CREDO project
Robert Kamiński, Tadeusz Wibig, David Alvarez Castillo, Kevin Almeida Cheminant, Aleksander Ćwikła, Alan R. Duffy, Dariusz Góra, Piotr Homola, Paweł Jagoda, Marcin Kasztelan, Marek Knap, Konrad Kopański, Peter Kovacs, Michał Krupiński, Marek Magryś, Vahabeddin Nazari, Michał Niedźwiecki, Wojciech Noga, Matias Rosas, Szymon Ryszkowski, Katarzyna Smelcerz, Karel Smolek, Jarosław Stasielak, Sławomir Stuglik, Mateusz Sułek, Oleksandr Sushchov, Krzysztof Woźniak
CCREDO project ∗ Robert Kami´nski a , Tadeusz Wibig b , David Alvarez Castillo a , Kevin Almeida Cheminant a ,Aleksander ´Cwik(cid:32)la c , Alan R. Duffy d , Dariusz G´ora a , Piotr Homola a , Pawe(cid:32)l Jagoda e ,Marcin Kasztelan f , Marek Knap g , Konrad Kopa´nski a , Peter Kovacs h , Micha(cid:32)l Krupi´nski a ,Marek Magry´s i , Vahabeddin Nazari j , Micha(cid:32)l Nied´zwiecki c , Wojciech Noga a , Mat`ıasRosas k , Szymon Ryszkowski g , Katarzyna Smelcerz c , Karel Smolek l , Jaros(cid:32)law Stasielak a ,S(cid:32)lawomir Stuglik a , Mateusz Su(cid:32)lek g Oleksandr Sushchov a , Krzysztof Wo´zniak a a Institute of Nuclear Physics PAS,Radzikowskiego 152, Krak´ow, b University of Lodz,Faculty of Physics and Applied Informatics,90-236 (cid:32)L´od´z, Pomorska 149/153, Poland, c Cracow University of Technology, Krak´ow, Poland d Swinburne University of Technology,Melbourne, Australia e AGH University of Science and Technology,Krak´ow, Poland f National Centre for Nuclear Research,Otwock-´Swierk, Poland g CREDO Collaboration, Poland h Wigner Research Centre for Physics,Budapest, Hungary i Academic Computer Center Cyfronet,AGH University of Science and Technology,Krak´ow, Poland j JINR Dubna, Russia k CREDO Collaboration, Uruguay l IEAP, Czech Technical University in Prague, Czech Republic
The Cosmic-Ray Extremely Distributed Observatory (CREDO) is a project created a few yearsago in the Institute of Nuclear Physics PAS in Krak´ow and dedicated is to global studies of extremelyextended cosmic-ray phenomena. The main reason for creating such a project was that the cosmic-ray ensembles (CRE) are beyond the capabilities of existing detectors and observatories. Until now,cosmic ray studies, even in major observatories, have been limited to the recording and analysisof individual air showers therefore ensembles of cosmic-rays, which may spread over a significantfraction of the Earth were neither recorded nor analyzed. In this paper the status and perspectivesof the CREDO project are presented.
