Effects of labile carbon and phosphorus addition on N transformations with N‐ vs. non‐N‐fixing tree plantations

Abstract

Phosphorus (P) is the primary factor limiting soil organic matter (SOM) decomposition and carbon (C) storage in the tropical P‐deficient plantations. However, the mineralization of SOM derived from N‐fixing and non‐N‐fixing plantations may respond differently to P addition within the laboratory‐ or field‐based incubation experiment. In this study, the effects of P addition (+P) combined with glucose (+Glu) on soil N mineralization within Eucalyptus urophylla (EU) and Acacia auriculiformis (AA) plantations were evaluated using a 60‐d laboratory aerobic incubation assay in subtropical China. The results showed that Glu addition significantly affected the net N mineralization and nitrification in both EU and AA plantations. In the 7th day incubation, +Glu treatment enhanced the N mineralization, whereas +Glu+P dramatically accelerated the N immobilization in EU plantation. However, +Glu+P treatment increased N mineralization while +Glu enhanced N immobilization in AA plantation, suggesting that P addition greatly affected the direction of N transformation within different plantations under the labile C input in the early stage of incubation. However, in the 15th day incubation, +Glu increased while +Glu+P greatly reduced the N immobilization in EU plantation. On the contrary, +Glu promoted while +Glu+P further accelerated the N immobilization in AA plantations, indicating that P addition had the contrasting effects on N immobilization within EU and AA plantations. After the 15th day incubation, Glu addition further facilitated the N immobilization in two plantations. Although +Glu increased the fungi and bacteria PLFA biomass, +Glu+P significantly increased F/B ratios, which may contribute to the direction and magnitude of N transformations in two plantations. However, in these two plantation forests with low soil P availability, single P addition would not significantly alter the N mineralization mediated by microbe in the laboratory incubation experiment without the labile C input. Therefore, the significant effects of P addition on the N mineralization or immobilization varied with the incubation stage and plantation types. We recommend that the P availability has profound implications on SOM decomposition and C sequestration in subtropical plantations.

Introduction

Plantation forests are well known to maintain relatively higher primary production in terrestrial ecosystems and play a significant role in global commercial timber industries (Cook et al. , Masiero et al. ). It is estimated that the area of global plantation is ~2.78 × 10⁸ ha (FAO ). However, according to the eighth national forest resources inventory survey, the total area of plantation forests in China was ~6.9 × 10⁷ ha, accounting for a quarter of the world's plantation area. Further, the plantation forests contributed ~80% of the total forest carbon (C) sink increment of China (Fang et al. ). Therefore, the huge area of plantation forests are believed to be the potential C pools, mitigating global warming caused by the increasing CO₂ emission from the fossil fuel combustion and industrial development.

The plantation forests greatly enhanced the soil organic C sequestration (Chen et al. , Cook et al. ), as well as soil nutrient cycles and biological process (Forrester et al. , Bini et al. , Zhu et al. ). However, various tree species in plantation had significant effects on soil fertility and microbial community (Templer et al. , West et al. , Wang et al. ), especially the tree species in restoration plantation. In southern China, thousands hectare of land use patterns are being converted from natural forests to degraded land because of the excessive development and increasing population (Peng et al. , Ren et al. ). Therefore, a series of ecological engineering methods have been used to restore the degraded land since the last decades (Li et al. , Ren et al. ). In early stage of reforestation, the pioneer tree species such as Eucalyptus and Acacia are generally proposed for restoring degraded land and timber production (Wang et al. ). These plantations composed by Eucalyptus and Acacia greatly influenced the microclimate, soil fertility, flora, and fauna (Wang et al. , Zhao et al. , ). Given that, numerous studies have been conducted to investigate the relationship and feedback between soil and tree species in plantations (Chen et al. , Mo et al. ). Previous studies showed that tree species constitute in plantations would greatly influence C allocation (Chen et al. ), soil C sequestration (Resh et al. ), and soil N transformation (Wang et al. , Mo et al. ). However, these results suggested that the contrasting effects of N‐fixing and non‐N‐fixing tree species on soil C, N, and other nutrient were widely spread in forests. Nonetheless, the feedback and underlying mechanisms between the soil and tree species are still controversial due to the alteration of soil nutrient status and plant growth characteristics under the increasing atmospheric N deposition.

