Effects of simulated nitrogen deposition on the ecophysiological responses of Populus beijingensis and P. cathayana under intra- and interspecific competition

High N deposition has been recognized as a major factor influencing plant-plant interactions. We aim to discover whether N deposition could affect competitive relationships between poplars to predict threats posed by the introduction of exotic hybrids to native relative species. Intra- and interspecific competition was investigated for an introduced hybrid poplar (Populus beijingensis) and the native paternal species P. cathayana under two N deposition regimes. Under control conditions, P. cathayana grown under either intra- or interspecific competition showed consistently greater above-ground biomass, root-to-shoot ratio (R/S), photosynthetic capacity, higher activities of N-assimilation enzymes in leaves, and preference for N-NO3− than corresponding P. beijingensis. Interspecific competition increased leaf area, specific leaf area (SLA), R/S, and photosynthetic nitrogen use efficiency (PNUE) in P. cathayana, but decreased PNUE in P. beijingensis compared to the values under monoculture. Thus, P. cathayana can gain a competitive advantage over P. beijingensis at low N availability. In contrast, under N deposition, P. beijingensis under mixture showed more positive responses in growth, higher plasticity in biomass allocation and root architecture, and the shift for N-NO3− preference when compared with P. cathayana, which resulted in relatively higher values in leaf area, SLA, and biomass accumulation in P. beijingensis. Nitrogen deposition increased the intrinsic water-use efficiency of conspecific P. beijingensis and heterospecific P. cahtayana. Our results suggest that N deposition could reverse competitive relationships between the poplars, and exotic P. beijingensis has the potential to outcompete native P. cathayana under growing N deposition.


Introduction
Competitive interactions are recognized as an important determinant for the structures, compositions, and dynamics of plant communities because of their great potential for shaping distribution patterns and functional traits of competing species (Eckstein 2010;Kraft et al. 2015). Functional traits' differentiation between competitors not only drives stabilizing niche differences and competitive coexistence and maintains species diversity (Kraft et al. 2015), but also underlies average fitness differences, promoting competitive exclusion (Godoy et al. 2014). Belowground competitive ability is highly correlated with functional traits, including fine root density, surface area, and architectural plasticity involved in nutrient uptake (Aschehoug and Callaway 2014), whereas aboveground competitive ability depends on leaf growth rate and ability to occupy aerial volume (Eckstein 2010). For example, for slow-growth plants in nutrient-poor habitats, competitive superiority can be achieved by allocating more biomass to root systems and adjusting root morphology, such as increasing root length per unit root mass, while in microsites with high nutrient availability, the roots of fast-growing species react rapidly by increasing nutrient uptake kinetics, conferring a competitive advantage (Aerts 1999). Actually, competition outcomes not only depend on nutrient availabilities in soil but can also be readily influenced by light resources in a forest canopy. Competition for limiting soil resources can result in resource allocation trade-offs between below-and above-ground parts of plants, which may consequently cause a light limitation or co-limitation of both light and soil resources (Tilman 1990). For example, under nutrient deficiency, plants that allocate more photosynthate to root production enable them to capture soil nutrients more efficiently at the expense of aboveground parts' growth, resulting in trade-offs in competitive ability for other resources (Aschehoug et al. 2016). However, for pioneer trees with high shade intolerance, like Populus, access to high light is essential to their growth and survival. Strong competitors with a higher height growth rate can occupy canopy space more efficiently, which possibly intensifies oneside competition for light and places competitive counterparts at a disadvantage (Fahey et al. 1998).
In recent decades, due to intensive agriculture and fossil fuel combustion, atmospheric N deposition has increased continuously in China, and this trend is predicted to continue in the coming decades. It is generally believed that excessive N deposition may exert profound effects on terrestrial ecosystem processes. Long-term and excessive N deposition has become a major driver of a decline in species richness and biodiversity loss (Stevens 2019). A large body of evidence has shown that increasing N deposition can alter resource availability and induce asymmetric competition between species, which benefits nitrophilous species but threatens species adapted to infertile conditions (Elias and Agrawal 2021), although a few studies found that N addition alleviated competitive effects (Luo et al. 2014) or that elevated N availability may not reinforce competitive advantages (Bradford et al. 2007). In general, for N-deficient ecosystems, N deposition often causes plants to invest more in shoots than in roots when other nutrients and water are not limited (Leith et al. 1999), thus increasing competition for light and living space (Hautier et al. 2009). However, N deposition accelerates the N cycle, increasing N availability, while other essential nutrients (such as P, K, Ca, and Mg) may become limiting factors . At the individual species' level, N deposition is expected to have a negative impact on the balance of mineral nutrients (such as higher foliar N but lower Ca and Mg, or skewed N:P) and susceptibility to stress conditions (Friedrich et al. 2012), therefore altering plant performance, survival, and fitness (Wedlich et al. 2016). Nonetheless, most of these studies mainly focused on biomass accumulation, growth plasticity, and the morphological reaction of plants to reflect the shift of competition patterns in response to N deposition. Therefore, more knowledge about N allocation and N assimilation physiology is necessary to reveal N use strategies adopted by competitive counterparts under N deposition.
As a result of competition and evolution, sympatric species in N-limited ecosystems can occupy distinct realized niches. For example, niches may differentiate when plants can take up N in different chemical forms, including NH 4 + and NO 3 − . Niche differentiation could promote species co-existence because plants reduce belowground competition for limited N by specializing in the uptake of different N forms and evolving species-specific strategies for N uptake (Andersen et al. 2017). Even in tropical rainforests with relatively high soil N availability and N cycling rates, there is evidence that niche differentiation promotes species co-existence because of different preferences for N forms (Schimann et al. 2008). However, some studies have shown that N form preferences vary little among co-occurring species. For example, three sympatric Eucalyptus species with ecological and taxonomical similarities show remarkably similar preferences for N forms and are likely to compete for N (Paulding et al. 2010), suggesting the absence of niche differentiation. Thus, it is still uncertain whether alterations in N availability and chemical composition in soils as a result of N deposition can change competitive relationships between co-occurring species with close genetic relationships and similar N preferences.
Populus species, a fast-growing and widely distributed tree species in the northern hemisphere, have become one of the most intensive plantation species in China. Under natural conditions, several species of the genus Populus are broadly sympatric and are known to hybridize extensively (Hamzeh and Dayanandan 2004). Natural hybridization among members of poplars has long been recognized as a crucial driving force for genetic exchange, genetic variation and evolution, the origin of new species, and population genetic structure (Abbott et al. 2016). However, some excellent hybrid cultivars of poplars originating from anthropogenic hybridization between exotic and native poplar species have been subjected to rigorous breeding processes and have been demonstrated to have over-parent heterosis (Chen et al. 2014b). It has been realized that intensive application of these poplar cultivars may pose a threat to native poplar species not only by occupying their habitats but also by influencing the genetic integrity of native poplar species via introgressive hybridization and gene flow (Paffetti et al. 2018;Talbot et al. 2012). More importantly, along with the growing habitat fragmentation of the native poplar species due to agriculture development and urbanization, the opportunities for contact between introduced poplar hybrids and the native species greatly increase. This increasing contact may build direct competitive relationships between them, alter the niche of native poplar species, and endanger the growth, survival, and reproduction of native poplars. However, little information is available about the competitive interactions between them, especially in the context of growing N deposition.
Populus beijingensis W. Y. Hsu (P. nigra L. var. italica (Moench.) Koehne × P. cathayana Rehd.), a hybrid bred in 1956, is an excellent hybrid and is commonly used as a major tree species for shelterbelt afforestation and road greening in central, northwest, and parts of northeast China. With higher requirements for nutrients and resources, this species grows faster than native poplar species in sites with sufficient nutrients and water. Populus cathayana Rehd., a pioneer tree species in bare land, is widely distributed in northern, central, and southwestern China, and P. beijingensis has a similar ecological niche to the native paternal species P. cathayana Rehd., which results in competition between poplars for living space and nutrients. Therefore, in the present study, we evaluated the competitive relationships between sympatric species and examined how N deposition affects intra-and interspecific competition by detecting growth characteristics, morphological traits, photosynthetic rates, N metabolism physiology, N preference, and long-term water use efficiency. We tested the following hypotheses: (i) There are species-specific differences in adaptive strategies reflected by morphological and physiological adjustments under intra-and interspecific competition. (ii) Under N deposition, P. beijingensis exhibits competitive advantages over P. cathayana due to its higher resource requirements.

Study site and plant material
This study was conducted in a naturally lit greenhouse providing only shelter from rainfall at the Chengdu campus of Sichuan Agricultural University, China (30°42′ N, 103°51′ E). The mean annual rainfall, temperature, potential evapotranspiration, and duration of the frost-free period in the region are 966.1 mm, 16℃, 838 mm, and 294 days, respectively.
Healthy 1-year-old cuttings of P. cathayana and P. beijingensis were selected from a state forest farm under normal conditions in Qinghai Province (Datong, 101°35'E, 35°56'N), P.R., China. After rinsing in a sterilizing agent (1.25% carbendazol, w/w) for a half hour, the cuttings were planted in a seedbed. For each species, 72 healthy cuttings approximately 15 cm in height and 10 cm in root length were chosen for use in the competition experiment.

Experimental design
Surface soil (0-20 cm) was collected from an old field near the Chengdu campus of Sichuan Agriculture University. The soil type is alluvial soil. The soil was fully homogenized and then used in our study. Two cuttings were transplanted to each cylindrical plastic pot, with a diameter of 40 cm and height of 35 cm, and filled with 45 kg of homogenized soil. The cuttings in each pot were arranged along a diametrical line, and the distance between each cutting and the center of the pot was 10 cm. All cuttings were planted at a similar soil depth (10-15 cm). The physical and chemical soil properties were as follows (kg − 1 dry soil): pH 7.86, soil organic matter 16.12 g, available potassium 21.62 mg, available phosphate 13.78 mg, total nitrogen 1.13 g, total phosphate 0.65 g. Three types of competition patterns were designed: P. cathayana + P. cathayana (CC, intraspecific competition between P. cathayana cuttings); P. beijingensis + P. beijingensis (BB, intraspecific competition between P. beijingensis cuttings); and P. cathayana + P. beijingensis (CB, interspecific competition between P. cathayana cuttings and P. beijingensis cuttings). Then, two regimes of N deposition (i.e., N − and N + ) were arranged in each competition pattern. N − and N + represent the control conditions and simulated N deposition, respectively.
The experimental layout was a completely randomized design with three factors (species, competition pattern, and N deposition). For each species, there were four treatments according to the combination of competition pattern and N deposition. Twelve pots per treatment (three replicates with four pots per replicate) were included in the experiment. Pots were arranged in three blocks in a naturally lit greenhouse. Four replicated pots of each treatment were completely randomized in each block. The blocks were rotated periodically to minimize environmental impact. One month after being transplanted, the plants were subjected to different N treatments. For the N + treatment, N amount was determined based on the area of the soil surface and N deposition level (about 15 g N m − 2 a − 1 ) in local habitats according to the local environmental monitoring bureau. The N forms in precipitation under natural conditions mainly include NH 4 + and NO 3 − with a ratio of nearly 1:1. Therefore, NH 4 NO 3 as exogenous N was used in our study to simulate N deposition. In addition, N deposition amount under field conditions is proportional to the amount of precipitation. Thus, in our study, 65% of annual N deposition was set as the nitrogen amount added in the period of the experiment because the precipitation in the period accounted for 60-70% of annual precipitation in the past five years. All N was divided 17 times to add to the surface of the soil. Each time, 235 ml of 3.75 mM NH 4 NO 3 solution was evenly sprayed onto the potted soil surface each week during the treatment period. A total of 1.20 g NH 4 NO 3 was added to each pot for the N + treatments. The treatment without an additional N supply was watered with 235 ml of deionized water at the same interval. The treatments were conducted under climate conditions with a daytime temperature of 14-22℃, a night-time temperature of 11-18℃, and an average relative humidity of 75% for 17 weeks at Sichuan Agriculture University.

