James M. R. Brock, George L. W. Perry, Tynan Burkhardt & Bruce R. Burns
Tree ferns can be a common component of the understorey of tropical and southern temperate forests. In New Zealand’s broad-leaved podocarp forests they are frequently abundant in the understorey, forming near-continuous sub-canopies representing up to 50% of the stems and 21% of biomass. Tree ferns form deep, slow to decompose leaf (frond) litter, intercept high proportions (up to 50%) of sub-canopy light, and indiscriminately destroy seedlings and saplings when up to 3 m long dead fronds disconnect from the trunk and fall to the forest floor. Despite many hypotheses considering the likely effect of these processes (e.g. strong negative effects on establishment of shade-intolerant conifers), there have been no attempts to experimentally test the response of the seedling community to these potential effects.
Our research was conducted in the forests of northern New Zealand which comprise mixed broad-leaf canopies at 10–20 m with emergent conifers supporting 27 species across 13 families. Average rainfall at the sites is 1,280 mm per year, and the soils comprise volcanic sandy loams and granular clays. In our study, we first undertook a field survey to describe where seedlings are growing in the landscape and consider what might influence where they are growing including shading, depth of leaf litter, how close nearby trees and tree ferns are, and soil moisture. We then experimentally manipulated 160 tree ferns above seedling plots, to examine how tree fern litter and shading contribute to the patterns observed in the field survey.
The field survey showed that landscape-level seedling density is affected by the presence of tree ferns. Seedling densities of both angiosperms and conifers are up to 50% lower within the drip-line of a tree fern (i.e. within the area covered by the fronds projecting from the top of the tree fern); this reduction is associated with a doubling of leaf litter depth. Our field manipulation of tree ferns removed either litter and/or fronds from within and above seedlings plots. In the litter and frond removal treatment, we found that seedlings of two species of podocarp (Podocarpus totara and Phyllocladus trichomanoides) were consistently turning up, as well as there being increased seedlings densities and species richness. Our study suggests that through macro-litter and shading, tree ferns influence seedling community composition by reducing the density of canopy angiosperm and conifer species that can establish, particularly suppressing conifers in seedling communities within their drip-lines.
This is a plain language summary for the paper of Brock et al. published in the Journal of Vegetation Science.