Modeling habitat suitability is beneficial for management and conservation of a species. Although data-rich models are commonly used, opinion-based models may be a beneficial alternative to estimate suitable habitat locations. Despite the increasing use of habitat models, few studies have linked habitat model covariates (i.e., land cover, weather, and normalized difference vegetation indexes (NDVI)) to demographic parameters. This study evaluates model performance and transferability of maximum entropy (MaxEnt) and expert opinion models for predicting American beaver (Castor canadensis) distribution in the southeastern US. I also investigated the relationship of environmental and habitat model covariates to beaver survival. The model’s predictive performance and transferability were evaluated using the area under the curve (AUC) index. Both model approaches performed well at predicting beaver presence. While MaxEnt had better performance, the expert models predicted greater areas as suitable for beaver. Beaver survival was estimated for northern Alabama and was found to be influenced by NDVI and weather covariates in this study.

Submitted by Izabel Franco (izabel-franco@ufscar.br) on 2016-09-12T20:08:21Z No. of bitstreams: 1 TeseSSD.pdf: 4196266 bytes, checksum: 852bdafbbad865e0ccb454503e54e11b (MD5) / Approved for entry into archive by Marina Freitas (marinapf@ufscar.br) on 2016-09-13T19:55:41Z (GMT) No. of bitstreams: 1 TeseSSD.pdf: 4196266 bytes, checksum: 852bdafbbad865e0ccb454503e54e11b (MD5) / Approved for entry into archive by Marina Freitas (marinapf@ufscar.br) on 2016-09-13T19:55:50Z (GMT) No. of bitstreams: 1 TeseSSD.pdf: 4196266 bytes, checksum: 852bdafbbad865e0ccb454503e54e11b (MD5) / Made available in DSpace on 2016-09-13T19:55:59Z (GMT). No. of bitstreams: 1 TeseSSD.pdf: 4196266 bytes, checksum: 852bdafbbad865e0ccb454503e54e11b (MD5) Previous issue date: 2015-06-12 / Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP) / Roadkill and isolation of populations due to the barrier effect can be considered the main direct impacts of roads on wildlife. Mammals by behavioral, size and charisma features are one of the groups of vertebrates that require mitigation for mortality, as they are also among the most seriously injured on roads. An important approach to effectively mitigate roadkill on highways is to locate the points where roadkill are concentrated. The objective of this study was to evaluate the variation in mortality of mammals due to the duplication of the BR 101 highway south. In addition to the K Ripley aggregation analysis, were generated habitat suitability models (MAH) to identify favorable locations roadkill species, using the model functional groups for prediction of highway stretches with more chances to roadkill. It was recorded 21 taxa of mammals get hit, being the most abundant gender Didelphis (n = 721), followed by Cerdocyon thous (n = 108). The abundance of carcasses decreases between before and after the duplication (F 18.04 p = <0.001). Hotspots were not overlapped between the periods analyzed, indicating that some explanatory factor has changed over the work. Observing the variables that contributed most to the models for each functional group, we noted that there were differences in the most influential variables for each functional group in different periods. Of the nine generated models, we note that six variables contributed more than 20% in different models. There were differences between the periods before, during and after the duplication of the variables that contributed most or contribution value in functional groups. These differences may reflect the change in the distribution of roadkill among highway doubling periods. Both approaches demonstrated that there have been changes in the magnitude and distribution of roadkill in the period prior to the period after duplication. So a highway with two paviments differ from a highway with four paviments about how the mammals react in terms of movement and relationship to landscape. / A morte por atropelamentos e o isolamento de populações devido ao efeito de barreira podem ser considerados os principais impactos diretos das rodovias sobre a fauna silvestre. Os mamíferos pelas características comportamentais, tamanho e carisma são um dos grupos de vertebrados que necessitam mitigações para a mortalidade, visto que também estão entre os que mais morrem em rodovias. Uma abordagem importante para mitigar efetivamente os atropelamentos em rodovias é localizar os pontos onde os atropelamentos estão concentrados. O objetivo deste trabalho foi avaliar a variação da mortalidade de mamíferos em função da duplicação da rodovia BR 101 sul. Além da análise de agregação de K de Ripley, foram gerados modelos de adequabilidade de habitat (MAH) para identificar locais favoráveis a atropelamentos de espécies, usando no modelo grupos funcionais para a predição de trechos da rodovia com mais chances de atropelamento. Registrou-se 21 taxóns de mamíferos atropelados, sendo o mais abundante o gênero Didelphis (n=721), seguido de Cerdocyon thous (n=108). A abundância de carcaças diminui entre o antes e o depois da duplicação (F=18,04; p<0,001). Os hotspots não foram sobrepostos entre os períodos analisados, indicando que algum fator explicativo mudou ao longo da obra. Observando as variáveis que mais contribuíram com os modelos para cada grupo funcional, notamos que houve diferença nas variáveis mais influentes para cada grupo funcional nos diferentes períodos. Dos nove modelos gerados, notamos que seis variáveis contribuíram em mais de 20% em diferentes modelos. Houve diferença entre os períodos de antes, durante e depois da duplicação das variáveis que mais contribuíram ou do valor de contribuição nos grupos funcionais. Estas diferenças podem ser reflexo da mudança na distribuição dos atropelamentos entre os períodos da duplicação da rodovia. As duas abordagens demonstraram que houveram mudanças na magnitude e distribuição dos atropelamentos no período antes em relação ao período após a duplicação. Portanto uma rodovia com dois pavimentos funciona diferente de uma rodovia de quatro pavimentos em relação a como a mastofauna reage em termos de movimentação e relação com paisagem.

No description available.