Growth measurement
At the end of the experiment, five cuttings from each treatment were selected randomly to measure plant height and basal diameter. Then, all plants were harvested and divided into leaves, stems, and roots. Furthermore, roots were isolated gently and strictly by branch order, following the methods of Pregitzer et al. (2002), i.e., the distal roots were designated as the first order (R1), roots from which two first-order roots branched were categorized as the second order (R2), and so on. Root samples of different orders, as well as stems and leaves, were dried at 75℃ to a constant weight, and root mass, leaf mass, and stem mass were determined separately. The root/shoot ratio (R/S) was calculated as root mass / (leaf mass + stem mass). In most cases, R1 and R2 are functionally defined as absorptive roots, while R3, R4, and R5 are considered transport roots (Guo et al. 2008). Thus, the biomass ratios of absorptive to transport roots were calculated. The leaf area mentioned below refers to the total leaf area per cutting, which was determined using a portable laser leaf area meter (CI-203, CID Inc., Camas, WA). The specific leaf area (SLA, leaf area per unit dry mass) was calculated as the ratio of the leaf area to its dry mass.

Fine root morphology determination
Before the root biomass assessment, different root orders were positioned to minimize overlap and placed on a flatbed scanner. Root morphological parameters, including total root length, total root surface area, average length of individual root, and average diameter of individual root for different orders, were measured using WinRHIZO image analysis software (Regent Instruments, Quebec, QC, Canada). Then, the specific root surface area (SRA) was calculated by dividing the root surface area by the dry weight of the roots used for scanning.
Determination of net photosynthetic rate and photosynthetic N-use efficiency At the end of the experiment, for each treatment, the fourth fully expanded and intact leaf from five randomly chosen individuals was used to measure the net photosynthetic rate using a portable photosynthesis system (LI-6400; LI-COR Inc. Lincoln, NE, USA). The net photosynthetic rate (P n ) and stomatal conductance (G s ) was measured under the following conditions: ambient CO 2 concentration, 400 µmol mol − 1 ; leaf temperature, 25 °C; photosynthetic photon flux density (PPFD), 1500 µmol m − 2 s − 1 ; and relative air humidity, 70%. The photosynthetic N use efficiency (P NUE ) was determined as the ratio of the P n to the leaf N content per area.

Analysis of N assimilation enzymes and N status in leaves and absorptive roots
At the end of the experiment, activity assays for N assimilation-related enzymes were conducted for the third fully expanded and intact leaf from five randomly chosen cuttings from each treatment. Fresh leaves (0.3 g) were ground to a fine powder in liquid nitrogen. The powder was extracted in a 2 mL icecold extraction buffer consisting of 50 mM Tris-HCl buffer (pH 7.5), 0.5 mM EDTA, 1 mM MgCl, 10 mM β-mercaptoethanol, and 0.5% PVPP (Polyvinylpolypyrrolidone). The extract was centrifuged at 12,000 g for 10 min at 4 °C as the crude enzyme extract.
Nitrate reductase (NR, EC 1.6.6.1) activity was measured according to the method described by Datta and Sharma (1999). NR activity was determined by monitoring the absorbance at 540 nm using a spectrophotometer (Unicam UV-330, Unicam, Cambridge, UK). The consumed NO 2 − was expressed as nmol per minute per g of fresh material.
For GS activity measurement, 0.1 mL of the crude enzyme extract was added to a 1.9 mL assay mixture, including 50 mM Tris-HCl buffer (pH 7.5), 13 mM hydroxylamine, 50 mM glutamic acid-Na, 1 mM ATP, 20 mM AsNa 3 O 4 , and 20 mM MgCl. The mixture was incubated for 30 min at 37 °C, and the reaction was stopped by adding 1 mL acidic FeCl 3 . The mixture was centrifuged at 5000 g for 15 min, and the absorbance of the supernatant was assayed at 540 nm. GS activity was expressed as nmol γ-glutamyl-hydroxamate (γ-GHM) produced per minute per g of fresh tissue.
For the measurement of AspAT activity, 100 µL of crude enzyme extract was added to 1.4 mL of an assay mixture consisting of 50 mM Tris-HCL buffer (pH 7.5), 5 mM EDTA, 0.2 mL of aspartic acid, 12 mM of α-ketoglutaric acid, 0.15 mM NADH, 0.5 M l-alanine, and 5 units of malic dehydrogenase. After the reaction started, the extinction value was detected in a 30 s interval at 340 nm using a spectrophotometer and was recorded continuously for 450 s. The slope of the scanning curve was used to calculate AspAT activity. AspAT activities were expressed as mmol NADH per minute per g of fresh tissue.
For the measurement of NAD-GDH activity, the crude enzyme extract (0.1 mL) was added to a 1.9 mL assay mixture containing 50 mM Tris-HCl buffer (pH 8.8), 80 mM L-glutamic acid, and 0.2 mM NAD + . The absorbance of the mixture was monitored at 340 nm for 450 s. The NAD-GDH activity was expressed as nmol NADH produced per minute per g of fresh leaves.
For NO 3 − -N determination, frozen dried leaf samples (0.2 g) were ground to a powder and extracted in 50 mL of 2 mol·L − 1 KCl for 30 min. After filtering, 2 mL of the solution was mixed with 1 mL of 1 mol·L − 1 HCl. The NO 3 − -N content was determined when monitored at 220 nm using a spectrophotometer. The values were quantified according to a standard curve. Dried samples (0.1 g) of leaves were used for the N and P content measurements. Nitrogen and P content were measured according to the Kjeldahl method and Mo-Sb colorimetric method after H 2 SO 4 digestion, respectively. Then, foliar N:P was determined.

