Visualizing Transboundary Shipments of E-waste
How are the international trade and traffic of electronic waste organized?
The interactive cartograms above are a partial answer to that question. They depict the organization of trade and traffic in 1996 and 2012, the earliest and latest dates for currently available data. These visualizations are an interactive version of work published as part of the Reassembling Rubbish project in academic journals. A brief introduction to that work and its findings is available here. For readers who may be interested, the full articles are available free without subscription at the following links:What do the cartograms show?
The cartograms depict data available from the United Nations Statistics Division's "COMTRADE" database. The data are organized into territories and the flows of e-waste that connect them. Territories are colour coded to reflect their status under Annex VII of the Basel Convention (blue for Annex VII territories and yellow for non-Annex VII territories). The larger the circle representing a given territory (e.g., Canada) the more other territories report receiving e-waste shipments from it. Readers not already familiar with the significance of the division between Annex VII and non-Annex VII territories are invited to download the publications above, both of which provide discussion about this issue.
The flow lines are also colour coded to reflect the Annex VII status of their source. In other words, flows shown in blue originate in Annex VII territories; flows shown in yellow originate in non-Annex VII territories. The thicker the flow lines, the larger the volume of the flow.
The data come with some extremely important caveats and limits. These are discussed fully in the Methods section of Lepawsky (2014, free access). In summary, the caveats and limits of the data include:
- The data show information only for "waste batteries and scrap of primary batteries and electrical accumulators" (technically referred to as Harmonised System 2002 code 854810).
- The data are import transactions reported by a given territory. In other words, these are data showing what Territory "A" says it receives as an import from any other territory(ies) in a given year (e.g., from Territories "B", "C", and "D").
- The data miss unreported trade (whether licit or illicit).
- The data visualized here are best understood as proxy measurements of overall transboundary shipments of electronic waste.
How do I use the cartograms?
Interacting with the cartograms is as easy as using any webpage (though they do not have full functionality on mobile platforms):
- The cartograms work best by clicking the "View Page" button which will open each cartogram in a full sized window.
- Click on a given territory (e.g., Canada) and its trade network will be highlighted and a side-bar with additional information will pop-up.
- Read flow lines in a clockwise direction from a given territory to the territory(ies) it reports receiving shipments of e-waste from.
- The years 1996 and 2012 are used to show changes in the organization of transboundary shipments of e-waste over time. A brief overview of those changes is available here. Detailed interpretations of these changes are available in Lepawsky (2014, free access).
How were the cartograms produced?
A full discussion of the methods used to build these visualizations is available in Lepawsky (2014, free access). To summarize the process:
- Data were downloaded from COMTRADE and organized in a spreadsheet.
- The organized data were imported into Gephi, a free open-source network analysis software.
- The networks created in Gephi were exported using Sigmajs Exporter, a Gephi plugin produced by the Oxford Internet Institute.
The Gephi networks exported using Sigmajs Exporter currently have some limitations that control the 'look' of the cartograms. These are largely aesthetic, but do have some implications for interpreting the cartograms. For example, it not currently possible with Sigmajs Exporter to have the value of trade flows show up (e.g., with a data value adjacent to a given flow line). Thus interpretation of trade volumes is currently limited to visual comparisons of flow line thickness (thicker lines equal greater volumes of flow and vice versa). Also, some flow line colours render as a darker shade than others. This is an artefact of the Sigmajs Exporter plugin and amounts to an aesthetic issue and should not be interpreted as having anything to do with the underlying data.