Determine list of climate features to provide by default

[albangossard]

Depending on the different flavours of MBM, we have been using various climatic features so far. The goal of this issue is to determine which climatic feature should be retrieved by default so that the user only has to select among them.
For example for the moment in the Iceland preprocessing scripts we have ["t2m", "tp", "slhf", "sshf", "ssrd", "fal", "str"]
We should add pcsr which have already been used by @marvande.

@marvande @khsjursen @JordiBolibar @JoachimPiret feel free to jump into this conversation to suggest variables that you would like to see available by default.

[khsjursen]

From the overview of variables provided in ERA5-Land, here are the climate variables briefly summarized (explanations of the variables here):

Temperature and humidity:

• 2m temperature
• 2m dewpoint temperature
• skin temperature (refers to the temperature of the surface of the Earth - depends on the surface type and therefore the derived variable snow cover)

Wind:

• 10m u-component of wind
• 10m v-component of wind

Radiation:

• surface solar radiation downwards
• surface net solar radiation
• surface thermal radiation downwards
• surface net thermal radiation
• surface latent heat flux
• surface sensible heat flux
• forecast albedo

Pressure:

• surface pressure

Precipitation

• total precipitation

Snow:

• snowfall
• snow cover
• snow depth
• snow depth water equivalent
• snow density
• snow albedo
• snowmelt
• temperature of snow layer
• snow evaporation

I would suggest to just retrieve all of these variables, perhaps except the snow category (and skin temperature?). Personally, I don't think snow variables should be used if the goal is to infer mass balance from climate (and not through derived variables by ERA5), but they could also be included for those who would argue differently or want to use MBM for something else.

[albangossard]

I would suggest to just retrieve all of these variables, perhaps except the snow category (and skin temperature?). Personally, I don't think snow variables should be used if the goal is to infer mass balance from climate (and not through derived variables by ERA5), but they could also be included for those who would argue differently or want to use MBM for something else.

This depends on the motivations to build a statistical model. For example if the goal is to do a re-analysis and not to predict future MB, then it makes sense to incorporate those.
Given the list of variables that you sent, I think we could provide all of them for the sake of completeness.