ASSESSMENT OF THE EARTHQUAKE VULNERABILITY OF MULTI-RESIDENTIAL BUILDINGS IN SLOVENIA OCENA POTRESNE OGROŽENOSTI VECSTANOVANJSKIH ZGRADB V SLOVENIJI Vojko Kilar, Domen Kusar This multi-residential building was severely damaged during the earthquake in Koaceli, Turkey, in 2002. V potresu leta 2002 po{kodovana ve~stanovanjska hi{a v kraju Koaceli vTur~iji Assessment of the earthquake vulnerability of multi-residential buildings in Slovenia DOI: 10.3986/AGS49103 UDC: 728.22:550.34(497.4) 699.841(497.4) COBISS: 1.01 ABSTRACT: The paper assesses the earthquake vulnerability of multi-residential buildings in Slovenia, although it is limited to the buildings that were built before 1981, in the time when the earthquake building codes were much less elaborated than today. In the paper, based on the building completion year, buildings are classified into different time periods, which are characterized by important historical events and bigger changes in earthquake building codes. The assessment of earthquake vulnerability is based on the data from the building completion year, number of storeys, prevailing structural material and the year of the last renovation as obtained from the last census of the population, households and apartments completed in 2002. The result is an estimation of the earthquake vulnerability of a building, also because for very similar buildings the earthquake resistance depends on the architectural design of a building, the amount and layout of its structural elements, foundations, soil profile and other influences. In the first part of the paper the multi-residential buildings are divided by age, material and number of storeys. In the second part, the assessment of the earthquake vulnerability of these buildings is divided into three classes: a) probably earthquake unsafe, b) probably earthquake less safe and c) probably earthquake safe. Additionally the earthquake vulnerability assessment is presented geographically by showing the earthquake less safe and unsafe buildings on the maps and charts for different communities in Slovenia. It has been concluded that the earthquake safety of many multi-residential buildings in Slovenia might be questionable, while we have detected also bigger differences between communities. KEY WORDS: architecture, structural engineering, geography, earthquake safety, earthquake vulnerability, earthquake unsafe buildings, multi-residential buildings, Slovenia The article was submitted for publication on April 20, 2009. ADDRESSES: Vojko Kilar, Ph.D. Faculty of architecture University of Ljubljana Zoisova cesta 12, SI - 1000 Ljubljana, Slovenia e-mail: vojko.kilar@fa.uni-lj.si Domen Kusar, Ph. D. Faculty of architecture University of Ljubljana Zoisova cesta 12, SI - 1000 Ljubljana, Slovenia e-mail: domen.kusar@fa.uni-lj.si Contents 1 Introduction 92 2 Methodology 92 3 Characteristics of multi-residential building 93 4 Development of earthquake building codes and building time periods 95 4.1 The period before 1894 (before the Ljubljana earthquake in 1895) 95 4.2 The period from 1895 to 1945 (before World War I and between the World Wars) 96 4.3 Period from 1946 to 1963 (the period soon after the Second World War) 96 4.4 Period from 1964 to 1981 (after the Skopje earthquake) 97 5 Criteria for earthquake vulnerability assessment of multi-residential buildings 97 5.1 Probably earthquake unsafe buildings 98 5.2 Probably earthquake less safe buildings 98 6 Conclusion 104 7 References 105 1 Introduction In Slovenia the majority of the population, especially in cities, lives in multi-residential buildings. These buildings are mainly multi-storey apartment blocks and skyscrapers built in the previous century, especially after the end of World War II, when the erection of such structures became more popular. The results of the census of the population, households and apartments performed by Statistical Bureau of Republic of Slovenia in 2002, show that at that time there were 18,005 multi-residential buildings in Slovenia, which represented only 3.9% of all residential buildings. However, in multi-residential buildings there were 242,011 apartments, which was almost one third of all apartments in the country. The area of all apartments in multi-residential buildings amounted to 13,491,714m2 which is 23.2% of the total area of all apartments in Slovenia. Before 1981, there were built 14,744 multi-residential buildings with 185,994 apartments with an area of 10,253,913 m2, which was as much as 76% of the total area in all multi-residential buildings. The first part of the paper describes characteristic building time periods, which were characterized by important historical events and major developments in earthquake building codes. Based on the year of the erection of a building we can make a judgement about its current condition and earthquake resistance. We have limited our study to the buildings built before 1981, because after the Montenegro earthquake (1979) the new Yugoslavian earthquake building codes, which assured much better earthquake safety, were introduced in 1981. By statistical analysis, all buildings were classified by age, material and number of storeys. Considering relatively logical assumption that the builders had followed valid building codes, we can estimate the level of earthquake resistance of the existing building by comparison of the code's requirements valid at the time of erection with the requirements that are valid today. The classification in the older group does not automatically mean a lower level of earthquake resistance, since in some older periods, building