Assessment of Post-Earthquake Damage : St . Salvatore Church in Acquapagana , Central Italy

This article proposes a multidisciplinary approach for the assessment of seismic damage from the perspective of conservation and prevention. A comparison of the state of damage has been carried out in a case study, St. Salvatore church in Acquapagana (MC), as an example of church, which underwent two important seismic events in the Central Italy area, the 1997 and the 2016 earthquakes. The comparison of the state of damage passes through the following stages: (a) the territorial seismic overview; (b) the historical description and material analysis; (c) the identification of macro-elements with activated damage mechanisms; (d) the comparison between the two seismic events both from a territorialand building-scale perspective. This work puts together the archived and the on-site survey data with those elaborated starting from seismogenic information, available from the National Seismological Institute, and it provides a strategy also for other similar conditions. This work is to be considered a contribution to a wider study that could be carried out in the areas hit by the 2016 earthquake. It could also represent a way to collect documentation in the post-earthquake phase, improving the effectiveness of procedures currently applied to the first level of damage assessment.


Introduction
Earthquakes pose a threat to our cultural heritage assets, constituted in part by churches.The seismic risk of the Italian territory is well known because of the many earthquakes that have occurred over the years.This type of risk is strictly related to the vulnerability of the buildings; indeed, working on them is necessary to preserve the national cultural heritage asset.In this regard, the efforts of researchers are focused on acquiring procedures to access in detail the vulnerability of churches and to think about intervention strategies.Since the 1976 Friuli earthquake, which caused severe damage to the heritage building stock, churches have started to be studied in relation to their evidence of vulnerability under seismic actions.The study of vulnerability is strictly related to the study of damage.The post-earthquake observation of damage led to the elaboration of the forms of damage assessment of churches, whose current version [1] was approved with Decree DPCM of 23 February 2006, becoming a national standard.This survey form associates a typical mechanism of collapse to each macro-element, a well-recognizable structural unit.The observation of damage after the Friuli earthquake of 1976 and its interpretation converged in a book, Le chiese e il terremoto [2], which can be considered the basis of all the subsequent studies on the vulnerability of churches.With the 1997 Umbria-Marche earthquake, the observations of the effectiveness of strengthening and repairing interventions started, allowing one to express considerations about interventions and to collect a large amount of data, as specified in [3].As a consequence of this post-earthquake experience, a manual of best practices was written, Codice di pratica (The Code of Practice) [4].It was published by the Regional Authority of Marche, as a tool in support of professionals that would operate in reconstruction and repairing, suggesting a suitable modus operandi.In addition to this manual, other similar ones were issued in those years, dealing with best practices for interventions, such as [5][6][7][8][9][10].
Moreover, several studies have been carried out to assess the seismic vulnerability of churches and to understand the relative structural behavior under seismic actions by numerical analysis [24][25][26][27].The vulnerability assessment of churches was also studied by some researches from a large-scale perspective (i.e., [28]).
Considering this context and following the recommendations of the Italian Guidelines for Cultural Heritage [29], in accordance with the Italian Building Code (N.T.C. 2008) [30], this article proposes a procedure for the assessment of the damage of churches with a multidisciplinary approach, in line with Principle 1.1 of the ICOMOS Charter [31].This methodology is inserted in a panorama of studies, such as [32][33][34], which believe in the importance of achieving deep knowledge of the historical building, the object of the study, through the contribution provided by different disciplines, with the final aim of preserving the cultural heritage asset.The multidisciplinary approach in the following work is achieved observing the impact of the earthquake on the church case study from different perspectives, taking into account several aspects: the seismogenic system, the registered peak ground acceleration (PGA) values, the history of the church with the analysis of the masonry walls and the executed interventions with the two post-earthquake states of damage (1997)(1998)(1999)(2000)(2001)(2002)(2003)(2004)(2005)(2006)(2007)(2008)(2009)(2010)(2011)(2012)(2013)(2014)(2015)(2016).Indeed, with such an approach, the observation of the post-earthquake damage and the study of the interventions executed on a historical building become part of a wider process, which contributes to achieve the aim of conservation and prevention of damage caused by earthquakes [35] (Chapter 29, paragraphs 1,2,4).
The procedure, exposed here, is applied to the single-nave St. Salvatore church, taken as a case study representative of those churches that underwent earthquakes and interventions in the time lapse between the events.Moreover, the following work intends to provide a contribution to ongoing studies (e.g., [36]) on the state of damage of the churches of Central Italy damaged by the 2016 earthquake.In particular, in [37], a set of single-nave churches, located in Umbria, Perugia, Ascoli Piceno, Macerata and Pesaro-Urbino, is used to compare the surveyed state of damage with the expected damage, according to predictive methods [38,39].
Figure 1 summarizes the procedure followed to investigate the state of damage of St. Salvatore church, which is reflected in the structure of the article.Starting from a territorial analysis (Section 2.1) of the seismicity of the area (the seismogenic source system, active faults, expected accelerations and the catalogue of seismicity), entering into the detail of the historical description and material characterization of the church (Section 2.2) and the identification of the existing macro-elements (Section 2.3), the article compares the two seismic events, the 1997 Umbria-Marche earthquake and the 2016 Central Italy earthquake: first, from a territorial perspective and, later, from a building-scale point of view.The final sections are dedicated to the discussion of the evidence and to sum up the results of the study.
All in all, this work will point out how from a multidisciplinary approach for the assessment of damage it is possible, not only to obtain information regarding the technical aspects about the effectiveness of interventions and situations to be avoided for historical masonry churches, but also to interpret the damage, including seismic and territorial aspects.Meanwhile, another indirect outcome of this approach is the collection of territorial-and building-scale information.

Territorial Overview
For the study, the starting point was the identification of a church in the territory.San Salvatore Church is the only church of Acquapagana, a small hamlet in the municipality of Serravalle di Chienti, in the province of Macerata (Italy).The development of the urban center follows a south-north axis, and the church is located in the southern part of the hamlet (Figure 2a).It is part of an isolated monastic complex, developing with a west-east direction (Figure 2b).This geographical orientation will be useful for the final considerations about the state of damage post the 1997 and post the 2016 earthquakes.The seismicity of this geographical area is well known, as the system of seismogenic sources (Figure 3a), the map of active faults (Figure 3b), the seismic hazard map of the Marche region (Figure 4a), the seismic catalogue (Table 1) and the histogram of seismicity of Serravalle di Chienti (Figure 4b) demonstrate.

Territorial Overview
For the study, the starting point was the identification of a church in the territory.San Salvatore Church is the only church of Acquapagana, a small hamlet in the municipality of Serravalle di Chienti, in the province of Macerata (Italy).The development of the urban center follows a south-north axis, and the church is located in the southern part of the hamlet (Figure 2a).It is part of an isolated monastic complex, developing with a west-east direction (Figure 2b).This geographical orientation will be useful for the final considerations about the state of damage post the 1997 and post the 2016 earthquakes.

Territorial Overview
For the study, the starting point was the identification of a church in the territory.San Salvatore Church is the only church of Acquapagana, a small hamlet in the municipality of Serravalle di Chienti, in the province of Macerata (Italy).The development of the urban center follows a south-north axis, and the church is located in the southern part of the hamlet (Figure 2a).It is part of an isolated monastic complex, developing with a west-east direction (Figure 2b).This geographical orientation will be useful for the final considerations about the state of damage post the 1997 and post the 2016 earthquakes.The seismicity of this geographical area is well known, as the system of seismogenic sources (Figure 3a), the map of active faults (Figure 3b), the seismic hazard map of the Marche region (Figure 4a), the seismic catalogue (Table 1) and the histogram of seismicity of Serravalle di Chienti (Figure 4b) demonstrate.The seismicity of this geographical area is well known, as the system of seismogenic sources (Figure 3a), the map of active faults (Figure 3b), the seismic hazard map of the Marche region (Figure 4a), the seismic catalogue (Table 1) and the histogram of seismicity of Serravalle di Chienti (Figure 4b) demonstrate.The territory of Serravalle di Chienti is in the Colfiorito-Campotosto seismogenic source and near that of Vettore-Porche-Bove (Figure 3a).In these areas, active faults are concentrated (Figure 3b), and earthquakes, therefore, are generated.For Serravalle di Chienti, values of peak ground acceleration (PGA) between 0.225 and 0.250 g are expected, according to the map in Figure 4a.These values are useful in the perspective of damage prevention, but also to express an evaluation about accelerations occurring during a seismic event.The recurrence of earthquakes in this area is confirmed by Table 1 and Figure 4b, where the correspondent microseismic intensity and the moment magnitude of the seismic events from 1279-2006 for Serravalle di Chienti are reported.This table helps to have a perceptive of how many earthquakes churches located in that territory could have undergone.Most of the listed events belong to the seismic sequence corresponding to the 1997 Umbria-Marche earthquake, labelled as Appennino Umbro-Marchigiano.

Historical Description and Material Analysis of the Building
The original nucleus of the monastic complex of Acquapagana goes back to the XI Century: Saint Romualdo built both the church and the monastery in 1007.This date is confirmed by Lodovico Jacobini, a historian who found the complex in 1653, thanks to a parchment of 1063, which has been lost [45].The church is an example of the Gothic-Umbrian style [46]: the smooth stone of the portal and of the rosette, the horizontal frame on the façade, the buttresses and the single windows on the lateral walls are clearly elements of this architectural period and of the geographical area under examination.Entering more into detail, on the west side of the church (Figure 5b), the principal façade appears, shaped as a simple hut with two openings: a portal, with a white stone frame, and a rose window.On the lateral sides (Figure 5a,c), the façades are articulated with buttresses to balance the thrust of the inner triumphal arches.In these walls, there are few openings: two small ones on the south side (Figure 5c) and one on the north and east side (Figure 5a,d).According to [45], other information on the church is dated 1737.At that time, according to the inventory of Camerino archbishopric's archive, the bell tower appeared on the main façade.It is assumed that the bell tower on the north side of the presbytery was built later.In 1960, the wooden roof was replaced with a concrete slab and a concrete curb [45], also visible in the graphic documentation of the 1997 earthquake, and then, after the evidence of the damage caused [46], the roof was rebuilt with a light timber structure [45].The church is an example of the Gothic-Umbrian style [46]: the smooth stone of the portal and of the rosette, the horizontal frame on the façade, the buttresses and the single windows on the lateral walls are clearly elements of this architectural period and of the geographical area under examination.Entering more into detail, on the west side of the church (Figure 5b), the principal façade appears, shaped as a simple hut with two openings: a portal, with a white stone frame, and a rose window.On the lateral sides (Figure 5a, c), the façades are articulated with buttresses to balance the thrust of the inner triumphal arches.In these walls, there are few openings: two small ones on the south side (Figure 5c) and one on the north and east side (Figure 5a, d).According to [45], other information on the church is dated 1737.At that time, according to the inventory of Camerino archbishopric's archive, the bell tower appeared on the main façade.It is assumed that the bell tower on the north side of the presbytery was built later.In 1960, the wooden roof was replaced with a concrete slab and a concrete curb [45], also visible in the graphic documentation of the 1997 earthquake, and then, after the evidence of the damage caused [46], the roof was rebuilt with a light timber structure [45].Regarding the planimetric configuration, the examined church consists of a single rectangular nave surrounded by continuous stone-framed masonry walls with an apse that ends the church on the back side and gives access to the bell tower, remaining perfectly in axis with the entrance (Figure 6a).A stair located in the southwest corner of the nave allows one to enter the timber choir above the principal entrance (Figure 6b).The nave is marked by triumphal arches with a regular inter-beam distance to provide support to the principal structure of the timber roof, which replaced the pre-existent concrete roof after the 1997 earthquake.Today, the roof structure is clearly visible: simple beams, running along the width of the church, bearing a secondary structure made of timber joists which provide the support to the roof (Figure 6c,d).Regarding the planimetric configuration, the examined church consists of a single rectangular nave surrounded by continuous stone-framed masonry walls with an apse that ends the church on the back side and gives access to the bell tower, remaining perfectly in axis with the entrance (Figure 6a).A stair located in the southwest corner of the nave allows one to enter the timber choir above the principal entrance (Figure 6b).The nave is marked by triumphal arches with a regular inter-beam distance to provide support to the principal structure of the timber roof, which replaced the pre-existent concrete roof after the 1997 earthquake.Today, the roof structure is clearly visible: simple beams, running along the width of the church, bearing a secondary structure made of timber joists which provide the support to the roof (Figure 6c,d).A fundamental step considered in the comprehension of the construction history of the church is the analysis of masonry.The first necessary action is the analysis of texture to understand how it was built.The correspondence to the "rule of the art" was used to assess the seismic vulnerability of the church masonry walls.With this term is meant the totality of all those constructive techniques that guarantee good behavior and ensure compactness and monolithism [48] (p.238), through quality evaluation criteria and the identification of possible phenomena of instability or degradation.
The focus was set on the façade, extracting and analyzing a portion of 200 cm of width and 90 cm of thickness (see the red square in Figure 7a and redrawn in Figure 7b).For the global assessment of masonry quality, the selected specimen has been evaluated, following the indications, given in [48], by which seven factors, which influence the good realization of masonry, are identified.Each factor is evaluated as: respected, or partially respected, or not respected.The seven factors are: A fundamental step considered in the comprehension of the construction history of the church is the analysis of masonry.The first necessary action is the analysis of texture to understand how it was built.The correspondence to the "rule of the art" was used to assess the seismic vulnerability of the church masonry walls.With this term is meant the totality of all those constructive techniques that guarantee good behavior and ensure compactness and monolithism [48] (p.238), through quality evaluation criteria and the identification of possible phenomena of instability or degradation.
The focus was set on the façade, extracting and analyzing a portion of 200 cm of width and 90 cm of thickness (see the red square in Figure 7a and redrawn in Figure 7b).A fundamental step considered in the comprehension of the construction history of the church is the analysis of masonry.The first necessary action is the analysis of texture to understand how it was built.The correspondence to the "rule of the art" was used to assess the seismic vulnerability of the church masonry walls.With this term is meant the totality of all those constructive techniques that guarantee good behavior and ensure compactness and monolithism [48] (p.238), through quality evaluation criteria and the identification of possible phenomena of instability or degradation.
The focus was set on the façade, extracting and analyzing a portion of 200 cm of width and 90 cm of thickness (see the red square in Figure 7a and redrawn in Figure 7b).For the global assessment of masonry quality, the selected specimen has been evaluated, following the indications, given in [48], by which seven factors, which influence the good realization of masonry, are identified.Each factor is evaluated as: respected, or partially respected, or not respected.The seven factors are: For the global assessment of masonry quality, the selected specimen has been evaluated, following the indications, given in [48], by which seven factors, which influence the good realization of masonry, are identified.Each factor is evaluated as: respected, or partially respected, or not respected.The seven factors are:

•
Good quality of the mortar.

•
Presence of bond stones, or stones passing through the entire thickness of the wall.

•
Shape of the elements.

•
Dimension of the elements.

•
Presence of staggering in the vertical joints.

•
Presence of horizontal rows.

•
Good quality of resistant elements.
These evaluation criteria help with the comprehension of the response of a masonry panel to gravity and seismic actions, and they allow one to identify parameters that, if present, guarantee the masonry quality.The considered panel is assumed to be subject to vertical loads (both concentrated and distributed) in the middle plane and orthogonal to it.Each parameter is described for every action with a score, which contributes to the definition of the final wall quality index.This index corresponds to different categories for the quality of masonry (A = good masonry behavior; B = average masonry quality; C = insufficient behavior).
Regarding St. Salvatore church, stone elements do not exhibit degradation phenomena, have a regular and squared shape and well-defined edges (Figure 8a).This is clearly visible in lateral walls and in the principal façade, while the state of conservation and the pattern of the material are hidden by the plaster in the east front and in the annexed body in the north side of the church.Prevalent dimensions are a 30-50-cm length, a 10-25-cm height and a 10-20-cm thickness, but many elements are also present with a major dimension of 70 cm.The wall section, analyzed with the image of the 1997 earthquake damage, shows the presence of a rubble masonry: three leaves, two external layers with a filling material, linked with some cross blocks (Figure 8b).Mortar joints are in good condition and well preserved: large stone elements and thin layers of mortar are well proportioned.Horizontal rows (Figure 8c) characterize all the length of the wall without interruption, while the splitting of vertical joints is only partially respected (Figure 8d), because some joints are not in an intermediate position with respect to the stone element.Analyzing the corners, it is possible to observe the presence of cross blocks, which help to assume a good connection between lateral walls and the façade.The global evaluation of the masonry is summarized in Table 2: a section is dedicated to show the photographic details of the masonry, another to show the schemes, one more to show the descriptions of the techniques of construction and materials and one to show the geometrical stone characteristics; finally, the last section is dedicated to the correspondence of the seven factors with a final evaluation of the quality of the masonry, resulting in being of good quality.

The Identification of Existing Macro-Elements
Earthquakes, which have occurred in Italy in the last few decades, have produced severe damage to the masonry building stock: especially churches have been shown to be particularly vulnerable also to low seismic accelerations.The reason for such vulnerability is connected to the church typological characteristics, such as the presence of slender walls, heavy construction elements producing thrust actions on walls and the absence of diaphragm slabs.

The Identification of Existing Macro-Elements
Earthquakes, which have occurred in Italy in the last few decades, have produced severe damage to the masonry building stock: especially churches have been shown to be particularly vulnerable also to low seismic accelerations.The reason for such vulnerability is connected to the church typological characteristics, such as the presence of slender walls, heavy construction elements producing thrust actions on walls and the absence of diaphragm slabs.
Systematic studies of church damage have been carried out since the 1976 Friuli earthquake.Indeed, a process of damage data collection has started from that occasion and has been continued also after the following major earthquakes.The observation of the structural behavior of churches clearly allows one to understand typical aspects of the response during earthquakes, identifying well-recognizable parts of the building that show unity in their behavior during earthquakes, labeled as macro-elements [2].
Since the study, Le chiese e il terremoto [2], a strong connection between macro-elements and collapse mechanisms has been pointed out, ending with the elaboration of the form of damage for churches [1], based on 28 kinematic mechanisms, including both out-of-plane and in-plane collapse modalities [50].The survey form for the damage of churches is used for the first-level damage assessment, to have accessibility and the state of damage produced by the seismic event.It is filled by a team of experts, following the procedure indicated in a specific manual [51].This is an essential action that helps to define the planning of safety interventions and the next repairing activities.For the present study, the existing macro-elements of the church have been identified.The tridimensional model (Figure 9) clearly shows the main structural units; out of all the possible macro-elements, only a few are present: façade, lateral walls, bell tower, apse, triumphal arch.Consequently, kinematic mechanisms have been associated with each macro-element.

The Identification of Existing Macro-Elements
Earthquakes, which have occurred in Italy in the last few decades, have produced severe damage to the masonry building stock: especially churches have been shown to be particularly vulnerable also to low seismic accelerations.The reason for such vulnerability is connected to the church typological characteristics, such as the presence of slender walls, heavy construction elements producing thrust actions on walls and the absence of diaphragm slabs.
Systematic studies of church damage have been carried out since the 1976 Friuli earthquake.Indeed, a process of damage data collection has started from that occasion and has been continued also after the following major earthquakes.The observation of the structural behavior of churches clearly allows one to understand typical aspects of the response during earthquakes, identifying well-recognizable parts of the building that show unity in their behavior during earthquakes, labeled as macro-elements [2].
Since the study, Le chiese e il terremoto [2], a strong connection between macro-elements and collapse mechanisms has been pointed out, ending with the elaboration of the form of damage for churches [1], based on 28 kinematic mechanisms, including both out-of-plane and in-plane collapse modalities [50].The survey form for the damage of churches is used for the first-level damage assessment, to have accessibility and the state of damage produced by the seismic event.It is filled by a team of experts, following the procedure indicated in a specific manual [51].This is an essential action that helps to define the planning of safety interventions and the next repairing activities.For the present study, the existing macro-elements of the church have been identified.The tridimensional model (Figure 9) clearly shows the main structural units; out of all the possible macro-elements, only a few are present: façade, lateral walls, bell tower, apse, triumphal arch.Consequently, kinematic mechanisms have been associated with each macro-element.Figure 10 shows the activated mechanisms and the relative damage, reported by the team of surveyors.This interpretation, made for each macro-element, leads to elaborate the general damage index of the church (I d ); in this case, the reported value is equal to 0.32 [52].Figure 10 shows the activated mechanisms and the relative damage, reported by the team of surveyors.This interpretation, made for each macro-element, leads to elaborate the general damage index of the church (Id); in this case, the reported value is equal to 0.32 [52].
As a result of the survey, the church was declared inaccessible [52].Some of the mechanisms, in the church façade and in the bell tower, that occurred in the apse area have a serious or rather diffused damage.To these aspects could be associated with the given assessment of accessibility.

Two Earthquake Episodes in Comparison: the 1997 Umbria-Marche Earthquake and the 2016 Central Italy Earthquake
The seismicity of the area of Serravalle di Chienti is well known, as pointed out in Section 2.1.The major recent events, which hit this area and are discussed in the present work, are the 1997 Umbria-Marche and the 2016 Central Italy earthquakes.St. Salvatore Church suffered damage in As a result of the survey, the church was declared inaccessible [52].Some of the mechanisms, in the church façade and in the bell tower, that occurred in the apse area have a serious or rather diffused damage.To these aspects could be associated with the given assessment of accessibility.

