Bycatch Audit¶
Test or assessment — instantiates Selectivity-Window Calibration
Examines non-target capture or collateral burden that ordinary success metrics ignore.
A success dashboard that counts only the target is a lie of omission — it makes an indiscriminate process look precise. Bycatch Audit is the operational accounting that fills the omission: it measures what the selector actually swept up during ordinary operation — the non-target captures and collateral burden the headline metric never records — and attaches a real number to the hidden cost. Its defining feature is that it counts real captures in the field, retrospectively and continuously, from a representative sample of live operation rather than from designed test cases. Where the target metric asks "how much target did we get?", the audit asks the complementary and usually unasked question: "and what else came up in the net?"
Example¶
A shrimp trawl fishery reports its performance in kilograms of shrimp landed — a clean, rising number. But a trawl net is not selective, and the landing metric is silent about everything else the net brings aboard. A bycatch audit places independent observers on a representative sample of trips (with cameras filling the gaps) to record the whole haul: juvenile finfish, crabs, and the occasional sea turtle, all normally discarded and uncounted. Tallied and normalized as non-target mass per unit of shrimp, the audit reveals several kilograms of discarded life for every kilogram landed, with turtle interactions concentrated on the longest tows. None of this existed in the shrimp dashboard; the audit is what converts an invisible externality into a tracked, attributable figure the fishery can actually manage against.
How it works¶
- Define a sampling frame of real operations. Choose a representative slice of live activity — trips, batches, decisions — to observe, since auditing everything is rarely feasible.
- Instrument independent observation. Place observers, cameras, or logs that record all captures, not just the target, and that the operator cannot quietly curate.
- Count and normalize the collateral. Tally each non-target category and express it per unit of effort or per unit of target, so burden is comparable across settings.
- Attribute to settings. Tie collateral levels back to the operating conditions that produced them, feeding the two-sided response the target metric omitted.
Tuning parameters¶
- Observer / sample coverage — what fraction of operations is watched. Higher coverage tightens the estimate but costs observers, cameras, and time.
- Collateral definition — which non-target categories and sublethal harms count. A broader definition captures more real burden but raises measurement effort and dispute.
- Normalization basis — burden per unit effort vs. per unit target. Each basis tells a different and legitimately contestable story.
- Audit cadence — one-off vs. continuous. Continuous auditing tracks change but is a standing operational cost.
When it helps, and when it misleads¶
Its strength is turning an externality nobody was counting into a number somebody now owns — the missing half of the response map that lets a process be judged on precision rather than on raw target throughput.
Its failure mode is coverage: low observer coverage or systematically unobserved effort undercounts the burden, and an operator who knows when the observer is aboard can behave differently on unwatched trips.[n1] The classic misuse is the target-only optimization the audit exists to correct — celebrating rising landings while the discard ratio quietly climbs out of frame. The guarding discipline is independent, adequately sampled observation (ideally tamper-resistant, like continuous camera coverage) so the audited slice honestly represents the whole.
How it implements the components¶
bycatch_and_cross_reactivity_monitor— its core function: the standing measurement of non-target capture recorded separately from target success.selectivity_response_map— it populates the collateral-burden layer of the map, tying observed non-target capture back to the operating conditions.
It counts what the selector really captured in the field but does not probe it with manufactured hard cases — the curated challenge_or_decoy_panel and target_and_non_target_reference_set are Challenge-Panel Cross-Reactivity Test's.
Related¶
- Instantiates: Selectivity-Window Calibration — this audit supplies the non-target-burden side of the response the target metric hides.
- Sibling mechanisms: Selectivity Curve Sweep · Operating Band Specification · Challenge-Panel Cross-Reactivity Test · ROC or Precision–Recall Surface Review · Window Drift Control Chart · Selective Admission Band Protocol
Editorial Notes¶
Form Classification¶
Form family: Assessment, Review & Assurance
Rationale: Examines non-target capture or collateral burden that ordinary success metrics ignore, making its operative form a bounded evaluation of existing evidence or work that produces a finding or disposition.
Independent corroboration: The frozen evidence defines Bycatch Audit as 'Examines non-target capture or collateral burden that ordinary success metrics ignore', so its operative form is Assessment, Review & Assurance.
Review outcome: Independent reviewer agreement; high confidence.
Origin Attribution¶
Primary origin: Marine Science & Oceanography
Origin pattern: Single lineage
Present-day reach: Specialized
Rationale: Fisheries science made bycatch a named non-target-capture problem measured through observer coverage and live catch accounting.
Related originating lineages:
- Environmental Science & Climate Studies — Environmental assessment generalizes the audit to collateral ecological burdens beyond a target yield.
Review resolution: Marine science is the agreed primary lineage because fisheries made bycatch a named non-target-capture problem measured through observer coverage and catch accounting. Environmental assessment is formative, while the mechanism's origin remains specialized even though its logic can be borrowed elsewhere.
Review outcome: Reconciled after independent review; high confidence.
Notes¶
The nearest twin is Challenge-Panel Cross-Reactivity Test; the difference is which world the evidence comes from. The bycatch audit observes real collateral capture already happening in operation; the challenge panel provokes collateral capture with designed decoys to find where it would happen. Observation of the actual versus provocation of the possible.
[n1] Bycatch is the non-target catch taken incidentally alongside the target; the discipline of measuring it depends entirely on observer or camera coverage, because unobserved effort is the standing route by which a burden stays invisible. ↩