SciELO - Scientific Electronic Library Online

vol.39 número1Larvas de Sergestes arcticus Kroyer, 1855, Neotrypaea uncinata (H. Milne-Edwards, 1837) y Munida gregaria (Fabricius, 1793), entre el seno Reloncaví y Boca del Guafo, sur de ChileInfluencia de aportes fluviales en playas arenosas de Higuerote, costa central de Venezuela índice de autoresíndice de materiabúsqueda de artículos
Home Pagelista alfabética de revistas  

Servicios Personalizados




Links relacionados


Latin american journal of aquatic research

versión On-line ISSN 0718-560X

Lat. Am. J. Aquat. Res. v.39 n.1 Valparaíso  2011 

Lat. Am. J. Aquat. Res., 39(1): 43-55, 2011
DOI: 10.3856/vol39-issuel-fulltext-5



Size structure of a heavily fished benthic/demersal community by shrimp trawling in the Colombian Caribbean Sea

Estructura de tamaños de una comunidad bentónica/demersal fuertemente impactada por la pesca de arrastre camaronero en el Mar Caribe de Colombia


Paúl Gómez-Canchong1,2,3, Renato A. Quiñones1,2 & Luis Manjarrés3

1 Graduate Program in Oceanography, Department of Oceanography, Universidad de Concepción Casilla 160-C, Concepción, Chile
2Center for Oceanographic Research in the Eastern South Pacific (COPAS), Facultad de Ciencias Naturales y Oceanógraficas, Universidad de Concepción, Casilla 160-C, Concepción, Chile
3Laboratorio de Investigaciones Pesqueras Tropicales (LIPET), Universidad del Magdalena Cr32 22-08 Santa Marta, Colombia

Dirección para Correspondencia

ABSTRACT. The benthic and demersal communities in the Colombian Caribbean Sea (CCS) are heavily fished by the shrimp trawling fishery, which presents very high discard levels. Here, we conducted an analysis of the size structure of these benthic and demersal communities in the northern and southern zones of the CCS. Sampling was conducted onboard shrimp trawlers throughout an entire year. No significant differences were found in the size distributions of the two zones, among sites within southern ecoregions, or among the analyzed cruises. This homogeneity in size structure is remarkable since the zones analyzed possess very different species compositions and environmental conditions. The observed size structures were adequately described by non-linear distributions rather than the traditionally employed linear normalized biomass size spectra. It is hypothesized that the non-linearity is due to the effect of fishing and particularly, of discarding. This study emphasizes the need for a greater understanding of the impacts that trawl fishing has on community size structure and the applicability of this knowledge towards fishery resource management in ecosystems with high diversity.

Keywords: fishing effects, bycatch, ecological indicators, size spectra, Pareto distribution, Caribbean Sea.

RESUMEN. Las comunidades bentónico-demersales en el Mar Caribe de Colombia (MCC) son fuertemente explotadas por la pesca de arrastre camaronero, presentando niveles de descarte muy altos. Se efectuó un análisis de la estructura de tamaños de estas comunidades bentónico-demersales en las zonas norte y sur del MCC. Se realizaron muéstreos a bordo de las embarcaciones de arrastre de camarón a lo largo de un año. No se encontraron diferencias significativas entre las distribuciones de tamaños de las diferentes zonas, ecorregiones de la zona sur y cruceros analizados. Esta homogeneidad en la estructura de tamaños es destacable ya que las areas analizadas difieren en composición de especies y condiciones medioambientales. Las estructuras de tamaño observadas, fueron descritas adecuadamente por distribución no-lineales, en lugar del espectro lineal de tamaños de biomasa normalizado tradicionalmente utilizado. Se plantea la hipótesis que la no-linealidad se debe al efecto de la pesca y, particularmente al descarte. Este estudio enfatiza la necesidad de entender los impactos de la pesca de arrastre sobre la estructura de tamaños de las comunidades y la aplicabilidad de este conocimiento para el manejo de los recursos pesqueros en ecosistemas con alta diversidad.

Palabras clave: efectos de la pesca, bycatch, indicadores ecológicos, espectro de tamaños, distribución de Pareto, Mar Caribe.


Tackling ecological complexity in marine ecosystems is a very difficult task due to high species diversity, numerous interactions among species, the scarce quantification of those interactions, and physical variability. Body size has been suggested as an aggregating criterion that reduces the complexity of the system to a manageable level (Sheldon et al., 1972; Platt, 1985; Quiñones, 1994) because of the numerous allometric relationships linking body size with attributes at the (i) individual level (e.g. metabolic rate, life span, growth; Peters, 1983; Brown et al., 2004), and (ii) at population and community levels (e.g. food needs, dispersal, space use; Peters, 1983; Cohen et al., 2003).

Size spectrum research has shown that, in the absence of major disturbances, the size distribution of a community will present regularities and be relatively stable over time (Sheldon et al., 1972; Platt, 1985; Quiñones, 1994). Regularities in the biomass size distribution of aquatic communities have been observed in (i) offshore systems (e.g. Sheldon et al., 1972; Rodríguez & Mullin, 1986a, b; Quiñones et al., 2003), (ii) lakes (e.g. Sprules & Munawar, 1986; Echevarría et al., 1990; Gaedke, 1993), (iii) salt marshes (Quintana et al., 2002), and (iv) benthic communities (e.g. Schwinghamer, 1981; Warwick, 1984; Duplisea, 2000; Quiroga et al., 2005). Sheldon et al. (1972) and Kerr (1974) proposed the "Linear Biomass Hypothesis", in which biomass is constant when organisms are organized in logarithmic size classes. In contrast, Platt and Denman's model (Platt & Denman, 1977, 1978) predicts a slight decrease of biomass with organism size, with a slope cióse to -0.22, and proposes an allometric structure for the pelagic ecosystem. The few biomass size spectra constructed in near steady state systems and in those that cover a wide size range (e.g. North Pacific Central Gyre, Rodríguez & Mullin, 1986a; oligotrophic areas of the Northwest Atlantic, Quiñones et al., 2003) support Platt and Denman's prediction. Nevertheless, the response of marine communities in terms of their biomass size distributions when confronting factors that could affect the resilience of marine ecosystems remains scarcely studied.