C and 15 N analysis
At the end of the experiment, the carbon isotope composition ( 13 C/ 12 C), as expressed by δ 13 C, was determined for the dried leaves. The dried leaves were ground into fine powder in a ball mill and passed through a 100-mesh sieve. Stable carbon isotope abundance in the combusted samples was determined with a mass spectrometer (Thermo Fisher Scientific, Inc., USA). The overall precision of the δ-values was better than 0.1‰, as determined from repetitive samples. The entire analysis was performed in the Stable Isotope Laboratory for Ecological and Environmental Research (SILEER), CAS.
At the end of the experiment, five randomly chosen cuttings per treatment were supplied with either labeled 15 NO 3 − -N as NH 4 15 NO 3 or labeled 15 NH 4 + -N as 15 NH 4 NO 3 (30 mg per plant). Then, 72 h later, the third fully expanded and intact leaf from the upper position of the cuttings was harvested, dried in an oven at 75 °C for 48 h, and ground into fine powder for the 15 N isotope composition analysis. The 15 N/ 14 N ratios, as expressed by δ 15 N, were determined by an Isotope Ratio Mass Spectrometer (Thermo Fisher Scientific, Inc., USA). The overall precision of δ 15 N was better than 0.1‰, as determined by repetitive samples.

Relative competitive intensity
The relative competitive index (RCI) of P. cathayana and P. beijingensis under control and N deposition conditions was calculated according to the methods of Chen et al. (2017): RCI = (B inter − B intra ) / B intra , where B inter stands for the average dry matter accumulation of a kind of organ or total biomass of cuttings from interspecific competition, and B intra stands for the average dry matter accumulation of the corresponding organ or total biomass of cuttings from intraspecific competition. Each block, as a replicate, was used to calculate the RCI values.

Statistical analyses
All data were analyzed with the Statistical Package for the Social Sciences software (SPSS, Chicago, IL, USA) version 20.0. One-way analysis of variance (anova) was performed for all parameters, and post hoc comparisons were applied using Tukey's test at a significance level of α = 0.05. All data were tested for normality and homogeneity of variances and logtransformed to correct deviations from these assumptions when needed. A paired-sample t-test was used to evaluate the significance of differences between the N treatments (P < 0.05). Three-way ANOVA was employed to test the interactive effects of species, N deposition, and competition mode on morphological, physiological, and biochemical parameters. A Twoway ANOVA was used to test the interactive effect of species and N deposition on the RCI indexes. The effects were considered significant if P < 0.05.

Growth characteristics
Based on ANOVA, plant height, leaf area, SLA, and stem mass were significantly affected by species as independent factors (Table 1). All growth and morphological parameters were significantly affected by N application. Competition patterns significantly affected SLA, leaf mass, stem mass, and total biomass. In addition, both basal diameter and leaf area were significantly affected by the interactive effect of species × N and N × competition mode. Leaf mass, stem mass, and total biomass were significantly affected by the interactive effect of species × N.
The plant height and basal diameter of both species were significantly increased after N addition, regardless of any competition pattern, except that N application did not induce a significant change in basal diameter of P. cathayana under monoculture mode (Table 1). Under either N − or N + conditions, there were no significant differences between the species in plant height and basal diameter when they grew in either competition mode. However, N addition significantly increased leaf area of P. beijingensis in both competition modes, which did not occur in P. cathayana. Under N − conditions, there was no difference in both leaf area and SLA between the species from the interspecific competition mode, whereas P. beijingensis had significantly higher leaf area and SLA than P. cathayana in the intraspecific competition mode (Table 1). In contrast, under N + conditions, there was a significantly higher leaf area and SLA in P. beijingensis than in P. cathayana in both competition modes.
Nitrogen addition significantly increased leaf mass, stem mass, and total biomass of P. beijingensis in both competition modes, whereas N addition did not change such parameters of P. cathayana, except that an increase in leaf mass occurred in interspecific competition mode when exposed to N addition. Under N − conditions, P. cathayana had significantly higher stem mass and total biomass than P. beijingensis in both competition modes. However, under N + conditions, P. beijingensis showed a significantly higher leaf mass, stem mass, and total biomass than P. cathayana in the interspecific competition pattern, while there was no significant difference in such indexes between the species in the intraspecific competition mode.

Morphology and biomass of different root orders
Based on ANOVA, species, N, and competition mode as independent factors all significantly affected the total root length, total root surface area, and root biomass of both R1 and R2, as well as the SRA of R1 and R4 (Table 2). Both species and competition mode significantly affected the average diameter of individual root of R1 and R2, and only N as an individual factor significantly affected the average length of individual root of R1, R2, and R3. Competition mode as an independent factor exerted a significant effect on the total root length, average diameter of individual root, and biomass of R3. Both species and N significantly affected total root length and total root surface area of R4 and SRA of R2 and R5, while average diameter of individual root and biomass of R4 were significantly affected by species and competition mode, and by N deposition, respectively. N deposition as an individual factor significantly affected the total root length, total root surface area, and root biomass of R5, and competition mode significantly affected both the total root length and average diameter of individual root of R5. Nitrogen, as Table 1 The effects of N addition on growth, biomass accumulation and allocation of P. cathayana and P. beijingensis under intraand interspecific competition Each value is the mean ± SE (n = 5). Values followed by the different lowercase letters in the same column represent for significant differences between the treatments under N − conditions, while values followed by the different uppercase letters in the same column represent for significant differences between the treatments under N + conditions according to Tukey's test (P < 0.05). The asterisks designate significant differences induced by N application according to the paired-samples t-test (*, 0.01 < P < 0.05; **, P ≤ 0.01). SLA, specific leaf area. F S , species effect; F N , N effect; F C , competition mode effect; F S×N , the interactive effect of species and N; F S×C , the interactive effect of species and competition mode; F N×C , the interactive effect of N and competition mode; F S×N×C , the interactive effect of species, N and competition mode. ns, not significant; *, 0.01 < P < 0.05; **, P ≤ 0.01. C/CC, P. cathayana cuttings from intraspecific competition; B/BB, P. beijingensis cuttings from intraspecific competition; C/CB, P. cathayana cuttings from interspecific competition; B/CB, P. beijingensis cuttings from interspecific competition; C/CC-N, P. cathayana cuttings from intraspecific competition under N deposition; B/BB-N, P. beijingensis cuttings from intraspecific competition under N deposition; C/ CB-N, P. cathayana cuttings from interspecific competition under N deposition; B/CB-N, P. beijingensis cuttings from interspecific competition under N deposition. Significant P values (i.e., P < 0.05) are in bold. The same as below an independent factor, did not significantly affect the average diameter of individual root of all roots, and the average length of individual root of all roots was not significantly affected by species and competition mode as independent factors. Under all treatments, total root length, total root surface area, and SRA showed a decreased trend along with an increase in root order for both species (Fig. 1b, c, and f). In contrast, an opposite trend for root biomass, average length of individual root, and average diameter of individual root was observed with increasing root order (Fig. 1a, d, and e). Regardless of the competition mode, a significant increase in root biomass, total root length, and total root surface area of R1 and R2 was observed in P. beijingensis but not in P. cathayana when exposed to N addition (Fig. 1a, b, and c). Additionally, N addition caused an increase in root biomass, total root length, and total root surface area of both R3 and R4 for P. beijingensis under mixed culture mode but caused a decrease in average length of individual root of R1 for P. beijingensis under pure culture mode.
Under N − conditions, there was no significant difference in root morphological indexes or root biomass between P. cathayana and P. beijingensis under either pure cultivation or mixed culture, except that P. cathayana had a greater biomass and total root surface  area in R3 but lower average diameter of individual root in R2 than in P. beijingensis under the interspecific competition mode (Fig. 1a, c, and e). However, under N + conditions, P. cathayana in mixed culture mode had greater root biomass and total root length of R1 but lower average diameter of individual root of R2 than those in pure culture mode, whereas P. beijingensis in mixed culture mode showed greater biomass, total root length, and total root surface area of R2 than those under pure culture conditions (Fig. 1a, b, and c). Regardless of any competition mode, P. beijingensis exhibited larger root biomass, total root length, and total root surface area of both R1 and R2 than P. cathayana. In addition, in the mixed culture mode, P. beijingensis had a greater total root surface area of R3 and R4 and a bigger average diameter of individual root and SRA of both R1 and R2 than P. cathayana, whereas there were no significant differences in such parameters between the species under the pure culture mode.

