gap_check.RmdA couple of simple functions to evaluate differences between updated survey databases from GAP and historical databases.
Step 1 - connect to VPN
# load ----
library(afscdata)
# globals ----
year = 2024
species = 30420 # northern rockfish
area = 'ai'
type = 'region'There is a function named gap_check_bio() in the
afscdata package. It produces a brief report of differences
in the databases. Note that some of the function inputs differ slightly
from the afscdata queries to be more in line with GAP
naming conventions.
The area inputs available are: ai, goa, ebs, bss, nbs
The type inputs available are: region, subarea, area, stat_area,
stratum, inpfc, inpfc_depth, depth, reg_area_depth
Of course some of these types are not available in some areas, its
assumed the user is aware of this.
out <- gap_check_bio(year = year, species = species, type = 'region', area = area)
outa <- gap_check_bio(year = year, species = species, type = 'area', area = area)
outd <- gap_check_bio(year = year, species = species, type = 'depth', area = area)
outi <- gap_check_bio(year = year, species = species, type = 'inpfc', area = area)
outid <- gap_check_bio(year = year, species = species, type = 'inpfc_depth', area = area)The reports generated from these queries should be the same (or really, really close) since they are summing over the same data, just with different objectives.
The output is a list and the biomass and variance estimates can be compared by calling the report. Here, we can see there are minor differences that can likely be attributed to slight computational differences.
outa$report
# A tibble: 16 × 7
year tot var tot_g var_g bio_diff var_diff
<dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
1 1980 37593. 1146927834. NA NA NA NA
2 1983 56368. 75319232. NA NA NA NA
3 1986 140405. 2255266444. NA NA NA NA
4 1991 212812. 1080676741. 212813. 1080676741. -0.338 0
5 1994 88463. 1930605884. 88463. 1930605884. -0.2 0
6 1997 87391. 730236669. 87391. 730236669. -0.171 0
7 2000 205373. 3627754487. 205373. 3627754487. 0.213 0
8 2002 178791. 2268494951. 178791. 2268494951. -0.029 0
9 2004 189446 1753885854. 189446. 1753885854. -0.112 0
10 2006 219350. 2833224664. 219350. 2833224664. -0.134 0
11 2010 217319. 2256385707. 217318. 2256385707. 0.115 0
12 2012 285164. 20603776739. 285165. 20603776739. -0.303 0
13 2014 472895. 21594362893. 472895. 21594362893. 0.049 0
14 2016 253217. 2177621012. 253216. 2177621012. 0.139 0
15 2018 212472. 1852635907. 212471. 1852635907. 0.212 0
16 2022 287315. 4239548374. 287316. 4239548380. -0.746 -6.27However if we change the area to the GOA, then we note a difference between the regional and INPFC by depth estimates, particularly for 2005.
out <- gap_check_bio(year = year, species = species, type = 'region', area = 'GOA' )
outid <- gap_check_bio(year = year, species = species, type = 'inpfc_depth', area = 'GOA' )
out$report# A tibble: 18 × 7
year tot var tot_g var_g bio_diff var_diff
<dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
1 1984 39334. 127857605. NA NA NA NA
2 1987 136417. 1532588438. NA NA NA NA
3 1990 107076. 2068646138. 107076. 2068646159. 0.121 -21.2
4 1993 104992. 1358147916. 104992. 1358148033. -0.007 -117.
5 1996 98965. 707334568. 98965. 707334567. 0.148 0.938
6 1999 242187. 21641102940. 242187. 21641102895. -0.271 44.6
7 2001 343614. 42219837391. 343614. 42219837394. -0.151 -3.63
8 2003 66310. 1021150676. 66310. 1021150323. -0.041 353.
9 2005 358999. 17538253085. 358999. 17538251713. -0.088 1373.
10 2007 221226. 7153639777. 221226. 7153639738. -0.057 39.0
11 2009 89896. 834489465. 89896. 834489567. 0.058 -102.
12 2011 173642. 4504728201. 173642. 4504728555. -0.154 -354.
13 2013 370454. 48669985128. 370454. 48669985125. 0.055 3.37
14 2015 48933. 278528587. 48933. 278528608. -0.049 -20.9
15 2017 150326. 4609076209. 150326. 4609075707. 0.016 502.
16 2019 86725 942842957. 86725. 942842996. -0.263 -38.8
17 2021 90670. 1004131755. 90670. 1004131900. 0.061 -145.
18 2023 31758. 290679663. 31758. 290679663. 0.245 -0.042outid$report
# A tibble: 18 × 7
year tot var tot_g var_g bio_diff var_diff
<dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
1 1984 39334. 127857605. NA NA NA NA
2 1987 136417. 1532588438. NA NA NA NA
3 1990 107076. 2068646138. 107076. 2068646159. 0.121 -21.2
4 1993 104992. 1358147916. 104992. 1358148033. -0.007 -117.
5 1996 98965. 707334568. 98965. 707334567. 0.148 0.938
6 1999 242187. 21641102940. 242187. 21641102895. -0.271 44.6
7 2001 343614. 42219837391. 343614. 42219837394. -0.151 -3.63
8 2003 66310. 1021150676. 66310. 1021150323. -0.041 353.
9 2005 358999. 17538253085. 358961. 17538250296. 37.5 2789.
10 2007 221226. 7153639777. 221226. 7153639738. -0.057 39.0
11 2009 89896. 834489465. 89896. 834489567. 0.058 -102.
12 2011 173642. 4504728201. 173642. 4504728555. -0.154 -354.
13 2013 370454. 48669985128. 370454. 48669985125. 0.055 3.37
14 2015 48933. 278528587. 48933. 278528608. -0.049 -20.9
15 2017 150326. 4609076209. 150326. 4609075707. 0.016 502.
16 2019 86725 942842957. 86725. 942842996. -0.263 -38.8
17 2021 90670. 1004131755. 90670. 1004131900. 0.061 -145.
18 2023 31758. 290679663. 31758. 290679663. 0.245 -0.042We can dril down to see where this difference may be coming from (and
chat with GAP if there are questions). Note the different variable names
in the datasets.
In this instance it looks like the number of hauls is being treated
differently within the two datasets.
outid$orig %>%
group_by(year) %>%
summarise(haul = sum(haul_count),
count = sum(catch_count)) %>%
left_join(
outid$gap %>%
group_by(year) %>%
summarise(haul_g = sum(n_haul),
count_g = sum(n_count))
) %>%
mutate(haul_diff = haul - haul_g,
count_diff = count - count_g)Joining with `by = join_by(year)`
# A tibble: 18 × 7
year haul count haul_g count_g haul_diff count_diff
<dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
1 1984 929 157 NA NA NA NA
2 1987 783 202 NA NA NA NA
3 1990 708 130 708 129 0 1
4 1993 774 148 774 148 0 0
5 1996 807 135 807 135 0 0
6 1999 764 119 717 119 47 0
7 2001 489 107 489 107 0 0
8 2003 809 126 794 126 15 0
9 2005 837 147 802 146 35 1
10 2007 816 139 777 139 39 0
11 2009 823 132 795 132 28 0
12 2011 670 89 653 89 17 0
13 2013 548 86 536 86 12 0
14 2015 771 95 743 95 28 0
15 2017 536 92 525 92 11 0
16 2019 541 74 531 74 10 0
17 2021 529 70 519 70 10 0
18 2023 526 52 516 52 10 0