Effects of fireworks on particulate matter concentration in a narrow valley: the case of the Medellín metropolitan area
The extensive use of fireworks generates large amounts of pollutants, deteriorating air quality and potentially causing adverse health impacts. In Medellín and its metropolitan area, although fireworks are banned during December, their use is widespread during the Christmas season, particularly during the midnight of November 30 (La Alborada) and New Year’s Eve (NYE). It is therefore essential to assess the effects of these celebrations on air quality in the region. Air-quality data from the official monitoring network and a low-cost particulate matter (PM) citizen science project, backscattering intensity (BI) retrievals from a ceilometer network, potential temperature from a microwave radiometer, and information from a radar wind profiler provide an excellent platform to study the spatio-temporal distribution of contaminants resulting from the La Alborada and NYE celebrations. Substantial increases in PM2.5 and PM10 mass concentrations due to La Alborada and NYE, ranging in some cases from 50 to 100 μgm−3, are observed in the Aburrá Valley and particularly in the densely populated communes of Medellín, with most concentration changes corresponding to ultrafine and fine particles. The PM increments resulting from fireworks show almost no increase in the net amount of black carbon in the atmosphere. Ceilometer BI profiles show a substantial change immediately after the La Alborada and NYE midnights, confined to the atmospheric boundary layer (ABL). Strong thermal inversions lead to fairly homogeneous increments in BI within the ABL, lasting until the onset of the convective boundary layer. In contrast, weak thermal inversions lead to rapid dispersion of aerosols, allowing them to episodically escape above the ABL.
KeywordsAtmospheric boundary layer Air quality Atmospheric stability Black carbon Fireworks Particulate matter
The massive use of fireworks during short periods generates large amounts of pollutants in the atmosphere, generally within the atmospheric boundary layer (ABL), causing a deterioration in air quality (Lin 2016). The highest recorded hourly concentration of fine particulate matter (PM2.5) corresponds to exposure to firework plumes, nearly reaching 10,000 μgm−3 (Joly et al. 2010). Fireworks used in national, regional, religious, and military celebrations and parades, as well as in cultural and sporting events, are composed of oxidant and fuel agents and other components such as agglutinants, coloring agents, smoke, and propellants, all of which, after an oxide-reduction reaction produces different pollutants and particulate matter (PM) into the atmosphere. In addition to the health risks associated with the manipulation of fireworks, the potential for wildfire and infrastructure fire ignition, and the distressing effect on wild and domestic animals because of the loud noise, a high concentration of fireworks-related pollutants during short periods could result in severe health impacts, depending on the PM composition (Bach et al. 1975; Salvi et al. 1999; Licudine et al. 2012; Kloog et al. 2013; Watanabe et al. 2016). The pioneering evidence presented by Bach et al. (1975) indicated that a 170% increase in pollutant concentration was associated with a 113% increase in respiratory illness. One of the potential health risks is associated with perchlorate, common in the vicinity of fireworks-manufacturing sites and firework display sites, contaminating the groundwater and the surface water (Sijimol and Mohan 2014). Perchlorate contamination could disrupt the thyroid and has an impact on ecology (Greer et al. 2002; Sijimol and Mohan 2014). Similarly, toxic byproducts appear as a result of atmospheric reactions between metal oxides and organic fuels (Fleischer et al. 1999).
Several authors have described adverse effects on air quality resulting from the use of fireworks as part of popular festivities around the world. Hourly concentrations of PM2.5 associated with the fireworks of Fourth of July, a national US holiday commemorating the Declaration of Independence, have been reported to be considerably higher than those of the preceding and following days, with an average increase of 21 μgm−3 in the 21:00–22:00 local time—LT-period (Seidel and Birnbaum 2015). The Hindu Diwali festival, in India, is characterized by the widespread use of firecrackers constituting a source of primary aerosols, black carbon (BC), organics, and trace gas emission. A two- to threefold increase in pollutants has been measured during these festivities compared with prior days, and coarse particulate matter (PM10) concentration has reached values over 750 μgm−3, more than five times higher than during a regular day (Barman et al. 2008; Yerramsetti et al. 2013). In addition to finding a considerable increment in PM10 (1.3- to fourfold increase), Mandal et al. (2012) also found an important increase in the ambient noise. Adverse meteorological conditions during the Diwali festival, characterized by a decrease in the ABL height and the magnitude of surface winds, also play a role in pollutant concentration associated with fireworks (Singh et al. 2010). The pollution generated during this festival has been reported to increase the concentrations of Ba, K, Sr, Mg, Na, S, Al, Cl, Mn, Ca, and EC by factors of 264, 18, 15, 5.8, 5, 4, 3.2, 3, 2.7, 1.6, and 4.3, respectively (Sarkar et al. 2010). Recently, the induced pollution due to Diwali fireworks has been assessed using in situ monitoring sites and MODIS retrievals (Kumar et al. 2016). The results suggest a nationwide increase in PM, over 50%, during the festival days in comparison with their background concentrations.
