Birds are conspicuous members of grassland ecosystems, as anyone entranced by the acrobatic and vocal territorial displays of Bobolinks can attest. However, the question remains, Do birds play any sort of significant ecological roles in grasslands? Existing data are contradictory. For example, based on their contribution to annual productivity (production of offspring, expenditure of energy, etc.) in shrub-steppe (shrubby grassland), ecologist John Wiens speculated that birds may be little more than "frills" in the ecosystem, not interacting with it in any particular way.
In contrast, several studies have found that grassland birds can significantly reduce the numbers of grasshoppers. Because grasshoppers can be abundant and voracious herbivores, their suppression has the potential to free grassland plant species from a major loss of photosynthetic tissue, which in turn may allow plants to increase their own annual productivity. This potential increased productivity may support a greater abundance and diversity of animals that ultimately depend upon primary production.
To address this issue, colleagues from the University of Illinois at Chicago and I set up an experiment in an ongoing prairie restoration experiment conducted at The Morton Arboretum in Lisle, IL. The original experiment was designed to look for effects of avian and mammalian granivores on initial prairie restoration establishment (effects of both birds and mammals were found). The main focus now is on how mammalian herbivores (voles) affect community composition and productivity.
In our experiment, we erected bird exclosure cages within the replicated prairie restoration plots. The cages are constructed with half-inch-diameter rebar, PVC plumbing pipe, and monofilament nylon gill netting. The mesh of the netting is large enough for virtually all insects to access the interior of the cage, but it is small enough to exclude birds. If birds significantly reduce the numbers of insects, then presumably herbivory should be of greater magnitude inside of cages than in comparable control areas outside of cages. If this is the case, then it is also possible that plants outside of cages will experience decreased insect herbivory, and in turn may experience elevated productivities.
To sample insects, we collected them with a gas-powered vacuum eight times during the growing season of 2001, with usually about two or three weeks separating collection dates. Insects were identified to Order (e.g., Orthoptera), counted, and returned to the site of capture. To sample herbivory and its potential effect on plant productivity, we examined all the stem leaves on individual plants of the species stiff goldenrod that were both inside and outside the cages. In addition to estimating loss of leaf tissue on each stem leaf of each stem of each individual, all flower heads were counted. Later, randomly selected flower heads will be collected and seed mass estimated.
Although the data have not yet been analyzed, several tantalizing findings seem apparent. First, on some sampling dates, but not others, more insects appeared to be captured inside than outside the cages. If this is so, it suggests that the effects of birds on insect population abundances may blink on and off during the growing season (this was also found in a similar experiment conducted in an oak forest in Missouri). Second, we found that the composition of the insect community could vary considerably between consecutive sampling periods separated by only two weeks. The changing insect community composition could be both a consequence and the cause of the variable bird effect. Third, although not insects, spiders were collected and counted. Spiders, which along with birds are also insect predators, usually appeared more numerous within cages. So when insect numbers were greater within cages, this is despite what appears to be increased spider predation within cages. If all of these impressions hold, our conclusion will be that birds, despite having low annual productivities themselves, are certainly more than just frills in these experimental communities.
Christopher J. Whelan, Center for Biodiversity
Please report any problems with or suggestions about this page to: