NICHE MODELING OF COASTAL MEDITERRANEAN FISHES UNDER CLIMATE CHANGE SCENARIO SIMULATED BY THE NEW NEMOMED8 MODEL: EXPECTED BIODIVERSITY LOSS AND SPECIES TURNOVER 1,4

1

2

3

Camille Albouy , François Guilhaumon ,Frida Ben Rais Lasram , Samuel Somot , 3 1 4 Roland Aznard , François Le Loc'h and David Mouillot 1 Laboratoire Ecosystèmes Lagunaires UMR 5119, Université Montpellier 2, cc 093, Place E. Bataillon, 34095 Montpellier Cedex 5, France 2 Laboratoire Ecosystèmes et Ressources Aquatiques UR03AGRO1, Institut National Agronomique de Tunisie, 43 avenue Charles Nicolle, 1082 Tunis, Tunisia 3 Météo-France, Centre National de Recherches Météorologiques CNRM-GAME, 42 avenue Gaspard Coriolis, 31057, Toulouse Cedex, France 4 Centre de recherche Halieutique Méditerranéenne et tropicale UMR 212, avenue Jean Monnet BP171,34203 Sète Cedex,France.

The Mediterranean Sea is one of the most impacted areas by human activities (Halpern et al. 2008) while being indubitably a marine biodiversity and a climate change hotspot (Coll et al. 2010). The biodiversity of the Mediterranean Sea is exceptional relative to its water volume: it represents only 0.32% of the global oceanic volume but holds 4-18% of known marine species and encompass 8.8% of endemism. In the present study, we aim at forecasting the potential impacts of climate warming on coastal fish diversity. Modelling Sea Surface Temperature (SST) Water temperature is recognized as one of the main driver shaping fish species distributions (Cheung et al. 2009). We used SST values averaged from an early period as a baseline to calibrate the species distribution models (SDMs): daily SST values were taken for the period 1961-1980 from the NEMOMED8 model. Projected SST values were obtained from the same model, based on the IPCC A2 scenario. We studied two time periods : 2040-2059 and 2080-2099. Present-day and projected temperature data were summarized using the K-means method and 8 synthetic variables were obtained.

A)

B)

Differences in SST between observed data and two time periods ( 2040-2059 A ,2080-2099 B) predicted on the continental shelf of the mediterranean sea, based on NEMOMED8 and the A2 IPCC scenario.

st Greatest differences in mean SST between 1961-1980 and the middle of the 21 century were observed in the Aegean Sea. At the end of the century, the Aegean sea also experienced the greatest change in mean SST. Several areas also showed noticeable increase in mean SST, namely the Adriatic Sea, the Levantine Basin and the Catalan sea .

Modelling species distributions Our dataset contained the occurrences of 339 coastal fish species. We corrected the current geographical distribution of each species according to the minimum and maximum depths at which species are encountered. We used 7 presence/absence statistical methods (BIOMOD, Thuiller et al. 2009), to project the climatic niche of each species for the two considered time periods 2040-2059 and 2080-2099. Specifically, we employed an ensemble forecasting approach (Thuiller et al. 2009). Differences between observed and the middle of the 21st century projected richness, showed that 64.3% of the continental shelf cells lost species. Area with greatest species loss was observed in the Western basin. Comparison between the present-day period and the end of the 21st showed that 6211 cells lost species. Areas with greatest species loss were the Western Mediterranean sea, a part of the Adriatic sea.

Differences in fish species richness between the current and two projected time periods ( 2040-2059 A, 2080-2099 B) on the continental shelf of the Mediterranean sea.

β diversity: β-diversity is here considered as a measure of dissimilarity between two time periods for each cell. The total amount of beta-diversity may be estimated using a pairwise dissimilarity metric (βSor ) based on the Sørensen index and partitioned into two components, the spatial turnover (βSIM) and the nestedness components (βNes).

Partitionning of beta diversity between observed and the first predicted period. A) Soresen diversity; B) Simpson diversity, C) Nesteness diversity.

The site 1 is completely nested because in time period 2 species are subsets of species in time period 1. Site 2 present 3 species in time period 1 and 2 new species in period 2,displaying a pattern of spatial turnover. Site 3 present both patterns, because in time period 2, there is a new species (5) and the 2 others are subsets of species in time period 1 . See Baselga (2010) for more details.

Highest values of beta diversity were found in the Alboran and Adriatic seas for which changes in fish assemblage composition mostly resulted from nestedness and spatial species turnover, respectively. Overall, changes in fish assemblage composition in the the South were almost completely caused by a pattern of species loss only (nestedness) whereas species replacement dominated in the Northern part of the Mediterranean sea.

Bibliography: Baselga A (2010) Partitioning the turnover and nestedness components of beta diversity. Global Ecology and Biogeography 19:134-143 Coll M, Piroddi C, Steenbeek J, Kaschner K, Ben Rais Lasram F et al. (2010) The biodiversity of the Mediterranean Sea: estimates, patterns, and threats. PLoS One 5: e11842. Cheung WWL, Lam VWY, Sarmiento JL, Kearney K, Watson R et al. (2009) Projecting global marine biodiversity impacts under climate change scenarios. Fish and Fisheries 10: 235-251 Halpern BS, Walbridge S, Selkoe KA, Kappel CV, Micheli F et al. (2008) A global map of human impact on marine ecosystems. Science 319: 948-952. Thuiller W, Lafourcade B, Engler R, Araujo MB (2009) BIOMOD – a platform forensemble forecasting of species distributions. Ecography, 32, 369–373