Forest Ecology and Management 337 (2015) 126–134

Contents lists available at ScienceDirect

Forest Ecology and Management

journal homepage: www.elsevier .com/ locate/ foreco

Understory management and fertilization affected soil greenhouse gasemissions and labile organic carbon pools in a Chinese chestnutplantation

http://dx.doi.org/10.1016/j.foreco.2014.11.0040378-1127/� 2014 Published by Elsevier B.V.

⇑ Corresponding author. Tel./fax: +86 571 637 40889.E-mail address: yongfuli@zafu.edu.cn (Y. Li).

Jiaojiao Zhang a,b, Yongfu Li a,⇑, Scott X. Chang c, Hua Qin a, Shenglei Fu d, Peikun Jiang a

a Zhejiang Provincial Key Laboratory of Carbon Cycling in Forest Ecosystems and Carbon Sequestration, Zhejiang A & F University, Lin’an 311300, Chinab Jiangsu Yanjiang Institute of Agricultural Sciences, Nantong 226541, Chinac Department of Renewable Resources, University of Alberta, 442 Earth Sciences Building, Edmonton, AB T6G 2E3, Canadad Key Laboratory of Vegetation Restoration and Management of Degraded Ecosystems, South China Botanical Garden, Chinese Academy of Sciences, Xingke Road 723,Tianhe District, Guangzhou 510650, China

a r t i c l e i n f o

Article history:Received 5 August 2014Received in revised form 9 November 2014Accepted 11 November 2014

Keywords:Castanea mollissimaFertilizationGreenhouse gas (GHG)Labile organic C poolsUnderstory replacement

a b s t r a c t

Management practices markedly impact carbon (C) cycling in forest plantations. However, the interactiveeffects of understory management and fertilization on soil greenhouse gas (GHG) fluxes and labileorganic C pools remain unclear in forest plantations. To investigate the effects of understory replacement,fertilization, and their interaction on soil GHG fluxes and labile organic C pools in a Chinese chestnutplantation, we conducted a 2 � 2 factorial experiment over a 12-month field study with four treatments:Control (understory removed without understory replacement or fertilization), understory replacement(understory removed and seeded with Medicago sativa L., MS), fertilization (F), and MS + F. The GHGfluxes were determined using a static chamber/GC technique. The seasonal pattern of GHG fluxes didnot change in any of the treatments in this one-year study; however, soil GHG fluxes, total global warm-ing potential (GWP) of GHG fluxes, and soil organic C (SOC), water soluble organic C (WSOC), microbialbiomass C (MBC), and NO3

�–N concentrations were significantly affected by MS, F, and their interaction. Inaddition, GHG fluxes, GWP, and SOC, WSOC, MBC and NO3

�–N concentrations were markedly increased byfertilization, regardless of the understory replacement treatment (P < 0.05), but they were increased byunderstory replacement only in the fertilized plots. The GHG fluxes were correlated with soil temperatureand WSOC in all plots (P < 0.05), but not with soil moisture and MBC. These findings suggest that under-story replacement likely is the optimum management technique for reducing/minimizing GHG fluxes,while F can enhance the effects of MS on increasing soil organic C and nutrient availability. We concludethat a combination of a moderate rate of fertilization and understory replacement with legume speciesshould be adopted to increase soil C sequestration, maintain soil fertility and sustainably developchestnut plantations.

� 2014 Published by Elsevier B.V.

1. Introduction

Anthropogenic emissions of greenhouse gases (GHG), particu-larly carbon (C) dioxide (CO2), methane (CH4) and nitrous oxide(N2O), have increasingly become major contributors to global cli-mate change (Manne and Richels, 2001; Jassal et al., 2011; Lavoieet al., 2013). Although CO2 has been considered the most importantGHG, CH4 and N2O also play major roles in global warming becausetheir global warming potential (GWP) are 25 and 298 times greater,respectively, than that of CO2 over a time span of 100 years (Forster

et al., 2007). In most forest soils, CO2, CH4 and N2O are formedthrough soil respiration, organic matter decomposition, and nitrifi-cation and denitrification, and these processes are strongly influ-enced by environmental factors and forest management practices(Tang et al., 2006; Wang et al., 2010, 2013; Lavoie et al., 2013).Forest management practices can greatly influence how forests,especially forest plantations, mitigate climate change throughlong-term C sequestration (Peng et al., 2008; Song et al., 2013).

