Differential Responses of Soil Nitrogen Forms to Climate Warming in Castanopsis hystrix and Quercus aliena Forests of China

Abstract

Climate warming impacts soil nitrogen cycling in forest ecosystems, thus influencing their productivity, but this has not yet been sufficiently studied. Experiments commenced in January 2012 in a subtropical Castanopsis hystrix Hook. f. and Thomson ex A. DC. plantation and in May 2011 in a temperate Quercus aliena Blume forest, China. Four treatments were established comprising trenching, artificial warming (up to 2 °C), artificial warming + trenching, and untreated control plots. The plots were 2 × 3 m in size. In 2021 and 2022, soil nitrogen mineralization, soil nutrient availability, fine root biomass and microbial biomass were measured at 0–20 cm soil depth in 6 replicate plots per treatment. Warming significantly increased soil temperature in both forests. In the C. hystrix plantation, warming significantly increased available phosphorus (AP) and fine root biomass (FRB), but it did not affect soil microbial biomass carbon (MBC), microbial biomass nitrogen (MBN), microbial biomass phosphorus (MBP) and their ratios. Warming depressed the net mineralization rate (NMR) and net ammonification rate (NAR) of the C. hystrix plantation, probably because the competition for nitrogen uptake by fine roots and microorganisms increased, thus decreasing substrates for nitrogen mineralization and ammonification processes. Trenching and warming + trenching increased the net nitrification rate (NNR), which might be related to decreased NH₄⁺-N absorption of trees in the trenched plots and the increased microbial activity involved in soil nitrification. In the Q. aliena forest, warming significantly increased NH₄⁺-N, MBC/MBN, Root C/N, Root N/P, and decreased pH, MBN, MBN/MBP and Root P; and there was no effect of trenching. Notably, the NAR, NNR and NMR were largely unaffected by long-term warming. We attributed this to the negative effect of increasing NH₄⁺-N and decreasing MBN/MBP offsetting the positive effect of soil warming. This study highlights the vulnerability of subtropical forest stands to long-term warming due to decreased soil N mineralization and increased NO₃⁻-N leaching. In contrast, the soil N cycle in the temperate forest was more resilient to a decade of continuous warming.

1. Introduction

According to the World Meteorological Organization, the earth’s temperature in 2023 was about 1.4 °C higher than in the early industrialization period of 1850–1900 [1]. However, there are significant regional differences, and ecosystems at middle and high latitudes and elevations are likely to encounter more serious warming [2]. It is projected that the global temperature will further increase by 1.0–3.7 °C by the end of the 21st century, and pose a significant threat to global sustainable development and human livelihood [2,3]. Forests play a critical role in sequestering carbon, maintaining biodiversity and providing other ecosystem services. Nitrogen is a key driver of forest growth and therefore nitrogen provides a useful index to evaluate forest ecosystem response to global warming [4,5]. Soil nitrogen pool size and flux, nitrogen utilization by plants and N₂O emissions in terrestrial ecosystems are influenced by soil nitrogen mineralization, which is highly sensitive to warming [6].

Many warming control experiments have been carried out in forest ecosystems [7], and these have increased our understanding of the response of forests to climate warming. In particular, soil nitrogen mineralization is mainly influenced by soil organic matter, nitrogen availability [8], as well as soil temperature and humidity [9]. Increasing soil temperature changes nitrogen conversion by soil microbes and roots, and accelerates nitrogen mineralization by promoting the activity of extracellular enzymes involved in nitrogen circulation [10,11]. However, forest ecosystems differ greatly in species composition, community structure and soil types. In addition, tree species may differ in root morphology, physiology and nutrient requirements [12]. As a result, the response modes and feedback mechanisms to warming may differ across forest ecosystems. For example, in a mixed deciduous temperate forest, the annual average soil net nitrogen mineralization (0–10 cm) increased by 45% in a 5 °C ambient warming experiment [13]. These authors proposed that the soil net nitrogen rate accelerated due to the comprehensive actions of the decreased unstable organic carbon, increased NH₄⁺-N and enhanced microbial metabolism in the soil. Furthermore, experimental warming significantly increased rates of net nitrogen mineralization, as well as net nitrification and denitrification in soil under Picea asperata Mast. and Abies faxoniana var. faxoniana (Rehder and Wilson) Tang S. Liu in plateau-climate zones in most sampling periods [12]. This is because increasing temperature and nutrient availability altered the morphology (root length and type) and activity (e.g., root exudates) of plant roots, thus modulating soil nitrogen mineralization. By contrast, the soil net nitrogen mineralization rates in superficial and deep soil layers of a temperate Pinus tabuliformis plantation decreased by 52% (0–10 cm) and 51% (10–20 cm), respectively, under warming conditions [14]. This was attributed to warming decreasing soil moisture content, reducing soil microbial activity, and restricting the diffusion of soluble carbon and nitrogen. Surprisingly, in a subtropical Cunninghamia lanceolata (Lamb.) Hook. plantation, increasing soil temperature by 4 °C had no significant effect on the in situ soil net nitrogen mineralization rate, the net nitrification rate or the in situ soil N₂O emission rate [15]. This may be partly due to increasing temperature promoting the activity of nitrogen-transforming microorganisms but at the same time decreasing microbial immobilization of nitrogen. The above range of responses of soil nitrogen mineralization in field warming experiments highlights uncertainty for climate change predictions in forest ecosystems and the necessity for more comprehensive studies to be undertaken.

Several meta-analysis studies discussed different responses of soil nitrogen mineralization of forest ecosystems to warming [6,16,17], and concluded that there remains significant uncertainty due to inconsistent warming techniques and observation periods. Our previous research showed that during the first two years of warming, soil warming increased heterotrophic respiration and inhibited root respiration in a C. hystrix forest, leading to an overall decrease in topsoil SOC content of about 16.9% [18]. After 5 years of warming, the labile C in the trenched plots decreased significantly, while the effect of warming on SOC content leveled off during the latter three years of the study [19]. However, in a Q. aliena forest, soil warming substantially elevated soil respiration and autotrophic respiration [20]. Furthermore, 5 years of continuous soil warming caused a decline in nitrogen availability in the Q. aliena forest, while microbial biomass-specific nitrogen-acquisition enzyme activity increased significantly [21]. However, how warming affects the soil nitrogen transformation process and its regulatory mechanisms in different climatic zones is still not clear. Further field warming studies in forest ecosystems in different climatic zones are conducive to improving our understanding of the response mechanism of nutrient cycling to climate change. In particular, they can provide reference data for the parameterization of large-scale climate-nitrogen cycling models.

Soil warming experiments established in January 2012 in a south subtropical C. hystrix plantation [18] and in May 2011 in a temperate Q. aliena natural forest [20] provide an opportunity to identify ten-year responses of forest soil nitrogen mineralization in two climatic regions to soil warming. In this study, it is hypothesized that: (1) soil warming will increase soil temperature and promote fine root growth and microbial activity, thus increasing the soil nitrogen mineralization rate; and (2) the soil net nitrogen mineralization rate in the temperate forest will be more sensitive to warming than in the subtropical forest.

2. Materials and Methods

2.1. Study Area Sites

There were two field experiments, one study was conducted in a south subtropical C. hystrix monoculture plantation and the other was in a temperate Q. aliena pure natural forest. The C. hystrix plantation is located in the Guangxi Youyiguan Forest Ecosystem Research Station of Experimental Center for Tropical Forestry, Chinese Academy of Forestry (22°10′ N, 106°50′ E), at an elevation of 550 m. The area is part of the south subtropical monsoon climate zone, with 1200–1500 mm annual average precipitation, mainly from April to September, 1200–1400 mm annual evaporation, 80%–84% relative humidity, and 20.5–21.7 °C average annual temperature. In the 2012 survey, the average diameter at breast height of C. hystrix was 25.7 cm, and the tree density was 333 trees ha⁻¹. The forest soil type is Oxisol, and the soil depth is >80 cm.

The Q. aliena forest is located in the Positioning Research Station of the Henan Neixiang Baotianman Forest Ecosystem, National Forestry and Grassland Administration (111°47′–112°04′ E, 33°20′–33°36′ N), with an elevation of 1400 m. It experiences a temperate monsoon climate and has four distinctive seasons. The average annual precipitation is 894 mm, mainly from June to August, and the average annual temperature is 15.1 °C. In 2012, the average diameter at breast height of Q. aliena was 20.1 cm, and the tree density was 1270 trees ha⁻¹. The forest soil type is Hapludalfs, with a soil depth of 20–60 cm [20].

2.2. Experimental Design

A split plot experiment was established at each site with warming as the main plot, and trenching as the split plot. The four treatments were: control, warming, trenching, and warming + trenching. Each treatment had six replications, forming 24 split plots. Six pairs of 4 × 3 m plots were randomly established on each site over a 30 × 70 m area, each pair containing a plot that was subject to warming using an infrared heater and a control plot that was not subject to warming. Each plot was further divided into two 2 × 3 m subplots, one of which was randomly assigned to be trenched while the other remained non-trenched. The trenches were dug 1 m deep to minimize the influence of roots entering the subplots. The warming experiments commenced in January 2012 (C. hystrix) and May 2011 (Q. aliena), respectively, and have been ongoing continuously since then, 24 hours a day. More detailed information on site preparation and treatment protocols is provided by Wang et al. [18] and Liu et al. [20].

2.3. Soil Collection and Analysis

Soil nitrogen mineralization rates were determined at the sites using the PVC tube method described by Tsui and Chen [22] in two incubation periods (20 July–20 August 2021, 22 July–22 August 2022). At the beginning of each incubation, three sample points were chosen randomly and the surface litter was removed in each split-plot. At each sample point, a pair of tubes was inserted 20 cm vertically into the soil using a rubber mallet, leaving 8 cm above the ground. The PVC tubes had an inner diameter of 5 cm and height of 28 cm, and there were four holes near the top for gas exchange. The tops were sealed with PVC caps to prevent the loss of nutrients by precipitation and to exclude litter fall. Three tubes were taken out immediately, the content was partitioned into 0–10 cm and 10–20 cm layers, and the layers from the three tubes were combined and mixed thoroughly providing two samples for determining the initial NH₄⁺-N and NO₃⁻-N concentrations. The remaining tubes with closed caps and cotton gauze bottoms to exclude roots were left in situ and collected at the end of each incubation period for the measurement of the final NH₄⁺-N and NO₃⁻-N concentrations. Meanwhile, temperatures of soil layers 0–10 cm and 10–20 cm were measured with a digital thermometer (Harvesting Science and Technology Co., Ltd., Beijing, China) at three positions (upper, middle and lower) near each incubation point.

The formulae used to calculate the soil net nitrogen mineralization rates are as follows:

Net ammonification rate (NAR, mg N kg⁻¹ d⁻¹) = (ammonium nitrogen after incubation − ammonium nitrogen before incubation)/days of incubation

(1)

Net nitrification rate (NNR, mg N kg⁻¹ d⁻¹) = (nitrate nitrogen after incubation − nitrate nitrogen before incubation)/days of incubation

(2)

Net nitrogen mineralization rate (NMR, mg N kg⁻¹ d⁻¹) = NAR + NNR

(3)

Soil samples collected at the same soil depth in each plot were combined on-site, placed in plastic bags and transported to the laboratory on ice. A total of 48 mixed soil samples (24 split plots × 2 soil layers) were collected from each forest type at each sampling time. After removing stones and coarse roots, each soil sample was sieved through a 2 mm sieve and divided into two parts. One part was air dried and the physical and chemical properties were determined: soil pH (soil:water mass ratio 1:2.5), soil organic matter (SOC) concentration (acid potassium dichromate oxidation), total nitrogen (TN) concentration (Kjeldahl method), total phosphorus (TP) concentration (alkali fusion-molybdenum blue method), and available phosphorus (AP) concentration (Olsen P method). The other was stored at 4 °C for 48 h before measuring the soil moisture (SM, after drying at 105 °C for 48 h), carbon (C), nitrogen (N) and phosphorus (P) concentrations in soil microorganisms, as well as NH₄⁺-N and NO₃⁻-N concentrations (SEAL Auto Analyzer 3, SEAL, Norderstedt, Germany). We calculated soil bulk density (BD) by the following equation: BD (g cm⁻³) = soil fresh weight × (1-soil moisture)/soil volume.

Soil microbial biomass carbon (MBC) and microbial biomass nitrogen (MBN) concentrations were obtained by the chloroform fumigation-potassium sulphate extraction method [23]. The dissolved organic carbon concentrations in fumigated and non-fumigated filtrates were measured with a total organic carbon analyser (TOC, VCPH/CPN, Kyoto, Japan). The total soluble nitrogen concentration in the filtrate was measured using a continuous flow analyser (Skalar San + +, Skalar, Delft, The Netherlands). Microbial biomass phosphorus (MBP) concentration was obtained using the chloroform fumigation-NaHCO₃ extraction method [24] followed by spectrophotometry at 880 nm. These parameters were calculated using the following formulas:

MBC = ΔE_C/K_C

(4)

MBN = ΔE_N/K_N

(5)

MBP = ΔE_P/K_P/K_Pi

(6)

where ΔEC is the difference in dissolved organic carbon between fumigated and non-fumigated soils, and K_C is the transformation coefficient (0.45). ΔE_N is the difference in total soluble nitrogen content between fumigated and non-fumigated soils, and K_N is the transformation coefficient (0.57). ΔE_P is the difference in AP between fumigated and non-fumigated soils, and K_P is the transformation coefficient (0.4). KPi = (test value of soil added with KH₂PO₄ solution − test value of non-fumigated soil)/25 × 100%.

2.4. Fine Root Collection and Analysis

The roots obtained during soil collection were taken back to the laboratory for cleaning. The fine root (diameter ≤ 2 mm) fraction was obtained and morphological traits were analyzed using the root analysis system WinRHIZO 2013 (Regent Instruments Inc., Quebec, Canada). Next, they were dried in an oven at 65 °C to constant weight (48 h) and weighed to determine fine root biomass. The dried roots were ground in a ball mill, and the powder was used for elemental analysis. The C and N concentrations were obtained using an element analyser (Elementar Vario EL III, Elementar, Germany). The P concentration was measured using a continuous flow analyser (Skalar San + +, Skalar, Delft, The Netherlands) after acid digestion (HClO₄, H₂SO₄).

2.5. Statistical Analysis

We conducted two identical experiments using the same methods and timing, one in a C. hystrix plantation and the other in a Q. aliena forest. Thus, the statistical analysis was the same for both forests. In each forest, differences in the indexes at the same soil layer in the two years under different treatments were explored through one-way analysis of variance (one-way ANOVA). These indexes included soil physical and chemical properties, MBC, MBN, MBP, fine root biomass and nutrient concentrations, and net nitrogen transformation rates. In addition, differences in these indexes among different soil layers under the same warming and trenching treatments were analyzed. Furthermore, effects of warming, trenching, soil depth and interaction on soil physical and chemical properties, MBC, MBN, MBP, fine root biomass and nutrient concentrations, and net nitrogen transformation rates were examined by multi-way analysis of variance (multi-way ANOVA). We used Duncan’s multiple range test for comparing means across treatments. The statistical analysis was carried out using SPSS 25.0 software (SPSS Inc., Chicago, IL, USA). Images were plotted using the R 4.4.0 software.

To quantify the effects of soil variables, microbial variables, and fine root variables on NAR, NNR and NMR in soil, we employed the linear mixture model using the “lme” function in the R package “nlme”. The fixed effect terms included soil, microbial and fine root variables. All variables were normalized before modeling so that each variable had a mean of zero and a standard deviation of one. To reduce the effects of multicollinearity, a variance inflation factor < 3 was used to identify multicollinearity variables in multiple regression models [25]. We calculated the variance inflation factor using the R package “car”. The pseudo-R2 was calculated using the function “r.squared GLMM” from the R package “MuMIn” to represent the variance of the fixed effects interpretation in a linear mixed model. The effect size of the fixed factor was measured by the regression coefficient in the linear mixed model.

3. Results

3.1. Soil Physicochemical Properties

The influences of warming, trenching, soil depth and interaction on soil physical and chemical properties varied with tree species (

Table 1. Multi-way ANOVA results for the effect of warming (W), trenching (T), soil depth (D) and their interaction on soil physicochemical properties in the Castanopsis hystrix (CH) plantation and Quercus aliena (QA) forest.

Forest	Source of Variation	ST	SM	BD	pH	SOC	TN	AN	NN	TP	AP	C/N	C/P	N/P
CH	W	0.000 **	0.758	0.987	0.367	0.670	0.730	0.851	0.344	0.007 **	0.055	0.331	0.395	0.131
	T	0.477	0.943	0.963	0.679	0.657	0.825	0.728	0.678	0.845	0.631	0.149	0.563	0.859
	D	0.000 **	0.008 **	0.000 **	0.000 **	0.000 **	0.000 **	0.001 **	0.217	0.000 **	0.000 **	0.185	0.000 **	0.000 **
	T × D	0.733	0.777	0.549	0.371	0.569	0.422	0.955	0.867	0.651	0.907	0.517	0.727	0.474
	W × D	0.144	0.762	0.665	0.546	0.738	0.302	0.978	0.684	0.470	0.684	0.592	0.255	0.085
	W × T	0.516	0.944	0.395	0.332	0.785	0.584	0.725	0.302	0.242	0.274	0.934	0.387	0.285
	W × T × D	0.814	0.907	0.412	0.815	0.534	0.868	0.944	0.510	0.335	0.721	0.539	0.853	0.795
QA	W	0.000 **	0.644	0.560	0.000 **	0.736	0.599	0.055	0.550	0.291	0.338	0.783	0.846	0.878
	T	0.192	0.684	0.054	0.567	0.068	0.147	0.080	0.813	0.908	0.382	0.445	0.220	0.329
	D	0.001 **	0.140	0.000**	0.000 **	0.000 **	0.000 **	0.023 *	0.598	0.011 *	0.000 **	0.000 **	0.000 **	0.000 **
	T × D	0.943	0.704	0.414	0.946	0.968	0.597	0.231	0.100	0.411	0.306	0.221	0.598	0.945
	W × D	0.830	0.194	0.407	0.267	0.557	0.352	0.026 *	0.046 *	0.373	0.038 *	0.458	0.412	0.225
	W × T	0.245	0.068	0.393	0.457	0.210	0.091	0.369	0.174	0.781	0.529	0.212	0.327	0.101
	W × T × D	0.793	0.357	0.386	0.305	0.717	0.895	0.394	0.190	0.689	0.991	0.460	0.727	0.910

ST: soil temperature; SM: soil moisture; BD: bulk density; SOC: soil organic carbon; TN: total nitrogen; AN: ammonium nitrogen; NN: nitrate nitrogen; TP: total phosphorus; AP: available phosphorus. C/N: soil organic carbon/total nitrogen; C/P: soil organic carbon/total phosphorus; N/P: total nitrogen/total phosphorus. Significant probabilities are denoted by * and ** for p < 0.05 and 0.01, respectively.

