B21E-01 INVITED
Optical Isotopic Measurements of CO2 and H2O for Investigating Biosphere-Atmosphere Interactions: Recent Progress and Challenges
The isotopic composition of atmospheric CO2 and H2O represent important signals of global change. Interpreting these signals, however, requires quantification of the transport of isotopic CO2 and H2O between the biosphere and atmosphere and an understanding of the underlying discrimination mechanisms at the appropriate temporal and spatial scales. Over the last 5 years we have experimented with optical techniques to measure CO2 and H2O isotopic mixing ratios and fluxes at a relatively high frequency and at spatial scales ranging from a small chamber to the region. The highlights and lessons learned from three different experiments will be presented. First, carbon isotopic flux measurements from a 244 m tall tower will be used to show how agricultural ecosystems and land surface heterogeneity influence net ecosystem CO2 exchange and biosphere-atmosphere discrimination at the regional scale in the Upper Midwestern, United States; Second, chamber and field-scale carbon isotope observations will be used to examine the isotopic partitioning of net ecosystem CO2 exchange; Finally, the feasibility of long-term continuous measurements of isotopic water vapor fluxes, based on the eddy covariance technique, will be examined. Our recent progress and the challenges that lay ahead will be discussed.
B21E-02 INVITED
A possible new role for atmospheric 13CO2 in global models
The promise of utilizing large-scale atmospheric δ13CO2 measurements to understand terrestrial processes has not been fully realized. Here, we will present recent progress in the use of measurements and simulations of atmospheric δ13C to better understand the signals of the biosphere contained within atmospheric data. The motivation for measuring atmospheric δ13C over the past 20 years has been to help partition global surface carbon fluxes into terrestrial and oceanic components. While this is still possible, it is probably not the most effective use of δ13C. First, isotopic disequilibria and their interannual variability introduce great uncertainty into land/ocean partitioning estimates. Second, the presence of a greatly increased density of terrestrial CO2 observations in the Northern Hemisphere obviate this application of δ13C to a large extent. What this suggests, however, is that in a system in which land/ocean uptake is already well constrained, δ13C can constrain terrestrial biosphere processes. For example, analysis of atmospheric CO2 and δ13C correlations at regional spatial scales (105 - 106 km22) suggests a seasonality in photosynthetic discrimination and stomatal conductance. Other analyses of CO2 and δ13C correlations show changes in C4 and C3 productivity. This suggests that within a carbon data assimilation system, such as NOAA/ESRL's CarbonTracker, atmospheric δ13C data could be used to optimize large-scale land surface characteristics like C3/C4 mixtures or stomatal conductance. We will present forward modeling simulations of atmospheric δ13C with comparisons to observations as well as model tests quantifying the sensitivity of atmospheric δ13C to C3 and C4 carbon flux variations and stomatal conductance parameterizations.
B21E-03
High frequency field measurements of diurnal carbon isotope discrimination and internal conductance in a semi-arid species, Juniperus monosperma
We present field observations of leaf gas exchange, carbon isotope discrimination (Δ) and internal conductance of CO2 to the sites of carboxylation (gi) collected during summer 2006 using tunable diode laser spectroscopy (TDL). Δ ranged from 27.4‰ to 12.6‰ over diurnal periods with daily means of 16.3 ± 0.2‰ during drought to 19.0 ± 0.5‰ during monsoon conditions. We observed a large range in gi, from 0.03-2.03 μmol m-2 s-1 Pa-1 among measured leaves. We tested the comprehensive Farquhar, OLeary & Berry (1982) model of Δ (Δcomp), a simplified form of Δcomp (Δsimple), and recently suggested amendments (Δrevised; Wingate et al. 2007). Sensitivity analyses demonstrated that incorporating variable gi had a substantial effect on Δcomp, resulting in mean differences between observed Δ (Δobs) and Δcomp predictions as low as 0.04‰ and as high as 9.6‰. We found first order linear models adequately described the relationship between Δ and the ratio of substomatal to atmospheric CO2 partial pressure (pi/pa) on all three days, but curvilinear second order models best described the relationship in July and August, potentially due to the dominance of respiration and associated isotopic signatures at high pi/pa. There was good agreement between Δobs and predictions from all models, with Δsimple producing the best fit of Δobs in June, Δcomp producing the best fit in July, and Δrevised producing the best fit in August.
