David A. Wedin

Species effects on nitrogen cycling: a test with perennial grasses.

SummaryTo test for differing effects of plant species on nitrogen dynamics, we planted monocultures of five perennial grasses (Agropyron repens, Agrostis scabra, Poa pratensis, Schizachyrium scoparium, and Andropogon gerardi) on a series of soils ranging from sand to black soil. In situ net N mineralization was measured in the monocultures for three years. By the third year, initially identical soils under different species had diverged up to 10-fold in annual net mineralization. This divergence corresponded to differences in the tissue N concentrations, belowground lignin concentrations, and belowground biomasses of the species. These results demonstrate the potential for strong feedbacks between the species composition of vegetation and N cycling. If individual plant species can affect N mineralization and N availability, then competition for N may lead to positive or negative feedbacks between the processes controlling species composition and ecosystem processes such as N and C cycling. These feedbacks create the potential for alternative stable states for the vegetation-soil system given the same initial abiotic conditions.

Influence of Nitrogen Loading and Species Composition on the Carbon Balance of Grasslands

In a 12-year experimental study of nitrogen (N) deposition on Minnesota grasslands, plots dominated by native warm-season grasses shifted to low-diversity mixtures dominated by cool-season grasses at all but the lowest N addition rates. This shift was associated with decreased biomass carbon (C):N ratios, increased N mineralization, increased soil nitrate, high N losses, and low C storage. In addition, plots originally dominated by nonnative cool-season grasses retained little added N and stored little C, even at low N input rates. Thus, grasslands with high N retention and C storage rates were the most vulnerable to species losses and major shifts in C and N cycling.

Competition Among Grasses Along a Nitrogen Gradient: Initial Conditions and Mechanisms of Competition

We grew four perennial grass species (Poapratensis, Agropyron repens, Agros- tis scabra, and Schizachyrium scoparium) for 5 yr in monocultures and in pairwise com- petition plots on an experimental nitrogen gradient. The gradient contained plots ranging from 100% sand to 100% black soil, plus plots that received additional N fertilizer. To examine the impact of initial conditions on the long-term outcome of interspecific com- petition, three competitive situations were created: seed vs. seed competition (both species planted simultaneously), seed invasions (each species added as seed to year-old monocul- tures of the other), and vegetative invasions (dividers separating adjacent monocultures of two species removed after 1 yr). Extractable soil NO3 and NH4, were measured to test if species differences in the concentration of available soil N in monoculture (i.e., R* for N, Tilman 1982) could predict the long-term outcome of competition. By year 5, Schizachyrium displaced or greatly reduced the biomass of both Poa and Agropyron on the soil mixture gradient (the mixed soils but not the added-N plots) inde- pendent of the wide range of starting conditions. On these soils, Schizachyrium monocul- tures had significantly lower soil concentrations of both NO3 and NH4, than either Poa or Agropyron monocultures. Similarly, Agropyron displaced or greatly reduced the biomass of Agrostis by year 5. Agropyron monocultures had significantly lower concentrations of N03- and NO3 + NH4+, but not NH4+, than Agrostis monocultures. In contrast, no competitive displacement occurred in competition between Poa and Agropyron, and initial differences persisted over 5 yr. Monocultures of these two species did not differ in NO3 concentration, but did differ for NH4+ and NO3 + NH4+. Thus, species differences in ability to deplete soil NO3 successfully predicted the outcome of competition for all four species pairs on the soil mixture gradient. If resource preemption or asymmetric compe- tition had been the mechanism of competition, initial conditions would have affected the long-term outcome of competition. Rather, these results support the R* (i.e., resource reduction) model for competition for soil N. In the added-N fertilizer plots, Schizachyrium had decreased biomass in competition with both Poa and Agropyron. However, neither Agropyron nor Poa appeared to have an advantage when they competed with each other in the added-N plots. For these three species pairs, the 5-yr results of competition in the added-N plots, which had greatly reduced light availability because of increased production and litter accumulation, depended on initial conditions. In the fourth pair, Agrostis was displaced by Agropyron in all competition treatments in the added-N plots. Thus, we cannot reject the hypothesis that resource preemption (i.e., asymmetric competition) is important in light competition.

PLANT TRAITS AND RESOURCE REDUCTION FOR FIVE GRASSES GROWING ON A NITROGEN GRADIENT

Five grass species (Agrostis scabra, Agropyron repens, Poa pratensis, Schiz- achyrium scoparium and Andropogon gerardi) were grown in monoculture for 3 yr on an experimental nitrogen gradient. The species differed significantly in the levels to which they reduced soil solution (0.01 mol/L KCl extractable) nitrate and ammonium concentrations and light penetration to the soil surface. Soil nitrate concentration was an inverse function of root mass, which explained 73% of the observed variance in nitrate. Other species differences explained an additional 9.2%, and total soil N an additional 5% of this variance. Extractable soil ammonium also depended on these variables, but total soil N explained the most variance. Light penetration to the soil surface in these monocultures was a negative exponential function of aboveground biomass (R2 = 0.79). Schizachyrium and Andropogon, the species that reduced soil solution N to the lowest levels on infertile soils, had lower vegetative growth rates, higher root allocation, lower reproductive allocation, and lower tissue N than the other species. Many of these traits are associated with plants of infertile habitats, suggesting a direct link between ecophysiology, resource reduction, and distributional patterns. Because all species survived on even our most nitrogen-poor soil (subsurface sand), differential nutrient reduction, not tolerance, may be the main mechanism favoring these traits in infertile habitats. On infertile soils, the three earlier successional species (Agrostis, Agropyron, and Poa) allocated more to reproduction (rhizome or seed) than the later successional species, but did not reduce soil solution nitrate and ammonium to as low levels. This suggests that our early successional species may be superior colonists but inferior nitrogen competitors compared to the prairie bunchgrasses. Our results can be used to make explicit predictions as to the outcome of nitrogen competition among all possible combinations of these five species.

Plant and Soil Controls on Mycorrhizal Fungal Communities

A field experiment was conducted to examine the relative importance of soil factors and plant species on communities of vesicular-arbuscular mycorrhizal (VAM) fungi. Populations of VAM fungal spores were studies in 4-yr-old monocultures of five successional grass species grown in gradient of soil mixtures ranging from pure subsurface sand to pure sandy loam topsoil. A total of 19 species of VAM fungi were found across all treatments. Of the 12 most abundant VAM fungal species, 6 species had a significant dependence on both soil mixture and host species, while 2 were dependent only on soil and 2 only on host. To our knowledge, these are the first results indicating that even closely related hosts (five grasses) may cause divergence in VAM fungal communities on initially identical soils. Cluster analysis of the similarity of fungal communities by host plant species showed that fungal communities in the two late successional grasses to be most similar to one another and least similar to the fungal communities in the early successional grass species. Cluster analysis of the similarity of fungal communities by soil mixture showed the fungal communities in the sandy end of the soil gradient diverged predictably from the fungal communities in the black soil end of the gradient. These results support the hypothesis that soil factors and plant species may be of equal importance in regulating the species composition of VAM fungal communities.

FIRE AND VEGETATION EFFECTS ON PRODUCTIVITY AND NITROGEN CYCLING ACROSS A FOREST-GRASSLAND CONTINUUM

Mixed tree-grass vegetation is important globally at ecotones between grass- lands and forests. To address uncertainties vis-a`-vis productivity and nitrogen (N) cycling in such systems we studied 20 mature oak savanna stands, ranging from 90% woody dominated to 80% herbaceous dominated, growing on comparable soils in a 32-yr-old fire frequency experiment in Minnesota, USA. Fire frequencies ranged from almost annual burning to complete fire protection. Across all stands, aboveground net primary productivity (ANPP) ranged from 2 to 12 Mg·ha 21 ·yr 21 , decreased with fire frequency (r 2 5 0.59), increased with woody canopy dominance (r 2 5 0.83), and increased with soil net N min- eralization rates (r 2 5 0.79), which varied from 25 to 150 kg·ha 21 ·yr 21 . ANPP was positively related to total biomass (r 2 5 0.95), total canopy leaf N content (r 2 5 0.88), leaf area index (LAI; r 2 5 0.87), annual litterfall N cycling ( r 2 5 0.70), foliage N concentration (r 2 5 0.62), and fine root N concentration (r 2 5 0.35), all of which also increased with increasing tree canopy cover. ANPP, soil N mineralization, and estimated root turnover rates increased with woody canopy cover even for stands with similar fire frequency. ANPP and N min- eralization both decreased with fire frequency for stands having a comparable percentage of woody canopy cover. Fine root standing biomass increased with increasing grass dom- inance. However, fine root turnover rate estimated with a nitrogen budget technique de- creased proportionally more with increasing grass dominance, and hence fine root produc- tivity decreased along the same gradient. Via several direct and indirect and mutually reinforcing (feedback) effects, the com- bination of low fire frequency and high tree dominance leads to high rates of N cycling, LAI, and productivity; while the opposite, high fire frequency and high grass dominance, leads to low rates of N cycling, LAI, and productivity. Carbon and N cycling were tightly coupled across the fire frequency and vegetation type gradients.

Dynamics of nitrogen competition between successional grasses.

Pairwise competition experiments were performed for 3 yr on an experi- mental nitrogen gradient at Cedar Creek Natural History Area, Minnesota, where N is the major limiting resource during early succession. Agrostis scabra, an early successional grass, competed against another early (Agropyron repens), a later (Schizachyrium scoparium), and an even later (Andropogon gerardi) successional species. On low N soils, Agrostis was competitively displaced by each of the later successional species, but persisted with Agro- pyron. On high N soils, Agrostis was displaced by all three of the other species. The inferior competitive ability for N of the early successional species refutes the resource ratio hy- pothesis of succession. Rather, the high allocation of Agrostis to seed and its rapid colo- nization of fields support a colonization-competition hypothesis of succession. For two of three cases, the outcome of competition on low nitrogen soils was predicted by R*, the nitrogen concentration to which monocultures of each species reduced extractable soil nitrate and ammonium on N-limited soils. In these cases, the species with the signif- icantly lower R* for nitrogen displaced the other species. In the third case, the species had more similar R* values, and the species with the lower R* had not displaced the other species within 3 yr (but it had done so after 5 yr; see Note added in proof).