Two Earthquake Episodes in Comparison: the 1997 Umbria-Marche Earthquake and the 2016 Central Italy Earthquake
The seismicity of the area of Serravalle di Chienti is well known, as pointed out in Section 2.1.The major recent events, which hit this area and are discussed in the present work, are the 1997 Umbria-Marche and the 2016 Central Italy earthquakes.St. Salvatore Church suffered damage in both cases, even if at different levels, as is clearly visible looking at the infographic documentation.Understanding what happened from a seismic territorial perspective is one of the essential steps for the comprehension of the state of damage in the analyzed case study.Indeed, a territorial overview, together with building-scale photographic documentation, allows a deeper comprehension of the state of damage.

The 1997 Umbria-Marche Earthquake
In the first days of September 1997, close to Colfiorito, an incremental seismic phenomenon was registered.On 26 September 1997 at 0:33 a.m., the first shock was felt with a 5.70 moment magnitude (Mw) (Figure 11a) at 3.51 km of depth, with the epicenter located at 43.022 latitude and 12.891 longitude [53].After some hours, at 9:40 a.m., another strong shock (Figure 11b), with 6.01 Mw and at 9.87 km of depth, hit the area with the epicenter slightly to the north (43.014 lat.and 12.853 long.)[54].Then, on 14 October 1997 at 3:23 p.m., a shock (Figure 11c) of 5.65 Mw at 7.33 km of depth hit Sellano, with the at 42.898 latitude and 12.898 longitude [55].Other shocks followed until the first months of 1998, as is visible in the historical catalogue of seismicity of Serravalle di Chienti (Table 1).All the epicenters of the main shocks belong to the Colfiorito seismogenic unit, indicated with the number 31 in Figure 3b.Understanding what happened from a seismic territorial perspective is one of the essential steps for the comprehension of the state of damage in the analyzed case study.Indeed, a territorial overview, together with building-scale photographic documentation, allows a deeper comprehension of the state of damage.

The 1997 Umbria-Marche Earthquake
In the first days of September 1997, close to Colfiorito, an incremental seismic phenomenon was registered.On 26 September 1997 at 0:33 a.m., the first shock was felt with a 5.70 moment magnitude (Mw) (Figure 11a) at 3.51 km of depth, with the epicenter located at 43.022 latitude and 12.891 longitude [53].After some hours, at 9:40 a.m., another strong shock (Figure 11b), with 6.01 Mw and at 9.87 km of depth, hit the area with the epicenter slightly to the north (43.014 lat.and 12.853 long.)[54].Then, on 14 October 1997 at 3:23 p.m., a shock (Figure 11c) of 5.65 Mw at 7.33 km of depth hit Sellano, with the epicenter at 42.898 latitude and 12.898 longitude [55].Other shocks followed until the first months of 1998, as is visible in the historical catalogue of seismicity of Serravalle di Chienti (Table 1).All the epicenters of the main shocks belong to the Colfiorito seismogenic unit, indicated with the number 31 in Figure 3b.After this series of shocks, the church suffered severe damage, as the collected images document.The south portion of the complex collapsed, while the upper part of the bell tower was strongly affected, losing portions of walls (Figure 12) in correspondence to the opening, probably due to the good connection of the transversal walls, which determined the migration of the damage in the weakest and unbound part of the structure.After this series of shocks, the church suffered severe damage, as the collected images document.The south portion of the complex collapsed, while the upper part of the bell tower was strongly affected, losing portions of walls (Figure 12) in correspondence to the opening, probably due to the good connection of the transversal walls, which determined the migration of the damage in the weakest and unbound part of the structure.Understanding what happened from a seismic territorial perspective is one of the essential steps for the comprehension of the state of damage in the analyzed case study.Indeed, a territorial overview, together with building-scale photographic documentation, allows a deeper comprehension of the state of damage.

The 1997 Umbria-Marche Earthquake
In the first days of September 1997, close to Colfiorito, an incremental seismic phenomenon was registered.On 26 September 1997 at 0:33 a.m., the first shock was felt with a 5.70 moment magnitude (Mw) (Figure 11a) at 3.51 km of depth, with the epicenter located at 43.022 latitude and 12.891 longitude [53].After some hours, at 9:40 a.m., another strong shock (Figure 11b), with 6.01 Mw and at 9.87 km of depth, hit the area with the epicenter slightly to the north (43.014 lat.and 12.853 long.)[54].Then, on 14 October 1997 at 3:23 p.m., a shock (Figure 11c) of 5.65 Mw at 7.33 km of depth hit Sellano, with the epicenter at 42.898 latitude and 12.898 longitude [55].Other shocks followed until the first months of 1998, as is visible in the historical catalogue of seismicity of Serravalle di Chienti (Table 1).All the epicenters of the main shocks belong to the Colfiorito seismogenic unit, indicated with the number 31 in Figure 3b.After this series of shocks, the church suffered severe damage, as the collected images document.The south portion of the complex collapsed, while the upper part of the bell tower was strongly affected, losing portions of walls (Figure 12) in correspondence to the opening, probably due to the good connection of the transversal walls, which determined the migration of the damage in the weakest and unbound part of the structure.

Figure 12.
The image shows the state of damage post the 1997 earthquake of the east side of the building.In the foreground, the portion of the collapsed monastic complex is visible, while in the background, the damaged bell tower appears [56].
As mentioned in Section 2.2, the connection between lateral walls of the nave and the façade is well realized (see the image of the corner in Table 2).This aspect influences the response of the building against horizontal forces, developing a box-like behavior.Indeed, the collapse of lateral walls is located in the upper area, in contact with the roof.In this regard, the information about the substitution of the roof in the 1960s helps to understand the damage.Under seismic cycles, the concrete roof pounded on the masonry walls, causing local damage.The correlated state of damage is clearly visible in the upper part of the lateral walls (north and south) and of the buttresses (Figure 13a).Moreover, the multi-leaf walls, shaken and pounded, show the of the external leaf from the standing interior one.This phenomenon is visible both in the north walls (Figure 13b) and in the gable part of the main façade (Figure 13c).As in the case of the north wall, in the portion above the rose window in contact with the roof, the pounding effect caused the loss of a part of the masonry.In Figure 13c, this is visible as rubble on the ground in front of the church.The main façade also showed in-plane shear cracks in the area, which goes from the gable to the lower south west corner.These kinds of cracks followed the path of the mortar, denouncing the lower strength capacity of the mortar compared to that of the masonry.Moreover, none of the cracks cross the blocks of stone, confirming the good quality of the masonry evaluated in Section 2.2.
Buildings 2018, 8, x FOR PEER REVIEW 13 of 23 The image shows the state of damage post the 1997 earthquake of the east side of the building.In the foreground, the portion of the collapsed monastic complex is visible, while in the background, the damaged bell tower appears [56].
As mentioned in Section 2.2, the connection between lateral walls of the nave and the façade is well realized (see the image of the corner in Table 2).This aspect influences the response of the building against horizontal forces, developing a box-like behavior.Indeed, the collapse of lateral walls is located in the upper area, in contact with the roof.In this regard, the information about the substitution of the roof in the 1960s helps to understand the damage.Under seismic cycles, the concrete roof pounded on the masonry walls, causing local damage.The correlated state of damage is clearly visible in the upper part of the lateral walls (north and south) and of the buttresses (Figure 13a).Moreover, the multi-leaf walls, shaken and pounded, show the separation of the external leaf from the standing interior one.This phenomenon is visible both in the north walls (Figure 13b) and in the gable part of the main façade (Figure 13c).As in the case of the north wall, in the portion above the rose window in contact with the roof, the pounding effect caused the loss of a part of the masonry.In Figure 13c, this is visible as rubble on the ground in front of the church.The main façade also showed in-plane shear cracks in the area, which goes from the gable to the lower south west corner.These kinds of cracks followed the path of the mortar, denouncing the lower strength capacity of the mortar compared to that of the masonry.Moreover, none of the cracks cross the blocks of stone, confirming the good quality of the masonry evaluated in Section 2.2.Finally, the damage also occurred the interior of the church.The stiffness of the concrete roof determined shear cracks in the triumphal masonry arches, as shown in Figure 14a.Furthermore, from the images of the church interior, the loss of both the internal leaf of the façade wall and of part of the upper portion of the lateral walls (Figure 14b) is visible.The explanation of this damage is the same as the exterior damage.Finally, the damage also occurred the interior of the church.The stiffness of the concrete roof determined shear cracks in the triumphal masonry arches, as shown in Figure 14a.Furthermore, from the images of the church interior, the loss of both the internal leaf of the façade wall and of part of the upper portion of the lateral walls (Figure 14b) is visible.The explanation of this damage is the same as the exterior damage.

