The Analysis of DPRK Nuclear Test of February 12, 2013 by Belbasi Nuclear Tests Monitoring Center-KOERIThe Analysis of DPRK Nuclear Test of February 12, 2013 by Belbasi Nuclear Tests Monitoring Center-KOERIK. Semin, O. Necmioglu, C. Destici, N. Ozel, S. Kocak, U. TeomanK. Semin, O. Necmioglu, C. Destici, N. Ozel, S. Kocak, U. Teoman

Kandilli Observatory and Earthquake Research Institute, Boğaziçi University, Istanbul, TurkeyKandilli Observatory and Earthquake Research Institute, Boğaziçi University, Istanbul, Turkey

T2-P71

IntroductionIntroduction Seismic AnalysisSeismic Analysis

On 12 February 2013, The Democratic People’s Republic of Korea (DPRK) announced the conduct of a nuclear test. Corresponding seismic event was recorded by IMS, upon which IDC released first automatic estimation (SEL1) of time (02:57:51 GMT), location (41.3386°N  and 129.0711°E ) and the magnitude (4.9 mb) of the event in less than two hours time. During the preliminary analysis of the 2013 DPRK event by the Turkish NDC, a very clear P arrival at 03:08:55 (GMT) at BRTR (Keskin SP array) has been observed, which was not associated to SEL3. The result of our analysis confirmed that the arrival belongs to the DPRK event. In this study, we present some of the analysis performed at the NDC.

Infrasound AnalysisInfrasound AnalysisAlthough Turkey doesn't have an infrasound station we have performed an infrasound analysis as an exercise. Analysis was done for the event of 12/02/2013 at 02:57:51 UTC using stations that recorded the event, I45RU, Russian and I30JP, Japan infrasound arrays. This analysis was performed by the PMCC software package provided by CEA/DASE, France. PMCC calculation value based on the infrasonic data for infrasound array, including the azimuth, apparent velocity, and waveform. The results here do not include ray tracing for infrasonic phases and Effective winds in Stratosphere (ECMWF data) as we don't have the capacity to perfom these steps.

Figure 9: Calculation process was set for the range frequency 0.1 – 0.8 Hz. for I30JP infrasound station. Wlenght 10-20 s and Nb of bands 20. The results of PMCC on the infrasonic data found as Velocity 0.267 km/s, Azimuth 306 deg, Amplitude 0.016 Pa RMS.

Figure 8: Calculation process was set for the range frequency 1 – 4 Hz for I45RU infrasound station. Wlenght 30 s and Nb of bands 20. The results of PMCC on the infrasonic data found as Velocity 0.329 km/s, Azimuth 214 deg, Amplitude 0.003 Pa RMS.

Figure 1: The great circle distance of the February 2013 event from BRTR IMS station.

BRTR has detected all of the previous nuclear tests by North Korea, and the latest nuclear test. We begin the analysis by obtaining the other IMS array stations' waveform data using the new IDC waveform retrieval script. Most of the stations had really good signal/Noise ratio for this event which indicated a big magnitude event. Automatic SEL3 bulletin did not included the BRTR array despite a very good signal. Later on analyst reviewed event bulletin (REB) add the BRTR signal to the final location. Unfiltered signal recorded at BRTR array can be seen in Figure 2.

The data from selected IMS arrays were used in Geotool software for location and body wave magnitude estimations. Geotool software only calculates the mb magnitude at the moment, but it will include the Ms magnitude calculations in the future releases.

Figure2: Unfiltered seismic recording of the event at BRTR array (left) and F-K analysis results (right).

The F-K analysis resulted in 53 degree back azimuth and 4.8 slowness values for BRTR array which is consistent with North Korea test. The final location results is given below in figure 3. The event has been located at 41.29 N, 129.07 E and a distance of 7608 km from BRTR array. The contributed IMS array stations and their paths are shown on the map in figure 4 and 5.

Figure 3: The final location results using only IMS array data.

O. Time Lat Lon Depth Mb

IDC 02:57:50.80 41.30 129.06 0f 5.0

USGS 02:57:51 41.30 129.07 0f 5.1

TR NDC 02:57:50.6 41.29 129.07 0f 5.0

Table 1: The comparison of final location parameters of different institutes

Figure 4: Selected IMS array stations in the location of TR NDC and te relative location results

Mb:Ms discriminationMb:Ms discriminationWe have calculated the body wave (m

b) and surface wave

magnitudes (MS) for 8 IMS array stations. Body wave

magnitude estimations were done in Geotool software using the IDC m

b magnitude procedures while surface

wave magnitude values calculated manually using the following formula;

Ms = log (A/T) + 1.66 log D + 3.3

where A is the maximum ground amplitude in micrometers (microns) of the surface wave within the period range 18 to 22 s. T is the period in seconds. D is the epicentral distance in degrees.

Calculated magnitude values were drawn on a graph in order to make a M

s:m

b discrimination of the event. A

provisional screening line Ms = m

b – 0.64 suggested by

Selby et al. (2012) were used to make the discrimination. As can be seen from the figure, all of the magnitudes stay in bottom part of the screening line which indicate an explosion source for this event.Figure 5:mb:Ms discrimination graph, the dashed line represents

provisional screening line by Selby et al. (2012). The red dots indicates individual magnitude results at each array site.

Yield EstimationYield Estimation

The yields of underground nuclear explosions are typically estimated from the related measure of seismic magnitude or similar measures of seismic signal strength.These estimations usually employ emprical relationships between the known yields of nuclear explosions conducted at different test sites with the relations given by Murphy (1981, 1996)

mb = 4.45 + 0.75 log Y ; where Y is the yield in kt.

Using this formula, 2006, 2009 and 2013 nuclear explosions give yield estimations as;

2006 (mb 4.3) = 0.65 kT2009 (mb 4.7) = 2.2 kT2013 (mb 5.1) = 7.4 kT

As a further analysis we can look for teleseismicP wave spectra.

Figure 6: We have shown the seismic recordings of 3 nuclear test by N.Korea on a common amplitude scale recorded at BRTR array. The gradual increase in the energy and amplitude of these events can be easily seen from the figure.

Figure 10: The final location of the infrasound analysis using only 2 stations and the REB location. Without the ECMWF corrections the amount of difference in location is normal.

ReferencesMurphy, J. R. (1981). P-wave coupling of underground explosions in various geologic media, in Identification of Seismic Sources—Earthquake or Underground Explosion, Proc. of the NATO Advanced Study Institute, D. Reidel Publishing Company, Dordrecht, The Netherlands,201–205.Murphy, J. R. (1996). Types of seismic events and their source descriptions, in Monitoring a Comprehensive Test Ban Treaty, Proc. of the NATO Advanced Study Institute, Kluwer Academic Publishers, Dordrecht, The Netherlands, 225–256.Selby N. D., Marshall P.D., Bowers D. mb:Ms Event Screening Revisited, Bulletin of the Seismological Society of America, Vol 102, No. 1, pp. 88-97, February 2012.