Fishing time and trap ghost fishing for Cancer johngarthi along the Baja California peninsula's southwestern coast, Mexico.
Article Type: Abstract
Subject: Crabs (Distribution)
Crabs (Environmental aspects)
Authors: Ramirez-Rodriguez, Mauricio
Arreguin-Sanchez, Francisco
Pub Date: 12/01/2008
Publication: Name: Journal of Shellfish Research Publisher: National Shellfisheries Association, Inc. Audience: Academic Format: Magazine/Journal Subject: Biological sciences; Zoology and wildlife conservation Copyright: COPYRIGHT 2008 National Shellfisheries Association, Inc. ISSN: 0730-8000
Issue: Date: Dec, 2008 Source Volume: 27 Source Issue: 5
Topic: Event Code: 690 Goods & services distribution Advertising Code: 59 Channels of Distribution Computer Subject: Company distribution practices
Product: Product Code: 0913040 Crabs NAICS Code: 114112 Shellfish Fishing SIC Code: 0913 Shellfish
Geographic: Geographic Scope: Mexico Geographic Code: 1MEX Mexico
Accession Number: 191646316
Full Text: ABSTRACT During trap-fishing investigations on the crab Cancer johngarthi along Baja California peninsula's southwestern coast, Baja California Sur, Mexico, conducted between 2002 and 2006, information was gathered to assess fishing efficiency in terms of the number of crabs caught per trap during one hour of operation (catch per unit of effort, CPUE = c/ht). As a result of vessel operation issues, some trap lines were abandoned, whereas vessels returned to land for repairing, and because the effective fishing time for these traps was significantly longer than normal, the information so obtained was regarded as reflecting the potential effect on fishing of traps lost during fishing operations (ghost fishing). Of 651 line sets analyzed, involving 45,152 traps, 77% had effective fishing times below 60 h, 14% between 70 and 150 h and 9% between 150 and 5,500 h. There is an exponential reduction in CPUE with increasing fishing time that could be associated to trap saturation. The number of crabs per trap also decreased with immersion rime, fitting a power function. After 1,000 h of immersion, each trap may contain 7-18 crabs. Four hundred traps were lost over the course of 14 fishing trips. However, because the fishery is in the early development phases, the implementation of measures to avoid or restrain ghost fishing is deemed convenient.

KEY WORDS: ghost fishing, traps, Cancer, crab, Mexico


The effects of fishing methods on exploited populations are related to their design and mode of operation, which are based on the life-stage characteristics of both the target species and the environment in which it grows, and aim at selectively catching the largest amount of specimens of the target species meeting the size-related characteristics (Krouse 1989). There is a broad variety of fishing techniques (FAO 1964, Nedelec 1975, FAO 1978).

In the case of traps, selectivity will depend on mesh size, and trap fishing is regarded as causing a low environmental impact compared with the effects of trawling (Jennings & Kaiser 1998). However, one of the issues associated with the former is "ghost fishing", or unintentional catch because of the continuous operation of fishing traps/nets lost as a result of bad weather, loss of floats or fouling in the seabed. This issue is related to the capacity of lost fishing gear to catch specimens of either the target species or others related to it, as well as to the amount of fishing gear that gets lost and the time it remains undamaged.

The Code of Conduct for Responsible Fisheries (FAO 1995) acknowledges the impact of lost fishing gear and sets forth that governments should implement measures aimed at minimizing it. Ghost fishing may last many years, depending on the environmental conditions and the durability of fishing-gear materials; however, little is known on this issue, partly because of the difficulty involved in conducting long-term studies (Breen 1990).

In the case of traps, as a result of their longer durability and higher selectivity, ghost fishing may occur continuously, with an unknown duration. Its importance has been acknowledged, given that the number of traps lost in different crab fisheries is not negligible. In general, investigating ghost fishing is challenging (Pawson 2003), and is based on either following up simulated lost traps or analyzing catch data from traps lost and recovered in specific projects (Breen 1990, Stevens et al. 2000, Bullimore et al. 2001, Al-Masroori et al. 2004).

Along the Baja California peninsula's southwestern coast, Mexico, the fishing potential of the crab Cancer johngarthi was assessed in 2002 and 2003 (Ramirez-Rodriguez et al. 2003; Lopez-Rocha et al. 2006; Ramirez-Rodriguez et al. 2007), and from 2004 the Mexican government approved the crab's commercial fishing using traps, limiting the fishing effort to the operation of two vessels (SAGARPA 2004). However, from the very beginning only one vessel has been operating with the participation of two on-board observers, so that information is available on fishing effort and catch data, including traps lost for variable time periods because of vessel operation issues.



