Tropical tree holes and other phytotelmata

Steve Yanoviak
Research - Tree holes / Phytotelmata

Home

Community ecology seeks to understand how abiotic and biotic factors influence species diversity within a given temporal and spatial domain. This can be particularly challenging in the tropics, where high species density (and limited taxonomic information) make biologically relevant changes in species richness difficult to detect or interpret. One way to overcome this problem while still testing interesting and relevant hypotheses is to focus on small systems with discrete boundaries. Phytotelmata (= "plant-held waters") provide the perfect solution; they are often locally abundant, they support a characteristic macrofauna, and they are fairly easy to manipulate for experiments.

Common types of phytotelmata in the neotropics include tree holes (water-filled cavities in the woody parts of trees), axils of bromeliads, fallen plant parts on the forest floor (fruit husks, palm fronds, etc.), and Heliconia bracts.

Aechmea sp., a bromeliad

water-filled husk of Lecythis sp.

Although some tree holes (e.g., in the buttresses of fallen trees) hold >50 liters of water, phytotelmata are generally small and easy to manipulate for experiments.

a small water-filled tree hole

Heliconia bracts collect rainwater and produce their own fluids

Larvae of true flies (Diptera) are ubiquitous and conspicuous in phytotelmata. The most common phytotelm dipterans are mosquitoes, including several species that are potential disease vectors. Thus, the fauna of phytotelmata are both ecologically relevant and economically important.

Aedes terrens (left) and Culex urichii (right) breed in tree holes

In the neotropics, mosquito larvae in tree holes (primarily Aedes spp. and Culex spp.) are the principal prey for several species of top predators, including larvae of giant damselflies (see Pseudostigmatidae on D. Paulson's tropical odonate site), Toxorhynchites mosquitoes, and poison dart frogs (Dendrobatidae). In Panama, Dr. Ola Fincke and I found five common predator species and only about twelve common prey species in tree holes. Given this relatively diverse suite of top predators, it seems that tropical tree holes should be under intense "top-down" control, especially if lack of refugia for mosquito larvae leads to high predator efficiency in this system. Results from tree holes (Yanoviak 2001 pdf) and similar container habitats in Panama provided some support for this hypothesis (Fincke et al. 1997 pdf, Yanoviak 1999 pdf).

Toxorhynchites sp. mosquito larva

giant damselfly larva

Leaf litter forms the base of the food web in tree holes, and the quality and quantity of this resource likely influence species richness in the system. Field experiments in Panama showed that this is true in natural and artificial tree holes, but that the type and frequency of litter inputs are potentially more important than quantity of litter input, at least over the short term (Yanoviak 1999 pdf).

a large tree hole full of leaf litter

Predation and resource availability have interactive effects on the number of species present in a tree hole. My studies in Panama showed that they function on different and somewhat counter-intuitive time scales. Specifically, odonates had small but protracted effects relative to litter inputs (Yanoviak 2001 pdf). This is complicated by the fact that litterfall intensity and predator breeding cycles vary seasonally, and different top predator species show measurable differences in their feeding behavior. Thus, all else equal, different combinations of predators at different times of the year should alter prey species richness and composition in the prey guild. Furthermore, if mosquito larvae facilitate the breakdown of leaf litter in tree holes, one would expect trophic cascades to occur in the presence of certain predators or at certain times of the year. A preliminary exper-iment did not support this hypothesis (Yanoviak 1999 pdf), but many variables remain to be tested.

Ongoing and future work on this system will examine variation in functional response curves among members of the top predator guild, and its consequences for species richness and composition in the prey assemblage. Larger-scale studies of the effects of deforestation on phytotelm abundance are in progress. For more on that, click here.

Page updated 1 April 2006

Home