The impact of diurnal variations of air traffic on contrail radiative forcing

Abstract

Abstract. We combined high resolution aircraft flight data from the EU Fifth Framework Programme project AERO2k with analysis data from the ECMWF's integrated forecast system to calculate diurnally resolved 3-D contrail cover. We scaled the contrail cover in order to match observational data for the Bakan area (eastern-Atlantic/western-Europe).We found that less than 40% of the global distance travelled by aircraft is due to flights during local night time. Yet, due to the cancellation of shortwave and longwave effects during daytime, night time flights contribute a disproportional 60% to the global annual mean forcing. Under clear sky conditions the night flights contribute even more disproportionally at 76%. There are pronounced regional variations in night flying and the associated radiative forcing. Over parts of the North Atlantic flight corridor 75% of air traffic and 84% of the forcing occurs during local night, whereas only 35% of flights are during local night in South-East Asia, yet these contribute 68% of the radiative forcing. In general, regions with a significant local contrail radiative forcing are also regions for which night time flights amount to less than half of the daily total of flights. Therefore, neglecting diurnal variations in air traffic/contrail cover by assuming a diurnal mean contrail cover can over-estimate the global mean radiative forcing by up to 30%.

Parameterization of contrails in the UK Met Office Climate Model

“…One such a region is the western Europe area of 10°W to 23°E and 40°N to 56°N from Meyer et al [2002], where data of the Advanced Very High Resolution Radiometer (AVHRR) sensor onboard the NOAA 14 satellite was analyzed for the 1995-1997 period using an operational contrail detection algorithm. The 1985 average contrail coverage for this region reported by Meyer et al [2002] is 0.5%, while Bakan et al [1994] and Stuber and Forster [2007] reported values of approximately 0.7% and 0.9%, respectively. Our value for the same region is 0.97% for the year 1985.…”

“…This is caused by the fact that the contrail formation is affected by two independent factors, one being the amount of air traffic, and the other one being the ambient meteorological conditions. Thus, if the winter months correspond to low air traffic (with a minimum in December) and the summer months correspond to high air traffic (with a maximum in August), the meteorological conditions are less favorable to contrail formation in the three summer months, mainly because in these months the northern hemisphere midlatitudes upper troposphere relative humidity reaches its minimum [Marquart et al, 2004;Stuber and Forster, 2007]. The combination of these two independent factors results in the largest global monthly mean contrail coverages being recorded in June and October.…”

“…For these calculations, the natural clouds are based on monthly averaged distributions for the 1983-2002 period, for low-level, midlevel and high-level cloud from the International Satellite Cloud Climatology (ISCCP) project. For comparison, Stuber and Forster [2007] also used ISCCP data, whilst Myhre and Stordal [2001] employed 1996 natural clouds from the European Centre for MediumRange Weather Forecasts (ECMWF) analysis.…”

“…Existing studies have shown that for these persistent contrails, their daily average LW radiative forcing (RF) effect is larger than their SW effect, meaning that contrails cause a positive net RF, and therefore a warming [see, e.g., Meerkötter et al, 1999]. The magnitude of this net positive RF estimated by various studies for the air traffic of the year 1985, varies from a value of 2.0 mW m −2 [Stuber and Forster, 2007] to a value of 17 mW m −2 . According to the Intergovernmental Panel on Climate Change (IPCC) fourth assessment , the linear contrail radiative forcing for the year 2005 is estimated at 10 mW m −2 , with an uncertainty factor of 3 caused by a low level of current scientific understanding.…”

“…[26] Another test for the radiation code is performed by considering the Myhre and Stordal [2001] case, also repeated by Stuber and Forster [2007], where a 1% homogeneous contrail coverage is assumed at 250 hPa, with an optical depth t = 0.3 at 0.55 mm, an ice water content IWC = 21 mg m −3 , and a generalized effective size D ge = 23mm. For these calculations, the natural clouds are based on monthly averaged distributions for the 1983-2002 period, for low-level, midlevel and high-level cloud from the International Satellite Cloud Climatology (ISCCP) project.…”