IVL Swedish Environmental Research Institute

The OP-FLYKLIM project investigated the potential to reduce the climate impact of aviation through climate optimization of flight routes to reduce the high-altitude effects of aviation with a focus on climate forcing from contrails and contrail cirrus under Scandinavian conditions. We have developed a calculation methodology where areas with potential to form persistent contrails are identified. The duration and climate forcing of contrails and contrail cirrus in these areas are calculated using data from SMHI's meteorological forecast model. Information on the position and climate forcing potential of these areas has been used to quantify climate forcing of flights on selected routes over a period of several months, and to test optimization of route planning for reduced climate effect with the flight planning system used by the airline Novair.

Climate forcing from contrails and contrail cirrus during the flight calculated with the OP-FLYKLIM methodology is compared with calculations of climate forcing from the CO2 emitted from combustion of the jet fuel. This enables a direct comparison of the climate benefit of avoidance of contrail formation with its fuel penalty. In the future this method could be deployed in flight planning systems to enable climate optimization. The method can also be used in cost-benefit analyses of climate-optimized flight planning.

We have also investigated several issues that are important for route optimization in general and for correct assessment of whether persistent contrails occur. Meteorological models of good quality in terms of forecasts of winds, temperature and humidity at flight altitude is of great importance both for ordinary route planning and for climate optimization. In OP-FLYKLIM, SMHI has tested streaming data from aircraft (so-called Mode-S EHS data) through air traffic control radars and local data receivers directly to their operational forecast model, which showed improved quality of forecasts.A persistent contrail occurs only if the humidity in the area of the flight is supersaturated relative to ice but is not already containing clouds.

In the project, we have thus investigated the distribution of ice-saturated areas across Scandinavia as an average over several years using data from the ECMWF global numerical weather prediction (NWP) model. The results show a quite high potential for the formation of persistent contrails and thus for high-altitude effects in the area. Comparison to published data on the frequency of occurrence of ice supersaturated layers over Sweden and Europe indicate that observations and model data are broadly consistent. However, when comparing to observed relative humidity with respect to ice (RHI) from radiosondes directly it becomes clear that both the ECMWF model and the MetCoOp model used by SMHI for short range forecasts underestimate RHI near the tropopause, where most flights take place.

As an additional means to evaluate the performance of NWP models with respect to ice supersaturation SMHI initiated observations of contrails by their climate observers. The observations were then matched against flights in the area and RHI calculated by the SMHI forecast model to determine if observations of persistent contrails also corresponded to ice supersaturation in the model. In agreement with the evaluation against radiosondes it was found that the NWP model underestimated RHI in connection with observed contrails.

A correct calculation of fuel consumption and emissions during the flight is a prerequisite both for calculating its high-altitude effects and for monitoring of aviation emissions by national and international authorities. In OP-FLYKLIM the fuel consumption calculated with FOI3 methodology, used for the Swedish reporting of the national emissions from aviation to the UNFCCC and other international reporting obligations, has been compared with true fuel consumption obtained from data from the flight data recorder (FDR data) onboard aircraft on several routes. Comparison showed differences below 10% that could be explained by differences between route plans and type of aircraft in the FDR data and the FOI calculation, respectively.