Direct Measurements, Acceleration and Propagation of Cosmic Rays

0.53 MB·English

by  Pasquale Blasi

30THINTERNATIONALCOSMICRAYCONFERENCE
DirectMeasurements,AccelerationandPropagationofCosmicRays
PASQUALEBLASI
INAF/OsservatorioAstrofisicodiArcetri
8
0 LargoE.Fermi5,50126Firenze(Italy)
0 [email protected]
2
Abstract: ThispapersummarizeshighlightsoftheOG1sessionofthe30thInternationalCosmicRay
n
Conference, held in Merida (Yucatan, Mexico). The subsessions (OG1.1, OG1.2, OG1.3, OG1.4 and
a
OG1.5)summarizedhereweremainlydevotedtodirectmeasurements,accelerationandpropagationof
J
cosmicrays.
9
2
Prologue Introduction

]
h
p The OG1 session was splitted in 5 subsessions, The origin of cosmic rays is a problem that has
- devoted to direct measurements of cosmic rays been haunting scientists for almost one century
o
r by balloons and satellites (OG1.1), cosmic ray now. Solvingthisscientificproblemmeansputting
t
s composition (OG1.2), cosmic ray propagation together numerous pieces of a complex puzzle, in
a (OG1.3),cosmicrayacceleration(OG1.4)andin- whichtheaccelerationprocesses,theinnerdynam-
[
trumentations and new projects (OG1.5). The ics of the sources, the propagation, the chemical

1 number of papers discussed in the OG1 session composition all fit together to provide a satisfac-
v gives a feeling of the huge number of results pre- toryandself-consistentglobalpicture. Wearenot
4
sented during the conference: there were 138 pa- thereyet,thoughanincreasinglylargernumberof
3
pers presented (66 of which were oral presenta- piecesarefindingtheirplaceinthepuzzle. TheIn-
5
4 tions). About37oftheoralpresentationsreported ternationalCosmicRayConferenceisaprivileged
. on observational results, while about 29 reported placetoobservethepuzzletakingshape.
1
0 ontheoreticalorphenomenologicalwork. Itisob- This rapporteur paper is organized as follows: in
8 viousthatameaningfulsummaryofthisworkim- §3 I present a subjective bird’s eye view of how
0 plies a selection of highlights. An apology is due thingsstandinthefield,inordertomakeiteasier
v: to all those that will not see their work properly for the reader to put in the right context the ex-
i discussed here. It should also be understood that tremelywiderangeofresultspresentedattheCon-
X
in order to avoid doing a simple list of the results ference. In§4Isummarizesomeimportantobser-
r
a itisneededtoputthingsincontext,andthisoften vational results about spectra and chemical com-
requiresweighingtheresultsaccordingtotheopin- position of cosmic rays (including the electronic
ionsofthewriter,whichdonotcoincidenecessar- component) that have been presented. In §5 I il-
ilywiththeopinionsoftheCommunityatlarge. I lustrate some recent developments in the under-
hope that this may be a starting point for further standing of the acceleration of cosmic rays. In
discussiononthetopicsthatIwilltouchuponbe- §6 I discuss the problem of relating the observed
low. cosmicraystothesourcesthroughpropagationin
theGalaxy. Thetransitionfromthegalacticcom-
ponent to the extragalactic cosmic ray component
is briefly discussed in §7 together with some is-
suesrelatedtothepropagationofultrahighenergy
ORIGINANDPROPAGATIONOFCOSMICRAYS
cosmic rays in the Galaxy and in the intergalactic the material which is accelerated and the actual
medium. Iconcludein§8. timewhenitisacceleratedtocosmicrayenergies.
Current direct measurements are filling the gap
to the knee region, thereby providing a way to
Abird’seyeviewoncosmicrays
match and cross-check the indirect measurements
of spectrum and composition carried out by ob-
Thebestknownpropertyofcosmicraysistheirall-
servingandmodellingextensiveairshowersaten-
particlespectrum.Mostexperimentsagree,atleast
ergies across and above the knee. The data col-
qualitatively,thatthespectrumconsistsoffourre-
lected by the KASCADE experiment suggest that
gions: 1) at low energies (below ∼ 10 GeV) the
the knee in the all-particle spectrum may be an
observed spectrum is flat as it is affected by so-
artifact of the superposition of sharper knees in
lar modulation. 2) For energies 10GeV ≤ E ≤
the single components, mainly in the light chem-
3×1015eV, thespectrumcanbefitwithapower
icals, such as H and He. The proton spectrum
law with slope ∼ 2.7. 3) At energies between
measured by KASCADE has a pronounced knee
3 × 1015GeV and ∼ 1018eV the slope grows to
at energy ∼ 106 GeV. A knee in the He compo-
∼ 3.1. 4) The upper boundary of region 3 is not
nent is also seen, but the fluxes become more un-
welldefined,soIwilltakesomelibertiesindefin-
certain while moving to larger masses. It is not
ing region 4. In terms of Physics this reflects our
knownasyetwhatcouldbetheexplanationforthe
uncertaintiesindefiningtheregionoftransitionto
knees in these components, though it is plausible
extragalacticcosmicrays(see§7).
thattheymayreflectinefficiencyoftheaccelerator.
Thechemicalcompositionofcosmicraysisacru- Inastrictlyrigidity-dependentapproach,akneein
cialpieceofinformationandisbecomingwellde- theironspectrumcouldbeexpectedat∼ 3×107
terminedatenergiesbelowtheknee, wheredirect GeV. This simple argument opens the way to an
measurements can still be carried out, while it is equally simple but far reaching implication: the
moreuncertainandmodeldependentathigheren- spectrumofGalacticcosmicraysshouldendaten-
ergies.Thecompositionoflowenergycosmicrays ergies around ∼ 1017 eV, so that cosmic rays of
provides important hints to the acceleration pro- largerenergiesmustbeacceleratedinextragalactic
cessesandthepropagationofcosmicraysthrough sources. UnfortunatelytheKASCADEresultsare
the interstellar medium (ISM). Especially impor- notfullyconsistentwiththoseofsomeotherexper-
tantinthisrespectaretheabundancesandspectra iments, especiallytheTibetarray. Themaindiffi-
ofelementssuchasBoron,BerylliumandLithium, culty of all ground based experiments, including
which are mainly produced as secondaries of pri- those operating in the ultra high energy range, is
mary cosmic rays. The ratio of secondary to pri- the poor understanding of cosmic ray interactions
mary(forinstanceB/C)cosmicrayfluxesprovides intheatmosphere,whichmaybeveryproblematic
a unique tool to characterize the diffusion proper- toinferbasicinformationasthechemicalcompo-
tiesoftheISM.Existingmeasurementsofthisra- sitionoftheprimaryparticles.
tio as a function of energy suggest that the diffu-
Three sets of measurements will help us figure
sioncoefficientscaleswithenergyasD(E)∝Eα,
out whether the physical picture suggested by the
withα≈0.6,atleastatrigiditiesbelow∼10GV,
KASCADEdataiscorrect(assumingthatthedata
whileitisnotclearwhetherathigherenergiesthe
themselves find proper confirmation): 1) direct
sloperemainsconstantorthereisaflattening.
measurementsofthecosmicrayspectraandchem-
ElementssuchasGeandGacanpotentiallyallow ical composition should extend as far as possible
ustodiscriminatebetweenvolatilitybasedandfirst towards the knee and possibly across it; 2) addi-
ionization potential based acceleration processes, tionaldataandreliablemodelsforthedescription
andthereforeprovideinformationaboutthedomi- oftheshowerdevelopments(possiblycheckedver-
nantaccelerationsites. Otherelements(e.g. 59Co susfutureLHCdata)arecrucialatenergiesaround
and 59Ni) also provide us with precious informa- and above the knee; 3) the chemical composition
tion on the mean time between the production of intheenergyregionbetween1017eVand1019eV
needstobemeasuredreliably.
30THINTERNATIONALCOSMICRAYCONFERENCE
Fortunately, important steps ahead in all these di- ical elements have been presented, from balloon
rectionsarebeingdoneandsomeimportantresults flights (CREAM, ATIC, Tracer, TIGER, Bess-
willbediscussedbelow. Polar,PPB-BETS),fromsatellites(preliminaryre-
What about the sources? The paradigm based on sults from PAMELA) and even some shower ex-
supernova remnants (SNRs) as the main sources periments (Tibet Array and HESS). The overlap
of galactic cosmic rays remains the most plau- between all these techniques in the energy region
sible, but the smoking gun that would turn the around the knee appears to be of crucial impor-
paradigm into a well established theory is still tance, especially to unveil the origin of the knee
missing. The most important news in this direc- inthecosmicrayspectrum.
tion is represented by the recent observations of
SNRsingammaraysbyCherenkovimagingtele- BalloonsandSatellites
scopes (see [31] and references therein for a re-
TheCREAMCollaborationhaspresentedimpres-
cent review) and the high resolution observations
sive results collected during a total of 70 days of
of the X-ray emission from the rims of several
flight of their balloon experiment. The data were
SNRs(see[59]andreferencesthereinforarecent
collected during the record breaking CREAM-I
review). Theformerhavepresenteduswithsome
flight (42 days between 12/16/05 and 01/27/05)
evidencesofTeVgammarayemission,possiblyof
and a second flight (CREAM-II) lasted 28 days
hadronic origin. The latter have provided us with
(between12/16/05and01/13/06). Convincingevi-
strongevidenceofefficientmagneticfieldamplifi-
dencewaspresentedoftheexcellentchargeresolu-
cation, which is in turn required in order to reach
tionoftheexperiment(∆q ∼0.2electroncharge),
thehighenergiesobservedincosmicrays,andare
precioustooltoreconstructthespectraofdifferent
aconsequenceofefficientcosmicrayacceleration.
chemicals(see[50,23]foratechnicaldiscussion).
Fromthetheoreticalpointofview,therehavebeen
In Fig. 1 I reproduce the plots as shown by [55]
severalnewdevelopmentsinourunderstandingof
(see also [63]) illustrating the spectra of protons
the mechanism of diffusive shock acceleration in
andheliumnucleiasmeasuredbyCREAM-I(red
SNRs but also related to the propagation of cos-
circles). The CREAM spectra are compared with
micraysbothintheGalaxyandintheintergalac-
those of other experiments as listed in the figure
ticmedium. Iwilldiscusssomeoftheseissuesin
withdifferentsymbols. ThespectrumofHenuclei
moredetailbelow.
as measured by CREAM is compatible with that
measuredbyATIC-2,butappearstobesomewhat
NewMeasurements flatterthanthatmeasuredbyATIC-1.
The spectra of Carbon and Oxygen measured by
The direct measurements of the flux of cosmic CREAM-II were presented by [65] and [2]. The
rays and of their chemical composition, carried results on the C/O ratio confirm the primary na-
out by using balloons and satellites, has always ture of both nuclear species [51]. An overview
playedacrucialroleinadvancingourunderstand- oftheCREAMresultsandfuturedevelopments(a
ing of both acceleration and propagation of cos- thirdandfourthflight)werepresentedby[55].The
mic rays, at least at energies below the knee. At spectraofthechemicalspresentedbytheCREAM
higher energies the low fluxes due to the steeply collaboration do not show appreciable differences
falling spectrum make it necessary to use ground intheslope,withthepossibleexceptionofthehe-
arrayswhichobservecosmicrayinducedairshow- lium spectrum that might be slightly harder than
ers.Thetwotechniquesareclearlycomplementary theprotonspectrum. Infactthistrendisprobably
and one of the biggest problems has always been present even in the ATIC-2 spectrum. In Fig. 1
to cross-calibrate the two. For the first time this the shaded area is supposed to show the range of
goal seems to be at least in sight, in that the di- fluxes as measured by ground arrays. The width
rect measurements are extending to ∼ 10 − 100 of the shaded region provides an estimate of the
TeV, therefore approaching the knee region. Nu- uncertainties due to either the interaction models
merous new results on spectra of different chem- adopted for the analysis or systematics in the dif-
ORIGINANDPROPAGATIONOFCOSMICRAYS
Figure1: Spectraofprotons(leftpanel)andheliumnuclei(rightpanel)asmeasuredbyCREAM-I(figure
from[55])comparedwiththeresultsofotherexperiments.
ferent experiments. An important point that this
figureshowsisthatfinallydirectandindirectmea-
surements are starting to overlap in the knee re-
gion. This will be of great importance to under-
standthereasonfortheappearanceofthekneein
theall-particlecosmicrayspectrum.
The ATIC balloon experiment has presented im-
pressiveresultsontheB/Cratioandmasscompo-
sition of cosmic rays below the knee. In fact the
ratiosN/OandC/Owerealsomeasured. TheB/C
ratio,asIdiscussin§6,isacrucialindicatorofthe
mode of cosmic ray propagation in the interstel-
lar medium. Low energy measurements (E < 30
GeV) carried out by HEAO-3 [30] show that the
B/Cratioscaleswithrigidityas∼ R−0.6, usually
interpreted as a proof of the primary (secondary)
originofCarbon(Boron)nuclei.
