David S. Ellsworth

GENERALITY OF LEAF TRAIT RELATIONSHIPS: A TEST ACROSS SIX BIOMES

Convergence in interspecific leaf trait relationships across diverse taxonomic groups and biomes would have important evolutionary and ecological implications. Such convergence has been hypothesized to result from trade-offs that limit the combination of plant traits for any species. Here we address this issue by testing for biome differences in the slope and intercept of interspecific relationships among leaf traits: longevity, net pho- tosynthetic capacity (Amax), leaf diffusive conductance (Gs), specific leaf area (SLA), and nitrogen (N) status, for more than 100 species in six distinct biomes of the Americas. The six biomes were: alpine tundra-subalpine forest ecotone, cold temperate forest-prairie ecotone, montane cool temperate forest, desert shrubland, subtropical forest, and tropical rain forest. Despite large differences in climate and evolutionary history, in all biomes mass-based leaf N (Nmass), SLA, Gs, and Amax were positively related to one another and decreased with increasing leaf life span. The relationships between pairs of leaf traits exhibited similar slopes among biomes, suggesting a predictable set of scaling relationships among key leaf morphological, chemical, and metabolic traits that are replicated globally among terrestrial ecosystems regardless of biome or vegetation type. However, the intercept (i.e., the overall elevation of regression lines) of relationships between pairs of leaf traits usually differed among biomes. With increasing aridity across sites, species had greater Amax for a given level of Gs and lower SLA for any given leaf life span. Using principal components analysis, most variation among species was explained by an axis related to mass-based leaf traits (Amax, N, and SLA) while a second axis reflected climate, Gs, and other area-based leaf traits.

Soil fertility limits carbon sequestration by forest ecosystems in a CO2-enriched atmosphere.

Northern mid-latitude forests are a large terrestrial carbon sink. Ignoring nutrient limitations, large increases in carbon sequestration from carbon dioxide (CO2) fertilization are expected in these forests. Yet, forests are usually relegated to sites of moderate to poor fertility, where tree growth is often limited by nutrient supply, in particular nitrogen. Here we present evidence that estimates of increases in carbon sequestration of forests, which is expected to partially compensate for increasing CO2 in the atmosphere, are unduly optimistic. In two forest experiments on maturing pines exposed to elevated atmospheric CO2, the CO2-induced biomass carbon increment without added nutrients was undetectable at a nutritionally poor site, and the stimulation at a nutritionally moderate site was transient, stabilizing at a marginal gain after three years. However, a large synergistic gain from higher CO2 and nutrients was detected with nutrients added. This gain was even larger at the poor site (threefold higher than the expected additive effect) than at the moderate site (twofold higher). Thus, fertility can restrain the response of wood carbon sequestration to increased atmospheric CO2. Assessment of future carbon sequestration should consider the limitations imposed by soil fertility, as well as interactions with nitrogen deposition.

Modeling and measuring the effects of disturbance history and climate on carbon and water budgets in evergreen needleleaf forests

The effects of disturbance history, climate, and changes in atmospheric carbon dioxide (CO2) concentration and nitrogen deposition (Ndep) on carbon and water fluxes in seven North American evergreen forests are assessed using a coupled water–carbon–nitrogen model, canopy-scale flux observations, and descriptions of the vegetation type, management practices, and disturbance histories at each site. The effects of interannual climate variability, disturbance history, and vegetation ecophysiology on carbon and water fluxes and storage are integrated by the ecosystem process model Biome-BGC, with results compared to site biometric analyses and eddy covariance observations aggregated by month and year. Model results suggest that variation between sites in net ecosystem carbon exchange (NEE) is largely a function of disturbance history, with important secondary effects from site climate, vegetation ecophysiology, and changing atmospheric CO2 and Ndep. The timing and magnitude of fluxes following disturbance depend on disturbance type and intensity, and on post-harvest management treatments such as burning, fertilization and replanting. The modeled effects of increasing atmospheric CO 2 on NEE are generally limited by N availability, but are greatly increased following disturbance due to increased N mineralization and reduced plant N demand. Modeled rates of carbon sequestration over the past 200 years are driven by the rate of change in CO2 concentration for old sites experiencing low rates of N dep. The model produced good estimates of between-site variation in leaf area index, with mixed performance for between- and within-site variation in evapotranspiration. There is a model bias

