The Possibility of Earthquake Forecasting, a Few Days before the Earthquake Using GNSS Satellites

Abstract

Today, large lengths and compressions of the Earth’s crust can be measured with great precision using GNSS satellites. This makes it possible to accurately determine how much the distances were shortened in 1 day under the action of compression, every day, 24 hours a day. If most of the distances were shortened in 1 day, compression of the Earth’s crust has occurred, and it can be expected that an earthquake will occur in 2 to 3 days. An earthquake occurred in Skopje after 3 days of compression. We have also analyzed several other earthquakes that occurred after 2 days of compression: the Kraljevo earthquake of 2010, the Drežnica earthquake of 2013, and the Zagreb earthquake of 2020. Based on the conclusions of our analyses of all of these earthquakes, we propose that the Zagreb office for crisis management organize a department that would use satellite monitoring in the vicinity of faults under compression every day.

1. Introduction

Today, large distances and compressions of terrain can be measured very precisely using satellites of the GNSS (global navigation satellite system). GNSS consists of GPS (American), GLONASS (Russian), GALILEO (European), and COMPAS (Chinese). GPS (global positioning system) satellites orbit the Earth at an altitude of 20,200 km from the Earth’s surface in six orbital planes, so that the GPS receiver on Earth can always receive signals from at least four satellites.

Satellites send signals to Earth, which contain the time of the satellite’s atomic clock at the moment the signal is sent from the satellite. The receiver on the Earth’s surface records the arrival time of the antenna signal according to its electronic clock. The satellite distance is calculated based on the difference in the time of transmission and the reception of the signal. In a similar way, satellite coordinates are measured from stations on the Earth’s surface whose coordinates have previously been established. To make the calculation easier, a positioning system on the Earth’s surface was installed in Macedonia. For this purpose, a system of permanent reference stations called MAKPOS (English abbreviation for Macedonian positional system) was set up in North Macedonia to determine the position of points on the ground. Similar terrestrial positioning systems have also been developed in Croatia (CROPOS), Slovenia (SIGNAL), Serbia (AGROS), Montenegro (MONTEPOS), and Bosnia and Herzegovina (BHPOS and RSPOS) [1, 2].

In order to measure compression in the Earth’s crust, four reference stations near the epicenter of the earthquake are selected and measured by GNSS satellites 24 hours a day for several days before and after the earthquake. If almost all the distances are shortened in 1 day, compression has occurred in the Earth’s crust, after which an earthquake occurs 3 or 2 days later. In 2016, an earthquake occurred in Skopje, 3 days after compression, while the aforementioned earthquakes in Kraljevo, Drežnica, and Zagreb occurred after 2 days of compression. This phenomenon can be used to predict earthquakes in the vicinity of the fault under compression.

2. Analysis of the deformations of the Earth’s crust before and after the Skopje earthquake in September 2016

In order to see the daily changes in the distances between reference stations, we obtained the coordinates of referent stations Skopje, Tetovo, Kumanovo, and Veles from the cadaster in Skopje (Figure 1).

Figure 1.

Reference stations near Skopje.

We calculated the daily distance changes between reference stations in the vicinity of Skopje and presented the changes graphically in Figure 2 [3].

Figure 2.

Daily changes of distances between the observed MAKPOS GNSS stations around Skopje.

Figure 2 shows the decrease in all the distances between the stations on the 8th of September 2016 (i.e., the terrain has compressed), and the strongest earthquake in this area with a magnitude of 5.2 according to the Richter scale occurred 3 days later.

3. The Kraljevo earthquake

The earthquake occurred on November 3, 2010, near Kraljevo. It had a magnitude of 5.4, according to the Richter scale. We received a yearbook from our Sarajevo colleagues and at the end of the yearbook, under the foreign works cited, we found a paper in which the results of measurements after the Kraljevo earthquake were published. Later, we also received standardized measurements published in foreign literature.

In our analysis of Kraljevo earthquake, we used an elongated quadrangle with diagonals with reference stations in the AGROS network (Figure 3).

Figure 3.

Daily changes in the side distance between the AGROS permanent stations are presented in quadrangle with diagonals around Kraljevo. Almost all the sides were slightly decreased 2 days before the earthquake, while the sides E, D, and I on the quadrangle, which were further away from what was to be the epicenter of the earthquake shows no significant daily changes in distances [4].

We calculated the lengths of the sides and the daily changes of lengths and diagonals in Excel and graphically presented them in Figure 3.

4. The 2013 Drežnica earthquake

On the 30th of July 2013, at 14:58 local time, a minor earthquake occurred in Croatia with an epicenter in Drežnica, 15 kilometres northeast of Senj [5]. The earthquake had a magnitude of 4.6, according to the Richter scale and its hypocenter was at a depth of 20 km. The intensity in the earthquake epicenter was VI-VII on the Mercalli-Cancan-Selberg (MCS) scale.

The earthquake was felt in the Ogulin, Karlovac, and Zagreb areas, as well as in southern and central Slovenia. The county emergency communications centre in Karlovac received only two reports of damage. Given the low intensity of the earthquake, only minor damage was possible in the epicenter area.

We calculated the daily changes in distances and pressure and presented them graphically in Figure 4. From the graph, it can be seen where the compression was.

Figure 4.

For our analysis, we chose four crops reference stations in Karlovac, Rijeka, Senj, and Slunj, calculated the daily distance changes and presented them graphically. We saw that almost all distances were shortened on the 28th of July 2013 and 2 days later, on the 30th of July 2013, there was an earthquake [6].

