The “Amadeu Amaral” Public Library is located in downtown São Carlos, SP, Brazil (22˚S, 48˚W), on the corner of two streets with intense vehicular and foot traffic. This location has high noise levels, which renders the library an inappropriate place for reading and studying. The library has nine employees, and approximately 100 people use the library daily. The library consists of three levels (Figure 1), which will be denoted Floor 1 (134 m²), Floor 2 (128 m²) and Floor 3 (285 m²). Measurements were taken on all three levels. The first floor of the building is on the ground level, and the main library collection is located on the third floor. The building is naturally ventilated through tilted windows and doors and has no ceiling fans or similar air circulation methods. During the study, the windows and doors remained open or closed, depending on the needs of the visitors. The city of São Carlos is situated in the geographical center of the state of São Paulo, in the southeastern region of Brazil. São Carlos is considered to be a medium-sized city (with 230,000 inhabitants, according to the 2010 Brazilian Census).

The temperature, relative humidity, noise, carbon dioxide level and airborne particle concentration (PM₁₀) were measured on each floor and outdoors. These parameters were monitored from May to September 2013. In each week of the sampling period, measurements were randomly taken on three normal workdays (08:00 to 17:00). The indoor sampling instruments were placed on supports 1.5 m above the ground, in the center of the room. The outdoor sampling sites were located between the library and the busiest street.

The temperature and relative humidity were measured every 15 minutes using a digital thermo hygrometer (HygroPalm-0, Rotronic Instrument Corp., New York, USA). CO₂ levels were measured using a continuous multigas monitor with a range of 0 to 5000 ppm (MultiRAE IR PGM-54, Rae Systems Inc., California, USA). Noise was measured using alogging sound meter with a range of 30 to 130 dB (DEC-490, Instrutherm, São Paulo, Brazil). A dust monitor (Aerocet 531, Met One Instruments, Inc., Oregon, USA) was used to measure the PM₁₀.

The monthly average temperature and relative humidity values during the study period are shown in Table 1 and Table 2, respectively. A daily average was obtained using the measurements taken throughout each day, and the monthly average was defined as the arithmetic mean of the daily averages obtained during that month.

The monthly average internal and external temperatures on the three levels were not significantly different. The standard deviation of the outdoor temperature was approximately 3˚C, and the standard deviation of the indoor temperature was approximately 2˚C. These deviations show that the building has little influence on the temperature. The relative humidity inside and outside the building were also very similar, which seems to be characteristic of buildings utilizing natural ventilation.

These results enabled us to evaluate the thermal comfort of the library’s visitors. Thermal comfort is defined as satisfaction with the thermal environment. The factors that influence the thermal comfort levels of the library’s

occupants can be divided into environmental (temperature, thermal radiation, humidity, airspeed) and personal factors (personal activity and condition, clothing).

The literature on thermal comfort usually focuses on humidity and temperature and their influence on personal well-being. One approach to defining a comfort zone and setting appropriate limits is using the standards defined in the ISO 7730-1994 [14] . The ISO 7730 presents a method to predict the degree of discomfort in individuals exposed to moderate thermal environments and specify the environmental conditions that achieve comfort. This standard can be used to design new environments or assess existing spaces.

Figure 2 presents a combination of the daily average temperature and relative humidity over the study period for the purposes of evaluating the thermal comfort level.

Outside the comfort zone, May and June had the lowest temperatures, with daily averages of approximately 18˚C. August had the lowest average daily relative humidity (approximately 20%) and a high average daily temperature (above 27˚C). Approximately 62% of the data lie within the comfort zone.

The monthly average noise levels are calculated in the same manner as the temperature and relative humidity averages and are shown in Table 3. The noise levels are slightly higher on the first floor than on the other two floors. A value of approximately 60 dBA was observed inside the library, which is significantly different from the value of 68 dBA measured outside the library. The standard deviation of the noise level was approximately 6 dBA outside the library and 4 dBA inside the library. The external noise levels reached values up to 92 dBA, while the internal noise levels did not exceed 85 dBA.

According to Willich et al. [10] , subjects exposed to noise levels of more than 60 decibels during the day experienced an increased risk for myocardial infarction compared to subjects with exposure to noise levels below 60 decibels. The currently employed threshold of 85 decibels may sufficiently protect the library’s visitors from

hearing damage but not from cardiovascular risks. Noise is a frequent burden in daily life, particularly in metropolitan areas and work sites. Figure 3 shows a comparison of the average monthly noise levels with the value recommended in the cited study.

