Rigorous modeling of the spatial species distributions is critical in biogeography, conservation, resource management, and assessment of climate change. The goal of chapter 2 of this dissertation was to evaluate the potential of using historical samples to develop high-resolution species distribution models (SDMs) of stream fishes of the United States. I explored the spatial transferability and temporal transferability of stream–fish distribution models in chapter 3 and chapter 4 respectively. Chapter 2 showed that the discrimination power of SDMs for 76 non-game fish species depended on data quality, species' rarity, statistical modeling technique, and incorporation of spatial autocorrelation. The area under the Receiver-Operating-Characteristic curve (AUC) in the cross validation tended to be higher in the logistic regression and boosted regression trees (BRT) than the presence-only MaxEnt models. AUC in the cross validation was also higher for species with large geographic ranges and small local populations. Species prevalence affected discrimination power in the model training but not in the validation. In chapter 3, spatial transferability of SDMs was low for over 70% of the 21 species examined. Only 24% of logistic regression, 12% of BRT, and 16% of MaxEnt had AUC > 0.6 in the spatial transfers. Friedman's rank sum test showed that there was no significant difference in the performance of the three modeling techniques. Spatial transferability could be improved by using spatial logistic regression under Lasso regularization in the training of SDMs and by matching the range and location of predictor variables between training and transfer regions. In chapter 4, testing of temporal SDM transfer on independent samples resulted in discrimination power of the moderate to good range, with AUC > 0.6 for 80% of species in all three types of models. Most cool water species had good temporal transferability. However, biases and misspecified spread occurred frequently in the temporal model transfers. To reduce under- or over-estimation bias, I suggest rescaling the predicted probability of species presence to ordinal ranks. To mitigate inappropriate spread of predictions in the climate change scenarios, I recommended to use large training datasets with good coverage of environmental gradients, and fine-tune predictor variables with regularization and cross validation. / Ph. D.

<p>Human use of the oceans is increasingly in conflict with conservation of endangered species. Methods for managing the spatial and temporal placement of industries such as military, fishing, transportation and offshore energy, have historically been post hoc; i.e. the time and place of human activity is often already determined before assessment of environmental impacts. In this dissertation, I build robust species distribution models in two case study areas, US Atlantic (Best et al. 2012) and British Columbia (Best et al. 2015), predicting presence and abundance respectively, from scientific surveys. These models are then applied to novel decision frameworks for preemptively suggesting optimal placement of human activities in space and time to minimize ecological impacts: siting for offshore wind energy development, and routing ships to minimize risk of striking whales. Both decision frameworks relate the tradeoff between conservation risk and industry profit with synchronized variable and map views as online spatial decision support systems.</p><p>For siting offshore wind energy development (OWED) in the U.S. Atlantic (chapter 4), bird density maps are combined across species with weights of OWED sensitivity to collision and displacement and 10 km2 sites are compared against OWED profitability based on average annual wind speed at 90m hub heights and distance to transmission grid. A spatial decision support system enables toggling between the map and tradeoff plot views by site. A selected site can be inspected for sensitivity to a cetaceans throughout the year, so as to capture months of the year which minimize episodic impacts of pre-operational activities such as seismic airgun surveying and pile driving.</p><p>Routing ships to avoid whale strikes (chapter 5) can be similarly viewed as a tradeoff, but is a different problem spatially. A cumulative cost surface is generated from density surface maps and conservation status of cetaceans, before applying as a resistance surface to calculate least-cost routes between start and end locations, i.e. ports and entrance locations to study areas. Varying a multiplier to the cost surface enables calculation of multiple routes with different costs to conservation of cetaceans versus cost to transportation industry, measured as distance. Similar to the siting chapter, a spatial decisions support system enables toggling between the map and tradeoff plot view of proposed routes. The user can also input arbitrary start and end locations to calculate the tradeoff on the fly.</p><p>Essential to the input of these decision frameworks are distributions of the species. The two preceding chapters comprise species distribution models from two case study areas, U.S. Atlantic (chapter 2) and British Columbia (chapter 3), predicting presence and density, respectively. Although density is preferred to estimate potential biological removal, per Marine Mammal Protection Act requirements in the U.S., all the necessary parameters, especially distance and angle of observation, are less readily available across publicly mined datasets.</p><p>In the case of predicting cetacean presence in the U.S. Atlantic (chapter 2), I extracted datasets from the online OBIS-SEAMAP geo-database, and integrated scientific surveys conducted by ship (n=36) and aircraft (n=16), weighting a Generalized Additive Model by minutes surveyed within space-time grid cells to harmonize effort between the two survey platforms. For each of 16 cetacean species guilds, I predicted the probability of occurrence from static environmental variables (water depth, distance to shore, distance to continental shelf break) and time-varying conditions (monthly sea-surface temperature). To generate maps of presence vs. absence, Receiver Operator Characteristic (ROC) curves were used to define the optimal threshold that minimizes false positive and false negative error rates. I integrated model outputs, including tables (species in guilds, input surveys) and plots (fit of environmental variables, ROC curve), into an online spatial decision support system, allowing for easy navigation of models by taxon, region, season, and data provider.