). For the C. hystrix plantation, the ST (2.1%) and AP (21.3%) increased significantly at 10–20 cm soil depth after warming compared to the unheated control (p < 0.05,

Table 2. Effect of warming, trenching, and soil depth on soil physicochemical properties in the Castanopsis hystrix (CH) plantation and Quercus aliena (QA) forest in 2021 and 2022.

Forest	Index	Depth	ST (℃)	SM (%)	BD (g cm−3)	pH	SOC (g kg−1)	TN (g kg−1)	AN (mg kg−1)	NN (mg kg−1)	TP (g kg−1)	AP (mg kg−1)	C/N	C/P	N/P
CH	C	0–10	25.2 ± 0.1 Aa	29.2 ± 1.6 A	1.7 ± 0.1 B	4.1 ± 0.0 B	44.6 ± 2.8 A	2.8 ± 0.2 A	11.6 ± 1.7	1.1 ± 0.5	0.4 ± 0.0 A	1.8 ± 0.1 Aa	16.1 ± 0.3	130.0 ± 8.2 A	8.1 ± 0.6 A
		10–20	24.8 ± 0.1 Bb	25.3 ± 0.9 B	2.2 ± 0.1 A	4.2 ± 0.0 A	23.2 ± 1.9 B	1.5 ± 0.0 B	7.2 ± 1.1	0.9 ± 0.3	0.3 ± 0.0 B	0.8 ± 0.0 Bb	15.0 ± 0.5	79.1 ± 5.6 B	5.2 ± 0.3 B
	W	0–10	25.9 ± 0.1 Aa	29.5 ± 2.1 A	1.7 ± 0.1 B	4.1 ± 0.0 B	46.8 ± 3.0 A	2.8 ± 0.2 A	11.0 ± 1.4	1.8 ± 0.6	0.4 ± 0.0 A	2.0 ± 0.2 Aa	16.6 ± 0.5	129.9 ± 5.3 A	7.9 ± 0.3 A
		10–20	25.4 ± 0.1 Ba	26.1 ± 1.1 A	2.4 ± 0.1 A	4.2 ± 0.0 A	25.6 ± 1.3 B	1.7 ± 0.1 B	7.0 ± 0.7	1.5 ± 0.4	0.3 ± 0.0 B	0.9 ± 0.0 Ba	15.5 ± 0.6	84.7 ± 2.9 B	5.5 ± 0.2 B
	T	0–10	25.2 ± 0.1 Aa	29.0 ± 2.2 A	1.7 ± 0.1 B	4.1 ± 0.0 B	46.1 ± 2.3 A	2.9 ± 0.2 A	11.5 ± 1.6	1.6 ± 0.6	0.3 ± 0.0 A	1.8 ± 0.1 Aa	16.5 ± 0.7	136.4 ± 5.2 A	8.5 ± 0.5 A
		10–20	24.8 ± 0.1 Bb	25.6 ± 1.3 A	2.3 ± 0.1 A	4.2 ± 0.0 A	24.3 ± 2.3 B	1.5 ± 0.1 B	6.6 ± 0.6	1.1 ± 0.3	0.3 ± 0.0 B	0.8 ± 0.1 Bab	16.2 ± 0.8	84.9 ± 6.9 B	5.2 ± 0.4 B
	WT	0–10	25.8 ± 0.1 Aa	28.8 ± 2.3 A	1.7 ± 0.1 B	4.1 ± 0.0 A	45.9 ± 1.9 A	2.7 ± 0.1 A	10.8 ± 1.4	1.8 ± 0.4	0.4 ± 0.0 A	1.9 ± 0.1 Aa	17.0 ± 0.6	127.9 ± 5.1 A	7.6 ± 0.3 A
		10–20	25.3 ± 0.1 Bab	26.3 ± 1.1 A	2.3 ± 0.1 A	4.2 ± 0.0 A	24.3 ± 1.1 B	1.5 ± 0.1 B	7.0 ± 0.8	1.4 ± 0.3	0.3 ± 0.0 B	0.8 ± 0.0 Bab	16.1 ± 0.4	80.8 ± 3.9 B	5.0 ± 0.2 B
QA	C	0–10	19.7 ± 0.4 b	27.8 ± 0.5	1.9 ± 0.1 B	5.0 ± 0.1 Aa	37.7 ± 2.0 A	3.1 ± 0.2 A	30.3 ± 4.8 Ab	1.1 ± 0.2 A	0.5 ± 0.0 A	3.5 ± 0.4 A	12.5 ± 0.4 A	78.8 ± 5.0 A	6.4 ± 0.4 A
		10–20	18.8 ± 0.4 b	27.3 ± 0.5	2.1 ± 0.1 A	5.1 ± 0.0 Aa	22.8 ± 2.0 B	1.9 ± 0.1 B	26.1 ± 1.3 Aa	1.4 ± 0.4 A	0.4 ± 0.0 B	2.1 ± 0.3 B	11.7 ± 0.3 A	57.3 ± 3.7 B	4.9 ± 0.3 B
	W	0–10	21.0 ± 0.4 ab	29.9 ± 1.4	1.9 ± 0.1 B	4.8 ± 0.0 Ab	41.2 ± 2.3 A	3.4 ± 0.2 A	38.7 ± 3.1 Aa	0.8 ± 0.2 A	0.5 ± 0.0 A	3.2 ± 0.2 A	12.1 ± 0.3 A	84.4 ± 4.5 A	7.0 ± 0.4 A
		10–20	20.0 ± 0.4 ab	29.7 ± 0.9	2.1 ± 0.1 A	4.9 ± 0.1 Ab	23.7 ± 2.0 B	2.1 ± 0.1 B	29.7 ± 2.8 Ba	0.9 ± 0.3 A	0.4 ± 0.0 B	2.5 ± 0.2 B	11.4 ± 0.3 A	56.2 ± 2.2 B	5.0 ± 0.2 B
	T	0–10	19.9 ± 0.4 b	28.5 ± 1.0	1.9 ± 0.1 A	5.0 ± 0.0 Ba	37.3 ± 2.7 A	3.0 ± 0.2 A	28.6 ± 1.7 Ab	0.9 ± 0.2 A	0.5 ± 0.0 A	3.0 ± 0.2 A	12.5 ± 0.3 A	80.6 ± 4.9 A	6.5 ± 0.4 A
		10–20	18.8 ± 0.4 b	28.8 ± 1.3	2.2 ± 0.1 A	5.2 ± 0.0 Aa	21.5 ± 1.9 B	2.0 ± 0.2 B	24.1 ± 3.4 Aa	1.1 ± 0.3 A	0.4 ± 0.0 B	2.0 ± 0.1 B	10.8 ± 0.3 B	54.1 ± 3.3 B	5.1 ± 0.3 B
	WT	0–10	21.6 ± 0.3 a	29.2 ± 1.1	1.9 ± 0.1 B	4.9 ± 0.1 Bab	36.3 ± 2.0 A	3.0 ± 0.2 A	33.8 ± 2.6 Aab	1.7 ± 0.3 A	0.5 ± 0.0 A	2.9 ± 0.2 A	12.4 ± 0.4 A	78.6 ± 3.8 A	6.4 ± 0.3 A
		10–20	20.7 ± 0.4 a	26.1 ± 1.5	2.4 ± 0.1 A	5.1 ± 0.1 Aab	19.9 ± 1.6 B	1.8 ± 0.1 B	23.3 ± 2.8 Ba	0.7 ± 0.1 B	0.5 ± 0.1 A	2.7 ± 0.4 A	11.4 ± 0.4 A	49.4 ± 4.3 B	4.3 ± 0.3 B

Control: C; Warming: W; Trenching: T; Warming and Trenching: WT. ST: soil temperature; SM: soil moisture; BD: bulk density; SOC: soil organic carbon; TN: total nitrogen; AN: ammonium nitrogen; NN: nitrate nitrogen; TP: total phosphorus; AP: available phosphorus; C/N: soil organic carbon/total nitrogen; C/P: soil organic carbon/total phosphorus; N/P: total nitrogen/total phosphorus. Within a column, different lowercase letters indicate significant differences at p < 0.05 between treatments for the same soil depth, and different uppercase letters indicate significant differences between soil depths for the same treatment, based on one-way ANOVA and Duncan’s multiple range test. Values are means with standard errors.

). However, the difference between different warming and trenching treatments in the same soil layer was not significant for the other soil properties (p > 0.05, Table 2). In addition, the BD (32.5%–44.0%) and pH (3.0%–4.2%) of the 10–20 cm soil layer increased significantly compared to the 0–10 cm soil layer irrespective of warming or trenching treatments. Constantly, SOC, TN, TP, AP, C/P and N/P ratios of the 10–20 cm soil layer decreased sharply to different extents compared to the 0–10 cm soil layer for all treatments (p < 0.05, Table 2).

For the Q. aliena forest, the ST was significantly increased by 6.3% (0–10 cm) and 6.5% (10–20 cm) after warming, and also increased by 10.2% (0–10 cm) and 10.2% (10–20 cm) in the warming + trenching treatment. In addition, the NH₄⁺-N concentration of the Q. aliena forest increased significantly by 27.6% (0–10 cm) after warming, but the pH decreased sharply by 4.4% (0–10 cm) and 4.5% (10–20 cm) (p < 0.001, Table 2). However, there were no significant differences in the other soil properties between any warming and trenching treatments in the same soil layer (p > 0.05, Table 2). Comparing the two soil layers, the BD (12.8%–23.3%) and pH (3.8%–4.9%) of the 10–20 cm soil layer increased significantly compared to those of 0–10 cm soil layer regardless of warming or trenching treatments. However, SOC, TN, TP, AP, C/P and N/P ratios of the 10–20 cm soil layer decreased sharply to different extents compared to those of the 0–10 cm soil layer in all treatments (p < 0.05, Table 2). In addition, there were significant interactive effects of warming and soil depth on NH₄⁺-N, NO₃⁻-N and AP (p < 0.05, Table 1).

3.2. Soil Microbial Biomass

The effects of warming, trenching, soil depth and their interactions on MBC, MBN and MBP, and their ratios were not obvious in the C. hystrix plantation (

Table 3. Multi-way ANOVA results for the effect of warming(W), trenching (T), soil depth (D), and their interaction on the concentration and ratios of microbial biomass carbon (MBC), microbial biomass nitrogen (MBN), and microbial biomass phosphorus (MBP) in the Castanopsis hystrix (CH) plantation and Quercus aliena (QA) forest.

Forest	Source of Variation	MBC	MBN	MBP	MBC/ MBN	MBC/ MBP	MBN/ MBP
CH	W	0.019 *	0.152	0.860	0.741	0.244	0.353
	T	0.706	0.855	0.488	0.718	0.353	0.365
	D	0.067	0.853	0.886	0.264	0.266	0.385
	T × D	0.346	0.680	0.121	0.581	0.521	0.377
	W × D	0.233	0.302	0.377	0.303	0.284	0.329
	W × T	0.349	0.201	0.550	0.042 *	0.495	0.348
	W × T × D	0.418	0.530	0.654	0.976	0.498	0.380
QA	W	0.089	0.001 **	0.816	0.008 **	0.759	0.049 *
	T	0.167	0.921	0.078	0.435	0.157	0.960
	D	0.000 **	0.218	0.008 **	0.200	0.507	0.368
	T × D	0.004 **	0.168	0.293	0.615	0.011 *	0.157
	W × D	0.000 **	0.097	0.292	0.059	0.357	0.615
	W × T	0.858	0.849	0.122	0.479	0.383	0.297
	W× T × D	0.210	0.005 **	0.040 *	0.362	0.358	0.061

Significant values are denoted by * and ** for p < 0.05 and 0.01, respectively.

). Only warming significantly affected the MBC concentration (p < 0.05, Table 3), while soil depth and trenching did not influence MBC, MBN and MBP, or their ratios (p > 0.05, Table 3). In addition, the interaction of warming and trenching significantly influenced the MBC/MBN ratio in the C. hystrix plantation (p < 0.05, Table 3). Compared with the control, warming, and warming + trenching slightly increased the MBC concentration by 14.3% and 63.3%, respectively, but this was not significant (0–10 cm, p > 0.05, Figure 1a). In comparison with the 0–10 cm soil layer, the MBC concentration at 10–20 cm decreased by 48.3% after warming + trenching (p < 0.05, Figure 1a).

