JAQU Air Quality Modelling Report - Portsmouth City Council
←
→
Page content transcription
If your browser does not render page correctly, please read the page content below
JAQU Air Quality Modelling Report AQ3 December 2020
JAQU Air Quality Modelling Report
Quality information
Prepared by Checked by Verified by Approved by
Alice Gurung Helen Venfield Anna Savage Gareth Collins
Graduate Air Quality Principal Air Quality Associate Air Quality Regional Director
Consultant Consultant Director
Revision History
Revision Revision date Details Authorized Name Position
1 June 2019 Working Draft GC Gareth Collins Regional
Director
2 September 2019 Updated GC Gareth Collins Regional
Working Draft Director
3 October 2019 Final GC Gareth Collins Regional
Director
4 February 2020 Updated with T- GC Gareth Collins Regional
IRP comments Director
5 November 2020 Updated draft for GC Gareth Collins Regional
FBC Director
6 December 2020 Final version for GC Gareth Collins Regional
FBC Director
Distribution List
# Hard Copies PDF Required Association / Company Name
Changes for FBC version highlighted in yellow.
2
JAQU Air Quality Modelling Report Prepared for: Portsmouth City Council and Joint Air Quality Unit (JAQU) Prepared by: Alice Gurung Graduate Air Quality Consultant T: +44(0)020 043 9340 E: alice.gurung@aecom.com AECOM Infrastructure & Environment UK Limited Sunley House 4 Bedford Park, Surrey Croydon CRO 2AP United Kingdom T: +44 20 8639 3500 aecom.com © 2019 AECOM Infrastructure & Environment UK Limited. All Rights Reserved. This document has been prepared by AECOM Infrastructure & Environment UK Limited (“AECOM”) for sole use of our client (the “Client”) in accordance with generally accepted consultancy principles, the budget for fees and the terms of reference agreed between AECOM and the Client. Any information provided by third parties and referred to herein has not been checked or verified by AECOM, unless otherwise expressly stated in the document. No third party may rely upon this document without the prior and express written agreement of AECOM. 3
JAQU Air Quality Modelling Report
Table of Contents
1. Introduction ................................................................................................................................... 7
2. Atmospheric Dispersion Modelling Approach ............................................................................... 9
2.1 Dispersion Model Selection and Tools ................................................................................ 9
2.2 Assessment Scenarios ....................................................................................................... 9
2.3 Model Input Data ................................................................................................................ 9
2.3.1 Traffic Data ......................................................................................................................... 9
2.3.2 Baseline Emissions Inventory Development (NO x and f-NO2) ........................................... 9
2.3.3 Road Width Data .............................................................................................................. 12
2.3.4 Road Source Emission Rates (NOX and f-NO2) ............................................................... 12
2.4 Gradients, Tunnels, Flyovers and Street Canyon Effects................................................. 12
2.4.1 Road Gradient Effects ...................................................................................................... 12
2.4.2 Street Canyons ................................................................................................................. 13
2.4.3 Flyovers and Tunnels ....................................................................................................... 13
2.5 Surface Roughness and Minimum Monin-Obukhov Length ............................................. 13
2.6 Meteorological Data .......................................................................................................... 13
2.7 Modelled Receptor Selection ............................................................................................ 14
2.8 Model Output Data ........................................................................................................... 14
2.8.1 Base year 2018 and Projected Base Year 2022 ............................................................... 14
2.8.2 Interim and Future Base Year Interpolation ...................................................................... 14
3. Model Verification and Adjustment .............................................................................................. 15
3.1 Comparison of Modelled (Unadjusted) and Monitored Road NOx ................................... 15
3.2 Verification ........................................................................................................................ 15
3.3 Modelled Road NOx Adjustment ...................................................................................... 15
3.4 Model Adjustment Summary ............................................................................................. 17
4. Baseline Results ......................................................................................................................... 18
4.1 Vehicle Fleet ..................................................................................................................... 18
4.2 NO2 concentrations ........................................................................................................... 18
5. Source Apportionment ................................................................................................................ 24
5.1 Road Vs Non-Road Contribution ...................................................................................... 24
5.2 Road Contributions by Vehicle Type ................................................................................. 25
6. Options Modelling ....................................................................................................................... 26
6.1 Shortlisted Options ........................................................................................................... 26
6.2 Option Assumptions .......................................................................................................... 28
6.3 Options Model Results ..................................................................................................... 29
6.3.1 Identification of Benchmark .............................................................................................. 29
6.3.2 Identification of Alternative Package................................................................................. 32