In China, the world’s largest fireworks producer, their adverse effects on air quality have been reported not only for large urban areas during the Lantern Festival (Wang et al. 2007), with a sixfold increase in PM2.5 reported during 2006, but also for the rural and suburban areas of the Henan and Shandong provinces (Zhang et al. 2017). During Taiwan’s Lantern Festival, PM10 has been recorded to be higher than 150 μgm−3, with 60% of the mass corresponding to PM2.5, and the concentration of metallic elements has been up to ten times higher than that of the typical values (Tsai et al. 2012). In Girona, Spain, during the Sant Joan fiesta, there have been reports of increased local background PM2.5 concentrations with Sr, K, Ba, Co, Pb, Cu, Zn, Bi, Mg, Rb, Sb, P, Ga, Mn, As, and Ti increasing 86, 26, 11, 9, 7, 5, 4, 4, 4, 4, 3, 3, 2, 2, 2, and 2 times, respectively (Moreno et al. 2010). During the 2006 soccer World Cup celebrations in Milan, Italy, PM10 concentrations increased by approximately 50%, with a 120-times increase in Sr, 22-times in Mg, 12-times in Ba, 11-times in K, and 6-times in Cu (Vecchi et al. 2008). Dutcher et al. (1999) reported concentrations of fine particles over 150 μgm−3 associated with indoor pyrotechnic displays in sporting events, dominated by an increase in K and S, which originate primarily from black powder. In a recent review article, Singh et al. (2019) present compelling evidence of the effect of fireworks on air quality, visibility, and human health, highlighting the fact that pollutant concentrations are usually 2–8 times higher than average during fireworks events, with significant changes in concentrations of elements such as Ba, Cu, Pb, Cr and Sr, and a decrease in visibility by as much as 92%. Singh et al. (2019) also stress the need for more research on the respiratory impacts of fireworks and more evidence from different regions of the world.
In Medellín, Colombia, and its surrounding metropolitan area in the Aburrá Valley, it has become a highly controversial tradition, apparently since 2003, to celebrate the beginning of December and the Christmas season with widespread use of firecrackers and fireworks by a large portion of the population. This tradition has been referred to as “La Alborada,” Spanish for “The Dawn”. The main difference with most of the previously mentioned cases around the globe is that, in Medellín, the use of fireworks is not a centralized event, but rather they are set off by people along and across the valley, with a tendency to be concentrated in the poorest neighborhoods. The generalized use of fireworks has made La Alborada even more unpopular given that the Aburrá Valley, home to approximately four million people, has endured the onset of critical air-quality episodes during the last 5 years as a result of significant anthropogenic emissions and adverse meteorological conditions for vertical dispersion of pollutants. The latter is associated with thin ABLs during certain seasons, as a result of regional scale forcing at the annual and interannual time scales (Herrera-Mejía and Hoyos 2019). In addition to La Alborada fireworks, there is also widespread fireworks use during New Year’s Eve (NYE) celebrations, with significantly less rejection by the community. Fireworks and firecrackers used in Colombia, in general, are made of black powder, which is made of potassium nitrate (the oxidant), charcoal (organic fuel), and sulfur (fuel) in a 75:15:10 mixture (Russell 2008; Mocella and Conkling 2019; Martín-Alberca and García-Ruiz 2014; Peña-Jiménez and Silva-Riaño 2008).
The goal of the present study is to evaluate the effects of fireworks’ use during La Alborada and NYE celebrations on PM concentration in the Aburrá Valley using information from 2015 to 2018, providing evidence to the local governments that, in addition to firework manipulation safety and the negative effects on wildlife and domestic animals, health-related issues associated with air quality also provide grounds for strictly enforcing the recurrent bans on fireworks fabrication, transportation, commercialization, distribution, use, possession, or carry, imposed by the national government (Gaviria-Uribe and Ospina-Martínez 2016), the City of Medellín (Gómez-Barrera 2016; Gaviria-Correa 2015), and the metropolitan environmental authority (Área Metropolitana del Valle de Aburrá, AMVA) (Gaviria-Correa and Elejalde-López 2015). Although from a different cultural background, strict enforcement of fireworks bans has helped to reduce the pollution peaks during the celebration of the Chinese new year, compared with cities without restrictions (Lai and Brimblecombe 2017).
Among the key differences of this work compared with other studies on the impact of fireworks in air quality is the use of multiple datasets of different nature, including records from the official air quality monitoring network, backscattering intensity (BI) retrievals from a ceilometer network, measurements from a dense low-cost PM citizen science network, and other in situ remote-sensing equipment to characterize the atmospheric conditions during La Alborada and NYE. As one of the novel aspects of our work, we assess pollution concentration near the surface and in the vertical, and we take into account the thermodynamic state of the atmosphere to better understand the residence time of pollution after it is emitted. The magnitude of the aerosol increase resulting from fireworks is highly variable from location to location; hence, using relatively sparse monitoring networks to quantify the risks could result in an underestimation of the PM increment. While the Aburrá Valley official PM network is considered dense, the hyper-dense low-cost network allows estimating in detail the probability density function (PDF) of the PM increments, a required input for accurate probabilistic health risks and econometric studies. We also consider socioeconomic aspects of the city of Medellín to assess the use of fireworks and the impact on local air quality. Section “Region of study and data” describes the geography and the socioeconomic characteristics of the region, as well as the different datasets used in this study. Section “Results” presents the results of the data analysis, highlighting the most critical areas within the metropolitan region where the effects are marked. Finally, Section ”Conclusions” presents the most important conclusions of the study.
Region of study and data
In this work, the assessment of the effects of fireworks during La Alborada and NYE on pollutant concentration relies on in situ air-quality measurements from the official Aburrá Valley network and from a citizen science PM monitoring project, as well as on remote-sensing information from elastic ceilometers that serve as a proxy for PM concentration, allowing for the assessment of the vertical distribution of firework pollutants within the ABL. Thermodynamic profiles based on retrievals from a microwave radiometer (MWR), dynamic profiles from a radar wind profiler (RPW), rainfall derived from weather radar reflectivity and in situ gauges, and meteorological information from automatic weather stations are also used to characterize the state of the atmosphere during La Alborada and NYE from 2015 to 2018.
Region of study: geographical context and socioeconomic information
Medellín’s geographical and socioeconomic information for 2018 (Restrepo et al. 2016). Rural communes do not have an official number. Values in italics in the 4th and 5th columns denote communes below the average population density (25610 persons per km2), and above the average QLMDI, respectively
Income > 4 MMS (%)
Doce de Octubre
San Antonio de Prado
Official air-quality monitoring network
Automatic air-quality monitoring network
Citizen science network
Ceilometers provide information regarding the laser-pulse energy that is backscattered by clouds and other atmospheric components, including PM, expressed as the backscattering attenuated coefficient (β) (Emeis et al. 2009; Münkel and Roininen 2010; Kambezidis et al. 2013)s. Lidars and ceilometers have been widely used to detect sand to characterize aerosols in the atmosphere (Muenkel et al. 2004; Morille et al. 2007; Wiegner et al. 2014), including PM from fireworks plumes (Calhoun et al. 2004; van der Kamp et al. 2008; Han et al. 2014). For example, van der Kamp et al. (2008) show evidence of an elevated aerosol plume using ceilometer retrievals, corresponding to a pyrotechnic competition in Vancouver, Canada. In Wuhan, China, the vertical distribution of aerosols emitted by firecrackers on the ground was reported to spread to a height of more than 450 m in the atmosphere (Han et al. 2014).
Retrievals from three Vaisala CL51 ceilometers (910-nm wavelength), installed at different locations inside the valley (see Fig. 1b), are used to assess changes in the vertical BI profile as a result of fireworks use. Ceilometer BI profiles are also useful to detect the height of the ABL, based on the assumption that a significant aerosol concentration reduction occurs at the top of the latter (Hayden et al. 1997; Steyn et al. 1999; Muenkel et al. 2004; Granados-Munoz et al. 2012; Emeis et al. 2012; Stachlewska et al. 2012; Herrera-Mejía and Hoyos 2019). During fair-weather days, BI profiles often provide a good depiction of the ABL evolution within the valley. Typically, the onset of the convective boundary layer (CBL), associated with an unstable atmosphere, generates a considerable reduction in the ceilometer BI. Ceilometer profiles are available continuously (with some missing data periods) since November 2015 in all sites.
Temperature and humidity profiles retrieved using a Radiometrics MP-3000A microwave radiometer (MWR) are used to estimate potential temperature profiles to assess the stability of the lower troposphere. The MWR is located at the top of the SIATA main operation center on the valley floor (see Fig. 1b), approximately 60 m from the surface, and provides profiles of the thermodynamic state of the atmosphere with a variable vertical resolution, depending on the atmospheric layer: 50-m resolution for retrievals in the layer from the surface to 500 m above the surface, 100-m resolution from 500 m to 2 km above the surface, and 250-m resolution from 2 km to 10 km. Profiles are available every 2 min since January 2013 with periods of missing data mostly due to lightning-related damages.
Radar wind profiler
Vertical wind profiles are used to assess the dynamic conditions of the atmosphere and the potential for dispersing pollutant from fireworks in the valley. The vertical wind profiles are obtained from a RAPTOR VAD-BL Doppler radar designed and manufactured by DeTect Inc. The radar relies on the refractive index variations caused by changes in humidity, temperature, and pressure. The Aburrá Valley wind profiler (see Fig. 1b) works at a nominal frequency of 915 MHz, reaching up to 8 km above the surface under high humidity conditions (Lau et al. 2013). The radar wind profiler (RWP) measures the wind profile in different modes with different vertical resolutions. We use two overlapping modes: in the higher resolution mode (60 m), the RWP measures the wind profile from 77 to 3500 m, and in the lower resolution mode (72 m), from 2500 to 8000 m. In this study, we only use the higher resolution mode since the ABL was never higher than 3500 m. The temporal resolution is 5 min and the information from the radar is available since January 2015.
Precipitation estimates from a weather radar are based on a technique described in Sepúlveda (2015) and Sepúlveda and Hoyos (2017) using weather radar and in situ disdrometer and rain gauge information. The technique allows the estimation of precipitation maps over the valley using retrievals from a C-band polarimetric Doppler weather radar operated by SIATA. The radar scanning strategy allows for obtaining precipitation information every 5 min at a spatial resolution of approximately 128 m using a 1∘ tilt plan position indicator sweep; the uncertainty associated with the quantitative precipitation estimates is relatively low in a 120-km radius from the installation site (Sepúlveda 2015).
La Alborada and NYE PM changes
Evidence from official monitoring networks
Monitoring stations 80 and 86 are located in Medellín on the eastern slope, in two residential areas in densely populated communes (see commune area and population in Table 1). From a socioeconomic point of view, these two communes are below average in terms of multi-dimensional quality-of-life indicators and per capita income (see Table 1 for socioeconomic information). Stations 80 and 86 PM2.5 records are available for the 2017 and 2018 La Alborada events; station 3 records are available since 2015 (see Table 2). Station 80 (Fig. 3a) shows a tripling of PM2.5 in less than 2 h during 2018, with concentrations increasing from 25 μgm−3 at 23:00 LT on November 30, to 75 μgm−3 at 01:00 LT on December 1; station 86 presents yet a more drastic increase with a fivefold change from 25 to 125 μgm−3 (see Fig. 3b). On the other hand, stations 80 and 86 do not show any significant increase in PM2.5 concentration during the 2017 La Alborada, although station 86 does show a PM2.5 doubling at approximately 01:00 LT. Station 3, in the municipality of Girardota (population 54219, area 82 km2), does not show an important increase in any of the years since 2015 most likely because of the low population density. The average PM2.5 record considering all available sensors each year, shows a similar behavior, with the largest increase during La Alborada 2014, followed by 2015 and 2018, and no average PM2.5 increase during 2016 and 2017 (Fig. 3g).
