MP43 focal hockeystick case
Status: focal case in the current analysis; not formally recommended or adopted. MP29 (HS+20) is retained as its annual-limit sensitivity case.
1 Identity and controls
| Component | Current 500-draw definition |
|---|---|
| Analysis code and label | tun43; MP43; HS-20 |
| Stock/index indicator | cpues.ind; mean of the five standardized indices |
| Reference and recent windows | 2019–2023 reference; 3-year recent average |
| Harvest control rule | Hockey-stick (hockeystick.hcr) |
| Minimum and target catch | 270 and 2,000 kt |
| Limit and refined trigger | 0.1 and 1.9375 |
| Annual decrease/increase limits | -20% / +15% (dlow = 0.80, dupp = 1.15) |
| Initial simulated catch | 1,385 kt |
2 Rule
Let \(I_t\) be the combined recent index. Before the annual change limit, the catch recommendation is
\[ C_t^* = \begin{cases} 270, & I_t \le 0.1,\\ 270 + (I_t-0.1)\dfrac{2000-270}{1.9375-0.1}, & 0.1 < I_t < 1.9375,\\ 2000, & I_t \ge 1.9375. \end{cases} \]
The annual constraint then limits advice to 80%–115% of the previous advice.
3 What the current analysis shows
MP43 was refined from the earlier 100-draw control using the same 500 posterior draws and tuning years (2041–2050) as the other seven CMPs. Its realized mean dynamic Kobe-green probability was 0.5918 against the 0.60 objective. The completed reference and nine-OM robustness results are included in the SC14 report and current Slick file.
Operational units, annual cutoffs, missing-data and rounding rules, and the formal exceptional-circumstances protocol still require approval before use.