Damage to Common Octopus (Octopus minor) Caught in Pot Fisheries

Abstract

Beyond continual reductions in catch, common octopus frequently suffer damage during pot fishing, which can reduce the quality of the product and consequently, its value. This study evaluated how pots with different mesh sizes affect the integrity of common octopuses captured by commercial fisheries. Experimental fishing was conducted in Taean-gun and Incheon-si using pots with different mesh sizes (16.3, 18.3, and 22.4 mm). Common octopuses were classified as either damaged or undamaged based on whether the number of injured arms exceeded a specific threshold value, and logistic regression was applied to estimate the probability of damage, based on the mesh size and region. Smaller mesh sizes significantly reduced the damage in common octopuses and increased the catch effectiveness; however, the degree of impact differed according to the region. This study provides quantitative estimates of the mesh sizes associated with specific damage probabilities, offering a scientific basis for refining regionally tailored management practices. By reducing the mesh size from the current legal standard of 22 mm to 16 mm, the probability of damage that leads to a decline in commercial value (i.e., threshold = 2) is projected to decrease from 77.8% to 46.5% in Taean-gun, and from 93.4% to 39.3% in Incheon-si.

1. Introduction

The common octopus (Octopus minor) is a valuable species of mollusk that is commercially exploited in South Korea [1,2,3]. However, the annual production of common octopuses has shown a clear decline from more than 10,000 tons in the early 1990s to under 7000 tons in the early 2020s [4]. Beyond reductions in catch, common octopus frequently suffer damage during pot fishing [5], which can reduce the quality of the product and consequently its value [6,7,8].

Over the past 10 years (2014–2023), 54% of the total production has been captured through coastal pot fishing [4]. Common octopus pots are cylindrical in shape with three entrances around the curve [9]. According to current fisheries regulations, the use of pots with a mesh size of 22 mm or less is prohibited to protect juvenile common octopuses by law [10]. However, pot fishermen claim that common octopuses are often caught with damaged arms, which can lower their commercial value, when using pots with legally required mesh sizes (>22 mm). These complaints suggest that the current regulations, which focus on resource conservation, may be inadequate in addressing quality-related concerns. Consequently, fishers have requested permission to use smaller mesh sizes to reduce damage [5,11].

Previous studies in feeding behaviors and bait preferences [12], common octopus behavior when entering pots and their catch efficiency [9,13], performance of biodegradable pots [14], and the fishing capacities of fishing fleets [15,16,17] help to develop an effective, sustainable, and stable management of common octopus. However, these studies did not account for the quality of captures since the damage suffered by common octopuses was not considered in the development of fishing practices.

Maintaining the economic value of the catch to ensure a stable income for fishermen is also an important consideration. In the fishing industry, a catch is considered of high quality when fish exhibit minimal physical injuries. Fishers, retailers, and scientists all agree that damage to the fish, such as marks, scale loss, bruising, and poor exsanguination, significantly impacts quality and can lead to a reduced market value. A decline in catch quality not only reduces potential revenue by limiting the range of marketable uses, but it also shortens the shelf life [6,18,19]. Once damage occurs during the catching process, restoring quality becomes extremely difficult or impossible [7,8]. Therefore, preventing physical injury during the catch process is essential for maintaining high catch quality, enhancing product value, and supporting sustainable fisheries.

This study focused on elucidating the effects of current fishing gear on the integrity of common octopus captured by commercial fishery and identifying other possible designs that could reduce the damage rates, thus maintaining or increasing the optimal capture.

2. Materials and Methods

2.1. Test Fishing Operation and Survey

The experimental fishing operations were conducted in two areas along the western coast of South Korea: the coastal waters off Taean-gun, located in the central region, and Incheon-si in the north (Figure 1). This study was carried out on board a 7.31-ton commercial fishing vessel in Taean-gun and a 7.93-ton vessel in Incheon-si. The experimental pots were retrieved after immersion for 2–7 days, depending on the seasonal weather conditions and fishing practices of each region. The experiment was conducted thrice between 25 April and 24 May 2023, during the major fishing season. The first trial was conducted at the end of April (25 April in Taean-gun and 27 April in Incheon-si). The second trial was conducted in the middle of May (16 May in Taean-gun and May 18 in Incheon-si), and the third trial took place at the end of May (23 May in Taean-gun and 24 May in Incheon-si).

