Virtual acoustic reconstruction of the Roman Theatre of Palmyra

   48º CONGRESO ESPAÑOL DE ACÚSTICA ENCUENTRO IBÉRICO DE ACÚSTICA EUROPEAN SYMPOSIUM ON UNDERWATER ACOUSTICS APPLICATIONS EUROPEAN SYMPOSIUM ON SUSTAINABLE BUILDING ACOUSTICS VIRTUAL ACOUSTIC RECONSTRUCTION OF THE ROMAN THEATRE OF PALMYRA PACS: 43.55. Ka Álvarez-Corbacho, A.1; Bustamante, P.2; Zamarreño, T.1; Galindo, M.1; Girón, S.1 1 Departamento de Física Aplicada II, Universidad de Sevilla. Instituto Universitario de Arquitectura y Ciencias de la Construcción (IUACC). Avda. Reina Mercedes 2. 41012 - Sevilla, España. Tel: +34 954556612, Fax: +34 954557892, E-mails: [email protected]; [email protected]; [email protected]; [email protected] 2 Departamento de Construcciones Arquitectónicas I, Universidad de Sevilla. Instituto Universitario de Arquitectura y Ciencias de la Construcción (IUACC). Avda. Reina Mercedes 2, 41012 - Sevilla, España. Tel: +34 954559517, Fax: +34 954557018, E-mail: [email protected] Keywords: Roman theatres, virtual acoustics, virtual reconstruction, computer models ABSTRACT In this paper, the creation process, adjustment, and validation of the 3D model of the Roman theatre of the monumental Syrian city of Palmyra is analysed in order to simulate its sound field. Given the impossibility of carrying out on-site measurements, the acoustic simulation is based on a 3D model made from photogrammetry, blueprints, and photographs obtained from the web and from the previous studies of the group on the Roman theatres of Italica (Seville, Spain) and Regina Turdulorum (Badajoz, Spain). Due to the special shape of the cross-section of the cavea of this theatre, a study has been carried out on the influence of the dispersion of reflected sound in the objective acoustic parameters. RESUMEN En esta comunicación se analiza el proceso de creación, ajuste y validación del modelo 3D del teatro romano de la ciudad monumental siria de Palmira para simular su campo sonoro. Dada la imposibilidad de realizar medidas in situ, la simulación acústica está basada en un modelo 3D realizado a partir de fotogrametría, planimetría y fotografías obtenidas de la web y en los trabajos anteriores del grupo sobre los teatros romanos de Itálica (Sevilla, España) y Regina Turdulorum (Badajoz, España). Debido a la especial forma de la sección de la cavea de este teatro, se ha realizado un estudio de la influencia de la dispersión del sonido reflejado en los parámetros acústicos objetivos. THE CITY OF PALMYRA According to Plinio's testimony [1], the city of Palmyra, located in the Syrian steppe, benefitted from its location between the Euphrates River and the cities of western Syria (Homs, Damascus, Bosra) and the Mediterranean coast (Figure 1).    48º CONGRESO ESPAÑOL DE ACÚSTICA ENCUENTRO IBÉRICO DE ACÚSTICA EUROPEAN SYMPOSIUM ON UNDERWATER ACOUSTICS APPLICATIONS EUROPEAN SYMPOSIUM ON SUSTAINABLE BUILDING ACOUSTICS Palmyra was originally a settlement located in an oasis called Tadmor in the northern part of the Syrian Desert. Although the Roman province of Syria was created in 64 BC, the inhabitants of Tadmor (Arabic name of the same Aramaic word "Palmyra"), mainly Arameans and Arabs, remained semi-independent for more than half a century, benefitting from their control of the caravan routes between the Roman coast of Syria and the empire of Parthia to the east of the Euphrates, which enabled them to provide the Roman Empire with goods from all directions. Palmyra was strategically located on two of the most important trade routes of the ancient world: one stretched from the Far East and India to the head of the Persian Gulf, while the other route (the Silk Road) extended all over the Eurasian continent up to China. Under the Roman emperor Tiberius (14- 37 AD), Tadmor was incorporated into the province of Syria and adopted the name Palmyra, "city of date trees" (Figures 2 and 3). After the Roman invasion of Nabataea in 106 AD, Palmyra replaced Petra as the leading Arab city in the Near East and its most important commercial centre. Around 129 AD, during the reign of Hadrian, Palmyra ascended to the rank of a free city, and in 212 AD to that of a Roman colony. With the founding of the Sasanian Empire of Iran in 224 AD, Palmyra lost control of the commercial routes, but Septimius Odaenathus, head of a prominent Arab family that was an ally of the Roman Empire, directed two campaigns against the Sasanians and drove them out of Syria. When Odaenathus was assassinated in 267 AD, his Arab queen, Zenobia, declared herself Augusta (Empress) and ruled in the name of her son, Vaballathus. Zenobia established Palmyra as the capital of an independent empire and expanded its borders by conquering Lower Egypt, Syria (region), Palestine, Anatolia and Lebanon. Her government did not last long however: in the year 272 AD, Emperor Aurelian reconquered Palmyra and captured Zenobia. To celebrate his triumph, he paraded its queen, Zenobia, dressed in jewels and fettered in gold chains so heavy that she needed servants to move them. 'The haloed prisoners of greatness constituted solid proof of the splendour of victory', writes Mary Beard [2]. Figure 1. General map of Palmyra. (Drawing from Thibaud Fournet).  Figure 2. Theatre of Palmyra in the Decumanus. (National Geospatial-Intelligence Agency, Bethesda, MD, USA). Figure 3. Theatre of Palmyra in the Decumanus. On the left the Tetrapylon. (www.historiayarqueología.com).    48º CONGRESO ESPAÑOL DE ACÚSTICA ENCUENTRO IBÉRICO DE ACÚSTICA EUROPEAN SYMPOSIUM ON UNDERWATER ACOUSTICS APPLICATIONS EUROPEAN SYMPOSIUM ON SUSTAINABLE BUILDING ACOUSTICS That date marked the end of Palmyra’s great prosperity and the beginning of its decline, but not its demise. Palmyra never regained its previous power, but did remain a major city. With the passage of time it became a bishopric and the city sent a bishop to the Council of Nicea in 325 AD and again in 351 AD to the council of Chalcedon. The recognition of the splendour of the ruins of Palmyra by travellers in the seventeenth and eighteenth centuries [3] contributed greatly to the later revival of classical architectural styles and urban design in the West. The first excavations of Palmyra were made in 1902 by Otto Puchstein and in 1917 by Theodor Wiegand. In 1929, the director general of antiquities in Syria and Lebanon, Henri Arnold Seyrig, began excavating the ruins and convinced the villagers to move to a new village built by the French near the site. The relocation was completed in 1932; the old Palmyra was ready to be excavated while its villagers settled in the new town of Tedmur. The excavation of the site, interrupted by the Second World War, resumed shortly after the end of the war. Since 1980, it has been registered as an archaeological site on the Unesco World Heritage List. On 20 June, 2013, Unesco included all the Syrian heritage sites on the list of endangered World Heritage Sites to warn of the risks to which they would be exposed due to the Syrian civil war [4]. THE ROMAN THEATRE OF PALMYRA The unfinished theatre of Palmyra [5] dates back to the 2nd century AD. It was built in the centre of a semi-circular arcaded square that opens to the south gate of Palmyra (Figures 3 and 4). The square, completely flanked by columns, measuring 82 by 104 m, is located southwest of the city's Decumanus, which runs from the monumental arch, access gate in the southeast of the city, to the Tetrapylon, in the northwest of the city (Figure 3). Its stage front, in carved stone, has the royal vault located in a wide curved recess and two hospitalarium doors in shallow rectangular recesses. It also has two additional doors at the ends of the scaenae frons. The wall of the stage therefore has five entrances through which actors acceded to the stage. The sides of the scaenae frons are closed by walls (versurae) that also have access doors that led to the stage from the rooms or side basilicas; their columns are of Corinthian order. The Decumanus, flanked by columns, functions as a type of porticus post scaena (Figure 5). Its proscenium has ten curved niches, nine rectangular niches and two access staircases to the pulpitum from the orchestra. This large number of niches is a characteristic of the second and third centuries, especially in the East. The pulpitum dimensions are 45.5 m x 10.5 m. The unfinished tier, executed in carved stone, only has the ima cavea of 12 rows, divided into 11 cunei by 10 staircases. It had a planned diameter of 92 m. The theatre has a central corridor in the cavea that bisects it (Figures 5 and 6) and allows direct access from the exterior square behind the theatre into the orchestra (Figures 5 and 6). This access must have been for the exclusive use of the most influential spectators of Palmyra who would sit in the proedria of the theatre and for those who occupied the lowest seats in the ima cavea. In other words, it complemented or replaced the aditus maximi as a means of privileged entry. The aditus maximi of the theatre, with its main entrances, is 3.5 m wide and leads to the orchestra paved in Figure 4. Ground plan of the Roman Theatre of Palmyra. (F. Sear. Roman Theatres. An Architectural Study).    48º CONGRESO ESPAÑOL DE ACÚSTICA ENCUENTRO IBÉRICO DE ACÚSTICA EUROPEAN SYMPOSIUM ON UNDERWATER ACOUSTICS APPLICATIONS EUROPEAN SYMPOSIUM ON SUSTAINABLE BUILDING ACOUSTICS carved stone, with a diameter of 23.50 m. It is bounded by a wall (balteus) of 20.30 m in diameter, which forms a praecintio of 1.60 m in width (Figures 6 and 7). In the 1950s, the theatre was cleared of sand and later underwent restoration work. It should be noted that the theatre of Palmyra, studied by Puchstein between 1902 and 1917, is highly reconstructed rather than restored and, therefore, has been the object of many academic concerns regarding the historical authenticity of the current structure of the building [6]. It served as the venue for the annual festival of Palmyra, hosting performances of folk music until the beginning of the civil war in Syria. EXPERIMENTAL METHOD In order to adapt the acoustic conditions of simulation to those of a real situation, the model has been adjusted using the absorption and dispersion coefficients of the materials used in the acoustic studies of the Roman theatres of Italica [7] and of Regina Turdulorum [8]. In both theatres, acoustic measurements were made in situ with the theatres unoccupied. In both measurements, the environmental conditions were monitored by measuring temperature and relative humidity, and the recommendations of ISO 3382-1:2009(E) [9] were followed. The process of generating the frequency exponential sweep signal to excite the theatre, the acquisition, and the analysis of the IR were carried out with the WinMSL2004 programme in Italica and with Dirac 5 in Regina Turdulorum through the Edirol UA-101 sound card. The signal generated was emitted by the AVM DO-12 dodecahedral source with a B&K 2734 power amplifier, the sources were located at 1.50 m above the floor of the proscenium and in the orchestra, and were recorded at the reception points distributed by the cavea, the proedria of the orchestra and in the proscenium. The monaural IRs were collected using a multi-pattern microphone (omnidirectional Figure 7. Wall of the stage. Roman theatre of Palmyra. Photograph from Guillaume Piolle, CC BY 3.0, https://commons.wikimedia.org/w/index.php?curid=11304940 Figure 5. The Roman theatre of Palmyra. Colonnade of the Decumanus. Top left: the Tetrapylon. (www.flickr.com/photos/newpalmyra). Figure 6. Cavea of the Roman theatre of Palmyra. Photograph from Bernard Gagnon, CC BY-SA 3.0, https://commons.wikimedia.org/w/index.php?curid=12198409    