Skip to main content

Performance-based approach to the earthquake protection of cultural heritage in European and Mediterranean countries

Final Report Summary - PERPETUATE (Performance-based approach to the earthquake protection of cultural heritage in European and Mediterranean countries)

Executive summary:

PERPETUATE project has intended to develop a methodology for the assessment of seismic risk to cultural heritage assets and design of interventions. The main goal was to develop European Guidelines for evaluation and mitigation of seismic risk to cultural heritage assets, applicable in the European and other Mediterranean countries.

The first driving idea of the project is that the protection of cultural heritage needs an improvement in methods of analysis and assessment procedures rather than in intervention techniques. The second driving idea is that a verification approach in terms of displacement capacity rather than in term of strength is more reliable and effective for heritage buildings and that flexibility, ductility and lightness are positive factors for their seismic behaviour. The third idea is that a reliable assessment procedure of heritage buildings requires the assessment of both architectonic and artistic assets contained in them.

Project Context and Objectives:
The damage assessment to cultural heritage assets after recent earthquakes, in particular in Italy (Umbria and Marche, 1997; L'Aquila 2009 and Emilia 2012), showed the high vulnerability of some types of historical structures (churches, towers). Earthquakes also proved that strengthening interventions adopted in the last decades are invasive, not effective and even increase the vulnerability. Thus, there is an urgent need for disseminating these discoveries and promoting a new and really effective strategy for the risk mitigation of cultural heritage.

The preservation of cultural heritage assets must guarantee their capacity of lasting over time against decay, natural hazards and extreme events, without losing their authenticity and usability. This means that the need of guaranteeing an "acceptable level" of structural safety for building's occupants should be always related to the principle of "minimum intervention" on the building itself. The definition of "acceptable" safety levels, as well as the concept of "safety", still represents an open issue for monumental buildings.

Furthermore, it should be considered that the intangible value of these buildings depends, besides their social and historical meaning, on both architectural and artistic factors. Thus, their risk assessment is a challenge regarding not only structural and architectural components, but also movable (paintings, statues, libraries…) and unmovable (frescos, stucco-works, pinnacles, battlements, banisters, balconies) artistic assets contained in it.

PERPETUATE project has intended to develop a methodology for the assessment of seismic risk to cultural heritage assets and design of interventions, based on the following principia:
- The protection of cultural heritage needs an improvement in methods of analysis and assessment procedures rather than improvement of intervention techniques. A reliable assessment procedure is the main tool to respect the principle of "minimum intervention" under the constraint of structural safety.
- If, by reliable methods of analysis and assessment procedures, it is demonstrated that the monument is not safe and that its retrofitting is unavoidable, an effective improvement of the intervention techniques does not necessarily need new materials and technologies. In many cases, traditional techniques, improved by the results of new methods of analysis, represent a sustainable, effective, cost-efficient and reversible solution.
- The displacement-based approach for vulnerability assessment of cultural heritage assets and design of interventions is adopted as standard method of analysis. The mechanical models available for the analysis of ancient masonry buildings or elements usually consider a verification approach in terms of strength, because in the past strengthening techniques were based on the concept of increasing stiffness and strength. This approach is correct in static conditions. However, earthquake is a dynamic phenomenon that induces deformations and dynamic amplification. The seismic response of the structure depends on its dynamic properties. Usually much more the construction is stiffer much higher are the equivalent static actions, which is subjected to: thus, flexibility is a first positive characteristic for a good seismic behaviour. Moreover, since inertial actions are proportional to the weight of the construction, the lightness is a second fundamental aspect. Finally, in the case of rare destructive earthquakes, it is impossible to bear the seismic actions without significant damage. Hence, the building must be ductile enough and be able to sustain vertical loads without collapse, even if large horizontal displacements are induced by the earthquake.
- Despite the complex nature of heritage buildings and artistic assets, the DBA (Displacement Based Assessment) calls for nonlinear models. Nonlinear static (pushover) analyses are considered as the main tool for the application of the assessment procedure. Nonlinear dynamic analyses are considered as an alternative tool only for certain types of assets.
- A reliable assessment procedure of heritage buildings requires that both architectonic and artistic assets contained in them are considered.

Definition of performance limit states
It is a shared opinion that it is not possible to define strict safety levels for cultural assets, since the "case by case" approach is always preferable. The definition of "acceptable" safety levels may be intended as a compromise: on the one side, they must prevent the damage deriving from environmental actions (earthquake, flood, aging); on the other side, they must prevent from invasive interventions, designed in order to avoid future damage from environmental actions but unwillingly producing immediate and significant loss in terms of conservation. To conciliate these issues is of fundamental relevance. "Acceptable" safety levels for cultural heritage assets aimed to be based on the concept of performance limit states.

For ordinary buildings, usually one refer to two fundamental limit states:
1) the serviceability limit state, corresponding to the occurrence of reparable damages limiting temporary the use;
2) the ultimate limit state, usually corresponding to near collapse conditions.

Evaluation of seismic hazard and soil-foundation interaction.
The evaluation of the seismic hazard plays a fundamental role in the seismic risk assessment of heritage buildings, since they are sensitive to: particular characteristics of the seismic input (earthquake duration, frequency content in the long period range, vertical component of the motion); soil amplification effects; soil-structure interaction phenomena (in the case of big, heavy and stiff masonry structures). PERPETUATE, after a detailed probabilistic analysis of available intensity measures, has intended to obtain proper Acceleration-Displacement Response Spectra (ADRS) and proper time-histories for cultural heritage assets (see WP2). Furthermore (see WP3), the project planned to develop an improved foundation model for the vulnerability assessment of massive masonry structures, taking into account the role of the foundation compliance, associated soil-foundation interaction (SFI) and soil-foundation-structure interaction (SFSI) effects. The role of the foundation and SFSI effects was supposed to be studied both for seismic ground shaking and induced permanent ground deformations. Specific experimental campaigns and on-situ surveys were planned to support the achievement of these objectives.