6.3.3 Other options not taken forward ......................................... Error! Bookmark not defined.
7. Limitations and Assumptions ...................................................................................................... 49
7.1 Local Air Quality Model Limitations................................................................................... 49
7.2 Transport Model Limitations ............................................................................................. 50
7.3 Sensitivity tests ................................................................................................................. 50
7.3.1 Change in transport assumptions ..................................................................................... 52
7.3.2 Change in air quality assumptions.................................................................................... 53
Appendix A Model Verification ............................................................................................................... 57
Appendix B Supporting Information ...................................................................................................... 59
Appendix C Quality Assurance of Monitoring Data ............................................................................... 80
QA / QC of automatic monitoring ................................................................................................ 80
QA / QC of diffusion tube monitoring .......................................................................................... 82
4
JAQU Air Quality Modelling Report Figures Figure 2-1 ANPR camera locations ....................................................................................................... 10 Figure 2-2 Comparison of Portsmouth Vehicle Fleet with National Fleet – Car Fuel Split ................... 11 Figure 2-3 Comparison of Portsmouth Vehicle Fleet with National Fleet – Euro Proportions .............. 11 Figure 3-1 - Modelled Road-NOx versus Monitored Road-NOx ........................................................... 16 Figure 3-2 - Adjusted NO2 versus Monitored NO2 Concentrations ....................................................... 16 Figure 4-1: Location of roadside receptor sites with modelled exceedances in 2022 baseline (EFT v1.9b) .................................................................................................................................................... 20 Figure 6-1: Indicative boundary for Portsea Island CAZ ....................................................................... 27 Figure 6-2: Indicative boundary for Small Area CAZ ............................................................................ 28 Figure 7-1: Wind roses (wind speed/direction) at Thorney Island Met Office site ................................ 53 Tables Table 2-1: Split of cars by compliant and non-compliant ...................................................................... 12 Table 4-1 Current and predicted vehicle fleet and non-compliance to CAZ emission standards, ........ 18 Table 4-2: Locations with modelled (or near) exceedances in 2022 Baseline ...................................... 21 Table 4-3: Comparison between modelled and observed traffic data, Church Street, 2019 ................ 23 Table 5-1 Percentage Contribution of Road and Non-Road Sources to background NOx in selected areas of Portsmouth, 2018 .................................................................................................................... 24 Table 5-2: Percentage Contribution of Road and Non-Road Sources to NOx in selected areas of Portsmouth, 2022 .................................................................................................................................. 25 Table 5-3 Contribution of vehicle type to road NO x emissions on exceedance road links, 2022 Future Base ...................................................................................................................................................... 25 Table 6-1: Options modelled for 2022 ................................................................................................... 26 Table 6-2: Assumed responses of LGVs and HGVs to a CAZ (based on JAQU data) ......................... 29 Table 6-3: Modelled NO2 concentrations (µg/m 3) in 2022 for different options (Based on Emissions Factor Toolkit, v9.1b) ............................................................................................................................. 33 Table 6-4: Annual mean NO2 concentrations for individual non-charging measures, 2022.................. 35 Table 7-1 HGV response assumptions for the core scenario and sensitivity tests ............................... 52 Table 7-2 Impact of sensitivity tests on modelled NO2 (µg/m3) concentrations in 2022 of the Alternative Package without Alfred Road signals – change in modelled NO2 (µg/m3) Error! Bookmark not defined. Appendix Figures Figure B- 1 Study Area showing PCM Links and 50m Buffer ............................................................... 59 Figure B- 2 Location of Street Canyons ................................................................................................ 60 Figure B- 3 Location of Flyovers and Bridges ....................................................................................... 61 Figure B- 4 Exceedances of Annual Mean NO2 Limit Value, 2018 ....................................................... 62 Figure B- 5 Monitoring Locations .......................................................................................................... 62 Figure B- 6 Wind Rose, Thorney Island (2018 data) ............................................................................ 63 Figure B- 7 Receptor Locations ............................................................................................................ 64 Figure B- 8 Source Apportionment- Diesel Cars 2018 (L) and 2022 (R) .............................................. 65 Figure B- 9 Source Apportionment- Petrol Cars 2018 (L) and 2022 (R) ............................................... 66 Figure B- 10 Source Apportionment- Full Hybrid Diesel Cars 2018 (L) and 2022 (R) .......................... 67 Figure B- 11 Source Apportionment- Full Hybrid Petrol Cars 2018 (L) and 2022 (R) ........................... 68 Figure B- 12 Source Apportionment- Artic HGVs 2018 (L) and 2022 (R) ............................................. 69 Figure B- 13 Source Apportionment- Rigid HGVs 2018 (L) and 2022 (R) ............................................ 70 Figure B- 14 Source Apportionment- Diesel LGVs 2018 (L) and 2022 (R) ........................................... 71 Figure B- 15 Source Apportionment- Petrol LGVs 2018 (L) and 2022 (R) ........................................... 72 Figure B- 16 Source Apportionment- Buses and Coaches 2018 (L) and 2022 (R) .............................. 73 Figure B- 17 Source Apportionment- Taxis 2018 (L) and 2022 (R) ....................................................... 74 Figure B- 18 Source Apportionment- Motorcycles 2018 (L) and 2022 (R) ........................................... 75 5
JAQU Air Quality Modelling Report Appendix Tables Table A- 1 Monitoring Sites Used in Verification ................................................................................... 57 Table B- 1 Details of automatic monitoring locations ............................................................................ 76 Table B- 2 Details of diffusion tube locations ........................................................................................ 77 6
JAQU Air Quality Modelling Report 1. Introduction This report constitutes the ‘Local Plan Air Quality Modelling Report (AQ3)’ and supplements the information provided in the Local Plan Air Quality Modelling Tracking Table (AQ1) and Local Plan Air Quality Modelling Methodology Report (AQ2). The documents remain ‘live’ documents and are updated as required during the course of the various stages of the study. On 26 July 2017, the government published the UK plan for tackling roadside nitrogen dioxide (NO 2) concentrations (‘the UK Plan’) to bring NO2 concentrations within the European Union (EU)’s statutory annual limit value of 40 micrograms per cubic metre (µg/m 3) in the shortest possible time, focussing on five key urban areas. The Department for Environment, Food and Rural Affairs and the Department for Transport’s Joint Air Quality Unit (JAQU) is responsible for overseeing the delivery of the UK Plan, which includes supporting local authorities and other organisations on the delivery of local measures in their area. On 23 March 2018 the government directed 33 additional English local authorities with projected annual mean NO2 exceedances in the short to medium term to undertake feasibility studies to establish whether there are measures they can take to reduce NO2 air pollution in their areas in the shortest possible time. On 5 October 2018 the government published a supplement to the UK Plan which highlighted that eight of the directed 33 local authorities had identified more persistent, longer term exceedances than were initially forecast by the Pollution Climate Mapping (PCM) model. Under the terms of the Environment Act 1995, the government has issued a Ministerial Direction to this group of local authorities to develop a Local Plan to identify measures that could bring forward compliance dates within the shortest possible time. Portsmouth City Council (PCC) is one of these local authorities. PCC has previously worked with AECOM to undertake an initial targeted feasibility study, submitted to JAQU in September 2018. This study included local modelling for the two non-compliant links and for AQMA 6 to identify the main causes and extent of exceedances, as well as to determine the level of emissions reductions required to achieve compliance, and potential measures that could bring this forward. Following this work, AECOM has provided further local air quality modelling support for the Local Plan study. This work has inputted into target determination to understand the extent of exceedances across the wider study area and further modelling of a range of measures against a Clean Air Zone (CAZ) baseline intervention has been conducted. This has identified a preferred package of measures that will bring forward compliance in the shortest time possible which feeds into the business case for the Local Plan. PCC submitted an Outline Business Case (OBC) to JAQU on 31 October 2019, and undertook public consultation in Summer 2020. This version of the document is part of the Full Business Case (FBC), submitted December 2020. 7
JAQU Air Quality Modelling Report
Initial Evidence Submission
8
JAQU Air Quality Modelling Report
2. Atmospheric Dispersion Modelling Approach
2.1 Dispersion Model Selection and Tools
The dispersion model selected for this study is CERC’s ADMS-Roads v4.1.1, which has been used to
simulate the dispersion of vehicle emissions of NOx from road links included in the model domain.