Biomass allocation
Based on ANOVA, both R/S and the biomass ratio of absorptive to transport roots were significantly affected by species, N deposition, competition mode as single factors, and the interactive effect of species × N deposition (Fig. 2). For P. beijingensis, N addition decreased R/S but increased the biomass ratio of absorptive to transport roots in both competition modes (Fig. 2). In contrast, N addition did not significantly affect R/S of P. cathayana in both competition modes but decreased the biomass ratio of absorptive to transport roots of P. cathayana under monoculture. Under N − conditions, P. cathayana had a significantly lower R/S than P. beijingensis in both competition modes, whereas there was no significant difference in R/S between the species under N + conditions. However, P. beijingensis had a significantly higher biomass ratio of absorptive to transport roots than P. cathayana in both competition modes under N + conditions, while there was no significant difference in this parameter between the species under N − conditions.

Gas exchange rate and N and water use efficiency
Based on ANOVA, both P n and G s were significantly affected by species, N, and competition mode, as well as the interactions of species × N and species × competition mode (Fig. 3). P NUE was significantly affected by species, the interactive effects of species × competition mode and species × N × competition mode. δ 13 C was significantly affected by N and the interactive effects of species × N × competition mode.
Nitrogen supply induced a significant increase in P n and G s in both species under both competition patterns ( Fig. 3a and b), except for P n of P. cathayana in the interspecific competition mode, while N application did not significantly influence P NUE of either species in both competition patterns (Fig. 3c). Under N − conditions, compared with P. beijingensis, P. cathayana showed a significantly higher P n in both competition modes, while under N + conditions, P n of P. cathayana was greater than that of P. beijingensis in intraspecific competition mode but not in interspecific competition mode. Under N + conditions, P. beijingensis in the mixed culture mode exhibited a higher P n than those in the pure cultivation mode, whereas the opposite results were found under the N − condition. Similarly, P. cathayana showed a consistently higher G s than P. beijingensis in intraspecific competition mode under both N conditions, whereas in interspecific competition mode, P. beijingensis had a higher G s than P. cathayana under N + conditions. Under N − conditions, P. cathayana showed a significantly lower P NUE than P. beijingensis in the intraspecific competition mode, while there was no significant difference between the species under interspecific competition due to the decrease in P NUE in P. beijingensis when compared to the values under monoculture. Under N + conditions, P. beijingensis showed a slightly higher P NUE than P. cathayana under both competition modes. Nitrogen deposition induced an increased δ 13 C in the leaves of P. beijingensis under monoculture and in the leaves of P. cathayana under mixture (Fig. 3d). However, there was no significant difference in this parameter between the species under all treatments.

N assimilation
Based on ANOVA (Table 3), all parameters related to N-assimilated enzymes were significantly affected by species, N, and competition pattern, except for NAD-GDH activity. GS activity was significantly affected by the interactive effects of species AspAT activity was significantly affected by the interactive effects of species × N, species × competition mode, N × competition mode, and species × N × competition mode. The application of N induced increases in the activities of N assimilation enzymes in the leaves of both species to varying degrees (Table 3), except for GS activities. For example, there was a significant increase in GOGAT and AspAT activities in both species when N was supplied, regardless of the competition mode. In addition, when exposed to N addition, NR activities in the leaves of P. beijingensis increased significantly in the pure cultivation mode. With or without N supply, there were consistently higher enzyme activities of NR, GOGAT, AspAT, and NAD-GDH in the leaves of P. cathayana than those of P. beijingensis in both competition modes.

N and P statuses in leaves
Based on ANOVA (Table 3), species as an independent factor significantly affected NO 3 − concentration, N content, and N:P in leaves. Competition mode as an independent factor significantly affected both NO 3 − concentration and P content in leaves. Nitrogen deposition as an independent factor significantly affected all parameters. In addition, the NO 3 − concentration in leaves was significantly affected by the interactive effect of species × N × competition mode. The P content in leaves was significantly affected by the interactive effect of species × competition mode.
Nitrogen deposition increased the NO 3 − concentration and N content in the leaves of P. beijingensis in both competition modes, whereas N addition did not significantly affect both parameters in the leaves of P. cathayana in either competition mode (Table 3), except that an increase in N content in leaves of P. cathayana occurred in intraspecific competition mode when exposed to N addition. Under N + conditions, NO 3 − concentration in the leaves of P. beijingensis was greater than those in the leaves of P. cathayana under monoculture, whereas there was no significant difference in this parameter between the species under mixture. When compared to the values under monoculture, the P content in the leaves of P. beijingensis under mixture increased significantly under N − conditions. There was no significant difference in either P content or N:P between the species under the corresponding conditions.

