Gilberto C. Pereira

Knowledge discovering for coastal waters classification

Since almost all anthropogenic activities ultimately affect the coastal waters, access properties and processes in this environment is the major issue in decision making and system management. Particularly, seasonal patterns are not clear in tropical areas, therefore, requiring environmental classification. The knowledge of long-term biogenic element dynamics, the biological response, and the selection of indicators connecting lower and higher trophic levels have became a real need for the sustainable management of marine resources. Under this scenario, this paper uses a machine-learning approach to determine the ecological status of coastal waters based on patterns of occurrence of meroplankton larvae of epibenthic fauna and its relationship with other environmental variables. The case studied is the upwelling influenced bay at Cabo Frio Island (Rio de Janeiro - Brazil) because this location has been suffering with anthropogenic impact. Models of crisp and fuzzy rules have been tested as classifiers. Results show it is possible to access hidden patterns of water masses within a set of association rules.

Data mining for environmental analysis and diagnostic: a case study of upwelling ecosystem of Arraial do Cabo

A zona costeira brasileira apresenta grande extensao e variedade de ambientes. Contudo, pouco se sabe sobre sua diversidade biologica e o funcionamento dos ecossistemas. Como mudancas ambientais sao constantes, e muito importante distinguir entre variabilidade natural e antropica. Nesse cenario, o objetivo deste trabalho e apresentar a metodologia para o desenvolvimento de um Sistema Inteligente de Gerenciamento Integrado do Ecossistema Costeiro (SIGIEC) capaz de acessar o nivel de qualidade e saude ambiental atraves do conceito de Integridade Biologica. Foram usadas series temporais de dez anos de parâmetros fisicos, quimicos e biologicos para extrair conhecimento e gerar modelos de regras de associacao para classificar sete diferentes tipos de condicoes ambientais, analisadas atraves da diversidade biologica e um novo indice trofico (PLIX). Redes neurais artificiais foram otimizadas por algoritmos geneticos para fazer predicoes desses indices e apresenta-se um diagnostico ambiental baseado na analise dos mecanismos de controle da topologia, estabilidade e propriedades do comportamento complexo de redes alimentares.

Virioplankton Abundance in Trophic Gradients of an Upwelling Field

This work correlates time series of biological and physical variables to the marine viruses across trophic gradients within Arraial do Cabo upwelling system, Southeast of Brazil. The objective is to investigate the major controlling factors of virioplankton dynamics among different water masses. It was used an in situ and ex situ flow cytometry for accessing the plankton community. Viruses were highly correlated to bacteria and phytoplankton, but although the lack of direct correlation with physicals, upwelling turned out to be the main contributing factor to the highest values of viral abundance and virus:bacterial ratio. Our data suggest that the lowest temperature of upwelled South Atlantic Central Waters would help to maintain a high viral abundance and higher temperatures of Coastal and Tropical Waters might be another ecological niche allowing the co-existence.

Yearly variation of bacterial production in the Arraial do Cabo protection area (Cabo Frio upwelling region): an evidence of anthropogenic pressure

Arraial do Cabo is where upwelling occurs more intensively on the Brazilian coast. Although it is a protection area it suffers anthropogenic pressure such as harbor activities and sporadic sewage emissions. Short-time studies showed a high variability of bacterial production (BP) in this region but none of them evaluated BP during long periods in a large spatial scale including stations under different natural (upwelling and cold fronts) and anthropogenic pressures. During 2006, we sampled surface waters 10 times (5 in upwelling and 5 in subsidence periods) in 8 stations and we measured BP, temperature as well as the concentrations of inorganic nutrients, pigments and particulate organic matter (POM). BP was up to 400 times higher when sewage emissions were observed visually and it had a positive correlation with ammonia concentrations. Therefore, in 2007, we did two samples (each during upwelling and subsidence periods) during sewage emissions in five stations under different anthropogenic pressure and we also measured particles abundance by flow cytometry. The 12 samples in the most impacted area confirmed that BP was highest when ammonia was higher than 2 μM, also reporting the highest concentrations of chlorophyll a and suspended particles. However, considering all measured variables, upwelling was the main disturbing factor but the pressure of fronts should not be neglected since it had consequences in the auto-heterotrophic coupling, increasing the concentrations of non fluorescent particles and POM. Stations clustered in function of natural and anthropogenic pressures degrees and both determined the temporal-spatial variability.

