Model output associated with "Inter-comparison of Atmospheric Trace Gas Dispersion Models: Barnett Shale Case Study", Atmospheric Chemistry and Physics, 2019.

Please refer to the publication, particularly the information in the
Supplementary Information, for details on the model configuration. Footprint files:
Footprints generated every 30 seconds along flight paths using different models are
archived as netCDF files. Each tar.gz file contains all the netCDF footprints for a
particular flight and model combination, and is named after the model combination and
flight date (YYYYMMDD). For example the WRF-HYSPLIT combination for the flight on
10/19/2013 referenced in the paper is in: WRF-HYSPLIT/WRFHYS_Footprints_20131019.tar.gz
Most flight footprints were run for 24 hours back in time but there are a few
exceptions, so the files must be opened to see how far back the footprints go. A few
flights use shorter times, and on 10/28/2013 WRF-HYSPLIT was run for 36 hours. The
HYSPLIT footprint files are numbered consecutively in order of time. The location and
time of the receptor is in the NetCDF file itself, not in the filename. The STILT
filenames have location and time information but only to the nearest minute, so the
numerical order in the filename indicates their time order, because there are two
footprints each minute. STILT and HYSPLIT footprints all also contain particle
trajectories. The receptor location and time is in the NetCDF file as origutctime,
origlat, origlong, etc. The files include footprints from basic model runs referenced in
Figures 4 and 7, that is NAMS-HYSPLIT, WRF-HYSPLIT, and WRF-STILT. WRF-STILT for flights
on 3/25, 3/27, 3/30 and 10/16 uses instantaneous wind output; WRF-STILT for 10/19,
10/20, 10/25, and 10/28 flights uses averaged wind fields (see the SI in the manuscript
for details). For WRF2-Flexpart footprints, please contact [e-mail]. For
CarbonTracker Lagrange WRF-STILT footprints, contact [e-mail]. For
WRF-LPDM footprints, contact Thomas Lauvaux ([e-mail]). WRF-Chem
output files: WRF-Chem output for 4 flights in October 2013 is also included here, both
in native NetCDF format and in ARL format. These contain 3-km resolution, hourly
WRF-Chem output. ARL files can be used to run HYSPLIT or STILT. The NetCDF formatted
files include the tracer variables for CH4 from both the EPA inventory (tracer_1
variable) and the EDF inventory used in the paper (tracer_2 variable). The third tracer
(tracer_3) is using the EPA inventory but masked so that emissions outside the domain of
the EDF inventory are zero (so that they can be directly compared). Only tracer_2 is
used in the manuscript. The NetCDF output files are combined into tar.gz files. Please
contact Thomas Lauvaux ([e-mail]) for details regarding WRF-Chem.
Observations: Observations from the Barnett flight campaign are made available by
NOAA/ESRL, please contact [e-mail].