David W. Welch

POST–the Pacific Ocean salmon tracking project

For most of history, the ocean has remained nearly opaque to study, and it has been difficult to understand where salmon or other marine animals go or how they use the ocean. This greatly limits the ability of oceanographers and fisheries biologists to improve the management of many marine resources. The technical and scientific basis now exists to track the ocean movements of individual marine fish for months or years at a time. In this article, we review how new technologies might be applied to salmon in particular. Our conclusion is that animals as small as juvenile Pacific salmon can be followed for months to years at sea, and thus over great distances. By identifying the migration pathways for individual salmon and specific populations of Pacific salmon, we can establish their ocean foraging grounds. We outline the approaches and initial results from the Census of Marine Life program pacific ocean salmon tracking (POST). The research program involves two distinct aspects: (1) the development of an acoustic array for tracking the movements of Pacific salmon during their shelf-resident phase of the life history and (2) the use of archival (data storage) tags to measure aspects of their local environment and to delineate their open ocean migration pathways off the shelf. We report on some of the preliminary findings from the first year of the field project using acoustic tags.

Ability of Archival Tags to Provide Estimates of Geographical Position Based on Light Intensity

We tested the ability of archival tags and their associated algorithms to estimate geographical position based on ambient light intensity by attaching six tags (three tags each from Northwest Marine Technologies [NMT] and Wildlife Computers [WC]) at different depths to a stationary mooring line in the Pacific Ocean (approx. 166°42′W, 24°00′N), for approximately one year (29-Aug-98 to 16-Aug-99). Upon retrieval, one tag each from the two vendors had malfunctioned: from these no data (NMT) or only partial data (WC) could be downloaded. An algorithm onboard the NMT tag automatically calculated geographical positions. For the WC tags, three different algorithms were used to estimate geographical positions from the recorded light intensity data. Estimates of longitude from all tags were significantly less variable than those for latitude. The mean absolute error for longitude estimates from the NMT tags ranged from 0.29 to 0.35°, and for the WC tags from 0.13 to 0.25°. The mean absolute error in latitude estimates from the NMT tags ranged from 1.5 to 5.5°, and for the WC tags from 0.78 to 3.50°. Ambient weather conditions and water clarity will obviously introduce errors into any geoposition algorithm based on light intensity. We show that by applying objective criteria to light level data, outliers can be removed and the variability of geographical position estimates reduced. We conclude that, although archival tags are suitable for questions of ocean basin-scale movements, they are not well suited for studies of daily fine scale movement patterns because of the likely magnitude of position estimate errors. For studies of fine scale movements in relation to specific oceanographic conditions, forage densities and distance scales of 100 km or less, other methods (e.g. acoustic tracking) remain the tool of choice.

Distribution and Migration of Juvenile Chinook Salmon Derived from Coded Wire Tag Recoveries along the Continental Shelf of Western North America

Abstract The effects of ocean conditions on highly migratory species such as salmon are difficult to assess owing to the diversity of environments they encounter during their marine life. In this study, we reconstructed the initial ocean migration routes of juvenile Chinook salmon Oncorhynchus tshawytscha originating from Oregon to Southeast Alaska using coded wire tag recovery data from Canadian Department of Fisheries and Oceans and National Marine Fisheries Service research surveys conducted between 1995 and 2006. Over this 12-year period, 1,862 coded-wire-tagged juvenile Chinook salmon were recovered along the coasts of Oregon, Washington, British Columbia, and Alaska from March to November. Except for those from the Columbia River, most juvenile Chinook salmon remained within 100–200 km of their natal rivers until their second year at sea, irrespective of their freshwater history and adult run timing. Northward migration of most coastal stocks was initiated during their second or possibly third year ...

Life History and Seasonal Stock-Specific Ocean Migration of Juvenile Chinook Salmon

Abstract The ocean feeding grounds of juvenile Pacific salmon Oncorhynchus spp. range over several thousand kilometers in which ocean conditions, prey quality and abundance, and predator assemblages vary greatly. Therefore, the fate of individual stocks may depend on where they migrate and how much time they spend in different regions. Juvenile (n = 6,266) and immature (n = 659) Chinook salmon Oncorhynchus tshawytscha were collected from coastal Washington to Southeast Alaska in coastal trawl surveys from February to November 1998–2008, which allowed us to reconstruct changes in stock composition for seasons and regions by means of DNA stock identification techniques. Individuals were allocated to 12 regional stocks. The genetic stock assignments were directly validated by showing that 96% of the 339 known-origin, coded-wire-tagged fish were accurately allocated to their region of origin. Overall, the analyses performed in this study support the main findings of previous work based on tagging. However, gi...

Modeling the oxygen consumption rates in Pacific salmon and steelhead: An assessment of current models and practices

Abstract Bioenergetic models of fish have been used to study a large number of processes. Like most models, bioenergetic models require the estimation of numerous parameters. As a consequence, they have often relied on parameters borrowed from other species or values extrapolated from other life stages or size-classes. The magnitude of the biases associated with these practices remains largely unknown. The objective of this study was to determine whether or not metabolic rates could be extrapolated between closely related species and life stages. We focused on Pacific salmon Oncorhynchus spp. and steelhead O. mykiss, as the metabolic rates of these species have been well documented. Our analyses showed that models derived from closely related species did not accurately predict the metabolic rates of salmon, indicating that the practice of “species borrowing” should be avoided in assessing fish metabolic rates. Our work also showed that allometric equations of metabolic rates were not stable when measured ...

Modeling the Oxygen Consumption Rates in Pacific Salmon and Steelhead: Model Development

Abstract We derived a series of models for estimating the standard metabolic rates, swimming costs, and total metabolic rates for sockeye salmon Oncorhynchus nerka and steelhead O. mykiss. The performance of these models was compared statistically and used to predict optimal cruising speeds. These predictions were tested with independent estimates of swimming speed obtained under field conditions. Standard metabolic rates were correlated with body mass and temperature. Swimming costs were correlated with body mass and swimming speed, whereas total metabolic rates were correlated with body mass, water temperature, and swimming speed. Swimming costs were also correlated with temperature and salinity in steelhead but not in sockeye salmon. Regression models accounted for 94–99% of the variance in standard metabolic rates, swimming costs, and total metabolic rates. The oxygen consumption rate models we derived for sockeye salmon were inadequate for describing oxygen consumption in other species of Pacific sal...

Recent Progress in Estimating Geoposition Using Daylight

In this paper we report on the results from a new experiment where we placed tags at three distinct depths on a fixed subsurface mooring. The accuracy of the daily geoposition estimates is substantially improved over our earlier work, with average positional errors of 30 km in longitude and 44 km in latitude. We attribute the result to improvements in both the tags and our software. From the resulting data, we were able to estimate a relationship between light and depth. We generated a simulated variable-depth light record, corrected the light values to a constant depth using the estimated relationship, and then applied our estimation algorithm to the depth-corrected values. This yielded an average absolute positional error of 42 km in longitude and 68 km in latitude, a significantly better result than previously reported. Finally, we provide a theoretical assessment of the expected differences that can be obtained by shielding the light sensor with a blue filter. Although this helps to linearize the relationship between light intensity and depth, the combination of the broad-band spectral sensitivity of the light sensor and the surface light spectrum means that a simple linear response should not be expected. Attempts to correct the light record for variations caused by changes in depth of a tagged animal are probably not substantially improved by filtering the light sensor. More attention needs to be given to the combined effect of the spectral sensitivity of the light sensor and the light spectrum present at twilight.