ARTÍCULO ORIGINAL

DOI: 10.17533/udea.redin.n76a07

Influence of precipitation scavenging on the PM_2.5/PM₁₀ ratio at the Kennedy locality of Bogotá, Colombia

Influencia del efecto barrido en la relación PM_2.5/PM₁₀ en la localidad de Kennedy de Bogotá, Colombia

¹ Facultad de Ingeniería Ambiental, Universidad Santo Tomas. Carrera 9 N.º 51-11. C. P. 110311. Bogotá, Colombia.

² Departamento de Salud Pública, Facultad de Salud, Universidad Industrial de Santander. Carrera 32 N.º 29-31. C. P. 680002. Bucaramanga, Colombia.

³ Departamento de Química e Ingeniería Ambiental, Universidad Nacional de Colombia. Carrera 30 N.º 45-03. C. P. 111321. Bogotá, Colombia.

* Corresponding author: Néstor Yezid Rojas Roa, e-mail: nyrojasr@unal.edu.co

 DOI: 10.17533/udea.redin.n76a07

(Received January 14, 2015; accepted May 11, 2015)

Objective: To establish whether the scavenging effect reduces the PM_2.5/PM₁₀ ratio in rainy periods in comparison with dry periods, at the Kennedy locality of Bogotá, Colombia. Materials and methods: Relationships among hourly and daily PM₁₀, PM_2.5, PM_2.5/PM₁₀ ratio, temperature, relative humidity and precipitation records from the Kennedy air quality station from January 2007 to September 2011 were analyzed. Results: The hourly mean PM_2.5/PM₁₀ ratio was 0.36 (SD= ± 0.12), with an hourly maximum of 0.96. In rainy hours, the PM_2.5/PM₁₀ ratio was 0.41 (SD= ± 0.13) and was reduced to 0.36 (DE= ± 0.12) in dry hours. On the other hand, the daily mean PM_2.5/PM₁₀ ratio was 0.36 (SD= ± 0.09) with a daily maximum of 0.79. The daily mean on rainy days was higher (0.39; SD= ± 0.09) than that recorded on dry days (0.34; SD= ±0.08). All these differences were statistically significant. Conclusions: Precipitation reduces PM_2.5 concentrations at a lower extent than it reduces PM₁₀ concentrations. The analysis was not conclusive about the effect of precipitation on PM_2.5 concentrations, so it is not possible to assert that precipitation reduces the risk associated with the exposure to airborne particulate matter. To deepen our knowledge about the effect of precipitation on particulate matter pollution, it is recommended to apply additional techniques such as particle counting and particulate matter chemical characterization.

Keywords: Precipitation, PM_2.5/PM₁₀ ratio, air pollution, scavenging

RESUMEN

Objetivo: Establecer si el efecto barrido disminuye el valor de la relación PM_2.5/PM₁₀ en horas o días lluviosos, en comparación con periodos secos, en la localidad de Kennedy de Bogotá, Colombia. Material y métodos: Se analizaron relaciones entre los registros horarios y diarios de PM₁₀, PM_2.5, relación PM_2.5/PM₁₀, temperatura, humedad relativa y precipitación, de la estación del sistema de vigilancia de calidad del aire de la localidad de Kennedy en Bogotá durante el periodo de enero de 2007 a septiembre de 2011. Resultados: La media horaria de la relación PM_2.5/PM₁₀ fue de 0.36 (DE= ± 0.12), con un máximo horario de 0.96. En horas lluviosas, la relación PM_2.5/PM₁₀ fue de 0.41 (DE= ± 0.13), disminuyendo a 0.36 (DE= ± 0.12) en horas secas. Por otra parte, la media diaria de la relación fue de 0.36 (DE= ± 0.09), con una media máxima diaria de 0,79. La media diaria en días lluviosos fue mayor (0.39; DE= ± 0.09), que la encontrada en días secos (0.34; DE= ± 0.08). Todas las diferencias de medianas fueron estadísticamente significativas. Conclusiones: La precipitación disminuye las concentraciones de PM₁₀, y en menor proporción las de PM_2.5, lo cual tiene como resultado un leve aumento de la relación PM_2.5/PM₁₀ con la precipitación. El análisis no es concluyente en cuanto a la reducción de PM_2.5, por lo cual no es posible afirmar que la precipitación reduce el riesgo a la salud asociado a la exposición a material particulado en el aire ambiente. Con el fin de profundizar sobre el conocimiento del efecto de la precipitación sobre la contaminación por material particulado, se recomienda hacer mediciones con técnicas complementarias, como el conteo y el análisis químico de las partículas.

