Transferability of environmental favourability models in geographic space: The case of the Iberian desman (Galemys pyrenaicus) in Portugal and Spain

Transferring distribution models between different geographical areas may be problematic, as the performance of models outside their original scope is hard to predict. A modelling procedure is needed that gets the gist of the environmental descriptors of a distribution area, without either overfitting to the training data or overestimating the species’ distribution potential.We tested the transferability power of the favourability function, a generalized linear model, on the distribution of the Iberian desman (Galemys pyrenaicus) in the Iberian territories of Portugal and Spain.We also tested the effects of two of the main potential constraints on model transferability: the analysed ranges of the predictor variables, and the completeness of the species distribution data. We modelled 10 km×10km presence/absence data from Portugal and Spain separately, extrapolated each model to the other country, and compared predictions with observations. The Spanish model, despite arguably containing more false absences, showed good predictive ability in Portugal. The Portuguese model, whose predictors ranged between only a subset of the values observed in Spain, overestimated desman distribution when transferred.We discuss possible reasons for this differential model behaviour, and highlight the importance of this kind of models for prediction and conservation applications

Otter (Lutra lutra) distribution modeling at two resolution scales suited to conservation planning in the Iberian Peninsula

We used the results of the Spanish Otter Survey of 1994–1996, a Geographic Information System and stepwise multiple logistic regression to model otter presence/absence data in the continental Spanish UTM 10 10-km squares. Geographic situation, indicators of human activity such as highways and major urban centers, and environmental variables related with productivity, water availability, altitude, and environmental energy were included in a logistic model that correctly classified about 73% of otter presences and absences. We extrapolated the model to the adjacent territory of Portugal, and increased the model’s spatial resolution by extrapolating it to 1 1-km squares in the whole Iberian Peninsula. The model turned out to be rather flexible, predicting, for instance, the species to be very restricted to the courses of rivers in some areas, and more widespread in others. This allowed us to determine areas where otter populations may be more vulnerable to habitat changes or harmful human interventions. # 2003 Elsevier Ltd. All rights reserved.

How well can models predict changes in species distributions? A 13-year-old otter model revisited

Species distribution and ecological niche models are increasingly used in biodiversity management and conservation. However, one thing that is important but rarely done is to follow up on the predictive performance of these models over time, to check if their predictions are fulfilled and maintain accuracy, or if they apply only to the set in which they were produced. In 2003, a distribution model of the Eurasian otter (Lutra lutra) in Spain was published, based on the results of a country-wide otter survey published in 1998. This model was built with logistic regression of otter presence-absence in UTM 10 km2 cells on a diverse set of environmental, human and spatial variables, selected according to statistical criteria. Here we evaluate this model against the results of the most recent otter survey, carried out a decade later and after a significant expansion of the otter distribution area in this country. Despite the time elapsed and the evident changes in this species’ distribution, the model maintained a good predictive capacity, considering both discrimination and calibration measures. Otter distribution did not expand randomly or simply towards vicinity areas,m but specifically towards the areas predicted as most favourable by the model based on data from 10 years before. This corroborates the utility of predictive distribution models, at least in the medium term and when they are made with robust methods and relevant predictor variables.

Obtaining environmental favourability functions from logistic regression

Logistic regression is a statistical tool widely used for predicting species’ potential distributions starting from presence/absence data and a set of independent variables. However, logistic regression equations compute probability values based not only on the values of the predictor variables but also on the relative proportion of presences and absences in the dataset, which does not adequately describe the environmental favourability for or against species presence. A few strategies have been used to circumvent this, but they usually imply an alteration of the original data or the discarding of potentially valuable information. We propose a way to obtain from logistic regression an environmental favourability function whose results are not affected by an uneven proportion of presences and absences. We tested the method on the distribution of virtual species in an imaginary territory. The favourability models yielded similar values regardless of the variation in the presence/absence ratio. We also illustrate with the example of the Pyrenean desman’s (Galemys pyrenaicus) distribution in Spain. The favourability model yielded more realistic potential distribution maps than the logistic regression model. Favourability values can be regarded as the degree of membership of the fuzzy set of sites whose environmental conditions are favourable to the species, which enables applying the rules of fuzzy logic to distribution modelling. They also allow for direct comparisons between models for species with different presence/absence ratios in the study area. This makes themmore useful to estimate the conservation value of areas, to design ecological corridors, or to select appropriate areas for species reintroductions.

