# This YAML file holds some of the content of the website, for convenience. # Learn YAML, it's worth it: http://sweetohm.net/article/introduction-yaml.html # /!. IF YOU BREAK THIS FILE YOU BREAK THE WEBSITE. TREAD CAREFULLY. # Metadata about this website meta: title: Travel Carbon Footprint Calculator description: A travel carbon footprint calculator for researchers. authors: - name: Didier Barret email: dbarret@irap.omp.eu role: Principal Investigator - name: Antoine Goutenoir email: agoutenoir@irap.omp.eu role: Software Ninja # Aka. Laws models: - name: Atmosfair_old # Slugged version of the display name above # Only lowercase alphanumeric, starting with a letter, using - as liaison # In other words, MUST match ^[a-z]([a-z0-9-]*[a-z0-9])?$ eg: icao-with-rfi # MUST be unique, two models MUST NOT have the same slug. slug: atmosfair_old # There MUST exist a python file named like this in `flaskr/laws` # And it MUST contain a class named EmissionModel # Please keep this lowercased, letters-only (or I will breathe fire) file: travel_emission_linear_fit # Color MUST be in the hex form, without alpha color: "#ff3399" # The configuration that will be fed to the model. # May be anything, really. Go bonkers! config: plane_emission_linear_fit: # A coefficient applied to the distance connecting_flights_scale: 1.05 # Radiative Forcing Index # Multiplier after scaling rfi: 1.9 # Flat scalar to add before scaling with intervals offset_before: 50 # Flat scalar to multiply before scaling with intervals scale_before: 1 # First interval found will apply its scale and offset intervals: # Distances [dmin, dmax[ are in kilometers, here. - dmin: 0 dmax: 1000 # The scaling law for this interval of distances. Defaults to 1. scale: 0.079107 # Offset is optional and defaults to 0. offset: 25.922 - dmin: 1000 dmax: 4000 scale: 0.066183 offset: 35.041 - dmin: 4000 dmax: 999999999 scale: 0.095998 offset: -80.835 # dmin dmax a b # for a fitting of (a*x + b) on the interval [dmin dmax[ # b is optional and defaults to 0 # - "0 500 4" # - 500 1000 8 - name: MyClimate # Slugged version of the display name above # Only lowercase alphanumeric, starting with a letter, using - as liaison # In other words, MUST match ^[a-z]([a-z0-9-]*[a-z0-9])?$ eg: icao-with-rfi # MUST be unique, two models MUST NOT have the same slug. slug: my-climate # There MUST exist a python file named like this in `flaskr/laws` # And it MUST contain a class named EmissionModel # Please keep this lowercased, letters-only (or I will breathe fire) file: travel_emission_lerp_fit # Color MUST be in the hex form, without alpha color: "#ff9933" # The configuration that will be fed to the model. # May be anything, really. Go bonkers! config: plane_emission_linear_fit: # A coefficient applied to the distance connecting_flights_scale: 1.05 # Radiative Forcing Index # Multiplier after scaling rfi: 1.9 # Flat scalar to add before scaling with laws offset_before: 0 # Flat scalar to multiply before scaling with laws scale_before: 1 # The travel_emission_lerp_fit uses points instead of intervals # List of (distance in km, footprint in ??) points: - [ 300.0, 120.42928309 ] - [ 1500.0, 278.88301355 ] - [ 2500.0, 438.31863895 ] - [ 20000.0, 3335.62849772 ] # This is the addition by Didier - name: atmosfair # Slugged version of the display name above # Only lowercase alphanumeric, starting with a letter, using - as liaison # In other words, MUST match ^[a-z]([a-z0-9-]*[a-z0-9])?$ eg: icao-with-rfi # MUST be unique, two models MUST NOT have the same slug. slug: atmosfair # There MUST exist a python file named like this in `flaskr/laws` # And it MUST contain a class named EmissionModel # Please keep this lowercased, letters-only (or I will breathe fire) file: travel_emission_lerp_fit # Color MUST be in the hex form, without alpha color: "#9933ff" # The configuration that will be fed to the model. # May be anything, really. Go bonkers! config: plane_emission_linear_fit: # A coefficient applied to the distance connecting_flights_scale: 1.05 # Radiative Forcing Index # Multiplier after scaling rfi: 1.0 # Flat scalar to add before scaling with laws offset_before: 0 # Flat scalar to multiply before scaling with laws scale_before: 1 # The travel_emission_lerp_fit uses points instead of intervals # List of (distance in km, footprint in ??) points: - [ 300.0, 420.42928309 ] - [ 1500.0, 478.88301355 ] - [ 2500.0, 838.31863895 ] - [ 20000.0, 6335.62849772 ] # The content is Markdown. HTML is also allowed. # If you also want Markdown in the titles, just ask. home: # The hero block (aka. jumbotron) is the top-level, salient block # It's like a welcoming mat :) hero: title: Estimate your travel carbon footprint # Using a pipe (|) allows you to set multiline content # Careful, indentation matters. description: | Travel footpint calculator provided by Didier Barret