Palabras clave: precipitación, relación PM_2.5/PM₁₀, contaminación del aire, barrido

1. Introduction

Air pollution is a concerning issue in several cities in Latin America and the Caribbean (LAC), owing to unplanned urbanization processes with a fast growth in population and vehicular fleet, poor maintenance practices and industrial facilities with obsolete technologies [1]. The main air pollutant is particulate matter (PM), measured both as PM₁₀ – particles with an aerodynamic diameter smaller than 10 micrometers – and PM_2.5 – smaller than 2.5 micrometers, or fine PM –. The difference between PM₁₀ and PM_2.5 is referred to as coarse particles. In general, fine and coarse particles have different chemical compositions, are generated by different processes, emitted by different sources, and removed from the atmosphere by different mechanisms [2]. The PM_2.5/PM₁₀ ratio represents the fraction of fine particles, mostly produced by incomplete fuel combustion, and the sum of fine and coarse particles [3]. Particles of different sizes are deposited in different sections of the respiratory tract and have different effects on human health [4]. Fine particles are more harmful than coarse particles because they have a higher fraction of toxic compounds [5] and present a higher surface area, interacting more easily with airway and alveolar cells.

One of the PM removal mechanisms from the atmosphere is the wet deposition or precipitation scavenging, a natural process in which the interaction between hydrometeors (cloud droplets, mist, rain, ice crystals) and particles interact, resulting in particle growth and deposition on the Earth's surface. Interactions include nucleation scavenging and impaction scavenging. Nucleation scavenging involves water condensation over particles that act as condensation nuclei and ice-formation nuclei. Impaction scavenging results from the particle capture by cloud droplets, ice crystals and falling hydrometeors [2]. Precipitation scavenging is the most important sink of PM in the troposphere [6] and coarse particles are removed more efficiently than fine particles through these mechanisms [7].

As it is usual in urban centers around the world, Bogotá and other Colombian cities have records of poor air quality, associated with high concentration levels of PM₁₀ [8]. It has been estimated that point sources in Bogotá emit nearly 1,440 ton/yr of total suspended particles (TSP), and mobile sources emit 1,100 ton/yr of PM_2.5 [9, 10]. The annual PM₁₀ average concentration for the city was 48 µg/m³ in 2012, but Kennedy, the most polluted locality in the SW area the city, had a much higher annual average for the same year, 71 µg/m³. The Kennedy air quality station is the only one in the city equipped with a PM_2.5 monitor for several years, since 2007. The PM_2.5 annual average for 2012 was 28.5 µg/m³ [11]. Both PM₁₀ and PM_2.5 concentration levels in Kennedy are well above the WHO air quality guideline values of 20 µg/m³ and 10 µg/m³, respectively.

In a situation like this, precipitation scavenging is supposed to mitigate the impacts of PM pollution. However, records of Bogotá's health authority have shown that the number of events related with respiratory illness increase during the wet seasons and decrease during the dry seasons. Therefore, it is worthwhile to examine the extent at which the precipitation scavenging effect is occurring.

Relationships among hourly and 24-h records of PM₁₀, PM_2.5 and meteorological parameters such as temperature, relative humidity, wind speed and precipitation at the Kennedy station, from January 2007 to September 2011, were analyzed with the aim to determine the impact of the precipitation scavenging on particulate matter concentrations and the fine/coarse particle ratio. Wet deposition is supposed to have a stronger effect on coarse particles than on fine particles, so the PM_2.5/PM₁₀ should increase during wet periods with respect to dry periods.

2. Methods

2.1. Study location

Bogotá D.C is located at a latitude of 4° 35' 56" a longitude of - 74° 4' 51" and an altitude between 2,500 and 2,800 masl, with an annual mean temperature between 12 ºC and 15 ºC over the urban area. The annual mean temperature is virtually constant throughout the year (Table 1). Monthly precipitation distribution is bimodal, with two rainy periods and two dry periods annually. The first rainy period occurs roughly in April and May and the second in October and November. The dryer periods occur in January-February and July-August. March, June, September and December are transition months [12].

Bogotá's official population in 2011 was 7,467,804 inhabitants, 14% of which (1,019,949) lived in the Kennedy locality. In Kennedy locality 8.5% of the population are children under 4 and 8.67%, adults aged 60 and over [13].

