Sound and noise in urban parks

Proceedings of Meetings on Acoustics Volume 18, 2012 http://acousticalsociety.org/ 164th Meeting of the Acoustical Society of America Kansas City, Missouri 22 - 26 October 2012 Session 2aNS: Noise 2aNS10. Sound and noise in urban parks Antonio P. Carvalho* and Ricardo A. Cleto *Corresponding author’s address: Laboratory of Acoustics, University of Porto, FEUP, Porto, P-4200-465, Porto, Portugal, [email protected] The main goal of this work is to study the soundscape of city gardens and urban parks using a sample of ten sites in Oporto, Portugal to analyze their soundscape through the acoustic characterization of the park’s exterior and interior noise levels (LAeq, LA10, LA50 and LA90) and by a socio-acoustic survey to the visitors to check their perception of acoustic quality. The measurements showed gardens/parks with interior noise levels from 47 to 61 dB(A) (with exterior noise levels up to 67 dB(A)). The difference between exterior and interior LAeq was between 3 and 19 dB. The gardens with lower noise levels are the largest and out of downtown. An "acoustic" classification for gardens/urban parks is proposed regarding their noise "isolation" capacity and their acoustic ambiance. Old 1990 measurements allow for the comparison of the acoustic evolution in the last 21 years. The socio-acoustic survey concludes that Oporto’s city parks are visited mostly by an elderly male population that regards these places as sites of gathering and to practice some physical activity rather than as an acoustic retreat. They seem accustomed to the dominant sound sources, classifying those spaces as pleasant and quiet, even when noise is over acceptable limits. Published by the Acoustical Society of America through the American Institute of Physics A. Carvalho and R. Cleto © 2012 Acoustical Society of America [DOI: 10.1121/1.4772735] Received 8 Oct 2012; published 5 Dec 2012 Proceedings of Meetings on Acoustics, Vol. 18, 040001 (2012) Page 1 1 Introduction The objectives of this work are the characterization and analysis of the sound levels in the most significant city gardens and urban parks of Oporto (the second largest town in Portugal with 238,000 inhabitants and 42 km 2 ) using two approaches [1]: •In situ sound level measurements to acoustically characterize the gardens' interior and exterior environment; •Questionnaires to visitors of the parks to assess their perceived acoustical quality and tranquility achieved in those places. To understand the evolution of noise levels in the last 21 years, a comparison with measurements done in 1990 is presented. This study also formulates an “acoustical” classification for city gardens and urban parks regarding their aptitude for “urban noise isolation” and to provide a calm and serene environment to their visitors. 2 Methodology 2.1 Parameters The acoustic parameters used in this study were L Aeq , L A10 , L A50 and L A90 . These statistical descriptors were used especially to allow a comparison with results measured on 1990 and to analyze the evolution in the last 21 years of the Oporto’s gardens. Measurements were done in the exterior and in the interior of ten gardens/parks. The sound level variation from the outside to the inside of the parks in each of those parameters was also analyzed. The equipment used for the in situ measurements was a Brüel & Kjær 2236 sound level meter with a B&K 4188 microphone (Figure 1). 2.2 Questionnaires A system of personal interviews to visitors within the parks was used to obtain concrete answers to subjective matters relating to the noise exposition (Figure 2). The visitors were randomly approached during their activity in the parks and asked to be a volunteer in this study. Not to allow any bias, the visitors were not told that the main goal was to assess the soundscape in the park. Instead, they were informed that the inquiry was about the environmental quality of that park, giving a much more general idea of the objective. The questionnaire used had ten closed questions and one open question, divided in three sections. The first section is composed by fields regarding the identification (genre and age); the second regards the sound and environmental quality of the park and the third involves the type of visitor concerning his/her park use and a final evaluation. The second section of the questionnaire is based in a normalized methodology [2] tested by Pereira [3], with adaptations for this case, and consisted of eight closed questions (nºs 1-5 and 79) in a Likert scale and