TheratioasmeasuredbyATIC-2isreportedinFig.
2[49]togetherwiththepreviousresultsofHEAO-
3 and the predictions of two theoretical models
(lines) (see [49] for some discussion and refer- Figure2: B/Cratioasafunctionofenergypernu-
ences). Despitetheapparentflatteningofthemea- cleon as measured by ATIC-2 (squares) [49] and
suredB/CratiointheATICdata,theerrorbarsare comparedwiththeresultsofHEAO-3andthepre-
stilltoolargetoinfersolidconclusionsonthisis- dictionsoftwotheoreticalmodels(lines).
sue.
The all-particle spectrum and the average mass
number as a function of energy as measured by
ATIC-1areshowninFig. 3[1]comparedwiththe
resultsofotherexperiments,asindicatedinthefig-
ure.Thereisasubstantialagreementamongallthe
data sets shown. At least in part the small offsets
30THINTERNATIONALCOSMICRAYCONFERENCE
may be explained in terms of systemaic errors in
the energy determination, amplified by the multi-
plicationbyE2.5.
The mean logarithmic mass number found by
ATICintheenergyrange102 −105GeVshowsa
generaltrendtoaheaviercompositionandappears
tomatchwellwiththeresultsofRUNJOB,CASA-
BLANCA, DICE and KASCADE in the knee re-
gion. At energies above the knee (not reached by
ATIC) the other experiments show quite different
trends. This prevents us from reaching a satisfac-
tory explanation of the origin of the knee and of
thetransitionfromgalactictoextragalacticcosmic
rays(see§7).
Veryaccuratemeasurementsofthefluxofdifferent
chemicalelementsbelowthekneewerepresented
bytheTRACERCollaboration[21]andareshown
in Fig. 4 (filled dots) for nuclei between Oxy-
gen and Iron. The fluxes are multiplied by differ-
entnormalizationfactorstomaketheplotclearer, Figure4:Differentialspectraofchemicalelements
andtheyarecomparedwiththeresultsofHEAO-3 from Oxygen to Iron as measured by Tracer [21]
and CRN. No appreciable difference between the andcomparedwithpreviousresultsfromCRNand
slopesofthespectraofthesenucleiwasdetected, HEAO-3.
allslopesbeingaround∼2.7.
The TRACER Collaboration also presented the
however recall that a standard solar system com-
B/C ratio in the energy region around ∼ 100
positionneartheacceleratorisallbutgranted.
GeV/amu, being fully consistent with other mea-
Importantinformationontheoriginofcosmicrays
surements. The predicted error bars on the B/C
and their propagation can also be indirectly gath-
ratio with a 30 days balloon flight of TRACER
eredbyobservingelectrons. Theirpropagationin
should allow to finally pin down the slope of the
the Galaxy is affected by diffusion, as for ions,
ratio as a function of energy up to ∼ 1 TeV/amu.
andenergylosses(mainlysynchrotronandinverse
While indicators such as the B/C ratio provide
Comptonscatteringlosses).Atenergieslargerthan
us with important information on the propagation
20-100 GeV the loss time due to ICS and syn-
of cosmic rays in the Galaxy, the abundances of
chrotronbecomesshorterthanthediffusiveescape
super-heavy elements (Z > 30) tell us about the
time from the Galaxy, so that the spectrum steep-
accelerationregions,thoughtheirfluxismorethan
ensbyonepowercomparedwiththesourcespec-
∼ 1000 times smaller than the Iron flux. The
trum. At energies larger than ∼ 1 TeV the flux
flux of these elements has been measured during
ofelectronsatEarthisexpected(andobserved)to
twoballoonflightsbytheTIGERCollaborationin
decline,presumablyasaresultofthediscreteness
2001and2003(atotalofabout50daysofobserva-
in the spatial distribution of the sources. In this
tions). Thefluxesintheenergyregionbetween0.3
senseseveralgroupshavebeenengagedinasearch
and 10 GeV/nucleon are reported in Fig. 5 (from
forapossiblesignaloftheclosestsourcesofelec-
[54]). The lines represent the predictions of two
tronsaroundthesolarsystem,themostpromising
modelsforacceleration,onebasedonfirstioniza-
sources being Vela, the cygnus loop and Mono-
tionpotential(FIP)andthesecondonvolatility,but
gem. Theexpectedsignalwouldconsistofabump
both with a standard solar system composition of
inthediffusefluxofelectronsatenergiesinexcess
theambientmedium.Theabundancesof31Gaand
of ∼ 1 TeV. The proximity of the source would
32Geappeartobeinconsistent(iftakenatthesame
also result in a small anisotropy in the arrival di-
time) with both theoretical models. One should
ORIGINANDPROPAGATIONOFCOSMICRAYS
Figure3: Allparticlespectrum(leftpanel)andmeanlogarithmicmassnumber(rightpanel)asmeasured
byATIC[1].