2.2. PM₁₀, PM_2.5 and meteorological variables

Hourly records of PM₁₀, PM_2.5, together with temperature, relative humidity (RH), wind speed, barometric pressure (BP) and precipitation were obtained from the Kennedy station, which is part of Bogotá's Air Quality Monitoring Network. PM₁₀ and PM_2.5 concentrations are measured using Beta attenuation monitors [11].

2.3. Statistical analysis

An exploratory analysis of hourly PM₁₀, PM_2.5 and PM_2.5/PM₁₀ ratio data was carried out to identify and discard outliers. Meteorological variables were also examined to check their quality. 24-h averages were then analyzed for all variables, except precipitation, which was analyzed as a 24-h cumulative variable. Only days with less than 25% of missing data were taken into account for the statistical analysis.

Days with a cumulative precipitation above 0 mm were considered as rainy days. The equivalent criterion was used for rainy hours. Since the pollutant concentrations and meteorological data did not have normal distributions, non-parametric tests were used to establish correlations (Spearman test) and difference of means (Kruskal-Wallis test) between rainy and dry periods. The STATA 11® software application was used to run the tests.

3. Results

3.1. Hourly data

Of the 41,616 hours of the study period, a total of 29,146 valid records of PM₁₀ and PM _2.5 (70%) concentration were counted. The hourly average of PM₁₀ concentration was 87.91 µg/m³ (standard deviation SD = ± 37.86), with a maximum hourly value of 240 µg/m³. For PM_2.5 the hourly average was 31.65 µg/m³ (SD = ± 17.13), with a maximum hourly concentration of 98 µg/m³. The Spearman correlation between PM₁₀ and PM_2.5 was 0.75 showing a statistically significant association between variables. The hourly average of the PM_2.5/PM₁₀ ratio was 0.36 (SD= ± 0.12) with a maximum hourly value of 0.96 (Table 2).

When stratified by rainy or dry hours, from the 29,146 obtained hourly records, 93% (n=27,098) corresponded to dry hours, while the remaining percentage (n=2,048) were rainy hours. In the rainy hours, a higher hourly average value of RH compared with dry hours (RH _rain=78.9%; RH _dry=68.6%) was observed. Temperature showed an inverse relation to hourly RH, which was evidenced in the correlation value (-0.77), which was statistically significant. The finding in the correlation was confirmed with the meteorological variables recorded values, which show that in dry hours the average temperatures was 14.6°C and the average RH was 68.55%, while in rainy hours the average temperature was 13.48°C and RH was 78.91%. All differences between average meteorological variables in the dry and rainy hours, except BP, were statistically significant. The decrease in temperature is one characteristic of the rain formation processes, which can be convective, orographic and frontal or cyclonic; clouds are formed by cooling humid air parcels (masses), which may be caused by an ascent of the parcel or a barometric process. The mass is cooled gradually during the ascent until the dew point temperature is reached at which the water vapor condensation occurs with the subsequent transformation from vapor to tiny droplets of water [15].

The hourly average of the PM_2.5/PM₁₀ ratio in rainy hours was 0.41 (SD= ± 0.13), higher than the one found in dry hours (0.36; SD= ± 0.12). The PM_2.5/PM₁₀ ratio showed a negative correlation with the temperature (-0.36) and positive with the RH (0.44). With an average temperature of 13.48°C, the average value of the PM_2.5/PM₁₀ ratio was 0.41 and decreased to 0.36 with a higher temperature (14.6°C in dry hours). For RH, with a value of 68.6% (dry hours) a PM_2.5/PM₁₀ ratio of 0.36 was obtained, which increased to 0.41 with a RH of 78.9% (Table 2). The correlations and medium difference were statistically significant.

3.2. Daily data

Of the 1,734 days of the study period, a total of 1,189 records of daily average concentration for PM₁₀ and PM_2.5 (69%) were obtained. The daily average concentration of PM₁₀ was 89.9 µg/m³ (SD= ± 25.76), with a maximum daily average value of 183.3 µg/m³. For PM_2.5, the daily average was 32.7 µg/m³ (SD= ± 12.29) with a maximum daily average of 75.1 µg/m³. The Spearman correlation between PM₁₀ and PM_2.5 was 0.74, showing an association between variables, which was statistically significant. The daily average PM_2.5/PM₁₀ ratio was 0.36 (SD= ± 0.09), with a maximum daily Average of 0.79 (Table 4).

