This package offers a handy tool to calculate carbon footprints from air travel based on IATA airport codes or latitude and longitude.
Package footprint
uses the the Haversine great-circle distance formula to calculate distance between airports or distance between latitude and longitude pairs. This distance is then used to derive a carbon footprint estimate, which is based on converstion factors from the Department for Environment, Food & Rural Affairs (UK) 2019 Greenhouse Gas Conversion Factors for Business Travel (air): https://www.gov.uk/government/publications/greenhouse-gas-reporting-conversion-factors-2019.
DEFRA’s conversion factors are a widely used tool for calculating emissions for a variety of industries. For business air travel, they consider trip length (domestic, short-haul, long-haul, and international), flight class (e.g. economy, first), and various types of emissions, with and without radiative forcing.
Their methodology for determining distance states that anything within the UK is “domestic”, and that flights up to 3,700km are “short-haul”. “Long haul” is over 3,700km. “International” flights are those that occur entirely outside of the UK. Neither domestic nor international was clearly defined, so definitions from the US Environmental Protection Agency were considered. Based on these two data sources, the footprint
calculates distance as follows:
EPA | DEFRA/footprint |
Distance |
---|---|---|
Short-haul | Domestic | < 483 km |
Medium-haul | Short-haul | 483 to 3,700 km |
Long-haul | Long-haul | > 3700 km |
You can use pairs of three-letter IATA airport codes to calculate distance. This function uses the airportr
package, which contains the data and does the work of getting the distance between airports. Note: the airportr
package offers a number of useful functions for looking up airports by city or name and getting the IATA airport codes.
The airport_footprint()
functions takes a three-letter IATA code for the departure airport (case insensitive), a three-letter IATA code for the arrival airport (case insensitive), a flightClass
(e.g. “Economy), and a emissions metric (e.g. ”co2e"). The latter two arguments are case sensitive. See ?airport_footprint
for more information on arguments.
The example below calculates a simple footprint estimation for an economy flight from Los Angeles International (LAX) to Heathrow (LHR). The estimate will be in CO2e (carbon dioxide equivalent, including radiative forcing). The output is always in kilograms.
If there is a layover in Chicago, you could calculate each leg of the trip as follows:
We can calculate the footprint for multiple itineraries at the same time and add to an existing data frame using mutate
. Here is some example data:
travel_data <- tibble(name = c("Mike", "Will", "Elle"),
from = c("LAX", "LGA", "TYS"),
to = c("PUS", "LHR", "TPA"))
name | from | to |
---|---|---|
Mike | LAX | PUS |
Will | LGA | LHR |
Elle | TYS | TPA |
Here is how you can take the from
and to
data and calculate emissions for each trip. The following function calculates an estimate for CO2 (carbon dioxide with radiative forcing).
name | from | to | emissions |
---|---|---|---|
Mike | LAX | PUS | 1434.663 |
Will | LGA | LHR | 825.497 |
Elle | TYS | TPA | 136.721 |
If you have a list of cities, it might be easier to calculate emissions based on longitude and latitude rather than trying to locate the airports used. For example, one could take city and state data and join that with data from maps::us.cities
to quickly get latitude and longitude. They can then use the latlong_footprint()
function to easily calculate emissions based on either a single itinerary or multiple itineraries:
The following example calculates the footprint of a flight from Los Angeles (34.052235, -118.243683) to Busan, South Korea (35.179554, 129.075638). It assumes an average passenger (no flightClass
argument is included) and its output will be in kilograms of CO2e (the default)
You can use mutate
to calculate emissions based on a dataframe of latitude and longitude pairs.
Here is some example data:
travel_data2 <- tribble(~name, ~departure_lat, ~departure_long, ~arrival_lat, ~arrival_long,
# Los Angeles -> Busan
"Mike", 34.052235, -118.243683, 35.179554, 129.075638,
# New York -> London
"Will", 40.712776, -74.005974, 51.52, -0.10)
name | departure_lat | departure_long | arrival_lat | arrival_long |
---|---|---|---|---|
Mike | 34.05224 | -118.24368 | 35.17955 | 129.0756 |
Will | 40.71278 | -74.00597 | 51.52000 | -0.1000 |
And here is code to apply it to a dataframe:
travel_data2 %>%
rowwise() %>%
mutate(emissions = latlong_footprint(departure_lat,
departure_long,
arrival_lat,
arrival_long))
name | departure_lat | departure_long | arrival_lat | arrival_long | emissions |
---|---|---|---|---|---|
Mike | 34.05224 | -118.24368 | 35.17955 | 129.0756 | 1881.589 |
Will | 40.71278 | -74.00597 | 51.52000 | -0.1000 | 1090.260 |