Effects of Plant Composition and Diversity on Nutrient Cycling

We evaluated the effects of plant functional group richness on seasonal patterns of soil nitrogen and phosphorus cycling, using serpentine grassland in south San Jose, California. We established experimental plots with four functional types of plants: early-season annual forbs (E), late-season annual forbs (L), nitrogen-fixers (N), and perennial bunchgrasses (P). These groups differ in several traits relevant to nutrient cycling, including phenology, rooting depth, root:shoot ratio, size, and leaf C:N content. Two or three species of each group were planted in single functional group (SFG) treatments, and in two-, three-, and four-way combinations of functional groups. We analyzed available nutrient pool sizes, microbial biomass nitrogen and phosphorus, microbial nitrogen immobilization, nitrification rates, and leaching losses. We used an index of “relative resource use” that incorporates the effects of plants on pool sizes of several depletable soil resources: inorganic nitrogen in all seasons, available phosphorus in all seasons, and water in the summer dry season. We found a significant positive relationship between increasing relative resource use (including both plant and microbial uptake) and increasing plant diversity. The increase in relative resource use results because different functional groups have their maximum effect on different resources in different seasons: E's dominate reduction of inorganic nitrogen pools in winter; L's have a stronger depletion of nitrogen in spring and a dominant reduction of water in summer; P's have a stronger nitrogen depletion in summer; N-fixers provide additional nitrogen in all seasons and have a significant phosphorus depletion in all seasons except fall. Single functional group treatments varied greatly in relative resource use; for example, the resource use index for the L treatment is as high as in the more diverse treatments. We expected a reduction of leaching losses as functional group richness increased because of differences in rooting depth and seasonal activity among these groups. However, measurements of nitrate in soil water leached below the rooting zone indicated that, apart from a strong reduction in losses in all vegetated treatments compared to the bare treatment, there were no effects of increasing plant diversity. While some single functional group treatments differed (P ≤ L, N), more diverse treatments did not. Early- and late-season annuals, but not perennial bunchgrasses, had significant positive effects on microbial immobilization of nitrogen in short-term (24 h) 15N experiments. We conclude that: (1) total resource use, across many resource axes and including both plant and microbial effects, does increase with increasing plant diversity on a yearly timescale due to seasonal complementarity; (2) while the presence of vegetation has a large effect on ecosystem nitrogen retention, nitrogen leaching losses do not necessarily decrease with increasing functional group richness; (3) indirect effects of plants on microbial processes such as immobilization can equal or exceed direct effects of plant uptake on nutrient retention; and (4) plant composition (i.e., the identity of the groups present in treatments) in general explains much more about the measured nutrient cycling processes than does functional group richness alone (i.e., the number of groups present).