Made available in DSpace on 2016-08-19T18:20:38Z (GMT). No. of bitstreams: 1 Clarissa Lobato da Costa.pdf: 1392819 bytes, checksum: 9ff49d89da37676c1ed441938de43e14 (MD5) Previous issue date: 2006-09-28 / Coordenação de Aperfeiçoamento de Pessoal de Nível Superior / The artisan fishing is one of the main economic activities in the town of Viana. It involves an expressive amount of people daily practicing that activity. The objective of this study was to evaluate the artisan fishing considering the seasonality of the region, approaching the socio-economical and ecological dimensions for the sustainability of the activity. The methodology used was questionnaires and later the following of the fishery unloading to determine the destiny of the fishing product. It was also used the hydro balance calculation to consider fishing sustainable sceneries. The results demonstrated that even having a guaranteed daily food the fisherman is the one who has the lowest financial gain from the activity, living in very poor conditions. It was recorded in Viana Lake the existence of twenty-six species belonging to seventeen families and twenty-four genera. The species Plagioscion squamosissimus (south american silver croaker), Pygocentrus nattereri (red piranha) and Prochilodus nigricans (black prochilodus) occured during the whole year, being meaningful in the captures, while Schizodon vittatus (aracu) and Pimelodus omatus (bloch´s catfish) appeared more frequently in the rainy season. It can also be observed that Hoplias malabaricus (trahira) and Loricariichthys sp (loricaria), known as black fish appeared more frequently in the dry season. This paper presents as an alternative to the sustainability in the region sceneries related to damming up water through building small dams into canals. When it comes to fishing activities, though, investments in aquaculture, to promote the creation of native species in community could be an alternative as the area shows an irregular hydro period. / A pesca artesanal é uma das principais atividades econômicas no município de Viana, envolvendo um contingente expressivo de pessoas que praticam diariamente esta atividade. Este estudo teve como objetivo avaliar a pesca artesanal considerando a sazonalidade regional, abordando as dimensões social, econômica e ecológica para a sustentabilidade da atividade. Para isso, foi utilizada a metodologia de aplicação de questionários e acompanhamento de desembarque pesqueiro, determinando o destino do produto da pesca. Foi utilizado também o cálculo do balanço hídrico para propor cenários de sustentabilidade da pesca. Os resultados revelaram que o pescador, embora tenha alimentação diária garantida, é o que menos ganha em termos financeiros com a atividade, tendo condições de vida precária. Foi registrada para o Lago de Viana a ocorrência de 26 espécies pertencentes a 17 famílias e 24 gêneros. As espécies Plagioscion squamosissimus (pescada branca), Pygocentrus nattereri (piranha) e Prochilodus nigricans (curimatá) ocorreram durante o ano inteiro sendo significativas nas capturas, ao passo que Schizodon vittatus (aracu) e Pimelodus omatus (mandi) tiveram maior representatividade no período chuvoso. Observa-se também que Hoplias malabaricus (traíra) e Loricariichthys sp (viola), os chamados peixes pretos tiveram maior representatividade no período seco. Cenários relacionados ao represamento de água por meio de construção de pequenas barragens ou represamento de canais foram aqui colocados como alternativa para a sustentabilidade na região. Entretanto, no que diz respeito à atividade de aqüicultura, o investimento viabilizando a criação de espécies nativas em consórcios comunitários poderia ser uma alternativa para a localidade, a qual tem um hidroperíodo irregular.