5. The 2020 Zagreb earthquake

The Croatian geodesy journal Geodetski list [7] provides a brief overview of the earthquakes that occurred in Zagreb and its surroundings in the past, as well as of the latest earthquakes, which occurred on the 22nd of March 2022, and had magnitudes of 5.4 and 5.0, respectively. These earthquakes, which caused extensive damage to more than 25,000 buildings, occurred 140 years after the great earthquake that devastated Zagreb in 1880.

We chose four reference stations (Figure 5) from which GNSS satellite signals were received 24 hours a day. The signals were processed in the Geodetic Administration with the Bernise program package. With the help of these measurements, the daily changes in length were determined. We presented them graphically in Figure 6.

Figure 5.

Crops reference stations for the analysis of daily distance changes before and after the earthquake the Zagreb earthquake on March 22, 2020.

Figure 6.

Daily changes in distances (the Zagreb earthquake on March 22, 2020) [7].

Figure 6 shows that the Earth’s crust was compressed on March 20th, 2020, which meant the earthquake would occur in 2 days. Since approximately half a day is spent on the calculation, this means that an earthquake could be announced one and a half days before.

A similar elongated quadrangle with diagonals was used in the analysis of the Kraljevo [4], Drežnica [5], and Zagreb [7] earthquakes (Table 1).

	Earthquake magnitude according to the Richter scale	Pre-earthquake compression (days)	Pre-earthquake compression if calculation time is deduced (days)
Kraljevo, 2010	5.4	2	1.5
Drežnica (Senj), 2013	4.6	2	1.5
Skopje, 2017	5.2	3	2.5
Zagreb, 2020	5.5	2	1.5

Table 1.

A list of earthquakes that have been analyzed.

6. Analysis of terrain deformations using daily distance changes in the MAKPOS network before and after the main earthquake, that is, from the 1st of January 2016 to the 30th of October 2016, around Skopje

In order to investigate whether there was previous compression of the terrain before the main earthquake on September 11, 2016, we analyzed the daily changes in the distance between GNSS stations in the vicinity of Skopje since the beginning of the year (i.e., 7 months before the main earthquake).

Several minor earthquakes occurred in Macedonia in the vicinity of Skopje on September 11, 2016. The third and largest of them occurred at 15:10 (local time) and had a moderate magnitude of 5.2 on the Richter scale. The epicentre of the earthquake was located at latitude φ = 41.98° and longitude λ = 21.50°, and its hypocentre was at a depth of 4 km. It was felt in Kosovska Mitrovica, Niš, Vranje, and Belgrade.

A similar quadrilateral with diagonals was used in the analysis of earthquakes in Kraljevo [4], Drežnica [5], and Zagreb [7].

We also calculated daily changes in the distance between MAKPOS reference stations in the vicinity of Skopje for the months of February, March, April, May, June, July, August, September, and October of 2016. They are graphically shown in Figures 7–15 in our paper [8].

7. Conclusion

This analysis shows that compression of terrain happened before each earthquake. Two to nine days after compression, an earthquake occurred. Only once, on the 29th of July 2016, was there no earthquake after compaction of soil, because the soil layers did not crack.

We determined the daily changes in the distance between GNSS stations in the vicinity of Skopje for January 2016. The graph of the daily changes for January is shown in Figure 7. This graph shows that on the 12th of January 2016, there was a relatively large extension of Skoplje-Tetovo and Tetovo-Kumanovo distances. These shifts were about 3.5 cm in length. The spaces between the plates were large, but after the collision of the plates, the direction of the plates’ motion changed. Nine days after the compression in January, a smaller earthquake with a magnitude of 2.0 on the Richter scale occurred (Figure 7).

Figure 7.

We also examined the daily changes in longitude from the beginning of 2016 until the main earthquake. For January 2016, we got the daily length changes shown in Figure 7. We can see that there was one small earthquake in January, which was preceded by a small compression. We can see that the two lengths became longer at the beginning of the year. This is typical for such situations because at that time the distance between the plates was large. After the collision, their direction changes, and they become shorter. When it happens later it is usually a sign that the earthquake will be weak. The same thing happened several other months before the earthquake. Graphs for other months in which minor earthquakes occurred can be found in source [8].

8. Proposal

Based on our analysis of the four earthquakes, my brother and I can confidently propose that a department be established at the Geodetic Administration and the Faculty of Geodesy that will monitor the daily changes in the length of the Earth’s crust using our method with GNSS measurements. The department would initially act in the Zagreb area, where the population is largest, and it would move on to other seismically active areas of Croatia once it has proved to be effective. That department of the Geodetic Administration would have almost all the instrumentation necessary for such analyses. There would be no great need for additional funding for the instruments. Furthermore, the Geodetic Administration already has an excellent expert, who would be ideal for the task of processing our geodetic measurements. All measurements and calculations should be automated. All of this should be included in the annual financial plan of the Geodetic Administration and the Faculty of Geodesy. Also, the geodetic expert from the Geodetic Administration should invite colleagues from the Faculty of Geodesy and the Seismology Department of the Faculty of Science to work with them. The Zagreb office for crisis management should financially support this initiative. In addition, the project could be EU-funded. We would be overjoyed if my friends, as well as all inhabitants of Zagreb and Croatia, become safer from, and more prepared for, earthquakes.

Acknowledgments

We would like to thank our colleagues Dimeski, Bogdanovski, and Postalovski from North Macedonia for the many measurements we received from them. We also received measurements from the Croatian Geodetic Administration, although, these (regrettably) did not include measurements for the Petrinja earthquake.

We would also like to thank our colleague Željko Gulija, who introduced us to the colleagues from Macedonia.

We would also like to thank our colleagues from Sarajevo for their yearbook, which included information about and the measurements of the Kraljevo earthquake.

We would also like to thank Professor Eugen Prelogović for his advice on the analysis of measurements.

We would also like to thank Professor Walter Salazar for the corrections he made to our texts.