Improved global estimates of terrestrial photosynthesis and respiration are critical for predicting the rate of change in atmospheric CO 2. The oxygen isotopic composition of atmospheric CO2 can be used to estimate these fluxes because oxygen isotopic exchange between CO 2 and water creates distinct isotopic flux signatures. The enzyme carbonic anhydrase (CA) is known to accelerate this exchange in leaves, but the possibility of CA activity in soils is commonly neglected. Here, we report widespread accelerated soil CO 2 hydration. Exchange was 10 -300 times faster than the uncatalyzed rate, consistent with typical population sizes for CAcontaining soil microorganisms. Including accelerated soil hydration in global model simulations modifies contributions from soil and foliage to the global CO 18 O budget and eliminates persistent discrepancies existing between model and atmospheric observations. This enhanced soil hydration also increases the differences between the isotopic signatures of photosynthesis and respiration, particularly in the tropics, increasing the precision of CO2 gross fluxes obtained by using the ␦ 18 O of atmospheric CO2 by 50%.carbon cycle ͉ water cycle ͉ carbonic anhydrase ͉ oxygen isotopes ͉ terrestrial biosphere

The terrestrial carbon cycle is influenced by environmental variability at scales ranging from diurnal to interannual. Here, we present 5-years of growing season (day 131-275) observations of the carbon isotope ratio of ecosystem respiration (d 13 C R ) from a semiarid woodland. This ecosystem has a large necromass component resulting from 97% Pinus edulis mortality in 2002, is dominated by drought-tolerant Juniperus monosperma trees, and experiences large variability in the timing and intensity of seasonal and synoptic water availability. Mean growing season d 13 C R was remarkably invariant (À23.57 AE 0.4%), with the exception of particularly enriched d 13 C R in 2006 following a winter with anomalously low snowfall. d 13 C R was strongly coupled to climate during premonsoon periods ($May to June), including fast ( 2 days) responses to changes in crown-level stomatal conductance (G c ) and vapor pressure deficit (vpd) following rain pulses. In contrast, d 13 C R was relatively decoupled from G c and environmental drivers during monsoon and postmonsoon periods (July-August and September, respectively), exhibiting only infrequent couplings of d 13 C R to vpd and soil water content (SWC) with longer lags ($ 8 days) and variable response slopes (both positive and negative). Notably, d 13 C R exhibited consistent dynamics after rainfall events, with depleted d 13 C R occurring within 1 h, progressive hourly d 13 C R enrichment over the remainder of the night, and net d 13 C R depletions over the multiple nights postevent in monsoon and postmonsoon periods. Overall this ecosystem demonstrated strong dependence of d 13 C R on precipitation, with an apparent dominance by the autotrophic d 13 C signal in premonsoon periods when deep soil moisture is abundant and surface soil moisture is low, and weaker coupling during monsoonal periods consistent with increasing heterotrophic dominance when deep soil moisture has declined and surface moisture is variable.

In semi-arid regions, where plants using both C(3) and C(4) photosynthetic pathways are common, the stable C isotope ratio (delta(13)C) of ecosystem respiration (delta(13)C(R)) is strongly variable seasonally and inter-annually. Improved understanding of physiological and environmental controls over these variations will improve C cycle models that rely on the isotopic composition of atmospheric CO(2). We hypothesized that timing of precipitation events and antecedent moisture interact with activity of C(3) and C(4) grasses to determine net ecosystem CO(2) exchange (NEE) and delta(13)C(R). Field measurements included CO(2) and delta(13)C fluxes from the whole ecosystem and from patches of different plant communities, biomass and delta(13)C of plants and soils over the 2000 and 2001 growing seasons. NEE shifted from C source to sink in response to rainfall events, but this shift occurred after a time lag of up to 2 weeks if a dry period preceded the rainfall. The seasonal average of delta(13)C(R) was higher in 2000 (-16 per thousand) than 2001 (20 per thousand), probably due to drier conditions during the 2000 growing season (79.7 mm of precipitation from April up to and including July) than in 2001 (189 mm). During moist conditions, delta(13)C averaged -22 per thousand from C(3) patches, -16 per thousand from C(4) patches, and -19 per thousand from mixed C(3) and C(4) patches. However, during dry conditions the apparent spatial differences were not obvious, suggesting reduced autotrophic activity in C(4) grasses with shallow rooting depth, soon after the onset of dry conditions. Air and soil temperatures were negatively correlated with delta(13)C(R); vapor pressure deficit was a poor predictor of delta(13)C(R), in contrast to more mesic ecosystems. Responses of respiration components to precipitation pulses were explained by differences in soil moisture thresholds between C(3) and C(4) species. Stable isotopic composition of respiration in semi-arid ecosystems is more temporally and spatially variable than in mesic ecosystems owing to dynamic aspects of pulse precipitation episodes and biological drivers.

No abstract

We conducted high frequency measurements of the <i>&delta;</i>¹⁸O value of atmospheric CO₂ from a juniper (<i>Juniperus monosperma</i>) woodland in New Mexico, USA, over a four-year period to investigate climatic and physiological regulation of the <i>&delta;</i>¹⁸O value of ecosystem respiration (<i>&delta;</i>_R). Rain pulses reset <i>&delta;</i>_R with the dominant water source isotope composition, followed by progressive enrichment of <i>&delta;</i>_R. Transpiration (<i>E</i>_T) was significantly related to post-pulse <i>&delta;</i>_R enrichment because leaf water <i>&delta;</i>¹⁸O value showed strong enrichment with increasing vapor pressure deficit that occurs following rain. Post-pulse <i>&delta;</i>_R enrichment was correlated with both <i>E</i>_T and the ratio of <i>E</i>_T to soil evaporation (<i>E</i>_T / <i>E</i>_S). In contrast, soil water <i>&delta;</i>¹⁸O value was relatively stable and <i>&delta;</i>_R enrichment was not correlated with <i>E</i>_S. Model simulations captured the large post-pulse <i>&delta;</i>_R enrichments only when the offset between xylem and leaf water <i>&delta;</i>¹⁸O value was modeled explicitly and when a gross flux model for CO₂ retro-diffusion was included. Drought impacts <i>&delta;</i>_R through the balance between evaporative demand, which enriches <i>&delta;</i>_R, and low soil moisture availability, which attenuates <i>&delta;</i>_R enrichment through reduced <i>E</i>_T. The net result, observed throughout all four years of our study, was a negative correlation of post-precipitation <i>&delta;</i>_R enrichment with increasing drought