of certain types of multi-residential buildings was better than in the periods that followed. For example, building quality of low-rise masonry and mixed multi-residential buildings in the first half of the 20th century taking into consideration Austrian codes and regulations were relatively good. On the contrary, building quality in the first year of the development of socialism after World War II was generally much worse and consequently most of the higher buildings from that period were probably earthquake less safe. The main contribution of this paper is the preparation of the criteria for the assessment of the earthquake vulnerability of multi-residential buildings built before 1981 and their division into 'probably earthquake unsafe', 'probably earthquake less safe' and 'probably earthquake safe' categories. It should be stressed, that on the basis of census data, only a general estimation of the earthquake vulnerability and the necessity for an earthquake retrofit can be given. A more accurate evaluation can only be obtained by detailed inspection and analysis of individual buildings or group of buildings in a smaller region and its generalization to the area of the whole of Slovenia. The obtained results for the state and regional level show share of probably earthquake unsafe and probably earthquake less safe building for all Slovene communities. Alas the application of higher safety standards remains problematic, since it depends mainly on financial resources, which are getting even more complicated in the new proprietorial relations recently formed in many multi-residential buildings. 2 Methodology Earthquake resistance of buildings is normally determined by studying the building plans and performing static calculations and analyses for each individual building (Tomaževic 1987 and 1998; Fajfar et al 2000; Dujič, Žarnic 2008; Bosilkov et al 2008). This method, however accurate, might be expensive or even unworkable, if we are trying to asses earthquake vulnerability of a larger area, region or even of the whole country. Several similar earthquake vulnerability assessments have already been made for Ljubljana and some other Slovenian cities and municipalities (Orožen Adamič 1995; Orožen Adamič and Perko 1996; Kilar 2004). In our study we have used the results of the last census of the population, households and apartments (Popis... 2002), which also contains some data about the buildings within which the apartments are located (Zupančič et al 2003). Data on year of completion, number of storeys, type of structural system and prevailing material of the load bearing system of the building, as well as the year of the last renovation of the apartment, purpose of use and number of apartments in a building are available. Assuming that the builders had followed the building code valid at the time of building, we can compare the code's demands with the demands that are valid today and estimate the number of apartments in »probably earthquake unsafe« buildings, number of apartments in »probably earthquake less safe« buildings and number of apartments in buildings which can be considered as earthquake safe (Kilar 2004). In this context the formulation »probably earthquake unsafe« building stands for a building which could have been dangerous during an earthquake and could be damaged beyond the repair limit (or could even endanger human lives), however, it is not necessary that this would actually happen since the majority of buildings have certain additional strength, which is the consequence of the amount and the disposition of load bearing elements (Tomaževic 1987; Kilar and Koren 2009), their interconnections, solid building according to sound engineering principles, fulfilment of minimal requirements, general work quality and quality of details (Žarnic 2005), as well as soil quality and other influences (Orožen Adamic and Hrvatin 2001; Slak and Kilar 2005). For these reasons buildings with the same height and age, built of the same material on similar ground, are not necessarily equally vulnerable during the same earthquake. Because of this additional strength many buildings survive an earthquake even if they are not built according to any earthquake building code. Exact determination of additional strength is only possible by structural analysis of each individual building and it cannot be included in this general assessment of earthquake vulnerability, based on a limited statistical input parameters. It is therefore necessary to interpret the results about »probably earthquake unsafe« buildings with some caution, understanding that these buildings could be dangerous during an earthquake, however, it is not certain that they would be destroyed or that they would cause danger to human lives. 