Quality codes A, B, C and insufficient
Every cell gets a quality code, A, B, C or insufficient, from the number of events and the width of its 95 % interval, and the b-value is shown only for A and B.
Interactive diagram · Real example from the atlas →Simple explanation
The atlas has one hard rule: no b-value without its interval and quality code.
If too few earthquakes are above the completeness threshold, the cell is insufficient. If there are enough but the interval is still very wide, the cell is C and b is hidden: a number that could be anything in a wide range would mislead. Narrow intervals give A, moderately narrow ones B.
So the grey and hatched areas on the map are not missing work. They are the honest answer for a global catalogue of moderate and large earthquakes.
An everyday picture
A weather station that reported only two days last year cannot give you a reliable yearly average. The atlas would rather leave the box empty and say why than print a guess.
Technical details
Rule (manifest.json, method.quality): at least {nMin} events above the per-event thresholds, otherwise insufficient; then A if the bootstrap 95 % interval is at most {wA} wide, B if at most {wB}, else C (b masked, only the width is published).
Why width and not only n: the same n can give different widths when Mc is poorly defined or the sample mixes periods; the bootstrap width includes the Mc uncertainty. A cutoff-stability flag (unstable) is added when b at threshold + 0.2 and + 0.4 both fall outside the interval on the same side.
Map conventions: A full colour, B the same colour at reduced opacity, C hatched, insufficient a light grey wash. The legend counts how many cells pass the gate.
Limitation: the gate controls precision, not accuracy. A cell can be A and still biased (for example by a wrong threshold), and neighbouring A cells share events (Wiemer & Wyss 2002; Marzocchi et al. 2020).
Further reading
- Woessner, J. & Wiemer, S. (2005). Assessing the quality of earthquake catalogues: estimating the magnitude of completeness and its uncertainty. BSSA, 95(2), 684–698. doi.org/10.1785/0120040007
- Marzocchi, W., Spassiani, I., Stallone, A. & Taroni, M. (2020). How to be fooled searching for significant variations of the b-value. Geophysical Journal International, 220, 1845–1856. doi.org/10.1093/gji/ggz541
- Wiemer, S. & Wyss, M. (2002). Mapping spatial variability of the frequency-magnitude distribution of earthquakes. Advances in Geophysics, 45, 259–302. doi.org/10.1016/S0065-2687(02)80007-3