The issue of the ecological partitioning of these species needs to be addressed. Very true that hatchery steelhead and coho juveniles are quite a bit larger than juvenile chinook, but I've always been lead to believe that these hatchery smolts do not rear and forage substantially in the streams after release, but move quickly to marine areas. And once in the marine environment, quickly move offshore.

I sample juvenile salmonids by beach seine just about every year in estuaries (river mouths and surrounding marine nearshore). Here are a seasons worth of results at the mouth of one river (sampling every other week at five stations April - July):

chinook - 1,026
chum - 262
pink - 102
coho - 15
steelhead - 2

My work concentrates in estuaries and nearshore marine areas, not rivers; but the above data reflect the different habitats that juvenile salmonids occupy. Chinook, chum, and pinks are beach huggers while larger coho and steelhead quickly move offshore. The data also reflect that chinook are the most nearshore dependent of all the anadromous salmonids. Although chum and pinks are also shoreline oriented, they occupy these shallow nearshore areas for a smaller time period than chinook.

I don't have the professional opportunities to sample streams as much (though I'd love too!) but I've always been lead to believe that you have a similar partitioning of species into respective habitats. For smolting hatchery fish, at least in theory, hatcheries release age 1+ (larger) coho and steelhead juveniles during spring runnoff periods when rivers are high. The high flows, coupled with the drive to smolt, carry the fish downstream fairly rapidly such that substantial foraging does not occur. In practice, I don't know how complete such partitioning is. After all, we've all caught smolts during the spring while fishing for summer runs.

The news piece posted by Ramon cites the Sacramento basin. With the dam at Shasta and all of the irrigation diversions, has this effectively reduced the ability for juveniles to partition into the habitats that they historically would occupy? (keeping them separate from one another?) If the migration behaviors of hatchery smolts on the Sacramento and Washington streams are different (perhaps due to seasonal flows?), then this would have a substantial affect on the level of predation experienced.

From a scientific perspective; from a life history perspective, this issue needs to be examined before hatcheries are simply eliminated. I don't think it would be hard or that expensive to collect hatchery smolts as they're migrating downstream and do stomach contents analyses on them.

Opinions from any fish managers out there?