In fisheries, the size spectrum approach has been applied to predict fish production firom lower trophic levels (Sheldon et al., 1977; Borgmann, 1983; Moloney & Field, 1985), to rank the productivity of many important fishing grounds and lakes (Boudreau & Dickie, 1992), and as an indicator of the impact of fisheries at a multispecies level (Gobert, 1994; Bianchi et al., 2000; Stobberup et al., 2005).

Several authors have proposed that fishing makes the slope of the size spectrum steeper (i.e. more negative) because fishing selectively removes larger individuals and reduces survival (Gislason & Rice, 1998; Bianchi et al., 2000). Obviously, this is based on the assumption that the linearity of the spectrum is not lost. However, ecosystems far firom steady state can display non-linear normalized biomass-spectra (Quiñones, 1994; Rodríguez, 1994) and high levels of fishing may cause the size distribution of the biota to be drastically modified (Jennings & Kaiser, 1998).

Worldwide, shrimp bottom trawling, due to its lack of selectivity and the physical damage that the trawl inflicts on the seabed, has a high impact on the ecosystem (Jennings & Kaiser, 1998; Cook, 2003; Kaiser & Hiddink, 2007; Caddy, 2007). For the specific case of the CCS, bycatch can be as high as 93%, of which around 80% is discarded (Viaña et al., 2004). A retrospective study in the northern CCS of the shrimp trawling fishery showed a general downward trend in the biomass of demersal resources (García et al., 2007) and bycatch-to-shrimp catch ratios fluctuating between 12:1 and 30:1 (Viaña et al., 2004). As a consequence, the CCS shrimp fishery has faced difficult times since the beginning of this decade (Zúñiga et al., 2006). This led to a reduction in fleet, and in 2005, the subsequent closure of the most important fishing industries (Zúñiga et al., 2006) after over 50% of the vessels in the fleet suffered financial losses in the 2004-2005 period.

It is well known that shrimp trawling has profound effects on demersal communities due to direct mortality in the form of bycatch (Kelleher, 2005). Here, we analyze the size distribution of biomass and abundance of the coastal benthic/demersal community, which has been heavily fished by the shrimp trawling fishery in the Colombian Caribbean Sea. It is hypothesized that the size structure of this community is disturbed and, consequently, size spectrum linearity is lost.


Study area

The study area includes the continental shelf of the CCS (Fig. 1), covering a coastline of 1,600 km. The Magdalena River naturally divides this area into northern and southern fishing grounds. The northern zone (12°40'N, 77°19'W) is characterized by the presence of seasonal upwelling events fuelled by the Trade Winds (Andrade, 2000), whereas the southern zone (11°50'N, 71°18'W) is significantly influenced by continental discharges (Patino & Florez, 1993).

Figure 1. a) Study area. Colombian Canbbean Sea, b) distribution of the analyzed trawl hauls in the northern and southern zones, c) localization of the southern ecoregions: (1) Uraba Gulf, (2) South of Morrosquillo, (3) Islands, (4) Morrosquillo Gulf, and (5) North of Morrosquillo.

Figura 1. a) Área de estudio. Mar Caribe de Colombia, b) distribución de los lances analizados en las zonas norte y sur, c) localización de las ecorregiones de la zona sur: (1) Golfo de Urabá, (2) Sur de Morrosquillo, (3) Islas, (4) Golfo de Morrosquillo, y (5) Norte de Morrosquillo.

Nearly the entire length of the northern coast is arid, with a wide continental shelf, and there are large extensions of sandy bottoms reaching Riohacha, where sand and coral grounds are found (Pineros & Sieguert, 1989). The southern coast has abundant vegetation and the continental shelf is narrow, no more than 84 m from the coast. The bottom is fíat and composed of mud and muddy sand with depths ranging between 25 and 50 m. The water exchange between the northern and southern zones of the CCS, separated by the mouth of the Magdalena River, is minimal because the river water always tends to move southwards due to the anticyclonic gyre that is generated between Central América and the southern Colombian Caribbean (Andrade et al., 2003). This constitutes a biogeographic barrier for most marine species.

The southern zone can be divided into three ecoregions (INVEMAR, 2000): i) the gulf of Morrosquillo ecoregion, influenced by continental discharges with low water transparency and low exposure to wind and wave action, seagrass beds, deltaic and estuarine systems with mangrove forests; ii) the Coralline Archipelagos ecoregion (Islands), encompassing islands, shallow carbonate platforms and coral shoals located off the mainland coast between Islas del Rosario and Isla Fuerte, as well as the continental shelf between the 40 and 130 m depth contours, and clear waters; and iii) the gulf of Urabá, a wide continental shelf, mainly terrigenous muds, high influence of continental discharges, predominantly

turbid waters, defined by alluvial plains with mangrove forests and a rapidly extending delta of the Atrato River. The gulf may be regarded as an estuary, with scattered coral reefs and seagrass beds along this mountainous coastline. North and South of Morrosquillo constitute Transition Zones.

Sources of information

Between August 2004 and July 2005, two observers were placed onboard commercial shrimp trawlers working in the CCS for a total of 10 fishing cruises (Table 1). Nine of those fishing cruises were conducted in the southern zone and one in the northern zone (September 2004). The vessels used, trips undertaken and station locations (Fig. 1) depended on the fleet activity. Historically, fishing activity has been circumscribed to very specific fishing grounds due to the fishermen's preference for traditional grounds with higher shrimp abundances and the spatial distribution of trawable grounds.