However, in the Q. aliena forest, warming greatly influenced the MBN, MBN/MBP and MBC/MBN; soil depth significantly affected the MBC and MBP (p < 0.05, Table 3); but trenching did not alter the MBC, MBN, MBP and their ratios (p > 0.05, Table 3). We also found that there were significant interaction effects on the MBC, MBN, MBP concentrations and the MBC/MBP ratio (p < 0.05, Table 3). The MBN concentration in the warming and warming + trenching treatments decreased by 49.4% and 42.4%, respectively, compared to the control (Figure 1h). This resulted in the MBC/MBN being increased by 207.6% and 226.2%, and the MBN/MBP being decreased by 69.6% and 80.1%, respectively (10–20 cm, p < 0.05, Figure 1j,k). Compared to the 0–10 cm soil layer, the MBN concentration at 10–20 cm decreased by 48.5% after warming (p < 0.05, Figure 1h), while MBC/MBN and MBC/MBP at 10–20 cm increased by 214.7% and 149.9%, respectively (p < 0.05, Figure 1j,l). In addition, the MBN/MBP of the 10–20 cm soil layer in the control plot increased by 152.5% compared to those at 0–10 cm (p < 0.05, Figure 1k).

3.3. Fine Root Biomass and Nutrients

Warming, trenching, soil depth and their interaction effects on the fine root biomass and nutrient concentrations varied with tree species (

Table 4. Multi-way ANOVA results for the effect of warming (W), trenching (T), depth (D) and their interaction on the concentration and ratios of root carbon (Root C), root nitrogen (Root N), root phosphorus (Root P) in the Castanopsis hystrix (CH) plantation and Quercus aliena (QA) forest, based on FRB: fine root biomass.

Forest	Source of Variation	FRB	Root C	Root N	Root P	Root C/N	Root C/P	Root N/P
CH	W	0.335	0.568	0.468	0.777	0.608	0.825	0.479
	T	0.000 **	0.179	0.357	0.787	0.231	0.648	0.736
	D	0.000 **	0.608	0.000 **	0.001 **	0.000 **	0.004 **	0.299
	T × D	0.840	0.752	0.863	0.552	0.995	0.570	0.442
	W × D	0.936	0.905	0.547	0.928	0.361	0.693	0.766
	W × T	0.476	0.727	0.192	0.294	0.320	0.445	0.721
	W ×T × D	0.231	0.355	0.099	0.229	0.604	0.706	0.908
QA	W	0.536	0.971	0.004 **	0.000 **	0.009 **	0.000 **	0.004 **
	T	0.001 **	0.418	0.748	0.985	0.687	0.640	0.828
	D	0.000 **	0.355	0.000 **	0.000 **	0.000 **	0.000 **	0.975
	T × D	0.312	0.488	0.662	0.403	0.879	0.679	0.487
	W × D	0.924	0.861	0.565	0.581	0.137	0.111	0.695
	W × T	0.560	0.592	0.770	0.465	0.741	0.276	0.316
	W × T × D	0.560	0.822	0.954	0.352	0.873	0.372	0.283

Significant probabilities are denoted by ** for p < 0.01.

). In the C. hystrix plantation, warming significantly increased the FRB of the 10–20 cm soil layer by 37.2% compared to the control (p < 0.05, Figure 2b), but had no apparent influence on the Root C, Root N, Root P concentration, and their ratios (p > 0.05, Figure 3a–f). In addition, trenching depressed fine root biomass of C. hystrix, although there was no statistical difference between trenching and the control (p > 0.05, Figure 2a,b). Overall, the FRB, Root N and Root P concentrations all decreased with soil depth, especially in the control and warming + trenching treatments (Figure 2 and Figure 3).

In the Q. aliena forest, warming significantly influenced Root N, Root P, Root C/N, Root C/P and Root N/P, while trenching apparently influenced the fine root biomass (p < 0.05, Table 4). The Root P concentration at 10–20 cm soil layer in the warming and the warming + trenching treatments decreased significantly by 26.2% and 23.8%, respectively, compared to the control (p < 0.05, Figure 3i). As a result, the Root C/P (37.2% and 27.4%) and Root N/P ratios (18.9% and 13.3%) increased significantly (p < 0.05, Figure 3k,l). Additionally, Root N (21.4%–28.8%) and Root P concentrations (17.7%–31.1%) were significantly lower in the 10–20 cm soil layer than in 0–10 cm soil layer (p < 0.05, Figure 3h,i). Consequently, the Root C/N (31.3%–42.2%) and Root C/P ratios (21.1%–43.0%) were far higher in the 10–20 cm soil layer than in the 0–10 cm soil layer (p < 0.05, Figure 3j,l).

3.4. Soil Net Nitrogen Mineralization Rate

In the C. hystrix plantation, compared to the control, the NAR and NMR of the 0–10 cm soil layer decreased by 100.0% and 69.8% after warming (p < 0.05, Figure 4a,c). However, the NNR of the 0–10 cm soil layer in the trenching and warming + trenching treatments increased by 100.0% and 106.3%, respectively (p < 0.05, Figure 4b). In addition, the NAR, NNR and NMR of the 10–20 cm soil layer decreased significantly by 24.3%–75.7%, 75.0%–87.5%, and 24.3%–75.7%, respectively, in comparison with those of the 0–10 cm soil layer (p < 0.05, Figure 4a–c). We also observed that there was an interactive effect of trenching and depth on the NNR; and the interactive effect of warming and soil depth on the NAR and NMR (p < 0.01,

Table 5. Multi-way ANOVA results for the effect of warming (W), trenching (T), depth (D) and their interaction on soil net nitrogen transformation rates in the Castanopsis hystrix (CH) plantation and Quercus aliena (QA) forest.

Forest	Source of Variation	NAR	NNR	NMR
CH	W	0.015 *	0.604	0.030 *
	T	0.846	0.000 **	0.244
	D	0.000 **	0.000 **	0.000 **
	T × D	0.900	0.000 **	0.216
	W × D	0.008 **	0.736	0.009 **
	W × T	0.783	0.731	0.875
	W × T × D	0.227	0.912	0.232
QA	W	0.538	0.795	0.528
	T	0.860	0.203	0.723
	D	0.000 **	0.000 **	0.000 **
	T × D	0.225	0.151	0.607
	W × D	0.855	0.706	0.757
	W × T	0.555	0.132	0.945
	W × T × D	0.659	0.092	0.304

NAR: net ammonification rate; NNR: net nitrification rate; NMR net nitrogen mineralization rate. Significant probabilities are denoted by * and ** for p < 0.05 and 0.01, respectively.

).

However, only soil depth had a significant influence on the NAR, NNR and NMR in the Q. aliena forest (p < 0.001, Table 5). For example, the NAR (83.3%–125.5%), NNR (82.4%–91.2%) and NMR (85.5%–107.3%) of the 10–20 cm soil layer significantly decreased compared to those of the 0–10 cm soil layer (p < 0.05, Figure 4d–f).

