February 1st to 5th 2016
Olhão, Portugal
Tweet #MaresConference2016

Three decades of sandy commercial bivalves surveys on the South coast of Portugal: a spatio-temporal analysis

Scientific Exhibition
Natural Resources
Wednesday, February 3, 2016 -
18:30 to 20:00

Rufino, M.M. 1 Pereira, F. 2 Moura, P. 3 Vasconcelos, P. 4 Gaspar, M.B. 5

11.Instituto Português do Mar e da Atmosfera (IPMA, I.P), Rua Alfredo Magalhães Ramalho Nº 6, 1495-006 Algés, Lisboa, Portugal.2. Centre of Marine Sciences (CCMAR), University of Algarve, Campus de Gambelas, 8005-139 Faro, Portugal
2Instituto Português do Mar e da Atmosfera (IPMA, I.P), Rua Alfredo Magalhães Ramalho Nº 6, 1495-006 Algés, Lisboa, Portugal
3Instituto Português do Mar e da Atmosfera (IPMA, I.P), Rua Alfredo Magalhães Ramalho Nº 6, 1495-006 Algés, Lisboa, Portugal
4Instituto Português do Mar e da Atmosfera (IPMA, I.P), Rua Alfredo Magalhães Ramalho Nº 6, 1495-006 Algés, Lisboa, Portugal
51. Instituto Português do Mar e da Atmosfera (IPMA, I.P), Rua Alfredo Magalhães Ramalho Nº 6, 1495-006 Algés, Lisboa, Portugal.2. Centre of Marine Sciences (CCMAR), University of Algarve, Campus de Gambelas, 8005-139 Faro, Portugal

Most marine species are known to be aggregated in spatial patches. Those patches vary through time, both in biomass and location, as a response to a set of drivers. Thus, to understand the ecological, environmental and anthropological drivers of species biomass, long term spatial analysis on species dynamics are essential. In particular, target species of small-scale fisheries are further influenced by governance and market drivers, with a significant impact on its dynamics, which in turn have considerable economic and social impacts. In the coast of the Algarve (southern Portugal) an important small-scale fishery targeting three commercial bivalve species takes place on sandy bottoms. In the current work almost three decades (1986-2014) of striped venus clam (Chamalea gallina) biomass was analysed using spatio-temporal geostatistical methods. The studied area comprises 119 km2 of coast, distributed between -8.1o to -7.4oW longitude and 36.95 - 37.17oN latitude, and from 3 to 15 m depth. Species biomass was modelled in function of space and time using a spatio-temporal variogram. The model was then used to produce interpolations over the studied area and period using spatio-temporal kriging. The striped venus clam (C. gallina) showed a simple sum metric covariance spatio-temporal model, explaining 53% of covariance. The model showed that bivalve species is aggregated in 2.9 km spatial patches, whose location and dimension varied throughout the sampling period. Spatio-temporal kriging predictions maps were used to extract the main summarizing statistical features, such as the mean, maximum and variance, for all time-series stacked, thus determining the persistent favourable areas for this species and corresponding spatial patterns of variation through time. The main spatio-temporal patterns of variation were further explored using Empirical Orthogonal Function analysis (EOF), is a particular principal component analysis, which decomposes the spatio-temporal variability in the time series into principal spatial orthogonal components. This analysis permits to identify the main types of spatial distribution and temporal evolution of a given variable.The stock estimates produced and the corresponding spatial distribution will be used within the European project SAFI (http://www.safiservices.eu/; REA Grant Agreement 607155), that aims to develop indicators between species and aspects of their environment to support fisheries operations and help in management decisions, assuring a sustainable exploration of the living marine resources by using satellite-derived information.

Log in to upload your presentation