48º CONGRESO ESPAÑOL DE ACÚSTICA ENCUENTRO IBÉRICO DE ACÚSTICA EUROPEAN SYMPOSIUM ON UNDERWATER ACOUSTICS APPLICATIONS EUROPEAN SYMPOSIUM ON SUSTAINABLE BUILDING ACOUSTICS and figure-of-eight) Audio-Technica AT4050/CM5 connected to a 4-channel Sound Field SMP200 polarization source. The binaural IRs were obtained with a Head Acoustics HMS III torso simulator (Code 1323) and the B&K-2829 micro-polarization source. In all cases, the microphone was placed at a height of 1.20 m from the ground. Table 1 contains a summary of the geometry of the three theatres. It also shows the three theatres studied in the ERATO project. ACOUSTIC MODEL AND SIMULATION The acoustic simulation has been carried out using the CATT-Acoustic v9.1a programme [10], based on geometric acoustic algorithms. The model is constructed from photogrammetry, planimetry, and photographs obtained from the web. For the geometric survey, a three-dimensional model has been generated by the SketchUp programme, and exported to CATT-Acoustic through the SU2CATT v 1.3 plugin. The final model consists of 5,629 plans. The simulations were carried out by considering the case of an open model, with the source centred on the proscenium, and by recording the signal at the 27 reception points distributed across the cavea (Figure 8). The meteorological conditions used for the simulation have been taken from the 2013 publication ASHRAE Handbook-Fundamentals [11]. Table 2 collects the data used: Table 2. Meteorological conditions of the simulation. Temperature (ºC) Relative Humidity (%) Pressure (Pa) Air density (kg/m 3 ) [12] 22.4 ºC 51 96,560 1.13 Results have been obtained with the TUCT v2 (The Universal Cone Tracer) calculation engine, which calculates the acoustic parameters from energy (E) and/or B-format and binaural impulse responses (h) based on the evaluation of sound pressure. In the simulation process, complete simulations were launched with 1,000,000 rays and an echogram duration of 1,500 ms, superior to the estimated reverberation time of 1,000 ms, and used the TUCT algorithm 2. Table 3 shows the absorption and dispersion coefficients finally assigned to the surfaces, the associated colours in Figure 9, and the bibliographic references used. Figure 8. Computer model of the theatre of Palmyra. Table 1. Geometry of the theatres [5]. Theatre Orchestra diameter (m) Cavea diameter (m) Regina Turdulorum 19.60 64.00 Palmyra 23.50 46.82 Italica 25.40 75.76 Jerash South 19.91 76.00 Syracuse 21.40 138.50 Aspendos 23.87 95.48    48º CONGRESO ESPAÑOL DE ACÚSTICA ENCUENTRO IBÉRICO DE ACÚSTICA EUROPEAN SYMPOSIUM ON UNDERWATER ACOUSTICS APPLICATIONS EUROPEAN SYMPOSIUM ON SUSTAINABLE BUILDING ACOUSTICS Table 3. Coefficients of absorption (top) and dispersion (below), at octave bands, of the materials for the simulation. Surface, [reference] Colour (Fig. 9) Absorption and dispersion coefficients 125 250 500 1k 2k 4k Land, [13] 210,179,102 0.15 0.35 0.40 0.50 0.55 0.80 0.20 0.25 0.30 0.40 0.50 0.60 Stone, [14] (Dispersion Italica theatre) 240, 200,100 0.02 0.03 0.03 0.03 0.04 0.07 0.61 0.61 0.57 0.44 0.37 0.33 Stone, [14] (Dispersion Regina theatre) 240, 200,100 0.02 0.03 0.03 0.03 0.04 0.07 0.55 0.58 0.62 0.62 0.60 0.56 Stone Cavea, [14] (Dispersion Italica theatre) 205,155,30 0.02 0.03 0.03 0.03 0.04 0.07 0.61 0.61 0.57 0.44 0.37 0.33 Stone Cavea, [14] (Dispersion Regina theatre) 205,155,30 0.02 0.03 0.03 0.03 0.04 0.07 0.55 0.58 0.62 0.62 0.60 0.56 Stone columns, [14] 250, 180, 10 0.02 0.03 0.03 0.03 0.04 0.07 0.35 0.35 0.45 0.40 0.45 0.50 Table 4 shows the results of the calibration process for the values of T30 spatially averaged for the source centred in the proscenium and in Figure 9 these values are displayed. Table 4. Results of the calibration. Values of T30 (s), spatially averaged, at octave bands. Frequency (Hz) 125 250 500 1000 2000 4000 Measured values Italica 1.00 0.98 0.98 1.16 