Data Requirements

To use mscatch you will need to compile two data sets for a specific region of interest, whether a stock area, an ecological production unit, or a larger area:

• Length data
• Catch data

If you want a sample data set to begin to use mscatch then you can use the function get_sample_data

Length data

Length data (lengths of sampled catch taken by port samplers) can be pulled from the oracle table (stockeff.mv_cf_len) under the stock efficiency initiative or from the annual (cflen, wolen) AA tables. It is then aggregated by YEAR, QTR, GEAR, MARKET_CODE to summarize two data sets

• The size of the biological samples taken on trips
• Individual fish lengths and their frequencies from these samples

Size of biological samples

• The number of length samples (number trips in which samples were taken)
• The total number of lengths (number of actual fish that were sampled)

This is then merged with the catch data.

Individual fish lengths

• The individual lengths of a sampled fish (1 cm increments)
• The number of times a fish of this length was sampled

The data set should have this form

#> # A tibble: 11 × 6
#>     YEAR   QTR NEGEAR MARKET_CODE LENGTH NUMLEN
#>    <int> <int> <chr>  <chr>        <dbl>  <int>
#>  1  1975     2 050    UN              37      4
#>  2  1975     2 050    UN              38      2
#>  3  1975     2 050    UN              25      1
#>  4  1975     2 050    UN              28      1
#>  5  1975     2 050    UN              30      1
#>  6  1975     2 050    UN              31      4
#>  7  1975     2 050    UN              32      8
#>  8  1975     2 050    UN              33      7
#>  9  1975     2 050    UN              34     14
#> 10  1975     2 050    UN              35      3
#> 11  1975     2 050    UN              36      4

where each row represents the number of fish sampled of a given length in YEAR, QTR, caught using NEGEAR and classed as MARKET_CODE

• LENGTH is the length in cm of sampled fish and
• NUMLEN is the number of fish of this size sampled

Catch data

Landings data can be pulled from the oracle table (stockeff.mv_cf_landings) under the stock efficiency initiative or from the annual (cfdet, wodet, woland) AA tables. If pulled from the AA tables, the MARKET_CODE field does not exist. NESPP4 codes will need to be mapped to MARKET_CODE categories outlined marketCodeLookupTable. (created from comlandr::get_species_itis)

Some trips will have missing information in the AREA field. These trips will need to be assigned an area using the methods outlined by (Palmer, 2008) prior to using mscatch. The package comlandr replicates these methods.

Trips with missing gear information is handled in mscatch. It is assumed that trips with unknown gear information are catching fish with lengths proportional to trips with known gear types.

Discards are not included in this data. At a future date both comlandr and the stock efficiency initiative will contain discard estimates using the methods of (Wigley & Tholke, 2020).

The resulting Catch data is then merged with the biological samples to summarize total catch (metric tons) by YEAR, QTR, GEAR, MARKET_CODE, the number of biological samples taken (landings_nn), and the total number of fish sampled (len_totalNumlen). The data set should have this form:

#> # A tibble: 15 × 8
#>     YEAR   QTR NEGEAR MARKET_CODE landings_land landings_nn len_totalNumLen
#>    <int> <int> <chr>  <chr>               <dbl>       <int>           <dbl>
#>  1  2000     1 010    UN                  10935           7              NA
#>  2  2000     1 050    UN                 185019          27              NA
#>  3  2000     1 100    UN                 138712          65              NA
#>  4  2000     1 360    UN                     34           1              NA
#>  5  2000     2 010    UN                 525450         117             250
#>  6  2000     2 020    UN                     81           3              NA
#>  7  2000     2 050    UN                  12754          45              NA
#>  8  2000     2 100    UN                 319747         433             125
#>  9  2000     2 360    UN                   2496          12              NA
#> 10  2000     3 010    UN                3122083         512             655
#> 11  2000     3 020    UN                 128054          41              NA
#> 12  2000     3 050    UN                   3680          11              NA
#> 13  2000     3 100    UN                 280584         266             100
#> 14  2000     4 050    UN                   2160           9              NA
#> 15  2000     4 100    UN                   4912          16              NA
#> # ℹ 1 more variable: len_numLengthSamples <int>

Use mscatch

Once the above data sets have been assembled they can now be used to aggregate the landings and length samples and then expand the commercial landings data to length distributions using mscatch

References

Palmer, M. C. (2008). A method to apportion landings with unknown area, month and unspecified market categories among landings with similar region and fleet characteristics. NOAA Working Paper.

Wigley, S. E., & Tholke, C. (2020). 2020 discard estimation, precision, and sample size analyses for 14 federally managed species groups in the waters off the northeastern united states. NOAA Working Paper, NMFS-NE-261.