Regarding PM10, stations 6, 37, and 38 records are available for the La Alborada events since 2014. Station 38, in a densely populated area in the city of Itagüí (population 279894, area 21.9 m2), shows an increase in PM10 in the range of 25 to 40 μgm−3 during 2015, 2016, and 2018, and a 75 μgm−3 increase during 2014, all associated with La Alborada fireworks (see Fig. 3d). Similar to the behavior of PM2.5, 2017 does not show an increase in PM10 concentration but rather a slight decrease. Stations 37 and 6, located within university campuses, do not show, in general, large relative PM10 increments during La Alborada events. Station 37 does show, however, a PM10 increase of 20 μgm−3 during 2014, coinciding with the year corresponding to the most substantial PM10 increase at station 38. Station 6 is directly influenced by traffic and it is located in Medellín in a commune with the highest socioeconomic conditions in the region (El Poblado, see Table 1), where the use of fireworks is not as widespread as in other communes. Anthropological evidence, exclusively for the City of Medellín, suggests that affluent communities tend to use less fireworks than the average family, with a preference for high-flying, colorful, pyrotechnics, while communities with less income mostly use firecrackers (Toro-Loaiza and Manrique-López 2015). The average PM10 record, considering all available sensors, also shows a similar behavior, with the largest increase during La Alborada 2014, followed by 2018, with a slight decrease during 2017, and no relevant trends during 2016 (Fig. 3h). The observed temporal evolution of the PM10 and PM2.5 concentrations during La Alborada suggests that most of the PM mass increment is a direct result of fireworks and the corresponding fine particles. It is also important to note that the fireworks-related PM dispersed faster during 2014 than during 2018, pointing to the need to assess the state of the atmosphere and its role modulating pollutant dispersion.
According to Figs. 5 and 6, 2016 and 2017 show the lowest increments in PM associated with La Alborada. During 2017, as previously mentioned, a rainfall event discouraged people in the region from using fireworks after midnight and generated below-cloud scavenging, effectively removing aerosols from the lower atmosphere. In different circumstances, during 2016, the use of fireworks in the La Alborada celebrations was considerably reduced due to a tragic event that spontaneously inspired a mourning period in Medellín and most of the metropolitan area: Chapecoense, a small Brazilian football club, reached the Copa Suramericana final to be played against Atlético Nacional 29\(\bigstar \), a team from Medellín. The plane carrying the Brazilian team to the match was en route to Colombia on November 28, 2016, when it crashed near Medellín just 11 miles from the international airport, killing 71 of the 77 people on board. The accident was reported globally by major news outlets1 and sparked tributes from the world of football. The people from Medellín was deeply moved by the tragedy, and as of different solidarity demonstrations with the Chapecoense club members and their families, there was a city-wide call through social media,2 including messages from the major and different personalities, for not using fireworks during La Alborada as a sign of respect for the peopled that perished in the accident (Duque-Suárez 2018). Most local and national news outlets reported on the success of the social campaign,3 also highlighting, citing sources from the public health authority of the city, that there were no injuries during the 2016 La Alborada due to manipulation of fireworks, compared with the number of injuries during La Alborada from 2012 to 2015, which had been 14, 17, 27, and 24. In 2017, the number of injuries was 5, and during 2018, there were 13 fireworks-related accidents. In contrast, 2018 was the year with the largest increments in PM2.5 concentration resulting from the La Alborada fireworks, and the spatial distribution shows the largest anomalies occurring in densely populated areas in Medellín (in the communes of Aranjuez, Belén, and Villa Hermosa).
PM10, PM2.5, PM1, and BC
The average hourly evolution of PM10, PM2.5, BC, and PM1 for December 2018, and the records for the morning of December 18, 2018, show evidence of environmental pollution due to high vehicular emissions, with a peak in PM concentration around 07:00 LT, with a large increase in PM1 and BC concentrations. Figure 10a and b suggest a large increase in the proportion of PM2.5, with the relative average concentration of PM1 changing from 35% (45% for December 18, 2018) at around 03:00 LT to 62% (80%) during the rush hour. The net increment in PM10 is, in average, approximately 50 μgm−3, of which 20 μgm−3 corresponds to the increase in PM1, 14 μgm−3 corresponds to the increase in particles larger than 1 μm and smaller than 2.5 μm, and 16 μgm−3 corresponds to the increment in particles larger than 2.5 μm and smaller than 10 μm. In other words, 7 out of each 10 μgm−3 increment is associated with the increase in ultrafine and fine PM. The BC also increases, in average, from 16% of the total PM10 at around 03:00 LT to approximately 23% at 07:00 LT.
The records during the evenings of November 30, and December 31, 2018, corresponding to La Alborada and NYE fireworks emissions, show a different behavior compared with a regular day, with almost no increase in the net amount of BC in the atmosphere, hence, a decrease in the relative proportion out of the total PM10, from 20 to 25% at 11:00 LT to approximately 10% at 01:00 LT, during both days. This result is different from the evidence presented in Yerramsetti et al. (2013) and Pathak et al. (2015) showing an increase in BC due to firework-related emissions. The relative proportion of ultrafine PM, out of the total PM10, remains relatively constant during La Alborada and NYE at about 45% of the mass. During NYE, 2018, the total increase in PM10 at station 48 is 80 μgm−3, out of which, 50 μgm−3 (62%) corresponds to PM1, 25 μgm−3 (31%) to particles larger than 1 μm and smaller than 2.5 μm, and 5 μgm−3 (7%) to particles larger than 2.5 μm and smaller than 10 μm. Another difference between fireworks-induced PM increments and those resulting from direct combustion engine vehicle emissions is the mass percentage of ultrafine particles; in addition to a higher BC content, fossil-fuel-powered vehicles generate higher ultrafine particle content compared with that of fireworks.