1.17 1.01 Measured values Regina Turdulorum 0.69 0.66 0.64 0.64 0.63 0.56 Palmyra simulated values coef. Italica 0.98 0.92 0.92 1.01 0.98 0.89 Palmyra simulated values coef. Regina 0.96 0.92 0.90 0.89 0.84 0.76 Simulation difference (%) 1.33% -0.48% 2.21% 12.73% 14.61% 14.01% It can be observed (Figure 10) that the differences obtained between the two simulations (referred to the values of the simulation carried out with the coefficients of Italica theatre) are less than 5%, that is, lower than 1 JND, in the bands 125-500 Hz. The most significant differences lie in the bands 1000-4000 Hz, whereby they are close to 3 JND for 2000 Hz for the reverberation time T30 since, in these bands, the most notable differences appear in the dispersion coefficients. ANALYSIS AS A WHOLE Figure 11 shows the spectral behaviour of the differences between the simulated values using different material dispersion coefficients, spatially averaged corresponding to the energy and binaural acoustic parameters. In order to compare the differences in the various parameters, they have been expressed in terms of their respective JND. Figure 9. Spectral behaviour, spatially averaged, of T30 values. Vertical bars value the spatial dispersion through the standard error. Figure 10. Spectral behaviour of the differences between the spatial average of simulated values of the acoustic p arameters T30 and EDT. Frequency (Hz) 125 250 500 1000 2000 4000 Reverberation time T 30 (s) 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 Measured Italica theatre Simulated Palmyra coef. Italica theatre Measured Regina theatre simulated Palmyra coef. Regina theatre Frecuency (Hz) 125 250 500 1000 2000 4000 JND -1 0 1 2 3 4 EDT T 30    48º CONGRESO ESPAÑOL DE ACÚSTICA ENCUENTRO IBÉRICO DE ACÚSTICA EUROPEAN SYMPOSIUM ON UNDERWATER ACOUSTICS APPLICATIONS EUROPEAN SYMPOSIUM ON SUSTAINABLE BUILDING ACOUSTICS We can see that these differences, for all parameters, are kept below ± 1 JND for all frequen-cies. Figure 12 shows the value of the sound strength Gm, at an average of 500 and 1000 Hz, versus source-receiver distance. We see that both have a linear drop with a slope of -0.48 dB/m and correlation coefficients r2 equal to 0.84 and 0.87, respectively. The simulated values remain about 6 dB above the trend marked by the theoretical value of G for a free field (Figure 12). This drop was not observed in the C80m and Tsm parameters, as happens in closed enclosures. The simulated values of these parameters remain between 4-8 dB and 30-55 ms, respectively. Table 5 summarizes the values of the acoustic parameters, spatially and spectrally averaged in accordance with ISO 3382-1 [9], of the unoccupied theatres, from the ERATO project [15] and those of Italica, Regina Turdulorum, and Palmyra. Only acoustic measurements have been carried out in Aspendos, Italica and Regina Turdulorum, the remaining measurements are simulated by computer. Table 5. Values of the acoustic parameters of theatres from ERATO project and those from Italica, Regina, and Palmyra spatially and spectrally averaged in accordance with ISO 3382-1. T30m [s] Gm [dB] C 80m [dB] T30m [s] Gm [dB] C 80m [dB] Aspendos 1.89 -4.37 2.68 Italica 1.07 -0.71 8.09 Jerash South (*) 1.21 -1.18 5 . 98 Regina Turdulorum 0.64 -0.10 (*) 10.62 Syracuse (*) 1.25 -10.60 12 . 