When stratified by rainy or dry day, from 1,189 days, 51% (n=610) were dry days, while 49% (n=579) were rainy days. As same as the hourly analysis, it was found that in rainy days there is a higher daily average of RH compared with dry days (RH rainy day=72.9%; RH dry day= 65.5%). Temperature showed a statistically significant inverse relation with daily RH (-0.69) equal to what was found in the hourly analysis. The finding in the correlation was confirmed with the meteorological variables records, in which the daily average temperature in dry days was 14.8°C, with a daily average RH of 65.5%, while in rainy days the daily average temperature was 14.2°C, with a RH of 72.9%; All differences between medium meteorological variables in the dry and rainy days, except BP, were statistically significant.

As in the hourly analysis, the daily average of PM_2.5/PM₁₀ ratio in rainy days was higher (0.39; SD= ± 0.09) that the one found in dry days (0.34; SD= ± 0.08). The PM_2.5/PM₁₀ ratio showed a negative correlation with temperature (-0.34) and positive with RH (0.54). With a temperature of 14.2°C, the value of the PM_2.5/PM₁₀ ratio was 0.39 decreasing to 0.34, with a higher temperature (14.8°C in dry days). For RH with a value of 65.5% (dry day) a PM_2.5/PM₁₀ ratio of 0.34 was found, which increased to 0.39 with a RH of 72.9% (Table 4). The correlations and medium difference were statistically significant.

In order to identify the effect of precipitation on daily concentrations of particulate matter, 120 days for PM₁₀ and PM_2.5 were examined; the percentage reduction was calculated in the concentration of a dry day followed by a rainy day. The daily reduction of PM₁₀ was 17% and for PM_2.5 was 11%, in which the PM_2.5/PM₁₀ ratio for a rainy day was higher than the one for a dry day (rainy day=0.37, dry day=0.35). The difference of medium by Kruskal-Wallis was statistically significant, indicating that the values found in dry and rainy days are different (Table 5).

3.3. Discussion

These results differ from those found by other researchers in the same monitoring site. In Bogotá [3] a study found an average PM_2.5/PM₁₀ ratio of 0.416, 0.290 for rainy conditions and 0.375 for dry conditions. Other monitoring stations localized in different localities of the city, showed a similar behavior. The difference may be due to the monitoring time of the year, since [3] analyzed data from August to October 2005, which are mostly dry, whereas this work analyzed 4 years of data, covering several rainy and dry seasons throughout that period.

The average PM_2.5/PM₁₀ ratio of 0.36 in this study is lower when compared with the values measured in several cities around the world, which are normally between 0.6 and 0.8 [3]. This is presumably due to a high contribution of coarse particulate matter associated with dust resuspension from the roads surrounding the park where the monitoring station is located. These roads regularly have a visible layer of dust and are not cleaned frequently. Other study [16] state that PM_2.5/PM₁₀ ratios higher than 0.60 would be expected under the direct influence of combustion sources, while ratios lower than 0.30 would have a strong influence of resuspended dust or erosion, which corresponds to this case at the Kennedy locality.

Similar studies conducted in other countries show dissimilar results. In Mexico City, researchers [17] found a PM_2.5/PM₁₀ ratio of 0.44 for the rainy season and 0.47 for the dry season, that is, a higher PM_2.5/PM₁₀ ratio during dry periods than during rainy periods, in contrast with the present study. However, Salazar et al. analyzed a 1-year period, from June 1998 to May 1999. Also in México City from 1992 to 1995, a study [18] determined a ratio of 0.8 during rainy season (May to November), versus 0.5 during dry season (December to April), in agreement with the results from the present study.

In Hong Kong, researchers [19] determined a ratio of 0.84 in Summer (April to September) and 0.69 in Winter (October to March), at a station located on the side of an urban road with heavy traffic. The summer months recorded a value of 1912.7 mm of rainfall compared to the winter months which registered a value of 201.5 mm; June was the rainiest month with 814.5 mm and a ratio of 0.83 compared to February where rainfall had a value of 0 mm and a ratio of 0.69. This study showed an apparent scavenging effect over Total Suspended Particles (TSP) and PM₁₀. PST and PM₁₀ concentrations increased when monthly precipitation was lower, whereas PM_2.5 concentrations did not show a significant correlation with rain.

In Thumba, India, a study [20] analyzed the responses of PM concentrations in a tropical coastal station strongly affected by rapid winds and precipitation associated with the Monsoon season, between October 1998 and December 2000. The PM_2.5/PM₁₀ ratio was 0.9 in Winter (December to February), 0.79 in pre-Monsoon (March to May), 0.8 in Monsoon (June to September) and 0.87 in post-Monsoon (October and November), showing a higher ratio in the Winter months with respect to those with higher precipitation in the Monsoon season. High wind speed and relative humidity, as well as torrential rains during the Monsoon reduce PM₁₀ and PM_2.5, whereas low wind speed and dry soil during the dry winter season are related with high PM concentrations, which affect PM_2.5/PM₁₀ ratios and explain the differences with the present results.

In China [7] a study established that coarse particles (PM_>10 and PM_2,5—10) can be washed by rain more efficiently than fine particles (PM_2,5 and PM_1,0) and, thus, wet deposition rates are generally lower for fine particles. This is because the rain scavenging coefficient for particles with a semi-diameter 0.1 < r < 1.0 µm is very low. Furthermore, precipitation moistens the soil, reducing dust resuspension by traffic and other activities.

Temperature and relative humidity (RH) in Kennedy agree with findings for Bogotá and other Colombian cities. A research [21] found that minimum RH is coincident with minimum precipitation and maximum temperature and viceversa. The increase in PM_2.5 during rainy periods may be attributed to an increase in RH. One study [22] found that for particles with an aerodynamic diameter between 0.3 and 1.0 µm, RH can affect the particles' mass more strongly than precipitation, potentially increasing their mass concentration.

Having seen that precipitation in Kennedy reduce PM₁₀ but not PM_2.5 concentrations, it is reasonable to assert that exposure to PM_2.5 during rainy periods could have a stronger effect on morbidity and mortality associated with cardio-respiratory illness in this locality. Coarse particles are deposited more efficiently than fine particles, affecting mainly urban areas near the sources [23]. Daily mortality is more strongly associated with fine particles [24]. Road transport, particularly based on diesel trucks and buses, is the main source accountable for excess mortality and morbidity associated with fine particles in urban areas [25]. In Santiago de Chile, a study [26] found a significant correlation between the number of acute bronchitis hospital visits and environmental factors such as an increase in PM_2.5 and lower ambient temperature. Fine aerosols are more effective than coarse particles in causing respiratory disease and premature deaths, owing to their ability to penetrate deeply into the lungs [27]. Some of the factors that increase morbidity and mortality associated with respiratory illness are biological materials and metals that are part of the fine fraction of PM. Biological material, including bacteria, polen, skin debris, spores and cellulose fragments, are in the coarse fraction, while viruses are in the fine fraction [3]. Studies suggest that metals like Fe, Ni, Cu, Zn and V play a role in causing physiological changes such as inflammatory response or injuries in the respiratory tissue. Metals have been mostly associated with fine particles [28, 29].

One of the strengths of this study was the availability of a large number of data recorded in the air quality network station in Bogotá located in Kennedy locality. A total of 29.146 hourly values and 1,189 daily values of PM₁₀ and PM_2.5 were obtained and made possible to capture the effect of precipitation on the PM_2.5/PM₁₀ ratio. Furthermore, the availability of these data sustained the difference in results with other studies, where monitoring was performed using manual equipment and only daily averages were obtained. In addition, such studies have not had continuous sampling during a complete year, but only on specific days or during a few months, unlike this study, which spanned four years of continuous data. On the other hand, the climate characteristics of Bogotá, represented by an average temperature and temperature range with little changes throughout the year, a marked period of rainfall and its height above sea level, made the obtained results innovative and show an uncommon scenario when compared with studies in other countries, where the presence of seasons is determining for the behavior of air pollutants, such as particulate matter. Therefore, it is necessary to analyze more deeply the physical phenomena that make less evident the wet scavenging of particulate matter when analyzing the available dataset.

Although this analysis is not conclusive about the PM_2.5 reduction due to the scavenging action of rain, it would be of utmost interest conducting studies with analytical techniques that can observe the response of specific characteristics of PM to precipitation in Bogotá, such as the black carbon fraction, the particle size distribution and the particle number concentration. From the health point of view, it is important to conduct epidemiological studies where the precipitation variable is included in the analysis, which would show the effect of this environmental factor in cardio-respiratory morbidity and mortality in tropical countries. This information would help health authorities to make public health decisions to prevent exposure to fine particles.

