Beijing Air Quality – Citizen-Science Approach to Mapping Levels?

A recent article in Environmental Technology Online reports on a community-based science project called ‘Float’ that is actually part-science and part-art project. The idea is that pollution-sensitive kites will be flown over Beijing. These kites contain Arduino pollution-sensing modules and LED lights and will indicate levels of volatile organic compounds, carbon monoxide and particulate matter by changing colour to green, yellow or red depending on the pollutant levels. The kites are attached to GPS device loggers and the real-time data website Cosm.

The project is designed by students Xiaowei Wang from Harvard’s Graduate School of Design and Deren Guler from Carneige Mellon and is designed to involve local residents in data collection. The project relies on public funding and is still raising funds. The project derives its funds from Kickstarter, a website devoted to creative projects and obtaining funding for such projects (The Float project on Kickstart). The project also has funding from the Black Rock Arts Foundation and the Awesome Foundation.

The project has generated a lot of interest on the Web:

Fighting China’s Pollution Propaganda, with Glowing Robot Kites For the People

Pollution-detecting kites to monitor Beijing's air quality
Glowing Pollution Sensor Equipped Kites Replace Beijing's Stars
Kickstarter Project Plans to Measure Beijing Pollution Using Kite Sensors

Only a couple of comments and an expression of interest in the results really.


The project is undoubtedly part of the growing and, in my view, superb trend towards more inclusive community or participatory science (choose whichever term you prefer, Guler uses citizen-science). The ideal of getting local communities involved in the data collection as well as involving them in all aspects of the research process is an excellent way to raise awareness of an issue as well as educate people about the scientific approach and its problems and potentials. The Float project has involved local communities, young and old, from the start with workshops in Beijing and as well as in the design of the kites. In terms of how to organise a community-based, participatory science project it is one that I will advice my students to look at. It is just a shame that the descriptions of the project veer from highlighting the science to highlighting the arts aspects as if the two are, or need to be, distinct. It should also be remembered that this project, as any project involved in monitoring pollution, is entering the political as well as the scientific arena. Involving local populations is a political act (as is their agreement to involvement) as much as the monitoring of pollution by the American Embassy or the siting of monitoring sites by the Chinese. Local is as political as the national or international, but the nature of the act does not necessarily mean the data is political bias only that data collection is for a purpose.

As with most community-based projects, however, there is the issue of belief, trust or confidence in the data collected. These projects do tend to illustrate quite nicely the continuing divide between the ‘specialist’ or ‘expert’ and the ‘public’ (I would say amateur, but much of British science in the nineteenth and early twentieth century only developed because of amateurs!) The expert has been trained and accepts certain methods as being appropriate for data collection. Control and standardization are essential in ensuring what is termed ‘intersubjectivity communication’ between researchers – basically it means  I know what you did because that is how I was trained to do it, so I trust your data as being real. Guler seems to downgrade the status of the data collected even before the project really begins by stating:

‘We’re trying to interact with people on the street and see what they’re tying to do with the information they see. I don’t plan to argue that this is the most accurate data because there are many potential reasons for differences in air quality reports. We want to just keep it up, upload the data, and focus on that more after we come back’.

My impression is this statement is a great get-out clause for ‘official’ monitoring be it by the Chinese or atop the American Embassy. I wouldn’t’ be so pessimistic. The aims of the project in terms of improving public understanding of air pollution, its impact on health and the visualization of pollution through the kites are all excellent and likely to be successful. The data collected is also of value. The ‘official’ pollution monitoring sites probably conform to national or international standards for static sites in terms of equipment and monitoring periods. The kite data does not necessarily provide comparable data to these sites. The kites are mobile and collect data on levels that can be spatially references (I assume in 4 dimensions). They provide a different perspective on atmospheric pollution rather as a spatially altering phenomenon, something the official monitoring sites can not provide.  It could even be argued that the kite data provides information on pollution as experienced by the population (although the population is unlikely to move across the sky at the height of the kites!) The important thing to remember is that there is not one, single correct measure of atmospheric pollution; there are merely different representations of atmospheric pollution. The official static sites have the advantage of having clearly defined protocols that ensure the data or information they collect is immediately comparable with data or information collected at similar monitoring sites globally. The Float project is generating a different and novel set of data or information. This may require a different approach to thinking about the information and its interpretation (Guler seems to suggest this with some hints at triangulation of trends) and in how confidence or belief in the information is assessed either qualitatively or quantitatively. I will be very interested to see what form the results and interpretation takes. Good luck with the project!

