Table 4: Input data structure for AGEPRO Recruitment Models

Table 4: Required input data for AGEPRO recruitment models, where spawning biomass and recruitment inputs are measured in units of RECRUIT the conversion factors SSBFac and RecFac respectively, which typically have units of SSBFac=RecFac=1000.
Model Number Recruitment Model Recruitment Model Input Description
1 Markov Matrix
  • Input the number of Recruitment States: \(K\)
  • On the next line, input the recruitment values: \(R_1,R_2,...,R_K\)
  • On the next line, input number of spawning biomass states: \(J\)
  • On the next line, input \(J-1\) cut points : \(B_{S,1},B_{S,2},...,B_{S,J-1}\)
  • On the next \(J\) lines, input the conditional recruitment probabilities for the spawning biomass states:
    • \(P_{1,1}, \quad P_{1,2}, \quad \dots, \quad P_{1,K}\)
      \(P_{2,1}, \quad P_{2,2}, \quad \dots, \quad P_{2,k}\)
      \(\quad \vdots \quad\quad \vdots \qquad\quad \vdots \qquad\quad \vdots\)
      \(P_{J,1}, \quad P_{J,2}, \quad \dots, \quad P_{J,K}\)
2 Empirical Recruits Per Spawning Biomass Distribution
  • Input the number of stock recruitment data points: \(T\)
  • On the next line, input the recruitments: \(R_1,R_2,...,R_T\)
  • On the next line, input the spawning biomasses: \(B_{S,1}, B_{S,2}, ..., B_{S,T}\)
3 Empirical Recruitment Distribution
  • Input the number of recruitment data points: \(T\)
  • On the next line, input the recruitments: \(R_1,R_2,...,R_T\)
4 Two-Stage Empirical Recruits Per Spawning Biomass Distribution
  • Input the number of low and high recruits per spawning biomass data points: \(T_{Low} \cdot T_{High}\)
  • On the next line, input the cutoff level of spawning biomass: \(B^*_S\)
  • On the next line, input the low state recruitments: \(R_1,R_2,...,R_{T_{Low}}\)
  • On the next line, input the low state spawning biomasses: \(B_{S,1}, B_{S,2}, ..., B_{S,T_{Low}}\)
  • On the next line, input the high state recruitments: \(R_1,R_2,...,R_{T_{High}}\)
  • On the next line, input the high state spawning biomasses: \(B_{S,1}, B_{S,2}, ..., B_{S,T_{High}}\)
5 Beverton-Holt Curve with Lognormal Error
  • Input the stock-recruitment parameters: \(\alpha, \beta, \sigma^2_w\)
6 Ricker Curve with Lognormal Error
  • Input the stock-recruitment parameters: \(\alpha, \beta, \sigma^2_w\)
7 Shepherd Curve with Lognormal Error
  • Input the stock-recruitment parameters: \(\alpha, \beta, k, \sigma^2_w\)
8 Lognormal Distribution
  • Input the log-scale mean and standard deviation: \(\mu_{\log(r)},\sigma_{\log(r)}\)
10 Beverton-Holt Curve with Autocorrected Lognornal Error
  • Input the stock-recruitment parameters: \(\alpha, \beta, \sigma^2_w\)
  • On the next line, input the autoregressive parameters: \(\phi, \varepsilon_{0}\)
11 Ricker Curve with Autocorrected Lognormal Error
  • Input the stock-recruitment parameters: \(\alpha, \beta, \sigma^2_w\)
  • On the next line, input the autoregressive parameters: \(\phi, \varepsilon_{0}\)
12 Shepherd Curve with Autocorrected Lognormal Error
  • Input the stock-recruitment parameters: \(\alpha, \beta, k, \sigma^2_w\)
  • On the next line, input the autoregressive parameters: \(\phi, \varepsilon_{0}\)
13 Autocorrected Lognormal Distribution
  • Input the log-scale mean and standard deviation: \(\mu_{\log(r)},\sigma_{\log(r)}\)
  • On the next line, input the autoregressive parameters: \(\phi, \varepsilon_{0}\)
14 Empirical Cumulative Distribution Function of Recruitment
  • Input the number of recruitment data points: \(T\)
  • On the next line, input the recruitments \(R_1,R_2,...,R_T\)
