Final Report Summary - PRE-EARTHQUAKES (Processing Russian and European Earth Observations for Earthquake Precursors Studies)
A large scientific documentation is today available about the appearance of anomalous space-time patterns of geophysical parameters measured from days to week before earthquakes occurrence. Nevertheless, up to now no one measurable parameter, no one observational methodology has demonstrated to be sufficiently reliable and effective for the implementation of an operational earthquake prediction system. To this aim, the combined use of different observations / parameters together with the refinement of data analysis methods, are expected to give major improvements in order to reduce false alarm rates and improve reliability and precision (in the space-time domain) of predictions.
These are the premises of PRE-EARTHQUAKES project which committed European Union (EU) and Russian researchers to integrate different observational data and technologies, to improve, by cross-validating, different data analysis methods, in order:
- to improve our knowledge of preparatory phases of earthquakes and of their possible precursors;
- to promote a worldwide earthquake observation system (EQuOS) as a dedicated component of the Global Earth Observation System of Systems (GEOSS);
- to develop and offer to the international scientific community an integration platform where independent observations and new data analysis methodologies devoted to the research on/of earthquake precursors can be collected and cross-validated.
To these aims the project followed three mainstreams devoted:
- to coordinate and realise a systematic data acquisition and product generation in predefined (standardised) formats;
- to define and implement a common integration platform where heterogeneous data inputs can be ingested, organised and compared;
- to validate and disseminate data analysis and integration methods / tools making them available and open to the contribution of the worldwide scientific community in order to further extend the number of contemporary monitored parameters and to improve quality of data analysis methods.
To this purposes, observations from more than 20 different satellite systems (particularly those from European, Russia and United States (US) space agencies) and about 100 ground based stations were systematically collected and analysed by using different data analysis methods, looking for anomalous variations of surface and atmospheric parameters (up to the ionosphere) possibly related to impending earthquakes.
The project developed in two main phases:
- A first learning phase devoted to integrate independent observations on selected events occurred in the past in order to perform an a posteriori evaluation of the performance exhibited by the selected observational techniques (parameters, observation and data analysis methods) and, in case, to try to enlarge / improve them for the following project development.
- A second real-time integration and monitoring phase (pre-earthquakes real-time integration and monitoring experiment (PRIME)), devoted to apply and evaluate, for a limited time period on selected test areas, the performance of a pre-operational system based on the best of the knowledge / capabilities developed in the first phase of the project.
The learning phase was performed on past events occurred in three main selected areas (Italy, Turkey and Sakhalin Peninsula). The PRIME duration was of five months and allowed to verify, in collaboration with selected institutional and scientific end-users, the possible contribute of PRE-EARTHQUAKES strategy, to the dynamic assessment of seismic risk (DASR) in a pre-operational context.
Even after a so short time of study (only two years for so challenging objectives) results achieved have far exceeded all expectations and from many points of view.
Data product generation and integration:
- The product generation (PG) capability expressed by the consortium was impressive: at the end of the second year of activity PG = 6165 is the number describing the variety of measurements performed (i.e. corresponding to different parameters or technologies or algorithms or testing areas or period of observation).
- Similarly impressive the products integration (PI) capability expressed by the consortium at the end of the second year, PI = 2011 represent the number of the different combination of independent observations which have been compared.
- At the end of the project, the capacity of the consortium to mobilise innovation was quite impressive in terms of new methodologies (more than 50 % of the algorithms were significantly improved, several completely new proposed) and new data products (at the end of the project quite doubled compared with the ones initially planned).
- The adoption of international standards for data exchange, the definition of an original information protocol (PRE-EARTHQUAKES policy on non disclosure of information ) particular important in a so delicate field, together with the development of the pre-earthquakes geoportal (PEG) as a common and open access real-time integration and visualisation tool, represented the key choices for facilitate the growing of the consortium and its possible continuation after the nominal end of the project.
Scientific results
Even if based on a too short experimentation time and limited observational capabilities, results so far achieved appear quite interesting under different points of view. We will mention here only the most significant.
- A clear evidence was reached of the importance of an integrated observational approach in reducing (up to zero) false alarms rates.
- Shocking evidence of the importance of the continuity of the observations which permitted to repeat, with an highest confidence, a second successful prediction in the same seismic area.
- Any difficulty to guarantee continuity of the monitoring systems, including natural factors which can often 'blind' them (e.g. cloud presence in case of satellite observations, geomagnetic or solar storms for ionospheric measurements) strongly increases the probability to 'miss' events. This circumstance can discourage, without any scientifically funded reason, other scientists or potential end-user to further consider the potential of such monitoring systems. It is instead quite clear from the PRE-EARTHQUAKES experience that insufficiency or discontinuous coverage of the monitoring techniques make only a small fraction of occurring earthquakes (possibly) predictable independently from their principal characteristics (like magnitude, depth etc.).
- However, in five months of real-time monitoring PRE-EARTHQUAKES practically demonstrated that (at least) in some case, earthquakes can be predicted (provided that a suitable observing system is in place). Like in the case of missed events, this circumstance depends not only on earthquakes characteristics but (often much more) on the actual availability of usable observations.
- The two last points are of fundamental importance for the future of the research in this field as, together with the high rate of false positives, represent one of the main point of criticism from scientists asking (without any apparent justification) 100 % of efficiency to systems that, instead, can have evident societal utility even at lower level of performance.
- A not minor results achieved by the direct collaboration with the PRE-EARTHQUAKES end-users is the possible use of PRE-EARTHQUAKES strategy within a pre-operational context. It seems quite evident that (and recent seismic crisis recently occurred in Italy demonstrated how much urgent is to quickly move in this direction) more than (or waiting for) a precise earthquake prediction system (presently not available) what decision makers actually need is a system which, integrating in real-time all relevant data and information, is able to guarantee a dynamic assessment of seismic risk (DASR) useful for adjusting the intensity of the prevention actions to be put in place by local authorities in presence of seismic crisis like the ones recently experimented in Italy in the case of l'Aquila, Emilia and Pollino earthquakes.
Enlarging the community looking for continuity
One of the main successes of PRE-EARTHQUAKES was its communication strategy and the capability to enlarge the boundaries of its initial (very limited for reason of budget) partnership. To define the conditions for allowing other institutions to participate to project efforts increasing its results and overall impacts was a not small effort. It required the definition of specific communication protocols and rule of enrolments ensuring the appropriate use of the information generated within the consortium which are potentially critical for their possible societal impact. Thanks to a dissemination effort well beyond the one initially planned and significantly supported by partners' internal resources, in the first week after the first call for enrolling PRE-EARTHQUAKES networking members, more than 20 request of enrolment were submitted and only the scarce capability to guarantee systematic observations (which are the fundament of PRE-EARTHQUAKES growing strategy) expressed by most of candidates, prevented the consortium to enrol more than four of them as networking members until now. The good news is that PRE-EARTHQUAKES efforts will continue (under the EQuOS-GEOSS umbrella and, hopefully, with a COST financial support for which a proposal already passed the first selection). Anyway the capital of expertises, integration tools, communication standards and data product generation automatic chains, is still there and do not want to be dispersed. Most of the partners and networking members are continuing to systematically generate their data products, PEG integration tool and website maintenance will be guaranteed by UNIBAS internal resources until the completion of the above mentioned passages to a different internal organisation. After only two year of (short but intense) activity PRE-EARTHQUAKES has been recognised as one of the worldwide best-practices of research in the field of short-term earthquakes prediction, to which several national projects (like the ones recently promoted by the Italian government or re-proposed in Taiwan) are presently looking with particular attention.
Project context and objectives:
A large scientific documentation is to-date available about the appearance of anomalous space-time patterns of geophysical parameters measured from days to week before earthquakes occurrence. Nevertheless, up to now no one measurable parameter, no one observational methodology has demonstrated to be sufficiently reliable and effective for the implementation of an operational earthquake prediction system. To this aim, the combined use of different observations / parameters together with the refinement of data analysis methods, are expected to give major improvements in order to reduce false alarm rates and improve reliability and precision (in the space-time domain) of predictions.
In this frontier research field, EU and Russia are playing, since years, a worldwide scientific leading role and dispose of unique, satellite and ground based, Earth observation systems and capabilities.
The project PRE-EARTHQUAKES co-funded by the EU under the Seventh Framework Programme (FP7) was devoted to contribute to consolidate European and Russian excellence in observation and research in this important field.
Started in January 2011, it committed EU and Russian researchers to integrate different observational data (including the European Space Agency (ESA) and Roskosmos satellite data) and to improve, by cross-validating, their methodologies, in order:
- to substantially improve our knowledge of preparatory phases of earthquakes and their possible precursors;
- to promote a worldwide EQuOS as a dedicated component of GEOSS;
- to develop and offer to the international scientific community an integration platform where independent observations and new data analysis methodologies devoted to the research on/of earthquake precursors can be collected and cross-validated.
To these aims the project followed three mainstreams devoted:
- to coordinate and realise a systematic data acquisition and product generation in predefined (standardised) formats;
- to define and implement a common integration platform where heterogeneous data inputs can be ingested, organised and compared;
- to validate and disseminate data analysis and integration methods / tools making them available and open to the contribution of the worldwide scientific community in order to further extend the number of contemporary monitored parameters and to improve quality of data analysis methods.
To this purpose, observations from more than 20 different satellite systems - particularly those from the ESA, Russian agency Roskosmos and NASA - have been used to study the anomalous variations of those, surface and atmospheric parameters (up to the ionosphere), that have been for a long time proposed as possible precursors of strong earthquakes.
The project developed in two main phases:
- A first learning phase devoted to integrate independent observations on selected events occurred in the past in order to perform an a posteriori evaluation of the performance exhibited by the selected observational techniques (parameters, observation and data analysis methods) and, in case, to try to enlarge / improve them for the following project development.
- A second real-time integration and monitoring phase (PRIME), devoted to apply and evaluate, for a limited time period on selected test areas, the performance of a pre-operational system based on the best of the knowledge / capabilities developed in the first phase of the project.
The main selected test areas were the Sakhalin peninsula in eastern Russia, Turkey (where more than 80 radon measurement stations were operative) and Italy. During the second year of the project (for a period of five months), the study area was extended to include also Greece Kamchatka Peninsula and Japan.
The project that ended on December 2012 has been coordinated by the University of Basilicata (the project coordinator is Dr Ass. Prof. Valerio Tramutoli) and involved as partners:
- the Institute of Marine and Earth Sciences (EMSI) of the Turkey National Council of Scientific and Technological Research (Tubitak) - Turkey - Project team leader: Dr Ass. Prof. Sedat Inan (since January 2012 substituted by Dr Ass.Prof. Erhan Alparslan)
- the Deutsches Zentrum Fuer Luft - Und Raumfahrt EV (DLR) - Germany - Project team leader: Dr Norbert Jakowski
- Geospazio Italia srl - Italy - Project team leader: Dr Carolina Filizzola
- the Institute Fiodorov of Applied Geophysics - Russia - Project team leader: Dr Ass. Prof. Sergey Pulinets
- the Joint Stock Company 'Russian Space Systems' - Russia - Project team leader: Prof. Alexey Romanov
- the Pushkov Institute of Terrestrial Magnetism, Ionosphere and Radio Wave Propagation (WD Izmiran) of the Russian Sciences Academy of Kaliningrad: Project team leader: Dr Irk Shagimuratov.
In order to enlarge the project impact involving as much as possible the worldwide scientific community involved in PRE-EARTHQUAKES related studies, since 2012 a call for new PRE-EARTHQUAKES networking members received more than 20 request of participation. Eight of them were considered relevant to the project scope and three were formalised with the following institutions:
1. the Laboratory of Physics and Chemistry of the Environment and Space (LPC2E) of the French National Center for Scientific Research (CNRS). Project team leader: Dr Michel Parrot
2. the Chapman University, US. Project team leader: Dr Dimitar Ouzounov
3. National Observatory of Athens Institute of Geodynamics. Project team leader: Dr Gerassimos A. Papadopoulos.
Project results:
Even if a much longer experimentation time (and wider observational capabilities) should be required to reinforce results so far achieved, however some very clear indication and (quite unexpected) significant results were obtained:
- clear evidence of the importance of an integrated approach reducing false alarms rate from a 50 % (at the best) in the case of the best performing technique used alone, up to zero when an integrated approach is applied at the higher level of reliability;
- discovery of a further (strategic) added value of real time integration which can give back significance to observations which, taken alone or for impossibility of their repetition, were in the past just discarded;
- shocking evidence of the importance of the continuity of the observations: a first successful prediction for an earthquake in Turkey strongly reinforced the second one (presented in a scientific conference two weeks before the quake) presenting quite identical precursors signal in the same (restricted) area;
- any difficulty to guarantee continuity of the monitoring systems, including natural factors which can often 'blind' them (e.g. cloud presence in case of satellite observations, geomagnetic or solar storms for ionospheric measurements) strongly increases the probability to 'miss' events. This circumstance can discourage, without any scientifically funded reason, other scientists or potential end-user to further consider the potential of such monitoring systems. It is instead quite clear from the PRE-EARTHQUAKES experience that insufficiency or discontinuous coverage of the monitoring techniques make only a small fraction of occurring earthquakes (possibly) predictable independently from their principal characteristics (like magnitude, depth etc.);
- however, in five months of real-time monitoring PRE-EARTHQUAKES practically demonstrated that (at least) in some case, earthquakes can be predicted (provided that a suitable observing system is in place). Like in the case of missed events, this circumstance depends not only on earthquakes characteristics but (often much more) on the actual availability of usable observations;
- the two last points are of fundamental importance for the future of the research in this field as, together with the high rate of false positives, represent one of the main point of criticism from scientists asking (without any apparent justification) 100 % of efficiency to systems that, instead, can have evident societal utility even at lower level of performance. It seems instead that, if an improved quality of data analyses coupled with real-time integration could already strongly (up to zero) reduce false alarms rate only by the extension of the observational capabilities is possible to expect the possibility to increase the number (presently limited) of earthquakes that could be predicted in the future;
- a not minor results achieved by the direct collaboration with the PRE-EARTHQUAKES end-users is the possible use of PRE-EARTHQUAKES strategy within a pre-operational context. It seems quite evident that (and recent seismic crisis recently occurred in Italy demonstrated how much urgent is to quickly move in this direction) more than (or waiting for) a precise earthquake prediction system (presently not available) what decision makers actually need is a system which, integrating in real-time all relevant data and information, is able to guarantee a dynamic assessment of seismic risk (DASR) useful for adjusting the intensity of the prevention actions to be put in place by local authorities in presence of seismic crisis like the ones recently experimented in Italy in the case of l'Aquila, Emilia and Pollino earthquakes;
- a number of useful indications were moreover obtained regarding the actual potential of different observational technologies (individually considered or in combination) their costs and possibility of a realistic use in a pre-operational context.
From the technological point of view it should be mentioned the following:
- The product generation (PG) capability expressed by the consortium was impressive: at the end of the second year of activity PG = 6165 was the number describing the variety of measurements performed (i.e. corresponding to different parameters or technologies or algorithms or testing areas or period of observation).
- Similarly high was the products integration (PI) capability expressed by the consortium at the end of the project, PI = 2011 represents the number of different combination of independent observations which have been compared.
- The capacity of the consortium to mobilise innovation was also impressive both in terms of new methodologies (more than 50 % of the algorithms were significantly improved, several completely new proposed) and new data products (at the end of the project quite doubled compared with the ones initially planned).
