Review of the 2009 Gauteng Air Quality Management Plan

Transcription

1 Review of the 2009 Gauteng Air Quality Management Plan GT/GDARD/154/2016 DISPERSION MODELLING PLAN OF STUDY To: From: Gauteng Provincial Treasury Imbumba House 75 Fox Street Marshalltown, Johannesburg umoya-nilu Consulting (Pty) Ltd P O Box 76400, Lynnwood, 0040 Tel: Mobile: benton@umoya-nilu.co.za Website: i

2 Report details: Client: Gauteng Department of Agriculture and Rural Development (GDARD) Report title: : Dispersion Modelling Plan of Study Report number: umn Project number: umn Project: GDARD AQMP Review Version: Final (16 July 2017) Author details: Author: Mark Zunckel Review: Atham Raghunandan This report has been produced for the Gauteng Department of Agriculture and Rural Development by umoya-nilu Consulting (Pty) Ltd. No part of the report may be reproduced in any manner without written permission from the Gauteng Department of Agriculture and Rural Development and umoya-nilu Consulting (Pty) Ltd. When used in a reference this document should be cited as follows: umoya-nilu (2017): : Dispersion Modelling Plan of Study, Report No. umn , July i

3 TABLE OF CONTENTS TABLE OF CONTENTS... ii LIST OF FIGURES... ii 1 INTRODUCTION Background Terms of Reference Project Location Land use determination in the modelling domain Elevation data (DEM) and resolution MODELLING OBJECTIVE DISPERSION MODEL SCENARIOS Ekurhuleni Industry Wind entrained dust from mine dust Motor vehicles emissions highway simulation Motor vehicle emissions BRT simulation METEOROLOGICAL DATA PROPOSED MODEL MODEL PARAMETERISATION MODEL ACCURACY... 8 REFERENCES... 9 LIST OF FIGURES Figure 1-1: Location map showing Gauteng and the metropolitan, district and local municipalities... 2 Figure 1-2: Land cover Map of Gauteng (GDARD, 2014)... 3 LIST OF TABLES Table 1-1: Land types, use and structures and vegetation cover... 3 Table 6-1: Parameterisation of key variables for CALMET... 7 Table 6-2: Parameterisation of key variables for CALPUFF... 8 ii

4 1 INTRODUCTION 1.1 Background The Gauteng Department of Agriculture and Rural Development (GDARD) recognised the need to review the 2009 Air Quality Management Plan (AQMP) in The overall objective of the project for the review of the GDARD AQMP is to establish the current status of air quality in the province and to develop a revised AQMP with objectives to ensure prevention and improvement in air quality in Gauteng, so as to fulfill Government s Constitutional mandate to ensure an environment that is not harmful to the health and well-being of all South Africans. A baseline assessment is the first step in the development or review of an AQMP and includes the assessment of the status of the air quality management tools and systems, including dispersion modelling. The focus of this report is the dispersion modelling plan of study according to the requirements of the DEA regulation for dispersion modelling, documented in the Code of Practice for Air Dispersion Modelling in Air Quality Management in South Africa (DEA, 2014). 1.2 Terms of Reference Dispersion modelling forms Task 4 of the Objective 2, i.e. the baseline assessment report. The terms of reference for the dispersion modelling are stated as: Based on the emission inventory developed, the Service Provider will use the appropriate modelling approach to provide air pollution dispersion model results that characterises the impact of air pollution on ambient air quality in both time and space. The service provider is expected to describe the modelling approach and demonstrate its applicability. A final decision on the modelling approach should be discussed and approved prior to any work commencement and should be in line with the Regulations regarding air dispersion modelling (Government Gazette, No ). Upon completion of the project the service provider will supply GDARD with copies of all input data, model parameterization and post-processing files. 1.3 Project Location Gauteng is located in the central parts of the country and is bordered by the provinces of the North West, Limpopo, Mpumalanga and Free State (Figure 1-1). 1

