Despite their wide distribution, burbot Lota lota have often been neglected by anglers and fish management agencies in North America (Quinn 2000). Only four of 20 states with burbot regulated their harvest as of 2000, yet growing angler interest in burbot in some regions, ecosystem-based management, and declines in some populations emphasize the need for greater focus on burbot by management entities. Several North American burbot populations have declined and the declines were attributed to over harvest (Bonar et al. 2000a; Paragamian et al. 2000; Quinn 2000; Ward et al. 2000), negative species interactions (Carl 1992; Bonar et al. 2000a), and impacts from hydropower development (Paragamian et al. 2000), decreased productivity (Paragamian et al. 2000), and reservoir fluctuations (Krueger and Hubert 1997).

In Washington State, populations of burbot are known to occur in 11 lakes and reservoirs in the upper Columbia River drainage (Bonar et al. 1997; Bonar et al. 2000a; Wydoski and Whitney 2003). A review of burbot stock status in Washington categorized the status of each population as healthy, depressed, critical, or unknown based on existing abundance, size structure, growth, and condition data (Bonar et al. 1997; Bonar et al. 2000a). The Lake Roosevelt population was considered healthy due to its stable abundance based on electrofishing and gill net catch-per-unit-effort (C/f) data, although nothing was known about size structure, growth, or condition. Subsequent to the status review, Polacek et al. (2006) analyzed data from burbot captured during electrofishing and gill net sampling conducted on Lake Roosevelt between 1988 and 2001. They indicated that abundance, based on C/f, had increased after 1994; though, all of the C/f values were relatively low and the trend analysis was qualitative. Growth and condition of Lake Roosevelt burbot were found to be low, which were attributed to poor rearing conditions in the reservoir (Polacek et al. 2006). They suggested that reservoir operations were unlikely to change, so if the Lake Roosevelt burbot population is a priority then it will likely have to be actively managed to achieve increases in abundance, survival, growth, and condition.

Standardized stock assessment data is needed to facilitate management of the Lake Roosevelt burbot population. Bonar et al. (1997; 2000a) pointed out the need for a standardized stock assessment program that monitors trends in abundance, growth, and condition of burbot in Washington State. Standardized fish sampling programs have been used to determine and monitor stock status of multiple fish populations (Ney 1993; Willis and Murphy 1996; Bonar et al. 2000b), including burbot (Bernard et al. 1993). Without a standardized stock assessment protocol, management biologists have no way of monitoring changes in burbot populations as a result of changes in management, exploitation, biological factors, or environmental factors. The Fall Walleye Index Netting (FWIN) program implemented annually in Lake Roosevelt may provide the standardized stock assessment program needed for monitoring burbot population status and trends (see the proposal for the Lake Roosevelt Fisheries Evaluation Project, BPA Project No. 1994-043-00) for a complete description of FWIN). The FWIN program, established in 2002, is used to monitor Lake Roosevelt walleye population trends, but burbot are regularly captured as bycatch in FWIN gill nets. Managers recognized the potential of FWIN for monitoring burbot stock status, so catch and biological data were also collected on burbot. Prior to 2013, the FWIN data for burbot had never been analyzed so it was unknown if FWIN would be adequate (a=0.05, ß=0.20, effect=25%) for monitoring status and trends in stock assessment metrics (C/f, growth, condition). In 2013, a research scientist and  biometrician were subcontracted to analyze the burbot by-catch data from the 10-year (2003-2012) Lake Roosevelt Fall Walleye Index Netting (FWIN) survey data set to assess its adequacy (appropriate statistical power) to monitor status and trends in burbot relative abundance, mortality, size and age structure, condition and growth. The results of this analysis indicate that the FWIN survey does result in the capture of enough burbot to monitor annual changes in population metrics (listed above) with adequate statistical power (80%). Thus, we will hire a subcontractor to analyze the 2013 FWIN burbot data during the FY2014 project.

The results of the 10 year analysis conducted in 2013 also indicated relatively low mean lengths of the burbot population each year and relatively few older fish (>5 years). It is unknown if the larger, older burbot in the population are under-represented by the FWIN gill nets or not. Thus, the question of whether or not the FWIN sampling accurately represents the size and age structure of the Lake Roosevelt burbot population needs to be answered for us to be to say that the FWIN sampling is adequate for monitoring burbot population status and trends. Thus, we will conduct supplemental sampling to answer this question using gears that are known to capture large, old burbot. The supplemental sampling will consist of fishing trammel nets and modified cod traps in coordination with the FWIN sampling. Trammel nets, commonly used for burbot stock assessment (Abrahamse 2009), are typically less size selective than gill nets (Hubert et al. 1996), so it would be reasonable to use them to evaluate for size and age bias in the FWIN gill net catch. The primary drawback of using trammel nets in Lake Roosevelt is the potential catch and mortality (through suffocation when gilled in the outer mesh [30.48 cm stretch]) of wild adult white sturgeon that are subject to a population recovery effort. Modified cod traps, also commonly used for burbot stock assessment (Spence 2000; McLellan and Hayes 2011), usually have low by-catch, pose no risk to white sturgeon, and select for large burbot (Horton and Strainer 2007) making them an obvious choice to evaluate size and age bias in the FWIN catch. However, it is unknown if the relative biases of the FWIN gill nets and cod traps are perfectly complimentary, or in other words, together provide an accurate representation of the size and age structure of the population. Thus, we will supplement the burbot size and age data set from FWIN gill net sampling with data from the supplemental fishing of modified cod traps. In order to evaluate the relative size (and associated age) bias between the FWIN nets and cod traps, we will conduct supplemental trammel net sampling in the areas of Lake Roosevelt that area outside of the distribution of white sturgeon (lower half of the reservoir during winter). We will assume that trammel net catch is representative of the population size (and associated age) distribution. The trammel net data will be used to develop length and age bias corrections for FWIN gill nets and/or modified cod traps in the event biases are observed. The bias corrected data will be used for the calculation of size/age structure indices during 2014 FWIN burbot data analysis.