15 Two-Stage Empirical Cumulative Distribution Function of Recruitment
  • Input the number of low and high recruits per spawning biomass data points: \(T_{Low} \cdot T_{High}\)
  • On the next line, input cutoff level of spawning biomass: \(B^*_S\)
  • On the next line, input the low state recruitments: \(R_1,R_2,...,R_{T_{Low}}\)
  • On the next line, input the high state recruitments: \(R_1,R_2,...,R_{T_{High}}\)
16 Linear Recruits Per Spawning Biomass Predictor with Normal Error
  • Input the predictors: \(N_P\)
  • On the next line, input the intercept coefficient: \(\beta_0\)
  • On the next line, input the slope coefficient for each predictor: \(\beta_1, \beta_2, ..., \beta_{N_p}\)
  • On the next line, input the error variance: \(\sigma^2\)
  • On the next \(N_P\) lines, input the expected value of the predictor through projection time horizon:
    • \(X_1(1), \quad\ \dots, \quad\ X_1(Y)\)
      \(X_2(1), \quad\ \dots, \quad\ X_2(Y)\)
      \(\ \vdots \qquad\qquad\ \ \vdots \qquad\quad\ \ \vdots\)
      \(X_P(1), \quad\ \dots, \quad\ X_P(Y)\)
17 Linear Recruits Per Spawning Biomass Predictor with Lognormal Error
  • Input the number of predictors: \(N_P\)
  • On the next line, input the intercept: \(\beta_0\)
  • On the next line, input the linear coefficient for each predictor: \(\beta_1, \beta_2, ..., \beta_{N_P}\)
  • On the next line, input the log-scale error variance: \(\sigma^2\)
  • And on the next \(N_P\) lines, input the expected predictor values over the forecast time horizon \(1, ..., Y\)
    • \(X_1(1) \quad X_1(2) \quad \dots \quad X_1(Y)\)
      \(X_2(1) \quad X_2(2) \quad \dots \quad X_2(Y)\)
      \(\quad \vdots \qquad\quad\ \vdots \qquad\quad \vdots \qquad\quad \vdots\)
      \(X_P(1) \quad X_P(2) \quad \dots \quad X_P(Y)\)
18 Linear Recruitment Predictor with Normal Error
  • Input the number of predictors: \(N_P\)
  • On the next line, input the intercept: \(\beta_0\)
  • On the next line, input the linear coefficient for each predictor: \(\beta_1, \beta_2, ..., \beta_{N_P}\)
  • On the next line, input the error variance: \(\sigma^2\)
  • And on the next \(N_P\) lines, input the expected predictor values over the forecast time horizon \(1, ..., Y\)
    • \(X_1(1) \quad X_1(2) \quad \dots \quad X_1(Y)\)
      \(X_2(1) \quad X_2(2) \quad \dots \quad X_2(Y)\)
      \(\quad \vdots \qquad\quad\ \vdots \qquad\quad \vdots \qquad\quad \vdots\)
      \(X_P(1) \quad X_P(2) \quad \dots \quad X_P(Y)\)
19 Loglinear Recruitment Predictor with Lognormal Error
  • Input the number of predictors: \(N_P\)
  • On the next line, input the intercept: \(\beta_0\)
  • On the next line, input the linear coefficient for each predictor: \(\beta_1, \beta_2, ..., \beta_{N_P}\)
  • On the next line, input the log-scale error variance: \(\sigma^2\)
  • And on the next \(N_P\) lines, input the expected predictor values over the forecast time horizon \(1, ..., Y\)
    • \(X_1(1) \quad X_1(2) \quad \dots \quad X_1(Y)\)
      \(X_2(1) \quad X_2(2) \quad \dots \quad X_2(Y)\)
      \(\quad \vdots \qquad\quad\ \vdots \qquad\quad \vdots \qquad\quad \vdots\)
      \(X_P(1) \quad X_P(2) \quad \dots \quad X_P(Y)\)
20 Fixed Recruitment
  • Input the number of recruitment data points: \(T\)
  • On the next line, input the Recruitments: \(R_1,R_2,...,R_T\)
21 Empirical Cumulative Distribution Function of Recruitment with Linear Decline to Zero
  • Input the number of number of observed recruitment values: \(T\)
  • On the next line, input the recruitment values: \(R_1, R_2, ..., R_T\)
  • And on the next line, input spawning biomass threshold: \(B^*_S\)