Description
Air pollution is the introduction into the atmosphere of chemicals, particulates, or biological materials that cause discomfort, disease, or death to humans, damage other living organisms such as food crops, or damage the natural environment or built environment.
Introduction Progress and examples Future developments
The Open-Source Air Pollution Project
Community tools for analysing air pollution data
David Carslaw
Institute for Transport Studies, University of Leeds
14 May 2008
Air Pollution Forecasting Seminar
The Open-Source Air Pollution Project
Introduction Progress and examples Future developments
Contents Openness Background
Summary
Background to the project Progress and examples Future directions and developments
Air Pollution Forecasting Seminar
The Open-Source Air Pollution Project
Introduction Progress and examples Future developments
Contents Openness Background
The bene?ts of openness
Transparency should be at the heart of environmental regulation
Those a?ected by environmental decisions should be able to scrutinise the tools that lead to those decisions
Air Pollution Forecasting Seminar
The Open-Source Air Pollution Project
Introduction Progress and examples Future developments
Contents Openness Background
The bene?ts of openness
Transparency should be at the heart of environmental regulation
Those a?ected by environmental decisions should be able to scrutinise the tools that lead to those decisions Many examples from the USA
Air Pollution Forecasting Seminar
The Open-Source Air Pollution Project
Introduction Progress and examples Future developments
Contents Openness Background
The bene?ts of openness
Transparency should be at the heart of environmental regulation
Those a?ected by environmental decisions should be able to scrutinise the tools that lead to those decisions Many examples from the USA
Open-source software
Air Pollution Forecasting Seminar
The Open-Source Air Pollution Project
Introduction Progress and examples Future developments
Contents Openness Background
The bene?ts of openness
Transparency should be at the heart of environmental regulation
Those a?ected by environmental decisions should be able to scrutinise the tools that lead to those decisions Many examples from the USA
Open-source software
All source code made available, free and can be modi?ed by anyone
Air Pollution Forecasting Seminar
The Open-Source Air Pollution Project
Introduction Progress and examples Future developments
Contents Openness Background
The bene?ts of openness
Transparency should be at the heart of environmental regulation
Those a?ected by environmental decisions should be able to scrutinise the tools that lead to those decisions Many examples from the USA
Open-source software
All source code made available, free and can be modi?ed by anyone No longer in the realm of enthusiasts e.g. Linux, MySQL, R
Air Pollution Forecasting Seminar
The Open-Source Air Pollution Project
Introduction Progress and examples Future developments
Contents Openness Background
The bene?ts of openness
Transparency should be at the heart of environmental regulation
Those a?ected by environmental decisions should be able to scrutinise the tools that lead to those decisions Many examples from the USA
Open-source software
All source code made available, free and can be modi?ed by anyone No longer in the realm of enthusiasts e.g. Linux, MySQL, R
Both promote participation and ownership
Air Pollution Forecasting Seminar
The Open-Source Air Pollution Project
Introduction Progress and examples Future developments
Contents Openness Background
The bene?ts of openness
Transparency should be at the heart of environmental regulation
Those a?ected by environmental decisions should be able to scrutinise the tools that lead to those decisions Many examples from the USA
Open-source software
All source code made available, free and can be modi?ed by anyone No longer in the realm of enthusiasts e.g. Linux, MySQL, R
Both promote participation and ownership
Air Pollution Forecasting Seminar
The Open-Source Air Pollution Project
Introduction Progress and examples Future developments
Contents Openness Background
Interesting aside – climate change and the “hockey stick”
Controversy over IPCC “hockey stick” temperature grapha Statistical methods shown to be ?awed (McIntyre and McKitrick) Code and data made available – some in R
Mann, M.E. et al. (1998). Global-scale temperature patterns and climate forcing over the past six centuries. Nature, Vol. 392, pp. 779787.
a
Air Pollution Forecasting Seminar
The Open-Source Air Pollution Project
Introduction Progress and examples Future developments
Contents Openness Background
Project aims
To build a set of free, open-source tools for the analysis of air pollution data
Use highly developed open-source statistical software ‘R’
Air Pollution Forecasting Seminar
The Open-Source Air Pollution Project
Introduction Progress and examples Future developments
Contents Openness Background
Project aims
To build a set of free, open-source tools for the analysis of air pollution data
Use highly developed open-source statistical software ‘R’ To make it easier to analyse data and to gain insights from it
Air Pollution Forecasting Seminar
The Open-Source Air Pollution Project
Introduction Progress and examples Future developments
Contents Openness Background
Project aims
To build a set of free, open-source tools for the analysis of air pollution data
Use highly developed open-source statistical software ‘R’ To make it easier to analyse data and to gain insights from it Exploit the enormous (and growing) amount of air pollution data available
Air Pollution Forecasting Seminar
The Open-Source Air Pollution Project
Introduction Progress and examples Future developments
Contents Openness Background
Project aims
To build a set of free, open-source tools for the analysis of air pollution data
Use highly developed open-source statistical software ‘R’ To make it easier to analyse data and to gain insights from it Exploit the enormous (and growing) amount of air pollution data available Progressively include advanced approaches generally not widely available
Air Pollution Forecasting Seminar
The Open-Source Air Pollution Project
Introduction Progress and examples Future developments
Contents Openness Background
Project aims
To build a set of free, open-source tools for the analysis of air pollution data
Use highly developed open-source statistical software ‘R’ To make it easier to analyse data and to gain insights from it Exploit the enormous (and growing) amount of air pollution data available Progressively include advanced approaches generally not widely available
Outputs
Air Pollution Forecasting Seminar
The Open-Source Air Pollution Project
Introduction Progress and examples Future developments
Contents Openness Background
Project aims
To build a set of free, open-source tools for the analysis of air pollution data
Use highly developed open-source statistical software ‘R’ To make it easier to analyse data and to gain insights from it Exploit the enormous (and growing) amount of air pollution data available Progressively include advanced approaches generally not widely available
Outputs
An R ‘package’ dedicated to air pollution analysis
Air Pollution Forecasting Seminar
