Estimation of Seismic Vulnerability of Buildings
Near Real-Time Estimation of Seismic Vulnerability of Buildings with Images Application
Tech Area / Field
- ENV-SEM/Seismic Monitoring/Environment
- ENV-MIN/Monitoring and Instrumentation/Environment
- INF-IMA/High-Definition Imaging and Displays/Information and Communications
- INF-SOF/Software/Information and Communications
- SAT-OTH/Other/Space, Aircraft and Surface Transportation
3 Approved without Funding
MISIS (Steel and Alloys), Russia, Moscow
- Institute of Environmental Geoscience / Seismological Center\nVNIITF, Russia, Chelyabinsk reg., Snezhinsk
- Karlsruhe University / Geophysikalisches Institut / Collaborative Research Center on Strong Earthquakes, Germany, Karlsruhe\nUniversity of East Anglia, UK, Norwich\nNational Institute of Building Sciences, USA, DC, Washington\nEcole et Observatoire des Sciences de la Terre, France, Strasbourg\nBRGM (French Geological Survey), France, ORLEANS CEDEX 2
Краткое описание проектаThe lessons learned from strong earthquakes occurred recently in Japan, Russia, Indonesia, Taiwan, Greece, Turkey, Afghanistan, and other countries give evidence that the losses due to these events are tremendous, in spite of preventive measures taken in different countries. According to the Earthquake Situation Report no. 23 issued on October 29, 1999 by UN DHA, the official death toll from the 17 August, 1999 earthquake in Turkey is 17,127 and the number of reported injured stands at 43,953; unofficial estimates place the death toll from this event between 30,000 and 40,000. The 1995 Kobe earthquake killed more than 6,000 people and injured over 40,000; deaths due to collapse of buildings had been almost negligible in number in Japan during the last 40 years before this event and the number of casualties proved to be quite unexpected for local authorities, as well as for scientists and engineers. These events show that even earthquake-resistant construction in highly developed countries cannot protect population against strong events. Analysis of statistics due to these events also gave evidence that, within 24 hours after events, the number of fatalities under building debris may reach 48% of total number of those trapped under debris unless the rescue operations are started in time at once.
Unacceptable increase in losses due to strong earthquakes all over the world results in changing the natural disaster management strategy in many countries. At present, together with earthquake-resistant design and construction, earthquake insurance, as well as other preventive measures which are implemented in order to make the society safer, great attention is paid to operative or near real time systems for damage and loss assessment due to strong events, which allow to take effective decisions about rescue and other urgent operations. Such systems of operative assessment of earthquake impact have been developed and are developing in many countries in order to help make decision about effective response measures at different levels: local, regional, national. Rapid estimate of potential damage due to strong earthquakes is considered by the Council of Europe's Eur-OPA Major Hazards Agreement as a crucial question, the answer for which must be given in “emergency” mode, in order to allow persons in charge to make the proper decisions, as adequate as possible.
Within the framework of the Council of Europe's Programmes EDRIM (“Electronic Discussion group for Risk Management”) and STRIM (“Space Technology Applications for Risk Management”), the testing of different systems for damage and loss estimation due to strong earthquakes, developed by prominent scientists and engineers, was undertaken.
The systems’ efficiency is mainly determined by the reliability of estimates obtained by applying them, cost and time needed for system development and updating, as well as level of confidence of results of computations.
Reliability of estimations in its turn depends on:
– databases on elements of risk (population and built environment) and hazard sources;
– vulnerability functions for different elements at risk due to strong earthquakes;
– discrepancies (errors) in strong earthquakes’ parameters determination by different seismological alert services;
– conformity of computer models to real events.
Cost is determined by expenses spent for creation and actualization of data and knowledge bases.
Time is very critical for such systems, especially time spent for getting information about:
– strong earthquakes’ parameters, issued by different seismological alert services;
– data on changes in population distribution and quality of existing building stock;
as well as on time spent for:
– expected damage and loss computations;
– collecting expert opinions;
– passing out the information about potential event impact to decision makers.
We expect to increase the efficiency of near real time systems in operation by putting efforts on the verification of building stock characteristics existing at the moment of a disaster. In “emergency mode”, such verification may be carried out by using only existing remote sensing data on the disaster-affected area. Therefore, the development of operative procedures for high-resolution space images processing is very vital for estimation or verification of buildings’ vulnerability to strong earthquake impact.
Scope of Proposal.
