The dataset features 8 hours of single layer shallow cumulus clouds with an ever increasing cloud deck. After 8 hours, we increased the output interval to 1s for 2 minutes. The domain was initialized with dropsonde data from the EUREC4A campaign averaged over the HALO-0128 flight. A key feature of the dataset is its very high spatial and temporal resolution of 3D output fields (10m/5m and 1 second). The vision is that the high temporal frequency and spatial resolution of cloud and wind variables will allow for a wide range of offline benchmarks. Applications that come to our mind are offline benchmark for 1D and 3D radiative heating rate computations, 3D radiative transfer effects on retrievals as well as cloud motion tracking algorithms.

PALM-LES / EUREC4A shallow cumulus dataset with 3D cloud output data

Authors: Fabian Jakub and Lea Volkmer (LMU - MIM)

Contact: Fabian.Jakub@physik.uni-muenchen.de, L.Volkmer@physik.uni-muenchen.de

The dataset features 8 hours of single layer shallow cumulus clouds with an ever increasing cloud deck. After 8 hours, we increased the output interval to 1s for 2 minutes. The domain was initialized with dropsonde data from the EUREC4A campaign averaged over the HALO-0128 flight. A key feature of the dataset is its very high spatial and temporal resolution of 3D output fields (10m/5m and 1 second). The vision is that the high temporal frequency and spatial resolution of cloud and wind variables will allow for a wide range of offline benchmarks. Applications that come to our mind are offline benchmark for 1D and 3D radiative heating rate computations, 3D radiative transfer effects on retrievals as well as cloud motion tracking algorithms.

---

The dataset is comprised of the following directories/files:

• spinup - PALM input files and outputs for the first phase of the simulations (8h)
• highres - PALM input files and outputs for 2 minutes with outputs every second
• overpass_images/ - virtual aircraft camera to get a feel for the cloud structure
• palm_input/ - PALM input files
• <>.ts.nc - Timeseries
• <>.pr.nc - Vertical profiles
• <>.3d.nc - 3D fields

---

---

Notable Variables in the 3D file:

varname	dim	description	units
z	512	vertical layers (stretched)	5m at the surface
x	2560	horizontal grid	dx 10m
y	1280	horizontal grid	dy 10m
q	time, z, y, x	Total water mixing ratio	kg/kg
qc	time, z, y, x	Liquid water mixing ratio	kg/kg
nc	time, z, y, x	Rain-drop number mixing ratio	#/m3
theta	time, z, y, x	Liquid Water Potential temperature	K
u	time, z, y, x	Horiz. wind velocity	m/s
v	time, z, y, x	Horiz. wind velocity	m/s
w	time, z, y, x	Vertical wind velocity	m/s

---

The folder plots/overpass_images/ contains images as they would be seen by the specMACS camera system on the HALO research aircraft. Images were generated by MonteCarlo raytracing with libRadtran/MYSTIC. For more details about the rendering contact L.Volkmer@physik.uni-muenchen.de.

Download_movie

---

Example to work with the data with python/xarray:

import xarray as xr
ds = xr.open_mfdataset('highres/data/w/*.nc', concat_dim='time', combine='nested', decode_timedelta=False)
ds.w.isel(time=-1).sel(zw_3d=1000).plot()

---