3.5. Factors Influencing the Net Nitrogen Transformation Rate

The 54%, 50% and 52% of the variation in NAR, NNR and NMR could be explained by the predictor in the C. hystrix plantation. Specifically, ST had a significant negative effect on NAR (standardized coefficient = −0.23 ± 0.09, p < 0.05), while Root P and FRB had a significant positive effect on NAR (standardized coefficient = 0.5 ± 0.2, p < 0.01; 0.4 ± 0.1, p < 0.05, Figure 5a). The MBN had a significant negative effect on NNR and NMR (standardized coefficient = −0.4 ± 0.2, −0.3 ± 0.1, p < 0.05, Figure 5b,c), while the FRB had a significant positive effect on NMR (standardized coefficient = 0.4 ± 0.2, p < 0.05).

In the Q. aliena forest, 56%, 49% and 73% of the variation in NAR, NNR and NMR, respectively, could be explained by the predictor. Among them, the ST had significant positive effects on NAR, NNR and NMR (standardized coefficient = 0.3 ± 0.1, 0.5 ± 0.2, p < 0.05, 0.5 ± 0.1, p < 0.01), and FRB had significant positive effects on NAR (standardized coefficient = 0.4 ± 0.1, p < 0.05), while NH₄⁺-N had a negative effect on NAR (standardization coefficient = −0.3 ± 0.2, p < 0.05, Figure 5d). The MBN had a significant positive effect on NNR (standardization coefficient = 0.2 ± 0.1, p < 0.05, Figure 5e); The influences of NH₄⁺-N, MBP and MBC on NMR were negative (standardized coefficient = −0.5 ± 0.1, p < 0.001, −0.2 ± 0.1, −0.1 ± 0.1, p < 0.05, Figure 5f).

4. Discussion

4.1. Warming Decreased Soil NAR and NMR in the Castanopsis Hystrix Plantation

In this study, the fine root biomass in the 10–20 cm soil layer in the C. hystrix plantation increased significantly after continuous soil warming for 11 years, indicating that trees probably increased the proportion of deep roots to cope with decreasing soil moisture content caused by the long-term warming process [26]. This was also observed in the studies of Wang et al. [27] and Zhao et al. [28] who found that fine root biomass had a stronger response to warming in deeper soil layers. Nevertheless, the fine root biomass at 0–10 cm in the C. hystrix plantation was not affected in the fourth and seventh years of warming, suggesting that short-term mild soil warming (+1.5 °C) had minimal influence on fine root growth in this soil horizon [18]. Previous studies have demonstrated that increasing the temperature by 4 °C had no influence on fine root biomass of 46-year-old Picea abies (L.) H. Karsten, but biomass declined sharply when the ambient temperature range was between +4 °C to +6 °C [29,30]. It is evident that the growth of fine roots is closely related to the duration and magnitude of temperature increases.

Soil microbial biomass or community structure plays a crucial role in soil nitrogen cycling [31]. In this study, the AP concentration in the 10–20 cm soil layer increased significantly after warming for 11 years, indicating that long-term climate warming will increase phosphorus availability in the south subtropical C. hystrix plantation. However, MBC, MBN, MBP and their ratios remained unchanged after warming, indicating that soil microbes did not benefit from the higher nutrient availability [32]. Unstable carbon consumption caused by long-term warming [19] might be a major reason for limited soil microbial biomass and nutrients as well as limited microbial reaction [33]. Previous studies demonstrated that the responses of MBC and MBN contents in soil to warming might be positive or negative. A meta-analysis study showed that MBC and MBN contents in soils had strong positive responses to mid-term (3–4 years) or short-term (1–2 years) warming, but their responses to long-term (5 years) warming tended to be weakened [34]. In this study, the soil microbes may have adapted to the warming episode up to 11 years, so they did not respond. In addition, the MBC content in soils shows relatively consistent positive responses to different increased amplitudes of temperature. In contrast, MBN’s response to a low increase amplitude of temperature (<1 °C) was negative, and its response to a high increase amplitude of temperature (>2 °C) was positive. This might be related to warming-induced changes in soil moisture and substrate supply [34].

Generally, the NMR of soil increases with an increase in temperature [35] because high temperatures enhance the activity of soil microbes [36] and extracellular enzymes, and accelerate the transformation of macromolecular organic matter into forms easily assimilated by plants and microorganisms [10]. However, Gao et al. [37] found that long-term warming (14 years) decreased soil nitrogen cycling of Mediterranean-climate annual grassland due to a reduction in the abundance of relevant functional genes. In the present study, the NAR of the 0–10 cm soil layer in the C. hystrix plantation also decreased significantly after 11 years of warming, which was opposite to our first hypothesis. On the one hand, the Root P and fine root biomass (FRB) had significant positive effects on NAR (Figure 5a). Specifically, warming increased the underground distribution of plant biomass by enhancing the growth and turnover of fine roots (Figure 2a,b) which secreted more carbon into the soil [38], stimulating the growth of nitrogen cycling-related microorganisms [39]. In this case, the microorganisms produce more extracellular enzymes, thus accelerating the soil NAR. On the other hand, ST had a significant negative impact on NAR (Figure 5a). The increase in temperature also would have accelerated the decomposition of labile carbon, further accelerating microbial nitrogen immobilization by heterotrophic microbes. This is consistent with the increase in MBN concentration in the 0–20 cm soil layer after warming (Figure 1b), which would have decreased the substrate for soil ammonification and lowered the NAR of soils [40]. Overall, the negative effect of ST on NAR was greater than the positive effect of Root P and FRB, resulting in a significant decrease in NAR for the 0–10 cm C. hystrix soil horizon under warming. The NMR and NAR of the 0–10 cm soil layer fell sharply, while the NNR remained unchanged under warming conditions.

