Refine
Has Fulltext
- no (3)
Year of publication
- 2013 (3) (remove)
Document Type
- Article (3) (remove)
Language
- English (3)
Is part of the Bibliography
- yes (3)
Keywords
- Allometry (1)
- Body size (1)
- Foraging movement (1)
- Fractal landscapes (1)
- Individual-based model (1)
- Locomotion costs (1)
Institute
Fragmentation and loss of habitat are major threats to animal communities and are therefore important to conservation. Due to the complexity of the interplay of spatial effects and community processes, our mechanistic understanding of how communities respond to such landscape changes is still poor. Modelling studies have mostly focused on elucidating the principles of community response to fragmentation and habitat loss at relatively large spatial and temporal scales relevant to metacommunity dynamics. Yet, it has been shown that also small scale processes, like foraging behaviour, space use by individuals and local resource competition are also important factors. However, most studies that consider these smaller scales are designed for single species and are characterized by high model complexity. Hence, they are not easily applicable to ecological communities of interacting individuals. To fill this gap, we apply an allometric model of individual home range formation to investigate the effects of habitat loss and fragmentation on mammal and bird communities, and, in this context, to investigate the role of interspecific competition and individual space use. Results show a similar response of both taxa to habitat loss. Community composition is shifted towards higher frequency of relatively small animals. The exponent and the 95%-quantile of the individual size distribution (ISD, described as a power law distribution) of the emerging communities show threshold behaviour with decreasing habitat area. Fragmentation per se has a similar and strong effect on mammals, but not on birds. The ISDs of bird communities were insensitive to fragmentation at the small scales considered here. These patterns can be explained by competitive release taking place in interacting animal communities, with the exception of bird's buffering response to fragmentation, presumably by adjusting the size of their home ranges. These results reflect consequences of higher mobility of birds compared to mammals of the same size and the importance of considering competitive interaction, particularly for mammal communities, in response to landscape fragmentation. Our allometric approach enables scaling up from individual physiology and foraging behaviour to terrestrial communities, and disentangling the role of individual space use and interspecific competition in controlling the response of mammal and bird communities to landscape changes.
Collinearity a review of methods to deal with it and a simulation study evaluating their performance
(2013)
Collinearity refers to the non independence of predictor variables, usually in a regression-type analysis. It is a common feature of any descriptive ecological data set and can be a problem for parameter estimation because it inflates the variance of regression parameters and hence potentially leads to the wrong identification of relevant predictors in a statistical model. Collinearity is a severe problem when a model is trained on data from one region or time, and predicted to another with a different or unknown structure of collinearity. To demonstrate the reach of the problem of collinearity in ecology, we show how relationships among predictors differ between biomes, change over spatial scales and through time. Across disciplines, different approaches to addressing collinearity problems have been developed, ranging from clustering of predictors, threshold-based pre-selection, through latent variable methods, to shrinkage and regularisation. Using simulated data with five predictor-response relationships of increasing complexity and eight levels of collinearity we compared ways to address collinearity with standard multiple regression and machine-learning approaches. We assessed the performance of each approach by testing its impact on prediction to new data. In the extreme, we tested whether the methods were able to identify the true underlying relationship in a training dataset with strong collinearity by evaluating its performance on a test dataset without any collinearity. We found that methods specifically designed for collinearity, such as latent variable methods and tree based models, did not outperform the traditional GLM and threshold-based pre-selection. Our results highlight the value of GLM in combination with penalised methods (particularly ridge) and threshold-based pre-selection when omitted variables are considered in the final interpretation. However, all approaches tested yielded degraded predictions under change in collinearity structure and the folk lore'-thresholds of correlation coefficients between predictor variables of |r| >0.7 was an appropriate indicator for when collinearity begins to severely distort model estimation and subsequent prediction. The use of ecological understanding of the system in pre-analysis variable selection and the choice of the least sensitive statistical approaches reduce the problems of collinearity, but cannot ultimately solve them.