J Soils Sediments DOI 10.1007/s11368-015-1121-7

SOILS, SEC 3 • REMEDIATION AND MANAGEMENT OF CONTAMINATED OR DEGRADED LANDS • RESEARCH ARTICLE

Variations of soil N transformation and N2O emissions in tropical secondary forests along an aridity gradient Yu Xie 1 & Jinbo Zhang 1,2 & Lei Meng 3 & Christoph Müller 4,5,6,7 & Zucong Cai 1,3

Received: 3 December 2014 / Accepted: 18 March 2015 # Springer-Verlag Berlin Heidelberg 2015

Abstract Purpose The aim of the study was to explore the effects of climate-induced soil properties on nitrogen (N) transformation and N2O emissions in the tropical region. Materials and methods Soil samples were collected from the soils developed from granite and covered with secondary forest vegetation along with the aridity gradient in the Hainan Islands, China, in which the annual mean temperature ranged from 24 to 25 °C, but the annual precipitation from 1,000 to 2, 166 mm, and the annual evaporation from 1,861 to 2,409 mm within about 180 km. The gross of N transformation rates and Responsible editor: Hong Jie Di Electronic supplementary material The online version of this article (doi:10.1007/s11368-015-1121-7) contains supplementary material, which is available to authorized users. * Zucong Cai [email protected] 1

School of Geography Sciences, Nanjing Normal University, Nanjing 210023, China

2

Jiangsu Provincial Key Laboratory of Materials Cycling and Pollution Control, Nanjing, China

3

Jiangsu Center for Collaborative Innovation in Geographical Information Resource Development and Application, Nanjing, Jiangsu, China

4

Key Laboratory of Virtual Geographical Environment (VGE), Ministry of Education, Nanjing Normal University, Nanjing, Jiangsu, China

5

College of Agriculture, Hainan University, Haikou 570208, China

6

Department of Plant Ecology (IFZ), Justus-Liebig University Giessen, Heinrich-Buff-Ring 26, 35392 Giessen, Germany

7

School of Biology and Environmental Science, University College Dublin, Dublin, Ireland

N2O emissions in the soils were determined by employing 15N tracing methods at laboratory incubation under conditions of 60 % WHC and 25 °C. Results and discussion The gross rate of mineralization and turnover rate of soil organic N increased significantly (P<0.05 for both), but net mineralization rate did not differ significantly with the increase of aridity index (Di) (P>0.05). Gross and net rate of nitrification increased significantly (P<0.05 for both), while cumulative N2O emissions over the incubation period did not differ significantly along the aridity gradient (P>0.05). Statistical analysis showed that soil pH, which was correlated significantly with Di, was correlated significantly with the gross rate of N mineralization, turnover rate of soil organic N, gross nitrification, and the ratio of N2O in nitrification products. Conclusions The investigation indicated that the climateinduced soil properties had significant effects on the gross and net N transformation rates, but not on net mineralization and N2O emissions in the tropical secondary forests in the Hainan Island. 15

N tracer . Aridity index . Gross rate of N transformation . Nitrous oxide

Keywords

1 Introduction The productivity in terrestrial ecosystem is directly related to the nutrient supply capacity (Vitousek and Reiners 1975). Nitrogen (N) is one of the key factors regulating the biomass growth and can be lost easily via leaching and nitrogenous gas emissions. N availability for plants and losses from soils are dependent on their forms, which are controlled by N transformation (Müller et al. 2007). It has been documented that soil N transformation shows its local climate characteristics. For

J Soils Sediments

instance, the organic N mineralization rate is higher and ammonium (NH4+) oxidation rate is lower in the humid subtropical soils than those in the soils of temperate regions (Zhang et al. 2013b). Soil N content and net N-mineralization rate was reported to increase generally with increasing humidity across a regional humidity gradient in the northern Patagonia and in the northern of China (Wang et al. 2005; Bertiller et al. 2006). It has been known that the soils across climate zones have climate-specific strategies for balancing between supply and conservation of N (Zhang et al. 2013b). For instances, the soils developed in the semiarid and arid zones are often alkaline and their inorganic N is dominated by nitrate (NO3−) due to the high rate of nitrification at high soil pH, thus reducing the NH4+ content and NH3 volatilization in high pH soil (Zhang et al. 2013b). Because of limited water percolation, NO3− loss via leaching and runoff in the arid and semiarid zones is negligible (Peterjohn and Schlesinger 1990), even though nutrient cycling rates in arid soils are high (Zaady et al. 1996; Zaady 2005). In contrast, in the soils developed in the humid subtropical and tropical regions, the inorganic N is dominated by NH4+ due to the low rate of nitrification at low soil pH, thus NO3− leaching and runoff is minimized by suppressing NO3− production, and the acidic conditions effectively suppress NH 3 volatilization in these soils as well (Zhang et al. 2013b). Therefore, to understand N transformation in soils and their relations with the local climate provides insights into the climate-specific strategies for balancing between supply and conservation of N. Soils are the most important source of atmospheric nitrous oxide (N2O), as a by-product of nitrification and intermediate product of denitrification (Mosier et al. 1998). Recent estimate showed that N2O emissions from soils under natural vegetation alone ranged between 3.3 and 9.0 Tg N2O–N yr−1 (Ciais et al. 2014). N2O is produced in soils not only from NH4+ pool through nitrification and NO3− pools through denitrification but also from organic N through heterotrophic nitrification (McLain and Martens 2006; Zhang et al. 2011; Müller et al. 2014). It has been reported that the N2O emissions from the soils under both natural vegetation and upland crops are positively correlated with precipitation across China (Lu et al. 2006; Cai 2012). However, the underlying mechanisms have not been explored completely. Precipitation affects N2O emissions from soils probably through two pathways: determining soil properties, such as soil pH, which affects N transformation and the ratio of N2O in the products of N transformation processes (Zhang et al. 2011) and directly driving nitrification and denitrification, which produce N2O (Liu et al. 2014). Soil properties vary greatly across China, from very acid in the humid tropical and subtropical zones to very alkaline in the semiarid and arid temperate zones (Hseung 1980). Accordingly, the patterns of N transformation in soils vary greatly from tropical and subtropical zones to temperate zones (Zhang et al. 2013b). The effects of climate-induced soil properties on N transformations and

N2O emissions are intrinsic and persistent. In contrast, the N2O emissions driven directly by rainfall events are instantaneous and peaked emissions of N2O from soils are frequently observed immediately after rainfall events (e.g., Stange et al. 2000). Therefore, the question arises whether the correlations of precipitation with N2O emissions from cropland and noncropland soils across the country of China are established because of its instantaneous effects of precipitation or intrinsic and persistent effects of precipitation-induced properties or intrinsic and persistent effects or both? Hainan Island (18° 9′–20° 11′ N and 108° 36′–111° 03′ E) is located in the southeast of China. Although the island covers only 35,100 km2, the climate varies from humid to semihumid and further to semiarid zones from the East to the West due to the obstruction of the Wuzhi Mountain laid on the middle island from the North to the South, but temperature varies very little. In line with the climate variations, the zonal soils vary from acidic in humid climate zone to neutral or even alkaline in semiarid climate zones. The characteristics of climate and soils in the island provide excellent materials for exploring the effects of climate-induced soil properties on N transformation and N2O emissions in soils distributed along with the climate zones. In order to explore the effects of aridity on N transformation and N2O emissions through the persistent effects, we collected soil samples from humid to semiarid climate zones, where the annual mean temperature varied within 1 °C and the soils were developed from the same parent materials (granite) under the secondary forest and their organic C and N contents were comparable. N transformation in and N2O emissions from the soils were determined in laboratory incubation at the same temperature and same soil moisture by using 15N tracing method. It was hypothesized that: (1) aridity change controls the N production and consumption mechanisms. NH4+ oxidation are much higher in semiarid regions than in humid regions; (2) soil pH driven by the aridity is a key factor regulating nitrogen transformation rates; and (3) the instantaneous effect is more important than the effect of climateinduced soil properties in regulating N2O emission along this aridity gradient.

2 Materials and methods 2.1 Soil samples In total, 18 soil samples were collected along the aridity gradient from the humid to semihumid and further to the semiarid zones in the Hainan Island, among which, six were collected from Wanning and Qionghai in the humid zone, six from Ledong and Sanya in the semihumid zone, and six from Dongfang in the semiarid zone (Fig. 1). The annual mean temperature across the gradient varies from 24 to 25 °C, the

J Soils Sediments

Fig. 1 Soil sampling sites in the Hainan Island, China

annual precipitation varies from 2,166 mm in the humid zone to 1,000 mm in the semiarid zone, and the evaporation varies from 1,861 mm in the humid zone to 2,409 mm in the semiarid zone (Jing 2006). Corresponding to the difference in precipitation and evaporation, the predominating vegetation also varies from tropical rain forest and tropical monsoon forest in the humid zone, tropical monsoon forest in the semihumid zone, and tropical bosks in the semiarid zone, respectively. The tropical rain forests are dominated by evergreen trees, such as Vatica astrotricha, Tarrietia parvifolia, and Diospyros hainanensis, and Leguminosae. The tropical monsoon are mainly arboreal evergreen trees such as Kleinhovia hospita, Spondias pinnata, Hainania triochosperma, Apocynaceae, Anacardiaceae, Tiliaceae, Myrtaceae, Altingiaceae, Meliaceae, and Euphorbiaceae. The tropical bosks are mainly Flacourtia, Pandanus, Phoenix, and Cactaceae (Mao et al. 2006). All the sampled soils are developed from granite materials and classified into humic acrisol except the samples collected in Dongfang which is ferric authrosol (FAO 1988). To avoid anthropogenic effects on N transformation rates, all the samples were collected in unfertilized secondary forests. The distance from one sampling site to another was at least 100 m. All the samples were taken from A horizon, passed through a 2-mm sieve, and then split into

two subsamples. One sample was stored at 4 °C for the incubation studies within 1 month, and the other was air-dried for soil properties analyses (Table 1). 2.2 15N tracing experiment in laboratory For exploring the effects of soil properties on N transformation and N2O emissions, the laboratory incubation was conducted under the same soil moisture and temperature following the methods described by Zhang et al. (2013a). The method was also applied by many researchers for the quantification of N cycling in soils (Müller et al. 2004, 2007; Huygens et al. 2007). There were two different 15N treatments including NH415NO3 and 15NH4NO3, each has three replicates. The abundance of 15N in both 15NH4+ and 15NO3− was 10.03 atom % excess. Before the N solution addition, 30 g soil in the 250ml flasks was preincubated for 24 h at 25 °C. The rate of N addition was 2.86 μmol N g−1 soil (equivalent to 20 μg NH4– N g−1 soil and 20 μg NO3–N g−1 soil), and the N solution was added through pipettes to assure the applied N evenly distributed in the soils. After the N solution addition, distilled water was added to adjust the soil water-holding capacity (WHC) to 60 %, and then the flasks were sealed with silicone rubber stoppers, which were removed 30 min every 2 days for

J Soils Sediments Table 1 Average air temperature (AMT), average precipitation (AP), average evaporation (AEV), aridity index (Di), dominant tree species, soil organic carbon (SOC), soil total N (TN), soil pH, soils structure and ammonium (NH4+), and nitrate (NO3−) in fresh Items

Sampling site Humid QiongHai

AMT (°C)a AP (mm)a AEV (mm)a Dia Vegetation typeb Soil moisture (%) TN (g N kg−1) SOC (g C kg−1) pH NH4+ (mg N kg−1) NO3− (mg N kg−1) Clay (%)<2 μm Silt (%) 2–20 μm Sand (%) 20–2,000 μm

Semihumid Wanning

Average

24 24.5 2,072 2,166 1,861 1,873 0.9 0.86 Tropical rain, tropical monsoon forest 25.68±± 8.07 24.18±1.45 25.18±6.34a 2.31±0.91 1.93±0.30 2.18±0.74a 20.73±2.94 21.82±3.01 21.09±5.58a 4.77±0.89 4.79±0.04 4.77±0.69a 3.39±0.82 3.71±0.38 3.39±0.70a 20.35±2.92 14.83±1.42 18.51±3.70a 22.75±2.22 23.75±1.77 23.08±1.96b 31.75±5.56 28.87±1.77 30.75±4.64a 45.50±5.97 47.45±3.54 46.17±5.00a

