Eric J. Ward

Acclimation of leaf hydraulic conductance and stomatal conductance of Pinus taeda (loblolly pine) to long‐term growth in elevated CO2 (free‐air CO2 enrichment) and N‐fertilization

We investigated how leaf hydraulic conductance (K(leaf)) of loblolly pine trees is influenced by soil nitrogen amendment (N) in stands subjected to ambient or elevated CO(2) concentrations (CO(2)(a) and CO(2)(e), respectively). We also examined how K(leaf) varies with changes in reference leaf water potential (Psi(leaf-ref)) and stomatal conductance (g(s-ref)) calculated at vapour pressure deficit, D of 1 kPa. We detected significant reductions in K(leaf) caused by N and CO(2)(e), but neither treatment affected pre-dawn or midday Psi(leaf). We also detected a significant CO(2)(e)-induced reduction in g(s-ref) and Psi(leaf-ref). Among treatments, the sensitivity of K(leaf) to Psi(leaf) was directly related to a reference K(leaf) (K(leaf-ref) computed at Psi(leaf-ref)). This liquid-phase response was reflected in a similar gas-phase response, with g(s) sensitivity to D proportional to g(s-ref). Because leaves represented a substantial component of the whole-tree conductance, reduction in K(leaf) under CO(2)(e) affected whole-tree water use by inducing a decline in g(s-ref). The consequences of the acclimation of leaves to the treatments were: (1) trees growing under CO(2)(e) controlled morning leaf water status less than CO(2)(a) trees resulting in a higher diurnal loss of K(leaf); (2) the effect of CO(2)(e) on g(s-ref) was manifested only during times of high soil moisture.

On the difference in the net ecosystem exchange of CO2 between deciduous and evergreen forests in the southeastern United States

The southeastern United States is experiencing a rapid regional increase in the ratio of pine to deciduous forest ecosystems at the same time it is experiencing changes in climate. This study is focused on exploring how these shifts will affect the carbon sink capacity of southeastern US forests, which we show here are among the strongest carbon sinks in the continental United States. Using eight-year-long eddy covariance records collected above a hardwood deciduous forest (HW) and a pine plantation (PP) co-located in North Carolina, USA, we show that the net ecosystem exchange of CO2 (NEE) was more variable in PP, contributing to variability in the difference in NEE between the two sites (ΔNEE) at a range of timescales, including the interannual timescale. Because the variability in evapotranspiration (ET) was nearly identical across the two sites over a range of timescales, the factors that determined the variability in ΔNEE were dominated by those that tend to decouple NEE from ET. One such factor was water use efficiency, which changed dramatically in response to drought and also tended to increase monotonically in nondrought years (P < 0.001 in PP). Factors that vary over seasonal timescales were strong determinants of the NEE in the HW site; however, seasonality was less important in the PP site, where significant amounts of carbon were assimilated outside of the active season, representing an important advantage of evergreen trees in warm, temperate climates. Additional variability in the fluxes at long-time scales may be attributable to slowly evolving factors, including canopy structure and increases in dormant season air temperature. Taken together, study results suggest that the carbon sink in the southeastern United States may become more variable in the future, owing to a predicted increase in drought frequency and an increase in the fractional cover of southern pines.

Increases in atmospheric CO2 have little influence on transpiration of a temperate forest canopy

Models of forest energy, water and carbon cycles assume decreased stomatal conductance with elevated atmospheric CO2 concentration ([CO2]) based on leaf-scale measurements, a response not directly translatable to canopies. Where canopy-atmosphere are well-coupled, [CO2 ]-induced structural changes, such as increasing leaf-area index (LD), may cause, or compensate for, reduced mean canopy stomatal conductance (GS), keeping transpiration (EC) and, hence, runoff unaltered. We investigated GS responses to increasing [CO2] of conifer and broadleaved trees in a temperate forest subjected to 17-yr free-air CO2 enrichment (FACE; + 200 μmol mol(-1)). During the final phase of the experiment, we employed step changes of [CO2] in four elevated-[CO2 ] plots, separating direct response to changing [CO2] in the leaf-internal air-space from indirect effects of slow changes via leaf hydraulic adjustments and canopy development. Short-term manipulations caused no direct response up to 1.8 × ambient [CO2], suggesting that the observed long-term 21% reduction of GS was an indirect effect of decreased leaf hydraulic conductance and increased leaf shading. Thus, EC was unaffected by [CO2] because 19% higher canopy LD nullified the effect of leaf hydraulic acclimation on GS . We advocate long-term experiments of duration sufficient for slow responses to manifest, and modifying models predicting forest water, energy and carbon cycles accordingly.