88 Palmyra coef. Italica (*) 0.97 0.17 6.33 (*) Values obtained by simulation. Palmyra coef. Regina (*) 0.89 -0.08 6.48 CONCLUSIONS The historical, archaeological, architectonic, and heritage value of the Roman theatre of Palmyra, Syria, has been documented from bibliographic bases, and the acoustic behaviour of the aforementioned theatre has been characterised. A 3D model has been developed for the simulation of the sound field. This theatre has been simulated using the coefficients of absorption and dispersion of the materials used in the acoustic studies of the Roman theatres of Italica and Regina Turdulorum, whose models were adjusted by testing dispersion coefficients for the different materials, especially in the cavea. We analysed the spectral behaviour of the acoustic parameters T30, EDT, C80, D50, and IACCE, spatially averaged and, where appropriate, the variation with the source-receiver distance of the Figure 11. Spectral behaviour of the differences between the spatial averages of simulated values of the acoustic parameters: C80, D50, and IACCE. Frecuency (Hz) 125 250 500 1000 2000 4000 JND -0.8 -0.6 -0.4 -0.2 0.0 0.2 0.4 C80 D50 IACCe Figure 12. Variation of the spectral simulated value of Gm (average of 0.5 and 1 kHz) against source-receiver distance. Source-reveiver distance (m) 12 14 16 18 20 22 24 26 28 30 G m (dB) -12 -10 -8 -6 -4 -2 0 2 4 6 Simulation Palmyra theatre coef. Italica theatre Simulación Teatro Palmira coef. Teatro Regina Campo libre    48º CONGRESO ESPAÑOL DE ACÚSTICA ENCUENTRO IBÉRICO DE ACÚSTICA EUROPEAN SYMPOSIUM ON UNDERWATER ACOUSTICS APPLICATIONS EUROPEAN SYMPOSIUM ON SUSTAINABLE BUILDING ACOUSTICS spectral parameters Gm, C80m, and TSm. In this respect, it has become clear that the Gm values present an attenuation with distance that is approximately linear, however, similar behaviour has not been observed with other parameters (such as those of clarity), which in closed spaces usually appear. Several of the parameters (simulated), spatially and spectrally averaged in accordance with ISO, have been compared with the values published for three theatres investigated in the ERATO project, and for Italica and Regina Turdulorum, and show that parameters from Palmyra are similar to other analogous classical spaces. One aspect to be studied in the future would involve the use of models generated for sensory evaluation through auralisations. REFERENCES [1] Plinio el viejo. Historia Natural. Libro V. Descripción de África y Asia. Gredos. Madrid. (1998). [2] M. Beard. El triunfo romano: una historia de Roma a través de la celebración de sus victorias. Crítica. Barcelona. (2008). [3] R. Wood. The Ruins of Palmyra otherwise Tedmor, in the Desart. Londres. (1753). [4] Wikipedia, visited 5th February 2017. [5] F. Sear. Roman Theatres. An Architectural Study. Oxford University Press. Oxford. (2006). [6] C. Finlayson. New Excavations and a Reexamination of the Great Roman Theater at Apamea, Syria. American Journal of Archaeology, 116(2), 277-319 (2012). [7] A. Álvarez-Corbacho, T. Zamarreño, M. Galindo, S. Girón. Virtual acoustics of the Roman Theatre of Italica (Acústica virtual del teatro romano de Itálica). Proceedings of Tecniacústica. Murcia, 1229-1236 (2014). [8] A. Álvarez-Corbacho, T. Zamarreño, M. Galindo, S. Girón. Virtual acoustics of the Roman Theatre of Regina Turdulorum (Acústica Virtual del teatro romano de Regina Turdulorum). Proceedings of Tecniacústica. Valencia, 1515-1522 (2015). [9] ISO 3382-1:2009(E), Acoustics-Measurement of room acoustic parameters, part 1: Performance rooms, International Organisation for Standardisation, Geneva, Switzerland 2009. [10] B. I. L. Dalenbäck. CATT-Acoustic v9 powered by TUCT user manuals. Computer Aided Theatre Technique, Gothenburg (Sweden), (2011). [11] American Society of Heating, Refrigerating, and Air-Conditioning Engineers. 2013 ASHRAE handbook: fundamentals (SI). Atlanta: ASHRAE (2013). [12] A. Picard, R. S. Davis, M. Gläser, K. Fujii, Revised formula for the density of moist air (CIPM- 2007), Metrologia 45, 149-155 (2008). [13] Physikalisch-Technische Bundesanstalt. http://www.ptb.de/en/org/1/17/172/datenbank.htm (visited 10/07/2015). [14] M. Vorländer. Auralization, fundamentals of acoustics, modelling, simulation, algorithms and acoustic virtual reality. Springer-Verlag, Berlín (2008). [15] J. H. Rindel, Roman Theatres and Revival of Their Acoustics in the ERATO Project. Acta Acustica United with Acustica, 99, 21–29 (2013).