2.2. Test Gear Design and Specification

The common octopus pot was 37 cm in diameter and 12 cm in height. It was constructed with a 9 mm iron frame that was covered with PE Td210 12-ply netting. The pot had three funnel-shaped entrances on its curved sides. Each entrance was made of plastic and had a rectangular inner opening with a 14 cm circumference.

Three experimental pots with different mesh sizes were used to investigate the correlation between the mesh size and damage in common octopuses captured by pots. For each pot type, thirty randomly selected mesh gaps were measured, and the average was calculated. The mean mesh sizes (±standard deviation) were 16.3 ± 0.36 mm, 18.3 ± 0.45 mm, and 22.4 ± 0.36 mm, respectively. In addition, they were marked with different paint colors to distinguish them during the experimental fishing (Figure 2). In the following sections, the mesh sizes are referred to by their mean values only for clarity and brevity.

The experimental setup consisted of the sequential arrangement of three common octopus pots of different mesh sizes, with a total of 60 to 75 pots per set (Figure 3). Four sets of gear were used in each region.

2.3. Measurement of Catch

Before being measured, the catches were separated and classified according to the gear set and the mesh size of the pot. For each individual of common octopuses, the mantle length (ML), total length (TL), total weight (TW), and sex were measured. We also recorded the number of injured arms, where injury was defined as visible physical damage, such as a loss of or severe wounds in each arm.

The common octopus was classified as a male if the tip of the third arm on the right was hectocotylized, significantly shorter than the third arm on the left, and exhibited a rounded tip, owing to the presence of a copulatory apparatus. Conversely, if the third arm on the right was similar in length to the third arm on the left and ended in a pointed tip, the specimen was classified as a female [1]. However, in cases where confirmation was impossible, owing to damage to the arms, the gender was classified as indeterminable. Similarly, injured common octopuses that could not be measured were excluded from the TL measurements.

2.4. Catching Performance Analysis

The catching performance of the pot gear was calculated under the assumption that the common octopus resources were uniformly distributed within the relevant sea area. Catch per unit effort (CPUE), which is commonly used as a measure of fishing efficiency, was defined as the number of common octopuses caught per pot deployed during each fishing operation. This method aligns with standard practices in fisheries science for pot/trap gear [15]. The CPUE values were calculated by dividing the total number of common octopuses caught by the total number of pots hauled. A one-way analysis of variance (ANOVA) was conducted to examine whether there were statistically significant differences in the CPUE values among the three mesh size groups (16.3 mm, 18.3 mm, and 22.4 mm). When significant differences were detected by ANOVA, Tukey’s Honestly Significant Difference (HSD) test was applied for pairwise comparisons to identify the specific mesh size groups that differed significantly from each other [20].

2.5. Classification of Common Octopuses by Degree of Damage

Based on a survey conducted among fishers engaged in common octopus fishing, injuries to one arm were reported to have little to no impact on market value. When two to three arms are injured, the value typically decreases by approximately 20–40%, and injuries to four or more arms significantly reduce marketability, making the common octopus nearly unfit for commercial distribution.

Reflecting these responses, we recorded the visible physical damage to each arm, including amputation (partial or full loss), severe lacerations, or deep puncture wounds that were likely caused by entanglement or escape attempts. These injuries were counted per individual to determine the total number of injured arms. Using this count, eight damage classification thresholds were applied based on the number of injured arms. Common octopuses were then classified as either damaged or undamaged according to the following threshold values (

Table 1. Damage classification criteria for the common octopus based on the number of injured arms.

Threshold Value	Criterion Level	Classification Rule	Market Value Impact
1	Low	Damaged if 1 or more arms are injured.	Little to no impact on market value
2	Moderate	Damaged if 2 or more arms are injured.	~20–40% decrease in value
3		Damaged if 3 or more arms are injured.
4	High	Damaged if 4 or more arms are injured.	Significant loss; nearly unfit for distribution
5		Damaged if 5 or more arms are injured.
6		Damaged if 6 or more arms are injured.
7		Damaged if 7 or more arms are injured.
8		Damaged if all arms are injured.

Note 1: A common octopus with fewer injured arms than the specified threshold was classified as undamaged. Note 2: Threshold values were based on fisher perceptions rather than market price validation.

). Additionally, these thresholds were grouped into three broader criterion levels—Low (threshold 1), Moderate (thresholds 2 and 3), and High (thresholds 4–8). These varying criteria allowed us to assess for damage prevalence under different levels of severity and to examine the sensitivity of damage patterns to the definition of damage.