Surface Water Flows and Flooding

The recent ASC report on flooding and water scarcity makes some interesting points about surface water flows, particularly those associated with flood hazard in urban areas. The report states that ‘Every millimetre of rainfall deposits a litre of water on a square metre of land’. Water falling onto a paved, impermeable will not infiltrate into the ground and so the volume has to move somewhere. The amount of paved surface is increasing as the report noted that green spaces in urban areas have been paved over and so surface water flows in urban areas are increasing even before the more intense rainfall associated with climate change is considered. The figures cited are that the proportion of paved gardens has increased from 20%in 2001 to 48% in 2011 of the total garden area of 340,000 hectares.

To combat this increase in paved area contributing to runoff the report suggest that urban creep should be minimized, sustainable urban drainage (SUDS) (SUDS) (SUDS-EA) should be improve to slow down water flows and store water above ground, and that conventional sewers should be maintained or upgraded: all good ideas. Recent floods in urban areas have highlighted the importance of such measures. Paved surfaces permit no storage of water, runoff is almost immediate and, with intense rainfall, the volumes of runoff involved can be huge within a short period of time. Overwhelmed urban drainage systems mean that the water moves rapidly across impermeable surfaces and flows through streets and roadways using them like predefined river channels. Similarly, when a river bursts its banks the water tends to use the paved, impermeable surface as a routeway for movement. The urban road network provides a convenient substitute for natural channels providing water with a rapid means of moving across an urban area.
A great deal of the potential damage from a flood and even flash floods could be mapped using a detailed digital elevation model (DEM) and a knowledge of past events in an urban area. This will help map out previous routeways that surface flows have used. Future events may be harder to predict, as the urban infrastructure changes and precautions are taken by planners to block or re-route surface flow, then the microtopogaphy of the urban area may be a guide to patterns of surface flow but other factors will also affect the detailed routes the water takes. The local detail is a bugger for modelling flow patterns. It will be interesting to see what, if any, use is made of the information about flood damage from the recent floods. There is a great deal of information online from Twitter, as well as local blogs and newspapers accounts that could provide a great deal of information about how surface water moved through urban areas. The potential for ‘citizen science’, for ordinary people (a horrible term that seems to imply scientists and planners are extraordinary) to contribute to the scientific investigation of flooding is immense. Co-ordination of this type of information, the mere exercise of collecting and collating information, or judging its quality and usefulness fro modelling and understanding urban surface flow is immense. Time, expertise and, potential funds, are needed for these activities but by who is unclear. Once the aftermath of the floods disappears from public view, the chances of funding such work drops dramatically. The need for people, the public (rather than the ordinary – anyone got a better term that isn’t condescending?) to be involved is important, however, if some of the recommendations of the ASC report are put into practice. In particular, the emphasis on households undertaking property-level flood protection measures might be enhanced if they were also actively involved in monitoring and in the feedback loop from modelling studies of their local areas. This would not only mean they were better informed about the risks of flooding but also more likely to act in the manner hoped for by planners if they felt they were an active part of preventing flood damage rather than passive victims in urban flooding.

COMMUNITIES AND ENVIROMENTAL GEOGRAPHY

ENVIRONMENTAL GEOGRAPHY THAT MATTERS

I outlined my initial view of environmental geography a couple of blogs ago. Since then I have been looking around for something that could clarify, expand and explain my view with a little more clarity and depth. I hope that my outline of the different approaches to studying hazards is beginning to show how environmental geography can be relevant.