However, the NNR in the C. hystrix plantation increased by 100.0% and 106.3% after trenching and warming + trenching treatments (Figure 4b). This suggests that the risk of soil NO₃⁻-N leaching loss during the wet season may increase after trenching, and it is further intensified by long-term warming. In other studies on carbon input control based on ditching, the NNR of soils also increased significantly after trenching was applied in coniferous forests dominated by 27-year-old Pinus taeda L. [41], Pseudotsuga menziesii (Mirb.) Franco and Tsuga heterophylla (Raf.) Sarg. [42]. The increase in NNR was explained as follows: First, the NH₄⁺-N absorption capacity of plants decreased after roots were cut, while the residual substrate for nitrification (NH₄⁺-N in soils) increased to accelerate the NNR of soils [43]. Second, after plant roots were broken, dead roots and root exudates decreased as carbon sources for heterotrophic microorganisms, resulting in a decrease in microbial biomass and the immobilization of NH₄⁺ by soil microorganisms [44]. Furthermore, root severing would cause an increase in NH₄⁺ available to nitrifying bacteria and an increase in the net nitrification rate due to the negative correlation between MBN and NNR (Figure 5b).

4.2. Warming Did Not Change the NMR of Soils in the Quercus Aliena Forest

It could be expected that fine root biomass in temperate forests might show stronger responses to warming than in subtropical or tropical forests [28,45]. However, we found that warming had little effect on the fine root biomass in the temperate Q. aliena forest. According to our early findings, the fine root biomass at 0–10 cm increased greatly in the first four years of warming, but such positive effects began to be weakened in the fifth year [20,21]. Short-term warming promotes fine root biomass, which may be caused by a variety of factors. First, warming can extend the growing season of plants [46], which has a positive impact on plant roots; Moreover, the experimental warming will increase evapotranspiration and decrease soil water [20], leading to an increase in fine root biomass to enhance water uptake. That fine root biomass was hardly affected in the 11th year of warming in this study demonstrates that the warming effect fine on root biomass gradually disappears as warming continues [27]. Warming can increase the availability of plant-usable nitrogen in the soil (e.g., NH₄⁺-N, Table 2). As a result, plants may not need to allocate additional photosynthetic resources to fine roots for nutrient acquisition, limiting the growth of fine roots under warming conditions [47].

We found that the MBN concentration in the 10–20 cm soil layer of the Q, aliena forest decreased significantly after 11 years of warming, thus increasing the MBC/MBN but decreasing the MBN/MBP ratios. However, the MBC and MBN of the 0–10 cm soil layer increased significantly in the first three years of warming [21]. This suggests that the positive response of soil microbial biomass also weakens gradually as warming continues. Short-term warming-induced increases in air and soil temperature and a fresh C supply from the plants’ above- and below-ground parts might have enhanced microbial growth [48], resulting in an increase in MBC and MBN. However, water losses due to long-term rising temperatures may inhibit microbial growth [49], which eventually decreases the MBN but not the MBC. The maintenance of MBC is probably because root-related C input had offset losses of carbohydrates [12]. The sharp reduction in soil pH (Table 2) might be a major reason for the decreased MBN in the 10–20 cm soil layer [50]. It has been suggested that climate warming in temperate zones will inhibit the nitrogen fixation rate of soil microbes and decrease soil microbial reserves [51], thus increasing the inorganic nitrogen content of soils [52]. The considerable increase in NH₄⁺-N concentration in the Q. aliena forest soils (Table 2) supports this view. In our study, the sharp reduction in Root P and sharp increase in Root N/P (Figure 3i,k) under warming conditions are consistent with previous research findings [27,28]. Ferric oxide, hydroxide and P adsorption by clay constituents in soils increases with temperature, while low soil pH (Table 2) enhances the fixation of soil P [53]. Thus, further long-term warming is likely to further decrease P absorption in the Q. aliena forest [54].

Warming usually improves the physiological activity of soil microbes, thus facilitating the decomposition and release of soil organic nitrogen and increasing the nitrogen content in soils [55]. Similarly, we found that NH₄⁺-N concentration in soils increased significantly after 11-year warming. However, warming had no significant influence on the NAR, NNR and NMR of the Q. aliena forest, indicating that nitrogen transformation of soil microbes in the temperate Q. aliena forest was less temperature sensitive than in the C. hystrix plantation. This disagrees with our second hypothesis. On the one hand, the increase in soil temperature promotes NH₄⁺-N concentration due to enhanced microbial metabolism and enzyme activity [56], thereby stimulating NMR and NAR. On the other hand, warming induces a reduction in SM which can hinder substrate accessibility [19], thus counteracting the temperature-induced stimulation of biochemical processes [51]. It seems that the negative effect of an increased NH₄⁺-N concentration offsets the positive effect of ST on the NMR and NAR of the Q. aliena forest, thus preventing significant fluctuation in NMR and NAR after warming.

The NNR of Q. aliena forest soils also did not change greatly after warming. The optimal temperature of nitrification is about 25–30 °C [57], which is higher than the soil temperature during the warming treatment of the two-year growing season (18.7–21.6 °C) in our site. Therefore, warming may promote NNR in the Q. aliena forest because it brings the soil temperature close to the optimum. However, a significantly lower pH not only leads to significant reductions in MBN (Figure 1h), but may also negatively affect the growth and activity of nitrifying bacteria [53], thereby reducing NNR. Overall, the positive effect of increased ST offsets the negative effect of decreased MBN, and therefore, the NNR was not changed under warming conditions in the Q. aliena forest.

It is worth mentioning that there might be seasonal differences in the influences of warming on soil nitrogen cycling of forests in different climatic zones. However, this was not examined in our study as we limited our investigation to the wet seasons in two successive years. Hence, further longitudinal studies concerning the influences of seasonal changes on soil nitrogen cycling of forests in different climatic zones are needed.

5. Conclusions

This study found that soil nitrogen mineralization was more affected by long-term warming in the subtropical C. hystrix plantation than in the temperate Q. aliena natural forest. The sharp reduction in NAR and NMR of the C. hystrix plantation was mainly because the increasing temperature promoted the nitrogen immobilization of heterotrophic microorganisms and resulted in fewer substrates for NAR. The sharp rise in NNR caused by trenching and warming + trenching was mainly related to the reduced NH₄⁺-N absorption of trenched plants and the reduced immobilization of NH₄⁺ by soil microorganisms. It was unexpected that NAR, NNR and NMR did not change in the Q. aliena forest after warming. This is because warming not only stimulates soil N mineralization by increasing NH₄⁺-N and reducing MBN, but also induces a reduction in SM which can hinder substrate accessibility, thus counteracting the temperature-induced stimulation of N mineralization. Therefore, the negative effect of increased NH₄⁺-N and decreased MBN would have offset the positive effect of ST on the NMR and NAR of Q. aliena forest. Overall, the different responses of soil nitrogen mineralization to warming could be attributed to differences in soil nutrient availability, fine root biomass and microbial biomass of forests in the two climatic zones.