The effects of elevated CO2 and nitrogen fertilization on stomatal conductance estimated from 11 years of scaled sap flux measurements at Duke FACE

In this study, we employ a network of thermal dissipation probes (TDPs) monitoring sap flux density to estimate leaf-specific transpiration (E(L)) and stomatal conductance (G(S)) in Pinus taeda (L.) and Liquidambar styraciflua L. exposed to +200 ppm atmospheric CO(2) levels (eCO(2)) and nitrogen fertilization. Scaling half-hourly measurements from hundreds of sensors over 11 years, we found that P. taeda in eCO(2) intermittently (49% of monthly values) decreased stomatal conductance (G(S)) relative to the control, with a mean reduction of 13% in both total E(L) and mean daytime G(S). This intermittent response was related to changes in a hydraulic allometry index (A(H)), defined as sapwood area per unit leaf area per unit canopy height, which decreased a mean of 15% with eCO(2) over the course of the study, due mostly to a mean 19% increase in leaf area (A(L)). In contrast, L. styraciflua showed a consistent (76% of monthly values) reduction in G(S) with eCO(2) with a total reduction of 32% E(L), 31% G(S) and 23% A(H) (due to increased A(L) per sapwood area). For L. styraciflua, like P. taeda, the relationship between A(H) and G(S) at reference conditions suggested a decrease in G(S) across the range of A(H). Our findings suggest an indirect structural effect of eCO(2) on G(S) in P. taeda and a direct leaf level effect in L. styraciflua. In the initial year of fertilization, P. taeda in both CO(2) treatments, as well as L. styraciflua in eCO(2), exhibited higher G(S) with N(F) than expected from shifts in A(H), suggesting a transient direct effect on G(S). Whether treatment effects on mean leaf-specific G(S) are direct or indirect, this paper highlights that long-term treatment effects on G(S) are generally reflected in A(H) as well.

On the complementary relationship between marginal nitrogen and water-use efficiencies among Pinus taeda leaves grown under ambient and CO2-enriched environments

BACKGROUND AND AIMS Water and nitrogen (N) are two limiting resources for biomass production of terrestrial vegetation. Water losses in transpiration (E) can be decreased by reducing leaf stomatal conductance (g(s)) at the expense of lowering CO(2) uptake (A), resulting in increased water-use efficiency. However, with more N available, higher allocation of N to photosynthetic proteins improves A so that N-use efficiency is reduced when g(s) declines. Hence, a trade-off is expected between these two resource-use efficiencies. In this study it is hypothesized that when foliar concentration (N) varies on time scales much longer than g(s), an explicit complementary relationship between the marginal water- and N-use efficiency emerges. Furthermore, a shift in this relationship is anticipated with increasing atmospheric CO(2) concentration (c(a)). METHODS Optimization theory is employed to quantify interactions between resource-use efficiencies under elevated c(a) and soil N amendments. The analyses are based on marginal water- and N-use efficiencies, λ = (∂A/∂g(s))/(∂E/∂g(s)) and η = ∂A/∂N, respectively. The relationship between the two efficiencies and related variation in intercellular CO(2) concentration (c(i)) were examined using A/c(i) curves and foliar N measured on Pinus taeda needles collected at various canopy locations at the Duke Forest Free Air CO(2) Enrichment experiment (North Carolina, USA). KEY RESULTS Optimality theory allowed the definition of a novel, explicit relationship between two intrinsic leaf-scale properties where η is complementary to the square-root of λ. The data support the model predictions that elevated c(a) increased η and λ, and at given c(a) and needle age-class, the two quantities varied among needles in an approximately complementary manner. CONCLUSIONS The derived analytical expressions can be employed in scaling-up carbon, water and N fluxes from leaf to ecosystem, but also to derive transpiration estimates from those of η, and assist in predicting how increasing c(a) influences ecosystem water use.