</p><p>For predicting cetacean density within the inner waters of British Columbia (chapter 3), I calculated density from systematic, line-transect marine mammal surveys over multiple years and seasons (summer 2004, 2005, 2008, and spring/autumn 2007) conducted by Raincoast Conservation Foundation. Abundance estimates were calculated using two different methods: Conventional Distance Sampling (CDS) and Density Surface Modelling (DSM). CDS generates a single density estimate for each stratum, whereas DSM explicitly models spatial variation and offers potential for greater precision by incorporating environmental predictors. Although DSM yields a more relevant product for the purposes of marine spatial planning, CDS has proven to be useful in cases where there are fewer observations available for seasonal and inter-annual comparison, particularly for the scarcely observed elephant seal. Abundance estimates are provided on a stratum-specific basis. Steller sea lions and harbour seals are further differentiated by ‘hauled out’ and ‘in water’. This analysis updates previous estimates (Williams & Thomas 2007) by including additional years of effort, providing greater spatial precision with the DSM method over CDS, novel reporting for spring and autumn seasons (rather than summer alone), and providing new abundance estimates for Steller sea lion and northern elephant seal. In addition to providing a baseline of marine mammal abundance and distribution, against which future changes can be compared, this information offers the opportunity to assess the risks posed to marine mammals by existing and emerging threats, such as fisheries bycatch, ship strikes, and increased oil spill and ocean noise issues associated with increases of container ship and oil tanker traffic in British Columbia’s continental shelf waters.</p><p>Starting with marine animal observations at specific coordinates and times, I combine these data with environmental data, often satellite derived, to produce seascape predictions generalizable in space and time. These habitat-based models enable prediction of encounter rates and, in the case of density surface models, abundance that can then be applied to management scenarios. Specific human activities, OWED and shipping, are then compared within a tradeoff decision support framework, enabling interchangeable map and tradeoff plot views. These products make complex processes transparent for gaming conservation, industry and stakeholders towards optimal marine spatial management, fundamental to the tenets of marine spatial planning, ecosystem-based management and dynamic ocean management.</p> / Dissertation

O Eragrostis plana Nees (capim-annoni-2 ou capim annoni) é uma gramínea exótica trazida da África do Sul nos anos cinquenta e atualmente tem presença em aproximadamente 10% da área total do bioma Pampa, sendo a espécie mais invasiva desse bioma. Tendo em conta a grande capacidade desta espécie para se estabelecer em uma ampla variedade de condições ambientais, os efeitos ambientais e econômicos negativos envolvidos, bem como sua dificuldade de erradicação, é importante identificar as áreas mais suscetíveis à invasão em um futuro próximo, para assim aprimorar os planos de manejo e evitar a expansão de áreas infestadas. O presente trabalho foi desenvolvido no município de Aceguá – RS, com o objetivo de identificar quais áreas são as mais suscetíveis à invasão. Foram aplicados os modelos de distribuição de espécies MAXENT e GARP tendo como dados de entrada as variáveis ambientais calculadas a partir imagens espectrais, modelo numérico de elevação, mapa de solos e mapa de vias. Como algumas variáveis originalmente têm resolução espacial de 250 m e outras de 30 m, foi feita uma reamostragem tanto a 30 m como a 250 m para comparar os resultados dos modelos nas duas resoluções espaciais. Para diminuir o número de variáveis de entrada foi feita uma análise de correlação para eliminar as variáveis com alta correlação. Também foi feito o teste Jackknife para avaliar quais variáveis contribuem mais na modelagem espacial da distribuição do capim annoni. Os dois modelos, tanto no treinamento como na validação, nas duas resoluções espaciais, apresentam valores médios de ajuste de AUC acima de 0,7, sendo considerado um bom ajuste. Foram empregados três métodos para calcular os limiares ótimos de corte para cada um dos modelos: 1) sensibilidade igual à especificidade; 2) soma entre a sensibilidade e a especificidade máxima; Os limiares obtidos foram 42 para MAXENT_250, 39 para MAXENT_30, 69 para GARP_250 e 68 para GARP_30. Após a aplicação dos limiares, verificou-se que o modelo GARP prediz uma área maior que o MAXENT, 33,20% em comparação com 24,60% na resolução espacial de 250 m, e 35,83% contra 27,17% na resolução espacial de 30 m. Verificou-se também que o GARP possui melhor capacidade de generalização, o qual é importante para modelar espécies invasoras. Os dois modelos predizem com presença uma área comum de 21,23% e 23,94% nas resoluções espaciais de 250 m e 30 m respectivamente. As pastagens são as classes de uso que apresentam uma maior suscetibilidade à invasão de capim anonni. Ao cruzar os resultados dos modelos de suscetibilidade à invasão de capim annoni, com resolução espacial de 30 m, e as áreas de pastagens que estão sob alta pressão de pastejo, verificou-se que o modelo MAXENT consegue predizer uma suscetibilidade à invasão em 24,51% das áreas e o modelo GARP prediz 37,95% de suscetibilidade à invasão. As comparações entre as duas resoluções espaciais demonstrou que não há muitas diferenças em termos de quantificação de área, sendo que o principal ganho foi o detalhamento espacial, o qual foi obtido com um alto custo computacional. / The Eragrostis plana Nees (South African lovegrass), is an exotic grassy plant originally from South Africa, introduced in the 50s and is currently present in approximately 10% of the total area of the Pampa biome, being the most invasive species in this biome. Considering the large capacity of the South African lovegrass establishing itself in a wide variety of environmental conditions, the negative effects, both environmental and economical that it involves, as well as its difficulty of eradication, it is important to identify the invasion most susceptible areas in the near future, in order to improve the management to prevent the spread of infested areas. This research was developed in the municipality of Aceguá – RS, with the objective of identifying which areas are most susceptible to invasion. The MAXENT and GARP models of distribution of species were applied, having as input data the environmental variables calculated from spectral images, digital elevation model, soil map and road map. As some variables originally had spatial resolution of 250m and others of 30m, a resample was done at both 30m and 250m in order to compare the models results in these two spatial resolutions. To reduce the input variables amount, a correlation analysis was performed to eliminate the high correlation variables. The Jackknife test was also used to evaluate which variables contribute most to the South African lovegrass distribution spatial modeling. Both models, at the two spatial resolutions, during the training and the validation steps, present mean values of AUC adjustment above 0.7, being considered a good fit. Three methods were used to calculate the optimal thresholds for each model: 1) the sensitivity equals to the specificity; 2) the sum between sensitivity and specificity is the maximum; 3) the distance between the ROC curve and left top corner is minimum. The calculated thresholds were 42 for MAXENT_250, 39 for MAXENT_30, 69 for GARP_250 and 68 for GARP_30. After applying these thresholds, it was verified that the GARP model predicts an area greater than MAXENT, 33.20% compared to 24.60% for the spatial resolution of 250m, and 35.83% against 27.17% in the spatial resolution of 30m. It was also verified that GARP has a better generalization capacity, which is important for modeling invasive species patterns. Both models predict a common area with susceptible to invasion of 21.23% and 23.94% in spatial resolutions of 250m and 30m respectively. The grasslands are the land cover that presents a South African lovegrass invasion greater susceptibility. Cross-referencing the susceptibility invasion models with the overgrazing areas at 30m of spatial resolution, it was verified that the model MAXENT can predict a susceptibility to invasion in 24.51% of the areas and the GARP model predicts 37.95% susceptibility to invasion. Comparisons between the two spatial resolutions showed that there are not many differences in terms of area quantification, where the main gain was spatial detailing, which was obtained with a high computational cost.

Species respond to environmental change by dynamically adjusting their geographical ranges. Robust predictions of these changes are prerequisites to inform dynamic and sustainable conservation strategies. Correlative species distribution models (SDMs) relate species’ occurrence records to prevailing environmental factors to describe the environmental niche. They have been widely applied in global change context as they have comparably low data requirements and allow for rapid assessments of potential future species’ distributions. However, due to their static nature, transient responses to environmental change are essentially ignored in SDMs. Furthermore, neither dispersal nor demographic processes and biotic interactions are explicitly incorporated. Therefore, it has often been suggested to link statistical and mechanistic modelling approaches in order to make more realistic predictions of species’ distributions for scenarios of environmental change. In this thesis, I present two different ways of such linkage. (i) Mechanistic modelling can act as virtual playground for testing statistical models and allows extensive exploration of specific questions. I promote this ‘virtual ecologist’ approach as a powerful evaluation framework for testing sampling protocols, analyses and modelling tools. Also, I employ such an approach to systematically assess the effects of transient dynamics and ecological properties and processes on the prediction accuracy of SDMs for climate change projections. That way, relevant mechanisms are identified that shape the species’ response to altered environmental conditions and which should hence be considered when trying to project species’ distribution through time. (ii) I supplement SDM projections of potential future habitat for black grouse in Switzerland with an individual-based population model. By explicitly considering complex interactions between habitat availability and demographic processes, this allows for a more direct assessment of expected population response to environmental change and associated extinction risks. However, predictions were highly variable across simulations emphasising the need for principal evaluation tools like sensitivity analysis to assess uncertainty and robustness in dynamic range predictions. Furthermore, I identify data coverage of the environmental niche as a likely cause for contrasted range predictions between SDM algorithms. SDMs may fail to make reliable predictions for truncated and edge niches, meaning that portions of the niche are not represented in the data or niche edges coincide with data limits. Overall, my thesis contributes to an improved understanding of uncertainty factors in predictions of range dynamics and presents ways how to deal with these. Finally I provide preliminary guidelines for predictive modelling of dynamic species’ response to environmental change, identify key challenges for future research and discuss emerging developments. / Das Vorkommen von Arten wird zunehmend bedroht durch Klima- und Landnutzungswandel. Robuste