
2
II. PROPAGATION OF COSMIC RAYS AND
GAMMA RAYS
UHECRs can interact with cosmological photon fields
such as the EBL, as well as the cosmic microwave back-
ground (CMB) and the cosmic radio background (CRB)
during their journey to Earth. For a given cosmic-ray
nucleus A
ZX of atomic number Zand mass A, the main
photonuclear processes that affect them are: photopion
production (e.g., p+γbg →p+π0,p+γbg →p+π+),
Bethe-Heitler pair production (A
ZX + γbg →A
ZX + e++
e−), and photodisintegration (A
ZX + γbg →A−1
ZX + n,
A
ZX + γbg →A−1
ZX + p, etc). Unstable nuclei undergo
alpha, beta, and gamma decays during the photodis-
integration chain. An additional subdominant energy-
loss channel for photon production is elastic scattering
(A
ZX + γbg →A
ZX + γ). The by-products of these inter-
actions are the still-undetected [28] cosmogenic particles,
whose existence has been predicted long ago [29, 30].
Electrons and photons, too, interact during intergalac-
tic propagation. The main ones are pair production
(γ+γbg →e++e−) and inverse Compton scattering
(e±+γbg →e±+γ), although higher-order processes such
as double pair production (γ+γbg →e++e−+e++e−)
and triplet pair production (e±+γbg →e±+e++e−) may
also contribute at some specific energy ranges. These
processes feed one another, thus constituting an electro-
magnetic cascade (see ref. [31] for a detailed review).
In addition to the aforementioned interactions, all par-
ticles lose energy losses due to the adiabatic expansion of
the universe. Moreover, charged particles can emit syn-
chrotron radiation in the presence of magnetic fields.
Cosmogenic particles can be produced approximately
along the line of sight, leading to interesting obser-
vational signatures. Such model has been invoked,
for example, to explain gamma-ray observations from
blazars [32–35], and also GRB 221009A [36–38]. This di-
rectional correlation with sources can only occur if mag-
netic fields are not exceedingly strong. For a magnetic
field of coherence length LB, the deflection of a charged
particle after travelling a distance `is ∆δB'arcsin RL/`
if `LBand √2`LB/3RLotherwise [39], with Bdenot-
ing the strength of the magnetic field and RLthe Larmor
radius of the particle. The associated time delay is [40]:
∆tB'
`
c[1 −cos(∆δB)] if `LB,
`∆δ2
B
12cif `LB.
(1)
The main source of uncertainty in ∆tBare the proper-
ties of magnetic fields (see, e.g., refs. [41–43] for reviews).
UHECRs deflections cannot be properly estimated be-
cause IGMFs are unknown, especially in cosmic voids.
Nevertheless, even in the most extreme scenarios they
can be bounded [44].
III. SIMULATIONS
To interpret the observations, one-dimensional Monte
Carlo simulations are performed using the CRPropa
code [45, 46]. UHECRs are assumed to be emitted by the
GRB with spectrum dN/dE∝E−α, with an exponen-
tial suppression factor exp(−E/ZRmax) for E≥ZRmax.
Here Rmax denotes the maximal rigidity of the emitted
cosmic rays, given by Rmax ≡Emax/Z. A fraction ηCR
of the isotropic-equivalent energy of the GRB is assumed
to be converted into cosmic rays (of all energies). To en-
sure that the secondary fluxes arrive within .1 day of
the burst, magnetic deflections ought to be small. Since
IGMFs are poorly know, a conservative and an optimistic
scenario are considered, depending on the rigidity of the
UHECRs. This established a lower bound on the rigidity
of the primary UHECRs that will be considered.
The interaction processes described in section II are
included in the analysis. As a benchmark for the simula-
tions, the EBL model by Gilmore et al. [47] is chosen, as
well as the CRB model by Protheroe & Biermann [48].
Naturally, these choices have a considerable impact on
the results of the simulation, together with other factors
like photonuclear cross sections, for example [49, 50]. It
is beyond the scope of the present Letter to discuss these
uncertainties in detail. It is already enough to prove that
the proposed interpretation of GRB 221009A is justified
considering at least one realistic model.
IV. ANALYSIS AND INTERPRETATION OF
THE OBSERVATIONS
GRB 221009A drew much attention because the events
detected by LHAASO and Carpet-2, in principle, should
not have arrived at Earth due to pair production with
the EBL. If primary gamma rays were emitted, to first
order, this flux (Φ0) would be exponentially suppressed
by a factor corresponding to the optical depth (τ):
Φ(E, z) = Φ0(E, z) exp [−τ(E, z)] (2)
where zis the redshift. It is understandable that a spec-
trum extending up to ≈18 TeV from a source at z'0.15
could be seen as a signature of New Physics (see ref. [51]
for a discussion of some models), especially with a pos-
sible coincident event with E≈251 TeV. But many
plausible hypotheses remain within the realm of conven-
tional explanations. An UHECR origin is one of them,
provided that the requirements (i), (ii), (iii) are fulfilled.
To obtain a rough estimate for the time delay of pho-
tons from GRB 221009A for the scenario here proposed,
it is important to know its precise location within the
large-scale structure of the universe. The contribution
of the host galaxy of the GRB can be ignored given the
jet’s extension. For this reason, it is a good approxima-
tion to consider only the contributions of galaxy clusters,