Radiation, Clouds, and Self-Aggregation in RCEMIP Simulations

This dataset includes the numpy arrays used in plotting the figures in the paper: "Radiation, Clouds, and Self-Aggregation in RCEMIP Simulations".

Each zipped file contains the arrays used to create one figure. The units are always SI. A description of the format of each array is given below.

Fig01

"<cloud_scheme>_cloud_components_for_<cloud_type>.npy"

Has shape (n_cloud_types, n_CRMs)

Cloud types are in the same order as displayed on the x-axis of the figure. CRMs are in order of those displayed in figure 3. DAM and MESONH are missing due to not having 3D data.

Fig02

"<cloud_scheme>_<model_type>_cloud_fraction_vs_FMSE.npy"

Shape = (100, <n_cloud_types>)

100 FMSE percentiles from 0 to 99, 4 or 8 cloud types depending on the cloud scheme. Value shows fraction of cloud type at that FMSE percentile.

Fig03 & Fig04

"<model>_<SST>_var_normFMSE.npy"

Time series of var(hn) without any smoothing in time. First 2 days (8 time steps) are not included.

Fig05

"<model_type>_<aggregation_phase>_<SST>_budget-terms.npy"

Shape = (5, n_models)

First dimension is var(hn) term, LW term, SW term, SEF term, Advection term.

Models are in order of those shown in figures 3 and 4 depending on the model_type.

"<model_type>_<aggregation_phase>_<SST>_budget-terms_ClearSky.npy"

Shape = (2, n_models)

First dimension is LW term, SW term.

Fig06

Each subplot can be made by plotting the arrays on either the x or y axis.

Fig07 & Fig08

"<model_type>_<aggregation_phase>_<SST>_<variable>.npy"

shape = (4, n_models)

first dimension is the cloud types: Clear, Shallow, Other, Deep

models in the same order as shown in figures 3 or 4 depending on the model_type.