Impact of 5G deployment on energy consumption and climate

Illustré par :

As predicted long ago by a few enlightened minds, for example by the geophysicist Marion King Hubbert who, as early as 1956, predicted the peak of US conventional oil extraction in 1970(1), the peak of world conventional oil was reached in 2008. Presumably, the peak of conventional and unconventional oil combined occurred in November 2018 at 84.6 million barrels per day(2).

In view of the predicted shortage, it would be good policy to consider that any technical innovation must be examined in terms of its energy impact and that the only acceptable ones should be those that would contribute to a reduction in the consumption of fossil fuels.

The emergence of each new generation of mobile telephony (2G, 3G and 4G) has inevitably been followed by a sharp increase in data flows and, consequently, in wireless network power consumption. There is no reason to believe that 5G will be any different; on the contrary, as with previous generations, the search for better performance in terms of throughput and capacity will lead to a new explosion in data traffic that will wipe out the benefits of any possible improvements in energy efficiency: this is the consequence of the rebound effect or Jevons paradox, named after the economist who stated it in the nineteenth century(3). Assuming 5G is more energy efficient than previous generations, which is promised but not proven, the rebound effect will negate its unlikely benefit.

In fact, a 5G antenna site consumes 3 to 3.5 times more electricity than the equivalent in 4G according to a Huawei document(4), which is also confirmed by Chinese operators precursors in this field(5). On the other hand, given the use of millimeter waves by 5G — waves that are strongly attenuated by the slightest obstacle such as tree leaves and rain — the deployment of 5G requires a multiplication of antennas, up to one antenna every 100 meters in urban areas, per operator. With antennas alone accounting for more than half of the operators’ power consumption, the rollout of 5G will triple their power consumption.

According to Hugues Ferreboeuf and Jean-Marc Jancovici, engineers and experts in energy transition, the impact of 5G antennas will therefore represent a 2% increase in the overall electricity consumption of a country like France (or Belgium). They state,  » To this must be added the energy needed to manufacture the network elements, and especially to produce the billions of terminals and connected objects that we wish to link via this network (worldwide, the energy needed to manufacture terminals, servers and network elements represents 3 times the energy needed to operate the networks, excluding data centers). While an increase in the duration of use of smartphones would be central to reducing their carbon footprint, the advent of 5G would accelerate their replacement, to the delight of equipment manufacturers « (6).

The 2% increase in a country’s electricity consumption linked to 5G antennas is therefore only the tip of an iceberg that is mainly made up of the energy required for all the industrial processes linked to the deployment of this technology, first and foremost the manufacture of terminals (smartphones, tablets, laptops, etc.) that are still and always promised to be rapidly obsolescent.

Deploying 5G would therefore contribute to waste ever faster this limited resource that is oil, which will thus be greatly and cruelly missed by future generations to ensure a transition to a sustainable and decent society.

CLIMATE

5G is promoted by the telecom industry as a major technical innovation with multiple qualities(7) which would even contribute to the fight against global warming: according to Agoria(8), in its 2019 brochure, 5G, IoT (internet of things) and  » The massive use of connected objects will improve energy and environmental management, thus contributing to the achievement of European climate objectives .

Yet wireless data transmission is inherently energy inefficient: for example, 4G is about 20 times more energy intensive than wired transmission (fiber optics or copper cable)(9). It currently accounts for a significant portion of the 4% of greenhouse gases emitted worldwide by the digital industry, whose energy consumption is growing strongly, at 9% per year(10). The deployment of 5G would further accelerate this already deleterious trend.

At a time when worrying signals are multiplying, such as the accelerated melting of glaciers around the world and temperature records that are repeated year after year, and as the goal of containing global warming to 1.5 degrees, as set out in the COP21 agreement in Paris in 2015, becomes more and more remote, it has become imperative to limit the use of wireless, or even to outlaw it. Clearly, the deployment of 5G would go exactly against this imperative.(11)

Francis Leboutte

Notes et références
  1. 9,6 millions de barils par jour pour les 48 États (sans l’Alaska et Hawaï).
  2. Données de l’EIA (US Energy Information Administration, www.eia.gov) acquises en août 2020. Il s’agit du pétrole brut auquel les condensats aux puits de gaz (pentane, etc.) sont ajoutés comme le veut la tradition – sont exclus de ces données les autres ersatz de pétrole comme les liquides de gaz naturels (butane, propane…) et les agrocarburants. Depuis 2008, la diminution de l’extraction du pétrole conventionnel est péniblement compensée par le non conventionnel (pétrole de schiste, sable bitumineux, pétrole extrait en mer à très grande profondeur…) et par l’augmentation des condensats liée à celle de l’extraction du gaz naturel. Le pic de 2018 à 84,6 millions de barils ne devrait plus jamais être atteint d’autant que les investissements dans le non conventionnel ont chuté récemment, notamment suite à la chute du cours du baril et la récession liées à la pandémie du covid-19 (en avril 2020, le volume extrait était de 82,5 millions de barils par jour). L’extraction du gaz naturel et du charbon devrait culminer cette décennie.
  3. Transposé à la situation qui nous préoccupe : si une innovation (la 5G) entraîne un gain énergétique pour une technique considérée (la téléphonie mobile), son utilisation va s’intensifier de telle sorte qu’il en résultera une augmentation de la consommation d’énergie. Pour plus d’information sur ce paradoxe, lire The Myth Of Resource Efficiency, Mike Hanis, thelandmagazine.org.uk/articles/myth-resource-efficiency.
  4. Huawei, 5G Telecom Power Target Network, carrier.huawei.com/~/media/CNBGV2/download/products/network-energy/5G-Telecom-Energy-Target-Network-White-Paper.pdf
  5. Operators Starting to Face Up to 5G Power Cost, Robert Clark, www.lightreading.com/asia-pacific/operators-starting-to-face-up-to-5g-power-cost-/d/d‑id/755255 Hugues Ferreboeuf et Jean-
  6. Marc Jancovici : « La 5G est-elle vraiment utile ? ». Le Monde du 9 janvier 2020.
  7. Everything You Need to Know About 5G, IEEE (Institute of Electrical and Electronics Engineers, « The world’s largest technical professional organization for the advancement of technology », www.ieee.org) : « Les utilisateurs de téléphones portables d’aujourd’hui veulent des débits de données plus rapides et un service plus fiable. La prochaine génération de réseaux sans fil – la 5G – promet d’offrir tout cela, et bien plus encore. Avec la 5G, les utilisateurs devraient pouvoir télécharger un film en haute définition en moins d’une seconde (une tâche qui pourrait prendre 10 minutes sur la 4G LTE). Et les ingénieurs du sans-fil affirment que ces réseaux vont également stimuler le développement d’autres nouvelles technologies, comme les véhicules autonomes, la réalité virtuelle et l’internet des objets… ».
  8. Agoria : une association belge de l’industrie et des services liés aux nouvelles technologies.
  9. Frédéric Bordage, Sobriété numérique, Buchet/Chastel, 2019.
  10. The Shift Project, www.theshiftproject.org.
  11. En prime, une autre projection de Marion King Hubbert : la consommation d’énergie fossile de l’humanité sur une échelle de 10 000 ans, diagramme établi en 1969 (représenté par le zéro sur l’axe horizontal).
 

Espace membre

Member area