The 2016 Central Italy Earthquake
Acquapagana in Serravalle di Chienti is among the areas hit by the several shocks that have affected Central Italy since August 2016.The major shocks happened on 24 August 2016, on 26 and 30 October 2016 and on 18 January 2017.The I.N.G.V. (Istituto Nazionale di Geofisica e Vulcanologia) identifies these events with the so-called Amatrice-Norcia-Visso seismic sequence.Maps in Figure 15 show the PGA curves for each seismic event together with the church location.For the series of shocks, other data, such as the shake maps, could be found in [58][59][60][61].

The 2016 Central Italy Earthquake
Acquapagana in Serravalle di Chienti is among the areas hit by the several shocks that have affected Central Italy since August 2016.The major shocks happened on 24 August 2016, on 26 and 30 October 2016 and on 18 January 2017.The I.N.G.V. (Istituto Nazionale di Geofisica e Vulcanologia) identifies these events with the so-called Amatrice-Norcia-Visso seismic sequence.Maps in Figure 15 show the PGA curves for each seismic event together with the church location.For the series of shocks, other data, such as the shake maps, could be found in [58][59][60][61].These maps are the result of a QGIS software [62] elaboration: the overlapping of geographical information-St.Salvatore Church coordinates (42°59′00.5′′N 12°55′49.2′′E) and administrative limits-with the curves of the PGA (% g) allows one, for each seismic event, to connect different kinds of information.Even if a more specific assessment would require the acceleration time history at the site, this step of knowledge is helpful for the comprehension of what happened in those areas.Indeed, these maps show that, on 24 August, the church reached a PGA curve of 12% g (Figure 15a), on the 26 October of 8% of g. (Figure 15b), on 30 October of 12% of g (Figure 15c), while, on 18 January, the church is outside the curve of 2.5% of g (Figure 15d).Even if in this last case, the PGA value is lower than in the preceding shocks, the church structure could have suffered damage because it was already compromised.In this regard, [63,64] could contribute to explaining the progression of damage caused by the repetitive shocks.Moreover, all the values stay in the range of the expected acceleration of the area, specified in Section 2.1.
The effects of the 2016 earthquake on the building have been studied according to: the onsite external survey, the collection of photographic documentation related to the moment before the provisional interventions and the survey form [52] of St. Salvatore filled out by a ReLUIS (Rete These maps are the result of a QGIS software [62] elaboration: the overlapping of geographical information-St.Salvatore Church coordinates (42 • 59 00.5 N 12 • 55 49.2 E) and administrative limits-with the curves of the PGA (% g) allows one, for each seismic event, to connect different kinds of information.Even if a more specific assessment would require the acceleration time history at the site, this step of knowledge is helpful for the comprehension of what happened in those areas.Indeed, these maps show that, on 24 August, the church reached a PGA curve of 12% g (Figure 15a), on the 26 October of 8% of g. (Figure 15b), on 30 October of 12% of g (Figure 15c), while, on 18 January, the church is outside the curve of 2.5% of g (Figure 15d).Even if in this last case, the PGA value is lower than in the preceding shocks, the church structure could have suffered damage because it was already compromised.In this regard, [63,64] could contribute to explaining the progression of damage caused by the repetitive shocks.Moreover, all the values stay in the range of the expected acceleration of the area, specified in Section 2.1.
The effects of the 2016 earthquake on the building have been studied according to: the onsite external survey, the collection of photographic documentation related to the moment before the provisional interventions and the survey form [52] of St. Salvatore filled out by a ReLUIS (Rete Laboratori Universitari di Ingegneria Sismica) team after the 2016 earthquake.Both the bell tower and the church building present a significant state of damage.Externally, in particular on the east side, the bell tower appears compromised: along the height, shear cracks are present; due to the plaster, it is not possible to affirm that the cracks follow the mortar.Probably the path of the cracks, at some points, provides evidence of the discontinuity of the masonry caused by reconstruction after seismic events (Figure 16a,b).This is visible in the upper part of the bell tower.Indeed, Figure 12 confirms that this portion was rebuilt after the 1997 earthquake.Moreover, the upper part of the bell tower, on the west side, shows not only shear cracks, but also torsional ones (Figure 16c).The presence of the tie rods efficiently works: indeed, this portion of the tower reacts with a box-like behavior effect, providing a contrast to the torsional action and causing cracks at the base of the terminal part of the bell tower.Figure 16d shows a diagonal and vertical crack that should be connected to the pounding effect between the annexed body and the buttress of the nave, which shows Mechanism 5 (see Section 2.3).Laboratori Universitari di Ingegneria Sismica) team after the 2016 earthquake.Both the bell tower and the church building present a significant state of damage.Externally, in particular on the east side, the bell tower appears compromised: along the height, shear cracks are present; due to the plaster, it is not possible to affirm that the cracks follow the mortar.Probably the path of the cracks, at some points, provides evidence of the discontinuity of the masonry caused by reconstruction after seismic events (Figure 16a, b).This is visible in the upper part of the bell tower.Indeed, Figure 12 confirms that this portion was rebuilt after the 1997 earthquake.Moreover, the upper part of the bell tower, on the west side, shows not only shear cracks, but also torsional ones (Figure 16c).The presence of the tie rods efficiently works: indeed, this portion of the tower reacts with a box-like behavior effect, providing a contrast to the torsional action and causing cracks at the base of the terminal part of the bell tower.Figure 16d shows a diagonal and vertical crack that should be connected to the pounding effect between the annexed body and the buttress of the nave, which shows Mechanism 5 (see Section 2.3).The damage of the church building appeared more concentrated in the principal façade.Indeed, from external images (Figure 18), the façade shows light visible lesions in correspondence with the timber beam of the roof, which causes shear cracks because of the pounding effect.Light lines of The state of damage observed outside has a correspondence with the interior part of the bell tower, where the cracks cross the different levels of the structure (Figures 16c and 17a,b).Laboratori Universitari di Ingegneria Sismica) team after the 2016 earthquake.Both the bell tower and the church building present a significant state of damage.Externally, in particular on the east side, the bell tower appears compromised: along the height, shear cracks are present; due to the plaster, it is not possible to affirm that the cracks follow the mortar.Probably the path of the cracks, at some points, provides evidence of the discontinuity of the masonry caused by reconstruction after seismic events (Figure 16a, b).This is visible in the upper part of the bell tower.Indeed, Figure 12 confirms that this portion was rebuilt after the 1997 earthquake.Moreover, the upper part of the bell tower, on the west side, shows not only shear cracks, but also torsional ones (Figure 16c).The presence of the tie rods efficiently works: indeed, this portion of the tower reacts with a box-like behavior effect, providing a contrast to the torsional action and causing cracks at the base of the terminal part of the bell tower.Figure 16d shows a diagonal and vertical crack that should be connected to the pounding effect between the annexed body and the buttress of the nave, which shows Mechanism 5 (see Section 2.3).The state of damage observed outside has a correspondence with the interior part of the bell tower, where the cracks cross the different levels of the structure (Figures 16c and 17a, b).The damage of the church building appeared more concentrated in the principal façade.Indeed, from external images (Figure 18), the façade shows light visible lesions in correspondence with the timber beam of the roof, which causes shear cracks because of the pounding effect.Light lines of The damage of the church building appeared more concentrated in the principal façade.Indeed, from external images (Figure 18), the façade shows light visible lesions in correspondence with the timber beam of the roof, which causes shear cracks because of the pounding effect.Light lines of damage are present, retracing the discontinuity of the masonry due to the post-1997 earthquake reconstruction of the gable of the façade.The pounding effect of the timber beams of the roof, introduced post-1997 earthquake, is more evident from the interior of the church (Figure 19a).The overturning mechanism of the façade is shown in Figure 19b.A vertical line of discontinuity denounces the overturning of the façade, evaluated in Figure 10, underlined also by the diagonal cracks in the lateral walls.