Construction knowledge.
The minimization of interventions on cultural heritage assets requires a comprehensive knowledge on the assets themselves (see WP4). In order to guarantee a shared and uniform approach at a European level, the project planned to define both standards concerning the survey of the asset (structural and technological survey, material characterization …) on the basis of traditional and innovative techniques and proper criteria to optimize the investigation plan (in order to minimize its invasiveness and impact on building). Moreover laboratory and on-situ tests have been made in order to define reference values of the masonry properties to adopt in the seismic verifications. Useful values have been proposed.

Structural models for the seismic analysis of masonry structures and artistic assets and the design of interventions.

In order to perform displacement-based verifications, models capable to describe the nonlinear behaviour of the assets under the external actions, up to their collapse, are required (see WP5). After a classification of different types of architectonic and artistic assets and a review of the tools available in the literature, PERPETUATE aimed to define a logical path to select the most reliable, accurate and computationally sustainable model and define proper seismic safety verification procedures for each class of asset. Moreover, it has intended to improve some of the mechanical models already available to evaluate the displacement capacity of buildings as well as single architectonic elements or artistic assets.

Application to case studies.
Finally, the project intended to apply the proposed assessment methodology to case studies for validation purposes (see WP7). In particular, the following cases studies were selected: Ardinghelli palace and Santa Maria Paganica church hit by the earthquake on 6th April 2009; St. Pardo Cathedral in Larino (Molise Region, Italy), hit by the earthquake on 2002; a Neoclassical School, the Arsenal de Milly and the Hassan Bey Mansion in the historic centre of Rhodes, Greece; the Great Mosque and the Casbah of Algiers; Kolizej Palace in Ljubljana; Slovenian examples of rural settlements from Posocje region and Old City Centre of Ljubljana.

Project Results:

Main scientific and technological results

3.1 The proposed assessment procedure
3.1.1 Single assets
The seismic assessment procedure for single assets is based on three main steps. In the first one, the building is known and the seismic input is defined in accordance with specific safety and conservation requirements. In the second one, the seismic response of the asset is described through mechanical model and its capability to satisfy the defined requirements is assessed. Finally, the results are discussed and rehabilitation decisions are taken.

In PERPETUATE procedure, the description of the seismic response of the asset is always performed taking into account the non-linear behaviour of the structure. In general, non-linear static (pushover) analyses are considered. The use of dynamic analyses is suggested for particular types of assets.

In the following, each step of the procedure is analysed in detail.

3.1.1.1 Classification of the asset and preliminary modelling choices
After a preliminary phase in which the asset is studied, its historical and architectonical relevance is analysed and its use is identified, the monument is classified according to PERPETUATE classes (D4). The classification is based on the prevailing seismic damage modes of assets (identified at the scale of "macroelements" in D4 and D5) and on the assumption that their occurrence is closely related to building morphology (architectural form, proportions) and technology (type of masonry, nature of horizontal diaphragms, effectiveness of wall-to-wall and floor-to-walls connections). It should be not intended in a strict way because of complexity of buildings and variety of related macroelements. It is clear that each class and related sub-class collect a wide variety of buildings and a single building may belong to different classes depending on its specific features (e.g. even if two buildings have the same function they may differ for the morphology or for the constructive and technological details/solutions). In this sense, as an example, it is important pointing out it is the prevailing seismic behaviour and not the use that plays the fundamental role in the assignment of a certain class.

The assignment to a class is functional to drive modelling choices and identify proper limit states on the capacity curve.

3.1. The type of models proposed have been analysed in D7, while in D26 the specific problems of each approach in relation to different building classes are discussed.
The analysis moves then to immovable artistic assets contained in the building. Once identified and once assessed their cultural relevance, they are assigned to a class. In this case, the classification is based on two main criteria (D4). The first one concerns the role of the asset in the building (if structural or not). The second one is related to the interaction with the architectonic asset, which they are linked to (in some cases they are strictly connected to structural elements, in others they show an own independent seismic response), and on the type of damage. The assignment to a class drives again modelling choices, as discussed in D23. Besides the model type, the relation with the building structures is also indicated, since in certain cases the seismic behaviour of assets is strictly related to that of the structure, while in other cases it is not.

3.1.1.2.Definition of safety and conservation requirements
In this sub-step, in the framework Performance Based Assessment (PBA), the target Performance Levels (PLs) to be fulfilled by the monument in the occasion of given earthquake hazard levels and the related acceptance criteria are defined.

In cultural heritage assets, conservation and safety of people should be considered in an integrated approach. According to this, in PERPETUATE project, the following targets have been considered: Use and human life (U), Building conservation (B); Artistic asset conservation (A). For each target, a primary and a secondary PL are identified. The verification of the primary ones is mandatory, while that of secondary ones is optional (depends on specific cases). It is assumed that, if the primary and the secondary PLs are fulfilled, the remaining ones are usually fulfilled, too.

The return periods may be modified by importance coefficients ( k, where k=U,B,A), which take into account the use and the architectonic and artistic relevance of the examined building (D4).

Acceptance criteria (to verify whether the asset is able to satisfy the required PLs for the considered hazard levels or not) are defined on the basis of Damage Levels (DLs). Although empiric and qualitative, they have the advantage to be directly related to the seismic response of architectonic and artistic assets and, thus, to be evaluable by structural analyses. Five levels (from negligible damage to the complete collapse) are considered, as for example adopted in the EMS98 scale; obviously in the PBA the last one is not interesting. By considering the pushover analysis as the standard method to describe the response of a cultural heritage asset, DLs should be identified on the pushover curve by the displacement dDLi in which the i-th DL is attained. This is done by a multi-criteria approach, as described in the second step of the procedure concerning modelling and safety verification method (D17).

3.1.1.3. Definition of seismic hazard
In this sub-step, seismic input to be considered in the assessment of the monument for the required return periods is discussed and defined. In particular, in agreement with the two types of analysis considered in the procedure, the following types of input are provided: Acceleration-Displacement Response Spectra (ADRS) for nonlinear static (pushover) analyses (D13) and Acceleration Time Histories for nonlinear dynamic analyses.

