EMA: Forecasting Pink Salmon Harvest in Southeast Alaska
Understanding how ocean conditions impact salmon year class strength is an objective of the Auke Bay Laboratories (ABL) Southeast Alaska Coastal Monitoring (SECM) project. The SECM project has collected a time series of indexes that include juvenile salmon and their associated biophysical data in coastal Southeast Alaska (SEAK) since 1997.
|
Juvenile and adult pink salmon captured during SECM sampling
|
|
|
Purse seiners harvest pink salmon worth
upwards of $90 million annually in SEAK. |
Pink salmon returns are notoriously difficult to forecast because their two-year life history cycle only involves one ocean winter and precludes the use of younger returning age classes to predict cohort abundance. Moreover, year-class success varies widely with harvests ranging from 3 to 78 million fish annually in SEAK since 1960.
Fortunately, direct measures of juvenile pink salmon abundance at sea from SECM sampling have provided a powerful tool to improve forecasts. Juvenile salmon marine mortality is reported to be high and variable during their initial spring nearshore residency. The SECM sampling of juveniles, which occurs later in the summer in seaward migration corridors, has proven to be an effective index for use in pink salmon forecast models for SEAK.
SECM has been using juvenile pink salmon catch and associated biophysical data to forecast adult pink salmon harvest in SEAK since 2004. Several models have been used in this annual forecasting. However, the primary factor that is most highly correlated with harvest each year was the peak average catch per unit effort (CPUE) of juvenile pink salmon in Icy Strait in June or July. Thus, summer juvenile pink salmon abundance in Icy Strait usually reflects the ensuing harvest of pink salmon to the region.
|
Correlation of juvenile pink peak June or July CPUE and SEAK adult pink harvest |
The SECM forecast models have performed well in 8 of the past 9 harvest years, giving estimates that averaged within 7% of actual harvests. However, the 2006 harvest was well below the forecast. Although the juvenile CPUE did indicate a low return for 2006 relative to previous years, this result suggests that a single-parameter CPUE model alone will not always predict harvest accurately. To address this issue, step-wise multiple regression models have been developed that include additional environmental parameters if they significantly improve model fit and help to explain later ocean conditions that affect year class strength. The SECM preferred forecast for 2013 is based on a two parameter CPUE model.
The SECM forecasts have been reported to the Southeast Alaska Purse Seine Task Force Meeting (PSTFM) since 2005. This annual meeting is open to the public and is jointly sponsored by the Alaska Department of Fish and Game and the Southeast Alaska Seiners Association. Resource stakeholders attending the meeting include fishing processors, fishermen, state managers, and researchers.
For more information on the 2013 forecast model, including a matrix of selected variables associated with SECM juvenile pink salmon data and SEAK, please refer to the presentation and handout given at the PSTFM held in Juneau, Alaska, on
December 6, 2012. (handout and presentation in pdf format)
Watch an underwater fish videography - inside a surface trawl (Windows media file)
| |
Contacts:
Joe Orsi, Emily Fergusson, Molly Sturdevant, Alex Wertheimer
Auke Bay Laboratories
Alaska Fisheries Science Center, NOAA Fisheries
Ted Stevens Marine Research Institute
17109 Pt Lena Loop Rd
Juneau AK 99801
(907) 789-6034
joe.orsi@noaa.gov
|
|
Previous forecast model predictions (with 80% confidence intervals) and actual harvests |
|