We propose to develop a set of operative (automatic) procedures for remote sensing material processing. These procedures will allow the characteristics of existing building stock to be verified, and seismic vulnerability classes of buildings and structures to be determined in “emergency mode” for any region of the world.
Image processing aimed at seismic vulnerability estimation includes detection, identification and determination of a set of specific characteristics of buildings from their image, namely size, material, structural peculiarities.
Detection means searching the places in an image where the objects under study are located with high probability.
Identification means evaluation of the “essence” of depicted and detected buildings and structures in the image.
Determination of characteristics of identified objects is the third stage of processing. Within this stage the following procedures are realized:
– analysis and averaging of object qualitative and quantitative parameters;
– evaluation of vulnerability class to which the object under study is likely to belong.
At present, planned (apparatus' optical axis is perpendicular to the Earth's surface at the moment when the picture is taken) and inclined (apparatus' optics “sees” the Earth under an angle equal to 10-80 degrees) images are used to study the characteristics of buildings and structures.
In addition, surveying may be done with and without images overlapping. It depends on the conditions and equipment used.
Buildings may be recognized on an image by their regular geometrical shapes. Buildings on an image may be depicted as compact (point) or areal (area) objects (primitives) depending on image resolution.
Building characteristics, for instance its height, may be identified by such indirect feature as shadow size.
There are three levels of details while estimating the buildings and structures vulnerability classes. Images with different resolution, corresponding to the different levels of details, are used.
The aim of the present Project is:
– to develop an automatic method of processing high-resolution (1, …4 m) images obtained with KLA application, in order to verify buildings vulnerability to strong earthquakes;
– to carry out a comparative analysis of the results of this procedure and land truth on the example of cities located in earthquake prone areas;
– to estimate the reliability of the procedures developed.
Tasks (objectives) of the present Project are the following:
– acquisition of material for carrying out scientific research and experimental study;
– analysis of information about earthquake resistance of buildings and structures which is available on images of the area under study, as well as possible approaches to estimation the photographic quality of remote sensing material, possible approach for evaluation of remote sensing material resolution, possible approach for estimating measurement properties of photographic images, possible approach for estimating the scale of images;
– specification of requirements to images which should provide reliable estimation of seismic vulnerability, including characterization features for buildings' earthquake-resistance assessment, requirements for resolution and scale of images, choice of optimal inclination of surveying system's optical axis;
– description of buildings vulnerability classes according to identified features;
– development of formalized procedures for characterization features selection;
– experimental processing of images in order to compile seismic vulnerability maps for the affected areas;
– comparison between results obtained by image interpretation and ground inventory;
– estimation of the reliability and efficiency of space image application for loss assessment due to earthquakes in urban areas and planning of preventive measures for the cases of emergency.
Expected output (effect).
Implementation of the procedures developed will provide more reliable operative estimations of expected losses due to strong earthquakes and will increase the efficiency of response measures. In earthquake prone areas characterized by high density of population, it is probably possible to increase the accuracy of computations of potential earthquake impact up to 50 %.
The project includes following stages:
1. Preparatory work.
2. Review of methods for vulnerability classes assessment.
3. Theoretical study in order to develop the requirements for images.
4. Theoretical study: development of methods for image processing in order to estimate buildings and structures vulnerability.
5. Formal description of vulnerability classes for chosen test areas.
6. Experimental researches.
7. Analysis of experimental researches results.
8. Study of the influence of error in building vulnerability determination through image analysis on the reliability of forecast of potential impact of strong earthquakes.
9. Development of recommendations on vulnerability estimation from remote sensing material exploitation.
10. Development of proposals for the creation of archives of digital images for cities located in earthquake prone areas all over the world.
The main expected results are following:
– Review of methods for buildings vulnerability class estimation.
– Knowledge bases on strong earthquakes containing detail descriptions of events, their social, engineering and economic consequences.
– Collection of remote sensing data characterized by different resolutions, various types of equipment used for surveying, different conditions for surveying chosen test areas.
– Requirements on resolution and scale of images, which should provide reliable estimation of seismic vulnerability of buildings and structures.
– Description of main features of buildings depicted on space images of the area under study, which are used for estimation of buildings earthquake vulnerability.
– Description of building vulnerability classes according to characterization features.
– Formal procedures of processing features selection.
– Report, which will include description of algorithms and operative procedure of remote sensing data processing.
– Draft software for operative estimation of buildings types and vulnerability classes.
– Analysis of results of evaluation of image processing reliability, recommendations on estimation of vulnerability from remote sensing data analysis, and proposals on creation of archives of digital space images for cities located in earthquake prone areas all over the world.