Superdroplet time series

[jbarr444]

Hi @slayoo,

I was wondering how one might get the history of a superdroplet (its radius time series, for example) as a product of a simulation. Is this possible?

Thanks,
Jason

[slayoo]

Hi @jbarr444,

PySDM uses the term product to denote (cell/column/grid)-aggregated quantities, while attribute is what can be used to analyze the timeseries of particle properties, e.g. radius.

Accessing both can be done through the instance of Particulator with particulator.products and particulator.attributes.
Here you can find one relevant example: https://github.com/open-atmos/PySDM/blob/main/examples/PySDM_examples/Grabowski_and_Pawlowska_2023/figure_1.ipynb

Attributes are being saved here:

PySDM/examples/PySDM_examples/Grabowski_and_Pawlowska_2023/simulation.py

Lines 109 to 112 in 9e153ee

	for key, attr in self.output_attributes.items():
	attr_data = self.particulator.attributes[key].to_ndarray()
	for drop_id in range(self.particulator.n_sd):
	attr[drop_id].append(attr_data[drop_id])

In the Shipway & Hill examples, picking which attributes to save is controlled from here now:

PySDM/examples/PySDM_examples/Shipway_and_Hill_2012/simulation.py

Lines 178 to 183 in 9e153ee

	self.output_attributes = {
	"cell origin": [],
	"position in cell": [],
	"radius": [],
	"multiplicity": [],
	}

... but it probably would be worth to improve it to be user-controlled and not hardcoded.

Finally, note that there is a question of particle identity preservation upon collisions/breakup, here's a relevant discussion: #1361

HTH,
Sylwester

[jbarr444]

That helps a lot! Thanks as always.