2.6. Statistical Analyses on Damage to Common Octopus

To investigate the effect of the mesh size on the probability of common octopus damage, a logistic regression model was fitted using the entire dataset [21]. Prior to model fitting, key assumptions of logistic regression were evaluated. Each common octopus captured from individual pots was treated as an independent observation. Although this approach assumes independence among common octopuses, potential intra-pot correlation arising from shared capture conditions cannot be fully ruled out. However, as all pots were deployed under similar spatiotemporal conditions and the number of common octopuses per pot was relatively small, any violation of independence is expected to have a limited impact on the results. Multicollinearity among the predictor variables was assessed using Variance Inflation Factors (VIF) that were calculated from logistic regression models, excluding interaction terms. All VIF values were close to 1, indicating no significant multicollinearity.

Subsequently, let y_i ∈ {0, 1} denote the binary response variable for the i-th individual, where 1 indicates that the common octopus was classified as damaged and 0 indicates undamaged. The conditional probability that an individual exhibits damage, given the explanatory variables X_i, is modeled as follows:

$$P\left(y_i=1|X_i\right)={\displaystyle \frac{1}{1+e^{(-{X_i}^T\mathsf{\beta })}}}$$

where X_i includes covariates, such as the mesh size, test region, and their interaction, and β is the corresponding parameter vector.

The log-likelihood function used for parameter estimation is expressed as follows:

$$l\left(\beta \right)=\sum _{i=1}^n[y_i\mathrm{l}\mathrm{o}\mathrm{g}(P\left(y_i=1\right|X_i))+\left(1-y_i\right)\mathrm{l}\mathrm{o}\mathrm{g}(1-P\left(y_i=1\right|X_i))]$$

where the model parameters were estimated by maximizing the log-likelihood.

The significance of the interaction term between the mesh size and test region was assessed using a likelihood ratio test (LRT) by comparing the full model (with interaction) to a reduced model (without interaction). The LRT statistic was calculated as follows

$$D=-2[l_{reduced}-l_{full}]$$

following a chi-square distribution with the degrees of freedom equal to the difference in the number of parameters between the two models. If the interaction was statistically significant (p < 0.05), this indicated that the effect of the mesh size on the damage probability differed between the regions. In such a case, separate logistic regression models were fitted for each region to evaluate the region-specific effects. If the interaction was not significant, a single combined model excluding the interaction term was used for subsequent analysis.

All statistical analyses were conducted in R version 4.3.2 (R Core Team, 2022).

3. Results

3.1. Catch Composition

During the survey period, 759 common octopuses were captured, and their total weight was 92 kg. Of the captured common octopuses, 249 were male, 209 were female, and 301 were of an undetermined sex owing to arm injuries. Their ML (min–max) averaged 60.6 mm (30–123 mm). The histogram of the ML for trapped common octopuses (bin width = 7.5 mm) showed a unimodal distribution, with most common octopuses concentrated within the 48.75–71.25 mm range. Common octopuses smaller than 33.75 or larger than 93.75 mm were rarely observed. Additionally, their TW averaged 121.2 g (20–340 g) and their TL averaged 418.4 mm (50–1000 mm) (Figure 4).

3.2. Number of Injured Arms of Common Octopus

Among the 759 common octopuses, 583 had one or more injured arms, including 181 common octopuses with all eight arms injured. In contrast, only 176 common octopuses were found to be completely intact (

Table 2. The number of common octopuses captured using common octopus pots with different mesh sizes counted on the basis of the number of injured arms.

