What is the Lowest Magnitude Threshold at Which an Earthquake can be Felt or Heard, or Objects Thrown into the Air? - Historical Seismology: Interdisciplinary Studies of Past and Recent Earthquakes - page 305

in operation at the epicentre. Since our Gutenberg-Richter analysis shows that all
M L 1.3 events can be located, we conclude that those aftershocks recorded at a

single station probably have a magnitude smaller than 1.3.
A second example is the Laffrey earthquake (maximum EMS intensity V-VI),
15 km south of Grenoble (Isere). Besides the fact that its focus was similarly shallow
(

>

3 km) although here located in the pre-Triassic micaschist basement, it should
be also pointed out that: (i) it also involved strike slip; (ii) glacial deposits along
the Drac river also produced site effects; (iii) it also generated a long series of
aftershocks over more than 15 months (Thouvenot et al. 2003), again an unusual
span for a 3.5 magnitude. Many of these aftershocks were locally felt, although the
information that reached us by e-mail (no on-line questionnaire was then available)
is necessarily biased. The smallest aftershock that could be located and was also
reported felt occurred 3 days after the mainshock. For this event, we compute a
magnitude of 1.1 only, whereas we estimate a maximum intensity of IV from the
fragmented received testimonies.
At short epicentral distance, the routine computation of the M L magnitude can
be questioned: Richter (1935) dealt with earthquakes assumed to be sited at a depth
of 15 km, and his flat attenuation curve for the first 5 km of epicentral distance ex-
presses this assumption. In the case of the aforementioned event, 4 Sismalp stations
at distances of 10, 35, 58, and 100 km were available for M L computation, which
yielded the respective values of 0.86, 1.29, 1.00, and 1.11 (mean value: 1.07

∼

0.18).
Although the 0.86 value obtained at a distance of 10 km is the lowest of the series, it
does not deviate significantly from the mean value if we take the standard deviation
into account. However at still shorter epicentral distance we can expect problems:
what would be the meaning of an M L -magnitude computation for a station sited just
above a 300-m-deep focus? The question seems academic, but such instances are
encountered when small, ultra-shallow earthquakes are felt or heard.

±

3 The 2002-3 Tricastin Earthquake Swarm and the 2006
Conand Aftershocks

3.1 The 2002-3 Tricastin Earthquake Swarm

The first instance of such small, ultra-shallow earthquakes is provided by the earth-
quake swarm that occurred in 2002-3 in Tricastin (France) close to Saint-Paul-
Trois-Ch ateaux (Dr ome). This area of the middle 'Sillon Rhodanien' (Fig. 1),
between the French Massif Central to the west and the Alps to the east, has
been known for centuries as the seat of long-lasting earthquake swarms. In
1772-3 such a swarm visited the village of Clansayes where the church tower
was knocked down by the strongest event of the sequence (maximum intensity:
VII-VIII); in 1933-6 another swarm visited several villages close to La Garde-
Adhemar, which suffered slight damage (maximum intensity: VII) during the 1934
climax (Rothe 1936).