For these traps, the effective fishing rime, or soak time (the period that traps remained on the seabed), was significantly longer than the one involved in routinary operations, so the information gathered was deemed likely to reflect the effect of fishing by traps lost during fishing operations (ghost fishing). Based on the earlier mentioned, the aim of this investigation is to determine the effect of variations in effective fishing rime on changes in catch per unit of effort, and estimate the potential effects of ghost fishing on the fishery of the Mexican crab Cancer johngarthi.


Information was gathered for 651 trap-line sets conducted during 14 fishing trips of the only crab vessel operating off the Baja California southwestern coast, mainly between 20[degrees] 50' and 26[degrees]30' North (Fig. 1), as follows: 7 from October 2002 to August 2003, 3 from August to October 2005 and, 4 from May to October 2006. Fifty-seven lines were set in one trip and collected in the next.


Trip duration varied depending on logistic issues, but followed a commercial fishing strategy. The fishing operation started by setting lines of up to 100 traps each, separated from one another by 30-40 m, at depths ranging from 100-400 m (Fig. 2). Traps were conical, 150 cm x 75 cm x 60 cm (base diameter, top diameter, and height respectively), with a metallic frame and a 41-mm mesh-nylon net on the body. The mouth (entrance) was located at the top. A net bag containing 1-2 kg frozen sardine was placed inside each trap, as bait. The decision on fishing sites and effective fishing time or soak time was made by fishing technicians and qualified staff with expertise in crab fishing.

For every fishing operation, including those completed in the same trip and those initiated in a trip and completed in the next, onboard observers gathered data on: date, depth, time, and geographical position at the beginning and end of each fishing operation; number of traps per line and number of specimens caught per trap. Soak time was estimated as the number of hours between the date and rime of trap placement and the date and time of trap collection. The effective fishing effort was estimated as the effective fishing time of one operation multiplied by the number of traps recovered in it (hour-trap = ht). The catch per unit of effort was estimated as the number of specimens caught divided by the effective fishing effort per fishing operation (CPUE = c/ht).

For each fishing operation, the mean CPUE and the mean number of crabs per trap were estimated and correlated to the corresponding soak time. The models to be used were determined based on the best-fit estimate of observed and calculated data through estimates of parameters by nonlinear regression and the use of Marquardt's algorithm (StatSoft 1999).


Overall, 77% of the 651 line sets had effective fishing rimes below 60 h, mostly between 30 and 40 h, 14% between 70-150 h; 6% between 150 and 500 h and 3% between 1,000 and 5,500 h (Fig. 3). As expected, CPUE noticeably decreased with longer soak times, but the decrease in the number of crabs per trap did not show the same relationship (Table 1). The mean crab size did not vary significantly with the effective fishing time.

The relationship between CPUE and soak time was fitted ([R.sup.2] = 0.53) to the negative exponential model.

y = [0.2765e.sup.-0.0012x]

where y is catch per hour-trap and x is soak time. When considering only line sets between 10 and 100 h of immersion, this relationship became clearer (Fig. 4).

As regards the mean number of crabs per trap, an apparent trend was observed towards a decrease, and the power function displayed the best fit (Fig. 5). After 1,000 h of immersion (42 days approx.), each trap may contain 7-18 crabs. When considering the deviations from the mean number of crabs per trap for all line sets (26 crabs per trap), a trend was observed towards a decrease with soak time, with figures above the average in the first 140 h and below it thereafter. The increase in crab number around 100-140 h and 500 h of immersion might reflect the trap's particular conditions in relation to the fishing season.

The information on normal fishing operations reveals a direct relationship between catch and fishing effort, but the data in terms of catch per hour-trap show no seasonal pattern in the abundance of C. johngarthi, which varies widely between years as well as between months (Fig. 6).



The results show that immersion time exerts a strong influence on trap efficiency to catch C. johngarthi, revealing an exponential decrease in the catch per hour-trap (CPUE) with increasing trap operation rime. This effect has been reported for other fisheries, and stresses the importance to avoid leaving traps submerged for periods when CPUE drops below the overall average (about 40 h for C. johngarthi). On the other hand, the number of crabs per trap decreased exponentially with longer soak times.