A test of the hydraulic vulnerability segmentation hypothesis in angiosperm and conifer tree species

Water transport from soils to the atmosphere is critical for plant growth and survival. However, we have a limited understanding about many portions of the whole-tree hydraulic pathway, because the vast majority of published information is on terminal branches. Our understanding of mature tree trunk hydraulic physiology, in particular, is limited. The hydraulic vulnerability segmentation hypothesis (HVSH) stipulates that distal portions of the plant (leaves, branches and roots) should be more vulnerable to embolism than trunks, which are nonredundant organs that require a massive carbon investment. In the current study, we compared vulnerability to loss of hydraulic function, leaf and xylem water potentials and the resulting hydraulic safety margins (in relation to the water potential causing 50% loss of hydraulic conductivity) in leaves, branches, trunks and roots of four angiosperms and four conifer tree species. Across all species, our results supported strongly the HVSH as leaves and roots were less resistant to embolism than branches or trunks. However, branches were consistently more resistant to embolism than any other portion of the plant, including trunks. Also, calculated whole-tree vulnerability to hydraulic dysfunction was much greater than vulnerability in branches. This was due to hydraulic dysfunction in roots and leaves at less negative water potentials than those causing branch or trunk dysfunction. Leaves and roots had narrow or negative hydraulic safety margins, but trunks and branches maintained positive safety margins. By using branch-based hydraulic information as a proxy for entire plants, much research has potentially overestimated embolism resistance, and possibly drought tolerance, for many species. This study highlights the necessity to reconsider past conclusions made about plant resistance to drought based on branch xylem only. This study also highlights the necessity for more research of whole-plant hydraulic physiology to better understand strategies of plant drought tolerance and the critical control points within the hydraulic pathway.

The effect of plant water storage on water fluxes within the coupled soil–plant system

In addition to buffering plants from water stress during severe droughts, plant water storage (PWS) alters many features of the spatio-temporal dynamics of water movement in the soil-plant system. How PWS impacts water dynamics and drought resilience is explored using a multi-layer porous media model. The model numerically resolves soil-plant hydrodynamics by coupling them to leaf-level gas exchange and soil-root interfacial layers. Novel features of the model are the considerations of a coordinated relationship between stomatal aperture variation and whole-system hydraulics and of the effects of PWS and nocturnal transpiration (Fe,night) on hydraulic redistribution (HR) in the soil. The model results suggest that daytime PWS usage and Fe,night generate a residual water potential gradient (Δψp,night) along the plant vascular system overnight. This Δψp,night represents a non-negligible competing sink strength that diminishes the significance of HR. Considering the co-occurrence of PWS usage and HR during a single extended dry-down, a wide range of plant attributes and environmental/soil conditions selected to enhance or suppress plant drought resilience is discussed. When compared with HR, model calculations suggest that increased root water influx into plant conducting-tissues overnight maintains a more favorable water status at the leaf, thereby delaying the onset of drought stress.

Hydraulic time constants for transpiration of loblolly pine at a free-air carbon dioxide enrichment site

The impact of stored water on estimates of transpiration from scaled sap flux measurements was assessed in mature Pinus taeda (L.) at the Duke Free-Air CO(2) Enrichment (FACE) site. We used a simple hydraulic model with measurements of sap flux (J) at breast height and the base of the live crown for 26 trees over 6 months to examine the effects of elevated CO(2) (eCO(2)) and fertilization (N(F)) treatments, as well as temporal variation in soil moisture (M(()(t)())), on estimates of the hydraulic time constant (κ). At low M(()(t)()), there was little (<12%) difference in κ of different treatments. At high M(()(t)()), differences were much greater, with κ reductions of 27, 52 and 34% in eCO(2), N(F) and eCO(2) × N(F) respective to the control. Incorporating κ with these effects into the analysis of a larger data set of previous J measurements at this site (1998-2008) improved agreement between modeled and measured values in 92% of cases. However, a simplified calibration of κ that neglected treatment and soil moisture effects performed more dependably, improving agreement in 98% of cases. Incorporating κ had the effect of increasing estimates of reference stomatal conductance at 1 kPa vapor pressure deficit (VPD) and saturating photosynthetic active radiation (PAR) an average of 12-14%, while increasing estimated sensitivities to VPD and PAR. A computationally efficient hydraulic model, such as the one presented here, incorporated into a hierarchical model of stomatal conductance presents a novel approach to including hydraulic time constants in estimates of stomatal responses from long-term sap flux data sets.