3 Characteristics of multi-residential building The2002 census of the population, households and apartments (Popis...2002) classified multi-residential buildings as apartment blocks, skyscrapers or older municipal multi-storey buildings, which are built one next to another and do not look like a modern multi-residential building. All considered multi-residential buildings have certain common characteristics, mostly related to building material and plan layout. Most of the buildings are made as a combination of reinforced concrete elements and shear masonry walls. The reinforced concrete is mostly used for ceilings, staircases and beams, while shear walls are mostly made of masonry units or prefabricated concrete panels. In older buildings, ceilings are made of wooden beams, while walls are built of masonry without any concrete confinement elements. Concrete became more popular for building walls after World War II. Wood as structural material has been mainly used for roofing, while wooden multi-residential buildings were only exceptional in the considered time period. Similarly, use of steel frame structures for multi-residential buildings had not been popular until the last decade. Most Slovene multi-residential buildings have an elevated ground floor with the main entrance followed by a small entrance hall (windbreak) with letter boxes. This space is usually separated from the main communications inside the buildings. Typically the communication corridors and staircases are positioned in the centre of the building, while the apartments are arranged on the perimeter of the building. In the higher buildings, one or two elevators are positioned next to the main communication shafts. Such an arrangement enables better illumination of the apartments. Due to the rationality of communal pipelines, plan layout remains practically the same on all storeys. Three types of multi-residential buildings are most common in Slovenia: multi-storey houses, skyscrapers and apartment blocks. Each type has its own characteristics; nevertheless the multi-storey house and apartment block seem to be structurally very similar to each other. A typical skyscraper has a rectangular or even square floor plan shape and ranges in height from 10 to 12 storeys. The apartment block is usually elongated in one direction and includes more than one communication shaft. It usually has a ground floor and three or four storeys, because for all buildings higher than four storeys the elevators were mandatory. Some apartment blocks have vertical communications in the centre, while in others they are positioned closer to the side with one wall on the perimeter of the building. In this way natural illumination of the stairs has been made possible during the daytime. Storey height in older buildings is about 3m, respecting the standards valid at that time (Building law, 1931) which required aminimal height of Figure 1: Share of multi-residential buildings in Slovene communities in2002 (100% = all residential buildings in the community). p p. 94 Percentage/delež 10.00% and more/in več H 7.50- 9.99% 5.00- 7.49% 2.50-4.99% Till/do 2.49% Statistical confidentialy/ statistična zaupnost 50.0 km 1 Authors of contents/avtorja vsebine: Vojko Kilar, Domen Kušar Author of map/avtor zemljevida: Domen Kušar Sources/viri: Statistični urad Republike Slovenije, 2002; Geodetska uprava Republike Slovenije, 2008 2.8 m. Newer buildings have lower storey height; nowadays the required height amounts to 2.5 m while the height is usually higher. We can conclude that height of older multi-residential buildings with a ground floor and four storeys is about 15 or 16 m and height of an average older skyscraper with twelve storeys around 36 m. Usually, multi-residential buildings from the considered period do not have parking areas under the building, the basements are usually used for storage and maintenance. The garages are arranged next to the building or separately. Newer buildings have more basement floors including parking areas. Most multi-residential buildings are in municipal communities with a developed industry. Many of such buildings can be found on the Slovenian coast and in the north part of Slovenia (Figure 1). The highest percentage was recorded in Trbovlje (14%), Maribor (13%), Ljubljana (13%) and Mežica (13%), while the largest number of such buildings can be found in Ljubljana (4291), Maribor (2094), Celje (682), Koper (569) and Kranj (528). 4 Development of earthquake building codes and building time periods The development of earthquake building codes has been gradual; they were usually extended and made stricter after every strong earthquake. The first code that included earthquake loading as a separate loading case was the Temporary technical code (Privremeni tehnički propisi - PTP), which was issued in the Federative Republic of Yugoslavia back in 1948. According to the overview of the code development in Slovenia in the past one hundred years we can establish the characteristic building time periods, which significantly differ from what was at that time the valid building code and requirements for earthquake resistant design (Bubnov et al. 1982; Bubnov 1996; Kilar 2004; Slak and Kilar 2005; Žarnic 2005). In this article we have determined four characteristic building time periods before 1981 taking into consideration the historical bench-marks and time of implementation of different building codes, Tables 1 to 4 present a review of the apartment area in buildings built from different materials. Total apartment area in the corresponding building time period is also given together with share percentages so the exact values can be calculated. The Statistical Bureau of the Republic of Slovenia strictly considers the provisions about data confidentiality required by the law, so all values smaller than 5 are not published. However, total sums of these low data values are included, so the sum of individual percentages is lower than 100%. 