Search

Physical models are essential to describe the behavior of pollutants especially in high latitudes, and they have been regarded as immensely precise. In the tropics, however, these models have lower accuracy due to the absence of a simple theoretical framework to describe tropical dynamics. Hence, the development of predictive nonlinear models with machine learning has increased, as they are able to quantify the different dynamic processes regarding air quality and to obtain accurate predictions in less computational time than their physical counterpart. This study constructs and evaluates a support vector machine (SVM) to forecast ground-level PM2.5 in a populated city with complex topography. The simulations were built for days with red Air Quality Index (AQI), to assess whether the model could represent the behavior of days with high values and data with fast and substantial changes in the PM2.5 tendency. The SVM is trained with an air quality monitoring network using the radial basis function kernel. A spatial interpolation is also conducted to determine and describe the behavior of the AQI in the city of Bogota. This work uses statistical scores (root mean square error (9.302 μg/m3), mean BIAS (1.405 μg/m3), index of agreement (0.732), and correlation coefficient (0.654)) to validate the capability of an SVM model of simulating, with high precision, the concentrations of PM2.5 in a city with complex terrain in the short term and also to demonstrate the potential of the SVM to be used as a forecast model in other tropical cities.

Light-absorbing aerosols emitted during open biomass burning (BB) events such as wildfires and agricultural burns have a strong impact on the Earth's radiation budget through both direct and indirect effects. Additionally, BB aerosols and gas-phase emissions can substantially reduce air quality at local, regional, and global scales, negatively affecting human health. South America is one of largest contributors to BB emissions globally. After Amazonia, the BB emissions from wildfires and agricultural burns in the grassland plains of northern South America (NSA) are the most significant in the region. However, few studies have analyzed the potential impact of NSA BB emissions on regional air quality. Recent evidence suggests that seasonal variations in air quality in several major cities in NSA could be associated with open biomass burning emissions, but it is still uncertain to what extent those sources impact air quality in the region. In this work, we report on 3 years of continuous equivalent black carbon (eBC) and brown carbon (BrC) observations at a hilltop site located upwind of the city of Bogotá, and we demonstrate its association with fires detected by the MODerate-resolution Imaging Spectroradiometer (MODIS) in a 3000 km × 2000 km domain. Offline PM2.5 filter samples collected during three field campaigns were analyzed to quantify water-soluble organic carbon (WSOC), organic and elemental carbon (OC∕EC), and biomass burning tracers such as levoglucosan, galactosan, and potassium. MODIS active fire data and HYSPLIT back trajectories were used to identify potential biomass burning plumes transported to the city. We analyzed the relationship between BrC, WSOC, water-soluble potassium, and levoglucosan to identify signals of the regional transport of BB aerosols. Our results confirm that regional biomass burning transport from wildfires occurs annually during the months of January and April. The seasonality of eBC closely followed that of PM2.5 at the city air quality stations; however, the observed seasonality of BrC is distinctly different to that of eBC and strongly associated with regional fire counts. The strong correlation between BrC and regional fire counts was observed at daily, weekly, and monthly timescales. WSOC at the measurement site was observed to increase linearly with levoglucosan during high BB periods and to remain constant at ∼2.5 µgC m−3 during the low BB seasons. Our findings show, for the first time in this region, that aged BB plumes can regularly reach densely populated areas in the Central Andes of northern South America. A source footprint analysis involving BrC observations, back trajectories, and remotely sensed fire activity shows that the eastern savannas in NSA are the main BB source region for the domain analyzed.

Biomass burning emissions have a substantial impact on regional air quality and climate. The region of Amazonia in South America has long been identified as one of the largest contributors to short lived pollutants globally. However, massive natural wildfires and agricultural burns also occur every year in the grassland plains of Northern South America during the dry season (November to April). The regional-scale air quality impact of these biomass burning emissions has not been studied in depth and is analyzed in this study. We used PM2.5 and PM10 concentrations from three large urban areas: Bogota, Medellin, and Bucaramanga, for the period 2006–2016. Carbon monoxide data was only available for the city of Bogota for the analysis period. These cities are located hundreds of kilometers westward of the emission areas. The spatio-temporal distribution of fires was obtained from MODIS Active Fire Data. The back-trajectories of air masses reaching the receptor sites were computed with two different meteorological datasets. Radiosonde data, available only for Bogota, was used to account for local meteorological factors impacting pollution dispersion. A novel analysis algorithm was developed to combine active fire data with back-trajectory locations to select those active fires in the vicinity of the air masses arriving at each city. This analysis allows the selection of only those upwind fires that can be related to the air quality in the selected locations. We show that anomalously high PM and CO levels occurred when air masses originated from the Orinoco grasslands during the times when the largest number of fires in the region were active. The correlation between number of fires and PM10 concentration was found to decrease with increasing distance from the sources, ranging from 0.6 to 0.25. Our results are insensitive to the meteorological dataset used to generate back-trajectories. For Bogota it was found that mixing height variations can explain an important fraction of the observed seasonal variations in PM10, PM2.5, and CO concentration. The number of upwind fires can explain 11% ± 5% of the seasonal variability in CO concentrations. Estimates of the seasonal variability of PM10 and PM2.5 explained by fires are 45% ± 7% and 39% ± 8% respectively. However, covariance between occurrence of fires and non-combustion local sources of PM imply that the latter estimates are likely an overestimation of the actual contribution. Our findings support the possibility that fires in the Orinoco river basin deteriorate air quality in highly populated urban centers hundreds of kilometers away from the sources.

Lockdown measures led to air pollution decrease in several countries around the world such as China and India, whereas other regions experimented an increase in pollutant concentrations. Northern South America (NSA) was one of those areas where pollution changed during lockdown due to high fire activity. This study aims to analyze, for the first time in NSA, the behavior of selected criteria air pollutants during the implementation of the SARS-CoV-2 lockdown in two high populated cities of the region: Bogotá and Medellín in Colombia. A set of tools including surface measurements, as well as satellite and modeled data were used. 24-hour average concentrations of PM10, PM2.5, and NO2 were collected from air quality stations for the lockdown period ranging from February 21 to June 30, 2020. The Copernicus Atmosphere Monitoring Service (CAMS) was used to analyze the fire flux OC as a biomass burning (BB) indicator, and tropospheric NO2 concentrations were retrieved from TROPOMI. The HYSPLIT model was used to analyze back trajectories and fire data were obtained from MODIS sensor measurements. Our analysis shows short-term background NO2, PM10, and PM2.5 concentration reductions of 60%, 44%, and 40%, respectively, for the strict lockdown; and 62%, 58%, and 69% for the relaxed lockdown. Corresponding long-term reductions were of 50%, 32%, and 9% for the strict lockdown; and 37%, 29%, and 19% for the relaxed lockdown. Regional BB increased PM2.5 concentrations by 20 μg/m3 during the strict lockdown, and the Saharan dust event increased PM10 concentrations up to 168 μg/m3 in Bogotá, and 104 μg/m3 in Medellín, bringing an additional risk of morbidity and mortality for population. Regional BB has several causes that need to be properly managed to benefit local air quality improvement plans. Future cleaner transport policies equivalent to reduced lockdown mobility could bring pollution close to WHO guidelines., Graphical abstract Unlabelled Image, Highlights • Short-term NO2, PM10, PM2.5 decreased by 60%, 44%, and 40% respectively during the strict lockdown. • Short-term NO2, PM10, PM2.5 decreased by 62%, 58%, and 69% respectively for the relaxed lockdown. • Long-term NO2, PM10, PM2.5 reductions were of 50%, 32%, and 9% respectively for the strict lockdown. • Regional biomass burning increased PM2.5 concentrations by 20 μg/m3 during the strict lockdown. • Policies equivalent to reduced lockdown mobility could bring pollution close to WHO guidelines.

BRT (Bus Rapid Transit) public transport systems, are highly frequented micro environments by the population of a city. In this research a TSI DustTrak II was used to determine the PM2.5 concentration inside the buses of the BRT system in Bogota, and a GPS app, MyCarTracks®, to record the speed and location data. Measurements were performed in three different segments of the system, and in different seats inside the buses. PM2.5 concentration levels above 120 µg/m3 were measured for all the segments. Time travels of the BRT passengers was determined to be around 20 to 40 minutes average, where they are exposed directly to these high levels of pollution. Thereafter, a CFD model was implemented to simulate and assess the behavior of the tailpipe emissions. Contours of concentration showed the self-pollution phenomenon as a fraction around 18% of the total concentration of PM 2.5 measured inside the cabin of the BRT.

