Measures of Overall Model Fit - In general, the Beaufort Assessment Model (BAM) fit well to the available data. Predicted length compositions from each fishery were reasonably close to observed data in most years, as were predicted age compositions (Figure 4). The model was configured to fit observed commercial and recreational removals closely (Figures 5–11). Fits to indices of abundance generally captured the observed trends but not all annual fluctuations (Figures 12–14). - Stock Abundance and Recruitment - In general, estimated abundance at age shows a structure that has been relatively consistent through time, reflecting effects of year-class strength and annual fishing mortality, but without severe age truncation (Figure 15; Table 6). Total estimated abundance decreased until 1990, then increased to a peak in 2004, then decreased again until reaching its lowest values at the end of the assessment period 2013-2015. The uptick in 2016 should be interpreted with caution, as it is primarily based on the forecast of recruitment (age-1 fish), which is uniformed by data on year-class strength. Instead, the forecasted 2016 value is that expected from the spawner-recruit curve precisely, which is larger than any in the last decade. Annual number of recruits is shown in Table 6 (age-1 column) and in Figure 16. The highest recruitment values were predicted to have occurred in 2003–2004. Since then, recruitment has been below expectation. - Total and Spawning Biomass - Estimated biomass at age followed a similar pattern as abundance at age (Figure 17; Table 7). Total biomass and spawning biomass showed similar trends—general decline until 1990, increase until the mid-2000s, and decrease since (Figure 18; Table 8). - Status of the Stock and Fishery - Estimated time series of stock status (SSB/MSST and SSB/SSBMSY) showed general decline throughout the beginning of the assessment period, increase starting about 1990, and then decrease since 2007 (Figure 34, Table 8). Base-run estimates of spawning biomass have remained below the threshold (MSST) for most of the assessment period. Current stock status was estimated in the base run to be SSB2015/MSST = 0.38 and SSB2015/SSBMSY = 0.29 (Table 18), indicating that the stock has not yet recovered to SSBMSY. Median values from the MCB analysis indicated similar results (SSB2015/MSST = 0.37 and SSB2015/SSBMSY = 0.27). The uncertainty analysis suggested that the terminal estimate of stock status is robust (Figures 35, 36). Of the MCB runs, 100% indicated that the stock in 2015 was below SSBMSY, and 97.7% that the stock was below MSST. Age structure estimated by the base run generally showed fewer fish of all ages than the (equilibrium) age structure expected at MSY (Figure 37). The 2015 age structure showed more old fish than in previous years, reflecting the strong recruitment pulse in the early 2000’s, and it showed fewer young fish, reflecting the poor recruitment in recent years. The estimated time series of F/FMSY suggests that overfishing has occurred throughout most of the assessment period (Table 8), but with some uncertainty in terminal years demonstrated by the MCB analysis (Figure 34). Current fishery status in the terminal year, with current F represented by the geometric mean from the period 2013–2015, was estimated by the base run to be F2013−2015/FMSY = 1.54, and the median value was F2013−2015/FMSY = 1.54 (Table 18). The fishery status was less robust than the stock status (Figures 35, 36). Of the MCB runs, approximately 89.1% agreed with the base run that the stock is currently experiencing overfishing.