INTEGRATED TRANSPORTATION AND LAND USE REGRESSION MODELLING FOR NO2 MITIGATION
Citation:
Ó Domhnaill, Aonghus, INTEGRATED TRANSPORTATION AND LAND USE REGRESSION MODELLING FOR NO2 MITIGATION, Trinity College Dublin.School of Engineering, 2022Download Item:
Abstract:
The transport industry has been identified as one of the main contributors to Nitrogen Dioxide (NO2) pollution in Ireland. Diesel fuelled vehicles emit significantly greater amounts of NO2 in comparison to any other fuel type. In 2008, car taxation in Ireland underwent a significant change from an engine size based system to a Carbon Dioxide (CO2) emission rate based system. This resulted in a significant transition towards diesel fuelled vehicles in response to taxation change which typically emit less CO2 than other fuel types. The majority of vehicle categories are now diesel powered such as small public service vehicles, large public service vehicles and heavy goods vehicles are diesel powered and the car fleet is also pre-dominantly diesel powered. Whilst air quality in Ireland is considered relatively good in comparison to other countries in Europe this dependency on diesel fuelled vehicles is of concern as NO2 monitoring stations in Ireland continue to record NO2 concentration levels close to the European Union s Directive on Ambient Air Quality and Cleaner Air for Europe (2008/50/EC) limit values and any change in conditions (such as adverse weather conditions or increased traffic levels or other pollution sources) could lead to an exceedance of these limit values. The World Health Organisation reduced their annual limit value for NO2 in September 2021, due to the increasing evidence in literature which identify links between NO2 exposure and various health effects (respiratory and cardiovascular) in the population. The revised limit by the World Health Organisation considers levels of NO2 in excess of 10 µg/m3 to be harmful to the population. In the Irish context, only rural locations are achieving this revised limit value, therefore, identifying mitigation measures to reduce NO2 across wide regions of the country are necessary to reduce the impacts of air pollution on the health of the population.
This research develops an enhanced Wind Sector Land Use Regression (WS-LUR) model to estimate NO2 concentrations across Ireland, in areas where monitoring of air pollution is not available. The model incorporates details of the vehicle fleet breakdown within the WS-LUR model equation to weight vehicle type flows based on the emission rates of the vehicle type, which differentiates routes with different proportions of heavier emitting vehicles (such as haulage route). The model was developed in Excel and provides a simpler approach for NO2 concentration estimates to the same level of accuracy as detailed emissions modelling software. The model has two modelling approaches, the pre-set approach which utilises stored variable information within the model and therefore can calculate NO2 concentrations automatically once a location is specified, and a manual entry approach which allows a modeller to analyse any location by inputting values manually for each of the predictor variables within the model.
The model was validated against measured concentrations from numerous locations in Ireland for the 2016 to 2018 period and also validated in an additional analysis of the unique scenario / environment presented by the COVID lockdown period in 2020. A number of mitigation measures to improve air quality in Ireland that target model variables were identified and analysed using the enhanced WS-LUR model. These mitigation measures included; the relocation of a major business hub to reduce commercial properties in an area and reduce traffic traveling to and from the area; the removal of diesel fuelled vehicles from the small public service vehicle and large public service vehicle fleets; the construction of a ring road around a major city to provide an alternative route for traffic to bypass the city centre area and finally the introduction of a Low Emission Zone (LEZ) within Dublin City to influence transport mode choice and the number of vehicles entering the city centre area.
The original model captured 78% of spatial variability in NO2 and when checked against 2016 to 2018 conditions both the original model approach and the enhanced WS-LUR achieved cross-validation R2 of 76% confirming the accuracy was not impacted by the addition of a weighting which accounts for the vehicle fleet breakdown. The analysis of the COVID lockdown period also reinforced the confidence in the enhanced WS-LUR model to accurately model concentrations in unique scenarios, with a cross-validation R2 of 82% when excluding an outlier in the analysis. The changes in major route flows were the main contributors to the reductions in concentrations whilst the changes in weather conditions during the COVID lockdown period contributed to increases in concentrations across the majority of locations. The model successfully estimated the changes in concentrations due a number of mitigation measures. Positive results were achieved in the most heavily polluted areas in each of the mitigation measures (Dublin City Centre for the LEZ, Blanchardstown business hub relocation and the public service vehicles diesel removal measures; South Cork City for the Cork Ring Road mitigation measure), areas which are currently experiencing pollution levels close to the Directive 2008/50/EC limits.
Sponsor
Grant Number
Environmental Protection Agency (EPA)
Author's Homepage:
https://tcdlocalportal.tcd.ie/pls/EnterApex/f?p=800:71:0::::P71_USERNAME:DOMHNAIADescription:
APPROVED
Author: Ó Domhnaill, Aonghus
Advisor:
Broderick, BrianPublisher:
Trinity College Dublin. School of Engineering. Disc of Civil Structural & Environmental EngType of material:
ThesisCollections
Availability:
Full text availableMetadata
Show full item recordLicences: