canyon heads, where smaller individuals concen- trate and where food resources would be enhanced due to the autumn plankton bloom and the materials carried downwards from shallower areas influenced by river outflows in this season. The higher activity of submarine canyons in the northwestern Mediter- ranean has been pointed out by Courp and Monaco (1990) in the Lacaze-Duthiers Canyon (to the north of Catalonia) and for our study region by Puig and Palanques (1998a,b). Furthermore, the influence of episodic fluxes on the behaviour of benthic species in the Mediterranean has been shown by Buscail et al. (1990), De Bovée (1990), Féral et al. (1990), Cartes et al. (2002) and Puig et al. (2001). New evi- dence on benthic-pelagic coupling and the relation- ship with particular benthic compartments such as bacteria, meiobenthos and macrobenthos has appeared recently (Angel and Smith, 2000). Compa- ny et al. (2001) showed the link between the popu- lation structure of different shrimp species in the genus Plesionika Bate, 1888, such as the presence of mature females and recruitment, and the presence of nepheloid layers at different depths. However, the relationship between the geomorphological struc- ture and environment of canyons and the exploited natural resources has not yet been demonstrated in situ as a cause-effect, and the results in this work are a case in this direction. Some aspects are not entire- ly clear yet. For instance, there is no hypothesis yet as to why females undergoing gonadal maturity would concentrate primarily on the middle and lower slope in spring and summer or why smaller sizes concentrate at canyon heads. Eco-physiologi- cal studies are needed in order to clarify the possible relationship between particle fluxes and different population structures of deep-water shrimp in rela- tion to submarine canyons.
Cartes and Sardà (1989) and Maynou and Cartes (1997, 1998) consider this species to occupy one of the lower positions in the benthopelagic food chain but to be atypical among deep-sea decapod crus- taceans in that it exhibits a relatively high proportion of full stomachs as compared to other deep-sea decapod crustaceans. The high metabolic and growth rate demonstrated for this species by Com- pany and Sardà (1998, 2000) is likewise indicative of this. Furthermore, more mobile species tend to have higher metabolic rates, that is, they have high- er energy requirements, which translates into a high- er daily ration (Koslow, 1996). Given the reduction in food sources in deep-sea habitats, causing dietary overlap and competition for food (Gage and Tyler,
30 F. SARDÀ et al.
1990), it seems reasonable to assume that A. anten- natus will have specific nutritional requirements during spawning and will therefore tend to adopt a distribution at optimum depths to fulfil those requirements. This could be one of the main reasons for the high level of dominance found for this species in the depth interval studied. In the Catalan Sea total consumption by bathyal decapod crus- tacean assemblages is higher on the upper middle slope (400-900 m) than on the lower middle slope (900-1200 m). The generally lower food consump- tion by decapod crustaceans with depth is consistent with the commonly accepted notion that food avail- ability also declines with depth, which holds both for the suprabenthos (one of the main sources of food for benthic decapod crustaceans) and for mesopelagic decapods and euphausiid crustaceans and other crustacean taxa (Carpine, 1970; Cartes, 1998; Cartes and Maynou, 1998; Mura et al., 1998; Carrassón and Cartes, 2002). The reduction in food resources takes place around the zonation boundary located at 900 m, with deep-water rose shrimp shoals being located above that depth.
A comparison of trophic level results with previ- ous studies in shallower waters shows that there are some differences between shallower and deep-sea species (Cartes et al., 2001; Cartes and Carrassón, 2004; Cartes et al., 2004b; Sardà et al., 2004b). Trophic levels of crustaceans from deep-sea area are higher than those obtained from ecological models of the shelf and upper slope area of South Catalan Sea and the coastal area of the Bay of Calvi, Corsi- ca (northwestern Mediterranean Sea). Trophic levels of deep-sea fishes are also in the high range of val- ues compared with the results from shallower areas mentioned above. However, this result cannot be aligned with a viable economical and ecological development of a deep-sea fishery, because the highest trophic levels have been identified to corre- spond to species with low fecundity and a low meta- bolic rate. Moreover, the food web structure is com- posed of highly-specialised organisms and displays low web-like features.
Temperature did not appear to be a determining factor in these processes, the temperature in the Mediterranean being constant at around 13 ± 0.5 ºC below 200 m (Hopkins, 1985), so the population structure and behaviour of A. antennatus can be con- sidered temperature-independent. In the deep-water habitat that concerns us here, food availability in the deep-sea food web would seem to be the main lim- iting factor (Gage and Tyler, 1990).