I. INTRODUCTION
The main goal of the CREDO project is to combine existing cosmic-ray detectors (largeprofessional arrays, educational instruments, individual detectors e.g. smartphones, etc.) intoone worldwide network, enabling global analysis of both individual cosmic showers and CREpresented on Fig. 1. A very important aspect of this project is geographical spread of thedetectors and access to big man power necessary to deal with vast amount of data to searchfor evidence for cosmic-ray ensembles. This is due to the fact that the apple of the eye of thisproject and an integral part of the chosen research method is citizen science enabling the useof work and the involvement of a huge number of people who are not necessarily scientists.How a CRE can be produced? A good candidate is a shower induced by an ultra high energy ∗ Presented at Matter To The Deepest Recent Developments In Physics Of Fundamental Interactions XLIIIInternational Conference of Theoretical Physics, September 2019, Katowice, Poland a r X i v : . [ a s t r o - ph . I M ] S e p photon interacting at least at some distance from the Earth e.g. close by the Sun. Such cascades,which are called preshowers [1], are a consequence of an interaction of ultra high energy photons(UHE, with energies larger than 10 eV) with solar or terrestrial magnetic fields. When thecascade arrives at the Earth, it comprises several thousand photons and leptons with a peculiarspatial distribution to which CREDO will be sensitive [2]. Another possible mechanism of theproduction of CRE can be decay of long-lived super-massive particles (M > eV) and maylead to a significant fraction of UHE photons [3] in cosmic ray flux and thus CRE.CREDO proposes operation under a planetary network, what is necessary for signatures ofCRE which may be spread over very large surface ( ≈ km). The components of CREmight have energies that practically span the whole cosmic-ray energy spectrum. Thus, all thecosmic ray detectors working in this range, beginning from smartphones and pocket scintillators,through numerous larger educational detectors and arrays to the professional infrastructure thatwill receive cosmic rays as a signal or as a background could contribute to a common efforttowards a hunt for CRE.In July 2019 number of institutions which are part of the CREDO collaboration was 23located in 11 countries [4]. Important is number of active users of CREDO application (forsmartphones), Fig. 2 presents state of art in July 2019. FIG. 1: Left: Typical strategy: search for one extensive air shower (EAS) - cascade of secondaryparticles initiated by a single high-energy cosmic ray; Right: Cosmic-Ray Ensembles: a novelty incosmic-ray research and target for CREDO.
II. METHOD
Analysis of scenarios for the origin of cosmic particles that can be verified on Earth throughan observation of an ensemble of at least two particles or photons with a globally spread andcoordinated network of detectors requires special and well-developed methods. The generalCREDO strategy includes searches for groups of spatially correlated cosmic-ray photons thatmight arrive at the Earth at significantly different times, with temporal dispersion of the orderof minutes or more. Such phenomena have been reported in the literature [5, 6], but they have
FIG. 2: The map of CREDO user locations as of middle July 2019. More than 7500 users spread overthe globe not been observed repeatedly until now. Of course spatial correlations of particles arrivingsimultaneously at the Earth must be tested at the very beginning.Cosmic-ray data are available everywhere (also inside buildings) they are free, and they canbe acquired with the minimum detection effort in an easy reach of the public - through a mobiledevice equipped with an application that turns it into a particle detector. CREDO is going touse the already existing applications, the CREDO Detector [7], with the source code open underthe MIT license that gives a freedom of development driven even by a wide community, andunlimited flexibility to implement all the features essential for the project. The other uniquefeature of the app is that it connects the user to the open server-side data processing, analysis,and visualization system [8], with open access to receive data. It concerns also detectors otherthan smartphones, and opens access to the data in close to real time.All the CREDO related codes developed so far are available on GitHub repository [9]. Itenables an easy teaming up and continuation of the existing projects together with activeparticipation and engagement of non-experts.Since a smartphone with the CREDO Detector is capable of detecting particle track can-didates (see Figure 3 for examples) including both the cosmic radiation (e.g. the penetratingparticles like muons) and the local radiation (e.g. X-rays) it opens a door to the public en-gagement model which is based not only on the data classification, like for most of the citizenscience projects, e.g. in zooniverse.org, but also on the data acquisition.Figure 2 shows the map of the CREDO user’s locations, from which one can see that CREDOsmartphones network already spreads over the planet. The number of registered users, with atleast one detection, in the middle of July 2019 was 7500 and about 2 918 000 images were storedin the database. The observing time for all users equates to 958 years searching for particles,which shows the large potential of such observations.The unique scientific advantages that the smartphone cloud has in comparison to other detec-tor systems is the geographical spread and the public availability at no additional investment.These two features make the smartphone cloud a critically important and efficient componentof the whole CREDO system.
