detection_plot {eDNAjoint} | R Documentation |
Plot the survey effort necessary to detect species presence, given the species expected catch rate.
Description
This function plots the number of survey effort units to necessary detect species presence, calculated using median estimated parameter values from joint_model(). Detecting species presence is defined as producing at least one true positive eDNA detection or catching at least one individual. See more examples in the Package Vignette.
Usage
detection_plot(
model_fit,
mu_min,
mu_max,
cov_val = NULL,
probability = 0.9,
pcr_n = 3
)
Arguments
model_fit |
An object of class |
mu_min |
A value indicating the minimum expected species catch rate for plotting. If multiple traditional gear types are represented in the model, mu is the catch rate of gear type 1. |
mu_max |
A value indicating the maximum expected species catch rate for plotting. If multiple traditional gear types are represented in the model, mu is the catch rate of gear type 1. |
cov_val |
A numeric vector indicating the values of site-level covariates to use for prediction. Default is NULL. |
probability |
A numeric value indicating the probability of detecting presence. The default is 0.9. |
pcr_n |
An integer indicating the number of PCR replicates per eDNA sample. The default is 3. |
Value
A plot displaying survey efforts necessary to detect species presence, given mu, for each survey type.
Note
Before fitting the model, this function checks to ensure that the function is possible given the inputs. These checks include:
Input model fit is an object of class 'stanfit'.
Input mu_min is a numeric value greater than 0.
Input mu_max is a numeric value.
If model fit contains alpha, cov_val must be provided.
Input cov_val is numeric.
Input cov_val is the same length as the number of estimated covariates.
Input probability is a univariate numeric value.
Input model fit has converged (i.e. no divergent transitions after warm-up).
If any of these checks fail, the function returns an error message.
Examples
# Ex. 1: Calculating necessary effort for detection with site-level
# covariates
# Load data
data(goby_data)
# Fit a model including 'Filter_time' and 'Salinity' site-level covariates
fit_cov <- joint_model(data = goby_data, cov = c('Filter_time','Salinity'),
family = "poisson", p10_priors = c(1,20), q = FALSE,
multicore = FALSE)
# Plot at the mean covariate values (covariates are standardized, so mean=0)
detection_plot(fit_cov$model, mu_min = 0.1, mu_max = 1,
cov_val = c(0,0), pcr_n = 3)
# Calculate mu_critical at salinity 0.5 z-scores greater than the mean
detection_plot(fit_cov$model, mu_min = 0.1, mu_max = 1, cov_val = c(0,0.5),
pcr_n = 3)
# Ex. 2: Calculating necessary effort for detection with multiple
# traditional gear types
# Load data
data(green_crab_data)
# Fit a model with no site-level covariates
fit_q <- joint_model(data = green_crab_data, cov = NULL, family = "negbin",
p10_priors = c(1,20), q = TRUE,
multicore = FALSE)
# Calculate
detection_plot(fit_q$model, mu_min = 0.1, mu_max = 1,
cov_val = NULL, pcr_n = 3)
# Change probability of detecting presence to 0.95
detection_plot(fit_q$model, mu_min = 0.1, mu_max = 1, cov_val = NULL,
probability = 0.95, pcr_n = 3)