Trawling vessels operating in the study area have steel hulls and are generally similar in terms of engines (450 HP in average), tonnage (around 100 ton), storage capacity, equipment, and fishing gear (Zúñiga et al., 2004). The fishing gear is composed of four Korean or Japanese type nets, two per side, each with a 12 m head rope and mesh size of 4.45 mm (Zúñiga et al., 2004). Excluding devices were not employed during the fishing operations. Similar tow speeds (2.8 knots) and identical electronic equipment (satellite navigator and echo-sounder) were employed throu-ghout (Viaña et al., 2004).

Onboard, one of four fishing nets was randomly selected from each tow for the assessment of catch composition. The following information was recorded: date, hour, depth, location, and tow duration. Target and incidental catches were sorted, identified and weighed onboard with a hand-held spring scale, 20% of the homogenized discards (Pauly, 1983) were stored for further identification and processing in the laboratory on land. The precisión of the onboard weighing procedure was verified by taking samples (30%) of target and incidental catches from 25% of the tows and re-analyzing these in the laboratory. Observed catches were extrapolated for the entire tow by multiplying the valúes of the selected net by four.

A total of 197 fishing trawl hauls were analyzed. Approximately 95,000 organisms were directly counted and measured, representing approximately 700,000 specimens caught in the fishing gear (Table 1). The majority of the tows lasted 3-4 h and were performed at night (between 18:00 and 6:00), covering a depth range of 12 to 65 m. Sporadically, some pelagic species were caught in the trawls, these species were not included in the calculations because they feed on resources allochthonous to the benthic system.

Construction and analysis of size spectra

The biomass and the number of individuals caught per hour (h) in each size range (expressed in grams) were calculated by the ratio estimator method (Scheaffer et al., 1990):

where n is the number of tows sampled, a¡ is the biomass (B) or the number of individuals (N) caught in the size range during the ith tow, and Oj is the duration of the ith tow. R provides unbiased and confident estimates, especially because the bycatch and the time of the tows are highly variable (Ye, 2002).

Size distribution analyses are only applicable to data in the size range over which the gear effectively samples the community (Jennings & Dulvy, 2005). The spectra were constructed taking into consideration gear selectivity and therefore limits the computation to those individuals within a size range that is not strongly affected by gear selectivity. The lower limit of this range was determined by using an adaptation of the linearized catch curve based on length composition data proposed by Pauly (1983) and Sparre & Venema (1995) (Fig. 2). This method calculates the range of lengths in which the gear does not select 100% of the individuals. The upper limit on the other hand, was established as the highest size range found in at least 90% of tows.

Figure 2. Pseudo-Catch Curve to determine the body mass on which gear selectivity has no effect.

Figura 2. Curva de pseudo-capturas para determinar el tamaño corporal a partir del cual la selectividad del arte no tienen ningún efecto.

Normalized size spectra for biomass (NBSS) and abundance (NASS) for pre-defined strata (zone, southern ecoregion, and monthly fishing cruises) were constructed following Blanco et al., (1994). The normalized size spectrum was obtained after dividing the biomass or abundance in every size class (M) by its amplitude. A base 2 logarithm was applied to the independent (weight) and dependent (abundance or biomass) variables as proposed by Platt & Denmann (1977) and Blanco et al., (1994). The least squares method was used to test the fit to a linear model (Zar, 1999).

To analyze the size distribution, we also used the Pareto distribution (Vidondo et al., 1997), a model widely used in other disciplines for dealing with size structured systems. The Pareto I distribution has a probability density function defined as:

where s is the size, and c and K are the distribution's shape and scale parameters, respectively. The parameter K is a constant that, in this case, represents the size of the smallest object. The parameter c is an empirical constant that describes the decline of probability as size increases.

The size distributions were also analyzed using a non-linear Pareto II distribution (Vidondo et al., 1997), whose probability density function is given by:

where D is an additive constant, different to zero in the type II Pareto distribution. The ecological meaning of the polynomial coefficients K, c, and D is far from obvious (Vidondo et al., 1997) and, accordingly, no attempt was made to analyze them.

In order to compare the Pareto distributions, a Kolmogorov-Smirnov test (Zar, 1999) was conducted. To obtain a comparative picture of the size distribution from different fishing cruises, we performed a dendrogram applying a single linkage agglomeration clustering method (Legendre & Legendre, 1998) based on a modified Bray-Curtis dissimilarity index (Rodríguez & Mullin, 1986b). All calculations were performed using the JMP© statistical software package.


The linear model was not able to represent adequately the NBSS (Fig. 3) or the NASS (plots not shown) of these benthic/demersal communities because, even when they adjusted significantly to the linear model (Table 2), the clear presence of residuals with tendency (Fig. 3) violated the assumption of the linear model that the residuals must be randomly scattered (Zar, 1999). Residuals peaked around mid-range size classes (16-64 g) and troughed on either side, displaying a dome-like pattern (Fig. 3).

Figure 3. Examples of some of the normalized biomass size spectra (left panel) for the different a) zones, b) ecoregions, and c) cruises in the Colombian Canbbean Sea. The residuals for each NBSS (normalized size spectra for biomass) are added in the right panel.

Figura 3. Ejemplos de los espectros de tamaño de biomasa normalizados (panel izquierdo) para las diferentes a) zonas, b) ecorregiones, y c) cruceros en el Mar Caribe de Colombia. Los residuos para cada ETBN (espectro de tamaño normalizado para la biomasa) se observan en el panel derecho.

