---
title: Synchrotron-Dominated Cascades Explained
url: https://www.emergentmind.com/topics/synchrotron-dominated-cascades
type: topic
---

# Synchrotron-Dominated Cascades Explained

Searching arXiv for relevant papers on synchrotron-dominated cascades and related cascade formalisms.
Synchrotron-dominated cascades are electromagnetic cascade regimes in which the secondary $e^\pm$ produced by high-energy photons or hadrons lose energy primarily through synchrotron radiation rather than inverse Compton scattering. Across blazar environments, compact $\gamma$-ray–opaque sources, intense magnetic fields, and magnetar magnetospheres, this regime arises when magnetic energy density and kinematic conditions force rapid synchrotron cooling of cascade pairs, yielding distinctive photon spectra, angular signatures, temporal behavior, and, in some cases, strong polarization. In the literature summarized here, the term encompasses several closely related but physically distinct settings: ultra-high-energy cosmic-ray–induced cascades in magnetised large-scale structure around blazars [1406.6075], time-dependent synchrotron-dominated electromagnetic cascades in relativistic jets [2001.04649], linear internal cascades in compact opaque sources [2509.00152], one-photon pair cascades in strong magnetic fields [1805.03040], and pair-synchrotron cascades in ultra-strong magnetar fields [2508.11552].

## 1. Defining physical regime

The common structural element is a feedback loop between pair creation and synchrotron emission. A high-energy primary photon, proton, or electron injects secondary $e^\pm$; these pairs radiate synchrotron photons; and those photons may in turn generate additional pairs if the local opacity permits it. The cascade is therefore controlled by the competition among pair injection, radiative cooling, escape, and further conversion.

In the blazar-cosmic-web scenario, ultra-high-energy cosmic rays (UHECRs; primarily protons) accelerated in a blazar jet escape into a magnetised region of size $R\sim$ a few Mpc and field strength $B\sim10^{-7}$ G, where they interact with CMB and EBL photons through Bethe–Heitler pair production and photomeson production [1406.6075]. The immediately produced secondary electrons and positrons of energy $E\gg m_e c^2$ then lose energy almost exclusively by synchrotron radiation in the local field. This establishes a cascade whose observable output is a synchrotron pair halo or echo.

In compact-source internal-cascade theory, the defining assumption is a $\gamma$-ray–opaque region of radius $R$ containing a dense target photon field $n_t(\epsilon_t)$ and a magnetic field $B$ such that electron losses are dominated by synchrotron rather than inverse Compton scattering. The source is “linear,” meaning cascade particles interact only with the fixed background $n_t$, and one requires $\tau_{\gamma\gamma}\equiv t_{\rm esc}\Gamma_{\gamma\gamma}\gg1$ so that the cascade is fully developed [2509.00152].

In intense magnetic fields, the regime changes from ordinary $\gamma\gamma$-driven cascades to one-photon pair production plus quantum synchrotron emission. A high-energy photon of energy $\epsilon_0$ propagating across a magnetic field $B$ can be absorbed if
\[
\chi_\gamma \equiv (\epsilon_0/2m_e c^2)\cdot(B_\perp/B_{\rm cr}) \gtrsim 1,
\]
with $B_{\rm cr}=4.414\times10^{13}$ G, producing an $e^+e^-$ pair whose members radiate high-energy photons via synchrotron emission and continue the cascade [1805.03040].

In magnetar magnetospheres, the cascade is initiated by resonant inverse Compton scattering (RICS) of thermal X-rays by relativistic particles on closed magnetic loops. The resulting photons are attenuated by one-photon pair production and photon splitting, with the synchrotron emission of the secondary pairs dominating the emergent sub-MeV spectrum for most observer angles [2508.11552].

A general regime criterion appears explicitly in compact-source treatments: synchrotron dominates if
\[
t_{\rm syn}(\epsilon_e)<t_{\rm IC}(\epsilon_e)
\quad \Leftrightarrow \quad
u_B>u_t(\epsilon\ {\rm in\ relevant\ band}),
\]
and, in jet applications, Klein–Nishina suppression of inverse Compton cooling can strengthen synchrotron dominance at high $\gamma$ [2001.04649, 2509.00152].

## 2. Transport equations and cascade closure

The electron distribution in synchrotron-dominated cascades is usually described by kinetic transport equations. For UHECR-induced blazar cascades, the differential number density $n_e(E,t)$ obeys
\[
\frac{\partial n_e(E,t)}{\partial t}
+ \frac{\partial}{\partial E}\Bigl[b(E)\,n_e(E,t)\Bigr]
= Q_{\rm pair}(E,t)
-\frac{n_e(E,t)}{\tau_{\rm esc}(E)} \, ,
\]
where $b(E)=|\dot E|$ is the total energy-loss rate, $Q_{\rm pair}(E,t)$ is the $e^\pm$ injection rate from $p\gamma$, and $\tau_{\rm esc}(E)\sim R/c$ is the escape timescale or diffusion time. In steady state and for slow escape,
\[
\frac{d}{dE}\bigl[b(E)\,n_e(E)\bigr]
=Q_{\rm pair}(E) \, .
\]
This form makes explicit that the steady lepton population is set by a balance between continuous synchrotron cooling and injection [1406.6075].

The corresponding synchrotron loss rate is
\[
b_{\rm syn}(E)
=\left|\frac{dE}{dt}\right|
=\frac{4}{3}\,\sigma_T\,c\,
\frac{B^2}{8\pi}\,
\Bigl(\frac{E}{m_e c^2}\Bigr)^2 ,
\]
and the characteristic synchrotron photon energy is
\[
\varepsilon_{\rm syn}
\simeq \frac{3e\hbar\,B}{2\,m_e c}\,\gamma^2,
\quad
\gamma=\frac{E}{m_e c^2} .
\]
For first-generation $e^\pm$ from photomeson production with $E_e\sim10^{19}$ eV in $B\sim10^{-7}$ G, one finds $\varepsilon_{\rm syn}\sim{\rm few}\times10^{11\mbox{--}12}$ eV, i.e. in the TeV band [1406.6075].

In time-dependent jet models, the evolution is written as a Fokker–Planck equation including stochastic acceleration:
\[
\frac{\partial N_e(E,t)}{\partial t}
=
\frac{\partial}{\partial E}
\Bigl[
D(E)\,\frac{\partial N_e}{\partial E}
+
b(E)\,N_e(E,t)
\Bigr]
+
Q_{\rm pri}(E,t)
+
Q_{\gamma\gamma}(E,t)
-
\frac{N_e(E,t)}{\tau_{\rm esc}} .
\]
Here $D(E)$ is the energy-diffusion coefficient, $Q_{\rm pri}$ the primary injection, and $Q_{\gamma\gamma}$ the secondary-pair source term from photon-photon annihilation. For the stochastic-acceleration “hard-sphere” case,
\[
D(E)=K E^2
\quad \Longrightarrow \quad
t_{\rm acc}(E)=\frac{E^2}{D(E)}=\frac{1}{K},
\]
so reacceleration applies equally to primary and secondary pairs [2001.04649].

In the linear compact-source theory, the steady-state electron and photon equations are given, in the $\delta$-function approximation, by
\[
\partial_{\epsilon}\,[b_{\rm syn}(\epsilon_e)n_e(\epsilon_e)]
+4\,n_\gamma(2\epsilon_e)\Gamma_{\gamma\gamma}(2\epsilon_e)=0
\]
and
\[
\frac{U_B\sigma_T m_e}{2\omega_B}
\sqrt{\frac{3}{4\omega_B\epsilon_\gamma}}\,
n_e\!\Bigl(m_e\sqrt{\frac{3\epsilon_\gamma}{4\omega_B}}\Bigr)
-\Gamma_{\gamma\gamma}(\epsilon_\gamma)n_\gamma(\epsilon_\gamma)=0.
\]
These equations codify, energy bin by energy bin, the balance between synchrotron cooling and pair injection on the lepton side, and between synchrotron photon production and pair-production losses on the photon side [2509.00152].

This suggests that “cascade closure” can be formulated in a unified way across environments: one needs sufficiently rapid pair creation to replenish the radiating lepton pool and sufficiently rapid synchrotron cooling to transfer energy downward before escape or competing channels remove it.

## 3. Numerical methodologies and approximations

The numerical treatment depends strongly on the astrophysical setting. In UHECR-seeded blazar cascades, propagation is handled with CRPropa 2 in 3D mode. Interaction lengths for $p\gamma$ processes on the CMB, EBL, and radio backgrounds are sampled using SOPHIA for photo-pion production and the Kelner & Aharonian parametrisation for Bethe–Heitler pair production. Deflections in the structured magnetic field, with coherence $\lambda_{\rm coh}\sim0.4$ Mpc, are tracked analytically. Every time a secondary photon or $e^\pm$ is produced, its position and momentum are recorded, and particles pointing within the observer cone ($\sim11^\circ$) are collected. The collected leptons are then propagated by solving the kinetic equation on the fly, including continuous synchrotron losses, inverse Compton scattering when relevant, and further pair production above threshold; the code of Murase & Dermer implements the fully coupled cascade [1406.6075].

