Please clarify a few things here:

"The ERA-Interim reanalysis data is also used in the calculation of 5-days backward air-mass trajectories with the HYSPLIT model using geopotential height, relative humidity, temperature, u/v wind components, vertical velocity at 37 pressure levels. The backward trajectories 5 are initialized at 12 UTC, at each grid point of the study area and at a subregional mean cloud-top altitude obtained from the CALIPSO Level-2 5 km layer cloud product (version 3, daytime). All meteorological variables are resampled from 0.5 degrees to the MODIS L3 resolution of 1 degrees and temporally averaged to the corresponding 8-day product."

1. The ERA-Interim data are provided at ~0.7 deg horizontal resolution, if it is the same dataset that I use in research.

2. The 37 pressure levels extend to 1 hPa (stratopause), presumably a small subset of these are used for trajectory analysis.

3. Presumably the trajectories are driven by 6-hourly analyses or perhaps interpolated to something finer in time, in any case, they are certainly not driven by 8-day averaged data.

A general suggestion: If possible, please identify supporting literature that would justify the use of 8-day meteorological averages. While cloud properties are justifiably driven by variations on all timescales, they also exhibit virtually instantaneous response to changes, e.g, near water-phase transitions!

Thank you, --Tim Dunkerton