A virtual conference is a green decision. Instead of traveling, staying in hotels, and using a convention center, Educators in VR prevented more than 7.5 million kg of carbon dioxide from being emitted into the atmosphere.

Article and analysis by Don Balanzat

Educators in VR made history a couple of weeks ago by holding the largest immersive virtual conference, the 2020 Educators in VR International Summit. It featured 170 speakers in over 150 events on 5 different virtual platforms over 6 days, and an estimated 6,000 people attended. A volunteer team of 75 people made this possible, as explained in a recent Behind the Scenes of the International Summit workshop.

In addition to being the first time a VR virtual conference of this size ever took place, the conference’s impact on anthropogenic climate change was exciting. With the rising concern over the amount of academic and professional travel that takes place to produce such conferences, mixed with the unfortunate timing of the COVID-19/coronavirus outbreak, the team asked me to explore the environmental impact of the conference. Travel and energy utilities are generally the highest percentages of any given individual’s carbon footprint.

Virtual conferences can be an incredibly impactful way in which we can individually and societally reduce our carbon output, and slow the human impact on the climate change process.

Considerations and Calculations

Discussing this with the Educators in VR team, we had to consider the comparison of the virtual conference with a real world conference, and make some considerations and assumptions for our calculations. To ensure we are all on the same page, let me explain.

Because of the international nature of attendees in Educators in VR, we had to consider a broad range of travel options – cars, rail, planes – and different distances. Virtually, people stayed wherever they were, attending from homes, offices, schools, even parks, rarely traveling out of reach of a stable WIFI connection. If this was a real-world conference, travel would be the largest impact on the environment.

As a reference “conference destination,” we hovered between New York and London, since Educators in VR is based in Wales and these cities are two of the most highly traveled cities in the area. Whatever carbon would have been output by travel to these locations, would then be prevented from entering the environment by a virtual conference.

We will estimate this output based on things like travel method fuel efficiency and travel distance.

While the International Summit was hosted on AltspaceVR, ENGAGE, rumii, Mozilla Hubs, and Somnium Space, the majority of the events were held in AltspaceVR, owned by Microsoft.

The summit operated primarily on Microsoft Azure servers out of Seattle, running for the entirety of the 6 days (144 hours).

We wanted to consider energy usage of hotel accommodations. Additionally, we couldn’t ignore the energy usage of an appropriate conference space in the real world.

The final tallies will be in terms of kilograms of carbon dioxide (kg CO₂).

Travel: Planes, Trains, and Automobiles

A good adventure begins with a trip, so let’s explore the planes, trains, and automobiles used to arrive at a real-world conference to our virtual world conference.

This diagram created by BBC using data from the UK’s Department for Business, Energy and Industrial Strategy (BEIS) shows the emissions from different modes of transport.

The lowest impact travel system is rail. As of this writing, it seems to be the most energy efficient form of mass transit with around 41 grams of CO₂ generated per passenger per kilometer traveled. For those wanting a more comprehensive look, it’s important to know that this number is highly dependent on whether the train is powered by natural gas or electricity, and if the train is powered by electricity, the number will depend on whether the electricity is green or not. You can read about all of those juicy bits in this UK Government PDF on 2019 Green House Gas Reporting Methodology for specifics.

For our calculations, we will say it is 41g CO₂/passenger/km traveled via rail.

Next in the real-world conference adventure is travel by car. From the above graphic, a 1 passenger car ride will produce 171g CO₂/passenger/km, while a 4 passenger car ride will produce 43g CO₂/passenger/km. This is based on the average diesel car, where the extra weight of passengers is negligible compared to the overall gas mileage of the car. For our calculations, we’ll stick with the carpooling number of 43g CO₂/passenger/km traveled by car.

Last in our travel plans is the more common plane travel to a real-world conference. This is the big one. A domestic flight is generally defined as a flight that departs and arrives within the same country, and a long haul flight is generally any flight over 7 hours. I found that the diagram above is reductionist in relation to that, so let’s move on to the following graphic from Vox that uses data from Green Car Congress:

These calculations are based on round-trip flights, and utilized the calculator at myclimate.org, which takes into account both direct and indirect emissions. You can read more about the calculation principles in their PDF support documentation.

Obviously, longer flights will produce higher total emissions, but it’s important to note that the emissions per mile per passenger significantly decreases as the length of the flight increases – the fuel efficiency gets better on longer flights. For scale, let’s dig into the following examples.

The San Francisco-Santa Rosa flight is a very short 50 minute flight and creates 2.98 lbs of CO₂/mile (840 grams of CO₂/km) per passenger. This seems very high, especially compared to the BEIS data, but this is an incredibly short flight where a majority of the CO₂ emissions inefficiencies come from taking off and landing.

Compare this to a long New York-Singapore flight, which is 19 hours long and generates around 0.73 lbs of CO₂/mile (205 grams of CO₂/km) per passenger.

This is assuming the passenger is flying on economy tickets – these numbers increase if you fly first class due to the increased accommodations!

So, we finally have the fuel efficiencies of all of these modes of transportation:

• 41g CO₂/passenger/km traveled via rail
• 43g CO₂/passenger/km traveled by car when carpooling with 4 people
• 205g-840g CO₂/passenger/km traveled by plane depending on how long or short the flight

Initial figures for attendees at the Educators in VR International Summit were 5,000, upon which I based my research. The latest numbers are higher, but let’s stick with that nice round number. Using more assumptions on those 5,000 people, most would have flown due to the international nature of our attendees – the longest flights being 20 hours and the shortest being about 45 minutes to an hour. If we split the difference and skew towards more people being centralized in big American or European countries, let’s say the average flight time of attendees will be 7 hours, and let’s say 3500 people are taking flights.

A 7 hour flight will land our flying travelers around the 366g CO₂/passenger/km CO₂ emission rate. For 3,500 people traveling 5567 km (the distance between London and New York), that will mean around 7,131,327 kg of CO₂ just from air travel. (For fun, let’s push the extreme and say all 5000 people are flying – that increases the number to about 10,000,000 kg of CO₂!)

The fuel efficiency of carpooling and train-riding folks are about the same, so let’s split the difference at 42g CO₂/passenger/km for the other 1,500 people. Let’s say they will only travel this way if it’s “convenient” – so they are, on average, 2 hours or around 450 kilometers away from the conference destination, and they are making a round trip. Then the CO₂ produced is about 56,700 kg. The emissions from cars and rail are crumbs compared to air travel, but every bit counts!

This brings our travel carbon total to 7,188,027 kg for the 5,000 people traveling by plane, rail, or car.

Energy Usage: Powering the Conference

How much energy does it take to produce a real-world conference? That’s the question, indeed. To figure this out, we need to explore the cost of, say, a hotel as the conference location, both for the actual conference as well as lodging. How do you estimate the electricity and gas usage of such a facility?

According to the U.S. Energy Information Administration (EIA), 62.7% of electricity in the United States comes from fossil fuels. This electricity production emits 2.21 lbs CO₂/kWh, or about 1 kg/kWh. Also, according to the EIA, 117 lbs CO₂ or about 53 kg CO₂ is produced per thousand cubic feet of natural gas.

Let’s consider accommodations for people staying in hotels. According to the EIA’s Commercial Buildings Energy Consumption Survey, hotels and motels use an average of 14 kWh of electricity/ft² and 49 cubic feet of natural gas/sqft annually. We can find numbers for the 6 day summit by dividing this annual number by 365 days and multiplying by 6. This gives us .23 kWh/ft² and .805 ft³ of natural gas/ft² for the 6 days that attendees would be at the conference. The question that remains is how much square footage our attendees would have occupied during a 6 day stay.

The average hotel room is roughly 330 square feet. Let’s say, on average, there will be 2 people in each hotel room. This means that 5,000 people would have occupied 825,000 square feet of hotel space.