one open question (nº 6): A. Carvalho and R. Cleto Proceedings of Meetings on Acoustics, Vol. 18, 040001 (2012) Page 2 •Question 1: Are the sounds that I hear in this park expected? •Question 2: Do I like the sounds that I hear? •Question 3: Does the volume of these sounds bother me? •Question 4: Do I consider this garden/park agreeable? •Question 5: Do I consider this garden/park tranquil? •Question 6: Identify three sounds that you hear in this garden/park. •Question 7: What brings you to this garden/park? •Question 8: What is the aspect of this garden/park that is the most important for you? •Question 9: How do you rate this garden/park? Question 2 wanted to identify the visitors' opinion about the aesthetic quality of the sound they heard to know if these sounds should or should not be preserved in the local soundscape. In question 6, all the main sounds identified by the visitors are registered to be listed and grouped concerning their reference aspects based on Schafer’s classification [4]: a) traffic, b) human sounds, c) natural sounds, d) bird sounds, e) equipments and machinery, f) music, g) traffic lights and h) other. Question 7 wants to identify the motives that brought the visitors to the park. In question 8 the visitor was invited to classify the importance of some aspects of the garden (choosing only one) as vegetation, clean air, cleanness, safety and silence. Finally the visitor was asked in question 9 to give an overall opinion of the garden. 2.3 Procedures for in situ measurements The following procedures were followed for the in situ measurements: • Meteorological conditions were taken into account not using rainy days or when the wind speed was above 5 m/s; • To compare with in situ measurements done in 1990 (in six gardens) a similar time schedule was used in this study (between 15 h and 18 h); • To chose the measuring positions within the parks two perpendicular axes were traced, if possible, oriented by the North-South lines, getting four points in the limits of the garden/park and two in the interior; • Measuring intervals of about 10 to 20 minutes were used to get representative values of the chosen parameters; • Each measurement position was chosen not to interfere with the visitors and at least 3.5 m from any reflective surface; • The sound level meter was placed in each position with a tripod at a height of 1.2 to 1.5 m (Figure 1). 3Sample The selection of gardens/urban parks used in this study followed a criterion of being representative in size and in use by the public (Table 1 and Figures 1 to 4). They were chosen based on their placement and significance within the town including the desire to include, in the sample, places with small and large areas, within and outside the downtown zone, that is, where diverse parameters could be studied as urban density, proximity with large road highways and the multiplicity of uses that those spaces bring to the city. A. Carvalho and R. Cleto Proceedings of Meetings on Acoustics, Vol. 18, 040001 (2012) Page 3 Table 1 - City gardens and urban parks studied (Jardim = Garden; Praça = Square, Quinta = Farm; Parque = Park). City garden/urban park full name (short name) Area (ha) Parish (within Porto county) Type of perimeter 1 Jardim Teófilo Braga (República) 1.3 Cedofeita free (no fence or other limit) 2 Praça do Marquês de Pombal (Marquês) 0.8 Santo Ildefonso free (no fence or other limit) 3 Praça Mouzinho de Albuquerque (Boavista) 3.1 Cedofeita free (no fence or other limit) 4 Jardim de João Chagas (Cordoaria) 1.6 Vitória free (no fence or other limit) 5 Jardim de São Lázaro (S. Lázaro) 0.7 Bonfim totally fenced 6 Jardim da Casa de Serralves (Serralves) 18.0 Lordelo do Ouro totally walled 7 Parque Ocidental da Cidade (Ocidental) 83.0 Aldoar partially walled 8 Quinta do Covelo (Covelo) 8.0 Paranhos with wall and metallic fence 9 Parque de São Roque (S. Roque) 4.5 Bonfim totally walled 10 Praça Francisco Sá Carneiro (Velásquez) 3.0 Bonfim free (no fence or other limit) Figure 1 (left): Measurement in Garden 1 (República, East side). Figure 2 (right): Visitors in Garden 1 (República) answering the questionnaire. A. Carvalho and R. Cleto Proceedings of Meetings on Acoustics, Vol. 18, 040001 (2012) Page 4 Figure 3 (left): Urban area of the Municipal Plan with Garden 1 in red [adapted from 5] Figure 4 (right): Aerial photo of Garden 1 (República) [6]. 