China has the largest plantation area in the world, coveringmore than 62 million ha and accounting for approximately one-third of the world’s total plantation area (Piao et al., 2009; Chenet al., 2011). Historically, forest management practices, such asunderstory vegetation management and fertilization, have been

J. Zhang et al. / Forest Ecology and Management 337 (2015) 126–134 127

used to increase forest productivity and enhance ecosystem C stor-age (Peng et al., 2008; Powers et al., 2013). In addition, these prac-tices can markedly affect soil physical, chemical, and biologicalproperties in plantations (Xiong et al., 2008; Wu et al., 2011b;Krause et al., 2013; Zhao et al., 2013), and consequently affect soilGHG fluxes and organic C pools (Wang et al., 2011; Zhang et al.,2014)

Fertilization, especially the application of chemical fertilizers, isone of the most important and commonly used methods for plan-tation management and has a significant effect on soil GHG emis-sions in forest plantations (Mo et al., 2008; Deng et al., 2010; Tuet al., 2013; Gao et al., 2014). Nitrogen (N) fertilizer applicationshave been reported to increase primary production in most terres-trial ecosystems across the world (Elser et al., 2007), but theireffects on soil GHG emissions were not fully understood or contra-dictory. Several previous studies have reported that N applicationincreased soil GHG emissions in tropical or subtropical plantations(Zhang et al., 2008a,b; Zhang et al., 2011; Tu et al., 2013; Gao et al.,2014). However, other studies found that N addition had no effector negative effects on GHG emissions in plantations (Janssens et al.,2010; Jassal et al., 2010; Krause et al., 2013). For example, Mo et al.(2007, 2008) found that N addition significantly decreased CO2

emissions in a mature forest but had no effect in a disturbed forestin southern China. In a Pacific Northwest Douglas-fir forest, Napplication increased N2O emissions in the first year but therewas a small N2O uptake in the second year (Jassal et al., 2010). Inaddition, Krause et al. (2013) found that long-term N additionincreased N2O and CH4 emissions, but did not significantly changeCO2 emissions in a mountain spruce forest. These contradictoryresults may be attributed to the differences in the initial C and Nstatus, the microbial community composition, the fertilizer typeand application rate (Mo et al., 2008; Zhang et al., 2008a; Iqbalet al., 2009; Janssens et al., 2010; Liu et al., 2011).

As a component of forest ecosystems, the understory vegetationplays a vital role in maintaining the status of biota as well as bio-diversity, microclimate and ecosystem nutrient cycling bothabove- and belowground (Wang et al., 2013, 2014; Zhao et al.,2013; Zhang et al., 2014). Commonly, understory vegetationmanagement is one of the most important measures of plantationmanagement (Powers et al., 2013; Zhao et al., 2013). Understoryvegetation removal decreases the surface coverage and leads tomore sunlight reaching the forest floor, and consequently increasessoil temperature and causes changes in soil nutrient availabilityand microbial activity (Zhao et al., 2011, 2013; Wu et al., 2011a;Wang et al., 2014). In addition, understory species can, to someextent, reduce tree productivity in plantations, due to their compe-titions for N and other resources with forest trees (Zhao et al.,2011; Wang et al., 2014). The ecological function of understoryvegetation and the effect of understory management on soil prop-erties and microbial communities have been studied in forest plan-tations (Zhao et al., 2012, 2013; Wang et al., 2013, 2014; Zhanget al., 2014), but the effects of understory management (e.g.,removal or addition) on soil GHG emissions in plantations areinconsistent among previous studies (Li et al., 2009, 2010; Li,2010; Zhang et al., 2014). For example, in a 23-year-old Acaciamangium plantation, understory removal and keeping the siteunderstory vegetation free did not affect soil organic C (SOC),microbial biomass C (MBC), or available N and P in the 0–20 cm soillayers (Xiong et al., 2008). Wang et al. (2014) found that removal ofunderstory twice a month had little influence on soil pH, C/N, avail-able P, NH4–N and NO3–N, but decreased SOC content over a six-month experiment in Eucalyptus urophylla and Acacia crassicarpaplantations in southern China. In addition, Wang et al. (2011)reported that soil respiration (Rs) rate was reduced by frequentmanual understory removal (UR), increased by understory vegeta-tion addition with Cassia alata (CA), but was not changed by

