Abade & Albuquerque example roadmap

[slayoo]

• [x] add ABIFM (time dependent) curves (water & ice) for Fig 2 (e.g., using other symbol and same color) - 5 datasets in both Homogeneous and Stochastic plot

• [x] consider running Fig 2 plot for different seeds and plotting range of values

• [x] depict how the realisation spread depends on n_sd

• [x] experiment with different cooling rates (new notebook: Fig 4?)

• [ ] start coding temperature attribute

• [ ] temperature attribute dynamics: relaxation, updraft (+fluctuation), phase changes

• [ ] refactor diffusion dynamics to use the temperature attribute

• [ ] start coding moisture fluctuation attribute (eqs 9, 10)

• [ ] refactor diffusion dynamics to use the new moisture attribute

so far, no change in result from the above 4 points

• [ ] draft a test with arbitrarily constructed (within reasonable limits) fluctuation and check departures of behaviour vs. no-fluctuation case
• [ ] draft logic foir the w' term (eq. 14)

[slayoo]

Design goals:

• [ ] avoid making the existing condensation/deposition code longer/more complex - all the new stuff should be optional and "external"

Questions:

• [ ] is mass and energy conservation ensured (i.e. total mass of vapour in all phases in the system should be constant, e.g.)?
• [ ] how is the ambient air <T> & <Q> calculated? does the weighted mean approach converge to PySDM original logic if no relaxation is done; how to implement it so that in case we switch the fluctuation logic off, there is no performance trade off