The bulk of the gap between projection and harvest came in Prince William Sound, which had a projected harvest of 38 million pink salmon, but a harvest of only 16 million. Of that harvest, 6.3 million went to cost recovery for the hatcheries.

The reason for the missing pinks could be many things, and also applied to chum salmon, according to Alaska Department of Fish and Game area management biologist Jeremy Botz.

"Our wild stocks came in weak for both chum and pink salmon," Botz said. He theorized that it could be a result of ocean conditions " leading to weak zooplankton blooms, or maybe something after the release (of the smolt), in the outmigration."

"A lot of the fish came back small this year," he added, "including chum, pink and coho species, so perhaps conditions in the Gulf of Alaska may not have been optimal, either."

There was a problem with the release of the smolt at the Valdez hatchery, as well, that resulted in the loss of about 10 percent, according to Botz. Generally they release around 220 million smolt, and the 2008 release was estimated at around 200 million. Botz said that number is not too far out of the range of normal, though.

"They usually don't have releases that low but it's not out of the question," he said.

He added that it shouldn't have affected the projected harvest, saying that number is based on what they released out of the pens.

The run failure was not system-wide, however, mostly affecting wild and hatchery runs in the northern areas of the district.

"It was an interesting phenomenon," he said. "Many of the northern reaches of the Sound did poorly, and the southern portions of the Sound, near the Gulf, did better."

Chinook salmon also were missing from the Sound, with only 9,000 caught out of a projected harvest of 157,000. Many rivers in Southcentral Alaska saw low returns of chinooks, causing many sport fishing closures and restrictions. The cause is unknown, but ocean survival is thought to play a large role.

Upper Cook Inlet is another area that came in below projection, for the second year in a row. The preseason projection of a 3 million fish harvest of sockeye salmon was off by a third, resulting in an actual harvest of only 2 million sockeye. The commercial fishery was shut down for two of what should have been peak openings while the dipnet and sport fisheries on the Kenai River continued unabated, the second year in a row that the commercial fleet bore the sole responsibility for resource conservation. The personal-use dipnet fishery is estimated by some to have harvested 500,000 to 600,000 sockeye, although voluntary reporting makes it hard to track accurately.

The Kenai River did reach the escapement goal this season, however, unlike in 2008.

The Bristol Bay sockeye season was the one fishery that came in well over its projected harvest of 24 million sockeye with a total of nearly 30 million sockeye actually landed. That ties with 2007 as the best season in the past 10 years, and is in sharp contrast with runs in the earlier part of the decade, which bottomed out in 2002 with a harvest of only 10.7 million sockeye, and a price per pound of 49 cents.

That's not the worst year on record though. In 1973 only 4.4 million sockeye were caught statewide, with a total of 22 million fish of all species. The best sockeye year since 1970, the furthest back online records go, was 1993, with 64 million sockeyes landed statewide. A breakdown by area was not available.

Cristy Fry has commercial fished in Homer since 1978. She also designs and builds gear for the industry. She currently longlines for halibut and gillnets salmon in upper Cook Inlet aboard the F/V Realist. She can be reached at realist468@gmail.com.