Vorhersagen der damit verbundenen Arealveränderungen sind ausschlaggebend für die Erarbeitung dynamischer und nachhaltiger Naturschutzstrategien. Habitateignungsmodelle erstellen statistische Zusammenhänge zwischen dem Vorkommen einer Art und relevanten Umweltvariablen und erlauben zügige Einschätzungen potentieller Arealveränderungen. Dabei werden jedoch transiente Dynamiken weitgehend ignoriert sowie demographische Prozesse und biotische Interaktionen. Daher wurden Vorschläge laut, diese statistischen Modelle mit mechanistischeren Ansätzen zu koppeln. In der vorliegenden Arbeit zeige ich zwei verschiedene Möglichkeiten solcher Kopplung auf. (i) Ich beschreibe den sogenannten ‚Virtuellen Ökologen’-Ansatz als mächtiges Validierungswerkzeug, in dem mechanistische Modelle virtuelle Testflächen bieten zur Erforschung verschiedener Probenahmedesigns oder statistischer Methoden sowie spezifischer Fragestellungen. Auch verwende ich diesen Ansatz, um systematisch zu untersuchen wie sich transiente Dynamiken sowie Arteigenschaften und ökologische Prozesse auf die Vorhersagegüte von Habitateignungsmodellen auswirken. So kann ich entscheidende Prozesse identifizieren welche in zukünftigen Modellen Berücksichtigung finden sollten. (ii) Darauf aufbauend koppele ich Vorhersagen von Habitateignungsmodellen mit einem individuen-basierten Populationsmodell, um die Entwicklung des Schweizer Birkhuhnbestandes unter Klimawandel vorherzusagen. Durch die explizite Berücksichtigung der Wechselwirkungen zwischen Habitat und demographischer Prozesse lassen sich direktere Aussagen über Populationsentwicklung und damit verbundener Extinktionsrisiken treffen. Allerdings führen verschiedene Simulationen auch zu hoher Variabilität zwischen Vorhersagen, was die Bedeutung von Sensitivitätsanalysen unterstreicht, um Unsicherheiten und Robustheit von Vorhersagen einzuschätzen. Außerdem identifiziere ich Restriktionen in der Datenabdeckung des Umweltraumes als möglichen Grund für kontrastierende Vorhersagen verschiedener Habitateignungsmodelle. Wenn die Nische einer Art nicht vollständig durch Daten beschrieben ist, kann dies zu unrealistischen Vorhersagen der Art-Habitat-Beziehung führen. Insgesamt trägt meine Arbeit erheblich bei zu einem besseren Verständnis der Auswirkung verschiedenster Unsicherheitsfaktoren auf Vorhersagen von Arealveränderungen und zeigt Wege auf, mit diesen umzugehen. Abschließend erstelle ich einen vorläufigen Leitfaden für Vorhersagemodelle und identifiziere Kernpunkte für weitere Forschung auf diesem Gebiet.

O Eragrostis plana Nees (capim-annoni-2 ou capim annoni) é uma gramínea exótica trazida da África do Sul nos anos cinquenta e atualmente tem presença em aproximadamente 10% da área total do bioma Pampa, sendo a espécie mais invasiva desse bioma. Tendo em conta a grande capacidade desta espécie para se estabelecer em uma ampla variedade de condições ambientais, os efeitos ambientais e econômicos negativos envolvidos, bem como sua dificuldade de erradicação, é importante identificar as áreas mais suscetíveis à invasão em um futuro próximo, para assim aprimorar os planos de manejo e evitar a expansão de áreas infestadas. O presente trabalho foi desenvolvido no município de Aceguá – RS, com o objetivo de identificar quais áreas são as mais suscetíveis à invasão. Foram aplicados os modelos de distribuição de espécies MAXENT e GARP tendo como dados de entrada as variáveis ambientais calculadas a partir imagens espectrais, modelo numérico de elevação, mapa de solos e mapa de vias. Como algumas variáveis originalmente têm resolução espacial de 250 m e outras de 30 m, foi feita uma reamostragem tanto a 30 m como a 250 m para comparar os resultados dos modelos nas duas resoluções espaciais. Para diminuir o número de variáveis de entrada foi feita uma análise de correlação para eliminar as variáveis com alta correlação. Também foi feito o teste Jackknife para avaliar quais variáveis contribuem mais na modelagem espacial da distribuição do capim annoni. Os dois modelos, tanto no treinamento como na validação, nas duas resoluções espaciais, apresentam valores médios de ajuste de AUC acima de 0,7, sendo considerado um bom ajuste. Foram empregados três métodos para calcular os limiares ótimos de corte para cada um dos modelos: 1) sensibilidade igual à especificidade; 2) soma entre a sensibilidade e a especificidade máxima; Os limiares obtidos foram 42 para MAXENT_250, 39 para MAXENT_30, 69 para GARP_250 e 68 para GARP_30. Após a aplicação dos limiares, verificou-se que o modelo GARP prediz uma área maior que o MAXENT, 33,20% em comparação com 24,60% na resolução espacial de 250 m, e 35,83% contra 27,17% na resolução espacial de 30 m. Verificou-se também que o GARP possui melhor capacidade de generalização, o qual é importante para modelar espécies invasoras. Os dois modelos predizem com presença uma área comum de 21,23% e 23,94% nas resoluções espaciais de 250 m e 30 m respectivamente. As pastagens são as classes de uso que apresentam uma maior suscetibilidade à invasão de capim anonni. Ao cruzar os resultados dos modelos de suscetibilidade à invasão de capim annoni, com resolução espacial de 30 m, e as áreas de pastagens que estão sob alta pressão de pastejo, verificou-se que o modelo MAXENT consegue predizer uma suscetibilidade à invasão em 24,51% das áreas e o modelo GARP prediz 37,95% de suscetibilidade à invasão. As comparações entre as duas resoluções espaciais demonstrou que não há muitas diferenças em termos de quantificação de área, sendo que o principal ganho foi o detalhamento espacial, o qual foi obtido com um alto custo computacional. / The Eragrostis plana Nees (South African lovegrass), is an exotic grassy plant originally from South Africa, introduced in the 50s and is currently present in approximately 10% of the total area of the Pampa biome, being the most invasive species in this biome. Considering the large capacity of the South African lovegrass establishing itself in a wide variety of environmental conditions, the negative effects, both environmental and economical that it involves, as well as its difficulty of eradication, it is important