**4. Analysis of the Kronotsky earthquake, taking into account the determination of the error in the depth of the event**

Let us consider the case when the error in determining the depth of the earthquake hypocenter Δ*hmist* ≤ 5 km is taken into account. **Figures 6** and **7** show the results of wavelet decompositions for 1994, 1995, and 1996 and from 1995-01-01 to 1997-12-04 for all the studied sectors. As can be seen from the figures, taking into account the error in determining the depth of the hypocenter leads to a sharp reduction in statistics. In some sectors, there are no events at all, which, of course, complicates the analysis. **Figures 8** and **9** for the region *S*<sup>Σ</sup> present the summed wavelet coefficients from scale level 1 to 32 for the probability distributions *P*(Δ*h*) of earthquakes with *KS* ≥ 8.5 over depth intervals Δ*h* = 1 km and the error in determining the depth of hypocenters for. For comparison, **Figures 8** and **9** show the intensity distribution of seismic events over the period of instrumental observations from 1962-01-01 to 2021-12-31, taking into account Δ*hmist* ≤ 5 km, against which events develop over the considered time period 1994–1997. As can be seen from **Figures 8** and **9**, taking into account the error leads to a significant change in the depth distribution of earthquake intensity. So in **Figure 8**, the peak of the maximum seismic activity at 40 km for 1994, which is clearly distinguished in **Figure 5**, is completely absent. Instead of a maximum at this depth, we have a minimum.

In addition, **Figure 8** for 1994 shows several clear intensity peaks in the altitude range of 18–40 km (Coeff ≈ 1.4, Coeff ≈ 1, and Coeff ≈ 1.2), 40–65 km (Coeff ≈ 0.7), and 65–90 km (Coeff ≈ 0.4). In turn, the intensity at shallow depths from 0 to 10 km for 1994-01-01/1995-12-31 slightly exceeds the intensity for the period 1962-01-01/ 2021-12-31. The absence of a maximum at a depth of 40 km in **Figure 8** indicates that the error in determining the depth for 1994 is mostly greater than the selected *Δhmist* ≤ 5 km, so these earthquakes were not included in the statistics and were sifted out. In turn, in 1996, the accuracy in determining the depth of hypocenters equal to 40 km

*Investigation of the Dynamics of the Seismic Regime in the Kamchatka… DOI: http://dx.doi.org/10.5772/intechopen.109069*

#### **Figure 6.**

*The summed wavelet coefficients from scale level 1 to 32 for the probability distributions* P*(Δ*h*) of earthquakes with energy class* KS ≥ *8.5 over depth intervals Δ*h *= 1 km for 1994 and 1995, taking into account the error in determining the depth Δ*h*mist* ≤ *5 km.*

#### **Figure 7.**

*The summed wavelet coefficients from scale level 1 to 32 for the probability distributions* P*(Δ*h*) of earthquakes with energy class* KS ≥ *8.5 over depth intervals Δ*h *= 1 km for 1996 and for the period from 01.01.1997 to 04.12.1997, taking into account the error in determining the depth Δ*hmist ≤ *5 km.*

mostly satisfied the accuracy of Δh*mist* ≤ 5 km, and as a result, the intensity peak was identified (see **Figure 9**).

It follows from the analysis of **Figure 8** that the intensity of the wavelet coefficients at depths from 0 to 10 km in 1996 increased sharply compared to 1994 and

#### **Figure 8.**

*The summed wavelet coefficients from 1 to 32 scale decomposition level for the probability distributions* P*(Δ*h*) of earthquakes with* KS ≥ *8.5 over depth intervals Δh = 1 km and an error in determining the depth of hypocenters Δ*hmist ≤ *5 km. The periods under consideration are 1994, 1995, and 1962–2021.*

#### **Figure 9.**

*The summed wavelet coefficients from scale level 1 to 32 for the probability distributions* P(*Δ*h) *of earthquakes with* KS ≥ *8.5 over depth intervals Δ*h *= 1 km and an error in determining the depth of hypocenters Δ*hmist ≤ *5 km. The periods under consideration are 1996, 1997.01.01–1997.12.04, and 1062–2021.*

1995. This happened, as we know, as a result of aftershock activity after the earthquake of June 21, 1996.

If we compare **Figure 5** with **Figures 8** and **9**, then the latter show some chaotization in the distribution of wavelet coefficient intensity peaks over depths with similar Coeff values. A similar picture is also observed in the averaged distributions of wavelet coefficients for 1962-01-01/2021-12-31, taking into account the determination of the error in the depth of hypocenters and without taking into account. This suggests that with such a choice of the numerical value of Δ*hmist*, we simply lose information about the intensity of the distribution of events over *h*.
