Browsing by Author "Fleming, Zöe"
Now showing 1 - 5 of 5
Results Per Page
Sort Options
Item Forecasting PM2.5 levels in Santiago de Chile using deep learning neural networks(2021) Menares, Camilo; Pérez, Patricio; Parraguez, Santiago; Fleming, ZöeAir pollution has been shown to have a direct effect on human health. In particular, PM2.5 has been proven to be related to cardiovascular and respiratory problems. Therefore, it is important to have accurate models to predict high pollution events for this and other pollutants. We present different models that forecast PM2.5 maximum concentrations using a Long Short-Term Memory (LSTM) based neural network and a Deep Feedforward Neural Network (DFFNN). Ten years of air pollution and meteorological measurements from the network of monitoring stations in the city of Santiago, Chile were used, focusing on the behaviour of three zones of the city. All missing values were rebuilt using a method based on discrete cosine transforms and photochemical predictors selected through unsupervised clustering. Deep learning techniques provide significant improvements compared to a traditional multi-layer neural networks, particularly the LSTM model configured with a 7-day memory window (synoptic scale of pollution patterns) can capture critical pollution events at sites with both primary and secondary air pollution problems. Furthermore, the LSTM model consistently outperform deterministic models currently used in Santiago, Chile.Publication Impact of mining on the metal content of dust in indigenous villages of northern Chile(2022) Zanetta-Colombo, Nicolás; Fleming, Zöe; Gayo, Eugenia M.; Manzano, Carlos A.; Panagi, Marios; Valdés, Jorge; Siegmund, AlexanderIndigenous communities from northern Chile have historically been exposed to the impacts of massive copper industrial activities conducted in the region. Some of the communities belonging to the Alto El Loa Indigenous Development Area are located less than 10 km from the “Talabre'' tailings dam, which contains residues from copper production and other metals that can be toxic to human health (e.g., As, Sb, Cd, Mo, Pb). Given the increasing demand of copper production to achieve net-zero emission scenarios and concomitant expansions of the tailings, the exposure to toxic metals is a latent risk to local communities. Despite the impact that copper production could generate on ancestral communities from northern Chile, studies and monitoring are limited and the results are often not made accessible for local communities. Here, we evaluate such risks by characterizing metal concentrations in dust collected from roofs and windows of houses from the Alto El Loa area. Our results showed that As, Sb, Cd, Cu, Mo, Ag, S, and Pb concentrations in these matrices can be connected to local copper mining activities. Additionally, air transport models indicate that high concentrations of toxic elements (As, Sb, and Cd) can be explained by the atmospheric transport of particles from the tailings in a NE direction up to 50 km away. Pollution indices and Health Risk Assessment suggested a highly contaminated region with a health risk for its inhabitants. Our analysis on a local scale seeks to make visible the case of northern Chile as a critical territory where actions should be taken to mitigate the effects of mining in the face of this new scenario of international demand for the raw materials necessary for the transition to a net-zero carbon global society.Publication Long-term airborne particle pollution assessment in the city of Coyhaique, Patagonia, Chile(2022) Solís, Rafael; Toro A., Richard; Gómez, Luis; Vélez-Pereira, Andrés M.; López, Manuela; Fleming, Zöe; Fierro, Nicolás; Leiva G., ManuelAn air pollution assessment in a small city located in the heart of Chilean Patagonia is presented. Seven years (2014–2020) of PM concentration levels retrieved from two monitoring stations permits an evaluation of the city’s pollution variability, the effect of meteorological variables and long-term trends of air pollution. The highest PM concentration levels observed during the coldest months are mainly related to an increasing emission associated with the intensive use of firewood for residential heating and cooking. The most polluted days are associated with low temperatures, low wind speed and high PM2.5/PM10 ratios, which is consistent with the predominance of local firewood sources over background emissions. A decrease in both PM fractions over time has been estimated (PM10: -4.1, CI99%: − 5.7 to − 2.9 and PM2.5: -2.2, CI99%: − 3.5 to − 1.3 μg m− 3 year− 1 ). However, the annual average PM mass concentrations in Coyhaique exceeded both national and international air quality thresholds. The city reported a percent of annual exceedances of the daily WHO guidelines of 57% for PM10 and 77% for PM2.5. These numbers highlight the serious air pollution problem of the city of Coyhaique, which exhibits air pollution levels comparable to those of many polluted megacities in the world.Publication Photochemical sensitivity to emissions and local meteorology in Bogotá, Santiago, and São Paulo: An analysis of the initial COVID-19 lockdowns(2022) Seguel, Rodrigo J.; Gallardo, Laura; Osses, Mauricio; Rojas, Néstor Y.; Nogueira, Thiago; Menares, Cmilo; Andrade, María Fátima; Belalcázar, Luis C.; Carrasco, Paula; Eskes, Henk; Fleming, Zöe; Huneeus, Nicolás; Ibarra-Espinosa, Sergio; Landulfo, Eduardo; Leiva, Manuel; Mangones, Sonia C.; ernando G. Morais8 , Gregori A. Moreira11, Nicola´ s Pantoja3 , Santiago Parraguez1,12, Jhojan P. Rojas13, Roberto Rondanelli1,2, Izabel da Silva Andrade8 , Richard Toro9 ,; Moreira, Gregori A.; Pantoja, Nicolás; Parraguez, Santiago; Rojas, Jhojan P.; Rondanelli, Roberto; Silva Andrade, Izabel da; Toro, Richard; Yoshida, Alexandre C.This study delves into the photochemical atmospheric changes reported globally during the pandemic by analyzing the change in emissions from mobile sources and the contribution of local meteorology to ozone (O3) and particle formation in Bogotá (Colombia), Santiago (Chile), and São Paulo (Brazil). The impact of mobility reductions (50%–80%) produced by the early coronavirus-imposed lockdown was assessed through high-resolution vehicular emission inventories, surface measurements, aerosol optical depth and size, and satellite observations of tropospheric nitrogen dioxide (NO2) columns. A generalized additive model (GAM) technique was also used to separate the local meteorology and urban patterns from other drivers relevant for O3 and NO2 formation. Volatile organic compounds, nitrogen oxides (NOx), and fine particulate matter (PM2.5) decreased significantly due to motorized trip reductions. In situ nitrogen oxide median surface mixing ratios declined by 70%, 67%, and 67% in Bogotá, Santiago, and São Paulo, respectively. NO2 column medians from satellite observations decreased by 40%, 35%, and 47%, respectively, which was consistent with the changes in mobility and surface mixing ratio reductions of 34%, 25%, and 34%. However, the ambient NO2 to NOx ratio increased, denoting a shift of the O3 formation regime that led to a 51%, 36%, and 30% increase in the median O3 surface mixing ratios in the 3 respective cities. O3 showed high sensitivity to slight temperature changes during the pandemic lockdown period analyzed. However, the GAM results indicate that O3 increases were mainly caused by emission changes. The lockdown led to an increase in the median of the maximum daily 8-h average O3 of between 56% and 90% in these cities.Item Road Traffic Noise on the Santa Marta City Tourist Route(2021) Jiménez-Uribe, Dámaris A.; Daniels, Darwin; Fleming, Zöe; Vélez-Pereira, Andrés M.The objective of this study was to determine the influence of vehicular traffic on the environmental noise levels of the Santa Marta City tourist route on the Colombian coast. An analysis of vehicle types and frequencies at various times of the day over nearly a year helped to track the main sources of environmental noise pollution. Five sampling points were selected, which were distributed over 12 km, with three classified as peripheral urban and two as suburban. The average traffic flow was 966 vehicles/h and was mainly composed of automobiles, with higher values in the peripheral urban area. The noise level was 103.3 dBA, with background and peak levels of 87.2 and 107.3 dBA, respectively. The noise level was higher during the day; however, there were no differences between weekdays and weekends. The results from the analysis of variance showed that the number of vehicles and the noise levels varied greatly according to the time of day and sampling point location. The peak and mean noise levels were correlated with the number of automobiles, buses and heavy vehicles. The mean noise levels were similar at all sample points despite the traffic flow varying, and the background noise was only correlated for automobiles (which varied much more than the heavy vehicles between day and night).