No, you are correct, it's just complicated and broad applications of afforestation will require very accurate modeling. The perceived benefit will depend on wha

No, you are correct, it's just complicated and broad applications of afforestation will require very accurate modeling. The perceived benefit will depend on what timescale you care about. Changing the albedo and local transpiration has an immediate effect (which can be positive or negative, depending on the existing albedo, downstream effects, etc). CO2 sequestration, on the other hand, takes time. If you plant trees, let them grow for 30 years, then remove a high fraction of that carbon from the cycle (e.g. lumber), then the break-even point can be pretty short (decades, probably, but I don't have a good number). This will depend on a lot of climate feedback effects, like heat leaking out of your forest through moisture evaporation (which otherwise would have been shed back into space), what the existing carbon uptake rate was of the location you planted (did you replace scrubland, grassland, desert, ...?), etc. If your permanent sequestration is primarily due to carbon cycle efficiency loss, maybe multiply that timescale by a factor of 10-50.