to identify the invasion most susceptible areas in the near future, in order to improve the management to prevent the spread of infested areas. This research was developed in the municipality of Aceguá – RS, with the objective of identifying which areas are most susceptible to invasion. The MAXENT and GARP models of distribution of species were applied, having as input data the environmental variables calculated from spectral images, digital elevation model, soil map and road map. As some variables originally had spatial resolution of 250m and others of 30m, a resample was done at both 30m and 250m in order to compare the models results in these two spatial resolutions. To reduce the input variables amount, a correlation analysis was performed to eliminate the high correlation variables. The Jackknife test was also used to evaluate which variables contribute most to the South African lovegrass distribution spatial modeling. Both models, at the two spatial resolutions, during the training and the validation steps, present mean values of AUC adjustment above 0.7, being considered a good fit. Three methods were used to calculate the optimal thresholds for each model: 1) the sensitivity equals to the specificity; 2) the sum between sensitivity and specificity is the maximum; 3) the distance between the ROC curve and left top corner is minimum. The calculated thresholds were 42 for MAXENT_250, 39 for MAXENT_30, 69 for GARP_250 and 68 for GARP_30. After applying these thresholds, it was verified that the GARP model predicts an area greater than MAXENT, 33.20% compared to 24.60% for the spatial resolution of 250m, and 35.83% against 27.17% in the spatial resolution of 30m. It was also verified that GARP has a better generalization capacity, which is important for modeling invasive species patterns. Both models predict a common area with susceptible to invasion of 21.23% and 23.94% in spatial resolutions of 250m and 30m respectively. The grasslands are the land cover that presents a South African lovegrass invasion greater susceptibility. Cross-referencing the susceptibility invasion models with the overgrazing areas at 30m of spatial resolution, it was verified that the model MAXENT can predict a susceptibility to invasion in 24.51% of the areas and the GARP model predicts 37.95% susceptibility to invasion. Comparisons between the two spatial resolutions showed that there are not many differences in terms of area quantification, where the main gain was spatial detailing, which was obtained with a high computational cost.

Coordenação de Aperfeiçoamento de Pessoal de Nível Superior / Knowing the distributional patterns and the mechanisms that lead to the actual levels of diversity and richness is a challenge given the ecosystems threats. Understanding how the threats coming from the human actions can affect this diversity is the key to protect ecosystems and species. This study toward (i) investigate the richness distribution patterns of the genus Aegla in four freshwater ecoregions in South America, as well as (ii) the mechanisms that lead to that distribution, and (iii) the human influences in these freshwater ecosystems. The study area comprised four freshwater ecoregions: Upper Uruguay, Low Uruguay, Laguna dos Patos, and Tramandaí-Mampituba. Through data from five scientific collections, UFRGS, UFSM, FZB-RS, PUC-RS, and URI-Erechim, we compiled 30 species of Aegla. We associated the species distribution in four ecoregions to the distributional patterns, conservation status, environmental and spatial variables, rugosity, and land cover. The protected area network not effectively protects the aquatic ecosystems. The species richness is associated to average values of rugosity. Finally, the species occurrence areas exhibit land cover categories of agriculture and urban areas. We consider the levels of endemism inside the protected area network are key factor to conservation of aquatic ecosystems. Historic process linked to the river network formation, in a spatial context, is an important approach to understand the distributional pattern of some freshwater groups. These patterns are associated to spatial questions, following the increasing of agriculture and urban lands, are committing the diversity in freshwater environments. / Conhecer os padrões de distribuição e os mecanismos que levam aos níveis de diversidade e riqueza atuais são desafios dado à ameaça dos ecossistemas. Entendermos como as ameaças oriundas das ações do homem podem impactar essa diversidade é a chave para conservarmos ecossistemas e espécies. Esse estudo teve como objetivo (i) investigar os padrões de distribuição da riqueza do gênero Aegla em quatro ecorregiões de água doce da América do Sul, bem como, (ii) quais os mecanismos que levaram a tal distribuição, e por fim, (iii) a influência das ações do homem nos ecossistemas de água doce. A área de estudo compreendeu quatro ecorregiões de água doce: Alto Uruguai, Baixo Uruguai, Laguna dos Patos e Tramandaí-Mampituba. Através de registros de cinco coleções científicas, UFRGS, UFSM, FZB-RS, PUC-RS e URI-Erechim, obtivemos dados de 30 espécies do gênero Aegla. Padrões de distribuição, status de conservação, variáveis ambientais e espaciais, rugosidade e cobertura do solo foram analisadas e associadas à distribuição das espécies nas ecorregiões. Ficou claro que a rede de unidades de conservação não protege efetivamente os ecossistemas aquáticos. Por sua vez, a riqueza de espécies está associada a valores médios de rugosidade do relevo. E, finalmente, as áreas de ocorrência das espécies apresentaram cobertura do solo com áreas agrícolas e urbanas. Considerarmos os níveis de endemismos dos grupos nas redes de unidades de conservação é um fator chave para proteção efetiva dos ecossistemas aquáticos. Processos históricos ligados à formação das redes de rios, no contexto espacial, é a chave para entendermos padrões de distribuição de alguns grupos de água doce. Esses padrões de distribuição associados às questões espaciais, ligados ao crescente aumento das áreas de agricultura e urbana, estão comprometendo a diversidade de ambientes dulcícolas.

O Eragrostis plana Nees (capim-annoni-2 ou capim annoni) é uma gramínea exótica trazida da África do Sul nos anos cinquenta e atualmente tem presença em aproximadamente 10% da área total do bioma Pampa, sendo a espécie mais invasiva desse bioma. Tendo em conta a grande capacidade desta espécie para se estabelecer em uma ampla variedade de condições ambientais, os efeitos ambientais e econômicos negativos envolvidos, bem como sua dificuldade de erradicação, é importante identificar as áreas mais suscetíveis à invasão em um futuro próximo, para assim aprimorar os planos de manejo e evitar a expansão de áreas infestadas. O presente trabalho foi desenvolvido no município de Aceguá – RS, com o objetivo de identificar quais áreas são as mais suscetíveis à invasão. Foram aplicados os modelos de distribuição de espécies MAXENT e GARP tendo como dados de entrada as variáveis ambientais calculadas a partir imagens espectrais, modelo numérico de elevação, mapa de solos e mapa de vias. Como algumas variáveis originalmente têm resolução espacial de 250 m e outras de 30 m, foi feita uma reamostragem tanto a 30 m como a 250 m para comparar os resultados dos modelos nas duas resoluções espaciais. Para diminuir o número de variáveis de entrada foi feita uma análise de correlação para eliminar as variáveis com alta correlação. Também foi feito o teste Jackknife para avaliar quais variáveis contribuem mais na modelagem espacial da distribuição do capim annoni. Os dois modelos, tanto no treinamento como na validação, nas duas resoluções espaciais, apresentam valores médios de ajuste de AUC acima de 0,7, sendo considerado um bom ajuste. Foram empregados três métodos para calcular os limiares ótimos de corte para cada um dos modelos: 1) sensibilidade igual à especificidade; 2) soma entre a sensibilidade e a especificidade máxima; Os limiares obtidos foram 42 para MAXENT_250, 39 para MAXENT_30, 69 para GARP_250 e 68 para GARP_30. Após a aplicação dos limiares, verificou-se que o modelo GARP prediz uma área maior que o MAXENT, 33,20% em comparação com 24,60% na resolução espacial de 250 m, e 35,83% contra 27,17% na resolução espacial de 30 m. Verificou-se também que o GARP possui melhor capacidade de generalização, o qual é importante para modelar espécies invasoras. Os dois modelos predizem com presença uma área comum de 21,23% e 23,94% nas resoluções espaciais de 250 m e 30 m respectivamente. As pastagens são as classes de uso que apresentam uma maior suscetibilidade à invasão de capim anonni. Ao cruzar os resultados dos modelos de suscetibilidade à invasão de capim annoni, com resolução espacial de 30 m, e as áreas de pastagens que estão sob alta pressão de pastejo, verificou-se que o modelo MAXENT consegue predizer uma suscetibilidade à invasão em 24,51% das áreas e o modelo GARP prediz 37,95% de suscetibilidade à invasão. As comparações entre as duas resoluções espaciais demonstrou que não há muitas diferenças em termos de quantificação de área, sendo que o principal ganho foi o detalhamento espacial, o qual foi obtido com um alto custo computacional. / The Eragrostis plana Nees (South African lovegrass), is an exotic grassy plant originally from South Africa, introduced in the 50s and is currently present in approximately 10% of the total area of the Pampa biome, being the most invasive species in this biome. Considering the large capacity of the South African lovegrass establishing itself in a wide variety of environmental conditions, the negative effects, both environmental and economical that it involves, as well as its difficulty of eradication, it is important to identify the invasion most susceptible areas in the near future, in order to improve the management to prevent the spread of infested areas. This research was developed in the municipality of Aceguá – RS, with the objective of identifying which areas are most susceptible to invasion. The MAXENT and GARP models of distribution of species were applied, having as input data the environmental variables calculated from spectral images, digital elevation model, soil map and road map. As some variables originally had spatial resolution of 250m and others of 30m, a resample was done at both 30m and 250m in order to compare the models results in these two spatial resolutions. To reduce the input variables amount, a correlation analysis was performed to eliminate the high correlation variables. The Jackknife test was also used to evaluate which variables contribute most to the South African lovegrass distribution spatial modeling. Both models, at the two spatial resolutions, during the training and the validation steps, present mean values of AUC adjustment above 0.7, being considered a good fit. Three methods were used to calculate the optimal thresholds for each model: 1) the sensitivity equals to the specificity; 2) the sum between sensitivity and specificity is the maximum; 3) the distance between the ROC curve and left top corner is minimum. The calculated thresholds were 42 for MAXENT_250, 39 for MAXENT_30, 69 for GARP_250 and 68 for GARP_30. After applying these thresholds, it was verified that the GARP model predicts an area greater than MAXENT, 33.20% compared to 24.60% for the spatial resolution of 250m, and 35.83% against 27.17% in the spatial resolution of 30m. It was also verified that GARP has a better generalization capacity, which is important for modeling invasive species patterns. Both models predict a common area with susceptible to invasion of 21.23% and 23.94% in spatial resolutions of 250m and 30m respectively. The grasslands are the land cover that presents a South African lovegrass invasion greater susceptibility. Cross-referencing the susceptibility invasion models with the overgrazing areas at 30m of spatial resolution, it was verified that the model MAXENT can predict a susceptibility to invasion in 24.51% of the areas and the GARP model predicts 37.95% susceptibility to invasion. Comparisons between the two spatial resolutions showed that there are not many differences in terms of area quantification, where the main gain was spatial detailing, which was obtained with a high computational cost.

La mise en place d'objectifs de conservation adéquats repose sur la définition d'états de référence appropriés pour la distribution et l'abondance des espèces. Cependant, l'étendue des impacts cumulés de l'homme sur les écosystèmes est aujourd'hui largement sous-estimée. Dans ce projet, je m'intéresse aux opportunités qu'offre l'utilisation de données historiques combinées à différentes méthodes analytiques pour définir ces états de référence ainsi qu'aux défis posés par ce type d'approche. Des données de présence ont été recueillies pour sept espèces de cétacés et trois espèces de pinnipèdes à partir de sources archéologiques, historiques et industrielles, révélant des réductions dans la distribution et l'abondance des espèces depuis la préhistoire à nos jours. Des modèles de distribution d'espèces ont été développés pour cinq espèces de cétacés, combinant des données de chasse baleinière du 19ème siècle à des variables environnementales afin d'estimer la distribution historique des espèces avant qu'elles n'aient été chassées. J'ai obtenu pour la baleine franche de l'Atlantique Nord (Eubalena glacialis) une estimation détaillée de sa distribution et de son abondance avant qu'elle ne soit exploitée, en extrapolant des connaissances sur la distribution et l'abondance d'une espèce congénérique, la baleine franche du Pacifique Nord (E. japonica). Ces résultats suggèrent que la baleine franche de l'Atlantique Nord occupe une portion réduite de sa distribution historique, et que son abondance actuelle ne représente qu'une infime portion (<5%) de son abondance passée. Plus généralement, ces résultats soulignent l'importance de considérer des données historiques pour comprendre le niveau d'impact par l'homme sur les espèces, évaluer leur niveau de déplétion et renseigner leur potentiel de rétablissement dans l'avenir. / Relevant baselines on the historical distribution and abundance of species are needed to support appropriate conservation targets for depleted species, but the full scale of cumulative human impacts on ecosystems is highly underestimated. In this project, I investigated the challenges and opportunities of combining historical data with analytical methods to improve these historical baselines. Occurrence data from archaeological, historical and industrial sources were reviewed for seven cetacean and three pinniped species, revealing range contractions and population depletions from prehistorical times to today. For five whale species, I used species distribution modelling to combine 19th Century whaling records with environmental data, to estimate pre-whaling distributions. For the highly depleted North Atlantic right whale, (Eubalaena glacialis), I obtained a detailed estimate of pre-whaling distribution and abundance by inferring from the historical distribution and abundance of its congeneric North Pacific right whale (E. japonica). These results suggest that the North Atlantic right whale occupies a small fraction of its historical range and that its current population represents <5% of its historical abundance, with implications for the management, monitoring and conservation targets of this species. More generally, these results emphasize the utility of considering historical data to understand the extent to which species have been impacted by humans, assess their current level of depletion, and inform the options available for their future recovery.

