Question on GCAM secondary output + cost structure

[One-12373]

Hello everyone,

I’m new to GCAM and would like to better understand how secondary outputs work within the model.

My goal is to add a secondary output to a technology so that it can supply a new market with existing demand. However, it seems that GCAM does not “call” a technology based solely on its secondary output. In other words, if the technology is not utilized in its primary market, the secondary market will not receive any supply, which could cause issues. Is that correct?

If so, when the technology is not used in the primary market, how can I force its utilization (through a subsidy?) to ensure the new market is supplied?
Finally, am I required to add a supply curve (which can be costly) for the new market to guarantee supply and allow price formation?

Thank you for your insights!

[pkyle]

A technology with secondary outputs differs from a standard technology in the cost calculation—the cost is equal to the sum of non-energy costs and input costs, minus the price of the secondary-output commodity times the output-ratio—and in the production of the secondary-output commodity, whose supply is increased by the primary output times the secondary output-ratio.
The secondary output market (i.e. commodity) needs to exist independent of the technology producing it as a secondary output. This can be hacked using fractional-secondary-output and making a resource with no supply (if you search the string DDGS in the xml input directory, you can see how this is set up structurally).
The primary output market also needs to have non-zero demand, for the technology to have non-zero outputs. While secondary outputs can reduce the levelized cost of producing a primary output, and thus may make a technology more profitable, increasing its market share, the production of the secondary output can be limited by the size of the market of the primary output commodity.
Note that it is arbitrary which output is assigned as primary versus secondary. Normally the primary output is the one that accounts for the majority of the revenue, but there isn’t anything stopping a model user from switching the two.

[One-12373]

Thanks for your answer Kyle.
If I understand correctly, when my main technology is not used in its primary sector, I need to ensure that a resource supply curve exists for the secondary market. This way, the model can still provide supply to that market, even if the technology isn’t being called in its primary role.
Conversely, I also need to make sure there is demand in the market. So if the demand-side technology isn’t being triggered in its sector, I would need to enforce a minimum output (e.g., via a policy constraint) to ensure that demand exists and the market functions properly.
Does that sound right?
Thanks again for your help!

[pkyle]

The empty resource supply curve (see what is done for DDGS) needs to be made if the secondary output market does not otherwise exist (i.e., if there is no supplysector or resource of that same name). This is true irrespective of whether the technology's outputs will be zero. And yes, to get non-zero output of the secondary commodity you will need to ensure non-zero output of the primary commodity.

[One-12373]

Thanks a lot !
I still have a last question.
From what I understand, when using the following input (from cement):
minicam-non-energy-input name="non-energy"
0.08667
/minicam-non-energy-input
the cost is expressed per kg of output and includes all unspecified costs (e.g., labor, capital).

However, what changes when both capital and input cost are specified, as in the case of refining cellulosic ethanol for example ?
tracking-non-energy-input name="non-energy">
capital-coef>0.1149693818768012
tracking-market>capital
depreciation-rate>0
input-cost>4.9918
/tracking-non-energy-input>
What does the capital-coef represent, and how is the total cost calculated in this case per output unit?

Thanks again for your help!