Weighing for Fisheries Science and Marine Research at Sea

Why weight data matter in fisheries science

Weight is one of the core measurements behind fish stock assessment: survey catch weights, individual fish weights and organ weights are converted into abundance indices, weight-at-age, maturity and condition, which in turn inform catch advice. A research vessel is therefore a floating measurement laboratory, and the scales on board are measuring instruments whose errors travel all the way into scientific advice.

In Europe, the collection of biological data on commercial stocks and the running of research surveys at sea are organised under the EU data collection framework, Regulation (EU) 2017/1004. Many of these surveys are coordinated internationally through the International Council for the Exploration of the Sea (ICES), which publishes survey manuals so that vessels from different countries sample in the same way. The North Sea International Bottom Trawl Survey (IBTS), for example, has been carried out in the first quarter of the year since the 1960s, and its current ICES survey manual specifies how catches are sorted, measured and sampled.

Weights enter the data at several levels:

• Catch weights per species and category, used to calculate catch rates and to raise subsamples to the whole catch.
• Subsample weights, which define the raising factor between the fish that were measured and the fish that were caught.
• Individual fish weights, which feed length–weight relationships, weight-at-age and condition indices.
• Organ and tissue weights, such as gonads, livers and stomach contents, used for maturity, energy reserves and diet studies.

The ICES DATRAS database, which holds trawl survey data from many nations, stores catch weights, subsample weights and individual weights in grams (DATRAS units). A value recorded to the nearest gram is only as good as the scale and the procedure that produced it.

What gets weighed on a research vessel

A research vessel weighs objects that span five orders of magnitude or more, from stomach contents of a few grams to catches of several hundred kilograms, and no single scale covers that range with useful resolution. The table below groups the typical weighing tasks on a fisheries survey and indicates which knowledge article covers each one in depth.

Weighing task	Typical mass range	What the value is used for	Typical scale type	Read more
Bulk catch or large species categories	10 kg to several hundred kg	Catch rates, raising of subsamples	Bench scale with large platform or hopper; deck platform scale	Biological sampling workflow
Baskets and subsamples	0.5–30 kg	Raising factors for numbers-at-length	Motion-compensated bench scale	Biological sampling workflow
Individual fish	1 g to over 10 kg	Length–weight relationships, weight-at-age, condition	Scientific marine scale at the measuring station	Length–weight relationships
Gonads, livers, stomach contents	Below 1 g to over 1 kg	Maturity, gonadosomatic and hepatosomatic indices, diet	High-resolution scientific marine scale with wind cover	Precision weighing of small samples
Otoliths	Milligram range	Age estimation support, otolith mass studies	Analytical balance in a laboratory ashore	Precision weighing of small samples
Benthos and marine litter	Grams to kilograms	Ecosystem and litter monitoring	Bench scale	This page

The survey weighing workflow in brief

On a bottom trawl survey, the catch is sorted by species, each species or size category is weighed, a representative subsample is measured for length, and selected fish are sampled individually for weight, sex, maturity and otoliths. The IBTS manual states that the catch from all valid hauls must be fully sorted where practically possible, and that very abundant species or size categories can be subsampled.

The subsample step is where weighing matters most. Numbers-at-length measured in a subsample are multiplied by the ratio of category weight to subsample weight. If either weight is wrong, every length class in that category is scaled by the same error. The manual also gives practical guidance on subsample size: at least 75 fish when more than 75 are caught, increasing to 150 fish for catches above 1,000 individuals, and separate size categories when small and large fish are clearly separated.

Individual fish sampled for age are also sampled for sex, maturity and weight. Electronic data capture systems on survey vessels can read these weights directly from marine scales: the Cefas research vessel data capture system described by Silva et al. (2013) records catch element weights from linked marine scales and prompts the operator to take otoliths when a sampling target is not yet met.

The full step-by-step workflow, including raising factors, tare handling and common error sources, is described in Biological Sampling on Research Vessels: The Weighing Workflow.

Length–weight relationships and condition

A length–weight relationship predicts the weight of a fish from its length, and its parameters are estimated from individual fish that were both measured and weighed. Because length is quicker to measure than weight, surveys and commercial sampling programmes often record many lengths and fewer weights, then use the relationship to estimate biomass from length frequencies.

Weight rises roughly with the cube of length, because a fish grows in height and width as well as in length: a fish 10% longer is about a third heavier, and a fish twice as long is about eight times as heavy. In a meta-analysis of 3,929 relationships for 1,773 species, Froese (2006) confirmed that the fitted exponent is expected to lie between 2.5 and 3.5, with a median of 3.03. Values outside this range deserve scrutiny for problems such as a narrow length range, mixed length types or weighing errors.

Weighing resolution has a visible effect on the quality of these relationships. Silva et al. (2013) compared two sprat datasets from different surveys and years: one recorded to 0.5 cm and 1 g gave a goodness of fit (r²) of 0.70, while another recorded to 0.1 cm and 0.1 g gave 0.865, and the authors recommended that surveys review the resolution used for small species. How the relationship is fitted, a worked example with published cod parameters and the use of condition factors are covered in Length–Weight Relationships in Fish: What They Tell Scientists.

Precision weighing of small samples

Small biological samples are the hardest weighing task at sea, because the error sources on a moving vessel stay roughly the same in absolute terms while the sample mass shrinks. A 0.2 g error is negligible on a 2 kg cod but amounts to 10 % of a 2 g gonad.

A practical rule is to choose a readability no larger than about 1 % of the lightest sample you intend to record. This is our working guidance rather than a formal standard, but it links the choice of scale directly to the data quality you need. Following it, a scale with 0.1 g readability suits samples from about 10 g upwards; lighter samples are better frozen and weighed ashore.