Stevens et al. (2000) consider that, in the case of the Tanner crab (Chionoecetes bairdi) fishery in Alaska, an average of 1.5 crabs per trap lost is not high, but point out that the number of traps, the extent of trap deterioration and the fishing mortality associated with each of them, may all lead to important differences in the number of crabs caught, so that keeping a record of each trap in each fishery is a basic rule when attempting to reduce these effects.

The decrease in the number of crabs per trap with longer soak times may be related to the escape of some animals, death by starvation, presence of predators, cannibalism, or decrease of trap retention rate as a result of deterioration (Breen 1990; Stevens et al. 2000). In the first case, trap selectivity is determined by trap shape and size, as well as by net mesh size; in the case of traps used for C. johngarthi, the likelihood of escape is low for specimens larger than 115 mm carapace width (Cervantes-Diaz et al. 2006). However, it is worth noting that currents and type of seabed substratum may facilitate the escape of crabs. These effects are mentioned by Bullimore et al. (2001).

The findings reported here indicate a trend towards a reduction in the number of crabs per trap with longer soak times; however, crabs remain trapped for a prolonged time, and the effect is likely not negligible if the number of lost traps is high, as reported for other crab fisheries (Breen 1987, 1990). In this regard, the vessel administration company informed that 400 traps were lost over the course of the 14 trips investigated, averaging 29 per trip, which were replaced as needed. The causes included loss of marking floats, theft of some fishing lines, and trap fouling in the seabed when lines are collected.


Given the low number of lost traps, a notorious effect of ghost fishing on the C. johngarthi population is not expected, but evidence showed that traps remained undamaged after three months of immersion, and although the number of crabs per trap decreases with longer soak times, after 1,000 h of immersion each contained 7-18 crabs, which may have been trapped since the beginning of the fishing period.

Grossly, ghost fishing associated with the 14 trips studied would amount to 4,200 crabs (400 traps x 18 crabs). This figure may vary because, similar to other fisheries, traps continue fishing with trapped specimens serving as bait (Breen 1990), in the absence of bait (Stevens et al. 2000) or with crabs moving in and out of traps (Godoy et al. 2003). In this respect it is worth noting that the likelihood of escaping is minimum, given the conical design of the traps used (Gagnon & Baicireau 1991).


There is no information about the presence of species likely to feed on C. johngarthi, but the scarce bycatch included octopuses and swellshark Cephaloscyllium ventriosum (Cervantes-Diaz et al. 2006), which might move into the trap, feed on crabs, and then leave.

The information was analyzed disregarding any likely effect of the seasons associated with the development of the life stages in C. johngarthi, which would be reflected in the catch composition, (i.e., number of crabs, sizes and sex ratio). However, ghost fishing has been found to be a dynamic process, related to the characteristics of the fishing areas, season of the year and per-trap mortality rate (Breen 1990; Guillory et al. 2001). Nevertheless, in terms of the number of crabs caught by ghost fishing, our findings suggest a potential effect on the C. johngarthi population, and given that the fishery is in its early development stages, the implementation of management measures aimed at avoiding or reducing such effects seems convenient, (e.g., by implementing strategies already used in other fisheries) like the one for Dungennes crab Cancer magister m California, United States, where the use of mechanisms to facilitate the escape of crabs from lost traps is mandatory (Breen 1990).


The authors thank The National Polytechnic Institute for support through COFAA, EDI, as well as projects SIP. The authors also thank CONACYT and their special thanks to Carlos Suarez, president of the enterprise Acuaproductos Baja S.A. de C.V.


Al-Masroori, H., H. Al-Oufi, J. L. McIIwain & E. McLean. 2004. Catches of lost fish traps (ghost fishing) from fishing grounds near Muscat, Sultanate of Oman. Fish. Res. 69:407-414.

Breen, P. A. 1987. Mortality of Dungeness crabs caused by lost traps in the Fraser River Estuary, British Columbia. Am. J. Fish. Management 7:429-435.

Breen, P. A. 1990. A review of ghost fishing by traps and gillnets. In: R. S. Shornura & M. L. Godfrey, editors. Proceedings of the Second International Conference on Marine Debris,. April 2-7, 1989. Honolulu, Hawaii. Memo. NHFS. NOAA-TM-NMFS-SUFSC154, 1990. U.S. Dep. Comer. N O M Tech.

Bullimore, B. A., P. B. Newman, M. J. Kaiser, S. E. Gilbert & K. M. Lock. 2001. A study of catches in a fleet of "ghost-fishing" pots. Fish. Bull. (Wash. D. C.) 99:247-253.