Annual mean NO2 concentrations were subsequently derived at identified receptor locations through
utilising the outputs of the model (road-NOx) in combination with the following tools published by
Defra1:
Emissions Factors Toolkit (EFT) v9.1.b;
2017 reference year Background Pollutant Maps (for NO x and NO2);
NO2 Adjustment for NOx Sector Removal Tool v7.0; and,
NOx to NO2 Calculator 2017 to 2030 v7.1.
Verification of the air quality model was completed with reference to Defra’s LAQM.TG16 document,
specifically:
Section 4: Dispersion Modelling of Emissions
─ Box 7.14: Initial Comparison of Modelled and Monitored Total NO2 Concentrations
─ Box 7.15: Comparison of Road-NOx Contributions Followed by Adjustment
─ Box 7.16: Importance of an Approach to Verifying Modelled NO 2 Concentrations from Road
Traffic
─ Box 7.17: Methods and Formulae for Description of Model Uncertainty
2.2 Assessment Scenarios
The assessment scenarios focused on for the Target Determination submission were:
Base Year 2018 (scenario to be used for air quality model verification);
Projected Base Year 2022 (future baseline ‘without measures’); and,
Interim Base Years 2019, 2020 and 2021 (using interpolation methodology).
The air quality modelling process followed a number of sequential steps to convert the vehicle
emissions from traffic on the modelled road network into annual mean concentrations of NO 2. An
overview of each step of the process is provided below.
2.3 Model Input Data
2.3.1 Traffic Data
Traffic data for the 2018 and 2022 baseline scenarios were obtained from the SATURN-based
Southern Regional Transport Model (SRTM) run by Systra. This provided period average flows along
with speeds for the AM peak, inter-peak, PM peak, and off-peak periods with associated link distances
for all PCM and non-PCM links included within the air quality model domain, as depicted in Figure B-1
in Appendix B. These were converted to 24 hourly data for the purpose of the air quality modelling.
The modelled road links were georeferenced prior to input to ADMS-Roads, with each road link
spatially matched to the Intelligent Transport Network (ITN) centre lines, thus ensuring a real-world
representation. Each road link was matched to the respective NOx link-specific emission rate derived
from the EFT using a common reference link.
2.3.2 Baseline Emissions Inventory Development (NOx and f-NO2)
An Automatic Number Plate Recognition (ANPR) camera survey was completed during a week’s
study in March 2019 at a large number of locations across the air quality model domain area (see
Figure 2-1).
1
LAQM tools published by Defra / JAQU specifically in relation to CAZ studies
9
JAQU Air Quality Modelling Report Figure 2-1 ANPR camera locations The raw datasets were sent to the Department for Transport (DfT) via JAQU to match the number plates against the DVLA database. The returned data of over 8 million vehicle captures were collated and aggregated to generate a locally representative and domain-wide vehicle fleet composition. This enabled total vehicle journeys on all modelled links to be proportioned according to characteristics such as: Vehicle size and class distributions; Fuel splits (e.g. petrol, diesel, Liquid Petroleum Gas -LPG, hybrid, electric); Estimated Euro emission standard based on year of manufacture; Rigid and articulated Heavy Goods Vehicle (HGV) split; and, Bus and coach split. Given the availability of ANPR data, the detailed input option within the EFT was utilised (Alternative Technology’) in combination with the use of a bespoke ‘simple User Euro’ work tab for the 2018 scenario. Therefore, the local fleet Euro composition relevant to the model domain was represented within the emissions inventory calculations and outputs. For the Projected Base Year (2022), Alternative Technology within the EFT was utilised along with the ‘Fleet Projection’ tool tab. ‘Option 1’ of the projection tool was utilised within EFT, which assumed the future year 2022 Euro fleet composition has the same difference in Euro classes as observed between the default base year profile and the ANPR data. Version 9.1b of the EFT incorporates an updated Petrol/Diesel Projection Tool for forecasting the fuel split of cars – as determined from the ANPR data – to future assessment years. The tool was used predict the relative proportions of conventional petrol and diesel cars, hybrid cars and electric cars in 2022. Figure 2-2 shows the output of the Petrol/Diesel Projection Tool, comparing the default car fleet fuel splits for 2019 and 2022 (from the EFT) with the ANPR observed fleet split for 2019 and projected fleet split for 2022. 10
JAQU Air Quality Modelling Report Figure 2-2 Comparison of Portsmouth Vehicle Fleet with National Fleet – Car Fuel Split A comparison of the Euro standards of main vehicle types identified in Portsmouth from the 2019 ANPR with the national fleet for the same year given in the EFT is shown in Figure 2-3. Figure 2-3 Comparison of Portsmouth Vehicle Fleet with National Fleet – Euro Proportions The traffic data provided numbers of cars, LGVs, HGVs and buses for each road link and were further disaggregated into compliant and non-compliant vehicles. The modelled traffic data did not distinguish taxis (including London-style black cabs and private hire vehicles) from cars. However, CAZ B and CAZ C configurations both cover taxis whilst excluding cars. Consequently, the proportion of taxis relative to cars was extracted from the ANPR survey data and used to calculate the numbers of taxis in the modelled traffic datasets. For each modelled scenario, emissions were therefore calculated for compliant and non-compliant vehicles in separate EFT spreadsheets, as follows: Compliant cars (excluding taxis), LGVs, HGVs and buses; Non-compliant cars (excluding taxis), LGVs, HGVs and buses; Compliant taxis; and Non-compliant taxis. Within the compliant EFT spreadsheets, the projected fleet composition as entered on the “Simple User Euro” worksheet was renormalised based on fuel type into the compliant Euro standard fields i.e. petrol-fuelled vehicles were renormalised across Euro 4/IV and newer, diesel vehicles of Euro 6/VI and newer. For the non-compliant EFT spreadsheets, the renormalisation was done across the non- compliant Euro standards (Euro 3/III and older for petrol; Euro 5/V and older for diesel). The vehicle fleet assumed in the non-compliant and compliant vehicles was different. This is illustrated in Table 2-1 for the 2022 Projected Base Year for cars. 11
JAQU Air Quality Modelling Report
Table 2-1: Split of cars by compliant and non-compliant
Category % split by car body and fuel type
Petrol car Diesel car Hackney carriage Alternative fuelled cars
Compliant cars (ex taxis) 74% 19% 0% 7%
Non-compliant cars (ex 26% 71% 0% 3%
taxis)
Compliant taxis 7% 82% 0% 11%
Non-compliant taxis 0% 96% 1% 3%
The outputs from the EFT for each assessment scenario included the following:
Link-specific NOx emissions rates for air quality model input (g/km/s);
Annual NOx emissions total for each link within the modelled road network (kg/annum);
Primary NO2 (f-NO2) emissions fraction for each link and, for the links included in the model, an
average-domain wide f- NO2 fraction; and,
Annual NOx emissions split by vehicle type for source apportionment.
