Ecological Archives E096-237-A2
Tom P. Bregman, Alexander C. Lees, Nathalie Seddon, Hannah E. A. Macgregor, Bianca Darski, Alexandre Aleixo, Michael B. Bonsall, and Joseph A. Tobias. 2015. Species interactions regulate the collapse of biodiversity and ecosystem function in tropical forest fragments. Ecology 96:2692–2704. http://dx.doi.org/10.1890/14-1731.1
Appendix B. Expanded materials and methods detailing trait and model selection Avian functional traits.
The relationship between avian morphology and ecology is well established: beak shape is related to the dietary niche (Miles et al. 1987, Grant and Grant 2006) while wing shape, tarsus and tail length are proxies for locomotion, i.e. dispersal limitation, substrate use and foraging technique (Miles et al. 1987, Dawideit et al. 2009, Claramunt et al. 2012). Thus, to capture variation in the ecological niche, we measured museum specimens. Three beak measurements were taken at or from the anterior edge of the nostrils: (1) width, (2) length to the tip of the beak and (3) depth (as vertical height). Tarsus length was measured as the distance between the final scale of the acrotarsium to the middle of the ankle joint. Tail length was taken from the point at which the two central rectrices meet the skin to the tip of the longest rectrix. Wing chord was measured as the distance between the carpal joint and wing tip of the unflattened wing. We also measured the distance between the first secondary to the longest primary and used this to calculate the hand-wing index of Claramunt et al. (2012): hand-wing index = wing chord – distance between the first secondary to longest primary divided by wing chord. All measurements were taken to the nearest 0.01 mm apart from wing chord and tail length which were measured to the nearest mm. To account for intraspecific variability, we attempted to measure at least two males and two females for each species, using additional unsexed specimens for 130 monomorphic species (39% of total) lacking 4 specimens of known sex. On average, we measured 7.9 ± 8.9 specimens per species (3.7 ± 4.0 males, 2.7 ± 1.6 females and 1.7 ± 5.2 unsexed). Before analysis, we averaged the data from unsexed, male and female specimens to create a species average for each morphological trait.
We then used ordination techniques (principal components analysis) to generate four categories of functional trait: overall body size, trophic traits, locomotory traits and dispersal traits. Trophic traits were derived from beak measurements and thus provide an index of diet; locomotory traits and dispersal traits were derived from wing, tail and tarsus measurements and thus capture important aspects of the foraging niche; dispersal traits were derived from wing length and primary projection. See Methods for details of procedure. The trait axes generated are described as ‘functional’ due to their importance in governing species interactions and fitness within ecosystems (Violle et al. 2007).
Details on model selection and averaging (analysis 2a)
Traditionally, null hypotheses are tested individually, with ‘best’ models chosen based on information criteria such as Akaike’s information criterion (AIC). While this method provides a means of selecting the final model it does not take into consideration model uncertainty (Whittingham et al. 2006). Recently, there has been increasing use of information theoretic (‘IT’) techniques as they allow the simultaneous testing of multiple hypotheses and generation of robust estimates for model parameters (Burnham and Anderson 2002). We follow techniques described in Grueber et al. (2011) to generate ‘top model sets’, which we then average to produce parameter estimates.
We first fitted an initial GLMM, where we included fragment size, proportion of fragment covered in good forest and distance to forest larger than 1000 ha, using the lme4 package (Bates and Maechler 2009). Input variables were standardized using Gelman’s (2008) approach implemented in the arm package (Gelman et al. 2009). We then use the dredge function in the MuMIn package (Bartoń 2009) to derive a complete set of models before selecting those which are within 2 AICC (AIC corrected for the number of model parameters) (Burnham and Anderson 2002) values of the top model to derive average parameter estimates.
Differences in fragmentation impacts across guilds
Our finding that the signature of over-dispersion in communities varied across guilds and traits perhaps arises because of differences in ecological characteristics. Insectivore communities change dramatically with a decrease in patch size as many forest specialists disappear, whereas the same process has limited impact on tropical frugivores, many of which are tolerant of land-use change (Bregman et al. 2014). The greater susceptibility of insectivores has often been attributed to factors such as territoriality and lower dispersal ability, both of which increase the minimum area requirements for viable species and reduce gap-crossing ability. These factors seem less likely to increase competition than shifts in food supply. In the aftermath of forest fragmentation, the abundance and availability of invertebrates may decline, leading to food shortages (Burke and Nol 1998, Zanette et al. 2000, but see Sekercioglu et al. 2002). The consequent decrease in the foraging success of insectivorous birds may intensify competition among species, potentially explaining over-dispersion in body size, an important proxy for prey-choice (Hespenheide 1971).