I have been, however, loking for something that would serve as a reference for dicsussion; something that might need expansion and correction from time to time but one which readers of the blog might like to mull over and consider. A useful starting point might be the quote below taken from a book by Bent Flyvbjerg. I have just replaced the words ‘social science’ with the words ‘environmental geography’.

‘.. we must take up problems that matter to the local, national, and global communities in which we live, and we must do it in ways that matter; we must focus on issues of values and power like great social scientists have advocated ….. Finally, we must effectively communicate the results of our research to fellow citizens. If we do this we may successfully transform [environmental geography] from what is fast becoming a sterile academic activity, which is undertaken mostly for its own sake and in increasing isolation from a society on which it has little effect and from which it gets little appreciation. We may transform [environmental geography] to an activity done in public for the public, sometimes to clarify, sometimes to intervene, sometimes to generate new perspectives, and always to serve as eyes and ears in our ongoing efforts at understanding the present and deliberating about the future.’

(Bent Flyvbjerg, 2001, Making social science matter – why social science inquiry fails and how it can succeed again. Cambridge University Press, Cambridge. p.166.)

Bent Flyvbjerg is professor at the University of Oxford, in the Said Business School (http://www.sbs.ox.ac.uk/research/people/Pages/BentFlyvbjerg.aspx). He has lead a debate calling for a rejection of the natural science model of research in the social sciences and making social sciences more relevant to people outside science such as citizens and policy makers. He has developed the phronetic approach to social sciences, i.e. studying of social phenomena with a focus on power and values. This approach asks four specific questions:

1. Where are we going?
2. Is this development desirable?
3. Who gains and who loses, and by which mechanisms of power?
4. What, if anything, should we do about it?

(see Wikipiedia for more details: http://en.wikipedia.org/wiki/Phronetic_social_science)

Whilst Flyvbjerg focuses on a idealised model of how physical science is done culminating in a predictive model of reality that is not necessarily how mdoern science with one eye on complexity views or understands reality, he does make an interesting point that predictability as understood in the natural sciences may not be achievable in the social sciences. The application of a model with relaible equations or laws may not be that useful in trying to predict human behaviour or in answering questions of what ought to be, of what is fair, questions of value and judgement that natural science, in the view of many social scientists, has trouble with.

My own view is that the physical sciences (for want of a better term) ask important, but different types of questions of the environment than social sciences so it is not a surprise that different types of answers are produced by each type of study. What the above quote does emphasis is that study for its own sake will produce a sterile subject. Although environmental geography has not wandered down this cul-de-sac yet, it is vital that it is practised and practised in a relevant context for it to develop and to provide communities with the perspective and power to improve their circumstances. In other words environmental geography must be relevant.

So what would a relevant environmental geography look like? Could it square the circle of incorporating both natural and social science? Could it inform and empower communities? A possible example of this type of environmental geography is provided by the South Durban Environmental Alliance (http://www.sdcea.co.za/). This is a community based organization, active since 1996, (an umbrella for 14 affiliate organizations) that lobbies, reports and researches industrial incidents in the South Durban area of South Africa. It is a good example of participatory science or democratic science where communities get involved in developing, logging, collating and interpreting scientific information and knowledge. The division between ‘expert’ and ‘local’ knowledge becomes deliberately blurred. The reporting of incidents, for example, is collated and mapped http://www.sdcea.co.za/images/stories/pdfs/mapsincidentstoscale0406.pdf . A set of data reliant on local knowledge, presented in a format understandable to local people and available for local communities to lobby on the basis of ‘scientific’ information.
Geography is central to this alliance and they have produced a brochure on their use of GIS in developing this community based science. http://www.sdcea.co.za/images/stories/pdfs/gisbrochurejuly08a.pdf
http://www.sdcea.co.za/images/stories/pdfs/gisbrochurejuly08b.pdf
Although this type of community based activity may not be translatable across the globe it does illustrate how individuals can use geography to monitor, interpret and lobby for action on their local environments. Environmental geography that really matters.