*2.3. Chemical Analysis*

The samples were analyzed using a sequential extraction technique according to Hilli et al. [34] and Preston et al. [14]. The samples were sequentially treated with trichloromethane to extract organic-soluble extractives (OSE), followed by boiling water to extract water-soluble extractives (WSE) and finally 72% (*v*/*v*) sulfuric acid to extract acidsoluble extractives (ASE), and the remaining residues were considered acid-unhydrolyzable residues (AUR). We performed 3 laboratory replicates for each sample during the experiment.

During solid-state 13C NMR measurement, all samples were treated with hydrofluoric acid (HF) to remove paramagnetic materials such as iron oxide from samples as follows: 0.5 g subsamples were weighed into 50-mL falcon tubes, and the tubes were tightly capped and horizontally shaken after adding 10 mL of 46% HF. Then, 40 mL of deionized water was added to the tubes, and the tubes were centrifuged at 3000 rpm for 10 min. The supernatant was removed and discarded, and the residue was washed 3–5 times with

deionized water to completely remove HF, and then dried at 60 ◦C in an oven. The HFtreated samples were used for solid-state 13C NMR analysis [19]. We did not perform laboratory replicate analyses of the samples. Generally, the measurements were better than the chemical extractions, and the replicates at each site were adequate to ensure the reliability of the data.

Solid-state 13C NMR spectra were obtained on a Bruker AVANCE III 400 spectrometer (Bruker BioSpin AG, Fällanden, Switzerland) at 100 MHz. The dried and finely powdered samples were packed in a zirconium dioxide (ZrO2) rotor topped with a Kel-F cap, which was spun at a 5 kHz rate. The contact time for the ramp sequence of the 1H ramp was 2 ms. A total of 2048 scans were recorded with a1srecycle delay for each sample, and the plotted spectral regions ranged from 0–220 ppm.

The contributions of various functional C groups to the total organic C (TOC) content were determined via the integration of their 13C signal intensity in the respective chemical shifts using MestreNova-9.0.1 (Mestrelab Research S.L., Santiago de Compostela, Spain). Functional C groups were assigned to aliphatic C (0–50 ppm), methoxyl C (50–60 ppm), O-alkyl C (60–95 ppm), di-O-alkyl C (95–110 ppm), aromatic C (110–140 ppm), phenolic C (140–165 ppm), and carboxyl C (165–220 ppm) [17,35]. Several hydrophobicity indices, including the ratio of aliphatic C to O-alkyl C (A/O-A) and the ratio of hydropholic-C to hydrophilic-C (HB/HI), i.e., (aliphatic C + functional C groups of the 110–165 ppm peak area)/(functional C groups of the 45–110 ppm peak area + functional C groups of the 165–190 ppm peak area), were used to reflect the stability of organic C. The higher these two ratios are, the higher the stability of organic C [20,36,37].

Total organic C, N and phosphorus (P) concentrations were determined via the dichromate oxidation, Kjeldahl determination (KDN, Top Ltd., Zhejiang, China) and phosphomolybdenum yellow spectrophotometry (TU-1901, Puxi Ltd., Beijing, China) methods, respectively [38]. Litter DOC was extracted by shaking 0.5 g of dried sample with 50 mL of deionized water for 30 min at room temperature, and the suspension was filtered through a 0.45-μm filter membrane. A total organic C analyzer (multi N/C 2100, Analytik Jena, Thüringen, Germany) was used to determine the litter DOC concentration [39]. The concentration of Mn via inductively coupled plasma-mass spectroscopy (ICP-MS, IRIS Advantage 1000, Thermo Elemental, Waltham, MA, USA) after litter materials (1.00 g) were digested in HNO3-HClO<sup>4</sup> (5:1, *v*/*v*) at 160 ◦C for 5 h [40].
