North Korean (DPRK) Nuclear Test of 25 May, 2009:

Summary of Ground Truth Information



Joe Bennett

Last Revised:


Joe Bennett

Summary of Event

The Korean Central News Agency (2009) issued an announcement on 25 May 2009 that another underground nuclear test had been successfully conducted by North Korea (Democratic Peoples Republic of Korea, DPRK). The explosion was located in the same area of northeastern North Korea where a previous nuclear test was conducted in 2006. Although rather small in magnitude, the associated seismic event was recorded by numerous global seismic stations, including 59 IMS stations. Seismic locations developed by the International Data Centre (IDC) (2009) and US Geological Survey (USGS) (2009) generally agree with the location for the test site established previously from satellite imagery analyses.


Geologic Setting: 

The 2009 North Korean nuclear test appears to have been conducted in the same tunnel complex (as the 2006 test) mined into Mount Mantap, a mountain located in the northeastern part of the DPRK. The area of the test is a relatively stable craton (the North China-Korean platform) with a basement of Archean (~2000 Ma BP) and Proterozoic (~1000 Ma BP) granite and metamorphic rocks (USGS, 1967). The basement in the vicinity of the test site is overlain by up to 1 km of Cenozoic (65 Ma BP – Present) volcanic basalts which are little deformed.     


Geographic Location: 

41.293° N  ± 1 km,  129.066° E  ± 1 km

Satellite imagery was used to establish the entrance to the tunnel (, 2006) where the North Korean nuclear explosions of 2006 and 2009 were conducted, and topography for the area has enabled estimation of setback from the tunnel entrance needed to provide explosion containment (assuming normal nuclear testing practice, see below). The preliminary analysis suggested that the location for the 2009 test could be taken as approximately the same as that for the 2006 North Korean nuclear test. These ground truth location estimates have subsequently been revised to produce the preferred location (shown above) based on more complete analyses of the relative seismic locations of the two events along with more careful assessment of the topography in the vicinity of the tunnel complex and conditions necessary to achieve event containment, as described below.





















Figure 1. Ground truth based on satellite imagery indicates the location of the underground tunnel complex in which the North Korean nuclear tests in 2006 and 2009 were conducted on the southern flank of Mount Mantap.





The seismic locations (Table 1, Figure 2) for the 2009 North Korean nuclear test tend to be better resolved (smaller location error ellipses) than for the 2006 event, but still have some uncertainty because of the small size of the event and the number and distribution of stations in the recording network. The USGS solution was based on a larger number of stations than that of the IDC REB (75 versus 59), including more within regional distances (10 versus 6). However, the maximum station azimuthal gap is somewhat smaller for the IDC REB solution (53° versus 72°). Although the USGS location error ellipse area is smaller (39 km2 versus 265 km2), it does not overlap the tunnel complex which is thought to be the likely ground truth location for the 2009 explosion. The ground truth location is just outside the error ellipse determined by the USGS (see Figure 3 below) and within the error ellipses for the IDC REB and SEL seismic solutions. We also developed an alternative location solution (SAIC CrossCor) based on relative timing of seismic signals at common stations for the 2009 and 2006 NK tests and using the 2006 event preferred location as a master event. The latter is both closer to what we believe to be ground truth (i.e. the Mount Mantap tunnel complex) and the small location error ellipse for the solution overlaps it.



Lat (°N)

Lon (°E)

Error Ellipse Area (km2)

No. Stations Used

No. Stations within 2000 km

Largest Azimuth Gap







~112° (50-162°)







~53° (238-291°)







~72° (49-121°)

SAIC CrossCor






~101° (194-295°)


Table 1. Seismic locations and their uncertainties from various sources.




Tunnel Entrance


1 km


Figure 2. Locations reported by various sources for the 2009 North Korean nuclear test.





Figure 3. Seismic locations and associated error ellipses from various sources for the North Korean test.





Murphy et al. (2010) applied a range of relative seismic location techniques (Joint Hypocenter Determination – JHD, Double Difference Location – DD, and Differential Waveform Interferometry – DWIF) to determine the location of the 2009 North Korean nuclear test relative to the 2006 test. The results of these analyses indicated that the 2009 test was located ~2.5 km west-northwest of the 2006 test. Then, based on analyses of nuclear test containment practice and the topography in the vicinity of the tunnel complex with constraints: (1) the 2006 event was placed at a depth with 100-300 m of overburden relative to the tunnel entrance elevation, (2) the 2009 event was placed at a depth with 350-750 m of overburden relative to the tunnel entrance, and (3) the 2009 event was located about 2.5 km to the west-northwest of the 2006 event, the preferred event locations were refined, as shown in Figure 4.

Figure 4. Best estimates for the preferred locations of the North Korean nuclear tests resulting from the topographic analysis.