Three sample types come up again and again:

• Gonads, weighed for the gonadosomatic index (gonad weight as a percentage of body weight), which supports maturity staging.
• Stomach contents, where the ICES stomach sampling manual asks for total prey weight and the weight of each prey group; the minimum sampling level can be analysed on board, while the extended level is frozen and analysed in the laboratory (ICES stomach manual).
• Otoliths, whose mass is in the milligram range and which are normally dried and weighed on laboratory analytical balances ashore rather than at sea.

Readability requirements per sample type, the effect of wind on small platforms, and a step-by-step protocol are set out in Precision Weighing of Small Samples at Sea.

Motion, wind and vibration: the physics of weighing at sea

A scale measures force, not mass, so any vertical acceleration of the vessel changes the reading. When the platform accelerates upwards, the sample presses harder on the load cell, the same effect that makes a person feel heavier in an accelerating lift (apparent weight). An upward acceleration of 0.5 m/s² therefore raises the reading by about 5 %, and a downward acceleration lowers it by the same amount.

On a vessel, these accelerations come from heave, pitch and roll, and from vibration transmitted by engines, winches and pumps. Marine scales compensate for vessel motion so that a stable, usable value is shown on a moving deck, but no compensation removes every effect instantly. There is always a trade-off between how quickly a stable value is shown and how much noise remains, which is why performance should be judged by testing at realistic accelerations and tilt.

Two further effects matter for scientific work:

• Air movement. Moving air exerts a pressure that grows with the square of wind speed (NASA); at 1 m/s it is about 0.6 Pa. Acting on a 120 × 120 mm platform, that pressure corresponds to a force equivalent to roughly 0.9 g if it acted fully in the vertical direction, and a draught twice as fast gives about four times as much. Only part of it does, but on a scale reading to 0.1 g, draughts from doors, fans and deck wind are clearly relevant.
• Tilt and gravity. A tilted platform changes the force component on the load cells, and the local value of gravity varies with latitude by roughly 0.5 % between the equator and the poles. Both matter most for large platform scales and for vessels working over a wide latitude range.

The engineering background of motion compensation and marine scale design is covered in the Weighing at Sea hub.

Choosing and integrating scientific scales

Choose a research scale by starting from the samples: list every weighing task, its mass range and the resolution the data need, and then select capacities, platforms and interfaces to match. A typical groundfish survey needs at least three weighing points: a sorting station for baskets and categories, a measuring station for individual fish, and a wet-lab position for small samples.

Beyond capacity and readability, the requirements that most often decide whether a scale works at sea are motion compensation performance in realistic conditions, stabilisation time, platform size relative to the fish, water resistance and cleanability, and direct data capture. Manual transcription of weights is slow and introduces errors, so most survey institutes connect scales to their data capture software through serial, USB, Ethernet or wireless links.

A full requirements checklist, including data capture and integration with survey software, is given in Choosing a Scientific Marine Scale for a Research Vessel. Integration patterns for weighing data in general are covered in the Data Integration hub.

Quality assurance: calibration, checks and metadata

Scientific weight data are credible when three things can be shown: the scale was calibrated traceably, its performance was checked during the survey, and each value can be linked to the scale and conditions that produced it.

1. Calibrate before the survey. Calibration of non-automatic weighing instruments is described in EURAMET Calibration Guide No. 18, which covers test loads, eccentricity, repeatability and measurement uncertainty.
2. Check at sea. Weigh a test weight of a mass similar to the samples at the start of each watch and after moving or cleaning a scale, and record the result.
3. Record metadata. Store scale identifier, readability, date and time, and a note of sea state or unusual conditions together with the weight values.
4. Screen the data. Plot weight against length for each species after each station; outliers are often the first sign of a scale, tare or transcription problem.

Scientific survey weighing is usually not a commercial transaction, so the legal metrology rules that apply to trade weighing are generally not the driver here. Where weights from research or observer programmes are used for control purposes, different rules can apply; see the Legal Metrology hub.

Beyond fish: benthos, litter and other samples

Research vessels also weigh non-fish material, and the same principles of resolution, motion and data capture apply. Two examples from the IBTS manual:

• Benthic invertebrates caught in the trawl may be recorded as presence/absence or as weights and numbers, at the discretion of the national institute. Published length–weight relationships for benthic invertebrates, such as those for 216 North Sea species by Robinson et al. (2010), allow biomass to be estimated from size measurements.
• Marine litter collected in the trawl has been recorded on North Sea IBTS surveys since 2011; items within a category are counted and weighed individually rather than grouped.

Multidisciplinary and ecosystem surveys add further tasks, such as weighing sediment or plankton samples, which usually call for the same small-sample precautions as stomach contents.

How WPL approaches research weighing

WPL Industries designs motion-compensated marine scales and has delivered them to fishing and research fleets across Europe, the Americas and beyond. For scientific work, the M3 Series Scientific Marine Scale is built in 316L stainless steel on a 120 × 120 mm or 270 × 270 mm platform, with capacities from 300 g (readability 0.1–0.2 g) to 6,000 g and an optional wind cover for the smaller platform. Baskets and catch categories are handled by the M2 Series Marine Scale, and the embedded WeightControl software logs every weighing and exports data as CSV or JSON through its API.

Articles in this hub

• Biological Sampling on Research Vessels: The Weighing Workflow – catch sorting, subsampling, raising factors and individual fish sampling.
• Length–Weight Relationships in Fish: What They Tell Scientists – the cube law in plain terms, fitting, worked example and condition factor.
• Precision Weighing of Small Samples at Sea: Otoliths, Gonads and Stomach Contents – readability, wind covers and motion.
• Choosing a Scientific Marine Scale for a Research Vessel – requirements checklist and data integration.
• Weighing Equipment for Fisheries Research Vessels – a practical station-by-station equipment overview.