Evidence from the citizen science network
Aerosol vertical profile
The BI from two ceilometers installed on the base of the Aburrá Valley is used to study the vertical structure of the aerosol plume from fireworks during La Alborada and NYE. BI is a proxy for aerosol concentration; large values of BI, excluding those resulting from the presence of clouds, are usually associated with large aerosol loads. Also, atmospheric humidity accentuates the size of hygroscopic particles, enhancing the lower-troposphere ceilometer BI signal (Emeis et al. 2012; Young and Whiteman 2015). As previously mentioned, BI profiles generally depict the marked diurnal cycle of the ABL height. Around midnight, the ABL height at the base of the Aburrá Valley is typically close to 500 m, limiting the available control volume for the pollutants to disperse (Herrera-Mejía and Hoyos 2019). By the time the CBL starts to develop due to the onset of atmospheric instability, around 10:00 LT, the concentration of aerosols near the surface and within the valley decreases.
Cloudiness and rainfall are less likely during NYE than during La Alborada: NYE is during one of the two dry seasons in the Aburrá Valley, while La Alborada is during one of the wet-to-dry transitional seasons. Figure 15 presents the NYE cases from 2014 to 2018, showing a marked increment in BI immediately after midnight during all years, with the most prominent peaks during 2016 and 2018. The BI increase is fairly homogeneous within the ABL and it is prolonged in time; this feature is different from previous reports, using ceilometers, of fireworks plumes that are localized in time and height (Calhoun et al. 2004; van der Kamp et al. 2008). The latter is most likely because, as previously mentioned, fireworks are not part of a centralized event, but their use is widespread in most communes. The vertical distribution of aerosols from fireworks is similar to the reports for Wuhan, China (Han et al. 2014). None of the cases are modulated by rainfall; however, cloudiness does vary from year-to-year, with 2014 and 2017 corresponding to cloudy conditions during NYE and 2016 and 2018 to cloud-free conditions during the night. Aerosols linked to fireworks use quickly disperse within the ABL, occupying the entire control volume. The most salient difference among the years is that, in most cases, the aerosol vertical distribution presents evidence of a marked and steady capping inversion, persistent throughout the night (e.g., during 2016 and 2018), while during other years (e.g., 2014 and 2015), aerosol plumes seem to vertically penetrate above 500 m. This is likely associated with the stability of the lower layers, and it is explored in the following section. As during La Alborada, the nearest PM2.5 monitoring station shows a very similar behavior to ceilometers BI near-surface, with peak hourly concentrations of 61, 63, 160, and 110 μgm−3, during La Alborada 2014, 2015, 2016, and 2018, respectively (PM2.5 at station 25 is not available for the 2017 NYE). Note that the largest BI near the surface is observed during 2016, matching the PM2.5 records.
Similar behavior is observed in the NYE cases presented in Fig. 16 for 2015, 2017, and 2018, with the inversion layer modulating the dispersion and recession of the aerosol concentration curves. The extent of the lower-troposphere stability is considerably greater during 2017 and 2018 than during 2015, with 2017 exhibiting the higher vertical gradient of potential temperature. Again, the average near-surface air temperature records (Fig. 17b), for the different years, show consistently, that the atmosphere was colder near the surface during 2017 compared with during 2015 and 2018, with 2015 showing the warmest conditions among the 3 years, lead to a larger magnitude of the stability during 2017 NYE. BI profiles in Fig. 15d and e, for 2017 and 2018, respectively, also show a distinctly stratified structure in the lower atmosphere, with the aerosol load confined to the lower 500 to 600 m, while in Fig. 15 b, for 2015, shows vertically penetrating aerosol plumes up to 1000–1500 m above the surface, allowing for the dispersion of most aerosols before January 1, 8:30 LT. As previously mentioned, Fig. 7 b shows a slower aerosol load recession during 2017 than that during 2018. This is because of the larger and extended potential temperature gradient during 2017 as shown in Fig. 16d.
Near-surface and vertical profiles of winds
In Medellín and its metropolitan area, the use of fireworks is widespread during the Christmas season, and in particular, during La Alborada and NYE, causing a deterioration in ambient air quality. The effects of La Alborada and NYE fireworks on PM concentration in the Aburrá Valley were assessed using records from the official air-quality monitoring network and a low-cost PM citizen science network, BI retrievals from a ceilometer network, a MWR to characterize the atmospheric stability, and a RWP to evaluate the dynamic structure of the atmosphere. We found significant increases in PM2.5 and PM10 mass concentrations resulting from the use of fireworks associated with the La Alborada and NYE celebrations.
The hourly evolution of the fine and coarse PM for the 2015 to 2018 period, during the La Alborada events, shows PM changes ranging from large concentration increments to almost no perceptible air-quality effects, depending on the location of the monitoring station and the meteorological conditions. Densely populated communes, with relatively low quality-of-life conditions, show hourly PM2.5 concentration increments ranging from 50 to 100 μgm−3. Areas with low population density do not show important PM increments after the La Alborada midnight. Overall, the range of the La Alborada PM2.5 increments is from as low as 7 to as high as 100 μgm−3, with significant variability from year to year depending on meteorological conditions and social factors. Precarious outdoor conditions due to weather events discourage people in the region from using fireworks. During NYE celebrations, the change in concentrations around midnight is more notable than that during La Alborada. NYE hourly anomalies are as high as 150 μgm−3 when compared with the previous day and up to 190 μgm−3 when compared with the concentration levels at approximately 18:00 LT, on December 31.
