Contrail coverage derived from 2001 AVHRR data over the continental United States of America and surrounding areas

Linear contrail coverage, optical depth, and longwave radiative forcing are derived from NOAA-15 and NOAA-16 Advanced Very High Resolution Radiometer data taken during daytime over the continental United States of America (USA), southern Canada, northern Mexico, and the adjacent oceans. Analyses were performed for all available overpasses during 2001, but for NOAA-15 were primarily limited to the eastern half and the northwestern corner of the domain. Contrail coverage averaged 1.17% and 0.65% from the early morning NOAA-15 and midafternoon NOAA-16, respectively, for the areas and month common to both satellites. The NOAA-16 contrail coverage and radiative properties for the limited NOAA-15 domain are, on average, nearly identical to those for the entire domain. The estimated combined maximum coverage for the entire domain was similar to 1.05% during February, while the minimum of 0.57% occurred during August. Mean optical depths varied by similar to 20% with winter minima and summer maxima. The annual mean optical depth of 0.27 translated to a normalized contrail longwave radiative forcing of 15.5 Wm(-2). The overall daytime longwave radiative forcing for the domain is 0.11 Wm(-2). The normalized radiative forcing peaked during summer while the overall forcing was at a maximum during winter because of the greater contrail coverage. A detailed error analysis showed that the linear contrail coverage was overestimated by similar to 40% for both satellites the true coverage is closer to 0.70 and 0.40% for NOAA-15 and 16, respectively. Errors in the mean NOAA-15 optical depths and radiative forcing were negligible while their NOAA-16 counterparts were overestimated by approximately 13%. Contrail coverage was dramatically lower than expected from previous studies, but is most likely due to the significant decrease in upper tropospheric humidity observed in numerical weather analysis data. Contrail optical depths are much greater than both theoretical estimates for this part of North America and empirical retrievals over Europe. The cause of the morning-afternoon difference in contrail coverage is not yet known. Further modelling studies and additional satellite analyses are needed to understand this diurnal cycle and to explain the differences between the present and previous results.