5 Figure 1-1: Location map showing Gauteng and the metropolitan, district and local municipalities It comprises three metropolitan municipalities namely, City of Johannesburg (CoJ), City of Tshwane (CoT) and Ekurhuleni Metropolitan Municipality (EMM), and two district municipalities, Sedibeng District Municipality and West Rand District Municipality. 1.4 Land use determination in the modelling domain The Code of Practice for Air Dispersion Modelling in Air Quality Management in South Africa (DEA, 2014) recommends the Land Use Procedure as sufficient for determining the urban/rural status of a modelling domain. The classification of the study area as urban or rural is based on the Auer method specified in the US EPA guideline on air dispersion models (US EPA, 2005). From the Auer s method, areas typically defined as rural include residences with grass lawns and trees, large estates, metropolitan parks and golf courses, agricultural areas, undeveloped land and water surfaces. An area is defined as urban if it has less than 35% vegetation coverage or the area falls into one of the use types in Table 1-1. The Land cover Map of Gauteng (GDARD, 2014) is shown in Figure

6 Table 1-1: Land types, use and structures and vegetation cover Urban Land Use Type Use and Structures Vegetation I1 Heavy industrial Less than 5 % I2 Light/moderate industrial Less than 5 % C1 Commercial Less than 15 % R2 Dense single / multi-family Less than 30 % R3 Multi-family, two-story Less than 35 % Figure 1-2: Land cover Map of Gauteng (GDARD, 2014) 3

7 1.5 Elevation data (DEM) and resolution The topographical and land use for the respective modelling domains is obtained from the dataset accompanying the CSIRO s The Air Pollution Model (TAPM) modelling package. This dataset includes global terrain elevation and land use classification data on a longitude/latitude grid at 30-second grid spacing from the US Geological Survey, Earth Resources Observation Systems (EROS) Data Center. 2 MODELLING OBJECTIVE There is a comprehensive ambient air quality monitoring network in Gauteng, providing a reasonably good understanding of concentrations for a range of pollutants. The objective of dispersion modelling for the 2009 AQMP Review is to add to the understanding of four key issues regarding air pollution in the province. The four issues to be assessed are to obtain an understanding of the: i. Spatial impact of emission from the concentration of industrial sources in Ekurhuleni on ambient air quality; ii. Spatial impact of wind entrained dust from mine dumps on ambient air quality; iii. Spatial impact of motor vehicle emissions from a selected highway in Gauteng on ambient air quality; and iv. Impact on air quality of the implementation of the Bus Rapid Transit (BRT) system. By adding insights to the four issues, the assessments will contribute to identifying priority issues and will provide input to the review of the AQMP. 3 DISPERSION MODEL SCENARIOS 3.1 Ekurhuleni Industry The following scenario will be modelled in order to simulate the dispersion of emissions from industrial sources in Ekurhuleni and the spatial extent of the impact: All industrial point source emissions reported to the National Atmospheric Emission Inventory System (NAEIS) for 2016 in Ekurhuleni will be included in the simulation; A modelling domain of 30 km x 30 km will be used, centred on Ekurhuleni; A regular grid spacing of 1 km will be applied; 3-years of hourly surface and upper air meteorological data will be applied; and Background concentrations are excluded. Emissions from Listed Activities (Section 21) and Controlled Emitters (Section 23) in Ekurhuleni for 2016 that were reported to the NAEIS will be included in modelling this scenario. The emission rates and stack parameters will therefore be obtained from the Provincial Air Quality Officer. 4

