Interclonal variation, coordination, and trade-offs between hydraulic conductance and gas exchange in Pinus radiata: consequences on plant growth and wood density.

Stem growth reflects genetic and phenotypic differences within a tree species. The plant hydraulic system regulates the carbon economy, and therefore variations in growth and wood density. A whole-organism perspective, by partitioning the hydraulic system, is crucial for understanding the physical and physiological processes that coordinately mediate plant growth. The aim of this study was to determine whether the relationships and trade-offs between (i) hydraulic traits and their relative contribution to the whole-plant hydraulic system, (ii) plant water transport, (iii) CO2 assimilation, (iv) plant growth, and (v) wood density are revealed at the interclonal level within a variable population of 10 Pinus radiata (D. Don) clones for these characters. We demonstrated a strong coordination between several plant organs regarding their hydraulic efficiency. Hydraulic efficiency, gas exchange, and plant growth were intimately linked. Small reductions in stem wood density were related to a large increase in sapwood hydraulic efficiency, and thus to plant growth. However, stem growth rate was negatively related to wood density. We discuss insights explaining the relationships and trade-offs of the plant traits examined in this study. These insights provide a better understanding of the existing coordination, likely to be dependent on genetics, between the biophysical structure of wood, plant growth, hydraulic partitioning, and physiological plant functions in P. radiata.

Manipulation of soil methane oxidation under drought stress.

Drought events are predicted to occur more frequently, but comprehensive knowledge of their effects on methane (CH4) oxidation by soil methanotrophs in upland ecosystems remains elusive. Here, we put forward a new conceptual model in which drought influences soil CH4 oxidation through a direct pathway (i.e., positive effects of soil CH4 oxidation via increasing soil aeration) and through an indirect pathway (i.e., negative effects of in planta ethylene (C2H4) production on soil CH4 oxidation). Through measuring soil CH4 efflux along a gradient of drought stress, we found that drought increases soil CH4 oxidation, as the former outweighs the latter on soil CH4 oxidation, based on a mesocosm experiment employing distinct levels of watering and a long-term drought field trial created by rainfall exclusion in a subtropical evergreen forest. Moreover, we used aminoethoxyvinylglycine (AVG), a C2H4 biosynthesis inhibitor, to reduce in planta C2H4 production under drought, and found that reducing in planta C2H4 production increased soil CH4 oxidation under drought. To confirm these findings, we found that inoculation of plant growth-promoting rhizobacteria containing the 1-aminocyclopropane-1-carboxylate deaminase alleviated the negative effects of drought-induced in planta C2H4, thus increasing soil CH4 oxidation rates. All these results provide strong evidence for the hypothesis that in planta C2H4 production inhibits soil CH4 oxidation under drought. To our knowledge, this is the first study to manipulate the negative feedback between C2H4 production and CH4 oxidation under drought stress. Given the current widespread extent of arid and semiarid regions in the world, combined with the projected increased frequency of drought stress in future climate scenarios, we provide a reliable means for increasing soil CH4 oxidation in the context of global warming.

Aquaporin regulation in roots controls plant hydraulic conductance, stomatal conductance, and leaf water potential in Pinus radiata under water stress.

Stomatal regulation is crucial for forest species performance and survival on drought-prone sites. We investigated the regulation of root and shoot hydraulics in three Pinus radiata clones exposed to drought stress and its coordination with stomatal conductance (gs ) and leaf water potential (Ψleaf ). All clones experienced a substantial decrease in root-specific root hydraulic conductance (Kroot-r ) in response to the water stress, but leaf-specific shoot hydraulic conductance (Kshoot-l ) did not change in any of the clones. The reduction in Kroot-r caused a decrease in leaf-specific whole-plant hydraulic conductance (Kplant-l ). Among clones, the larger the decrease in Kplant-l , the more stomata closed in response to drought. Rewatering resulted in a quick recovery of Kroot-r and gs . Our results demonstrated that the reduction in Kplant-l , attributed to a down regulation of aquaporin activity in roots, was linked to the isohydric stomatal behaviour, resulting in a nearly constant Ψleaf as water stress started. We concluded that higher Kplant-l is associated with water stress resistance by sustaining a less negative Ψleaf and delaying stomatal closure.

Phenotyping Whole Forests Will Help to Track Genetic Performance.