N isotope concentrations in leaves
Based on ANOVA, both δ 15 NO 3 − -N and δ 15 NH 4 + -N were significantly affected by species, N, and competition mode as independent factors and the interactive effects of N × competition mode and species × N × competition mode (Fig. 4). For both species, N addition significantly increased both δ 15 NO 3 − -N and δ 15 NH 4 + -N under either of the competition modes ( Fig. 4a and b), except for δ 15 NH 4 + -N in leaves of P. beijingensis from the mixture. Under N − conditions, when compared to P. cathayana, P. beijingensis had significantly higher δ 15 NO 3 − -N in monoculture conditions, and higher δ 15 NH 4 + -N in mixed culture conditions. In contrast, under N addition, P. beijingensis had significantly significant higher δ 15 NH 4 + -N in monoculture conditions and higher δ 15 NO 3 − -N than P. cathayana in mixture conditions.

Relative competition intensity
As shown in Fig. 5, based on ANOVA, both RCIstem and RCI-root were significantly affected by species as an independent factor, while both RCIleaf and RCI-total were significantly affected by N deposition as an independent factor. In addition, all RCI values were significantly affected by the interaction of species × N deposition.
Under N − conditions, the RCI of all organs in both species were negative, except for a positive Fig. 1 The effects of N addition on biomass and morphological traits of different root orders in P. cathayana and P. beijingensis under intra-(monoculture) and interspecific competition (mixture). a root biomass; b total root length; c total root surface area; d average length of individual root; e average diameter of individual root; and f SRA, specific root surface area. Different lowercase letters above the bars represent for significant differences between the treatments under N − conditions, while different uppercase letters above the bars represent for significant differences between the treatments under N + conditions according to Tukey's test (P < 0.05). The asterisks designate significant differences induced by N addition according to the paired-samples t-test (*, 0.01 < P < 0.05; **, P ≤ 0.01). Values are given as mean ± SE (n = 5) ◂ value detected in the roots of P. cathayana (Fig. 5). There were no significant differences in both RCIleaf and RCI-stem between the species, whereas P. cathayana had greater values in both RCI-root and RCI-total than P. beijingensis. However, under N deposition conditions, for both species, the RCIs of aboveground organs (RCI-leaf and RCI-stem) and total biomass (RCI-total) were still negative, but the RCI roots of both species were positive. Populus beijingensis had greater RCI values of all organs than P. cathayana, except for RCI-root. Thus, the negative effects exerted on P. cathayana were more apparent when compared to P. beijingensis under N + conditions.

Discussion
In our study, we found that the effects of competitive interaction on the species-specific growth of poplars depend on N regimes. Under N − conditions, P. cathayana grown under either of the competition modes always showed a better performance in growth than its counterpart, as seen in the consistently higher leaf mass, stem mass, total biomass, and P n , suggesting that P. cathayana has a relatively lower N demand. However, P. beijingensis under the mixture had greater growth responses to N deposition, with a significant increase in plant height, basal diameter, leaf area, leaf mass, stem mass, root mass, total biomass, and R/S, whereas the corresponding P. cathayana was comparatively insensitive to N deposition, especially dry mass accumulation. Therefore, P. beijingensis showed competitive superiority over P. cathayana under N deposition, as seen in the higher values in leaf area, SLA, leaf mass, stem mass, and total biomass. Occupying more aerial volume as a result of the higher biomass and leaf area, as well as greater height growth may endow P. beijingensis with asymmetrical advantage over P. cathayana because of the high shade intolerance. Although the photosynthetic rate per unit leaf area in P. beijingensis is relatively lower than that in P. cathayana, the higher leaf area, SLA, and P NUE can confer greater capability of light interception and light use to P. beijingensis, which could not only compensate for its inferiority in carbon assimilation in the unit leaf area but also enhance competitiveness by shading neighbors. Such results imply that it could be important to identify functional traits at both leaf and whole plant scales to fully characterize the competitive ability of plants.
Previous studies have indicated that most neighbor effects are mainly belowground at low N availability (Wilson and Tilman 1991). Some studies have reported that interspecific competition can promote root growth of the stronger competitors while inhibiting root growth of the counterparts under resource deficiency (Lei et al. 2012;Hommel et al. 2016), suggesting that soil resource deficiency can intensify belowground competition. In our study, Fig. 2 The effects of N addition on the root/shoot ratio (R/S) (a) and the biomass ratio of absorptive to transport roots (b) of P. cathayana and P. beijingensis under intra-(monoculture) and interspecific competition (mixture). Significant P values (i.e., P < 0.05) are in bold. The same as below investing more in belowground growth, as shown by the significantly increased R/S, can be considered a strategy adopted by P. cathayana under heterospecific competition, which enables P. cathayana to capture more nutrients from the soil. The positive RCI-root also suggested that the growth of roots in P. cathayana benefited from the presence of P. beijingensis, which may help P. cathayana to be a stronger competitor in belowground competition at low N availability (Kołodziejek 2019). In contrast, such facilitation effects on root growth were not detected in P. beijingensis when growing with distinct neighbors, indicating that P. beijingensis is less responsive to belowground competition at low N availability. However, it is well known that biomass and productivity increase with the increasing availability of soil nutrients (especially N), causing competition to shift from belowground resources to aboveground resources (especially light) (Herbert et al. 2004). Plants generally decrease investment in belowground organs when supplied with high levels of soil nutrients, and the resulting decrease in R/S can enhance the competitive capacity of plants by reducing root respiratory carbon loss without influencing soil resource uptake and increasing carbon and resource allocation to aboveground organs (Liao et al. 2013). In our study, in response to N deposition, P. beijingensis decreased investment in roots, as shown by the significantly decreased R/S, whereas R/S of P. cathayana did not change Fig. 3 The effects of N addition on net photosynthetic rate (P n ) (a), stomatal conductance (G s ) (b), photosynthetic nitrogen use efficiency (P NUE ) (c), foliar carbon isotope composi-tion (δ 13 C) (d) of P. cathayana and P. beijingensis under intra-(monoculture) and interspecific competition (mixture) Vol:. (1234567890) Table 3 The effects of N addition on activities of N assimilation-related enzymes, and N and P status in P. cathayana and P. beijingensis under intra-and interspecific competition Each value is the mean ± SE (n = 5). NR, nitrate reductase; GS, glutamine synthetase; GOGAT, glutamate synthetase; AspAT, aspartate aminotransferase; NAD-GDH,  significantly. Such higher plasticity in biomass allocation may help P. beijingensis gain a competitive advantage over P. cathayana under N deposition. Nutrient availability has been found to affect the competitive ability of plants and consequent species composition in communities because of interspecific differences in N uptake and use efficiency (Song et al. 2017). In our study, P NUE was mainly determined by species and indirectly affected by N availability and competition mode. Under N − conditions, compared to the values under intraspecific competition, interspecific competition increased P NUE of P. cathayana but decreased P NUE of P. beijingensis, which could benefit the competitive performance of P. cathayana. However, we detected species-specific physiological reactions in N assimilation in response to Relative competitive intensity of P. cathayana and P. beijingensis exposed to interspecific competition and N deposition. Values are given as mean ± SE (n = 3). RCI-leaf, relative competitive intensity calculated from responses in leaf mass. RCI-stem, relative competitive intensity calculated from responses in stem mass. RCI-root, relative competitive intensity calculated from responses in root mass. RCI-total, relative competitive intensity calculated from responses in total biomass N deposition and competition mode, which may partly contribute to the competitive ability of poplar. In accordance with responses in the activities of N assimilation-related enzymes to N addition (Li et al. 2015), we found that N deposition increased the activities of all N-related enzymes in both P. beijingensis and P. cathayana to varying degrees. Interestingly, P. cathayana exhibited consistently higher activity of these enzymes than P. beijingensis in both competition modes, regardless of N availability. It is reasonable to suppose that P. cathayana has a higher constitutive expression of specific genes coding for the N assimilating pathway (Tischner 2000), conferring an inherently greater capacity for N assimilation under low N availability. Thus, under N − conditions, the higher ability in N assimilation can facilitate P. cathayana to acquire more inorganic N under interspecific competition, which can be reinforced by the fact that the N content in R1 and R2 increased in P. cathayana, not in P. beijingensis (data not shown), under mixture when compared to the values under monoculture. Nevertheless, under N deposition, mixture cultivation induced greater increases in both GS and AspAT activities in P. beijingensis than in P. cathayana when compared to monoculture, which might be responsible for the decrease in NO 3 − -N but a slight increase in N content in the leaves of P. beijingensis. In view of greater leaf area and leaf mass, these results indicate that P. beijingensis from the mixture assimilates more inorganic N in leaves than P. cathayana under N deposition. In addition, we found that interspecific competition increased the N content of both R1 and R2 in P. beijingensis, but not in those of P. cathayana under N deposition, when compared to individuals subjected to intraspecific competition (data not shown). Therefore, these results collectively suggest that P. beijingensis grown in mixture conditions has a higher capability of utilizing the increasing N in soils than its counterpart.
Populus cathayana and P. beijingensis are typical deciduous species distributed in the temperate zone of China, where forest ecosystems are considered to be N-limited (Wu 2011). For example, in natural conditions, previous studies indicated that growth of mature P. cathayana is N-limited in growing season (Liu and Wang 2021). Compared to soil resource availability in a typical plantation of P. cathayana in Datong (Lin et al. 2021), N availability in soils of the present study is relatively low, while the P supply is sufficient. In our study, N and P contents in leaves of poplar saplings are comparative to those reported by previous studies (Chen et al. 2015;Liu and Wang 2021). Short-term N deposition as an independent factor significantly affected both N and P content, suggesting that N deposition not only has affected N uptake and assimilation in plants but also potentially affected P mineralization and cycle. The P content in the leaves also significantly affected by competition mode as an independent factor, which can be explained by the fact that individual species under interspecific competition may exploit different P fractions using their specific P-acquisition mechanisms (Güsewell 2004), while plants under monoculture with the same P preference would compete for the same P forms. In addition, we found that shortterm N deposition may mitigate N limitation in both poplars, as shown by the slight increase in N:P ratio in poplar leaves, especially in P. beijingensis leaves. However, the effects of N deposition on N or P limitation in poplars should be further studied in future by combining analysis in N and P dynamics in soils.