The high density of the low-cost PM citizen science monitoring network allows one to unequivocally identify the communes with the highest PM increments resulting from fireworks. The spatial distribution of the PM2.5 concentration anomalies during La Alborada shows that the largest PM changes are confined to Medellín, in particular to the eastern hill communes, with the exception of the highest quality of life area. The northeastern Medellín communes show an average increase greater than 42 μgm−3, while the western communes show an average increase of nearly 27 μgm−3. The increase during NYE is greater than that during La Alborada, and it is observed not only in Medellín but also in all the municipalities in the metropolitan area. The median increments relative to the 22:00 LT concentrations, which correspond to a representative value for the entire valley, are 15 and 20 μgm−3 during La Alborada and NYE, respectively. Despite the cited educational campaigns and bans to reduce fireworks use, the year with the largest anomalies in PM2.5 concentration since 2015 is 2018. These anomalies are larger than the increments resulting from fireworks reported in most cities throughout the world.
The observed PM10 and PM2.5 concentration changes during La Alborada and NYE suggest that most of the PM mass increments associated with fireworks correspond to fine particles. The PM increments resulting from fireworks show a different behavior compared to rush hour PM peaks on a regular weekday, with almost no increase in the net amount of BC in the atmosphere. Another difference between fireworks-induced PM increments and those resulting from direct combustion engine vehicle emissions is the mass percentage of ultrafine particles; in addition to higher BC content, fossil fuel–powered vehicles generate a higher content of ultrafine particles compared to that of fireworks.
The vertical structure of the aerosol plumes from fireworks during La Alborada and NYE, and their residence time in the atmosphere, strongly depends on the structure of the ABL, which in turn is modulated by the vertical gradient of the potential temperature. In other words, the prolongation of the nocturnal stable layer determines the aerosol load recession. A pronounced thermal inversion leads to a slower dispersion and recession of the aerosol concentration. Ceilometer BI profiles show a substantial change immediately after the La Alborada and NYE midnights, confined to the first 550 m, which is precisely the height of the ABL at the base of the Aburrá Valley, and acts as a lid limiting the vertical dispersion of the pollutants. Under stagnant conditions, and given the topographic setting of Medellín, aerosols emitted during the use of fireworks are sufficient to significantly increase the PM concentration in the entire ABL. Under strong inversion conditions, aerosols present evidence of a marked vertical gradient, with the increments in BI being fairly homogeneous within the ABL, lasting until the onset of the CBL. Weak thermal inversions lead to a fast dispersion of aerosols, with aerosol plumes from fireworks episodically penetrating above the ABL.
- 1.See reports by CNN and BBC as examples (accessed September 5, 2019):
- 3.See reports by (accessed September 5, 2019):
The authors would also like to acknowledge the use of free software including Python and Matplotlib (Hunter 2007).
This work was financially supported by SIATA (Sistema de Alerta Temprana de Medellín y el Valle de Aburrá) funds, provided by Área Metropolitana del Valle de Aburrá (AMVA), Municipio de Medellín, Grupo EPM, and ISAGEN under the Research and Technology Contract CD511, 2017, with Universidad EAFIT, institution that operates the early warning system. Financial assistance to Laura Herrera, Natalia Roldán, and Alejandra Isaza was also provided by Universidad Nacional de Colombia under the Facultad de Minas graduate scholarship program.
- Austen, K. (2015). Environmental science: pollution patrol. Nature News, 517(7533), 136–138.Google Scholar
- Barman, S.C., Singh, R., Negi, M.P.S., Bhargava, S.K. (2008). Ambient air quality of Lucknow City (India) during use of fireworks on Diwali Festival. Environmental Monitoring and Assessment, 137(1), 495–504.Google Scholar
- Bonney, R., Shirk, J.L., Phillips, T.B., Wiggins, A., Ballard, H.L., Miller-Rushing, A.J., Parrish, J.K. (2014). Next steps for citizen science. Science, 343(6178), 1436–1437.Google Scholar
- Calhoun, R.J., Heap, R., Sommer, J., Princevac, M., Peccia, J., Fernando, H. (2004). Tracking aerosol plumes - lidar, modeling, and in situ measurement. In Carapezza, E (Ed.) Proceedings of SPIE - the international society for optical engineering, (Vol. 5403 pp. 683–694), DOI https://doi.org/10.1117/12.539910.
- Castell, N., Dauge, F.R., Schneider, P., Vogt, M., Lerner, U., Fishbain, B., Broday, D., Bartonova, A. (2017). Can commercial low-cost sensor platforms contribute to air quality monitoring and exposure estimates? Environment International, 99, 293–302.Google Scholar
- Conrad, C.C., & Hilchey, K.G. (2011). A review of citizen science and community-based environmental monitoring: issues and opportunities. Environmental Monitoring and Assessment, 176(1), 273–291.Google Scholar
- Duque-Suárez, L.F. (2018). El mes de la parranda: El papel de la música parrandera en el valle de aburrá durante las festividades decembrinas. Master’s thesis, Universidad de Antioquia - Medellín.Google Scholar
- Gaviria-Correa, A. (2015). Decreto 1084 de 2015 de la alcaldí,a de Medellín, gaceta oficial no. 4336. https://normograma.info/medellin/normograma/docs/pdf/d_alcamed_1804_2015.pdf. accessed: August 30, 2019.
- Gaviria-Correa, A., & Elejalde-López, H.D. (2015). Resolució,n 18 de 2015 del área metropolitana del valle de aburrá, gaceta oficial no. 4332. https://normograma.info/medellin/normograma/docs/pdf/a_amva_0018_2015.pdf. accessed: August 30, 2019.
- Gaviria-Uribe, A., & Ospina-Martínez, M.L. (2016). Circular conjunta externa 47 de 2016 del ministerio de salud y protecció,n social de colombia: Más conciencia, cero pólvora, prende la fiesta sin pólvora. https://www.medellin.gov.co/normograma/docs/pdf/circular_minsaludps_0047_2016.pdf. accessed: August 30, 2019.