It is worth noting that the definition of seismic input for cultural heritage assets should be faced by considering the specific characteristics of such type of buildings. Many differences with modern buildings considered in building codes may be highlighted. As an example, historical buildings are sometimes built at locations where strong local site effects occur due to steep topography (e.g. hilltops), basin effects or foundations built on the remains from previous structures. Moreover, the long period of vibration of some classes of asset (consider the case of rocking behaviour in single stone columns) requires a reliable definition of the acceleration-displacement response spectrum also in this range.

The following main points have been analysed within PERPETUATE project:
- Probabilistic definition of the seismic hazard by single or multiple earthquake intensity parameters. A methodology for the identification of the most suitable earthquake Intensity Measures (IM) for historical buildings is proposed (D10). To this aim, many nonlinear dynamic analyses on selected structures have been performed in order to obtain fragility curves related to a great number of intensity parameters (highly-correlated with the seismic response and poorly-correlated together in order to minimize dispersion of results). Then, a "variable clustering" procedure to create several clusters from all parameters, based on principal components analysis of a matrix containing parameter values for each accelerogram, has been developed. Finally, the use of a vector-valued ground motion intensity measures to reduce the uncertainties related to the representation of the seismic action by only one parameter has been discussed (D24). With the support of numerical and experimental tests, it has been demonstrated that, for some architectonic classes, parameters related to the structural behaviour (e.g. spectral displacement) show a better correlation with the observed damage than parameters such as peak ground acceleration (PGA) (this is the case of blocky structures).
- Selection of elastic response spectra. In common practice, elastic response spectra are usually just scaled up or down depending on different earthquake magnitudes and distances, the actual spectrum shape remaining unvaried. However, this might not be the case of inelastic and degrading systems, where the shaking duration will have an effect. Hence, it is not possible to compensate for magnitude effects by scaling only, since the spectral shape changes and the scaling may not be of a linear form. This important problem has been thoroughly discussed through numerous examples. In particular, a great number of numerical dynamic analyses have been carried out and the results were compiled so as to demonstrate the ineffectiveness of the commonly used damped response spectrum in predicting the seismic behaviour of certain type of monuments (consider the case of tall statues, multi-drum columns, slender towers and isolated walls). It is concluded that the use of elastic spectrum as a representative index of destructiveness for any system is limited. The need of more resilient indicators, capable to include the inelastic behaviour of structural systems, has emerged (D13).
- Definition of proper amplification factors and soil categories. New amplification factors and soil categories referring to EC8, defined on the basis of a subset of SHARE's global strong-motion database (related FP7 project Seismic Hazard Harmonization in Europe (SHARE) - WP4 SHARE, 2011), are proposed (D13). A new classification scheme, based on parameters such as average velocity to seismic bedrock, fundamental period and depth of soil deposits is also proposed. Normalized elastic acceleration response spectra are proposed for the new classification system by fitting the spectra equations more closely to the 84th percentile of the data. Period-independent amplification factors are also proposed for the soil classes of the new classification scheme. The proposed acceleration response spectra are then used to derive demand spectra for all the new site classes. All the necessary equations for the derivation of demand spectra for any other level of ground acceleration are provided.
- Definition of permanent ground displacements effects (D21). Besides being the generation source of earthquakes, tectonic faults may also directly affect surface structures by means of permanent ground displacements. Given their significantly longer lifetime expectancy, their immovability, and in many cases their relatively large size, monuments are more likely than most other structures to experience such tectonic hazards. The response of historic masonry buildings subjected to tectonically induced ground distress was studied both experimentally and analytically. A hybrid methodology is proposed to rigorously account for the multiple nonlinearities in the response of the rupturing soil and the distressed masonry structure, combining an experimentally validated 3D FE model for simulation of fault rupture–soil–foundation interaction with a well-established nonlinear macro-element model for simulation of the response of masonry walls.

It comprises two steps:
(a) calculation of the foundation displacements resulting from their interaction with the fault rupture, by assuming elastic response of the superstructure, and
(b) analysis of the nonlinear response of the masonry wall (through the use of TREMURI program) subjected to the foundation displacements obtained by the first step, to predict its consequent damage. Results highlight the determinative role of the exact location of the structure with respect to the fault.

3.1.1.4. As built information
In this sub-step, geometrical, technological and mechanical features of the asset are analysed in depth, with the aim of defining the structural model of the building and related artistic assets.

The as built information process concerns the acquisition of several data related to: geometry of the building; foundations; mechanical parameters estimate; historical data on transformation and damage (with particular reference to past earthquakes); state of maintenance and damage mechanisms identification (in case of post-earthquake assessment); dynamical behaviour. In PERPETUATE procedure, tools to optimize the investigation program are proposed.

3.1.1.5. Modelling and safety verifications
In this step, the final model of the structure is defined (on basis of results achieved by the as built information), its seismic response is calculated and the fulfilment of PLs is verified.

The outcome of the seismic assessment procedure proposed in PERPETUATE is the maximum seismic Intensity Measure (IM) compatible with the fulfilment of a given performance level (IMPLi). To this aim, the following methods of analysis and verification procedures are considered (D35):
Nonlinear Static Analyses and Capacity Spectrum Method (CSM), based on the comparison between the displacement demand, obtained by a properly reduced acceleration-displacement response spectrum, and the displacement capacity.
Incremental Dynamic Analysis (IDA) or Nonlinear Dynamic Analyses with a large amount of records (cloud method), based on the statistical evaluation of IMPLi from the results of nonlinear dynamic analyses with properly selected time histories.

The first method (CSM) is assumed as the standard one; the second (IDA or cloud method), even if more accurate, is necessary only for some classes of assets (e.g. Class F), for which it is applicable with a reasonable computational.

It is worth noting that linear elastic analysis may be considered as possible alternative only in case of very complex assets for which nonlinear analyses are unfeasible (e.g. complex assets in Class B). In these cases, instead of referring to the use of a behaviour factor (q), specific hints to define a simplified capacity curve, based on the combined use of linear elastic analyses with conventional heuristic approaches, are proposed.

In the adoption of CSM, the procedure aimed to evaluate IMPLi may be summarized by the following main steps: execution of the pushover analysis; identification of the DLs, and related PLs, on the pushover curve; conversion of the pushover curve in capacity curve; given the seismic demand (in terms of proper IM), computation of the maximum IM value compatible with the i-th PL (IMPLi).