4 Results and analyses 4.1 In situ measurements 4.1.1 Measured values Using the L Aeq values measured in the exterior and interior of the gardens studied it is possible to find an effective reduction in the interior L Aeq compared with the gardens’ exterior, in all cases studied (Figures 5 and 6). This can be justified by the fact that the involved sound sources are in the gardens’ periphery making possible a decrease of sound pressure level (even if small) with the progressive increase in distance from the noise source. In Figure 5, concerning the surrounding noise levels (exterior of the gardens/urban parks), it is possible to see that the most noisy environment was measured around Garden 3 (Boavista), followed by the exterior of Park 7 (Ocidental), with the highest L Aeq value (67 dB). The quietest was the exterior of Garden 9 (S. Roque) with the lowest L Aeq (54 dB). Having the noisiest environment around Garden 3 (Boavista) can be justified by the urban net in that place because it is a major round square (with a circular turnaround) with main streets getting to and out of that square (an important commercial area that canalizes a large amount of compact traffic in and out of it). For Park 9 (S. Roque), its urban environment is characterized as a residential area with low traffic, which justifies the differences found in noise levels. A. Carvalho and R. Cleto Proceedings of Meetings on Acoustics, Vol. 18, 040001 (2012) Page 5 Figure 5 - Average sound levels values (L Aeq , L A10 , L A50 and L A90 ) measured in the exterior of ten gardens/urban parks of Oporto, Portugal. Figure 6 - Average sound levels values (L Aeq , L A10 , L A50 and L A90 ) measured in the interior of ten gardens/urban parks of Oporto, Portugal. The most noisy interior in the tested gardens (Figure 6) were numbers 2 and 5 with L Aeq of 61 dB (Marquês and S. Lázaro) that can be justified by their reduced garden size and by the heavy concentration of visitors that intensely use those places. The quietest garden was number 6 A. Carvalho and R. Cleto Proceedings of Meetings on Acoustics, Vol. 18, 040001 (2012) Page 6 (Serralves) with L Aeq of 47 dB. This value can be justified by the extensive garden area (18 ha) and its insertion within a residential area distant from large traffic roads. It is possible to observe a maximum variation, on average, of about 15 dB(A) in the various parameters, between the most and the least noisy parks. 4.1.2 Difference values To analyze how gardens/parks can decrease noise, regarding the emergence of extreme noise levels (L A10 ) to the background levels (L A90 ), the comparison of their measured exterior and interior values was done. Regarding exterior values (Figure 5) a difference between those two parameters’ values (L A10 - L A90 ) of about 10 dB was found on average. The largest difference (17 dB) was registered in Park 7 (Ocidental) and the smallest (6 dB) in Garden 1 (República). These differences, measured outside the gardens, can be explained, when high, by the road traffic having a small number of vehicles that makes the background noise relevant; when those differences are small they indicate that the road traffic is more intense, masking the background noise. In the gardens' interior (Figure 6) the difference between the extreme noise levels (L A10 ) and the background levels (L A90 ), is about 5 dB on average. The largest difference (6 dB) was measured in Garden 5 (S. Lázaro) and the smallest (3 dB) in Park 9 (S. Roque). These differences are small; however, the difference in Garden 5 (S. Lázaro) is higher due to the concentrated presence of visitors. The smallest can be justified by the stability and regularity in the measured values because the garden has less visitor use and is placed within a residential area without heavy traffic roads. The values in other gardens are close to the average (5 dBA). It can be expected that without major noisy sound sources in a garden’s interior, this will also be the average difference between its extreme levels (L A10 ) and background noise levels (L A90 ). In three urban parks a great reduction capacity was verified by a difference larger than 9 dB between exterior and interior L Aeq values: Park 7 (Ocidental) (19 dB), Park 6 (Serralves) (17 dB) and Park 8 (Covelo) (11 dB). The gardens with smaller areas show a smaller variation between the exterior L Aeq and the interior L Aeq , such as Garden 1 (República) (3 dB), Garden 2 (Marquês) (4 dB), Garden 4 (Cordoaria) (7 dB), Garden 5 (S. Lázaro) (3 dB) and Garden 10 (Velásquez) (6 dB). Two gardens must be distinguished: numbers 1 and 5 (República and S. Lázaro) show the smallest noise level variation between the exterior and the interior (ǻL A = 3 dB). This can perhaps be explained for garden 5 (S. Lázaro) by its elevated terrain level relative to the surrounding street with the largest traffic (on average about 1 m), and by the large traffic volume that the surrounding streets have. The fact that within the gardens several groups of retired persons are playing cards together is also a factor that can increase the overall L Aeq value in the interior of those gardens. 