In the 3C 279 flare model, the simulations are explicitly time-dependent and one-zone. The synchrotron-dominated electromagnetic cascade appears in a stochastic-acceleration scenario in which secondary pairs from $\gamma\gamma$ annihilation are reaccelerated in the same emitting region. Klein–Nishina suppression is incorporated in the inverse Compton loss term through either $\sigma_T\to\sigma_{\rm KN}(\gamma)$ or a suppression factor
\[
f_{\rm KN}(\gamma)\simeq \frac{3}{8x}\bigl[\ln(2x)+1/2\bigr]
\quad
{\rm for}\quad
x=4\gamma h\nu_{\rm ext}/(m_e c^2)\gg1,
\]
so that
\[
b_{\rm IC}(E)
=
\frac{4}{3}\sigma_T c\,U'_{\rm ext}\,\gamma^2\,f_{\rm KN}(\gamma).
\]
This treatment is essential because the synchrotron-dominated state is reached precisely where inverse Compton losses are suppressed at high Lorentz factor [2001.04649].

In magnetar calculations, Monte Carlo methods propagate “weighted” photons emitted from RICS spectral bins in the tangent direction to the local field. Along each photon path, opacities to one-photon pair production and photon splitting are accumulated, and a rejection-sampling scheme determines whether the photon splits, pair produces, or escapes. Secondary photons, whether split or synchrotron, are followed recursively until all weight escapes. For synchrotron emission from pairs created in excited Landau states, the code uses the classical form
\[
P(\nu)=\sqrt{3}\,(e^3B/m_e c^2)\,F(\nu/\nu_c)
\]
in the “circular” frame when appropriate, but switches to full QED rates in supercritical fields $(\gamma B'>0.1)$ [2508.11552].

In strong-field one-dimensional magnetic cascades, a dedicated Monte Carlo code is used to study the spectrum after the cascade reaches saturated status, when almost all the energy of the primary particles transfers to photons. The resulting spectral energy distribution is then fit analytically as a function only of the product $R B_\perp$ [1805.03040].

A recurring methodological theme is the hybridization of event-based Monte Carlo interaction sampling with deterministic or semi-deterministic kinetic evolution of secondaries. This is especially pronounced in the blazar-halo framework, where proton interactions are stochastic while lepton cooling is treated through transport equations [1406.6075].

## 4. Spectral structure and universal behaviors

Several distinct universal or quasi-universal spectral patterns appear in synchrotron-dominated cascades, but they are not identical across regimes.

For linear internal cascades in compact opaque sources, a key analytic result is the emergence of two photon power-law regimes when the injection is effectively monochromatic at $\epsilon_{\gamma,\rm inj}\gg\epsilon_{\rm thr}$. At high photon energies $\epsilon_{\gamma,\rm syn}<\epsilon_\gamma<\epsilon_{\gamma,\rm inj}$,
\[
n_e(\epsilon_e)\propto \epsilon_e^{-3}
\quad \Rightarrow \quad
n_\gamma(\epsilon_\gamma)\propto \epsilon_\gamma^{-2},
\]
where $\epsilon_{\gamma,\rm syn}\simeq \omega_B\gamma_{\rm thr}^2$ and $\gamma_{\rm thr}=\epsilon_{\rm thr}/m_e$. At lower photon energies $\epsilon_\gamma<\epsilon_{\gamma,\rm syn}$, in the cooling-only regime,
\[
n_e(\epsilon_e)\propto \epsilon_e^{-2}
\quad \Rightarrow \quad
n_\gamma(\epsilon_\gamma)\propto \epsilon_\gamma^{-3/2}.
\]
The approximate cascade spectrum can therefore be written as
\[
n_\gamma(\epsilon_\gamma)\simeq
\frac{L_{\gamma,\rm inj}\,t_{\rm esc}}
{2+\ln(\epsilon_{\gamma,\rm thr}/\epsilon_{\gamma,\rm syn})}
\times
\min\!\Bigl[
(\epsilon_\gamma/\epsilon_{\gamma,\rm syn})^{-2},
(\epsilon_\gamma/\epsilon_{\gamma,\rm syn})^{-3/2}
\Bigr].
\]
The reported simulations confirm the analytic indices $s_\gamma=2$ above $\epsilon_{\gamma,\rm syn}$ and $s_\gamma=3/2$ below it in AGN coronae, GRB internal shocks, some compact blazar models, and TDE environments [2509.00152].