Nitrogen limitation constrains sustainability of ecosystem response to CO2

Enhanced plant biomass accumulation in response to elevated atmospheric CO2 concentration could dampen the future rate of increase in CO2 levels and associated climate warming. However, it is unknown whether CO2-induced stimulation of plant growth and biomass accumulation will be sustained or whether limited nitrogen (N) availability constrains greater plant growth in a CO2-enriched world. Here we show, after a six-year field study of perennial grassland species grown under ambient and elevated levels of CO2 and N, that low availability of N progressively suppresses the positive response of plant biomass to elevated CO2. Initially, the stimulation of total plant biomass by elevated CO2 was no greater at enriched than at ambient N supply. After four to six years, however, elevated CO2 stimulated plant biomass much less under ambient than enriched N supply. This response was consistent with the temporally divergent effects of elevated CO2 on soil and plant N dynamics at differing levels of N supply. Our results indicate that variability in availability of soil N and deposition of atmospheric N are both likely to influence the response of plant biomass accumulation to elevated atmospheric CO2. Given that limitations to productivity resulting from the insufficient availability of N are widespread in both unmanaged and managed vegetation, soil N supply is probably an important constraint on global terrestrial responses to elevated CO2.

Relationships of leaf dark respiration to leaf nitrogen, specific leaf area and leaf life-span: a test across biomes and functional groups

Abstract Based on prior evidence of coordinated multiple leaf trait scaling, we hypothesized that variation among species in leaf dark respiration rate (Rd) should scale with variation in traits such as leaf nitrogen (N), leaf life-span, specific leaf area (SLA), and net photosynthetic capacity (Amax). However, it is not known whether such scaling, if it exists, is similar among disparate biomes and plant functional types. We tested this idea by examining the interspecific relationships between Rd measured at a standard temperature and leaf life-span, N, SLA and Amax for 69 species from four functional groups (forbs, broad-leafed trees and shrubs, and needle-leafed conifers) in six biomes traversing the Americas: alpine tundra/subalpine forest, Colorado; cold temperate forest/grassland, Wisconsin; cool temperate forest, North Carolina; desert/shrubland, New Mexico; subtropical forest, South Carolina; and tropical rain forest, Amazonas, Venezuela. Area-based Rd was positively related to area-based leaf N within functional groups and for all species pooled, but not when comparing among species within any site. At all sites, mass-based Rd (Rd-mass) decreased sharply with increasing leaf life-span and was positively related to SLA and mass-based Amax and leaf N (leaf Nmass). These intra-biome relationships were similar in shape and slope among sites, where in each case we compared species belonging to different plant functional groups. Significant Rd-mass−Nmass relationships were observed in all functional groups (pooled across sites), but the relationships differed, with higher Rd at any given leaf N in functional groups (such as forbs) with higher SLA and shorter leaf life-span. Regardless of biome or functional group, Rd-mass was well predicted by all combinations of leaf life-span, Nmass and/or SLA (r2≥ 0.79, P < 0.0001). At any given SLA, Rd-mass rises with increasing Nmass and/or decreasing leaf life-span; and at any level of Nmass, Rd-mass rises with increasing SLA and/or decreasing leaf life-span. The relationships between Rd and leaf traits observed in this study support the idea of a global set of predictable interrelationships between key leaf morphological, chemical and metabolic traits.

Different photosynthesis-nitrogen relations in deciduous hardwood and evergreen coniferous tree species

The relationship between photosynthetic capacity (Amax) and leaf nitrogen concentration (N) among all C3 species can be described roughly with one general equation, yet within that overall pattern species groups or individual species may have markedly different Amax-N relationships. To determine whether one or several predictive, fundamental Amax-N relationships exist for temperate trees we measured Amax, specific leaf area (SLA) and N in 22 broad-leaved deciduous and 9 needle-leaved evergreen tree species in Wisconsin, United States. For broad-leaved deciduous trees, mass-based Amax was highly correlated with leaf N (r2=0.75, P<0.001). For evergreen conifers, mass-based Amax was also correlated with leaf N (r2=0.59, P<0.001) and the slope of the regression (rate of increase of Amax per unit increase in N) was lower (P<0.001) by two-thirds than in the broad-leaved species (1.9 vs. 6.4 μmol CO2 g−1 N s−1), consistent with predictions based on tropical rain forest trees of short vs. long leaf life-span. On an area basis, there was a strong Amax-N correlation among deciduous species (r2=0.78, P<0.001) and no correlation (r2=0.03, P>0.25) in the evergreen conifers. Compared to deciduous trees at a common leaf N (mass or area basis), evergreen trees had lower Amax and SLA. For all data pooled, both leaf N and Amax on a mass basis were correlated (r2=0.6) with SLA; in contrast, area-based leaf N scaled tightly with SLA (r2=0.81), but area-based Amax did not (r2=0.06) because of low Amax per unit N in the evergreen conifers. Multiple regression analysis of all data pooled showed that both N (mass or area basis) and SLA were significantly (P<0.001) related to Amax on mass (r2=0.80) and area (r2=0.55) bases, respectively. These results provide further evidence that Amax-N relationships are fundamentally different for ecologically distinct species groups with differing suites of foliage characteristics: species with long leaf life-spans and low SLA, whether broad-leaved or needle-leaved, tend to have lower Amax per unit leaf N and a lower slope and higher intercept of the Amax-N relation than do species with shorter leaf life-span and higher SLA. A single global Amax-N equation overestimates and underestimates Amax for temperate trees at the upper and lower end of their leaf N range, respectively. Users of Amax-N relationships in modeling photosynthesis in different ecosystems should appreciate the strengths and limitations of regression equations based on different species groupings.

Influence of soil porosity on water use in Pinus taeda

Abstract We analyzed the hydraulic constraints imposed on water uptake from soils of different porosities in loblolly pine (Pinus taeda L.) by comparing genetically related and even-aged plantations growing in loam versus sand soil. Water use was evaluated relative to the maximum transpiration rate (Ecrit) allowed by the soil-leaf continuum. We expected that trees on both soils would approach Ecrit during drought. Trees in sand, however, should face greater drought limitation because of steeply declining hydraulic conductivity in sand at high soil water potential (ΨS). Transport considerations suggest that trees in sand should have higher root to leaf area ratios (AR:AL), less negative leaf xylem pressure (ΨL), and be more vulnerable to xylem cavitation than trees in loam. The AR:AL was greater in sand versus loam (9.8 vs 1.7, respectively). This adjustment maintained about 86% of the water extraction potential for both soils. Trees in sand were more deeply rooted (>1.9 m) than in loam (95% of roots <0.2 m), allowing them to shift water uptake to deeper layers during drought and avoid hydraulic failure. Midday ΨL was constant for days of high evaporative demand, but was less negative in sand (–1.6 MPa) versus loam (–2.1 MPa). Xylem was more vulnerable to cavitation in sand versus loam trees. Roots in both soils were more vulnerable than stems, and experienced the greatest predicted loss of conductivity during drought. Trees on both soils approached Ecrit during drought, but at much higher ΨS in sand (<–0.4 MPa) than in loam (<–1.0 MPa). Results suggest considerable phenotypic plasticity in water use traits for P. taeda which are adaptive to differences in soil porosity.

