bristol city council clean air plan outline business case

Bristol City Council Clean Air Plan Outline Business Case

Air Quality Modelling Report (AQ3)

Document: OBC-19

October 2019

Bristol City Council

Document Title


Document No. i

Bristol Clean Air Plan

Project No: 673846.ER.20.01

Document Title: Air Quality Modelling Report (AQ3)

Document No.: OBC-19

Revision: 3

Date: October 2019

Client Name: Bristol City Council

Project Manager: HO

Author: DW & KT

Jacobs Engineering Group Inc. 1 The Square Temple Quay 2nd Floor Bristol BS1 6DG GB +44 117 910 2580 +44 117 910 2581 www.jacobs.com

© Copyright 2019 Jacobs Engineering Group Inc. The concepts and information contained in this document are the property of Jacobs. Use or copying of this document in whole or in part without the written permission of Jacobs constitutes an infringement of copyright.

Limitation: This document has been prepared on behalf of, and for the exclusive use of Jacobs’ client, and is subject to, and issued in accordance with, the provisions of the contract between Jacobs and the client. Jacobs accepts no liability or responsibility whatsoever for, or in respect of, any use of, or reliance upon, this document by any third party.

Document history and status

Revision Date Description By Review Approved

1 04/09/2018 Draft AD PS & CB BL

2 28/01/2019 Final AD PS HO

3 10/09/2019 Draft DW KT & HP HO

4 28/10/2019 Draft DW KT HO


Document No. ii

Contents

Acronyms and Abbreviations ................................................................................................................. 3

1. Introduction .................................................................................................................................. 1

1.1 Background to the Project ............................................................................................................. 1

1.2 Purpose of Report .......................................................................................................................... 2

2. AQ Monitoring .............................................................................................................................. 4

2.1 Background Concentrations .......................................................................................................... 5

3. Base Year 2015 Model Verification ............................................................................................ 6

3.1 Introduction .................................................................................................................................... 6

3.2 Nitrogen Dioxide ............................................................................................................................ 6

4. Target Determination ................................................................................................................... 9

4.1 2021 Reference Case Dispersion Model Results: Nitrogen Dioxide ............................................. 9

4.1.1 Reference Case 2021 .................................................................................................................... 9

4.2 Reference Case 2031 .................................................................................................................. 16

4.3 Final Reference Case .................................................................................................................. 16

5. Source Apportionment .............................................................................................................. 22

6. Options Assessment ................................................................................................................. 24

7. References .................................................................................................................................. 27


Document No. iii

Acronyms and Abbreviations

ADMS Atmospheric Dispersion Modelling System

ANPR Automatic Number Plate Recognition

ATC Automatic Traffic Counters

AQMA Air Quality Management Area

BCC Bristol City Council

CAZ Clean Air Zone

COPERT Computer Programme to calculate Emissions from Road Transport

Defra Department for Environment, Food & Rural Affairs

DfT Department for Transport

EFT Emission Factor Toolkit

GBATS Greater Bristol Area Transport Model

HGV Heavy Goods Vehicle

JAQU Joint Air Quality Unit (Defra and the Department for Transport)

LDV Light Duty Vehicle

LSOA Lower Super Output Area

μg/m3 micrograms per cubic metre

NO2 Nitrogen dioxide

NOx Nitrogen oxides (taken to be NO2 + NO)

PCM Pollution Climate Mapping

PM10 Particulate Matter with an aerodynamic diameter of less than 10 micrometres

PM2.5 Particulate Matter with an aerodynamic diameter of less than 2.5 micrometres

SGC South Gloucestershire Council


1

1. Introduction

1.1 Background to the Project

Poor air quality is the largest known environmental risk to public health in the UK1. Investing in cleaner air and doing more to tackle air pollution are priorities for the EU and UK governments, as well as for Bristol City Council (BCC). BCC has monitored and endeavoured to address air quality in Bristol for decade and declared their first Air Quality Management Area in 2001. Despite this, Bristol has ongoing exceedances of the legal limits for Nitrogen Dioxide (NO2) and these are predicted to continue until around 2029 without intervention.

The UK has in place legislation transposing requirements in European Union law, to ensure that certain standards of air quality are met, by setting Limit Values on the concentrations of specific air pollutants. In common with many EU member states, the EU limit value for annual mean nitrogen dioxide (NO2) is breached in the UK and there are on-going breaches of the NO2 limit value in Bristol. The UK government is taking steps to remedy this breach in as short a time as possible, with the aim of reducing the harmful impacts on public health. Within this objective, the government has published a UK Air Quality Plan and a Clean Air Zone Framework, both published in 2017. The latter document provides the expected approach for local authorities when implementing and operating a Clean Air Zone (CAZ).

Due to forecast air quality exceedances, in 2017 Bristol City Council has been directed by the Minister Therese Coffey (Defra) and Minister Jesse Norman (DfT) to produce a Clean Air Plan to achieve air quality improvements in the shortest possible time. In line with Government guidance, as part of the Plan, Bristol City Council has considered a range of options for the implementation of a Clean Air Zone (CAZ), including both charging and non-charging measures, in order to achieve sufficient improvement in air quality and public health and in line with legal requirements as set out below. This process requires the production of a Strategic Outline Case, an Outline Business Case (this report) and a Full business Case, that will be prepared following the Outline Business Case.

Part of identifying the 28 local authorities, the Pollution and Climate Mapping (PCM) model has been used by Defra to predict annual mean NO2 concentrations at roadsides across the UK. Results are predicted from 2015 to 2031, to identify any exceedances of the annual mean NO2 EU Limit Value (i.e. 40µg/m3) and to provide an indication of when each road link (termed Census ID) will achieve compliance. The PCM modelled results for Census IDs that are exceeding within the BCC administrative boundary are presented in Table 1-1 and Figure 1-1. The year 2021 was selected because this is the latest year where a Census ID is predicted to be non-compliant with regards to the air quality criteria set out in the European Air Quality Directive (AQD), which includes the EU Limit Value for NO2.

Table 1-1. Total NO2 Concentrations at PCM links JAQU 2015 base at 2021 - 1 link

Local Authority Road PCM Census ID 2021 (µg/m3)

Bristol City Council A4032 Newfoundland Way 57291 42.3

1 Public Health England (2014) Estimating local mortality burdens associated with particular air pollution.

https://www.gov.uk/government/publications/estimating-local-mortality-burdens-associated-with-particulate-air-pollution


2

Figure 1-1. Locations of PCM Census IDs, with exceedances of the annual mean NO2 Limit Value in 2021 coloured red

As indicated in Figure 1-1, the Census ID showing non-compliance with the AQD is a section of the A4032 Newfoundland Way (PCM ID 57291). According to the PCM model, all other modelled road links are likely to have achieved compliance with the AQD by 2021. Furthermore the PCM modelling indicates that if local authorities adopt a charging scheme, the Government’s suggests that local authorities could achieve statutory NO2 limit values in most cases by 20212.

1.2 Purpose of Report

Jacobs has been commissioned to support BCC produce an Outline Business Case (OBC) for the delivery of the CAP; a package of measures which will bring about compliance with the EU Limit Value for annual mean NO2 in the shortest possible time in Bristol. The OBC assesses the shortlist of options set out in the Strategic Outline Case3, and proposes a preferred option including details for delivery. The OBC forms a bid to central government for funding to implement the CAP.

This report presents the validation process of the air quality modelling for the BCC Clean Air Plan (CAP) Feasibility study. It includes the future year reference 2021 and 2031 results (i.e. without the CAP measures).

A detailed comparison between the local air quality modelling presented in this report and the national PCM modelling is provided through the Target Determination (TD1) process which has been submitted to JAQU as a separate submission.

2 JAQU Inception Package – Section 2 UK Plan for tackling roadside nitrogen dioxide. 3 Bristol Council Clean Air Plan: Strategic Outline Case, March 2018

(https://democracy.bristol.gov.uk/documents/s19804/Clean%20Air%20Plan%20-%20Cabinet%20Report%20and%20Appendices%20-%20Final%20with%20Early%20Measures%20Fund%20included%20-with%20legal.pdf)


3

AQ3 includes the results of the option testing work (i.e. the results of the modelling for proposed CAP measures) which assists to define the preferred option feeding into the OBC. The preferred option is the one which brings about compliance of the annual mean NO2 EU Limit Value in the shortest possible timeframe.

A draft version of this report was published in January 2019, which supported the draft economic case that was also published at this time. Since this report, further work has been undertaken to develop the scheme options (i.e. packages of measures), and this work is reported in the Option Assessment Report, appended to the OBC.


4

2. Air Quality Monitoring

Air quality monitoring, undertaken by local authorities for local air quality management (LAQM) purposes, has been used in this assessment to provide indications of annual mean NO2 concentrations in areas of interest, and in the model verification process. Monitoring information from both BCC and neighbouring South Gloucestershire Council (SGC) has been used.

Figure 2-1 provides the locations of the monitoring sites in these local authorities and highlights which sites monitored annual mean NO2 concentrations which exceeded the EU Limit Value. The concentrations presented are for 2017, as this is the most recent year that data were available for both local authorities. This data are provided in detail in Appendix B.

Figure 2-1. Air Quality Monitoring Within BCC & SGC – 2017 Annual Mean NO2

The 2017 NO2 monitoring indicates several exceedances of the EU Limit Value, particularly in the city centre (represented by red in Figure 2-1). The likely cause of the exceedances at these locations is a combination of the traffic mix (particularly diesel vehicles), road speed (i.e. slower speeds tending to increase emissions) and presence of canyons (generally tall buildings on either side of the road which prevent pollutants from dispersing as effectively as they would in an open area). BCC diffusion tube locations BCC20 and BCC374 are located alongside the exceeding PCM Census ID 57291 shown in Figure 1-1 and indicate that this location was exceeding in 2017. Note that the map of air quality monitoring indicates that exceedances are not limited to locations around PCM Census ID’s.

As well as providing an indication of pollutant concentrations, a number of these monitoring locations have been used in the model verification process and are presented in Figure 2-2. A total of 85 monitoring sites across BCC and SGC were used in the process. For a full description of the modelling process, see Section 3 of the AQ2 report.


5

Figure 2-2. BCC & SGC Monitoring Sites Used in Model Verification

2.1 Background Concentrations

Estimated background concentrations in the study area have been determined for 2015 and the future year 2021 using Defra’s background maps (Defra, 2018b), with the NO2 values interpolated and adjusted based on a comparison with local measurements. The range of background concentration values interpolated for the different 1x1 km grid squares covering the study area are set out in Table 2-1 and have been derived as described in the Air Quality Modelling Methodology Report (AQ2). The background concentrations are all well below the EU Limit Values.

Table 2-1. Estimated Annual Mean Background Pollutant Concentrations in 2015 and 2021 (µg/m3)

Year NOx NO2

2015 14.3 – 37.3 10.7 – 24.2

2021 11.4 – 36.3 8.7 – 23.6

EU Limit Value - 40


6

3. Base Year 2015 Model Verification

3.1 Introduction

To ensure that ADMS-Roads accurately predicts local concentrations, it is necessary to verify the model against local measurements.

