How Many of the Fish We Catch Are Still Juveniles? 8,790 Logged Lengths Against Length at Maturity

When anglers log a length they are usually thinking "how big". A fish biologist reads the same number and asks something else: has this fish had a chance to spawn? Every species has a length at maturity (L50), the size at which half the population is sexually mature. Setting 8,790 logged lengths against it answers a question recreational fisheries keep asking and rarely have large samples for: how many of the fish we catch, and keep, are still juveniles?

Finding: Two-thirds of marine native catches are below length at maturity, over 90% for mangrove red snapper, bigeye trevally and Malabar grouper. Anglers do decide by size, but only for natives; for invasive species release is size-blind. About half of kept fish are immature, and release shows no jump at L50: anglers judge 'big enough', not 'mature'.

Analysis date 2026-09-05 · Data 2025-01-01 to 2026-09-04 · 中文版

• H1 Juvenile share: most (>50%) logged marine native catches are below length at maturity. → supported: 66% (59%–71%)
• H2 Size-selective release: release probability falls with length for marine natives but not for freshwater invasives (negative control). → supported: natives odds ratio 3.713 per halving (p < 0.001), invasives 1.215; pairs 72% vs 52%
• H3 Kept juveniles: most kept marine natives are below maturity. → not established: 53%, interval 44%–60% spans one half
• H4 No discontinuity at L50: with no legal minimum size, release should not jump at L50. → consistent: jump -0.187 (SE 0.215, p 0.386)

Malabar grouper, Bigeye trevally, Mangrove red snapper: over 90% of catches are below maturity, and so are the kept fish. Mangrove red snapper matures at about 57 cm (female); our median logged length is 27 cm. That matches monitoring in Queensland, Australia almost exactly, where over 90% of the 2017 recreational catch was below the female L50. Taiwanese anglers are not unusually fond of small fish; the juveniles of these species live in estuaries and harbours while adults are on offshore reefs. Malabar grouper is more extreme still at 98%, with maturity at 64 cm and a median of 25 cm.

At the other end is orange-spotted spinefoot: 9% below L50 and a release rate of 12%, almost all adults and almost all eaten. The seabreams and grey mullet sit in between at forty to sixty percent. The three freshwater invasives below the rule are the control: they mature small (14–19 cm), nearly every catch is mature, yet release rates all exceed 87%.

Marine native release falls from 78% in the smallest bin to 24% in the largest; the invasives are a flat line from 86% to 93%. Three methods agree. Species-fixed-effect logit: halving the length multiplies the odds of release by 3.713 for natives (p < 0.001; 6.436 on the 1,047 ruler-flagged records) and by 1.215 for invasives (p 0.036). Within-angler, within-species pairs: in 117 native pairs the kept fish was larger 72% of the time (median difference 4 cm, p < 0.001); in 175 invasive pairs only 52% (p 0.630).

This control answers a question left open by our earlier release study, which found that 87% release is a method culture rather than a species judgement. This study adds: within a species, anglers do judge, and what they judge is size. That release of invasives is size-blind means releasing an invasive is a reflex, not a decision about that fish.

But the judgement is not about maturity. In a piecewise-linear fit within ±0.6 log units of L50, the jump at L50 is -0.187 (SE 0.215, p 0.386) with slopes -1.509 and -2.067 on either side: a smooth line. None of these species has a legal minimum size in Taiwan, so anglers have no reason to know L50; they use "worth eating".

The last column is size selectivity at its plainest: Mangrove red snapper 54% vs 31%, Giant trevally 56% vs 30%, Great barracuda 53% vs 17%, Blackhead seabream (black porgy) 59% vs 16%, while the three invasives are nearly equal on both sides.

With the Queensland wild female (51.2) or male (44.9) L50, 94% and 86% of mangrove red snapper are still below threshold. Converting black seabream's 29.1 cm standard length to about 33 cm total length raises its immature share from 60% to 75%; that is the correct total-to-total comparison, and it shows the main table understates immature shares for SL-referenced species. Grey mullet against FishBase's Taiwan row of 27.8 cm drops to 25%; mullet has the widest regional spread of L50 (27 to 45 cm) and its figure should be read as a range.

58% of lengths are multiples of 5 cm and 33% multiples of 10, against about 20% and 10% at random. Manual entries: 62%; records flagged as ruler-measured: 47%, so even measured fish are being rounded. Jittering every multiple of 5 uniformly within ±2.5 cm, 200 times, moves the marine native immature share from 66% to 66% (66%–67%). Heaping does not change the conclusions, but it shows most lengths are eyeballed; the size-selectivity result is stronger on the ruler subset (6.436×), so it is not a measurement artefact.