4.1 The period before 1894 (before the Ljubljana earthquake in 1895) In this period, the earthquake resistance was mainly achieved by experience, such as by reducing the building height, increasing the wall thicknesses in lower storeys, lowering the mass centre of the building, etc. Some buildings that were built in this period might have already reached their life-time limit and should have probably been renovated or demolished. Historically protected buildings need a special approach in this manner. The area of multi-residential buildings from this period amounts to 6.5% of all multi-residential buildings in Slovenia (13,491,714 m2). They are made of masonry (41.1%), combined materials (20.6%), stone (17.9%) and concrete (16.8%). The division of multi-residential buildings based on material and number Table 1: Division of the area of multi-residential buildings built before 1894, according to the prevailing structural material and number of storeys. Total area of reviewed buildings in this time period is 874,993 m2. Single From 1 to 3 storeys 4 storeys From 5 to 8 storeys 9 and more storeys Masonry 0.50 35.41 2.83 1.73 0.63 Concrete 0.48 4.82 4.75 6.77 Combination of different materials 0.46 17.41 2.00 0.50 0.22 Wood 0.25 Stone 0.56 16.60 0.75 Other 0.01 0.06 0.02 of storeys is presented in Table 1. Most of the apartment area (70.2%) is in one to three-storeys buildings, while 14.6% of apartments can be found in higher buildings with five or more storeys. According to census data 66,500 m2 of buildings with nine or more storeys were built in this period, which does not seem very likely, since such buildings from this period do not exist excluding belfries and towers. It is also possible that there were some errors or inconsistencies during the census. 4.2 The period from 1895 to 1945 (before World War I and between the World Wars) Buildings were built according to Austrian and old Yugoslavian building codes in this period, which prescribed the thickness of the masonry walls for different storeys, width of walls between windows, procedures for fabrication of ceilings, fire walls and massive floor plates (Gradbeni zakon 1931). For horizontal loading only the wind loading was considered. Most buildings from this period are solidly built, relatively regular in plan and elevation with prescribed details and carefully selected materials. They had started to use reinforced concrete in this period to build the first higher building and sky scrapers which react to earthquake loading completely differently from rigid masonry buildings from previous centuries. The most well known example in Slovenia is the Ljubljana skyscraper from 1933. Area of multi-residential buildings from this period amounts to 9.2% of all multi-residential buildings in Slovenia. They are mostly made of masonry (67.9%) and combined materials (18.0%). The division of multi-residential buildings based on material and number of storeys is presented in Table 2. Most of the apartment area (76.9%) is in one- to three-storey buildings, while 8.6% of apartments are in buildings with five or more storeys. Table 2: The division of area of multi-residential buildings built from 1895 to 1945, according to the prevailing structural material and number of storeys. Total area of the reviewed buildings in this time period is 1,247,127 m2. Single From 1 to 3 storeys 4 storeys From 5 to 8 storeys 9 and more storeys Masonry 1.40 55.16 6.71 4.17 0.51 Concrete 2.20 2.08 1.61 1.35 Combination of different materials 0.63 14.85 1.69 0.84 0.06 Wood 0.06 0.14 Stone 0.05 3.89 Other 0.12 4.3 Period from 1946 to 1963 (the period soon after the second world war) Most of the buildings from this period were built according to the first Yugoslavian codes for imposed building loads (PTP - Privremeni tehnički propisi, 1948). Yugoslavia was divided into three earthquake zones according to this code: • Zone a) of smaller damage, • Zone b) of bigger damage and • Zone c) of possible catastrophic destruction. According to this code maximum earthquake force for Zone (c) amounted to 3% of the dead load and half of the live load. These values are up to five to ten times smaller than the forces used in modern standards. For this period of growing socialism the quality of building was generally not very high. This code was valid until 1963. Area of multi-residential buildings from this period amounts to 22.6% of all multi-residential buildings in Slovenia. They are mostly made of masonry (62.9%), concrete (21.5%) and combined materials (13.8%). The division of multi-residential buildings based on material and number of storeys is presented in Table 3. Most of the apartment area (45.6%) is in one to three-storeys buildings, while 29.1% of apartments are in buildings with five or more storeys. Table 3: Division of area of multi-residential buildings built from 1946 and 1963, according to the prevailing structural material and number of storeys. The total area of the reviewed buildings in this time period is 3,053,960 m2. Single From 1 to 3 storeys 4 storeys From 5 to 8 storeys 9 and more storeys Masonry 0.62 33.97 14.47 11.53 2.17 Concrete 0.01 4.46 5.78 6.87 4.41 Combination of different materials 0.06 6.27 3.83 2.81 0.82 Wood 0.03 0.18 Stone 0.03 0.38 0.02 Other 0.28 0.45 0.47 4.4 Period from 1964 to 1981 (after the Skopje earthquake) After the catastrophic earthquake in Skopje in 1963, a new earthquake building code was introduced in 1964. This code significantly increased earthquake forces for all types of buildings, prescribed distribution of horizontal forces over the height of a building and included influence of soil quality on the determination of horizontal forces. Also, in the same year, a new seismic hazard map of Slovenia was issued, which presented the division of earthquake prone areas in Slovenia much better. Code requirements for building masonry buildings in earthquake prone areas were completely changed. For the first time vertical reinforced concrete confinement elements at the corners and at the junctions of masonry walls were prescribed. The