Appendix 1
POSSIBLE FORECASTS OF SOME NATURAL CATACLYSMS AND COSMIC PROCESSES
Introduction
The first IC GEOCHANGE GCGE report can hardly be called complete without this appendix. The analysis conducted on the dynamics of statistics of many natural hazards, geophysical and cosmic parameters has showed their tendency to increase substantially since about 1998. However, it is obvious that the main point of this study is not about a formal statement of facts but rather in a possible prediction of future events.
Not only must we demonstrate the evolution of the dynamics of the number and scale of natural disasters, but we must also suggest possible patterns of future development of events, that is, provide a longterm forecast for Earth’s most dangerous hazards. This section does not address other types of disasters, which is planned for the next IC GCGE reports.
Attitude to the problem of forecasting natural disasters may be disputable; therefore this section is not included in the basic contents of the first IC GCGE report, being instead given as a special Appendix 1.
We are not trying to predict specific events since it is too complicated and controversial an issue. Our objective is longterm forecasting of changes in the next decade’s dynamics of global seismic and volcanic activity and tsunami manifestations. Tsunamis typically result from seismic and volcanic activity except for rare cases when they may be caused by other geological processes.
In producing longterm forecasts, we have been relying on the wellknown principle which longterm forecasts in all areas of science are based on. The principle is as follows: “To look into the future, one must study the past well”.
Longterm forecast of Earth’s seismic activity dynamics
Methodology
Following the basic principle of longterm forecasting, we have attempted to examine the regularity in the evolution of monthly numbers of earthquakes of different magnitudes and for different time intervals.
The main purpose of this research is to identify Earth’s seismic activity dynamics patterns.
One of the fundamental regularities of all natural processes is cyclicity. Revealing the objectively existing cyclicity in Earth’s seismic activity dynamics is an important aspect for longterm forecasting. Meanwhile, there are special techniques for detecting hidden periodicities in the time series of different processes. These techniques include linear and nonlinear transformations of time series. Linear transformations may refer to various types of averaging of time series for different time intervals. As to this problem, we apply the running average method. Another approach is based on spectral analysis, which helps identify different harmonics in the time series. If both methods are employed without any special technology, an incorrect result may be obtained because it heavily depends on the length of the filter and other preset parameters. However, there are special methods to use spectral analysis with as objective a result as possible. This technique is described by the author in a number of works (E.N.Khalilov, 1987; V.E.Khain, E.N.Khalilov, 2008; 2009).
As numerous studies by various authors (Sh.F. Mehdiyev, E.N. Khalilov, 1987; V.E. Khain, E.N. Khalilov, 2009) have demonstrated, there are cycles of different orders in volcanic and seismic activity, ranging from hundreds of millions of years to several months.
However, this chapter does not address detailed time series analyses. In this case, we have only applied initial treatment and trend analysis to identify a general tendency in processes as they progress with time by approximating them with simpler functions (straight line, sine curve, polynomial trend, exponential function). We find it very interesting to reveal trends in global seismic activity for different time intervals and earthquakes of different energy.
Research
During the first phase, the evolution of monthly numbers of earthquakes with M> 6.5 was studied for the period between 1976 to May 2010 smoothed with 11month running averages. Fig. 84 provides a diagram for variations of monthly numbers of earthquakes with M> 6.5 together with a straightline trend and a sinusoidal trend both reflecting the dynamics and cyclical nature of the studied process. The cycles described by the sine curve have a period of about 18 years. Drawing the sine curve further along the straightline trend from May 2010 to 2016 allows us to forecast the general dynamics of changes in monthly numbers of earthquakes. Thus, according to the sine curve’s projected segment, a rise in the seismic activity level is expected for 2010 to 2016.
The straightline trend also indicates the stable dynamics of increase in monthly numbers of earthquakes in the course of time. Thus, the overlapping straightline and sinusoidal trends intensify the total effect of increased earthquake numbers from 2010 to 2016.
Fig. 84. Graph for monthly numbers of M >6,5 earthquakes between 1976 and 2010 projected to 2016
by highlighting sinusoidal trend (by E.N.Khalilov, 2010, according to USGS data) Sinusoidal trend with projected seismic activity segment is marked in red; number of earthquakes graph smoothed with 11month running averages is marked in black; straightline trend is marked in green; figures 117 denote 23 year seismic activity cycles.
At the same time, the cycles with periods ranging from 1.5 to 3 years and numbered 117 are clearly seen in the diagram. By superimposing those cycles on the sine curve, we obtained a forecasted graph for global seismic activity from 2010 to 2016 which contains two minor seismic activity cycles with average periods of 23 years. Within the first cycle, peak numbers of large earthquakes are expected for 2011 with a subsequent relative decrease in activity in 2012, and the second, higher peak of seismic activity is forecast for 20132015 to be followed by an expected decline.
For greater research objectivity, we tried a different approach to longterm forecasting of global seismic activity. Fig. 85 contains a graph for monthly numbers of M> 6.5 earthquakes for the period between 1976 and May 2010. The highest peak values of the number of earthquakes are indicated with red dots. If we take a closer look at the graph, we will notice that the distances between the peak values (red dots) correspond to the periods of the cycles highlighted above, 1,53 years on average. We have drawn a trend enveloping the peak values marked with red dots, which is described by a sine curve as well with a period of 1718 years, as seen from the image.
Fig.85. Graph for monthly numbers of M> 6,5 earthquakes between 1976 and 2010 forecasted until 2015 by highlighting sinusoidal trend (by E.N.Khalilov, 2010, according to USGS data) Graph for monthly earthquake figures is marked in red; sinusoidal trend enveloping highest peaks of monthly earthquake figures is marked in green; number of earthquakes graph smoothed with 11month averages is marked in black; straightline trend is marked in yellow; projected segment of seismic activity graph is marked in blue.
The highest part of the enveloping sine curve falls within the time period between 2010 and 2015 as well. Employing the same principle that sums up the dynamics of the straightline and sinusoidal trends and the cycles with a period of 23 years, we get the forecasted segment of the graph where 2011 and 2013 indicate the highest levels of Earth’s global seismic activity.
Longterm forecasting of catastrophic earthquakes with M> 8 is also a matter of interest. To that effect, we have drawn a graph for the dynamics of annual numbers of large M> 8 earthquakes for the period from 1980 to 2010, Fig. 86. The diagram shows a straightline trend and a sinusoidal trend both describing the general nature of the dynamics of the global seismic process. The straightline trend points to the stable dynamics of growth in the number of large earthquakes in the course of time. The sinusoidal trend helps reveal some cyclicity with a period of 17 years. Thus, the periods of cycles revealed for large earthquakes and for M> 6.5 earthquakes coincide for the period between 1976 and May 2010. In addition, the diagram also exposes some cycles with periods of 1,53 years on average, which is also in line with the results obtained earlier.
Fig.86. Graph for numbers of M> 8 earthquakes between 1980 and 2010 forecasted until 2016 (by E.N.Khalilov, 2010, according to USGS data) Annual number of earthquakes graph is marked in blue; sinusoidal trend is marked in red; straightline trend is marked in black; forecasted seismic activity graph for M>8 earthquakes is marked in green.
Using the principle that sums up the dynamics of the straightline and sinusoidal trends and the cycles with a period of 23 years, we get the graph’s forecasted part (marked in green) where 2011 and 2013 indicate the highest levels of Earth’s global seismic activity, with a relative minimum in 2012. 2016 is expected to see a substantial decline in seismic activity.
Let us investigate the dynamics of catastrophic M> 8 earthquakes for the period between 1900 and May 2010 for the purpose of longterm forecasting, Fig. 87.
Fig.87. Graph for numbers of M> 8 earthquakes between 1900 and 2010 forecasted until 2016 (by E.N.Khalilov, 2010, according to USGS data) Annual number of earthquakes graph is marked in darker blue; sinusoidal trend is marked in red; curve enveloping peak values of earthquake numbers graph is marked in black; straightline trend is marked in lighter blue; forecasted seismic activity graph for M> 8 earthquakes is marked in green.
As in the previous cases, the graph here is approximated with a straightline (lighter blue) and sinusoidal (red) trends, Fig. 87. The straightline trend indicates a steady evolution of numbers of catastrophic earthquakes in the course of time. The sinusoidal trend reveals longer cycles of seismic activity with a period of 75 years (19051980).
The peak of the next global seismic activity cycle falls within the period between 2011 and 2015. The diagram also clearly shows the cycles with an average period of 2 to 3 years. Summing up the dynamics of the straightline trend, sinusoidal trend and the cycles with a period of 23 years, we get the forecasted part of the graph (marked in green) with peaks in 2011 and 2013 and a relative minimum in 2012.
As stated above, the peak values of the highest seismic activity cycles can be yet another indicator as marked with red dots on the diagram. Peak values distribution is most effectively described by a parabolic trend marked in black in the diagram. The parabolic trend has allowed us to determine the approximate amplitudes of the forecasted cycles of seismic activity with peaks in 2011 and 2013.
Longterm forecast of Earth’s volcanic activity dynamics
In forecasting global volcanic activity, we have employed the same methods and approaches that have been used in forecasting global seismic activity.
Fig. 88 contains a graph for annual numbers of world volcanic eruptions between 1900 and 2009 forecasted up to 2016. The graph is approximated with a sinusoidal trend and a straightline trend. The latter reflects the stable dynamics of annual increase in the number of volcanic eruptions while the former indicates a certain cyclicity in the observed process. The sine curve allowed us to identify cycles with a period of about 26 years. Of course, these cycles are not as apparent as the double cycles with a period of 57 years consisting of shorter cycles with a period of 2,5 to 3,5 years on average. Thus, these cycles are similar to the cycles of global seismic activity with a period of 23 years. Summing up the effects of superimposing the straightline and sinusoidal trends, the diagram shows the projected part with two activity cycles highlighted as well, with peaks in 2011 and 2013 and a local minimum in 2012.
Fig.88. Graph for annual numbers of world’s volcanic eruptions between 1900 and 2009 forecasted until 2020 (by E.N.Khalilov, 2010, according to Global Volcanism Program) http://www.volcano.si.edu/world/find_eruptions.cfm Graph for annual volcanic eruption numbers is marked in blue; sinusoidal trend is marked in red; straightline trend is marked in green; 117 are cycles of volcanic activity.
What is the reason of such a high amplitude of the forecasted global volcanic activity cycles? The answer to this question has a logical basis. For the period between January 01, 2010 and May 31, 2010, 52 officially confirmed volcanic eruptions occurred according to the Global Volcanism Program. Therefore it can be expected that by the end of 2010 at least 90 eruptions will have taken place (in a year). In our view, cycle amplitudes indicating 100 eruptions in 2011 and about 110 eruptions in 2013 are quite acceptable.
Longterm forecast of the dynamics of major tsunami numbers
Forecasting of major tsunamis depends to some degree on forecasting of strong, tsunamigenerating earthquakes and volcanic eruptions. We have studied a possible pattern of evolution of numbers of large tsunamis for the next five years.
Fig. 89 provides a graph for the numbers of catastrophic tsunamis which have occurred between 1990 and May 2010, according to ITIC (International Tsunami Information Centre). The analysis of tsunami dynamics demonstrated that for the considered time period, two distinct cycles with a period of 3 years can be singled out, which correspond to the cycles identified in the dynamics of annual numbers of large earthquakes and volcanic eruptions. This is quite logical since for the most part, tsunamis are directly related to large earthquakes and (submarine) volcanic eruptions.
Fig.89. Graph for tsunami numbers between 1900 and 2010(by E.N.Khalilov, 2010, according to ITIC – International Tsunami Information Centre data) Tsunami numbers graph is marked in darker blue; straightline trend is marked in lighter blue; forecasted graph for 20102015 tsunami numbers is marked in red.
The straightline trend points to a steady increase in the number of catastrophic tsunamis in the course of time. By analogy with longterm forecasting of global seismic and volcanic activity, the graph’s projected part is marked in red, with two highlighted cycles of increased catastrophic tsunami numbers with peaks in 2011 and 2013 and a local minimum in 2012.