In order to carry out the pushover analysis, a model of the structure should be defined. To this aim, it is useful to make a distinction between simple assets made by a single macroelement and complex assets made by a set of macroelements. For assets of the first type, the single macroelement is modelled, the assessment being performed on its capacity curve.

For complex assets, it is necessary to distinguish the following two sub-cases:
Buildings characterized by box behaviour. In this case, a 3D model of the whole building is required (global scale approach), the assessment being performed on its overall capacity curve.
Buildings made by a set of N macroelements, which exhibit an almost independent behaviour. In this case, each macroelement is modelled independently (macroelement scale approach) and N capacity curves are calculated, the seismic load being assigned by a proper redistribution. The assessment of whole asset is then made through proper combination criteria.

3.1.1.6 Rehabilitation decisions
In case safety verification highlights the need of improving the seismic capacity of the building, different rehabilitation alternatives may be considered. The design of strengthening interventions is not the only possible choice. Conservation without interventions may be also considered, if strengthening actions would be too invasive. In this case, the usable life of the building is evaluated and further decisions are postponed. Another alternative is the revision of safety requirements that, in practice, means the change of use of the building. Finally, the building may be monitored and models upgraded. Also in this case the choice is forecast in the future, when more accurate and validated tools will be available for sure, due to the improvements of applied research in the field.

3.1.2. Building aggregates
When undertaking the seismic vulnerability assessment for building aggregates of a historic city centre (that may be included in class G of PERPETUATE classification) or for the portfolio of historic monumental assets of a Conservation Authority, the level of detail and the extension to which such an assessment can be performed are conditioned on one hand by the availability of economic resources and time and by the level of technical expertise of the staff appointed, on the other by the level of access to and number and types of buildings within the urban context that can be surveyed to create a realistic inventory profile. One other important factor is the time and purpose for the assessment: in other words, whether this is part of a preventative programme and first step towards definition of a mitigation strategy to reduce losses, or whether it is carried out after a damaging seismic event. In the second case the extent and depth of the assessment is further influenced by the level and distribution of damage, and by whether the assessment is carried out to enact emergency safety measures, such as shoring and debris removal, or whether is carried out at a later stage, aimed at documenting and interpreting the damage occurred, to define repair and retrofitting interventions, and quantify requirements of economic resources and distribution of financial aid.

Within PERPETUATE project, a procedure for the vulnerability assessment of historic centres, methodologically similar to that proposed for single assets, have been developed. Similarly to single asset procedure, the following main steps are considered:
a) definition of safety and conservation requirements;
b) definition of the seismic hazard;
c) analysis of study area and identification of typologies and construction details;
d) identification of size of sample and choice of buildings to be surveyed;
e) data collection and analysis;
f) definition of performance points;
g) evaluation of cumulative fragility functions.

3.2 Experimental tests
Laboratory experimental tests performed within PERPETUATE project were oriented to different objectives.
Shaking table tests performed in the laboratories of NTUA in Greece (D33) and ENEA in Italy (D12) were mainly oriented to: analyse the dynamic behaviour of multi-blocks systems (a multi-drum column, an arch-pier system and a 3-block obelisk), stress their displacement capacity under different seismic inputs, evaluate limit states and validate numerical models typically adopted for this type of structures (belonging to class F); analyse the shear behaviour of cross vaults (real scale masonry vault) under asymmetric boundary conditions; evaluate the effectiveness of retrofitting interventions, such as the introduction of rigid or flexible tie-rods in masonry arches (such as in the case of the arch-pier system) and the insertion of base isolation systems for artistic assets (such as in the case of statues).

Shear tests on masonry walls (D15 and D23) were oriented to two main objectives. On the one side, the mechanical properties of ancient-like brick and stone walls were measured and the reliability of strength criteria were stressed. On the other side, the interaction between the structural behaviour of the wall and the mechanical behaviour of coverings (plasters) was analysed. This latter point was linked to the will to consider the problem of artistic assets strictly connected to structural elements (class Q according to PERPETUATE classification).

Finally, many in situ tests were performed in Slovenia by UL and ZRMK (D15 and D34). Such tests mainly aimed to evaluate the effectiveness of different techniques of interventions (CFRP diagonal or perpendicular stripes and GFRP meshes).

In the following paragraphs, the tests set-up and the main results of each test campaign will be illustrated.

3.2.1. Shaking table tests
3.2.1.1. Multi-block obelisk
The multi-block obelisk tested is a 1:6 scale reproduction of the "Obelisco Lateranense", the tallest and most damaged Egyptian obelisk in Rome. The Obelisk is 32 m tall, and is composed by three blocks of granite connected each other and at the base by three "crux joints" whose mechanical characteristics have been investigated. The monument has been widely studied by mean of ambient vibrations test, seismic monitoring, as well as many in situ non-destructive analyses for mechanical characterization of the material.

3.2.1.2. Arch-piers system
The behaviour of arch-piers systems reinforced with different types of tie-rods has been analysed by two experimental campaigns. The first one, carried out at the University of Genoa, has been performed by adopting the static inclined plane test; the second one, carried out at ENEA Casaccia Research Center, has been performed on the shaking table. The same scale model has been adopted for the static and dynamic tests. In both cases, the results have been interpreted through simple analytical models developed in the framework of displacement-based design methods. The research work had two aims. On the one hand, the influence of the stiffness of tie-rods on arch-piers systems is analysed. On the other hand, the reliability of a displacement-based procedure of analysis based on MBM models for masonry structures subjected to rigid blocks collapse mechanisms is assessed.

The 1:10 model of the arch-pier system was composed by discrete blocks made of plastic material modelled by a numerical control machine. Stiffness was calibrated in order to let the theoretical hypothesis of rigid blocks be fulfilled and to guarantee the repeatability of the tests, thus minimizing damage to blocks due to impacts at collapse. Tie-rods of different sections and materials were considered in the tests.

Through static tests, it has been observed that in free-standing arches the collapse was always produced by the formation of 4 hinges while in reinforced arches, the collapse was produced by 5-hinges mechanisms. This behaviour may be easily explained considering that, in reinforced arches, tie-rods couple the displacements of the top of piers, constraining one degree of freedom. Thus, one more hinge is needed in the arch in order to produce the collapse. The comparative analysis of the results in terms of collapse modes and collapse multipliers has led to the following main considerations: tie-rods increase remarkably the collapse multipliers of free-standing arches, while their stiffness does not influence meaningfully neither the collapse multiplier, nor the collapse mode; the pre-stress applied on tie-rods does scarcely influences the collapse. Finally, by mean of a detailed displacement analysis, it has been assessed the limited capability of MBM model to describe the response of the structure, mainly due to its elastic deformability.

3.2.1.3. Masonry vault
The objective of the tests on the masonry vault was to simulate a specific damage condition that is the drift of vault in its horizontal plane. The idea was to reproduce the real geometry of a generic cross vault of the Mosque of Dey, located in the Citadel of the Casbah of Algiers.

The cross vault was built in real scale, resulting from the intersection of two semi-circular barrel vaults 0.12 m thick. On one side, there was set on a rigid wall, while on the opposite side was set on two short columns, inducing thus a asymmetric boundary condition. The wall was fixed while the base of the two columns was allowed to move along the longitudinal direction, by mean of Teflon bases. A particular technological system of wooden logs inserted at the top of the columns, typical of Algerian building tradition, was reproduced in the model.

3.2.1.4. Multi-drum columns
One of the most common types of ancient monuments around the Mediterranean is the Greek classical temple. Unfortunately, since then, very few of them have remained intact and most of them now consist of a couple of free-standing multi-drum columns. Since these structures are made of stones resting on top of each other without mortar and due to their extreme slenderness, multi-drum columns are particularly vulnerable to seismic shaking. However, their dynamic response is highly non-linear and extremely complicated to be predicted by analytical methods.

3.2.1.5. Base isolation system for statues and artistic assets
A new anti-seismic basement developed by ENEA for earthquake protection of artworks and high vulnerable statues standing on their original basements was proposed within PERPETUATE project. The design approach was to confer the seismic isolation property by means of the geometrical properties of the basements, according the following design targets: maximum seismic isolation of the three components of the earthquake achieved by large displacements; low stiffness and low dissipation; decouple the in plane horizontal components and the vertical component of the motion of the basement; reversibility of the intervention; fully compatibility of the materials; easy maintenance.

The design result is a basement made by two blocks made by marble type "Carrara"; the surfaces of the blocks are modelled as ellipsoid of revolution where 4 spheres made by the same material of the blocks are located. The two blocks are connected by dissipative device made by stainless steel cables for horizontal displacement limitations and recentering.

Shaking table tests on this basement have been performed at the ENEA Casaccia Research Centre. Time histories were natural accelerograms from strong motion database normalized and rescaled to the max hazard of Reggio Calabria and return period of 2475 years.
All tests reached the reduction coefficient of 15-20% for the horizontal accelerations and 2.5-3% for the vertical accelerations.

3.2.2. Shear tests on walls
Shear-compression tests performed at the University of Ljubljana
A first experimental campaign on 18 three-leaf stone masonry walls under compression and combined in-plane shear and compression and on its constituents was performed in Laboratory of Faculty of Civil and Geodetic Engineering and Geodesy, University of Ljubljana.

In this case, the specimens were designed to study the behaviour of multi-leaf stone masonry walls and the influence of presence of stones, which are connecting the outer leaves. Therefore, half of the specimens had header stones going through the whole depth of the specimens, and the other half had no such connecting stones. External leaves were constructed from regular coursed squared ashlars, while the internal core was filled with stone rubble and lime mortar. A two layers lime plaster was applied on one side of the specimens.

3.2.3. In situ tests
In-situ tests were performed on two typical old masonry buildings in Slovenia. First one (labelled: MB) was from 1930-ies. The second one (labelled C) was from 1874.

In the first building (MB), load bearing masonry walls were made with solid clay bricks and weak lime mortar. Walls were of two thicknesses: 30 cm and 45 cm. On each thickness, three specimens were prepared: one unreinforced specimen (labelled O30 and O45), one strengthened with diagonal stripes (D30 and D45) and one with combined horizontal and vertical stripes (H30 and H45). Sheet cut on 10 cm wide stripes were used. They were bonded with epoxy resin on both sides of the masonry. Specimens were prepared by cutting walls on 2.0 m high and 1.0 m wide pieces.

3.3. Software
3.3.1. Software for the analysis of single assets
The displacement-based assessment of complex masonry assets requires the availability of reliable and versatile freeware software packages, to be used by architects and engineers. To this aim, a new tool has been specifically developed for the macro-block modelling of collapse mechanisms (MB-PERPETUATE) and an existing software for the global analysis of masonry buildings (TREMURI) has been significantly improved.

MB-PERPETUATE is a software package based on the macro-blocks model (MBM). It is written in MATLAB and is downloadable from the project site D29. The macro-blocks approach is used to perform the Limit Analysis of structures or the dynamic analyses of simple single or multi-body systems. This approach is particularly interesting for the seismic analysis of out-of-plane response of masonry walls, arch mechanisms, response of churches macroelements and single or multiplied isolated masonry "bodies" behaving as rigid blocks. The observation of damages in post-earthquake survey highlighted the propensity of historical masonry structures to develop damage models involving the formation of macro-blocks. The identification of the structural masonry portions (blocks) is defined "a priori" on the basis of recurrent cracks patterns observed in post-earthquake survey or by CCLM.

3.3.2. Software for the analysis of building aggregates
LOG-IDEAH (D27 and D30 for implementation) is an interactive system with a web-based data collection tool which allows the record of the seismic damage of architectural assets in a semantic database and a processor which uses logic trees for the identification of the collapse mechanisms on the basis of the collected data. It is based upon a hierarchical approach in which the architectural asset is deconstructed into macroelements, the macroelements into structural elements and the structural elements are identified as being linked to or being in itself artistic assets.