This dissertation presents research on the value of using different sources of data to explore the factors determining invasiveness of introduced species. The research draws upon the availability of data on the historical trial plantings of alien species and other sources. The focus of the study is on Australian Acacia species as a taxon introduced into southern Africa (Lesotho, South Africa and Swaziland). The first component of the study focused on understanding the factors determining introduction outcome of species in historical trial plantings and invasion success of Australian Acacia species using Species Distribution Models (SDMs) and classification tree techniques. SDMs were calibrated using the native range occurrence records (Australia) and were validated using results of 150 years of South African government forestry trial planting records and invaded range data from the Southern African Plant Invaders Atlas. To understand factors associated with survival (‘trial success’) or failure to survive (‘trial failure’) of species in historical trial plantings, classification and regression tree analysis was used. The results indicate climate as one of the factors that explains introduction and/or invasion success of Australian Acacia species in southern Africa. However, the results also indicate that for ‘trial failures’ there are factors other than climate that could have influenced the trial outcome. This study emphasizes the need to integrate data on whether the species has been recorded to be invasive elsewhere with climate matching for invasion risk assessment. The second component of the study focused on understanding the distribution patterns of Australian Acacia species that are not known as invasive in southern Africa. The specific aims were to determine which species still exist at previously recorded sites and determine the current invasion status. This was done by collating data from different sources that list species introduced into southern Africa and then conducting revisits. For the purpose of this study, revisits means conducting field surveys based on recorded occurrences of introduced species. The known occurrence data for species on the list were obtained from different data sources and various invasion biology experts. As it was not practical to do revisits for all species on the list, three ornamental species (Acacia floribunda, A. pendula and A. retinodes) were selected as part of the pilot study for the conducted revisits in this study. Acacia retinodes trees were not found during the revisits. The results provided data that could be used to characterize species based on the Blackburn et al., (2011) scheme. However, it is not clear whether observed Acacia pendula or A. floribunda trees will spread away from the sites hence the need to continuously monitor sites for spread. The methods used in this research establish a protocol for future work on conducting revisits at known localities of introduced species to determine their population dynamics and thereby characterize the species according to the scheme for management purposes. / Dissertation (MSc)--University of Pretoria, 2014. / National Research Foundation (NRF) / Zoology and Entomology / MSc (Zoology) / Unrestricted

Estamos vivenciando o que pode ser considerado como o sexto evento de extinção global da Biodiversidade. Os anfíbios são os vertebrados mais ameaçados do mundo e ainda o número de espécies ameaçadas pode estar subestimado, pois 22% estão classificadas na categoria Dados Insuficientes (DD). O Brasil possui alta riqueza de anfíbios, porém inúmeras lacunas de conhecimento dificultam a elaboração de listas completas e prejudica a avalição do estado de conservação e o planejamento da conservação dos anfíbios brasileiros. Assim, visando contribuir efetivamente no direcionamento das ações de conservação dos anfíbios brasileiros avaliamos a influência da atitude de especialistas na classificação das espécies, e também identificamos e adaptamos ferramentas para melhor explorar os dados de distribuição disponíveis sobre as espécies DD. Foi proposto um framework de modelagem adaptativa para lidar com a escassez de dados de distribuição destas espécies e incluir a capacidade de dispersão nos modelos de distribuição de espécies. Além disso, para preencher lacunas de conhecimento dos aspectos biológicos e ecológicos das espécies de anfíbios da Mata Atlântica considerados como DD foi realizado busca na literatura e em coleções científicas, bem como a indicação de áreas prioritárias para obter informações adicionais sobre essas espécies. Identificamos que a linha de atuação dos avaliadores influencia na determinação da categoria DD, aumentando ou reduzindo a probabilidade de classificar uma espécie nesta categoria. Com isso, ressalta-se a necessidade de compor equipes multidisciplinares para avaliar o estado de conservação das espécies. O framework aqui proposto tem o potencial de inovar o processo de modelagem com poucos dados disponíveis a partir da inclusão de um dos aspectos mais difíceis de mensurar, a capacidade de dispersão da espécie. O conjunto de informações sobre as espécies DD da Mata Atlântica aumentou consideravelmente com o levantamento de dados e os modelos de distribuição de espécies, e ainda foi possível obter áreas prioritárias para aumentar o conhecimento empírico em mais de 180 municípios. Dentre as categorias de unidades de conservação mais frequentes como áreas prioritárias, destacam-se as Reservas Particulares do Patrimônio Natural nas regiões nordeste e sul, e as Áreas de Preservação permanente na região sudeste. Os resultados deste estudo contribuem efetivamente para o processo de avaliação do estado de conservação dos anfíbios brasileiros, especialmente das espécies DD da Mata Atlântica, de modo que possa ser utilizado no planejamento sistemático da conservação deste grupo. As abordagens utilizadas neste estudo podem servir de modelos para outras espécies ou grupos taxonômicos, reduzindo lacunas e incertezas no processo de avaliação do estado de conservação de espécies. / We are experiencing what could be considered as the sixth global biodiversity extinction event. Amphibians are the most threatened vertebrates in the world, and still the number of endangered species might be underestimated because 22% are classified in the category data deficient (DD). Brazil has a high richness of amphibians, but several knowledge gaps make it difficult to compile complete lists and impair the evaluation of the conservation status and the conservation planning of Brazilian amphibians. Thus, in order to contribute effectively in directing conservation actions for Brazilian anphibians we evaluated the influence of experts attitude on species classification, and also identified and adapted tools to better explore the available data on DD species. It was proposed an adaptive modeling framework to deal with the scarcity of these species distribution data and include dispersion capacity in species distribution modeling. In addition, to fill knowledge gaps of biological and ecological aspects of amphibian species of the Atlantic forest considered as DD literature search and in scientific collections were conducted, as well as the indication of priority areas for gathering additional information about these species. We identified that the line of action of the evaluators influences the determination of the DD category, increasing or reducing the probability of classifying a species in this category. Thereby, it`s emphasized the need to compose multidisciplinary teams to assess species conservation status. The framework proposed here has the potential to innovate the modeling process through the inclusion of one of the most difficult aspects to be measured, the species dispersion capacity. The set of information about DD species of the Atlantic forest has increased considerably with the survey and the species distribution models, and it was still possible to obtain priority areas to increase the empirical knowledge in more than 180 municipalities. Among the categories of protected areas most frequent as priority areas, Private Reserves stands out in the Northeast and South regions and Areas of Permanent Preservation the Southeast. Results of this study contribute effectively to the process conservation status assessment of amphibians, especially DD species from the Atlantic forest, so that it could be used in conservation systematic planning of this group. The approaches used in this study could serve as models for other species or taxonomic groups, reducing gaps and uncertainties in the process of evaluation of species conservation status.