CNRS, [IRAP](http://www.irap.omp.eu), [@DidierBarret](https://twitter.com/DidierBarret), [didier.barret@gmail.com](mailto:didier.barret@gmail.com) sections: # Add as many sections as you want. # Three blocks per section ; if you want another amount, it can be done, please ask. - blocks: - title: What does this do? content: | The tool helps computing your travel carbon footprint for a **round trip** from a given location, for a set of visited cities (listed in the input excel sheet). Similarly, the tool allows to compute the travel footprint of a conference/meeting/etc. (the originating city of each participant is provided in the input excel sheet). It can also return the city that would minimize the travel footprint assuming the same audience of the conference/meeting. While online CO2 calculators enable to compute the footprint for a limited number of trips with detailed trip information (e.g. connecting flights), this tool enables to compute the footprint of a larger number of travels, making some assumptions, e.g. to model connecting flights. For short distance travels, the CO2 footprint associated with trains is calculated and added to the air travel footprint. ### Original Motivation Global warming is a threat for life on our planet. Emissions of carbon dioxide by aircraft were 0.14 Gt C/year in 1992. This was between 2% of total anthropogenic carbon dioxide emissions in 1992 or about 13% of carbon dioxide emissions from all transportation sources (Intergovernmental panel on climate change, report 1999). Due to an increasing demand and the growth of the world economy, the number has grown since then, reaching closer to 3% nowadays, a number which will keep growing. Carrying scientific research requires traveling all across the world, but time has come to critically look at the way we carry research, with the aim of raising awareness and reducing our environmental impact, whenever possible. In this context, it is worth looking at what the carbon footprint of travels associated with the development of a large project, such as the Athena X-ray Integral Field Unit, is and implement actions to reduce it. The consortium involved in the development of the X-ray Integral Field Unit for the flagship Athena Space X-ray observatory of the European Space Agency involves currently 13 countries, 11 in Europe plus Japan and the United States. The amount of traveling required for the project is necessarily large. As PI of the X-IFU, my original motivation was to estimate what large meant, as an input to discussing actions to reduce our footprint with the members of the consortium. - title: Which data are used? content: | There is a growing interest in getting the travel footprint of scientific events, such as conferences or large meetings. Those are based on freely available CO2 calculators, some being relatively to use, requiring very limited user inputs. However, just by running some of them (including those from Carbon offset companies), it is amazing to see how their estimates can differ quite significantly (up to a factor of a few for the same trip). This is because they use different input data and consider different perimeters and assumptions, e.g. excluding freight or not, assuming different radiative forcing indices, seat accommodation in the plane… The straight numbers provided should therefore not be taken at face value, but should be looked at, for what they include and mean. It is also striking to me that there is hardly any scientific literature on the comparison between CO2 calculators, although often discrepancies are noticed in some communications, more particularly for long distance flights. Relying on one calculator is therefore not possible. This is why I am computing estimates based on different methods. This tool thus enables to compute your travel carbon footprint (for round trips), based on data provided by 4 independent state-of-the-art emission calculators: 1. the International Civil Aviation Organization (ICAO), 2. the UK Department for Environment, Food & Rural Affairs (DEFRA), 3. the ATMOSFAIR German Carbon offsetting company, and finally 4. the French Environment & Energy Management Agency (ADEME). ICAO, DEFRA and ATMOSFAIR have their methodology very well described (see resources section below). I have enquired to ADEME to get more about their methodology. ADEME and DEFRA provide mean emission factors as a function of flight distance. ADEME considers seat capacity and DEFRA gives emission factors as a function of seat type (from economy to first class seat). The ADEME emission factors are averaged over the seat capacity the coefficients are provided for. ATMOSFAIR and ICAO provide on line emission calculators requiring limited user inputs. Those on-line calculators have been run for a variety of flight distances, so that the estimates (without error bars) could be fitted with linear functions, between boundaries arbitrarily set (below 1000 km, between 1000 and 4000 km