However, the N deposition has approached to 30–70 kg/yr in southern China (Fang et al. ). The increasing N deposition has significant effect on forest development (Fang et al. ), especially on the huge area of plantation forests in southern China (Liu et al. , ). Generally, the active N input potentially enhanced the tree growth (Lu et al. ), simultaneously affected the decomposition of litter and dead wood, and then stimulated the nutrient cycles in plantations (Chen et al. ). In addition, rapidly increasing N deposition would intensity the limitation of other element for plant growth in forests, especially for the P limitation (Bachmann et al. ). Hence, a long‐term field experiment has been established to investigate the effects of N and/or P addition on the above‐ and below‐ground ecological process (Mirmanto et al. , Ceccon et al. , Alvarez‐Clare et al. , Mayor et al. , Zhao et al. ). Therefore, assessing the potential effect of P addition on microbial mineralization process is critical for predicting soil C accumulation and nutrient retention in plantation ecosystems.

Phosphorus is the fundamental element for ecosystem and always limited the primary production in tropical forests (Alvarez‐Clare et al. , Sullivan et al. ). The variation of tree species may significantly affect the P cycles in plantation because of the different decomposition rates of litter and dead wood, as well as the microbial community under various plantations (Chen et al. ). For instance, N‐fixing tree species in tropical forest were believed to enhance the available P contents because of the N investing to the enzymes for active P (Huang et al. ). In addition, N‐fixing species were along with readily decomposable litter and high rates of nutrient cycling, as well as high rates of N fixation (Forrester et al. ). On the contrary, the non‐N‐fixing tree species, Eucalyptus urophylla, might experience the slower decomposition of litter and inner P cycles (Bouillet et al. ). In addition, in tropical P‐deficient soils, C or N combined with P addition could stimulate soil ammonia‐oxidizing bacteria (AOB) activity, suggesting that P availability may be vital for N transformations (Chu et al. ). Therefore, the mineralization of soil organic matter (SOM) derived from N‐fixing and non‐N‐fixing plantations may respond differently to P addition within the laboratory‐ or field‐based incubation experiment.

Nevertheless, the effect of field P addition treatment on microbial mineralization is determined by multiple factors (Zhang et al. , Zhu et al. ). Previous study showed that P addition might improve the microbial biomass (Liu et al. , Aronson et al. , Li et al. ). Yet, the continuous input of litter and root residual potentially increase the C and energy for microbe (Lemma et al. , Rubino et al. ). In addition, the microbial activities may also be influenced by other elements in plantation forests (Allen and Schlesinger , Paterson et al. ). Therefore, field‐based treatment is inadequate to distinguish the direct or indirect effect of P addition on microbial activities and community structure. Hence, it is very necessary to conduct the laboratory‐based incubation to investigate the underlying mechanisms of P addition on the microbial mineralization. We pose the following questions: (1) Does the single P addition enhance/suppress the soil net N mineralization and nitrification of E. urophylla and Acacia auriculiformis plantations based on laboratory study? (2) How the soil N mineralization and nitrification of these two plantations (N‐fixing vs. non‐N‐fixing tree species) respond to glucose addition with/without P under laboratory incubation experiment?

In this study, we conducted a laboratory‐based incubation experiment to investigate the interactive effect of glucose (+Glu) and phosphorus (+P) addition on the temporal N transformations of two mineral soil sampled from E. urophylla (EU, non‐N‐fixing tree species) and A. auriculiformis (AA, N‐fixing tree species) plantation forests in southern China. The study addresses the following hypothesis: (1) Single P addition would not significantly alter the N mineralization mediated by microbe within the soil sampled from these two plantation forests in the laboratory incubation experiment because of the absence of organic matter input (litter or root residual); (2) in these two plantation forests with low P availability, Glu addition would stimulate the soil N mineralization, simultaneously aggravate the microbial P limitation; however, P addition combined with glucose would enhance/suppress the microbial mineralization in N‐fixing/non‐N‐fixing plantation forests.

Materials and Methods

Site descriptions and soil sampling

The soils were sampled from two plantations in the Heshan National Field Research Station of Forest Ecosystems (112°50 E, 22°34 N), which is located in the middle of Guangdong Province, South China. The climate of this region is typical subtropical monsoon with mean annual temperature of 22.6°C and highest average temperature of 28.7°C in July and lowest average temperature of 14.5°C in January. The annual precipitation in this region is 1700 mm and over ~85% rainfall in the wet season. In this region, there is a distinct wet season (April–September) and dry season (October–March; Mo et al. ). Ambient atmospheric N deposition in precipitation was about 43.1 ± 3.9 kg N·ha⁻¹·yr⁻¹ in 2011 with 1:1 ratio for NH₄⁺ and NO₃⁻, which is almost fivefold higher than that in 1995 (8.31 kg N·ha⁻¹·yr⁻¹; Huang et al. , Zhu et al. ).

These two plantations, including N‐fixing A. auriculiformis (AA) and non‐N‐fixing E. urophylla (EU), were established on a degraded grassland site in 1984 with the area of ~5–8 ha. The two plantation forests were apart from 500 m (Zhang et al. ). The soil in this area has eroded seriously due to long‐term disturbances. The mean annual litter production was 694 ± 25 g air‐dried mass·m⁻²·yr⁻¹ and 590 ± 14 g air‐dried mass·m⁻²·yr⁻¹ in the AA and EU plantations, respectively (Zhu et al. ).

In each plantation, we established three blocks with 20 × 20 m for sampling soil. Each block was at least 20 m apart away. After removing the litter layer, the mineral soils from the top 0–10 cm of two plantations were sampled in July 2014. In each block, twenty cores of top 0–10 cm mineral soils were randomly sampled and pooled as a composite sample then sent to laboratory immediately. The visual roots, rocks, and litter residual of the sampled soils were removed and sieved by 2‐mm mesh. Each soil sample of two plantations was divided into three parts, one was used to determine the concentrations of NH₄⁺‐N, NO₃⁻‐N, dissolved organic carbon (DOC), dissolved organic nitrogen (DON), microbial biomass carbon (MBC), and microbial biomass nitrogen (MBN) within 48 h, the other one was air‐dried and detected the initial physicochemical properties, and the third part was stored at 4°C to prepare for the incubation experiments.

The soil sampled from AA plantation had significantly higher SOM and TN concentrations than that from EU plantation, while the EU plantation soil had the higher TP and available P concentrations rather than AA (Table ). The soil NH₄⁺‐N and NO₃⁻‐N concentrations in AA plantation were 2.85 and 10.86 mg/g, respectively, while those in EU plantation were 2.72 and 13.28 mg/g, respectively. In addition, the soil pH of EU was 3.72, which was significantly larger than that of AA plantation (Table ).

Initial soil chemical properties (0–10 cm) of two subtropical plantations with N‐fixing vs. non‐N‐fixing tree species

Forest type	EU	AA
pH (H2O)	3.72a ± 0.00	3.09b ± 0.05
SOM (%)	3.62b ± 0.06	4.22a ± 0.01
TN (mg/g)	2.04b ± 0.06	2.29a ± 0.04
TP (mg/g)	0.29a ± 0.07	0.23a ± 0.03
Available P (mg/kg)	2.87a ± 0.31	2.21a ± 0.22
NH4+‐N (mg/g)	2.72a ± 0.03	2.85a ± 0.37
NO3−‐N (mg/g)	13.28a ± 0.49	10.86b ± 0.15

Notes

EU, E. urophylla monoculture plantation; AA, A. auriculiformis monoculture plantation; SOM, soil organic matter; TN, soil total nitrogen; TP, soil total phosphorus. Different lowercase letters indicate significant difference between the two plantations (P < 0.05).

Experimental design

A labile C (added glucose) and P addition incubation experiment was designed as a randomized block design. The soils from two plantations were assigned to four treatments: soil alone (CT, control treatment), soil with glucose (+Glu), soil with glucose and P (+Glu+P), and soil with P (+P), each treatment had three replicates. The glucose was added to the soil at a rate of 2 g/100 g soil (dry weight), which was equal 0.8% of the SOC, while the mineral P (NaH₂PO₄) was added to the soil at a rate of 50 mg/100 g soil (dry weight), which was equivalent 0.01% of the total P concentration in forest soil.

Fifty grams (dry weight) of 2 mm sieved soil samples was added to 100‐mL plastic bottle and adjusted to 50% of the water‐holding capacity (WHC). A set of 120 microcosms (2 plantations × 4 treatment × 3 replicates × 5 sampling time) were set up for 60‐d incubation experiment. All soils were pre‐incubated at 20°C for 7 d. Thereafter, water or glucose and/or P (NaH₂PO₄) solution were added evenly dropwise to the surface using a pipette to uniform distribution. These additions raised soils to 60% of WHC, and they were maintained at this level throughout the experiment. A diaphragm with some needle holes was sealed on each plastic bottle to make gas exchange. In the 7th, 15th, 30th, 45th, and 60th day, the incubated soil samples of three plastic bottles of each treatment were collected to determine the contents of NH₄⁺‐N, NO₃⁻‐N, DOC, DON, MBC, and MBN at each time interval during 60‐d incubation. Regular water addition was carried on during the whole incubated experiment to keep constant soil moisture.

Measurements

The aired dry subsamples of two plantation forests were used for detecting the physicochemical properties in Table . Soil moisture was measured by oven‐drying for 24 h at 105°C. Soil pH was determined in 1:2.5 (soil: water, w/v) soil solutions. Soil NH₄⁺‐N and NO₃⁻‐N in filtered 2 mol/L KCl extracts of fresh soil sample were measured with a flow injection autoanalyzer (FIA, Lachat Instrument, Loveland, Colorado, USA). Soil available P was extracted with Bray‐2 solution and determined by the molybdate blue colorimetric method. The concentrations of MBC, MBN, DOC, and DON were measured by chloroform fumigation/extraction (modified after Vance et al. ). Briefly, 10 g fresh soil was extracted with 30 mL 0.5 mol/L K₂SO₄. An additional 10 g soil was fumigated with ethanol‐free chloroform for 24 h and then was extracted again in the same manner. Total organic C concentrations in the K₂SO₄ extracts were measured with Dimatec‐100 TOC/TIC analyzer (Dimatec Analysentechnik GmbH, Essen, Germany). Total organic C/N concentrations in the K₂SO₄ extracts from non‐fumigated soils were defined as DOC/DON. Soil total organic C (SOC) concentrations were measured using H₂SO₄‐K₂Cr₂O₇ oxidation method (Nelson and Sommers ). Soil total N (TN) was determined using the Kjeldahl acid‐digestion method with an Alpkem autoanalyzer (Kjektec System 1026 Distilling Unit, Hoganas, Sweden). Soil total P (TP) concentration was measured via the molybdate blue colorimetric method after samples were digested with sulfuric acid (H₂SO₄).

Soil phospholipid fatty acids (PLFAs) were measured and analyzed following the methods described by Bossio and Scow () and Moore‐Kucera and Dick () and a gas chromatograph (GC7890; Agilent, Wilmington, Delaware, USA). Microbial groups were identified and classified in accordance with bacteria PLFA biomarker (Frostegård and Bååth , Frostegård et al. ). The sum of saturated unsubstituted fatty acids of i14:0, i15:0, a15:0, i16:0, i17:0, a17:0 were considered as G⁺ bacteria; the sum of monounsaturated fatty acids of 16:1ω7c, 15:0 3OH, cy17:0, 16:1 2OH, 18:1ω7c, and cy19:0ω8c were chosen as G⁻ bacteria; the sum of G⁺, G⁻ bacterial biomass, and 14:0, 15:0, 17:0 were calculated as indicators of bacterial biomass; and the PLFAs used as fungal (F) biomarkers were 18:2ω6,9 and 18:1ω9c. Total microbial PLFAs were estimated as the sum of the all determined PLFAs plus 16:1ω5c, 10Me16:0, 10Me17:0, and 10Me18:0 (Frostegård et al. ). The ratio of fungal to bacterial PLFAs (F/B) was also calculated to indicate microbial community composition. All PLFAs were calculated as nanomole per gram dry soil in this study.

Calculations

In this study, the rates of net N mineralization and nitrification during the incubation period were calculated from the differences of inorganic N (NH₄⁺‐N+NO₃⁻‐N) contents divided by the time interval between the initial and after incubation. Cumulative net N mineralization and nitrification were calculated by summing the rates of net N mineralization and nitrification of each incubation period.

Data analysis

The effects of Glu and P addition and their interactions on NH₄⁺‐N, NO₃⁻‐N, net N mineralization and nitrification, cumulative N mineralization and nitrification, DOC, DON, MBC, MBN, and PLFA biomass for each plantation were explored by repeated‐measures two‐way ANOVA with Glu addition and P addition as the main effects. Results are reported as significant at P < 0.05. All data were performed using the SPSS 16.0 for Windows (SPSS Inc., Chicago, Illinois, USA).

Results

Dynamics of NH₄⁺‐N and NO₃⁻‐N, and dissolved organic carbon (DOC) and nitrogen (DON)

During the 60‐d incubation course, the NH₄⁺‐N and NO₃⁻‐N concentrations performed consistent pattern in CT treatment of the two plantation soils. In CT, the NH₄⁺‐N concentrations first slightly decreased and then gradually increased in EU and AA (Fig. a, b), while NO₃⁻‐N concentration continuously increased from the 1st to 60th day in two plantation soils (Fig. c, d). In CT and +P treatments of the incubation experiment, the NO₃⁻‐N was the dominating N form, which was ~4–10‐fold as high as NH₄⁺‐N in two plantations (Fig. ). RM‐ANOVA showed that glucose (Glu) addition significantly affected the NH₄⁺‐N and NO₃⁻‐N concentrations in both EU and AA soils (Fig. ). However, P addition and Glu × P only significantly influenced NH₄⁺‐N and NO₃⁻‐N contents in EU during the whole incubation experiment. For the 7th day incubation, two‐way ANOVA suggested that Glu addition significantly decreased the NH₄⁺‐N concentrations of these two soils (P = 0.000 for both) while only significantly affected the NO₃⁻‐N concentrations in EU (P = 0.016; Appendix S1: Table S1). Specially, compared with CT, +Glu significantly increased the NO₃⁻‐N concentrations while +Glu+P or +P had no significant effect on NO₃⁻‐N concentrations in EU soil (Appendix S1: Table S1). However, +Glu+P significantly increased the NO₃⁻‐N concentrations while +Glu or +P had no significant influence on NO₃‐N concentrations in AA soil (Appendix S1: Table S1).

RM‐ANOVA showed that Glu addition significantly enhanced the DOC concentrations while reduced the DON concentrations in two plantation soils. However, P addition or Glu × P had no significant effects on DOC or DON concentrations during the entire incubation experiment (Fig. ).

Rates and accumulation of net N mineralization and nitrification

RM‐ANOVA showed that Glu addition significantly affected the rate of net N mineralization and nitrification in EU and AA plantation soils (P < 0.000 for both). Nevertheless, there was no significant difference of P addition or Glu × P on the rate of net N mineralization and nitrification in these two plantations. Specifically, in EU soil, Glu addition significantly influenced the net N mineralization and nitrification rate across the whole incubation process, while P addition and Glu × P significantly affected the net N mineralization and nitrification rates only in the 7th and 15th day experiment (Appendix S1: Table S2). On the contrary, in AA soil, Glu addition significantly influenced the rates of net N mineralization and nitrification in 15th, 30th, 45th, and 60th day incubation course, while P addition significantly affected the net N mineralization and nitrification rates only in 15th day experiment. However, the Glu × P effects were significant only in the 7th and 15th day incubation (Appendix S1: Table S2).

In the 7th day incubation, the net N mineralization rate under +Glu treatment was significantly higher than that under +Glu+P treatment in EU soil (Fig. c), oppositely, the rate of net N mineralization under +Glu was significantly lower (negative value) than that under +Glu+P treatment in AA soil (Fig. b). However, in the 15th day incubation, the values of net N mineralization and nitrification in these two plantations soil were negative. Especially, the net N mineralization rate in +Glu+P treatment was significantly higher than that in +Glu treatment in EU soil (Fig. c). On the contrary, the net N mineralization rate in +Glu+P treatment was significantly lower than that in +Glu treatment in AA (Fig. d). The net rate of nitrification in 7th and 15th day experiment performed the similar pattern like the 7th and 15th day pattern of net N mineralization in EU and AA plantation soil (Fig. a, b). After 30th day incubation, there was no significant difference on the net N mineralization between +Glu and +Glu+P treatment in EU and AA mineral soil (Fig. ). However, there was either no significant difference on net N mineralization rate between CT and +P treatment across the 60‐d incubation experiment (Appendix S1: Table S2).