Discussion
The multidisciplinary approach, exemplified in this article, for the assessment of the state of damage has provided a more exhaustive vision, not limited to the building scale.Several aspects have been considered, as summarized in Figure 1.Information on the seismogenic source systems, on the PGA accelerations, on the localization of the epicenters, archived data of the historical seismicity of the area, the documentation of the state of damage from the survey form used for churches and images of the state of damage have been employed.
The comparison between the two seismic episodes, the 1997 Umbria-Marche and 2016 Central Italy earthquake, shows a milder state of damage in the last one.In summary, this could be caused by: (a) a lower distance from the epicenters of the several shocks in the 1997 Umbria-Marche earthquake compared to those of the 2016 Central Italy earthquake (Figure 20); (b) the different propagation direction of waves, from the northwest (26 and 27 September) and from the south (14 October) in the 1997 earthquake, while southeast/east in the 2016 earthquake, in reference to the position of the church explained in Section 2.1 and in Figure 2c, as well as the effectiveness of the interventions done after the 1997 earthquake.The pounding effect of the timber beams of the roof, introduced post-1997 earthquake, is more evident from the interior of the church (Figure 19a).The overturning mechanism of the façade is shown in Figure 19b.A vertical line of discontinuity denounces the overturning of the façade, evaluated in Figure 10, underlined also by the diagonal cracks in the lateral walls.The pounding effect of the timber beams of the roof, introduced post-1997 earthquake, is more evident from the interior of the church (Figure 19a).The overturning mechanism of the façade is shown in Figure 19b.A vertical line of discontinuity denounces the overturning of the façade, evaluated in Figure 10, underlined also by the diagonal cracks in the lateral walls.

Discussion
The multidisciplinary approach, exemplified in this article, for the assessment of the state of damage has provided a more exhaustive vision, not limited to the building scale.Several aspects have been considered, as summarized in Figure 1.Information on the seismogenic source systems, on the PGA accelerations, on the localization of the epicenters, archived data of the historical seismicity of the area, the documentation of the state of damage from the survey form used for churches and images of the state of damage have been employed.
The comparison between the two seismic episodes, the 1997 Umbria-Marche and 2016 Central Italy earthquake, shows a milder state of damage in the last one.In summary, this could be caused by: (a) a lower distance from the epicenters of the several shocks in the 1997 Umbria-Marche earthquake compared to those of the 2016 Central Italy earthquake (Figure 20); (b) the different propagation direction of waves, from the northwest (26 and 27 September) and from the south (14 October) in the 1997 earthquake, while southeast/east in the 2016 earthquake, in reference to the position of the church explained in Section 2.1 and in Figure 2c, as well as the effectiveness of the interventions done after the 1997 earthquake.

Discussion
The multidisciplinary approach, exemplified in this article, for the assessment of the state of damage has provided a more exhaustive vision, not limited to the building scale.Several aspects have been considered, as summarized in Figure 1.Information on the seismogenic source systems, on the PGA accelerations, on the localization of the epicenters, archived data of the historical seismicity of the area, the documentation of the state of damage from the survey form used for churches and images of the state of damage have been employed.
The comparison between the two seismic episodes, the 1997 Umbria-Marche and 2016 Central Italy earthquake, shows a milder state of damage in the last one.In summary, this could be caused by: (a) a lower distance from the epicenters of the several shocks in the 1997 Umbria-Marche earthquake compared to those of the 2016 Central Italy earthquake (Figure 20); (b) the different propagation direction of waves, from the northwest (26 and 27 September) and from the south (14 October) in the 1997 earthquake, while southeast/east in the 2016 earthquake, in reference to the position of the church explained in Section 2.1 and in Figure 2c, as well as the effectiveness of the interventions done after the 1997 earthquake.At the building scale, from the comparison of the state of damage between the two earthquakes, some considerations can be made.Observing Figure 21, the shear cracks on the plane of the façade, below the rose window are lightly marked from the recent earthquake.On the contrary, the upper portion of the gable, in contact with the roof, is in good condition, and it does not show the separation of the leaves.It seems that a good interlocking between the internal and the external leaf of the façade has been performed with the post-1997 earthquake intervention.Moreover, the substitution of the heavy concrete roof with a light timber structure (Figure 22) has contributed to reducing the mass, which during the cyclic dynamic effects of the earthquake, could have severely damaged the façade.At the building scale, from the comparison of the state of damage between the two earthquakes, some considerations can be made.Observing Figure 21, the shear cracks on the plane of the façade, below the rose window are lightly marked from the recent earthquake.At the building scale, from the comparison of the state of damage between the two earthquakes, some considerations can be made.Observing Figure 21, the shear cracks on the plane of the façade, below the rose window are lightly marked from the recent earthquake.On the contrary, the upper portion of the gable, in contact with the roof, is in good condition, and it does not show the separation of the leaves.It seems that a good interlocking between the internal and the external leaf of the façade has been performed with the post-1997 earthquake intervention.Moreover, the substitution of the heavy concrete roof with a light timber structure (Figure 22) has contributed to reducing the mass, which during the cyclic dynamic effects of the earthquake, could have severely damaged the façade.On the contrary, the upper portion of the gable, in contact with the roof, is in good condition, and it does not show the separation of the leaves.It seems that a good interlocking between the internal and the external leaf of the façade has been performed with the post-1997 earthquake intervention.Moreover, the substitution of the heavy concrete roof with a light timber structure (Figure 22) has contributed to reducing the mass, which during the cyclic dynamic effects of the earthquake, could have severely damaged the façade.The effectiveness of these interventions seems to be demonstrated, even if further studies should be done in this regard.An improvement in the connection between the timber beams and the wall is recommended.A good box-like effect of the building was observed in both events.Nevertheless, the rather high level of damage of Mechanisms 1 and 2 (Figure 10) suggests the use of tie rods at the level of the gable, in correspondence with the transversal walls.In the case of the bell tower, the tie rods helped to keep the four walls together and to limit the damage.The interventions performed on the main façade, on the lateral masonry walls and on the buttresses have provided good seismic performance.
Finally, other general considerations can be added:


In the bell tower, an additional steel structure, tying together the four walls of the bell tower, could guarantee ductility and reduce torsional movement of the upper portion.