8 At each facility for each stack the following data required are required. Stack name and location (latitude and longitude); Stack height; Emission temperature and exit velocity; Emissions rates of pollutants relevant to each stack. 3.2 Wind entrained dust from mine dumps The dispersion and spatial impact of wind entrained dust from mine dumps on ambient air quality will be assessed by modelling the dispersion and deposition of particulates from a single selected mine dump. The following will apply in the modelling: Criteria for the selection of the mine dump will be defined with the PSC who will then assist with the selection; A modelling domain of 10 km x 10 km will be used, centred on the selected mine dump; A regular grid spacing of 500 m will be applied; 1-year of hourly surface and upper air meteorological data will be applied; Emissions from the selected mine dump only will be included in the simulation; Background concentrations will be excluded. Emissions resulting from wind entrainment from the mine dump will be estimated using US- EPA methodology and hourly wind speed and direction. The information required for the selected mine dump are: Location (latitude and longitude), lateral and vertical dimensions; Orientation; Area vegetated or compacted; and Silt and moisture content of the surface material. 3.3 Motor vehicles emissions highway simulation A scenario assessment to simulate the dispersion of pollutants from motor vehicle emissions from a selected highway in Gauteng and to assess the spatial extent of the impact on ambient air quality will focus on the Gillooly s Interchange. This is a typical Gauteng interchange and is located in Bedfordview where the N3 Eastern Bypass and the R24 Airport Freeway intersect. The Bedfordview ambient air quality monitoring station is located at the interchange. The following will apply to the dispersion modelling: A modelling domain providing a 5 km zone either side of the Gillooy s Interchange will be used; A regular grid spacing of 100 m will be applied; 1-year of hourly surface and upper air meteorological data will be applied; Hourly emissions for a typical day will be used in the simulation. 5

9 o Emissions will be estimated using vehicle count data, vehicle kilometres travelled for different vehicle categories and emission factors. o Emissions will be estimated for a year and averaged to provide hourly emissions for a typical day. Background concentrations will be excluded. 3.4 Motor vehicle emissions BRT simulation A scenario assessment to simulate the change in air quality resulting from motor vehicle emissions before and after the implementation of the BRT will consider a case study on a section of road in Johannesburg. The following will apply to the dispersion modelling: Review of feasibility studies undertaken by Ria Vaya for the Johannesburg BRT; Definition of criteria for the selection of a section of roadway with the assistance of the PSC who will then assist with the selection; A modelling domain extending 1 km either side of the roadway will be used, centred along the length of the selected roadway; A regular grid spacing of 100 m will be applied; 1-year of hourly surface and upper air meteorological data will be applied; The simulation will consider emissions from the vehicle traffic on the road with and without the BRT system in place. o Feasibility studies for the Ria Vaya BRT in Johannesburg will be reviewed; o Emissions for the vehicle traffic on the roadway before the implementation of the BRT will be estimated using vehicle count data, vehicle kilometres travelled for different vehicle categories and emission factors. o Emissions for the vehicle traffic on the roadway after the implementation of the BRT will be estimated using vehicle count data, vehicle kilometres travelled for different vehicle categories (including BRT busses) and emission factors. Background concentrations will be excluded. 4 METEOROLOGICAL DATA The South African Weather Service (SAWS) and municipalities operate meteorological monitoring stations within Gauteng. These do not provide comprehensive data coverage across the province. To address this issue the TAPM meteorological data will be used to supplement the meteorology in the modelling domain (CSIRO, 2008). Three years of hourly observed meteorological data for the period from selected SAWS meteorological stations will be input to TAPM to nudge the modelled meteorology towards the observations and to create a continuous meteorological input file for the domain. The datasets will be chosen based on the completeness of the data record and the representativeness of the surrounding areas. TAPM is set-up in a nested configuration of two domains. The outer domain is 400 km by 400 km at a 20 km grid resolution and the inner domain is 200 km by 200 km at a 10 km grid 6