- The adoption of international standards for data exchange, the definition of an original information protocol (PRE-EARTHQUAKES Policy on non disclosure of information) particular important in a so delicate field, together with the development of the pre-earthquakes geoportal (PEG) as a common and open access real-time integration and visualisation tool, represent an important heritage for facilitating a further growing of the consortium and its possible continuation after the nominal end of the project.
Potential impact:
PRE-EARTHQUAKES is a first attempt to join researchers of different countries to face rigorously and without prejudices the study of preparatory phases of earthquakes.
Continuity, and real-time integration of observations, before, during and after earthquakes, were the focus of the project.
The project exploited and reinforced the long and prestigious tradition of collaboration in seismology and geophysics among European and Russian scientists. It moreover enlarged the collaboration to US scientists looking to establish a wider, worldwide network (EQuOS) under the GEOSS umbrella.
The project contributed in this way to the cooperation process started within the EU-Russian space dialogues Earth observation working group, by consolidating and further developing scientific cooperation among European and Russian researchers involved in the study of earthquake preparatory phenomena on the basis of satellite and ground based Earth observation systems. The project have indeed profited of the unique opportunity, offered by ESA and Roskosmos, to have access for free to their satellite data archives in order to integrate, compare, improve different observational techniques.
The project was obviously not expected to solve the problem of earthquakes prediction and also in that case, it was clear that to build safe houses remains the best prevention action to put in place.
However some of its results seem could have significant scientific and societal impacts. Let we mention only the most important:
1. During the short period of the real-time implementation, successful predictions were performed (one during one of the most important conference in the field) with zero false alarms in case of violet alerts (the higher level in the PRE-EARTHQUAKES alert colour code). At the immediately lower level of alert (red alert) which corresponds also to a lower level of integration, the success rate was 1/3.
2. In numerous cases, it was evident (a posteriori), that the integration of all the available information (including for instance long-medium terms seismological forecasts) could significantly enlarge the number and reliability of predictions.
3. Finally, after the interaction with project end-users, it resulted quite evident that even prediction having lower level of reliability can be usefully exploited in an operational context. The new concept of dynamic assessment of seismic risk (DASR) was indeed successfully experimented during PRIME demonstrating as the real-time integration of all available observations / information could offer a significant improvement in the quality of decision that institutional bodies have to take during seismic crises like the ones recently experimented in Italy.
4. The achieved successes (even for a small fraction of earthquakes occurred during the PRIME period) together with the new possibilities offered by the use of such technologies within a DASR approach, will surely encourage not only the interdisciplinary research in the field but also a wider use of them with immediate advantages in the management of seismic crises by civil protection authorities.
A strong dissemination effort (well beyond the one initially planned) has been developed, especially during the second year of the project, by PRE-EARTH-QUAKES partners and networking members promoting the project in the most important international conferences (EGU, AGU, IGARSS, AOGS, EMSEV, ESC, etc.), publishing scientific papers and soliciting the collaboration of the worldwide scientific community. After only two years, the project is now well know all around the world in all the scientific communities differently involved in earthquakes prediction and/or forecast. PRE-EARTHQUAKES is recognised today as one of the best practices to be followed by national government moving (like recently the Italian one) in the same direction. The development of PEG designed for ingesting, visualising and comparing heterogeneous data products generated by the project was successfully released and implemented to support PRIME activities. A video on its use is presently available on the project website in order to encourage additional scientists to join to the observational network that even after the nominal end of the project will likely continue its effort under the EQuOS-GEOSS umbrella.
More in details the PRE-EARTHQUAKES dissemination numbers are:
- project website (see http://www.pre-earthquakes.org online) 43097 visits from 100 countries and 921 cities (statistics as of 1 February 2013);
- 3 workshops organised with 121 participants, 22 participating institutions, 3 thematic sub-groups and 1 education and training initiative hosted;
- the project was presented in 30 relevant scientific conferences, meetings and events;
- project press release produced 7 newspaper articles, 39 internet articles, brochures and contributions to several exhibitions relevant to the project and to the FP7 initiative;
- 7 TV interviews;
- 18 scientific papers presenting project results.
Great interest of the scientific community was raised in the project, in dissemination and project networking, in participation to EquOS, as testified in particular by:
a. eight meetings with representatives of connected projects (SEMEP, Twin-Sat project, Infrer Initiative, Demeter) for enlarging the PRE-EARTHQUAKES impact through an enlarged cooperation;
b. 21 expressions of interest in becoming networking members, following the establishment of the PRE-EARTHQUAKES networking membership procedure;
c. 2 published papers on international scientific journals quoting PRE-EARTHQUAKES;
d. the establishment of the EQuOS network promoting group (ENPG), set up by the SC on 18 November 2011;
e. the organisation of a EGU-Wien splinter meeting on 10 April 2013 devoted to: 'From PRE-EARTHQUAKES to EQUOS: Building the earthquakes observing system community as a dedicated component of GEOSS' (public).
Wide interest of the identified end-users in being involved in the project, providing information and feedbacks regarding the project activities as well as advice on how to outline potential operational solutions, as testified by the five letters of acceptance to be involved in the project received from relevant end users and the participation of 11 representatives of end-users in the workshop organised in Sakhalin (test area).
Project website: http://www.pre-earthquakes.org/index.php?lang=en
Dr Valerio Tramutoli
School of Engineering - University of Basilicata
Via dell'Ateneo Lucano, 10, 85100 - Potenza - Italy
Tel: +39-097-1205205
Fax: +39-097-1205215
Mobile: +39-329-3178385
These are the premises of PRE-EARTHQUAKES project which committed European Union (EU) and Russian researchers to integrate different observational data and technologies, to improve, by cross-validating, different data analysis methods, in order:
- to improve our knowledge of preparatory phases of earthquakes and of their possible precursors;
- to promote a worldwide earthquake observation system (EQuOS) as a dedicated component of the Global Earth Observation System of Systems (GEOSS);
- to develop and offer to the international scientific community an integration platform where independent observations and new data analysis methodologies devoted to the research on/of earthquake precursors can be collected and cross-validated.
To these aims the project followed three mainstreams devoted:
- to coordinate and realise a systematic data acquisition and product generation in predefined (standardised) formats;
- to define and implement a common integration platform where heterogeneous data inputs can be ingested, organised and compared;
- to validate and disseminate data analysis and integration methods / tools making them available and open to the contribution of the worldwide scientific community in order to further extend the number of contemporary monitored parameters and to improve quality of data analysis methods.
To this purposes, observations from more than 20 different satellite systems (particularly those from European, Russia and United States (US) space agencies) and about 100 ground based stations were systematically collected and analysed by using different data analysis methods, looking for anomalous variations of surface and atmospheric parameters (up to the ionosphere) possibly related to impending earthquakes.
The project developed in two main phases:
- A first learning phase devoted to integrate independent observations on selected events occurred in the past in order to perform an a posteriori evaluation of the performance exhibited by the selected observational techniques (parameters, observation and data analysis methods) and, in case, to try to enlarge / improve them for the following project development.
- A second real-time integration and monitoring phase (pre-earthquakes real-time integration and monitoring experiment (PRIME)), devoted to apply and evaluate, for a limited time period on selected test areas, the performance of a pre-operational system based on the best of the knowledge / capabilities developed in the first phase of the project.
The learning phase was performed on past events occurred in three main selected areas (Italy, Turkey and Sakhalin Peninsula). The PRIME duration was of five months and allowed to verify, in collaboration with selected institutional and scientific end-users, the possible contribute of PRE-EARTHQUAKES strategy, to the dynamic assessment of seismic risk (DASR) in a pre-operational context.
Even after a so short time of study (only two years for so challenging objectives) results achieved have far exceeded all expectations and from many points of view.
Data product generation and integration:
- The product generation (PG) capability expressed by the consortium was impressive: at the end of the second year of activity PG = 6165 is the number describing the variety of measurements performed (i.e. corresponding to different parameters or technologies or algorithms or testing areas or period of observation).
- Similarly impressive the products integration (PI) capability expressed by the consortium at the end of the second year, PI = 2011 represent the number of the different combination of independent observations which have been compared.
- At the end of the project, the capacity of the consortium to mobilise innovation was quite impressive in terms of new methodologies (more than 50 % of the algorithms were significantly improved, several completely new proposed) and new data products (at the end of the project quite doubled compared with the ones initially planned).
- The adoption of international standards for data exchange, the definition of an original information protocol (PRE-EARTHQUAKES policy on non disclosure of information ) particular important in a so delicate field, together with the development of the pre-earthquakes geoportal (PEG) as a common and open access real-time integration and visualisation tool, represented the key choices for facilitate the growing of the consortium and its possible continuation after the nominal end of the project.
Scientific results
Even if based on a too short experimentation time and limited observational capabilities, results so far achieved appear quite interesting under different points of view. We will mention here only the most significant.
- A clear evidence was reached of the importance of an integrated observational approach in reducing (up to zero) false alarms rates.
- Shocking evidence of the importance of the continuity of the observations which permitted to repeat, with an highest confidence, a second successful prediction in the same seismic area.
- Any difficulty to guarantee continuity of the monitoring systems, including natural factors which can often 'blind' them (e.g. cloud presence in case of satellite observations, geomagnetic or solar storms for ionospheric measurements) strongly increases the probability to 'miss' events. This circumstance can discourage, without any scientifically funded reason, other scientists or potential end-user to further consider the potential of such monitoring systems. It is instead quite clear from the PRE-EARTHQUAKES experience that insufficiency or discontinuous coverage of the monitoring techniques make only a small fraction of occurring earthquakes (possibly) predictable independently from their principal characteristics (like magnitude, depth etc.).
- However, in five months of real-time monitoring PRE-EARTHQUAKES practically demonstrated that (at least) in some case, earthquakes can be predicted (provided that a suitable observing system is in place). Like in the case of missed events, this circumstance depends not only on earthquakes characteristics but (often much more) on the actual availability of usable observations.
- The two last points are of fundamental importance for the future of the research in this field as, together with the high rate of false positives, represent one of the main point of criticism from scientists asking (without any apparent justification) 100 % of efficiency to systems that, instead, can have evident societal utility even at lower level of performance.