Background concentrations of Nitrogen Dioxide (NO2) for the verification sites have been derived and interpolated from the national maps, using the approach as described in Section 2.10 of AQ2 Local Plan Air Quality Modelling Methodology Report. The background concentrations have then been adjusted based on a comparison with local monitoring.

Annual Average Daily Traffic (AADT) flows, and the proportions of different types of vehicles in the fleet, for all roads, including those adjacent to the monitoring sites, have been determined from the traffic model utilised for this feasibility study (and described in Report T3 Local Plan Transport Modelling Methodology Report). The specifics of the vehicle fleet (i.e. vehicle type, weight, fuel type and emission standard) have been derived from Automatic Number Plate Recognition (ANPR) data collected around Bristol and have been described in Report T3 Local Plan Transport Modelling Methodology Report.

3.2 Nitrogen Dioxide

Most NO2 is produced in the atmosphere by the reaction of nitric oxide (NO) with ozone. It is therefore most appropriate to verify the model in terms of primary pollutant emissions of nitrogen oxides (NOx = NO + NO2). It is also important to only verify that portion of the total concentration which is predicted by the dispersion model (i.e. the background component has been verified and adjusted separately). This is because the alternative (i.e. verifying against the total concentration) risks hiding poor performance in the dispersion model.

The model is unlikely to perform equally well at all locations. However, verifying the model at a large number of sites is preferable to only using a small number of sites, which means making use of a combination of automatic monitors as well as diffusion tube monitors. While measurements made with diffusion tubes are generally of lower quality than those from automatic monitors, they are still considered to be more reliable than the raw results from a dispersion model, particularly one which is only verified for a small number of monitoring sites. The model has thus been run to predict the annual mean NOx concentrations during 2015 at 80 diffusion tube monitoring sites and 5 automatic monitoring sites. Concentrations have been modelled at the inlet height of the monitors, as presented in the BCC and SGC Air Quality Annual Status Reports (ASRs).

The model output of road-NOx has been compared with the ‘measured’ road-NOx, as set out in the JAQU Guidance. Measured road-NOx has been calculated from the measured NO2 concentrations and the predicted background NO2 concentration using the NOx from NO2 calculator (Version 6.1) available on the Defra LAQM Support website4. The results of this comparison indicate that the model consistently under-predicted road-NOx concentrations at the monitoring sites. This is a common experience when predicting NOx concentrations in urban environments using a dispersion model such as ADMS-Roads and emissions from Defra’s EFT, which is the method recommended by the JAQU and thus followed. The accepted approach to addressing this under-prediction, following both the Defra and JAQU guidance, is to increase the model outputs in line with measurements. An adjustment factor of 2.281 has therefore been applied to modelled NOx emissions.

The EFT (v8.0.1a) has been used to estimate road NOx and NOx as primary NO2 (i.e. the primary NO2 (f-NO2) emitted directly from motor vehicles, as a factor of NOx). Road NOx and NOx as NO2 were input to ADMS Roads to estimate location specific Road NOx and NOx as NO2 concentration values for each receptor. These location specific values have then been used within the conversion of NOx to NO2 (using the NOx to NO2 calculator (Version 6.1) supplied by Defra). This is the approach recommended by JAQU.

4 Defra. (2018b). Local Air Quality Management (LAQM) Support Website. Retrieved from http://laqm.defra.gov.uk/


7

The total NO2 concentrations were determined by combining the adjusted modelled road-NOx concentrations with the predicted background NO2 concentrations and f-NO2 values within the NOx to NO2 calculator.

Figure 3-1 compares final adjusted modelled total NO2 at each of the monitoring sites to measured total NO2.

Figure 3-1. Comparison of Measured Total NO2 to Final Adjusted Modelled Total NO2 Concentrations. The dashed lines show ± 25%.

Table 3-1 shows the statistical parameters relating to the performance of the model, as well as ‘ideal’ values5, based on total annual mean NO2 concentrations. The values calculated for the model demonstrate that it is performing acceptably.

5 Defra. (2018b). Review & Assessment: Technical Guidance LAQM.TG16 February 2018 Version. Defra. Retrieved from

https://laqm.defra.gov.uk/documents/LAQM-TG16-February-18-v1.pdf


8

Table 3-1. Statistical Model Performance

Statistical Parameter Model-Specific Value ‘Ideal’ Value

Correlation Coefficient a 0.6718 1

Root Mean Square Error (RMSE) b 7.56 0

Fractional Bias c 0.03 0 a Used to measure the linear relationship between predicted and observed data. A value of zero means no relationship and a value

of 1 means absolute relationship. b Used to define the average error or uncertainty of the model. The units of RMSE are the same as the quantities compared (i.e.

µg/m3). TG(16) (Defra, 2018b) outlines that, ideally, a RMSE value within 10% of the air quality objective (4 µg/m3) would be derived. If RMSE values are higher than 25% of the objective (10 µg/m3) it is recommended that the model is revisited.

c Used to identify if the model shows a systematic tendency to over or under predict. Negative values suggest a model over-prediction and positive values suggest a model under-prediction.


9

4. Target Determination At this point its worth being reminded that much effort was afforded to ensure that modelling receptors complied with criteria included in the AQD. Receptors which were automatically located adjacent to Census IDs were individually revisited and stamped as reportable or non-reportable. In other words, all receptors identified as being within 25m of a junction and / or not within 100m of uninhibited road space were discounted. Careful attention was also given to receptors placed in areas with no public access. Clearly in constricted urban areas, these criteria would lead to discounting many receptors and so any removals were carefully considered.

