Observatories and Research Facilities for EUropean Seismology
Volume 4, no 2 September 2002 Orfeus Newsletter

Destructive earthquake in Afghanistan (Hindu Kush) on March 25, 2002

O. Starovoit, S. Yunga, I. Gabsatatova and L. Chepkunas

Geophysical Survey of Russian Academy of Science, 249020, Kaluga region, Obninsk, Lenin str. 189

Introduction - Epicenter Location - Macroseismic observations
Discussion and Conclusions - Acknowledgements - References


On March 25, 2002 at 14:56 GMT a destructive earthquake occured in the Northern part of Afghanistan (35.97 N 69,18 E Ms = 6.0 NEIS). This shallow earthquake occurred in the Pamir-Hindu Kush region, located at the western end of the Minalayan Arch (Figure 1). Seismicity in this region is associated with the deformation belt created by northward collision of the Indian plate into the Eurasian plate. This collision occurs at a rate of about 4.5 centimeters per year (Windley, 1988; DeMets et al., 1990). Many different models have been presented to interprete the intermediate-depth seismicity (e.g. Dewey et al., 1989; Molnar and Tapponier, 1975; Fan et al., 1994; Hamburger et al., 1992; Lukk et al., 1995). A summary of the features displayed in the seismicity has been presented by Pegler and Das (1998), who relocated about 6000 earthquakes for the period 1964-1992. The seismicity is dominated by earthquakes extending in depth from the surface to approximately 300 km deep. On average, there are about five earthquakes per year with a magnitude 5 and larger whose epicenters are within 150 kilometers of the event of March 25. Major tectonic features in the region include the Darvaz-Karakul Fault and the Main Karakoram Thrust (Figure 1).

The epicenter of the March 25 earthquake can be associated with the southern branch of the Darvaz-Karakul fault. The seismic history shows an earlier large shallow earthquakes that occured at approximately the same location, i.e. December 16, 1982 at 00:40:48.72 UT (36.148 N, 69.011 E, Mb = 6.2). It is interesting to note that this earthquake was preceded by a deep large earthquake on May 02, 1981, 16:04:52.8 UT (36.30 N, 71.20 E, Mb = 6.3).

The large shallow earthquake of March 25, 2002 occured three weeks after two large deep earthquakes on March 3, with magnitudes Mb of 6.2 and 6.6 respectively (locations 35.98 N, 69.17 E and 36.65 N, 70.45 E respectively).

Figure 1. Map of the earthquakes in Afghanistan. The location of the epicenters is determined by and published in the Obninsk seismological bulletin for the period 1955-2001. Blue is depth more 70 km, purple is depth in crust. The focal mechanisms of the earthquakes on March 3 and 25, 2002 come from the database of Harvard University Centroid Moment Tensor. Faults are marked yellow: DKF is the Darvaz-Karakul Fault and MKT is the Main Karakoram Thrust. The black arrow shows the direction of the plate motion of India relative to Eurasia (DeMets et al., 1990).

Epicenter Location

The shallow earthquake of March 25, 2002

Preliminary earthquake parameters were determined by the Alert Survey (AS) of the Data Processing Center (DPC) of the Geophysical Survey of the Russian Academy of Sciences in Obninsk (Starovoit and Mishatkin, 2002). Digital and analog data from seismic stations of Russia, Commonwealth of Independent States (CIS) countries (Kirghizia, Kazakhstan, Turkmenistan, and Ukraine) and abroad(Norway and Alaska, USA) have been used. Subsequently, an alert message of this earthquake was sent to the Ministry of Emergency of the Russian Federation (EMERCOM) within 50 minutes after its occurence.
Preliminary analysis of this event was done with data from 15 stations and a later improvement included data from another 10 stations. The azimuthal coverage of these stations were limited to the first (Az = 2°-54°) and fourth (Az = 286°-344°) quadrant with respect to the epicenter and epicentral distances in the range of 8°-75°. Out of the 25 stations, 15 were Russian, 8 were CIS stations (Kirghizia 1, Kazakhstan 2, Turkmenistan 2 and Ukraine 3), 1 was in Norway (KONO) and 1 in Alaska, USA (COLA). A summary of the obtained parameters compared with those provided by the NEIC and CSEM is provided in Table 1.
Figure 2 shows the vertical broadband registrations of 11 of the above stations (Figure 3) which were obtained by the DPC in near real-time on March 25, 2002. The first motion polarities corroborate the Harvard centroid-moment tensor (CMT) solution. The north-north-west oriented nodal plane coincides roughly with the orientation of the aftershock sequence.