new code also improved building of concrete structures by prescribing the reinforcement details such as shape and distance between stirrups, overlapping of reinforcing bars, anchorage. Nevertheless, quality of the prescribed details was still much lower than in present codes. In general, earthquake resistance of buildings built in this period was higher than for older buildings. This code was again critically analyzed after the Montenegro earthquake in 1979, resulting in a new Yugoslavian earthquake building code which was issued 1981. Table 4: The division of the area of multi-residential buildings built from 1964 and 1981, according to the prevailing structural material and number of storeys. The total area of the reviewed buildings in this time period is 5,478,708 m2. Single From 1 to 3 storeys 4 storeys From 5 to 8 storeys 9 and more storeys Masonry 0.09 5.50 6.16 3.80 1.17 Concrete 5.90 21.80 21.41 24.56 Combination of different materials 2.01 3.23 2.49 0.43 Wood 0.07 0.03 0.05 Stone 0.11 Other 0.42 Area of the multi-residential buildings from this period amounts to 40.6% of all multi-residential buildings in Slovenia. The majority of them are built of concrete (73.7%), much less of masonry (16.7%) and of combined materials (8.2%). The division of the multi-residential buildings based on material and number of storeys is presented in Table 4. In this period approximately one third of apartments (31.6%) were built in four storey multi-residential buildings, while more than half of the apartments from this period (53.9%) are in buildings with five or more storeys. 5 Criteria for earthquake vulnerability assessment of multi-residential buildings The data collected by the 2002 census of population, households and apartments are unfortunately not complete enough to analytically evaluate the earthquake resistance of a building. It is however possible to make a general assessment of earthquake vulnerability based on year of building completion, prevailing material of the structural system, number of storeys and year of the eventual renovation. We also wanted to include actual earthquake hazard as an influencing parameter, as it is shown on the seismic hazard map Figure 2: Seismic hazard map of Slovene communities based on the seismic hazard map of Slovenia including the proposed weight factor (CEN (2004): Eurocode 8: Design of structures for earthquake resistance - Part 1: General rules, seismic actions and rules for buildings, EN 1998-1). P p. 99 of expected ground accelerations in Slovenia (CEN (2004): Eurocode 8: Design of structures for earthquake resistance - Part 1: General rules, seismic actions and rules for buildings, EN 1998-1). The seismic hazard map divides municipalities into areas with different expected ground acceleration (Figure 2). We anticipated that the probability that a building (which was not built according to modern standards) is earthquake unsafe, is much higher in the communities with larger than expected ground acceleration. Influence of the expected ground acceleration has been considered by a weight coefficient. Its value is 1.0 for areas with the highest ground acceleration (> 0.25 g), and smaller than 1.0 in all other areas. If a specific area is not prone to earthquakes (e. g. North West of Slovenia) the coefficient amounts to 0.4. We therefore simply considered that the weight coefficient is linearly proportional to the design ground acceleration (i. e. earthquake forces) in our analysis. Actually the relation between damage and ground acceleration is a non-linear one as it also depends on dissipated energy and other parameters, such as soil quality and distance from epicentre. Chapter 5 presents the criteria for the division of concrete, masonry and combined buildings in three classes with different level of earthquake vulnerability which were derived from the demands and descriptions of characteristic time periods in Chapter 4. 5.1 Probably earthquake unsafe buildings These buildings have a higher probability to be earthquake unsafe and it is very likely that their structural system should be seismically retrofitted. Masonry buildings: • Buildings with five or more storeys built before 1981 (too high in respect to current codes, which for such buildings require the usage of reinforced masonry, but only in the areas with lower seismicity. • Buildings with four or less storeys built before 1964 (there were no reinforced concrete vertical confinement elements used, probably also no horizontal confinement connections on top of the walls were built in). All newer building codes require vertical and horizontal reinforcing ties, which should be concreted at the prescribed distances after building the masonry walls. They bind the entire building in a homogenous unit and increase its strength (Tomaževic 1987; Slak and Kilar 2005). Combined buildings: • Buildings built before 1894 and between the years 1946 and 1963 (too small earthquake design loads, in combination with masonry it is very likely that vertical and horizontal reinforced concrete confining elements are missing). Concrete buildings: • Buildings with four or more storeys built before 1894 and between the years 1946 and 1963 (much smaller earthquake design loads, in some cases only concrete blocks without any reinforcement were used). 