RM‐ANOVA also showed that Glu addition significantly influenced rates of net N mineralization and nitrification in EU and AA plantation soils (Fig. ). However, P addition and Glu × P significantly affected the net N mineralization and nitrification rate only in EU soil (Fig. a, c). There was no significant effect of P addition or Glu × P on the net N mineralization and nitrification in AA plantation soil (Fig. b, d). Additionally, the cumulative N mineralization of EU and AA soil in CT and +P treatment was continuously increased, while in +Glu and +Glu+P treatment was gradually decreased across the 60‐d incubation experiment (Fig. ). Specifically, in EU soil, the cumulative N mineralization in +Glu was significantly higher than that in +Glu+P, while in AA soil, the cumulative N mineralization in +Glu+P was significantly larger than that in +Glu in 7th day incubation. From the 15th to 60th day incubation, the cumulative N mineralization in +Glu and +Glu+P treatment was negative with no significant difference between these two treatments in EU and AA (Fig. c, d), suggesting Glu addition greatly enhanced the N immobilization by microorganisms. The cumulative nitrification in two plantation soil was quite similar as the cumulative N mineralization in the study (Fig. a, b).

Soil microbial biomass carbon (MBC), microbial biomass nitrogen (MBN), and PLFAs

RM‐ANOVA showed that Glu addition significant increased the MBC and MBN (P < 0.001 and P < 0.05, respectively), while P addition or Glu × P interaction did not significantly affect the MBC and MBN in EU plantation soil (Fig. a, c). However, Glu addition only significantly increased the MBC and did not significantly affect MBN in AA soil during the 60‐d incubation experiment. There was either no significant effect of P addition or Glu × P on the MBC or MBN in the AA plantation mineral soil (Fig. ).

In addition, the soil microbial PLFA biomass in the 7th day was measured in EU and AA plantation soils. The results of two‐way ANOVA showed that Glu addition had no significant effect on the fungi and G⁻ PLFA biomass, while significantly influenced the G⁺, bacteria, total PLFAs, and F/B ratios in EU and AA soils (Table ). However, there was no significant difference of P addition or Glu × P interaction on the soil microbial PLFA biomass in the 7th day incubation of these two plantation soils. Compared with CT treatment, +Glu significantly increased the fungi biomass; however, the fungi biomass under +Glu was higher than that under +Glu+P treatment in EU and AA soils. For bacteria biomass, +Glu+P had a positive effect in EU while a negative effect in AA soil. In addition, +Glu+P and +Glu significantly enhanced the total PLFA biomass, with an order of +Glu > +Glu+P. And +Glu and +Glu+P treatment also significantly increased the F/B ratios in EU and AA soil, with an order of +Glu > +Glu+P (Table ).

Soil microbial PLFA biomass (nmol/g dry soil) in the 7th day incubation after glucose (+Glu) and phosphorus (+P) addition of two subtropical plantations

Forest types	Microbial PLFA biomass	CT	+Glu	+Glu+P	+P	Two‐Way ANOVA
Glu	P	Glu × P
EU	Fungi	0.30b ± 0.01	2.82a ± 0.31	1.40b ± 0.72	0.32b ± 0.00	0.789	0.748	0.248
G−	3.51 ± 0.05	3.88 ± 0.18	3.90 ± 0.19	3.64 ± 0.26	0.130	0.688	0.766
G+	5.72b ± 0.04	6.13ab ± 0.19	6.28a ± 0.16	5.90ab ± 0.15	0.029	0.279	0.923
Bacteria	9.34b ± 0.08	10.15ab ± 0.38	10.32a ± 0.06	9.67ab ± 0.38	0.029	0.380	0.783
Total	16.32c ± 0.15	29.01a ± 1.50	25.22b ± 1.49	17.11c ± 0.53	0.000	0.205	0.069
F/B	0.03b ± 0.00	0.28a ± 0.02	0.14b ± 0.07	0.03b ± 0.00	0.002	0.095	0.090
AA	Fungi	0.31b ± 0.00	2.24a ± 0.28	1.88a ± 0.05	0.31b ± 0.00	0.878	0.414	0.824
G−	3.85 ± 0.02	3.42 ± 0.42	3.25 ± 0.16	3.50 ± 0.02	0.129	0.386	0.860
G+	6.08a ± 0.09	5.39ab ± 0.56	4.86b ± 0.20	5.85ab ± 0.02	0.026	0.230	0.667
Bacteria	10.09a ± 0.11	8.96ab ± 0.99	8.24b ± 0.37	9.50ab ± 0.02	0.051	0.274	0.862
Total	16.95b ± 0.20	24.81a ± 2.77	21.88a ± 0.77	16.12b ± 0.19	0.002	0.263	0.437
F/B	0.03c ± 0.00	0.25a ± 0.01	0.23b ± 0.01	0.03c ± 0.00	0.019	0.772	0.839