B21E-04
Photodegradation of Leaf Litter in Water-Limited Ecosystems
The longstanding view of terrestrial decomposition holds that heterotrophic respiration drives release of CO2, but recent studies, such as Austin and Vivanco (2006) have shown that in water-limited environments, photochemical decomposition of leaf litter may be equally or more effective than microbial decomposition. Although initial studies have concluded that photochemical degradation can be important in some environments, it has been difficult to quantify and the oxidative mechanisms involved remain unknown. Thus, the objectives of our study were to (1) quantify the CO2 emitted during photochemical degradation of leaf litter and (2) use the stable isotopic signatures of evolved CO2 to elucidate pathways of production. Emitted CO2 and its isotopic signature were measured using a tunable diode laser (TDL) to assess the pool of photochemically-labile plant matter (δ13C-CO2) in a given sample and to assess the source of the oxygen (δ18O-CO2). We quantified the photochemical release of CO2 and its isotopic signature from dried leaf litter of 10 tree and grass species prevalent in major biotic zones of New Mexico. The cumulative CO2 released upon exposure of 0.1-0.3 g of dried leaf litter to three hours of simulated sunlight ranged from 8-25 mg CO2-C g-1 dried litter, corresponding to 1-2% mass loss. Generally, the δ13C-CO2 was more depleted (4-7 ± 2 per mil) than the average δ13C of the respective leaf litter sample. The δ18O-CO2 evolved is approximately equal to δ18O of atmospheric O2, suggesting that the oxidation mechanism involves direct reaction with atmospheric O2.
B21E-05
Soil Drying Effects on the Carbon Isotope Composition of Soil Respiration
Stable isotopes are used widely as a tool for determining sources of carbon (C) fluxes in ecosystem C studies. Environmental factors that change over time, such as moisture, can create dynamic changes in the isotopic composition of C assimilated by plants, and offers a unique opportunity to distinguish fast- responding plant C from slower-responding soil C pools, which under steady-state conditions may be too similar isotopically to partition. Monitoring the isotopic composition of soil respiration over a period of changing moisture conditions is potentially a useful approach for characterizing plant contributions to soil respiration. But this partitioning hinges on the assumption that any change in the isotopic signature of soil respiration is solely due to recent photosynthetic discrimination, and that post-photosynthetic processes, such as microbial respiration, do not discriminate as moisture decreases. The purpose of the present study is to test the assumption that δ13CO2 from microbial respiration remains static as soil dries. We conducted a series of greenhouse experiments employing different techniques to isolate microbial respiration from root respiration. The first involves removing roots from soil, and showed that when roots are present, respiration from dry soil is enriched in 13C relative to moist soil, but when roots are absent, respiration is isotopically similar from moist and dry soils. This indicates that rhizospheric respiration changes isotopically with moisture whereas soil microbial respiration does not. In contrast, a second experiment in which soil columns without plants were monitored as they dried, showed respiration from very dry soil to be enriched by 8‰ relative to moist soil. However, simulations with an isotopologue-based soil gas diffusion model demonstrate that at least a portion of the apparent enrichment is due to non-steady state gas transport processes. Careful sampling methodologies which prevent or account for non-steady-state effects are necessary to avoid spurious correlations between measured δ13CO2 and soil moisture. A third experiment, using closed-system soil incubations to avoid non-steady state mixing with atmospheric CO2, indicates that the isotopic composition of microbial soil respiration appears to be unchanging under a large range of soil moisture contents.
