Domain-average cloud Cover and cloud-radiative effect are derived with these python scripts
cd /vols/talos/home/fabian/proj/2017-07_icon-nawdex/inout
ll save_nwcsaf_histograms_nawdex_analysis.py save_ave_simulated_cre.py save_ave_observed_cre.py
These scripts contain analysis and output functions that should be merged into this library.
We have a set of notebooks
http://localhost:8787/notebooks/proj/2017-07_icon-nawdex/nbooks/12-plot_cc_tseries_for_simulationsets.ipynb
http://localhost:8787/notebooks/proj/2017-07_icon-nawdex/nbooks/15-plot_cloudtype_statistics.ipynb
http://localhost:8787/notebooks/proj/2017-07_icon-nawdex/nbooks/16-plot_cre_statistics.ipynb
that are very dirty - a lot of different tasks and duplciates in there.
They do
- input of CC and CRE data,
- stacking,
- set selection and collection --> output rearranged data
- plotting of CC and CRE vs. cloud type plots
It would be great
- to have a common format for set collections and full collections
- to include clear-sky fractions and CRE
- to define an extra data dimension of variables like LCRE, SCRE, etc or/and an extra dimension for experiments
- analyze clear-sky CRE --> bias in short- and longwave will give information about cloud masking, clearsky estimation
- check different scaling factors for observed CRE (1. vs. 0.85)
- develop a relative measure how a certain configuration change improves CC and CRE compared to a standard setup -- this means a standard or reference needs to be defined.