The existing independent steel structure, used to carry the bells, has shown a satisfactory response to seismic actions and probably has reduced the damage to the masonry structure.

Conclusions
The multidisciplinary procedure for the assessment of seismic damage, presented in this work, demonstrates the advantages that such an approach can provide in terms of a more exhaustive vision of the damage.
This methodology could be applied to other churches recently affected by earthquakes in Central Italy and to other similar situations, thanks to its replicability and adaptability.Indeed, inserting churches with their own coordinates in a GIS work space, together with seismic data of an earthquake, accessible from the INGV website [65] or in general from similar authorities, the damage of the building can be studied; moreover, seismic accelerations, directions and distances can be considered.In keeping with such an approach, a knowledge of the territorial scenario could be achieved, in terms both of the vulnerability of the geographical area and of the interventions performed after seismic events.Meanwhile, this methodology for the assessment of damage could help to store and collect territorial-and building-scale information, which would be useful for future advancement of knowledge.The effectiveness of these interventions seems to be demonstrated, even if further studies should be done in this regard.An improvement in the connection between the timber beams and the wall is recommended.A good box-like effect of the building was observed in both events.Nevertheless, the rather high level of damage of Mechanisms 1 and 2 (Figure 10) suggests the use of tie rods at the level of the gable, in correspondence with the transversal walls.In the case of the bell tower, the tie rods helped to keep the four walls together and to limit the damage.The interventions performed on the main façade, on the lateral masonry walls and on the buttresses have provided good seismic performance.
Finally, other general considerations can be added:

•
In the bell tower, an additional steel structure, tying together the four walls of the bell tower, could guarantee ductility and reduce torsional movement of the upper portion.

•
The existing independent steel structure, used to carry the bells, has shown a satisfactory response to seismic actions and probably has reduced the damage to the masonry structure.

Conclusions
The multidisciplinary procedure for the assessment of seismic damage, presented in this work, demonstrates the advantages that such an approach can provide in terms of a more exhaustive vision of the damage.
This methodology could be applied to other churches recently affected by earthquakes in Central Italy and to other similar situations, thanks to its replicability and adaptability.Indeed, inserting churches with their own coordinates in a GIS work space, together with seismic data of an earthquake, accessible from the INGV website [65] or in general from similar authorities, the damage of the building can be studied; moreover, seismic accelerations, directions and distances can be considered.In keeping with such an approach, a knowledge of the territorial scenario could be achieved, in terms both of the vulnerability of the geographical area and of the interventions performed after seismic events.Meanwhile, this methodology for the assessment of damage could help to store and collect territorialand building-scale information, which would be useful for future advancement of knowledge.

Figure 1 .
Figure 1.Diagram of the multidisciplinary procedure followed in the article.

Figure 1 .
Figure 1.Diagram of the multidisciplinary procedure followed in the article.

Buildings 2018, 8 , 23 Figure 1 .
Figure 1.Diagram of the multidisciplinary procedure followed in the article.

Figure 4 .
Figure 4. (a) Marche seismic hazard map.Serravalle di Chienti's value of peak ground acceleration (PGA) (% g) is pointed out [43].(b) The histogram represents in the x axis the year of occurrence of the earthquake episode and on the y axis the intensity value [44].

Figure 4 .
Figure 4. (a) Marche seismic hazard map.Serravalle di Chienti's value of peak ground acceleration (PGA) (% g) is pointed out [43].(b) The histogram represents in the x axis the year of occurrence of the earthquake episode and on the y axis the intensity value [44].

Figure 4 .
Figure 4. (a) Marche seismic hazard map.Serravalle di Chienti's value of peak ground acceleration (PGA) (% g) is pointed out [43].(b) The histogram represents in the x axis the year of occurrence of the earthquake episode and on the y axis the intensity value [44].

Figure 5 .
Figure 5. Geometrical survey of the church [47].(a) North elevation of St. Salvatore church; (b) principal (west) façade of St. Salvatore church; (c) south elevation of St. Salvatore church; (d) east elevation of St. Salvatore church.

Figure 5 .
Figure 5. Geometrical survey of the church [47].(a) North elevation of St. Salvatore church; (b) principal (west) façade of St. Salvatore church; (c) south elevation of St. Salvatore church; (d) east elevation of St. Salvatore church.

Figure 6 .
Figure 6.Geometrical survey of the church [47].(a) Entrance-level plan of St. Salvatore church; (b) choir-level plan of St. Salvatore church; (c) plan at the bell tower opening level; (d) roof plan.

Figure 7 .
Figure 7. (a) Part of the investigated façade; (b) redrawn of a portion of the wall of the façade of the church (2.00 × 2.00 m).

Figure 6 .
Figure 6.Geometrical survey of the church [47].(a) Entrance-level plan of St. Salvatore church; (b) choir-level plan of St. Salvatore church; (c) plan at the bell tower opening level; (d) roof plan.

Figure 6 .
Figure 6.Geometrical survey of the church [47].(a) Entrance-level plan of St. Salvatore church; (b) choir-level plan of St. Salvatore church; (c) plan at the bell tower opening level; (d) roof plan.

Figure 7 .
Figure 7. (a) Part of the investigated façade; (b) redrawn of a portion of the wall of the façade of the church (2.00 × 2.00 m).

Figure 7 .
Figure 7. (a) Part of the investigated façade; (b) redrawn of a portion of the wall of the façade of the church (2.00 × 2.00 m).

Figure 8 .
Figure 8.(a) Quality of mortar, form of resistant elements, dimensions of elements, quality of resistant elements; (b) presence of cross blocks [49]; (c) splitting of vertical joints; (d) presence of horizontal rows.

Figure 8 .
Figure 8.(a) Quality of mortar, form of resistant elements, dimensions of elements, quality of resistant elements; (b) presence of cross blocks [49]; (c) splitting of vertical joints; (d) presence of horizontal rows.

Figure 10 .
Figure 10.Activated mechanisms and level of damage.Source data: [52].(a) Overturning of the façade; (b) mechanisms of the upper part of the façade; (c) in-plane mechanisms of the façade; (d) longitudinal response of the lateral wall, in plane; (e) shear mechanism of the aps; (f) vaults of the aps; (g) bell tower; (h) upper portion of the bell tower.

Figure 11 .
Figure 11.The 1997 major shocks with the location of St. Salvatore church (green square) in Acquapagana and the seismogenic fault system of Colfiorito (purple area).(a) The red star points out the epicenter of the 26 September 1997 event at 0:33 a.m.[53]; (b) the blue star specifies the epicenter of the 26 September 1997 event at 9:40 a.m.[54]; (c) the yellow star indicates the epicenter of the 14 October 1997 event [55].