The Open-Source Air Pollution Project
Introduction Progress and examples Future developments
Contents Openness Background
Project aims
To build a set of free, open-source tools for the analysis of air pollution data
Use highly developed open-source statistical software ‘R’ To make it easier to analyse data and to gain insights from it Exploit the enormous (and growing) amount of air pollution data available Progressively include advanced approaches generally not widely available
Outputs
An R ‘package’ dedicated to air pollution analysis A web site to act as a central resource
Air Pollution Forecasting Seminar
The Open-Source Air Pollution Project
Introduction Progress and examples Future developments
Contents Openness Background
Project aims
To build a set of free, open-source tools for the analysis of air pollution data
Use highly developed open-source statistical software ‘R’ To make it easier to analyse data and to gain insights from it Exploit the enormous (and growing) amount of air pollution data available Progressively include advanced approaches generally not widely available
Outputs
An R ‘package’ dedicated to air pollution analysis A web site to act as a central resource Comprehensive documentation
Air Pollution Forecasting Seminar
The Open-Source Air Pollution Project
Introduction Progress and examples Future developments
Contents Openness Background
Project aims
To build a set of free, open-source tools for the analysis of air pollution data
Use highly developed open-source statistical software ‘R’ To make it easier to analyse data and to gain insights from it Exploit the enormous (and growing) amount of air pollution data available Progressively include advanced approaches generally not widely available
Outputs
An R ‘package’ dedicated to air pollution analysis A web site to act as a central resource Comprehensive documentation
Air Pollution Forecasting Seminar
The Open-Source Air Pollution Project
Introduction Progress and examples Future developments
Contents Openness Background
Progress – case studies and user involvement
Sefton Council
Support as part of their ‘Beacon’ status
Air Pollution Forecasting Seminar
The Open-Source Air Pollution Project
Introduction Progress and examples Future developments
Contents Openness Background
Progress – case studies and user involvement
Sefton Council
Support as part of their ‘Beacon’ status Help inform continuing air quality management activities
Air Pollution Forecasting Seminar
The Open-Source Air Pollution Project
Introduction Progress and examples Future developments
Contents Openness Background
Progress – case studies and user involvement
Sefton Council
Support as part of their ‘Beacon’ status Help inform continuing air quality management activities
North Lincolnshire/AEA
Air Pollution Forecasting Seminar
The Open-Source Air Pollution Project
Introduction Progress and examples Future developments
Contents Openness Background
Progress – case studies and user involvement
Sefton Council
Support as part of their ‘Beacon’ status Help inform continuing air quality management activities
North Lincolnshire/AEA
Scunthorpe steel works
Air Pollution Forecasting Seminar
The Open-Source Air Pollution Project
Introduction Progress and examples Future developments
Contents Openness Background
Progress – case studies and user involvement
Sefton Council
Support as part of their ‘Beacon’ status Help inform continuing air quality management activities
North Lincolnshire/AEA
Scunthorpe steel works Highly complex mixture of sources
Air Pollution Forecasting Seminar
The Open-Source Air Pollution Project
Introduction Progress and examples Future developments
Contents Openness Background
Progress – case studies and user involvement
Sefton Council
Support as part of their ‘Beacon’ status Help inform continuing air quality management activities
North Lincolnshire/AEA
Scunthorpe steel works Highly complex mixture of sources Many non-road tra?c sources
Air Pollution Forecasting Seminar
The Open-Source Air Pollution Project
Introduction Progress and examples Future developments
Contents Openness Background
Progress – case studies and user involvement
Sefton Council
Support as part of their ‘Beacon’ status Help inform continuing air quality management activities
North Lincolnshire/AEA
Scunthorpe steel works Highly complex mixture of sources Many non-road tra?c sources
Defra/AEA
Air Pollution Forecasting Seminar
The Open-Source Air Pollution Project
Introduction Progress and examples Future developments
Contents Openness Background
Progress – case studies and user involvement
Sefton Council
Support as part of their ‘Beacon’ status Help inform continuing air quality management activities
North Lincolnshire/AEA
Scunthorpe steel works Highly complex mixture of sources Many non-road tra?c sources
Defra/AEA
Developments to enhance informatics value of AURN data
Air Pollution Forecasting Seminar
The Open-Source Air Pollution Project
Introduction Progress and examples Future developments
Contents Openness Background
Progress – case studies and user involvement
Sefton Council
Support as part of their ‘Beacon’ status Help inform continuing air quality management activities
North Lincolnshire/AEA
Scunthorpe steel works Highly complex mixture of sources Many non-road tra?c sources
Defra/AEA
Developments to enhance informatics value of AURN data
NERC Knowledge Transfer bid
Air Pollution Forecasting Seminar
The Open-Source Air Pollution Project
Introduction Progress and examples Future developments
Contents Openness Background
Progress – case studies and user involvement
Sefton Council
Support as part of their ‘Beacon’ status Help inform continuing air quality management activities
North Lincolnshire/AEA
Scunthorpe steel works Highly complex mixture of sources Many non-road tra?c sources
Defra/AEA
Developments to enhance informatics value of AURN data
NERC Knowledge Transfer bid
Would provide signi?cant funding
Air Pollution Forecasting Seminar
The Open-Source Air Pollution Project
Introduction Progress and examples Future developments
Contents Openness Background
Progress – case studies and user involvement
Sefton Council
Support as part of their ‘Beacon’ status Help inform continuing air quality management activities
North Lincolnshire/AEA
Scunthorpe steel works Highly complex mixture of sources Many non-road tra?c sources
Defra/AEA
Developments to enhance informatics value of AURN data
NERC Knowledge Transfer bid
Would provide signi?cant funding Widely supported
Air Pollution Forecasting Seminar
The Open-Source Air Pollution Project
Introduction Progress and examples Future developments
Contents Openness Background
Progress – case studies and user involvement
Sefton Council
Support as part of their ‘Beacon’ status Help inform continuing air quality management activities
North Lincolnshire/AEA
Scunthorpe steel works Highly complex mixture of sources Many non-road tra?c sources
Defra/AEA
Developments to enhance informatics value of AURN data
NERC Knowledge Transfer bid
Would provide signi?cant funding Widely supported
Air Pollution Forecasting Seminar
The Open-Source Air Pollution Project
Introduction Progress and examples Future developments
Contents Openness Background
What e?ort is required to use these tools?
