Abstract
Aerosols (PM\(_{2.5}\) and PM\(_{10}\)) represent one of the most critical pollutants due to their negative effects on human health. This research analyzed the relationship of PM and its PM\(_{2.5}\)/PM\(_{10}\) ratios with climatic variables in the austral spring (2016–2018) in Metropolitan Lima. Overall, there was an average PM\(_{2.5}\)/PM\(_{10}\) ratio of 0.33 with fluctuations from 0.30 to 0.35. However, there have also been high point values that reached ratios greater than one. This situation indicates a moderate condition of contamination by particulate matter with a predominance of coarse aerosols in spring, with an increasing trend over the years. The locations Ate and Villa Maria del Triunfo, especially Ate, presented poor quality conditions. Thursdays showed outstanding pollution peaks by PM\(_{10}\), and a decrease is visible on Sundays. On the other hand, the PM\(_{2.5}\) showed a similar pattern every day, including Sundays. The maximum peaks occurred in the morning and night hours. The increase in anthropogenic emissions associated with the formation of secondary aerosols has been evident, being the case of the location Campo de Marte, the one that had a significant increase in ratios PM\(_{2.5}\)/PM\(_{10}\), which confirms a greater intensity of secondary formations of carbonaceous particles from industrial oil sources, vehicle exhaust, as well as aerosols from metal smelting and biomass burning. There were negative correlations of the ratios with PM\(_{10}\), temperature, wind speed, and direction, and positive correlations with PM\(_{2.5}\) and relative humidity. Contour lines were successfully developed that demonstrated the interaction of climate with PM\(_{2.5}\)/PM\(_{10}\) ratios. This will deepen the exploration of emission sources and modeling, which allows for optimizing air quality indices to control emissions and adequately manage air quality in Metropolitan Lima.
Similar content being viewed by others
Introduction
Deteriorating air quality has been linked to respiratory and circulatory problems experienced by residents1. It is estimated that each year, exposure to air pollution causes 7 million premature deaths and result in the loss of millions more healthy years of life2. Fine particles penetrate directly through the respiratory tract, increasing the risk of mortality over time. This not only causes infections but also promotes lung cancer, diseases such as asthma and cardiovascular diseases2,3. Arregocés et al. predicted up to 1.95% mortality due to prolonged exposure to PM\(_{2.5}\) for regions of Colombia favoured by winds from the industrial zone4. According to Han et al., there is a synergistic effect between pollution sources and meteorological variables, such as temperature, relative humidity, wind speed and intensity on a seasonal basis that influences the contributions to mineral aerosols5. Likewise, topographic and land use factors are added, coastal areas are subject to the permanent influence of the marine flow and the continental breeze, masses of marine air that transport coarse sea salt to the continental part during the day6. Zhang et al., studied the sources and contributions of PAHs linked to the PM\(_{2.5}\)/PM\(_{10}\) ratio as an indicator of air pollution for the winter and summer periods in Hotan (China)7. Zhang et al., identified vehicle emissions, coal combustion and biomass as the main emissions from local sources whose local trajectory is favoured by air masses during winter7. The study of PM\(_{2.5}\)/PM\(_{10}\) fractions helps to understand the physical and chemical characteristics of aerosols and natural or anthropogenic emission sources. This research helps to evaluate their spatiotemporal distribution and the effects they have on the environment and human health8. In China, PM\(_{2.5}\)/PM\(_{10}\) ratios > 0.5 are related to anthropogenic sources, coming from coal energy consumption or heavy industries that contribute higher proportions of PM\(_{2.5}\)9. Franzin et al., calculated ratios of 0.33 to 0.47 for the city of Sao Paulo and related them to agricultural and industrial activities10. Likewise, the use of PM\(_{2.5}\)/PM\(_{10}\) ratios has demonstrated the effectiveness of air quality management or the lack of control when increasing annual trends in the ratios are evident11; indicating the increase in anthropogenic emissions6. Therefore, it is important to deepen research to understand the relationship that exists between the sources of particulate matter emissions and the climate; explain the relationship between PM\(_{10}\) and PM\(_{2.5}\) in an urban environment subject to diverse microclimates, geographies, populations and industries, among others3.
Air pollution in the urban areas of Lima (Peru) is fundamentally due to the vehicle fleet, the open burning of biomass, the poor disposal of solid waste, and industrial emissions12. The city of metropolitan Lima has a population of ten million inhabitants13; this is considered one of the most polluted cities in Latin America14. Population growth coincides with the increase in air pollution15. Among the main polluting agents are PM\(_{10}\) (less than 10 microns) and PM\(_{2.5}\) (less than 2.5 microns) particles, which exert negative effects on human health16,17. The concentrations of these aerosols vary according to the seasonality of the year depending on meteorological conditions18. Steenland et al., studied the effect of maximum ambient temperatures associated with PM\(_{2.5}\) on cardiorespiratory mortality of patients admitted to emergency rooms in Lima19. Velasquez et al., reported 2300 premature deaths from particulate matter in Lima and almost 3,000 deaths from biomass burning20.