5.2 Probably earthquake less safe buildings These buildings are probably earthquake less safe and it is possible that their structural resistance to horizontal loads needs some improvement. Masonry buildings: • Buildings with five or more storeys built between 1982 and 1999 (too high in respect to Eurocode 8, which requires the use of reinforced masonry in these cases). Our investigation shows that there are not many such buildings in Slovenia. Since our research is limited to buildings built before 1981, this group of buildings was not included. Combined buildings: • Buildings built between 1894 and 1945 (too small earthquake design forces, probably higher building quality than in the period before and after that). Weight Design ground koeficient/ acceleration/projektni utežni pospešek tal [g] koeficient 0.250 1.0 0.225 0.9 0.200 0.8 0.175 0.7 0.150 0.6 0.125 0.5 0.100 0.4 50.0 km Authors of contents/avtorja vsebine: Vojko Kilar, Domen Kušar Author of map /avtor zemljevida: Domen Kušar Sources/viri: ARSO. Urad za seizmologijo in geologijo, 2001; Geodetska uprava Republike Slovenije, 2008 s s "o C g ^ T3 3 %-908070605040302010- Combination of different materials/kombinacija Masonry/opeka različnih materialov (37,9%) (12,3%) Probably earthquake safe/ verjetno potresno varna Probably earthquake less safe/ verjetno potresno manj varna Probably earthquake unsafe/ verjetno potresno nevarna Percentage of area multi residental houses/ delež glede na površino večstanovanjskih stavb 0 Figure 3: Earthquake vulnerability of multi-residential buildings made from different building materials (apartment area in %). • Buildings with four or more storeys built between 1964 and 1981 (too small earthquake design forces, the details/connections/stirrups of concrete parts of these structures prescribed at that time are disputable from the modern code point of view). Concrete buildings: • Buildings with four or more storeys built between 1894 and 1945 (too small earthquake design forces but probably higher building quality). • Buildings with five or more storeys built between 1964 and 1981 (too small earthquake design forces, the prescribed details/connections/stirrups for concrete structures are disputable from the modern code point of view). Table 5: Earthquake vulnerability of multi-residential buildings in Slovenia built before 1981 Buildings Apartments Number Percent (%) area (m2) Percent (%) Number Percent (%) Probably earthquake safe 4,545 30.8 2,657,215 25.9 49,543 26.6 Probably earthquake less safe 2,287 15.5 3,383,236 33.0 61,043 32.9 Probably earthquake unsafe 7,912 53.7 4,213,462 41.1 75,408 40.5 Total 14,744 100.0 10,253,913 100.0 185,994 100.0 The results show that according to the selected criteria 41.1% of apartment's area are in probably earthquake unsafe and 33.0% in probably earthquake less safe buildings (see Table 5 and Figure 3). The results for different communities are presented in Figures 4 and 5 and in Tables 6 and 7. Figures 4 and 5 present the area of apartments in earthquake unsafe and less safe buildings geographically as a percentage of the total area in multi-residential buildings in proper municipality. The municipalities with the higher share of earthquake vulnerable apartment areas are also listed in Tables 6 and 7. It is not surprising that most municipalities stand out, even if they do not seem to be critical by share of multi-residential buildings presented in Figure 1. Figure 4: Area of apartments in earthquake unsafe buildings as a share of total multi-residential buildings area in this community. p p. 101 Percentage/delež .0% and more/in več 70.0-79.9% 60.0-79.9% 50.0-59.9% 40.0-49.9% 30.0-39.9% 20.0-29.9% less than/do 19.9% statistical confidentiality/ statistična zaupnost 50.0 km i Authors of contents/avtorja vsebine: Vojko Kilar, Domen Kušar Author of map/avtor zemljevida: Domen Kušar Sources/viri: Statistični urad Republike Slovenije, 2002; Geodetska uprava Republike Slovenije, 2008 Table 6: Communities with a higher percentage of probably earthquake unsafe multi-residential buildings (communities with a design ground acceleration ag > 0.175 are marked in grey). Municipality Area (m2) Earthquake intensity weight factor Ljubljana 1,223,045 1 Maribor 655,300 0.4 Celje 215,478 0.6 Kranj Trhnvl ip 128,537 122 080 0.9 0.6 11 UUVIJC Jesenice 1 tltljUOU 104,413 u.u 0.7 Koper 91,964 0.5 Velenje 91,059 0.5 Nova Gorica Ptuj 69,492 64,928 0.7 0.5 Novo mesto 61,675 0.7 Postojna Raunp na Knrnškpm 52,875 49,261 0.8 0.5 navi ic i id i\ui Uoi\ciu Kamnik 43,053 0.9 Piran 42,720 0.4 Hrastnik 39,555 0.6 Tržič 39,217 0.7 Krško Kočevje 37,224 36.480 0.8 0.6 l\UOGVJG Idrija 34,649 0.8 Izola 34,169 0.4 Žalec 28,495 0.6 Domžale 28,230 1 Skofja Loka 27,372 0.9 Slovenska Bistrica 27,262 0.5 Table 7: Municipalities with a higher share of probably earthquake less safe multi-residential buildings (municipalities with a design ground acceleration ag > 0.175 are marked in grey). Municipality Area (m2) Earthquake intensity weight factor Ljubljana 1,223,045 1 Maribor 655,300 0.4 Celje 215,478 0.6 Kranj Trhnvl ip 128,537 122 080 0.9 0.6 11 UU V1JC Jesenice 104,413 0.7 Koper 91,964 0.5 Velenje 91,059 0.5 Nova Gorica Ptuj 69,492 64,928 0.7 0.5 Novo mesto 61,675 0.7 Postojna Raunp na Knrnškpm 52,875 49,261 0.8 0.5 navi ic i id i\ui usmili Kamnik 43,053 0.9 Piran 42,720 0.4 Hrastnik 39,555 0.6 Tržič 39,217 0.7 Krško Kočevje 37,224 36,480 0.8 0.6 l\UOGVJG Idrija 34,649 0.8 Izola 34,169 0.4 Žalec 28,495 0.6 Domžale 28,230 1 Skofja Loka 27,372 0.9 Slovenska Bistrica 27,262 0.5 Figure 5: Area of apartments in earthquake less safe buildings as a percentage of the total multi-residential buildings area in this community. p p. 103 0.0 I— Percentage/delež 80.0% and more/in več 70.0-79.9% 60.0-69.9% 50.0-59.9% 40.0-49.9% 30.0-39.9% 20.0-29.9% Till/do 20.0% Statistical confidentialy/ statistična zaupnost 50.0 km -1 Authors of contents/avtorja vsebine: Vojko Kilar, Domen Kušar Author of map/avtor zemljevida: Domen Kušar Sources/viri: Statistični urad Republike Slovenije, 2002; Geodetska uprava Republike Slovenije, 2008 It can be seen that the number of probably earthquake unsafe buildings, for which it is very likely that their structural system should be seismically retrofitted, amounts to 53.7% of the total number of all multi-residential buildings in Slovenia. The apartment area in these buildings amounts to 4,213,462 m2 (41.1%). Most of these buildings were built of brick or other combined materials before 1963. Share of buildings which are probably earthquake less safe is smaller and amounts to 15.5%. Apartment area in these buildings is 3,383,236 m2, which amounts to approximately to one third of all multi-residential buildings in Slovenia. 