If we plug in these numbers, we get that the total electricity usage in hotels would be:

.23 kWh/sqft * 825,000 sqft = 189,750 kWh.

This translates to 189,750 kg of CO₂ emissions for hotel accommodations for 5,000 people over 6 days.

The natural gas isn’t negligible either. Using that same space number, 0.805 ft³ natural gas/sqft * 825,000 sqft = 664,125 ft³ of natural gas would be used, which translates to:

(117 lbs CO₂/1000 ft³) * (664,125 ft³) = 77,703 lbs CO₂ or 35,245 kg of CO₂ produced

In total, this brings us to 189,750kg + 35,425kg = 225,175 kg of CO₂ produced from energy usage in hotel accommodations.

Our last consideration is conference venue energy usage. According to Statista, the biggest convention centers are between 1 and 2.5 million ft² big. The convention centers on the lower end of the scale are appropriate to support our attendance number of 5,000, so we will use the 1 million ft² number for our calculations. Convention centers are considered “public assembly” buildings in the U.S., and according to the EPA’s Center for Corporate Climate Leadership (PDF), the average public assembly electricity and natural gas consumption are .0389 kWh/ft²/day and 0.093 ft³ natural gas/ft²/day. Multiplying by 1 million ft² and 6 days, we get that the conference venue energy usage would be:

(0.00389 kWh/ft²/day) * (1 million ft²) * (6 days)= 233,400 kWh, which translates to 233,400 kg of CO₂ emitted from electricity

and

(0.093 ft³ natural gas/ft²/day) * (1 million ft²) * (6 days) = 558,150 ft² natural gas which translates to 65,303 kg of CO₂ emitted from natural gas.

This adds up to 298,703 kg of CO₂ emissions for the 6 days of conference venue usage.

The energy needs of the conference and accommodations for people would be:

233,400 kg CO₂ + 298,703 kg CO₂ = 523,878 kg of CO₂.

Energy Usage: Powering the Virtual Conference

The Educators in VR International Summit ran mostly off of Microsoft Azure servers, which currently operate 100% sustainably with carbon offsets and energy certificates since 2012, with commitment to being carbon negative by 2030. Turns out, we don’t need to worry about this at all in our calculations, especially since we have the report on the state of Data Center Energy Use in 2018.

Let’s consider the electricity required by the devices of the virtual conference attendees. We don’t have the exact numbers, but there were more than 150 events and over 5,000 people – this averages out to about 30 people online at any given time during the 6 day summit. The average power consumption of a desktop computer is 200 W/hour. Some arithmetic yields (30 people) x (200 W / hour) x (144 hours) = 864 kW over 6 days. This translates to an energy usage of 124,416 kWh, and we will need to subtract this from the emissions that were avoided by the hotels and conference venues.

Therefore, the total emissions from energy usage from a virtual conference instead of staying in hotels and using a convention center
523,878 kg CO₂ – 124,416 kg CO₂ = 399,462 kg CO₂

Real-World Conference vs Virtual Conference Impact?

If we add up the total CO₂ emissions of 5,000 people traveling by air, rail, and car with the emissions of hotel accommodations and an appropriate conference venue, we get a grand total of the following CO² that was not emitted into the atmosphere by the virtual conference. Drum roll, please:

7,188,027 kg + 399,462 kg of CO₂ = 7,587,489 kg of CO₂

It’s important to understand that this is by no means a perfect calculation – there is a lot of nuance in the considerations here, as well as some things that were glossed over. I think that this is a decent starting point for a conversation about how virtual conferences can impact human generated greenhouse gas emissions.

About the Author

Don Balanzat has a degree in Physics from Rutgers University and is employed at Arizona State University (ASU). His work focuses on enhancing traditional science classes with novel apparatus and technology, as well as designing science outreach content. He is currently earning a Master’s in Geology at ASU studying the effects of virtual field trips (VFTs) as an educational platform and is actively participating in projects within the university’s virtual/augmented reality labs.