Comparative study of the influence of environmental factors on the distribution of the fish Prionace glauca (blue shark) and Xiphias gladius (sword fish) of the eastern Mediteranean pelagic ecosystem

Swordfish (Xiphias gladius) and blue shark (Prionace glauca) are two apex predators in the pelagic ecosystem of the eastern Mediterranean Sea. Our current knowledge on their spatio-temporal distribution patterns and the effect that the surrounding environment has on them is limited. During the period of 1998-2001 and 2003-2005, data gathered from a series of surveys on the Greek surface drifting longline fleet were analyzed in order to estimate the effect of spatial, temporal, environmental and operational factors on the abovementioned species distribution and relative abundance (expressed in number of fish caught per 1000 hooks deployed). Application of generalized additive models allowed the identification of various parameters influencing catches as well as the contribution of each one. The optimum model was derived using a cross validation procedure which was based on the principal of parsimony: this principal tends to select those variables that depict more closely the actual trends than those that fit best the underlying dataset. Three sets were investigated for swordfish: (i) all specimens; (ii) individuals of size less than 100 cm of lower jaw fork length; (iii) individuals of size greater than 100 cm. The latter two partitions were considered with the intention to confirm whether maturation differentiates the animal’s behavior. The low number of blue shark catches, compared to swordfish, led us to a diverse modeling approach. Firstly, as a first step we modeled the probability of making a catch and secondly we modeled the positive catch rates separately, as in swordfish. It was deduced that spatio-temporal variables (Longitude, Latitude, Month, year) along with the fishing gear type were the most influential on the catches. Environmental constituents (Lunar index, thermal fronts, sea surface temperature) were of less importance and had only some partial effect. Distance from coast was proven to be a very significant factor, bearing a dual property as a variable: spatial and environmental. In particular, immature and mature swordfish differentiated significantly: adult specimens were more abundant to the south-east (Levantine), at the open seas preferring cooler water masses. In contrast, juveniles were more frequently observed to the north-west (Ionian), close to warmer coastal waters. Seasonally, juvenile swordfish were more proliferate in catches during the start and end of the fishing period, and during years of high water temperatures. Increased concentrations of swordfish were detected in the vicinity of thermal fronts. The lunar disc illumination affected positively all year classes, with maximum catches observed around full moon. Although no obvious decline of swordfish abundance was noticed, average size of specimens caught did decrease considerably during the study period. An extended migratory pattern was not confirmed. However, there were strong indications suggesting that Ionian and Levantine actually represent spawning grounds. Blue sharks were encountered more frequently to the north-west of our study area, during late summer in open sea cool water masses. However, the probability of locally high concentrations was higher to the south-east, during springtime, in coastal warm waters. Their presence in catches showed a constant plummeting trend through the years. Annual spring movements towards the shore are most probably related to reproductive activity. Both species are considered to be under severe fishing pressure. Annual closure of swordfish fishery during the winter months, enforced the past 20 years, represent a set of beneficial measures. Introduction of a minimum landins size for blue sharks, appears to be absolutely essential, since the declining annual abundance trends presented in this study, come as a confirmation of other analogous research