INTRODUCTION
Climate change, mainly manifested as temperature rise and variation in precipitation patterns, has already induced changes in forest ecosystems (Allen et al., 2015). Tree mortality events are expected to increase in large areas of the planet under the current scenario of climate change, with higher temperature and water stress as a common causal factors (Allen et al., 2010; Berry et al., 2014a). These factors affect plant physiology, metabolism and growth, as well as tree productivity (Giammarchi et al., 2017) since forest productivity is the result of convergent physiological and biochemical aspects acting in a coordinated way. Such climate variations have had a negative effect on the adaptive response of forest ecosystems (Cristiansen, 2016).
Nowadays fast and precise analytical techniques, such as isotopic analyses, have been used in forest science to understand mechanisms of plant adaptation to climate change and productivity. These techniques are mainly based on the evaluation of photosynthetic activity, water relations and nitrogen use efficiency (Di Matteo et al., 2017). The use of stable isotope analysis is an important research line since 1927, showing significant advances in the last 20 years; nevertheless, there are still several limitations for the understanding of its relationship with plant development promotion and the explanation of possible physiological mechanisms of adaptation to conditions commonly present in forest ecosystems.
The present review compiles some of the scientific results achieved by various researchers that focus their studies on the forest species response to the environment. The topics to be addressed are on the use of stable isotopes for the measurement of photosynthetic activity, water absorption, transpiration, and water and nitrogen use efficiency. In addition, some medium-term perspectives of the isotopic composition and discrimination of 13C, 18O, and 15N involved in forest trees metabolism and productivity will be considered.
HISTORY OF STABLE ISOTOPES AND THEORETICAL ASPECTS
Carbon- 13C
Around 1927 and 1932, light bioelements such as carbon, nitrogen, hydrogen and oxygen isotopes were discovered (Lehmann, 2017). The first studies of the carbon isotope composition in plants showed that it had lower 13C value than carbonates. Later, studies of variations in abundance of 13C in marine and terrestrial plants lead to propose a schematic representation of isotopic fractionation during the photosynthetic cycle for terrestrial plants (Craig and Gordon, 1965). This fractionation on plants is based on the existence of two stable carbon isotopes (12C and 13C) forming part of the atmospheric carbon dioxide (CO2), which are in approximate proportions of 99 % and 1%, respectively. Plant tissues have a different distribution of 13C/12C isotopes in comparison to the atmosphere, mainly associated to the biologic carbon incorporation into the plants, which includes processes of intracellular uptake and diffusion, and photosynthetic fixation of atmospheric CO2 (O’Leary, 1981). Farquhar et al. (1982) developed the theory of isotopic composition in plants, where isotopic discrimination was modeled for diffusion, carboxylation, photorespiration and respiration effects.
The composition of 13C in a plant tissue is determined through mass spectrometry, as an isotopic ratio (R) of the sample of interest in relation to a standard ratio (O’Leary, 1981) as described in Equation 1.
Where δ13C is an isotopic signature, a measure of the ratio of stable isotopes 13C:12C, reported in parts per thousand; R is the molar abundance ratio 13CO2/12CO2 for the sample of interest and the standard, where the standard relation is 0.01124.
It is recommended to use the term isotopic discrimination of 13C, as this provides directly an integrated value of the biologic processes that interacted throughout the plants biological cycle (Dawson et al., 2002; Farquhar et al., 1989), and also provides an independent value of the standard isotopic relation and the source isotopic ratio (atmosphere) (Farquhar et al., 1989). Then, the isotopic discrimination value is expressed as a difference between the source and sample isotopic compositions, where the atmosphere is the source (O’Leary, 1981) (Equation 2).
For plants with C3 metabolism, a comprehensive model of photosynthetic discrimination was developed (Equation 3), where the discrimination is a consequence of Rubisco fractionation, stomatal conductance, mesophyll conductance, respiration, and photorespiration, as well as ternary effects (Farquhar and Cernusak, 2012; Ubierna and Farquhar, 2014). Nevertheless, because the equation has several parameters that are to estimate or to measure directly, a simplified discrimination model becomes relevant (Equation 4) (Ubierna and Farquhar, 2014). In this expression, the discrimination values are obtained for the variation of photosynthetic capacity or stomatal aperture (conductance).
Where a is the discrimination caused by CO 2 diffusion through the boundary layer and the stomatal pore, which has 4.4 % and b is the discrimination realized by the carboxylation enzyme ribulose-1,5-bisphosphate carboxylase oxygenase (Rubisco, C 3 metabolism) which has a 27 % value; C i and C a are the CO 2 partial pressures inside the leaf and in the atmosphere, respectively (Farquhar et al., 1989).
Oxigen- 18O
Gonfiantini et al. (1965) demonstrated that 18O be used for studies on water-plant relationships, as isotopic water composition on the leaf is enriched in heavy isotopes when transpiration occurs. 16O and 18O, used in plant tissues studies have a proportion of 99.7 % and 0.21 %, respectively (Barbour, 2007).
Analogous to the carbon 13 (δ13C) composition, oxygen composition on vegetal tissue is determined as the isotopic ratio of the sample related to a standard ratio which, for oxygen, is commonly Vienna-Standard Mean Oceanic Water (VSMOW), the mean concentration of present-day ocean water with a molar abundance ratio value 18O/16O of 2.0052 × 10-3 (Equation 5) (Barbour, 2007; Dansgaard, 1964).
Nevertheless, it is recommended to use the term isotopic enrichment of oxygen 18, which eliminates the variability factor in isotopic composition terms of water source (Barbour, 2007), it also provides an independent value of the isotopic standard relation. Therefore, the isotopic enrichment value is expressed as a difference between the source and the sample isotopic compositions, where soil water is the source (Equation 6).
Oxygen stable isotopes in plants can be used for identification of water source (Ehleringer et al.,1991), a record of precipitation patterns in growth rings in woody plants and evaporative enrichment of leaf water as a consequence of physical and physiological factors (Barbour and Farquhar, 2000; Roden and Ehleringer, 1999). Specifically, evaporative enrichment of leaf water above source water, can be represented by applying the Craig and Gordon (1965) model for evaporative enrichment for the free water surface, adding some modifications (Equation 7) (Farquhar and Lloyd,1993).
Where, ea and ei are vapor pressures for atmosphere and intercellular spaces, ∆Ov is the isotopic relationship of water vapor relative to a water source, εk is the diffusive fractionation trough boundary layer and stomata and ε+ is a proportional depression of equilibrium vapor pressure of H2 18O respect to H2 16O.
Nitrogen- 15N
Two stable isotopes conform to atmospheric nitrogen (14N and 15N), the light isotope is 99.6337 %, while the heavy one is 0.3663 % approximately (Handley and Raven, 1992). In atmosphere 15N/14N ratio is very constant; therefore, atmospheric N2 is used as a standard to report values of 15N isotopic composition (Shearer and Kohl, 1993; Stewart, 2013).
Analogous to isotopic discrimination of 13C, for 15N will exist fractionation during biochemical, biogeochemical and physiological processes. Isotopic composition of N can provide information about inputs of N fertilizer, sources of N available for plant growth, N inputs through N2 fixation by free-living and symbiotic organisms, and estimates of plant fixation, because non-fixing plants have higher δ15N than fixing plants (Boddey et al., 2000; Dawson et al., 2002); however, 15N measurements have been less used than other isotopes (Robinson et al., 2000), but these measurements are most useful when there is a multiple isotope approximation (Griffiths, 1991).
Stable isotopes studies in the measurement of photosynthesis
Plants may react to rising levels of atmospheric CO2 and temperature by increasing the water-use efficiency (WUE) through either an increased photosynthetic rate or by a reduced water transpiration at a higher vapor pressure deficit (VPD) (Farquhar et al., 1989). These two physiological processes result in carbon and oxygen isotopic fractionation. The response of tree growth to increased CO2 is far from being straightforward and seems to be strongly dependent on site conditions, as it can interact with other drivers, such as warming-induced drought and physiological acclimation to high carbon dioxide levels, potentially reducing the ability of forests to act in the long-term as carbon sinks (Gómez-Guerrero et al., 2013). Tree growth and intrinsic water-use efficiency (iWUE) have been observed not to increase as expected or even decline (Lévesque et al., 2014).
It is well known that stable isotopes can serve as integrators or tracers of many key physical and biological processes. Carbon fixation during the process of photosynthesis discriminates against the heavier stable isotope of carbon (13C) in favor of the lighter isotope (12C), but the intensity of this discrimination (∆13C) depends on environmental conditions such as vapor pressure deficit, soil-water availability, as well as physiological responses such as stomatal conductance. The carbon isotopic composition of C3 plant tissues is often expressed as carbon isotope discrimination, a parameter used to track how environmental conditions affect leaf gas exchange (Cornejo-Oviedo et al., 2017).