Notes

Values are mean ± SE for n = 3. Different superscript letters with a row indicate significant differences among four treatment at P < 0.05. And the bold value in the right parts of the table specially indicated the significant differences within Glu, P or Glu × P treatment. EU, E. urophylla monoculture plantation; AA, A. auriculiformis monoculture plantation.

Discussion

Variation of N form and transformation in legume and non‐legume tree species

It is generally considered that the soil N form and transformation were always governed by tree species in plantations (Maithani et al. , Ross et al. , Wei et al. ). In this study, the dominant type of N was nitrate, which was about fourfold to 10‐fold as high as ammonium in both plantations (30 yr) during the 60‐d incubation experiment. However, there was no significant effects of plantation type (legume and non‐legume) on nitrate and ammonium contents in our incubation experiment, which was inconsistent with the results from Li et al. () who reported that legume plantation (13 yr) was dominated by nitrate N while non‐legume plantation (13 yr) was dominated by ammonium N in an aerobic incubation experiment. The stand age (30 yr vs. 13 yr) might cause the different N forms in legume and non‐legume between our study and the Li's study (Li et al. ). Although the plantation soil dominated by legume species always had the higher N content rather than that dominated by non‐legume species (Forrester et al. , Zhu et al. ), the N supply originated from litter decomposition with various C:N ratios might dramatically change with the increasing stand age of forests (Idol et al. , Carrillo et al. ).

In tropical red soil, P addition at lower levels accelerated soil nitrification via increased AOB activity, while higher levels of P addition might maintain higher soil ammonium from litter decomposition in soil through a reduction of N mineralization enzymes and the stabilization of the AOB community structure (Chen et al. ). This supported our results that N nitrification was greater than ammonification under P addition in two plantations of subtropical regions.

Contrasting effects of glucose with/without phosphorus addition on net N mineralization and nitrification

It is generally considered that soil C was a dominating determinant of N cycling in the terrestrial ecosystems due to all N mineralization and uptake were always mediated by enzymatic systems that require C and energy for their synthesis and expression (Thomas et al. ). In laboratory‐based incubation experiment, labile C input would stimulate the mineralization of SOM and microbial activities in forest soil (Fontaine et al. , Nottingham et al. ). As expected in our hypothesis, Glu addition significantly stimulated the N mineralization in the EU and AA soils in laboratory incubation experiment. However, the direction and magnitude of N mineralization in these two plantation soils were greatly depended on the soil nutrient status (Schmidt et al. , Chen et al. ). In our study, the SOM content in AA soil was significantly higher than that in EU soil (Table ), which might result in the different response to the exogenous P addition. Previous study also showed that the SOM mineralization mediated by microbe was first limited by the C source (Fontaine et al. ); hence, the soil varied with SOM contents may respond differently following the litter or labile C addition. In the 7th day incubation of this study, Glu addition significantly increased the net N mineralization in EU soil while greatly reduced those in AA soil. This was in accordance with prior research that the C limited the decomposer in forest soil (Barantal et al. ). However, in the 15th day incubation, the Glu addition had converted the direction of N mineralization in these two plantation soils. The higher N immobilization rate was observed in EU soil than that in AA soil, suggesting that higher inorganic N concentration supported the larger microbial activity in EU soil (Table ). This may also be attributed to the C limitation combined with nutrient supply together driving the microbial mineralization of SOM (Barantal et al. , Nottingham et al. ).

