This fits Crutzen's position, which seems well considered.

It also fits into situation (1) in this comment, and avoids the trap of situation (2).

Words and ideas I offer here may be used freely and without attribution.

Planting trees...

Oh, how cute.  It just doesn't help.  To capture CO2, you must also be willing to cut them down and cover them with enough dirt to prevent them from rotting.

Planting new forests captures CO2. Many trees grow for up to a century during which they will capture and store CO2. This is roughly the same as the atmospheric lifetime of CO2. As long as the forest doesn't burn or isn't cut down, the sink remains (though after 100 years little to nothing is added).

I did say that all geoengineering options are transitional. Planting trees on the scale of, like, half the Sahara will have a big effect (and might have some questionable ecological consequences).

On seeding the seas: I think that algae have a very important role in the self-regulation of our planet that we don't fully understand yet. For that reason, I think that tinkering with algae on a large scale is more dangerous than an option that has a natural experiment as a precedent which indicates that it will not have catastrophic side-effects. Otherwise, I think that it might lead to mass extinctions among other sea life through eutrophication.

Iron dust is definitely considered, see here and here, and here.  

Obviously, you can't foresee all effects of either option (neither can I), so how can you rank them?

Being able to see all the effects (of anything) is too high a standard. Ranking options and choosing actions always goes forward with imperfect predictions and best judgement.

From what I can see, the iron approach isn't very attractive, but I'm glad that it was taken seriously enough to be investigated through open-sea experiments.

Words and ideas I offer here may be used freely and without attribution.