*4.3. Relationships between the* δ*13C,* δ*15N Values and Leaf Nutrients*

Multiple studies have reported various correlations between the δ13C and leaf nutrients [20,21,38]. In the present work, the positive relationship between δ13C and N over all aged leaves together (Table 2) is in accordance with most previous studies [20,39]. Moreover, our conclusions suggest that the relative contribution of leaf N concentrations on δ13C was significant (*p* < 0.01). The main cause of the positive relationships was that photosynthetic capacity increased with leaf N concentrations [20], and there was a positive correlation between leaf δ13C and photosynthetic capacity [4]. However, other studies have found a negative correlation between leaf δ13C and leaf N concentrations, and this was likely attributed to the presence of nitrogen-fixing species in samples such as *Caragana microphylla* [5] or an autocorrelation with water availability in a semiarid environment. In addition, in high altitude areas, low atmospheric pressure and temperature could alter the expression of the relationship between N and photosynthetic and thus, the δ13C-N relation [38].

Research about the relationship between leaf δ13C and P concentrations is relatively limited, and the results have been inconsistent. Some studies have demonstrated the positive relationship between leaf δ13C and P owing to the effect of leaf P concentrations on photosynthetic via Rubisco, while other works have found leaf δ13C to be negatively related with leaf P concentration. Our study observed that leaf δ13C was positively related to all aged leaves' P concentrations in simple regression (Table 2), but the effect of leaf P concentration on δ13C was not significant in multiple regression (Figure 4A). This indicates that the variations in the leaf δ13C values were likely caused by stomatal limitation rather than P-related changes in photosynthetic efficiency [18,20]. In addition, our findings of the negative correlation between leaf δ13C and the C:N ratios is consistent with the results from multiple previous studies and suggests that *P. crassifolia* may achieve higher water use efficiency (WUE) at the expense of decreased nitrogen use efficiency (NUE) [20,38].

The uptake and discrimination of 15N are also significantly related to plant N demand and assimilation capacity [1,40]. N availability in ecosystem, N re-translocation in plants, and N fractionation after plant uptake is known to influence leaf δ15N [1,41]. Multiple previous studies have reported that there is a positive relationship between plant δ15N and leaf N concentrations at various spatial scales [17]. In our work, leaf δ15N was also positively related to leaf N concentrations. Furthermore, the result of Figure 4B suggested that leaf nitrogen concentrations play an importance role in accounting

for the variations of leaf δ15N. Changes in environmental nitrogen demand or supply could influence whole plant and organ level nitrogen isotope discrimination [12,34]. Likewise, leaf N concentrations of current-year leaves were significantly higher than other old leaves in our study area (*p* < 0.001, Figure S1). If the nitrogen supply of current-year leaves increased, discrimination could be expected to increase. However, with important questions still remaining about the relationship of leaf N and leaf δ15N, more comparative data are need to evaluate the potential drivers of leaf δ15N with increasing leaf N concentrations in the future [11].

#### **5. Conclusions**

In summary, the carbon and nitrogen assimilation in *P. crassifolia* leaves resulted in the same gradient of stable isotope compositions: young *P*. *crassifolia* leaves were more enriched in 13C and 15N compared with the older leaves at each tree age level. No significant difference in δ13C values among different tree ages was observed at each leaf age level, while the δ15N values of middle-aged (51–100 years old) were significantly more enriched than juvenile trees (<50 years old) at each leaf age level except for 1-year-old leaves. Based on the relative importance analysis, we identified that leaf age compared to tree age plays a dominant role in variation in leaf δ13C and δ15N values. Leaf nutrients such as leaf nitrogen concentrations are also important determinant factors for leaf δ13C and δ15N. However, our knowledge on the mechanism and effects of these biotic and abiotic factors on leaf δ13C and δ15N values at large scales are still limited. Further investigation is necessary to consider combinations of different drivers and their relative importance on the δ13C and δ15N values.

**Supplementary Materials:** The following are available online at http://www.mdpi.com/1999-4907/10/4/310/s1, Figure S1: Differences in leaf N concentrations (A) and leaf P concentrations (B) between four years' leaves (from current-year-old and up to 3-year-old) collected along different tree ages levels. Different uppercase letters represent significant differences among leaf ages at each tree age level, while different lowercase letters represent significant differences among tree ages at each leaf age level.

**Author Contributions:** Writing—original draft preparation, writing—review and editing, data curation, formal analysis and validation, C.L.; methodology, software, visualization and supervision, B.W.; conceptualization, project administration and funding acquisition, T.C.; investigation, resources and visualization, G.X.; investigation and software, M.W.; resources, G.W.; investigation, J.W.

**Funding:** This research was funded by National Natural Science Foundation of China, grant numbers 31670475, 41421061.

**Acknowledgments:** We appreciate three anonymous reviewers and editors for their helpful comments to improve the manuscript. We thank Gaosen Zhang for helping us with leaves sampling. We also thank American Journal Experts help us to improve the language.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


© 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).