The cumulative biomass distribution function adjusted well to a type II Pareto model (i.e. Pareto's D parameter was different from 0 in all the cases; Table 2; Fig. 4). On the other hand, the cumulative biomass distribution function did not fit a type I Pareto model since the probability density of the bigger individuals was lower than that predicted by the model (Fig. 4).


Figure 4. Cumulative Pareto distribution. a) North vs south, b) southern ecoregions, and c) cruises in the southern zone.

Figura 4. Distribución acumulada de Pareto. a) Norte vs sur, b) ecorregiones del sur, y c) cruceros en la zona sur.

No significant differences were found for the type II Pareto size distributions (Fig. 4) between the different zones (Table 3), among the southern ecoregions (Table 4), or among monthly fishing cruises (Table 5).

For the SVFA11 cruise, the size structure was discontinuous, with few or no individuals in the upper size classes (Figs. 3 and 4). All type II Pareto size distributions showed increased dispersal for the size range containing individuals larger than 256 g.

The most closely associated group of monthly cruises in the dendrogram (Fig. 5) had no temporal coherence or association with dry or rainy seasons. All cruises were associated under a Bray-Curtis dissimilarity index of 0.5.

Figure 5. Dendrogram of dissimilarities (Bray-Curtis index) of size distribution between southern cruises.

Figura 5. Dendrograma de disimilaridades (índice de Bray-Curtis) de la distribución de tamaños entre los cruceros de la región sur.


Our results show that the linear model is inadequate for representing the size structure of this heavily fished benthic/demersal community of the CCS. Non-linear Pareto II distribution provides a proper description of the observed size structure. Several studies have shown non-linear trends (e.g. dome-like patterns) in size spectra, including empirical analyses (e.g. Sprules & Munawar, 1986; Sprules et al., 1991; Boudreau et al., 1991) and simulations (Maury et al., 2007). A variety of reasons have put forward to explain this pattern, such as the propagation of a biomass or energy peak through the size spectrum (Silvert & Platt, 1978; Han & Straskraba, 2001) or the slowdown of growth in larger sizes cióse to the asymptotic length (Maury et al., 2007), among others. Dickie et al., (1987) and Boudreau et al., (1991) suggested that "there may exist a variety of smaller-scale strategies" that might explain the existence of a dome for each trophic group of organisms with similar production efficiencies (e.g. fish, meiofauna, large invertebrates; Dickie et al., 1987). This argument does not apply to the benthic/demersal community analyzed herein because in this case the domes observed are composed of many species from several trophic groups and possessing many different production efficiencies. In fact, over 250 taxa of fish and invertebrates were found during our fishing trips, and at least 100 species were captured in each trawl haul.

The fitting of biomass size distributions to the Pareto type II model in this overexploited community emphasizes the need to explore the use of non-linear size spectra as a tool for analyzing community dynamics in heavily fished ecosystems. In fact, Benoit & Rochet (2004) and Shin & Cury (2004) suggest that fishing effects may be better captured by the curvature of the size spectrum than by its slope, proposing that this curvature is a candidate for determining a reference point at the community level that could be useful in detecting ecosystem overexploitation. In this context, the D valué in the Pareto type II distribution

Size structure of a heavily fished community in the Colombian Caribean Sea is an interesting candidate for working with the curvature of the size structure, especially if we consider that a D = 0 means an absence of curvature (i.e. the Pareto type I or linear model is considered to be a special case of the Pareto type II when D=0; Vidondo et al., 1997).

No statistical differences were found in the size structure (Pareto distributions) between southern and northern zones, within the southern ecoregions, or among the analyzed months (Tables 3 to 5). This homogeneity in size structure is remarkable because the analyzed areas differ (Duarte et al., 2006) in: (i) species composition, (ii) depth ranges of the fishing areas, (iii) levels of precipitation and river discharge in the different ecoregions, (iv) the presence of an upwelling focus in the northern zone, and (v) isolation through distance and múltiple geographical barriere that reduces the exchange of individuals. Nevertheless, this result is consistent with the conservative nature of the size structure of marine communities proposed by several authors (e.g. Pope et al., 1987; Quiñones et al., 2003; Sweeting et al., 2009). In addition, our results support the idea that species composition and size structure may respond differently to environmental disturbances. Several examples are found in the literature in which — unlike ours — species composition remains almost unaltered despite changes in the size structure. For instance, a negative temperature anomaly clearly affected the size distribution of zooplanktonic biomass in the central gyre of the North Pacific Ocean in the summer of 1969 (Rodríguez & Mullin, 1986b), nevertheless, the taxonomic approach did not show any effect of this temperature anomaly on the species composition of the macrozooplanktonic community (Rodríguez & Mullin, 1986b).

Most size spectra presented high dispersión within the relatively low abundance found in the size ranges corresponding to larger individuals. In fact, in one case (cruise SVFA11, Figs. 3 and 4), no individuals were found in the upper size classes. This could be explained by long-term overfishing of large sizes, as suggested for the CCS by Duarte et al., (2006), by the escape of larger individuals due to slow tow speeds, and/or by the migrations or nictimeral vertical movements of larger individuals to areas or depths beyond the reach of the fleet and/or the fishing gear. In fact, migratory events by larger individuals of some species (e.g. lañe snapper, murtón snapper) are commonly observed in the study area (Manj arres et al., 2001; Gómez-Canchong et al., 2004).

We are aware that one of the limitations of our research is the relatively short time period covered (i.e. one year). Nevertheless, the wide spatial coverage of the sampling, as well as the consistency of the non-linear pattern found, suggests that the results obtained are representative of the size structure of the study area for the period analyzed. Unfortunately, we have not had the opportunity to conduct new research onboard the shrimp trawling fleet. However, we have had the opportunity to sample the size spectra of the benthic/demersal community during two scientific cruises (August and December 2005) in the northern Morrosquillo ecoregion (data not shown), finding a non-linear size distribution and no significant differences with the size distribution found during the sampled year (Figs. 3 and 4).

Shrimp trawling fisheries kill benthic invertebrates as the fishing gear is dragged across the seabed (Cook, 2003; Duarte et al., 2006; Kaiser & Hiddink, 2007) reducing the overall biomass and production of benthic invertebrate communities (Hiddink et al., 2006). Most of the carbón released by discarding could fuel microbial pathways rather than metazoan fish food pathways (e.g. Jennings et al., 2001; Kaiser et al., 2003). Furthermore, the discarded fish have low chances of survival, unlike most crustaceans and mollusks (40-50%), and starfish (90%) (Groenewold & Fondss, 2000). Due to the shallowness (12 to 65 m depth) of the fishing grounds, these discards produce an important energy input into the benthic system. Both trawling and discarding seem to favor the dominance of small benthic organisms and scavengers (Cook, 2003; Duarte et al., 2006; Kaiser & Hiddink, 2007), even when no real increase in benthic scavenger abundance has been observed (Ramsay et al., 2000). In other words, bottom trawling removes other populations but does not alter scavenger populations (Kaiser & Hiddink, 2007). The deterioration of the benthic/demersal communities due to trawling and discarding has been put forward as a possible cause for the decreased catches and biomasses of shrimp and demersal resources in the CCS (Duarte et al., 2006; García et al., 2007). Figure 6 represents a conceptual scheme of how the biomass size spectra of a community could be affected by fishing, producing non-linear spectra as a possible output. The key factors in the case of the shrimp trawling fishery (Fig. 6d-h) are the more or less uniform removal of individuals (non-selective gear) across the size spectrum and the high levels of discards that increase the relative biomass of scavengers.

Figure 6. Conceptual diagram of the changes expected in a community size spectram when dealing with fishing impacts, prey release and bycatch. a) Unexploited state, b) as fishing mortality mercases and the relative abundance of large individuals decreases, a steeper biomass size spectram is expected, c) on the other hand, fishing may indirectly increase the abundance of small prey species through the removal of their predators, making the slope of the size spectram even steeper, (d-i) non size-selective fishing (e.g. shrimp trawling). Different processes are analyzed including (c and e) prey release, especially of smaller individuals, d) the escape of larger individuals from the fishing gear, f) increased scavengers due to discarding, g) the carbón released by discarding, which could also fuel microbial pathways and not just the metazoan fish food web, thereby potentially reducing fisheries production, and h) continued fishing forcing over time that causes a resilience threshold to be crossed after which the community size stracture is lost, increasing the dispersal and gaps in the size stracture. Plots (a) to (c) are based on Sweeting et al. (2009). Dashed lines represent the slope of a theoretical size spectram from an unexploited system. Arrows indicate the trend followed by the biomass in different size ranges due to the analyzed forcing process.

Figura 6. Diagrama conceptual de los cambios esperados en el espectro de tamaños de una comunidad cuando enfrenta impactos de la pesca, liberación de presas, y bycatch. a) Estado no explotado, b) conforme se incrementa la mortalidad por pesca y la abundancia relativa de los individuos más grandes decrece, se espera un ETBN más empinado, c) por otro lado, la pesca puede indirectamente incrementar la abundancia de las pequeñas presas al remover a sus depredadores, haciendo más empinada la pendiente, d-i) pesca no selectiva por tamaño (e.g. arrastre camaronero). Diferentes procesos son analizados incluyendo (c y e) liberación de la presa, especialmente de los pequeños individuos, d) el escape de individuos grandes del arte de pesca, f) incremento de los carroñeros debido al descarte, g) el carbón liberado por el descarte, puede alimentar las ratas microbianas y no solamente la rata clásica hacia los depredadores, potencialmente reduciendo la producción pesquera, y h) la continuación del forzante pesquero a lo largo del tiempo puede causar cruzar un umbral de resiliencia luego del cual la estructura de tamaños de la comunidad se pierde, incrementando la dispersión y los espacios en la estructura de tamaños. Gráficos (a) a (c) están basados en Sweeting et al. (2009). Líneas entrecortadas representan la pendiente del espectro de tamaños teórico de un sistema no explotado. Las flechas indican la tendencia seguida por la biomasa en las diferentes clases de tamaño debido al forzamiento analizado.

The case of the CCS shrimp trawling fishery is not unique; trawling fisheries are considered to be among those with the highest impact on the ecosystem (Jennings & Kaiser, 1998; Caddy, 2007). Accordingly, there is growing interest in generating ecological indicators that support an ecosystem approach to fisheries (e.g. Rice, 2000; Garcia et al., 2003; Rice & Rochet, 2005). In line with what Rodríguez (1994) has suggested, "only when we understand the sources, scales and mechanisms that control variability will we be able to make biomass spectrum a useful approach in biological oceanography", we believe that the case presented here (the size structure of these benthic/demersal communities associated with the shrimp fishery in the CCS and especially the fact that their spectra were far from linear), emphasizes the need to achieve greater insight into the dynamics of non-linearities in size spectra analyses. It also emphasizes the need for a size stracture theory capable of dealing with ecosystems subject to strong disturbances, in order to use size spectram as an

Size structure of a heavily fished community in the Colombian Caribean Sea appropriate and successful fisheries management indicator so as to contribute to the development of an ecosystem approach to fisheries.



The sampling program and data acquisition was financed by the Instituto Colombiano para el Desarrollo de la Ciencia y la Tecnología COLCIENCIAS (Grant No. 1117-09-13273 "Valoración Biológico-Pesquera y Ecológica de la Pesca Industrial de Arrastre Camaronero e Impacto de la Introducción de Dispositivos Reductores de Fauna Acompañante, en el Mar Caribe Colombiano"). We wish to thank the valuable field and laboratory support of the members of the Grupo de Investigación Evaluación y Ecología Pesquera (GIEEP, Universidad del Magdalena, Colombia). Paúl Gómez-Canchong was funded by a Doctoral scholarship of the Deutscher Akademisher Austausch Dienst, (DAAD, Germany). The size structure analysis and Renato A. Quiñones were funded by the COPAS Center (FONDAP, CONICYT, Chile).



Andrade, C.A. 2000. The circulation and variability of the Colombian Basin in the Caribbean Sea. PhD Thesis, University of Wales, 223 pp.        [ Links ]

Andrade, C.A., E.D. Barton & C.N.K. Mooers. 2003. Evidence for an eastward flow along the Central and South American Caribbean Coast. J. Geophys. Res., 108(C6):3185.        [ Links ]

Benoit, E. & MJ. Rochet. 2004. A continuous model of biomass size spectra governed by predation and the effects of fishing on them. J. Theor. Biol., 226: 9-21.        [ Links ]

Bianchi, G., H. Gislason, K. Graham, L. Hill, X. Jin, K. Koranteng, S. Manickchand-Heileman, I. Paya, K. Sainsbury, F. Sánchez & K. Zwanenburg. 2000. Impact of fishing on size composition and diversity of demersal fish communities. ICES J. Mar. Sci., 57: 558-571.        [ Links ]

Blanco, J.M., F. Echevarría & C. García. 1994. Dealing with size-spectra: some conceptual and mathematical problems. Sci. Mar., 58: 17-29.        [ Links ]

Borgmann, U. 1983. Effect of somatic growth and reproduction on biomass transfer up pelagic food web as calculated from particle-size-conversion efficiency. Can. J. Fish. Aquat. Sci., 44: 136-140.        [ Links ]

Boudreau, P.R. & L.M. Dickie. 1992. Biomass spectra of aquatic ecosystems in relation to fisheries yield. Can. J. Fish. Aquat. Sci., 49: 1528-1538.        [ Links ]

Boudreau, P.R., L.M. Dickie & S. Kerr. 1991. Body-size-spectra of production and biomass as system-level indicators of ecological dynamics. J. Theor. Biol., 152: 329-339.        [ Links ]

Brown, J.H., J.P. Gillooly, A.P. Alien, V.M. Savage & GB. West. 2004. Toward a metabolic theory of ecology. Ecology, 85: 771-1789.        [ Links ]

Caddy, J. 2007. Marine habitat and cover: their importance for productive coastal fishery resources. UNESCO Publishing, Paris, 256 pp.        [ Links ]

Cohen, J.E., T. Jonsson & S.R. Carpenter. 2003. Ecological community description using the food web, species abundance, and body size. Proc. Nati. Acad. Sci. U.S., 100: 1781-1786.        [ Links ]

Cook, R. 2003. The magnitude and impact of bycatch mortality by fishing gear. In: M. Sinclair & G Valdimarsson (eds.). Responsible fisheries in the marine ecosystem. FAO, Rome, pp. 219-233.        [ Links ]

Dickie, L.M., S.R. Kerr & P.R. Boudreau. 1987. Size-dependent processes underlying regularities in ecosystem structure. Ecol. Monogr., 57: 233-250.        [ Links ]

Duarte, L.O., P.G Gómez-Canchong, L. Manjarrés, C.B. García, F. Escobar, J. Altamar, J.E. Viaña, K. Tejada, J. Sánchez & F. Cuello. 2006. Variabilidad circadiana de la tasa de captura y la estructura de tallas en camarones e ictiofauna acompañante en la pesquería de arrastre del Mar Caribe de Colombia. Invest. Mar, Valparaíso, 34(1): 23-42.        [ Links ]

Duplisea, D.E. 2000. Benthic organism biomass size-spectra in the Baltic Sea in relation to the sediment environment. Limnol. Oceanogr., 45(3): 558-568.        [ Links ]

Echevarría, F., P. Carrillo, F. Jiménez, P. Sánchez-Castillo, L. Cruz-Pizarro & J. Rodríguez. 1990. The size-abundance distribution and taxonomic composition of plankton in an oligotrophic, high mountain lake (La Caldera, Sierra Nevada, Spain). J. Plankton Res., 12: 415-422.        [ Links ]

Gaedke, U. 1993. Ecosystem analysis based on biomass size distributions: a case study of a plankton community in a large lake. Limnol. Oceanogr., 38: 112-127.        [ Links ]

García, C.B., L.O. Duarte, L. Altamar & L. Manjares. 2007. Demersal fish density in the upwelling ecosystem off Colombia, Caribbean Sea: historie outlook. Fish. Res., 85: 68-73.        [ Links ]

García, S., A. Zerbi, C. Aliaume, T. Do Chi & G Lasserre. 2003. The ecosystem approach to fisheries. Issues, terminology, principles, institutional foundations, implementation and outlook. FAO Fish. Tech. Pap., 443: 1-71.        [ Links ]

Gislason, H. & J. Rice. 1998. Modelling the response of size and diversity spectra of fish assemblages to changes in exploitation. ICES J. Mar. Sci., 55: 362-370.        [ Links ]

Gobert, B. 1994. Size structures of demersal catches in a multispecies multigear tropical fishery. Fish. Res., 19: 87-104.        [ Links ]

Gómez-Canchong, P.G., J. Rodríguez, J.C. Arévalo & L. Manjarrés. 2004. Identificación de stocks de pargo rayado Lutjanus synagris (Linnaeus, 1758) en el área norte del caribe colombiano. In: L. Manjarrés (ed.). Pesquerías demersales del área norte del mar Caribe de Colombia y parámetros biológico-pesqueros y poblacionales del recurso pargo. Editorial Gente Nueva, Bogotá, pp. 155-179.        [ Links ]

Groenewold, S. & M. Fonds. 2000. Effects on benthic scavengers of discards and damaged benthos produced by the beam trawl fishery in the southern North Sea. ICES J. Mar. Sci., 57: 1395-1406.        [ Links ]

Han, B.P. & M. Straskraba. 2001. Size dependence of biomass spectra and abundance spectra: the optimal distributions. Ecol. Model., 145: 175-187.        [ Links ]

Hiddink, J.G., S. Jennings, M.J. Kaiser, A.M. Queirós, D.E. Duplisea & G.J. Piet. 2006. Cumulative impacts of seabed trawl disturbance on benthic biomass, production and species richness in different habitats. Can. J. Fish. Aquat. Sci., 63: 721-736.        [ Links ]

Instituto de Investigaciones Marinas y Costeras (INVEMAR). 2000. Programa Nacional de Investigaciones en Biodiversidad Marina y Costera-PNIBM. Instituto de Investigaciones Marinas y Costeras-INVEMAR-FONADE, MMA, Santa Marta, Colombia, 83 pp.        [ Links ]

Jennings, S. & N.K. Dulvy. 2005. Reference points and reference directions for size-based indicators of community structure. ICES J. Mar. Sci., 62: 397-404.        [ Links ]

Jennings, S., T.A. Dinmore, D.E. Duplisea, K.J. Warr & J.E. Lancaster. 2001. Trawling disturbance can modify benthic production processes. J. Anim. Ecol., 70: 459-475.        [ Links ]

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

Kaiser, M.J. & J.G. Hiddink. 2007. Food subsidies from fisheries to continental shelf benthic scavengers: are they important? Mar. Ecol. Prog. Ser., 318: 267-276.        [ Links ]

Kaiser, M.J., J.S. Collie, S.J. Hall, S. Jennings & I.R. Poiner. 2003. Impacts of fishing gear on marine benthic habitats. In: M. Sinclair & G. Valdimarsson (eds.). Responsible fisheries in the marine ecosystem. FAO, Rome,pp. 197-216.        [ Links ]

Kelleher, K. 2005. Discards in the world's marine fisheries: an update. FAO Fish. Tech. Pap., 470: 1-131.        [ Links ]

Kerr, S.R. 1974. Theory of size distribution in ecological communities. J. Fish. Res. Bd. Can., 31: 1859-1862.        [ Links ]

Legendre, L. & P. Legendre. 1998. Numerical ecology. Elsevier Science, New York, 853 pp.        [ Links ]

Manjarrés, L., C.B. García & A. Acero. 2001. Caracterización ecológica de las asociaciones de peces demersales del Caribe colombiano norte con énfasis en pargos (Lutjanidae). Bol. Inst. Invest. Mar., 30: 77-107.        [ Links ]

Maury, O., B. Faugeras, Y.J. Shin, J.C. Poggiale, T. Ben Ari & F. Marsac. 2007. Modeling environmental effects on the size-structured energy flow through marine ecosystems. Part 1: the model. Prog. Oceanogr., 74(4): 479-499.        [ Links ]

Moloney, C.L. & J.G. Field. 1985. Use of particle-size data to predict potential pelagic-fish yields of some southern african areas. S. Afr. J. Mar. Sci., 3: 119-128.        [ Links ]

Patino, F. & F. Florez. 1993. Ecología marina del Golfo de Morrosquillo. Universidad Nacional de Colombia, FEN. Bogotá, 109 pp.        [ Links ]

Pauly, D. 1983. A selection of a simple methods for the assessment of tropical fish stocks. FAO Fish. Cira, 729: 1-54.        [ Links ]

Peters, R.H. 1983. The ecological implications of body size. Cambridge University Press, Cambridge, 329 pp.        [ Links ]

Pineros, V. & P. Sieguert. 1989. La pesquería del camarón en el Caribe colombiano. Tesis, Universidad Jorge Tadeo Lozano, Colombia, 132 pp.        [ Links ]

Platt, T. 1985. Structure of the marine ecosystem: Its allometric basis. In: R.E. Ulanowicz, & T. Platt (eds.). Ecosystem theory for biological oceanography. Can. Bull. Fish. Aquat. Sci., 213: 55-64.        [ Links ]

Platt, T. & K. Denman. 1977. Organization in the pelagic ecosystem. Helgol. Wiss. Meresunter., 30: 575-581.        [ Links ]

Platt, T. & K. Denman. 1978. The structure of the pelagic marine ecosystems. Rapp. Proces-Verbaux Reun. Cons. Perm. Int. Explor. Mer, 173: 60-65.        [ Links ]

Pope, J.G., T.K. Stokes, S.A. Murawski & S.I. Idoine. 1987. A comparison of fish size composition in the North Sea and on Georges Bank. In: W. Wolff, C.J. Soeder & F.R. Drepper (eds.). Ecodynamics: contributions to theoretical ecology. Springer-Verlag, Berlin, pp. 146-152.        [ Links ]

Quiñones, R.A. 1994. A comment on the use of allometry in the study of pelagic ecosystem processes. Sci. Mar., 58: 11-16.        [ Links ]

Quiñones, R.A., T. Platt & J. Rodríguez. 2003. Patterns of biomass size spectra from oligotrophic waters of the Northwest Atlantic. Prog. Oceanogr., 57: 405-427.        [ Links ]

Quintana, X.D., F.A. Comín & R. Moreno-Amich. 2002. Biomass-size spectra in aquatic communities in shallow fluctuating Mediterranean salt marshes (Empordà wetlands, NE Spain). J. Plankton Res., 24: 1149-1161.        [ Links ]

Quiroga, E., R.A. Quiñones, M. Palma, J. Sellanes, V.A. Gallardo, D. Gerdes & G. Rowe. 2005. Biomass size-spectra of macrobenthic communities in the oxygen minimum zone off Chile. Estuar. Coast. Shelf Sci., 62:217-231.        [ Links ]

Ramsay, K., M.J. Kaiser, A.D. Rijnsdorp, J. Craeymeersch & J. Ellis. 2000. The impact of beam trawling on populations of the benthic scavenger Asterias rubens. In: M.J. Kaiser & S.J. de Groot (eds.). The effects of trawling on non-target species and habitats: biological, conservation and socioeconomic issues. Blackwell Science, Oxford, pp. 151-162.        [ Links ]

Rice, J.C. 2000. Evaluating fishery impacts using metrics of community structure. ICES J. Mar. Sci., 57: 682-688.        [ Links ]

Rice, J.C. & M.J. Rochet. 2005. A framework for selecting a suite of indicators for fisheries management. ICES J. Mar. Sci., 62: 516-527.        [ Links ]

Rodríguez, J. 1994. Some comments on the size based structural analysis of the pelagic ecosystem. Sci. Mar., 58: 1-10.        [ Links ]

Rodríguez, J. & M. Mullin. 1986a. Relation between biomass and body weight of plankton in a steady state oceanic ecosystem. Limnol. Oceanogr., 31(2): 361-370.        [ Links ]

Rodríguez, J. & M. Mullin. 1986b. Diel and interannual variation of size distribution of oceanic zooplanktonic biomass. Ecology, 67: 215-222.        [ Links ]

Scheaffer, R.L., W. Mendenhall & L. Ott. 1990. Elementary survey sampling. Duxbury Press, New York, 321 pp.        [ Links ]

Schwinghamer, P. 1981. Characteristic size distributions of integral benthic communities. Can. J. Fish. Aquat. Sci., 38: 1255-1263.        [ Links ]

Sheldon, R.W., A. Prakash & W.H. Sutcliffe Jr. 1972. The size distribution of particles in the ocean. Limnol. Oceanogr., 17: 327-340.        [ Links ]

Sheldon, R.W., W.H. Sutcliffe Jr. & A. Prakash. 1977. Structure of pelagic food chain and relationships between plankton and fish production. J. Fish. Res. Bd. Can., 34: 2334-2353.        [ Links ]

Shin, Y.J. & P.M. Cury. 2004. Using an individual-based model of fish assemblages to study the response of size spectra to changes in fishing. Can. J. Fish. Aquat. Sci., 61: 414-431.        [ Links ]

Silvert, W. & T. Platt. 1978. Energy flux in the pelagic ecosystem: a time dependent equation. Limnol. Oceanogr., 18: 813-816.        [ Links ]

Sparre, P. & S. Venema. 1995. Introduction to tropical fish stock assessment. Manual. FAO Fish. Tech. Pap., 306(1): 1-420.        [ Links ]

Sprules, W.G. & M. Munawar. 1986. Plankton size spectra in relation to ecosystem productivity, size and perturbation. Can. J. Fish. Aquat. Sci., 43: 1789-1794.        [ Links ]

Sprules, W.G., S.B. Brandt, D.J. Stewart, M. Munawar, E.H. Jin & J. Love. 1991. Biomass size-spectrum of the Lake Michigan pelagic food web. Can. J. Fish. Aquat. Sci., 48: 105-115.        [ Links ]

Stobberup, K.A., C.A.O. Inejih, S. Traore, C. Monteiro, P. Amorim & K. Erzini. 2005. Analysis of size spectra off northwest Africa: a useful indicator in tropical areas? ICES J. Mar. Sci., 62: 424-429.        [ Links ]

Sweeting, C.J., F. Badalamenti, G. D'Anna, C. Pipitone & N.V.C. Polunin. 2009. Steeper biomass spectra of demersal fish communities after trawler exclusion in Sicily. ICES J. Mar. Sci., 66: 195-202.        [ Links ]

Viaña, J., A. Medina, M. Barros, L. Manjarrés, J. Altamar & M. Solano. 2004. Inventario y caracterización general de la flota industrial de arrastre que operó en la zona norte del Caribe colombiano durante el año 2000. In: L. Manjarrés (ed.). Pesquerías demersales del área norte del mar Caribe de Colombia y parámetros biológico-pesqueros y poblacionales del recurso pargo. Editorial Gente Nueva. Bogotá, pp. 115-151.        [ Links ]

Vidondo, B., Y. Prairie, J.M. Blanco & C.M. Duarte. 1997. Some aspects of the analisis of size spectra in aquatic ecology. Limnol. Oceanogr., 42(1): 184-192.        [ Links ]

Warwick, R.M. 1984. Species-size distributions in marine benthic communities. Oecologia, 61: 32-41.        [ Links ]

Ye, Y. 2002. Bias in estimating bycatch-to-shrimp ratios. Aquat. Living Resour., 15: 149-154.        [ Links ]

Zar, J. 1999. Biostatistical Analysis. Prentice Hall, New Jersey, 718 pp.        [ Links ]

Zúñiga, H., J. Altamar & L. Manjarrés. 2004. Colombia: Caracterización tecnológica de la flota de arrastre camaronero del Mar Caribe de Colombia. Technical report. REBYC (EP/GLO/201/GEF). FAO, UNEP. Available at: [Revised: 9 June 2008].        [ Links ]


Received: 29 March 2010; Accepted: 2 December 2010

Corresponding author: Paúl Gómez-Canchong (