The ecological importance of small, terrestrial insectivores such as litter frogs and small geckos is a topic that I’ve been curious about for years. While my dissertation research does not include anything about it, I am still quite curious about how these small, diverse and abundant vertebrates fit into ecosystems. Anoles possess all of the attributes that seem to predispose them to strong interactions and soon after arriving in Miami to begin my Ph.D., I decided to launch a small, side-project using tried-and-true food web tools, stable isotopes and gut content analysis to try to illustrate if and how these small, rather inconspicuous predators might affect ecosystem structure and function. The results of this study were just published online in Functional Ecology.
Basically, we found that anoles couple adjacent food webs by consuming insects that move across habitat boundaries. While food web linkages are a potentially important ecological dynamic and our study yielded some unique findings, there are other bits of information for those more generally interested in anole biology. Therefore, I’ve decided to break this post into two parts. Part 1 deals with our primary findings and contextualizes them within current understanding of food web ecology linkages. It’s sort of a geeky treatment of the subject. Part 2 will illustrate some of the other data that we’ve collected that were not dealt with explicitly in the paper that will be of interest to, I suspect, AA readers.
Bidirectional trophic linkages couple canopy and understory food webs
Sean T. Giery,Nathan P. Lemoine, Caroline M. Hammerschlag-Peyer, Robin N. Abbey-Lee, and Craig A. Layman
1. Cross-system resource flux is a fundamental component of ecological systems. Allochthonous material flows generate trophic linkages between adjacent food webs, thereby affecting community structure and stability in recipient systems.
2. We investigated cross-habitat trophic linkages between canopy and understory food webs in a terrestrial, wooded, ecosystem in South Florida, USA. The focal community consisted of three species of Anolis lizards and their prey. We described interspecific differences among Anolis species in the strength and routing of these cross-habitat flows using stable isotope analysis, stomach content analysis, and habitat use data.
3. All three Anolis species in this study consumed different prey, and occupied vertically distinct arboreal habitats. Despite these differences, carbon isotope and stomach content analysis revealed strong integration with understory and canopy food webs for all Anolis species. Modes of resource flux contributing to the observed cross-habitat trophic linkages included prey movement and the gravity-driven transport of detritus.
4. Our study shows that terrestrial systems are linked by considerable bidirectional cross-system resource flux. Our results also suggest that considering species-specific interactions between predator and prey are necessary to fully understand the diversity of material and energy flows between spatially separated habitats.
Some basics - The community was composed of four anole species, Anolis sagrei, A. distichus, A. carolinensis*, and A. equestris. The study site was recently featured in AA. Generally, the purpose of the study was to describe variation among species in resource use using stomach contents, habitat use, and stable isotope analysis. But based on some initial observations and a bit of stable isotope data, we had considered that there might be a role for anoles in ecosystems via linking spatially distinct food webs. That is, anole diets might be sourced, in part, by primary production originating outside their respective microhabitats. Basically, we knew that anoles occupy distinct arboreal habitats, but when we examined the stomach contents of each, we found that some prey were from habitats spatially distinct from the ones used by each anole species (e.g., How do terrestrial grasshoppers get inside a canopy giant anole such as A. equestris?),which spawned a more in-depth investigation. Additionally, some initial stable isotope data strongly supported the same interpretation – that is, anole diets might be at least partially derived from allochthonous resources.
Based on what we know about anole habitat partitioning, we predicted that because each species occupied a different perch height, they would integrate carbon differently; specifically we thought the degree of integration would be correlated with proximity to the allochthonous resources: A. sagrei – a trunk-ground ecomorph – would have the highest contribution from understory sources, followed by A. distichus – a trunk ecomorph, and A. equestris – a crown giant ecomorph, respectively. It seemed a rather basic prediction and I felt pretty good about it. If correct, then A. sagrei would be important ‘couplers’ of food webs and ecosystems. We were totally wrong about that one! In fact, what we found was that all of the species we examined were strongly linked by their prey to adjacent habitats. In addition, all of these links were species-specific (Figure 4):
- A. distichus was linked to understory production by mobile carpenter ants (carpenter ants forage in both canopies and understory vegetation for honeydew, insects and vegetable foods. In transit between these resource patches they are consumed by bole-inhabiting A. distichus).