Sanya

Semiarid Ledong

25.5 25 1,400 1,600 2,361 2,300 1.69 1.44 Tropical monsoon forest 11.92±1.63 13.71±1.70 1.83±0.29 1.00±0.16 17.56±2.88 10.18±1.44 6.34±0.11 6.43±0.36 8.07±2.67 5.61±1.33 12.59±1.99 18.34±9.63 18.13±1.25 20.00±3.54 38.13±3.75 37.45±0.07 43.75±4.33 42.50±3.54

Average

Dongfang

Average

12.51±1.74b 1.56±0.49a 15.10±4.47a 6.37±0.19ab 7.25±2.50b 14.51±5.45a 18.75±2.09a 37.90±2.93b 43.33±3.76a

25 1,000 2,409 2.41 Tropical bosks 10.36±4.65 10.36±4.65b 1.93±0.73 1.93±0.73 18.77±7.03 18.77±7.03a 6.73±0.37 6.73±0.37b 6.05±3.61 6.05±3.61ab 18.84±5.55 18.84±5.55a 26.38±1.96 26.38±1.96c 29.38±3.70 29.38±3.70a 44.25±2.09 44.25±2.09a

The values in the rows with different letters represent the statistically significant differences among the climate zones at 95 % level a

The climate parameters were cites from Jing (2006)

b

The vegetation types were cited from Mao et al. (2006)

aerating. The concentration and abundance of NH4+ and NO3− was also sampled at 0.5, 24, 72, and 120 h immediately after the gas sampling. For these measurements, three flasks from each treatment were extracted with 150 ml 2 M KCl solution by shaking for 1 h at 200 rpm at 25 °C. After shaking, the suspensions were passed through a filter paper (Qualitative Filter Paper, GB/T1914-2007) and the extracts were then stored at 4 °C for analyses of concentrations and 15N abundance of NH4+ and NO3− within 1 week. Headspace gas was also sampled at 0.5, 24, 72, and 120 h after the incubation for measuring N2O emission and 15N abundance of N2O. Before gas sampling, the flasks were connected to a multiport vacuum manifold and then flushed with ambient air for 30 s for three times and then sealed for 6 h. At the end of 6-h incubation, headspace gas was collected by a 20-ml syringe and then injected into two preevacuated vials with a volume of 18.5 ml for determining the concentration and 15N abundance of N2O. The N2O emission was measured in six flasks (three in 15 N-labeled NH4+ treatment and another three in 15N-labeled NO3− treatment) for all replicates. The concentration of the N2O was measured immediately after the gas sampling and the abundance of the N2O was measured within 1 month.

The contents of total N and C were determined using SerCon SL C/N Elemental Analyzer (Sercon, Crewe, UK). The concentrations of NH4+ and NO3− in KCl extraction were determined using a continuous flow analyzer (Skalar, Breda, The Netherlands). The concentration of N2O was analyzed by the Agilent 7890 gas chromatography (Agilent, USA) using a 2-mm ID stainless steel column, 3 m long, packed with Porapak Q (80/100 mesh), fitted with an electron capture detector (ECD) set at 300 °C. The column temperature was maintained at 65 °C and the carrier gas was an argon 5 % methane mixture at a flow rate of 40 ml min−1. 15N abundance in NH 4+, NO 3−, and N2O was analyzed by a Sercon SL Elemental Analyzer coupled to a 20–20 isotope ratio mass spectrometer (IRMS) (Sercon). The separation of NH4+ and NO3− in KCl solution was through distillation with magnesium oxide and Devarda’s alloy, followed the procedure described by Zhang et al. (2011).

2.3 Analyses

An aridity index (Di) was employed to compare the degree of drought of the sampled sites. The drought increased with increasing Di value. The Di value in the range of 0 to 1 was regarded as humid, in the range of 1 to 1.7 as semihumid, in

Soil pH was measured in deionized water (1:2.5 v:v ratio) using a DMP-2 mV/pH detector (Quark, Nanjing, China).

2.4 Calculations 2.4.1 Aridity index

J Soils Sediments

the range of 1.7 to 3 as semiarid climate, and >3 as arid climate (Meng et al. 2004) using the following equation: Di ¼

E0 p

ð1Þ

where E0 is the annual average of evaporation, and P is the annual average precipitation. 2.4.2 Net and gross nitrogen transformation rates Gross N transformation rates were simulated by a MCMC routine based on 15N tracing model by Müller et al. (2007), which was programmed in the software MatLab (version 7.7, The Math Works). Ten gross rates of N transformation were obtained by the model simulation (see Müller et al. 2007 for detail). According to Müller et al. (2004), the intial concentration of the mineral N pools were estimated by backing extrapolation of data at 0.5 h. The difference of the applied NH4+ and the measured NH4+ was attributed to the NH4+ adsorbed to the NH4+ exchange sites (NH4+ads). Net mineralization rate was calculated as the total mineral N production rates minus the total mineral N consumption rates, and net nitrification rate was calculated as the total nitrification rate minus the total NO3− consumption rate. The turnover rate of organic N was calculated by the total N divided by the gross N mineralization rate. The gross and net N transformation rates were expressed as mg N kg−1 day−1. 2.4.3 N2O emissions N2O fluxes were calculated as follows (Lang et al. 2011): F¼

ρ  ΔC  V  273 W  Δt  ð273 þ T Þ

ð2Þ

where F is the flux of N2O (μg N2O–N kg−1 h−1), ρ is the density of N2O under standard conditions (kg m−3), ΔC is the production of N2O during the 6 h cover period (ppb), V is the headspace volume of the flasks (m3), T is the incubation temperature (25 °C in the study), and Δt was the incubation time (6 h). W is the weight of soil on ovendry basis (kg). The mean N2O fluxes rates during the first 120-h incubation period were weighted by an interval of two adjacent measurements, and the cumulative N2O was calculated by mean N2O flux and a period of 120 h. In our research, the 15N enrichment of the N2O was between that of NH 4+ and NO 3− so that we use the method described by Stevens et al. (1997). In this model, N2O emissions were attributed to nitrification (Cn) and denitrification (C d ), and the relative contributions of

these two processes were calculated by the following equations: Cd ¼

ðaN 2O −aNH 4 Þ ðaN O3 ≠aNH4 Þ ðaN O3 −aN H4 Þ

C n ¼ 1−C d

ð3  1Þ ð3  2Þ

where aN2O is the 15N atom % enrichment of the N2O emitted, aNO3 is the enrichment of NO3− pool, and aNH4+ is the enrichment of NH4+ pool. The relative contributions of N2O emission were obtained from the NH415NO3 treatment (Stevens et al. 1997). Then the rates of N2O emission from NH4+ nitrification (N 2 O n ) and denitrification (N 2 O d ) were calculated by Eqs. (4-1) and (4-2): N 2 Od ¼ C d  N 2 OT

ð4  1Þ

N 2 On ¼ C n  N 2 OT

ð4  2Þ

where N2OT was the average N2O emission rate from soil during the first 120-h incubation period. The ratio of N2O (PN2O) to the products of NH4+ nitrification was calculated by the equation as follows: PN 2O ¼

N 2 On ON H4

ð5Þ

where ONH4 is the gross rate of nitrification of NH4+. 2.5 Statistical analyses The means over climate zones were used in all analyses in order to avoid pseudoreplications and the differences in soil properties, and N transformation rates between the humid, semihumid, and semiarid zones were performed by one-way ANOVA analysis of multiple comparisons (LSD). Pearson correlation analyses with two-tailed significance tests were performed at the individual soil level. All analyses were performed using SPSS 17.0 software for windows. Statistically significant differences were all set at P<0.05.

3 Results 3.1 Soil properties Even though the soil samples were all collected randomly on the days without rain, the moisture of collected soils significantly and exponentially decreased with increasing aridity index (Di) (Fig. 2, P<0.01). The soil pH varied significantly from 4.1 to 7.2 and correlated significantly with both Di and soil moisture (Fig. 3), indicating that aridity was a key factor determining soil pH. There were no significant differences in averaged soil organic C and N contents between the humid,

J Soils Sediments

Fig. 2 Relationship between soil moisture and aridity index across the aridity gradient in the Hainan Island

semihumid, and semiarid soils (P>0.05 for both, Table 1), although there were large variations within the same climate zones. 3.2 Gross and net rates of N transformation along the humid–semiarid gradient The changes in NH4+ and NO3− content and their 15 N atom % excess in each soil during the incubation are presented in Table S1. Ten gross rates of N transformation in soil could be successfully optimized by the N tracing model (Müller et al. 2007). However, we focused on the N transformation Fig. 3 Relationship between pH and soil moisture (a) or aridity index (b) across the aridity gradient in the Hainan Island

related to mineral N content and nitrification in the investigation. The sum of gross rates of labile and recalcitrant organic N mineralization (M) were, on average, 5.40 (±0.45), 3.57 (±0.53), and 2.16 (±0.29) mg kg−1 day−1 in the semiarid, semihumid, and humid soils, respectively. The differences were significant between the climate zones (P < 0.05, Table 2). The gross rate of organic N mineralization was positively and significantly correlated with soil pH (r=0.66, P<0.01) and aridity index (DI) (r=0.81, P<0.01), but correlated insignificantly with total N (r=0.25, P>0.05) and soil organic carbon (r=0.25, P>0.05, Table 3), respectively. The turnover rate of organic N, calculated by dividing soil organic N by the gross rate of organic N mineralization, was much faster in semiarid soils (347 days on average) than in semihumid soils (472 days) and in humid soils (1,027 days), and the differences were statistically significant (P<0.05). The turnover rate of organic N was negatively and significantly correlated with soil pH (Table 3, r=−0.88, P<0.01) and Di (r=−0.84, P<0.01) and positively and significantly correlated with soil moisture measured at sampling (r=0.80, P<0.01, Table 3), respectively. There was no significant change of net mineralization rates along the aridity gradient, although the gross rate of organic N mineralization and turnover rate of organic N were significantly faster in semiarid soils than in humid and semihumid soils (Table 2). The gross rate of NH4+ nitrification (ONH4) increased from 2.81 mg N kg−1 day−1 in humid soils to 6.25 mg N kg−1 day−1 in the semiarid soils, and the rate was significantly larger in the semiarid soils than those in the humid and semihumid soils (P<0.05, Table 2). ONH4 was positively and significantly correlated with clay content (r=0.65, P<0.01), soil pH (r=0.68,

J Soils Sediments Table 2 Gross and net nitrogen transformation rates in soils along the aridity gradient in the Hainan Island, China (values in means± standard deviation)

Code

Humid

M (mg N kg−1 day−1) Mnlab (mg N kg−1 day−1) Mnrec (mg N kg−1 day−1) INH4 (mg N kg−1 day−1) INH4−Nrec (mg N kg−1 day−1)

2.16±0.29a 1.70±0.64a 0.47±0.14a 0.52±0.16a 0.30±0.19a

INH4−Nlab (mg N kg−1 day−1) NT (mg N kg−1 day−1) ONH4 (mg N kg−1 day−1) ONrec (mg N kg−1 day−1) DNO3 (mg N kg−1 day−1) INO3 (mg N kg−1 day−1) RNH4a (mg N kg−1 day−1) NH4+ ads (mg N kg−1 day−1) NetM (mg N kg−1 day−1) NetN (mg N kg−1 day−1) Torg (day)

0.22±0.18a 4.13±0.82a 2.81±0.42a 1.32±0.50ab 0.07±0.00a 0.07±0.03a 0.19±0.05a 0.63±0.25a 3.34±0.63a 3.98±1.82a 1,027±174a

Semihumid

Semiarid

3.56±0.53b 2.72±0.1.61a 0.84±0.30a 1.62±0.40b 0.97±73b

5.40±0.45c 4.92±1.24b 0.47±0.38a 0.46±0.26a 0.33±0.51ab

0.65±0.37b 3.77±0.76a 3.55±0.72a 0.22±0.12a 0.03±0.02a 0.51±0.05a 0.44±0.20a 1.40±0.59a 1.78±2.39a 4.27±1.64b 472±147b

0.14±0.14a 8.30±0.79b 6.25±0.51b 2.05±0.82b 0.10±0.05a 1.52±1.31a 0.22±0.05a 3.85±1.09b 1.49±4.84a 7.83±2.81c 347±65c

The values in the rows with different letters (a, b, and c) represent the statistically significant differences among the climate zones at the 95 % level M total mineralization rates, Mnlab mineralization of labile organic N, Mnrec mineralization of reluctant organic N, INH4+ microbial immobilization of NH4+ , INH4−Nrec immobilization of NH4+ to the reluctant organic N pool, INH4−Nlab immobilization of NH4+ to the labile organic N pool, NT total nitrification rates, ONH4 oxidation of NH4+ to NO3− , Nhete oxidation of organic N to NO3− , DNO3 dissimilatory NO3− to reduction to NH4+ , INO3− microbial immobilization of NO3− , RNH4a to the release of adsorbed NH4+ , NH4+ ads adsorption of NH4+ , NetM net mineralization rate; NetN net nitrification rate; Torg turnover rate of organic N

observed between the measured soil properties and the gross rate of organic N oxidation to NO3− (Table 3). Similar to the change of gross rates of NH4+ nitrification with aridity gradient, the net rates of nitrification also increased significantly from humid soils (3.98±1.82 mg N kg−1 day−1), to semihumid soils (4.27±1.64 mg N kg−1 day−1), and further to semiarid soils (7.83±2.81 mg N kg−1 day−1) (P<0.05, Table 2). In contrast to the gross rate of NH4+ nitrification, no significant correlation was observed between the net rate of NH4+ nitrification and soil pH (r=0.42, P>0.05).

P<0.01), and DI (r=0.71, P<0.01), but negatively with soil moisture (r=−0.53, P<0.05, Table 3), respectively. The organic N oxidation to NO3− through heterotrophic nitrification (ONrec) took place as well, and the rate was significantly higher in semiarid soils (2.05±0.82 mg N kg−1 day−1) than in semihumid soils (0.22±0.12 mg N kg−1 day−1) (P<0.05, Table 2). The differences in the gross rate of oxidation of organic N to NO3− were not significant between the humid soils and semihumid soils and between the semihumid soils and semiarid soils (P>0.05, Table 2). There were no significant correlations Table 3 Pearson correlations between soil properties and N transformation in the secondary forest soils along the aridity gradient in the Hainan Island (n=18)

N transformation

Clay

Silt

Sand

TN

SOC

pH

Di

Moisture

M INH4 NT ONH4 ONrec RNH4a NH4+ads Torg

0.40 −0.55 0.60** 0.65** 0.26 −0.28 0.25 −0.17

−0.02 0.35 −0.20 −0.19 −0.12 0.44 0.06 −0.13

−0.37 0.05 −0.31 −0.37 −0.09 −0.34 −0.32 0.35

0.25 −0.12 0.19 −0.01 0.37 0.13 0.36 0.36

0.25 −0.13 0.16 −0.05 0.36 0.10 0.37 0.37

0.66** 0.10 0.58* 0.68** 0.17 0.13 0.59** −0.88**

0.81** −0.03 0.63** 0.71** 0.22 0.06 0.64** −0.84**

−0.44 −0.07 −0.35 −0.53* 0.06 −0.14 −0.33 0.80**

0.02 0.45

0.38 0.42

0.12 −0.14

−0.04 0.00

−0.06 −0.01

−0.32 0.42

−0.11 0.47

0.10 −0.35

NetM NetN

TN soil total N, SOC soil organic C; Di aridity index *

Significance level was at 0.05; ** Significance level was at 0.01

J Soils Sediments

3.3 N2O emissions On average, the N2O emission rate during the first 120-h incubation period was 24.01±9.91 μg N2O–N kg−1 day−1 from the humid soils, 12.47±1.12 μg N2O–N kg−1 day−1 from the semihumid soils, and 17.69±3.41 μg N2O–N kg−1 day−1 from the semiarid soils, but the differences were not statistically significant (P>0.05, Table 4). The relative contributions of NH4+ pool through nitrification and NO3− pool through denitrification to N2O emissions were calculated by Eqs. (3-1) and (3-2) (Table 4). Under the conditions of aerobic incubation (60 % WHC), the relative contributions of nitrification to total N2O emissions were, on average, 55, 57, and 69 % in the humid, semihumid, and semiarid soils, respectively. Correspondingly, the relative contribution of denitrification to N2O emissions decreased from the humid soils (45 %) to the semihumid soils (42 %) and further to the semiarid soils (31 %). But the differences were not statistically significant (P>0.05, Table 4). The ratio of N2O emission rate from the NH4+ nitrification to the gross rate of NH4+ nitrification ranged from 0.01 to 0.62 % and was correlated significantly with soil pH (Fig. 4). In the current version of the 15N tracing model, the rate of denitrification is not included in the NO3− immobilization rate. Therefore, the ratio of N2O emission rate to total N2 could not be calculated in the current study.

4 Discussion Climate affects N transformation in soils both instantaneously and persistently. The instantaneous effects on N transformation in soils, which are observed immediately after altering soil moisture and temperature, have been investigated intensively and introduced into many N cycling models (Cheng Table 4 Time-weighted average rates of N2O emission (N2OT), contribution percentages of nitrification (Cn) and denitrification (Cd) to N2O emission rates, N2O emission rate from nitrification (N2Oa) and denitrification (N2Od), and the ratio of N2O in the nitrification products (PN2O) during the 120-h incubation under 60 % WHC and 25 °C N2O

Climate zone Humid

Semihumid

Semiarid

N2OT (μg kg−1 day−1) Cd (%)

24.01±9.91a 45.19±7.41a

12.47±1.12a 42.65±8.29a

17.69±3.41a 30.75±14.49a

Cn (%) N2Od (μg kg−1 day−1) N2Oa (μg kg−1 day−1) PN2O (%)

54.81±0.07a 14.16±8.61a 9.84±1.60a 0.38±0.05a

57.35±0.04a 5.45±1.49a 7.02±1.16a 0.21±0.04a

69.25±0.12a 4.57±1.69a 13.12±3.65a 0.22±0.05a

The value in the rows with the same letter denotes values that are not significantly different from each other

