LDN Air Analysis

This analysis shows the air quality in greather London area. An average measure of each pollutanat were calculated in order to measure their impact in the air quality. Currently, The UK has a lot AURN statiosn to detect pollutant matrials and report their concentrations, but this analysis will include all AURN networks located in the Great London area, wich includes 39 stations that monitor 57 pollutants.

Although all of them cause effect in humans, animals, structures and vegetation, some of them are considered highly harmful and due to their effects in human health. These pollutants are the particulates PM2.5 and PM10, nitrogen dioxide NO2 and ozone O3 which are regulated by the European Environmental Agency EEA.

Tools:

• R
• Open Air (R Library for air analysis)
• ggplot2 for R
• MySQL database.

Dataset

The datasets used for this analysis can be accessed through UK Air web site, which is part of the Department for Environment Food and Rural Affairs of U.K. There is two datasets: The stations and the measures. The stations dataset contains metadata related to the station like location, name, time of use, time off, parameters for calibration, etc. Despite the stations are phisycally located in different areas of London and each station measures a set of pollutants, not all the stations measures the same.

The pollutnats concentrations are measured per hour and stored in the station. After a considerable time, the information is extracted from the stations al=nd published in the UK Air Website. The interval of time used in this analysis were 6 years (2013-2018).

How to run

1. In Rstudio, open the file LdnAirAnalysis.Rproj
2. Install dependencies: openair and dplyr
3. Create folders data and images in the projet root
4. Run the extraction.R. It will generate a .csv file in the folder data/
5. Execute dataExploration.R

Pollutant selection criteria

The European Union (EU) through the European Environment Agency has developed an extensive legislation which establishes standards and objectives for a number of pollutants in air. The selection criteria for the pollutant to analyze will be that one that has exceeded this quality index limit more frequently.

Output

A couple of .csv file that contains the average measures of the most significant pollutant in London per hour and per day.

Some Plots

Correlation matrix between pollutants and variables monitored by the sensors in the stations.

Summary for O3

Summary for No2

Data trend non deseasonalized

Data trend deseasonalized