Sea Age Exploration
LGCarlson
7/10/2019
Read in growth and release datasets.
Carlin_full_growth<-readRDS(paste(shared.path, "Salmon/Data/Growth/Data/Carlin_growth_fulldata.rds", sep = ""))
releasedata<- readRDS(paste(shared.path, "Salmon/Data/Mark-recap/Carlin_all_releases.rds", sep = ""))Calculate number of circuli deposited after the last marine annulus
How many circuli are deposited after the annulus?
Carlin_full_growth %>%
mutate(M1.circ = ifelse(is.na(M1) == FALSE, (smolt.circ + FS.circ + FW.circ), NA)) %>%
mutate(M2.circ = ifelse(is.na(M2) == FALSE, (M1.circ + SS.circ + SW.circ), NA)) %>%
select(JoinID,FreshAge, SeaAge, Circ, M1.circ, M2.circ, RecaptureYear, RecaptureMonthNum) %>%
filter(!is.na(M1.circ)) %>%
filter(!is.na(RecaptureMonthNum)) %>%
filter(SeaAge == 1 | SeaAge == 2) %>%
mutate(postannulus.circ = ifelse(is.na(M2.circ) == FALSE, (Circ - M2.circ), (Circ - M1.circ))) %>%
ggplot(aes(postannulus.circ)) + geom_histogram() + facet_wrap(~SeaAge)
Do fish caught later in the year have more circuli after the annulus?
It doesn’t seem that way. Months are pretty even.
Carlin_full_growth %>%
mutate(M1.circ = ifelse(is.na(M1) == FALSE, (smolt.circ + FS.circ + FW.circ), NA)) %>%
mutate(M2.circ = ifelse(is.na(M2) == FALSE, (M1.circ + SS.circ + SW.circ), NA)) %>%
select(JoinID,FreshAge, SeaAge, Circ, M1.circ, M2.circ, RecaptureYear, RecaptureMonthNum) %>%
filter(!is.na(M1.circ)) %>%
filter(!is.na(RecaptureMonthNum)) %>%
filter(SeaAge == 1) %>%
mutate(postannulus.circ = ifelse(is.na(M2.circ) == FALSE, (Circ - M2.circ), (Circ - M1.circ))) %>%
tidyr::complete(postannulus.circ,RecaptureMonthNum, fill = list(n=0)) %>%
group_by(postannulus.circ, RecaptureMonthNum) %>%
dplyr::summarise(n = n()) %>%
mutate(percent = (n / sum(n) * 100)) %>%
ggplot(aes(postannulus.circ,percent, fill = as.factor(RecaptureMonthNum))) + geom_area() + xlim(5,25) +
labs(x = "Number of postannulus circuli", y = "Percent", fill = "Month caught", title = "1SW") + theme(panel.background = element_blank()) + theme(panel.grid = element_blank())
Carlin_full_growth %>%
mutate(M1.circ = ifelse(is.na(M1) == FALSE, (smolt.circ + FS.circ + FW.circ), NA)) %>%
mutate(M2.circ = ifelse(is.na(M2) == FALSE, (M1.circ + SS.circ + SW.circ), NA)) %>%
select(JoinID,FreshAge, SeaAge, Circ, M1.circ, M2.circ, RecaptureYear, RecaptureMonthNum) %>%
filter(!is.na(M1.circ)) %>%
filter(!is.na(RecaptureMonthNum)) %>%
filter(SeaAge == 2) %>%
mutate(postannulus.circ = ifelse(is.na(M2.circ) == FALSE, (Circ - M2.circ), (Circ - M1.circ))) %>%
tidyr::complete(postannulus.circ,RecaptureMonthNum, fill = list(n = 0)) %>%
group_by(postannulus.circ, RecaptureMonthNum) %>%
dplyr::summarise(n = n()) %>%
mutate(percent = (n / sum(n)*100)) %>%
ggplot(aes(postannulus.circ,percent, fill = as.factor(RecaptureMonthNum))) + geom_area() + xlim(3,15) +
labs(x = "Number of postannulus circuli", y = "Percent", fill = "Month caught" , title = "2SW") +
theme(panel.background = element_blank()) + theme(panel.grid = element_blank())
Sea Age errors?
There are 5 fish classified as 0SW that have a marine annulus.
## JoinID FreshAge SeaAge Circ M1.circ M2.circ
## 1 1968_6931 H1 0 71 61 NA
## 2 1970_44155 H2 0 74 61 NA
## 3 1973_61966 H2 0 98 72 NA
## 4 1984_79462 H1 0 108 84 NA
## 5 1984_89176 H1 0 93 74 NAThere are 6 fish classified as 1SW that have a second marine annulus.
## JoinID FreshAge SeaAge Circ M1.circ M2.circ
## 1 1966_661008 H2 1 111 71 100
## 2 1973_66686 H2 1 119 73 101
## 3 1975_89317 H2 1 81 64 77
## 4 1979_152340 H2 1 81 48 74
## 5 1981_285462 H2 1 78 52 71
## 6 1984_55278 H1 1 103 55 82There are 18 fish classified as 2SW that don’t have a second marine annulus.
## JoinID FreshAge SeaAge Circ M1.circ M2.circ
## 1 1966_663052 H2 2 90 60 NA
## 2 1966_663067 H2 2 99 78 NA
## 3 1966_664021 H1 2 106 79 NA
## 4 1968_7138 H2 2 107 75 NA
## 5 1969_5314 H2 2 92 78 NA
## 6 1969_9178 H2 2 90 75 NA
## 7 1971_74901 H1 2 111 81 NA
## 8 1971_87680 H1 2 93 79 NA
## 9 1972_31365 H2 2 111 90 NA
## 10 1972_53511 H2 2 95 75 NA
## 11 1972_56068 H2 2 84 67 NA
## 12 1972_6446 H2 2 105 85 NA
## 13 1972_8910 H2 2 73 59 NA
## 14 1973_87760 H2 2 101 80 NA
## 15 1975_81371 H2 2 77 63 NA
## 16 1979_182389 H2 2 69 54 NA
## 17 1980_206348 H1 2 97 84 NA
## 18 1984_87190 H1 2 93 71 NAHow many circuli are deposited after the annulus?
Carlin_full_growth %>%
mutate(M1.circ = ifelse(is.na(M1) == FALSE, (smolt.circ + FS.circ + FW.circ), NA)) %>%
mutate(M2.circ = ifelse(is.na(M2) == FALSE, (M1.circ + SS.circ + SW.circ), NA)) %>%
select(JoinID,FreshAge, SeaAge, Circ, M1.circ, M2.circ) %>%
filter(!is.na(M1.circ)) %>%
filter(SeaAge == 1 | SeaAge == 2) %>%
mutate(postannulus.circ = ifelse(is.na(M2.circ) == FALSE, (Circ - M2.circ), (Circ - M1.circ))) %>%
ggplot() + geom_histogram(aes(postannulus.circ, fill = SeaAge)) + labs(x = "Number of circuli after last winter annulus", y = "n") + theme(panel.grid = element_blank())
Number of circuli between second annulus and first annulus ~ 24.5 circuli.
## mean(M2.circ - M1.circ, na.rm = TRUE)
## 1 24.62264Calculate mean recapture dates
Create “RecaptureDate” formatted as a date.
Create and format range of release dates (Date 1 and Date 2). Calculate # days between Release Date 1 and Release Date 2.
Write function to calculate midpoint of interval. Calculate midpoint of release interval using “int_midpoint” function. Add it to the dataframe as “EmigrationDate.”
Add release dates dataframe to full growth dataframe. Remove fish with unknown recapture or emigration dates.
Calculate mean release dates
library(dplyr)
meanrelease<-Carlin_simpledata %>%
mutate(ReleaseMonth = month(EmigrationDate)) %>%
mutate(ReleaseYear = year(EmigrationDate)) %>%
dplyr::group_by(ReleaseYear) %>%
dplyr::summarise(meanReleaseDate = mean(EmigrationDate), meanReleaseMonth = mean(ReleaseMonth), sdMonth = sd(ReleaseMonth), n = n())
meanrelease## # A tibble: 25 x 5
## ReleaseYear meanReleaseDate meanReleaseMonth sdMonth n
## <dbl> <date> <dbl> <dbl> <int>
## 1 1966 1966-04-23 4.2 1.10 30
## 2 1967 1967-04-11 4 0 10
## 3 1968 1968-04-22 4.4 0.507 15
## 4 1969 1969-05-18 5.35 0.521 188
## 5 1970 1970-05-05 5 0 244
## 6 1971 1971-05-14 5 0 68
## 7 1972 1972-05-13 5 0 103
## 8 1973 1973-05-09 5 0 158
## 9 1974 1974-05-05 4.93 0.264 81
## 10 1975 1975-05-05 4.97 0.184 58
## # ... with 15 more rows#write.csv(meanrelease, "meanrelease.csv")Postsmolt increment relationship with emigration DOY
## Warning: Column `JoinID` joining character vector and factor, coercing into
## character vectorThe relationship between emigration and postsmolt increment doesn’t seem particularily strong, though DOY is significant in the lm.
| term | estimate | std.error | statistic | p.value |
|---|---|---|---|---|
| (Intercept) | 1.85 | 0.12 | 15.59 | 0.0000 |
| smoltgrowth$EmigrationDOY | 0.00 | 0.00 | -3.03 | 0.0025 |
Marine increment relationship with emigration DOY
The relationship between emigration and marine increment doesn’t seem particularily strong, though DOY is significant in the lm.
| term | estimate | std.error | statistic | p.value |
|---|---|---|---|---|
| (Intercept) | 2.83 | 0.17 | 16.29 | 0.0000 |
| smoltgrowth$EmigrationDOY | 0.00 | 0.00 | -2.87 | 0.0042 |
Perhaps being an early emigrator doesn’t make you grow any faster or larger!