Area	Trial	Gear Set	Mesh Size	Number of Common Octopuses by the Number of Injured Arms									Sum of Individuals	Total Weight (g)	No. of Pots
				0	1	2	3	4	5	6	7	8
Taean-gun	1	1	16.3	6	5	1	1	0	0	0	0	0	13	1435	20
			18.3	1	5	2	2	1	0	0	0	0	11	1385	20
			22.4	5	2	2	1	0	0	0	0	0	10	1170	20
		2	16.3	11	2	0	0	0	0	0	1	0	14	1967	20
			18.3	4	5	2	1	1	0	0	0	0	13	1735	20
			22.4	1	1	0	2	2	0	0	0	0	6	750	20
		3	16.3	3	3	1	1	1	0	4	0	1	14	1635	20
			18.3	1	1	1	2	1	2	0	1	0	9	1000	20
			22.4	0	1	1	0	2	0	1	0	1	6	800	20
		4	16.3	8	1	2	0	1	0	0	0	0	12	1105	20
			18.3	8	4	0	0	0	0	0	0	0	12	1365	20
			22.4	1	2	2	0	0	0	0	0	0	5	610	20
	2 1	1	16.3	3	6	0	2	1	2	1	0	2	17	1875	20
			18.3	4	1	2	0	2	0	1	1	2	13	1205	20
			22.4	0	0	0	0	1	1	1	1	2	6	805	20
		2	16.3	1	2	3	4	1	0	1	1	3	16	2015	20
			18.3	1	4	1	0	1	0	1	4	2	14	1420	20
			22.4	0	1	0	2	0	1	4	0	3	11	1290	20
		3	16.3	7	2	2	5	2	1	2	0	0	21	2545	20
			18.3	3	2	1	2	1	2	2	1	1	15	1935	20
			22.4	0	0	0	0	2	1	2	1	2	8	1130	20
		4	16.3	5	4	2	0	0	4	0	0	3	18	1882	17
			18.3	3	1	1	0	1	1	1	0	5	13	1242	17
			22.4	1	0	0	0	0	0	0	1	5	7	930	17
	3 2	1	16.3	2	2	3	0	1	0	1	2	1	12	1355	20
			18.3	1	0	0	1	0	1	1	1	4	9	900	20
			22.4	0	0	1	0	1	1	1	4	3	11	1110	20
		2	16.3	8	1	1	0	0	1	1	0	0	12	1320	20
			18.3	4	3	2	1	3	0	0	0	0	13	1660	20
			22.4	0	0	1	1	0	2	0	1	0	5	482	20
		3	16.3	0	1	1	1	5	2	0	0	5	15	1830	20
			18.3	2	0	0	1	1	1	1	3	6	15	1830	20
			22.4	0	1	0	0	1	0	0	0	5	7	555	20
Incheon-si	1	1	16.3	4	0	0	0	0	0	1	0	0	5	735	25
			18.3	1	1	1	0	0	2	1	1	0	7	970	25
			22.4	1	1	1	0	0	0	0	0	4	7	845	25
		2	16.3	3	1	0	0	0	0	0	0	0	4	735	25
			18.3	4	1	0	0	1	1	1	0	0	8	1035	25
			22.4	0	0	0	1	2	0	1	0	4	8	670	25
		3	16.3	3	1	0	0	0	0	0	0	0	4	620	25
			18.3	1	2	0	0	1	2	1	0	0	7	785	25
			22.4	0	0	0	0	0	1	0	0	4	5	355	25
		4	16.3	3	1	0	0	0	0	0	0	0	4	625	25
			18.3	2	0	1	0	0	1	0	2	2	8	1300	25
			22.4	0	0	0	0	0	0	0	0	7	7	435	25
	2	1	16.3	7	1	1	1	0	1	0	0	3	14	2280	25
			18.3	5	1	1	0	3	1	1	1	1	14	1715	25
			22.4	0	0	0	0	0	1	0	1	10	12	949	25
		2	16.3	4	2	0	1	0	1	0	2	1	11	1675	25
			18.3	2	2	1	1	0	3	0	0	3	12	1660	25
			22.4	0	0	0	0	0	1	1	0	9	11	980	25
		3	16.3	7	3	1	1	1	0	2	0	1	16	2620	25
			18.3	3	1	0	2	2	0	0	1	3	12	1670	25
			22.4	0	0	0	0	0	0	0	0	9	9	625	25
		4	16.3	7	0	1	1	0	4	2	0	0	15	2475	25
			18.3	2	2	1	0	0	3	0	0	5	13	1540	25
			22.4	0	0	0	0	0	1	0	0	7	8	605	25
	3	1	16.3	3	0	1	1	0	0	3	0	1	9	1740	25
			18.3	1	2	1	1	1	2	2	0	4	14	2055	25
			22.4	1	0	0	0	0	0	0	0	11	12	1125	25
		2	16.3	0	2	0	1	0	3	0	1	1	8	1275	25
			18.3	4	3	0	0	0	1	1	3	0	12	1685	25
			22.4	0	1	0	0	0	1	1	1	9	13	1220	25
		3	16.3	3	1	2	1	2	1	1	1	0	12	2055	25
			18.3	5	2	3	1	0	1	2	4	0	18	2520	25
			22.4	0	0	0	0	0	0	1	3	12	16	1360	25
		4	16.3	2	1	3	0	0	2	2	1	0	11	1775	25
			18.3	4	2	5	0	1	1	0	1	1	15	2025	25
			22.4	0	0	0	0	0	0	0	2	13	15	995	25
Total				176	97	59	43	47	58	50	48	181	759	92,007	-

¹ During this test session in Taean-gun, some pots of the fourth gear set were lost. ² During this test session in Taean-gun, the fourth gear set was lost.