Fig. 4 Relationship between pH and the ratio of N2O to the nitrification rate in the soils along the aridity gradient in the Hainan Island

et al. 2014; Grant 2014; Lan et al. 2014). The persistent effects on N transformation in soils refer to those regulated by the inherent soil properties. Climate is one of the most important factors determining soil properties and microbial community, which determine the intrinsic characteristics of N transformation in soils (Zhang et al. 2013b). The effects of climate on N transformation through regulating soil properties are also named as the effects of climate-induced soil properties in the current study. The soils distributed along the aridity gradient in the Hainan Island provide an excellent object for investigating persistent effects of aridity on N transformation. Our results showed that the gross rates of many N transformation processes varied significantly along the aridity gradient in the Hainan Island (Table 2). It has been demonstrated that there are many inherent soil properties engaged in the regulation of N transformation in soils, such as soil pH, texture, and microbial community (Gallo et al. 2013; Zhang et al. 2013b). The correlation analysis suggested clearly that some soil properties, such as pH, which was affected by aridity (Fig. 3), was a crucial factor determining the gross rates of organic N mineralization, turnover rate of organic N, and nitrification in the soils (Table 3). However, other inherent soil properties that are affected by climate are expected to affect N transformation in soils to a certain extent as well, For instance, a significant correlation was also observed between clay content and gross NH4+ nitrification rate in the study (r=0.65, P<0.01, Table 3). Therefore, the persistent effects of aridity on N transformation are influenced by soil properties comprehensively, which regulate soil microbial community (Fierer and Jackson 2006). Other interesting phenomena observed in the investigation were that although the gross rates of N transformation varied significantly from the humid soils to the semiarid soils in the aridity gradient, the net rates of mineralization and N2O

J Soils Sediments

emission did not vary significantly in the soils along the gradient (Tables 2 and 4). These phenomena imply that the inherent soil properties, mainly determined by aridity, affect the gross rates but are not necessary to affect the net rates of N transformation and N2O emissions under the same incubation conditions. This can, for instance, be explained by simultaneous changes of production and consumption processes for one soil N pool in question. 4.1 N transformation processes regulating the net mineralization rate The gross N transformation rates between different climate zones may be confounded by soil parent material, geographic position, vegetation, and some other factors (Laverman et al. 2000; Kooijman and Martinez-Hernandez 2009; Tokuchi et al. 2014; Zeng et al. 2014). Therefore, we limited the vegetation type to the secondary forest, constrained the soil parent material to granite, and bound the geographic location to top of the hill (200–300 m). The results showed that the gross mineralization rates and the turnover time of organic N in semiarid soils were significantly higher than in the humid soils incubated under the same conditions of the temperature and moisture (Table 2), being in high agreement with previous studies (Vanderbilt et al. 2008; Bontti et al. 2009; Austin 2011). However, our results further showed that the net rate of organic N mineralization did not vary significantly along the aridity gradient under the incubation conditions (Table 2). Because the net organic N mineralization rate is the net product of inorganic N production and consumption, the increased production in the semiarid soils must be balanced by the increased inorganic N consumption. To get insight into the gross rates of inorganic consumption, we observed that the rates of NH4+ and NO3− immobilization were not significantly different between the humid soils and semiarid soils. However, the NH4+ adsorption rates, which were calculated by the change of applied NH4+ and the measured NH4+ after N applied at 0.5 h, accounted for 80 % of the gross mineralization rates in the semiarid soils and were significantly higher than those in humid soils (Table 2). The correlation analysis showed that NH4+ adsorption rate was correlated significantly with soil pH (r=0.59, P<0.01) and Di (r=0.64, P<0.01), but not significantly with clay content (r=0.25, P>0.05, Table 3), respectively. No significant correlation with clay content suggested that the high NH4+ adsorption rates in the semiarid soils may be related to the composition of the clay minerals. For soils in the humid zones, the predominant clay mineral is kaolinite, which belongs to 1:1 clay minerals. However, the soils in the semiarid zones were mainly dominated by 2:1 clay minerals (Chang 1963). The NH4+ can be fixed by 2:1 clay minerals and becomes nonexchangeable, but not by 1:1 clay minerals (Zhu 1998). Fixation of NH4+ by soil clay

can act as an effective buffer for the supply and retention of NH4+ (Huygens et al. 2007). It is typical for soil moisture in the field to vary along an aridity gradient, such as the one shown in Fig. 2. Previous studies have suggested that temperature and precipitation are the critical factors affecting soil N cycling in different climates (Schütt et al. 2014). The instantaneous effects of soil moisture and temperature on gross N mineralization rates were documented for forest soils (e.g., Cheng et al. 2014) and paddy soils (e.g., Lan et al. 2014). Because the incubation was carried out after adjusting soil moisture to the same level (60 % WHC), the effects of soil moisture in the field along the aridity gradient on N mineralization were minimized in the present study. Therefore, the present investigation could not judge whether net mineralization rate were significantly different in field along the aridity gradient, and in situ investigations are essential to understand the persistent effects of climateinduced soil properties on N transformation in soils. 4.2 Limited effect of climate-induced soil properties on N2O emissions There have been numerous investigations carried out on the effects of soil moisture and temperature on N2O emissions from soils. Across China, N2O emissions from both croplands and noncroplands increased linearly with the increase in the annual precipitation (Lu et al. 2006; Cai 2012). The instantaneous effects of altering soil moisture on N2O emissions from soils have been well documented (e.g., Cheng et al. 2014). García-Méndez et al. (1991) found that variation of soil water through the season was the primary factor controlling N2O emissions from two forests and one pasture in tropical deciduous forest near Chamela in Mexico. However, it is not clear how the climate-induced soil properties regulate the N2O emissions. Our results showed that there were no significant differences in N2O emission rates between the humid soils (24.01 ± 9.91 μg N 2 O–N kg−1 day −1 ), semihumid soils (12.47 ±1.12 μg N2O–N kg−1 day−1), and semiarid soils (17.69±3.41 μg N2O–N kg−1 day−1) when incubated under the same soil moisture and temperature conditions, although the relative contributions of nitrification and denitrification to N2O emissions varied (but not significantly) along the aridity gradient (Table 4). This result clearly indicated that if the difference in soil moisture in field along the aridity gradient was not taken into account, instantaneous rather than persistent effects of aridity regulated N2O emissions from the soils, although the gross and net rates of NH4+ nitrification increased significantly from the humid soils to the semiarid soils (Table 2). Calculations based on the relationship between the 15N enrichments in N2O and NH4+ and NO3− (Stevens et al. 1997) showed that nitrification was the dominant N2O-producing process in the soils under incubation conditions. This

