2020 Shortlist

Drowning in plastic

Category: Best visualization (small and large newsrooms)

Country/area: Singapore

Organisation: Reuters

Organisation size: Big

Publication date: 9 Apr 2019

Credit: Simon Scarr, Marco Hernandez

Project description:

It’s one thing to say the world is drowning in plastic. It’s another to show it. Around the world, almost 1 million plastic bottles are purchased every minute and they are having a devastating effect on the environment. That equates to 480 billion bottles per year. The numbers are mind boggling. But what do they look like?

This project puts those numbers into perspective by visualising them in real time, as well as piling the bottles up next to landmarks to give a true sense of scale.

Impact reached:

The project was instantly picked up and shared by prominent environmental groups including Greenpeace. It was tweeted by influential figures ranging from a U.S. House Representative to a Bollywood movie star. The piece grabbed headlines in other media outlets and online platforms which wrote articles on how the visualisation brought home the shocking scale of plastic waste. The piece is still shared widely as a vivid explanation of the world’s addiction to single-use plastic.


Techniques/technologies used:

The main piece of software at the heart of this project was Cinema 4D. The team made a 3D physics engine in order to allow the bottles to drop, collide, and roll in a realistic space. The first inputs into the model were the mass and other properties of a single bottle from a vending machine in our Hong Kong bureau.

An early projection showed that the almost 17,000 bottles consumed every second would quickly bury first the human figure and the garbage truck we placed for a sense of scale.

Adobe After Effects was used to add final polishing to the animation.

NOTE: Viewing this project is seemless on any device.

What was the hardest part of this project?

Building a 3D physics engine! As the team continued to work on the models and renderings, they hit a snag: the computer was not powerful enough to execute the visualisations were looking for. Even running a single computer around the clock, it took 9 days to render the first visualization.

The work-around was to remotely connect a dozen Reuters computers in Hong Kong, Singapore, Bengaluru and London into what became an ad hoc “render farm” using wifi to harness the computational power needed to make and edit the animation.

Finding the middle ground between realistic rendering and illustrative style was also difficult. Making the bold decision to alter any style or aesthetics of the final project was very challenging as we knew this would add time. Moving the camera angle a few degrees meant rerendering the entire animation and/or images which took many hours and sometimes days.

What can others learn from this project?

An effective and powerful visualisation can be made from a very small set of numbers. There isn’t always a need for massively complicated data sets. This piece started with a single figure, how many bottles were sold in 2018.

Project links: