Elizabeth O'neill

NET PRIMARY PRODUCTIVITY OF A CO2‐ENRICHED DECIDUOUS FOREST AND THE IMPLICATIONS FOR CARBON STORAGE

A central question concerning the response of terrestrial ecosystems to a changing atmosphere is whether increased uptake of carbon in response to increasing at- mospheric carbon dioxide concentration results in greater plant biomass and carbon storage or, alternatively, faster cycling of C through the ecosystem. Net primary productivity (NPP) of a closed-canopy Liquidambar styraciflua (sweetgum) forest stand was assessed for three years in a free-air CO2-enrichment (FACE) experiment. NPP increased 21% in stands ex- posed to elevated CO2, and there was no loss of response over time. Wood increment increased significantly during the first year of exposure, but subsequently most of the extra C was allocated to production of leaves and fine roots. These pools turn over more rapidly than wood, thereby reducing the potential of the forest stand to sequester additional C in response to atmospheric CO2 enrichment. Hence, while this experiment provides the first evidence that CO2 enrichment can increase productivity in a closed-canopy deciduous forest, the implications of this result must be tempered because the increase in productivity resulted in faster cycling of C through the system rather than increased C storage in wood. The fate of the additional C entering the soil system and the environmental interactions that influence allocation need further investigation.

Responses of soil biota to elevated atmospheric carbon dioxide

Increasing concentrations of atmospheric CO2 could have dramatic effects upon terrestrial ecosystems including changes in ecosystem structure, nutrient cycling rates, net primary production, C source-sink relationships and successional patterns. All of these potential changes will be constrained to some degree by below ground processes and mediated by responses of soil biota to indirect effects of CO2 enrichment. A review of our current state of knowledge regarding responses of soil biota is presented, covering responses of mycorrhizae, N-fixing bacteria and actinomycetes, soil microbiota, plant pathogens, and soil fauna. Emphasis will be placed on consequences to biota of increasing C input through the rhizosphere and resulting feedbacks to above ground systems. Rising CO2 may also result in altered nutrient concentrations of plant litter, potentially changing decomposition rates through indirect effects upon decomposer communities. Thus, this review will also cover current information on decomposition of litter produced at elevated CO2.SummaryPredictably, the responses of soil biota to CO2 enrichment and the degree of experimental emphasis on them increase with proximity to, and intimacy with, roots. Symbiotic associations are all stimulated to some degree. Total plant mycorrhization increases with elevated CO2. VAM fungi increase proportionately with fine root length/mass increase. ECM fungi, however, exhibit greater colonization per unit root length/mass at elevated CO2 than at current atmospheric levels. Total N-fixation per plant increases in all species examined, although the mechanisms of increase, as well as the eventual benefit to the host relative to N uptake may vary. Microbial responses are unclear. The assumption that changes in root exudation will drive increased mineralization and facilitate nutrient uptake should be examined experimentally, in light of recent models. Microbial results to date suggest that metabolic activity (measured as changes in process rates) is stimulated by root C input, rather than population size (measured by cell or colony counts). Insufficient evidence exists to predict responses of either soil-borne plant pathogens or soil fauna (i.e., food web responses). These are areas requiring attention, the first for its potential to limit ecosystem production through disease and the second because of its importance to nutrient cycling processes. Preliminary data on foliar litter decomposition suggests that neither nutrient ratios nor decomposition rates will be affected by rising CO2. This is another important area that may be better understood as the number of longer term studies with more realistic CO2 exposures increase. Evidence continues to mount that C fixation increases with CO2 enrichment and that the bulk of this C enters the belowground component of ecosystems. The global fate and effects of this additional C may affect all hierarchical levels, from organisms to ecosystems, and will be largely determined by responses of soil biota.

Nitrogen resorption in senescing tree leaves in a warmer, CO2-enriched atmosephere

The prediction that litter quality, and hence litter decomposition rates, would be reduced when plants are grown in a CO2-enriched atmosphere has been based on the observation that foliar N concentrations usually are lower in elevated [CO2]. The implicit assumption is that the N concentration in leaf litter reflects the N concentration in green leaves. Here we evaluate that assumption by exploring whether the process of seasonal nutrient resorption is different in CO2-enriched plants. Nitrogen resorption was studied in two species of maple trees (Acer rubrum L. and A. saccharum Marsh.), which were planted in unfertilized soil and grown in open-top chambers with ambient or elevated [CO2] in combination with ambient or elevated temperature. In the second growing season, prior to autumn senescence, individual leaves were collected and analyzed for N and dry matter content. Other leaves at the same and an adjacent node were collected for analysis as they senesced and abscised. This data set was augmented with litter samples from the first growing season and with green leaves and leaf litter collected from white oak (Quercus alba L.) saplings grown in ambient and elevated [CO2] in open-top chambers. In chambers maintained at ambient temperature, CO2 enrichment reduced green leaf N concentrations by 25% in A. rubrum and 19% in A. saccharum. CO2 enrichment did not significantly reduce resorption efficiency so the N concentration also was reduced in litter. There were, however, few effects of [CO2] on N dynamics in these leaves; differences in N concentration usually were the result of increased dry matter content of leaves. The effects of elevated [CO2] on litter N are inherently more difficult to detect than differences in green leaves because factors that affect senescence and resorption increase variability. This is especially so when other environmental factors cause a disruption in the normal progress of resorption, such as in the first year when warming delayed senescence until leaves were killed by an early frost. The results of this experiment support the approach used in ecosystem models in which resorption efficiency is constant in ambient and elevated [CO2], but the results also indicate that other factors can alter resorption efficiency.

Elevated atmospheric CO2 effects on seedling growth, nutrient uptake, and rhizosphere bacterial populations ofLiriodendron tulipifera L.

Yellow-poplar (Liriodendron tulipifera L.) seedlings were planted in unfertilized forest soil in boxes with a removable side panel and grown in atmospheres containing either ambient (367 μl l−1) or elevated (692 μl l−1) CO2. Numbers of total bacteria, nitrifiers, and phosphate-dissolving bacteria in the rhizosphere and in nonrhizosphere soil were measured every 6 weeks for 24 weeks. Seedling growth and nutrient content were measured at a final whole-plant harvest. Root, leaf, and total dry weights were significantly greater, and specific leaf area was significantly less, in 692 ml l−1 than in ambient CO2. Uptake per gram plant dry weight of N, S, and B was lower at elevated CO2, whereas uptake of P, K, Cu, Al, and Fe was proportional to growth in both CO2 treatments. Total uptake and uptake per g plant dry weight of Ca, Mg, Sr, Ba, Zn, and Mn were not affected by CO2 treatment. Bacterial populations differed due to CO2 only at the final harvest, where there were significantly fewer nitrite-oxidizers and phosphate-dissolving bacteria in the rhizosphere of seedlings grown at 692 μl l−1 CO2.