The project aims to overcome barriers to analysis
Lack of time, money, specialist software, know-how or all four!
Air Pollution Forecasting Seminar
The Open-Source Air Pollution Project
Introduction Progress and examples Future developments
Contents Openness Background
What e?ort is required to use these tools?
The project aims to overcome barriers to analysis
Lack of time, money, specialist software, know-how or all four!
Not required to learn R
Air Pollution Forecasting Seminar
The Open-Source Air Pollution Project
Introduction Progress and examples Future developments
Contents Openness Background
What e?ort is required to use these tools?
The project aims to overcome barriers to analysis
Lack of time, money, specialist software, know-how or all four!
Not required to learn R
Learning R can be hard work
Air Pollution Forecasting Seminar
The Open-Source Air Pollution Project
Introduction Progress and examples Future developments
Contents Openness Background
What e?ort is required to use these tools?
The project aims to overcome barriers to analysis
Lack of time, money, specialist software, know-how or all four!
Not required to learn R
Learning R can be hard work Make things as simple as possible for the user
Air Pollution Forecasting Seminar
The Open-Source Air Pollution Project
Introduction Progress and examples Future developments
Contents Openness Background
What e?ort is required to use these tools?
The project aims to overcome barriers to analysis
Lack of time, money, specialist software, know-how or all four!
Not required to learn R
Learning R can be hard work Make things as simple as possible for the user
Example of code required to make a polar plot polar.plot(mydata, pollutant = "so2")
Some examples of current capabilities to follow . . .
Air Pollution Forecasting Seminar
The Open-Source Air Pollution Project
Introduction Progress and examples Future developments
Contents Openness Background
What e?ort is required to use these tools?
The project aims to overcome barriers to analysis
Lack of time, money, specialist software, know-how or all four!
Not required to learn R
Learning R can be hard work Make things as simple as possible for the user
Example of code required to make a polar plot polar.plot(mydata, pollutant = "so2")
Some examples of current capabilities to follow . . .
Air Pollution Forecasting Seminar
The Open-Source Air Pollution Project
Introduction Progress and examples Future developments
Source detection Displaying lots of information Trends
Diurnal variation in concentrations
50
q q q q
Diurnal and day of week variations can provide clues as to the source Function diurnal.error produces three plots Uncertainty bands can help determine whether one source is di?erent from another
45 40 40
q
conc
35 30 25 20 0 5 10 15 20 25
conc
35
q
q
Sun Mon Tue Wed Thu
Fri
Sat
hour of day
day of week
60
50
conc
40
30
20
Sun
Mon
Tue
Wed
Thu
Fri
Sat
day of week
Air Pollution Forecasting Seminar
The Open-Source Air Pollution Project
Introduction Progress and examples Future developments
Source detection Displaying lots of information Trends
Bivariate polar plots
Sulphur dioxide at Marylebone Rd
N
8
Useful for source detection Methods have been extended to ‘model’ surface concentrations Can usefully be combined with other methods
S W E
7
6
5
4
3
2
1
Air Pollution Forecasting Seminar
The Open-Source Air Pollution Project
Introduction Progress and examples Future developments
Source detection Displaying lots of information Trends
Concentrations by wind direction, year and month
Jan
Apr
Jul
Oct
Jan
Apr
Jul
Oct
Jan
Apr
Jul
Oct
Jan
Apr
Jul
Oct 120 100 80
1998 wind direction (°)
300 200 100
1999
2000
2001
2002
2003
2004
2005
60 40 20 0 Jan Apr Jul Oct Jan Apr Jul Oct Jan Apr Jul Oct Jan Apr Jul Oct
month
Air Pollution Forecasting Seminar
The Open-Source Air Pollution Project
Introduction Progress and examples Future developments
Source detection Displaying lots of information Trends
Maximum hourly ozone concentrations by hour of day, year and month
Jan
Apr
Jul
Oct
Jan
Apr
Jul
Oct
Jan
Apr
Jul
Oct
Jan
Apr
Jul
Oct 70 60
1998
20
1999
2000
2001
2002
2003
2004
2005
50 40 30 20 10
hour
15 10 5 0 Jan Apr Jul Oct Jan Apr Jul Oct Jan Apr Jul Oct Jan Apr Jul Oct
month
Air Pollution Forecasting Seminar
The Open-Source Air Pollution Project
Introduction Progress and examples Future developments
Source detection Displaying lots of information Trends
Diurnal and day of week variation – PMcoarse
0 1998 Monday Monday 30 20 10 5 10 15 20 1999 Monday 2000 Monday 0 5 10 15 20 2001 Monday 2002 Monday 0 5 10 15 20 2003 Monday 2004 Monday 0 5 10 15 20 2005
1998 Tuesday Tuesday
1999 Tuesday
2000 Tuesday
2001 Tuesday
2002 Tuesday
2003 Tuesday
2004 Tuesday
2005 30 20 10 2005 Wednesday
1998 Wednesday Wednesday 30 20 10
1999 Wednesday
2000 Wednesday
2001 Wednesday
2002 Wednesday
2003 Wednesday
2004
coarse particles
1998 Thursday Thursday
1999 Thursday
2000 Thursday
2001 Thursday
2002 Thursday
2003 Thursday
2004 Thursday
2005 30 20 10 2005 Friday
1998 Friday Friday 30 20 10
1999 Friday
2000 Friday
2001 Friday
2002 Friday
2003 Friday
2004
1998 Saturday Saturday
1999 Saturday
2000 Saturday
2001 Saturday
2002 Saturday
2003 Saturday
2004 Saturday
2005 30 20 10 2005 Sunday
1998 Sunday Sunday 30 20 10
1999 Sunday
2000 Sunday
2001 Sunday
2002 Sunday
2003 Sunday
2004
0
5 10 15 20
0
5 10 15 20
0
5 10 15 20
0
5 10 15 20
hour of day
Air Pollution Forecasting Seminar
The Open-Source Air Pollution Project
Introduction Progress and examples Future developments
