Reanalysis-driven simulations may overestimate persistent contrail formation by 100-250%

Abstract Model-based estimates of aviation’s climate impacts have found that contrails contribute 36 – 81% of aviation’s instantaneous radiative forcing. These estimates depend on the accuracy of meteorological data provided by reanalyses like ERA5 and MERRA-2. Using data from 793,044 radiosondes, we find persistent contrails forming at cruise altitudes in 30N – 60S are overestimated by factors of 2.0 and 3.5 for ERA5 and MERRA-2 respectively. Seasonal and inter-annual trends are well-reproduced by both models (R2 = 0.79 and 0.74). We also find a contrail lifetime metric is overestimated by 17% in ERA5 and 45% in MERRA-2. Finally, the reanalyses incorrectly identify individual regions that could form persistent contrails 87% and 52% of the time respectively. These results suggest that contrail models currently overestimate the number and lifetime of persistent contrails. Additional observations are needed for future models to provide locally accurate estimates of contrails or support mitigation strategies.