Bag Harbour, Haida Gwaii · one salmon, followed past its death
A chum salmon grows up in the North Pacific and comes back to die in the stream where it hatched. A bear carries it into the forest. Part of what it was made of ends up in the needles of a spruce.
The North Pacific
He left his stream as a fry, days or weeks after coming up out of the gravel. Since then he has lived in open water. Chum salmon from this coast have been found across the northern Gulf of Alaska, the eastern North Pacific and the southern Bering Sea.
More than 90 per cent of an adult salmon’s body is put on out here, from what it eats in the ocean. Each gold dot stands for nitrogen that came from the sea.
Early summer
Most chum salmon are 4 years old when they are ready to breed. In May and June they begin to move toward the coast.
How a salmon steers across open ocean is not settled. The strongest evidence, from sockeye and pink salmon, points to a memory of the Earth’s magnetic field at the place where the young fish entered the sea. The last stretch is better understood: a salmon finds its own stream by the smell of the water.
Late September
The first chum reach Bag Harbour, in Haida Gwaii off the coast of British Columbia, in the last week of September. They average 2.96 kilograms.
In fresh water a salmon stops feeding. The fat and muscle it laid down at sea pay for everything that follows. The silver goes: a male turns olive and brown, with wavy bars of red and purple, his snout grows into a hook and large teeth line his jaws.
Bag Harbour
The stream is five to ten metres wide and less than half a metre deep. It runs through old forest of Sitka spruce and western hemlock.
In the autumns of 1992 to 1994 the biologist Tom Reimchen counted between 2,300 and 6,300 salmon coming in, more than nine in ten of them chum.
October
The female lays in a hollow in the gravel, he fertilises the eggs beside her, and she covers them. A North American chum carries 2,000 to 4,000 eggs. In one creek where the nests were measured, the eggs lay 22.6 centimetres under the surface on average.
She guards the nest until she is too weak to hold her place in the current.
Dusk
Every chum salmon dies after spawning. By 10 November it is over at Bag Harbour and only carcasses are left.
At Bag Harbour most do not live long enough to wear out. In one of those autumns the bears took 74 per cent of the salmon that entered the stream, and 70 to 80 per cent of the fish they caught had already spawned, in part or in full.
After dark
No more than 8 black bears fished this stream, mostly at night. Over the 45 days of the run each of them caught 13 salmon a day.
Worn-out fish the bears mostly ate where they found them, in the water. A fish still in good condition they carried off: four out of five of those went up the bank and in among the trees.
On the moss
It bites through the top of the skull and eats the brain, then the muscle along the back. It eats about 1.6 kilograms and leaves roughly half the fish.
In 1993 the bears carried 3,700 salmon into this forest. Most of them, 80 per cent, were dropped within 10 metres of the water. A few were found 80 metres in.
5 to 7 days
That year blowflies had laid eggs on nearly every carcass within two days. In 5 to 7 days their larvae had eaten everything soft. Then the larvae crawled off to pupate in the ground; some were found under bark several metres up the trunks.
Crows, gulls and martens take a share. What all of them leave behind, as bones, droppings and urine, goes into the ground with the rain. Reimchen put the nitrogen that reached the forest floor along this stream in the eight weeks of that run at 352 kilograms.
Under the moss
In the soil, microbes turn what is left into the simple forms of nitrogen that roots and fungal threads take up. Within 10 metres of the stream, that one run came to 17.6 grams of nitrogen on every square metre.
Not all of it stays. Some washes back into the stream, some goes off into the air, and some leaves on the wings of next year’s flies.
The following seasons
The Sitka spruces cored beside this stream were on average about four centuries old: 380 years in one group and 445 in the other.
Water drawn up through the trunk carries the nitrogen to the crown. There it is built into the needles, in the green pigment and the proteins that turn light into sugar.
In the crown
Nitrogen comes in a light form and a heavy one, and the sea holds more of the heavy one than the land does. A salmon’s flesh carries that marine mark, and so does a leaf fed on salmon. That is how the fish can be traced into the tree.
Measured this way beside salmon streams on Chichagof Island in Alaska, about 24 per cent of the nitrogen in the needles of Sitka spruce came from salmon; the range was 16 to 32. On this twig, the gold needles are that share.
A core from the trunk
Each year’s ring holds a little of that year’s nitrogen. At Bag Harbour the rings of 14 old spruces were read one by one from 1947 to 2000 and set beside the salmon counts for the same years.
The record is faint: wood is only about 0.05 per cent nitrogen. In 9 of the trees the marine mark rose and fell with the salmon, clearly so in 4. The largest run on record here was 35,000 fish, in 1947.
The valley
Seen from above, the salmon’s nitrogen keeps to the water. It is thickest in the first 10 metres, where the bears eat, and thins out behind them. On Chichagof Island the marine mark was still in trees 100 metres from the stream.
Where a waterfall stops the fish, the mark fades above it.
British Columbia
It is not one valley. The rings of 282 old Sitka spruces on 79 salmon streams along this coast tell the same story: trees standing where the carcasses are left grew faster than their neighbours a short way off.
The gain in wood laid down each year was 55 per cent on Haida Gwaii, 80 per cent on the mainland and 150 per cent on the islands between.
The following spring
Under the gravel the eggs hatch after three or four months, and the young stay down there two or three months more. In spring they come up into the stream, under the spruces. Fewer than one egg in ten gets this far.
Within days or weeks they leave for the sea. Those that return will carry it back up this stream.
The scene above is drawn. These are photographs and a plate from life.
What is measured and what is drawn. The fish is not one tagged animal. His life is put together from what is recorded about chum salmon in general and about Bag Harbour in particular; the numbers at each step are the ones in the papers below. The sea, the stream, the bear and the spruce are drawings. The gold dots are a way of showing nitrogen from the sea, not a measurement: nobody has followed single atoms. Their number means something only on the twig, where the gold needles are 24 in 100.
Where the numbers come from other places. The 24 per cent, and the 100 metres, were measured on Chichagof Island in south-east Alaska, on streams used mostly by pink salmon, not at Bag Harbour. The depth of the eggs is from a creek in Washington. The evidence for a magnetic memory is from sockeye and pink salmon. The roughly 100 grams of nitrogen in one fish is my arithmetic: 2.96 kilograms at 3.3 per cent.
How firm it is. That nitrogen from salmon is in the plants beside salmon streams is well established and has been found many times. Percentages such as 24 per cent rest on a calculation with two sources, sea and land, and soil microbes can shift the heavy form of nitrogen on their own, so they are estimates. Whether the fish make the trees grow faster was argued over: a 2001 claim that growth more than tripled was criticised because the sites differed in soil and in other ways, and its authors agreed the finding was an inference. The coast-wide ring study used here compares trees a short distance apart on the same streams. From one year to the next, in the paper’s own words, the size of the salmon run was “positively but weakly” tied to growth.
Numbers that differ within a source. Reimchen (2000) gives the bears’ take in one autumn as 74 per cent in its summary and as 67.7 to 75 per cent in its results; the page uses the summary. In 1992, unlike 1993, he saw no blowfly eggs on the carcasses at all.