Improving contrail avoidance through targeted deployment of humidity sensors on aircraft and advanced weather data assimilation.

Approximately two-thirds of aviation’s climate impact are attributed to contrail cirrus, the clouds resulting from the persistence of initially linear contrails [1]. This persistence can only occur in regions of the atmosphere where the relative humidity w.r.t. ice (RHi) is larger than 100%: these regions are known as ice supersaturated regions (ISSRs). The climate impact of contrail cirrus could be mitigated by means of minor trajectory adjustments that re-route aircraft around ISSRs (or subsets of these regions) at the cost of minor increases in fuel burn [2, 3]. The viability and effectiveness of this mitigation option rely on skillful forecasts of these regions: existing approaches to numerical weather prediction (NWP) have been found to be relatively poor at forecasting such regions of contrail persistence [4, 5]. This lack of skill is attributed to the coarse spatial resolution of such NWP model, simplified treatment of ice clouds, and the scarcity of high-quality measurements of humidity in the upper troposphere. 

This study evaluates the ability of in-flight humidity measurements to improve humidity forecasts, and contrail avoidance relying on these forecasts, through an Observation System Simulation Experiment (OSSE). An original RHi forecast is updated by employing 4D-Var data assimilation of simulated humidity data into ERA5 ensembles. A proof-of-concept case study is first presented, using real IAGOS flights: humidity observations from one transatlantic IAGOS flight are used to improve the humidity forecast for another, temporally adjacent, IAGOS flight. The updated forecast, based on the first flight observations, is validated against the independent RHi observations from this second flight, showing an improvement in RMSE relative to the original forecast. 

Next, the climate impact and operational cost of the improvements to the forecast are assessed for various simulated scenarios at a fleet-wide scale. The scenarios consist of: 1) 6 different levels of allocation of aircraft in the fleet that are equipped with humidity sensors (0%, 20%,40%, 60%, 80%, 100% of fleet penetration), 2) 4 levels of discrepancy between humidity forecasts and ground truths (10%, 30%, 50%, 70% recall on ISSR prediction), and 3) 3 levels of simulated accuracy of the humidity sensors used (3%, 6%, 12%). 

Results show that targeted sensor allocation among a fleet yield increased persistent contrail reduction under realistic forecast discrepancies, supporting scalable aviation climate strategies using on-situ humidity measurements. 

[1] Lee et al. (2021). The contribution of global aviation to anthropogenic climate forcing for 2000 to 2018. Atmospheric Environment 244, 117834. https://doi.org/10.1016/j.atmosenv.2020.117834 

[2] Teoh et al. (2020). Beyond contrail avoidance: Efficacy of flight altitude changes to minimise contrail climate forcing. Aerospace, 7(9), 121. https://doi.org/10.3390/aerospace7090121 

[3] Frias et al. (2024). Feasibility of contrail avoidance in a commercial flight planning system: An operational analysis. Environmental Research: Infrastructure and Sustainability, 4(1), 015013. https://doi.org/10.1088/2634-4505/ad310c 

[4] Gierens et al. (2020). How well can persistent contrails be predicted? Aerospace, 7(12), 169. https://doi.org/10.3390/aerospace7120169 

[5] Geraedts et al. (2024). A scalable system to measure contrail formation on a per-flight basis. Environmental Research Communications, 6(1), 015008. https://doi.org/10.1088/2515-7620/ad11ab