Direct emissions chart and dashboard items, mapping file, 1990 data

[kndehaan]

Context

This PR adds the queries for a new chart and dashboard items, using a newly introduced mapping file structure.

Implemented changes

• Queries for GHG emissions chart and dashboard items
• New mapping structure including mapping file and new sector label attribute
• Modified setup of the data export using the mapping structure
• 1990 emission data added to emissions.csv, following the emissions data pipeline

Related

Goes with pull requests:

• Emissions atlas#199
• Emissions etengine#1779
• Direct emissions chart and dashboard items, mapping file, 1990 data #3463
• Direct emissions chart and dashboard items etmodel#4753
• https://github.com/quintel/etdataset/pull/1095
• Direct emissions chart and dashboard items documentation#343

Checklist

• I have tested these changes
• I have updated documentation as needed
• I have tagged the relevant people for review

[mabijkerk]

I had AI do a full review of the branches, it flagged the following:

LULUCF removals get the wrong sign in the mapped form

The two 1990 LULUCF queries disagree about what "LULUCF" means:

• direct_emissions_lulucf_co2_1990 names the pairs explicitly and negates
  removals:
  SUM(EMISSIONS(lulucf_emissions, …), NEG(EMISSIONS(lulucf_removals, …)))
• direct_emissions_lulucf_other_ghg_1990 uses the new mapped form
  EMISSIONS(emissions_sector, "LULUCF", other_ghg, 1990), and sum_pairs
  sums over both lulucf_emissions and lulucf_removals.

It's harmless today only because no dataset has an
LULUCF,Removals,…,other_ghg row — nl has removals for co2 only. The day
someone adds one, it gets added instead of subtracted, silently. Worth either
using the explicit-pair form for consistency, or leaving a comment noting the
mapped form is only safe while removals stay CO2-only.

[mabijkerk]

Is this still valid?

[kndehaan]

It's still valid, it gives an example of how we could add a new column to the data export and how this relates to the columns in the mapping file.

[kndehaan]

It's harmless today only because no dataset has an
LULUCF,Removals,…,other_ghg row — nl has removals for co2 only. The day
someone adds one, it gets added instead of subtracted, silently.

By IPCC definition, removals in LULUCF only apply for CO2. It's not possible to have removals for other GHG. I updated the descriptions in the doc strings to further clarify this.

Note to ourselves: good practice to provide clear doc strings when queries deviate from "standard queries". This will also gives context to the AI review to understand why the query is different.

[mabijkerk]

Let's do it!