さあ！諸君！捕鯨問題だ！

>CV＝0.21なら「常識の範囲」には十分勝てたんだけどねえｗ
>目視調査と比べてみ、CV＝0.21ならそんなに悪かねえよ。

CV＝0.21が悪くない、というのは目視で生息数を推定する場合ね。
幅のある生息数推定から捕獲枠を算出する場合、信頼区間の下側の値を
採用するから、RMP（改訂管理方式）の安全装置は機能します。

ところが自然死亡率のCVだと、安全性を確保しながら「常識の範囲」
を狭めるというときに0.21では使えない、というのがデラマーレ、
クック、ホルト、バターワースおよびIWC年報匿名査読者の共通の
判断だ。シンプルに、0.21では「常識の範囲」が狭まらないとも言える。
バターワースはほとんどいつも日本擁護の発言をする人だね。

デラマーレ論文の結論部と参照論文貼っときます。
CONCLUSION AND DISCUSSION

The analyses in this paper confirm the findings of
Sakuramoto and Tanaka (1989) that the trend in natural
mortality with age can be correctly determined from
catch-at-age data alone. However, this is far short of the
information required to be useful for the management of
whaling, which is the magnitude of natural mortality, not
just the difference in mortality between different
age-classes. It is shown in de la Mare (1985) that precise,
accurate estimates of natural mortality are required if they
are to be used in refining estimates of the likely sustainable
yield from a whale population.
The simulations by Sakuramoto and Tanaka (1989) are
presented as if the question at issue was whether natural
mortality was age-dependent or not. This is a secondary
consideration compared with estimating the actual level of
natural mortality. Even if this was the question, the
simulations show that the selected sample size of 825 is only
just sufficient to detect natural mortality increasing at
0.008 per year of age. De la Mare (1985) gives curves
showing the apparent mortality rate from commercial
catch-at-age data for Southern Hemisphere minke whales.
These show that age-dependent trends in mortality are
likely to be less than about half the value used by
Sakuramoto and Tanaka.
（つづく）