).

3.3. Catching Performance

Based on the number of common octopuses, the means of the CPUEs for the mesh sizes of 16.3, 18.3, and 22.4 mm were 0.56, 0.55, and 0.39 individuals/pot, respectively (Figure 5). To evaluate the differences in fishing intensity based on the changes in the mesh size of common octopus pots, the means of the CPUEs were compared using ANOVA, revealing a statistically significant difference in the average catch according to the mesh size (p < 0.05) (

Table 3. ANOVA results for CPUE.

Source	Df	Sum Sq	Mean Sq	F Value	Pr (>F)
Mesh size	2	0.3826	0.19310	5.516	0.0061
Residuals	66	2.3104	0.03501

). Tukey’s HSD test showed no significant difference in catch between pots with mesh sizes of 16.3 and 18.3 mm (p >0.05). However, pots with mesh sizes of 16.3 or 18.3 mm had significantly higher catches than pots with a mesh size of 22.4 mm (p < 0.05) (

Table 4. Tukey’s HSD test results for CPUE.

Comparison	Difference	95% CI Lower	95% CI Upper	Adjusted p-Value
18.3 mm–16.3 mm	−0.0137	−0.1459	0.1186	0.9668
22.4 mm–16.3 mm	−0.1651	−0.2974	−0.0328	0.0107
22.4 mm–18.3 mm	−0.1514	−0.2837	−0.0191	0.0210

).

3.4. Analysis of Damage to Common Octopus

To evaluate whether the effect of the mesh size on common octopus damage differed between fishing locations, we compared logistic regression models with and without an interaction term between the mesh size and location. The likelihood ratio tests revealed that the interaction term significantly improved model fit across all thresholds, indicating that the effect of the mesh size on damage probability differed between Taean-gun and Incheon-si (e.g., threshold = 1: ∆Deviance = 7.59, p = 0.0059; threshold = 8: ∆Deviance = 40.88, p-value < 0.0001) (

Table 5. Likelihood ratio test for interaction effect between mesh size and fishing location.

Model Comparison	Damage Threshold	∆Df	∆Deviance	p-Value
Reduced model vs. Full model	1	1	7.59	0.0059
	2	1	11.36	0.0008
	3	1	16.26	<0.0001
	4	1	19.43	<0.0001
	5	1	20.95	<0.0001
	6	1	25.44	<0.0001
	7	1	34.91	<0.0001
	8	1	40.88	<0.0001

). These findings confirm that the impact of the mesh size on damage probability varies according to the region, thereby justifying the use of separate logistic regression models for each location in subsequent analyses.

Given this significant interaction, logistic regression models were fitted separately according to the fishing region (Taean-gun and Incheon-si) and damage thresholds (from 1 to 8) (Figure 6 and

Table 6. Estimated parameters for the relationship between mesh size and damage to common octopus across fishing locations and threshold values.

Location	Threshold	Logistic Parameters		Odds Ratio	OR 95% Confidence Interval (Lower–Upper)	p-Value	AIC	Pseudo R2
		β0	β1
Taean-gun	1	−3.3168	0.2468	1.28	1.13–1.45	<0.001	413.2	0.041
	2	−3.8582	0.2324	1.26	1.14–1.39	<0.001	499.4	0.045
	3	−4.0777	0.2228	1.25	1.14–1.37	<0.001	511.5	0.044
	4	−4.5336	0.2293	1.26	1.15–1.38	<0.001	499.9	0.049
	5	−4.6542	0.2146	1.24	1.13–1.36	<0.001	470.6	0.044
	6	−5.1363	0.2242	1.25	1.14–1.38	<0.001	432.3	0.048
	7	−5.4433	0.2195	1.25	1.12–1.38	<0.001	376.2	0.045
	8	−5.4276	0.1959	1.22	1.09–1.36	<0.001	311.7	0.035
Incheon-si	1	−8.4423	0.534	1.71	1.44–2.02	<0.001	334.5	0.162
	2	−8.6868	0.5156	1.67	1.46–1.92	<0.001	385.9	0.178
	3	−9.4721	0.5391	1.71	1.51–1.95	<0.001	398.6	0.205
	4	−10.2188	0.5696	1.77	1.55–2.01	<0.001	395.4	0.230
	5	−10.1797	0.5558	1.74	1.54–1.97	<0.001	400.6	0.232
	6	−11.529	0.599	1.82	1.62–2.05	<0.001	377.0	0.282
	7	−13.5393	0.6821	1.98	1.75–2.24	<0.001	331.4	0.353
	8	−15.1366	0.7365	2.09	1.84–2.38	<0.001	290.8	0.400

). In all models, the mesh size was significantly and positively associated with the probability of damage, with an increasing mesh size leading to a higher likelihood of damage. For example, at threshold = 2, the damage probabilities in Taean-gun were predicted to be 46.5%, 58.0%, 68.7%, and 77.8% for mesh sizes of 16 mm, 18 mm, 20 mm, and 22 mm, respectively. The corresponding probabilities in Incheon-si were 39.3%, 64.4%, 83.6%, and 93.4%. This pattern was consistent across all thresholds, indicating a clear and positive relationship between the mesh size and damage risk.

Looking at regional differences, the magnitude of effect, expressed as odds ratios (OR), was consistently higher in Incheon-si than in Taean-gun. For instance, at threshold = 2, the OR for the mesh size in Incheon-si was 1.67, indicating a 67% increase in odds of damage per 1 mm increase in mesh size, compared to 1.26 in Taean-gun. This trend was also observed across all thresholds, indicating a stronger association between the mesh size and damage probability in Incheon-si than in Taean-gun.

4. Discussion

This study demonstrates that the mesh size has a significant and positive relationship with the probability of damage in common octopuses (Octopus minor), and that this relationship varies by fishing region. Logistic regression models showed that, across all thresholds of injury severity (threshold = 1–8), larger mesh sizes consistently increased the likelihood of damage. Moreover, the magnitude of this effect was notably higher in Incheon-si compared to Taean-gun, as reflected by consistently higher odds ratios. These findings strongly support the need for region-specific management strategies in common octopus pot fisheries. For instance, at threshold = 2, which represents a level of injury that may impact the commercial value of the catch, a reduction in mesh size from 22 mm to 16 mm is predicted to decrease the damage probability from 77.8% to 46.5% in Taean-gun and from 93.4% to 39.3% in Incheon-si.

Reducing the mesh size appears to be an effective strategy for minimizing damage to the common octopus. However, this benefit comes with a trade-off: smaller mesh sizes may increase the CPUE and the retention of juvenile common octopuses [13,22], potentially undermining efforts for stock sustainability. Moreover, prolonged exposure of small common octopuses to confined spaces may alter their natural behavior and stress physiology, potentially reducing their chances of survival even if they are returned to the sea. From a management perspective, these results underscore the need for a multifaceted approach that balances conservation goals with fishery productivity.

In the present study, the CPUEs of common octopus pots, measured as the number of common octopuses per pot, varied significantly according to the mesh size. Pots with mesh sizes of 16.3 mm and 18.3 mm yielded similar catch rates, averaging 0.56 and 0.55 individuals per pot, respectively, while pots with a 22.4 mm mesh size showed a notably lower CPUE of 0.39 individuals per pot. These findings are consistent with previous research by Kwon and Kim [13], who reported that the CPUE, based on the total catch weight, decreased as the mesh size increased in pots deployed in Deukyang Bay. Specifically, CPUE values of 59 g, 44 g, and 30 g per pot were recorded for mesh sizes of 18 mm, 20 mm, and 22 mm, respectively. Although their study evaluated the CPUE in terms of biomass rather than individual count, the overall trend aligns with our results: larger mesh sizes are associated with lower catch efficiency. This inverse relationship between mesh size and catch rate likely reflects the reduced probability of retaining smaller common octopuses in pots with wider openings. Kim et al. [22] also support the observed pattern that octopus catches decreased as the mesh size increased, with a general tendency for larger common octopuses to be caught.