Source detection Displaying lots of information Trends
Flexible treatment of trends – ozone at Marylebone Road
Trend with smooth ?t and 95 % con?dence intervals
Air Pollution Forecasting Seminar
The Open-Source Air Pollution Project
Introduction Progress and examples Future developments
Source detection Displaying lots of information Trends
Flexible treatment of trends – ozone at Marylebone Road
Trend with smooth ?t and 95 % con?dence intervals Trend with smooth ?t and 95 % con?dence intervals – deseasonalise ?rst
concentration (ppb)
5 1998
10
15
2000
2002 date
2004
2006
2008
Air Pollution Forecasting Seminar
The Open-Source Air Pollution Project
Introduction Progress and examples Future developments
Source detection Displaying lots of information Trends
Flexible treatment of trends – ozone at Marylebone Road
Trend with smooth ?t and 95 % con?dence intervals Trend with smooth ?t and 95 % con?dence intervals – deseasonalise ?rst
Air Pollution Forecasting Seminar
The Open-Source Air Pollution Project
Introduction Progress and examples Future developments
Source detection Displaying lots of information Trends
Flexible treatment of trends – ozone at Marylebone Road
Trend with smooth ?t and 95 % con?dence intervals Trend with smooth ?t and 95 % con?dence intervals – deseasonalise ?rst Consider “no trend” hypothesis through bootstrap resampling
concentration (ppb) 14 4 1998 6 8 10 12
2000
2002 date
2004
2006
2008
Air Pollution Forecasting Seminar
The Open-Source Air Pollution Project
Introduction Progress and examples Future developments
Source detection Displaying lots of information Trends
Flexible treatment of trends – ozone at Marylebone Road
Trend with smooth ?t and 95 % con?dence intervals Trend with smooth ?t and 95 % con?dence intervals – deseasonalise ?rst Consider “no trend” hypothesis through bootstrap resampling
Air Pollution Forecasting Seminar
The Open-Source Air Pollution Project
Introduction Progress and examples Future developments
Source detection Displaying lots of information Trends
Flexible treatment of trends – ozone at Marylebone Road
Trend with smooth ?t and 95 % con?dence intervals Trend with smooth ?t and 95 % con?dence intervals – deseasonalise ?rst Consider “no trend” hypothesis through bootstrap resampling
concentration (ppb) 14
4
6
8
10
12
Measurements No trend hypothesis
1998
2000
2002 date
2004
2006
2008
Air Pollution Forecasting Seminar
The Open-Source Air Pollution Project
Introduction Progress and examples Future developments
Documentation support Concluding remarks
Documentation
Documentation has been started
Part 1: Introduction to using R to analyse monitoring data Part II: Dedicated functions to analyse monitoring data
Longer term aims
Develop a Framework for analysing data Case studies spanning a range of contemporary problems
Air Pollution Forecasting Seminar The Open-Source Air Pollution Project
Introduction Progress and examples Future developments
Documentation support Concluding remarks
Longer-term aims
The big picture
Change the way we do things
Air Pollution Forecasting Seminar
The Open-Source Air Pollution Project
Introduction Progress and examples Future developments
Documentation support Concluding remarks
Longer-term aims
The big picture
Change the way we do things Environmental models?
Air Pollution Forecasting Seminar
The Open-Source Air Pollution Project
Introduction Progress and examples Future developments
Documentation support Concluding remarks
Longer-term aims
The big picture
Change the way we do things Environmental models? Work that is truly reproducible
Air Pollution Forecasting Seminar
The Open-Source Air Pollution Project
Introduction Progress and examples Future developments
Documentation support Concluding remarks
Longer-term aims
The big picture
Change the way we do things Environmental models? Work that is truly reproducible
Wider issues
Air Pollution Forecasting Seminar
The Open-Source Air Pollution Project
Introduction Progress and examples Future developments
Documentation support Concluding remarks
Longer-term aims
The big picture
Change the way we do things Environmental models? Work that is truly reproducible
Wider issues
Developing countries
Air Pollution Forecasting Seminar
The Open-Source Air Pollution Project
Introduction Progress and examples Future developments
Documentation support Concluding remarks
Longer-term aims
The big picture
Change the way we do things Environmental models? Work that is truly reproducible
Wider issues
Developing countries Actively seek participation of researchers elsewhere in the world
Air Pollution Forecasting Seminar
The Open-Source Air Pollution Project
Introduction Progress and examples Future developments
Documentation support Concluding remarks
Longer-term aims
The big picture
Change the way we do things Environmental models? Work that is truly reproducible
Wider issues
Developing countries Actively seek participation of researchers elsewhere in the world
Air Pollution Forecasting Seminar
The Open-Source Air Pollution Project
Introduction Progress and examples Future developments
Documentation support Concluding remarks
Longer-term aims
The big picture
Change the way we do things Environmental models? Work that is truly reproducible
Wider issues
Developing countries Actively seek participation of researchers elsewhere in the world
Air Pollution Forecasting Seminar
The Open-Source Air Pollution Project
Introduction Progress and examples Future developments
Documentation support Concluding remarks
Thank you for your attention!
