ENV_ACT troubles

[Neha Jaggeshar]

Hi all, 

I'm having a little problem with env_act in my model: for a process FORStock with input FOR-Managed and emission FORSINKCO2, i have defined ENV_ACT~FORSINKCO2~2018. Also included the known emission intensity for years 2020 and 2025. 

The problem is that the model keeps using the 2018 value for 2020 and 2025, ignoring their input entries (the values show up in the items detail table). 



I am not sure what could be the reason. Any idea what might be causing this? 

Thank you, 
Neha 

[Antti-L]

The only idea I have is that the process might be vintaged, and the capacity might consist of a single vintage only (e.g. 2018), which would result in the FLO_EMIS being applied according to the 2018 vintage. But the information you have given is not sufficient for making any such conclusion.

When a TIMES process is vintaged, any FLO_EMIS defined for it is also vintaged. As we have discussed before, that is very useful for land-use modeling, because with vintaged processes one can well model the annual growth in the forest stock by the age of each forest vintage. But to do so, you would need to use a SHAPE parameter for shaping the FLO_EMIS parameter by the age of each vintage, and you would also need to make sure that the existing forests and any regenerated forests after harvesting are properly allocated to vintages. But otherwise, with vintaged processes it should be quite feasible to model any such land-use emissions (representing the growth or decrease in the carbon stocks) by forest age and taking into account forest management.

However, should the process happen to be only inadvertently vintaged, then just defne it non-vintaged, and then your FLO_EMIS will be period-dependent instead of vintage-dependent. You would thus see the FORSINKCO2 emissions changing by the "operating" year (instead of the vintage year) according to your parameter specification.

Notwithstanding, if it so happens that the process is indeed vintaged on purpose, but you would still like to model the net growth in the forest stock with only a non-vintaged FORSINKCO2 output flow parameter (ignoring the age of each forest vintage), that is also possible (I am not going into the details of that option now, because ignoring the age of a vintaged process would seem rather unusual for forest modeling).

[Neha Jaggeshar]

The only idea I have is that the process might be vintaged, and the capacity might consist of a single vintage only (e.g. 2018), which would result in the FLO_EMIS being applied according to the 2018 vintage. But the information you have given is not sufficient for making any such conclusion.

When a TIMES process is vintaged, any FLO_EMIS defined for it is also vintaged. As we have discussed before, that is very useful for land-use modeling, because with vintaged processes one can well model the annual growth in the forest stock by the age of each forest vintage. But to do so, you would need to use a SHAPE parameter for shaping the FLO_EMIS parameter by the age of each vintage, and you would also need to make sure that the existing forests and any regenerated forests after harvesting are properly allocated to vintages. But otherwise, with vintaged processes it should be quite feasible to model any such land-use emissions (representing the growth or decrease in the carbon stocks) by forest age and taking into account forest management.

However, should the process happen to be only inadvertently vintaged, then just defne it non-vintaged, and then your FLO_EMIS will be period-dependent instead of vintage-dependent. You would thus see the FORSINKCO2 emissions changing by the "operating" year (instead of the vintage year) according to your parameter specification.

Notwithstanding, if it so happens that the process is indeed vintaged on purpose, but you would still like to model the net growth in the forest stock with only a non-vintaged FORSINKCO2 output flow parameter (ignoring the age of each forest vintage), that is also possible (I am not going into the details of that option now, because ignoring the age of a vintaged process would seem rather unusual for forest modeling).

Ah, i understand it now. Thank you, Antti! 
Indeed, I do have the process as vintaged and a shape profile defined.

I have also included the historical past forest investments (NCAP_Pasti for 1990-2018) and known investments for 2020 and 2025.

Wrt what you've said "capacity might consist of a single vintage only (e.g. 2018), which would result in the FLO_EMIS being applied according to the 2018 vintage" - can it be said that defined FLO_EMIS values for later years (say 2020) will be applied to capacity built in that year? Or it still retains the 2018 FLO_EMIS value and applies the shape factor accordingly? 
This is just to clarify my understanding of this part 

[Antti-L]

> "Indeed, I do have the process as vintaged and a shape profile defined."

Good, but then it sounds a bit strange that you say "the model keeps using the 2018 value for 2020 and 2025".

> "I have also included the historical past forest investments (NCAP_Pasti for 1990-2018) and known investments for 2020 and 2025."

Good, but again a bit strange if "the model keeps using the 2018 value for 2020 and 2025". That would suggest that only the 2018 vintage is nonetheless getting used? I guess you should make sure that the capacity of each vintage is also getting used, for example a fixed AF(FX) might be useful.

> "can it be said that defined FLO_EMIS values for later years (say 2020) will be applied to capacity built in that year?"

Any FLO_EMIS(r,v,p,cg,com,ts), where v stands for the vintage, is applied to the activity of that vintage (which is in turn limited by the corresponding capacity), with any SHAPE profile defined so that the emission factor will have an age profile.  So, strictly speaking it is of course applied to the activity, but the activity of vintage v is constrained by the capacity of that vintage.

[Neha Jaggeshar]

> "Good, but then it sounds a bit strange that you say "the model keeps using the 2018 value for 2020 and 2025."

In that run I had vintage and shape turned off, which still is strange that the FLO_EMIS values for 2020 and 2025 were not applied. Quite possible I made a mistake somewhere.

> "Good, but again a bit strange if the model keeps using the 2018 value for 2020 and 2025. That would suggest that only the 2018 vintage is nonetheless getting used? I guess you should make sure that the capacity of each vintage is also getting used, for example a fixed AF(FX) might be useful."

As far as I’ve checked, the capacity for each vintage is being used. I’ll double-check on that. 

Thanks, Antti