From catch composition, the mantle size distribution of captured common octopuses was skewed toward a mid-size range, with a sharp decline in the number of common octopuses exceeding 71.25 mm mantle length. This pattern is likely influenced by the physical constraints imposed by the pot entrance, which was constructed from rigid plastic with a fixed circumference of 14 cm and a funnel-like shape.

Interestingly, the reduced capture of smaller common octopuses may also be attributed to the entrance functioning as an escape opening for juvenile common octopuses. Although the entrance in this study was a funnel-shaped rectangular opening designed for entry, its size and structure may have allowed smaller common octopuses to exit the pot after entry. According to Nagano et al. [23], small-sized Enteroctopus dofleini were able to escape through openings in basket traps. Despite the structural differences between the circular escape rings in their study and the rectangular funnel entrance used here, it is plausible that the entrance may have served a similar role in enabling the escape of small Octopus minor individuals. Therefore, the observed decline in the smallest size classes may reflect post-entry escape behavior through the entrance funnel.

On the other hand, the reduced number of large common octopuses likely results from size-based selectivity, as larger common octopuses may be physically unable to enter the pot owing to the fixed entrance dimensions. This interpretation aligns with Petetta et al. [24], who demonstrated that traps with larger entrance surfaces captured significantly larger individuals of Octopus minor, as indicated by a greater Catch-based Index. Their findings emphasize that the entrance size plays a critical role in determining the size composition. Accordingly, the fixed rectangular opening used in the present study may have acted as a physical barrier, limiting the entry of larger common octopuses and concentrating the catch within a specific size range.

In this regard, the funnel-shaped plastic entrance (opening) on its curved side of the pot deserves closer attention. While originally designed to facilitate entry into the pot, it may also function similarly to an escape gap. Escape gaps play a key role in determining gear selectivity by regulating the size composition of retained individuals. Previous studies have shown that incorporating escape openings in pot designs can reduce the capture of undersized or non-target individuals while maintaining or even improving the retention of commercially valuable sizes [25,26].

Given these parallels, adjusting the size and shape of the plastic entrance could serve as a complementary strategy to mesh size reduction. Specifically, while smaller mesh sizes can be implemented to reduce arm injury, the plastic entrance could be optimized to allow for the escape of undersized common octopuses, thereby protecting juvenile common octopuses. Both Nagano et al. [23] and Kim [27] investigated circular (ring-shaped) escape openings in octopus pots. Their results showed that appropriately sized ring openings effectively reduced the capture of small Octopus dofleini, improving size selectivity and reducing juvenile bycatch without significantly affecting the catch of market-sized individuals. Similarly, Berzosa et al. [28] demonstrated that funnel-shaped entrances with appropriate dimensions (e.g., inclination angle, length, and opening width) can improve the entry rate and size selectivity of fishes. Incorporating such features could enhance selective retention and reduce the bycatch of juveniles. A combined approach—reducing the mesh size to minimize injury, and modifying the entrance to maintain size selectivity—could enhance both the increase in its productivity and resource sustainability.

Such integrated gear modifications should be tested experimentally to evaluate their effectiveness under commercial fishing conditions. Future studies should also explore the behavioral responses of Octopus minor to varying entrance dimensions and their potential interactions with the mesh size so as to guide the development of best practices for gear design in this fishery. This dual approach could optimize gear design to both reduce physical injury and enhance conservation of juvenile common octopuses, aligning with sustainable fisheries management.

5. Conclusions

This study assessed the effects of pot mesh size on the integrity of common octopuses (Octopus minor) captured in commercial fisheries. The results indicate that smaller mesh sizes reduce the likelihood of arm injuries while enhancing catch efficiency. However, these effects were found to vary according to the location, underscoring the importance of region-specific management strategies.

Based on logistic regression models, reducing the mesh size from the current legal standard of 22 mm to 16 mm, the probability of damage that leads to a decline in commercial value (i.e., threshold = 2) is projected to decrease from 77.8% to 46.5% in Taean-gun, and from 93.4% to 39.3% in Incheon-si.

These findings suggest that adjusting the mesh size could lead to both improved product quality and increased economic returns for fishers. However, the potential benefits of such changes must be carefully balanced with ecological considerations, such as the protection of juvenile stocks and the possible rise in fishing intensity. A balanced approach that combines injury reduction, juvenile protection, and appropriate fishing intensity can contribute to more sustainable and efficient octopus pot fisheries.

We believe that the insights gained from this study can contribute to the development of more adaptive, evidence-based policies for the sustainable management of Octopus minor in South Korea.