A eutrofização de ambientes aquáticos continentais é um processo lento e natural, mas que tem se intensificado sobremaneira devido a influência das atividades antrópicas. Uma das consequências da eutrofização artificial são florações de algas cada vez mais frequentes e intensas. Alguns grupos de algas, como as cianobactérias, podem produzir toxinas que representam um risco para o homem e para a biota aquática e, por isso, sua floração merece atenção especial. O Lago Guaíba é um importante corpo hídrico do Estado do Rio Grande do Sul e o principal manancial do município de Porto Alegre. A sua bacia hidrográfica ocupa cerca de 30% da área do estado e os principais rios - Jacuí, Caí, Sinos e Gravataí - drenam áreas de intenso uso agrícola e industrial. No Lago Guaíba, as florações têm sido frequentes, principalmente na última década e representam um sério desafio ambiental. Neste contexto, o objetivo da presente pesquisa foi avaliar o potencial de identificação dos episódios de floração de algas no Lago Guaíba a partir da integração de dados radiométricos in situ adquiridos simultaneamente a dados limnológicos (clorofila-a, total de sólidos em suspensão, transparência da água, entre outros) e imagens de satélite. Para a obtenção dos espectros de reflectância foi utilizado o espectrorradiômetro portátil FieldSpec® HandHeld, com coleta em 16 pontos amostrais definidos a priori. Os trabalhos de campo ocorreram em março de 2012, maio de 2013 e abril de 2014. Os espectros foram correlacionados com medidas de variáveis limnológicas obtidas simultaneamente. Para a análise espaço-temporal das florações foram selecionadas 10 imagens produzidas pelos satélites Landsat 5 e 7 (sensores TM e ETM+) disponibilizadas pelo Serviço Geológico dos Estados Unidos (USGS), corrigidas para os efeitos da atmosfera e convertidas em reflectância de superfície. Estas imagens foram obtidas nos meses de verão e outono, entre os anos de 2005 e 2012, e coincidiram com as datas das florações identificadas pelo Departamento Municipal de Água e Esgotos (DMAE) do município de Porto Alegre. Três rotinas de processamento foram aplicadas às imagens: 1) o Índice de Vegetação por Diferença Normalizada (NDVI); 2) o Modelo Linear de Mistura Espectral (MLME) e 3) a Análise por Componentes Principais (ACP). Durante a coleta dos espectros em campo não houve florações, mas, ainda assim, estes permitiram identificar padrões de reflectância do lago em diferentes áreas e situações climáticas, bem como a influência da clorofila-a e dos sólidos em suspensão na resposta espectral da água. Os principais resultados do processamento das imagens de satélite foram: a) o NDVI é adequado para o mapeamento da abrangência e intensidade das florações e que resultados superiores a -0,2 indicam áreas de floração; b) o MLME apresentou limitações para o mapeamento das florações relacionadas à dificuldade de obtenção de bons espectros de referência diretamente sobre as imagens, mas permitiu avaliar as variações espaciais das características da água; c) a técnica da ACP ampliou as correlações dos dados das imagens com a Chl-a e as cianobactérias em comparação ao NDVI, e a classificação nãosupervisionada das principais componentes permitiu identificar as áreas (clusters) de floração, assim como áreas onde outros componentes atuavam. Os fatores que contribuíram para o estabelecimento de florações no lago foram: menor velocidade do fluxo da água, menor profundidade da coluna d´água, menor velocidade do vento e menores índices de precipitação pluviométrica, associados à maior disponibilidade de luminosidade e oferta de nutrientes. Concluí-se que as técnicas empregadas foram adequadas para o mapeamento de florações e que análises que utilizam dados de diferentes naturezas podem incrementar as metodologias existentes e abrir novas possibilidades de identificação e monitoramento de algas. / The eutrophication of freshwater ecosystems is a slow and natural process, but that has intensified greatly because of the influence of human activity. One of the consequences of artificial eutrophication are increasingly frequent and intense algae blooms. Some genera groups of algae, and cyanobacteria can produce toxins that pose a risk to humans and aquatic biota and therefore blooms deserve special attention. The Guaiba Lake is an important water body in the State of Rio Grande do Sul and the main source of Porto Alegre city. Its basin covers about 30% of the state area and major rivers - Jacuí, Caí, Sinos and Gravataí - drain areas of intense agricultural and industrial use. In the Guaiba Lake, blooms have been frequent, especially in the last decade and represent a serious environmental challenge. In this context, the objective of this study was to evaluate the potential for identification of episodes of algal bloom in the Guaiba Lake from the integration of radiometric data in situ simultaneously acquired with limnology data (chlorophyll-a, suspended solids total, water transparency, etc.) and satellite images. To obtain the reflectance spectra a portable spectroradiometer FieldSpec® HandHeld was used, with a collection of 16 sampling points defined a priori. The fieldwork took place in March 2012, May 2013 and April 2014. The spectra were correlated with limnological variables obtained simultaneously. For the spatio-temporal analysis of the blooms 10 images produced by Landsat satellites 5 and 7 (TM and ETM+ sensors) were selected and provided by United States Geological Survey (USGS), adjusted for the effects of the atmosphere and converted to surface reflectance. These images were produced in the months of summer and fall, between the years 2005 and 2012, and coincided with the dates of the blooms identified by the Municipal Department of Water and Sewerage (DMAE) in the city of Porto Alegre. Three processing routines were applied to images: 1) Normalized Difference Vegetation Index (NDVI); 2) Linear Spectral Mixture Model (LSMM) and 3) the Analysis by Principal Component (ACP). During the collection of spectra in the field no bloom was identified, but even so it was possible to identificaty reflectance standards in different areas and climatic conditions in the Lake as well as the influence of chlorophyll-a and suspended solids in the spectral response of the water. The main results of the satellite images processing were: a) the NDVI is suitable for mapping the scope and intensity of blooms and results greater than -0.2 indicate flowering areas; b) the LSMM presented limitations for mapping the blooms related to the difficulty of obtaining good reference spectra directly on the images but allowed the evaluation of spatial variations of water features; c) the technique of ACP increased correlations of image data with Chl-a and cyanobacteria compared to NDVI, and non-supervised classification of the principal component identified in the flowering cluster areas as well as areas where other components were at play. Factors that contributed to the establishment of blooms in the lake were slower rate of water flow, the less depth of the water column, lower wind speed and lower levels of rainfall, combined with the greater availability of light and nutrient supply. It was concluded that the techniques used were suitable for mapping blooms and the analysis using data from different natures can improve existing methodologies and open new possibilities for identification and monitoring of algae.

O Lago das Garças (23°39\'S, 46°37\'W) está situado dentro da área do Parque Estadual das Fontes do Ipiranga, cidade de São Paulo (SP). Com o objetivo de identificar as principais fontes de aporte de nutrientes (nitrogênio e fósforo) e a quantificação das cargas do entorno do corpo d\'água, o lago foi dividido em dois compartimentos. Foram realizadas amostragens mensais de janeiro de 1997 a junho de 1998, em dez estações de coleta. Foram feitas medidas de temperatura, pH, condutividade elétrica e concentrações de fósforo total, nitrogênio total, amônio, clorofila a, material em suspensão total, oxigênio dissolvido, em cinco profundidades no lago. Nos afluentes foram determinadas as concentrações de nitrogênio total e fósforo total. Os perfis obtidos na coluna d\'água em conjunto com os dados climatológicos e morfométricos do lago, indicaram que o aquecimento térmico pela radiação solar e a profundidade do sistema foram as principais funções de força nos processos de estratificação térmica e química. No período de estudo, as cargas de entrada no reservatório, foram de 6.519 Kg de fósforo e 37.156 Kg de nitrogênio, sendo que os afluentes representaram 99% e 34% da carga total de fósforo e nitrogênio, respectivamente. A eficiência de retenção do Lago das Garças foi de 61% para o fósforo total e 85% para o nitrogênio total. Em relação a trofia o lago foi classificado como eutrófico em todo o período de estudo. Através de equação matemática baseada na carga anual de fósforo, tempo de residência e profundidade média, foi possível estimar as concentrações de fósforo no reservatório, além de fazer simulações na redução das cargas. Considerando que a carga média de fósforo total foi de 4.891 mg/m2 ano, a simulação matemática indicou que reduções superiores a 90% da carga (valores máximos de 300 mg/m2 ano) alterariam o nível de trofia no sistema para oligotrófico/mesotrófico. / The Lago das Garças (23°39\'S, 46°37\'W) is located in Parque Estadual das Fontes do Ipiranga, São Paufo city (SP). In order to identify the main sources of nutrients (nitrogen and phosphorus) and to quantify of the nutrient loads of its tributaries, the lake was divided in two compartments. Water sample were collected monthly from January 1997 to June 1998, in ten sampling station. Measurements of temperature, pH and electric conductivity and concentrations of totaf phosphorus, total nitrogen, ammonium, chlorophyll a,suspended solids, dissolved oxigen, were done in five depth of in the lake. ln each tributarys it was established only one sampling station where it was measured the concentration of total phosphorus and nitrogen. The profiles obtained in the water column, associated with the climatological data and morphological features of the lake, showed that the thermaI heating by the solar irradiation and depth of the system were the main force functions in the processes of thermal and chemical stratification. In the period study, the entrance loads into the Iake were of 6,519 Kg of pnosphorus and 37,156 Kg of nitrogen. The tributaries represented 99% and 34% of the total load of phosphorus and nitrogen, respectively. The retention efficiency of the Lago das Garças was 61% for total phosphorus and 85% for total nitrogen. In relation to the trophic status, this lake was cfassified as eutrophic all over the studied time. Using a mathematical equation based on the annual load of phosphorus, residence time and medium depth, it was possible to evaluate the phosphorus concentration in the lake and also to verify the effect of the redution of the loads. Considering that average load of total phosphorus was of 4.891 mg/m2.year, the mathematical model indicated there more than 90% of the load (maximum 300 mg/m2.year) should be reduced, to alter the trophic status in the system for oligotrophic/mesotrophic.

Se describen una serie de métodos pasivos para la mejora de la calidad de las aguas de los lagos de mina. Se hace una descripción de la formación de los lagos de mina una vez cesan las actividades mineras.

Before 1910, the fish assemblage of Banyoles Lake (Catalonia, Spain) consisted of: eel (Anguilla anguilla), Mediterranean barbel (Barbus meridionalis), freshwater blenny (Bennius fluviatilis), three-spined stickleback (Gasterosteus aculeatus), chub (Leuciscus cephalus), and (perhaps introduced long time ago) tench (Tinca tinca). Presently, two of these species, i.e., three-spined stickleback and tench, seem to have disappeared and the remaining native fish are uncommon, mostly eel and barbel / Abans de 1910, la comunitat de peixos de l’Estany de Banyoles constava de: anguila (Anguilla anguilla), barb de muntanya (Barbus meridionalis), bavosa de riu (Bennius fluviatilis), espinós (Gasterosteus aculeatus), bagra comuna (Leuciscus cephalus) i, potser introduïda antigament, tenca (Tinca tinca). Actualment, dues d’aquestes espècies (espinós i tenca) sembla que han desaparegut i les altres autòctones són poc abundants, sobretot l’anguila i el barb de muntanya. La comunitat està molt dominada per les espècies introduïdes: d’origen nord-americà, com ara perca americana (Micropterus salmoides), peix sol (Lepomis gibbosus), gambúsia (Gambusia holbrooki); o d’origen europeu, com gardí (Scardinius erythrophthalmus), madrilleta vera (Rutilus rutilus), carpa (Cyprinus carpio) i perca (Perca fluviatilis)

This research is motivated by a drug discovery problem -- the AIDS anti-viral database from the National Cancer Institute. The objective of the study is to develop effective statistical methods to model the relationship between the chemical structure of a compound and its activity against the HIV-1 virus. And as a result, the structure-activity model can be used to predict the activity of new compounds and thus helps identify those active chemical compounds that can be used as drug candidates. Since active compounds are generally rare in a compound library, we recognize the drug discovery problem as an application of the so-called statistical detection problem. In a typical statistical detection problem, we have data {Xi,Yi}, where Xi is the predictor vector of the ith observation and Yi={0,1} is its class label. The objective of a statistical detection problem is to identify class-1 observations, which are extremely rare. Besides drug discovery problem, other applications of statistical detection include direct marketing and fraud detection. We propose a computationally efficient detection method called LAGO, which stands for "locally adjusted GO estimator". The original idea is inspired by an ancient game known today as "GO". The construction of LAGO consists of two steps. In the first step, we estimate the density of class 1 with an adaptive bandwidth kernel density estimator. The kernel functions are located at and only at the class-1 observations. The bandwidth of the kernel function centered at a certain class-1 observation is calculated as the average distance between this class-1 observation and its K-nearest class-0 neighbors. In the second step, we adjust the density estimated in the first step locally according to the density of class 0. It can be shown that the amount of adjustment in the second step is approximately inversely proportional to the bandwidth calculated in the first step. Application to the NCI data demonstrates that LAGO is superior to methods such as K nearest neighbors and support vector machines. One drawback of the existing LAGO is that it only provides a point estimate of a test point's possibility of being class 1, ignoring the uncertainty of the model. In the second part of this thesis, we present a Bayesian framework for LAGO, referred to as BLAGO. This Bayesian approach enables quantification of uncertainty. Non-informative priors are adopted. The posterior distribution is calculated over a grid of (K, alpha) pairs by integrating out beta0 and beta1 using the Laplace approximation, where K and alpha are two parameters to construct the LAGO score. The parameters beta0, beta1 are the coefficients of the logistic transformation that converts the LAGO score to the probability scale. BLAGO provides proper probabilistic predictions that have support on (0,1) and captures uncertainty of the predictions as well. By avoiding Markov chain Monte Carlo algorithms and using the Laplace approximation, BLAGO is computationally very efficient. Without the need of cross-validation, BLAGO is even more computationally efficient than LAGO.

This research is motivated by a drug discovery problem -- the AIDS anti-viral database from the National Cancer Institute. The objective of the study is to develop effective statistical methods to model the relationship between the chemical structure of a compound and its activity against the HIV-1 virus. And as a result, the structure-activity model can be used to predict the activity of new compounds and thus helps identify those active chemical compounds that can be used as drug candidates. Since active compounds are generally rare in a compound library, we recognize the drug discovery problem as an application of the so-called statistical detection problem. In a typical statistical detection problem, we have data {Xi,Yi}, where Xi is the predictor vector of the ith observation and Yi={0,1} is its class label. The objective of a statistical detection problem is to identify class-1 observations, which are extremely rare. Besides drug discovery problem, other applications of statistical detection include direct marketing and fraud detection. We propose a computationally efficient detection method called LAGO, which stands for "locally adjusted GO estimator". The original idea is inspired by an ancient game known today as "GO". The construction of LAGO consists of two steps. In the first step, we estimate the density of class 1 with an adaptive bandwidth kernel density estimator. The kernel functions are located at and only at the class-1 observations. The bandwidth of the kernel function centered at a certain class-1 observation is calculated as the average distance between this class-1 observation and its K-nearest class-0 neighbors. In the second step, we adjust the density estimated in the first step locally according to the density of class 0. It can be shown that the amount of adjustment in the second step is approximately inversely proportional to the bandwidth calculated in the first step. Application to the NCI data demonstrates that LAGO is superior to methods such as K nearest neighbors and support vector machines. One drawback of the existing LAGO is that it only provides a point estimate of a test point's possibility of being class 1, ignoring the uncertainty of the model. In the second part of this thesis, we present a Bayesian framework for LAGO, referred to as BLAGO. This Bayesian approach enables quantification of uncertainty. Non-informative priors are adopted. The posterior distribution is calculated over a grid of (K, alpha) pairs by integrating out beta0 and beta1 using the Laplace approximation, where K and alpha are two parameters to construct the LAGO score. The parameters beta0, beta1 are the coefficients of the logistic transformation that converts the LAGO score to the probability scale. BLAGO provides proper probabilistic predictions that have support on (0,1) and captures uncertainty of the predictions as well. By avoiding Markov chain Monte Carlo algorithms and using the Laplace approximation, BLAGO is computationally very efficient. Without the need of cross-validation, BLAGO is even more computationally efficient than LAGO.

Un mayor conocimiento de la hidrodinámica del lago Titicaca en el Perú, es la discusión del presente trabajo. El estudio se realiza para una escala de tiempo de ciclo diario. Usamos ELCOM (Estuary, Lake and Coastal Ocean Model), un modelo numérico en tres dimensiones para realizar simulaciones en un periodo comprendido entre el primero y quinto día del mes de enero del 2005. Para iniciar las simulaciones se obtuvo la batimetría del lago Titicaca a partir de una carta de sondajes. Se presentaron cuatro escenarios diferentes, cambiando las condiciones iniciales de temperatura y considerando el efecto coriolis. El modelo resuelve las ecuaciones hidrodinámicas de Navier-Stokes con aproximaciones mediante el método semi-implícito de diferencias finitas, para un flujo incompresible, barotrópico, rotacional y estratificado por temperatura. Las condiciones meteorológicas, entre ellas, el viento tiene un efecto en la circulación y la estratificación en capas superficiales del lago. Las corrientes en el lago Grande que ocupa el mayor volumen de agua, se debe principalmente al forzamiento del fluido debido a la temperatura. Los afluentes y efluentes no se consideraron en las simulaciones realizadas en el presente trabajo. La fuerza de coriolis es importante en la dinámica del lago Grande por que ayuda a mezclar las capas estratificadas térmicamente.

Cholera remains a threat to public health in many developing countries, including Mozambique. Although the disease is easily preventable by practices of hygiene and sanitation, cases are reported in the country every year, as for example in the Lago district in 2015. This qualitative research project set out to explore in what ways the promotion of hygiene and sanitation practices on district level in Mozambique is carried out. Therefore, actors, messages and channels involved in these communication processes were explored via a field study in Lago and a review of relevant literature. Subsequently, the results of the field study and literature review were analysed by applying the concepts of one-way and two-way communication which are part of public relations theory. This analytical framework allowed the researchers to fill a gap identified in the existing literature about hygiene and sanitation promotion, which did not seem to include communication theories linked to public relation practices when it came to hygiene and sanitation promotion in developing countries as a method to prevent cholera. It has been found that the one-way communication approach towards the public was successful in handling the recent cholera outbreak of 2015, however, the approach is not substantial and should be improved into a two-way communication approach, which would allow the local population to express their needs in hygiene and sanitation, as well as their capabilities to implement change in these matters. Simultaneously, a lack of resources within the district authorities involved in hygiene and sanitation promotion seems to encourage one-way communication towards the public from their side, as two-way communication would demand further resources for research into the above mentioned needs and capabilities of communities.

La contaminación de los ambientes acuáticos conlleva a la afectación directa de las especies que se desarrollan en estos medios, tal es así que las especies endémicas son las que sufren más estos efectos en niveles aún no determinados, pero que definitivamente afectan su población y subsistencia. Esto motivó el presente trabajo de investigación, donde se tuvo como objetivo determinar la concentración letal media (CL50-96) y efecto histopatológico del sulfato de cobre en alevinos de Orestias luteus “carachi amarillo”. El trabajo se realizó en el Laboratorio de Acuicultura y el Laboratorio de Muestras Biológicas de la Dirección de Recursos Hidrobiológicos del Proyecto Especial Binacional Lago Titicaca (PELT), ubicado en Muelle Barco S/N, distrito de Chucuito, Puno y en el Laboratorio de Histopatología de la Facultad de Medicina de la Universidad Nacional del Altiplano, durante los meses de Abril a Noviembre del año 2010. Un total de 280 alevinos del Orestias luteus fueron expuestos a concentraciones de 20 µg/l; 10 µg/l; 5 µg/l; 2,5 µg/l; 1,25 µg/l y 0,63 µg/l de sulfato de cobre y un control durante un periodo de 96 horas. Se observó una mortalidad mínima de12 alevinos en la concentración de 0,63 µg/l de sulfato de cobre, mientras que el máximo de mortalidad fue 40 alevinos al final de las primeras 72 horas en las concentraciones de 20 µg/l y 10 µg/l de sulfato de cobre. La Concentración Letal Media (CL50-96) para el sulfato de cobre mediante el método Probit fue 1,65 µg/l al final de todo el experimento. Posteriormente y para determinar los efectos histopatológicos en los alevinos, estos fueron procesados, realizando cortes histológicos a diferentes niveles anatómicos y teñidos con la técnica de hematoxilina-eosina y conservados de manera permanente utilizando Bálsamo de Canadá. El sulfato de cobre afectó la estructura ocular, produciendo una atrofia celular del cristalino y el iris en la concentración de 20 µg/l, lo mismo en la estructura del iris a 10 µg/l en comparación al control.