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%.

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.

Recreational shore fishing in southern Portugal: biological and socio-economic aspects and perspectives for management

Although overfishing is a concern for many fish stocks, it was for a long time only associated with commercial fishing exploitation, with less or no attention being given to the recreational fisheries. Recent research has shown however that the impact of recreational fishing on particular species can be considerable, and that the recreational harvest needs to be taken into account if fisheries are to be accurately assessed and effectively managed. In Portugal, the first recreational fishing regulations were only recently implemented. However, mirroring other European countries, regulations lacked scientific support, and specific knowledge of the activity was limited to a few studies with limited coverage. This thesis aimed to characterize the biological and socioeconomic aspects of the recreational shore angling activity in southern Portugal, to investigate whether the regulations in place were adequate and effective, and to provide recommendations for improved management and conservation of the inshore fisheries resources. A combined aerial-roving survey was conducted to gather data on fishing effort, catch, fishing trips and socioeconomic aspects (including anglers’ perceptions of regulations) of the recreational angling activity. The analysis of anglers’ catches suggested that compliance with daily bag limits was high, with less than 0.5% of creels exceeding the 10 kg angler-1 day-1 bag limit. Overall, 11.5% of the retained fishes were undersized, but non-compliance with minimum size limits was found to be high for some species (e.g. seabass, 73% undersized). In terms of the impact of recreational shore angling, the total estimated catches corresponded to less than 1% of the commercial landings for the same period (shared species). However, shore angling catches for white sea bream (Diplodus sargus) were found to be considerable, corresponding to 65% of the commercial landings (39.4% of total catch). In terms of anglers’ perceptions about the recreational fishing regulations in Portugal, the present study has shown that the majority of anglers accepted the existence of some kind of SRF regulations, but in general there was a partial or total disagreement with the recreational fishing restrictions recently put in place. Most anglers perceived themselves as not being involved in the decision-making process and claimed that some restrictions lacked a meaningful rationale (e.g. prohibition of fishing from piers/jetties). Fishers’ awareness with regard to specific aspects of the restrictions (such as the rationale for minimum size limits) was found to be very limited. During the same period, catches from sport fishing competitions were examined to test for differences with the recreational activity in terms of catches, and evaluate long term trends in catch and mean size of fish. Catches of the sport fishing competitions were found to be different from those observed for recreational fishing, being dominated by different species (e.g. garfish, mullets), and suggesting different fishing strategies of the the two types of anglers. High percentages of undersized fish were observed to be captured (and retained) during the competitions (in particular seabass, with 100% undersized), probably as a result of a single allowable minimum size (AMS) of 15 cm for all species in use in competitions. Lastly, catch and release fishing experiments were carried out to assess post-release mortality of three recreationally important species: two banded sea bream Diplodus vulgaris; black sea bream Spondyliosoma cantharus; and gilthead sea bream Sparus aurata. Post-release mortalities were found to be low (0-12%). The main predictor of mortality for Sparus aurata was anatomical hooking location, with 63% of the fishes that died being deeply hooked. The results support the release of fish, either from mandatory (e.g. minimum landing sizes) or voluntary practices. In summary, this thesis has demonstrated that the impact of recreational fishing for particular species is significant and needs to be taken into account for more effective management and stock assessment purposes. It has also highlighted several management issues that should be addressed in order to promote more adequate regulations in the future and prevent noncompliance issues. A periodic monitoring of the recreational fishing activity including all fishing modes (i.e. spear fishing, boat, and shore angling) would also be beneficial to ensure a timely knowledge on the global recreational fishing activity and support future management actions.