Why are non-photosynthetic tissues generally 13C enriched compared with leaves in C3 plants? Review and synthesis of current hypotheses

Non-photosynthetic, or heterotrophic, tissues in C3 plants tend to be enriched in 13C compared with the leaves that supply them with photosynthate. This isotopic pattern has been observed for woody stems, roots, seeds and fruits, emerging leaves, and parasitic plants incapable of net CO2 fixation. Unlike in C3 plants, roots of herbaceous C4 plants are generally not 13C-enriched compared with leaves. We review six hypotheses aimed at explaining this isotopic pattern in C3 plants: (1) variation in biochemical composition of heterotrophic tissues compared with leaves; (2) seasonal separation of growth of leaves and heterotrophic tissues, with corresponding variation in photosynthetic discrimination against 13C; (3) differential use of day v. night sucrose between leaves and sink tissues, with day sucrose being relatively 13C-depleted and night sucrose 13C-enriched; (4) isotopic fractionation during dark respiration; (5) carbon fixation by PEP carboxylase; and (6) developmental variation in photosynthetic discrimination against 13C during leaf expansion. Although hypotheses (1) and (2) may contribute to the general pattern, they cannot explain all observations. Some evidence exists in support of hypotheses (3) through to (6), although for hypothesis (6) it is largely circumstantial. Hypothesis (3) provides a promising avenue for future research. Direct tests of these hypotheses should be carried out to provide insight into the mechanisms causing within-plant variation in carbon isotope composition.

Belowground carbon allocation in forests estimated from litterfall and IRGA-based soil respiration measurements

Allocation of C to belowground plant structures is one of the most important, yet least well quantified fluxes of C in terrestrial ecosystems. In a literature review of mature forests worldwide, Raich and Nadelhoffer (1989) suggested that total belowground carbon allocation (TBCA) could be estimated from the difference between annual rates of soil respiration and aboveground litterfall. Here we analyze new measurements of soil respiration and litterfall, including data from the Ameriflux network. Our results generally agree with Raich and Nadelhoffer’s previous work. A regression analysis of data from mature forests produced the following relationship: annual soil respiration = 287 + 2.80 × annual litterfall. This regression slope indicates that, on average, soil respiration is roughly three times ab oveground litterfall-C, which further implies that TBCA is roughly twice annual aboveground litterfall-C. These inferences are based on the uncertain assumption of soil C stocks being at steady state. Nevertheless, changes in soil C would have to be very large to modify the conclusion that TBCA is generally much larger than litterfall. Among only mature temperate hardwood forests, however, the correlation between litterfall and soil respiration was poor, and the correlation among years for a single site was also poor. Therefore, the regression cannot be relied upon to provide accurate estimates of soil respiration or TBCA for individual sites. Moreover, interannual variation in TBCA, short-term changes in C stocks, or different temporal scales controlling leaf litter production and soil respiration may cause important deviations from the global average. The regression slope for data from young forests is steeper, possibly indicating proportionally greater TBCA, but the steady-state assumption is more problematic for young forests. This method

Application-controlled demand paging for out-of-core visualization

In the area of scientific visualization, input data sets are often very large. In visualization of computational fluid dynamics (CFD) in particular, input data sets today can surpass 100 Gbytes, and are expected to scale with the ability of supercomputers to generate them. Some visualization tools already partition large data sets into segments, and load appropriate segments as they are needed. However, this does not remove the problem for two reasons: 1) there are data sets for which even the individual segments are too large for the largest graphics workstations, 2) many practitioners do not have access to workstations with the memory capacity required to load even a segment, especially since the state-of-the-art visualization tools tend to be developed by researchers with much more powerful machines. When the size of the data that must be accessed is larger than the size of memory, some form of virtual memory is simply required. This may be by segmentation, paging, or by paged segments. The authors demonstrate that complete reliance on operating system virtual memory for out-of-core visualization leads to egregious performance. They then describe a paged segment system that they have implemented, and explore the principles of memory management that can be employed by the application for out-of-core visualization. They show that application control over some of these can significantly improve performance. They show that sparse traversal can be exploited by loading only those data actually required.