On-line operation of an intelligent seismic detector

This study describes the on-line operation of a seismic detection system to act at the level of a seismic station providing similar role to that of a STA /LTA ratio-based detection algorithms. The intelligent detector is a Support Vector Machine (SVM), trained with data consisting of 2903 patterns extracted from records of the PVAQ station, one of the seismographic network's stations of the Institute of Meteorology of Portugal (IM). Records' spectral variations in time and characteristics were reflected in the SVM input patterns, as a set of values of power spectral density at selected frequencies. To ensure that all patterns of the sample data were within the range of variation of the training set, we used an algorithm to separate the universe of data by hyper-convex polyhedrons, determining in this manner a set of patterns that have a mandatory part of the training set. Additionally, an active learning strategy was conducted, by iteratively incorporating poorly classified cases in the training set. After having been trained, the proposed system was experimented in continuous operation for unseen (out of sample) data, and the SVM detector obtained 97.7% and 98.7% of sensitivity and selectivity, respectively. The same type of ANN presented 88.4 % and 99.4% of sensitivity and selectivity when applied to data of a different seismic station of IM. © 2013 Springer-Verlag Berlin Heidelberg.

Short-term hooking mortality of three marine fish species (Sparidae) caught by recreational angling in the south Portugal

Short-term hooking mortality was evaluated for three sparid species [Diplodus vulgaris (Geoffroy Saint-Hilaire), Spar-us aurata L. and Spondyliosoma cantharus (L.)] in the Algarve, south Portugal. Fishes were caught from the shore during October 2009 at a fish farm reservoir (Ria Formosa), using three different hook sizes. The relationships between hooking mortality and seven independent variables were analyzed using logistic regression models. In all,384 fishes representing the three target species were caught during the angling sessions. The most caught species was S. cantharus (n = 181; 100% undersized), followed by S. aurata (n = 137; 89% undersized) and D. vulgaris (n = 66; 97% undersized). Mortalities ranged between 0% for D. vulgaris and 12% for S. aurata (S. cantharus, 3%). For S. aurora, anatomical hooking location was the main predictor of mortality, with 63% of the fishes that died being deeply hooked. Our results support the current mandatory practices of releasing undersized fish for the studied species, given the low post-release mortality rates observed. (C) 2010 Elsevier B.V. All rights reserved.

A neural network seismic detector

This experimental study focuses on a detection system at the seismic station level that should have a similar role to the detection algorithms based on the ratio STA/LTA. We tested two types of neural network: Multi-Layer Perceptrons and Support Vector Machines, trained in supervised mode. The universe of data consisted of 2903 patterns extracted from records of the PVAQ station, of the seismography network of the Institute of Meteorology of Portugal. The spectral characteristics of the records and its variation in time were reflected in the input patterns, consisting in a set of values of power spectral density in selected frequencies, extracted from a spectro gram calculated over a segment of record of pre-determined duration. The universe of data was divided, with about 60% for the training and the remainder reserved for testing and validation. To ensure that all patterns in the universe of data were within the range of variation of the training set, we used an algorithm to separate the universe of data by hyper-convex polyhedrons, determining in this manner a set of patterns that have a mandatory part of the training set. Additionally, an active learning strategy was conducted, by iteratively incorporating poorly classified cases in the training set. The best results, in terms of sensitivity and selectivity in the whole data ranged between 98% and 100%. These results compare very favorably with the ones obtained by the existing detection system, 50%.

A divulgação de informação do governo societário e o seu efeito no mercado de capitais. Evidência no PSI20 - Euronext Lisbon

Dissertação de Mestrado, Finanças Empresariais, Faculdade de Economia, Universidade do Algarve, 2016

Using a modified Nordmore grid for by-catch reduction in the Portuguese crustacean-trawl fishery

By-catch and discards are a common problem to all fisheries. Recent studies of the Portuguese crustacean-trawl fishery show that discards average 70% of the total catch. This situation poses a threat not only to the management of target and commercial fish by-catch species stocks, but also in terms of the ecosystem impact of the discarded species whose survival is generally low. Previous attempts to address this problem by using oblique separator mesh panels associated with square-mesh windows, or square-mesh windows alone, provided encouraging results, but these are unlikely to have commercial applicability at present. Therefore, we tested the efficiency of a modified Nordmore grid in excluding some of the most captured non-commercial by-catch species, such as the blue whiting, Micromesistius poutassou, and the boarfish, Capros aper, evaluating simultaneously the losses of the crustacean target species (rose shrimp, Parapenaeus longirostris, Norway lobster, Nephrops norvegicus and red shrimp, Aristeus antennatus) and commercially valuable fish by-catch. A total of 41 valid hauls were carried out during 2001, and 15 in 2002 on board the R/V "Noruega". Results from both years varied considerably, with higher by-catch exclusion rates, in weight, registered in the second survey (73-74 and 48-63% for blue whiting and boarfish, respectively). The corresponding percentage reduction for target species was 8-15% for the Norway lobster, 4-9% for rose shrimp and 7-10% for red shrimp. The level of exclusion of the non-commercial by-catch is encouraging, although short-term loss of Norway lobster, where the escapees comprised larger individuals, raises some concern with regards to fishers' acceptance of mandatory use of grids. (C) 2004 Elsevier B.V. All rights reserved.