Explosion of a bomb in the river Danube

S&T 2013, Vienna
Explosion of a bomb in the river Danube
Helmut Hausmann, Wolfgang Lenhardt, Ulrike Mitterbauer
Austrian National Data Centre (NDC-AT) at the Central Institute for Meteorology and Geodynamics, Vienna
Josef Havir, Jan Svancara
Institute of Physics of the Earth, Brno
Event
Fig.1: Explosion - Bomb in River Danube (Wien Heute)
On the afternoon of Saturday August 25, 2012, a
ground motion was recorded by several
monitoring stations at the Austrian Seismological
Service that was not characterized as an
earthquake due to its wave signature. It was
found that this event was caused by the intense
detonation of an old aircraft bomb from World
War II in the riverbed of the Danube in Vienna
Donaustadt / 22nd District (Fig.1, 2). This event
was well felt in the 1st, 2nd, 3rd, 4th, 10th, 11th
and 22nd Districts in Vienna. Small building
tremors were reported and some people were
awakened from sleep. A bang was also
observed in some places close to the detonation.
Fig.2: Location of the event
Bombing of Vienna in World War II
In the afternoon of the 25th of August 2012 five strong-motion stations located in Vienna
and four broadband stations of the Austrian Seismic Network registered a shock, which
occurred due to the detonation of an aircraft bomb from the Second World War. Stations
of the IPE Seismic Network, including the auxiliary seismic IMS station VRAC and
stations of the Temelin Network, also recorded the event. Independent calculated
locations were conducted by the Zentralanstalt für Meteorologie und Geophysik (ZAMG)
and by the Institute of Physics of the Earth (IPE) and the hypocenter of the event was
determined close to the Danube in Vienna by both institues. The event could be localized
with an accuracy of approximately one kilometer. Due to the registrations the exact time
of detonation could be calculated.
Vienna Strong Motion Network
In the years 1992/93 five strong motion station were installed in the framework of the project
“Seismisches Strong-Motion Meßnetz in Wien”. Fig.9 shows the location of the stations, the
groundtruth position and as well the result of the IPE location. In the year 2012 the stations were
upgraded and the old data acquisition systems were replaced by new BASALT-systems with integrated
Episensors / Kinemetrics (Fig. 10).
Fig.10: BASALT-system
Bombing of Vienna started in 1944. The city of Vienna was bombed fifty-three times during World
War II and 87 000 houses of the city were damaged (Ulrich, 1994). The estimated number of
aircraft bombs dropped over Vienna is 100 000. In the period between 1945 and 2011 around 26
000 t war relicts and 21 000 unexploded bombs were recovered and destroyed
(Entminungsdienst, 2011). Fig.3 shows the entrance of the allied airforces into Austria by the end
of the war. Via personal communication we were informed that most of the bombs thrown on
Vienna were 250 kg bombs (Fig.4).
Fig.11: Unfiltered data
Fig.4: 250 kg aircraft bomb
Analysis of Velocities
Using the travel times (Fig.14) we estimated the apparent
velocity for the P-wave at 5,5 km/s, for Onset1 1,04 km/s
and for Onset2 0,56 km/s. Both velocities do not
correspond with the speed of sound in the air (0,3 km/s)
nor with the speed in water (1,5 km/s). Even if the
common velocities of surface-waves (Love wave: 3,5 to 4
km/s, Rayleigh wave: close to 3,0 km/s) are different to the
observed velocities, the shape of the signals (Fig.13a to
13c) look like surface waves. Having no experience with
signals generated by an underwater explosion in the close
vicinity of seismic stations, we do not know the behavior of
surface waves in these conditions. Observing the signal it
is obvious that the difference in the Onsets can be found in
Fig.14: Plot of Travel times for the Strong the particle motion. For station WIWA, which is lying
Motion Stations
westwards of the epicenter the polarization of the Onsets
could indicate a Rayleigh- and a Love wave-type of
waves. Fig.15 shows a similar plot as Fig.14, but some
further stations of the ZAMG network are included. From
the plot one can see that Onset1 and 2 can be found as
well in signals from stations at a distance of more than 150
km. All analysed traces can be traced back to the
explosion itself and not to refracted waves.
Another explanation of the velocity of Onset1 could be an
upheaval of the river bed due to the explosion of the
buried aircraft bomb, whereas Onset2 could possibly
indicate a “ringing” of the water body in the river bed. With
increasing distance of the stations a slight dispersion of
Fig.15: Plot of Travel times including the waveform can be observed.
seismic stations of the ZAMG Network
The effect of the water outburst is considered negligible
when compared with the actual release of energy by the
bomb itself.
Fig.9: Strong motion network
Fig.3: Entry of allied airforces
Waveform Analysis at the IPE
Seismic traces of 22 seismic stations were analysed and the event was picked on 18 stations.
Most of the stations belong to the IPE Seismic Network, but also five stations of the Temelin
Network, four stations of the Austrian network and one station in Hungary were processed. Fig.5
shows the distribution of the stations and the two independent locations of the IPE and ZAMG.
No signal could be detected for stations marked with red triangles. The vertical traces of the
stations are displayed in Fig.6, whereas Fig.7 shows data of all three channels of the auxiliary seismic
IMS station VRAC. The result of the calculated location including data of the error ellipse is shown in Fig.8.
All stations registered the event. The data was processed and
analysed with Antelope Software. Unfiltered acceleration data from five
stations of the Vienna strong motion network are shown in Fig.11. One
can see the high frequency signal with high amplitudes of the body
waves between 12:47:10 UTC and 12:47:15 UTC followed by low
frequency signals around 12:47:18 UTC. To enhance the low
frequency part of the signal a 1.0 LP-Filter was applied (Fig.12).
Characteristic phases, which were found on all traces, are indicated
(Onset1 in red, Onset2 in blue).
Fig.12: Filtered data (1.0 LP)
Analysis of Spectrograms
Figures 13a to 13c show the filtered seismograms (1.0 LP) of all three components of stations KMWA,
WIWA and UMWA which are positioned 5 km NW, 7 km W and 7 km WNW of the explosion,
respectively. We observe that the first part of the signal from the vertical component of station KMWA
is reflected in the NS-component whereas the second part can be found in the EW-component. For
station WIWA we observe the reverse effect. For station UMWA Onset1 and Onset2 can be found in
both components. Station KMWA is situated close to the Danube river, whereas station WIWA was
installed a few kilometers from the river side. This could be one of the reasons why the observed
signals differ by so much.
Fig.13a: Traces of station KMWA
Fig.13b: Traces of station WIWA
Spectrograms can provide us an overview for wave energy distribution in frequency range for a
time series. In addition, difference of dependence of seismic phases on frequency can be
observed in spectrograms as well. To better understand signature of seismic phase energy in
addition to the seismograms spectrograms of the vertical channels of all strong motion stations
were calculated (Fig.16a-16c). Spectrograms for the stations which are lying within a radius of 10
km show that most of the body wave energy concentrates in a frequency range between 5 and
40 Hz and within a time interval of ~ 2 s. For station SNWA (13km) the energy concentrates in a
range up to 10 Hz. Low frequency signals follow with energy concentrations up to 5 Hz and a
duration of ~ 10 s.
Fig.16a: KMWA (distance = 4,6km)
Fig.16b: WIWA (distance = 6,42km)
Fig.16c: SNWA (distance = 13,1km)
Fig.13c: Traces of station UMWA
Fig. 7: Seismic traces - VRAC
Conclusions
- On August 25, 2012, an old aircraft bomb from World War II detonated in the riverbed of
Danube in Vienna
Fig.5: Distribution of stations
- The event was located independently by IPE and ZAMG using data of national networks.
Fig.6: Signals of the explosion
Fig.8: Location Details
References
Bundesministerium für Inneres (2011), Jahresbericht Entminungsdienst 2011
(see http://www.bmlv.gv.at/organisation/gattung/pdf/emd_bericht2011.pdf)
Johann Ulrich (1994), Der Luftkrieg über Österreich 1939-1945, Militärhistorische Schriftenreihe, Heft 5/6
- Stations of the Vienna Strong Motion Network registered the event as well.
- On all signals of the strong motion stations two characteristic phases were found in the low
frequency part.
- Velocities for the P-wave and for both Onsets were calculated.
- Descriptions and possible explanations of the Onsets were presented.