4. Acknowledgements

The authors would like to thank the Bogotá District Environmental Agency (Secretaría Distrital de Ambiente) for providing the information used in this work. The main author would like to dedicate this work to Orlando Blanco Casteñeda (R.I.P.) and Ana de Jesús Becerra de Blanco, his loving, exemplary and faithful parents.

5. References

1. M. Bell, L. Cifuentes, D. Davis, E. Cushing, A. Telles, N. Gouveia. "Environmental health indicators and a case study of air pollution in Latin American cities". Environmental Research. Vol. 111. 2011. pp. 57-66.         [ Links ]

2. J. Seinfeld, S. Pandis. Atmospheric chemistry and physics: from air pollution to climate change. 2^nd ed. Ed. Wiley. New Jersey, USA. 2006. pp. 1-1232.         [ Links ]

3. N. Rojas, B. Galvis. "Relación entre PM_2.5 y PM₁₀ en la ciudad de Bogotá". Revista de Ingeniería. N.° 22. 2005. pp. 54-60.         [ Links ]

4. M. Berico, A. Luciani, M. Formignani. "Atmospheric aerosol in an urban area measurements of TSP and PM₁₀ standards and pulmonary deposition assessments". Atmospheric Environment. Vol. 31. 1997. pp. 3659-3665.         [ Links ]

5. S. Wang, X. Feng, X. Zeng, Y. Ma, K. Shang. "A study on variations of concentrations of particulate matter with different sizes in Lanzhou, China". Atmospheric Environment. Vol. 43. 2009. pp. 2823-2828.         [ Links ]

6. G. Loosmore, R. Cederwall. "Precipitation scavenging of atmospheric aerosols for emergency response applications: testing an updated model with new real-time data". Atmospheric Environment. Vol. 38. 2004. pp. 993-1003.         [ Links ]

7. X. Feng, W. Shigong. "Influence of different weather events on concentrations of particulate matter with different sizes in Lanzhou, China". Journal of Environmental Sciences. Vol. 24. 2012. pp. 665-674.         [ Links ]

8. Instituto de Hidrología, Meteorología y Estudios Ambientales de Colombia (IDEAM). Informe del estado de la calidad del aire en Colombia. IDEAM. 2012. Bogotá, Colombia. Available on: https://www.siac.gov.co/documentos/DOC_Portal/DOC_Siac/20130626_Informe_calidad_aire_%202007-2010.pdf. Accessed: May 31, 2015.         [ Links ]

9. M. Fandiño, E. Behrentz. Actualización del inventario de emisiones atmosféricas provenientes de fuentes fijas en Bogotá a través de mediciones directas. Universidad de los Andes. Bogotá, Colombia. 2009. Available on: http://biblioteca.uniandes.edu.co/Tesis_12009_primer_semestre/525.pdf. Accessed: May 31, 2015.         [ Links ]

10. P. Rodríguez, E Behrentz. Actualización del inventario de emisiones de fuentes móviles para la ciudad de Bogotá a través de mediciones directas. Universidad de los Andes. Bogotá, Colombia. 2009. Available on: https://biblioteca.uniandes.edu.co/Tesis_12009_primer_semestre/524.pdf. Accessed: May 31, 2015.         [ Links ]

11. Red de Monitoreo de Calidad del Aire de Bogotá (RMCAB). Informe Anual Calidad del Aire de Bogotá Año 2012. RMCAB. Bogotá, Colombia. 2013. pp. 1-176.         [ Links ]

12. Instituto de Hidrología, Meteorología y Estudios Ambientales de Colombia (IDEAM). Estudio de la caracterización climática de Bogotá y cuenca alta del río Tunjuelo. IDEAM. Bogotá, Colombia. 2007. pp. 1-123.         [ Links ]

13. Secretaria de Cultura, Recreación y Deporte. Localidad de Kennedy. Ficha Básica. Bogotá, Colombia. 2008. Available on: http://www.culturarecreacionydeporte.gov.co/observatorio/documentos/localidades/kenedy.pdf. Accessed: May 31, 2015.         [ Links ]

14. Ministerio de Ambiente, Vivienda y Desarrollo Territorial (MAVD). Protocolo para el monitoreo y seguimiento de la calidad del aire. MAVD. Bogotá, Colombia. 2007. Available on: https://www.minambiente.gov.co/images/AsuntosambientalesySectorialyUrbana/pdf/aire/res_2154_021110_manual_diseno.pdf. Accessed: May 31, 2015.         [ Links ]