OPEN
Air Pollution Forecasting Seminar
source
The Open-Source Air Pollution Project
doc_425221775.pdf
Air pollution is the introduction into the atmosphere of chemicals, particulates, or biological materials that cause discomfort, disease, or death to humans, damage other living organisms such as food crops, or damage the natural environment or built environment.
Introduction Progress and examples Future developments
The Open-Source Air Pollution Project
Community tools for analysing air pollution data
David Carslaw
Institute for Transport Studies, University of Leeds
14 May 2008
Air Pollution Forecasting Seminar
The Open-Source Air Pollution Project
Introduction Progress and examples Future developments
Contents Openness Background
Summary
Background to the project Progress and examples Future directions and developments
Air Pollution Forecasting Seminar
The Open-Source Air Pollution Project
Introduction Progress and examples Future developments
Contents Openness Background
The bene?ts of openness
Transparency should be at the heart of environmental regulation
Those a?ected by environmental decisions should be able to scrutinise the tools that lead to those decisions
Air Pollution Forecasting Seminar
The Open-Source Air Pollution Project
Introduction Progress and examples Future developments
Contents Openness Background
The bene?ts of openness
Transparency should be at the heart of environmental regulation
Those a?ected by environmental decisions should be able to scrutinise the tools that lead to those decisions Many examples from the USA
Air Pollution Forecasting Seminar
The Open-Source Air Pollution Project
Introduction Progress and examples Future developments
Contents Openness Background
The bene?ts of openness
Transparency should be at the heart of environmental regulation
Those a?ected by environmental decisions should be able to scrutinise the tools that lead to those decisions Many examples from the USA
Open-source software
Air Pollution Forecasting Seminar
The Open-Source Air Pollution Project
Introduction Progress and examples Future developments
Contents Openness Background
The bene?ts of openness
Transparency should be at the heart of environmental regulation
Those a?ected by environmental decisions should be able to scrutinise the tools that lead to those decisions Many examples from the USA
Open-source software
All source code made available, free and can be modi?ed by anyone
Air Pollution Forecasting Seminar
The Open-Source Air Pollution Project
Introduction Progress and examples Future developments
Contents Openness Background
The bene?ts of openness
Transparency should be at the heart of environmental regulation
Those a?ected by environmental decisions should be able to scrutinise the tools that lead to those decisions Many examples from the USA
Open-source software
All source code made available, free and can be modi?ed by anyone No longer in the realm of enthusiasts e.g. Linux, MySQL, R
Air Pollution Forecasting Seminar
The Open-Source Air Pollution Project
Introduction Progress and examples Future developments
Contents Openness Background
The bene?ts of openness
Transparency should be at the heart of environmental regulation
Those a?ected by environmental decisions should be able to scrutinise the tools that lead to those decisions Many examples from the USA
Open-source software
All source code made available, free and can be modi?ed by anyone No longer in the realm of enthusiasts e.g. Linux, MySQL, R
Both promote participation and ownership
Air Pollution Forecasting Seminar
The Open-Source Air Pollution Project
Introduction Progress and examples Future developments
Contents Openness Background
The bene?ts of openness
Transparency should be at the heart of environmental regulation
Those a?ected by environmental decisions should be able to scrutinise the tools that lead to those decisions Many examples from the USA
Open-source software
All source code made available, free and can be modi?ed by anyone No longer in the realm of enthusiasts e.g. Linux, MySQL, R
Both promote participation and ownership
Air Pollution Forecasting Seminar
The Open-Source Air Pollution Project
Introduction Progress and examples Future developments
Contents Openness Background
Interesting aside – climate change and the “hockey stick”
Controversy over IPCC “hockey stick” temperature grapha Statistical methods shown to be ?awed (McIntyre and McKitrick) Code and data made available – some in R
Mann, M.E. et al. (1998). Global-scale temperature patterns and climate forcing over the past six centuries. Nature, Vol. 392, pp. 779787.
a
Air Pollution Forecasting Seminar
The Open-Source Air Pollution Project
Introduction Progress and examples Future developments
Contents Openness Background
Project aims
To build a set of free, open-source tools for the analysis of air pollution data
Use highly developed open-source statistical software ‘R’
Air Pollution Forecasting Seminar
The Open-Source Air Pollution Project
Introduction Progress and examples Future developments
Contents Openness Background
Project aims
To build a set of free, open-source tools for the analysis of air pollution data
Use highly developed open-source statistical software ‘R’ To make it easier to analyse data and to gain insights from it
Air Pollution Forecasting Seminar
The Open-Source Air Pollution Project
Introduction Progress and examples Future developments
Contents Openness Background
Project aims
To build a set of free, open-source tools for the analysis of air pollution data
Use highly developed open-source statistical software ‘R’ To make it easier to analyse data and to gain insights from it Exploit the enormous (and growing) amount of air pollution data available
Air Pollution Forecasting Seminar
The Open-Source Air Pollution Project
Introduction Progress and examples Future developments
Contents Openness Background
Project aims
To build a set of free, open-source tools for the analysis of air pollution data
Use highly developed open-source statistical software ‘R’ To make it easier to analyse data and to gain insights from it Exploit the enormous (and growing) amount of air pollution data available Progressively include advanced approaches generally not widely available
Air Pollution Forecasting Seminar
The Open-Source Air Pollution Project
Introduction Progress and examples Future developments
Contents Openness Background
Project aims
To build a set of free, open-source tools for the analysis of air pollution data
Use highly developed open-source statistical software ‘R’ To make it easier to analyse data and to gain insights from it Exploit the enormous (and growing) amount of air pollution data available Progressively include advanced approaches generally not widely available
Outputs
Air Pollution Forecasting Seminar
The Open-Source Air Pollution Project
Introduction Progress and examples Future developments
Contents Openness Background
Project aims
To build a set of free, open-source tools for the analysis of air pollution data
Use highly developed open-source statistical software ‘R’ To make it easier to analyse data and to gain insights from it Exploit the enormous (and growing) amount of air pollution data available Progressively include advanced approaches generally not widely available
Outputs
