Prediction of PM10 concentrations using Fuzzy c-Means and ANN

Salamanca has been considered among the most polluted cities in Mexico. The vehicular park, the industry and the emissions produced by agriculture, as well as orography and climatic characteristics have propitiated the increment in pollutant concentration of Particulate Matter less than 10 μg/m3 in diameter (PM10). In this work, a Multilayer Perceptron Neural Network has been used to make the prediction of an hour ahead of pollutant concentration. A database used to train the Neural Network corresponds to historical time series of meteorological variables (wind speed, wind direction, temperature and relative humidity) and air pollutant concentrations of PM10. Before the prediction, Fuzzy c-Means clustering algorithm have been implemented in order to find relationship among pollutant and meteorological variables. These relationship help us to get additional information that will be used for predicting. Our experiments with the proposed system show the importance of this set of meteorological variables on the prediction of PM10 pollutant concentrations and the neural network efficiency. The performance estimation is determined using the Root Mean Square Error (RMSE) and Mean Absolute Error (MAE). The results shown that the information obtained in the clustering step allows a prediction of an hour ahead, with data from past 2 hours

ANN and Fuzzy c-Means applied to environmental pollution prediction

Salamanca, situated in center of Mexico is among the cities which suffer most from the air pollution in Mexico. The vehicular park and the industry, as well as orography and climatic characteristics have propitiated the increment in pollutant concentration of Sulphur Dioxide (SO2). In this work, a Multilayer Perceptron Neural Network has been used to make the prediction of an hour ahead of pollutant concentration. A database used to train the Neural Network corresponds to historical time series of meteorological variables and air pollutant concentrations of SO2. Before the prediction, Fuzzy c-Means and K-means clustering algorithms have been implemented in order to find relationship among pollutant and meteorological variables. Our experiments with the proposed system show the importance of this set of meteorological variables on the prediction of SO2 pollutant concentrations and the neural network efficiency. The performance estimation is determined using the Root Mean Square Error (RMSE) and Mean Absolute Error (MAE). The results showed that the information obtained in the clustering step allows a prediction of an hour ahead, with data from past 2 hours.

How lowering speed limits in an urban highway affects traffics performance and emissions: the case of Madrid M-30 ring-road

The city of Madrid keeps not meeting the GHG and air pollutant limits set by the European legislation. A broad range of strategies have being taken into account to reduce both types of emissions; however traffic management meas ures are usually consigned to the sidelines. In 2004, Madrid City Council launched a plan to re-design its inner ring-road supported by a socioeconomic study that evaluated the environmental and operational benefits of the project. For safety reasons the planned speed limit for the tunnel section was finally reduced from 90km/h to 70km/h. Using a Macroscopic Traffic Model and the European Air Pollutant and Emissions Inventory Guidebook (EMEP/EEA), this paper examines the environmental and traffic performance consequences of this decision. Results support the thesis that reduced speed limits leads to GHG and air pollution reductions in the area affected by the measure without substantially altering traffic performance. The implementation of the new speed limit policy brings about a 15% and 16% reduction in both CO2 and NOx emissions respectively. Emissions’ reduction during off-peak hours is larger than during peak hours.

Urban trees and atmospheric pollutants in big cities: Effects in Madrid

Una investigación sobre la mejora de la contaminación del aire (CA) por medio de arbolado urbano se realizó en Madrid, una ciudad con casi 4 M de habitantes, 2,8 M de vehículos y casi 3 M de árboles de mantenimiento público. La mayoría de los árboles estaban en dos bosques periurbanos. Los 650.000 restantes era pies de alineación y parques. Los taxones estudiados fueron Platanus orientalis (97.205 árboles), Ulmus sp. (70.557), Pinus pinea (49.038), Aesculus hippocastanum (22.266), Cedrus sp. (13.678) y Quercus ilex (1.650), de calles y parques. Muestras foliares se analizaron en diferentes épocas del año, así como datos de contaminación por PM10 de 28 estaciones de medición de la contaminación durante 30 años, y también la intensidad del tráfico (IMD) en 2.660 calles. La acumulación de metales pesados (MP) sobre hojas y dentro de estas se estimó en relación con la CA y del suelo y la IMD del tráfico. La concentración media de Ba, Cd, Cr, Cu, Mn, Ni, Pb y Zn en suelo (materia seca) alcanzó: 489,5, 0,7, 49,4, 60,9, 460,9, 12,8, 155,9 y 190,3 mg kg-1 respectivamente. Los árboles urbanos, particularmente coníferas (debido a la mayor CA en invierno) contribuyen significativamente a mejorar la CA sobre todo en calles con alta IMD. La capacidad de las seis sp. para capturar partículas de polvo en su superficies foliares está relacionada con la IMD del tráfico y se estimó en 16,8 kg/año de MP tóxicos. Pb y Zn resultaron ser buenos marcadores antrópicos en la ciudad en relación con el tráfico, que fue la principal fuente de contaminación en los árboles y suelos de Madrid. Las especies de árboles variaron en función de su capacidad para capturar partículas (dependiendo de las propiedades de sus superficies foliares) y acumular los MP absorbidos de los suelos. Las concentraciones foliares de Pb y Zn estuvieron por encima de los límites establecidos en diferentes sitios de la ciudad. La microlocalización de Zn mediante microscópico mostró la translocación al xilema y floema. Se detectaron puntos de contaminación puntual de Cu and Cr en antiguos polígonos industriales y la distribución espacial de los MP en los suelos de Madrid mostró que en incluso en zonas interiores del El Retiro había ciertos niveles elevados de [Pb] en suelo, tal vez por el emplazamiento la Real Fábrica de Porcelana en la misma zona hace 200 años. Distintas áreas del centro de la ciudad también alcanzaron niveles altos de [Pb] en suelo. Según los resultados, el empleo de una combinación de Pinus pinea con un estrato intermedio de Ulmus sp. y Cedrus sp. puede ser la mejor recomendación como filtro verde eficiente. El efecto del ozono (O3) sobre el arbolado en Madrid fue también objeto de este estudio. A pesar de la reducción de precursores aplicada en muchos países industrializados, O3 sigue siendo la principal causa de CA en el hemisferio norte, con el aumento de [O3] de fondo. Las mayores [O3] se alcanzaron en regiones mediterráneas, donde el efecto sobre la vegetación natural es compensado por el xeromorfismo y la baja conductancia estomática en respuesta los episodios de sequía estival característicos de este clima. Durante una campaña de monitoreo, se identificaron daños abióticos en hojas de encina parecidos a los de O3 que estaban plantadas en una franja de césped con riego del centro de Madrid. Dada la poca evidencia disponible de los síntomas de O3 en frondosas perennifolias, se hizo un estudio que trató de 1) confirman el diagnóstico de daño de O3, 2) investigar el grado de los síntomas en encinas y 3) analizar los factores ambientales que contribuyeron a los daños por O3, en particular en lo relacionado con el riego. Se analizaron los marcadores macro y micromorfológicos de estrés por O3, utilizando las mencionadas encinas a modo de parcela experimental. Los síntomas consistieron en punteado intercostal del haz, que aumentó con la edad. Además de un punteado subyacente, donde las células superiores del mesófilo mostraron reacciones características de daños por O3. Las células próximas a las zonas dañadas, presentaron marcadores adicionales de estrés oxidativo. Estos marcadores morfológicos y micromorfológicos de estrés por O3 fueron similares a otras frondosas caducifolias con daños por O3. Sin embargo, en nuestro caso el punteado fue evidente con AOT40 de 21 ppm•h, asociada a riego. Análisis posteriores mostraron que los árboles con riego aumentaron su conductancia estomática, con aumento de senescencia, manteniéndose sin cambios sus características xeromórficas foliares. Estos hallazgos ponen de relieve el papel primordial de la disponibilidad de agua frente a las características xeromórficas a la hora de manifestarse los síntomas en las células por daños de O3 en encina. ABSTRACT Research about air pollution mitigation by urban trees was conducted in Madrid (Spain), a southern European city with almost 4 M inhabitants, 2.8 M daily vehicles and 3 M trees under public maintenance. Most trees were located in two urban forests, while 650'000 trees along urban streets and in parks. The urban taxa included Platanus orientalis (97'205 trees), Ulmus sp. (70’557), Pinus pinea (49'038), Aesculus hippocastanum (22’266), Cedrus sp. (13'678 and Quercus ilex (1'650) along streets and parks. Leave samples were analysed sequentially in different seasons, PM10 data from 28 air monitoring stations during 30 years and traffic density estimated from 2’660 streets. Heavy metal (HM) accumulation on the leaf surface and within leaves was estimated per tree related to air and soil pollution, and traffic intensity. Mean concentration of Ba, Cd, Cr, Cu, Mn, Ni, Pb and Zn in topsoil samples (dry mass) amounted in Madrid: 489.5, 0.7, 49.4, 60.9, 460.9, 12.8, 155.9 and 190.3 mg kg-1 respectively. Urban trees, particularly conifers (due to higher pollution in winter) contributed significantly to alleviate air pollution especially near to high ADT roads. The capacity of the six urban street trees species to capture air-born dust on the foliage surface as related to traffic intensity was estimated to 16.8 kg of noxious metals from exhausts per year. Pb and Zn pointed to be tracers of anthropic activity in the city with vehicle traffic as the main source of diffuse pollution on trees and soils. Tree species differed by their capacity to capture air-borne dust (by different leaf surface properties) and to allocate HM from soils. Pb and Zn concentrations in the foliage were above limits in different urban sites and microscopic Zn revelation showed translocation in xylem and phloem tissue. Punctual contamination in soils by Cu and Cr was identified in former industrial areas and spatial trace element mapping showed for central Retiro Park certain high values of [Pb] in soils even related to a Royal pottery 200 years ago. Different areas in the city centre also reached high levels [Pb] in soils. According to the results, a combination of Pinus pinea with understorey Ulmus sp. and Cedrus sp. layers can be recommended for the best air filter efficiency. The effects of ozone (O3) on trees in different areas of Madrid were also part of this study. Despite abatement programs of precursors implemented in many industrialized countries, ozone remained the main air pollutant throughout the northern hemisphere with background [O3] increasing. Some of the highest ozone concentrations were measured in regions with a Mediterranean climate but the effect on the natural vegetation is alleviated by low stomatal uptake and frequent leaf xeromorphy in response to summer drought episodes characteristic of this climate. During a bioindication survey, abiotic O3-like injury was identified in foliage. Trees were growing on an irrigated lawn strip in the centre of Madrid. Given the little structural evidence available for O3 symptoms in broadleaved evergreen species, a study was undertaken in 2007 with the following objectives 1) confirm the diagnosis, 2) investigate the extent of symptoms in holm oaks growing in Madrid and 3) analyse the environmental factors contributing to O3 injury, particularly, the site water supply. Therefore, macro- and micromorphological markers of O3 stress were analysed, using the aforementioned lawn strip as an intensive study site. Symptoms consisted of adaxial and intercostal stippling increasing with leaf age. Underlying stippling, cells in the upper mesophyll showed HR-like reactions typical of ozone stress. The surrounding cells showed further oxidative stress markers. These morphological and micromorphological markers of ozone stress were similar to those recorded in deciduous broadleaved species. However, stippling became obvious already at an AOT40 of 21 ppm•h and was primarily found at irrigated sites. Subsequent analyses showed that irrigated trees had their stomatal conductance increased and leaf life-span reduced whereas their leaf xeromorphy remained unchanged. These findings suggest a central role of water availability versus leaf xeromorphy for ozone symptom expression by cell injury in holm oak.