The link-specific emission rates output from each of the four EFT spreadsheets were added together
to form the emissions dataset for input into ADMS-Roads. Annual total pollutant emissions for each
link were aggregated in the same way. For f-NO2, the EFT outputs were combined and emissions-
weighted average f-NO2 values were calculated.
2.3.3 Road Width Data
Road width data for each modelled link were derived based on an automated GIS approach, which
utilised the georeferenced road centreline to link the respective mapped road polygon that each
centreline was within. This identified the road boundaries. Subsequently, GIS was used to draw lines
to the centreline that extended to the road edge. For each link, an average width was calculated
based on lengths of each of these lines.
2.3.4 Road Source Emission Rates (NOX and f-NO2)
The geometry of each road link from the PCM network were entered into ADMS-Roads, including
road width. The link specific NOx emission rates were input into the ADMS-Roads model for all road
links included in the air quality domain.
The link specific f-NO2 outputs from the EFT were reviewed and used within air quality modelling. For
2018, these f-NO2 values ranged from 0.098 to 0.329. For 2022, the f- NO2 values ranged from 0.063
to 0.299.
2.4 Gradients, Tunnels, Flyovers and Street Canyon Effects
2.4.1 Road Gradient Effects
The effects of road gradients on vehicle emissions, particularly heavy duty vehicles (HDVs), should be
represented in the air quality modelling appropriately. OS DTM data was used to calculate road
gradients for all modelled road links within the study area. Gradient effects were calculated and
applied to all road links where the gradient exceeds 2.5%, in accordance with Defra’s LAQM.TG16
methodology and associated information provided by JAQU.
PCC were consulted on the locations of the identified gradients, and it was concluded that gradient
effects only needed to be considered for Portsdown Hill Road, north of the A27/M27. After further
analysis this road was considered to be outside of the study area.
12JAQU Air Quality Modelling Report 2.4.2 Street Canyons With respect to street canyon effects, the road network and detailed OS mapping with address base and building layer data were used to facilitate use of the ‘Advanced Street Canyon’ module within ADMS-Roads. Figure B-2 in Appendix B shows an indication of those streets initially identified as being street canyons, based on TG16 (paragraph 7.408), which states that: “..Although street canyons can generally be defined as narrow streets where the height of buildings on both sides of the road is greater than the road width, there are numerous example whereby broader streets may also be considered as street canyons where buildings result in reduced dispersion and elevated concentrations (which may be demonstrated by monitoring data).” Street canyons were identified by measuring the road width (building façade to building façade) and the heights of the building. Narrow streets where the height of buildings on both sides of the road was greater than the road width were classed as a street canyon. Google Streetview and Google Earth were used to measure the road widths and building heights. A consideration of street canyons close to monitoring sites was made to determine whether it would be appropriate to apply a separate verification factor to those roads with street canyons. However, as the model verification factor is already low (1.61), and there are only a few street canyons in the modelled road network, it was considered that the model was performing well so no further consideration or modification to the verification was needed. 2.4.3 Flyovers and Tunnels The locations of bridges and flyovers have been reviewed to identify the relevance of these with respect to air quality modelling for the PCM links being investigated. Flyovers are represented within ADMS-Roads by assigning road elevations to the respective links using elevations from OS digital terrain model (DTM) data. The road elevations in metres are extracted using GIS and are cross- checked against Google Earth elevation. The source (elevated road) height is determined in relation to the receptor height. The elevation of source is calculated using the following formula: Elevation of source= Measured source height - Measured receptor height For the Portsmouth local model, the elevated sections of road were modelled at a height of 5 m above the receptor at that road. The identified flyovers are presented in Figure B-3 in Appendix B. No tunnels were identified within the modelled road domain. 2.5 Surface Roughness and Minimum Monin-Obukhov Length Given that most of the study domain encompasses a suburban area, a single surface roughness length of 0.5 m across the modelled area was assigned. Similarly, a minimum Monin-Obukhov length of 30 m was assigned within ADMS-Roads to provide a measure of atmospheric stability, which is considered representative of the landscape. 2.6 Meteorological Data Hourly sequential meteorological data were obtained from Thorney Island meteorological station (Lat. 50.817; Lon. -0.917; elevation: 3m), which is approximately 15 km east of the study area. A wind rose based on 2018 is shown in Figure B-6 in Appendix B. The dominant wind direction in this year was from the southwest (270 degrees). The data were obtained for the same year as the Base Year model scenario (2018) to maintain consistency. These data were used in all air quality modelling scenarios. The following parameters were included in the meteorological data file: Temperature; Wind speed; Wind direction; Relative humidity; Cloud cover extent; and 13
JAQU Air Quality Modelling Report Precipitation. 2.7 Modelled Receptor Selection Each PCM link has a unique Census ID and a grid reference typically describing the DfT traffic count points on each link. This location may not be where the highest roadside concentrations are occurring along the entire link length when using a more detailed local modelling method, with more detailed traffic data. For the purposes of Target Determination and with a focus on the primary objective, a suite of discrete receptor points was identified adjacent to each PCM link and local road link from the SRTM within the air quality model domain. To comply with the PCM model and to facilitate direct comparison for Target Determination, each receptor was modelled at 4 m from the kerb at a height of 2 m above ground level on either side of the road link. The receptors adhered to the criteria referenced by Annex III of EU Directive 2008/50/EC, which state that the receptor should be: Representative of at least 100 m of road length; At least 25 m from the edge of a major junction (one that interrupts flow of traffic); and, Within 10 m of the kerbside. The locations of the discrete receptors included in the air quality model are presented in Figure B-7 in Appendix B. 2.8 Model Output Data 2.8.1 Base year 2018 and Projected Base Year 2022 The ADMS-Roads model provides annual mean NOx concentration values at each identified receptor point. Defra’s NOx to NO2 calculator v7.1 was used to convert annual mean road-NOx to total annual mean NO2 at each point. This calculation required the background annual mean NOx and NO 2 value to be known, which were obtained from Defra’s national 1 km x 1 km grid pollutant maps for the respective years (2018 and 2022). These background values incorporated contributions from non-road sources of NOx and NO2. Given the extent of the study area, the background pollutant values vary across the model domain and thus were mapped using GIS and the relevant value assigned to the modelled receptors. Background NOx and NO2 were adjusted to remove contributions from roads included in the ADMS- Roads model (e.g. Trunk roads, A-roads), where applicable, thereby avoiding double-counting of emissions. The calculator also incorporates the domain-wide average f-NO2 fraction, which was derived from the EFT outputs and applied to each receptor point to determine the proportion of the road-NOx concentration as primary NO2. 2.8.2 Interim and Future Base Year Interpolation The ADMS-Roads model was used to predict NO2 concentrations at sensitive receptor locations for the Base Year (2018) and Projected Base Year (2022). The modelled road networks for the Base Year and Projected Base Year were the same. To interpolate concentrations to interim years, the approach for estimating roadside NO2 concentrations as described on the LAQM support website was initially used. However, these yearly factors were found to result in a greater reduction in concentrations as predicted by the local modelling. Therefore for this study, a set of yearly factors specific to each road link was calculated based on a linear change in concentration from 2018 to 2022. These factors were extrapolated to beyond 2022 to identify the year of compliance without any intervention up to 2030. 14
JAQU Air Quality Modelling Report
3. Model Verification and Adjustment
This section provides an overview of the dispersion model verification process and outcomes for the
2018 baseline year. A fuller description of the process is given in Appendix A.
