Confronting collinearity: comparing methods for disentangling the effects of habitat loss and fragmentation

By Adam C. Smith, N. Koper, Charles M. Francis, L. Fahrig. Landscape Ecology, 2009

332 citations recorded by the source corpus.

Subfields: Ecology, Global and Planetary Change, Nature and Landscape Conservation

View Paper: https://www.semanticscholar.org/paper/0cdb342078c9b794c833f7b315cc58b43b283995

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Ecology

1. J. Metzger
2. R. Pardini
3. L. Fahrig
4. M. Tabarelli
5. C. A. Peres
6. D. Faria
7. M. Vieira
8. J. Prevedello

In Global and Planetary Change

1. M. Tabarelli
2. I. Leal
3. R. Pardini
4. L. Fahrig
5. C. A. Peres
6. M. I. Hernández
7. J. Prevedello
8. M. Vieira

In Nature and Landscape Conservation

1. R. Pardini
2. J. Metzger
3. M. Tabarelli
4. L. Fahrig
5. C. A. Peres
6. I. Leal
7. Elizabeth S Nichols
8. J. Prevedello

Citations in the corpus

• Landscape moderation of biodiversity patterns and processes ‐ eight hypotheses (2012, neutral): “…been described as ‘‘virtually logistically impossible’’ (Damschen et al., 2006; Collinge, 2009) but is an innovative and, we think, still manageable enterprise (see Krauss, Steffan-Dewenter & Tscharntke, 2003a; Ewers & Didham, 2008; Smith et al., 2009; Brückmann, Krauss & SteffanDewenter, 2010).”
• The effect of spatial resolution on projected responses to climate warming (2012, methodology): “These methods have been found to be effective when dealing with correlated explanatory variables (Smith et al., 2009).”
• of construction probability on word durations during spontaneous incremental sentence production (2012, result): “This is in contrast to several other algorithms of model selection for which collinearity is demonstrably harmful (Smith et al., 2009).”
• Optimizing landscape selection for estimating relative effects of landscape variables on ecological responses (2013, background): “, confounding) can lead to weak inferences (McGarigal and Cushman 2002; Smith et al. 2009; Eigenbrod et al. 2011), independence between predictor variables was the third criterion for landscape selection.”
• A farewell to the sum of Akaike weights: The benefits of alternative metrics for variable importance estimations in model selection (2017, background): “The imprecision of SW revealed in previ - ous studies (Cade, 2015; Galipaud et al., 2014; Giam & Olden, 2016; Murray & Conner, 2009; Smith, Koper, Francis, & Fahrig, 2009),”
• Indirect effects of habitat loss via habitat fragmentation: A cross-taxa analysis of forest-dependent species (2020, methodology): “mating fragmentation effects is the most commonly applied method in studies investigating independent effects of both variables (see Fahrig, 2017), and proved to be the statistical method providing most reliable estimates when compared to other methods (Smith et al., 2009).”
• Biodiversity Response To Habitat Loss And Fragmentation (2017, background): “This makes the statistical isolation of their effects difficult in practice (Smith et al., 2009) and has led to biased conclusions, suggesting stronger effects of habitat loss than habitat fragmentation (Didham et al., 2012; Villard and Metzger, 2014).”
• Beyond occurrence: Body condition and stress hormone as integrative indicators of habitat availability and fragmentation in the common toad (2011, neutral): “Nonetheless, methods to estimate relative effect for correlated landscape predictors, like residual regression, were biased by correlations among predictors (Smith et al., 2009).”
• Multi-scale effects of landscape complexity and crop management on pollen beetle parasitism rate (2011, methodology): “Predictor variables used in models were tested for collinearity using the pairwise Pearson correlation matrix and the Variance Inflation Factor (VIF) at each circular sector size (Smith et al. 2009).”
• Independent effects of habitat loss, habitat fragmentation and structural connectivity on forest‐dependent birds (2010, background): “…as hierarchical partitioning, have also been successfully applied to evaluate the independent contribution of correlated variables (e.g. Heikkinen et al., 2004; Radford & Bennett, 2007; but see Smith et al., 2009 for a thorough analysis on statistical treatment of correlated landscape variables).”
• The pond network: can structural connectivity reflect on (amphibian) biodiversity patterns? (2011, neutral): “Thus, landscape connectivity and, specifically, pond connectivity is of extreme importance for the long term persistence of these populations, since it may buffer the stochastic events to which ponds (particularly Mediterranean ponds) are extremely prone (Cushman 2006; Compton et al. 2007).”
• Landscape heterogeneity as an ecological filter of species traits (2014, neutral): “This may be due to correlations between descriptors of composition and 325 configuration, but, as all descriptors were included, their relative importance in driving 326 functional composition could be identified (Smith et al., 2009).”