4.1 2021 Reference Case Dispersion Model Results: Nitrogen Dioxide

The following outputs were required in order to fulfill the Target Determination submission:

Maximum NO2 concentrations for all PCM Census IDs for the base year 2015, compliance year 2021 and years in between. The format was in line with the template in Annex D of the Evidence Package.

A shapefile or list of receptor coordinates so that the maximum concentration can be mapped to each Census ID.

4.1.1 Reference Case 2021

Target Determination Report 1 (TDR1) was forwarded to BCC from the JAQU on 15th April. This report was compiled in response to 2021 TDS1 Submission Revision 2 (REV2). This report identified 14 Census IDs warranting further investigation and 1 Census ID which was not included in the local modelling. In responding to the TDR1 BCC noticed a few irregularities in the TDS1 REV2 database. These irregularities which included the incorrect selection of receptors representing the maximum concentration for each PCM Census ID, were systematic of an incorrect linkage to a specific result file. It was decided therefore that to understand the implications of these irregularities TDS1 would need to be recompiled. The resulting compilation is presented in Table 7-1. BCC & SGC 2017 Monitoring Results

Local Authority

Monitoring Type ID X Y Annual Mean NO2 (µg/m3)

BCC Continuous Analyser 463 362926 175590 39.1

BCC Continuous Analyser 375 359645 173683 N/A




BCC Diffusion Tube BCC10 361217 171429 51.6















10

Local Authority














































11

Local Authority














































12

Local Authority








SGC Diffusion Tube SGC17 364830 173878 N/A

SGC Diffusion Tube SGC22 364116 172413 29.6






























13

Appendix A. Target Determination

Table 7-2 (Appendix C). The table shows the maximum annual mean NO2 concentration for REV2 compared to the REV3 (i.e. Appendix C of this report). Receptors selected for TDS1 were those considered reportable in line with Guidance issued by the JAQU and allowing target determination analysis to be conducted.

Under the compliance heading Table 4-1 indicates the status of the relationship between the datasets (e.g. “match” means same compliance status in REV2 and REV3). For example, for PCM Census ID 6401 the concentration increased by 55.4% to 34.3 µg/m3. However, the annual mean is still below the AQD EU Limit Value of 40.0 µg/m3. It is worth noting that the assessment considers compliance to be below 40.0 µg/m3, which is commensurate with UK Air Quality Objectives (AQO). This threshold is in accordance with how BCC interprets compliance. The status of compliance for REV3 therefore matches that of REV2 (i.e. match-compliant). For PCM Census ID 7097 there isn’t a compliance match between REV2 and REV3. The same systematic irregularity occurred for 2015. A summary of the changes observed between REV2 and REV3 for all 126 TDS1 receptors are shown in Table 4-1.

Table 4-1. Summary of the differences between REV2 and REV3 (number of receptors)

Statistical Parameter 2015 2021

Match 83 106

Wasn’t compliant, now is 8 2

Was compliant, now isn’t 35 18

Note that there were three Census IDs omitted from REV3 compared with REV2. These were as follows and the reason for their omission in brackets:

• 70103 (This was an elevated-on slip to the M5 with no relevant exposure/reportable locations).

• 74766 (M32 subway section under roundabout, hence no reportable locations).

• 74769 (Easton Way Junction with M32 and on off slips north of junction. No reportable receptors).

Questions raised by the JAQU in response to REV3 are presented in Table 4-2.

Table 4-2. Questions raised, and responses returned concerning the REV3 submission

JAQU Question Response

1 The data is significantly different to the previous data submitted to us.

Agreed, there are differences between REV2 and REV3. These differences are explained in the

Discussion Note issued to the JAQU 13th June 2019.

2 Concentrations are, on average, higher in the new data.

Agreed, on average they are higher - the Discussion Note issued to the JAQU 13th June 2019 provides a comparison of the REV2 and

REV3.

3 The maximum concentration on a PCM model road link is lower in the new data.

Agreed, the maximum concentration on the PCM network in REV2 was link 36409 at 62.8 µg/m3, in

REV3 this is now 41.1 µg/m3. For REV3 the highest concentration is estimated to be 58.8 µg/m3 at location Census ID 7617 which is

adjacent to the Wills Building on Queens Road.


14

JAQU Question Response

4 Several census IDs included in the previous dataset are not in the new dataset: 70103, 74766

and 74769.

Agreed, these have been explained in the Discussion Note issued to the JAQU 13th June

2019.

5 I believe an issue was raised that the highest exceedances were not included in the TD

submission. As well as PCM roads, exceedances on all road links will need to be tackled.

This is fully understood. Compliance is to be achieved across the whole study area

irrespective of whether it is a PCM Census ID or not.

In summary the implications of changes to the 2021 reference case (REV3) included the following:

Issues were associated with extracting data from the dispersion model – rather than the model itself. The changes to the target determination data did not result in the need to update the model

calibration work set out in the AQ2 report. The future year option assessment work was not affected by the change in target determination

data.

Having confirmed TD1 REV3 and hence the 2021 reference case BCC responded to the JAQU with a TD2 data submission on the 31st July 2019.

The locally predicted annual mean concentrations of NO2 (REV3) at PCM equivalent receptors showed exceedances of the annual mean EU Limit Value in 2021 at various locations, predominantly in central Bristol. In particular, the following 6 hotspots were identified: Rupert Street; Park Street; Queen's Road; College Green; Newfoundland Way; and Church Road. The predicted NO2 concentrations in 2021 at PCM equivalent receptor locations are shown in Figure 4-1. Census IDs were highlighted along its length where the maximum annual mean NO2 concentration exceeded the EU Limit Value. In addition to PCM equivalent receptors, another 4 localised hotspots were identified at Marlborough Street; Upper Maudlin Street; Park Row and Cheltenham Road. The CAP is to consider all hotspots equally irrespective of the presence of a Census ID.


15

Figure 4-1 Predicted NO2 concentrations in 2021 at PCM-equivalent receptor locations (REV3).


16

4.2 Reference Case 2031

The predicted annual mean concentrations of NO2 at PCM and non-PCM equivalent receptors show that all reportable locations are predicted to be compliant in 2031.

4.3 Final Reference Case 2021

REV3 was submitted to the JAQU in May 2019 and the subsequent TD2 data report at the end of July. Further post consideration of the 2021 reference case was discussed by the client team and as a result it was confirmed that in 2021, owing to ongoing policies, emission standards of buses and taxis were likely to be cleaner than had previously been estimated for REV3. On this basis REV3 was updated to become the Final Reference Case (termed Reference Case in the following section of this Report) against which all options have been compared. The JAQU were informed of this revision via various meetings. Owing to the relatively small impact these updates had on REV3 it was considered that no commensurate revision of the TD submission would be required. The Reference Case is shown in Figure 4-2. The Reference Case and REV3 are very similar apart from the absent Census ID’s located on the Fishponds Road between Lodge Causeway and the Cross Hands junction and Brislington Hill on the A4 Bath Road. Whilst the full length of the Census ID is shown, the prevalence of an exceedance only occurs at isolated receptors.

As mentioned in Section 4.1.1 ten locations were identified in the Reference Case driving the compliance year of the NO2 EU Limit Value. These locations are shown in Figure 4-3 and Figure 4-4 respectively. The maps show reportable receptors at PCM-equivalent locations (i.e. receptors meeting AQD siting criteria). The red receptors were predicted to exceed the EU Limit Value in 2021 and the green to comply. What is evident from the plots is that some receptors within the boxes show compliance whilst others show noncompliance. For example, Upper Maudlin Street is showing noncompliance on the northern side of the road and compliance on the southern side. Pollution on Upper Maudlin Street is confounded by street canyon effects which can act to elevate concentrations in very specific locations. The Rupert Street area largely encompasses the access to the City Centre area and as such has a high concentration of bus services constrained within street canyons. The City Centre itself is reasonably open and has better dispersion characteristics. There are several Plane trees on the City Centre which can limit dispersion in the spring and summer periods, but these effects cannot be easily accounted for using Gaussian based pollution dispersion models (such as ADMS). Hence, the City Centre was estimated to be compliant in 2021. From the City Centre traffic traverses through College Green and onto Anchor Road and or onto Park Street. The latter is a focus area largely owing to its steep gradient and street canyons. Most maximum noncompliance is towards the upper reaches of Park Street and onto Queens Road where a single reportable noncompliance was estimated. Park Row on to Maudlin and Marlborough Street are non PCM road links. These areas are particularly sensitive locations as they provide access to Bristol’s Royal Infirmary. Other locations driving the earliest compliance year included Cheltenham Road between Ashley Road and Colston Girls School, Newfoundland Way which leads onto the M32 and Church Road (shown separately in Figure 4-4).

A very detailed examination was conducted on Church Road to understand what was driving the exceedance of the EU Limit Value. The study team considered that perhaps adjustment in the dispersion model at this location could have been considered independently of other locations taking the view that Church Road had morphology which was in some way unique. In general, models tend to underestimate emissions. At Upper Maudlin Street (UMS) for example emissions were underestimated more so than on Church Road (CR). After applying a global adjustment, the modelled verses measured at UMS was very close but as shown in Table 4-3 the model appeared to be over estimating on Church Road. The relationship is variable across monitoring sites, so the modelling will be closer to the monitoring at some locations, and further at others. The study team had to decide how to best set up the dispersion model, to represent the study area and achieve specific confidence criteria. Church Road has street canyon issues that elevated concentrations. The background NO2 concentration was lower at Church Road compared to UMS which was to be expected. The Road NOx was as expected given the traffic forecast. However it is plausible that the traffic flows on Church Road have been over estimated.


17

Table 4-3. Monitored verses measured NO2 concentrations.

Site ID 2015 Annual Mean NO2 Concentration (µg/m3)

Background Monitored Modelled

BCC9 (UMS) 20.0 48.0 48.4

BCC496 (Church Road) 15.0 39.3 49.5


18

Figure 4-2 Predicted NO2 concentrations in the 2021 Final Reference Case at PCM-equivalent receptors


19

Figure 4-3 Predicted NO2 concentrations in the 2021 Final Reference Case at PCM-equivalent receptor locations in the City Centre Focus Areas


20

Figure 4-4 Predicted NO2 concentrations in the 2021 Final Reference Case at PCM-equivalent receptors in the Church Road Focus Area


21

Statistics relating to the list of priority areas driving the compliance year are shown in Table 4-4. The table indicates where a location is associated with a PCM Census ID. It shows the AADT combined two-way flow, the percentage of HGVs, the NOX emissions rate used in the pollution dispersion model and the maximum NO2 concentration. Newfoundland Way has the highest traffic flows being the main access route into the City Centre and the percentage of HGVs there reflect levels of goods vehicles, as opposed to Rupert Street which reflects more so the numbers of buses. Upper Maudlin and Marlborough Street have relatively low levels of HDV’s (the majority of which are goods vehicles) but in combination with their associated traffic flows have the highest emission rates (i.e. discounting Newfoundland Way). The lowest percentage of HDVs occurring on Park Row. Comparing emissions rates between Park Street, which has the lowest traffic flows, and for example Rupert Street, the rates appear to reflect the gradient conditions on Park Street. Emission rates for Church Road and College Green are very similar. The traffic on Church Road is somewhat higher but the levels of HDVs are lower. This points to HDVs being disproportionate in terms of emissions. Also, road speed on Church Road is higher than on College Green, lower average roads speeds generally increase emissions from vehicles.

Table 4-4. Localised predicted hotspots at 2021 and air quality source conditions

Road Name PCM Census ID

AADT (2 Way)

% of HDVs NOx emission rate (g/km/s)

Maximum NO2 (µg/m3)

Rupert Street 74772 24,986 13.2% 0.1099 50.0

Marlborough Street - 25,989 1.9% 0.1524 59.4

Upper Maudlin Street - 29,716 2.0% 0.1721 46.9

Park Row - 18,750 0.9% 0.1123 50.5

Park Street 7617 13,298 9.4% 0.1238 49.1

Queens Road 7617 21,476 6.4% 0.1439 41.9

College Green 74770 15,903 10.0% 0.1356 49.1

Cheltenham Road - 17,509 4.2% 0.1166 40.5

Newfoundland Way 57291 52,877 2.2% 0.2695 50.6

Church Road 47131 27,932 3.0% 0.1340 53.3

4.4 Conclusion

In conclusion, if the traffic situation was deemed to be reasonable and the background concentrations acceptable then (in the absence of further analysis of these parameters in combination with the assumption of street canyons) the consensus was that the modelling represented the operational expectancy of emissions in 2021.


22

5. Source Apportionment

A source apportionment for year 2021 was undertaken as the basis of understanding the types of measures to be considered at each of the locations driving the compliance year. The results are shown in Table 5-1. At each location the annual mean road NOx at the maximum reportable receptor is provided as a percentage of the total concentration shown in the second column (µg/m3). The following lists the key observations:

Diesel cars followed by diesel LGVs have the highest proportional NOx impact across all locations.

Petrol cars have a relatively low NOx impact given that they represent around half of the car fleet.

Bus and coach impacts are greatest on Rupert and Park Streets. The lowest impact from buses occurs at Marlborough Street and Church Road.

The highest impact from Rigid HGVs is on Queens Road, Park Street, and College Green. These roads form a contiguous mini route and Park Street has a gradient affecting emissions from heavy vehicles (as applied in the modelling methodology).

The highest articulated HGV emissions are on Park Street and are affected by the gradient.

LGV NOx emission are highest on Church Road. This may reflect the high number of small shop traders. Cheltenham Road has a similar morphology and similar LGV emissions.

Lower proportionate petrol and diesel car emissions on Park Street combined with higher bus and coach emissions (irrespective of the gradient) may reflect the importance of accessing this location using public transport.

Table 5-1. Road NOx Source Apportionment for the Receptor with the Highest Modelled Annual Mean NO2 Concentration in the 2021 Final Reference Scenario

Road Name Road NOx

(µg/m3)

NOx Concentration Source Apportionment (% of Road NOx) a

Petrol Cars

Diesel Cars

Petrol LGVs

Diesel LGVs

Rigid HGVs

Artic HGVs

Buses / Coaches

Rupert Street 71.8 7.8 42.0 0.2 30.1 4.3 1.0 14.5

Marlborough Street 93.9 8.4 49.2 0.2 29.8 7.9 1.9 2.6

Upper Maudlin Street 59.6 8.5 48.7 0.2 29.4 8.3 1.9 3.0

Park Row 69.4 8.5 50.9 0.2 30.7 6.0 1.4 8.5

Park Street 76.0 5.0 29.8 0.1 20.6 23.5 5.9 15.1

Queens Road 53.4 6.7 39.5 0.2 26.5 12.9 3.1 11.1

College Green 71.4 6.5 37.9 0.2 26.2 16.2 3.8 9.3

Cheltenham Road 50.0 7.0 40.5 0.2 31.6 5.2 1.3 14.1

Newfoundland Way 73.0 8.9 49.2 0.2 32.3 4.6 1.1 3.7

Church Road 84.4 8.3 45.3 0.3 35.2 6.6 1.5 2.8 a This does not take the background contribution into account

It’s worth noting that any measure which reduces the amount of traffic at all locations would be beneficial but not particularly effective. This can be achieved by using traffic signal to effectively hold traffic back at different locations on a cordon basis. The source apportionment suggests that measures limiting diesel vehicles would have the greatest impact. Even so, measures perhaps affecting discretionary journeys would


23

be more desirable. A solution perhaps would be one which limits diesel cars allowing for a transition to clean public transport. There maybe scope to limit the levels of HGVs accessing the City Centre at least on Park Street and College Green. There would appear to be no particular gain in restricting petrol vehicles, although restricting petrols to >Euro 4 would be sensible. LGVs do contribute to a third of NOx emissions but equally are important in terms of making deliveries to business and shops. It is clear that measures to restrict diesel LGVs would be beneficial, in order to retain economic vitality, it would be sensible to allow only the cleaner vehicles such as that proposed for a CAZ C.