6 Conclusion In the paper we tried to assess earthquake vulnerability of existing multi-residential buildings in Slovenia and confirm the concerns which have also been expressed by other experts for earthquake resistant design (see for example Kubelj 2009). Our research confirmed that the requirements of modern earthquake resistant codes are basically only fulfilled by less than one half of the existing multi-residential buildings. The difference from other similar studies, which were based on a rough estimation on the number and resistance of buildings, is that our study is based on statistical data collected by a census of population, households and apartments from 2002. It should be stressed that results obtained provide only an estimation of the number of probably earthquake unsafe and less safe buildings and the possible extent of retrofit measures. The earthquakes in north-western Slovenia showed that damage to buildings does not depend only on Table 8: Five or more storey buildings are the most earthquake-vulnerable part of the Slovenian residential fund. Number of unsafe buildings, number of apartments and apartment area in these buildings are presented for each municipality (municipalities with design ground acceleration ag > 0.175 are marked in grey). Number Percent Number Percent Area Percent Earthquake intensity of buildings (%) of apartments (%) (m2) (%) weight factor SLOVENIA 1,188 8.1 23,721 12.8 1,263,921 12.0 - Ljubljana 393 11.0 7,992 14.3 429,555 14.0 1.0 Maribor 164 9.4 3,205 13.8 177,728 14.0 0.4 Celje 71 12.5 1,376 16.5 67,779 15.0 0.6 Velenje 65 24.2 1,162 23.0 63,325 22.0 0.5 Kranj 51 12.1 896 15.0 47,774 14.0 0.9 Koper 40 9.4 785 16.7 40,564 15.0 0.5 Trbovlje 41 13.0 692 18.0 39,691 20.0 0.6 Jesenice 17 6.5 619 14.3 31.450 14.0 0.7 Piran 20 6.2 417 19.3 25,281 20.0 0.4 Izola 22 9.9 410 23.2 22,482 22.0 0.4 Ravne na Koro{kem 17 12.1 398 20,8 21,134 20.0 0.5 Novo mesto 17 7.6 389 13.2 20,476 13.0 0.7 Nova Gorica 15 6.0 305 7.4 17,337 7.0 0.7 Postojna 11 6.7 242 12.2 13,895 12.0 0.8 Kamnik 12 7.4 260 12.2 13,639 12.0 0.9 Sežana 12 12.9 197 16,8 10,980 17.0 0.6 Laško 7 8.6 188 26.4 9,337 26.0 0.6 Idrija 12 6.8 178 11.3 9,185 11.0 0.8 Zagorje ob Savi Miir^ka 9nhnta 12 7 9.7 7.1 169 146 11.1 8.1 8,176 7,772 10.0 8.2 0.7 0.4 iviuiot\a ouuuia Skofja Loka 6 5.6 120 6.2 7,158 6.8 0.9 Domžale 5 4.3 163 8.3 7,091 6.7 1.0 Bled Korevie 5 5 5.8 2.9 124 145 20.2 8.0 6,947 6,582 21.0 6.9 0,7 0.6 l\UL 0,175) so posebej označene). občina površina (m2) faktor izpostavljenosti občine Ljubljana 1.223.045 1 Maribor 655.300 0,4 Celje 215.478 0,6 Kranj 128.537 0,9 Trbovlje 122.080 0,6 Jesenice 104.413 0,7 Koper 91.964 0,5 Velenje 91.059 0,5 Nova Gorica 69.492 0,7 Ptuj 64.928 0,5 Novo mesto 61.675 0,7 Postojna 52.875 0,8 Ravne na Koroškem 49.261 0,5 Kamnik 43.053 0,9 Piran 42.720 0,4 Hrastnik 39.555 0,6 Tržič 39.217 0,7 Krško 37.224 0,8 Kočevje 36.480 0,6 Idrija 34.649 0,8 Izola 34.169 0,4 Žalec 28.495 0,6 Domžale 28.230 1 Skofja Loka 27.372 0,9 Slovenska Bistrica 27.262 0,5 Preglednica 7: Občine z največjo površino potresno verjetno manj varnih večstanovanjskih stavb (potresno bolj ogrožene občine (ag > 0,175) so posebej označene). občina površina (m2) faktor izpostavljenosti občine Ljubljana 1.223.045 1 Maribor 655.300 0,4 Celje 215.478 0,6 Kranj Trhn\/I ip 128.537 122.080 122.080 0,9 11 bovlje Jesenice 104.413 0,6 0,6 0,7 Koper 91.964 0,5 Velenje 91.059 0,5 Nova Gorica Ptuj 69.492 64.928 0,7 0,5 Novo mesto 61.675 0,7 Postojna 52.875 0,8 Ravne na Koroškem 49.261 0,5 Kamnik 43.053 0,9 Piran 42.720 0,4 Hrastnik 39.555 0,6 Tržič 39.217 0,7 Krško 37.224 0,8 Kočevje 36.480 0,6 Idrija 34.649 0,8 Izola 34.169 0,4 Žalec 28.495 0,6 Domžale 28.230 1 Skofja Loka 27.372 0,9 Slovenska Bistrica 27.262 0,5 posamezne občine so prikazani tudi na slikah 4 in 5 ter v preglednicah 6 in 7 (gre za stanovanjsko površino). Sliki 4 in 5 prikazujeta delež površin ogroženih stavb glede na površine vseh večstanovanjskih stavb v določeni občini. Občine z največjo površino ogroženih stanovanj so navedene v preglednicah 6 in 7. Izstopajo seveda večje mestne občine, v katerih je največ večstanovanjskih stavb, čeprav se morda na zemljevidu Delež večstanovanjskih stavb po občinah od vseh stanovanjskih stavb leta 2002 (slika 1) niso zdele kritične. Vidimo lahko, da je delež potresno verjetno nevarnih večstanovanjskih stavb, pri katerih bi bilo verjetno potrebno ojačiti nosilno konstrukcijo, kar 53,7 % od vseh večstanovanjskih stavb. Njihova skupna površina je 4.213.462 m2 (41,1 %). Večina teh stanovanj je v stavbah iz opeke ali iz različnih gradbenih materialov, zgrajenih pred letom 1963. Delež večstanovanjskih stavb, ki so potresno verjetno manj varne, je manjši, 15,5%. Njihova skupna površina je 3.383.236 m2, kar je tretjina površin stanovanj v večstanovanjskih stavbah. Slika 4: Delež površin potresno verjetno nevarnih večstanovanjskih stavb po občinah od celotne površine večstanovanjskih stavb v občini. Glej angleški del prispevka. Slika 5: Delež površin potresno verjetno manj varnih večstanovanjskih zgradb po občinah od celotne površine večstanovanjskih stavb v občini. Glej angleški del prispevka. 