Phenotyping is the accurate and precise physical description of organisms. Accurate and quantitative phenotyping underpins the delivery of benefits from genetic improvement programs in agriculture. In forest trees, phenotyping at an equivalent precision has been impossible because trees and forests are large, long-lived, and highly variable. These facts have restricted the delivery of genetic gains in forestry compared to other agricultural sectors. We describe a landscape-scale phenotyping platform that integrates remote sensing, spatial information systems, and genomics to facilitate the delivery of greater gains enabling forestry to catch up with other sectors. Combining remote sensing at a range of spatial and temporal scales with genomics will ultimately impact on tree breeding globally.

Genotypic variation in Pinus radiata responses to nitrogen source are related to changes in the root microbiome.

Variation in traits within a plant species contributes to differences in soil physicochemistry and rhizosphere microbial communities. However, how intraspecific variation in plant responses to nitrogen (N) shapes these communities remains unclear. We studied whether plant responses to organic and inorganic N forms vary among genotypes, and if these responses were associated with variation in root-associated communities. We investigated how the root microbiomes of two Pinus radiata D. Don genotypes were altered by two years of N-fertilisation in field conditions. We characterised rhizosphere bacterial and fungal communities, as well as root-associated fungal communities, of trees receiving yearly additions of NH4NO3 or L-arginine, and control trees. We also measured plant traits and rhizosphere soil physicochemical properties. Two main findings emerged: (i) N form and tree genotype affected soil physicochemical properties as well as plant measures, and these responses were associated with variation in microbial communities, and (ii) rhizosphere and root-associated communities differed in their responses to N form and host genotype. Our results suggest that N forms have different influences on N and carbon dynamics at the plant-soil interface by inducing root-mediated responses that are associated with shifts in the root microbiome such that communities more closely associated with roots are more sensitive to genotype-specific responses.

Host Genotype and Nitrogen Form Shape the Root Microbiome of Pinus radiata.

A central challenge in community ecology is understanding the role that phenotypic variation among genotypes plays in structuring host-associated communities. While recent studies have investigated the relationship between plant genotype and the composition of soil microbial communities, the effect of genotype-by-environment interactions on the plant microbiome remains unclear. In this study, we assessed the influence of tree genetics (G), nitrogen (N) form and genotype-by-environment interaction (G x N) on the composition of the root microbiome. Rhizosphere communities (bacteria and fungi) and root-associated fungi (including ectomycorrhizal and saprotrophic guilds) were characterised in two genotypes of Pinus radiata with contrasting physiological responses to exogenous organic or inorganic N supply. Genotype-specific responses to N form influenced the composition of the root microbiome. Specifically, (1) diversity and composition of rhizosphere bacterial and root-associated fungal communities differed between genotypes that had distinct responses to N form, (2) shifts in the relative abundance of individual taxa were driven by the main effects of N form or host genotype and (3) the root microbiome of the P. radiata genotype with the most divergent growth responses to organic and inorganic N was most sensitive to differences in N form. Our results show that intraspecific variation in tree response to N form has significant consequences for the root microbiome of P. radiata, demonstrating the importance of genotype-by-environment interactions in shaping host-associated communities.

The right tree for the job? perceptions of species suitability for the provision of ecosystem services.

Stakeholders in plantation forestry are increasingly aware of the importance of the ecosystem services and non-market values associated with forests. In New Zealand, there is significant interest in establishing species other than Pinus radiata D. Don (the dominant plantation species) in the belief that alternative species are better suited to deliver these services. Significant risk is associated with this position as there is little objective data to support these views. To identify which species were likely to be planted to deliver ecosystem services, a survey was distributed to examine stakeholder perceptions. Stakeholders were asked which of 15 tree attributes contributed to the provision of five ecosystem services (amenity value, bioenergy production, carbon capture, the diversity of native habitat, and erosion control/water quality) and to identify which of 22 candidate tree species possessed those attributes. These data were combined to identify the species perceived most suitable for the delivery of each ecosystem service. Sequoia sempervirens (D.Don) Endl. closely matched the stakeholder derived ideotypes associated with all five ecosystem services. Comparisons to data from growth, physiological and ecological studies demonstrated that many of the opinions held by stakeholders were inaccurate, leading to erroneous assumptions regarding the suitability of most candidate species. Stakeholder perceptions substantially influence tree species selection, and plantations established on the basis of inaccurate opinions are unlikely to deliver the desired outcomes. Attitudinal surveys associated with engagement campaigns are essential to improve stakeholder knowledge, advancing the development of fit-for-purpose forest management that provides the required ecosystem services.

Methane oxidation needs less stressed plants.

Methane oxidation rates in soil are liable to be reduced by plant stress responses to climate change. Stressed plants exude ethylene into soil, which inhibits methane oxidation when present in the soil atmosphere. Here we discuss opportunities to use 1-aminocyclopropane-1-carboxylate deaminase to manage methane oxidation by regulating plant stress responses.

Effect of nitrogen and waterlogging on denitrifier gene abundance, community structure and activity in the rhizosphere of wheat.

Microbial denitrification plays a key role in determining the availability of soil nitrogen (N) to plants. However, factors influencing the structure and function of denitrifier communities in the rhizosphere remain unclear. Waterlogging can result in root anoxia and increased denitrification, leading to significant N loss from soil and potential nitrous oxide (N(2)O) emissions. This study investigated denitrifier gene abundance, community structure and activity in the rhizosphere of wheat in response to anoxia and N limitation. Denitrifier community structure in the rhizosphere differed from that in bulk soil, and denitrifier gene copy numbers (nirS, nirK, nosZ) and potential denitrification activity were greater in the rhizosphere. Anoxia and N limitation, and in particular a combination of both, reduced the magnitude of this effect on gene abundance (in particular nirS) and activity, with N limitation having greater impact than waterlogging in rhizosphere soil, in contrast to bulk soil where the impact of waterlogging was greater. Increased N supply to anoxic plants improved plant health and increased rhizosphere soil pH, which resulted in enhanced reduction of N(2)O. Both anoxia and N limitation significantly influenced the structure and function of denitrifier communities in the rhizosphere, with reduced root-derived carbon postulated to play an important role.

Influence of sewage and pharmaceuticals on soil microbial function.

Although sewage effluent application to land is a common approach to recycle water and provide nutrients to plants, bioactive pharmaceuticals contained in sewage may change soil quality by affecting soil microbial communities. Establishing causal effects, however, is difficult, because trace levels of pharmaceuticals are confounded with other effluent constituents. Therefore, two originally similar soil microbial communities, one irrigated in situ with sewage effluent for 12 years and another nonirrigated, were exposed to high levels of acetaminophen, aspirin, carbamazepine, chlorpromazine, and tetracycline. The objectives of the current study were to determine the influence of high levels of pharmaceuticals on several soil microbial properties, the effect that prolonged effluent irrigation with ambient levels of pharmaceuticals had on soil microbial function, and how this effect would change in response to pharmaceutical exposure. Several pharmaceuticals, at high exposure levels, imposed stress on the soil microbial community as judged by increased CO(2) respiration, decreased biomass carbon, and altered substrate utilization affinities. Prolonged effluent irrigation, which altered the genetic fingerprint of the microbial community, also mitigated the response that exposure to pharmaceuticals had on the microbial community and enabled degradation of the antimicrobial salicylic acid after aspirin exposure. In conclusion, prolonged irrigation with sewage effluent containing pharmaceuticals at ambient levels influenced the microbial community so that they were able to better cope with sudden exposure to high levels of pharmaceuticals.

Relating nutritional and physiological characteristics to growth of Pinus radiata clones planted on a range of sites in New Zealand.

Six clones of radiata pine with known differences in growth rate were examined for clonal nutritional characteristics and for physiological determinants of clonal growth rate. We compared growth, foliar characteristics and nutrient, ¹³C and ¹5N concentration data for the six clones in 4- to 6-year-old field trials planted over a range of nutritionally contrasting sites. These data were also compared with growth, nutrient uptake and remobilization, foliar characteristic and gas exchange data from intensive physiological glasshouse experiments using 1- and 2-year-old plants of the same clones. Significant genotype x environment interactions in our field experiments conducted over strong nutritional gradients allowed us to identify radiata pine clones with consistent, superior growth and nutritional characteristics and clones that may be suited to particular site conditions. Our results suggest that the opportunity exists to exploit clone x site variation for site-specific clonal deployment and planting of fast-growing clones could be accompanied by planting of clones able to take relative advantage of site nutritional characteristics. Faster tree growth was not strongly related to any physiological characteristic, and the factors influencing growth rate differed among clones. The fastest-growing clone had consistent, high uptake of all nutrients, high fascicle weights and high water-use efficiency.

The influence of nitrogen and phosphorus supply and genotype on mesophyll conductance limitations to photosynthesis in Pinus radiata.