Predominance of ecophysiological controls on soil CO2 flux in a Minnesota grassland

Ecosystem studies often study soil CO2 flux as a function of environmental factors, such as temperature, that affect respiration rates by changing the rate of utilization of carbon substrates. These studies tend not to include factors, such as photosynthesis, that affect the supply of carbon substrates to roots and root-associated processes. We examined the role of decreased carbohydrate source on soil CO2 flux and root respiration in an annually-burned grassland through manipulations of light intensity and removal of above ground biomass. We also quantified the contribution of root respiration to soil CO2 flux by measuring the respiration rates of excised roots. Two days of shading caused a 40% reduction in soil CO2 flux, while clipping was associated with a 19% reduction in soil CO2 flux. Both reductions were independent of soil and air temperature at the time of measurement. The relative decrease in soil CO2 flux observed in the clipping experiment was similar in magnitude to an observed decrease in root respiration per gram of root, linking decreased root activity and soil CO2 flux. From these experiments, we conclude that variation in factors that affect carbon availability to roots can be important determinants of soil CO2 flux and should be included explicitly in studies that measure or model soil CO2 flux.

SOIL CARBON, NUTRIENTS, AND MYCORRHIZAE DURING CONVERSION OF DRY TROPICAL FOREST TO GRASSLAND

Wildfires and alien grass invasion threaten dry tropical forests throughout Central America. Efforts to preserve and restore these forests will require a better under- standing of how conversion to grassland changes key belowground processes and organisms such as soil organic matter, nutrient cycling, and mycorrhizae. We studied forest, edge, and grassland soils from five 60-m transects perpendicular to abrupt forest-grassland boundaries in Guanacaste Province, Costa Rica. Nutrient concentrations, N mineralization dynamics, and mycorrhizal fungal communities were compared across vegetation type (forest, edge, and grassland). The dynamics of N mineralization were measured in year-long laboratory incubations, and the diversity of mycorrhizal fungal communities was assessed from pop- ulations of soil-borne spores. Soil C, N, and K were lower, while many base cations and micronutrients were higher in grassland plots than in forest plots. Although differences in the quantity of total soil C and N occurred mainly in the forest-to-edge transition, differences in the quality of soil organic matter, as reflected by soil C:N ratios and mineralization rates, occurred in the edge-to-grassland transition. Beta diversity of mycorrhizal spore commu- nities (measured by Sorensons similarity index) was lower in the grassland plots than in the forest plots, indicating that grass invasion had caused some convergence. However, total spore density and alpha diversity of mycorrhizal spore communities (measured by species richness and Simpsons diversity index) were not altered by wildfires and grass invasion. These results suggest that persistence and regeneration of forest plant species in the grasslands may not be constrained to a significant degree by the lack of mycorrhizal symbionts. These grasslands appear to be sustainable, alternative stable states for these areas. Positive feedbacks between the alien grassland vegetation and both fire and nutrient cycling maintain and reinforce this alternative state.

Relationship between the structure of root systems and resource use for 11 North American grassland plants

Eleven Midwest North American grassland plant species differed in theirconstruction, production, and placement of fine and coarse belowground biomassin the soil profile after having been grown in containers in the field for twoand a half growing seasons. Based on the patterns of root system structure andresource utilization, the species we examined could be classified as 1)legumes,2) high-nitrogen rhizomatous C3 species, and 3) a separategradient of differentiation from tall- to short-statured species(i.e. tallgrass to shortgrass species). Legumes depleted water evenlythroughoutthe soil profile, with little capacity for acquisition of inorganic nitrogenthroughout the 1m soil profile. The three rhizomatous species had shallow fineroot distributions, a large relative investment in shallow rhizomes, andmoisture and NO3− levels were low in shallow soils,but high at depth. Tallgrass species maintained a large standing root biomassofhigh-density, low-nitrogen fine roots, and acquire nitrogen andwater from a large, deep volume of soil, in which inorganic nitrogen is presentin low concentrations. Root systems ofshortgrass species lacked coarse belowground biomass, had fine roots that werefiner than those of the tallgrass species, and had a shallow root distribution.There was little support for functional dichotomies between the C3and C4 species or between the grasses and forbs. For example,Solidago rigida (C3 forb) andAndropogon gerardii (C4 grass) were moresimilarto each other than to other C3 forbs or C4 grasses,respectively.Across all species and depths examined, there were strong relationships betweenthe amount of fine root biomass present in a unit of volume of soil and thedepletion of soil water and nitrogen, but there were no relationships withcoarse belowground biomass. This reaffirms that differentiation of coarse andfine root biomass is as important as differentiating stems and leaves inevaluating plant allocation and ecosystem functioning.