*2.1. Site Description*

The study site is in the Heilongjiang Liangshui National Nature Reserve (47◦1050" N, 128◦5320" E) in Dailing, Heilongjiang Province, Northeast China. The annual mean temperature is −0.3 ◦C, and the annual mean maximum temperature is 7.5 ◦C. The annual mean minimum temperature is −6.6 ◦C. The ≥ 0 ◦C accumulated temperature is 2200 ◦C–2600 ◦C. The annual mean precipitation is 676 mm, and >90% of the precipitation falls during the growing season from May to October. There are 100–120 frost-free days and 130–150 snow-cover days. The average elevation is 400 m, and the difference in the relative elevation is 100–200 m. The zonal soil is a dark brown forest soil. The mixed broadleaved-Korean pine (*Pinus koraiensis*) forest is the zonal vegetation of the reserve. The stand age is between 200 and 300 years, with some individual ages older than 400 years.

### *2.2. Experimental Design and Measurements*

The N addition experiment began during May 2008 in the mixed broadleaved-Korean pine (*Pinus koraiensis*) forest, an old-growth temperate forest. Twelve 20-m × 20-m sequential plots were established in a random manner. To prevent mutual interference, 10-m barriers between adjacent plots were retained. In every plot, species with a diameter at breast height (DBH, 1.3 m) >2 cm were surveyed; species, DBH, and coordinates were recorded and labeled in May 2008 and surveyed again in July 2014. The background conditions of each plot surveyed in 2008 are shown in Table 1. Using urea (CO(NH2)2) as the N source, four treatment groups were established, with three replicates—control (N0, 0 kg·ha−1·yr<sup>−</sup>1), low N (N1, 30 kg·ha−1·yr<sup>−</sup>1), medium N (N2, 60 kg·ha−1·yr<sup>−</sup>1), and high N (N3, 120 kg·ha−1·yr<sup>−</sup>1). The background N deposition (inorganic and organic N) was 12.9 kg·N·ha−1·yr<sup>−</sup>1, as recorded by a dry–wet deposition collector (New Star Environmental LLC, Roswell, GA, USA). N additions were applied at the beginning of June, July, and August every year and began in June 2008. The CO(NH2)2 for each treatment was dissolved in 20 L of water, and the solutions were sprayed onto the soil surface using a backpack sprayer. The control group was sprayed with the same amount of water.


**Table 1.** Natural conditions in each plot before the N addition experiment in 2008.

When collecting samples from tall trees, a branch with many leaves and sufficient sunlight was cut by a person specialized in climbing trees. The healthy and green leaves were selected and placed into a plastic vacuum packaging bag. Seventeen species were collected randomly in 3 plots for each treatment in the middle of May, July, and September during 2014. However, insufficient material was collected for a proportion of the species because of natural and anthropogenic reasons; for example, most of the *Fraxinus mandshurica* individuals had not renewed their foliage in May, and their leaves had already fallen by September, or a few trees were so high that samples could not be obtained. In this study, May, July, and September represented spring, summer, and autumn, respectively. The 17 species were of three types— deciduous broad-leaved trees, deciduous broad-leaved shrubs and evergreen conifers (hereafter trees, shrubs and conifers, respectively). Conifers were assigned their own group because they have unique traits compared to those of other trees [7]. The mycorrhizal types of species in this study were according to Shi et al. [37], and species belonging to the same family were classified into the same mycorrhizal type. The detailed sampling information is listed in Table 2.

Soil samples were also collected at the same time as leaf samples. In each plot, soils were randomly sampled by taking three 5-cm-diameter soil cores from depths of 0–10, 10–20, 20–30, and 30–40 cm. The litter on the soil surface was removed before sampling. The samples were placed in plastic bags and transported to the lab after being air-dried.


**Table 2.** Sampling condition of each species in 2014.

Each leaf sample was ground to a fine powder, passed through a 0.149-mm mesh, and enveloped. The sample was oven-dried at 65 ◦C to a constant weight for further analysis. The soil sample was passed through a 2-mm mesh; a portion was conserved in an envelope, and the remaining material was conserved after passing through a 0.149-mm mesh. In October 2014, leaf N concentration was analyzed using a Kjeltec KTM 2300 analyzer (Foss Teactor AB, Hoganas, Sweden) after H2SO4 + K2SO4 + CuSO4 digestion. Soil total N (TN) in every layer was analyzed using the same method but with H2SO4 + H2O2 digestion. Leaf and soil total P (TP) were measured using the molybdenum antimony colorimetric spectrophotometer method after H2SO4 + H2O2 digestion. These samples were oven-dried again at 65 ◦C for 2 h before digestion. For soil available P (AP), 2-mm of soil from every layer was pooled and fully mixed, and the air-dried soil was extracted using 0.05 mol·L−<sup>1</sup> HCl + 0.025 mol·L−<sup>1</sup> H2SO4. The P concentration in the extract was measured using the molybdenum antimony colorimetric spectrophotometer method. N and P concentrations are expressed as mg·g-1, and a mass basis was used to calculate stoichiometric ratios.