Mesophyll conductance, g(m), may pose significant limitations to photosynthesis and may be differentially affected by nutrition and genotype in Pinus radiata D. Don. Simultaneous measurements of gas exchange and chlorophyll fluorescence were made to determine g(m), using the constant J method (Harley, P.C., F. Loreto, G. Di Marco and T.D. Sharkey. 1992. Theoretical considerations when estimating the mesophyll conductance to CO(2) flux by analysis of the response of photosynthesis to CO(2). Plant Physiol. 98:1429-1436), in a fast- and a slow-growing clone of P. radiata grown in a greenhouse with a factorial combination of nitrogen (N) and phosphorus (P) supply. Values of g(m) increased linearly with the rate of photosynthesis at saturating irradiance and ambient CO(2) concentration, A(sat) (g(m) = 0.020A(sat), r(2) = 0.25, P < 0.001) and with stomatal conductance to CO(2) transfer, g(s) (g(m) = 1.16g(s), r(2) = 0.14, P < 0.001). Values of g(m) were greater than those of stomatal conductance, g(s), and the ratio (g(m)/g(s)) was not influenced by single or combined N and P additions or clone with a mean (+/-SE) value of 1.22 +/- 0.06. Relative limitations to mesophyll conductance, L(m) (16%) to photosynthesis, were generally greater than those imposed by stomata, L(s) (13%). The mean (+/-SE) CO(2) concentration in the intercellular air spaces (C(i)) was 53 +/- 3 mumol mol(-1) lower than that in the atmosphere (C(a)). Mean (+/-SE) CO(2) concentration in the chloroplasts (C(c)) was 48 +/- 2 mumol mol(-1) lower than C(i). Values of L(s), L(m) and CO(2) diffusion gradients posed by g(s) (C(a) - C(i)) and g(m) (C(i) - C(c)) did not significantly differ with nutrient supply or clone. Mean values of V(cmax) and J(max) calculated on a C(c) basis were 15.4% and 3.1% greater than those calculated on a C(i) basis, which translated into different slopes of the J(max)/V(cmax) relationship (C(c) basis: J(max) = 2.11V(cmax), r(2) = 0.88, P < 0.001; C(i) basis: J(max) = 2.43V(cmax), r(2) = 0.86, P < 0.001). These results will be useful for correcting estimates of V(cmax) and J(max) used to characterize the biochemical properties of photosynthesis for P. radiata.

The influence of N and P supply and genotype on carbon flux and partitioning in potted Pinus radiata plants.

Carbon (C) flux and partitioning responses of Pinus radiata (D. Don) clones to a factorial combination of nitrogen (N) and phosphorus (P) supply were estimated in small trees growing in a greenhouse over 44 weeks. Our objective was to use a C budget approach at the plant level to examine how a factorial combination of N and P additions and genotype modify gross primary production (GPP), net primary production (NPP), absolute C fluxes apportioned to aboveground net primary production (ANPP), aboveground plant respiration (APR), total belowground carbon flux (TBCF) and the partitioning of GPP to ANPP, APR and TBCF. Single N or P additions increased plant NPP and GPP similarly, but their combined effects exceeded those of their individual contributions. Nitrogen and to a lesser extent P additions enhanced carbon-use efficiency (CUE, NPP:GPP) and C partitioning to ANPP at the expense of TBCF. The fraction of GPP partitioned to APR was invariant to N or P additions. The ratio of soil respiration (FS) to TBCF was significantly greater in the low-N low-P addition treatment (61%) than in those treatments with single or combined N and P additions (49%). The slowest growing clone partitioned a significantly smaller fraction of GPP to ANPP (29%) than one of the faster-growing genotypes (33%). This research provides insight into how N and P regulate the C fluxes and partitioning in individual plants. Our results contribute to explaining clonal variation in aboveground growth rates and suggest that greater gains in CUE and partitioning to ANPP occur with addition of N rather than P supply.

Factors impacting on pharmaceutical leaching following sewage application to land.

Sewage effluent application to land is a treatment technology that requires appropriate consideration of various design factors. Soil type, level of sewage pre-treatment and irrigation rate were assessed for their influence on the success of soil treatment in removing pharmaceuticals remaining after conventional sewage treatment. A large scale experimental site was built to assess treatment performance in a realistic environment. Of the factors investigated, soil type had the biggest impact on treatment performance. In particular, carbamazepine was very efficiently removed (>99%) when irrigated onto a volcanic sandy loam soil. This was in contrast to irrigation onto a sandy soil where no carbamazepine removal occurred after irrigation. Differences were likely caused by the presence of allophane in the volcanic soil which is able to accumulate a high level of organic matter. Carbamazepine apparent adsorption distribution coefficients (K(d)) for both soils when irrigated with treated sewage effluent were determined as 25 L kg(-1) for the volcanic soil and 0.08 L kg(-1) for the sandy soil. Overall, a volcanic soil was reasonably efficient in removing carbamazepine while soil type was not a major factor for caffeine removal. Removal of caffeine, however, was more efficient when a partially treated rather than fully treated effluent was applied. Based on the investigated pharmaceuticals and given an appropriate design, effluent irrigation onto land, in conjunction with conventional sewage treatment may be considered a beneficial treatment for pharmaceutical removal.

Technological options for the management of biosolids.

BACKGROUND, AIM, AND SCOPE: Large quantities of biosolids (sewage sludge), which are produced from municipal wastewater treatment, are ever-increasing because of the commissioning of new treatment plants and continuous upgrades of the existing facilities. A large proportion of biosolids are currently landfilled. With increasing pressure from regulators and the general public, landfilling of biosolids is being phased out in many countries because of potential secondary pollution caused by leachate and the emission of methane, a potent greenhouse gas. Biosolids contain nutrients and energy that can be used beneficially. Significant efforts have been made recently to develop new technologies to manage biosolids and make useful products from them. In this paper, we provide a review of the technologies in biosolids management. MATERIALS AND METHODS: A survey of literature was conducted. RESULTS: At present, the most common beneficial use of biosolids is agricultural land application because of inherent fertilizer values found in biosolids. Expansion of land application, however, may be limited in the future because of more stringent regulatory requirements and public concern about food chain contamination in some countries. Perceived as a green energy source, the combustion of biosolids has received renewed interest. Anaerobic digestion is generally a more effective method than incineration for energy recovery, and digested biosolids are suitable for further beneficial use through land application. Although conventional incineration systems for biosolid management generally consume more energy than they produce because of the high moisture content in the biosolids, it is expected that more combustion systems, either monocombustion or cocombustion, will be built to cope with the increasing quantity of biosolids. DISCUSSION: Under the increasingly popular low-carbon economy policy, biosolids may be recognized as a renewable fuel and be eligible for 'carbon credits'. Because ash can be used to manufacture construction materials, combustion can provide a complete management for biosolids. A number of advanced thermal conversion technologies (e.g., supercritical water oxidation process and pyrolysis) are under development for biosolids management with a goal to generate useful products, such as higher quality fuels and recovery of phosphorus. With an ever-increasing demand for renewable energy, growing bioenergy crops and forests using biosolids as a fertilizer and soil amendment can not only contribute to the low-carbon economy but also maximize the nutrient and carbon value of the biosolids. CONCLUSIONS: Land application of biosolids achieves a complete reuse of its nutrients and organic carbon at a relatively low cost. Therefore, land application should become a preferred management option where there is available land, the quality of biosolids meet regulatory requirements, and it is socially acceptable. Intensive energy cropping and forest production using biosolids can help us meet the ever-increasing demand for renewable energy, which can eliminate the contamination potential for food sources, a common social concern about land application of biosolids. In recent years, increasing numbers of national and local governments have adopted more stringent regulations toward biosolid management. Under such a political climate, biosolids producers will have to develop multireuse strategies for biosolids to avoid being caught because a single route management practice might be under pressure at a short notice. Conventional incineration systems for biosolids management generally consume more energy than they produce and, although by-products may be used in manufacturing, this process cannot be regarded as a beneficial use of biosolids. However, biosolids are likely to become a source of renewable energy and produce 'carbon credits' under the increasingly popular, low-carbon economy policy. RECOMMENDATIONS AND PERSPECTIVES: To manage biosolids in a sustainable manner, there is a need for further research in the following areas: achieving a higher degree of public understanding and acceptance for the beneficial use of biosolids, developing cost-efficient and effective thermal conversions for direct energy recovery from biosolids, advancing technology for phosphorus recovery, and selecting or breeding crops for efficient biofuel production.

Partititioning concurrent influences of nitrogen and phosphorus supply on photosynthetic model parameters of Pinus radiata.