and above 4000 km). - title: How does this tool work? content: | For estimating your own travel footprint, the only thing you have to provide is an excel file which contains only the different final destinations that you traveled to, while the main city (origin) is entered on the form below. For estimating the travel footprint of a conference, you must provide the city from which each participant departs from and enter the host city of the conference in the form below. The tool then decodes the list of cities in the input excel sheet (please check the spelling when filling it up), and finds the closest airport within 100 km (large airport first and and if no large airports exists, it finds a medium one) using the google geolocator. This returns the longitude and latitude of the closest airport. Cities that cannot be located are ignored from the computation. To resolve ambiguity between cities of similar names (e.g. Cambridge), I require the name of the country. From the longitude and latitude of two airports (e.g. the host city of a conference and the departing city of an attendee), the tool computes the great circle distance (GCD). This is the shortest path a plane can follow. Short trips (e.g. shorter than 300 km) are accounted by trains. Traveling between cities often involve connexions. Here I consider a 5% increase of the GCD, as a mean value, understanding that it may be less on long flights and more on shorter flights. The emission factors of DEFRA include a 8% uplift to account for the fact that planes, even during direct flight, do not strictly follow the shortest path, e.g. to avoid bad weather conditions. ICAO adds some constants depending on the flight distance: the GCD correction factor is 50 km for GCD less than to 550 km, 100 km for GCD between 550 and 5500 km, and 125 km for for GCD above 5500 km. ATMOSFAIR adds a systematic 50 km to the GCD. For ADEME, there is no indication on whether any GCD correction applies, so I will assume the ICAO GCD correction. CO2 emissions per passenger take into consideration the load factor and are based only on passenger operations (i.e. fuel burn associated with belly freight is not charged to the passenger). Given the corrected GCD so computed, I use either the mean emission coefficients or the linear functions fitting the data of the on-line calculators. Note that there are no error bars on the estimates and very little literature on the topic. Only ATMOSFAIR returns whenever available, the data from different possible flights, considering different airlines. It also returns the the average over all airlines. # Second row of "blocks" - blocks: - title: What about Radiative Forcing? content: | CO2 emissions is computed from the total fuel burnt during the flight. For ICAO one of ton of fuel leads to `3.16` tons of CO2 emissions (this number accounts for the fuel burning itself and the prep-production and transport phase). A radiative forcing index of `1.9` is then applied as a multiplier to account for the fact that aviation contributes to climate change more than just from the emission of CO2 from burning fuels, by releasing gases and particles directly into the upper troposphere and lower stratosphere where they have an impact on atmospheric composition. These gases and particles include carbon dioxide (CO2), ozone (O3), and methane (CH4); trigger formation of condensation trails (contrails); and may increase cirrus cloudiness; all of which contribute to climate change. 1.9/2 is recommended by DEFRA and ADEME. ATMOSFAIR considers a multiplier of 3, for all emissions above 9000 km. ICAO, on the other hand does not include a multiplier, waiting for the scientific community to settle on a value. So 1.9 seems reasonable, but keep in mind that it comes with some uncertainty. You may want to select the ATMOSFAIR forcing index, instead of the 1.9 constant. The multiplier reaches about 2.8 for flights longer than 5000 km. - title: Which seat category are you considering? content: | I am considering only economy seats for the time being. Note that DEFRA emission factors are a factor of 3 larger for business class seat (which occupy a larger area in the plane). A factor of 1.5 should be considered when flying on Premium economy seats (ICAO would consider a factor of 2). - title: What about uncertainties? content: | I believe that each estimate has an uncertainty between 10 and 25% (really this is a best guess, and not substantiated by any statistical analysis and it could be more probably), the uncertainty being smaller for shorter flights for which there are more data. I would therefore refrain from using the numbers derived as absolute values. The results given should be considered indicative, although likely in the right ball park, for the assumptions that I make, and the approximation I use, e.g. in fitting ICAO and ATMOSFAIR data. For flights of average distance less than a few