This situation reflects a serious PM pollution problem and the Peruvian state maintains the Air Quality Monitoring Network in the Lima—Callao Metropolitan Area, developed by Servicio Nacional de Meteorología e Hidrología del Perú (SENAMHI). The institution reported from 2016 to 2018 an increase in PM\(_{2.5}\) and PM\(_{10}\) values, especially during hot periods and stable weather conditions21. This extensive network includes three key urban districts such as Ate (located in Lima Este), Villa María del Triunfo (VMT) (located in Lima Sur) and Campo de Marte (CDM) (located in the centre of Lima). These districts were home to 1 million 200 thousand inhabitants in 2018. Its location and topography is associated with very varied micro-climates, as well as a growing demographic expansion, extensive commercial demand, internal demand and transportation. Likewise, there have been incidents of air pollution that have worsened the situation, which is favoured by the stability of weather conditions and emissions from the vehicle fleet. The National Information System for Disaster Prevention and Attention recorded 2,924 fire incidents from 1993 to 2018 only in Lima, followed by Ate (334 reports), VMT (125 reports) and to a lesser extent Jesús María (25 reports)22.
A practical way to evaluate air quality and emission sources is the application of the PM\(_{2.5}\)/PM\(_{10}\) ratio because it can provide a relevant source of information on the causes of pollution. The processes that develop and affect the atmospheric environment. When the PM\(_{2.5}\)/PM\(_{10}\) ratio is low < 0.5, the predominance of course PM\(_{10}\) particles from natural sources is considered23. However, if the proportion is high then the emission sources are anthropogenic24. Some researchers have demonstrated the relationship of PM\(_{10}\) and PM\(_{2.5}\) with meteorological variables in certain areas of Lima12,18,25, but the PM\(_{2.5}\)/PM\(_{10}\) relationship that exists between both polluting agents in the meteorological conditions of Lima has not yet been studied. Previously, Pereira et al. demonstrated that in CDM, PM\(_{10}\) were made up of organic and elemental carbon, monosaccharide, PAHs, oxy- and nitro-PAHs, water-soluble ions; and Ni that comes from metal smelting and oil combustion sources. The researcher also demonstrated that the presence of marine and vehicular aerosols in an area of green areas was due to the environmental transport of contaminants from secondary formations26.
The objective of this research is to provide an updated environmental management instrument based on the PM\(_{2.5}\), PM\(_{10}\), PM\(_{2.5}\), PM\(_{10}\) relationship and meteorological variables using historical data from three key districts of Metropolitan Lima located in strategic locations of the great megacity. The main contributions of this study are described:
-
Application of air quality standards, current national air quality indexes and world health organization guidelines on PM\(_{2.5}\) and PM\(_{10}\) levels to assess the health risk status of inhabitants of metropolitan Lima.
-
Evaluation of the spatio-temporal variability of aerosols using the PM\(_{2.5}\)/PM\(_{10}\) indicator in synergy with PM\(_{2.5}\), PM\(_{10}\) and meteorological variables to identify pollution sources and the seasonal influence of austral spring (2016–2018).
-
Implementation of the analysis of dispersion, correlational, and contour lines to demonstrate the interaction between meteorological variables and PM\(_{2.5}\)/PM\(_{10}\) ratios by identifying emission sources.
The rest of the paper is organized as follows. Sections presents the “Methods” that detail the methodology applied in evaluating the ratios, relations with the meteorological variables, and identification of the emission sources. Then, Section outlines show the “Results and discussion”, which details the relevant research findings and discusses the theoretical approaches and historical background of other studies. Finally, Sections provide the “Conclusions” of this study, with recommendations for future research related to the application of indices based on PM\(_{2.5}\)/PM\(_{10}\) ratios.
Methods
Study area
The characteristics of PM\(_{2.5}\) and PM\(_{10}\) and their ratios (PM\(_{2.5}\)/PM\(_{10}\)) have been investigated in three districts of the Metropolitan network of Lima (Peru). These districts are located in strategic places in Lima: Ate, VMT and CDM. These are subject to different microclimates and a population of approximately one million inhabitants. The monitoring sites are classified as places of intense urban traffic, however CDM is a fully paved commercial area with the presence of a green area, Ate is further from the coast, has industries and less paved land, it is a district adjacent to the central highway where more than 7 thousand vehicles circulate daily towards the interior of the country. On the other hand, VMT is desert in nature, with sandy and poorly paved soils, located near the South Pan-American Highway, one of the main roads with intense national and international traffic. The CDM station is located in the Jesús María District at 117 meters above sea level with coordinates: 12\(^\circ\)4ʹ14.03″ AL—77\(^\circ\)2ʹ35.3″ WL. The Ate station is located in the district of the same name at 362 meters above sea level with coordinates: 12\(^\circ\)1ʹ34″ SL—76\(^\circ\)55ʹ7″ WL and the VMT District station is located at 272 meters above sea level with coordinates: 12\(^\circ\)9ʹ59.01″ SL and 76\(^\circ\)55ʹ11.99″ WL. Figure 1 shows the position chart used in this research, during the austral spring 2016–2018.
Air quality monitoring stations and data collection
The monitoring period covered the austral spring season (September 22 to December 22) from 2016 to 2018. Data on PM\(_{2.5}\) and PM\(_{10}\) concentrations, ambient temperature (°C), relative humidity (RH, %) have been processed, wind speed (WS; m/s) and wind direction (WD; degrees), the data have been provided by the SENAMHI Air Quality and Meteorology Monitoring network. Only the VMT site does not have WS and WD data. The research has considered the comparison of PM\(_{2.5}\) and PM\(_{10}\) data with the reference values of the WHO Guide27 for PM\(_{2.5}\) (10 \(\upmu\)g/m\(^{3}\) annual average; 25 \(\upmu\)g/m\(^{3}\) average of 24 hours) and PM\(_{10}\) (20 \(\upmu\)g/m\(^{3}\) annual average and 50 \(\upmu\)g/m\(^{3}\) 24-hour average). Additionally, PM data have been compared to current national air quality standards (NAAQS: PM\(_{2.5}\)=50 \(\upmu\)g/m\(^{3}\); PM\(_{10}\)=100 \(\upmu\)g/m\(^{3}\)) and National Air Quality Indices (NAQI). The NAQI is calculated using the following expressions:
where: I(PM\(_{10}\)) and I(PM\(_{2.5}\)) express the calculated NAAQI and the value inside the parentheses corresponds to the observed PM (PM\(_{10}\), PM\(_{2.5}\)). Table 1 shows the national NAAQI classification. Furthermore, the application of the PM\(_{2.5}\)/PM\(_{10}\) ratios has been evaluated as a new air quality management instrument that can be implemented to optimize pollution control and the identification of the main emitting sources of PM\(_{2.5}\) and PM\(_{10}\).