FIG. 3: Tracks of particles detected by the CREDO Detector application
III. FIRST SCIENTIFIC RESULTSA. Do CREDO smartphones really register single cosmic ray muons?
Among the images in the CREDO database there are photographs of events where they arevisible very long traces which are supposed to be images of tracks of cosmic ray muons thatpassed the matrix of the smartphone camera with large angles. The distribution of arrivingzenith angles of single, incoherent muons is known for years. It is still a subject of measurementsby the small, even portable muon telescopes often created and operated by students [10–13].The data used to obtain the results presented below consist of about 5% of registrations fromthe CREDO database collected in a long series on several different phones by two Collaborationmembers. In the database they are available as PNG files, cut from the whole camera frame to60 ×
60 pixels box around the lightest pixel in the whole frame.The crucial step in the analysis is to determine the main symmetry axis of the potential track.First all pixels that exceeded the average noise (by ten times its average dispersion) are chosenand for them the main axis of the track is determined. There are, in principle, many variouspossibilities to find it. CREDO tested inertia ellipse, the Hough algorithm line, the smallestsum of squared distances weighted by squares of the brightness or by just the brightness. Theyall gave slightly different results, but the images of the tracks in the pictures are clear andregardless of how they are linearized, the large axis of the track is almost always the same.Divergences sometimes appear in ”multi-track” cases. The distribution of the track length isshown as the histogram in Fig. 4.
FIG. 4: Measured track length distribution measured. The lines show comparison with the predictionsfor the various of the value of camera matrix height (h): 3 pixels – dotted line, 5 pixels – continuousline, 10 pixels – dotted line and 15 pixels dotted line.
The track length l relates to the zenith angles of incoming particles:Θ = ArcT an ( l/h ) (1)where h is the depth of the active layer of the matrix in the smartphone camera. The thicknessof this layer is not exactly known and the cameras used for recording were different leaving thevalue of a relatively free parameter which effective value can by adjusted using the data. Themuon zenith angle distribution is known for 80 years and many different measurements confirmthat it can be accurately described as dN (Θ) d Θ = cos γ (Θ) (2)with the index of γ ∼ .
0. Fig. 4 shows that the values of above a dozen pixel sizes or soare rather unacceptable (if we are measuring real cosmic ray muons), because they are notable to explain the observed large number of short traces. For the value of h = 3 one can seethe opposite: the number of long traces seen is definitely too large. The value of the camerathickness of ∼ B. CREDO: EAS measurement
However, the most interesting and attractive would be studying the phenomenon of ExtensiveAir Shower (EAS), a cascade of elementary particles, mostly photons and electrons, but alsosome muons or even high energy hadrons traveling almost at the speed of light from the upperatmosphere to the surface of the Earth. They arrive as a disk of millions of particles for oneshort instance. The source of very high energy particles that initiated such showers is, ingeneral, not known as well as the mechanism of their acceleration from astrophysical sources.The mystery of (high energy) cosmic rays has stood for almost a century and observing andstudying these Extensive Air Showers can be exciting and stimulating for the young minds. TheCREDO EAS array, called CREDO-Maze, was constructed and the prototype has registeredits first showers. The concept of the CREDO-Maze array was developed based on the 20years old Roland Maze Project [14, 15]. The technology today has developed greatly and thelocal shower array idea of Linsley (1985) [16] can now be implemented much more easily and,critically, much more cheaply. Eventually CREDO wishes to present high-schools with sets of atleast four professional cosmic ray detectors connected locally and forming the small school EASarray. These will use plastic scintillators instead of smartphone cameras, bespoke fast electronicsinstead of the smartphone application, and a more convenient connection to the wider CREDOdatabase. But this is not all. CREDO established some physical arrangements that can be usedby physics teachers in the standard physics education course showing properties of elementaryparticles, radiation attenuation, effect of interaction of particles with matter etc. There aremany historical experiments (dating back to the 40s) that can be repeated in the classroom orduring after hours activities. The final design of the small local EAS array is still in development,but to test the working principle CREDO have used small detector available on the market:the CosmicWatch Desktop Muon Detector. One set of Cosmic Watch contains two small (5cmx 5cm) scintillators monitored by silicon photomultipliers (SiPM) and slow electronics based onan Arduino microcontroller, that allows students to connect them to the computer and analyzesimultaneous registration of signals from both detector laying one above the other, which flagsmuon passing through both scintillators. CREDO have used two CosmicWatch sets, that is fourindividual scintillation detectors, bypassed the CosmicWatch electronics and used the raw SiPMsignal. Four available individual detector signals into 6 pares of fast coincidence units have beedcombined. The coincidence window width of 100 ns was used. This short time ensured that thesignal rate from uncorrelated muon (and other noise sources) was of order of tens per minuteto ensure that non-EAS signals will not disrupt the shower registrations. The logic of the arrayis such that if any coincidences appear in all 6 outputs, the data is immediately stored in afast register and the trigger starts the slower electronics (based on an Arduino microcontroller).This last stage reads the register and transfers all 6 bits to the computer where they are storedtogether with the actual time stamp in a file for further analysis.
