16 Matching Annotations
  1. Jun 2025
    1. In general, a sample size of 100 is sufficient for age, sex, length, and weight data, except for yelloweye rockfish in Cook Inlet (CI) and Prince William Sound Outside (PWSO), where estimated sample sizes of 123 and 111 individuals, respectively, are required for length data. Recent sampling efforts in Cook Inlet have yielded very low numbers of yelloweye rockfish—1, 4, and 15 individuals in 2022, 2023, and 2024, respectively—highlighting the practical challenges of achieving the recommended sample sizes in this area.

      In general, I think we should develop this analysis further and explain your approach here.

    2. # the number of age bins; m = 3 is the worst case

      worst case in that the three age classes show similar representation (e.g., A1=.33, A2=.33, A3=.33)? Why was three chosen and not 2 or 4?

    3. age_breaks <- c(seq(1, 31, by = bin_width), 100) age_breaks

      Maybe overlay this with the Inside management area to show potential differences. Also expound upon why the bin sizes are difference between these two management areas.

    4. Sample Size Determination (Length and Weight)

      Your approach to satisfy the objective needs further explanation. I am not sure if the sample sizes derived here are based on historical methods for this project or just for this assessment (I know the answer but others who read this may not.)

    5. mydata %>% filter(Sex.Code %in% c(1, 2)) %>% mutate(Sex.Code = as.factor(Sex.Code)) %>% ggplot(aes(x = Sex.Code, fill = District)) + geom_bar(position = "dodge") + facet_wrap(~ Species, scales = "free_y") + labs(title = "Distribution of Sex by Species and Management Area", x = "Sex", y = "Count") + theme_minimal()

      A legend for sex ID is needed here. Also facet by management area and species. This would be much clearer. Also change the factor levels to "Male", "Female", "Unknown".

    6. For some areas, the sizes of fish were similar; others were not.

      A summary table of length, weight, and age data would be beneficial.

      It is also worth mentioning that the length and weight distributions for each species and area show similar trends, indicating quality data.

    7. It seems sample sizes for some species were too small to make meaningful estimates of population parameters (weight, length, age compositions, etc).

      Which species and areas were too small and what should the sample sizes have been to be meaningful? Also please standardize the thickness of the bars in each facet.

      Please add figure captions to all figures. There are various ways to accomplish this but I can send you code if you would like help with this.

    8. This report describes the data cleaning, exploration, and statistical methods used to determine appropriate sample sizes for assessment of groundfish harvest in Central Alaska. It includes three management areas: Cook Inlet (CI), Prince William Sound Inside (PWSI), and Prince William Sound Outside (PWSO). Resulting sample size estimates are intended to support the operational project plan: Assessment of Groundfish Harvest in Central Alaska, 2025–2026.

      Please expand on the introduction. Things to include would be a brief description of the of the sampling project, the historical sampling design (e.g., sample sizes) and what potential changes staff are seeking to make to the historical design based on this analysis (e.g., what is the motivation for this analysis). It would also be good to include your objectives here.

    Annotators

    1. This document is a draft and has not undergone review or approval through the Alaska Department of Fish and Game’s statewide publications System. Its contents are preliminary and subject to change.

      Please provide thoughts on this statement. Andrew suggests something like "deliberative".

    Annotators

    1. Late-run Kenai River sockeye salmon estimates derived from the age allocation model are used inseason to inform the Kenai River sockeye salmon inriver goal

      This is also really cool