Changes in food supply are thought to be less important to frugivores, although rapid shifts in tree species composition in fragmented landscapes (Michalski et al. 2007) may lead to local extinctions of key food resources. This could result in increased competition among frugivores, particularly in smaller habitat patches where resources are most limited (Carnicer et al. 2009) , potentially explaining our finding of over-dispersion in beak shape in smaller patches. We also found over-dispersion in frugivore body size, which correlates with gape size and thus the maximum fruit size in the diet (Wheelwright 1985). Finally, evidence of competition among frugivores inhabiting small, isolated habitat patches was also detected in the dispersal trait axis. It is not apparent why competition should be mediated by this axis, but it may capture the signal of competition from other traits (i.e. species with similar dispersal traits may share other life history characteristics related to resource acquisition).
Bartoń, K. 2009. MuMIn: multi-model inference. R package.
Bates, D. and M. Maechler. 2009. lme4: Linear mixed-effects models using S4 classes.
Bregman, T. P., C. H. Sekercioglu, and J. A. Tobias. 2014. Global patterns and predictors of bird species responses to forest fragmentation: Implications for ecosystem function and conservation. Biological Conservation 169:372–383.
Burke, D. M. and E. Nol. 1998. Influence of food abundance, nest-site habitat, and forest fragmentation on breeding ovenbirds. Auk 115:96–104.
Burnham, K. P. and D. R. Anderson. 2002. Model selection and multimodel inference : a practical information-theoretic approach. Second edition. Springer, New York, New York, USA.
Carnicer, J., P. Jordano, and C. J. Melian. 2009. The temporal dynamics of resource use by frugivorous birds: a network approach. Ecology 90:1958–1970.
Claramunt, S., E. P. Derryberry, J. V. Remsen, and R. T. Brumfield. 2012. High dispersal ability inhibits speciation in a continental radiation of passerine birds. Proceedings of the Royal Society B-Biological Sciences 279:1567–1574.
Dawideit, B. A., A. B. Phillimore, I. Laube, B. Leisler, and K. Bohning-Gaese. 2009. Ecomorphological predictors of natal dispersal distances in birds. Journal of Animal Ecology 78:388–395.
Gelman, A. 2008. Scaling regression inputs by dividing by two standard deviations. Statistics in Medicine 27:2865–2873.
Gelman, A., Y.-S. Su, M. Yajima, J. Hill, M. G. Pittau, J. Kerman, and et al. 2009. arm: data analysis using regression and multilevel hierarchical models. R package.
Grant, P. R. and B. R. Grant. 2006. Evolution of character displacement in Darwin's finches. Science 313:224–226.
Grueber, C. E., S. Nakagawa, R. J. Laws, and I. G. Jamieson. 2011. Multimodel inference in ecology and evolution: challenges and solutions. Journal of Evolutionary Biology 24:699–711.
Hespenheide, H. 1971. Food preference and extent of overlap in some insectivorous birds, with special reference to Tyrannidae. Ibis 113:59–72.
Michalski, F., I. Nishi, and C. A. Peres. 2007. Disturbance-mediated drift in tree functional groups in Amazonian forest fragments. Biotropica 39:691–701.
Miles, D. B., R. E. Ricklefs, and J. Travis. 1987. Concordance of ecomorphological relationships in 3 assemblages of Passerine birds. American Naturalist 129:347–364.
Sekercioglu, C. H., P. R. Ehrlich, G. C. Daily, D. Aygen, D. Goehring, and R. F. Sandi. 2002. Disappearance of insectivorous birds from tropical forest fragments. Proceedings of the National Academy of Sciences of the United States of America 99:263–267.
Violle, C., M. L. Navas, D. Vile, E. Kazakou, C. Fortunel, I. Hummel, and E. Garnier. 2007. Let the concept of trait be functional! Oikos 116:882–892.
Wheelwright, N. T. 1985. Fruit size, gape width, and the diets of fruit-eating birds. Ecology 66:808–818.
Whittingham, M. J., P. A. Stephens, R. B. Bradbury, and R. P. Freckleton. 2006. Why do we still use stepwise modelling in ecology and behaviour? Journal of Animal Ecology 75:1182–1189.
Zanette, L., P. Doyle, and S. M. Tremont. 2000. Food shortage in small fragments: Evidence from an area-sensitive passerine. Ecology 81:1654–1666.
[Back to E096-237]