Please use this identifier to cite or link to this item: http://hdl.handle.net/10609/133666
Title: Estimació de la variació horària de la concentració d'ozó fent ús de dades meteorològiques i models LUR
Author: Camps Pons, Guillermo
Director: Solé-Ribalta, Albert  
Tutor: Ramírez Jávega, Francisco
Abstract: The rise in air pollutant emissions in the last centuries has turned air pollution into one of the main environmental threats. Ahead of the risks that air pollution means for the planet and for human health, steps must be taken to reduce the concentration of air pollutants in the atmosphere so that they don't exceed critical limits. One way to improve the efficiency of these actions is to create forecast models of air pollutants to be able to anticipate the issue using statistical, numeric or artificial-intelligence-based predictions. In this work, land use regressions (LUR) are created with GBM models to estimate the concentration of tropospheric ozone (O3) using data of the concentrations of air pollutants, data from weather stations and land use data. A selection of variables reveals that the 8-hour mean downward UV radiation at the surface and NO2 concentration are the most important factors to predict the concentration of ozone. Adding more variables based on time, wind, radiation and lane density of nearby highways, a 7-variable model is built with a R2 value of 0.911 and a RMSE value of 9.884g/m3. Adding a temperature-related variable and the year, the model obtains a R2 value of 0.933 and a RMSE value of 8.548g/m3. Finally, high resolution predictions are calculated from a downscaling process.
Keywords: ozone
air quality
LUR
Document type: info:eu-repo/semantics/masterThesis
Issue Date: Jun-2021
Publication license: http://creativecommons.org/licenses/by-nc-nd/3.0/es/  
Appears in Collections:Bachelor thesis, research projects, etc.

Files in This Item:
File Description SizeFormat 
gcampspo8TFM0621memoria.pdfMemoria del TFM11,92 MBAdobe PDFThumbnail
View/Open
gcampspo8TFM0621presentació.pdfPresentació del TFM4,08 MBAdobe PDFThumbnail
View/Open