Figure6:Spectrumofelectrons(multipliedbyE3)
asobservedbyPPB-BETS[64]comparedwiththe
resultsofpreviousmeasurements.
rections. Thespectrumandanisotropieshavebeen
presentedbyPPB-BETS.
Todate,thespectrum[64](Fig. 6)doesnotshow
evidence for the possible appearance of a nearby
electronsource. Eventheanisotropyisclaimedto
befullyconsistentwithanisotropicdistributionof
Figure 5: Relative abundances of heavy elements
arrivaldirections[64].
asobservedbytheTIGERexperiment[54].
TheBESS-Polarexperimenthaspresentedseveral
interesting results, ranging from low energy cos-
mic ray spectra to spectra of antiprotons and lim-
its to the flux of anti-helium nuclei. The flux of
cosmic ray protons in the energy region below
∼ 10−30GeVisheavilyaffectedbysolarmod-
30THINTERNATIONALCOSMICRAYCONFERENCE
tiproton/proton flux is compared with predictions
ofadriftmodelwithdifferenttiltanglesoftheso-
larmagneticfield.Thethickdotsandemptycircles
referagaintoBESS-PolarandBESS-1997respec-
tively.Intherightpaneltheeffectofadifferentso-
larmodulationinthetwoperiodsisevidentagain.
As expected, this effect almost completely disap-
pearsintheratiooftheantiproton/protonflux.The
fluxesofantiprotonsobservedbyBESS-1997and
BESS-Polararebothfullyconsistentwiththeone
expectedonthebasisofapropagationmodelwith
solarmodulation,therebyimplyingsignificantup-
per limits to the flux of antiprotons from, for in-
stance,neutralinoannihilationintheGalacticdark
matterhalo. BESShasalsofocusedonthedetec-
tionofanti-nuclei. Thequestforwhyouruniverse
is(almost)completelymadeofmatterinsteadofa
Figure 7: Spectrum of protons as observed by
mixture of matter and anti-matter is a fundamen-
BESS-1997(emptycircles)andBESS-Polar(filled
tal one, as Nature is expected to have produced a
circles)[32].Thedifferencebetweenthetwospec-
(almost)symmetricaluniverse. Thoughitislikely
traisduetothedifferentlevelsofsolarmodulation
that the small excess of matter over anti-matter is
atthetimeofthemeasurements.
whatnowmakesmostofthepresentuniverse,the
search for islands of antimatter or traces of anti-
ulation. InFig. 7Ireporttheresultspresentedby nucleihasneverquitefinished. Thelimitimposed
[32]: thethickdotsaretherecentresultsofBESS- by BESS-Polar and the limit that will possibly be
Polar, compared with those of BESS-1997 (open reachedbythenextBESSflightareshowninFig.
circles),atthetimeofthesolarminimum. Itisev- 9, together with the results of previous measure-
ident the effect of the enhanced solar modulation ments.
