The role of complementarity and competition in ecosystem responses to variation in plant diversity

Ecology, March, 1998 by David U. Hooper

INTRODUCTION

We know relatively little about how biological diversity affects the way ecosystems work, either in terms of processes at the ecosystem level, like primary production and nutrient cycling, or in terms of the long term sustainability of natural and managed ecosystems (Schulze and Mooney 1993, United Nations Environmental Programme 1995). Differences in plant species richness could affect ecosystem processes through partitioning of resources (Trenbath 1974, Harper 1977, Vandermeer 1990), whereby plants in more diverse communities may increase total resource capture, and thus increase net primary production. Such complementary resource use may occur in space, in time, or in types of resources used (Ewel 1986). Species that are deeply rooted have access to water and nutrients not available to more shallowly rooted species (Berendse 1979, Berendse 1982, Chapin et al. 1996). Differences in shoot architectures may allow species to fill aboveground space more efficiently with leaves, increasing overall leaf-area index and light interception (Vandermeer 1990, Naeem et al. 1994, Naeem et al. 1995, Tilman et al. 1996). Phenological differences may allow species to harvest resources at different times of the year (Steiner 1982, Gulmon et al. 1983). Different species may also utilize different nutrient sources, e.g., plants with symbiotic nitrogen-fixing bacteria, plants able to utilize organic nitrogen (Chapin et al. 1993, Kielland 1994, Northup et al. 1995), and plants with mycorrhizal mutualisms that allow greater access to organically bound phosphorus (Hogberg 1989, Perry et al. 1989). While these examples illustrate potential axes for differentiation, plants are also known to compete strongly for a relatively limited suite of resources (Tilman 1988). Relatively few experiments have explicitly tested how primary productivity responds to plant diversity in natural ecosystems (McNaughton 1977, McNaughton 1985, Naeem et al. 1994, Tilman and Downing 1994, Naeem et al. 1995, Tilman et al. 1996), and none have directly tested the degree to which complementary resource use is involved.

Whether or not complementary resource use leads to increased productivity in natural systems will depend on several factors, including the relative efficiencies of species in converting resources into biomass, differences in allocation (e.g., to growth, defense, storage, or reproduction; Chapin et al. 1996), and the degree of complementary and competitive interactions among species. Partial complementarity may increase productivity, at least marginally. Reduced competition by this mechanism has been suggested to be a primary reason for improved yields in intercropping (Vandermeer 1990), as well as allowing for species coexistence in diverse natural communities (Grubb 1977, Fowler 1982, Bazzaz 1987, McKane and Grigal 1990). However, because the resource requirements of all plants are fairly similar, the effects of complementary resource use on yearly productivity at the scale of alpha-diversity are thought to saturate at relatively low species richness (Swift and Anderson 1993, Vitousek and Hooper 1993, Field 1995).

In addition to complementary resource use, productivity may increase with increasing diversity because some species can aid the growth of others, either through provision of resources or amelioration of harsh environmental conditions (Bertness and Callaway 1994). Such facilitation is common for nitrogen fixers, which can increase resource availability for other species (Vitousek and Walker 1989, Vandermeer 1990). In other instances, canopy trees and shrubs can facilitate growth of understory species by positive effects on soil moisture, nutrients, and microclimate in both natural and complex agricultural systems (Beets 1982, Steiner 1982, Ewel 1986, Caldwell et al. 1991, Bertness and Callaway 1994). Pathogen/herbivore protection (by various mechanisms), enhanced pollination, and structural attributes could all play facilitative roles for certain species at certain times (Trenbath 1974, Vandermeer 1990, Bertness and Callaway 1994, Lawton and Jones 1995). However, these interactions may change over the course of time (both developmentally and successionally) or with plant density (Fowler 1986, Vandermeer 1990, Chapin et al. 1996). Whether facilitative and complementary interactions that increase productivity are a general attribute of increasing species diversity or are idiosyncratic properties of certain mixtures remains an important question (Morgan et al. 1992, Naeem et al. 1995).

Complementary resource use has received much attention in the literature on competition and intercropping (e.g., Trenbath 1974, Fowler 1982, Steiner 1982, Berendse 1983, Vandermeer 1988, McKane and Grigal 1990, Vandermeer 1990, Hetrick et al. 1994). However, most agricultural experiments use only two or three species. Of these, one is often a nitrogen fixer, a functional type for which little doubt remains as to its positive effects on growth of other species (Vandermeer 1990). Despite indications of complementarity, few experimental mixtures other than those with N fixers significantly increase primary production above that of the most productive monoculture, and almost none do so consistently (Trenbath 1974, Berendse 1979, Berendse 1982, Fowler 1982).