A gamma-ray burst is like a shot from a cosmic cannon. First a brief flash of light, then a fading afterglow. But inside that flash, things aren't so simple: its color and brightness don't change at the same time.
Scientists took 20 single bursts caught by the Fermi satellite Fermi and broke them down over time. They looked at two things: when the burst becomes brightest and when its emission becomes hardest (that is, with the highest energy). It turns out there are three scenarios. In five bursts, hardness and brightness peaked together, as if someone switched on a lamp and immediately turned the brightness knob all the way up. In thirteen, hardness came before brightness, as if the lamp first switched to blue light and then got brighter. And in only two, it was the other way around: brightness before hardness.
This behavior is like three different springs. The first spring compresses and expands in sync with brightness — everything is synchronized. The second spring first becomes stiffer and then pushes harder. The third is rare: it's the opposite. Scientists don't yet know why, but this helps distinguish different radiation mechanisms in explosions.
as written by the authors arXiv:2604.17835
The properties of the spectral evolution during the prompt emission phase of gamma-ray bursts (GRBs), which are closely related to the radiation mechanism (synchrotron or photosphere), are still a subject of debate. Two spectral evolution patterns (``hard-to-soft'' and ``intensity-tracking'') have been commonly observed in GRB prompt emission spectra. Here we present a well-defined sample of 20 single-pulse GRBs detected by \emph{Fermi} whose prompt emission spectra exhibit the intensity-tracking pattern. By performing a time-resolved spectral analysis, we derive $$E_{\rm p}$$ and the energy flux $$F$$ from the same time bins and introduce a matched-bin lag, $$t_{\rm lag}^{\rm F} \equiv t_{\rm p}(E_{\rm p})-t_{\rm p}(F)$$, where $$t_{\rm p}$$ denotes the time at which each quantity reaches its maximum. We find that the intensity-tracking pattern subdivides into three distinct subclasses: Type I (5/20), with aligned $$E_{\rm p}$$ and flux peaks; Type II (13/20), with $$E_{\rm p}$$ peaking before the flux; and Type III (2/20), with $$E_{\rm p}$$ peaking after the flux. The early-peaking Type~II subclass dominates the sample. The subclasses also exhibit systematic differences in their spectral and temporal properties. Type II bursts are systematically harder than Type I, show broader flux pulses, and more often display asymmetric rising and decaying $$E_{\rm p}$$-$$F$$ branches. Type I is consistent with tightly coupled spectral and power evolution, whereas Type II is more naturally explained by nonthermal or hybrid prompt-emission scenarios in which spectral hardening precedes the peak radiative output. Type III appears to form a rare positive-lag tail whose physical origin remains uncertain.
The hardest gamma rays can have energies billions of times greater than visible light.