15. Instituto de Hidrología, Meteorología y Estudios Ambientales de Colombia (IDEAM). Atlas Climatológico de Colombia. IDEAM. Bogotá, Colombia. 2005. pp. 1-217.         [ Links ]

16. X. Querol, C. Alastuey, B. Artiñano, H. Hansson, R. Harrison, E. Buringh, et al. "Speciation and origin of PM₁₀ and PM_2.5 in selected European cities". Atmospheric Environment. Vol. 38. 2004. pp. 6547-6555.         [ Links ]

17. S. Salazar, L. Lemoyne, I. Rosas. "Preliminary results of size distribution airborne particles in Mexico City". Water, Air, & Soil Pollution. Vol. 64. 1992. pp. 649-658.         [ Links ]

18. M. Castillejos, V. Borja, D. Dockery, D. Gold, D. Loomis. "Airborne coarse particles and mortality". Inhalation Toxicology. Vol. 12. 2000. pp. 61-72.         [ Links ]

19. L. Chan, W. Kwok. "Roadside suspended particulates at heavily trafficked urban sites of Hong Kong. Seasonal variation and dependence on meteorological conditions". Atmospheric Environment. Vol. 35. 2001. pp. 3177-3182.         [ Links ]

20. P. Pillai, S. Babu, K. Moorthy. "A study of PM, PM₁₀ and PM_2.5 concentration at a tropical coastal station". Atmospheric Research. Vol. 61. 2002. pp. 149-167.         [ Links ]

21. L. Marín, G. Poveda, H. Moreno. Análisis del ciclo diurno de la humedad relativa en 17 aeropuertos de Colombia. Proceedings of the XVI Seminario Nacional de Hidráulica e Hidrología. Armenia, Colombia. 2004. pp. 1-13.         [ Links ]

22. D. Sisterson, S. Johnson, R. Kumar. "The Influence of Humidity on Fine-Particle Aerosol Dynamics and Precipitation Scavenging". Aerosol Science and Technology. Vol. 4. 1985. pp. 287-300.         [ Links ]

23. B. Gomiscek, H. Hauck, S. Stopper, O. Preining. "Spatial and temporal variations of PM₁, PM_2.5, PM₁₀ and particle number concentration during the AUPHEP-project". Atmospheric Environment. Vol. 38. 2004. pp. 3917-3934.         [ Links ]

24. J. Schwartz, D. Dockery, L. Neas. "Is daily mortality associated specifically with fine particles?". Journal of the Air & Waste Management Association. Vol. 46. 1996. pp. 927-939.         [ Links ]

25. M. Koch. Airborne Fine Particulates in the Environment: A review of Health Effect Studies, Monitoring Data and Emission Inventories. International Institute for Applied Systems Analysis. Interim Report IR-00-004. Laxenburg, Austria. 2000. pp. 1-45.         [ Links ]

26. W. Zamorano, S. Márquez, J. Aránguiz, P. Bedregal, I. Sánchez. "Relación entre bronquiolitis aguda con factores climáticos y contaminación ambiental". Revista médica de Chile. Vol. 131. 2003. pp. 1117-1122.         [ Links ]

27. A. Zakey, M. Abdel, J. Pettersson, M. Gatari, M. Hallquist. "Seasonal and spatial variation of atmospheric particulate matter in a developing megacity, the Greater Cairo, Egypt". Atmósfera. Vol. 21. 2008. pp. 171-189.         [ Links ]

28. D. Costa, K. Dreher. "Bioavailable transition metals in particulate matter mediate cardiopulmonary injury in healthy and compromised animal models". Environ Health Perspect. Vol. 105. 1997. pp. 1053-1060.         [ Links ]

29. K. Dreher, R. Jaskot, J. Lehmann, J. Richards, J. McGee, A. Ghio, D. Costa. "Soluble transition metals mediate residual oil fly ash induced acute lung injury". J Toxicol Environ Health. Vol. 50. 1997. pp. 285-305.         [ Links ]

30. F. Vargas, N. Rojas, J. Pachón, A. Russell. "PM₁₀ Characterization and Source Apportionment at Two Residential Areas in Bogotá". Atmospheric Pollution Research. Vol. 3. 2012. pp. 72-80.         [ Links ]