RESULTS OF SEISMIC STUDY ON THE UPPER PART OF THE SECTION AT SEMIPALATINSK TEST SITE AREA (FOR OSI PURPOSES)

A. Belyashov1, V. Suvorov

2, E. Melnik

2

1Institute of Geophysical Research, Ministry of Energy, Republic of Kazakhstan

2Institute of Petroleum-and-Gas Geology and Geophysics, Siberian Branch of the Russian Academy of Science, Russian Federation

(T3.4-P3)

Underground nuclear explosions (UNEs) forming the areas of enclosing geological media disintegration with different size and

configuration (Figure 1): camouflet cavity (radius 10÷13.6 m/kt1/3

), rock contortion zone (to 14÷17 m/kt1/3

), zone of rock fragmentation (to

24÷34 m/kt1/3

), zone of intense fracturing (to 50÷55 m/kt1/3

), zone of block fracturing (to 65÷70 m/kt1/3

), and spall zone which is observed

at epicentral distances to 100÷200 m/kt1/3

and on depth to 10÷32 m/kt1/3

(Adushkin and Spivak, 2004). This paper describes the results of

seismic study of one of above mentioned zones—spall zone, marked by red color on Figure 1. The spall zones, as usual, forming for all

the UNEs, conducted in standard technical-geological conditions. And these zones are simple to reveal because of their shallow depth

and notable by long life period, so, they could be one of the main target objects during OSI fulfillment.

In 1997 a seismic first-arrival method of observation was applied at a UNE area in Semipalatinsk Test Site (STS) in order to study

velocity properties of the geological environment affected by multiple underground tests and discover regional post-explosion permeable

structures. Seismic observations were implemented on the site 6000 × 3500 m of size, consisting of eight parallel profiles (Fig. 2). The

distance between the profiles was 500 m, shot-point interval (SP) was 500 m, and observation interval was 125 m. For the purposes of

seismic energy injection, chemical explosions with the charge of up to 40 kg were used.

2D velocity sections on P-waves

Based on SeisWide Program [Zelt and Smith, 1992], adjusted into the Windows OS (Dr. Deping Chian), by forward ray tracing method,

a retrieval of velocity models along seismic profiles has been conducted (Fig. 3). The depth of the obtained sections is limited by 300 m,

which is determined by the greatest depth of P-waves’ rays by the travel-time curve with the length of up to 6 km from the extreme SPs.

According to experimental data it is known that the thickness of spall zones is proportional to the yield of underground explosion.

In this connection it is of great interest to study this correlation based on obtained seismic data, which will confirm the justification of

the isolation of spall layer. Data in this respect is presented in Figure 6, where a high-level correlation is characteristic of the domi-

nant number of data. However, the data on six boreholes sharply deviates from such a correlation. Basement rocks uncovered by

the boreholes with deviating values are characterized by anomalous strength parameters (Young modulus, tension and compres-

sion capacity) that impact the thickness of spall area. Hence, the obtained trend can be considered as reasonably sufficient and it

can be used when assessing seismic thickness of spall zones at other regions of the Semipalatinsk Test Site.

Conclusion

The results of seismic study of the construction of the upper part of section in Semipalatinsk Test Site area allowed us to isolate

and locate technogenic region of impact of underground nuclear explosions. First of all, it is a case of zones of spallation occur-

rences formed in the vicinity of emplacement boreholes under the bottom layer of LVZ and which are characterized by thickness

from 30 to 90 m and velocity of 3.4-3.8 km/s, which at average is 1.5 km/s lower than the velocity of rocks undisturbed by explo-

sions at the same depth. During the dense carrying out of nuclear tests, spall regions form continuous structures (layers) of up to 4

km from separate emplacement boreholes. Direct correlation has been established between the thickness of spall layer and nuclear

charge weight.

For the lower lying layers at the depth of up to 120 – 150 m the velocity of rocks in modern condition is lowered with respect to its

initial values at 0.4 – 0.7 km/s, which allows reference of these areas to the excessive fissuring zones.

Taking into account a small depth of spall zones laying and, therefore, simplicity of their detection, these zones could be regarded

as one of the target objects at the stage of triggering event epicenter searching and localization during OSI. Used in this work seis-

mic technique of refracted waves and forward ray tracing method of velocity section building could be used for the spall zones de-

tection during OSI.

Figure 1. Typical configuration of the borehole UNE’s central zone Figure 2. Configuration of seismic observations

Red triangles - shot-points; blue triangles - receivers; red stars - UNE epicenters.

Figure 3. 2D velocity sections along the seismic profiles. Projections of emplacement boreholes are shown for the profiles with borehole number and yield of explosion in Kt. Red color denotes the borehole distanced from the pro-file no more than 400 m, blue - at 400-700 m. Thin lines represent velocity isolines (km/s), thick lines - seismic bor-

ders.