Longterm forecast of solar activity
Forecasting of solar activity is one of the most important tasks solar activity studies face. It can be divided into three basic types, which are shortterm (up to 10 days), mediumterm (up to several months) and longterm (up to several decades) forecasting. Solar activity forecasts are of great practical importance since the impact on people of solar activity manifestations, first of all magnetic storms and increased solar radiation penetrating to Earth’s surface, can be considered proven for now. That is why people in many countries are warned of the approaching magnetic storms and those periods are deemed most dangerous for persons engaged in a highrisk professional activity (operating all kinds of sea, land and air transport, etc.). The increased solar activity expressed by powerful solar flares, solar wind and magnetic storms can have very dangerous consequences for the stability of human activity and affect the stable operation of radio communication systems and sophisticated electronic equipment. However, the greatest danger of high solar activity comes from its effect on climate and many natural disasters as evidenced by various scientists’ research findings described in section 5.3.
As far as this study is concerned, we are interested in longterm forecasting only. Despite the fact that sufficiently pronounced cycles have been revealed in solar activity, longterm forecasting even for wellstudied 11year cycles is quite complicated a task. This is evidenced by the fact that virtually not a single prediction made by various world scientists and organizations in forecasting the 24^{th} 11year cycle has been verified yet. Many forecasts are based on creating physicomathematical models to describe the process of solar activity growth. We do not aim to discuss these models, confining ourselves to just giving the evolution of NASA (United States National Aeronautics and Space Administration) forecasts shown in Fig. 90. As can be seen from the image, the 24^{th} solar activity cycle forecasted in March 2006 had a peak value in 2012. The predicted amplitude of the 24^{th} cycle was considerably higher than that of the 23^{rd }cycle. The forecast of January 2009 demonstrated a more moderate amplitude – at the same level as or slightly lower than the 23^{rd} cycle’s amplitude. In June 2010, the peak of the forecasted 24^{th} 11year solar activity cycle shifted to 2013, with its amplitude shown as being significantly lower than that of the 23^{rd} cycle.
Fig.90. Evolution of NASA predictions for solar activity (1) is number of sunspots in 23^{rd} cycle and prediction for 24^{th} cycle (NASA, March 2006); (2) is number of sunspots in 23^{rd} cycle and prediction for 24^{th} cycle (NASA, January 2009); (3) is number of sunspots in 23^{rd} cycle and prediction for 24^{th} cycle (NASA, June 2010).
What caused those changes in NASA predictions in different years? First of all, from the very start the progress of the 24^{th} solar activity cycle followed a different scenario than had been presumed in various models. In the first place, the 24^{th} cycle did not begin in 2008 as expected, but rather at the end of 2009. As a result, the physicomathematical models previously considered the most successful were refuted.
Another service providing solar activity forecasts is NOAA (United States National Oceanic and Atmospheric Administration). The predictions presented by NOAA in different years had similar dynamics, which is quite logical. Fig. 91 contains one of the forecasts provided by NOAA in May 2009.
In our view, for a more objective forecasting of solar activity it would be useful to consider a longer period of display of one of the most important parameters of solar activity, the solar constant. The point is that unlike the Wolf numbers (for sunspots) resting on the rather formalized solar activity index which cannot be clearly expressed in terms of energy, the solar constant reflects the changes in solar radiation per unit area.
The graph for solar constant variations has both similarities and significant differences with the Wolf numbers. The similarity lies in the fact that this graph also shows up the 11year solar activity cycles fully correlating with the similar cycles in the Wolf numbers.
At the same time, as can be seen from the solar constant graph for 1611 to May 2010 (Fig. 92), the amounts of radiated solar energy at the maximum and minimum values of the 11year solar cycles vary considerably for different years, which is not observed in the Wolf numbers. Thus, the solar constant has a pronounced amplitude modulation, apparently due to superimposition of larger solar cycles with a different scale.
