For example, imagine a solar panel that costs $1000. They probably cost about $1000 to make, roughly. There may be some subsidy that allows them to sell at a loss but it’s safe to say it costs around $1000 to make, at a maximum.
You can also presume that energy is a factor in the cost to make a solar panel. So if 100% of the cost of the solar panel is energy, then some factory spent $1000 on energy to make that panel, max.
So that’s your hard limit on how much CO2 is being produced in the production of that solar panel: $1000 dollars’ worth.
Which would make sense. At least in the UK, onshore and offshore wind are the cheapest sources going. Would suspect the amount of energy put into creating and maintaining them would mean that the LCOE is lower.
If the economy is not yet close to 100% renewable, what matters is not small CO2 emission from making the renewable sources, it's how rapidly the remaining fossil sources can be displaced.
We can always offset. I mean, burning fossil fuels in itself is not a bad thing, but we know they're a finite resource and we simply have to offset those emissions with carbon sequestration.
The reduction in fossil fuel usage needed to get to a climate-sustainable level is so large that any tolerable residuum could instead be handled by biomass.
But notice that most research accessible online is old and solar is improving fast (while the fossil fuels can do nothing but get worse with usage).
There's a comprehensive comparison here that is interesting:
http://newmaeweb.ucsd.edu/courses/MAE119/WI_2018/ewExternalF...
While this is a fair question, I love how these types of questions seem to only pop up with solar/wind/batteries (and by proxy EVs).
Nobody ever askes how much GRG is used to produce an engine block, cooling tower, reactor vessel, etc etc.