New EAT Lancet Diet Implementation

[FelicitasBeier]

🐦 Description of this PR 🐦

• New EATLancet diet implementation (required as part of the EAT2.0 project)
• The previous EATLancet (EAT 1.0) implementation was not a direct implementation of EAT targets, but based on scenario runs from IMPACT which in turn used the EAT Lancet targets. Now, the EAT targets (transformed to kcal) constrain p15_intake_detail directly in MAgPIE
• Currently, this new implementation can be selected via cfg$gms$s15_exo_diet <- 3, but the goal is to replace the old implementation by this one, i.e. delete code of cfg$gms$s15_exo_diet <- 1 to have no parallel implementations. Note is included in default.cfg for the switch
• Changes in the old diet implementation / default:

1. alcohol is reduced to zero by default when EATLancet diet is activated. This is in line with the EATLancet definition. If for a project, you would like to divert from this recommendation, you have to set the switch s15_exo_alcohol to 1.
2. brans does not have a target in EATLancet. It stays at its current level by default. If a reduction to 0 is wanted, s15_exo_brans has to be activated (set to 1) now.

• Differences between EAT1.0 and EAT2.0 (beyond technical implementation differences):

1. Starchy fruit didn’t have max. target
2. Roots didn’t have separate target (were parts of staples)
3. Soybean does not have separate target anymore. It’s combined with the legumes target
4. Peanuts does not have separate target anymore. It’s combined with nuts & seeds.

• Special treatment of EAT2.0 recommendation within MAgPIE:

1. We do not use gram targets, but kcal targets because MAgPIE calculates with kcal and treats the food products as homogenous good. We used the conversion from the Willet et al. paper
2. We do not scale up nuts in others following the EAT nut target, but they are scaled together with the fruits and vegetables because “other” is a homogenous good. “Fruits and Vegetables” together with “Bananas & plantains” have to meet EAT Fruit + Veggie target (with bananas & plantains having a maximum for EAT starchy fruit). The “nuts in other” then count towards the nut target and rapeseed and sunflower are scaled accordingly. The respective shares (of nuts in other and of bananas in cassav_sp) are country-specific because especially the bananas in cassav_sp vary by country. This is in line with EAT2.0 idea of regionalized diets and is consistent with iso-level food intake calculations

🔧 Checklist for PR creator 🔧

• Label pull request from the label list.

  • Low risk: Simple bugfixes (missing files, updated documentation, typos) or changes in start or output scripts
  • Medium risk: Uncritical changes in the model core (e.g. moderate modifications in non-default realizations)
  • High risk: Critical changes in model core or default settings (e.g. changing a model default or adjusting a core mechanic in the model)

• Self-review own code

  • No hard coded numbers and cluster/country/region names.
  • The new code doesn't contain declared but unused parameters or variables.
  • magpie4 R library has been updated accordingly and backwards compatible where necessary.
  • scenario_config.csv has been updated accordingly (important if default.cfg has been updated)

• Document changes

  • Add changes to CHANGELOG.md
  • Where relevant, put In-code documentation comments
  • Properly address updates in interfaces in the module documentations
  • run goxygen::goxygen() and verify the modified code is properly documented

• Perform test runs

  • Low risk:
    • Run a compilation check via Rscript start.R --> "compilation check"
  • Medium risk:
    • Run test runs via Rscript start.R --> "test runs"
    • Check logs for errors/warnings
  • High risk:
    • Run test runs via Rscript start.R --> "test runs"
    • Check logs for errors/warnings
    • Default run from the PR target branch for comparison
    • Provide relevant comparison plots (land-use, emissions, food prices, land-use intensity,...)

📉 Performance changes 📈

• Current develop branch default : ** mins
• This PR's default : ** mins

🚨 Checklist for reviewer 🚨

• PR is labeled correctly
• Code changes look reasonable
  • No hard coded numbers and cluster/country/region names.
  • No unnecessary increase in module interfaces
  • model behavior/performance is satisfactory.
• Changes are properly documented
  • CHANGELOG is updated correctly
  • Updates in interfaces have been properly addressed in the module documentations
  • In-code documentation looks appropriate
• content review done (at least 1)
• RSE review done (at least 1)

[bodirsky]

Mostly minor changes. Please separate fruitvegnuts and roots switch

[weindl]

Looks good in general. For more details, see comments.

[weindl]

replace "included in the diet shift towards cfg$gms$c15_EAT_scen" with
"included in the shift towards the selected scenario diet"

[weindl]

"their amount has been added to cassav_sp already". Here I am not sure. I think their amount will be added, in line 551.

[FelicitasBeier]

It is part of p15_intake_detail_fruitveg and p15_intake_detailed_scen_fruitveg already

[weindl]

I think you mean the condition $(sum(kfo_ns2, p15_intake_detail(t,iso,kfo_ns2)) < (f15_rec_EATLancet(iso,"t_nutseeds","min") - i15_intake_detailed_scen_nuts(t,iso))?

I think it is a good idea to avoid that i15_intake_detailed_scen_nuts over-fulfill the nuts and seeds target. However, your solution would not work in the case, that even without the scaling-up of the i15_intake_detailed_scen_nuts, the sum of the old "others-nuts" plus remaining nuts would be above the target. In this case, the current implementation would correct this unnecessarily down.

Maybe one can first substract i15_intake_detailed_scen_nuts from sum(kfo_ns2, p15_intake_detail(t,iso,kfo_ns2)):

p15_intake_detailed_scen_target(t,iso,kfo_ns)$(...)
= p15_intake_detail(t,iso,kfo_ns) / sum(kfo_ns2, p15_intake_detail(t,iso,kfo_ns2))
* (sum(kfo_ns2, p15_intake_detail(t,iso,kfo_ns2)) - i15_intake_detailed_scen_nuts(t,iso))
;
and then apply the constraint:

p15_intake_detailed_scen_target(t,iso,kfo_ns)$(sum(kfo_ns2, p15_intake_detailed_scen_targett,iso,kfo_ns)) < (f15_rec_EATLancet(iso,"t_nutseeds","min") - i15_intake_detailed_scen_nuts(t,iso)) = p15_intake_detailed_scen_target(t,iso,kfo_ns) / sum(kfo_ns2, p15_intake_detailed_scen_target((t,iso,kfo_ns2))
* (f15_rec_EATLancet(iso,"t_nutseeds","min") - i15_intake_detailed_scen_nuts(t,iso))
;

[bodirsky]

super feli!

[FelicitasBeier]

Thank you, @bodirsky and @weindl for this extensive review and discussions!