Responses of photosynthesis (A) to intercellular CO(2) concentration (C(i)) were measured in a fast- and a slow-growing clone of Pinus radiata D. Don cultivated in a greenhouse with a factorial combination of nitrogen and phosphorus supply. Stomatal limitations scaled with nitrogen and phosphorus supply as a fixed proportion of the light-saturated photosynthetic rate (18.5%) independent of clone. Photosynthetic rates at ambient CO(2) concentration were mainly in the V(cmax)-limited portion of the CO(2) response curve at low-nitrogen supply and at the transition between V(cmax) and J(max) at high-nitrogen supply. Nutrient limitations to photosynthesis were partitioned based on the ratio of foliage nitrogen to phosphorus expressed on a leaf area basis (N(a)/P(a)), by minimizing the mean square error of segmented linear models relating photosynthetic parameters (V(cmax), J(max), T(p)) to foliar nitrogen and phosphorus concentrations. A value of N(a)/P(a) equal to 23 (mole basis) was identified as the threshold separating nitrogen (N(a)/P(a) < or = 23) from phosphorus (N(a)/P(a) > 23) limitations independent of clones. On an area basis, there were significant positive linear relationships between the parameters, V(cmax), J(max), T(p) and N(a) and P(a), but only the relationships between T(p) and N(a) and P(a) differed significantly between clones. These findings suggest that, in genotypes with contrasting growth, the responses of V(cmax) and J(max) to nutrient limitation are equivalent. The relationships between the parameters V(cmax), J(max), T(p) and foliage nutrient concentration on a mass basis were unaffected by clone, because the slow-growing clone had a significantly greater leaf area to mass ratio than the fast-growing clone. These results may be useful in discriminating nitrogen-limited photosynthesis from phosphorus-limited photosynthesis.

Modelling environmental variation in Young's modulus for Pinus radiata and implications for determination of critical buckling height.

BACKGROUND AND AIMS: Although density-specific stiffness, E/rho, (where E is Young's modulus and rho is wood density) is often assumed constant by the elastic similarity model, and in determination of critical buckling height (H(crit)), few studies have tested this assumption within species. Here this assumption is tested for Pinus radiata growing across an environmental gradient, and theory is combined with data to develop a model of Young's modulus. METHODS: Analyses use an extensive series of environmental plots covering the range of climatic and edaphic conditions over which P. radiata is grown in New Zealand. Reduced major axis regression was used to determine scaling exponents between log-log plots of H(crit) vs. groundline diameter (D), and E/rho vs. D. Path analysis was used to identify significant direct and indirect (through stem slenderness) edaphic and climatic influences on E. KEY RESULTS: Density-specific stiffness exhibited 3-fold variation. As E/rho scaled positively with D, the exponent of 0.95 between H(crit) and D exceeded the assumed value of 0.67 under constant E/rho. The final path analysis model included mean air temperature in early autumn (T(aut)) and slenderness as significant (P < 0.05) positive direct influences on E. Tree leaf area index and T(aut) were indirectly associated with E through their significant (P < 0.05) positive direct relationship with stem slenderness. Young's modulus was most sensitive to T(aut), followed by stem slenderness then leaf area index, and the final model explained 76 % of the variance in E. CONCLUSIONS: The findings suggest that within species E/rho variation may influence H(crit) and the scaling exponent between D and H(crit) so important in assumptions regarding allometric relationships. The model presented may provide a useful means of determining variation in E, E/rho and H(crit) across environmental gradients.

Transpiration rates and canopy conductance of Pinus radiata growing with different pasture understories in agroforestry systems.

We measured tree transpiration and canopy conductance in Pinus radiata D. Don at two low rainfall sites of differing soil fertility in Canterbury, New Zealand. At the more fertile Lincoln site, we also assessed the effects of two common pasture grasses on tree transpiration and canopy conductance. At the less fertile Eyrewell Forest site, the effect of no understory, and the effects of irrigation in combination with mixtures of grass or legume species were determined. Tree xylem sap flux (F(d)') was measured by the heat pulse method. Total canopy conductance to diffusion of water vapor (G(t)) was calculated by inverting a simplified Penman-Monteith model. The different treatment effects were modeled by the simple decaying exponential relationship G(t) = G(tmax)e((-bD)), where D = air saturation deficit. At the Lincoln site, trees with an understory of cocksfoot had lower F(d)' and G(tmax) than trees with an understory of ryegrass, although the sensitivity of G(t) to increasing D (i.e., the value of b) did not differ between treatments. At the Eyrewell site, irrigation only increased F(d)' in the absence of an understory, whereas the presence of understory vegetation, or lack of irrigation, or both, significantly reduced G(tmax) and increased b. We conclude that the selection of understory species is critical in designing successful agroforestry systems for low rainfall areas.