10 resolution. The subset of the entire TAPM model output in the form of pre-processed gridded surface meteorological data fields will be input into CALMET. This approach eliminates potential issues associated with missing observational data. Upper air data is included in the pre-processed TAPM meteorological fields. 5 PROPOSED MODEL A Level 3 air quality assessment is conducted in situations where the purpose of the assessment requires a detailed understanding of the air quality impacts (time and space variation of the concentrations) and when it is important to account for causality effects, calms, non-linear plume trajectories, spatial variations in turbulent mixing, multiple source types and chemical transformations (DEA, 2014). A Level 3 assessment may be used in situations where there is a need to evaluate air quality consequences under a permitting or environmental assessment process for large industrial developments that have considerable social, economic and potential environmental consequences. Under these circumstances, this study clearly demonstrates the need for a Level 3 assessment. CALPUFF is a US EPA approved air dispersion model ( and is recommended by the DEA for Level 3 assessments (DEA, 2014). CALPUFF is considered to be an appropriate air dispersion model for the purpose of this assessment as it is well suited to simulate dispersion from the large array of sources and source types. More information about the model can be found in the User s Guide for the CALPUFF Dispersion Model (US EPA, 1995). 6 MODEL PARAMETERISATION The parameterisation of key variables that will apply in CALMET and CALPUFF for each of the four scenarios are indicated in Table 6-1 and Table 6-2 respectively. Table 6-1: Parameterisation of key variables for CALMET Parameter Model value 12 vertical cell face heights (m) 0, 20, 40, 80, 160, 320, 640, 1000, 1500, 2000, 2500, 3000, 4000 Coriolis parameter (per second) Empirical constants for mixing height equation Minimum potential temperature lapse rate (K/m) Depth of layer above convective mixing height through which lapse rate is computed (m) Wind field model Neutral, mechanical: 1.41 Convective: 0.15 Stable: 2400 Overwater, mechanical: Diagnostic wind module 7

11 Parameter Surface wind extrapolation Restrictions on extrapolation of surface data Radius of influence of terrain features (km) Radius of influence of surface stations (km) Similarity theory Model value No extrapolation as modelled upper air data field is applied 5 Not used as continuous surface data field is applied Table 6-2: Parameterisation of key variables for CALPUFF Parameter Chemical transformation Wind speed profile Calm conditions Plume rise Dispersion Dispersion option Terrain adjustment method Model value Default NO2 conversion factor is applied Rural Wind speed < 0.5 m/s Transitional plume rise, stack tip downwash, and partial plume penetration is modelled CALPUFF used in PUFF mode Pasquill-Gifford coefficients are used for rural and McElroy-Pooler coefficients are used for urban Partial plume path adjustment 7 MODEL ACCURACY Air quality models attempt to predict ambient concentrations based on known or measured parameters, such as wind speed, temperature profiles, solar radiation and emissions. There are however, variations in the parameters that are not measured, the so-called unknown parameters as well as unresolved details of atmospheric turbulent flow. Variations in these unknown parameters can result in deviations of the predicted concentrations of the same event, even though the known parameters are fixed. There are also reducible uncertainties that result from inaccuracies in the model, errors in input values and errors in the measured concentrations. These might include poor quality or unrepresentative meteorological, geophysical and source emission data, errors in the measured concentrations that are used to compare with model predictions and inadequate model physics and formulation used to predict the concentrations. Reducible uncertainties can be controlled or minimised. This is done by using accurate input data, preparing the input files correctly, checking and re-checking for errors, correcting for odd model behaviour, ensuring that the errors in the measured data are minimised and applying appropriate model physics. 8

12 Models recommended in the DEA dispersion modelling guideline (DEA, 2014) have been evaluated using a range of modelling test kits ( CALPUFF is one of the models that have been evaluated and it is therefore not mandatory to perform any modelling evaluations. Rather the accuracy of the modelling in this assessment is enhanced by every effort to minimise the reducible uncertainties in input data and model parameterisation. REFERENCES CSIRO (2008): TAPM V4. User Manual, CSIRO Marine and Atmospheric Research Internal Report No. 5, October 2008,ISBN: , a21befc593b9&dsid=ds1. DEA (2014): Code of Practice for Air Dispersion Modelling in Air Quality Management in South Africa, Government Notice R.533, Government Gazette, no , 11 July GDARD (2014): Gauteng Environmental Management Framework Report, November 2014, oduction%20and%20status%20quo.pdf. US EPA (2005): Revision to the Guideline on Air Quality Models: Adoption of a Preferred General Purpose (Flat and Complex Terrain) Dispersion Model and Other Revisions; Final Rule. US EPA. US EPA (1995): A User s Guide for the CALPUFF Dispersion Model. EPA-454/B US EPA, Research Triangle Park, N.C. 9