- A not minor results achieved by the direct collaboration with the PRE-EARTHQUAKES end-users is the possible use of PRE-EARTHQUAKES strategy within a pre-operational context. It seems quite evident that (and recent seismic crisis recently occurred in Italy demonstrated how much urgent is to quickly move in this direction) more than (or waiting for) a precise earthquake prediction system (presently not available) what decision makers actually need is a system which, integrating in real-time all relevant data and information, is able to guarantee a dynamic assessment of seismic risk (DASR) useful for adjusting the intensity of the prevention actions to be put in place by local authorities in presence of seismic crisis like the ones recently experimented in Italy in the case of l'Aquila, Emilia and Pollino earthquakes.
Enlarging the community looking for continuity
One of the main successes of PRE-EARTHQUAKES was its communication strategy and the capability to enlarge the boundaries of its initial (very limited for reason of budget) partnership. To define the conditions for allowing other institutions to participate to project efforts increasing its results and overall impacts was a not small effort. It required the definition of specific communication protocols and rule of enrolments ensuring the appropriate use of the information generated within the consortium which are potentially critical for their possible societal impact. Thanks to a dissemination effort well beyond the one initially planned and significantly supported by partners' internal resources, in the first week after the first call for enrolling PRE-EARTHQUAKES networking members, more than 20 request of enrolment were submitted and only the scarce capability to guarantee systematic observations (which are the fundament of PRE-EARTHQUAKES growing strategy) expressed by most of candidates, prevented the consortium to enrol more than four of them as networking members until now. The good news is that PRE-EARTHQUAKES efforts will continue (under the EQuOS-GEOSS umbrella and, hopefully, with a COST financial support for which a proposal already passed the first selection). Anyway the capital of expertises, integration tools, communication standards and data product generation automatic chains, is still there and do not want to be dispersed. Most of the partners and networking members are continuing to systematically generate their data products, PEG integration tool and website maintenance will be guaranteed by UNIBAS internal resources until the completion of the above mentioned passages to a different internal organisation. After only two year of (short but intense) activity PRE-EARTHQUAKES has been recognised as one of the worldwide best-practices of research in the field of short-term earthquakes prediction, to which several national projects (like the ones recently promoted by the Italian government or re-proposed in Taiwan) are presently looking with particular attention.
Project context and objectives:
A large scientific documentation is to-date available about the appearance of anomalous space-time patterns of geophysical parameters measured from days to week before earthquakes occurrence. Nevertheless, up to now no one measurable parameter, no one observational methodology has demonstrated to be sufficiently reliable and effective for the implementation of an operational earthquake prediction system. To this aim, the combined use of different observations / parameters together with the refinement of data analysis methods, are expected to give major improvements in order to reduce false alarm rates and improve reliability and precision (in the space-time domain) of predictions.
In this frontier research field, EU and Russia are playing, since years, a worldwide scientific leading role and dispose of unique, satellite and ground based, Earth observation systems and capabilities.
The project PRE-EARTHQUAKES co-funded by the EU under the Seventh Framework Programme (FP7) was devoted to contribute to consolidate European and Russian excellence in observation and research in this important field.
Started in January 2011, it committed EU and Russian researchers to integrate different observational data (including the European Space Agency (ESA) and Roskosmos satellite data) and to improve, by cross-validating, their methodologies, in order:
- to substantially improve our knowledge of preparatory phases of earthquakes and their possible precursors;
- to promote a worldwide EQuOS as a dedicated component of GEOSS;
- to develop and offer to the international scientific community an integration platform where independent observations and new data analysis methodologies devoted to the research on/of earthquake precursors can be collected and cross-validated.
To these aims the project followed three mainstreams devoted:
- to coordinate and realise a systematic data acquisition and product generation in predefined (standardised) formats;
- to define and implement a common integration platform where heterogeneous data inputs can be ingested, organised and compared;
- to validate and disseminate data analysis and integration methods / tools making them available and open to the contribution of the worldwide scientific community in order to further extend the number of contemporary monitored parameters and to improve quality of data analysis methods.
To this purpose, observations from more than 20 different satellite systems - particularly those from the ESA, Russian agency Roskosmos and NASA - have been used to study the anomalous variations of those, surface and atmospheric parameters (up to the ionosphere), that have been for a long time proposed as possible precursors of strong earthquakes.
The project developed in two main phases:
- A first learning phase devoted to integrate independent observations on selected events occurred in the past in order to perform an a posteriori evaluation of the performance exhibited by the selected observational techniques (parameters, observation and data analysis methods) and, in case, to try to enlarge / improve them for the following project development.
- A second real-time integration and monitoring phase (PRIME), devoted to apply and evaluate, for a limited time period on selected test areas, the performance of a pre-operational system based on the best of the knowledge / capabilities developed in the first phase of the project.
The main selected test areas were the Sakhalin peninsula in eastern Russia, Turkey (where more than 80 radon measurement stations were operative) and Italy. During the second year of the project (for a period of five months), the study area was extended to include also Greece Kamchatka Peninsula and Japan.
The project that ended on December 2012 has been coordinated by the University of Basilicata (the project coordinator is Dr Ass. Prof. Valerio Tramutoli) and involved as partners:
- the Institute of Marine and Earth Sciences (EMSI) of the Turkey National Council of Scientific and Technological Research (Tubitak) - Turkey - Project team leader: Dr Ass. Prof. Sedat Inan (since January 2012 substituted by Dr Ass.Prof. Erhan Alparslan)
- the Deutsches Zentrum Fuer Luft - Und Raumfahrt EV (DLR) - Germany - Project team leader: Dr Norbert Jakowski
- Geospazio Italia srl - Italy - Project team leader: Dr Carolina Filizzola
- the Institute Fiodorov of Applied Geophysics - Russia - Project team leader: Dr Ass. Prof. Sergey Pulinets
- the Joint Stock Company 'Russian Space Systems' - Russia - Project team leader: Prof. Alexey Romanov
- the Pushkov Institute of Terrestrial Magnetism, Ionosphere and Radio Wave Propagation (WD Izmiran) of the Russian Sciences Academy of Kaliningrad: Project team leader: Dr Irk Shagimuratov.
In order to enlarge the project impact involving as much as possible the worldwide scientific community involved in PRE-EARTHQUAKES related studies, since 2012 a call for new PRE-EARTHQUAKES networking members received more than 20 request of participation. Eight of them were considered relevant to the project scope and three were formalised with the following institutions:
1. the Laboratory of Physics and Chemistry of the Environment and Space (LPC2E) of the French National Center for Scientific Research (CNRS). Project team leader: Dr Michel Parrot
2. the Chapman University, US. Project team leader: Dr Dimitar Ouzounov
3. National Observatory of Athens Institute of Geodynamics. Project team leader: Dr Gerassimos A. Papadopoulos.
Project results:
Even if a much longer experimentation time (and wider observational capabilities) should be required to reinforce results so far achieved, however some very clear indication and (quite unexpected) significant results were obtained:
- clear evidence of the importance of an integrated approach reducing false alarms rate from a 50 % (at the best) in the case of the best performing technique used alone, up to zero when an integrated approach is applied at the higher level of reliability;
- discovery of a further (strategic) added value of real time integration which can give back significance to observations which, taken alone or for impossibility of their repetition, were in the past just discarded;
- shocking evidence of the importance of the continuity of the observations: a first successful prediction for an earthquake in Turkey strongly reinforced the second one (presented in a scientific conference two weeks before the quake) presenting quite identical precursors signal in the same (restricted) area;
- any difficulty to guarantee continuity of the monitoring systems, including natural factors which can often 'blind' them (e.g. cloud presence in case of satellite observations, geomagnetic or solar storms for ionospheric measurements) strongly increases the probability to 'miss' events. This circumstance can discourage, without any scientifically funded reason, other scientists or potential end-user to further consider the potential of such monitoring systems. It is instead quite clear from the PRE-EARTHQUAKES experience that insufficiency or discontinuous coverage of the monitoring techniques make only a small fraction of occurring earthquakes (possibly) predictable independently from their principal characteristics (like magnitude, depth etc.);
- however, in five months of real-time monitoring PRE-EARTHQUAKES practically demonstrated that (at least) in some case, earthquakes can be predicted (provided that a suitable observing system is in place). Like in the case of missed events, this circumstance depends not only on earthquakes characteristics but (often much more) on the actual availability of usable observations;
- the two last points are of fundamental importance for the future of the research in this field as, together with the high rate of false positives, represent one of the main point of criticism from scientists asking (without any apparent justification) 100 % of efficiency to systems that, instead, can have evident societal utility even at lower level of performance. It seems instead that, if an improved quality of data analyses coupled with real-time integration could already strongly (up to zero) reduce false alarms rate only by the extension of the observational capabilities is possible to expect the possibility to increase the number (presently limited) of earthquakes that could be predicted in the future;
- a not minor results achieved by the direct collaboration with the PRE-EARTHQUAKES end-users is the possible use of PRE-EARTHQUAKES strategy within a pre-operational context. It seems quite evident that (and recent seismic crisis recently occurred in Italy demonstrated how much urgent is to quickly move in this direction) more than (or waiting for) a precise earthquake prediction system (presently not available) what decision makers actually need is a system which, integrating in real-time all relevant data and information, is able to guarantee a dynamic assessment of seismic risk (DASR) useful for adjusting the intensity of the prevention actions to be put in place by local authorities in presence of seismic crisis like the ones recently experimented in Italy in the case of l'Aquila, Emilia and Pollino earthquakes;
- a number of useful indications were moreover obtained regarding the actual potential of different observational technologies (individually considered or in combination) their costs and possibility of a realistic use in a pre-operational context.
From the technological point of view it should be mentioned the following:
- The product generation (PG) capability expressed by the consortium was impressive: at the end of the second year of activity PG = 6165 was the number describing the variety of measurements performed (i.e. corresponding to different parameters or technologies or algorithms or testing areas or period of observation).
- Similarly high was the products integration (PI) capability expressed by the consortium at the end of the project, PI = 2011 represents the number of different combination of independent observations which have been compared.
- The capacity of the consortium to mobilise innovation was also impressive both in terms of new methodologies (more than 50 % of the algorithms were significantly improved, several completely new proposed) and new data products (at the end of the project quite doubled compared with the ones initially planned).
- The adoption of international standards for data exchange, the definition of an original information protocol (PRE-EARTHQUAKES Policy on non disclosure of information) particular important in a so delicate field, together with the development of the pre-earthquakes geoportal (PEG) as a common and open access real-time integration and visualisation tool, represent an important heritage for facilitating a further growing of the consortium and its possible continuation after the nominal end of the project.