The framework of the website connected to LOG-IDEAH has made accessible on line at the following link: http://perpetuate.cs.bath.ac.uk/.This web-tool allows to: logging-in from different locations, creating new records which refer to architectonic assets inspected by on site surveys and pictures, designing simplified sketches of inspected façades, recording damage types and damage levels of damaged structural elements, recording damage types and damage levels of damaged artistic assets, assessing probable collapse mechanisms, uploading pictures at levels of architectonic assets and artistic assets, storing seismic damage in a semantic database and interpreting the collected data by using the decision trees of LOG-IDEAH. The aim of the logic trees is to identify the failure modes of architectural assets on the basis of the collected data. In order to evaluate the performance target at global level in a specified architectural asset, it is necessary to define the global damage of all the facades which compose it, if this is possible. For instance, in the case in which the inspected architectural asset is a mansion, the global damage type and level of the mansion can be obtained as the combination of the seismic damage identification of the structural elements and artistic assets which compose all the facades of the mansion under consideration.

3.4. Case studies
The application to several case studies in different countries (D28 and D40) has been fundamental for the validation of the new assessment procedure proposed by PERPETUATE. The problem of seismic protection of cultural heritage was faced at different scales and conditions of damage: on the one hand, both single monuments and historic centres have been considered; on the other, both undamaged and recently damaged buildings have been considered (in the first case, the objective was the prevention, in the second, the reparation/reconstruction). The analysis of case studies recently damage by an earthquake has been mainly addressed to simulate the actual seismic response shown by the structure and validate the models adopted.

3.4.1. Ardinghelli Palaçe
Ardinghelli Palaçe was seriously damaged by 6th of April 2009 L'Aquila earthquake. Many vaults and extensive portions of the roof collapsed and the out-of-plane response of the top of the main façade was activated. The poor state of maintenance of the palace played a decisive role in the damage occurred. In this case, both nonlinear static and dynamic analyses have been performed. Two different types of model was adopted: a Structural Element Model (SEM) (by using the TREMURI software), for the global analysis; a Macro-Block Models (MBM) (by using the MB-PERPETUATE software) for the analysis of local mechanisms. A first set of analyses has been mainly addressed to simulate the actual seismic response shown by the structure during 2009 earthquake, validate the model and evaluate the pre-earthquake vulnerability condition of the building. A second set of analyses, has allowed us to define the criteria of Ardinghelli Palace's strengthening. Since the safety indexes computed in the pre-earthquake state were not so low, the SEM model adopted – together to the use of non linear analyses - represented a powerful tool to define strengthening interventions aimed to enhance the response of the structure, however without substantially altering its original structural system.

3.4.2. Santa Maria Paganica Church
Santa Maria Paganica church collapsed almost completely during 2009 L'Aquila earthquake. This building is particularly interesting since its seismic response was strongly influenced by the important structural modifications suffered in past. The complex reasons of the extensive collapses, the role of structural modifications (in particular, the replacement of the original wooden roof with a heavy reinforced concrete slab), the interaction between structural components and the structural solutions to be adopted in case of reconstruction, have been analysed through sensitivity analyses performed on a 3D detailed finite element model (CCLM).

3.4.3. St. Pardo Cathedral in Larino
St. Pardo Cathedral suffered in 2002 two earthquakes of moderate magnitude. Despite severe damage occurred to many churches in the centre of Larino, this church only was not seriously damaged. Damage occurred in those parts that were recently retrofitted. In this case, MBM models were firstly defined for the most vulnerable macroelements to identify meaningful parameters and addressing the plan of in-situ investigations. Then, the assessment has been carried out by combining the results of a 3D linear finite element model (CCLM), aimed to evaluate the load distribution among macroelements, with the full set of MBM analyses on macroelements.

3.4.4. Neoclassical School
The Neoclassical School, sited in Rhodes, Greece, is very interesting because of the complex foundation system, partially realized over the debris of ancient pre-existing structures. The original results obtained by PERPETUATE on the problem of soil-foundation interaction, implemented in TREMURI program, allowed us to simulate their effects on the seismic assessment of the building. Results highlighted that building should require severe strengthening interventions in order to fulfil the selected performance levels according to seismic demands. To this aim, several strategies (stiffening of floors, improvement of material characteristics by applying grout injections, improvement of wall connection) have been numerically simulated in order to evaluate their effectiveness.

3.4.5. Arsenal de Milly
The Arsenal de Milly is a quite simple structure, but it is strictly connected, on one side, to the huge defensive city walls of ancient Rhodes. Ambient vibration measurement has highlighted relevant torsional components in the response, own to the asymmetrical boundary condition. In this case, the results obtained by different linear and nonlinear models have been compared. The seismic analysis – by applying the whole procedure proposed in PERPETUATE - has been performed by using four different models: a continuum nonlinear finite element model of a typical cross section of the structure (CCLM-2D) and a continuum nonlinear finite element model of the entire structure (CCLM-3D) (by the code OPENSEES); a Structural Element Model (SEM) of the entire 3D structure (by the code TREMURI); a Macro-Block model (MBM) for the description of cross response of system vault-masonry walls (by using the Mc4Loc software). In the 2D finite element model of the cross section of the structure, soil-foundation interaction effects have been analyzed.

3.4.6. Hassan Bey Mansion
Hassan Bey Mansion is a typical ottoman mansion. The building is now not in use and is in a bad maintenance state (some floors are collapsed and deep cracks compromise the connection between walls). It is going to be interested by quite significant and articulate strengthening interventions, which are not yet defined by the administrators. In this case study, the whole procedure proposed in PERPETUATE project has been applied. In fact, by sensitivity analyses, it was possible to identify the main parameters affecting the structural response and their confidence factor, and establish proper criteria to optimize the plan of investigations. The final assessment performed by a Structural Element Model (SEM) (by the code TREMURI) has showed the actual safety level of the structure is too low and allowed us to identify the critical aspects of its seismic response.

