Authors Hao Wu, Nick Rand, Brian Stacey
Compilation date 20 July 2018
Customer Heathrow Airport Limited
Approved by Nick Rand
Copyright Ricardo Energy & Environment
EULA http://ee.ricardo.com/cms/eula/

Contract reference ED59405 Report reference Q2 2018

1 Introduction

This is a quarterly summary report for Heathrow Airport Limited for the period April to June 2018.

This is a dynamic report containing embedded data from which the report tables, plots and graphs are generated. The embedded data allows the reader a level of interaction with some of the report findings, providing additional insight. This approach enables a more easily navigated and streamlined report providing an engaging and intuitive reader experience.

Maps for example can be panned and zoomed with different layers and markers turned on and off and with popup information by clicking on markers or hovering the mouse over them. Tables may contain much more information than initially displayed and can be set to show different numbers of rows, and can be filtered sorted or searched to display only specific information of interest. Some time series plots can be tracked with the mouse cursor to obtain specific time/date stamps and values and the reader can zoom into a specific time window by dragging with the cursor and double left clicking to return to the full plot extent.

The report is easily navigable using the floating table of contents on the left pane which tracks with the reader’s progress through the report and expands and collapses to a level of detail related to the subheadings used. The layout of the report is also dynamic, with some sections split into specific ‘tabs’ (e.g. per pollutant or per site) for ease of access to those sections.

2 Monitoring stations

A summary of site information is presented in the interactive map and table below . Full site information can be obtained using the URL hyperlink in the popup for each monitoring station on the map.


3 Data description

Data contained within this report are managed by Ricardo and stored in a dedicated, secure database. Data are ratified quarterly on a calendar year basis and according to a regular schedule of audits and calibrations. As a result, data are marked as either provisional (and subject to change) or ratified. The data within this report are [fully ratified/provisional pending final ratification].

Some of the plots provided in this report use associated meteorological data (typically wind speed and wind direction) in order to put measured pollutant concentrations in a dispersion context. Very few monitoring stations include dedicated met equipment and those that do typically rely on relatively short met masts and so only provide very localised representation of dispersion. Instead of relying on limited local met data, these reports have incorporated modelled met data from Ricardo’s WRF model (at 10km resolution) 1. (Currently the report still uses the worldmet to import met data.)

Gaseous pollutant mass units are at 20 °C and 1013mb. NOx mass units are NOx as NO2 μg m-3. Particulate matter concentrations are reported at ambient temperature and pressure. Note that PM10 is referred to in the data tables and plots as ‘GE10’ if it is derived from an instrument that has demonstrated equivalence with the gravimetric reference method 2 (e.g. a TEOM FDMS, a FIDAS or a BAM with a correction factor applied).

3.1 Relevant pollution Limit Values

The European Air Quality Directive and Fourth Daughter Directive set out legal limits for different pollutants as Limit Values, Target Values or Long Term Objectives to protect human health. These are summarised in the table below. Local authorities don’t typically measure ozone, benzene, B[a]P or metals that are captured within Defra’s national networks. All pollutants measured have been included in this data summary for completeness, irrespective of their significance for local authority policy interests.


4 Data Analysis

4.1 Summary statistics

The following tables present some basic pollutant statistics for the period selected in the report. The Low, Moderate and High indicates the percentage of times when pollution concentrations are in the corresponding AQI bands defined by DEFRA.

NO2

Figure 1: NO2 exceedence.

Figure 2 shows the rolling annual mean of NO2 (calculated based on the preceding 12 monthly mean) for all the Heathrow sites from 2008.

Figure 2: NO2 12-month rolling mean for the Heathrow sites from 2008.

PM2.5

PM10

Figure 3: PM10 exceedence.

O3

Figure 4: O3 exceedance.

4.2 AQ index distribution

The plots below illustrate the distribution of AQ index values for each site by pollutant. It shows the number of occasions (hours) that each site reported concentrations are in each index. More information on the AQ Index is available from UK-Air 3. Note that this summary presents data at hourly resolution which differs from Defra’s Daily AQ Index (DAQI) which takes the worst across all pollutants and hours to represent each day.

NO2

Figure 5: Distribution of AQI for NO2.

PM2.5

Figure 6: Distribution of AQI for PM2.5.

PM10

Figure 7: Distribution of AQI for PM10.

O3

Figure 8: Distribution of AQI for O3.

4.3 Boxplots

The plots below are box and whisker plots to show the distribution in concentrations for each monitoring station. The boxes demarcate the lower quartile, median and upper quartile. The whiskers extend to the maximum and minimum values within median ± 1.5 times interquartile range (IQR). Values outside the median ± 1.5 times IQR are generally considered as outliers.

NO2

Figure 9: Boxplot for hourly NO2 concentrations.

PM2.5

Figure 10: Boxplot for daily PM2.5 concentrations.

PM10

Figure 11: Boxplot for daily PM10 concentrations.

O3

Figure 12: Boxplot for 8-hour rolling mean O3 concentrations.

4.4 Time series plot

The plots below show the time series of concentrations. Each pollutant is presented on a different tab and all sites are shown on each plot for comparison. The period covered is the complete data record to show trends over the medium and long term. Each iteration of this regular report (e.g. monthly, or quarterly) has the most recent data appended to the plot (not implemented in the code yet). A daily average resolution has been chosen as the most appropriate metric over a variety of different time windows. Zooming in on specific periods of the plot can be done by dragging a box over the section of the main plot frame or by using the sliders in the narrow plot frame below. To return to the default (all data) zoom level, double click the plot. Holding the mouse over the lines will highlight specific values and time stamp for that record for each station.

NO2

Figure 13: Time series plot of daily average NO2 concentration.

PM2.5

Figure 14: Time series plot of daily average PM2.5 concentration.

PM10

Figure 15: Time series plot of daily average PM10 concentration.

O3

Figure 16: Time series plot of daily maximum 8-hour rolling mean O3 concentration.

4.5 Time Variation plot

These plots show concentrations over different time intervals such as diurnal, day of week and month of year. The plot showing seasonal variation will show only the three months of the quarter or a dot showing one month depending on the time frame covered in this report. The topmost frame shows the concentrations as they vary by hour of the day and day of the week. The hour of the day variation is summarised on its own in the lower left pane and the variation by day of the week is shown in the lower right pane. These plots often help explain variations in concentration according to the emissions activity associated with them. For example, NOx concentrations at roadside sites tend to exhibit peaks according to morning and evening traffic rush hours and tend to decline over weekends when there is generally lower traffic volumes.

NO2

Figure  17: Time variation of hourly NO~2~ cocenctrations.

Figure 17: Time variation of hourly NO2 cocenctrations.

PM2.5

Figure  18: Time variation of hourly PM~2.5~ cocenctrations.

Figure 18: Time variation of hourly PM2.5 cocenctrations.

PM10

Figure  19: Time variation of hourly PM~10~ cocenctrations.

Figure 19: Time variation of hourly PM10 cocenctrations.

O3

Figure  20: Time variation of hourly O~3~ concentrations.

Figure 20: Time variation of hourly O3 concentrations.

4.6 Calendar plot

The plot below shows daily variation in concentrations by pollutant (one on each tab) across the period of the report, as laid out in a calendar style. This allows intuitive viewing of day to day headline trends in the wider context of the period. The background colours shown for each day relate to the concentration. The date is coloured by the wind speed for that day. The actual value can also be seen by hovering the mouse on the cell.

The site with highest average concentrations is used to produce the calendar plot.

NO2

Figure 21: NO2 calendar plot.

PM2.5

Figure 22: PM2.5 calendar plot.

PM10

Figure 23: PM10 calendar plot.

O3

Figure 24: O3 calender plot of daily maximum 8-hour rolling mean.

4.7 Back trajectory analysis

The back trajectory plot shows data from the HYSPLIT model (NOAA HYSPLIT 4) run in analysis mode. This shows the air mass back trajectories for the period covered by the report. Three different kinds of plot are shown. One statistically groups the trajectories into simioklar clusters (six) and shows the proportion of time during the report period that each represents. (Currently code only plots the clusters for a whole year.) This is useful to get an overview of air mass origins during the report period. On additional tabs, the trajectories associated with exceedances of the LV (1-hr for NO2 and 24-hr for PM10) are shown. A plot of the trajectories associated with the top 10 measured concentrations is also presented.

4.7.1 Trajectory clusters

4-day air mass back trajectories arriving at London for 2018 are grouped into 6 clusters.

Air mass back trajectories over these spatial scales do not vary locally so the receptor location used in this report has been selected from a range of national receptor locations maintained by Ricardo Energy & Environment. The receptor point is used here is London Bridge Place.

4.7.2 Trajectories associated with pollutant exceedances

Daily PM10

No daily average PM10 concentration exceeded 50 μg m-3.

4.7.3 Trajectories associated with top ten most polluted days

The average daily concentration for each pollutant across all the sites is calculated, with the top 10 most polluted days identified and linked to its back trajectory data in the plot below.

NO2

Figure 25: Trajectory plot for top ten highest daily NO2 concentrations.

PM2.5

Figure 26: Trajectory plot for top ten highest daily PM2.5 concentrations.

PM10

Figure 27: Trajectory plot for top ten highest daily PM10 concentrations.

O3

Figure 28: Trajectory plot for top ten highest daily O3 concentrations.

4.8 Polar plot map

The maps below show polar plots for each pollutant at each monitoring station superimposed on the map to help understand the relative strength and direction of sources. The distance from the plot origin shows the wind speed.

For primary pollutants directly emitted to the atmosphere these can often help to identify a local source especially when multiple plots are used in conjunction to ‘triangulate’ a common source. For pollutants with a secondary component (i.e. formed through chemical reactions in the atmosphere, e.g. NO2, PM2.5 and O3) the directional signature as seen in the measurements may not be as strong.

NO2

Figure 28: Polar plot for NO2.

PM2.5

Figure 29: Polar plot for PM2.5.

PM10

Figure 30: Polar plot for PM10.

O3

Figure 31: Polar plot for O3.



For further information, please contact:

Name Nick Rand
Address Ricardo Energy & Environment, Gemini Building, Harwell, Didcot, OX11 0QR, United Kingdom
Telephone 01235 753484
Email nick.rand@ricardo.com