- Gómez-Barrera, S. (2016). Decreto 1765 de 2016 de la alcaldí,a de Medellín, gaceta oficial no. 4410. https://normograma.info/medellin/normograma/docs/pdf/d_alcamed_1765_2016.pdf. accessed: August 30, 2019.
- Granados-Munoz, M.J., Navas-Guzmán, F., Bravo-Aranda, Ja, Guerrero-Rascado, J.L., Lyamani, H., Fernández-Gálvez, J., Alados-Arboledas, L. (2012). Automatic determination of the planetary boundary layer height using lidar: One-year analysis over southeastern Spain. Journal of Geophysical Research: Atmospheres, 117, 1–10. https://doi.org/10.1029/2012JD017524.Google Scholar
- Greenfield, S.M. (1957). Rain scavenging of radioactive particulate matter from the atmosphere. Journal of Meteorology, 14(2), 115–125. https://doi.org/10.1175/1520-0469(1957)014.0115:RSORPM.2.0.CO;2.Google Scholar
- Greer, M.A., Goodman, G., Pleus, R.C., Greer, S.E. (2002). Health effects assessment for environmental perchlorate contamination: the dose response for inhibition of thyroidal radioiodine uptake in humans. Environmental Health Perspectives, 110(9), 927–937. https://doi.org/10.1289/ehp.02110927.Google Scholar
- Han, G., Gong, W., Quan, J.H., Li, J., Zhang, M. (2014). Spatial and temporal distributions of contaminants emitted because of Chinese New Year’s Eve celebrations in Wuhan. Environ Sci: Processes Impacts, 16, 916–923.Google Scholar
- Hayden, K.L., Anlauf, K.G., Hoff, R.M., Strapp, J.W., Bottenheim, J.W., Wiebe, Ha, Froude, Fa, Martin, J.B., Steyn, D.G., McKendry, I.G. (1997). The vertical chemical and meteorological structure of the boundary layer in the Lower Fraser Valley during Pacific ’93. Atmospheric Environment, 31(14), 2089–2105. https://doi.org/10.1016/S1352-2310(96)00300-7.Google Scholar
- Herrera-Mejía, L., & Hoyos, C.D. (2019). Characterization of the atmospheric boundary layer in a narrow tropical valley using remote-sensing and radiosonde observations and the WRF model: the Aburrá Valley case-study. Quarterly Journal of the Royal Meteorological Society 0(0) https://doi.org/10.1002/qj.3583.Google Scholar
- Joly, A., Smargiassi, A., Kosatsky, T., Fournier, M., Dabek-Zlotorzynska, E., Celo, V., Mathieu, D., Servranckx, R., D’amours, R., Malo, A., Brook, J. (2010). Characterisation of particulate exposure during fireworks displays. Atmospheric Environment, 44(34), 4325–4329.Google Scholar
- Kambezidis, H.D., Paliatsos, A.G., Kappos, N., Kasselouri, B. (2013). A case of african dust transport over athens captured by a ceilometer. Berlin: Springer.Google Scholar
- van der Kamp, D., McKendry, I., Wong, M., Stull, R. (2008). Lidar ceilometer observations and modeling of a fireworks plume in Vancouver, British Columbia. Atmospheric Environment, 42(30), 7174–7178.Google Scholar
- Kloog, I., Ridgway, B., Koutrakis, P., Coull, B.A., Schwartz, J.D. (2013). Long- and short-term exposure to PM2.5 and mortality: using novel exposure models. Epidemiology (Cambridge, Mass), 24(4), 555–561.Google Scholar
- Kumar, M., Singh, R.K., Murari, V., Singh, A.K., Singh, R.S., Banerjee, T. (2016). Fireworks induced particle pollution: a spatio-temporal analysis. Atmospheric Research, 180, 78–91.Google Scholar
- Leukauf, D., Gohm, A., Rotach, M.W., Wagner, J.S. (2015). The impact of the temperature inversion breakup on the exchange of heat and mass in an idealized valley: sensitivity to the radiative forcing. Journal of Applied Meteorology and Climatology, 54(11), 2199–2216.Google Scholar
- Mocella, C., & Conkling, J. (2019). Chemistry of pyrotechnics: basic principles and theory, 3rd edn. Boca Raton: CRC Press.Google Scholar
- Moreno, T., Querol, X., Alastuey, A., Amato, F., Pey, J., Pandolfi, M., Kuenzli, N., Bouso, L., Rivera, M., Gibbons, W. (2010). Effect of fireworks events on urban background trace metal aerosol concentrations: is the cocktail worth the show? Journal of Hazardous Materials, 183 (1), 945–949.Google Scholar
- Morille, Y., Haeffelin, M., Drobinski, P., Pelon, J. (2007). STRAT: an automated algorithm to retrieve the vertical structure of the atmosphere from Single-Channel lidar data. Journal of Atmospheric and Oceanic Technology, 24(5), 761–775.Google Scholar
- Münkel, C., & Roininen, R. (2010). Automatic monitoring of boundary layer structures with ceilometer. Vaisala, Tech rep.Google Scholar
- Pathak, B., Biswas, J., Bharali, C., Bhuyan, P.K. (2015). Short term introduction of pollutants into the atmosphere at a location in the Brahmaputra Basin: a case study. Atmospheric Pollution Research, 6(2), 220–229.Google Scholar
- Peña-Jiménez, Y.Z., & Silva-Riaño, R.E. (2008). Determinación del impacto ambiental al recurso agua ocasionado por la desactivación de los explosivos pólvora y anfo con el método de disolución química y valoración del ruido producido por la destrucción de los explosivos incautados por la policía nacional de Colombia. Master’s thesis, Universidad de Lasalle.Google Scholar
- Restrepo, P.P., Hernández, P.A., Roldán, L.M. (2016). Informe de calidad de vida de medellín 2016. Medellín Cómo Vamos, Tech rep.Google Scholar
- Roldán-Henao, N., Hoyos, C.D., Herrera-Mejía, L., Isaza, A. (2019). An investigation of the precipitation net effect on pollutant concentration in a narrow valley: role of lower troposphere stability. Journal of Applied Meteorology and Climatology In revisions.Google Scholar
- Russell, M. (2008). The chemistry of fireworks. The Royal Society of Chemistry. https://doi.org/10.1039/9781847558916.
