Literature Review: Biodiversity, Evolution and Time

by Carl Strang

Today I hope to illustrate how different studies can be made to illuminate one another. The starting point is a paper that I think has the potential to be profoundly influential in the field of community ecology.

Lowland tropical forests are renowned for their high biodiversity. They also are old.

Lowland tropical forests are renowned for their high biodiversity. They also are old.

Reich, Peter B. 2012. Impacts of biodiversity loss escalate through time as redundancy fades. Science 336:589-592.

Studies of biodiversity in plant communities have suggested that the number of species climbs to a saturation point, after which additional species do not add more biomass productivity and apparently could be removed without affecting ecosystem function. In contrast, this long-term study shows that the added species refine their niches over time, and by sorting out and separating niches they eventually produce a unique portion of niche space for every species.

This result reminds us that our focus on the day-to-day or year-to-year dynamics of ecosystems needs to be tempered by the fact that ecological roles are flexible in evolutionary time. Species that begin as competitors, but which persist together, can subdivide whatever resources are the basis for their competition, as selective pressures nudge them apart. This process provides a mechanism which in part explains the results of the next featured study.

Jetz W, Fine PVA (2012) Global Gradients in Vertebrate Diversity Predicted by Historical Area-Productivity Dynamics and Contemporary Environment. PLoS Biol 10(3): e1001292. doi:10.1371/journal.pbio.1001292

(Also interpreted in an accompanying commentary by another author). They found that 80% of the variation in vertebrate species diversity between different terrestrial biomes is accounted for by a combination of ecosystem area, age, productivity and temperature, with the highest diversity in warm tropical forests. Thus evolutionary as well as ecological factors are important. Area alone provided a poor fit. Age was considered up to 55 million years, and age combined with area improved the fit greatly (for instance, grassland ecosystems are extensive but relatively young at 8 million years). Productivity and temperature separated deserts from tropical forests, for example, and further improved the model. Productive areas have more individuals, and thus greater potential for evolutionary diversification, and also have greater ecological space for niche diversification. Finally, they looked at ecological influences by dividing biomes into finer-grained divisions (down to a 110 km grid), and found that this further refined the results, again by variation in productivity of the smaller areas. This points to ecological interactions having an influence on local biodiversity.

The connection to the ideas in the first paper should be clear. The connection to the next one is less obvious.

Franzén M, Schweiger O, Betzholtz P-E (2012) Species-Area Relationships Are Controlled by Species Traits. PLoS ONE 7(5): e37359. doi:10.1371/journal.pone.0037359

They studied Lepidoptera on islands off the coast of Sweden to consider species richness-island area relationships and the niche and other ecological or physiological traits that may contribute to the overall pattern. They found a number of traits that had an impact on species-area relationships, and in general these made ecological sense. Examples of traits which were particularly sensitive to island area were low reproductive potential, small range size, narrow diet breadth, and low abundance. Thus classical island biogeography, which simply measures species area relationships and considers gross immigration and extinction rates, is improved in its explanatory power by consideration of traits that contribute to ease of immigration and resistance to extinction.

Here the communities under consideration are on islands, and once immigration has occurred the stage is set for the kinds of ecological-evolutionary processes outlined in the first paper above. This produces phenomena like the famous Darwin finches of the Galapagos, and innumerable other examples.

Moreno-Mateos D, Power ME, Comín FA, Yockteng R (2012) Structural and Functional Loss in Restored Wetland Ecosystems. PLoS Biol 10(1): e1001247. doi:10.1371/journal.pbio.1001247

Schmitz OJ (2012) Restoration of Ailing Wetlands. PLoS Biol 10(1): e1001248. doi:10.1371/journal.pbio.1001248

These two papers provide an interesting contrast in interpretations. The first is a review of studies in which wetland restorations were compared to undisturbed baselines. The authors concluded that restoration has some success but falls significantly short of the return to ecosystem structure and function that is the goal of restoration. Thus they caution that wetland destruction should not be regarded as something that can be mitigated by restoration. Schmitz looks at their results and comes to the opposite conclusion, suggesting that wetland restoration is a thoroughly successful process, but makes no attempt to reconcile his conclusion with theirs.

Here the value of the first paper at the top of this post is to remind us that there is a pitfall in thinking about results over too narrow a time frame. Restoration is valuable and good, but in the short term it cannot be thought of as completely replacing the original. There is no way to re-create all the intricacies of the co-evolved relationships Reich wrote about. It’s much better to preserve the original. Nevertheless, restoration is worth doing, and given enough time there is hope that something as good as the original will evolve. If we can at least preserve as many species as possible, that will minimize the time needed for such a dynamic endpoint to be reached.


Literature Review: Oaks as Islands

by Carl Strang

One paper in the journal Ecology this year caught my eye because of its relevance to my study of leaf miners in maples (Barber, Nicholas A., and Robert J. Marquis. 2011. Leaf quality, predators, and stochastic processes in the assembly of a diverse herbivore community. Ecology 92:699-708). They studied white oaks in Missouri.

White oak leaves

Barber and Marquis looked at leaf chewing herbivores on accessible lower tree branches, as I have done in sugar/black maples, but with much greater elaboration. They grouped the species they found into guilds based on their feeding style: free-feeding (e.g., chewers of holes or gouges in leaves), shelter building (insects that roll leaves or tie them with silk to form little hiding places), leaf-mining (tiny insects that live between the upper and lower surfaces of the leaves), and arthropod predators. They recognized separate herbivore assemblages divided by season with peaks in May, early July, and late August-early September, and found little or no species overlap between these periods.

They assessed predation by birds, using exclosures to prevent avian access to some branches, and found that birds had a negligible impact on insect abundance or community structure.

They also measured the nutritional value of the individual trees, and found that most of the invertebrate guilds preferred plants with high nitrogen and low tannin levels. This was especially true late in the season. Higher quality plants had larger communities, and sometimes higher species counts. On the other hand, similarity between communities on different trees was not based on similarity of nutritional value, but rather on how far apart the trees were. They concluded that community species composition is driven more by stochastic processes (the likelihood that dispersing insects will find a tree) than by host plant quality. This tied their study to the venerable body of research on island biogeography. Their insects behave as though these white oaks are islands, with some trees more remote than others and thus less likely to be reached by the less effective dispersers.

Literature Review: Population Edges

by Carl Strang

Today I want to share my notes from a couple papers published last year in the open on-line scientific journal PLoS ONE (Public Library of Science). These studies looked at the geography of populations. The first considered the ecological factors that determine the edges of a species’ range [Rhainds M, Fagan WF (2010) Broad-Scale Latitudinal Variation in Female Reproductive Success Contributes to the Maintenance of a Geographic Range Boundary in Bagworms (Lepidoptera: Psychidae). PLoS ONE 5(11): e14166. doi:10.1371/journal.pone.0014166].

Here is a bagworm case, formerly the home of a female but now containing her eggs, in a crabapple tree at Mayslake Forest Preserve.

They looked at bagworm populations from Tennessee to Michigan to determine what limits their range. Caterpillars of this moth live in protective cases that they build around themselves. Wingless females mate in fall with flighted males, lay eggs in their bags, then drop to the ground and die. The eggs hatch in the spring. Female mating success dropped from near 100% to near 0% as the edge of the range was reached (apparently as the population thins out, males have a hard time finding females). Other factors were declines in fecundity, egg and pupal survivorship (all perhaps attributable to stresses resulting from the northern climate). The population density thins and local extinctions become more frequent toward the edge of the range.

Here’s another bagworm egg case, this one on a willow at Willowbrook Forest Preserve.

This study was of personal interest because I have been looking at several instances of range extensions in singing insects.

The other paper reported a study of mammal communities in forest fragments in Brazil [Pardini R, Bueno AdA, Gardner TA, Prado PI, Metzger JP (2010) Beyond the Fragmentation Threshold Hypothesis: Regime Shifts in Biodiversity Across Fragmented Landscapes. PLoS ONE 5(10): e13666. doi: 10.1371/journal.pone.0013666].

This scene is from Panama, not Brazil, but may be similar to the kind of landscape Pardini and company were studying.

Community ecologists are concerned about what happens to species diversity as large blocks of habitat become broken into smaller pieces by human activity. These scientists found a landscape-wide threshold in total forest coverage below which species went extinct throughout the system. In other words, local extinction in island-like patches of habitat no longer could be replenished through immigration from other patches because there weren’t enough of these to serve as sources. I found similar patterns in my study of periodical cicadas in DuPage County in 2007.

Where the Periodical Cicadas Weren’t

by Carl Strang

In my earlier description of the 17-year cicada emergence in northeast Illinois, I showed how the periodical cicadas appeared in large numbers in parts of eastern DuPage County that either (1) were well forested for a long period of time; or (2) were associated with the rail lines where communities of Chicago commuters built up early, establishing urban forests that allowed the spread of the insects from source forest areas. Below is a map of presettlement DuPage, and it shows that potential source forest areas were as much a presence in western DuPage as in the eastern half of the county.


(map prepared by Rachel Reklau)

(map prepared by Rachel Reklau)





So, why weren’t cicadas there as well in 2007? My inquiry into this question took me to three kinds of sources: old newspaper accounts, old county atlases and other historical books, and a series of aerial photos covering the entire county in 1939.


Here are some quotes from the Wheaton Illinoian newspaper. Bartlett, 1888: “The locusts are singing so loud we can’t hear the frogs.” Turner Junction (now West Chicago), 1888: “The locusts are in full battle array, and will soon get in their work.” Downers Grove, 1888 (eastern DuPage, but a worthy quote): “If it’s locusts you want you can have them by the scoop shovel full.” Wheaton, 1905: “There is no longer any doubt about this being the seventeen-year locust date. They are hatching out in countless numbers.” None of the west DuPage towns had cicadas emerging beyond a handful of scattered individuals in 2007. Clearly they once were there.


So, I followed the 18 significant forest areas of the original land survey, measuring how they changed from their 1840 size to their size in the 1874 county atlas to the 1939 aerial photos to their size today. The pattern was dramatic, and clear. All forests were cut back for agriculture and development, but the process was delayed and less dramatic in eastern DuPage County where there was an earlier transition from agriculture to a suburban landscape. Chicago and the railways made the difference. By the time the eastern forest blocks were significantly diminished, the urban forests had grown and the periodical cicadas had spread into them.




In western DuPage County, the forests were cut back severely before urban forests of any size could grow, and before lands were protected by forest preserves and parks. County wide, the median forest depletion was 33% by 1874, 70% by 1939, with the greatest impact in western DuPage. I looked at the bottlenecks, the smallest known sizes to which the forest areas were trimmed since 1840, and compared them for presence and absence of cicadas in 2007. Forests that always have remained above 62 hectares in size all had cicadas. Forests that at one time or another were reduced below 50 hectares all were lacking cicadas. Between those sizes, there were 3 forest areas. Two of them had cicadas, and were relatively close to other forest remnants that had cicadas in 2007. The third, which had only a handful of periodical cicadas, was an isolated 7.7 kilometers (nearly 5 miles) from the nearest forest with cicadas.


So, now we have an accounting for the distribution of periodical cicadas in the county. Understanding them required a study of presettlement conditions, railroad history, suburban history and agricultural history.


But that is not the whole story. There were more surprises in 2007, and I will get into them in future entries.

%d bloggers like this: