Just as a reality check, phosphorus is potentially renewable.  Not only is phosphate rock being actively laid down in a few places around the world, anadromous fish such as salmon take up phosphorus from oceanic sources as they grow and bring it upstream when they return to their home streams to breed.

This is one more reason we should remove water diversions and other threats to salmon runs.  They are far more important than this year's crops.

Like many I like the idea of giving wild salmon better odds every chance we get (I've netted a 'few' reds leaving the Bering Sea over the years myself). Just curious though, what percentage of the phosphorus salmon gather could be actually mined without tearing up the salmon habitat? A related second question, how much phosphorus would be the maximum anadromous fish could put into our system a year and how does that compare to what we take out? I wouldn't be surprised if most of those numbers are tough to find.

I'm guessing catching the fish and eating them doesn't help the phosphorus to build up either. The big runs are spectacular, but we are not going to just leave the fish alone even if by some miracle we start to manage a whole lot more of the world's waterways in ways which allow full wild runs to return. I have a feeling we will never manage fish for maximum phosphorus production, but I could be wrong.

I'm guessing phosphate rock isn't being laid down near as fast as we are tearing it up, but I could be wrong there too. Just a reality check.

A lot of the salmon already go to feed mammals and birds (both before and after spawning).  Bears and such remove the nutrients from the streams and deposit them on land; they eventually wash back out again, but that's part of a loop, not a one-way traffic.

If we caught the fish, ate them and cycled the nutrients back to fields, we could build phosphorus stores on land.  Our biggest problem at this time may be keeping the oceans productive enough to have the salmon return.

The health of the oceans is a huge concern, right now it looks pretty grim, but not hopeless. I'm all for the return loops you mention, the more the better that is for sure. What I was pointing up was that it is very unlikely we could use the sources you indicated to replace the accessible phosphorus stores at near the rate we are now depleting them.