*2.2. Experimental Design*

The experiment began in November 2011, and four treatments (control: CK, N addition, P addition, and N and P addition together) were established. These four treatments were arranged in a random complete block design, with a total of four blocks. There were 16 of 20 × 20 m plots. To prevent influence between the plots, the spacing between plots was greater than 20 m. Nitrogen and P were applied first in December 2011 and then four times each year (March, June, September, and December). Nitrogen (100 kg ha−<sup>1</sup> year<sup>−</sup>1) was added in the form of NH4NO3, and P (50 kg ha−<sup>1</sup> year<sup>−</sup>1) was added in the form of NaH2PO4 [43]. To more evenly distribute N and P addition plots in the study site, we added sand (8 kg plot−1) in the N plots, phosphate plots, and control (CK) plots. For fertilization we chose a date where the forecast predicted no rain for four consecutive days [27].