24

6. Options Assessment Following on from examining the causes of non-compliance through source apportionment (as demonstrated in Section 5), the next step was to propose and assess different air quality mitigation measures. The primary objective of the CAP is to bring compliance across Bristol in the shortest possible timeframe, and the key success factor is therefore the earliest year where all modelled annual mean NO2 concentrations are below 40 µg/m3 (i.e. at PCM equivalent reportable receptors). The source apportionment is used to identify the direct contribution of emissions from the different sources to the total concentration6 at each receptor. This then enables bespoke solutions to be assessed.

After the Final Reference scenario was modelled and the source apportionment undertaken, the 4 core options assessed were:

Option 1 (CAZ C);

Option 2 (Diesel car ban);

Medium CAZ D +; and

Hybrid (Option 1 + Option 2).

Option 1 and Option 2 were the first scenarios that were fully assessed and presented at public consultation. Following the public consultation, BCC considered that an earlier compliance year was desirable which meant that more stringent measures needed to be developed. The outcomes of this process were the Medium CAZ D + and Hybrid scenarios. Further details of the option development work are reported in the Option Assessment Report appended to the OBC.

The results of these scenarios are presented in Table 6-1. Modelling of the 2021 Final Reference scenario indicated a total of 138 non-compliant, reportable receptors. The largest exceedances of the EU Limit Value were located across the focus areas presented in Section 4. It is worth noting that non-compliant receptors were identified in other locations outside of the focus areas. These however were less significant, predominantly because the exceedances were smaller and estimated to achieve compliance by natural means, or as a consequence of measures aimed at tackling the exceedances within the focus areas.

Table 6-1. Number of Non-Compliant Receptors Per Year, Per Scenario

Road Name Number of Non-Compliant Receptors in Each Scenario (2021)

Final Reference

Option 1 Option 2 Medium CAZ D +

Hybrid

2021 138 65 104 57 62

2022 97 47 72 45 37

2023 64 40 46 37 25

2024 48 32 37 22 18

2025 33 23 21 14 7

2026 26 14 14 12 37

2027 10 4 4 0 0

2028 6 4 0 0 0

2029 5 0 0 0 0

2030 0 0 0 0 0

6 The total concentration in this respect is the direction contribution from road sources and does not consider the background component. 7 Exceedances in 2026 and 2027 are on Church Road.


25

Further detail is provided on the options assessment and their compliance years in the following sections.

6.1 Option 1

Option 1 comprises the following measures:

Medium CAZ C – i.e. a charging scheme for non-compliant taxis (£9), LGVs (£9), HGVs (£100) and buses & coaches (£100) covering a medium sized area over Bristol city centre;

HGV weight restriction (3.5 tonnes) operating 24/7 on Rupert St, Baldwin St, Park Row, Upper Maudlin St, Marlborough St and Lewins Mead;

Diesel car ban on Upper Maudlin Street & Park Row operating 7am – 3pm, 7 days a week (not applicable to taxis, private hire vehicles or emergency vehicles);

Bus and local traffic interventions on the M32, Cumberland Road, and others;

M32 Park and Ride (P&R) with bus lane inbound; and

Scrappage scheme (up to £2,000) for private diesel cars.

The implementation of Option 1 has a significant impact in 2021, as the 65 anticipated non-compliances are less than half of the number modelled in the Final Reference case. The Option 1 compliance year is anticipated to be 2029. It achieves compliance on Marlborough Street and Church Road (the two most significant locations in this assessment) by 2027, but is held back by non-compliance at 4 receptors on Park Street. All other locations are anticipated to be compliant prior to 2027.

Park Street’s high annual mean NO2 concentrations in the 2021 Final Reference scenario are predominantly caused by high proportions of diesel cars, LGVs and HGVs, as well as the presence of a gradient, which exacerbates pollutant emissions from HGVs in particular. Unfortunately, the Upper Maudlin Street / Park Row diesel ban was predicted to cause a percentage of non-compliant vehicles to reroute onto Park Street increasing annual mean NO2 concentrations at reportable receptors. This is one of the main reasons that Option 1 by itself was not taken forwards.

6.2 Option 2

As Option 1 was still being considered, other options were being developed to see if a compliance year of 2029 could be bettered. Option 2 comprises a diesel car ban within a specific boundary (i.e. smaller than the medium area in Option 1) in central Bristol operating between 7am and 3pm, 7 days a week. This is a simpler but more stringent solution.

Option 2 did not appear to have an immediate significant impact on compliance across the network when compared with Option 1. It reduced the number of non-compliances by 34 to a total of 104 in 2021, which is attributable to the fact that it covers a much smaller area than the measures in Option 1.

However, Option 2 achieves a compliance year of 2028, which is one year earlier than Option 1. The additional number of non-compliances in the 2021 Option 2 scenario are locations outside of the diesel ban zone, with less significant exceedances of the EU Limit Value. Locations with smaller exceedances of the EU Limit Value are more likely to be brought to compliance naturally (i.e. owing to improvements in vehicle emissions technology), without specific intervention. The diesel ban targets the areas with the largest modelled annual mean NO2 concentrations where diesel emissions have been shown to significantly contribute to the non-compliance.

Table 6-1 indicates that there are 4 non-compliant, reportable receptors in 2027 with Option 2 implemented. One of these is located on Marlborough Street, with the remaining three located along Church Road. The Marlborough Street exceedance is considered to be caused by a combination of high levels of diesel cars and LGVs, which is exacerbated by the presence of tall ‘canyons’ which prevent pollutant dispersion.


26

Pollutant concentrations in canyons are often much worse than those outside of canyons with similar traffic characteristics (e.g. flows and speeds). The non-compliances on Church Road are considered to be caused by elevated road traffic emissions which are exacerbated by the presence of a shorter canyon.