Potential impact:
PRE-EARTHQUAKES is a first attempt to join researchers of different countries to face rigorously and without prejudices the study of preparatory phases of earthquakes.
Continuity, and real-time integration of observations, before, during and after earthquakes, were the focus of the project.
The project exploited and reinforced the long and prestigious tradition of collaboration in seismology and geophysics among European and Russian scientists. It moreover enlarged the collaboration to US scientists looking to establish a wider, worldwide network (EQuOS) under the GEOSS umbrella.
The project contributed in this way to the cooperation process started within the EU-Russian space dialogues Earth observation working group, by consolidating and further developing scientific cooperation among European and Russian researchers involved in the study of earthquake preparatory phenomena on the basis of satellite and ground based Earth observation systems. The project have indeed profited of the unique opportunity, offered by ESA and Roskosmos, to have access for free to their satellite data archives in order to integrate, compare, improve different observational techniques.
The project was obviously not expected to solve the problem of earthquakes prediction and also in that case, it was clear that to build safe houses remains the best prevention action to put in place.
However some of its results seem could have significant scientific and societal impacts. Let we mention only the most important:
1. During the short period of the real-time implementation, successful predictions were performed (one during one of the most important conference in the field) with zero false alarms in case of violet alerts (the higher level in the PRE-EARTHQUAKES alert colour code). At the immediately lower level of alert (red alert) which corresponds also to a lower level of integration, the success rate was 1/3.
2. In numerous cases, it was evident (a posteriori), that the integration of all the available information (including for instance long-medium terms seismological forecasts) could significantly enlarge the number and reliability of predictions.
3. Finally, after the interaction with project end-users, it resulted quite evident that even prediction having lower level of reliability can be usefully exploited in an operational context. The new concept of dynamic assessment of seismic risk (DASR) was indeed successfully experimented during PRIME demonstrating as the real-time integration of all available observations / information could offer a significant improvement in the quality of decision that institutional bodies have to take during seismic crises like the ones recently experimented in Italy.
4. The achieved successes (even for a small fraction of earthquakes occurred during the PRIME period) together with the new possibilities offered by the use of such technologies within a DASR approach, will surely encourage not only the interdisciplinary research in the field but also a wider use of them with immediate advantages in the management of seismic crises by civil protection authorities.
A strong dissemination effort (well beyond the one initially planned) has been developed, especially during the second year of the project, by PRE-EARTH-QUAKES partners and networking members promoting the project in the most important international conferences (EGU, AGU, IGARSS, AOGS, EMSEV, ESC, etc.), publishing scientific papers and soliciting the collaboration of the worldwide scientific community. After only two years, the project is now well know all around the world in all the scientific communities differently involved in earthquakes prediction and/or forecast. PRE-EARTHQUAKES is recognised today as one of the best practices to be followed by national government moving (like recently the Italian one) in the same direction. The development of PEG designed for ingesting, visualising and comparing heterogeneous data products generated by the project was successfully released and implemented to support PRIME activities. A video on its use is presently available on the project website in order to encourage additional scientists to join to the observational network that even after the nominal end of the project will likely continue its effort under the EQuOS-GEOSS umbrella.
More in details the PRE-EARTHQUAKES dissemination numbers are:
- project website (see http://www.pre-earthquakes.org online) 43097 visits from 100 countries and 921 cities (statistics as of 1 February 2013);
- 3 workshops organised with 121 participants, 22 participating institutions, 3 thematic sub-groups and 1 education and training initiative hosted;
- the project was presented in 30 relevant scientific conferences, meetings and events;
- project press release produced 7 newspaper articles, 39 internet articles, brochures and contributions to several exhibitions relevant to the project and to the FP7 initiative;
- 7 TV interviews;
- 18 scientific papers presenting project results.
Great interest of the scientific community was raised in the project, in dissemination and project networking, in participation to EquOS, as testified in particular by:
a. eight meetings with representatives of connected projects (SEMEP, Twin-Sat project, Infrer Initiative, Demeter) for enlarging the PRE-EARTHQUAKES impact through an enlarged cooperation;
b. 21 expressions of interest in becoming networking members, following the establishment of the PRE-EARTHQUAKES networking membership procedure;
c. 2 published papers on international scientific journals quoting PRE-EARTHQUAKES;
d. the establishment of the EQuOS network promoting group (ENPG), set up by the SC on 18 November 2011;
e. the organisation of a EGU-Wien splinter meeting on 10 April 2013 devoted to: 'From PRE-EARTHQUAKES to EQUOS: Building the earthquakes observing system community as a dedicated component of GEOSS' (public).
Wide interest of the identified end-users in being involved in the project, providing information and feedbacks regarding the project activities as well as advice on how to outline potential operational solutions, as testified by the five letters of acceptance to be involved in the project received from relevant end users and the participation of 11 representatives of end-users in the workshop organised in Sakhalin (test area).
Project website: http://www.pre-earthquakes.org/index.php?lang=en
Dr Valerio Tramutoli
School of Engineering - University of Basilicata
Via dell'Ateneo Lucano, 10, 85100 - Potenza - Italy
Tel: +39-097-1205205
Fax: +39-097-1205215
Mobile: +39-329-3178385