Statistical analysis
In this research, 124,488 data provided by SENAMHI have been analysed. This amount of information corresponds to 90.5% of the data on PM\(_{2.5}\), PM\(_{10}\) and meteorological parameters (T, HR, WS, WD), since VMT does not have WS and WD data. To process the information, the R Studio program was used and to analyse the descriptive and inferential statistics of the PM and climate data, the Pearson correlation coefficient and a multivariate Gaussian regression analysis were applied considering the p-value for analyse the level of significance (p < 0.05). The dispersion of the PM\(_{2.5}\)/PM\(_{10}\) ratios was analysed and contour isolines with Z axis were applied to analyse the bivariate effect of meteorological conditions on the PM\(_{2.5}\)/PM\(_{10}\) ratio (supplement). Python’s Windrose package was used to show how wind speed and direction are normally distributed at a given location.
Results and discussion
Concentrations of PM\(_{2.5}\), PM\(_{10}\), PM\(_{2.5}\)/PM\(_{10}\) ratios and meteorological variables during the austral spring
Table 1 presents a general statistical summary of the concentrations of particulate matter (PM\(_{2.5}\) and PM\(_{10}\)), meteorological variables and PM\(_{2.5}\)/PM\(_{10}\) ratios recorded during the austral spring 2016-2018 (September 22 to December 22) in the three sampling sites Ate, VMT and CDM.
National Ambient Air Quality Standards (NAAQS) of Peru (50 \(\upmu\)g/m\(^{3}\) for PM\(_{2.5}\) and 100 \(\upmu\)g/m\(^{3}\) for PM\(_{10}\)). T=atmospheric temperature (\(^\circ\)C); RH=relative humidity (%); WS=wind speed (m/s); WD=wind direction (degrees).
The average concentrations of PM\(_{2.5}\) during 2016-2018 fluctuated from 3 to 262.4 \(\upmu\)g/m\(^{3}\) (Ate) and for PM\(_{10}\) the value ranged from 6 to 503.90 \(\upmu\)g/m\(^{3}\) (Ate). Ate has stood out for maintaining the maximum PM values, compared to the other sites: Ate > VMT > CDM. Table 1 also shows the percentage of PM values above the NAAQS. Regarding PM\(_{10}\), this sequence has been recorded: Ate (24%) > VMT (6%) > CDM (0.26%), while for PM\(_{2.5}\): Ate (62%)> VMT (20%)> CDM (5.5%). On the other hand, 97% of the PM\(_{2.5}\) values and 71% of the PM\(_{10}\) values in Ate exceeded the WHO guideline values for 24-h measurements, showing a high risk for human health27. Likewise, the general average of the PM\(_{2.5}\)/PM\(_{10}\) ratio was 0.33 (range: 0.30 in CDM - 0.35 in Ate). The values were similar to those reported in other regions of the world for warm periods such as in Italy (0.38) considering that this proportion varies seasonally with increases in cold periods28. In Bahrain (0.353) where the contribution of PM\(_{2.5}\) emissions from densely located local industries stands out29. In contrast, ratios > 0.5 have also been reported7,30. A PM\(_{2.5}\)/PM\(_{10}\) ratio equal to 0.5 has previously been used as a threshold value for air pollution from PM emission sources7,31. In this research, the general average of the PM\(_{2.5}\)/PM\(_{10}\) ratios is less than 0.5, however 10.4% of the ratios > 0.5 followed the following order: Ate (16%, Max: 5.53) > VMT (14%, Max: 3.29) > CDM (11%, Max: 2.92). Previously, Valdivia reported ratios of 0.65 considering the winter periods of greatest pollution32.
Regarding the meteorological variables, Fig. 2 shows the average values, with a temperature range of 18 to 19 °C and a typical rise from September to December 22 (beginning of summer). The HR was similar for Ate and CDM (81% and 83% respectively) and somewhat higher for VMT (87%). This situation was associated with weak and moderate wind flows (averages < 5 m/s) in the morning, moderate to strong in the afternoon and weak winds at night; due to persistent atmospheric stability that limits vertical dispersion with horizontal transport directions in a northerly, north-easterly direction.
Interannual distribution of PM\(_{2.5}\), PM\(_{10}\) and PM\(_{2.5}\)/PM\(_{10}\) ratio during the austral spring (2016–2018)
Table 2 summarizes the interannual results of PM\(_{2.5}\) and PM\(_{10}\) concentrations during the spring of 2016 to 2018 at the selected sampling sites. The results of the annual averages summarize the percentage of PM>NAAQS concentrations in the following order: (i) PM\(_{10}\): Ate (60–65%) > VMT (12–29.5%) > CDM (1.5–10%) (ii) PM\(_{2.5}\): Ate (20–24%) > VMT (4–8%) > CDM (0–0.6%). It is important to highlight that Ate is located on both sides of the country’s central highway, where vehicular traffic has increased33. Likewise, VMT has a sandy desert soil with a road saturated with public and private transportation that includes large cargo trucks34. On the other hand, CDM is located in the district of Jesús María, this is a 100% paved district, and is mainly affected by the automobile fleet35. It has been shown that sources of air pollution are concentrated in urban areas such that winds transport these pollutants to surrounding areas36.
According to Table 2, in spring the ratios remain from 0.31 to 0.37 except for CDM which in 2016 presented the lowest ratio of all (PM\(_{2.5}\)/PM\(_{10}\) = 0.2), indicating a lower contribution of PMs to the atmosphere.
It is important to highlight those previous studies carried out on the chemical composition of PM\(_{10}\) in CDM showed contributions from the sea breeze (sulphates, nitrate, ammonium and chlorides). In addition, to terrestrial contributions from direct vehicle emissions, the resuspension of road dust, metal smelting, oil combustion, oil refineries, PAHs emissions and secondary organic carbon formations, due to combustion of biomass and solid waste26.
Regarding the NAAQI calculated for the PM, the results showed the PM\(_{2.5}\) levels of poor quality in the following order: Ate (70.5%) > VMT (36.4%) > CDM (11%) and levels of PM\(_{10}\) of moderate quality in the following order: Ate (67%) > VMT (44%) > CDM (15%) (Fig. 3). Attempts to optimize air quality management have been related to the 4-year strategic plan (2021–2025), to reduce air pollution within the Lima-Callao Metropolitan Area, through the management of emission sources. This agrees with the National Environmental Policy37. However, in practice there are still no positive results for the protection of human health and improvement of quality of life.
Spatiotemporal variations of PM\(_{2.5}\), PM\(_{10}\), PM\(_{2.5}\)/PM\(_{10}\) ratio and meteorological variables
Figure 4 shows the hourly concentrations of PM\(_{10}\) during the week, generally showing two very well-marked daily peaks around 08:00 hours and 10:00 hours Ate develops a first peak around 08:30 hours (194.29 \(\upmu\)g/m\(^{3}\)) that decreases towards the afternoon (14:00 hours–16:00 hours), due to the increase in wind intensity (maximum: 3.63 m/s). The wind disperses and decreases PM concentrations and then a second peak (maximum: 126.76 \(\upmu\)g/m\(^{3}\)) occurs around 22:00 hours as the WS decreases. In VMT the first peak occurs at 07:30 hours until 10:00 hours (maximum: 114.5 \(\upmu\)g/m\(^{3}\)). Then the wind disperses the particles around 13:00 hours38, until the second peak is formed from 20:00 to 22:00 hours (maximum: 87.14 \(\upmu\)g/m\(^{3}\)). For its part, CDM produces peaks at 08:30 hours and 20:00 hours (maximum averages of 74.29 \(\upmu\)g/m\(^{3}\) and 78.17 \(\upmu\)g/m\(^{3}\)). The decreases in concentrations are recorded in the early hours of the morning (around 03:00 hours) and in the afternoon (around 16:00 hours), associated with strong winds (maximum 8.5 m/s). This is because part of the coarse material is re-suspended dust, these would be carried by the wind in the early hours of the morning, when atmospheric instability and the greatest vertical and horizontal movement of the wind occur11. Aerosols increase during the night due to the lower intensity of the night wind (PM\(_{10}\): 69.33 \(\upmu\)g/m\(^{3}\), 19:00 hours).
In relation to PM\(_{2.5}\), certain similarities were also shown in the study areas, with an increasing trend from midnight and early morning towards (from 08:00 hours to 10:00 hours). In Ate a peak appears from 08:00 hours to 09:30 hours (maximum: 63.15 \(\upmu\)g/m\(^{3}\)), then a decrease in values occurs until 17:00 hours when the new cycle begins. In VMT the maximum peak (39.54 \(\upmu\)g/m\(^{3}\)) occurs around 10:00 hours and in CDM at 08:30 hours (maximum: 19.9 \(\upmu\)g/m\(^{3}\)), although a second is also observed peak in CDM around 22:00 hours (19.01 \(\upmu\)g/m\(^{3}\)) (Fig. 4b–d). Obviously, vehicular traffic activity increases in the morning hours and to a lesser extent after 20:00 hours. The vehicle fleet plays a relevant role in increasing PM concentrations11.
On the other hand, the monthly variation (Fig. 4e,f) showed a certain decrease in PM between October and November, favoured by higher relative humidity, and higher wind intensities (in November; Ate: 1.26 m/s; CDM: 2.78 m/s) compared to the months of December21. Regarding the days of the week (Fig. 4g,h), these were divided into weekdays (Monday to Friday) and weekends (Saturday and Sunday). In Ate and VMT on Thursdays a peak of higher average concentration of PM\(_{2.5}\) develops (127 \(\upmu\)g/m\(^{3}\) and 83.7 \(\upmu\)g/m\(^{3}\) respectively); associated with the supply of goods and services to industries, businesses and urban transportation, while on Sundays the decrease in values is observed. In CDM, a decrease in PM\(_{10}\) values was also observed over the weekend (Sunday: 54.6 \(\upmu\)g/m\(^{3}\)). These findings reaffirm that the highest PM contents during the day occur in the warmest hours11, but spring is an intermediate season where a climate that ascends from cold to warm temperatures can be observed. During this period, the rise of the thermal inversion layer occurs, this favours atmospheric instability, without significant contributions of humidity and increases in ambient temperature dispersing the pollutant25.