By contrast, saturated cascades in intense magnetic fields exhibit a different universal broken-power-law structure. Defining the normalized SED
\[
f(\chi)\equiv \epsilon^2\,dn/(\epsilon_0\,d\epsilon),
\qquad
\chi\equiv \epsilon B_\perp/(m_e c^2 B_{\rm cr}),
\]
the shape below the cutoff depends only on $R\cdot B_\perp$ and is fit by
\[
f(\chi)=
\biggl[
(a\sqrt{\chi})^{-4}
+
(0.323\,\chi^{0.125})^{-4}
\biggr]^{-1/4}
\quad
{\rm for}\quad \chi\lesssim0.95\,\chi_c,
\]
with
\[
\chi_c=\frac{2.67}{2.30\,\log_{10}[R\,B_\perp/({\rm cm\,G})]-13.74},
\qquad
a=0.225\,\log_{10}[R\,B_\perp/({\rm cm\,G})]-0.67.
\]
The break is at $\chi_b\simeq0.38\,\chi_c^2$. Below $\epsilon_b$, one has $f\propto\chi^{0.5}$; between $\epsilon_b$ and $\epsilon_c$, $f\propto\chi^{0.125}$; and above $\epsilon_c$ the spectrum displays a very sharp cutoff, much steeper than a simple exponential [1805.03040].

In UHECR-driven structured-region cascades, the spectrum is described as a broad bump peaking at $\sim1$–10 TeV with the high-energy cutoff set by EBL absorption. Below the peak, the slope is governed by the $p\gamma$ injection spectrum and synchrotron cooling, with $dN/dE\propto E^{-s}$ and $s\sim1.5$–2 [1406.6075]. In the 3C 279 stochastic-cascade model, the resulting electron distribution is described as very hard and nearly flat, $N_e(\gamma)\sim\gamma^{-2}$, feeding back into a flat $\nu F_\nu$ continuum from X-rays to GeV $\gamma$ rays, with a cooling break at $\sim2$ MeV and a sharp cutoff above $10$ GeV from internal $\gamma\gamma$ absorption and EBL attenuation [2001.04649].

These results caution against treating “synchrotron-dominated cascade spectrum” as a single universal template. The asymptotic form depends on whether the cascade is linear or nonlinear, whether it is driven by $\gamma\gamma$ absorption or one-photon magnetic conversion, whether the synchrotron process is classical or QED, and whether the photon field is external, internal, or negligible.

## 5. Source classes and phenomenology

The best-developed astrophysical application in the provided material concerns extreme TeV blazars. For 1ES 0229+200, RGB J0710+591, and 1ES 1218+304, synchrotron emission of UHECR secondaries produced in blazars embedded in structured regions with magnetic field strengths of order $10^{-7}$ G is presented as an alternative to both standard leptonic scenarios and inverse-Compton cascade channels [1406.6075]. The observed specific flux at Earth is written as
\[
F_{\nu_0}
=
\frac{1+z_s}{4\pi D_L^2(z_s)}
\int_V dV \int dE_e\;
P_{\nu'}(E_e)\,n_e(E_e)\;
e^{-\tau_{\gamma\gamma}(\nu_0,z_s)},
\]
with $\nu'=\nu_0(1+z_s)$ and an obvious generalization to radially varying $B(r)$. The total synchrotron luminosity in the halo is approximated as
\[
L_{\rm syn}\simeq \chi_{\rm em}L_{\rm UHECR},
\]
where $\chi_{\rm em}\lesssim1$ is the fraction of proton energy converted through $p\gamma$ interactions within the few-Mpc region [1406.6075].

The angular extent and time-delay phenomenology are specific and potentially diagnostic. The parent pairs are deflected over their cooling length by an angle
\[
\theta_e\simeq D_{\rm syn}/r_L
\sim10^{-4}(E_e/10^{19}{\rm eV})^{-2}(B/10^{-7}{\rm G})^{-1},
\]
so the halo is $\lesssim0.1^\circ$ for sources beyond $\sim100$ Mpc, implying a point-like image for current IACTs but a resolvable source for CTA at TeV energies. The delay of the synchrotron echo is
\[
\Delta t\sim\frac{D_{\rm syn}\theta_e^2}{2c}
\sim0.1\mbox{--}1\ {\rm yr}
\times
\Bigl(\tfrac{\varepsilon_{\rm syn}}{1\,{\rm TeV}}\Bigr)^{-2}
\Bigl(\tfrac{d}{1\,{\rm Mpc}}\Bigr),
\]
which yields month-to-year echoes [1406.6075].

In the 3C 279 June 2015 flare, the synchrotron-dominated electromagnetic cascade appears in a distinct jet context. The stochastic-acceleration solution uses $\Gamma=100$, $R_0=7.1\times10^{16}$ cm, $B'=0.1$ G, $K=4.2\times10^{-4}\,{\rm s}^{-1}$, primary monoenergetic injection at $\gamma'_{\rm inj}=10$, and one-sided jet luminosities $L_\gamma\sim10^{49}$ erg s$^{-1}$, $L_e\sim10^{47}$ erg s$^{-1}$, and $L_B\sim10^{44}$ erg s$^{-1}$ [2001.04649]. The observed maximum synchrotron photon energy follows from
\[
t_{\rm acc}=t_{\rm syn}
\quad \Longrightarrow \quad
\gamma'_{\max}
=\sqrt{\frac{6\pi e}{\xi\,\sigma_T\,B'}}
\]
and
\[
E_{\rm syn,max}
=
\Gamma\,
\frac{3eB'}{4\pi m_e c}\,
\gamma_{\max}'{}^2
\simeq
24\,
\frac{\Gamma}{100}\,
\frac{1}{\xi}\ {\rm GeV},
\]
with $\xi\approx3$ in Model C, implying $E_{\rm syn,max}\sim{\rm few}\times10$ GeV [2001.04649].

In magnetars, the relevant observables shift to hard X-rays and polarization. For a single loop with maximum radius $r_{\max}=4R_{\rm NS}$, surface field $B_0=(10$–$50)\,B_{\rm cr}$, and $\gamma_0=10^2$–$10^3$, the secondary synchrotron component dominates at 5–200 keV when RICS losses are included, with spectral index $\simeq1.5$ and cutoff at $0.3$–$0.6$ MeV, while split photons fill the MeV window up to $\sim3$ MeV. The synchrotron photons are predicted to yield net polarization 40%–80% in $\perp$ mode below the SR cutoff, depending on viewing angle, with the possibility of a mode reversal when split photons dominate at higher energies [2508.11552].

## 6. Parameter dependence, robustness, and comparisons with inverse-Compton cascades

A central result of the blazar-halo scenario is that the generated synchrotron pair halo or echo flux at the peak energy is not sensitive to variations in the overall IGMF strength. The reason given is geometrical and environmental: filaments occupy only $\sim10\%$ of void volume, so cascade photons produced far outside the structured region suffer large deflections in void IGMFs and are isotropised, suppressing their contribution to the beamed signal. The TeV synchrotron emission is therefore determined almost entirely by the few-Mpc, $B\sim10^{-7}$ G region around the source and is insensitive to weaker fields in voids [1406.6075].

The same work reports several robustness tests. Varying the filament magnetic field between $10^{-8}$ and $10^{-6}$ G shifts the synchrotron bump energy only within the observed TeV range. The absolute flux at peak depends mainly on the isotropic UHECR luminosity, $L_{\rm cr}\sim10^{46\mbox{--}47}$ erg s$^{-1}$, but not on the void field strength. Changing the UHECR injection index from $\alpha=2.0$ to $2.3$ requires only modest $\lesssim$ factor-few adjustments in $L_{\rm cr}$ to fit the same data. Scans in maximum energy from $10^{20.5}$ to $10^{21}$ eV show that $E_{p,\max}\gtrsim10^{20.5}$ eV is required to reproduce the highest-energy tail above $\gtrsim10$ TeV after EBL attenuation, and the fit is only weakly sensitive to the choice among current EBL models [1406.6075].

In the 3C 279 jet model, the defining comparison is instead with prompt-injection or reconnection-based synchrotron scenarios. A too strong magnetic field yields a too bright synchrotron X-ray flux due to secondary electron–positron pairs, whereas the stochastic-cascade scenario operates in a very low-magnetization regime, with $U_B/U_e\sim10^{-4}$ at $2R_0$, and uses KN suppression of IC to force synchrotron-dominated cooling of the highest-energy particles [2001.04649]. This indicates that synchrotron-dominated does not necessarily imply magnetically dominated in the dynamical sense; it may instead arise from radiative suppression of the IC channel.