In the vast majority of Latin American and South American countries, global emission inventories (EIs) are often used for modelling air quality. In particular the Emission Database for Global Atmospheric Research EDGAR is widely deployed but several studies have pointed to some gaps in comparison with national/regional inventories which incur errors in interpreting results. In Cuba, due to scarcity of a spatially distributed national inventory, EDGAR has been used as entry for air quality modelling without verifying their reliability over the region. Our goal in this article is to compare and contrast EDGAR with a local inventory and to evaluate similarities or discrepancies. We use advanced comparison techniques developed by the Forum for Air Quality Modelling in Europe–FAIRMODE. This approach differs from others in the detailed way in which it points out the differences and gets insights in possible explanations. Overall, EDGAR provided spatially smoother results and relatively lower values in hotspot areas. Coarse differences in terms of activities were low for all analyzed sectors. However, EDGAR overestimates emission factors (EFs) of stationary sources for CO by a factor of 3 and SO2 by a factor of 1.5 while underestimates those of PPM10 by a factor of 25. Most of the road transport EFs are overestimated in EDGAR; PM10, CO and NOx are 2 times higher, while CH4 and SO2 are 5 to 20 times higher. Large differences were found on the spatial distribution of energy and industrial sources. EDGAR can be regionally accepted as a reference but it is not recommended for air quality simulation over Cuba. A more complete reporting must be expected when more official national data are due. A review and evaluation of local emission inventories over Cuba can be useful for identifying potential areas for future improvement.

Road traffic emission factors (EFs) are one of the main sources of uncertainties in emission inventories; it is necessary to develop methods to reduce these uncertainties to manage air quality more efficiently. Recently an alternative method has been proposed to estimate the EFs. In that work the emission factors were estimated from a long term tracer study developed in Ho Chi Minh City (HCMC) Vietnam. A passive tracer was continuously emitted from a finite line source placed in one side of an urban street canyon. Simultaneously, the resulting tracer concentrations were monitored at the other side of the street. The results of this experiment were used to calculate the dispersion factors and afterwards, these dispersion factors were used to estimate the EFs. In this paper we use the Computational Fluids Dynamics (CFD) model WinMISKAM to critically evaluate the proposed methodology. In a first step, we use the results of the tracer study to validate the CFD model. Results show that the model is able to simulate quite well the tracer dispersion in most of the cases. The model is then used to evaluate the effect of varying the source configuration and to correct the EFs. A comparison with available studies shows that the corrected EFs are within the range of the EFs reported in other studies. Finally, the CFD model is used to find a source configuration that better represents the vehicle emissions and that may be used in future studies to estimate the EFs more accurately. Results show that a 200 m line placed in the center of the street would represent very well the vehicle emissions. This work shows that it is possible to accurately estimate the EFs from tracer studies.

Road traffic emissions, one of the largest source categories in megacity inventories, are highly uncertain. It is essential to develop methodologies to reduce these uncertainties to manage air quality more effectively. In this paper, we propose a methodology to estimate road traffic emission factors (EFs) from a tracer experiment and from roadside pollutants measurements. We emitted continuously during about 300 non-consecutive hours a passive tracer from a finite line source placed on one site of an urban street. At the same time, we measured continuously the resulting tracer concentrations at the other side of the street with a portable on-line gas chromatograph. We used n-propane contained in commercial liquid petroleum gas (LPG) as a passive tracer. Propane offers several advantages to traditional tracers (SF6, N2O, CFCs): low price, easily available, non-reactive, negligible global warming potential, and easy to detect with commercial on-line gas chromatographs. The tracer experiment was carried out from January to March 2007 in a busy street of Ho Chi Minh City (Vietnam). Traffic volume, weather information and pollutant concentrations were also measured at the measurement site. We used the results of the tracer experiment to calculate the dilution factors and afterwards we used these dilution factors, the traffic counts and the pollutant concentrations to estimate the EFs. The proposed method assumes that the finite emission line represents the emission produced by traffic in the full area of the street and therefore there is an error associated to this assumption. We use the Computational Fluids Dynamics (CFD) model MISKAM to calculate this error and to correct the HCMC EFs. EFs for 15 volatile organic compounds (VOCs) and NO are reported here. A comparison with available studies reveals that most of the EFs estimated here are within the range of EFs reported in other studies.