thousands kilometers, the results agree reasonably well, between the various estimators, which is a good sign. Larger differences are found when long distance flights are considered. In all cases, the estimates can be used for relative comparisons. # Third row, etc. - blocks: - title: Considering train travel for short travel distance content: | There is a minimum distance (by default 300 km) under which the calculator excludes flight travel. The calculator proposes instead to compute the travel footprint associated with train. The French emission factors are between 4 and 5 grams of CO2eq per km per passenger. This low value is likely due to the fact that electricity is provided by nuclear plants. It is larger by some factor accross Europe. Here I am assuming that the French factor are multiplied by a factor of 5. This makes train typically 10% less emitting than plane (including radiative forcing). - title: Additional resources content: | - Offset your flight with [atmosfair](https://www.atmosfair.de/en/offset/flight) - [ADEME](https://www.ecologique-solidaire.gouv.fr/sites/default/files/Info%20GES_Guide%20m%C3%A9thodo.pdf) (French Environment & Energy Management Agency) - [DEFRA](https://www.gov.uk/government/publications/greenhouse-gas-reporting-conversion-factors-2019) emission conversion factors 2019 - [DGAC](https://eco-calculateur.dta.aviation-civile.gouv.fr) Direction Générale de l'Aviation Civile - [ICAO](https://www.icao.int/environmental-protection/carbonoffset/pages/default.aspx) Carbon Emissions Calculator - L. Hackel [travel footprint calculator](https://lhackel.shinyapps.io/travel_footprint/) based on DEFRA emission factors - content: | ### Disclaimer I am obviously not a carbon footprint authority. I provide this as a service to members of the science community to get some ideas of travel footprints associated with scientific projects and activities. ### Confidentiality The data provided will remain confidential, as will be the results. There is a minimum distance (by default 300 km) under which the calculator excludes flight travel. The calculator proposes instead to compute the travel footprint associated with train. The French emission factors are between 4 and 5 grams of CO2eq per km per passenger. This low value is likely due to the fact that electricity is provided by nuclear plants. It is larger by some factor accross Europe. Here I am assuming that the French factor are multiplied by a factor of 5. This makes train typically 10% less emitting than plane (including radiative forcing). estimate: hero: title: Request an estimation description: | The results will available (almost) immediately.
It may take from a few seconds up to a few hours, depending on the amount of locations you provided. # Labels accept HTML, but not markdown # Descriptions accept neither, since we use the HTML title attribute form: email: label: Email Address description: Make sure you provide a valid address or you won't receive the results! first_name: label: First Name description: Also known as given name, eg. `Didier`. placeholder: Adèle last_name: label: Last Name description: Also known as family name, eg. `Barret`. placeholder: Bellego institution: label: Institution / Enterprise description: If any? comment: label: Leave a comment description: Any input is appreciated. Everyone's a critic. origin_addresses: label: Origin Cities description: | One address per line, in the form `City, Country`. Make sure your addresses are correctly spelled. # We MUST use the dumb CRLF pair for windows users placeholder: "Paris, France\r\nBerlin, Germany" # placeholder: | # Paris, France # Berlin, Germany destination_addresses: label: Destination Cities description: | One address per line, in the form `City, Country`. Make sure your addresses are correctly spelled. placeholder: | Washington, United States of America compute_optimal_destination: label: | Compute the destination city that will minimize emissions
(useful when setting up a meeting/conference) description: | We will only look through Cities specified in the Destination Cities. use_atmosfair_rfi: label: | Use the RFI multiplier recommended by atmosfair (i.e. 3 for all emissions above 9km)
For long flights, the multiplier may reach 2.8 or so. Otherwise, by default, 1.9 will be used. description: | We will only look through Cities specified in the Destination Cities. estimate_queue: hero: title: Please wait… description: | Your estimation is being computed. This may take several minutes.
This webpage will automatically update when it is done. estimation: hero: title: Your estimation is now available! description: | Thank you for using our service.
Bookmark this webpage, it is private and unlisted. failure: hero: title: Your estimation has failed! description: | Sorry about that. Please find the error message below.
Thank you for using our service. footer: credits: | Didier Barret © 2019