When we talk to people about streamkeeping, and in particular monitoring salmon that return to spawn in Byrne Creek, a common question that arises is: "How the heck do you count fish in the creek? How do you know you're not counting the same ones over and over again?"
I usually start off my reply by saying that we can recognize individual fish -- that sets the hook as people stare at me in disbelief. OK, I'll explain how we really do it, however sadly it is true that we can sometimes recognize fish from day to day.
I say sadly, because Byrne Creek is still recovering from a toxic spill several years ago that entered a storm drain and killed everything in the creek, and is constantly battling the pressures of existing in an urban environment. So far the most spawning salmon streamkeepers have seen return in one year was 72 in 2002. So yes, we do occasionally recognize individual fish from day to day, particularly if they have some distinctive marking or injury.
But back to how we really count fish.
Salmon die after they spawn, so we find their carcasses. Of course a few get washed away, a few get buried in silt, and a few are dragged off into the bush by coyotes and raccoons, but we patrol often enough that we likely spot over 95% of the "morts," or dead fish.
When we find dead spawners, we measure them, cut them open to confirm sex and see if they have spawned, and -- here is the key point -- we cut the bodies in half so we do not double count. We return the processed carcasses to the creek where they provide nutrients and food for other animals.
So, particularly in a situation like ours where we are dealing with at most a hundred or so fish -- and we hope several hundred in future years -- we get pretty accurate counts.
"But on your website you have counts for 'spotted' and 'processed.' How does that work?"
That's also quite simple. There are distinct areas in the creek, such as the artificial spawning habitat, the sediment pond, spans between bridges, culverts, etc., in which it is fairly easy to spot fish and count them, especially if you have at least two sets of eyes at work. We patrol in an upstream direction, because it is much more likely for a spooked fish to run downstream than up.
Initially the "spotted" count is likely not that accurate, however our data starts to firm up over time as the salmon spawn and begin to die.
Eventually we can simply add live fish spotted during a patrol to the total of dead fish processed to get a "spotted" total. And so far the "spotted" totals and the "processed" totals over the years have been very close when the spawning season ends, though we know that we don't find and process every dead spawner.
Posted by Paul at November 25, 2004 07:18 PM