- Salvi, S., Blomberg, A., Rudell, B., Kelly, F., Sandström, T., Holgate, S.T., Few, A. (1999). Acute inflammatory responses in the airways and peripheral blood after short-term exposure to diesel exhaust in healthy human volunteers. American Journal of Respiratory and Critical Care Medicine, 159(3), 702–709.Google Scholar
- Sarkar, S., Khillare, P.S., Jyethi, D.S., Hasan, A., Parween, M. (2010). Chemical speciation of respirable suspended particulate matter during a major firework festival in India. Journal of Hazardous Materials, 184(1), 321–330.Google Scholar
- Schneider, P., Castell, N., Vogt, M., Dauge, F.R., Lahoz, W.A., Bartonova, A. (2017). Mapping urban air quality in near real-time using observations from low-cost sensors and model information. Environment International, 106, 234–247.Google Scholar
- Seidel, D.J., & Birnbaum, A.N. (2015). Effects of Independence Day fireworks on atmospheric concentrations of fine particulate matter in the United States. Atmospheric Environment, 115, 192–198.Google Scholar
- Sepúlveda, J. (2015). Estimación cuantitativa de precipitación a partir de la información de radar meteorológico del área metropolitana del valle de aburrá. Master’s thesis, Universidad Nacional de Colombia - Sede Medellín.Google Scholar
- Sepúlveda, J., & Hoyos, C.D. (2017). Disdrometer-based C-band radar quantitative precipitation estimation (QPE) in a highly complex terrain region in tropical Colombia. AGU Fall Meeting Abstracts.Google Scholar
- Silvertown, J. (2009). A new dawn for citizen science. Trends in Ecology & Evolution, 24(9), 467–471.Google Scholar
- Singh, D.P., Gadi, R., Mandal, T.K., Dixit, C.K., Singh, K., Saud, T., Singh, N., Gupta, P.K. (2010). Study of temporal variation in ambient air quality during Diwali festival in India. Environmental Monitoring and Assessment, 169(1), 1–13.Google Scholar
- Steyn, D.G., Baldi, M., Hoff, R.M. (1999). The detection of mixed layer depth and entrainment zone thickness from lidar backscatter profiles. Journal of Atmospheric and Oceanic Technology, 16(1979), 953–959. https://doi.org/10.1175/1520-0426(1999)016.0953:TDOMLD.2.0.CO;2.Google Scholar
- Toro-Loaiza, D.M., & Manrique-López, L.V. (2015). La vigencia del consumo de la pólvora en la ciudad de Medellín. Master’s thesis, Universidad Pontificia Bolivariana.Google Scholar
- Tsai, H.H., Chien, L.H., Yuan, C.S., Lin, Y.C., Jen, Y.H., Ie, I.R. (2012). Influences of fireworks on chemical characteristics of atmospheric fine and coarse particles during Taiwan’s Lantern Festival. Atmospheric Environment, 62, 256–264.Google Scholar
- Vecchi, R., Bernardoni, V., Cricchio, D., D‘Alessandro, A., Fermo, P., Lucarelli, F., Nava, S., Piazzalunga, A., Valli, G. (2008). The impact of fireworks on airborne particles. Atmospheric Environment, 42(6), 1121–1132.Google Scholar
- Wang, Y., Zhuang, G., Xu, C., An, Z. (2007). The air pollution caused by the burning of fireworks during the lantern festival in Beijing. Atmospheric Environment, 41(2), 417– 431.Google Scholar
- Watanabe, M., Noma, H., Kurai, J., Sano, H., Hantan, D., Ueki, M., Kitano, H., Shimizu, E. (2016). Effects of short-term exposure to particulate air pollutants on the inflammatory response and respiratory symptoms: a panel study in schoolchildren from rural areas of Japan. International journal of environmental research and public health 13(10).Google Scholar
- Whiteman, C.D. (2000). Mountain meteorology: fundamentals and applications. Oxford: Oxford University Press. https://doi.org/10.1659/0276-4741(2001)021[0095:MMFAA]2.0.CO;2.Google Scholar
- Wiegner, M., Madonna, F., Binietoglou, I., Forkel, R., Gasteiger, J., Geiß, A, Pappalardo, G., Schäfer, K., Thomas, W. (2014). What is the benefit of ceilometers for aerosol remote sensing? an answer from earlinet. Atmospheric Measurement Techniques, 7(7), 1979–1997. https://doi.org/10.5194/amt-7-1979-2014.Google Scholar
- Yerramsetti, V.S., Sharma, A.R., Gauravarapu Navlur, N., Rapolu, V., Dhulipala, N.S.K.C., Sinha, P.R. (2013). The impact assessment of Diwali fireworks emissions on the air quality of a tropical urban site, Hyderabad, India, during three consecutive years. Environmental Monitoring and Assessment, 185(9), 7309–7325.Google Scholar
- Zhang, J., Yang, L., Chen, J., Mellouki, A., Jiang, P., Gao, Y., Li, Y., Yang, Y., Wang, W. (2017). Influence of fireworks displays on the chemical characteristics of PM2.5 in rural and suburban areas in Central and East China. Science of the Total Environment, The, 578, 476–484.Google Scholar
Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.