Figure 2. Registrations of the earthquake on March 25. Shown are the vertical components (BHZ) of 11 digital stations: BRVK is Borovoe, Kazakhstan (delta is 17°), ARU is Arti, Ural (21°), KIV is Kislovodsk, Caucasus (23°), TLY is Talaya, Baikal (29°), OBN is Obninsk, central European part of Russia (29°), PUL is Pulkovo, Northwestern Russia (35°), LVZ is Lovozero, Kola Peninsula Northwestern Russia (37°), TIXI is Tiksi, Northern Russia (47°), YSS is Yuzno-Sakhalinsk (54°), MA2 is Magadan (56°), BILL is Bilibino, Northeastern Russia (60°).

Figure 3. 12 Russian broadband stations that send the digital records in near real-time to Data Processing Center GS RAS in Obninsk.

Table 1.
Preliminary source parameters as determined by the GS RAS, NEIC and CSEM.
Center Origin time (GMT), Latitude Longitude Depth (km) Ns Mb / N MS / N I0








NEIC 14:56:33.3 35.93 69.19 8 72 6.0/30 6.3/55 .
EMSC 14:56:38.1 36.04 69.20 33 150 5.9-6.2 . .
N - number of stations used in the magnitude determination.
Ns - Number of stations used in the location
I0 - Intensity

The March 25 earthquake was followed by more than 20 aftershocks with Mb > 4. The strongest aftershock was on April 12, 2002, with magnitude Mb = 5.6 (GS-RAS) and M0 = 1.5·1025 dyne-cm (from P-wave spectra of OBN) corresponding to MW = 6.1. This values are close to the values obtained by the NEIC, i.e. M0 = 1.2·1025 dyne-cm and MW = 6.0, and those by Harvard, M0 = 1.5·1025 dyne-cm and MW = 6.1.

The deep earthquakes of March 3, 2002

The epicenters of the two earthquakes on March 3 were located 135 km northeast of the earthquake of March 25. Figure 4 shows the vertical broadband registrations of the same 11 digital station as shown in Figure 2.

Figure 4. Recordings of the earthquake of March 3, 2002. Shown are the vertical broadband components (BHZ) of the same stations as in Figure 2. See this figure for the station names and azimuthal distances

Analysis of the waveform data revealed phases (10-12 sec after the first arrival) that cannot be identified. It is therefore assumed that we recorded two deep events with a 10-12 sec difference in origin time and approximately the same location. This was confirmed in a more detailed analysis. The foreshock was slightly more shallow, i.e. 200 km depth, than the main shock, i.e. about 250 km. The parameters of both earthquakes are listed in Table 2. The depth estimation was made using the sP phases of the following stations: for the foreshock ARU, TLY, OBN, LVZ, KONO, MA2, BILL and PET and for the main shock TLY, MA2 and BILL. At the other stations the sP-phase interfered with the PP or PcP phase of the first event.
The values of M0 according to the NEIC and Harvard are M0 = 1.1·1027 dyne-cm and M0 = 1.2·1027 dyne-cm, respectively. Both centers had the same moment magnitude of Mw = 7.3. This magnitude estimation is done using both events. At present we do not have evaluated the two earthquakes separately.

