Mapping Pollution in China: an update on the CHIPOMAP Project

[:fr]Matteo Tarantino

The CHIPOMAP research project  was started by the Confucius Institute of the University of Geneva in June 2014. Funded by a Swiss National Science Foundation grant #153291, the three-year project analyzes the Chinese environmental crisis. Specifically, CHIPOMAP focuses on the China Pollution Map (“CPM”), a software system built and operated by the Beijing NGO Institute for Public and Environmental Affairs (“IPE”). IPE’s core mission is to increase public participation into environmental affairs through the disclosure of pollution information.

The knowledge infrastructure that underlies CPM contains official data on environmental violations by factories. The database feeds a website, a mobile app and other query tools targeted at professional users. Each of these avenues offers users different options for visualizing the data. In particular, both on the website and the mobile app, data can be visualized on a map of China, allowing users to see nearby polluters, have a bird’s eye view of the situation of industrial pollution in the country, and access data on polluters, including their violations and emission quantities.

Our research interest stems from the fact that the CPM connects with many social actors. For example, the data is supplied by government institutions (especially by local and provincial Environmental Protection Bureaus or EPBs), many of which are very supportive of IPE’s work. Several international brands use CPM to push their Chinese suppliers to address their negative environmental impact, making it a condition for retaining business relations. Chinese companies show a more complicated, often conflictual relationship with the CPM, as the cost of being included in the database has recently been rising for businesses (on the other hand, institutional fines for environmental violations have remained comparatively modest, albeit the situation has been changing recently). Also, after CPM’s mobile app was mentioned in Chai Jing’s breakthrough documentary on pollution “Under the Dome” (released in February 2015), national and international media has paid more and more attention to IPE’s activities.

This project focuses particularly on such questions as: how is the software system informing CPM built and operated? How does it impact different social actors? How do technical choices and constraints influence its evolution? Who is using CPM and how? What kind of data is included or excluded? What kind of knowledge and expertise informs its design?

Various disciplines, including Chinese studies, software studies, science and technology studies (as developed by scholars such as Trevor J. Pinch, Harry Collins and Bruno Latour), risk studies as well as critical cartography provide the analytical framework for this project. Such an interdisciplinary approach enables us to analyze the methods and significance of the sociological analysis of software objects; gain insight into the Chinese internet’s specificities and social impacts; and how technical objects – such as software, data and maps are constructed, operated, interpreted, negotiated, adopted, or eliminated; and how they co-produce a cartography of risk in the age of « big data.

CHIPOMAP is a collaborative effort. The project has benefitted from the participation of Beijing’s Renmin University’s Sociology Department (with renowed sociologist of social movements  LI Lulu 李路路),  the LATTS laboratory of ParisTech (with Risk Studies specialist Valérie November), and Cornell University’s department of Science and Technology Studies (with the social studies of cartography specialist Christine Leuenberger). The project has been led by the IC’s director Basile Zimmermann, and I, Matteo Tarantino, have served as the principal investigator. The project has benefitted from the generous and extensive cooperation of IPE itself and of its director Ma Jun.

In our first year we conducted a ethnographic fieldwork in IPE’s offices for 4 months, and analyzed IPE’s internal and public documents, Internet analytics, the code and architecture of the CHIPOMAP system, media coverage, and social media interactions.

Provisional results have been presented at the 2015’s AAS conference in Chicago, during a panel organised by the IC called “Making Sense of Pollution” , and will be disseminated in forthcoming papers.

Preliminary findings suggest that:

a) NGOs increasingly become information processing centres in order to perform “informational governance” in “information-poor environments”[1] such as in contemporary China. IPE established, refined and socialised a vast array of techniques and procedures in order to make data on pollution suitable for public disclosure and dissemination. While more and more data is increasingly available due to the Chinese Government’s growing efforts towards environmental transparency (especially since the 2008’s Open Government law), such data can seldomly just be fed into the CPM system, but needs considerable, often painstakingly manual processing efforts before the system can “read” and present that data to the public. Paradoxically, as environmental data becomes more available, disseminating it has become increasingly complex. For example, real time updates of data in mobile apps is theoretically the optimum in terms of transparency, but require a lot of battery power. This discourages users from adopting the app, thus impeding the disclosure effort. Secondly, the government’s push towards real time disclosure of emission data means companies develop various techniques for tampering with sensors, and thereby they compromise the quality of the data. IPE needs to therefore continuously respond to such challenges in order to remain effective

b) The CPM’s disclosure efforts to a “lay” public are constantly evolving and adapting to ever more complex knowledge networks.  IPE’s core objective has always been to increase public involvement into its disclosure efforts, and Chinese users have proven to be elusive and constantly migrating between different software platforms (from QQ to Weibo to Wechat and so on). Therefore IPE employs various dissemination strategies that entail not only its own avenues (website and app), but also traditional media sources and collaborations with IT’s players such as Sina Weibo and WeChat. Since 2015, these effort have started to pay off with a growing group of users taking part into IPE’s activities through microblogging by reporting on polluters and putting pressure on local EPBs. These networks also impact the technical nature of IPE’s software system, as it must connect with the systems, protocols and languages of these other social actor (e.g. making its content shareable on Weibo).

These examples reveal how a technically-grounded and scientifically informed approach can contribute to discussions of governmental and non-governmental environmental governance in the age of “big data”.

TARANTINO, Matteo. «Mapping Pollution in China: an Update on the CHIPOMAP Project ». In Blog Scientifique de l’Institut Confucius, Université de Genève. Lien permanent: https://ic.unige.ch/?p=703, consulté le 05/25/2024.

Cette contribution a été relue par Christine Leuenberger.


[1] Both “Informational Governance” and “Information-poor environment” concepts have been developed by Arthur P.J. Mol.  The former refers to the involvement of information technologies in governance issues; the latter to contexts in which access or quality of pertinent information is restricted by technical, political or social constraints.

[:en]Matteo Tarantino

The CHIPOMAP research project  was started by the Confucius Institute of the University of Geneva in June 2014. Funded by a Swiss National Science Foundation grant #153291, the three-year project analyzes the Chinese environmental crisis. Specifically, CHIPOMAP focuses on the China Pollution Map (“CPM”), a software system built and operated by the Beijing NGO Institute for Public and Environmental Affairs (“IPE”). IPE’s core mission is to increase public participation into environmental affairs through the disclosure of pollution information.

The knowledge infrastructure that underlies CPM contains official data on environmental violations by factories. The database feeds a website, a mobile app and other query tools targeted at professional users. Each of these avenues offers users different options for visualizing the data. In particular, both on the website and the mobile app, data can be visualized on a map of China, allowing users to see nearby polluters, have a bird’s eye view of the situation of industrial pollution in the country, and access data on polluters, including their violations and emission quantities.

Our research interest stems from the fact that the CPM connects with many social actors. For example, the data is supplied by government institutions (especially by local and provincial Environmental Protection Bureaus or EPBs), many of which are very supportive of IPE’s work. Several international brands use CPM to push their Chinese suppliers to address their negative environmental impact, making it a condition for retaining business relations. Chinese companies show a more complicated, often conflictual relationship with the CPM, as the cost of being included in the database has recently been rising for businesses (on the other hand, institutional fines for environmental violations have remained comparatively modest, albeit the situation has been changing recently). Also, after CPM’s mobile app was mentioned in Chai Jing’s breakthrough documentary on pollution “Under the Dome” (released in February 2015), national and international media has paid more and more attention to IPE’s activities.

This project focuses particularly on such questions as: how is the software system informing CPM built and operated? How does it impact different social actors? How do technical choices and constraints influence its evolution? Who is using CPM and how? What kind of data is included or excluded? What kind of knowledge and expertise informs its design?