Figure 11 .
Figure 11.The 1997 major shocks with the location of St. Salvatore church (green square) in Acquapagana and the seismogenic fault system of Colfiorito (purple area).(a) The red star points out the epicenter of the 26 September 1997 event at 0:33 a.m.[53]; (b) the blue star specifies the epicenter of the 26 September 1997 event at 9:40 a.m.[54]; (c) the yellow star indicates the epicenter of the 14 October 1997 event [55].

Figure 11 .
Figure 11.The 1997 major shocks with the location of St. Salvatore church (green square) in Acquapagana and the seismogenic fault system of Colfiorito (purple area).(a) The red star points out the epicenter of the 26 September 1997 event at 0:33 a.m.[53]; (b) the blue star specifies the epicenter of the 26 September 1997 event at 9:40 a.m.[54]; (c) the yellow star indicates the epicenter of the 14 October 1997 event [55].

Figure 13 .
Figure 13.Post-1997 earthquake damage of the exterior of the church.Photo source [56].(a) Collapse of the upper part of the buttresses and of the lateral walls; (b) partial collapse of the north wall of St. Salvatore church; (c) damaged gable, left portion of the upper side of the façade, shear crack in the central part under the rose window and collapse of the upper portion of the buttresses of the south side.

Buildings 2018, 8 , 23 Figure 13 .
Figure 13.Post-1997 earthquake damage of the exterior of the church.Photo source [56].(a) Collapse of the upper part of the buttresses and of the lateral walls; (b) partial collapse of the north wall of St. Salvatore church; (c) damaged gable, left portion of the upper side of the façade, shear crack in the central part under the rose window and collapse of the upper portion of the buttresses of the south side.

Figure 14 .
Figure 14.Post-1997 earthquake damage of the interior of the church.Photo source: [57].(a) View of the triumphal arches towards the apse area.Diffuse shear cracks and pounding damage.(b) Inside view of St. Salvatore church.Partial collapse of the wall and of part of the internal leaf of the wall of the main façade at the tympanum level.

Figure 14 .
Figure 14.Post-1997 earthquake damage of the interior of the church.Photo source: [57].(a) View of the triumphal arches towards the apse area.Diffuse shear cracks and pounding damage.(b) Inside view of St. Salvatore church.Partial collapse of the wall and of part of the internal leaf of the wall of the main façade at the tympanum level.

Figure 16 .
Figure 16.Exterior post-2016 earthquake damage.(a) Shear cracks along the height of the bell tower [47]; (b) damaged bell tower with provisional interventions; (c) torsional and shear cracks [47]; (d) diagonal and vertical crack in the annexed building in contact with the buttress of the north wall of the nave [47].The state of damage observed outside has a correspondence with the interior part of the bell tower, where the cracks cross the different levels of the structure (Figures 16c and 17a, b).

Figure 17 .
Figure 17.Interior state of damage of the bell tower.(a) Shear cracks of the ground floor [47]; (b) continuous part of the shear crack of the first floor [47]; (c) shear crack on the north side [47].

Figure 16 .
Figure 16.Exterior post-2016 earthquake damage.(a) Shear cracks along the height of the bell tower [47]; (b) damaged bell tower with provisional interventions; (c) torsional and shear cracks [47]; (d) diagonal and vertical crack in the annexed building in contact with the buttress of the north wall of the nave [47].

Figure 16 .
Figure 16.Exterior post-2016 earthquake damage.(a) Shear cracks along the height of the bell tower [47]; (b) damaged bell tower with provisional interventions; (c) torsional and shear cracks [47]; (d) diagonal and vertical crack in the annexed building in contact with the buttress of the north wall of the nave [47].

Figure 17 .
Figure 17.Interior state of damage of the bell tower.(a) Shear cracks of the ground floor [47]; (b) continuous part of the shear crack of the first floor [47]; (c) shear crack on the north side [47].

Figure 17 .
Figure 17.Interior state of damage of the bell tower.(a) Shear cracks of the ground floor [47]; (b) continuous part of the shear crack of the first floor [47]; (c) shear crack on the north side [47].

Buildings 2018, 8 ,
x FOR PEER REVIEW 17 of 23 damage are present, retracing the discontinuity of the masonry due to the post-1997 earthquake reconstruction of the gable of the façade.

Figure 18 .
Figure 18.Post-2016 earthquake damage of the external façade with the damage pattern.

Figure 19 .
Figure 19.Post-2016 earthquake damage of the inside view of the west wall of the church.(a) Cracks at the intersection between the beams of the roof and the facade[47]; (b) vertical crack between the façade and the lateral wall where there are also diagonal cracks[47].

Figure 18 .
Figure 18.Post-2016 earthquake damage of the external façade with the damage pattern.

Buildings 2018, 8 ,
x FOR PEER REVIEW 17 of 23 damage are present, retracing the discontinuity of the masonry due to the post-1997 earthquake reconstruction of the gable of the façade.

Figure 18 .
Figure 18.Post-2016 earthquake damage of the external façade with the damage pattern.

Figure 19 .
Figure 19.Post-2016 earthquake damage of the inside view of the west wall of the church.(a) Cracks at the intersection between the beams of the roof and the facade [47]; (b) vertical crack between the façade and the lateral wall where there are also diagonal cracks [47].

Figure 19 .
Figure 19.Post-2016 earthquake damage of the inside view of the west wall of the church.(a) Cracks at the intersection between the beams of the roof and the facade [47]; (b) vertical crack between the façade and the lateral wall where there are also diagonal cracks [47].

Figure 21 .
Figure 21.Comparison of the state of damage on the façade after the 1997 earthquake (on the left) and the 2016 one (on the right).

Figure 21 .
Figure 21.Comparison of the state of damage on the façade after the 1997 earthquake (on the left) and the 2016 one (on the right).

Figure 21 .
Figure 21.Comparison of the state of damage on the façade after the 1997 earthquake (on the left) and the 2016 one (on the right).

Figure 22 .
Figure 22.Comparison of the types of roof and the relative state of damage after the 1997 earthquake and the 2016 one.

Figure 22 .
Figure 22.Comparison of the types of roof and the relative state of damage after the 1997 earthquake and the 2016 one.

Table 1 .
Table of major seismic events for the Serravalle di Chienti area.Data source [44].

Table 2 .
[49] quality index evaluation.Source of the axonometry and section images:[49]. of two facades built with squared blocks of stones.with the interposition of a pebble filling.Resistant elements with variable dimensions, but with defined edges.Wall texture which respects horizontal lines, partial offsets of vertical joints and presence of transversal stone joints.
MASONRY BLOCKS OF SQUARED STONE AND PEBBLES, TRIPLE FACADE B = average masonry quality; C = insufficient behavior).Regarding St. Salvatore church, stone elements do not exhibit degradation phenomena, have a re PHOTOS Front Corner Inner masonry B = average masonry quality; C = insufficient behavior).Regarding St. Salvatore church, stone elements do not exhibit degradation phenomena, have a re PHOTOS Front Corner Inner masonry B = average masonry quality; C = insufficient behavior).Regarding St. Salvatore church, stone elements do not exhibit degradation phenomena, have a re Buildings 2018, 8, x FOR PEER REVIEW 12 of 23 both cases, even if at different levels, as is clearly visible looking at the infographic documentation.