Measured

• Marine natives: 66% (59%–71%) of catches below L50; kept fish 53% (44%–60%).
• Odds of release per halving of length: natives 3.713, invasives 1.215; pairs 72% vs 52%.
• Jump at L50 -0.187 (p 0.386).
• Multiples of 5: 58%; jitter changes shares by under one point.
• Power: bootstrap half-width for the pooled kept share 8%; a true log-slope of −0.5 (odds ratio about 1.4) would be detected 100% of the time.

Inferred, to be read separately

• Most maturity lengths are not Taiwan-population estimates; L50 for one species can differ twofold across regions (mullet). Read shares by species, not to the unit.
• App lengths are treated as total length; for SL/FL references the immature share is understated. Black seabream shows the size of that gap (60% → 75%).
• For protandrous species (black seabream, yellowfin seabream, barramundi), fish "below L50" are mostly mature males; the better reading is "keeping the big ones removes females".
• Malabar grouper and barramundi include aquaculture escapees or stocked fish, for which wild maturity lengths may not apply.
• Lengths are logged by choice (about three in ten catches carry one) and skew toward memorable fish, which may understate the immature share.
• "Releasing invasives is a reflex" is our reading of the control; the data show only that release is unrelated to size.

That recreational catches are dominated by juveniles is documented where legal sizes exist and are monitored (Queensland mangrove jack, over 90%). For Taiwan's recreational fishery, with no size rules and no effort survey, this is the first large-sample measurement from self-reported lengths. Size-selective release has mostly been studied where minimum sizes fix the angler's threshold; what this study adds is a measurement of the threshold anglers set for themselves when no rule exists: smooth, edibility-based, and blind to maturity.

• Froese, R. & Pauly, D. (eds.). FishBase, Reproduction › Maturity (length at first maturity, Lm) tables per species; e.g. Lutjanus argentimaculatus (ID 1407), Acanthopagrus schlegelii (6531), Mugil cephalus (785), Epinephelus malabaricus (6439). Retrieved 2026-09-05. sourceSource of the primary maturity lengths for all 14 species, each with length type, sex and locality.
• Law, C.S.W. & Sadovy de Mitcheson, Y. (2017). Reproductive biology of black seabream Acanthopagrus schlegelii, threadfin porgy Evynnis cardinalis and red pargo Pagrus major in the northern South China Sea with consideration of fishery status and management needs. Journal of Fish Biology 91(1): 101–125. sourceBlack seabream is protandrous; standard length at 50% sex change is 291 mm, so fish below it are almost all male.
• Status of Australian Fish Stocks Report: Mangrove Jack (Lutjanus argentimaculatus), 2020. Fisheries Research and Development Corporation. sourceQueensland wild L50 of 51.2 cm (female) and 44.9 cm (male) LCF; local monitoring found >90% of the 2017 recreational catch below the female L50, matching the 93–95% here.
• Heat, hurricanes, and health: Effects of natural disturbances on angling effort. PLOS ONE (2023). source(Method context) a recent example of angler-behaviour research from self-reported effort data.

1. Sample. Catches from 2025-01-01 to 2026-09-04 with a length of 1–200 cm and a species in the maturity table: 8,790 records, 14 species (11 marine natives, 3,676 records; 3 freshwater invasives, 5,114), 3,003 anglers; 2,663 flagged as ruler-measured. Species threshold: at least 100 records and a FishBase Lm.
2. Length at maturity. Primary Lm from each species' FishBase Maturity table (with length type, sex, locality); black seabream from the 50% sex-change SL in Law & Sadovy 2017. Alternatives in the sensitivity table.
3. Shares and intervals. Share below L50 (all and kept), 1,000 angler-level bootstrap replicates for 95% intervals.
4. Size selectivity. (a) Within angler and species where both kept and released exist, median kept length minus median released length, sign test; (b) logit of release on log(length/L50) with species fixed effects and angler-clustered SE, for marine natives, invasives and the ruler subset; (c) piecewise-linear fit within |log(length/L50)| < 0.6 testing a jump at L50.
5. Heaping. Shares at multiples of 5 and 10; multiples of 5 jittered uniformly within ±2.5 cm, 200 times.
6. Power. Bootstrap interval widths; simulated detection rate for a log-slope of −0.5.

Read-only analysis; no production data modified. Per-species length distributions (5 cm bins) are available to researchers on request.

The aggregates, figures and methods here are free to read and cite; please retain the data period, sample size and analysis date. Public pages are free to read and cite page by page. Raw records are not available for download, and bulk reuse licensing for aggregated data is still under review.

We are looking for collaborators. What is missing most is length at maturity for Taiwan populations. If you hold reproductive-biology data (L50, sex-change length) for mangrove red snapper, black seabream, yellowfin seabream or grey mullet in Taiwanese waters, we will recompute every share with your values and co-author the update. Write to support@dudefishing.net.