Sabre's flight carbon calculator described by Oxford study as the most detailed and accurate available on GreenAir Online

Mon 23 Mar 2009 – Sabre Holdings, the US travel technology organization, has developed an airline carbon calculator which has been described as the most detailed currently in existence in a new study by the University of Oxford’s Environmental Change Institute (ECI). The model draws on Sabre’s database used for its worldwide computer reservations system that contains information on all flights including date of travel, airline, departure point and destination, as well as model and seating configuration of the plane used for the flight.

The carbon calculator powers four CO2 reports now available through Sabre’s ‘Traveler Security and Data Suite’. Launched in 2007, it is used around the world by Sabre travel agencies and corporations to determine the impact their travel has on the environment. Users can view the total volume of carbon emissions produced for air, car and hotel travel across a date range or specific destinations travelled.

Sabre says it researched international guidelines and third party offerings before opting to build its own carbon calculation model in-house, drawing from protocols established by the Intergovernmental Panel on Climate Change (IPCC) and ICAO, as well as data and models from the FAA and Eurocontrol.

“In developing our solution, we recognized the need for a single industry standard on how air emissions and other elements of travel are measured and reported,” commented Tom Klein, Executive Vice President of Sabre Holdings. “This is something we would like to explore with our customers and industry associations, particularly as the demand for carbon and environmental reporting continues to grow.”

The study, ‘Calculating the Carbon Dioxide Emissions of Flights’, has been carried out by Dr Christian Jardine, a senior researcher at the ECI. He says that whilst calculator developers are increasingly transparent about the assumptions they make, and the reasoning behind them, there is as yet no internationally agreed and adopted methodology for the calculation of aviation emissions.

“Different greenhouse gas emissions calculators give widely varying results for the same flight due to variations in the underlying assumptions made in the calculator methodology,” says the study. Cited is the example of a return flight from London to New York in which two different calculators estimate the emissions to be 1.53 or 3.48 tonnes of CO2e, a variation of more than a factor of 2.

The problem arises because of the number of assumptions that have to be made and also uncertainties over factors such as the use of a multiplier to account for the effects of aviation emissions at altitude.

These effects will average themselves out over multiple flights so that the calculated value will still represent a good estimate of an average flight. However, there are a series of factors that influence per passenger emissions including:

Average emissions calculators therefore have to make assumptions about each of these factors, which introduce considerable errors and variations between methodologies, says the study. A standard methodology might make assumptions about which type of planes fly short-haul and long-haul routes, and how many seats would be on board a ‘typical’ plane. Freight load data, by weight, is also extremely rare in the public domain, it claims, so allocating a proportion of emissions to freight is also a loose approximation.

Emissions are also highly sensitive to the chosen plane model and there can be a factor of 2 between the most and least efficient plane models flying the same distances. Fuel burn data is publicly available for many plane models but these datasets are now becoming dated and do not include more modern models, such as the Boeing 737-800 or Airbus A380. “This is likely to lead to an overestimate in emissions as newer, more efficient planes are not represented,” says Dr Jardine.

There is also variation in sophistication between emission calculator methodologies in the way emissions are calculated as a function of distance and there is also a distinction as to whether emissions are allocated per passenger or per seat.

Emissions allocated per passenger will account for all emissions from the plane and allocate them to a sold ticket, but requires an assumption to be made about the likely percentage plane occupancy. Emissions allocated per seat make no assumptions about flight occupancy and allocate a proportion of emissions to those filled seats, but emissions allocated to unfilled seats are not accounted for. For offsetting and reporting purposes, allocating emissions to seats is preferable, suggests the study, because the customer is not answerable for how the airline is filling the other seats on the aircraft. “The traveller is responsible solely for the carbon emissions for the seat they occupy.”

The study reviews the approach of some of the more commonly used emissions calculators, including the UK Government’s Defra model and the ICAO calculator.

The study (which, it should be noted, was funded by Sabre Holdings) concludes that the Sabre Holdings model achieves a more detailed and accurate estimation of emissions because many of the unknown parameters from the passenger viewpoint are known in the Sabre database. This is because of the availability of two high-quality and detailed data sources, namely the SAGE model and the Passenger Name Record (PNR).

The System for assessing Aviation’s Global Emissions (SAGE) was developed by the FAA as a tool to examine annual global emissions, but because global emissions are calculated as the sum of many individual flights, it allows scenarios to be disaggregated to regional, national, airport and individual flight levels. Scenarios may, for example, examine the influence of policy measures, technological development, changes in the fleet stock and operational practices.

“This high fidelity of the model is also ideally suited for emissions calculators,” says the study. “Fuel burn data at the individual flight level, based on aircraft type, represents a quantum leap in sophistication for emissions calculators, hitherto based on crude averages and assumptions.”

The PNR contains information about individual flights and is utilized for booking flights for passengers, with data on the point of origin, destination, airlines, plane type used and can access seating configuration. The latter two parameters, used in conjunction with SAGE, can provide accurate CO2 emissions calculations on a flight by flight basis, claims the study.

A number of other advantages are claimed for the Sabre Holdings model. Because the information is known months in advance when the flights are scheduled, CO2 emissions can be calculated and displayed to the customer in advance, allowing them to incorporate the environmental impact of the flight into their purchasing decision. Furthermore, says the study, the extra detail in the model alters the decision-making process around the environmental impact of a flight from one of “fly vs. don’t fly” to one that is more refined. It can also allow customers to choose lower carbon flights, which would create a market pull towards more environmentally benign airlines. Another benefit is that it allows comparison with the CO2 emissions of other transport modes, such as rail.

“Because the Sabre Holdings model is more sophisticated than other aviation CO2 emissions calculators, this could provide a potentially unifying approach to calculator methodologies and remove the misleading variation between calculators that is observed today.

“At a point when there is a demand for greater consistency between them, so as not to confuse the market, the development of a detailed, high accuracy carbon calculator is exceedingly timely,” concludes the study.