UR + CA in a mixed species forest plantation in southern China.Removal of understory vegetation and addition of N-fixing speciescan exert variable influences on GHG emissions, depending on theunderstory vegetation type and structure, soil temperature, mois-ture content, SOC and nutrient status in plantation forests (Liet al., 2009; Wu et al., 2011a; Wang et al., 2011; Zhang et al., 2014).

Chinese chestnut (Castanea mollissima Blume) is a member ofthe Fagaceae family and is an important economic forest tree spe-cies. This species is distributed across 26 provinces and occupiesabout 1.25 million ha in China, which comprises 38% of the world’stotal chestnut plantation area and 75% of the total chestnut pro-duction (Chen et al., 2011; Li et al., 2014). Over the past two dec-ades, intensive management practices such as fertilization, tillageand understory vegetation management have increasingly becomecommon in Chinese chestnut plantations because of the demandfor high yield and high economic return (Wu et al., 2010). Our pre-vious study demonstrated that understory removal withoutreplacement with legume species increased soil CO2 and N2O emis-sions and CH4 uptake in a one-year field study, but had no effect onwater soluble organic C (WSOC) and MBC (Zhang et al., 2014).However, the interactive effect between understory vegetationmanagement and fertilization on soil GHG emissions and labileorganic C pools in these types of intensively managed plantationshas not been studied.

Replacement of understory vegetation with N-fixing speciesincreased soil GHG emissions in tropical and subtropical planta-tions (Li et al., 2009, 2010; Zhang et al., 2014), but these studiesrarely considered if the understory replacement effects on soilGHG emissions and C pools are affected by fertilization. In addition,the effects of fertilization on soil GHG emissions were rather com-plicated, with conflicting results reported for forest plantations(Zhang et al., 2008a; Janssens et al., 2010; Zhang et al., 2011;Krause et al., 2013; Tu et al., 2013). To improve our understandingof how understory replacement with legume species and/or fertil-ization influence soil GHG emissions and soil C and N pools, weconducted an one-year field study in which understory wasremoved or replaced in plots with or without fertilization in aChinese chestnut plantation, and we measured their effects on soiltemperature, moisture content, WSOC and MBC concentrations,and GHG fluxes. The objectives of this study were (1) to examinethe short-term (one-year) effects of fertilization, replacement ofunderstory with legume species, and their interaction on soilGHG fluxes and C and N pools in a Chinese chestnut plantationand (2) to determine the relationships between soil GHG fluxesand biotic and abiotic factors such as soil temperature, moisturecontent, WSOC, and MBC.

2. Materials and methods

2.1. Study area

The experiment was conducted in typical Chinese chestnutstands located at Tianze Forest Farm (30�250N, 119�860E) in Qing-Shan Township, Lin’an City, Zhejiang Province, China. This regionhas a typical subtropical humid monsoon climate with four distinctseasons. The mean annual air temperature and precipitation were15.9 �C and 1450 mm, respectively, between 2001 and 2010. Themonthly precipitation and mean air temperature during the exper-imental period are shown in Fig. 1. The annual sunshine and frost-free period averaged 1774 h and 239 days, respectively. The sitewas located in a hilly area at 150–250 m above sea level. The soilsat this site were classified as Ferralsols according to the FAO soilclassification system (WRB, 2006).

The Chinese chestnut plantation at this site was converted fromnatural evergreen broadleaf forests by planting chestnut trees after

2011

-6

2011

-7

2011

-8

2011

-9

2011

-10

2011

-11

2011

-12

2012

-1

2012

-2

2012

-3

2012

-4

2012

-5

2012

-6

0

100

200

300

400

500

600

Date (year-month)

Prec

ipita

tion

(mm

)

Monthly precipitation Monthly mean air temperature

0

5

10

15

20

25

30

35

Tem

pera

ture

(o C

)

Fig. 1. Monthly mean air temperature and precipitation during the experimentalperiod in the studied Chinese chestnut plantation.

128 J. Zhang et al. / Forest Ecology and Management 337 (2015) 126–134

clear-cutting the evergreen broadleaf forest. The frequency offertilization for the management of chestnut plantations is oncea year (in May or June) with broadcast application of fertilizersincluding urea (85 kg N ha�1), super phosphate (25 kg P ha�1) andpotassium chloride (65 kg K ha�1), and understory vegetation wasremoved several times a year. Tillage to 20–30 cm was conductedfollowing fertilization. The experiment was established in May2011 and the selected Chinese chestnut stands were 18 years old,with approximately 540 stems ha�1 and a mean diameter at breastheight of 14.1 cm. The dominant understory species that were nat-urally invaded include Dicranopteris dichotoma, Imperata cylindrica,Vaccinium bracteatum, and Artemisia argyi in the chestnut planta-tions. In addition, soil sampling (0–20 cm) of the studied chestnutforest was conducted in May 2011 (before the beginning of thisstudy) and in June 2012 (after the termination of this study) fromeach plot to determine soil chemical and physical properties. Priorto the experiment, the soils in the study area had a bulk density of1.17 g cm�3 and a pH of 4.61, and they contained 15.8 g organicC kg�1, 1.96 g total N kg�1, 9.97 mg available P kg�1, 100.3 mgavailable K kg�1, 2.35 mg NH4

+–N kg�1, and 2.18 mg NO3�–N kg�1.

2.2. Experimental design

The experiment was established in May 2011 using a 2 � 2factorial design. The four treatment combinations were Control(understory vegetation removed without understory replacementor fertilization), seeding of Medicago sativa L. after understoryremoval (we also call this an understory replacement treatment)(MS), fertilization after understory removal (F), and MS + F. Theplots were arranged in a randomized block design with three rep-licates per treatment (one block for each replicate). Four treatmentplots (12 � 12 m) were randomly assigned in each block with a3-m buffer zone to eliminate the edge effect. The native understoryvegetation in all plots was removed several times by hand with thehelp of a machete knife without disturbing the soil in early June.The forest floor was present in all plots throughout this study. M.sativa L. was seeded manually by broadcasting at a rate of22.5 kg ha�1 after removing the understory in the MS and MS + Ftreatments, and fertilizers, including urea (85 kg N ha�1), superphosphate (25 kg P ha�1) and potassium chloride (65 kg K ha�1),were broadcast applied in the F and MS + F treatments.

2.3. Measurement of GHG fluxes

Fluxes of soil CO2, CH4, and N2O were determined from June2011 to June 2012 using a static chamber technique (Ni et al.,2012; Krause et al., 2013). Several weeks before the first sample

collection, a specially designed rectangular polyvinyl chloride(PVC) static chamber was constructed using 5 mm PVC boards.The chamber has a base box (0.3 � 0.3 � 0.1 m) and a chamberbox (0.3 � 0.3 � 0.3 m). The base box was inserted 0.1 m into thesoil. In this study, three days before the commencement of gassampling, one chamber box was placed in the center of each block.At gas sampling, the chamber box was placed in the U-shaped slotof the base box, and the slot was filled with 0.02 m distilled water.Gas samples were collected using a 60 mL plastic syringe 0, 10, 20,and 30 min after closing the chamber. All gas sampling was con-ducted between 09:00 and 11:00 am when the GHG fluxes mea-sured are generally used to represent the mean daily GHG fluxrates in forest plantations (Alves et al., 2012). Each gas samplewas injected into a pre-evacuated container using a butyl rubberstopper and then transported to the laboratory for analysis (Niet al., 2012). To homogenize the gas in the chamber, we pumpedthe internal gas several times using the sampling syringe prior totaking a sample (Liu et al., 2011). Soil GHG fluxes were measuredtwice in the first month and once a month thereafter. At each gassampling, the soil temperature at 5 cm depth was measured by athermometer, and soil samples were collected by removing the for-est floor, digging soil pits, and then taking the 0–20 cm soil layerfrom five points located in the four corners and the center areain each plot to determine soil moisture content, and WSOC andMBC concentrations. To determine soil bulk density, soil samples(0–20 cm) were collected using a 200 cm3 bulk density corer inall plots.