Stable isotope ratios also provide time-integrated information on plant ecophysiological responses to changing abiotic conditions and can also help to characterize species-specific plant water use strategies. Plant δ13C provides an integrated record of the ratio of intercellular to atmospheric CO2 concentrations (Ci/Ca) during the period in which the carbon was fixed (Herrero et al., 2013). Factors affecting net photosynthetic rate and stomatal conductance (g s ) influence Ci, and thus, plant carbon isotope ratio (Martin-Benito et al., 2017).
Carbon isotope discrimination is typically employed as a proxy for a plant intrinsic water-use efficiency, or a ratio of net assimilation to stomatal conductance. Tree-rings are composed of annual increments of xylem tissue, so increment cores can be used to retrospectively estimate past diameter or basal area growth (Cornejo-Oviedo et al., 2017). Furthermore, carbon stable isotopes in the tree ring record a canopy-integrated signal of annual leaf gas-exchange (Martin-Benito et al., 2017).
Measurements of the stable carbon isotope ratio (δ13C) on annual tree rings offer new opportunities to evaluate mechanisms of variations in photosynthesis and stomatal conductance under changing CO2 and climate conditions, especially in conjunction with process-based biogeochemical model simulations.
A possible perspective for the mid-future should be to estimate photosynthesis along time using stable isotope analysis. Although it has been a recurrent theme for physiologists, biochemists and geneticists there is no agreement and the use of IRGA (Infrared gas analyzer system) for its measurement has been continued.
Stable isotope studies on water uptake and water use efficiency determination (iWUE)
Several forest species worldwide are strongly impacted by different scenarios of climate change (Feng et al., 2016); for example, in arid and semi-arid regions precipitation patterns have changed in relation to species seasonal-phenology. Under such conditions, water availability decreases considerably because precipitation takes place during plant dormancy (Wang et al., 2017). When plants recover cellular dynamism, a significant metabolic readjustment takes place to guarantee water uptake (Hsiao and Acevedo, 1974). The rainfall can be used by plants although a significant volume of water may be lost (Ehleringer and Dawson, 1992; Saiter et al., 2016). Even though there is relatively low precipitation in arid and semi-arid regions, individual rainfall events cause water source isotopic enrichment in the short-term, which plays important roles in shaping plant adaptation in water use strategies (Farquhar et al., 1989).
Stable isotope studies of 13C, 18O and 2H have proved useful in identifying the origin of water consumed, in knowing the photosynthetic pathway and studying possible changes in the transpiration rate in many species; an example of this is in Prosopis tamarugo Phil. (Garrido et al., 2016).
The isotopic composition of the water source, plant water, and growth rings cellulose can be used to analyze the water use efficiency patterns in short and long periods. As a medium-term perspective for ecosystems, which have been affected by climate change scenarios, stable isotope studies can provide accurate and consistent information on the plants stress intensity and its relationship with lumber quality.
Foliar water uptake describes the process by which plants absorb water into their leaves, resulting in a net increase in the mass of water in the leaf (Ehleringer and Dawson, 1992; Cernusak and Kahmen, 2013). This occurs when saturated atmospheric water vapor conditions result in a driving gradient for water to enter the leaf that is at a more negative water potential (Vesala et al., 2017). The conditions necessary for this phenomenon to occur are often observed in dew- and fog-affected ecosystems such as coastal Mediterranean ecosystems (Baguskas et al., 2016) and tropical montane cloud forests (Schwerbrock and Leuschner, 2017), where fog, often leading to leaf wetting, serves as an alternative plant water source during the dry season. The effects of precipitation events are similar and foliar water uptake has now been described as affecting plant water and carbon relations in more than 70 species from several different ecosystems (Berry et al., 2014b). The capacity for species to do foliar water uptake has frequently been established by means of water isotope labeling experiments (Cassana et al., 2016).
Stable isotope signatures in trees wooden tissues have been successfully used both in natural and controlled conditions, to detect and understand the physiological causes of iWUE changes (Guerrieri et al., 2015). On this respect, Scheidegger et al. (2000) suggested a conceptual physiological model that links changes in C and O isotopic ratios to clarify the role of net assimilation and stomatal conductance in determining iWUE changes, based on the response of the plant to different VPD scenarios. More recently, other authors such as Battipaglia et al. (2013) used this conceptual model to better understand the effect of the environmental drivers and climate change on iWUE changes. Tree size and age can also affect iWUE, probably because of a decrease in stomatal conductance with increasing tree height.
The oxygen isotope compositions of plant tissue in foliar organs have been of interest because current understanding suggests that a sequential environmental and plant physiological record may be preserved from oldest tissues to the newest tissue. Such results could be used to evaluate lumber quality; although several advances have been achieved, still some quality lumber elements remain unclear.
