Southeast Coastal Monitoring Pink Salmon Harvest Forecast Models: Accurate for 2010 and Predicting a Strong Harvest for 2011
Figure 5. Researchers process a large catch of juvenile salmon from a 20-minute surface trawl haul 4 miles offshore in Icy Strait, June 2010. Photo by Molly Sturdevant.
Fourteen years ago ABL launched the Southeast Alaska Coastal Monitoring (SECM) project to study the effects of marine ecosystem dynamics on salmon recruitment in Southeast Alaska (SEAK). This annual fisheries and oceanographic survey has been conducted during the months of May to August in inside waters of SEAK utilizing government and chartered vessels. Fishes, including juvenile salmon, Pacific herring, and walleye pollock are collected using surface trawls; oceanographic measurements include zooplankton biomass, conductivity-temperature-at depth, and surface nutrients and chlorophyll-a. The extended time series of annual SECM ecosystem metrics offers a unique opportunity to determine the effect of climate change and variability on marine ecosystems in SEAK.
One of the major objectives of these SECM annual surveys is to reduce uncertainty in pink salmon forecasts for SEAK. Pink salmon is a vital component of the commercial fishery in Alaska providing an average ex-vessel commercial value of $24 million to the SEAK economy from 2005 to 2010. Reducing uncertainty in salmon forecasts increases economic benefits to Alaska fishers and communities and also helps maintain biological sustainability of the salmon resource for future generations. Determining the appropriate metrics to include in salmon forecasts is not an easy task, and many of the current salmon forecast models do not capture the high variability in adult returns. This uncertainty in salmon forecast models has lead to a number of economic disaster declarations for Alaska in the past two decades, emphasizing the need for research efforts that examine marine ecosystem metrics that can help explain salmon recruitment.
Pink salmon, which have a simple 2-year life cycle, are notoriously difficult to forecast because unlike most other salmon species they lack leading indicator data necessary to construct sibling forecast models. Despite this lack of leading indicator data, SECM pink salmon forecasts have successfully predicted harvests within 8% of the actual annual harvests in SEAK (12-59 million) during 6 of the past 7 years by relying upon marine ecosystem metrics. SECM surveys (Fig. 5) have determined that the peak abundance of juvenile pink salmon during either June or July is one of the primary metrics explaining future pink salmon returns to SEAK. However, forecast model accuracy has also been improved in some years by adding oceanographic parameters such as regional May water temperature or later ocean basin index parameters such as the El Niño/Southern Oscillation Index (ENSO); these auxiliary parameters are believed to affect salmon early marine growth, condition, and marine survival.
Last year the SECM forecast model for 2010 predicted a harvest of 26.8 million fish, very close to the actual 24.2 million fish harvested that year. The SECM forecast model used to predict the 2010 harvest was a stepwise regression model using the factors of log normal-transformed peak juvenile pink salmon catch per unit effort (CPUE), May integrated temperatures in the upper 20 m waters of Icy Strait, and the ENSO Index lagged to the juvenile ocean entry year. For 2011, the SECM forecast model predicts a large harvest of 56.2 million pink salmon, more than double the harvest realized in 2010. The SECM forecast model used for the 2011 harvest year is based solely on juvenile pink CPUE.
The SECM research has been supported over the years by federal and state agencies through the Pacific Salmon Commission (Alaska Sustainable Salmon Fund or Northern Fund), and SECM researchers provide annual pink salmon forecasts at the Southeast Alaska Purse Seine Task Force (PSTF) meetings soon after the commercial salmon fishing season. The PSTF meetings are co-sponsored by the ADF&G and the Southeast Alaska Seiners Association and are held rotationally in Juneau, Ketchikan, Petersburg, or Sitka. The PSTF meetings are attended by commercial salmon fishermen, industry processors, ADF&G managers and researchers, private non-profit salmon aquaculture associations, and the general public.
Marine ecosystems are complex and difficult to understand, but the SECM research has shown that systematic sampling done on an annual basis can be applied to better understand mechanisms responsible for salmon recruitment and meet the needs of fishery resource managers and stakeholders.
By Joe Orsi, Molly Sturdevant, Emily Fergusson, and Bill Heard