The bedding of constructed sections is determined by sharp changes of the apparent velocity, correlated along the profiles whereas

the gradient – by the regions of the travel-time curve segments of tracing in the surroundings of first arrivals [Belyashov et al, 2013]. As

a result, layered-gradient models have been obtained, with lateral changes of velocity in the isolated layers, which are observed in nar-

row zones, emphasizing the block structure.

Low velocity zone (LVZ) with the fixed velocity of 1.4 km/s (interval between seismic receivers of 125 m does not allow obtaining the

data on velocity at LVZ) is traced in all profiles and its thickness is changing from 5 to 125 m. At the beginning of the profiles the velocity

slowed down to 3.0 km/s corresponds to soft Jurassic deposits, limited by spillage fault from carboniferous deposits of Paleozoic base-

ment.

The number and yield of seismic layers at the sections change by density of boreholes location. Thus, profiles 1 and 3 (Fig. 3), with

dense location of emplacement boreholes, under the bottom layer of LVZ demonstrate a layer with the thickness of 50 – 80 m with ve-

locity of 3.5-3.8 km/s along their entire length. At the same time the underlying layer is characterized by velocity 4.3-4.6 km/s and the

section is described by a three-layered superincumbent medium. Profile 4 illustrates the mentioned layer in central part of the profile as

well as in the region of emplacement boreholes location. The rest of the profiles show a layer with velocity of 3.4-3.8 km/s under the bot-

tom layer of LVL that has a limited occurrence with horizontal sizes from 500 m to 2 km also in the surroundings of emplacement bore-

holes; and, in general, the structure of velocity section is presented by two-layered superincumbent model. The roof of underburden half-

space for most of the profiles underlies at the depth of 120 – 170 m and is characterized by the velocity of 4.8 – 5.3 km/s. Therefore, dis-

continuous layer with the interbedded velocity of 3.4-3.8 km/s between LVZ (1.4 km/s) and underburden layer (4.3 – 4.8 km/s) is ob-

served in the regions of emplacement boreholes location.

Technogenic changes of velocity

For the purposes of justifiable reference of the discontinuous layer based on seismic data regarding the lowered velocity to the tech-

nogenic formation, it is necessary to assess possible changes in velocity as a consequence of the underground explosions’ impact in or-

der to separate them from natural heterogeneities. Such a task was solved by comparison of changes in seismic velocity with the depth

(Fig. 3) in the surroundings of 12 emplacement boreholes and laboratory changes on the core samples from the same boreholes, col-

lected during their drilling prior to carrying out the explosions. The results are illustrated in Figure 4 for the three of them with the most

common types of rocks uncovered by boreholes and for the fourth one, where there was no explosion carried out. For all studied bore-

holes represented with different rocks, a P-waves velocity slow-down is noted for the rocks that underwent the explosive impact regard-

ing their initial condition by an average of 1.5 – 2.1 km/s within a seismic layer underlying under the bottom layer of LVZ at the depth of

from 5 to 120 m (Fig. 3). The velocity slow-down of up to 0.4-0.7 km is traced even deeper into the underburden third layer (at the depth

of up to 200 m) and corresponds to the zone of excessive fissuring that is stretched from the spallation region towards the center of ex-

plosion [Adushkin and Spivak, 2004]. For the rocks in the vicinity of borehole 1316 that have not been influenced by an explosion, the P-

waves velocity based on seismic and borehole data practically coincide.

Mapping of spall layer

Mapping of the changes in the thickness of spall layer at the studied area has been conducted along the network of parallel seismic

profiles (Fig. 5). The increase of its thickness in the left part of the map of up to 200 m is determined by low-velocity soft rocks of Juras-

sic age, separated from the rest of the area by regional spillage. For the rest of the area the thickness of the spillage layer in the rocks of

Paleozoic basement reaches up to 30-90 m, correlating with the location of emplacement boreholes. With the dense location of em-

placement boreholes, continuous stretched spall regions have been formed with lateral dimensions of up to 4 km. Spall areas have not

been detected at the places of no explosions. Areal distribution of technogenically altered rocks is controlled by the region of velocity

lowered to 3.4-3.8 km/s (Fig. 5b). Beyond the places of emplacement boreholes location the velocities increased up to 4.2 – 4.6 km/s.

Figure 4. Examples of comparison of velocity values based on laboratory measurements on the samples from boreholes prior to explosions (black dashed lines) with the results of seismic observations (red lines).

Figure 5. Maps of spall layer thickness (a) and distribution of velocity values at its roof (b). Red stars denote UNE boreholes. Dashed lines denote seismic profiles.

a b

Figure 6. Dependency of spall zone thickness on charge yield. Black diamond shapes denote normal values, red square shapes denote anomalous (with the number of a borehole) ones. Black line represents approximating