Strange data in tutorial for aggregating and stacked plots

[danielhuppmann]

Using the stack_plot() feature and experiencing some strange behaviour, I went to the aggregating_variables_and_plotting_with_negative_values tutorial notebook (#198) to see check whether I was doing something wrong.

I noticed that the test data are a bit ... counter-intuitive. Is this dataset based on real scenario data, or just completely made up?

Specifically:

• Units/values are not in line with actual global data
• Is 'Emissions|CO2|Agg' supposed to be agriculture?
• Why are specifically 'Emissions|CO2|Energy|BECCS' both positive and negative? I would expect this variable in particular to be negative-only...

@rossursino @znicholls, any comments?

[znicholls]

Is this dataset based on real scenario data, or just completely made up?

Completely made up, can edit however makes sense to you. We just chose the data to make sure the features we cared about all worked (e.g. it could handle timeseries which were only positive, only negative, started positive then went negative, started positive then went negative then went positive again).

[danielhuppmann]

Thanks @znicholls - see related issue #266 - stack_plot() completely removes timeseries that start with zero values.

Will see if I can fix that and will update the illustrative dataset in the tutorial as well.