The following describes the workflow to obtain spatially-explicit
input sample size (ISS) for composition data provided by bottom trawl
surveys that occur in the GOA and AI. Please see the ‘Getting started’
vignette for installing and loading the surveyISS package,
as well as querying data. To obtain spatially-explicit ISS you will
follow the same steps outlined in the ‘Getting started’ vignette from
Step 1 through Step 4, here we provide instructions on
how to run the spatially-explicit cases to obtain ISS that would be
conducted in Step 5.
There are four spatially-explicit functions that have been developed, three for the GOA, and one for the AI. The same generic workflow is followed in each of these cases, they are simply applied to different spatial subregions. First, based upon the bottom trawl survey strata definitions (locations) the data are aggregated into the defined subregions. Second, this subregion data is resampled within the bootstrap simulation. Third, for each bootstrap replicate, length/age populations numbers are expanded within each of these subregions. Finally, the ISS is computed across the bootstrap replicates in the standard way.
The four spatially-explicit cases run in Step 5 of the
workflow described in the ‘Getting started’ vignette are as follows
(where the function arguments are the same as in the standard
surveyISS::srvy_iss() exampled provided, and are not
repeated here).
Case 1. This case was developed using the GOA rex sole stock
assessment as an example, and provides spatially-explicit ISS that
combines the western and central GOA separate from the eastern GOA. The
function for this case is
surveyISS::srvy_iss_goa_wc_e().
Case 2. This case was developed using the GOA
northern/southern rock sole stock assessment as an example, and provides
spatially-explicit ISS for western, central, and eastern GOA separately.
The function for this case is
surveyISS::srvy_iss_goa_w_c_e().
Case 3. This case was developed using the GOA walleye
pollock stock assessment as an example, and provides spatially-explicit
ISS for the GOA west of
140.
The function for this case is
surveyISS::srvy_iss_w140().
Case 4. This case was developed using the AI northern
rockfish stock assessment as an example, and provides spatially-explicit
ISS for the western AI, central AI, eastern AI, and southern BS
subregions within the AI bottom trawl survey. The function for this case
is surveyISS::srvy_iss_ai_subreg().
The output of these cases are in many ways similar to the output
provided by the standard surveyISS::srvy_iss() function,
with several nuanced differences. First, similar to the standard output,
there will be two output files, one for length composition ISS (denoted
by iss_ln in the file name) and the other for age
composition ISS (denoted by iss_ag in the file name). The
start of the file name will be whatever you have defined the
save argument in the function, and the end of the file name
will denote which spatially-explicit function the output has been
provided by (for example, if you have run the
surveyISS::srvy_iss_goa_wc_e() function, the end of the
output file name will be wc_egoa). Second, all the columns
in the output are identical to the standard
surveyISS::srvy_iss() function output (and we refer you to
the ‘Getting started’ vignette for column descriptions) with an
additional column region. The region column
denotes which region the specific ISS has been computed for. For
example, using the surveyISS::srvy_iss_goa_wc_e() function
there will be wcgoa and egoa regions within
the region column, or, for the
surveyISS::srvy_iss_goa_w_c_e() function there will be
wgoa, cgoa and egoa regions
within the region column. Within this output we have also
computed the survey-wide region ISS, which is a result of summing the
length/age population numbers across the subregions to the survey-wide
region (we computed this in order to compare the ISS as provided by
subregions to the standard method were data is pooled across the entire
survey region). Thus, within the region column there is an
additional region defined as either goa or ai,
depending on which spatially-explicit function that has been used. We
note that for the surveyISS::srvy_iss_w140() function there
is no region column, as the ISS provided is just for the
GOA west of
140
and the output is formatted identically to that provided by the standard
surveyISS::srvy_iss() function.