0.6 ± 0.4 km

The source depth for the 2009 North Korean nuclear test has been estimated from analyses of the tunnel adit location with respect to topography in the vicinity of the tunnel entrance (Figure 4) and nuclear explosion containment practice (Murphy, 1977), following the same methodology as for the 2006 test. The latter indicates that containment (based on a nominal depth of burial, h ≈ 0.120 W1/3) would have been achieved for a 2.2-4.8 kT explosion (see below) at a depth of 0.16-0.21 km below the topographic surface elevation, at a location within about 0.6 km of the tunnel entrance. At even greater horizontal extent of the adit, a depth of 0.4 km is reached about 1 km from the tunnel entrance along the maximum topographic gradient (approximately due north) and a maximum depth of about 0.8 km about 2 km from the entrance. Bermudez (2006) in Jane’s Defence Weekly cited reports that North Korea had excavated a horizontal tunnel extending ~0.7 km under Mount Mantap in the build-up preceding the 2006 nuclear test. The refined location (Figure 4) for the 2006 test is consistent with the latter report and a depth of burial of 0.18 km, and the corresponding depth at the preferred location for the 2009 test is 0.6 km.  





1.400 km ± 0.4 km

The region surrounding the North Korean nuclear test location at Mount Mantap has high topographic relief. The elevation at the tunnel entrance, determined from the regional topography, is ~1.4 km. As noted above, the surface elevation reaches ~1.8 km about 1 km in from the tunnel entrance and ~2.2 km at the peak of the mountain. The surface elevation at the preferred location of the 2009 test is ~2.0 km (as shown in Figure 4, above).


Origin Time (UTC): 

2009-May-25  00:54:43.3 ± 0.5 sec

The USGS origin time (USGS, 2009) for the North Korean nuclear test, as shown, is based on the seismic location solution described above. The event origin time for this solution is probably accurate to within ± 0.5 sec based on comparisons with other seismic location solutions (e.g. IDC REB and IDC SEL).


Event Magnitude

4.5 – 4.7 mb

The magnitude measures for the 2009 North Korean nuclear test vary depending on differences in the seismic magnitude measurement scheme and stations used. Teleseismic network mb values are between 4.5 (IDC REB based on 45 stations) and 4.7 (USGS based on 54 stations). The surface-wave magnitude Ms for the explosion, determined from the IDC REB based on 15 stations, was 3.6 (significantly larger than expected for an explosion with that small mb).


Explosion Yield

2.2 – 4.8 kT

Yield estimates for the 2009 North Korean nuclear test cover a range of about a factor of 10 depending on different yield relationships used and the magnitude measures applied. Using the mb/Yield relationship (Murphy, 1996) for the Soviet Semipalatinsk test site (i.e. relatively low attenuation in the upper mantle), the observed USGS teleseismic mb value (4.7 mb) suggests an explosion yield of about 2.2 kT for the North Korean nuclear test. A much larger yield estimate (~30 kT) comes from applying the 3.6 Ms using the nominal Ms/Yield relationship (Stevens and Murphy, 2001). A model-based yield-estimation method based on teleseismic P spectral measurements (Murphy and Barker, 2001) indicates a yield near 2.7 kT, if the explosion was detonated at a depth of ~200 m, and a yield near 4.8 kT, if the explosion was at a depth of ~800 m, as shown in Figure 5. That is, the short-period data in the frequency band from 0.5 to 2.5 Hz are consistent with a yield range extending from about 2.2 to 4.8 kT.
















Figure 5. Teleseismic P spectral observations from the 2009 North Korean nuclear test indicate a match with the model (based on Semipalatinsk stable craton) for an explosion yield of approximately 2.7 kiloton for a depth of burial of 200 m or 4.8 kiloton for a depth of 800 m..








USGS/NEIC (2009), Online Bulletin of the USGS/NEIC Earthquake Hazards Program (


USGS (1966-1969), Military Geology Branch Atlases of Asia and Eastern Europe to Support Detection of Underground Nuclear Testing. (2006), Weapons of Mass Destruction (WMD): Possible North Korean Nuclear Test Site near Punggye-yok (


Bermudez, J. (2006), North Korea Claims Nuclear Test, Jane’s Defence Weekly (


Korean Central News Agency-KCNA (2009), KCNA Report on One More Successful Underground Nuclear Test (


Murphy, J. (1977), Seismic Source Functions and Magnitude Determinations for Underground Nuclear Detonations, Bull. Seism. Soc. Am., 67, 135-158.


Murphy, J. (1996), Types of Seismic Events and Their Source Descriptions, in Monitoring Comprehensive Test Ban Treaty, edited by E. Husebye and A. Dainty, 225-256.


Murphy, J.,  and B. Barker (2001), Application of Network-Averaged Teleseismic P-Wave Spectra to Seismic Yield Estimation of Underground Nuclear Explosions, Pure and Applied Geophysics (PAGEOPH), 158, 2123-2171.

Murphy, J., B. Kohl, T. Bennett, and H. Israelsson (2010), An Analysis of Seismic Characteristics of the 25 May 2009 North Korean Underground Nuclear Test, Seism. Res. Letters., 81, 296.


Stevens, J., and J. Murphy (2001), Yield Estimation from Surface Wave Amplitudes, Pure and Applied Geophysics (PAGEOPH), 158, 2227-2251.