A growing number of studies suggest that morphological and architectural traits of roots, which are shaped by myriad environmental signals (Morris et al. 2017), play a major role in mediating plant-plant interactions and resource capture from soils (Colom and Baucom 2020), and can be used to predict the ability of plants to tolerate their competitors (Semchenko et al. 2018). In our study, when exposed to N − conditions, there were rarely species-specific divergences in biomass and morphology at the root rank scale, especially for absorptive roots (i.e., R1 and R2), under both competition modes. Such results confirm that species-specific differences in N assimilation rate and N use efficiency, rather than N uptake capacity, may contribute to P. cathayana outcompeting P. beijingensis under N − conditions. However, under either of the competition modes, P. beijingensis exhibited higher plasticity in root architecture than P. cathayana, as shown by increases in root biomass, total root length, and total root surface area of absorptive fine roots (i.e., R1 and R2) in response to N deposition. Combined with the higher average diameter of individual root and SRA in both R1 and R2, it can be inferred that P. beijingensis from the mixture can take up more space and explore soil volume by altering root architecture, and thereafter increase its belowground competitive ability (Lei et al. 2012). It is noted that lower ranks of roots, especially R1 and R2, were more responsive to N deposition and the competition mode than higher ranks of roots, which agrees with the findings of Salahuddin et al. (2018). In response to N deposition, the ratio of absorptive to transport roots increased in P. beijingensis in both competition modes, not in P. cathayana in either competition mode, which can promote nutrient absorption capacity of P. beijingensis. Taken together, our results suggested that P. beijingensis under N deposition is more flexible in the morphological and architectural adjustment of root system to meet nutrient requirements even if biomass allocation to roots decreased.
Our results revealed that N deposition affects carbon gain and long-term water use efficiency by regulating stomatal behavior. Under N − conditions, P. beijingensis from the mixture showed a lower P n but a higher G s than those from monoculture, suggesting that suppression in photosynthetic capacity in P. beijingensis may result from a decline in carboxylation capacity rather than stomatal limitation. Previous studies on the effects of N deposition on water use efficiency have revealed variable results; for example, N deposition can lead to increased (Lu et al. 2019), decreased (Huang et al. 2016), or invariant foliar iWUE (Diao et al. 2021). Such variable results can be explained by the fact that the effects of N on water use efficiency may depend on the extent and duration of N addition, as well as on the sensitivity of plants to N and the background N availability in soil. In our study, N deposition induced an increased δ 13 C in leaves of both species, regardless of competition mode, accompanied by increases in P n and G s , which agrees with the findings of Guo et al. (2016) and Song et al. (2017). These results indicate that the increased carboxylation rate, rather than stomatal limitation, contributed to the increased long-term water use efficiency in response to N deposition, as shown by the increased δ 13 C. In addition, we noted that N deposition resulted in an increased δ 13 C value for the conspecific P. beijingensis and for the heterospecific P. cahtayana when compared with individuals under N − conditions, possibly because P. beijingensis, with a relatively larger transpiring area, depleted soil moisture rapidly and forced their neighbors to promote water use efficiency. Consistently, under N deposition, we observed that competition with P. beijingensis induced stomatal closure of P. cathayana, which might be responsible for the increased water use efficiency of P. cathayana. Increased water use efficiency was not at the expense of carbon gain, suggesting that partial stomatal closure did not affect CO 2 supply and mesophyll diffusion of CO 2 , which is in line with the observations of Broeckx et al. (2014).
Previous studies have demonstrated that the preference for N resources in plants depends on species, plant sex, and soil nature, such as pH (Li et al. 2015). Some poplars have been proven to prefer N-NO 3 − to N-NH 4 + because N-NO 3 − can be highly efficient for uptake and assimilation (Min et al. 1998). In our study, we observed that both poplars prefer N-NO 3 − over N-NH 4 + because both species showed higher content of N-NO 3 − , which is in line with the findings of Chen et al. (2014a). More interestingly, we found that competition mode exerted an impact on poplars' preference for N forms, as reflected by the changing δ 15 N-NO 3 − and δ 15 N-NH 4 + in poplar leaves. Overall, under N − conditions, when compared to monoculture, interspecific competition increased δ 15 N-NO 3 − in P. cathayana leaves but decreased the δ 15 N-NO 3 − content in P. beijingensis leaves. The concomitant increase in δ 15 N-NH 4 + indicates that P. beijingensis may absorb more N-NH 4 + , whereas P. cathayana showed a preference for N-NO 3 − . However, under N + conditions, the opposite pattern of N preference between the poplars was detected. In view of the findings that poplars with better adaptation to N-NO 3 − are prone to achieve a competitive advantage (Chen et al. 2014a), our results suggest that selectively absorbing more N-NO 3 − is a possible cause of the competitive advantage of P. cathayana under N − conditions and P. beijingensis under N deposition.
High niche overlap could lead to fierce competition between vigorously introduced hybrids and closely related native species that occupy the sympatric range (Eckhart et al. 2017), thereby exerting adverse effects on species' coexistence. Novel recombinant genotypes with wide niche breadth have shown the potential to outcompete and displace the parent species in the introduced habitats (McCartney et al. 2019), and ultimately decrease biodiversity (Thum and Lennon 2010). In our study, under both N conditions, we detected an adverse competition relationship between the species, as shown by the negative RCI values, suggesting that both species are sensitive to distinct neighbors and that interspecific competition may decrease their fitness due to growth inhibition. Under N − conditions, growing with P. cathayana exerted more negative effects on P. beijingensis, as indicated by the more negative values in the RCI of all organs, whereas the root growth of P. cathayana was promoted in the presence of P. beijingensis, which could place P. cathayana at an advantage. Nevertheless, under N deposition, P. cathayana suffered more negative effects from the distinct neighbor, and aboveground growth was obviously inhibited, as reflected by the negative leaf and stem RCI values, while P. beijingensis was slightly affected in the presence of P. cathayana, suggesting that N deposition altered the competitive relationships between the poplars.
In conclusion, our study indicates that the competition mode could affect species-specific differences between P. cathayana and P. beijingensis in growth traits, biomass allocation, root architecture, photosynthetic rate, N-related metabolism, long-term water use efficiency, and preference for N forms. The competitive interactions between the poplars are regulated by N deposition. Under N − conditions, native P. cathayana is a competitively superior species, while the exotic P. beijingensis has shown the potential to outcompete P. cathayana when exposed to N deposition, illustrated by its more positive responses in growth, better light capture ability, and higher flexibility in fine root architecture, as well as a shift for N-NO 3 − preference compared with P. cathayana. Therefore, it is expected that native species might be replaced at an accelerated rate by species with a suite of functional traits beneficial for using excessive N in soils, thus showing a more plastic reaction to growth under the background of growing N deposition. Based on the above-mentioned results in our controlled experiment, fast-growing exotic species with a higher N requirement and N use efficiency should be carefully assessed before introducing them to ecosystems with a heavy N load. Nitrogen management is essential for the areas where such species have been introduced to maintain biodiversity. In addition, long-term observation of the interactive relationships between introduced species and native species should be performed to estimate the competitive outcomes, and more field research about the responses of mature trees to long-term N deposition should be conducted to estimate the ecological consequences of introduced tree species on native relatives because greenhouse studies cannot simulate complex environmental changes in natural conditions.