3.1 Comparison of Modelled (Unadjusted) and Monitored Road
NOx
A comparison of the unadjusted modelled annual mean road NOx and total NO2 concentrations at all
of the Council’s monitoring locations was undertaken for 2018. These were reviewed and some
discounted due to low data capture and specific siting issues. A total of 34 monitoring sites from
across the air quality domain were included in the initial comparison, comprising of two real-time
continuous analysers and 32 passive diffusion tubes. Information on monitoring locations are given in
Appendix B
There was an overall tendency for the model to underestimate the monitored road-NOx and total NO2
equivalent. The model is shown to under predict at a large number of sites, with 26 out of the 34 sites
under predicting concentrations.
3.2 Verification
Following review of the model performance at monitoring sites and liaison with PCC’s monitoring
team and JAQU, the model was adjusted by a single verification across the study area.
The modelled road-NOx adjustment factor was applied to the modelled road-NOx values for the Base
Year 2018 and Projected Base Year 2022 as well as all modelled options at all receptors.
3.3 Modelled Road NOx Adjustment
The modelled road-NOx values were plotted graphically versus the monitored road-NOx equivalent
for each site within the respective zone. A road-NOx adjustment factor was derived for each zone
based on a ‘y=mx’ line of best fit, forced through a zero intercept. This graph is presented in Figure
3-1 which show the modelled road-NOx value versus the monitored road-NOx value before and after
the adjustment. The adjustment factor based on the line of best fit were derived to be 1.61. Once the
derived factor was applied to the modelled road-NOx value, the NOx to NO2 calculator was utilised to
calculate the total adjusted annual mean NO2 at each site. A secondary adjustment factor was not
applied. The monitored NO2 concentrations versus modelled total NO2 concentrations are presented
in Figure 3-2.
15JAQU Air Quality Modelling Report Figure 3-1 - Modelled Road-NOx versus Monitored Road-NOx Figure 3-2 - Adjusted NO2 versus Monitored NO2 Concentrations 16
JAQU Air Quality Modelling Report 3.4 Model Adjustment Summary Following model adjustment, there was no apparent tendency for the dispersion model to over or under predict within each of the verification zones. Of the 34 monitoring sites considered, 32 were shown to return adjusted modelled total NO2 concentrations within +/-25% of the monitored equivalent, with 25 performing within +/-10%. A statistical analysis was completed for the adjusted model road-NOx to facilitate comparison with the unadjusted model road-NOx. The RMSE (average model uncertainty) value was within 10% of the air quality limit value (3.4 µg/m3). The statistical analysis of the adjusted model performance and uncertainty demonstrates that the atmospheric dispersion model is robust and representative for the prediction of annual mean road-NOx concentrations at identified receptor locations throughout the domain. The use of a single verification factor across the large study area was requested by JAQU and PCC as it was considered that there were not sufficient differences in the traffic network to warrant zoning of the model and the use of multiple adjustment factors. Although the model performs well across the study area, there are some monitoring locations where the outputs under or over-predict road NOx concentrations to a greater extent than others. For example, the model over-predicts at Church Street monitoring sites (DT32a, 32b and DT34) by 30-40%, but under-predicts on London Road (e.g. by more than 40% at monitoring site DT26 and C2). It is important to be mindful of this when considering the results. A consideration of street canyons close to monitoring sites was also made to determine whether it would be appropriate to apply a separate verification factor to those roads with street canyons. However, as the model verification factor is already low (1.61), and there are only a few street canyons in the modelled road network, it was considered that the model was performing well so no further consideration or modification to the verification was needed. 17
JAQU Air Quality Modelling Report
4. Baseline Results
4.1 Vehicle Fleet
The data obtained from the ANPR camera survey was used to identify the Euro emission standard of
the current vehicle fleet captured in 2019. Table 4-1 provides a summary of the number of vehicles
captured in the week’s survey and the proportion that are currently non-compliant to a CAZ emission
standard (Euro 4 petrol or Euro 6/VI diesel).
Of the 8 million vehicle movements captured, some 45% relate to non-compliant vehicles and 42% of
fleet movements are undertaken by non-compliant diesel cars, 9% by non-compliant petrol cars, 9%
by non-compliant diesel LGVs, and 2% by non-compliant taxis.
