One Fish, Two Fish, Three Hundred Thousand Fish

When you are staring at an empty river bed, ten minutes seems like an eternity. But soon, you see them. Dark shadows, moving across the bottom of the river. At first, a few here and there, but their numbers grow by the day. And then, one day, you see a solid band of fish, five to ten across, two or three deep.

"When the fish are streaming by by the thousands, you count by fives," says Fred West, who has been counting fish on the Ugashik River for four years. "When you can't count them by fives, you start counting by tens."

A hundred miles south of Naknek/King Salmon, the Ugashik River supports the fourth largest run of sockeye salmon in Bristol Bay. Every year, the Alaska Department of Fish and Game (ADFG) tries to get 700,000 fish past its escapement towers, and allow the commercial gillnet fleet to harvest the rest. The problem is, 300,000 fish can go past the towers in a single day. And once they pass the towers, they have escaped--it is too late to do anything about them.

Starting at 9:00 a.m. on June 23 each year, fishermen may only take fish in the Ugashik District when the Fishery Manager opens a fishing period by emergency order. Managing the Ugashik River fishery so that fishermen catch as many fish as possible, but enough escape to spawn, takes a squadron of field camps and observers.

At the start of the emergency order period, ADFG sends two technicians to the village of Pilot Point, near the mouth of the Ugashik River, to test-fish in the district. ADFG charters boats from the fleet to drift nets at specified locations in the district. The technicians serve as ADFG's representatives on board the chartered boat.

"Usually we do a dozen or so drifts [each day that the fishery is closed]," says Richard Russell, the Fishery Manager for the Ugashik and Egegik rivers from 1980 to 1995. "Sometimes we fish scattered areas throughout the district, sometimes we do a bunch of sets in a certain area where we think there may be more fish."

Sets are usually short, ten or fifteen minutes. As the crew brings in the net, the technician counts the number of fish coming over the rollers, and sometimes takes scale samples. Other technicians in King Salmon will determine the age structure of the returning salmon by reading the scales with a microfiche.

"With test-fishing in the district, we aren't trying to determine the number of fish, so much as whether there are fish there at all," says Russell. ADFG pays test-fish vessels a set amount per day, plus a little for each fish, and then sells the fish to the first available tender.

When fish arrive in the district, the manager lets a spurt of fish enter the river, and then opens the fishery. By alternating fishing periods with closures, he tries to maintain a steady stream of fish in the river. When the emergency order period starts, the Inside Test Fish crew in the village of Ugashik, 12 miles upriver from Pilot Point, has already been fishing for several days. Twice a day, before each high tide, test-fishers drift 25 fathoms of net from a set-net skiff to estimate how many fish have entered the river and escaped the fishermen.

The in-river test-fishers calculate an index based on the length of the net and the duration of the set. When the fish reach the towers several days later, the manager in King Salmon can calculate a rough estimate of how many fish the index represents. For example, if the index on a given tide is 1,200, and the tides before and after had a much lower index, and three days later 50,000 fish pass the towers, the manager can then estimate that some 25,000 fish pass the test-fish site when the index is 600.

"On the Egegik River, the fish generally reach the tower two to three days after passing the test-fish site, like clockwork," says Russell. "But on the Ugashik River, the lag time between test-fish and the towers can be anywhere from two to twelve days."

Finally, the fish pass the towers at Lower Ugashik Lake. A three-person crew monitors the river twenty-four hours a day, beginning July 3. When the federal government first started measuring escapement in the early 1950's, workers built a fence across the river, called a weir, which allowed water to pass but not fish. Salmon gathered into the part of the weir furthest upstream, where they could be counted individually as they passed through a gate which the technicians controlled.

But some days on the Ugashik river 300,000 fish can pass in a single day. In order to allow that many fish through, a technician would have to count 3.5 fish per second, for 24 hours, by individual fish. Not only was this hard for the workers, but it was also hard on the fish, as thousands trying to get upstream smashed their brothers and sisters against the gate. The managers quickly learned that weirs were impractical, observed that sockeyes migrated along the banks of the river, and started using towers.