Combining in situ flow cytometry and artificial neural networks for aquatic systems monitoring

In order to produce a system to automatically identify field water samples, it is essential to cover the entire spectrum of biological variation that a species can be found in the natural environments. This information must be available for modeling within specific training data sets. Thus, the one of the objectives of this work is to build a set of flow cytometric data containing this information in order to develop artificial neural network models that learn the patterns of biological variation induced by some environmental parameters. The second goal is to test the model in near real time recognition of phytoplankton. Twelve isolated groups were assayed in order to define their optical signature boundaries. Our results show high performance of a Radial Basis Function Neural Network in the test data set and its recognition and enumeration capability when assessing field data. It also suggests that it would be better to use a more generalist model for the different phytoplankton groups and more specialized networks to deal with specific organisms within a taxon. A discussion about the use of this type of model in monitoring programs is presented.

Bacterial and Archaeal Communities Variability Associated with Upwelling and Anthropogenic Pressures in the Protection Area of Arraial do Cabo (Cabo Frio region - RJ)

Upwelling systems contain a high diversity of pelagic microorganisms and their composition and activity are defined by factors like temperature and nutrient concentration. Denaturing gradient gel electrophoresis (DGGE) technique was used to verify the spatial and temporal genetic variability of Bacteria and Archaea in two stations of the Arraial do Cabo coastal region, one under upwelling pressure and another under anthropogenic pressure. In addition, biotic and abiotic variables were measured in surface and deep waters from three other stations between these stations. Six samplings were done during a year and adequately represented the degrees of upwelling and anthropogenic pressures to the system. Principal Component Analysis (PCA) showed negative correlations between the concentrations of ammonia and phosphorous with prokaryotic secondary production and the total heterotrophic bacteria. PCA also showed negative correlation between temperature and the abundance of prokaryotic cells. Bacterial and archaeal compositions were changeable as were the oceanographic conditions, and upwelling had a regional pressure while anthropogenic pressure was punctual. We suggest that the measurement of prokaryotic secondary production was associated with both Bacteria and Archaea activities, and that substrate availability and temperature determine nutrients cycling.

Coastal environmental management by fuzzy system modeling

Environmental data often need to be analyzed in order to obtain information necessary for environmental management decision. The main task today is to shift what is natural and atrophic variability and the assessment of trophic status to forecast the future ecosystem behavior. Sometimes it is necessary to use the data to build a model of the environmental processes that we want to manage. In other cases, we must identify and understand the interrelationships of different physical, chemical and biological parameters. The case studied is the algal community growth in coastal upwelling area of Cabo Frio Island at Rio de Janeiro state, southeastern of Brazil as a bioindicator since this place is becoming a operational support base of a oil drilling companies. We used a machine learning approach to elicit regularities/spl bsol/dependences that include both numerical and logical conditions.

Using in situ flow cytometry images of ciliates and dinoflagellates for aquatic system monitoring

Short-period variability in plankton communities is poorly documented, especially for variations occurring in specific groups in the assemblage because traditional analysis is laborious and time-consuming. Moreover, it does not allow the high sampling frequency required for decision making. To overcome this limitation, we tested the submersible CytoSub flow cytometer. This device was anchored at a distance of approximately 10 metres from the low tide line at a depth of 1.5 metres for 12 hours to monitor the plankton at a site in the biological reserve of Barra da Tijuca beach, Rio de Janeiro. Data analysis was performed with two-dimensional scatter plots, individual pulse shapes and micro images acquisition. High-frequency monitoring results of two interesting groups are shown. The abundance and carbon biomass of ciliates were relatively stable, whereas those from dinoflagellates were highly variable along the day. The linear regression of biovolume measures between classical microscopy and in situ flow cytometry demonstrate high degree of adjustment. Despite the success of the trial and the promising results obtained, the large volume of images generated by the method also creates a need to develop pattern recognition models for automatic classification of in situ cytometric images.

Exploring complex networks in the plankton

Aiming to contribute to environmental management activities and coastal ecosystems preservation, this paper presents a set of computational and methodological procedures to networks analysis of interactions between marine plankton organisms from cytometric data. As ecological interactions have dynamic character, fuzzy multigraphs were designed to capture trophic relations of competition, predation and herbivory. Complex networks indexes were applied to the topologies found revealing their structural patterns while the detection of subnetworks was made by an ant colony algorithm. The results show compliance with the literature. This approach allows access to the environmental status in a timely manner for the managerial decision making.

Centrality and Network Analysis in a Natural Perturbed Ecosystem

The aim of this work is to gain knowledge on the interactions between the chlorophyll-a and nine meroplankton larvae of epibenthonic fauna. The studied case is the Arraial do Cabo upwelling system, Southeastern of Brazil, which provides different environmental conditions. To assess this information a network approach based in probability estimative was used. Comparisons among the generated graphs are made in the light of different water masses, application of Shannon biodiversity index, and the closeness and betweenness centralities measurements. Our results show the main pattern among different water masses and how the core organisms belonging to the network skeleton are correlated to the main environmental variable. We conclude that the approach of complex networks is a promising tool for environmental diagnostic.