Table 4-1 Current and predicted vehicle fleet and non-compliance to CAZ emission standards,
across all ANPR sites
Vehicle type Non- Compliant Total vehicle % non- What % of the Predicted %
compliant vehicle movements compliant total fleet do non-
vehicle movements (2019) vehicle non-compliant compliant
movements (2019) movements vehicles vehicle
(2019) (2019) account for movements
(2019)? (2022 future
base)
Diesel cars 1,896,439 816,376 2,712,815 70% 23.5% 47%
Petrol cars 715,954 2,838,207 3,554,161 20% 8.9% 6%
Diesel black cabs 1,337 28 1,365 98% 0.0% 51%
Diesel taxi cars 170,113 200,417 370,530 46% 2.1% 32%
Petrol taxi cars 0 19746 19,746 0% 0.0% 0%
Other taxi cars 174 16566 16,740 1% 0.0% 0%
Electric cars 0 10011 10,011 0% 0.0% 0%
Hybrid cars 1546 102172 103,718 1% 0.0% 0%
Gas cars 2,625 0 2,625 100% 0.0% 100%
Diesel LGVs 730,820 282,869 1,013,689 72% 9.0% 45%
Petrol LGVs 3,155 3,922 7,077 45% 0.0% 6%
Other LGVs 679 2241 2,920 23% 0.0% 0%
Rigid HGVs 40,218 52,313 92,531 43% 0.5% 21%
Artic HGVs 13,633 32,543 46,176 30% 0.2% 10%
Mini buses 15,822 11,317 27,139 58% 0.2% n/a*
Diesel 62,220 42,479 104,699 59% 0.8% 11%
buses/coaches
Total 3,654,735 4,431,207 8,085,942 45% - -
A summary of the ANPR data for the camera sites closest to the two exceedance locations is provided
in Table 4-2.
18JAQU Air Quality Modelling Report
Table 4-2 Current and predicted vehicle fleet and non-compliance to CAZ emission standards,
based on ANPR Camera 35 (Commercial Road) and ANPR Camera 32 (Marketway, close to
Alfred Road)
Camera 35 (Commercial Road) Camera 32a,b (Marketway)
Vehicle type Total vehicle % non- What % of the Total vehicle % non- What % of
movements compliant total fleet do movements compliant the total fleet
(2019) vehicle non-compliant (2019) vehicle do non-
movements vehicles movements compliant
(2019) account for (2019) vehicles
(2019)? account for
(2019)?
Diesel cars 59,494 76% 28.3% 108,772 67% 23.9%
Petrol cars 73,950 20% 9.4% 131,953 17% 7.6%
Diesel black cabs 48 96% 0.0% 93 97% 0.0%
Diesel taxi cars 5,387 63% 2.1% 15,419 47% 2.4%
Petrol taxi cars 126 0% 0.0% 775 0% 0.0%
Other taxi cars 134 1% 0.0% 641 0% 0.0%
Electric cars 254 0% 0.0% 601 0% 0.0%
Hybrid cars 1,882 1% 0.0% 4,624 4% 0.1%
Gas cars 33 100% 0.0% 65 100% 0.0%
Diesel LGVs 14,113 80% 7.1% 27,836 70% 6.4%
Petrol LGVs 126 41% 0.0% 189 42% 0.0%
Other LGVs 63 14% 0.0% 95 25% 0.0%
Rigid HGVs 1,084 67% 0.5% 2,796 46% 0.4%
Artic HGVs 290 51% 0.1% 1,020 26% 0.1%
Mini buses 548 72% 0.2% 1,155 55% 0.2%
Diesel
buses/coaches 2,053 84% 1.1% 7,840 56% 1.4%
Total 159,585 48% - 303,874 41% -
4.2 NO2 concentrations
Total annual mean NO2 concentrations were derived for all receptor locations identified in Figure B-7
in Appendix B for the 2018 Base Year and 2022 Projected Base Year. There are 41 Census IDs
present in the modelled road domain. Across the wider model domain, there were also a large number
of local roads without an associated Census ID.
Based on the local model results, there are predicted to be a total of 70 individual receptors
demonstrating exceedances of the annual mean EU limit value in the Base Year 2018, reducing to 11
receptors in the Projected Base Year 2022 scenario (see Figure 4-1 a).
The 2022 future baseline shows that there are three road links within the city centre where the NO 2
EU Limit Value is predicted to be exceeded on Portsmouth controlled roads as shown in Figure 4-1 b.
This is indicated in Table 4-3 alongside other areas with concentrations close to the EU Limit Value.
19JAQU Air Quality Modelling Report Figure 4-1: Location of roadside receptor sites with modelled exceedances in 2022 baseline (EFT v1.9b) a) All receptors 20
JAQU Air Quality Modelling Report
b) City centre roads
Table 4-3: Locations with modelled (or near) exceedances in 2022 Baseline
Receptor ID Unique Link Road Name Modelled Modelled % Road NOx Year
ID (Census NO2 (µg/m3) Road-NOx reduction to compliance
ID if – (µg/m3) – meet EU limit would be
applicable) 2022 2022 achieved,
baseline baseline assuming no
intervention
Road sections on the local network modelled as exceeding the EU limit (40 µg/m3) in 2022
573 51842 A3 Alfred Road (Unicorn Rd 41.7 47.3 -6.7% 2023
(18114) to Queen St, s/b)
546 51448 A3 Commercial Road (south 41.1 39.6 -3.8% 2023
(80848) of Church St Rbt, s/b)
Road sections on the local network not exceeding the EU limit, but still above 37 µg/m 3 in 2022
526 51411 Church Street (east of Church 40.4 37.6 (+0.6%) -
St Rbt, n/b)
526 51411 Church Street (sensitivity test) 38.7 33.4 (+1.0%) -
– described below
536 51546 A3 Hope Street (south of 38.9 34.9 (+11.0%) -
(74735) Church St R'bout, s/b)
824 51828 (8250) A2030 Eastern Road Water 38.8 43.9 (+9.5%) -
Bridge (s/b)
21JAQU Air Quality Modelling Report
Receptor ID Unique Link Road Name Modelled Modelled % Road NOx Year
ID (Census NO2 (µg/m3) Road-NOx reduction to compliance
ID if – (µg/m3) – meet EU limit would be
applicable) 2022 2022 achieved,
baseline baseline assuming no
intervention
648 51601 A2047 London Road 38.5 33.1 (+14.3%) -
(38333) (Stubbington Ave to Kingston
Crescent, s/b)
520 51399 Mile End Road (north of 37.6 30.9 (+22.2%) -
(48196) Church St R'bout, s/b)
557 51461 A3 Marketway (Hope St Rbt to 37.4 38.5 (+19.8%)
(18114) Unicorn Rd)
Road sections on the Strategic Road Network exceeding the EU limit (40 µg/m 3) in 2022
986 52157 A27 (north of Portsea Island, 48.5 68.6 -29.5% 2026
w/b)
1089 52408 A27 (east of Portsea Island, 46.1 65.3 -21.3% 2025
w/b)
11 51817 M27 (west of Portsea Island, 45.3 68.0 -17.9% 2025
w/b)
968 53122 A27 (north of Portsea Island, 43.7 59.9 -14.7% 2024
e/b)
834 51837 A27 (east of Portsea Island, 41.1 49.0 -3.0% 2023
w/b)
For target determination, a sub-set of receptors was chosen through a process of joining road links
and receptors in GIS to identify those with the maximum predicted annual mean NO2 concentrations
on each modelled road link.