6 Sklep Članek ocenjuje potresno ogroženost obstoječih večstanovanjskih zgradb v Sloveni in potrjuje zaskrbljenost, ki so jo izrazili tudi drugi strokovnjaki za potresno gradnjo (glej npr. Kubelj 2009). Naša raziskava je potrdila zaskrbljujočo potresno varnost večstanovanjskih stavb, saj sodobnim zahtevam ustreza le nekaj manj kot polovica obstoječih večstanovanjskih stavb. Za razliko od drugih tovrstnih študij predstavljena raziskava ne temelji na ocenah o številu stavb in njihovih površinah, temveč na statističnih podatkih iz popisa prebivalstva, gospodinjstev in stanovanj leta 2002. Vendar je treba še enkrat poudariti, da gre za zelo splošne ocene, temelječe na dostopnih statističnih podatkih, ki za natančnejšo oceno ne zadostujejo, omogočajo pa statistično oceno in prikaz grobe slike stanja ter obsega verjetno potrebnih potresnih sanacij. Potresi v Posočju so pokazali tudi, da poškodbe na stavbah niso odvisne le od leta njihove izgradnje, višine in materiala nosilne konstrukcije, ampak tudi od količine, povezanosti in tlorisne razporeditve sten, ki lahko daje stavbi pri potresu še kako zaželeno dodatno nosilnost. Posledice potresa so odvisne tudi od lokalnih razmer (Orožen Adamič, Hrvatin 2001; Gosar 2007), kvalitete temeljnih tal in podobnega. Bolj natančno sliko bi bilo mogoče dobiti z analizo vsake stavbe posebej, ki bi omogočala račun dejanske dodatne nosilnosti in s tem točno oceno dejanske ogroženosti, kar pa je zamudno in predvsem drago opravilo. Rezultati po občinah so razkrili presenetljivo veliko potresno problematičnih objektov v severovzhodnem delu države, v pasu, ki se začne na Koroškem in se nadaljuje na obeh straneh Drave. Delež potresno verjetno nevarnih večstanovanjskih objektov tu presega 50 %, kar je praviloma posledica dokaj zgodnje industrializacije in potreb po večjem številu stanovanj na tem območju. Stavbe so bile grajene v času, ko so veljala drugačna - manj zahtevna varnostna načela. Na srečo je to območje po sedanjih seizmičnih zemljevidih potresno manj ogroženo. Treba pa je izpostaviti skupino zgradb, ki so najbolj nevarne za bivanje. To so stavbe s petimi ali več etažami, ki so bile zgrajene pred letom 1981 in bi jih bilo treba najprej sanirati. Teh stavb je v Sloveniji 1188. V njih je 23.721 stanovanj s skupno površino 1.263.921 m2. Po površini tovrstnih stavb in deležu glede na celotno površino večstanovanjskih stavb najbolj ogrožene občine so predstavljene v preglednici 8. Skrb za izboljšanje tovrstne varnosti oziroma statično sanacijo zgradb bo verjetno morala prevzeti država. Učinkovito sanacijo zelo otežujejo zapletena lastninska razmerja v večstanovanjskih stavbah, čeprav Pravilnik o merilih za določitev prispevka etažnega lastnika v rezervni sklad in najnižji vrednosti prispevka (Ur.l. RS 11/2004) predvideva zbiranje mesečnih prispevkov vseh lastnikov stanovanj v stavbah, starejših od 10 let za tako imenovani »rezervni sklad« za izvajanje potrebnih vzdrževalnih in obnovitvenih del. Glede na to, da gre pri protipotresni sanaciji za velike denarne zneske, bo tako zahteven projekt možno izvesti le z ustrezno zakonodajo in stimulacijami lastnikov oziroma olajšavami zanje. Druga možnost bi bila sofinanciranje občin v zameno za povečanje deleža lastništva v tovrstnihh zgradbah. Na kakršenkoli način se bo to stanje v prihodnosti urejalo, bo poleg gradbenih del odprlo še zelo zahtevno pravno in ekonomsko problematiko. Preglednica 8: Občine po številu stavb, številom stanovanj v njih in njihovo površino v večstanovanjskih stavbah, višjih od štiri etaže, ki so potresno verjetno najbolj ogroženi del slovenskega večstanovanjskega fonda. Poleg absolutnih vrednosti so navedeni tudi odstotni deleži glede na vse večstanovanjske objekte v občini (potresno bolj ogrožene občine (ag >0,175) so posebej označene). število delež število delež površina delež faktor izpostavljenosti stavb (%) stanovanj (%) (m2) (%) občine SLOVENIJA 1188 8,1 23.721 12,8 1.263.921 12,0 - Ljubljana 393 11,0 7992 14,3 429.555 14,0 1,0 Maribor 164 9,4 3205 13,8 177.728 14,0 0,4 Celje 71 12,5 1376 16,5 67.779 15,0 0,6 Velenje 65 24,2 1162 23,0 63.325 22,0 0,5 Kranj 51 12,1 896 15,0 47.774 14,0 0,9 Koper 40 9,4 785 16,7 40.564 15,0 0,5 Trbovlje 41 13,0 692 18,0 39.691 20,0 0,6 Jesenice 17 6,5 619 14,3 31.450 14,0 0,7 Piran 20 6,2 417 19,3 25.281 20,0 0,4 Izola 22 9,9 410 23,2 22.482 22,0 0,4 Ravne na Koroškem 17 12,1 398 20,8 21.134 20,0 0,5 Novo mesto 17 7,6 389 13,2 20.476 13,0 0,7 Nova Gorica 15 6,0 305 7,4 17.337 7,0 0,7 Postojna 11 6,7 242 12,2 13.895 12,0 0,8 Kamnik 12 7,4 260 12,2 13.639 12,0 0,9 Sežana 12 12,9 197 16,8 10.980 17,0 0,6 Laško 7 8,6 188 26,4 9337 26,0 0,6 Idrija 12 6,8 178 11,3 9185 11,0 0,8 Zagorje ob Savi Mi ircl/a Qnhntsi 12 7 7 9,7 169 146 11,1 8176 7772 7772 10,0 0,7 iviuiska sobota Skofja Loka 6 7,1 7,1 5,6 146 120 8,1 8,1 6,2 7158 8,2 8,2 6,8 0,4 0,4 0,9 Domžale 5 4,3 163 8,3 7091 6,7 1,0 Bled k'nrpuip Kočevje 5 5 5 5,8 124 145 20,2 6947 6582 6582 21,0 0,7 Litija 7 2,9 2,9 7,5 145 118 8,0 8,0 9,1 6209 6,9 6,9 9,2 0,6 0,6 0,8 Slovenska Bistrica 5 3,5 121 9,6 5184 8,1 0,5 Hrastnik 5 4,6 99 6,3 4969 6,3 0,6 Krško 5 3,8 87 5,6 4380 5,0 0,8 Vrhnika 5 6,0 66 8,4 3966 9,0 0,9 Ptuj 5 2,0 45 1,8 2633 1,8 0,5 7 Literatura Glej angleški del prispevka.