Evaluation of Alternative Approaches for the Calculation of the CESI Air Quality Indicators and Analyses of Ozone Trends

1.  Title

Evaluation of Alternative Approaches for the Calculation of the CESI Air Quality Indicators and Analyses of Ozone Trends.



2.  Background

The Canadian Environmental Sustainability Indicators (CESI ) provides data and information to Canadians on the state of the environment and describes Canada’s progress on key environmental sustainability issues such as air quality.  The air quality indicators focus on reporting the ambient concentration of fine particles (PM2.5), ozone, sulphur dioxide (SO2), nitrogen dioxide (NO2) and volatile organic compounds (VOC).  

The air quality indicators are reported nationally and regionally for six regions consisting of the Atlantic provinces, Southern Québec, Southern Ontario, the Prairies and Northern Ontario, and British Columbia.  Environment Canada is considering the development of regional air quality indicators that are based on the six regional airsheds (Figure 1) of the new Air Quality Management System (AQMS ), which is being implemented by the federal, provincial, and territorial governments to further protect the environment and the health of Canadians.   Environment Canada wishes to evaluate the differences in regional indicator values and trends using the CESI air regions and the airsheds. 



















Figure 1: The airshed regions under the AQMS




















 
Currently, the national CESI air quality indicator value for a given air pollutant statistic (e.g. an annual average concentration) in a given year is the average of the value of the statistic over all the monitoring stations considered in the year (the station-averaging approach).  Similarly, the regional indicator value is calculated using data from all of the monitoring stations in a given region.  The national and regional indicator values can also be calculated through an alternate approach, called the city-averaging approach.  Under this approach,  cities with two or more monitoring stations are first assigned a single indicator value, being the average of the value of the statistic over all monitoring stations in the city. The national CESI air quality indicator value for a given air pollutant statistic is then obtained by averaging over the city indicator values and the values of the statistic from the remaining individual stations which were not part of a city indicator values.  Environment Canada wishes to explore the differences in indicator values and their trends between the station-averaging and the city-averaging approaches. 

Ground-level ozone (ozone) is formed by chemical reactions between primarily oxides of nitrogen  and VOC in the presence of sunlight, and can be influenced by many factors including weather conditions and origins of air masses as evaluated through back-trajectory analyses .  In some areas of Canada, the average ozone concentrations3 are affected by transboundary air pollution from the United States.  A number of reports have also shown that trends in ozone concentrations are not uniform across the country.  For example, in some locations peak ozone concentrations are decreasing while the average ozone concentrations are increasing.  In Ontario, summer average ozone concentrations have decreased during the past years, while the winter average ozone concentrations have increased.     Ozone trends also differ between urban and rural locations, and for locations downwind of large urban centres.  Environment Canada wishes to identify some of the possible factors influencing the different trends in ozone across the country. 



3.  Objectives

The main objectives of this contract are for the Contractor to:  

1)	Objective 1:  Develop regional air quality indicators using the current CESI air regions and the airsheds of the new AQMS and evaluate the differences, the benefits, and the issues of using them. 
2)	Objective 2:  Develop national and regional air quality indicators and trends using the station-averaging and the city-averaging approaches and evaluate the advantages and disadvantages of each considering the intent and the goals of the CESI program. 
3)	Objective 3:  Perform ozone trend analyses and other in-depth analyses to identify some of the factors that influence the ozone concentrations and trends across the country.  
 

4.  Data Source and Methodologies


4.1  Data Source

The Canadian ambient air pollutant concentrations that are required to accomplish the tasks for this contract will be provided to the Contractor by Environment Canada.  The data will be extracted from the Canada-wide Air Quality Database (managed by Environment Canada), which includes data from the National Air Pollution Surveillance (NAPS) program, from the Canadian Air and Precipitation Monitoring Network (CAPMoN), and provincial, territorial and municipal monitoring stations which are not part of NAPS.  The Contractor will be responsible for acquiring the ambient concentration data from monitoring stations in the United States that is required to accomplish the tasks in relation to Objective (iii).  


4.2  Calculation Methods and Data Completeness

The fine particulate matter (PM2.5) daily 24-hour average concentrations (daily 24-hr PM2.5) 
and the ozone daily maximum 8-hour average concentrations (daily max 8-hr O3) must be calculated by the Contractor using the procedures and data completeness requirements specified in the Guidance Document on Achievement Determination for the Canadian Ambient Air Quality Standards for Fine Particulate Matter and Ozone  (GDAD) published by the Canadian Council of Ministers of the Environment, in 2012.

The annual average of the daily 24-hr PM2.5, the annual 98th percentile of the daily 24-hr PM2.5 and the annual 4th highest of the daily max 8-hr O3 must be calculated by the Contractor using the procedures and data completeness requirements specified in the GDAD.

The annual average and quarterly average concentrations of nitric oxide (NO) and nitrogen dioxide (NO2) must be calculated by the Contractor from the respective 1-hour average concentrations.  The annual average and quarterly average concentrations of volatile organic compounds (VOC) will be calculated by the Contractor from the 24-hour average concentrations. 

The data completeness requirements for all annual averages will be identical to those applicable to the annual average of the daily 24-hr PM2.5 as specified in the GDAD.

All quarterly statistics required of the Contractor under Objective (3) will be for the following quarters:

Winter:     January 1 to March 31
Spring:     April 1 to June 30
Summer:  July 1 to September 30
Fall:         October 1 to December 31 

A given quarterly statistic will be considered valid if the data completeness for the quarter is at least 75%.   For example, the winter average NO2 concentration at a station is considered valid if there are at least 75% of all 1-hour concentrations in the quarter. For the calculation of trends, and the inclusion of a station in the national, regional and city indicator values, a station must be included by the Contractor if the value of the statistic being considered is available in at least 75% of the years over the trend period, corresponding to 11 years for a 15-year trend period, and for 10 years for a 13-year trend period.  This criteria also applies for the trends by calendar quarters.  


4.3  Trend Method 

The temporal trends and the statistical significance of the trend at the 95% confidence level must be evaluated using the non-parametric Sen’s approach as described in  Estimates of the regression coefficient based on Kendall's tau”, by Pranab Kumar Sen, in Journal of the American Statistical Association, 63, pages 1379-1389, December, 1968.  



5.  Scope of Work for Objectives 1 and 2   

For Objectives 1 and 2, the Contractor must undertake the necessary calculations and analyses, and provide a single combined report to summarise the work undertaken and one Power Point presentation to summarise the content of the report as described in Tables 2 and 3 below. 


5.1  Statistics and Trend Periods


For Objectives 1 and 2, the Contractor must calculate the statistics described in Table 1 for the 13-year period from 2000 to 2012 for PM2.5, and for the 15-year period from 1998 to 2012 for ozone.  The statistics in Table 1 must be compiled by the Contractor for all monitoring stations that meet the requirements specified in Section 4.  


Table 1:  Annual statistics to be calculated for Objectives 1 and 2.
Pollutant	Statistic
PM2.5	The annual average of the daily 24-hour average PM2.5 concentrations (daily 24-hr PM2.5).
PM2.5	The annual 98th percentile of the daily 24-hr PM2.5
Ozone	The 1-year average of the daily maximum 8-hour average concentrations  (daily max 8-hr O3)
Ozone	The annual 4th highest of daily max 8-hr O3



 

5.2  Tasks for Objective 1

The purpose of Objective 1 is to develop regional air quality indicators for PM2.5 and ozone using the current CESI air regions and the airsheds of the new AQMS, and to evaluate the differences, the benefits, and the issues of using them.  The tasks the Contractor must complete for Objective 1 for both PM2.5 and ozone are outlined in Table 2.  Environment Canada will provide the Contractor with a list of monitoring stations for the CESI air regions and for the AQMS airsheds. 
 

Table 2:  Tasks to be completed by Contractor for Objective 1
Task 	Task Description 
1.1	Calculate the annual indicator values for the CESI air regions and for the airsheds for each year in the trend periods and for each statistic in Table 1 using the station-averaging approach,
 
For the purpose of the work under Objective 1, the East Central airshed is to be separated into the East Central – Ontario airshed and the East-Central - Quebec airshed.
1.2	Show charts of the regional indicator values and their trend (including the statistical significance) for the CESI air regions and the airsheds (with the East Central airshed separated into the East Central – Ontario airshed and the East-Central - Quebec airshed).
1.3	Indicate on maps of Canada the direction (increasing or decreasing) and magnitude of the trend on a station basis for all considered stations.  For the preparation of these maps, the Contractor should consider the format used in Figure 19 in the report Fine Particles and Ozone in Canada, A Canada-wide Standards Perspective – 2003 National Summary available at:  http://www.ccme.ca/assets/pdf/2003_pm_oz_ntnlsmryrpt_e.pdf 

For these maps, the Contractor must identify the monitoring stations according to the population of the community in which they are located as follows:

•	large urban (LU) stations, located in cities with population greater or equal to 500,000 
•	urban stations (U), located in cities with population between 100,000 and less than 500,000
•	small urban (SU), located in cities with population less than 100,000
•	non-urban (NU), these are non-urban stations
•	CAPMoN, for the CAPMoN stations 
1.4	Compare and discuss the trends for the indicator values between the CESI regions and the airsheds, including a discussion on the spatial representativeness of the regional indicator values.
1.5	Tabulate and present the number of monitoring stations used in the calculation of the regional indicators for the CESI air regions and for the airsheds.
1.6	Prepare a report which must summarise the work undertaken to complete Objective 1, including describing the tasks and related analyses conducted and methodologies used, presenting the results of the analyses and comparisons, and a synthesis/discussion of the results obtained and conclusions.  Appendices can be used to provide more detailed information on the analyses performed, the methodologies used, and the results obtained.  
1.7	Prepare a presentation in Microsoft PowerPoint to summarise the content of the report.



5.3  Tasks for Objective 2   

The purpose of Objective 2 is to develop national and regional air quality indicators and trends for PM2.5 and ozone using the station-averaging and the city-averaging approaches and evaluate the advantages and disadvantages of each considering the intent and the goals of the CESI program.  The tasks to be completed by the Contractor for Objective 2 are outlined in Table 3.  Annual regional indicator values must only be compiled by the Contractor for the CESI air regions.   

For the purpose of Objective 2, “city ” means communities with a population of 5000 and more as defined by Statistic Canada’s census subdivisions which are available at:  http://www12.statcan.gc.ca/census-recensement/2011/dp-pd/hlt-fst/pd-pl/Tables-Tableaux.cfm?LANG=Eng&T=300.   The Contractor and the Scientific Authority will agree on which cities to use for the calculation of the city indicators before beginning the work. 
  
   
Table 3:  Tasks to be completed by the Contractor for Objective 2
Task	Description of Task
2.1	Calculate the annual indicator values for the CESI air regions for each year of the trend periods and for each statistic provided in Table 1 using both the station-averaging and the city-indicator averaging approaches. 
2.2	Show charts of the national and regional indicator values and their trends for each of the two averaging approaches. 
2.3	For the national indicator, calculate, show and discuss how much each province and each city with two or more monitoring stations contributes (in absolute value and percentage) to the national indicator values for each year of the trend periods, under each of the two averaging approaches. 
2.4	For the regional indicators, calculate, show and discuss how much each city with two or more monitoring stations contributes (in absolute value and percentage) to the regional indicator values for each year of the trend periods, under each of the two averaging approaches. 
2.5	Compare and discuss the indicator values and trends between the station-averaging approach and the city-averaging approach.
2.6	Discuss the advantages and disadvantages of the two approaches from a purely statistical context, a population exposure context, and as indicators to gauge the effectiveness of implemented emission reduction measures.   
2.7	Prepare a report which must summarise the work undertaken to complete Objective 2, including describing the tasks and related analyses conducted and methodologies used, present the results of the analyses, and a synthesis/discussion of the results obtained and conclusions.  Appendices may be used to provide more detailed information on the analyses performed, the methodologies used, and the results obtained.  
2.8	Prepare a presentation in Microsoft PowerPoint to summarise the content of the report.



6.  Scope of Work for Objective  3  

The purpose of Objective 3 is to perform ozone trend analyses and other in-depth analyses to assist identify some of the factors influencing the ozone concentrations and trends.   The tasks to be completed for Objective 3 are outlined in Table 4.  The Contractor will identify representative urban and rural monitoring stations in the United States (U.S.) located within 500 km of the border and discuss with the Scientific Authority their suitability for use in the analyses before beginning the work.  Unless otherwise mentioned, only stations that satisfy the requirements specified in Section 4 will be used for Objective 3.

  
Table 4:  Tasks to be completed by the Contractor for Objective 3
Task 	Task Description 
3.1	Calculate the following annual statistics and their trends (including the statistical significance) for the 15-year period from 1998 to 2012 for each monitoring station in the Canada-wide Air Quality Database and for selected monitoring stations in the U.S. 

•	annual highest 1-hour O3,
•	annual highest daily max 8-hr O3,
•	annual 4th highest daily max 8-hr O3,
•	annual average daily max 8-hr O3,
•	annual average 1-hour nitric oxide (NO),
•	annual average 1-hour NO2,
•	annual average 24-hour VOC.

Show on maps the direction (increasing or decreasing) and the magnitude of the trends for all of the considered stations.  For the preparation of these maps, the service provider will consider the format used in Figure 19 in the report Fine Particles and Ozone in Canada, A Canada-wide Standards Perspective – 2003 National Summary available at:  http://www.ccme.ca/assets/pdf/2003_pm_oz_ntnlsmryrpt_e.pdf 

Provide tabulated trend data (i.e. rate of change and its statistical significance and the % change over the trend period) for each station.   

For these maps, the Contractor must identify the monitoring stations according to the population of the community in which they are located: 

•	large urban (LU) stations, located in cities with population greater or equal to 500,000 
•	urban stations (U), located in cities with population between 100,000 and less than 500,000
•	small urban (SU), located in cities with population less than 100,000
•	non-urban (NU), these are non-urban stations
•	CAPMoN, for the CAPMoN stations
3.2	Calculate quarterly statistics and their trends for the 15-year period from 1998 to 2012 for each monitoring station in the Canada-wide Air Quality Database and for selected monitoring stations in the U.S.  The statistics to calculate in each quarter are: 

•	the average 1-hour O3,
•	the average daily max 8-hr O3,
•	the average 1-hour NO,
•	the average 1-hour NO2, 
•	the average 24-hour VOC.  

Show on maps the direction (increasing or decreasing) and magnitude of the trends by quarter for all considered stations as in Task 3.1.  For these maps, identify the ozone monitoring stations according to the population of the community in which they are located as identified in Task 3.1. 

Calculate the annual composite quarterly average and its trend by province for the above statistics, and discuss how they compare to the data from individual stations. 

Show individual charts of the quarterly time series and their trends for representative key stations in Canada and the U.S. for each quarter.  Representative stations would include, for example, a comparison of ozone concentrations in downtown areas with those in residential areas  a comparison of concentrations between large urban areas and locations downwind of the large urban area (e.g., Toronto vs Peterborough);  a comparison of concentrations between large urban areas and rural areas;  and stations close to the Canada-U.S. border;  and any other other analyses that would assist in identifying the factors that influence the changes in the ozone concentrations.
3.3	Compute 48-hour back-trajectories using a peer-reviewed model for each year in the period from 1998 to 2012 for Simcoe, Toronto, Dorset, Halifax, Winnipeg and Edmonton; and perform clustering analyses using a peer-reviewed approach to cluster the trajectories into not more than eight clusters considering both direction of movement and speed.  

The Contractor must discuss with the Scientific Authority the trajectory model to use; the clustering technique; the heights of the trajectories; and any other relevant information that will be used to complete this task.

The Contractor must calculate and show, using tables and charts, the following statistics for each of the six  locations:

1.	The quarterly average and standard deviation for 1-hour ozone and daily max 8-hr O3 associated with each cluster for each year in the trend period. 
2.	The number of back-trajectories per cluster per quarter for each year in the trend period.
3.	The trend in the average 1-hour ozone and the average daily max 8-hr O3 by cluster and season.

Discuss how the trends by cluster and quarter compare with the overall trends. 

Discuss the results obtained and identify the possible factors which may have influenced any differences in concentrations and trends between clusters and between quarters.  Discuss, in particular, how the annual variability in the number of trajectories within a given cluster might influenced the annual variability in ozone concentrations.    

Discuss the emissions of NOX and VOC along each cluster and how their changes over the years might have influenced the ozone trends.  Discuss the typical weather conditions associated with each cluster and their possible influence in favouring or hindering ozone formation.  
3.4	Compile and show in charts the trends in emissions of NOX for on-road vehicles at the provincial level, and as feasible, scale provincial emissions to urban centers with population of > 500,000 for years within the trend period.  Discuss how these relate to the trends of ambient concentrations of NO, NO2 and ozone.
3.5	Compile and show in charts the trends in emissions of total anthropogenic VOC at the provincial level, and as feasible, scale provincial emissions to urban centers with population of > 500,000 for years within the trend period.   Discuss how these relate to the trends of ambient concentrations of VOC and ozone measured.
3.6	Previous work indicated that emissions from states located within the U.S. Pollutant Emission Management Area (PEMA) region contribute to ozone in southern Ontario and Southern Quebec.  The PEMA states are shown in Figure 11 of the Canada-U.S.  2012 Air Quality Agreement progress report available at:
http://www.ec.gc.ca/Publications/default.asp?lang=En&xml=D9D6380B-4834-41C4-9D36-B6E3348F1A39.
Compile and show using charts the anthropogenic emissions of NOX and VOC for the PEMA states and discuss how the changes in emissions might have influenced ozone concentrations at downwind locations in Ontario and Quebec.
3.7	Conduct analyses to identify the monitoring stations located downwind of major urban areas (i.e. population > 500,000) or downwind of industrial complexes, and evaluate and discuss how the trends in ozone concentrations at these downwind locations may have been affected by the trend in upwind emissions of NOX and VOC.
3.8	Prepare a report which will address the objectives of the work, describes the tasks and related analyses conducted and methodologies used, present the results of the analyses, and a synthesis/discussion of the results and conclusions.  Appendices can be used to provide more detailed information on the analyses performed, the methodologies used, and the results obtained.  

The report is to include discussions, considering the known chemistry of ozone, its precursors and transport, on the relationship between trends in ozone concentrations, NOX concentrations, VOC concentrations, and freshly emitted and aged NOX emissions, with the aim of elucidating any opposing trends, any difference in seasonal trends, difference in spatial trends, and differences in trends between traffic stations, urban stations and rural stations. 

Discuss how emission of NO in urban areas may suppress ozone already present in the air, while enhancing ozone formation downwind.  Discuss how reductions in NO emissions in urban areas may have influenced ozone in both the urban area and in downwind areas.  Discuss how emission trends of NOX and VOC in the considered U.S. states may have influenced ozone in downwind areas in Canada.  Discuss any influence of “global background” ozone and provide information to support any claim made.

Perform and discuss the results of any other analyses conducted as needed to support the discussions.  
3.9	Prepare a presentation in Microsoft Power Point to summarise the content of the report.


 

7.  Deliverables and timelines

The list of Deliverables to be provided by the Contractor to the Scientific Authority and the associated timelines are provided below.  The deliverables produced by the Contractor will be reviewed by the Scientific Authority and other experts at Environment Canada.  


Deliverable  1 

A first draft report for Objectives 1 and 2 in Microsoft Word format no later than 10 weeks after contract award. 

The Scientific Authority will provide timely written and verbal comments to the Contractor on these draft reports.  The Scientific Authority will schedule conference calls with the Contractor to discuss the comments.  The Contractor must indicate how the comments will be addressed and modify the draft report in consideration of the comments and discussions.  


Deliverable  2 

A first draft report for Objective 3 in Microsoft Word format no later than 3 weeks after acceptance of Deliverable 1.  

The Scientific Authority will provide timely written and verbal comments to the Contractor on the draft report.  The Scientific Authority will schedule conferences calls with the Contractor to discuss the comments.  The Contractor will indicate how the comments will be addressed and modify the draft report in consideration of the comments and discussions.         


Deliverable  3

A 2nd draft report for Objectives 1 and 2, and for Objective 3, and a first draft of their respective PowerPoint presentations no later than four weeks before the termination date of the contract. The 2nd draft reports for Objectives 1 and 2, and for Objective 3, must incorporate the discussed comments under Deliverables 1 and 2 respectively.

The Scientific Authority will provide timely written and verbal comments to the Contractor on the 2nd draft reports and presentations.  The Scientific Authority will schedule conference calls with the Contractor to discuss the comments.  The Contractor must indicate how the comments will be addressed and modify the draft reports and draft presentations in consideration of the comments and discussions.        


Deliverable  4

Two hard copies of each of the final reports, electronic copies of the final reports (in Microsoft Word format), and electronic copies of each final presentation (in Microsoft Power Point) by the termination date of the contract.  The two hard copies of the reports are to contain print on both sides of a page.  The final reports for Objectives 1 and 2, and for Objective 3, and the final presentations must incorporate the discussed comments under Deliverables 3.


Deliverable  5 

By the termination date of the contract, an electronic copy of: 

a)	all raw data used for this contract in Microsoft Excel.

b)	all analysed data used to generate the charts, maps and tables appearing in the final report in a file format accessible with  Microsoft Excel. 

The detail of the data specified above must be sufficient enough to allow the generation of all charts, maps and tables appearing in the final report.


Deliverable  6 

At the discretion of the Contracting Authority, an electronic copy of any other data used or generated under this contract not already provided under deliverable 5.  This provision will remain in effect until three years after the termination date of the contract.  



8.  Budget, Meetings and Travel

Environment Canada has established funding for this project at a maximum amount of $30,170  (excluding applicable taxes).  

There will be no in-person meetings for this contract.  All meetings will be by teleconference or video-conferences.  No additional funds will be provided for attending the teleconferences and video-conferences.  No travel expenses will be paid through this contract.