The δ13C measured in the dry matter has been a precise estimator of the intrinsic time-integrated water use efficiency for C3 plants (Farquhar et al., 1989). Also, δ13C has been successfully used in the iWUE study of eucalyptus (Eucalyptus) forests (Zolfaghar et al., 2017) and Prosopis tamarugo (Garrido et al., 2016) submitted to depth gradients of water level, as well as roots depth estimation and use of groundwater in different forest communities (Rumman et al., 2018).
Since the observed variation in δ13C may be due to the increase of net assimilation (with a constant stomatal conductance), or to the stomatal conductance decrease (with net assimilation constant), Scheidegger et al. (2000) proposed an analysis model based on the use of δ13C and δ18O: the Dual Isotope Model. The use of δ18O is based on its negative association with stomatal conductance when the latter is the limiting factor for transpiration (Farquhar et al., 2007). This scenario has been observed in forest communities where relative humidity and water source have a similar isotopic composition (Saurer et al., 2001).
Although Scheidegger et al. (2000) developed their interpretation model in herbaceous species under natural conditions, it has also been applied in woody species such as Quercus frainetto Ten. under natural conditions, as well as in retrospective studies in Fagus sylvatica and Nothofagus spp. (Tognetti et al., 2014), and in Scots pine (Pinus sylvestris) (Voltas et al., 2013).
The efficacy of the model has been evaluated by Roden and Farquhar (2012) by measuring δ13C and δ18O in growth rings of Eucalyptus globulus Labill. and Pinus radiata D. Don. under controlled conditions. The authors evaluated the effect of different abiotic stresses (low irradiance, nitrogen deficit, heat and drought), simulating climate change, on the isotopic composition of wood and cellulose. In that assay, the authors hypothesized that it was possible to predict net assimilation and stomatal conductance behavior from the obtained values. Although the dual isotopic model did not perform well under all the evaluated scenarios, it was particularly efficient under drought and heat stress (under low relative humidity conditions and light restriction), while it had no predictive power under nitrogen deficit stress.
The application of the dual isotope model depends to a great extent on the fulfillment of a series of assumptions, which were summarized in Roden and Siegwolfs (2012). One of the most relevant assumptions is that the variation observed in δ18O must be explained by stomatal conductance, which implies that the variables that affect the evaporative enrichment of water and organic tissue must be constant (mainly water source, leaf temperature and relative humidity). On the other hand, it is necessary to consider the buffer effect that wood capacitance and the presence of reserves can have on the isotopic composition of δ18O and δ13C respectively, which also perform independently.
Although the use of the dual isotopic model implies a careful experimental design and an extensive knowledge of the experimental conditions, its use allows understanding δ13C and δ18O variability under climate change scenarios in forest ecosystems.
Stable isotope studies on nitrogen nutrition measurement
There are different definitions and perspectives for nitrogen use efficiency (NUE) measurement. One of the definitions states that the NUE can be calculated by the yield ratio per unit of available nitrogen in the soil, including residual nitrogen present in the soil and nitrogen fertilizer in forest ecosystems where nitrogen fertilizer is applied, or available nitrogen in non-fertilized soil (Tarvainen and Näsholm, 2017); however, not all the available nitrogen in the plant comes from nitrogen fertilizer. The NUE is a function of the edaphic structure, climatic conditions, interactions between soil and bacterial processes, and the nature of organic and inorganic nitrogen sources, which generally is not included during NUE measurement from (plants nitrogen uptake from the field)/(nitrogen applied). The easiest way to calculate the NUE is based on a partial nitrogen balance (Salamanca-Jimenez et al., 2017).
Given the complexity of the NUE measurements and the variability of loss rates among ecosystems, innovative techniques have been generated for their evaluation, such as stable isotope measurements (Cornejo-Oviedo et al., 2017).
Nitrogen pattern valuation is important given their impact on plant biochemistry and physiology, and their consequences for the structuring of plant communities (Lambers et al., 2008), particularly when it is well known that most forest ecosystems are generally affected by low rainfall distribution, affecting nitrogen nutrition (Balzotti et al., 2016).
Plant uptake of dissolved organic nitrogen (DON) has been proposed to explain some inconsistencies of N balance of semi-arid ecosystem (Houlton et al., 2007) but the direct evidence for the importance of the role of DON in plant nutrition in these ecosystems remains elusive under field conditions, particularly when water availability is low (Pardo et al., 2013). Plant N-limitation is a widespread phenomenon in these ecosystems by the fact that the N supply and the N plants and microbes demand may be discontinuous and temporally asynchronous in arid and semiarid ecosystems (Yahdjian et al., 2011) due to the variability of water availability (Calvo-Rodriguez et al., 2017).
Nitrogen is readily available at higher concentrations during the time that plants and microbes are relatively inactive due to dry soil conditions (Austin et al., 2004), but the large increases in plant N uptake and bursts of microbial activity rapidly exhaust the available N when water stress is alleviated (Evans and Burke, 2013). Nowadays new and efficient methodologies are applied to study the natural abundance of 15N measurements (δ15N) of specific soil N sources, root, shoot and leaf material (Huygens et al., 2016). On the strict condition that the isotopic abundance significantly differs among potential plant N sources, this method provides time-integrated information on preferential N uptake patterns under undisturbed plant rhizosphere conditions and natural resource availabilities. Intrinsically, this approach has the potential to overcome the methodological limitations that call into question the validity of many previous assessments of plant N source partitioning (Jones et al., 2013).
Huygens et al. (2016) published the first report of quantitative time-integrated information on plant N source partitioning patterns under in situ conditions for a semi-arid ecosystem. It was found that all plant species showed similar N preferences and dominantly relied on NO3 - for their N nutrition. Dissolved organic N was an insignificant plant N source in this semi-arid model ecosystem. Additionally, the observed δ15N patterns of soil N pools and plant biomass provide further insight into the soil N cycle and competitive interactions among plants and microbes for N sources (Cornejo-Oviedo et al., 2017).
To the extent that growth responses to N fertilization are influenced by net assimilation/stomatal conductance, differences in carbon isotope discrimination should provide insight into the mechanisms of fertilization response; for example, Brooks and Coulombe (2009) measured tree-ring growth and both carbon and oxygen stable isotopes in response to three levels of nitrogen fertilization (157, 314, 417 kg ha-1) in an 85 year-old Douglas-fir plantation; the annual basal area increment of these trees peaked in the third growing season after fertilization, after which the values decreased slowly back to control levels over the next 20 years. In response to nitrogen fertilization, ∆13C was reduced and iWUE was increased in both earlywood and latewood components, but only for the first three years before returning to pretreatment levels. They interpreted this short-term ∆13C response to an increase in leaf nitrogen and photosynthesis, while the longer-term growth response was attributed to an increase in leaf area. Brooks and Mitchell (2011) found a similar response to nitrogen fertilization (448 kg N ha-1 as urea) in a 41 year-old Douglas-fir plantation. The direct effect of nitrogen on tree growth lasted six years after application, but ∆13C was only reduced for the first few years, again related to an increase in leaf nitrogen and photosynthesis, prior to a subsequent increase in leaf area, which sustained the longer-term increase in growth; in contrast, Balster et al. (2009) did not find a decrease in ∆13C in fertilized Douglas-fir plantations.
Analyses of tissue N concentration and N stable isotopic composition have provided very important information on resource acquisition (Givnish, 2002). A high tissue N concentration could benefit in tropical dry forest plants by enabling high rates of photosynthesis and maximizing carbon gain opportunities during the short-wet season (Santiago et al., 2017). Additionally, a high tissue N concentration has the potential to maximize carbon gain for a given stomatal conductance (Wright et al., 2003) during water deficit. N isotopic composition (δ15N) of plant tissue reflects N sources, and alternative N sources such as biological N-fixation or atmospheric deposition of N might allow contrasting growth forms unique mechanisms to support carbon gain (Craine et al., 2015).
Although studies conducted during the last 10 years have shown significant advances, even in relation to nitrogen discrimination, there are still some hints in carbon and oxygen isotopes that need to be clarified. An important aspect to treat as a future perspective would be to evaluate the nitrogen use efficiency in plants in the decomposed way between the structural and protein nitrogen accumulation.
CONCLUSIONS
This review gathers information on the most recent uses of stable isotope techniques in forest science, mainly based on the variability of response of the physiological variables: gas exchange, water use efficiency and nitrogen nutrition. These analyses have been useful to study the response of trees to climate change in most latitudes, and for the explanation of biochemical mechanisms activated during plant adaptation and productivity. Significant advances in the use of stable isotope techniques in forest science have been reported in scientific literature, we consider that such studies will allow us to calculate and model photosynthetic activity and nitrogen nutrition over time but not specific measurements on those areas are currently being carried out. Such studies will allow a retrospective assessment of possible climatic and biochemical variations that could affect metabolism and adopt forestry alternatives for productivity increase.