In jBernardADFG/LakeLouiseBurbotOP2020: Functions used in the 2020 Lake Louise OP

library(rlist)
library(prevR)
library(LakeLouiseBurbotOP2020)

To get the minimum sample size so that the length composition proportions are within 10 percentage points of the true values 95% of the time.

128 fish should do it.

length.comp.ss(0.05, 0.10) 

To set the precision criterion for CPUE and abundance

The tables give the percentage of time that an accuracy criterion is satisfied for diffent levels of true abundance (rows) and between transect log scale standard deviation (columns). Note: abundance was roughly 5000 fish and the between transect log scale standard deviation was 0.34 in 2005. With 492 sets, use "estimate mean CPUE within 30% of the asymptotic value 90% of the time" for objective 2; use "estimate abundance within 40% of the true value 90% of the time" for objective 3.

load("S:/Jordy/louiseOP2020/Data/lakelines.Rdata") # Load Lake Louise lakelines
n_max <- 492
seed <- set.seed(666)
ts_mat <- pick.coords(lakelines, n_max, seed)$ts_mat
N_vec <- seq(1000, 10000, by=1000) 
sig_vec <- seq(.1, 1, by=.1)
q_bar <- 0.6478999
sd_q <- 0.07966257
A <- 6519
acc_lev <- c(0.3, 0.3)
# 30 percent accuracy
assess.accuracy(ts_mat, N_vec, sig_vec, q_bar, sd_q, acc_lev, A)
acc_lev <- c(0.4, 0.4)
# 40 percent accuracy
assess.accuracy(ts_mat, N_vec, sig_vec, q_bar, sd_q, acc_lev, A)
acc_lev <- c(0.5, 0.5)
# 50 percent accuracy
print("50 percent accuracy")
assess.accuracy(ts_mat, N_vec, sig_vec, q_bar, sd_q, acc_lev, A)

To get power calculations

The table gives the power of concluding that mean CPUE has increased at the $\alpha=0.10$ significance level for different levels of 2020 mean CPUE (rows) and between transect log scale standard deviation (columns). The 2020 true mean CPUE will need to be ~0.65, to detect an increase 80 percent of the time.

set.seed(666)
ct_mat <- get.ct.mat("S:/Jordy/louiseOP2020/Data/CPUE.xls", "CPUEL")
ct_mat <- ct_mat[,-11]
mu_vec <- seq(0.1, 1, by=.1)
sig_vec <- c(0.05, 0.35, 0.85)
alpha <- 0.10
calc.power(ct_mat, ts_mat, mu_vec, sig_vec, alpha)

To generate a map of the sampling design and get transect/set coordinate information

trans_set_info <- pick.coords(lakelines, n_max, seed, quiet=F)
df <- data.frame(nsets=trans_set_info$table)
names(df) <- c("lon", "n_sets")
df
coords <- trans_set_info$set_coords
names(coords) <- c("lon", "lat")
head(coords)

jBernardADFG/LakeLouiseBurbotOP2020 documentation built on March 25, 2020, 12:06 a.m.