Church Street sensitivity test
It was apparent from the model results, that NO2 concentrations on Church Street (receptor 526) were
higher than expected from the Council’s monitoring. Therefore, following a comparison between the
strategic transport model outputs and observed traffic counts in the city centre, it is apparent that the
SRTM2 traffic model substantially over-estimates flows on Church Street, primarily as a result of
the modelled link capturing movements on other local roads which are not represented in the strategic
model network. Table 4-4 summarises the comparison of modelled traffic against observed data. The
comparison draws on the two available sources of observed data:
the vehicles counted by the ANPR camera installed on the northern part of Church Street for the
week of 18/03/19 to 24/03/19 providing two-way all day coverage; and
a classified count on a single day (Thursday 04/04/19) for the AM and PM peak periods at the
junction between Church Street, Holbrook Road and Lake Road to the south of the link.
The Council undertook a more comprehensive two week traffic count in September 2019 which
provides additional supporting evidence.
2
Sub Regional Transport Model
22JAQU Air Quality Modelling Report
Table 4-4: Comparison between modelled and observed traffic data, Church Street, 2019
Section - Modelled Baseline 2019 (SRTM) ANPR 2019* One day Classified
Direction Turning Count 2019 **
AM IP PM 24hr AM IP PM 24hr AM peak PM peak
peak peak peak AADT peak peak peak AADT hour hour
hour hour hour hour hour hour
North – NB 1,044 764 787 10533 - - - - - -
North – SB 983 758 1,169 11,506 - - - - - -
North – 2way 2,028 1,523 1,957 22,037 1,057 880 768 14,225 - -
South – NB 1,303 768 1,260 12,735 - - - - 517 440
South – SB 765 650 756 8,884 - - - - 584 334
South – 2way 2,067 1,352 2,017 21,620 - - - - 1,101 774
North refers to the short section between Church Street Roundabout and Wingfield Street and most
closely represents the conditions at receptor ID526. South refers to the section between Wingfield
Street and Lake Road Roundabout, a 350 metre section south of ID526.
* ANPR data is a 7 day average, adjusted to account for average 93% capture rate over the week.
** Turning count undertaken on Thursday 04/04/19
The data shows that two-way modelled flows on the short section between Church Street Roundabout
and Wingfield Street, which most closely represents the conditions at receptor ID526, are 22,037
compared with an ANPR count of 14,225.
A comparison of modelled speeds against available data from Trafficmaster, TomTom and Google
mapping showed that modelled speeds appear to be close to observed speeds, despite the difference
in flow levels. For example, TomTom GPS journey speed data for 2018 (24hr flow weighted average)
provides the following comparison for the section between Church Street Roundabout and Wingfield
Street:
southbound: median speed = 29kph and mean speed = 29kph, compared to a modelled speed of
31kph;
northbound: median speed = 13kph and mean speed = 15kph, compared to a modelled speed of
9kph.
As a result of the overestimate of traffic flows, the air quality model over-estimates NO2
concentrations compared to the measured data on Church Street. A sensitivity modelling test was
conducted whereby the observed traffic flows were growthed to 2022 levels (using the forecast growth
from SRTM and a further 15% uplift to allow for uncertainty) to provide a more realistic, lower future
flow estimate for Church Street. Using these revised traffic flows, the predicted modelled NO2
concentration at receptor 526 is forecast to be lower in 2022 (38.7 µg/m3) compared to the predictions
from the SRTM baseline forecast traffic outputs (40.4 µg/m3) as presented in Table 4-4 above.
Based on the evidence from the traffic count data presented in Table 4-4, the sensitivity test is judged
to be a more accurate representation of concentrations on Church Street, and from this point on we
will assume the revised baseline figure of 38.7 µg/m3 for Church Street.
23JAQU Air Quality Modelling Report
5. Source Apportionment
NO2 concentrations are affected by NOx emissions from both non-road and road sources within and
outside Portsmouth. Further information on the relative contribution from these sources at selected
locations with the city is given in this section.
5.1 Road vs Non-Road Contribution
The contribution of all non-road sources, both those within and outside the city have been included
within the model as part of the background. This is represented by mapping data within 1km grid
squares (see Section 2.8.1). Between 2018 and 2022, there is a predicted reduction in background
NOx concentrations in the city, but the relative contribution of each source type is similar.
In some areas of the city, the contribution of road sources makes up 40% of total modelled NOx, with
non-road sources, such as combustion process, domestic and commercial heating, railway, off-road
vehicles and shipping from the port making up a similar amount. JAQU have provided further
disaggregation of the background mapping for 2018. This additional information shows that there are
areas of the city, such as close to the coast where the contribution from roads is lower and shipping
emissions at the port are much greater. For example, at Mile end Road/Church Street/Commercial
Road, shipping emissions make up more than 40% of background road NO x in 2018. Close to Alfred
Road, the combined contribution from shipping (34% and off-road industrial activities – i.e. portside
operation (11%) is of a similar magnitude (see Table 5-1). There is also around 20-25% of NOx from
rural sources from outside the city, which are outside of the control of the Council.
Table 5-1 and Table 5-2 provide selected examples of the range of contribution of different sources to
NOx within selected 1km grid squares for the Base Year of 2018 and Projected Base Year of 2022
respectively. The tables show that the background concentration declines from 2018 to 2022, but the
relative contribution by source is similar in both years.
Table 5-1 Percentage Contribution of Road and Non-Road Sources to background NOx in
selected areas of Portsmouth, 2018
Off-road industrial
Domestic heating
Total Non-Road
Off-road other
Point sources
(outside PCC)
Road Sources
Shipping
Industry
Railway
Area
Sources
Rural
(Background grid Total bkd
Total
square and NOx
receptors) (µg/m3)
M275/A3 Mile End 46.8 1.9% 6.3% 0.1% 43.8% 2.9% 0.1% 1.9% 18.4% 75.3% 24.7%
Rd/ Church
St/Commercial Road
(incl. Portsmouth
Port)
Grid square:
464500, 101500
Receptors: 526, 546
Road link 18114
A3 Alfred 38.3 1.9% 5.7% 0.1% 34.3% 10.9% 0.1% 5.7% 22.6% 81.4% 18.6%
Rd/Marketway
(incl. the Naval
Dockyard)
Grid square:
463500. 100500
Receptor: 573
Road link 80848
Portsea Island 28.7 3.0% 7.1% 0.2% 21.7% 4.4% 0.2% 3.5% 30.1% 70.2 29.8
(average of grid
squares)
AECOM
24JAQU Air Quality Modelling Report
Table 5-2: Percentage Contribution of Road and Non-Road Sources to NOx in selected areas of
Portsmouth, 2022
Off-road industrial
Domestic heating
Total Non-Road
Off-road other
Point sources
(outside PCC)
Road Sources
Shipping
Industry
Railway
Area
Sources
(Background grid Total bkd
Rural
Total
square and NOx
receptors) (µg/m3)
M275/A3 Mile End 40.4 1.9% 6.8% 0.0% 45.2% 3.1% 0.1% 2.2% 18.6% 77.9% 22.1%
Rd/ Church
St/Commercial Road
(incl. Portsmouth
Port)
Grid square:
464500, 101500
Receptors: 526, 546
Road link 18114
A3 Alfred 33.0 2.2% 6.2% 0.1% 34.0% 11.6% 0.1% 6.4% 22.7% 83.3% 16.7%
Rd/Marketway
(incl. the Naval
Dockyard)
Grid square:
463500. 100500
Receptor: 573
Road link 80848
Portsea Island 24.8 3.4% 7.9% 0.1% 22.7% 4.7% 0.2% 4.1% 30.3% 73.4% 26.6%
(average of grid
squares)
Local estimate of port emissions
At the time of submission, reliable data on emissions associated with Portsmouth International Port
activity was not available. Prior to the pandemic, there were plans to expand shipping activity, but the
timescales and extent of any future growth plans are currently unconfirmed. The modelling presented
in this document is therefore based on the above JAQU estimates of background NOx concentrations.
5.2 Road Contributions by Vehicle Type
Road transport sources are the only source to be explicitly modelled in this study, as there is currently
not sufficient local data available to model the other non-road sources.
The contribution of road NOx emissions broken down by vehicle and fuel type is presented in Figures
B-8 to B-19 in Appendix B for each modelled road link in for Base 2018 situation and Future Base
2022. The figures show that it is the diesel cars that have the greatest contribution to road NO x, with
more than 50% on some roads, particularly routes down the western corridor into the city. In some
areas of the city, there is a much higher contribution from HGVs such as around Anchorage Park (at
least 50%) and from buses (which contribute to 18% on London Road).
The contribution by the main vehicle types to road NOx emissions at each of the receptors predicted
to exceed the EU Limit Value in 2022 is given Table 5-3.
Table 5-3 Contribution of vehicle type to road NOx emissions on exceedance road links, 2022
Future Base
PCM Petrol Diesel Taxis Petrol Diesel Rigid Artic Buses & M’cycle Hybri
Road Cars (%) Cars (%) (%) LGVs (%) LGVs (%) HGVs (%) HGVs Coaches (%) s (%) d (%)
Link (%)
18114 9.24 47.10 0.03 0.03 21.61 13.73 7.27 0.00 0.01 0.98
80848 10.64 49.80 0.03 0.03 22.63 7.24 3.64 4.91 0.02 1.07
AECOM
25PORTSMOUTH CITY COUNCIL
6. Options Modelling
6.1 Shortlisted Options at Strategic Outline Case (SOC) Stage
The study initially considered a Benchmark Charging Clean Air Zone (CAZ) option and three non-
charging air quality improvement package options as presented in the Strategic Outline Case (SOC)
submitted in January 2019.
6.2 Shortlisted Options at Outline Business Case (OBC) Stage
6.2.1 Options Shortlisted (at OBC stage)
Following further review of options and packages that took into account the more detailed evidence
and current understanding of exceedances across the city, these options were further refined as part
of the Outline Business Case (OBC) process. This process was based on the following activities:
A PCC workshop with officers to discuss further options
Input from the Air Quality Stakeholder Group and the Air Quality Project Board
Inputs from Members
Initial modelling of traffic and emissions impact, prior to detail transport and air quality modelling*
Further research and data collection relating to the baseline (including ANPR data) various
options.
This process has resulted in a shortlist of options for comparison (see Table 6-1).
Table 6-1: Options modelled for 2022 (at OBC stage)
Model Test Name Detail
0. 2022 Baseline 2022 Projected Base Year including committed developments.
1. Portsea Island CAZ B Targeting taxis and private hire vehicles (PHV), buses, coaches,
HGVs across Portsea Island.
2. Portsea Island CAZ C Targeting taxis and Private Hire Vehicles (PHVs), buses and
coaches, HGVs, LGVs on Portsea Island.
3. Small Area CAZ B Targeting taxis and private hire vehicles (PHV), buses, coaches,
HGVs within a smaller area of the city.
4. Small Area CAZ B with non-charging measures As Test 3 + parking measures + strategic cycling routes +
modification to the traffic signal timings at the Alfred Road / Queen
Street junction.
5. Portsea Island CAZ B external trips only As Test 1 but charge only applied to trips into / out of Portsea Island
(i.e. not including internal trips), as these trips make up the vast
majority of movements on the two exceedance links.
6. City Centre Transport Link Modification to the road layout in the city centre to support wider
ambitions for the City. No CAZ charges assumed.
The indicative boundary for the Portsea Island CAZ is shown in Figure 6-1. It is focused on the whole
of the Portsea Island area, excluding the M275 and the western arm of Rudmore Roundabout
(providing the option to exempt traffic to Portsmouth International Port).
The indicative boundary for the Small Area CAZ is shown in Figure 6-2. It includes key destinations
for targeted traffic on the two exceeding links including the City Centre, and Gunwharf Quay /
Wightlink Terminal, and is intended to minimise re-routing to avoid the CAZ. In particular, the
inclusion of Kingston Crescent and Fratton Road was intended to minimise re-routing along London
Road and Fratton Road which could result in new exceedances.
26You can also read