Scatter plot, correlation and determination factor
A scatter plot of PM\(_{10}\) versus PM\(_{2.5}\) at the monitoring sites using hourly data from the austral spring from 2016 to 2018 is shown in Fig. 5. Scatter plots show 7 colours represented by blue for ratios < 0.2, light blue for ratios < 0.4, green for ratios < 0.6, mustard for ratios < 0.8; brown for ratios < 1 and red for ratios > 1, these form fan-shaped distributions. According to Munir et al.11, the blue region suggests the predominance of coarse aerosols related to higher concentrations of PM\(_{10}\) that include material generated by mechanical routes related to the construction industry. Lima has industries producing non-metallic minerals, cement factories, limestone, dolomite and concrete extraction, especially in Ate and VMT. On the other hand, from beige to red, high values of the PM\(_{2.5}\)/PM\(_{10}\) ratio (>0.8) are observed, these would be composed mainly of carbonaceous species from contributions of fossil fuels, organic carbon from secondary emissions from the burning of biomass. This coincides with what was previously reported by Pereyra et al.26 for CDM and coincides with what was found in other cities where direct combustion sources predominate or present pollutants formed in secondary reactions in the atmosphere, sulphate and nitrate ions are also included in this group11.
The application of the Pearson correlation statistical method and multivariate Gaussian regression is then evaluated to evaluate the existing relationships. Figure 6 shows the correlation between the PM\(_{10}\), PM\(_{2.5}\), PM\(_{2.5}\)/PM\(_{10}\) ratio; temperature (°C), relative humidity %), wind speed (m/s), wind direction (degrees) per hour during the austral spring season (2016–2018) in the districts of Ate, VMT, CDM.
Regarding the correlations for PM and meteorological variables, our results disagree with the findings of Sirithian and Thanatrakolsri3, for rural and urban cities in Taiwan, since no strong correlations were established for PM and RH/T. The PM\(_{10}\)-RH/T correlations for Ate are the only consistent values (rPM\(_{10}\)-T: 0.1603; rPM\(_{10}\)-RH: – 0.17) and although the other areas did not reflect appreciable correlations and linear adjustments (r < 0.1; R\(^{2}\) < 0.03), the correlations observed for T and RH with PM\(_{10}\) were always inverse, VMT also showed a weak PM\(_{10}\)-T correlation (r: – 0.1796). Lima is a very particular place and in this analysis the topographic aspect has not been taken into account. The presence of hills or slopes, of basins (Chillón and Rímac) capable of confining humidity and winds from the sea, promote the appearance of local microclimates that cause chaotic behaviour with atmospheric pollutants39. Likewise, the correlation analysis for PM\(_{2.5}\) did not produce significant values. Regarding the correlations of the PM\(_{2.5}\)/PM\(_{10}\) ratios with RH and T, the values were inverse, however only Ate and CDM showed weak correlations (Ate rPM\(_{2.5}\)/PM\(_{10}\)-T = – 0.1179 and Ate rPM\(_{2.5}\)/PM\(_{10}\)-RH = 0.1068; CDM rPM\(_{2.5}\)/PM\(_{10}\)-T = 0.0937 and CDM RH-ratio = – 0.1191).
Likewise, wind flows, at the local level only Ate presented a weak correlation for PM\(_{10}\)-WS (r= 0.162). It is important to highlight that in this research 90.5% of data validated by SENAMHI was used due to the absence of information about WS and WD in VMT. Despite this, at a general level, better negative correlations have been established for PM\(_{10}\), PM\(_{2.5}\) and WS (rPM\(_{10}\)-WS: – 0.3363; rPM\(_{2.5}\)-WS: – 0.3168) and positive correlations for WD (rPM\(_{10}\)-WD: 0.2343; rPM\(_{2.5}\)-WD: 0.2166).
This indicates that wind is the variable that maintains a relevant role in the environmental transport, dispersion and removal of PM through dry or humid atmospheric deposition processes. Positive correlations suggest atmospheric transports from non-local sources, while negative correlations would indicate local dust re-suspension. Munir et al.11, points out that in arid areas that receive little rain (and this occurs in Lima) high wind speed and temperature promote wind turbulence and the re-suspension of dust and sand particles and increase the concentrations of particles in the atmosphere40. These findings indicate that it is possible to relate RH and T with PM\(_{10}\) at the local level, and that the PM\(_{2.5}\)/PM\(_{10}\) ratio can be used in Ate and CDM to correlate with RH and T. The method also serves to establish correlations between PM\(_{10}\) and PM\(_{2.5}\) with the WS and WD at the local level.
For PM\(_{10}\)–PM\(_{2.5}\)/PM\(_{10}\), negative correlations were obtained for the three sampling sites with values that ranged from r= -0.2585 to r= -0.396 (average r: -0.1898). The relationship for PM\(_{2.5}\)–PM\(_{2.5}\)/PM\(_{10}\) was higher (r: 0.5439 to 0.6828) and presented a direct average correlation coefficient equal to 0.5397. When applying the determination factor R\(^{2}\) (Fig. 7) for PM\(_{2.5}\) and PM\(_{2.5}\)/PM\(_{10}\), significant and moderate values were produced (R\(^{2}\)Ate = 0.2039–0.5483; R\(^{2}\)VMT = 0.3893–0.5612, R\(^{2}\)CDM = 0.267–0.4367; p < 0.01), the biggest factors corresponded to the 2016.
This finding is important because it establishes in a practical and reliable way the identification of the sources that put the health of the inhabitants of the area at risk. Previously, Pereira et al.26, characterized PM\(_{10}\) in CDM, however, they warned about the formation of secondary particles in the atmosphere. Likewise, Huyen et al.41, pointed out that fine aerosols (PM\(_{2.5}\)) that include ultrafine PM come from a significant contribution of secondary organic carbon, so the growth of these particles depends on the amount of surface water molecules present in the aerosol (associated to the relative humidity), since relative humidity exerts a different effect on the concentrations of fine and coarse particles42. In general, the levels of PM\(_{2.5}\) and PM\(_{10}\) observed in the southern spring of 2016-2018 were observed in the towns of VMT, CDM and Ate in Metropolitan Lima. To evaluate the health risk status of the inhabitants of the big city, the current standards that regulate air quality in the country, and the WHO guideline values, have been used.
Likewise, the spatiotemporal variability of the particulate material has been evaluated and the correlation coefficient (r) and the determination factor R\(^{2}\) have been applied to establish relationships of meteorological variables such as HR, T, WS and WD with the aerosol concentration. With this objective, the PM\(_{2.5}\)/PM\(_{10}\) ratio has been incorporated as a new indicator whose values allow identifying the sources of pollutant emissions, whether primary or secondary. This study has allowed us to better understand the local and macro-level climatic influence on the distribution of pollutants and their sources of pollution. The findings have been supported with previous information from Ilizarbe-Gonzáles et al.43, who analysed the physicochemical composition of PM\(_{10}\) samples in a district in the north of Lima and in San Juan de Lurigancho (eastern Lima) in autumn 2017 and identified the polluting sources that affect air quality. Also Pereira et al.26, analysed the physicochemical composition of PM\(_{10}\) particles and the presence of secondary and primary aerosols such as marine aerosols, demonstrating the influence of wind transport on the load of fine material towards CDM. Other complementary studies related to PM modelling14,20,32,44 and others related to the epidemiological effect12,45 have been reviewed. These indicators with meteorological variables can be applied locally by regional governments and can be applied to evaluate levels of air pollution and risk to human health. Likewise, the sources of polluting emissions can be identified, which will help optimize air quality control and management instruments.
Contour lines are shown below to represent the distribution of PM\(_{2.5}\)/PM\(_{10}\) ratios associated with the climate variables. Figure 8 shows for Ate an increasing distribution between relative humidity and the PM\(_{2.5}\)/PM\(_{10}\) ratio over the years between 2016 and 2018 (mean ratios: 0.33 to 0.37), but not temperature, which presented a significant decrease: in Ate (– 0.81\(^\circ\)C), in CDM (– 1.22\(^\circ\)C) and VMT (– 0.92\(^\circ\)C), developing inverse distributions between T\(^\circ\)C and HR. Figures 9 and 10 shows a dense distribution of wind direction between 200 and 250 degrees for Ate and CDM indicating the horizontal transport of air masses towards the north and northeast of Lima. Likewise, with respect to the intensity of these winds, the higher speeds dispersed the aerosols and, conversely, in more stable conditions, the PM\(_{2.5}\)/PM\(_{10}\) ratios increase due to the predominance of secondary aerosols. A positive correlation of the wind speed and direction with PM\(_{10}\) indicating greater primary aerosol transport. These directions are possible sources of coarse particles. On the other hand, CDM receives lower intensity winds, favouring the transport of fine PM\(_{2.5}\) particles that reflect higher PM\(_{2.5}\)/PM\(_{10}\) ratios (dominated by road traffic or other combustion sources).
Limitations
Strong but weak correlations were not established for PM\(_{10}\), and the meteorological variables of RH and T. Likewise, no significant correlation values were obtained for PM\(_{2.5}\) with the meteorological variables. This is because this research has not considered other variables such as the topographic aspect due to the presence of two basins and hills that generate local microclimates. It is also important to highlight that in this research 90.5% of data validated by SENAMHI was used due to the absence of information on WS and WD in VMT.
Conclusions
The concentration of airborne particles has been shown to increase during morning and evening peak hours, driven by the intensification of vehicle fleet activities and anthropogenic activities. The highest PM10 contents are observed with greater emphasis on Thursdays at 22:00 hours and decrease on Sundays, while PM\(_{2.5}\) remains practically constant throughout the week.
The dominant type of aerosol distributed in the three sampling sites has been identified using the color scatter plot method of the PM\(_{2.5}\)/PM\(_{10}\) data. The contribution of the non-metallic mineral production industry stood out as one of the sources, followed by natural contributions and the resuspension of dust. Likewise, the presence of PM\(_{2.5}\) has been classified, composed of carbonaceous species from fossil fuel sources, secondary formations and biomass burning.
Weak direct correlations for temperature and inverse correlations for relative humidity have been established with PM\(_{10}\) at the local level, but no significant correlation with PM\(_{2.5}\) has been achieved. However, correlations improved for PM\(_{10}\) and PM\(_{2.5}\), as wind flows influence environmental transport, dispersion and removal of PM. The correlation of PM\(_{10}\) and PM\(_{2.5}\) with PM\(_{2.5}\)/PM\(_{10}\) produced significant moderate and strong correlations respectively, which makes it possible to identify the emission sources and the levels that put the health of the inhabitants of the area at risk.
These indicators can be applied locally by local governments and can also work at a macro level, which will help optimize air quality management and control instruments.