Distinguishing the distributions of two cryptic frogs (Anura: Discoglossidae) using molecular data and environmental modeling

Currently, the identification of two cryptic Iberian amphibians, Discoglossus galganoi Capula, Nascetti, Lanza, Bullini and Crespo, 1985 and Discoglossus jeanneae Busack, 1986, relies on molecular characterization. To provide a means to discern the distributions of these species, we used 385-base-pair sequences of the cytochrome b gene to identify 54 Spanish populations of Discoglossus. These data and a series of environmental variables were used to build up a logistic regression model capable of probabilistically designating a specimen of Discoglossus found in any Universal Transverse Mercator (UTM) grid cell of 10 km × 10 km to one of the two species. Western longitudes, wide river basins, and semipermeable (mainly siliceous) and sandstone substrates favored the presence of D. galganoi, while eastern longitudes, mountainous areas, severe floodings, and impermeable (mainly clay) or basic (limestone and gypsum) substrates favored D. jeanneae. Fifteen percent of the UTM cells were predicted to be shared by both species, whereas 51% were clearly in favor of D. galganoi and 34% were in favor of D. jeanneae, considering odds of 4:1. These results suggest that these two species have parapatric distributions and allow for preliminary identification of potential secondary contact areas. The method applied here can be generalized and used for other geographic problems posed by cryptic species.

Allergological evaluation of a dog population in a veterinary immuno-allergology consultation: What correlates in a canine model

Introduction: Allergic dermatitis (AD) is the most common canine pruritic condition in veterinary dermatology. Allergic dermatitis to flea bites presents the highest prevalence, followed by atopic dermatitis and food AD. This study aimed to identify possible correlation between data from clinical signs, intradermal tests (IDT) and specific IgE levels, which are used in dog AD assessment. Methods: Fifty five dogs from the Veterinary Hospital of the University of Évora (Portugal) and Rof Codina University Hospital (Lugo, Spain) outpatient consultations were studied by means of clinical inquiry, IDT and specific IgE determination. Thirty five of the patients belonged to predisposed breeds, 30 were females and 25 males. Forty one (74%) were indoor. Results: In 82% of cases first clinical signs appeared before the age of 3 years and 24% even before 1 year old. In 70% of the individuals clinical signs included itching, which was generalized in 66%, with 78% of paw licking and chewing. Clinical profile showed seasonal worsening in 64% of cases. From the 69.1% of dogs already presenting with dermatitis, 50% also presented external otitis and 28.9% self-inflicted alopecia. "Intense itching" was found in 10.5%, "medium itching" in 81.6% and “mild itching” in 5.26% of the patients. Prevalence of positive IDT was 37.3 % to Lep d, 29.41% to Der f, 27.5% to Der p, 25.5% to Dac g and 21.6% to Malassezia sp. From the 37 dogs submitted to food IDT, 40.5% revealed positive to beef, 27% to chicken, 27% to porc and 5.4% to lamb. Specific IgE > 150 EAU was found in 84% of dogs to indoor allergen sources and in 68% to pollens. A negative correlation was found between an outdoor life and the intensity (p = 0.033) and precocity (p = 0.026) of clinical signs. Sensitization to pollens was found positively correlated with the seasonality of clinical signs (p = 0.001) and the positivity for Dac g (p = 0.007). The prevalence of chronic otitis correlated positively with alopecia and reactivity to Lep d (p = 0.008), Plantago lanceolata (p = 0.026) and Platanus acerifolia (p = 0.017). There was no correlation between the results of ITD and specific IgE. Conclusion: We can conclude that correlation between different clinical signs and positive testing for some allergenic sources may occur, as well as between sensitization to pollens and the beginning, the intensity and the seasonality of dog patient clinical signs.

Modelação do cancro da mama na região do Alentejo

Estudos epidemiológicos são estudos estatísticos onde se procura relacionar ocorrências de eventos de saúde com uma ou várias causas específicas. A importância que os modelos epidemiológicos assumem hoje no estudo de doenças de foro oncológico, em particular no estabelecimento das suas etiologias, é incontornável. Segundo Ogden, J. (1999) o cancro é "um crescimento incontrolável de células anormais que produzem tumores chamados neoplasias". Estes tumores podem ter origem benigna (não se espalham pelo corpo) ou maligna (apresentam metastização de outros órgãos). Sendo uma doença actual, com uma elevada taxa de incidência em Portugal quando comparada com outras doenças (Instituto Nacional de Estatística- INE, 2009), aumentando esta taxa com a idade tal como refere Marques, L. (2003), podendo ocorrer o diagnóstico desta doença em qualquer idade. De acordo com INE (2000) pode dizer-se que o cancro está entre as três principais causas de morte em Portugal, registando-se um aumento progressivo do seu peso proporcional, sendo o cancro da mama o tipo de cancro mais comum entre as mulheres e uma das doenças com maior impacto na nossa sociedade. O objectivo principal deste trabalho é a estimação e modelação do risco de contrair uma doença de natureza não contagiosa e rara (neste caso, cancro da mama), usando dados da região do Alentejo. Pretende-se fazer um apanhado das metodologias mais empregues nesta área e aplicá-las na prática, com ênfase nos estudos caso-controlo e nos modelos lineares generalizados (GLM) - mais concretamente regressão logística. Os estudos caso-controlo são usados para identificar os factores que podem contribuir para uma condição médica, comparando indivíduos que têm essa condição (casos) com pacientes que não têm a condição, mas que de resto são semelhantes (controlos). Neste trabalho utilizou-se essa metodologia para estudar a associação entre o viver em ambiente rural/urbano e o cancro da mama. Tendo em conta que o objectivo principal deste estudo se prende com o estudo da relação entre variáveis, mais propriamente, análise de influência que uma ou mais variáveis (explicativas) têm sobre uma variável de interesse (resposta), para esse efeito são estudados os modelos lineares generalizados - GLM - unificados na mesma moldura teórica pela primeira vez por Nelder & Wedderburn (1972) - e, posteriormente aplicados ao conjunto de dados sobre cancro da mama na Região do Alentejo. O presente trabalho pretende assim, ser um contributo na identificação de factores de risco do cancro da mama na região do Alentejo. ABSTRACT: Epidemiological studies are statistical studies where attempts to relate occurrences of health events with one or more specific causes. The importance of epidemiological models that are far in the study of diseases of cancer forum, particularly in establishing their etiology, is inescapable. According to Ogden, J. (1999) cancer is "an incontrollable growth of abnormal cells that produce tumors called cancer". These tumors may be benign (not spread throughout the body) or malignant (show metastasis to other organs). Being a current illness with a high incidence rate in Portugal compared with the same respect to other diseases (National Statistics 1nstitute -1NE, 2009) having an increasing rate with age as mentioned Marques, L. (2003), and can possibly be diagnosed at any age. According to 1NE (2000) the cancer is among the top three causes of death in Portugal and there is a progressive increase of its proportional weight. Breast cancer is the most common form of cancer among women and the diseases with major impact in our society. The main objective of this work is to model and estimate the risk of contracting a non-contagious and rare disease (in this case, breast cancer), using data from the Alentejo region. It is intended to summarize some of the methodologies employed in this area and apply them in practice, with emphasis on case-control studies and generalized linear models (GLM) - more specifically the logistic regression. The case-control studies are used to identify factors that may contribute to a medical condition, comparing individuals who have this condition (cases) with patients who have not the condition but that are otherwise similar (controls). ln this work we used this methodology to study the association between living in a rural/urban and breast cancer. Given that the main objective of this study rather relates to the study of the relationship between variables to analyze the influence that one or more variables (explanatory) have on a variable (response), for this purpose we study the generalized linear models - GLM - first unified in the same theoretical framework by Nelder and Wedderburn (1972) and subsequently applied to the data set on breast cancer in the Alentejo region. This work intends to be a contribution in identifying risk factors for breast cancer in the Alentejo region.

Single diode model parameters analysis of photovoltaic cell

In this paper it is proposed to obtain enhanced and more efficient parameters model from generalized five parameters (single diode) model of PV cells. The paper also introduces, describes and implements a seven parameter model for photovoltaic cell (PV cell) which includes two internal parameters and five external parameters. To obtain the model the mathematical equations and an equivalent circuit consisting of a photo generated current source, a series resistor, a shunt resistor and a diode is used. The fundamental equation of PV cell is used to analyse and best fit the observation data. Especially bisection iteration method is used to obtain the expected result and to understand the deviation of changes in different parameters situation at various conditions respectively. The produced model can be used of measuring and understanding the actions of photovoltaic cells for certain changes and parameters extraction. The effect is also studied with I-V and P-V characteristics of PV cells though it is a challenge to optimize the output with real time simulation. The working procedure is also discussed and an experiment presented to get the closure and insight about the produced model and to decide upon the model validity. At the end, we observed that the result of the simulation is very close to the produced model.