C. Measurements
Before using the array to search for an EAS, it was first tested to be sure that the registeredevents were real. The first test was to place four detectors in one tower arrangement. TheCosmicWatch detectors are expected to give a signal for single muons. The threshold was setfor all detectors individual signal forming amplifier/comparators on the same level of about 30mV chosen by comparing CosmicWatch original counting rate with the signal rate. The singlemuon passing at least two detectors can trigger the first and the second detectors, looking fromthe top of the tower and nothing more, leaving the tower from the side: (1, 2), can travers(1,2,3), (1,2,3,4), but also (2,3), (2,3,4) and (3,4). Other possibilities for the ideal geometry areforbidden. The extremely low count for forbidden coincidence (with the gap between the fireddetectors) configurations gives confidence that in the majority of cases CosmicWatch detectorin the array respond correctly to the passage of the single relativistic charged particle.
FIG. 5: Rate of different coincidences recorded in the tower geometry
The asymmetry between first (1,2) and sixth (3,4) configurations, and (1,2,3) and (2,3,4)gives as a signal that the detector thresholds are not set perfectly. The number of counts shownin Fig. 5 suggests that the detector number 1 seems to have the threshold set a little too high.However, this imperfection does not affect the general reasonable behavior of the whole array.After the test, four detectors in the plain with irregular (not rectilinear) configuration of thestep of order of 1 meter were arranged. Different geometries were tested without significantdifferences. The particles which could produce coincidence in the mini array have to come inone instant (within 100 ns) at the surface of about 1 m . Considering the small effective area ofCosmicWatch detectors, the particle density of particles in such event has to be rather high, or ifit is actually low, the rate of such low-density events should be very high. Respective simulationcalculations were performed. Millions of test densities according to one of the known examplesof measured shower particle densities available on the market [17] were generated. For eachdensity was checked if each single detector was triggered or not.The single hits were, of course, lost in the background of single muons counts. In spite of thenormalization, which is most problematic here, the ratios of three-fold to two-fold and four-foldto two-fold coincidences rate are observables which are independent on the overall normalization(actual solid angle, shielding etc.) and depend mostly on the density spectrum index.The mini-array was active for just over a week, and during this time 94 two-fold coincidences,2 three-fold and one all four hits event were registered.The number of more than 2-fold coincidences is much too low, even if taking into accountabsolutely insignificant statistics. To perform an accurate measurement it was necessary tohave detectors which respond in the same way to particle traversing the scintillator, which isnot exactly this case (see Fig. 5). Nevertheless, the comparison of the simulation result withmade measurement gives confidence that the CREDO-Maze mini array registered real ExtensiveAir Showers, and one can say that it has been shown that the CREDO-Maze mini array usingCosmicWatch detectors registered first Extensive Air Showers. IV. CONCLUSIONS
Pursuing the research strategy proposed in CREDO project will have a large impact onastroparticle physics and possibly also on fundamental physics. If CRE are found, they couldpoint back to the interactions at energies close to the Grand Unified Theories (GUT) scale. Thiswould give an unprecedented chance to test experimentally for example dark matter models. IfCRE are not observed it would valuably constrain the current and future theories. Apart fromaddressing fundamental physics questions CREDO has a number of additional applications:integrating the scientific community (variety of science goals, detection techniques, wide cosmic-ray energy ranges, etc.), helping non-scientists to explore Nature on a fundamental but stillunderstandable level. The high social and educational potential of the project gives confidencein its contributing to a progress in physics.
Acknowledgments