4.1.3 Gardens behaving as noise barriers Analyzing the noise stability in the gardens’ interiors, with intrusions of emerging noise (from the background noise level) that can have a perturbing effect in its calm soundscape it is possible to check that gardens 6, 7 and 8 (Serralves, Ocidental and Covelo) are the ones that show the largest noise level variations (Table 2). This indicates that these spaces have the potential of reducing those higher intensity and short duration noises. Although the three parks are surrounded by high traffic roads, their land extension seems to allow the reducing of those A. Carvalho and R. Cleto Proceedings of Meetings on Acoustics, Vol. 18, 040001 (2012) Page 7 noises. It is not possible to disregard also the existence of walls in the perimeter of these urban parks functioning, even if partially, as noise barriers. In the opposite extreme are gardens 1, 2 and 5 (República, Marquês and S. Lázaro) which reveal an incapacity of significantly reducing the peak noise levels due to their reduced land extension (less than 1.4 ha), proximity to traffic roads and the inexistence of walled perimeters that could attenuate the outside noise. Note the zero value achieved in the variation regarding the background noise (L A90 ) in park 9 (S. Roque) justified by the relatively constant acoustic environment in its exterior and interior, caused by its insertion in a residential area with low traffic volume. Table 2 - Variations in the average values of: L Aeq , L A10 (extreme values), L A50 (median) and L A90 (background), in the exterior and interior of gardens/urban parks in Oporto, Portugal. Gardens/Parks ǻL Aeq (dB) (=L ext. – L int. ) ǻL A10 (dB) (=L ext. – L int. ) ǻL A50 (dB) (=L ext. – L int. ) ǻL A90 (dB) (=L ext. - L int. ) 1 República 2.9 3.6 3.5 2.6 2 Marquês 4.3 5.1 4.0 2.6 3 Boavista 8.7 9.6 7.9 6.0 4 Cordoaria 6.6 7.4 6.5 3.5 5 S. Lázaro 2.9 3.5 2.6 0.9 6 Serralves 17.3 18.8 13.5 7.9 7 Ocidental 19.3 21.8 16.0 9.5 8 Covelo 11.4 12.4 10.1 6.9 9 S. Roque 5.6 7.3 2.3 0.0 10 Velásquez 5.8 7.0 4.8 3.1 Arithmetic average 8.5 9.7 7.1 4.3 4.2 Questionnaires 4.2.1 Results The sample has 85 inquiries done to visitors of the gardens (66% male) with 26% between 46 and 65 years old and 36% older than 65 years old. This reveals that the common user of the gardens/parks in Oporto is usually adult (only 14% were younger than 18 years old). When questioned if the sounds that they heard were expected (question 1) the majority (66%) answered that they agreed largely or totally, and only 2% said they were indifferent. 13% disagreed (Figure 7). When asked if they liked the sound that they heard (question 2), the majority answered that they agreed (49%), while others showed some indifference (13%) or even some form of annoyance (21%) (Figure 8). When questioned if the “volume” of these sounds caused annoyance (question 3), the majority disagreed strongly (32%) or totally (21%), 16% showed indifference and the remaining 25% were annoyed in some degree by those sounds (Figure 9). When questioned about the tranquility given by the park (question 5), the visitors agreed mildly (19%) or a lot (52%), and 14% disagreed in some form (Figure 10). A. Carvalho and R. Cleto Proceedings of Meetings on Acoustics, Vol. 18, 040001 (2012) Page 8 Figure 7 - Answers to: The sounds that I hear in this park are expected. (1: Disagreement, 7: Agreement) Figure 8 - Answers to: I like the sounds I hear. (1: Disagreement, 7: Agreement) Figure 9 - Answers to: The volume of these sounds bothers me. (1: Disagreement, 7: Agreement) Figure 10 - Answers to: I consider this garden/park tranquil. (1: Disagreement, 7: Agreement)                          Figure 11 - Distribution of the visitors’ global evaluations of the gardens/parks. Visitors were asked to identify three sounds in the garden. Using Schafer classification [3] those sounds are grouped as: traffic (34%), human sounds (28%), natural sounds (10%), birds (24%), machinery (2%), music (0%), signals (2%) and others (1%). A. Carvalho and R. Cleto Proceedings of Meetings on Acoustics, Vol. 18, 040001 (2012) Page 9