Measuring water fluxes in forests: the need for integrative platforms of analysis

To understand the importance of analytical tools such as those provided by Berdanier et al. (2016) in this issue of Tree Physiology, one must understand both the grand challenges facing Earth system modelers, as well as the minutia of engaging in ecophysiological research in the field. It is between these two extremes of scale that many ecologists struggle to translate empirical research into useful conclusions that guide our understanding of how ecosystems currently function and how they are likely to change in the future. Likewise, modelers struggle to build complexity into their models that match this sophisticated understanding of how ecosystems function, so that necessary simplifications required by large scales do not themselves change the conclusions drawn from these simulations. As both monitoring technology and computational power increase, along with the continual effort in both empirical and modeling research, the gap between the scale of Earth system models and ecological observations continually closes. This creates a need for platforms of model–data interaction that incorporate uncertainties in both simulations and observations when scaling from one to the other, moving beyond simple comparisons of monthly or annual sums and means. Global models of the Earth system are now approaching the complexity and resolution at which the dynamics of vegetation begin to play a crucial role. The next generation of dynamic global vegetation models (DGVMs) will not just include regional representations of typical ecosystems of different plant functional types (PFTs), but will take into account variation in important plant traits within them (Wullschleger et al. 2014), such as carbon allocation, stomatal regulation, xylem hydraulics, photosynthetic potential and respiration. Indeed, trait variability within PFTs has been shown to change projections of future terrestrial carbon sinks by up to 33% or more (Verheijen et al. 2015), while almost 70% of the variance in FLUXNET (a global database of ecosystem eddy-covariance observations) remain unexplained by classical PFT approach (Reichstein et al. 2014). This level of sophistication is understood as a prerequisite for predicting how ecosystems will adapt to a changing climate, including areas whose climate envelope is pushed beyond the range observed in recorded history. To achieve this, models must move beyond using aggregated observations of different PFTs to fully utilizing the complex record of observations acquired over decades of ecological field and laboratory research. Achieving such data–model interaction requires collaborations between empirical scientists and modelers, as well as those conversant in both types of research. Approaches that harness model–data interaction can be useful in multi-model comparisons (McDowell et al. 2013, Walker et al. 2015), in the design of observational and experimental studies (Medlyn et al. 2016, Norby et al. 2016), and even in selecting the most valuable measurements to transmit to dataloggers (Clark et al. 2011). The software that translates ecophysiological measurements from electrical signals (e.g. millivolts measured by a sensor) to state variables found in DGVMs (e.g. millimeters per day of transpiration) forms an often-overlooked part of model–data interaction and represents a key platform for this integration between physiological ecologists and ecological modelers. It is important to realize that the vast majority of researchers lack formal training in software development yet they may devote 30% of their time to developing domainspecific software to address such issues (Wilson et al. 2014). The assumptions inherent within such software forms a crucial

TRACC: An open source software for processing sap flux data from thermal dissipation probes

Key messageTRACC is an open-source software for standardizing the cleaning, conversion, and calibration of sap flux density data from thermal dissipation probes, which addresses issues of nighttime transpiration and water storage.AbstractThermal dissipation probes (TDPs) have become a widely used method of monitoring plant water use in recent years. The use of TDPs requires calibration to a theoretical zero-flow value (∆T0); usually based upon the assumption that at least some nighttime measurements represent zero-flow conditions. Fully automating the processing of data from TDPs is made exceedingly difficult due to errors arising from many sources. However, it is desirable to minimize variation arising from different researchers’ processing data, and thus, a common platform for processing data, including editing raw data and determination of ∆T0, is useful and increases the transparency and replicability of TDP-based research. Here, we present the TDP data processing software TRACC (Thermal dissipation Review Assessment Cleaning and Conversion) to serve this purpose. TRACC is an open-source software written in the language R, using graphical presentation of data and on screen prompts with yes/no or simple numerical responses. It allows the user to select several important options, such as calibration coefficients and the exclusion of nights when vapor pressure deficit does not approach zero. Although it is designed for users with no coding experience, the outputs of TRACC could be easily incorporated into more complex models or software.