Table 2.
Preliminary event parameters for the March 25 double events as determined by the GS RAS and the NEIC
Center Origin time (GMT), Latitude Longitude Depth (km) Ns Mb / N MS / N I0
12:08:07.1 36.51 70.53 200 30 6.4 / 11 6.6 / 10 6.5-7.0
12:08:06.16 36.44 70.45 195 52 6.2 / 33 . .
Main shock
12:08:22.9 36.65 70.45 250 18 6.9 / 10 . 6.5-7.0
Main shock
12:08:22.9 36.53 70.44 256 66 6.6 / 49 MW=7.3 .
N - number of stations used in the magnitude determination.
Ns - Number of stations used in the location
I0 - Intensity

Macroseismic observations

The shallow earthquake of March 25

The earthquake occured in the night between March 25 and 26 in the mountains of the highly seismic active region of Hindu Kush, 170 km North of the Afghanistan capital Kabul. Strong shaking was felt in Northern Afghanistan in the cities Kabul and Mazar-I-sharif and in Pakistan in Islamabad and Peshevar. In Dushanbe, Tajikistan, it was felt with intensity 3-4. Preliminary reports mention that about 2000 people were killed, nearly 4000 injured and more than 10,000 people became homeless due to this earthquake. Many people live in the streets, afraid to enter their house. Others live in or close to the desert without meals and water. The city Nahrin in the Baglan province was totally desolated and severe destructions has been observed in neighbouring Burka. The earthquake was followed by numerous aftershocks during the whole night. In the period between March 25 and 27 nine aftershocks with magnitudes 4 and higher are registered. These aftershocks raised the number of casualties.

The Afghan Government asked the World community for help. The United Nations sent humanitarian aid to the disastrous regions, including tents, blankets, food and medicine. Russia sent two airplanes with humanitarian cargo to Kabul.

The deep earthquakes of March 3

This double earthquake occured in the mountains 240 km north-northeast of Kabul and was felt in many countries of Central Asia: Afghanistan, Pakistan, India, Tadzhikistan, Uzbekistan, Kyrgyzstan , Kazakhstan and Xinjiang, China. However, relatively little damage was reported as compared to the March 25 earthquake. Preliminary reports mention about 100 people killed in the Afghan province Samangan.

Discussion and Conclusions

All the events occured in the highly seismic active region on the border of the Eurasian and Indian plate. Numerous earthquakes occur here both shallow, in the crust, and at medium depths up to 330 km. The destructions from the shallow earthquake on March 25 were much more severe than that from the deep earthquakes om March 3, although the March 25 earthquake had a smaller magnitude and generated 100 times less energy than the deep earthquakes of March 3.

The data used in this paper are the data used for the Alert messages. This data set will be updated with time using data from other Russian and worldwide stations.


This work was partly supported by grants of the RFBR 02-05-64333, 01-05-65340, and INTAS N 456.


  • DeMets, C., R.G. Gordon, D.F. Argus and S. Stein. 1990. Current plate motions. Geophysical Journal International, 101, 425-478.
  • Dewey, J.F., S. Cande and W.C. Pitman. 1989. III, Tectonic evolution of the India/Eurasian collision zone. Eclogae Geol. Helv., 82-3, 717-734.
  • Fan, G., J.F. Ni and T.C. Wallace. 1994. Active tectonics of the Pamirs and Karakoram. Journal of Geophysical Research, 99, 7131-7160.
  • Hamburger, M.W., D.E. Sarewitz, T.L. Pavlis and G.A. Popandopulo. ???. Structural and seismic evidence for intracontinental subduction in the Peter the First Range, Soviet Central Asia. Geol. Soc. Am. Bull., 104, 397-408.
  • Lukk, A.A., S.L. Yunga, V.I. Shevchenko and M.W. Hamburger. 1995. Earthquake focal mechanisms, deformation state and seismotectonics of the Pamir-Tien Shan region, Central Asia. Journal of Geophysical Research, 100, 20321-20343.
  • Molnar, P. and P. Tapponier. 1975. Cenozoic tectonics of Asia, effects of a continental collision. Science, 189, 419-426.
  • Pegler, G. and S. Das. 1998. An enhances image of the Pamir-Hindu Kush seismic zone from relocated earthquake hypocenter sources. Geophysical Journal International, 134, 573-595.
  • Starovoit, O. and V. Mishatkin, 2002. The seismic network of the Geophysical Survey of the Russian Academy of Sciences. Orfeus Electronic Newsletter, 4, 4.
  • Windley, B.F. 1988. Tectonic framework of the Himalaya, Karakoram and Tibet, and the problem of their evolution. Phil. Trans. R. Soc. Lond., A326, 3-16.

page 11
Copyright © 2002. Orfeus. All rights reserved.