Unsupervised system to classify SO2 pollutant concentrations in Salamanca, Mexico

Salamanca is cataloged as one of the most polluted cities in Mexico. In order to observe the behavior and clarify the influence of wind parameters on the Sulphur Dioxide (SO2) concentrations a Self-Organizing Maps (SOM) Neural Network have been implemented at three monitoring locations for the period from January 1 to December 31, 2006. The maximum and minimum daily values of SO2 concentrations measured during the year of 2006 were correlated with the wind parameters of the same period. The main advantages of the SOM Neural Network is that it allows to integrate data from different sensors and provide readily interpretation results. Especially, it is powerful mapping and classification tool, which others information in an easier way and facilitates the task of establishing an order of priority between the distinguished groups of concentrations depending on their need for further research or remediation actions in subsequent management steps. For each monitoring location, SOM classifications were evaluated with respect to pollution levels established by Health Authorities. The classification system can help to establish a better air quality monitoring methodology that is essential for assessing the effectiveness of imposed pollution controls, strategies, and facilitate the pollutants reduction.

Pollutant concentrations and Meteorological data classification by Neural Networks

This paper present an environmental contingency forecasting tool based on Neural Networks (NN). Forecasting tool analyzes every hour and daily Sulphur Dioxide (SO2) concentrations and Meteorological data time series. Pollutant concentrations and meteorological variables are self-organized applying a Self-organizing Map (SOM) NN in different classes. Classes are used in training phase of a General Regression Neural Network (GRNN) classifier to provide an air quality forecast. In this case a time series set obtained from Environmental Monitoring Network (EMN) of the city of Salamanca, Guanajuato, México is used. Results verify the potential of this method versus other statistical classification methods and also variables correlation is solved.

Unsupervised method to classify PM10 pollutant concentrations

In this paper a method based mainly on Data Fusion and Artificial Neural Networks to classify one of the most important pollutants such as Particulate Matter less than 10 micrometer in diameter (PM10) concentrations is proposed. The main objective is to classify in two pollution levels (Non-Contingency and Contingency) the pollutant concentration. Pollutant concentrations and meteorological variables have been considered in order to build a Representative Vector (RV) of pollution. RV is used to train an Artificial Neural Network in order to classify pollutant events determined by meteorological variables. In the experiments, real time series gathered from the Automatic Environmental Monitoring Network (AEMN) in Salamanca Guanajuato Mexico have been used. The method can help to establish a better air quality monitoring methodology that is essential for assessing the effectiveness of imposed pollution controls, strategies, and facilitate the pollutants reduction.

Air pollution data classification by SOM Neural Network

Over the last ten years, Salamanca has been considered among the most polluted cities in México. This paper presents a Self-Organizing Maps (SOM) Neural Network application to classify pollution data and automatize the air pollution level determination for Sulphur Dioxide (SO2) in Salamanca. Meteorological parameters are well known to be important factors contributing to air quality estimation and prediction. In order to observe the behavior and clarify the influence of wind parameters on the SO2 concentrations a SOM Neural Network have been implemented along a year. The main advantages of the SOM is that it allows to integrate data from different sensors and provide readily interpretation results. Especially, it is powerful mapping and classification tool, which others information in an easier way and facilitates the task of establishing an order of priority between the distinguished groups of concentrations depending on their need for further research or remediation actions in subsequent management steps. The results show a significative correlation between pollutant concentrations and some environmental variables.