Cervantes-Diaz, J. L., M. Ramirez-Rodriguez, F. Arreguin-Sanchez & J. Fiol-Ortiz. 2006. Selectivity of traps for the deep-water crab, Cancerjohngarthi Carvacho, 1989 in the Northern Mexican Pacific. Crustaceana 79:217-225.

FAO. 1964. Modern fishing gear of the world 2. FAO--Fish. News Books. 603 pp.

FAO. 1978. Catalogue of fishing gear designs. FAO--Fish. News Books. 160 pp.

FAO. 1995. Code of conduct for responsible fisheries. 41 pp. Gagnon, M. & M. Baicireau. 1991. Sea trials of a galvanic corrosion delayed release mechanism for snow crab traps. Can. Tech. Rep. Fish. Aquat. Sci. 1803. 17 pp.

Godoy, H., D. M. Furevik & S. Stiatnsen. 2003. Unaccounted mortality of red king crab (Paralithodes camtschaticus) in deliberately lost pots off Northern Norway. Fish. Res. 64:171-177.

Guillory, V., A. McMillen-Jackson, L. Hartman, H. Perry, T. Floyd, T. Wagner & G. Graham. 2001. Blue Crab Derelict Traps and Trap Removal Programs. Gulf States Marine Fisheries Commission, NOAA. Publication No. 88. 13 pp.

Jennings, S. & M. J. Kaiser. 1998. The effects of fishing on marine ecosystems. Adv. Mar. Biol. 34:201-352.

Krouse, J. S. 1989. Performance and selectivity of trap fisheries for crustaceans. In: J. F. Caddy, editor. Marine invertebrate fisheries: Their assessment and management. New York: John Wiley & Sons. pp. 307-326.

Lopez-Rocha, J. A., F. Arreguin-S & M. Ramirez-R. 2006. Stock size estimation of the Mexican crab, Cancer johngarthi Carvacho, 1989 on the southwestern peninsula of Baja California, Mexico. Crustaceana 79:1033-1047.

Nedelec, C. 1975. Catalogo de artes de pesca artesanal. FAO--Fish. News Books. 191 pp.

Pawson, M. G. 2003. The catching capacity of lost static fishing gears: Introduction. Fish. Res. 64:101-105.

Ramirez-Rodriguez., M., F. Arreguin-Sanchez, G. De la Cruz-A & P. E. Balart. 2003. Distribucion y abundancia de Cancer johngarthi Carvacho, 1989 (Decapoda: Brachyura: Cancridae), en la costa occidental de Baja California Sur, Mexico. In: M. E. Hendrickx, editor. Contribuciones al estudio de los crustaceos del Pacifico este 2. Instituto de Ciencias del Mar y Limnologia. UNAM. pp. 165-168.

Ramirez-Rodriguez, M., F. Arreguin-Sanchez, J. Lopez-Rocha, C. Lopez-Ferreira & E. Balart-Perez. 2007. Cancer johngarthi crabs in the southwest region off the Baja California Peninsula, Mexico: A potential new fishery. Am. Fish. Soc. Symp. 49:587-592.

SAGARPA (Secretaria de Agricultura. Ganaderia, Desarrollo Rural, Pesca y Alimentacion). 2004. Actualizacion de la Carta Nacional Pesquera y su anexo. Diario Oficial (Mexico), lunes 15 de marzo de 2004, Segunda Seccion. pp. 1-112.

StatSoft, 1999. Statistica for Windows, vol. III. Statistics, 2nd ed.

Stevens, B. G., I. Vining, S. Byersdorfer & W. Donaldson. 2000. Ghost fishing by Tanner crab (Chionoecetes bairdi) pots off Kodiak, Alaska: pot density and catch per trap as determined from sidescan and pot recovery data. Fish. Bull. (Wash. D. C.) 98:389-399.


Centro Interdisciplinario de Ciencias Matinas, Instituto Politecnico Nacional Ar. IPN s/n, Col. Playa Palo de Santa Rita, La Paz, Baja California Sur, Mexico, 23096

* Corresponding author,
Size and mean catch per hour-trap, trap and line set for
C. johngarthi according to the effective fishing time.

     Effective fishing
       time (hours)                                 Carapace
                        No. of   C/hour-             width
Min.    Max.    Mean     sets     trap     C/trap     (mm)

    9    60      34      499      0.96       29       137
   61    147     90       91      0.34       29       131
  152    522     347      43      0.06       20       130
1,296   5,253   3,329     18      0.005      15       137
Gale Copyright: Copyright 2008 Gale, Cengage Learning. All rights reserved.