In generalized linear-cascade theory, universality is contingent rather than automatic. Under sufficiently high $B$ and $u_B\gg u_t$, a synchrotron cascade with $n_\gamma\propto\epsilon_\gamma^{-2}$ can emerge in the optical–X-ray band in compact blazar environments. However, in many blazar models Bethe–Heitler pair injection over a broad range or partial $\gamma$ attenuation spoils the universality [2509.00152]. A plausible implication is that observationally inferred departures from the idealized $-2$ and $-3/2$ indices need not falsify the synchrotron-cascade mechanism itself; they may instead indicate non-monochromatic injection, incomplete opacity, mixed cooling, or additional pair-production channels.

The comparison with inverse-Compton cascades is recurrent. In voids, inverse-Compton pair echoes are highly sensitive to the poorly known intergalactic magnetic field: for $B\lesssim10^{-16}$ G delays are $\lesssim$ months, but for $B\gtrsim10^{-14}$ G delays become $\gg$ years and flux is suppressed. By contrast, the structured-region synchrotron signal is described as unavoidable and robust against void-field uncertainty [1406.6075]. In compact-source theory, inverse Compton losses must be subdominant throughout the cascade for the synchrotron-universal spectrum to appear [2509.00152]. In strong magnetic fields and magnetars, inverse Compton is not the competing closure process at all; instead, the competition is with magnetic pair creation and photon splitting [1805.03040, 2508.11552].

## 7. Conceptual distinctions and common misconceptions

One common source of confusion is the use of the same label for physically different cascade architectures. In extragalactic blazar environments, the cascade may be seeded by UHECR interactions on the CMB and EBL and radiate in $\sim10^{-7}$ G structured regions, producing TeV synchrotron halos or echoes [1406.6075]. In compact jets, the cascade can instead be an internal $\gamma\gamma$-driven process with reacceleration of secondary pairs in a one-zone emission region [2001.04649]. In intense magnetic fields, the cascade is driven by one-photon pair production and quantum synchrotron emission rather than by $\gamma\gamma$ absorption [1805.03040]. In magnetars, photon splitting is a major participant in the cascade and shapes both spectrum and polarization [2508.11552]. These are not merely different parameter choices of one theory.

A second misconception is that synchrotron-dominated cascades necessarily produce the same spectral slope. The material summarized here shows at least three distinct asymptotic families: the $n_\gamma\propto\epsilon_\gamma^{-2}$ and $\epsilon_\gamma^{-3/2}$ branches of linear internal cascades [2509.00152]; the broken SED with indices $0.5$ and $0.125$ in the saturated strong-field QED cascade, expressed in $f(\chi)$ rather than directly in $n_\gamma$ [1805.03040]; and the broad TeV bump or flat $\nu F_\nu$ continuum characteristic of specific source applications [1406.6075, 2001.04649]. The distinction reflects differences in kernel, opacity, geometry, and observable variable.

A third misconception is that synchrotron dominance requires extreme magnetic field strength in an absolute sense. The blazar-halo framework achieves TeV synchrotron emission at only $B\sim10^{-7}$ G because the pairs are ultra-relativistic, with $E_e\sim10^{19}$ eV [1406.6075]. The 3C 279 stochastic-cascade model operates with $B'=0.1$ G, but KN suppression renders inverse Compton inefficient at the relevant energies [2001.04649]. By contrast, the strong-field and magnetar cases truly involve QED-scale fields where one-photon pair creation and photon splitting become central [1805.03040, 2508.11552].

Finally, the observational appearance of synchrotron-dominated cascades is not uniformly extended or slowly varying. In structured extragalactic regions, the halo can be $\lesssim0.1^\circ$ and point-like for current IACTs, with month-to-year echoes [1406.6075]. In the 3C 279 flare model, the same broad physical category accommodates minute-scale variability, with $t_{\rm var}\simeq R_0/(c\Gamma^2)\sim2$ min [2001.04649]. In magnetars, the distinguishing diagnostics are instead hard-X-ray tails and polarization fractions of 40%–80% in the keV band [2508.11552].

Taken together, these results establish synchrotron-dominated cascades as a broad class of radiation-mediated pair cascades whose detailed manifestation is controlled by magnetic geometry, opacity, injection channel, and radiative hierarchy. The unifying principle is rapid synchrotron cooling of secondary pairs; the diversity lies in how those pairs are generated, how the cascade closes, and which observables carry the imprint.

Source: https://www.emergentmind.com/topics/synchrotron-dominated-cascades