Other studies have observed the harmful effects caused by exposure to high noise levels, demonstrating the need for proper standards to preserve public health [15] - [19] .

The average PM₁₀ concentrations during the study period are shown in Table 4 and Figure 4. The averages were calculated in the same manner as the other parameters. The highest outdoor PM₁₀ averages were observed during August and September and were approximately twice as high as the averages during May, June and July. This increased PM₁₀ concentration can partially be explained by decreased rainfall.

The indoor PM₁₀ data exhibited similar behavior, with the highest averages in the driest months, most noticeably on the first and second floors. However, the activities performed within the environment also influence the PM₁₀ concentration. Manipulating objects (books, magazines and newspapers), displacing people and cleaning activities can all generate and resuspend particulate matter and contribute to an increased internal PM₁₀ concentration (i.e., the internal PM₁₀ concentration depends on both the external environment and the internal activities).

The equipment (Hi-Vol) used by Bruno et al. [20] was located in a public square approximately 150 m away from the “Amadeu Amaral” Public Library. Their measured outdoor PM₁₀ concentrations were similar to those obtained in this work. According to the authors, the PM₁₀ concentration in the atmosphere of São Carlos varies seasonally and is higher in the dry period (from May to October) than in the rainy season (from November to April).

The WHO updated its Air Quality Guidelines in 2005 [21] based on a systematic review of the literature on the adverse health effects of air pollution. The current update is intended to be relevant and applicable worldwide and considers the large regional inequalities in air pollution exposure. It recommends guidelines for particulate matter, ozone, nitrogen dioxide and sulfur dioxide and a set of interim targets for these pollutants’ concentrations to encourage a gradual improvement in air quality and a reduction in the number of health impacts. The WHO air quality guidelines for particulate matter (PM₁₀) concentration recommend an annual mean of 20 μg/m³ and a 24-hour mean of 50 μg/m³.

The monthly average particulate matter concentrations (PM₁₀) obtained in this study were calculated from measurements taken over 8 hours. The indoor PM₁₀ was usually below the 50 μg/m³ value recommended by the WHO.

The monthly average CO₂ concentrations in the library are shown in Figure 5. The internal CO₂ levels were slightly greater than the external values, highlighting the fact that the natural ventilation of the building renders the interior air quality similar to the exterior air quality.

The mean daily peak CO₂ concentration was 600 ppm, with a standard deviation of 30 ppm. ASHRAE Standard 62.1-2010 [13] recommends that CO₂ concentrations be kept below 1000 ppm, and, thus, the average concentrations in the library are below the ASHRAE maximum.

In a building with constant occupancy, the daily maximum CO₂ concentration is related to the building's air exchange rate. The relationship between the CO₂ concentration and the air exchange rate was extensively discussed by Persily and Dols [22] .

To understand people’s opinions regarding the air quality of the library, eighteen people (nine users and nine employees) were interviewed. Each interviewee was asked the following two questions: What do you think about the thermal comfort of the library? What symptoms have you experienced because of the air quality of the library?

The answers provided by respondents are summarized in Figure 6 and Figure 7. On hot days, 89% of people think that the library is warm or hot. On cold days, 63% of people think that the library is cool or cold. The most frequently cited symptoms were sneezing, a running nose, headaches, a dry throat and dry eyes. A person could express more than one symptom. More than 55% of the respondents said they had a headache after staying in place, and more than 44% of the respondents had a runny nose and dry throat. Another common complaint from the employees is the high noise level in the library, which often disturbs communication between people and forces them to speak more loudly. Although the measured noise levels were below the recommended value of 85 dBA for an 8-hour workday, the staff felt uncomfortable with the overall level of noise in the library.

According to Witterseh et al. [23] , the exposure to warm temperatures (around 30˚C) and high noise levels (more than 55 dBA) had negative effects on the performance of office work. The overall acceptability of working conditions was lower during exposure to either factor.

Temperature and humidity have a strong and significant impact on the perception of indoor air quality; at a constant pollution level, the perceived air quality decreases with increasing air temperature and humidity [24] . Intensity of fatigue, headache and difficulty in thinking clearly decreased when subjects worked at slightly lower levels of air temperature and humidity [25] . Poor indoor air quality can reduce the performance of office work and there is a relationship between the percentage dissatisfied with IAQ and the measured decrement in performance. Negative indoor environmental effects on performance were accompanied by negative effects on general symptoms such as headache and concentration [26] .