The Fluorescence Detector (FD) of the Pierre Auger Observatory has been detecting elves with a dedicated trigger since 2013, and its 100 ns time resolution enables unprecedented precision in resolving their internal structure. Relative to other transient luminous event (TLE) detectors, this capability allowed the identification of more multi-ring events, which constitute about 23% of the elves observed at Auger. In a recent publication, we present evidence that the time gap between these multiple rings remains constant, suggesting that the current understanding of their generation mechanism may be incomplete and that further investigation is required.
Elves are TLEs characterised by expanding rings of light, triggered when the electromagnetic pulse from an intense lightning discharge excites ionospheric molecules. Elves’ emission pulses typically last about 1 ms. In some events, two or more pulses occur in close temporal succession, separated by only tens to a few hundred microseconds. The origin of these multiple rings remains under investigation; they cannot be attributed to multiple return strokes, as the typical interstroke interval is about 60 ms. One hypothesis is that they may be generated by compact intracloud lightning (CID), with the time separation between rings (∆T) depending on the height of the discharge (hs).
Figure 1: a) All traces of a double elve event in Coihueco (CO), with the first peak aligned to the lightning source time (t = 0). b) Time difference of the event at CO and Los Leones (LL), with a mean value of (87 ± 1) µs. The bouncing mechanism model curves for 5, 10, and 20 km lightning heights are also displayed. c) Traces of a triple elve event in CO. d) Time difference of the event from LL and CO, with mean values of (19 ± 2) µs and (98 ± 8) µs. e) Traces from an event showing multiple pulses that are significantly wider (halo) compared to typical elve pulses, occurring between 20 and 150 km of Darc. Beyond 150 km, traces of a double elve are observed (see zoomed-in plot). f) Time difference for the double elve, with a mean value of (10 ± 1) µs. Other values represent the time difference between the double elve and halo traces (tH - tE).
To investigate this, we analyzed traces from 70 double and 24 triple elves across four storms and found that the ∆T remains nearly constant, independent of the arc distance to the lightning (Darc). Figure 1a shows an example of pulses from a typical double elve, with constant ∆T at about 87 µs in two independent FD sites (see Fig. 1b). The open curves in this graph indicate the expected ∆T for different lightning heights, which should decrease with Darc. This constant behavior is observed over a wide range of ∆T values, from 7 µs to 260 µs, in double and triple elves detected at Auger (see the triple-event example in Fig. 1c-d). This may indicate that the ground and atmospheric conditions within the FD field of view are not favorable for CIDs producing multi-elves, or that a different underlying process is involved. The study of multielves also revealed events with a constant ∆T but several outliers (see Fig. 1f). Inspection of the traces showed that these outliers resulted from calculating ∆T between elves and another TLE, likely halos, detected within the same time window (see Fig. 1e). After separating the multi-elves from the halos, the constant behavior of ∆T is recovered.
An alternative explanation for the origin of multi-elves is that ∆T is related to the rise and fall time of the lightning current density pulse. By analyzing Earth Networks lightning waveforms correlated with our events, we found that the skywave base time (tb2) for single elves is generally narrower than for double and triple-elves (Fig. 2b). Moreover, tb2 correlates with the separation of the rings as shown in Fig. 3. For ∆T ≥ 50 µs, it increases with tb2 in double and triple events, indicating that multiple rings are more likely to be related to the lightning waveform than to its height. Although no clear trend appears in ∆T < 50 µs, these results advance the understanding of multi-elves and provide a basis for further, more in-depth investigation into this phenomenon.
Figure 2: (a) Definition of the base time of the lightning waveform pulse given by Earth Networks. (b) Examples of the second component (skywave) of lightning waveforms correlated with single, double, and triple elves. Distribution of the base time for the (c) first component and (d) second component of the waveform pulses correlated with elves detected on April 27-28, 2020. The mean value t¯b2 for single elves is (74 ± 21) µs, which is significantly shorter than the mean values observed for double and triple elves, (109 ± 26) µs and (112 ± 19) µs, respectively.
Figure 3: Base time of the skywave of waveforms (tb2) correlated with double elves (left) and triple elves (right) detected on April 27-28, 2020. In both cases, it can be observed that, beyond 50 µs of peak separation time for multi-elves (∆T), there is a high correlation with tb2: r = 0.78 in the fitting of doubles and, r = 0.75 in the triple events. As the base time increases, the temporal gap between the peaks of multi-elves also widens.
Related Paper:
Quasi-constant time gap in multiple rings of elves
The Pierre Auger Collaboration, Earth and Space Science 12 (2025) e2025EA004321
[doi: 10.1029/2025EA004321]
