Carbon Trust releases white paper on CO2 impact of streamingFriday, June 11th, 2021
Updated calculation released on the carbon impact of online video streaming
- New white paper published by the Carbon Trust estimates an average carbon footprint of 55gCO2e per hour of video streaming in Europe in 2020, with viewing device as the largest determining factor
The Carbon Trust, an independent organisation with a mission to accelerate the move to a sustainable, low carbon economy has today published a new white paper examining the carbon impact of streaming video on demand to improve the understanding of the carbon emissions associated with video streaming and enable future decisions to be based on informed understanding of the methods, complexity, and uncertainties involved. The study is released at a time of high demand for video streaming, but little consensus of video streaming’s impact on the climate.
This latest research from the Carbon Trust includes the lifecycle energy use of the different components involved in the distribution and viewing of video content1 to estimate the carbon impact per one hour of video streaming, based on video-on-demand viewing (not including live streaming).
The findings revealed that at an individual level, the carbon footprint of viewing one hour of video-on-demand streaming is very small compared to other everyday activities. The average carbon footprint in Europe per hour of video streaming is approximately 55gCO2e, equivalent to boiling an average electric kettle three times.2
Whilst the findings are comparable with some other recent estimates, they are lower than some previous studies, mainly due to older energy intensity figures being significantly higher than figures relevant to 2020.
The analysis also showed that the viewing device is typically responsible for the largest part (more than 50%) of the overall carbon footprint. For example, the footprint (related specifically to the energy use of the viewing device) of watching on a 50-inch TV is shown to be roughly 4.5 times that of watching on a laptop, and roughly 90 times that of watching on a smartphone. End-user viewing devices are also becoming more energy efficient due to a mix of technology advances, regulation and standards (e.g. around stand-by power and maximum power thresholds). In particular, energy efficiency gains have been achieved through improvements in device display technology. Specifically, television displays consume less energy per surface area, so are more energy efficient, offsetting the continued growth of average TV panel sizes. Some studies report that the shift to more efficient TVs has led to decreased TV related energy use in some households.3
Additionally, changes in video quality, due to different viewing resolutions and settings – like changing from high-definition to standard-definition – affect the bitrate required to transmit video data, but were found to have only a very small change in the carbon impact.
The white paper focuses its main results on a conventional approach, established in existing studies which represents the average energy intensity of the transmission network per unit of data transmitted.4 Additionally, it offers some insights based on an emerging power model approach which more closely represents the instantaneous use of energy in the network, primarily to assess the short-term marginal impact of changes in video quality. Together these methodologies serve to bring new insight into the links and influencing factors between video streaming and carbon emissions. By better understanding the footprint of video streaming, the sector can work together to reduce video streaming’s carbon impact.
Improvements in technology mean that network equipment required to deliver the internet is continually becoming more energy efficient. In tandem with this reduction in energy intensity, it is notable that the actions and trends in the information communication technology (ICT) sector, such as increased procurement of renewable electricity, are expected to continue to drive down the carbon intensity of ICT services over time, including video streaming.
The white paper also sets out many opportunities for further research to continue to improve and add to the evidence-base in this area. It is clear, however, that a strong understanding of the impact and context of video streaming is vital to informing future decision making on the use of video streaming and the use of ICT more broadly.
The white paper has been developed by the Carbon Trust, in close consultation with members of DIMPACT – a collaborative project of world-class researchers from the University of Bristol and 13 of the world’s most innovative media companies – and made possible by funding from Netflix.
Commenting on behalf of DIMPACT, Glynn Roberts, Director at Carnstone, stated: “DIMPACT welcomes the Carbon Trust’s whitepaper on the carbon impact of video-on-demand streaming.
DIMPACT is a pioneering initiative, convening leading media companies along with researchers from the University of Bristol, to help the digital media industry map and manage its carbon impacts. Together, we have developed a robust methodology and measurement tool that allows us to better understand ‘downstream’ carbon impacts, right through to the end-user.
Accurate measurement is essential for the industry to correctly account for the emissions along the value chain, as well as understand how these emissions can be practically reduced. This timely publication is an important milestone for the DIMPACT project.”
Andie Stephens, Associate Director at the Carbon Trust and lead author of the white paper, commented: “Our white paper shows that the carbon footprint of watching an hour of streamed video content is minor compared with other daily activities. As the electricity grids continue to decarbonise, and telecoms network operators increasingly power their networks with renewable electricity this impact is set to reduce even further. By undertaking this research with the support of the industry and academic experts, we hope to help inform discussions about the carbon impact of video streaming and of wider ICT use, and address some misunderstandings and outdated estimates that have been previously reported.”