6.3 Medium CAZ D + Option 1 (Benchmark Option)

Whilst Options 1 & 2 were presented at public consultation, BCC further considered running a benchmark CAZ D type scenario as a comparison. The Medium CAZ D + scenario comprises the same measures as Option 1, but with the addition of a £9 charge for non-compliant cars that enter the area.

The source apportionment results presented in Table 6-1 indicates that diesel cars are a significant contributor to poor air quality, often the most significant source. Upgrading Option 1’s CAZ C to a CAZ D has an immediate benefit, as the 65 non-compliances are reduced to 57 in this scenario.

Implementing a CAZ D instead of a CAZ C has additional benefits, as it also brings the compliance year forwards by 2 years (compared to Option 1), so that compliance is now achieved in 2027. In this case though, there are 12 receptors which prevent a compliance year of 2026. These are located on Park Street, Baldwin Street, Marlborough Street and Church Road. The non-compliances on Park Street, Marlborough Street and Church Road have been previously discussed, and the reasoning is relevant here. The only differences at these locations are that the Medium CAZ D leads to small improvements at these three locations, which bring compliance forward by a year.

Baldwin Street is an example of a non-focused area, which still presents issues in this scenario. There are two receptors at this location which are anticipated to be non-compliant in 2026. The largest contributors to the high modelled annual mean NO2 concentrations at this location are diesel cars, diesel LGVs and rigid HGVs, although it is likely that these are worsened by the presence of a canyon.

The purpose of the Medium CAZ D + was to act as a new benchmark case for future scenarios, and to compare previously assessed scenarios against.

6.4 Hybrid

The hybrid scenario combines all the aspects of Option 1, with Option 2 (i.e. Option 1 with a diesel car ban). It is similar to the Medium CAZ D + scenario, as it builds on previous solutions, but introduces a diesel car ban, rather than upgrading the CAZ C to CAZ D. It has slightly less immediate impact on compliance than the Medium CAZ D + scenario, as it reduces the number of non-compliances to 62. Similarly, it achieves a compliance year of 2027.

The Hybrid scenario can be considered more successful than the others assessed and reported here, as there are only 3 anticipated exceedances of the annual mean NO2 EU Limit Value in 2026, all of which are located along Church Road. This indicates that the Hybrid scenario is anticipated to bring compliance in the City center forwards to 2025, which is an improvement compared to all other modelled scenarios. The Hybrid option is therefore the best scenario to take forward. If additional measures can be focused on Church Road it may be possible to achieve total compliance of the EU Limit Value in 2025.

6.5 Conclusion

The four options proposed and reported in here are evidence based and evolved over time and focused on where improvements were needed most. Option 1 acted as a strong building block for further proposals, with more effective solutions being developed as a result.

The aim of the Bristol Clean Air Plan is to achieve compliance with the annual mean NO2 EU Limit Value in the shortest possible timeframe, which is in line with the Guidance provided by the JAQU. The Hybrid option is therefore the best proposition to take forward, as it achieves the joint earliest total compliance year (2027), and compliance in the City Center by 2025.

Following the completion of the modelling work reported in this document, additional modelling work has been undertaken to confirm the compliance data – see OBC-20 “Church Road assessment and 2025 modelling” appended to the OBC for details.


27

7. References Defra. (2017a). Air quality plan for nitrogen dioxide (NO2) in the UK. Retrieved from

https://www.gov.uk/government/publications/air-quality-plan-for-nitrogen-dioxide-no2-in-uk-2017 Defra. (2018b). Local Air Quality Management (LAQM) Support Website. Retrieved from

http://laqm.defra.gov.uk/ Defra. (2018b). Review & Assessment: Technical Guidance LAQM.TG16 February 2018 Version. Defra.

Retrieved from https://laqm.defra.gov.uk/documents/LAQM-TG16-February-18-v1.pdf Defra. (2018b). Review & Assessment: Technical Guidance LAQM.TG16 February 2018 Version. Defra.

Retrieved from https://laqm.defra.gov.uk/documents/LAQM-TG16-February-18-v1.pdf


28

Appendix B. BCC & SGC 2017 Monitoring Results

Table 7-1. BCC & SGC 2017 Monitoring Results

Local Authority











































29

Local Authority














































30

Local Authority














































31

Local Authority


















32

Appendix C. Target Determination

Table 7-2. TDS1 – Comparison between REV2 and REV3 (2021)

Census

ID

Rev 2 (Old) Rev 3 Difference

(ug/m3)

Difference

(% of Old)

Compliance

Local NO2 Local NO2 Rev2 Rev3 Compliant / Non‐Comp Status

6401 22.1 34.3 12.2 55.4% Yes Yes Match8 ‐ Compliant

7046 25.9 35.5 9.7 37.4% Yes Yes Match ‐ Compliant

7097 41.5 23.9 ‐17.6 ‐42.4% No Yes Doesn't Match Wasn't


7617 27.5 58.8 31.4 114.2% Yes No Doesn't Match Was





8391 35.3 35.2 ‐0.1 ‐0.4% Yes Yes Match ‐ Compliant



16123 41.6 41.6 0.0 0.0% No No Match ‐ Non‐Compliant





























8 Match means same compliance status in REV2 and REV3.


33

Census

ID


(ug/m3)

Difference

(% of Old)

Compliance







36409 62.8 41.1 ‐21.7 ‐34.5% No No Match ‐ Non‐Compliant










































34

Census

ID


(ug/m3)

Difference

(% of Old)

Compliance


















74768 40.9 37.2 ‐3.8 ‐9.2% No Yes Doesn't Match Wasn't























bristol city council clean air plan outline business case

Documents