Data availability
The collection and statistical processing of the data was carried out under the authorization of Servicio Nacional de Meteorología e Hidrología del Perú, is a specialized technical agency of the Peruvian State that provides information on weather forecasting, as well as scientific studies in the areas of hydrology, meteorology, and environmental issues. The datasets are available in the repository, https://www.senamhi.gob.pe/site/descarga-datos/.
References
Kowalska, M. et al. The relationship between daily concentration of fine particulate matter in ambient air and exacerbation of respiratory diseases in Silesian agglomeration, Poland. Int. J. Environ. Res. Public Health 16, 1131 (2019).
WHO. New who global air quality guidelines aim to save millions of lives from air pollution. Tech. Rep. (World Health Organization, 2021).
Sirithian, D. & Thanatrakolsri, P. Relationships between meteorological and particulate matter concentrations (PM\(_{2.5}\) and PM\(_{10}\)) during the haze period in urban and rural areas, northern Thailand. Air Soil Water Res. 15, 11786 (2022).
Arregocés, H. A., Rojano, R. & Restrepo, G. Health risk assessment for particulate matter: Application of airq+ model in the northern Caribbean region of Colombia. Air Qual. Atmos. Health 1, 1–16 (2023).
Han, J. et al. Characterization of PM\(_{2.5}\) mass in relation to PM\(_{1.0}\) and PM\(_{10}\) in megacity Seoul. Asian J. Atmos. Environ. 16, 2021124 (2022).
Spandana, B., Rao, S. S., Upadhya, A. R., Kulkarni, P. & Sreekanth, V. PM\(_{2.5}\)/PM\(_{10}\) ratio characteristics over urban sites of India. Adv. Space Res. 67, 3134–3146 (2021).
Zha, H., Wang, R., Feng, X., An, C. & Qian, J. Spatial characteristics of the PM\(_{2.5}\)/PM\(_{10}\) ratio and its indicative significance regarding air pollution in Hebei province, China. Environ. Monit. Assess. 193, 1–12 (2021).
Usman, F. et al. Exploring the mass concentration of particulate matter and its relationship with meteorological parameters in the hindu-kush range. Atmosphere 13, 1628 (2022).
Zhao, D., Chen, H., Yu, E. & Luo, T. PM\(_{2.5}\)/PM\(_{10}\) ratios in eight economic regions and their relationship with meteorology in China. Adv. Meteorol. 2019, 1–10 (2019).
Franzin, B. T. et al. Characterization of atmospheric aerosol (PM\(_{10}\) and PM\(_{25}\)) from a medium sized city in São Paulo state, Brazil. J. Environ. Sci. 89, 238–251 (2020).
Munir, S. et al. Analysing PM\(_{2.5}\) and its association with PM\(_{10}\) and meteorology in the arid climate of Makkah, Saudi Arabia. Aerosol Air Qual. Res. 17, 453–464 (2017).
Velásquez, R. M. A. & Lara, J. V. M. Gaussian approach for probability and correlation between the number of covid-19 cases and the air pollution in Lima. Urban Clim. 33, 100664 (2020).
Vasquez-Apestegui, B. V. et al. Association between air pollution in Lima and the high incidence of covid-19: Findings from a post hoc analysis. BMC Public Health 21, 1–13 (2021).
Rojas, J. P. et al. Effects of covid-19 pandemic control measures on air pollution in Lima metropolitan area, Peru in South America. Air Qual. Atmos. Health 14, 925–933 (2021).
Peláez, L. M. G. et al. Air quality status and trends over large cities in South America. Environ. Sci. Policy 114, 422–435 (2020).
Tapia, V. et al. Time-series analysis of ambient PM\(_{2.5}\) and cardiorespiratory emergency room visits in Lima, Peru during 2010–2016. J. Exposure Sci. Environ. Epidemiol. 30, 680–688 (2020).
DavilaCordova, J. E. et al. Association of PM\(_{2.5}\) concentration with health center outpatient visits for respiratory diseases of children under 5 years old in LimaPeru. Environ. Health 19, 1–6 (2020).
Cordova, C. H. et al. Air quality assessment and pollution forecasting using artificial neural networks in metropolitan Lima-Peru. Sci. Rep. 11, 24232 (2021).
Steenland, K., Vu, B. & Scovronick, N. Effect modification by maximum temperature of the association between PM\(_{2.5}\) and short-term cardiorespiratory mortality and emergency room visits in Lima, Peru, 2010–2016. J. Exposure Sci. Environ. Epidemiol. 32, 590–595 (2022).
Velásquez, R. M. A. Insulation failure caused by special pollution around industrial environments. Eng. Fail. Anal. 102, 123–135 (2019).
SENAMHI. Boletín mensual vigilancia de la calidad del aire Lima metropolitana. Tech. Rep. 16 (Servicio Nacional de Meteorología e Hidrología del Perú, 2016).
de Lima, M. Plan de contingencia ante incendios en la provincia de Lima 2022–2023. Tech. Rep. (2022).
Tian, Y. et al. Variability of depolarization of aerosol particles in the megacity of Beijing: Implications for the interaction between anthropogenic pollutants and mineral dust particles. Atmos. Chem. Phys. 18, 18203–18217 (2018).
Fan, H., Zhao, C., Yang, Y. & Yang, X. Spatio-temporal variations of the PM\(_{2.5}\)/PM\(_{10}\) ratios and its application to air pollution type classification in China. Front. Environ. Sci. 9, 692440 (2021).
Cabello-Torres, R. J. et al. Statistical modeling approach for PM\(_{10}\) prediction before and during confinement by covid-19 in south Lima, Perú. Sci. Rep. 12, 16737 (2022).
Pereira, G. M. et al. A comparative chemical study of PM\(_{10}\) in three latin american cities: Lima, Medellín, and São Paulo. Air Qual. Atmos. Health 12, 1141–1152 (2019).
WHO. Who air quality guidelines for particulate matter, ozone, nitrogen dioxide and sulfur dioxide global update 2005. Tech. Rep. (World Health Organization, 2005).
Maciejewska, K. Short-term impact of PM\(_{25}\), PM\(_{10}\), and pmc on mortality and morbidity in the agglomeration of Warsaw, Poland. Air Qual. Atmos. Health 13, 659–672 (2020).
Coskuner, G., Jassim, M. S. & Munir, S. Characterizing temporal variability of PM\(_{2.5}\)/PM\(_{10}\) ratio and its relationship with meteorological parameters in Bahrain. Environ. Forensics 19, 315–326 (2018).
Kaleta, D. & Kozielska, B. Spatial and temporal volatility of PM\(_{2.5}\), PM\(_{10}\) and PM\(_{10}\)-bound b [a] p concentrations and assessment of the exposure of the population of Silesia in 2018–2021. Int. J. Environ. Res. Public Health 20, 138 (2022).
Xu, L. et al. Spatiotemporal characteristics of PM\(_{2.5}\) and PM\(_{10}\) at urban and corresponding background sites in 23 cities in China. Sci. Total Environ. 599, 2074–2084 (2017).
Valdivia, S. A. P. Análisis temporal y espacial de la calidad del aire determinado por material particulado PM\(_{10}\) y PM\(_{25}\) en Lima metropolitana. Anales Científicos 77, 273–283 (2016).
Carbo-Bustinza, N. et al. A machine learning approach to analyse ozone concentration in metropolitan area of Lima, Peru. Sci. Rep. 12, 22084 (2022).
Añaños-Valladares, L. et al. Variation of air quality in four districts of Lima in the period of the covid-19 pandemic. in 2022 Congreso Internacional de Innovación y Tendencias en Ingeniería (CONIITI), 1–4 (IEEE, 2022).
de Jesus Maria, M. Plan anual de evaluacion y fiscalizacion ambiental 2018 (Tech. Rep, Municipalidad de Jesus Maria, 2018).
Strosnider, H., Kennedy, C., Monti, M. & Yip, F. Rural and urban differences in air quality, 2008–2012, and community drinking water quality, 2010–2015-United States. MMWR Surv. Summ. 66, 1 (2017).
MINAM. Política nacional del ambiente al 2030. Tech. Rep. (Ministerio del Ambiente, 2021).
SENAMHI. Senamhi pronostica condiciones normales en la primavera 2018. Tech. Rep. (Servicio Nacional de Meteorología e Hidrología del Perú, 2018).
Pacheco, P. & Mera, E. Relations between urban entropies, geographical configurations, habitability and sustainability. Atmosphere 13, 1639 (2022).
Lancaster, N. On the formation of desert loess. Q. Res. 96, 105–122 (2020).
Huyen, T. T. et al. Characteristics of chemical components in fine particles (PM\(_{2.5}\)) and ultrafine particles (PM\(_{0.1}\)) in Hanoi, Vietnam: a case study in two seasons with different humidity. Water Air Soil Pollut. 232, 183 (2021).
Phairuang, W., Hongtieab, S., Suwattiga, P., Furuuchi, M. & Hata, M. Atmospheric ultrafine particulate matter (PM\(_{0.1}\))-bound carbon composition in Bangkok, Thailand. Atmosphere 13, 1676 (2022).
Ilizarbe-Gonzáles, G. M. et al. Chemical characteristics and identification of PM\(_{10}\) sources in two districts of Lima, Peru. Dyna 87, 57–65 (2020).
Rojas, F. J., Pacsi-Valdivia, S. & Sánchez-Ccoyllo, O. R. Simulación computacional e influencia de las variables meteorológicas en las concentraciones de PM\(_{10}\) y PM\(_{25}\) en Lima metropolitana. Inf. Tecnol. 33, 223–238 (2022).
Vu, B. N. et al. Developing an advanced PM\(_{25}\) exposure model in Lima, Peru. Remote Sens. 11, 641 (2019).
Acknowledgements
The authors would like to thank “Investiga UCV” of the “Universidad César Vallejo” for financial support for these publication and the “Reseacrh Group ICAMB -UCV” for the support in the development of this research. Likewise, P.C. Rodrigues acknowledges financial support from the CNPq grant “bolsa de produtividade PQ-2” 309359/2022-8, Federal University of Bahia and CAPES-PRINT-UFBA, under the topic “Modelos Matemáticos, Estatísticos e Computacionais Aplicados às Ciências da Natureza.
Author information
Authors and Affiliations
Contributions
All authors participated in the conceptualization, methodology, software, and manuscript writing.
Corresponding author
Ethics declarations
Competing interests
The authors declare no competing interests.
Additional information
Publisher's note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
About this article
Cite this article
Cabello-Torres, R.J., Carbo-Bustinza, N., Romero-Cabello, E.A. et al. An exploratory analysis of PM\(_{2.5}\)/PM\(_{10}\) ratio during spring 2016–2018 in Metropolitan Lima. Sci Rep 14, 9285 (2024). https://doi.org/10.1038/s41598-024-59831-9
Received:
Accepted:
Published:
DOI: https://doi.org/10.1038/s41598-024-59831-9
Comments
By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.