Fig.92. Possible models for longterm solar activity forecasting (1) is model 1 for solar activity forecasting; (2) is model 2 for solar activity forecasting; graph for actually registered values of solarconstant from 1611 to May 2010 is marked in yellow; forecast graphs of solar activity are marked in blue; A, B, C are 8090year cycles of solar activity.
In particular, Fig. 92 (1) demonstrates the presence of three major cycles  A, B, C with a period of 8090 years. The maximum values of cycle A (1780) and cycle B (1838) have almost the same amplitude whereas the amplitude of the peak values of cycle C (1959) is much higher.
Thus, as stated by many scientists engaged in solar activity studies, there are larger cycles standing out against the background of the 11year solar activity cycles. However, in our view, the specifics of largescale variations of the solar constant may help in forecasting the amplitude of the 24^{th} solar activity cycle. To do that, we used the method of trend mirroring (E.N.Khalilov, 2010). The essence of the method is that any trend can be viewed as part of a larger cycle of the process in question. In this case, to predict the possible development of the process, the trend can be mirrored as a continuation of the actually observed process, i.e. as its forecasted part. That is the way how one or another model of the possible development of a process can be formed if we are not aware of the considered process’ development pattern for a longer time interval.
Fig. 92 (1 and 2) provides a review of two possible models for further evolution of solar activity. By mirroring the left side of the graph on its extension, it is assumed that the time span between 1675 and 1975 is half the period of a larger solar activity cycle with a period of 610 years, Fig. 92 (1). In that case, the low amplitude of the 24^{th} cycle indeed becomes obvious. This cycle may reflect another cycle with a period of 554 years, highlighted by D. Schove (Y.I.Vitinskii, 1976).
The second model of solar activity evolution suggests that the trend observed in solar constant variations reflects part of a cycle longer than 610 years, which we may be unaware of. In that case, the 24^{th} solar activity cycle will have a higher amplitude than the 23^{rd}.
So, we have two conceptually possible solar activity evolution models where largerscale cycles are described by the solar constant trend. Both models definitely contain 11year and 85year solar activity cycles.
The first model (1) is unambiguous since in such a course of events, the large cycle’s total period is just about 610 years. This cycle’s symmetry axis falls approximately on 1975. With such developments as mentioned above, it is to be expected that the 24^{th} solar cycle will be lower than the 23^{rd}.
The second model (2) is ambiguous in terms of the largescale cycle’s period length. The second model’s symmetry axis falls approximately on 2071, but it can move to the right if the period of the major cycle is still longer. Therefore, we cannot speak definitely about the time period for the largescale cycle in the second model. At the same time, the amplitude of the 24^{th} cycle in the second model is expected to be higher than that of 23^{rd}.
For the present (prior to May 31, 2010), it is not possible to state that the evolution of solar activity exactly follows one of the models. The continuing low activity level at the beginning of the 24^{th} solar cycle may not be an indicator of its low amplitude in 2013. Within the next few years, nature will answer this question more precisely.Прослушать
CONCLUSIONS
We have carried out longterm forecasting of the evolution of global seismic, volcanic, tsunami and solar activity. The forecasting was based on identification of cyclicities and other regularities in the distribution of numbers of earthquakes, volcanic eruptions and tsunamis for past periods of time and use of the established regularities in the development patterns for future processes.
All longterm forecasts for natural disasters have been made for the period between 2010 and 2016. Two cycles of increased activity with peaks in 2011 and 2013 and a local minimum in 2012 have been identified in longterm forecasts for large earthquakes, volcanic eruptions and tsunamis. By 2016, a decline in activity of all geodynamic cataclysms is expected.
Global changes in a number of geophysical parameters and the high correlation of the period of “explosive intensification” of natural disasters throughout the entire volume of Earth including the lithosphere, hydrosphere and atmosphere over the past two decades – all are indicative of release of an unusually high level of extra endogenous and exogenous energy.
The expected activity of natural disasters may have very serious negative consequences for the stable progress of civilization, leading to death and destruction unprecedented in human history. Economic implications for countries prone to natural disasters may be catastrophic.
It is necessary to unite scientists, international organizations and governments of various states under UN auspices in order to take effective measures to counter natural disasters and minimize casualties and damage they cause to humanity.