3.4.7. Great Mosque in Algiers
The assessment of the Great Mosque has showed how it is possible to integrate the results of different modelling strategies. In this case study, the attention has been focused on the study of the two systems of arcades supporting the roof (as the roof system cannot be considered as rigid in its horizontal plane, these macro-elements results to be most vulnerable). In order to define the capacity curve of these macro-elements, nonlinear static analyses have been performed by adopting both CCLM and MBM modelling approaches. The results of these two models have been combined in order to define the capacity curve of the unstrengthen building. A further 3D model according to SEM approach (by the code TREMURI) has been then defined in order to evaluate its seismic behaviour after a possible strengthening intervention consisting in roof stiffening. Such intervention aims to improve the box behavior of the building and to increase the interaction among macroelements.

3.4.8. Casbah of Algiers
The Casbah of Algiers is made up of a very complex aggregation of buildings. About 150 buildings have been directly surveyed, promoting in this way a close collaboration with local architects and authorities. The building samples in the Casbah of Algiers have been chosen by taking into account the different typologies and typical features of the Algerian constructions. An in-depth analysis of the geometrical and structural characteristics of the buildings has been made, discovering interesting ancient constructive solutions and valuable artistic assets, also in a rather poor environment. Local damage mechanisms have been identified, mostly related to the overturning of the façade due to foundation settlement and, in some cases, to previous earthquakes. The FAMIVE and LOG-IDEAH tools have been widely applied in this case. This allowed to get a quantitative assessment at large scale, useful for planning mitigation strategies and priorities of intervention. The performance of the building samples has been evaluated by considering the restraining effect of the cohesion or not. Moreover, a strengthening by ties has been supposed.

3.4.9. Kolizej Palaçe
Kolizej Palace was demolished in 2011, despite the overall opinion and battle among conservators upon its preservation. The demolition was decided on the basis of several analyses (based on minimum knowledge level), demonstrating that the asset cannot fulfil current codes and that rehabilitation measures would be enormous and uneconomic. However, none of the carried analysis did not convince the conservators and public. The goal of PERPETUATE for this case study was to perform detailed in-situ investigation and to perform reliable seismic analyses of the structure. Microtremor measurements, in-situ shear and compressive tests, shove tests, moisture measurements as well as numerous laboratory tests on masonry components and assemblages were carried out. Due to the poor state of masonry, mainly due to the moisture conditions and freezing-thawing cycles, results have turned to be an important reference to assess the effect of deterioration on mechanical properties.

3.4.10. Rural buildings in Posocje region
Posocje (Bovec) region was hit in 1998 and 2004 by two severe earthquakes. Although 2004 earthquake had 10 times less energy, some of the buildings that were already retrofitted after 1998 earthquake suffered major structural damage. Within PERPETUATE, 33 of these buildings were investigated with FaMIVE. Aim of this study was to evaluate the efficiency of PERPETUATE methodology, compare the predicted failure mechanisms with observed ones and evaluate vulnerability curves.

3.4.11. Historic centre of Ljubljana
The city of Ljubljana lies in area with high PGA and it has been struck by a severe earthquake in yr. 1895. For this study, 34 aggregated buildings were assessed. They were chosen since they represent the most valuable part of old city center and also in dependence to available documentation in respect to their layouts, cross-sections and record of damage following 1895 seismic event. Since for this case study the level of knowledge of the buildings from aggregate was very low, FAMIVE analysis was carried out considering different uncertainties for input data (like as different masonry bond and overlapping, different values of friction coefficient, different hypotheses on the efficiency of tie- rods, ..). Taking into account both the results analyses and the evidence of the response occurred from the earthquake of 1895, it was concluded to focus decisions on rehabilitation/strengthening measures oriented to prevent out-of-plane failure mechanisms.

Potential Impact:
4.1 Potential impact and exploitation of results
PERPETUATE project developed the following main products:
- Complete procedure for the seismic assessment of cultural heritage assets, described in the European Guidelines for the seismic preservation of cultural heritage assets (D41).
- Knowledge advancement on specific topics of earthquake engineering applied to cultural heritage assets, obtained by both experimental tests and theoretical analyses.
- Set of software for the modelling of masonry structures or urban aggregates, developed ex-novo or improved from a pre-existing basis.

The dissemination of these products among the European and Mediterranean countries will produce different types of impacts.

On the one hand, the European Guidelines for the seismic preservation of cultural heritage assets will represent a reliable and operative basis for eventual future developments of European standard or codes in the seismic and preservation field. Moreover, the methodology proposed by PERPETUATE is quite general so that can be easily exported also in other countries which share similar construction techniques (such as Mediterranean extra-European countries). To this aim, contents of Guidelines will guarantee the transferability of results also beyond Europe.

On the other hand, the availability of an accurate assessment procedure, based on the optimization of investigation plans on the assets (to get information and parameters useful for the seismic analyses with the minimum impact) and on the input (to accurately describe seismic forces), the improvement of the models and verification procedures, the design of compatible interventions, will lead to an increased protection of cultural heritage from earthquake events. In this sense, Guidelines can help architects, engineering and conservation officers employed in public institutions involved in preservation of cultural heritage. The improvement in methods of analysis and assessment procedures proposed will support them in a proper optimization of strengthening intervention. To this aim, the complete set of deliverables constitute the main reference to deepen the steps and seismic assessment criteria summarized in the Guidelines.

4.2 Dissemination activities
The methodology developed during PERPETUATE project and the entire set of results obtained has been spread in all these three years by a wide set of dissemination initiatives planned at the beginning of the Project.

Diversified tools of dissemination have been adopted specifically addressed to different user groups.

In particular, the following four types of relevant end-users have been identified:
a. conservation officers employed in public institutions;
b. engineers and architects involved in the design of strengthening interventions;
c. enterprises working in the field of restoration;
d. scientific community.

Therefore, dissemination actions performed distinguished in:
- general tools, transversally addressed to different classes of end users, like: website, brochure of the project, the "European Guidelines for the protection of cultural heritage assets" (the final product of PERPETUATE project), DVD;
- tools addressed to specific classes of end-users, such as: seminars, training workshops, on-line available freeware software, special session in International conferences, publication of results on International journals.
Following, a summary scheme of dissemination actions taken in these three years.