The towers are aluminum scaffolding, anchored by cables and stakes, used since the 1950's. Every hour a technician climbs the tower on one side of the river. Wearing polarized sunglasses so he can see through the surface glare, he counts every fish that passes for ten minutes. Then he crosses to the other side, and counts for another ten minutes. By adding the two counts and multiplying by six (since he has counted for one-sixth of the hour), he determines a count for the hour.

At night, the technician uses an automobile headlight hooked up to a twelve-volt marine battery and a solar panel to light the river. When the light is aimed offshore, the fish avoid the bright spot and pass through the glow around the edges. "It's easier to count at night, because the fish show up so clearly," says West. "But sometimes they spook and go all over the place, like popcorn. Then you just have to estimate how many there were."

During each ten-minute count, the technician will see anywhere from zero to over 3,000 fish. Up to about 1,500, he can count individual fish, but beyond that he must visually break them into groups of five or ten.

The towers give the manager a final count of how many fish escaped into the river to spawn. But he doesn't want to get the entire escapement at once. Fish that arrive early in the run are different from those that show up later. Males, and older, larger fish tend to come early. Later in the run, younger fish and more females return. The tower crews collect samples of fish throughout the season using a beach seine. By keeping track of the ratio of males to females, the manager gets another clue to how far the run has progressed.

The trick is to get short spurts of fish up the river throughout the run--enough to reach the escapement goal, but not exceed it. That's easier said than done, when the run peaks anywhere between July 5 and July 19, and half of the escapement goal can enter the river in a single day. The fish are fickle, staying well outside the district for days and then dashing for the river.

"[Field camps] are my eyes and ears," says Russell. "I rely on [them] to tell me what the fish are doing. And when I have a question, I go flying." From the air, the manager can see where groups of fish are, and sometimes which direction they are moving. "If the water is extremely clear, I can see individual schools of fish. You see a jumper in the air, and when it goes back in the water, you can still see it swimming. And then you can make out more dark shapes all around it. But usually, you can only see the jumpers, or sometimes fins breaking in the waves."

While he is flying around, the manager also assesses the fishing fleet to see how many boats there are, how close they are to the fish, and how effective he thinks they would be if he declared an opening.

Back in King Salmon, he must sift through a small mountain of information, determine what is important, and devise a strategy. Every morning he gathers his staff and reviews what they know. Before the season, the research biologists have forecasted the return by evaluating the number of smolt leaving the river several years earlier, and the age composition of fish the year before. If three years ago a large number of smolt left the Ugashik system after spending one or two years in the lake, and last year fishermen caught many four-year-old fish, then this year, there will likely be a large number of five-year-old fish.

At Port Moller, 200 miles down the Alaska Peninsula, ADFG employs another test-fish vessel to verify the numbers of fish returning to Bristol Bay, intercepting the fish seven to twelve days before they arrive in Ugashik.

Other technicians in King Salmon read the thumbprint-like scales collected from field camps and the commercial catch to determine the age of the returning fish. If the numbers and age composition fit the forecast, the manager can rely on the amount of fish caught to regulate the fishery. Every day, each processor reports its catch figures to ADFG. By monitoring the percentage of the forecast caught and the percentage past the towers, he can alternate fishing periods with closures, and keep the fish trickling in.

But forecasting the fish return is like forecasting the weather: biologists can get a general picture of what will probably happen, based on the data and what has happened in the past, but they cannot know what will happen in the future. If a different age class of fish returns than was expected, the entire forecast may be wrong.

When it comes down to it, the manager has to guess what the fish will do, before they do it. "You ask yourself, if I open it now, will I get enough fish up the river to make my escapement?" says Russell. "If I leave it closed a little longer, will too many go by? ... Sometimes you just have to go for it, take a chance." Managing the largest sockeye fishery in the world is no easy task, but through careful observation by a squadron of technicians, and informed, diligent management, Bristol Bay has remained one of the last healthy fisheries in the world.