Parameter Setting in modules 15_food/ exodietmacro.gms (line 382)

[Light-XY]

I am a user of the MAgPIE model (4.10.1). While reading the MAgPIE model code, I came across a parameter that I do not fully understand.
Specifically, in the file magpie\modules\15_food\anthro_iso_jun22\exodietmacro.gms at line 382, the following line appears:
i15_intake_EATLancet(iso,kfo) = i15_intake_EATLancet_all(iso,"2100kcal","%c15_EAT_scen%",kfo)
I am wondering whether the parameter "2100kcal" is correct in this context. Should it perhaps be "%c15_kcal_scen%" instead?

[tscheypidi]

Hi, could you please provide some context. What settings did you use to run MAgPIE? What config settings differed in your setup from the default.cfg?

[Light-XY]

Actually, this is a problem I found when reading the code to understand the algorithm. When running the demo, I found that no matter whether 2100kcal or 2500kcal is passed into the variable cfg$gms$c15_kcal_scen, it reads 2100kcal data in i15_intake_EATLancet(iso,kfo). In addition, I found that 2100kcal is also written in the model documentation (https://rse.pik-potsdam.de/doc/magpie/4.11.0/15_food.htm). This means that during operation, if cfg$gms$s15_exo_diet <- 1, the parameter cfg$gms$c15_kcal_scen does not play a role when reading i15_intake_EATLancet_all data.

[bodirsky]

Thanks for pointing to this.
The switch cfg$gms$c15_kcal_scen is indeed only used to set the total calories for a diet, not the composition.
It plays a role here:
$else
i15_intake_scen_target(t,iso) = sum(kfo,i15_intake_EATLancet_all(iso,"%c15_kcal_scen%","%c15_EAT_scen%",kfo));
p15_bmi_shr_target(t,iso,sex,age,bmi_group15)=0;
p15_bmi_shr_target(t,iso,sex,age,"medium")=1;
$endif
Putting into the equation you mentioned the parameter cfg$gms$c15_kcal_scen instead of "2100kcal" would not work for the other settings of c15_kcal_scen which are not included in this file.

Using the 2100 kcal should not make a big difference, because the approach of Marco Springmann to fill up the remaining calories is the same as ours: filling it up with staples. Apparently, there is only evidence that a certain amount of calories of specific items is healthy, but not that this quantity depends on bodyweight. Therefore the rest can be filled with staples.

We dont usually use this scenario setting, because we rather trust on our own estimates of total calories, which are consistent with underweight/overweight estimates. Nor sure what your application of the model is, but I would recommend this setting for you, too.

[Light-XY]

Got it, Thank you very much!