while the sun gradually approaches the next so- Themeasurementofthecosmicrayspectrumand
lar minimum. The upper solid line represents the chemical composition up to energies of ∼ 1 TeV,
spectrum of protons in the interstellar medium as the search for dark matter, anti-matter and ex-
inferredfromBESS-98measurements. otic particles, and finally bits of solar and mag-
The propagation of cosmic ray nuclei in the in- netospheric physics are among the goals of the
terstellar medium also results in the production PAMELA satellite, which has been succesfully
of many secondary products (nuclei, antiprotons, launched on june 15, 2006. The instruments on-
electron-positron pairs, gamma rays, neutrinos). boardarecollectingdata,butonlypreliminaryre-
Accurate measurements of these secondaries, es- sults were presented at the conference. A review
pecially antiprotons and positrons is instrumental of PAMELA preliminary results and expectations
tothesearchforfaintsignals,suchasthosecoming has been presented in [52]. Particularly interest-
from the annihilation of non-baryonic dark mat- ing,andsuggestiveofthequalityoftheresultsthat
ter. The flux of anti-protons and the ratio of anti- we should expect, is the spectrum of protons and
protonstoprotonsintheenergyregionbelowfew helium,reportedinFig. 10. Thispreliminaryanal-
GeV as measured by BESS-Polar is reported in ysis does not show evidence for any difference in
Fig. 8. Intheleftpanel,theantiprotonspectrumis theslopesofthespectraofHandHeintheenergy
comparedwithsometheoreticalpredictionsofthe regionbelow∼500GeV.Inbothcasesthebestfit
samequantitiesforsomemodels(solidline:propa- powerlawhasaslopeof2.73.
gationwithGALPROP;dottedline:standardleaky Several simulated results were also presented by
boxwithsolarmodulation;dash-dottedline: black the PAMELA Collaboration, mainly aimed at
holeevaporation). Intherightpanel, theratioan- showingthepotentialfordiscoveryofindirectsig-
ORIGINANDPROPAGATIONOFCOSMICRAYS
Figure8: LeftPanel: AntiprotonspectrummeasuredinBESS-PolarcomparedwithBESS(95+97)results.
Thelinesillustratepredictionsofdifferenttheoreticalscenarios. Thelinesthatpassthroughthedatapoints
arethepredictionsofstandardpropagationcalculations(GALPROPandleakyboxmodels). RightPanel:
p¯/pratiomeasuredbyBESS(95+97)andBESS-PolarI(2004)comparedwithdriftmodelcalculationsat
variousSolarmagneticfieldtiltangles.
Figure10: PreliminaryspectraofprotonsandheliumfromPAMELA.
30THINTERNATIONALCOSMICRAYCONFERENCE
Figure 11: Spectrum of iron nuclei (in fact nuclei
withZ >24obtainedbyHESSusingthedetection
ofDClight. TheHESSresults(withSYBILLand
QGSJETasmodelsforinteractions)arecompared
withtheresultsofothermeasurements.
theatmosphere.Thistechnique,proposedasapos-
sibletooltomeasurethechemicalcompositionof
cosmicraysabove10TeVby[58], hasbeenused
Figure 9: Limits on the flux of anti-Helium ob-
byHESStomeasurethefluxofheavynuclei(Fe-
tainedbyBESS-Polar[32].
like). The direct Cherenkov light is produced by
theprimarynucleuswhileenteringtheatmosphere
nals of dark matter annihilation (in the form of and before the first interaction that generates the
anomalous features in the antiproton and positron shower. High up in the atmosphere the density
spectra),andthelimitsachievableinthesearchfor of air is lower and therefore the Cherenkov cone
anti-matter. The limit on the ratio H¯e/He is ex- is more collimated with respect to the Cherenkov
pected to improve by about 1 order of magnitude cone of the light generated by lower energy parti-
in about three years of PAMELA operation with clesintheshower. AsaresultthetotalCherenkov
respect to that achieved by BESS-Polar (see Fig. emissionfromashowerisexpectedtohaveabroad
9). region with a hot pixel corresponding to the DC
PAMELA is expected to play a crucial role in the light. This emission has in fact been success-
identification of spectral features in the positron fullydetectedbyHESS.Sincetheintensityofthe
and antiproton spectra as they can possibly result Cherenkov signal scales with the square of the
from dark matter annihilation, provided the mass chargeoftheparentnucleus,itisbesttoseachfor
ofthedarkmattercandidateparticleisintheright thesignalfromFenucleiprovidedtheenergydoes
energyrange(below500−1000GeV). not exceed a few hundred TeV, in order to avoid
that the shower emission overshines the DC light.
ThespectrumofIronnucleiasmeasuredbyHESS
GroundExperiments
is reported in Fig. 11 [22] where it is compared
Since most results from ground experiments were withtheresultsofotherexperiments. Thetwosets
presented in other sessions of the ICRC, the talks ofdatapointsrefertoQGSJETandSIBYLLasin-
onthissubjectintheOG1sessionweresparse. teractionmodels.
The HESS Collaboration has presented the re- The search for the DC light is also being pursued
sults of a very interesting detection of the direct byVERITAS[62],whilededicatedinstrumentsare
Cherenkov(DC)lightfromcosmicraysimpacting beingplanned(e.g. TrICE[61]).