The first dissemination action taken has been the creation of a Website (see http://www.PERPETUATE.eu online) just at the beginning of the project - as described in D2 - as a gate for information on the project itself, its development and, more in general, cultural heritage topic. The website has therefore been designed in order to meet both internal and external needs as follows:
- Internal needs: improve communication between partners, work together and share confidential data.
- External needs: disseminate the main contents of the project to specialists, potential end-users, public officers, politicians and public funding authorities, as well as the general public.

It is structured in three main parts: Open access Area, Restricted Area and Social Area.

The majority of pages are open to all visitors, in order to promote the dissemination of the research and to permit everybody to browse freely. This part represents the key tool to disseminate the main contents of the project to specialists, potential end-users, public officers, politicians and public funding authorities, as well as the general public. Nevertheless, there are some documents which need to be protected and restricted only to members.

The website has been updated on a regular basis and will be kept active for at least three full years after the project end with a public and private sector in order to guarantee a long-lasting dissemination of the achieved results.

Moreover, from October 2010, we started monitoring visitors data on the website, verifying its increasing number of visitors from almost every country in the world.

Then, in deliverable D1, a brochure of the project was developed, in order to be a useful and nice vehicle of information of PERPETUATE Project.

According to the web site style, the brochure has been developed in the two colours (blue for the external sheet and orange for the internal one ), the same used in the website (blue for the public part of the site and orange for the private one).

One thousand copies of it have been printed and distributed within all the project years in order to promote its dissemination amongst scientific community (through international conferences, seminars and workshop), public institutions and enterprises working in the field of restorations not only by the Coordinator but also from each member partner.

Then, as requested from EC, a synthetic fact sheet of the project has been produced in order to be included in EC website, in the page related to Cultural Heritage Projects (see http://ec.europa.eu/research/environment/index_en.cfm?pg=projects&area=cultural online) together with the other funded projects of the area.

Still moving on outer communication tools, we have created a Project DVD (described in D43) as an attractive and up-to-date instrument to disseminate the main goals and results of PERPETUATE Project.

The aim of the DVD was to synthesize the results of the project and make them available to a large number of final users, in particular to professionals, researchers, public officers and students. The editing choice was to produce an attractive movie, in which scientific contents are alternated with interviews and pictures. In many cases the voice of some of the partners drive the viewer within the project. In some others, a professional voice follows the images. A "soundtrack" has been appositely composed for the movie by a professional musician. The movie is approximately one hour long. The main menu allows to search for single sections separately. A hundred and fifty hardware copies have been prepared and distributed amongst partners, EC and experts.

Regarding contacts with ECTP focus group and activities, in these three years we have been invited to Cultural Heritage events: the NET HERITAGE project conference in March 2011 and the Final event, held in Rome in September 2011. The final product of NET HERITAGE project was the launch of the website "the Heritage Portal", an online new resource for Cultural Heritage research, where PERPETUATE published 8 articles in the last 18 months. (for more detail see Deliverable D19).

In November 2010, PERPETUATE participated - with a single stand - at the AR&PA event (Bienal de la Restauration Y Gestion del Patrimonio) held in Valladolid, Spain. This event unites the representatives and managers of various social and institutional cultural heritage players – restoration companies, universities, suppliers, public authorities, private foundations and a large number of cultural industries together with EC members and representative of many intergovernmental organizations.

Moreover, together with other FP7 European Projects, PERPETUATE has been acknowledged to a clustering between CfC, SMooHS, 3ENCULT and EU-CHIC projects and joined the final Meeting of EU-Chic.

The participation to meeting/conference/events planned by this consortium - according also to aims of ECTP platform – allowed us to disseminate aims of PERPETUATE project at European scale by promoting the exploitation of its research results in a wide context. In fact, the classification of architectonic assets proposed in PERPETUATE (D4) has been assumed as reference also by EU-CHIC project, as in their final brochure.

A series of three Seminars has then been organized on Rhodes (July 2011), Ljubljana (October 2012) and Algiers (February 2013). All of the Seminars had a good success in terms of interests and participation, involving large groups of specialists (mostly architects and engineers) involved in the field of preservation and restoration of Cultural Heritage (see more details on D19).

In all these Seminar the main effort of the speakers has always been to disseminate not only the content of the project itself but a certain awareness of the importance of conservation and safety of masonry cultural heritage buildings and major errors and possible corrective actions in preservation of buildings projects. All partners actively participated in presenting different aspects of the same main theme, derived from one's background and specialization, adding great value to the common dissemination actions.

Some of the abovementioned activities – such as seminars, training workshops – represent, of course, good opportunities to disseminate results achieved in PERTUATE Project also to ICPC countries.

In particular, having Algeria as a ICPC country in the partnership, the project focuses its attention on which could be the best actions to be taken also in response to the great need and demand of know-how coming from that Country.

The first action taken then was the organization of one of the Meeting in Algiers (November 2010) where all the staff had the opportunity to visit and observe "lively" one of the main case studies of the project, the Citadel and the Casbah of Algiers. Following this first visit a team coordinated by UBATH began a field campaign in the Casbah, that continued on April 2011, and which permitted to create a solid team work amongst people from Algiers and European experts and which strongly promoted a direct involvement of local people.

Finally, as abovementioned, the last action has been the organization of a Seminar directly ion the Citadel, the oldest part of the city, open to the public.

Moving to scientific community dissemination actions, PERPETUATE Project has led to a lot of publications, either on journal or on proceedings of conferences.

In particular 3 papers have been published on international Journals (Advanced Materials Research, Materials and Structures and Bulletin of Earthquake Engineering) and four more have just been submitted or is going to in early 2013 (see deliverable D32 for references).

For what concerns the participation to International Conference, results related to PERPETUATE Project have been presented in conferences during these 3 years, all around the world. All 32 papers have been constantly uploaded in PERPETUATE website, in order to spread the first results obtained and to disseminate widely the content of the Project itself.

Moreover, in 2013 PERPETUATE staff is working on two special issues on International Journals: the first regarding the application to Case Studies to Engineering Structure, the other on PERPETUATE assessment procedure on the Bulletin of Earthquake Engineering.

List of Websites:
http://www.perpetuate.eu

Related documents