimarycosmic rays flux. The number of expected events per 100 m2 ·year·sr in the energy range 1-100 TeV is shown in Figure 4. The proposed direct measurement up to 100 TeV (or higher) is neededtocalibratethe indirectEASmeasurementmadewiththe samedetector atCha- caltaya. In addition, these measurements of protons with good statistics can be directly compared with balloon measurements. . 32.Pureevents.–”Pureevents”arethosethatinteractabout2collisionm.f.p. above the observation level and have not suffered the further complicated process of cascade development. A two nuclear interaction m.f.p. correspond to ≈ 4 radiation lengths (r.l.), i.e. only ≈ 2 km above Chacaltaya. Therefore, a very collimated hadronic jet is produced and ◦ the electromagnetic cascade originated by π ’s is at a very early stage of development. Therefore, this experiment provides a great opportunity to study jet production in the very forward direction. These events are impossible to study with collider detectors up to present. Fig.4.–Theintegralspectrumofsurvivalprimaryprotons(a),ofpureevents(b)andofprimary particles (c) at Chacaltaya per 100 m2·year From figure 4, it is seen that few events/year are expected at 1015 eV. These events are also due to pure protons contained in the primary cosmic ray spectrum. In order to distinguish those events from the normal EAS background we can use the information CHACALTAYA:TOWARDSASOLUTIONOFTHEKNEE....? 7 of the carpet detector as well as of the EAS array and (at night) the Cherenkov light detectors. In these events the electromagnetic cascade is at the very early stage of its development and would have a very steep lateral distribution; it should be well concen- trated within the 100 m2 carpet detector and not detected in the (104 m2) surrounding EAS array. From Figure 4 it is seen that about 100events per yearare expected at over 200 TeV. This would be a new approach to the study of the high-energy hadronic interactions through jet production in the very forward direction together with the associated elec- tromagnetic component. . 33. Multihadron surviving events. – The detection of multihadrons (≥ 2) by the calorimetercouldleadustostudytheprimarycompositioninasemi-directway. Infact, Chacaltaya is at only 540/λ(E) interaction m.f.p. and 14 r.l. from the top of atmo- sphere. Thereforethestudyofeventswithasimultaneousmeasurementsofthehadronic energy and multiplicity, the associated EAS and the Cherenkov light seems feasible at Chacaltaya. Moreover,specialtypes ofevents,likeCentauroandotherexoticeventslikeChirons[12] having a high hadronic multiplicity and high transverse momentum with no or low elec- tromagnetic component are still waiting to be understood. The fine granularity detec- tor (emulsion chambers, silicon detector planes or scintillating fibers) inserted into the calorimeter will measure such events. With silicon or scintillating fibers, these events could be detected in real time, together with the associated EAS event. Up to the present, such events have only been studied with emulsion technique. Ameasurementofspectraofsuchmultihadroneventswithenergiesupto100TeVwould be extremely helpful in order to better understand the phenomenology of these higher- energy interactions. A quantitative evaluation of such a study would be welcomed. . 34. Primary composition at the “knee” region. – As noted above, the “knee” in the energy spectrum holds a key to the understanding of the origin of cosmic ray. However, although a great effort has been dedicated to this problem over the past 40 years, con- fusion still reigns. Lindner [13] proposed a method based on the analysis of Cherenkov light and particle densities registered relatively close to the shower core; he shows that the b vs. X re- max lationship, (where b is the “slope” of the lateral distribution of the Cherenkov light and X (gr/cm2) is the depth of the maximum development of the Cherenkov shower), max does not depend on A, the primary mass, for a given model. Ontheotherhand,ProcureurandStamenov [14]haveintroducedanewparameterα(r) such that a shower selected with fixed values of this parameter would be generated by primaries with different masses but with the same primary energy. It is worthwhile to note that r for Chacaltaya is 35 m. Both methods pursue anunbiased determinationof the primaryenergy spectrum. How- ever, in order to significantly improve the unbiased determination of the primary mass composition it has to be done by adding observables related to the non-electromagnetic shower components. Since at Chacaltaya elevation the showers are at the minimum of their fluctuations for all components, with the proposed combined techniques (EAS and Cherenkov array andhadron-muondetector)itcouldbepossibletofacethe“knee”probleminadefinitive way. In particular, for lower energy showers the measurement will overlapdirect results 8 O.SAAVEDRAand L.JONES obtained by satellites or balloons. This possibility has a basic importance because, up to now, EAS data have not been calibrated against direct measurements. 4. – conclusions In this paper we have shownthat a possible experiment could be done at Chacaltaya that would have a great potential, thanks to its high altitude location. This presents a very great opportunity for cosmic ray physicists to exploit the unique conditions of this very high altitude mountain laboratory. Asuitabledesignofacollectionofdetectors,perhapsmodeledonthatsketchedhereand undertaken by an international collaboration, would demostrate the unique possibility to provide very significant cosmic ray physics results. ∗ ∗ ∗ Many people from several countries have contribuited to the revival of Chacaltaya Laboratory through their letters, conversations and discussions. O.S. would specialy like to thank the following: A. Watson, from Leeds University, E. Lorenz from HEGRA experiment,G.SchatzandH.RebelfromKASCADEexperiment,A.OhsawafromICRR, University of Tokyo, for their continued interest in this work about Chacaltaya. REFERENCES [1] N. Kawasumi et al., Phys. Rev. D, 53 (1996) 3634. [2] C. Aguirre et al., Phys. Rev. D, 62 (2000) 032003. [3] K. Werner,Phys. Rep., 232 (1993) 87. [4] N. Kalmykov and S. Ostapchenko, Yad. Fiz.,56 (1993) 105. [5] J.N. Capdevielle,J. Phys. G, 15 (1989) 909. [6] Y. Niihori et al., Phys. Rev. D, 36 (1987) 783. [7] D. Heck et al., FZKA report, 6019 (Fortshungszentrum Karlsruhe) 1998. [8] M. Risse et al. (KASCADE collaboration), 26th ICRC,1 (1999) 135. [9] A. Haungs et al., Nuovo Cimento, C (in press) . [10] L.W. Jones, Nucl.Phys. B (Proc. Suppl.), 75A (1999) 54 [11] G. Schatz,talk presented at the Cosmolep Conference at Sodankyla, Finland, 1999 (.) [12] S. Hasegawa, ICRR Report, 151-87-5 (1987) J.A. Chinellato et al. Prog. Theo. Phys. Suppl.,76(1983) seealsoC.Lattes,Y.FujimotoandS.HasegawaPhys. Reports, 65(1980) 152. [13] A. Lindner,Astrop. Phys., 8 (1998) 235. [14] J. Procureurs and J.N. Stamenov,J. Phys. G.,20(1994) 1665 and“Newpossibilities for the Chacaltaya array”, private communication, June 1999, presented also at the Chacaltaya Meeting, La Paz, Bolivia, July 23-27, 2000.