about ERA52ARL conversion and output precipitation along trajectories

[muyl]

I am using ERA5 data as input for Hysplit model. I used the branch program you provide in ‘/hysplit/data2arl/era52arl’ to convert ERA5 GRIB files to ARL, at first, the ‘era52arl.cfg’ is produced autimativelly. I think it is right, so I did not change it. I use this command ’./era52arl -dera52arl.cfg -iDATA.GRIB -aSFC.GRIB -fSFC2.GRIB -oDATA.ARL’ to do the converting without any problem( the SFC.GRIB and SFC2.GRIB is the same, because I am a little confused about ” -f[input grib1 forecast surface fields name {SFC2.GRIB}]” in the Readme_era52arl.txt.), but when I use the ARL file to calculate trajectories, I found that the precipitation along trajectories is always zero, then I thought that maybe the reason is in ERA5 GRIB data, the unit of precipitation is ‘m’, so the number is too small. Then I multiply 1000 to ‘precipitation in ERA5 GRIB file, did the conversion again, then I used this new ARL in calculation, now the precipitation along trajectories seems right, but the value is not the same with in ERA5 GRIB or ERA5 netcdf file. For example, when I start a back-trajectory calculation for 22 03 01 00, at S32E160, along the trajectory, the output precipitation is 4.1, but in ERA5, it is 6.776.(To confirm this, the python code I use to read precipitation from ERA5 single pressure layer dataset it given below:

f = Dataset('/scratch/w40/ym7079/ERA5_nc/sfc/tp_era5_oper_sfc_20220301-20220331.nc',mode = 'r') # S32E160

pre_dataset = f.variables['tp'][0,488,1360]*1000

print('rainfall in dataset',pre_dataset)
6.7764 # in Hysplit, precipitation along trajectory at 22030100 S32E160 is 4.1mm

I

The other question is that in ARL, there is a new variable ‘DIFW’, I can not completely understand this, but the ‘Hysplit User guide’ does not give a detailed explaination.