Abstract/Summary

Multiple studies have considered whether increased anthropogenic CO2 will affect the wind speeds and turbulence associated with the winter North Atlantic polar front jet stream in the upper atmosphere. Key questions are whether any effects can already be seen and, if so, can they be seen independent of computer models of the atmosphere. In this study we use two reanalyses, NCEP/NCAR and the ECMWF ERA5, and two large observational archives, AMDAR/ACARS and the Global Aircraft Data Set (GADS), to try to answer these questions for the period 2002-2020 when automated aircraft observations were plentiful over the North Atlantic. We focus on eastbound, New York to London, flights. No significant increase appears in reanalyses during the last roughly 40 years (1979-2020) which is our best estimate for the modern satellite era. In contrast, for the last roughly 20 years (2002-2020) both the ERA5 reanalysis (2.5% per year) and the GADS archive (1.2% to 1.4% per year) show a statistically significant rise in the wind speed in the North Atlantic jet streak exit region. These results must be considered in the context of atmospheric oscillations, changes to the North Atlantic Track System (NATS), and the effects of aircraft step climbs. We estimate that up to 0.5% of the rise may be due to improvements in the NATS operations and an unknown additional amount may be due to the substantial increase in automated aircraft observations starting in 1997. We also examine the impact of aircraft observations on one’s confidence in drawing conclusions from secular changes in the reanalyses. For turbulence, the Light turbulence trends are not statistically significant. Our confidence in the turbulence results is more limited since these observations reflect medium-term changes in tactical and strategic aircraft operational procedures as well as the underlying prevalence of turbulence.