B21E-06
Fine Root Longevity Still Under Debate
Assuming that fine roots (< 2 mm in diameter) turn over once per year, they represent a third of the global annual net primary productivity. These turnover estimates are based on rhizotron studies, where root longevity is determined by monitoring the appearance/disappearance of roots on a screen, which is inserted into the soil. Much slower fine root turnover rates were found using carbon (C) isotope methods (either 14C dating or continuous 13C-labelling), resulting in root longevities of several years. Stable C isotope tracer experiments, are argued to overestimate fine root longevities, mainly because the smallest roots with the highest turn over, are easily missed during sampling. The goal of the present study was therefore to carry out a C-labelling experiment, and specifically focus on the finest roots, namely root tips. In addition we sampled whole fine roots (<1 mm and 1-3 mm in diameter), as in other studies. We pulse labelled 14-year-old Pinus sylvestris (Pine) trees in the field for only three hours with highly 13C-enriched CO2 (24 atom percent). The mean residence time (MRT) of recently assimilated C in root tips was determined, as a measure for root longevity. Already two days after labelling, recent C had been translocated from the crowns to fine roots indicating rapid belowground C allocation. 13C signals in root tips were stronger than in whole roots, which shows that they are the most active part of the root system. MRT of C calculated using first order exponential decay functions of C in bulk roots were around 20 days in both <1mm and 1-3mm roots and 29 days in root tips. A rapid decline in 13C signals was observed which could be explained by a rapid decrease in the signal of the sucrose pool, which had a MRT of 5 days. However, part of the labelled C had been allocated to a pool with a slower turnover rate (most likely structural compounds such as cellulose) as indicated by persisting 13C signals measured 120 days after labelling. MRT of C in those slow turnover pools will be estimated by continuously monitoring the remaining 13C signals. Those MRTs are expected to offer more realistic fine root turnover estimates and could possibly end the root longevity debate.
B21E-07
Variability of oxygen isotopes in water vapor, soil and leaf water and leaf cellulose along a steep environmental gradient in Hawaii
The stable oxygen isotope ratio (δ18O) of plant organic mater (OM) has been shown to contain essential information on the ecophysiological performance of plants and for the understanding of water and carbon fluxes at the plant and ecosystem scales. The primary source of δ18O variability in plant OM is the isotopic composition of leaf water. In general, the isotopic composition of leaf water depends on the isotopic signature of the plant's source water, the isotopic composition of atmospheric vapor surrounding the leaf, the leaf to atmosphere vapor pressure gradient, the leaf's stomatal conductance and transpiration rates as well as leaf hydraulic properties. The combined influence of all these factors on isotopic leaf water enrichment clearly complicates the interpretation of plant δ18O signatures. However, significant recent improvements in mechanistic leaf water models have much advanced our understanding of the interplay of all these factors and their combined effects on isotopic leaf water enrichment. These models now give the theoretical background to constrain which of the multiple factors that potentially drive the isotopic enrichment of leaf water are finally recorded in plant OM. Along one of the steepest moisture gradients in the world – the eastern slope of the 4200 m volcano Mauna Loa (Hawaii) we tested the applicability of these models to constrain which environmental factors drive the isotopic variability in leaf water and OM of Metrosideros polymorpha, a tree species that inhabits the entire moisture gradient. On seven sites along the eastern slope of the 4200 m volcano Mauna Loa, we determined the isotopic signature of source water, leaf water and leaf cellulose. In addition, we also determined the diurnal isotopic enrichment of leaf water, the diurnal variability of the plant's stomatal conductance and transpiration rates and well as air temperature and humidity and the isotopic composition of atmospheric vapor on three core sites along the gradient. Using the data from these core sites in combination with state of the art isotopic leaf water models our study will allow to constrain which of the several parameters that potentially drive isotopic enrichment of leaf water long environmental gradients are finally recorded in OM of M. polymorpha. Going beyond typical correlative investigations, the mechanistic investigation and understanding of the variability in δ18O of plant OM along environmental gradients has important implications for the general application of oxygen isotopes in studies at the plant and ecosystem scale.