However, in P‐poor tropical soils, the mineralization of SOM was determined by C source and nutrient availability (Cleveland et al. , Barantal et al. ). Previous study indicated that P availability was more important in promoting microbial mineralization of sucrose C while N availability was more crucial in driving the priming of pre‐existing soil organic C (Nottingham et al. ), suggesting that P availability was greatly correlated with the N mineralization together with C amendment in P‐poor tropical plantation soil. As expected in hypothesis 2, P addition may enhance/suppress the net N mineralization of these plantation soils under laboratory incubation experiment. In our study, P addition together with glucose enhanced the net N mineralization in AA soil while reduced the net N mineralization in EU soil of the 7th day incubation. However, in the 15th day, P addition combined with glucose had higher N immobilization in AA soil rather than in EU soil. This suggested that the mineralization of SOM mediated by microbe might be limited by C and/or P availability in different stages and soil status (Cleveland et al. ). Inorganic P addition could accelerate litter N release and soil N mineralization through stimulating soil urease and dehydrogenase enzymatic activities in red acid P‐deficient soil (Chen et al. ). Our study also indicated that Glu addition significantly increased the fungi PLFA biomass and F/B ratios, which was closely related with the SOM mineralization, whereas Glu addition together with P might significantly decrease the fungi PLFA biomass and F/B ratios in the 7th day incubation. These contrasting effects of Glu addition with/without P addition resulted in the direction and magnitude of SOM mineralization in legume and non‐legume plantation of this study. In addition, our results also showed that P addition significantly promoted N immobilization in EU soil while enhanced N mineralization in AA soil in the early incubation stage (Fig. ; Appendix S1: Table S2).

Phosphorus addition alone did not significantly alter the net N mineralization and nitrification

In the field‐based study, P addition may enhance the microbial biomass, litter production, and nutrient input in tropical forests (Mirmanto et al. , Li et al. , Zhu et al. ). However, combined results of laboratory and field experiment demonstrated that C mineralization was strongly constrained by P availability in tropical forest soil (Cleveland et al. ). Except that, P addition to tropical soils (no C added) also suggested that microbial utilization of at least labile fractions of SOC was also P limited (Cleveland et al. ). In this study, P addition alone did not alter the direction and magnitude of net N mineralization and nitrification in these plantation soils, which was consistent with our hypothesis. Moreover, P addition alone did not change the soil microbial PLFA biomass of two plantations in this study (Table ). This may be due to the C supplement first limited the decomposer then the P availability also contributed to the microbial mineralization of plantation soil (Barantal et al. ).

On the contrary, the C limitation was the first factor determining the N mineralization mediated by microbe (Cleveland et al. ), causing no variation effects of P addition alone on microbial mineralization within these two plantation soils in this study. However, increased C input through leaf litter or labile SOC can stimulate SOM mineralization, whereas atmospheric P deposition can reduce this stimulatory effect and promote soil C storage in tropical forests (Wang et al. ). Given that, we suggest that P limitation of microbial decomposition may have profound implications for C and N cycling in tropical forest, including their potential response to increasing atmospheric CO₂ and N deposition.

Conclusions

In these two plantation forests with low soil P availability, single P addition would not significantly alter the N transformation mediated by microbe in the laboratory incubation experiment because of the absence of labile C input (litter or root residual); however, labile C addition (glucose) would stimulate the soil N transformations, which simultaneously aggravate the microbial P limitation. Therefore, P addition combined with glucose would enhance/suppress the microbial N transformations in N‐fixing/non‐N‐fixing plantation forests. Our results indicated that P addition combined with labile C (glucose) addition dramatically altered the direction and magnitude of N transformations, which varied with the incubation stage and plantation types. We recommend that P availability had significant implications on SOM mineralization and C storage in tropical plantations and forests for the rapidly increasing atmospheric N deposition.

Acknowledgments

This work was funded by Natural Science Foundation of China (31700371, 31670621, 31600353), the Natural Science Foundation of Guangdong Province, China (2016A030310450). We thank Dr. Faming Wang for his comments on the experiment design and draft. We also appreciate the anonymous reviewers for their helpful and detailed comments and advice for improving the manuscript.