- A. sagrei was linked to canopy resources by detritus falls (that source the detritivores consumed by A. sagrei).
So in the end, we found very little variation (i.e., none) in the contribution of allochthonous resources among species, but the magnitude of allochthonous resource flux was obviously large (~20-35% of consumed carbon was allochthonous for all species examined; Fig 3.). The significance of this result is that while all are apparently redundant in their basic role of coupling, because each species generates a distinct linkage each species produces a unique trophic link potentially further strengthening the coupling of adjacent food webs.
* A. carolinensis was a member of the community, but, they were infrequently encountered and not included in our study. However, we predicted that they would have a similar coupling strength as knight anoles due to their canopy affinities (unpublished data on A. carolinensis will be discussed later).
Other ecology-type findings and a bit of shameless speculation - More generally, our findings show that anoles eat things that move between habitats and that each of the three species of anole in our study creates unique or species-specific linkages between spatially distinct food webs due to their specific ecologies. This, I think, is pretty cool, but within our paper are a few other interesting nuggets.
Bidirectional flux - This is the first paper to show bidirectional linkages in a terrestrial ecosystem. Ecosystems tend to be linked in space by unidirectional flows. Prevailing physical forces tend to drive materials in one direction – as can be seen from other anole studies here, here and here. In the case of gravity, it’s pretty obvious. But organisms, particularly animals, can work against or modify prevailing forces to affect their trajectory and magnitude. In our case, insects move against a strong, prevailing force, gravity, to generate a parallel channel of material transport. Each channel – active movement and gravity – have been detailed previously from terrestrial systems, but not in concert, which is something that our paper demonstrates well.
Stomach Contents - This is also one of the few coupling papers that used diet data to explicitly bridge the coupling consumer with the allochthonous resource. Typically, stable isotopes are used to trace the flux of carbon, but one of the limitations of such tracers is that they typically do not reflect the specific identity of the link or how energy and material are routed to the consumer. Coming from a food web lab, my coauthors and I are quite familiar with the opportunities and limitations of stable isotope analyses and routinely include stomach content data simply because they provide the detail necessary to understand more specifically how organisms are linked in food webs (i.e., food web structure) and what the potential effects of altering web/community composition might be.
Taxonomic scope and resolution - This is one of the few papers that examines food web interconnectivity using a community-level approach. That is, we included several consumer species, and incorporate high resolution (often to species) prey data in our study. To finalize our paper we include the following rationale for this degree of resolution that I believe highlights the value of this approach:
“Each of the allochthonous flows in our ecosystem is generated by the unique ecologies of species, i.e., habitat use and the diet of predators and prey. These results are relevant for understanding the potential effects of species diversity on food web structure and ecological stability (Rooney & McCann 2011) because in concert, these species-specific interactions generate multiple trophic linkages that result in the coupling of adjacent food webs. We believe that multiple linkages are probably quite common among spatially distinct systems, yet several approaches to illustrating the biological vectors of allochthonous flux might mask important underlying diversity in trophic linkages. For example, organisms are often lumped into groups of functionally and/or taxonomically similar species such as ‘seabirds’, or ‘waterbirds’ (e.g., Sanchez-Pinero & Polis 2000). Another common approach is to focus on individual species, while not examining the larger community (e.g., examining a single species of salmonid where several may coexist; Holtgrieve & Schindler 2011). The results we present here suggest that the taxonomic scope and resolution used when studying food web linkages, whether lumping organisms into mixed taxa groups or taking a narrow perspective on community level analysis, may mask important diversity in trophic interactions.”
New Methods - Perhaps less exciting are some new (to our knowledge at least) calculations that we used in our study. First is a new approach for estimating trophic shifts in isotope values for consumers. It’s basically just a weighted-average of prey trophic shifts that gives a trophic correction value (TC) that can then be incorporated into mixture models. This is a simple procedure and could be integrated with Bayesian methods for more robust estimation. That said, it still requires a significant amount of information on prey ecology (trophic level, in particular) in order to give reasonable estimates for TC. See eqn. 3. Second, prompted by a reviewer’s desire for uncertainty estimates for our diet overlap statistic, we developed a procedure for testing differences between species based on stomach content data. It’s really just a re-sampling procedure applied to Schoener’s overlap Index (D) calculated between individuals of the population, but it allows for confidence interval calculation. See page 3, ‘Stomach Contents’ for more detail. Dan Bolnick et al. make a brief mention of this approach or something like it in their 2002 Ecology paper, but we’ve never seen it implemented. There are some limitations to this approach such as the large-ish sample sizes needed to overcome the large number of zeros that come from individual-level proportional similarity measures based on diet ‘snapshots.’ But this is a well-known issue with diet analyses and even with our somewhat modest sample sizes we were able to show low levels of interspecific diet overlap with confidence.
So what? - Are these linkages important? Well, this is a bit more difficult to answer. In some ways the linkages that we illustrated in our paper might not be that important for overall ecological function. Surely not every ecological process is fundamentally important. That said, there is some indication from other studies that the flow of material and energy across system boundaries can be important for those ecosystems receiving that flux. Generally, it seems that the importance of allochthonous flux increases with decreasing productivity. Examples include marine subsidies to nutrient poor, or water-limited ecosystems in which the flux of seaweed and other detritus can fuel a diversity of structuring forces (other Anolis studies here, here and here). In our case, the different habitats that we investigated appear more or less similar (and high) in productivity making this dynamic less apparent. However, there is another way that cross-system linkages are thought to be important – stability. There is a body of theory (coming out of early collaborations between Gary Polis and Gary Huxel and now led by Kevin McCann and Neil Rooney) suggesting that cross-boundary flows between habitats can have stabilizing effects on higher order systems. The specific predictions vary depending on the particulars of what is flowing back and forth as well as the magnitude and temporal variation of the flux, but there is reason to believe that diverse (and weak) trophic linkages may help stabilize ecosystems. So what do anoles have to do with ecosystem stability? Well, this might be a bit of a stretch, but if the incredible ecological diversity of anoles results in equally diverse trophic linkages between food webs, then it seems to me that there is the potential for anoles – and other diverse faunas – to stabilize ecosystems by linking lower-order processes and structures. More research, especially empirical study, is needed to evaluate these predictions. Again, this is only speculation, but it’s something exciting to think about.
For more information on these types of linkages and their functional consequences it’s worth looking into a few different sources. Below are a few good papers on the topic.
Baxter, C.V., Fausch, K.D., Murikami M. & Chapman P.L. 2004. Nonnative stream fish invasion restructures stream and forest food webs by interrupting reciprocal prey subsidies. Ecology, 85, 2656–2663.
Nakano, S., & Murakami, M. (2001) Reciprocal subsidies: dynamic interdependence between terrestrial and aquatic food webs. Proceeding of the National Academy of Sciences, 98, 166–170.
Polis, G.A., Power, M.E. & Huxel, G.R. (2004) Food webs at the landscape level. University of Chicago Press, Chicago, Illinois, USA.
Rooney, N. & McCann, K.S. (2011) Integrating diversity, food web structure and stability. – Trends Ecology and Evolution, 27, 40-46.
Pringle, R.M., & Fox-Dobbs, K. (2008) Coupling of canopy and understory food webs by ground-dwelling predators. Ecology Letters, 11, 1328–1337
Vadeboncoeur, Y., Vander Zanden, M.J., & Lodge, D.M. (2002) Putting the lake back together: reintegrating benthic pathways into lake food web models. Bioscience, 52, 44–54.