J Soils Sediments

is understandable because the incubation was conducted under aerobic conditions (60 % WHC). However, the results showed that a substantial contribution was made by denitrification to N2O emission, accounting for 45 % in the humid soils to 30 % in the semiarid soils (Table 4), which is consistent with previous reports that denitrification contributed 52 to 56 % of N2O emission from the subtropical acid forest soils under aerobic incubation (Zhang et al. 2011). The occurrence of denitrification under aerobic condition is usually attributed to the existence of anaerobic microsites created by microbial activities (Renault and Stengel 1994) or aerobic denitrification (Müller et al. 2002; Wolf and Brumme 2002). In line with previous observations (Ambus et al. 2006; Baggs et al. 2010), the relative contribution of denitrification to N2O emissions decreased with increasing soil pH, although the differences were not significant (Table 4). In this study, we did not know whether the gross rate of denitrification varied with soil pH. However, it has been well documented that the ratio of N2O/N2 in denitrification products increased with decreasing soil pH, because the nitrous oxide reductase is inhibited to a certain extent at low soil pH (Tiedje et al. 1979; Skiba et al. 1993; Bakken et al. 2012). Therefore, one possible way that the contribution of denitrification to N2O emission decreased from the humid soils to the semiarid soils was the increase in soil pH, which decreased the N2O/N2 ratio in the denitrification products along the aridity gradient. In contrast, the N2O ratio in nitrification products increased with decreasing soil pH (r2 =0.35, P<0.01, Fig. 4). Therefore, although the gross rate of nitrification increased with the increase in soil pH (r= 0.68, P<0.01, Table 3), the contribution of nitrification did not increase correspondingly. These balanced relationships would explain that N2O emissions from the soil did not vary significantly along with the aridity gradient.

5 Conclusions The investigation conducted in the study showed that the gross rates of N transformation in soils varied greatly along with the aridity gradient in the Hainan Island. Soil pH, determined mainly by aridity, was a crucial factor regulating N transformation. The gross rate of mineralization and the turnover rate of organic N increased significantly from the humid soils to the semiarid soils. However, the net rate of mineralization did not vary significantly along the aridity gradient. Ammonium fixation by 2:1 clay minerals in the semiarid soils might buffer the mineralized inorganic N under the incubation conditions. N2O emissions did not vary significantly from the humid soils to the semiarid soils with the variations of gross and net NH4+ nitrification along the aridity gradient. Balanced relationships between the gross rates of NH4+ nitrification and the ratio of N2O in the products responding to the change in soil pH are likely the main explanation for why N 2 O

emissions did not vary with the significant variations of gross and net nitrification rates among the climate regions. Instantaneous effects of climate, rather than the effect of climate-induced soil properties, play an important role in regulating N2O emissions from soils. Acknowledgments This work was supported by grants from the National Natural Science Foundation of China (41330744), the B973^ project (2014CB953800), the Natural Science Foundation of Jiangsu Province (BK20140062), the Qing Lan Project (184080H102142), the Outstanding Innovation Team in Colleges and Universities in Jiangsu Province, and the Priority Academic Program Development of Jiangsu Higher Education Institutions (PAPD, 164320H116).

References Ambus P, Zechmeister-Boltenstern S, Butterbach-Bahl K (2006) Sources of nitrous oxide emitted from European forest soils. Biogeosciences 3:135–145 Austin AT (2011) Has water limited our imagination for aridland biogeochemistry? Trends Ecol Evol 26:229–235 Baggs EM, Smales CL, Bateman EJ (2010) Changing pH shifts the microbial sourceas well as the magnitude of N2O emission from soil. Biol Fertil Soils 46:793–805 Bakken LR, Bergaust L, Liu B, Frostegård Å (2012) Regulation of denitrification at the cellular level: a clue to the understanding of N2O emissions from soils. Philos Trans R Soc Lond B Biol Sci 367: 1226–1234 Bertiller M, Mazzarino M, Carrera A, Diehl P, Satti P, Gobbi M, Sain C (2006) Leaf strategies and soil N across a regional humidity gradient in Patagonia. Oecologia 148:612–624 Bontti EE, Decant JP, Munson SM et al (2009) Litter decomposition in grasslands of central North America (US Great Plains). Glob Chang Biol 15:1356–1363 Cai Z (2012) Greenhouse gas budget for terrestrial ecosystems in China. Sci China Earth Sci 55:173–182 Chang S (1963) Clay minerals of tropical soils of Hainnan Island (in Chinese). Acta Pedol Sin 11:36–52 Cheng Y, Wang J, Wang S, Zhang J, Cai Z (2014) Effects of soil moisture on gross N transformations and N2O emission in acid subtropical forest soils. Biol Fertil Soils 50:1099–1108 Ciais P, Sabine C, Bala G, Bopp L, Brovkin V, Canadell J, Chhabra A, DeFries R, Gralloway J, Heimann M, Jones C, Le Quéré C, Myneni RB, Piao S, Thornto P (2013) Carbon and other biogeochemical cycles. In: Stocker T, Qin D, Plattner GK, Tignor M, Allen S, Boschung J, Nauels A, Xia Y, Bex V, Midgley P (eds) Climate change 2013: the physical science basis. Contribution of working group I to the fifth assessment report of the intergovernmental panel on climate change, Cambridge University Press, Cambrdige, pp 465–570 FAO (1988) FAO-UNESCO Soil Map of the World, Revised Legend (with corrections and updates), ISRIC, Wageningen Fierer N, Jackson RB (2006) The diversity and biogeography of soil bacterial communities. Proc Natl Acad Sci, USA 103:626–631 Gallo EL, Lohse KA, Ferlin CM, Meixner T, Brooks PD (2013) Physical and biological controls on trace gas fluxes in semi-arid urban ephemeral waterways. Biogeochemistry 121:189–207 García-Méndez G, Maass JM, Matson P, Vitousek P (1991) Nitrogen transformations and nitrous oxide flux in a tropical deciduous forest in México. Oecologia 88:362–366

J Soils Sediments Grant R (2014) Nitrogen mineralization drives the response of forest productivity to soil warming: modelling in ecosys vs measurements from the Harvard soil heating experiment. Ecol Model 288:38–46 Hseung Y (1980) General aspects of the soils in China. Report p 11 Huygens D, Rütting T, Boeckx P, Van Cleemput O, Godoy R, Müller C (2007) Soil nitrogen conservation mechanisms in a pristine south Chilean Nothofagus forest ecosystem. Soil Biol Biochem 39:2448– 2458 Jing Y (2006) Atlas of Hainan province. Cartographic Publishing House, Guangdong Kooijman AM, Martinez-Hernandez GB (2009) Effects of litter quality and parent material on organic matter characteristics and Ndynamics in Luxembourg beech and hornbeam forests. For Ecol Manag 257:1732–1739 Lan T, Han Y, Roelcke M, Nieder R, Cai Z (2014) Temperature dependence of gross N transformation rates in two Chinese paddy soils under aerobic condition. Biol Fertil Soils 50:949–959 Lang M, Cai Z, Chang S (2011) Effects of land use type and incubation temperature on greenhouse gas emissions from Chinese and Canadian soils. J Soils Sediments 11:15–24 Laverman AM, Zoomer HR, Van Verseveld HW, Verhoef HA (2000) Temporal and spatial variation of nitrogen transformations in a coniferous forest soil. Soil Biol Biochem 32:1661–1670 Liu X, Qi Y, Dong Y, Peng Q, He Y, Sun L, Jia J, Cai C (2014) Response of soil N2O emissions to precipitation pulses under different nitrogen availabilities in a semiarid temperate steppe of Inner Mongolia. China J Arid Land 6:410–422 Lu Y, Huang Y, Zou J, Zheng X (2006) An inventory of N2O emissions from agriculture in China using precipitation-rectified emission factor and background emission. Chemosphere 65:1915–1924 Mao L, Zhang Y, Bi H (2006) Modern pollen deposits in coastal mangrove swamps from northern Hainan Island, China. J Coast Res 22: 1423–1436 McLain JET, Martens DA (2006) N2O production by heterotrophic N transformations in a semiarid soil. Appl Soil Ecol 32:253–263 Meng M, Ni J, Zhang Z (2004) Aridity index and its applications in geoecological study (in Chinese). Acta Phytoecol Sin 28:853–861 Mosier AR, Duxbury JM, Freney JR, Heinemeyer O, Minami K (1998) Assessing and mitigating N2O emissions from agricultural soils. Clim Chang 40:7–38 Müller C, Martin M, Stevens RJ, Laughlin RJ, Kammann C, Ottow CG, Jäger HJ (2002) Processes leading to N2O emissions in grassland soil during freezing and thawing. Soil Biol Biochem 34:1325–1331 Müller C, Stevens RJ, Laughlin RJ, Jäger HJ (2004) Microbial processes and the site of N2O production in a temperate grassland soil. Soil Biol Biochem 36:453–461 Müller C, Rütting T, Kattge J, Laughlin RJ, Stevens RJ (2007) Estimation of parameters in complex 15N tracing models by Monte Carlo sampling. Soil Biol Biochem 39:715–726 Müller C, Laughli RJ, Spott O, Rütting T (2014) Quantification of N2O emission pathways via a 15N tracing model. Soil Biol Biochem 72: 44–54 Peterjohn W, Schlesinger W (1990) Nitrogen loss from deserts in the southwestern United States. Biogeochemistry 10:67–79

Renault P, Stengel P (1994) Modeling oxygen diffusion in aggregated soils: I. Anaerobiosis inside the aggregates. Soil Sci Soc Am J 58: 1017–1023 Schütt M, Borken W, Spott O, Stange CF, Matzner E (2014) Temperature sensitivity of C and N mineralization in temperate forest soils at low temperatures. Soil Biol Biochem 69:320–327 Skiba U, Smith KA, Fowler D (1993) Nitrification and denitrification as sources of nitric oxide and nitrous oxide in a sandy loam soil. Soil Biol Biochem 25:1527–1536 Stange F, Butterbach-Bahl K, Papen H, Zechneister-Boltenstrern S, Li CS, Aber J (2000) A process-oriented model of N2O and NO emissions from forest soils. Sensitivity analysis and validation. J Geophys Res 105:4385–4398 Stevens RJ, Laughlin RJ, Burns LC, Arah JRM, Hood RC (1997) Measuring the contributions of nitrification and denitrification to the flux of nitrous oxide from soil. Soil Biol Biochem 29:139–151 Tiedje J, Firestone R, Firestone M, Betlach M, Smith M, Caskey W (1979) Methods for the production and use of nitrogen-13 in studies of denitrification. Soil Sci Soc Am J 43:709–716 Tokuchi N, Yoneda S, Ohte N et al (2014) Seasonal changes and controlling factors of gross N transformation in an evergreen plantation forest in central Japan. J For Res 19:77–85 Vanderbilt KL, White CS, Hopkins O, Craig JA (2008) Aboveground decomposition in arid environments: results of a long-term study in central New Mexico. J Arid Environ 72:696–709 Vitousek PM, Reiners WA (1975) Ecosystem succession and nutrient retention: a hypothesis. Bioscience 25:376–381 Wang S, Zhou G, Gao S, Guo J (2005) Gradient distribution of soil nitrogen and its response to climate change along the Northeast China Transect (in Chinese with English abstract). J Appl Ecol 16: 279–283 Wolf I, Brumme R (2002) Contribution of nitrification and denitrification sources for seasonal N2O emissions in an acid German forest soil. Soil Biol Biochem 34:741–744 Zaady E (2005) Seasonal change and nitrogen cycling in a patchy Negev Desert: a review. Arid Land Res Manag 19:111–124 Zaady E, Groffman PM, Shachak M (1996) Litter as a regulator of N and C dynamics in macrophytic patches in Negev desert soils. Soil Biol Biochem 28:39–46 Zeng Y, Xiang W, Deng X, Fang X, Liu C, Peng C (2014) Soil N forms and gross transformation rates in Chinese subtropical forests dominated by different tree species. Plant Soil 384:231–242 Zhang J, Cai Z, Zhu T (2011) N2O production pathways in the subtropical acid forest soils in China. Environ Res 111:643–649 Zhang J, Zhu T, Meng T, Zhang Y, Yang J, Yang W et al (2013a) Agricultural land use affects nitrate production and conservation in humid subtropical soils in China. Soil Biol Biochem 62: 107–114 Zhang J, Cai Z, Zhu T, Yang W, Müller C (2013b) Mechanisms for the retention of inorganic N in acidic forest soils of southern China. Sci Rep 3:2342 Zhu Z (1998) The status, problems and countermeasures of nitrogen fertilizer application in China fertilizer issues of sustainable agriculture development in China (in Chinese). Jiangsu Sci Technol Publ, Nanjing, pp 28–51