Date

2014

Author

Sakınan, Serdar

Metadata

Show full item record

Item Usage Stats

257
views

198
downloads

Cite This

Stocks of traditionally targeted fish species are in decline in the north eastern Levant Sea due to overfishing during recent decades, while catch statistics show a simultaneous increase in landings of commercially less valuable small pelagic species. Moreover, the number of species in the eastern Mediterranean is increasing due to Lessepsian migration and the response of native species to the continuous migration of lessepsian species is in question. Additionally, warming associated with climate change is expected to affect populations of small pelagic fish. This hypothesis may be of particular importance in the north eastern Mediterranean as it is the warmest basin of the Mediterranean. In this study, the aim was to determine the environmental factors that may have an effect on the distribution of small pelagic fish populations and to predict their possible response to future climate changes. This study took place within the continental shelf area between Taşucu and the Turkish - Syrian border (33.8º E - 36.2º E). The data were collected between 2009 and 2011 during five hydro-acoustic surveys conducted for the project 108O566 (supported by Tubitak) by METU - IMS during June and October each year. Data consist of fisheries acoustics records, trawl samplings, CTD casts and satellite data. The overall workflow basically involved four phases,: I) As this was the very first attempt to study small pelagic fishes in the NE Mediterranean by a hydroacoustic method, much emphasis was placed on the development of a methodology to identify the species in the pelagic fish fauna; II) mapping the distribution of acoustically identified fishes; III) examining their relationship with environmental factors and finally; IV) predicting the effect of future warming based on habitat suitability. iv In the analysis part; the workflow starts with scrutinization which involved; application of calibration parameters to raw data, elimination of noise, and detection and removal of the sea bottom and surface reverberation layers. Subsequently the fish schools and their descriptors were extracted and meaningful school parameters were selected for statistical analysis. Additionally, trawl hauls and CTD data were analysed. School forming typologies were then utilised to distinguish particular fish schools. Finally, information from trawl hauls and clustering of acoustic fish school records were combined and analysed using supervised classification based on artificial neural network algorithm. Species distribution maps were then evaluated using hydrographic data and satellite data in order to determine the factors affecting fish distributions. Generalized additive models (GAMs) were used to investigate the observed relationships and these models were then applied to the construction of habitat suitability maps and used to predict the effect of the warming based on climate change scenarios. The results showed that the most dominant species in the region was Sardinella aurita while the distribution and density of the species Sardina pilchardus, Dussumieria elopsoides, Etrumeus teres, Trachurus trachurus and Trachurus mediterraneus also exhibited noticeable biomass in the region. Engraulis encrasicolus, Herklotsichthys punctatus and Sardinella madarensis were found to be less abundant. Thermal stratification was found to be an important determining factor of the distribution of different species as this creates two different pelagic habitats, inhabited by the species according to their temperature preference. The species with warm water preference that occupied a temperature range between 24ºC -27º C were mainly concentrated in the regions around river plumes and eutrophic - shallow waters affected by urban runoff. The distribution of this group was mainly associated with high chlorophyll concentration. In particular, the schools of S. aurita juveniles found during October preferred the most productive regions between the Seyhan and Ceyhan River inflows. This suggested that this area is an essential habitat for small pelagic fishes supporting the growth of the young fishes. The species inhabiting cooler water displayed two different temperature range preferences, 17 ºC -19 ºC and 19 ºC -21 ºC where the thermocline constituted an upper boundary as the surface layer is substantially less favorable. Due to such thermal limitations the species were located at some distance from the coast but shoreward of the 100 m depth contour. The results of the predictions performed by the GAM analysis suggested that the most important area with regards to habitat suitability for dominant species particularly Sardinella aurita, was the inshore parts of the region around the Mersin Bay and the Gulf of Iskenderun along the north eastern corner of the v Levantine Sea. In addition for the entire Levantine Sea the predictions based on GAM analysis indicated that the warm waters of the Nile delta region with its wide shelf area and high chlorophyll concentration exhibited the best environmental conditions. Finally the scenarios tested to foresee the possible impacts of temperature rise demonstrated that warming may result in a remarkable decline in Sardinella aurita populations which are already at its marginal temperature limits.

Subject Keywords

Fishes, Fishes, Fishes, Fishes, Fishes, Fishes

URI

http://etd.lib.metu.edu.tr/upload/12617029/index.pdf
https://hdl.handle.net/11511/23470

Collections

Graduate School of Marine Sciences, Thesis