ORIGINANDPROPAGATIONOFCOSMICRAYS
The HESS Collaboration has also presented the Oneofthemostrecentadvancementsinthetheory
preliminary results of their measurement of the of particle acceleration at non-relativistic shocks,
electron spectrum [29], consistent with measure- especially for SNR shocks, has been the calcula-
mentscarriedoutbyotherexperimentswithinthe tion of the dynamical reaction of the accelerated
systematicuncertainties. particles onto the background plasma. The effect
Arelevantcontributiontotheunderstandingofthe has been discussed already in the 80’s in the con-
nature of the knee in the all-particle spectrum has text of two-fluid models (see [44] for a review),
been provided by Tibet Array. The Collaboration andlateraddressednumericallybydirectsolution
haspresentedthemostrecentmeasurementsofthe of the time dependent equation of diffusive tran-
protonandheliumspectrainthekneeregion, and port coupled with the equations of mass, momen-
thefractionofheavynucleiasafunctionofenergy, tum and energy conservation of the background
intherange106 ≤E ≤107GeV. plasma(seeforinstance[11]).Inthelate90’sasta-
tionarysemi-analyticalsolutionofthissetofequa-
The measurements of the chemical abundances in
tionswasfoundintheformofanintegralequation
thisenergyregionarestillcontroversial. TheTibet
[43]foraspatiallyconstantdiffusioncoefficient.A
dataontheprotonspectrumappearinroughagree-
general semi-analytical method was recently pro-
ment with those of RUNJOB and JACEE (both
posedby[8], validforanychoiceofthediffusion
with QGSJET and SIBYLL as interaction mod-
coefficient.
els) but in rather apparent contradiction with the
KASCADEdata,especiallywhentheKASCADE Thestructureoftheshockischangedbytheaccel-
dataareanalyzedusingSIBYLL.Thelattershowa erated particles due to the pressure they exert on
steeperspectrumofprotonsandasomewhathigher the background plasma, though in a collisionless
normalization of the flux. The discrepancies are manner. Forordinarydiffusioncoefficients,which
even moreevident in the caseof the helium spec- are increasing functions of particle momentum,
trum. The Tibet results on helium roughly match higherenergyparticlescantravelfurtherawaythan
with those of RUNJOB, but not with data from low energy particles: a fluid element approach-
JACEE and KASCADE. The absolute flux of he- ingtheshocksurfacethereforefeelsanincreasing
liumasmeasuredbyKASCADEisalmostoneor- pressureduetoacceleratedparticlesandasacon-
derofmagnitudehigherthanthatmeasuredbyTi- sequence it slows down. This leads to the forma-
betArray. Thesituationinthekneeregioncontin- tionofaprecursorupstreamoftheshock(whichis
uestobeconfusedandthispreventsthepossibility nowcalledsubshock)inwhichthefluidvelocity,as
ofunveilingtheoriginoftheknee. seen in the (sub)shock reference frame decreases
while approaching the shock. The effective com-
pressionfactorfeltbyparticlescrossingtheshock
AccelerationofCosmicRays andprecursorisnowafunctionofmomentum,be-
ing bigger at large momenta and smaller at low
The main mechanism for the acceleration of cos- momenta. Theimmediateconsequenceisthatthe
micrays(notonlygalactic)remainsdiffusivepar- spectrumofparticlesacceleratedatshocksisnota
ticle acceleration at shocks. Many contributions power law, as expected in test-particle theory, but
presented at the conference have focused on the ratheraconcavefunctionofmomentum,harderat
determinationoftheefficiencyofparticleacceler- high momenta and softer at low momenta. This
ation,thedynamicaleffectoftheacceleratedparti- nonlinearsystemisfoundtoself-regulateitselfand
clesandthemagneticfieldamplificationgenerated tobeabletoreachrelativelyhighefficiencyofpar-
by the accelerated particles through streaming in- ticleacceleration.
stabilityorduetoturbulentamplification. This complex nonlinear system has however ad-
I will start by discussing some issues related to ditional interesting aspects, concerning the maxi-
the acceleration of Galactic cosmic rays in SNRs, mum momentum that the particles can be accel-
though some of the conclusions will be of wider erated to. As was recognized first in [39, 40],
applicability. the maximum momentum achieveable in SNRs is
exceedingly low compared with the knee energy