Various disciplines, including Chinese studies, software studies, science and technology studies (as developed by scholars such as Trevor J. Pinch, Harry Collins and Bruno Latour), risk studies as well as critical cartography provide the analytical framework for this project. Such an interdisciplinary approach enables us to analyze the methods and significance of the sociological analysis of software objects; gain insight into the Chinese internet’s specificities and social impacts; and how technical objects – such as software, data and maps are constructed, operated, interpreted, negotiated, adopted, or eliminated; and how they co-produce a cartography of risk in the age of « big data.

CHIPOMAP is a collaborative effort. The project has benefitted from the participation of Beijing’s Renmin University’s Sociology Department (with renowed sociologist of social movements  LI Lulu 李路路),  the LATTS laboratory of ParisTech (with Risk Studies specialist Valérie November), and Cornell University’s department of Science and Technology Studies (with the social studies of cartography specialist Christine Leuenberger). The project has been led by the IC’s director Basile Zimmermann, and I, Matteo Tarantino, have served as the principal investigator. The project has benefitted from the generous and extensive cooperation of IPE itself and of its director Ma Jun.

In our first year we conducted a ethnographic fieldwork in IPE’s offices for 4 months, and analyzed IPE’s internal and public documents, Internet analytics, the code and architecture of the CHIPOMAP system, media coverage, and social media interactions.

Provisional results have been presented at the 2015’s AAS conference in Chicago, during a panel organised by the IC called “Making Sense of Pollution” , and will be disseminated in forthcoming papers.

Preliminary findings suggest that:

a) NGOs increasingly become information processing centres in order to perform “informational governance” in “information-poor environments”[1] such as in contemporary China. IPE established, refined and socialised a vast array of techniques and procedures in order to make data on pollution suitable for public disclosure and dissemination. While more and more data is increasingly available due to the Chinese Government’s growing efforts towards environmental transparency (especially since the 2008’s Open Government law), such data can seldomly just be fed into the CPM system, but needs considerable, often painstakingly manual processing efforts before the system can “read” and present that data to the public. Paradoxically, as environmental data becomes more available, disseminating it has become increasingly complex. For example, real time updates of data in mobile apps is theoretically the optimum in terms of transparency, but require a lot of battery power. This discourages users from adopting the app, thus impeding the disclosure effort. Secondly, the government’s push towards real time disclosure of emission data means companies develop various techniques for tampering with sensors, and thereby they compromise the quality of the data. IPE needs to therefore continuously respond to such challenges in order to remain effective

b) The CPM’s disclosure efforts to a “lay” public are constantly evolving and adapting to ever more complex knowledge networks.  IPE’s core objective has always been to increase public involvement into its disclosure efforts, and Chinese users have proven to be elusive and constantly migrating between different software platforms (from QQ to Weibo to Wechat and so on). Therefore IPE employs various dissemination strategies that entail not only its own avenues (website and app), but also traditional media sources and collaborations with IT’s players such as Sina Weibo and WeChat. Since 2015, these effort have started to pay off with a growing group of users taking part into IPE’s activities through microblogging by reporting on polluters and putting pressure on local EPBs. These networks also impact the technical nature of IPE’s software system, as it must connect with the systems, protocols and languages of these other social actor (e.g. making its content shareable on Weibo).

These examples reveal how a technically-grounded and scientifically informed approach can contribute to discussions of governmental and non-governmental environmental governance in the age of “big data”.

TARANTINO, Matteo. «Mapping Pollution in China: an Update on the CHIPOMAP Project ». In Blog Scientifique de l’Institut Confucius, Université de Genève. Lien permanent: https://ic.unige.ch/?p=703, consulté le 05/25/2024.

Cette contribution a été relue par Christine Leuenberger.


[1] Both “Informational Governance” and “Information-poor environment” concepts have been developed by Arthur P.J. Mol.  The former refers to the involvement of information technologies in governance issues; the latter to contexts in which access or quality of pertinent information is restricted by technical, political or social constraints.

[:]