An R ‘package’ dedicated to air pollution analysis
Air Pollution Forecasting Seminar
The Open-Source Air Pollution Project
Introduction Progress and examples Future developments
Contents Openness Background
Project aims
To build a set of free, open-source tools for the analysis of air pollution data
Use highly developed open-source statistical software ‘R’ To make it easier to analyse data and to gain insights from it Exploit the enormous (and growing) amount of air pollution data available Progressively include advanced approaches generally not widely available
Outputs
An R ‘package’ dedicated to air pollution analysis A web site to act as a central resource
Air Pollution Forecasting Seminar
The Open-Source Air Pollution Project
Introduction Progress and examples Future developments
Contents Openness Background
Project aims
To build a set of free, open-source tools for the analysis of air pollution data
Use highly developed open-source statistical software ‘R’ To make it easier to analyse data and to gain insights from it Exploit the enormous (and growing) amount of air pollution data available Progressively include advanced approaches generally not widely available
Outputs
An R ‘package’ dedicated to air pollution analysis A web site to act as a central resource Comprehensive documentation
Air Pollution Forecasting Seminar
The Open-Source Air Pollution Project
Introduction Progress and examples Future developments
Contents Openness Background
Project aims
To build a set of free, open-source tools for the analysis of air pollution data
Use highly developed open-source statistical software ‘R’ To make it easier to analyse data and to gain insights from it Exploit the enormous (and growing) amount of air pollution data available Progressively include advanced approaches generally not widely available
Outputs
An R ‘package’ dedicated to air pollution analysis A web site to act as a central resource Comprehensive documentation
Air Pollution Forecasting Seminar
The Open-Source Air Pollution Project
Introduction Progress and examples Future developments
Contents Openness Background
Progress – case studies and user involvement
Sefton Council
Support as part of their ‘Beacon’ status
Air Pollution Forecasting Seminar
The Open-Source Air Pollution Project
Introduction Progress and examples Future developments
Contents Openness Background
Progress – case studies and user involvement
Sefton Council
Support as part of their ‘Beacon’ status Help inform continuing air quality management activities
Air Pollution Forecasting Seminar
The Open-Source Air Pollution Project
Introduction Progress and examples Future developments
Contents Openness Background
Progress – case studies and user involvement
Sefton Council
Support as part of their ‘Beacon’ status Help inform continuing air quality management activities
North Lincolnshire/AEA
Air Pollution Forecasting Seminar
The Open-Source Air Pollution Project
Introduction Progress and examples Future developments
Contents Openness Background
Progress – case studies and user involvement
Sefton Council
Support as part of their ‘Beacon’ status Help inform continuing air quality management activities
North Lincolnshire/AEA
Scunthorpe steel works
Air Pollution Forecasting Seminar
The Open-Source Air Pollution Project
Introduction Progress and examples Future developments
Contents Openness Background
Progress – case studies and user involvement
Sefton Council
Support as part of their ‘Beacon’ status Help inform continuing air quality management activities
North Lincolnshire/AEA
Scunthorpe steel works Highly complex mixture of sources
Air Pollution Forecasting Seminar
The Open-Source Air Pollution Project
Introduction Progress and examples Future developments
Contents Openness Background
Progress – case studies and user involvement
Sefton Council
Support as part of their ‘Beacon’ status Help inform continuing air quality management activities
North Lincolnshire/AEA
Scunthorpe steel works Highly complex mixture of sources Many non-road tra?c sources
Air Pollution Forecasting Seminar
The Open-Source Air Pollution Project
Introduction Progress and examples Future developments
Contents Openness Background
Progress – case studies and user involvement
Sefton Council
Support as part of their ‘Beacon’ status Help inform continuing air quality management activities
North Lincolnshire/AEA
Scunthorpe steel works Highly complex mixture of sources Many non-road tra?c sources
Defra/AEA
Air Pollution Forecasting Seminar
The Open-Source Air Pollution Project
Introduction Progress and examples Future developments
Contents Openness Background
Progress – case studies and user involvement
Sefton Council
Support as part of their ‘Beacon’ status Help inform continuing air quality management activities
North Lincolnshire/AEA
Scunthorpe steel works Highly complex mixture of sources Many non-road tra?c sources
Defra/AEA
Developments to enhance informatics value of AURN data
Air Pollution Forecasting Seminar
The Open-Source Air Pollution Project
Introduction Progress and examples Future developments
Contents Openness Background
Progress – case studies and user involvement
Sefton Council
Support as part of their ‘Beacon’ status Help inform continuing air quality management activities
North Lincolnshire/AEA
Scunthorpe steel works Highly complex mixture of sources Many non-road tra?c sources
Defra/AEA
Developments to enhance informatics value of AURN data
NERC Knowledge Transfer bid
Air Pollution Forecasting Seminar
The Open-Source Air Pollution Project
Introduction Progress and examples Future developments
Contents Openness Background
Progress – case studies and user involvement
Sefton Council
Support as part of their ‘Beacon’ status Help inform continuing air quality management activities
North Lincolnshire/AEA
Scunthorpe steel works Highly complex mixture of sources Many non-road tra?c sources
Defra/AEA
Developments to enhance informatics value of AURN data
NERC Knowledge Transfer bid
Would provide signi?cant funding
Air Pollution Forecasting Seminar
The Open-Source Air Pollution Project
Introduction Progress and examples Future developments
Contents Openness Background
Progress – case studies and user involvement
Sefton Council
Support as part of their ‘Beacon’ status Help inform continuing air quality management activities
North Lincolnshire/AEA
Scunthorpe steel works Highly complex mixture of sources Many non-road tra?c sources
Defra/AEA
Developments to enhance informatics value of AURN data
NERC Knowledge Transfer bid
Would provide signi?cant funding Widely supported
Air Pollution Forecasting Seminar
The Open-Source Air Pollution Project
Introduction Progress and examples Future developments
Contents Openness Background
Progress – case studies and user involvement
Sefton Council
Support as part of their ‘Beacon’ status Help inform continuing air quality management activities
North Lincolnshire/AEA
Scunthorpe steel works Highly complex mixture of sources Many non-road tra?c sources
Defra/AEA
Developments to enhance informatics value of AURN data
NERC Knowledge Transfer bid
Would provide signi?cant funding Widely supported
Air Pollution Forecasting Seminar
The Open-Source Air Pollution Project
Introduction Progress and examples Future developments
Contents Openness Background
What e?ort is required to use these tools?
The project aims to overcome barriers to analysis
Lack of time, money, specialist software, know-how or all four!
Air Pollution Forecasting Seminar
The Open-Source Air Pollution Project
Introduction Progress and examples Future developments
Contents Openness Background
What e?ort is required to use these tools?
The project aims to overcome barriers to analysis
Lack of time, money, specialist software, know-how or all four!
Not required to learn R
Air Pollution Forecasting Seminar
The Open-Source Air Pollution Project
Introduction Progress and examples Future developments
Contents Openness Background
What e?ort is required to use these tools?
The project aims to overcome barriers to analysis
Lack of time, money, specialist software, know-how or all four!
Not required to learn R
Learning R can be hard work
Air Pollution Forecasting Seminar
The Open-Source Air Pollution Project
Introduction Progress and examples Future developments
Contents Openness Background
What e?ort is required to use these tools?
The project aims to overcome barriers to analysis
Lack of time, money, specialist software, know-how or all four!
Not required to learn R
Learning R can be hard work Make things as simple as possible for the user
Air Pollution Forecasting Seminar
The Open-Source Air Pollution Project
Introduction Progress and examples Future developments
Contents Openness Background
What e?ort is required to use these tools?
The project aims to overcome barriers to analysis
Lack of time, money, specialist software, know-how or all four!
Not required to learn R
Learning R can be hard work Make things as simple as possible for the user
Example of code required to make a polar plot polar.plot(mydata, pollutant = "so2")
Some examples of current capabilities to follow . . .
Air Pollution Forecasting Seminar
The Open-Source Air Pollution Project
Introduction Progress and examples Future developments
Contents Openness Background
What e?ort is required to use these tools?
The project aims to overcome barriers to analysis
Lack of time, money, specialist software, know-how or all four!
Not required to learn R
Learning R can be hard work Make things as simple as possible for the user
Example of code required to make a polar plot polar.plot(mydata, pollutant = "so2")
Some examples of current capabilities to follow . . .
Air Pollution Forecasting Seminar
The Open-Source Air Pollution Project
Introduction Progress and examples Future developments
Source detection Displaying lots of information Trends
Diurnal variation in concentrations
50
q q q q
Diurnal and day of week variations can provide clues as to the source Function diurnal.error produces three plots Uncertainty bands can help determine whether one source is di?erent from another
45 40 40
q
conc
35 30 25 20 0 5 10 15 20 25
conc
35
q
q
Sun Mon Tue Wed Thu
Fri
Sat
hour of day
day of week
60
50
conc
40
30
20
Sun
Mon
Tue
Wed
Thu
Fri
Sat
day of week
Air Pollution Forecasting Seminar
The Open-Source Air Pollution Project
Introduction Progress and examples Future developments
Source detection Displaying lots of information Trends
Bivariate polar plots
Sulphur dioxide at Marylebone Rd
N
8
Useful for source detection Methods have been extended to ‘model’ surface concentrations Can usefully be combined with other methods
S W E
7
6
5
4
3
2
1
Air Pollution Forecasting Seminar
The Open-Source Air Pollution Project
Introduction Progress and examples Future developments
Source detection Displaying lots of information Trends
Concentrations by wind direction, year and month
Jan
Apr
Jul
Oct
Jan
Apr
Jul
Oct
Jan
Apr
Jul
Oct
Jan
Apr
Jul
Oct 120 100 80
1998 wind direction (°)
300 200 100
1999
2000
2001
2002
2003
2004
2005
60 40 20 0 Jan Apr Jul Oct Jan Apr Jul Oct Jan Apr Jul Oct Jan Apr Jul Oct
month
Air Pollution Forecasting Seminar
The Open-Source Air Pollution Project
Introduction Progress and examples Future developments
Source detection Displaying lots of information Trends
Maximum hourly ozone concentrations by hour of day, year and month
Jan
Apr
Jul
Oct
Jan
Apr
Jul
Oct
Jan
Apr
Jul
Oct
Jan
Apr
Jul
Oct 70 60
1998
20
1999
2000
2001
2002
2003
2004
2005
50 40 30 20 10
hour
15 10 5 0 Jan Apr Jul Oct Jan Apr Jul Oct Jan Apr Jul Oct Jan Apr Jul Oct
month
Air Pollution Forecasting Seminar
The Open-Source Air Pollution Project
Introduction Progress and examples Future developments
Source detection Displaying lots of information Trends
Diurnal and day of week variation – PMcoarse
0 1998 Monday Monday 30 20 10 5 10 15 20 1999 Monday 2000 Monday 0 5 10 15 20 2001 Monday 2002 Monday 0 5 10 15 20 2003 Monday 2004 Monday 0 5 10 15 20 2005
1998 Tuesday Tuesday
1999 Tuesday
2000 Tuesday
2001 Tuesday
2002 Tuesday
2003 Tuesday
2004 Tuesday
2005 30 20 10 2005 Wednesday
1998 Wednesday Wednesday 30 20 10
1999 Wednesday
2000 Wednesday
2001 Wednesday
2002 Wednesday
2003 Wednesday
2004
coarse particles
1998 Thursday Thursday
1999 Thursday
2000 Thursday
2001 Thursday
2002 Thursday
2003 Thursday
2004 Thursday
2005 30 20 10 2005 Friday
1998 Friday Friday 30 20 10
1999 Friday
2000 Friday
2001 Friday
2002 Friday
2003 Friday
2004
1998 Saturday Saturday
1999 Saturday
2000 Saturday
2001 Saturday
2002 Saturday
2003 Saturday
2004 Saturday
2005 30 20 10 2005 Sunday
1998 Sunday Sunday 30 20 10
1999 Sunday
2000 Sunday
2001 Sunday
2002 Sunday
2003 Sunday
2004
0
5 10 15 20
0
5 10 15 20
0
5 10 15 20
0
5 10 15 20
hour of day
Air Pollution Forecasting Seminar
The Open-Source Air Pollution Project
Introduction Progress and examples Future developments
Source detection Displaying lots of information Trends
Flexible treatment of trends – ozone at Marylebone Road
Trend with smooth ?t and 95 % con?dence intervals
Air Pollution Forecasting Seminar
The Open-Source Air Pollution Project
Introduction Progress and examples Future developments
Source detection Displaying lots of information Trends
Flexible treatment of trends – ozone at Marylebone Road
Trend with smooth ?t and 95 % con?dence intervals Trend with smooth ?t and 95 % con?dence intervals – deseasonalise ?rst
concentration (ppb)
5 1998
10
15
2000
2002 date
2004
2006
2008
Air Pollution Forecasting Seminar
The Open-Source Air Pollution Project
Introduction Progress and examples Future developments
Source detection Displaying lots of information Trends
Flexible treatment of trends – ozone at Marylebone Road
Trend with smooth ?t and 95 % con?dence intervals Trend with smooth ?t and 95 % con?dence intervals – deseasonalise ?rst
Air Pollution Forecasting Seminar
The Open-Source Air Pollution Project
Introduction Progress and examples Future developments
Source detection Displaying lots of information Trends
Flexible treatment of trends – ozone at Marylebone Road
Trend with smooth ?t and 95 % con?dence intervals Trend with smooth ?t and 95 % con?dence intervals – deseasonalise ?rst Consider “no trend” hypothesis through bootstrap resampling
concentration (ppb) 14 4 1998 6 8 10 12
2000
2002 date
2004
2006
2008
Air Pollution Forecasting Seminar
The Open-Source Air Pollution Project
Introduction Progress and examples Future developments
Source detection Displaying lots of information Trends
Flexible treatment of trends – ozone at Marylebone Road
Trend with smooth ?t and 95 % con?dence intervals Trend with smooth ?t and 95 % con?dence intervals – deseasonalise ?rst Consider “no trend” hypothesis through bootstrap resampling
Air Pollution Forecasting Seminar
The Open-Source Air Pollution Project
Introduction Progress and examples Future developments
Source detection Displaying lots of information Trends
Flexible treatment of trends – ozone at Marylebone Road
Trend with smooth ?t and 95 % con?dence intervals Trend with smooth ?t and 95 % con?dence intervals – deseasonalise ?rst Consider “no trend” hypothesis through bootstrap resampling
concentration (ppb) 14
4
6
8
10
12
Measurements No trend hypothesis
1998
2000
2002 date
2004
2006
2008
Air Pollution Forecasting Seminar
The Open-Source Air Pollution Project
Introduction Progress and examples Future developments
Documentation support Concluding remarks
Documentation
Documentation has been started
Part 1: Introduction to using R to analyse monitoring data Part II: Dedicated functions to analyse monitoring data
Longer term aims
Develop a Framework for analysing data Case studies spanning a range of contemporary problems
Air Pollution Forecasting Seminar The Open-Source Air Pollution Project
Introduction Progress and examples Future developments
Documentation support Concluding remarks
Longer-term aims
The big picture
Change the way we do things
Air Pollution Forecasting Seminar
The Open-Source Air Pollution Project
Introduction Progress and examples Future developments
Documentation support Concluding remarks
Longer-term aims
The big picture
Change the way we do things Environmental models?
Air Pollution Forecasting Seminar
The Open-Source Air Pollution Project
Introduction Progress and examples Future developments
Documentation support Concluding remarks
Longer-term aims
The big picture
Change the way we do things Environmental models? Work that is truly reproducible
Air Pollution Forecasting Seminar
The Open-Source Air Pollution Project
Introduction Progress and examples Future developments
Documentation support Concluding remarks
Longer-term aims
The big picture
Change the way we do things Environmental models? Work that is truly reproducible
Wider issues
Air Pollution Forecasting Seminar
The Open-Source Air Pollution Project
Introduction Progress and examples Future developments
Documentation support Concluding remarks
Longer-term aims
The big picture
Change the way we do things Environmental models? Work that is truly reproducible
Wider issues
Developing countries
Air Pollution Forecasting Seminar
The Open-Source Air Pollution Project
Introduction Progress and examples Future developments
Documentation support Concluding remarks
Longer-term aims
The big picture
Change the way we do things Environmental models? Work that is truly reproducible
Wider issues
Developing countries Actively seek participation of researchers elsewhere in the world
Air Pollution Forecasting Seminar
The Open-Source Air Pollution Project
Introduction Progress and examples Future developments
Documentation support Concluding remarks
Longer-term aims
The big picture
Change the way we do things Environmental models? Work that is truly reproducible
Wider issues
Developing countries Actively seek participation of researchers elsewhere in the world
Air Pollution Forecasting Seminar
The Open-Source Air Pollution Project
Introduction Progress and examples Future developments
Documentation support Concluding remarks
Longer-term aims
The big picture
Change the way we do things Environmental models? Work that is truly reproducible
Wider issues
Developing countries Actively seek participation of researchers elsewhere in the world
Air Pollution Forecasting Seminar
The Open-Source Air Pollution Project
Introduction Progress and examples Future developments
Documentation support Concluding remarks
Thank you for your attention!
OPEN
Air Pollution Forecasting Seminar
source
The Open-Source Air Pollution Project
doc_425221775.pdf