I think what happens is that the chinook have parasites and pathogens that they co-evolve with. It's like how if you leave your fish in the refrigerator, it will stay fresh until you're ready to eat it, but if you leave it out where it's warm, then those pathogens start growing faster than the chinook are prepared for and they can debilitate the chinook or kill them before they get back to the spawning grounds. If they kill them at sea, that means they're not available for whales to eat.
We've seen what's known as the blob. It's a very large patch of warm water off the Pacific coast of the U.S. and Canada. In the years when that's been present, salmon survivorship at sea has been very low. That's something we weren't aware was happening in previous years. It's something that may become a lot more common.
We've also had a lot of natural cycles like El NiƱo and Pacific decadal oscillation, which are known to have impacts on salmon survival. We can go back to tree rings from hundreds of years ago to detect whether salmon returned to the streams near those trees. They suggest salmon populations have fluctuated naturally by, say, a factor of two just due to natural conditions.
What we need to be careful about is that people are going to start seeing a lot more bad years relative to the number of good years than we have in the past.