The concentrations of CO2, CH4, and N2O in gas samples weredetermined using a Shimadzu GC-2014 gas chromatograph (GC)(Shimadzu, Japan) within 24 h of the gas sampling, followingKrause et al. (2013). The following formula (1) was used to calcu-late soil CO2, CH4 and N2O flux rates (Liu et al., 2011):

F ¼ qVA

PP0

T0

TdCt

dtð1Þ

where F (mg CO2 m�2 h�1, mg CH4 m�2 h�1, or mg N2O m�2 h�1) isthe soil GHG efflux, q (g m�3) is the density of CO2, CH4, or N2Ounder the standard condition, V (m3) is the effective volume ofthe chamber, A (m2) is the effective cross-sectional area of thechamber, T0 is the absolute temperature under the standard condi-tion, T is the absolute temperature in the chamber, and dCt

dtis the

change in the concentrations (m3 m�3) of CO2, CH4, or N2O in thechamber during the sampling period (h).

The cumulative soil CO2, CH4, and N2O fluxes were calculated asfollows (Liu et al., 2011; Ni et al., 2012):

Mg ¼X½ðRiþ1 þ RiÞ=2� � ðtiþ1 � tiÞ � 24� 10�5 ð2Þ

where Mg is the cumulative soil CO2, CH4, or N2O flux (t CO2 ha�1 yr�1,kg CH4 ha�1 yr�1, or kg N2O ha�1 yr�1, respectively), R is the soil CO2,CH4, or N2O fluxes (mg CO2 m�2 h�1, lg CH4 m�2 h�1, orlg N2O m�2 h�1) determined at each sampling time, i is the samplingnumber, and t is the sampling time based on the Julian day.

2.4. Analyses of soil physicochemical properties

The soil samples collected from five points in each plot weremixed and homogenized, with roots and stones removed. A subsetof the sieved (<2 mm) fresh soil sample was air-dried to determinesoil chemical properties, with the remaining sieved fresh sample(<2 mm) was stored in a refrigerator at 4 �C for determining soilNH4

+–N and NO3�–N concentrations. Fresh soil samples were

extracted using 2 mol L�1 KCl, and the concentrations of NH4+–N

and NO3�–N in the extracts were then determined using a Dionex

ICS 1500 ion chromatograph (Dionex Corp., Atlanta, USA) (Zhanget al., 2013). Soil pH was measured in 1:2.5 soil:water (w:v) using

J. Zhang et al. / Forest Ecology and Management 337 (2015) 126–134 129

a pH meter (PB-10, Sartorius, Germany). The available P concentra-tion in the soil sample was determined using the Bray method(Bray and Kurtz, 1945), and the available K concentration wasdetermined using the flame photometric method after extractionwith a NH4OAc (1 mol L�1, pH 7.0) solution (Stanford andEnglish, 1949). A small portion of each soil sample was groundand passed through a 0.25-mm mesh for the analyses of soilorganic C (SOC) and total nitrogen (TN) concentrations using theK2Cr2O7–H2SO4 wet combustion (Walkley and Black, 1934) andthe semi-micro Kjeldahl (Kirk, 1950) methods, respectively. Inaddition, the soil moisture content of each sample was determinedby oven-drying a sub-sample of the soils at 105 �C to a constantweight.

2.5. Analyses of WSOC and MBC concentrations

The WSOC concentration was measured following Wu et al.(2010). Briefly, the soil WSOC was extracted from fresh soils usingan equivalent of 20 g oven-dry soil, to which 40 mL of distilledwater was added. The mixture was shaken at 120 rpm for 30 minat 25 �C, centrifuged for 20 min at 4000g, filtered through a0.45-lm membrane filter (Millipore Corp, USA) and analyzed usingan automated TOC-TN analyzer (Shimadzu, TOC-Vcph, Japan).

The MBC concentration was measured using the fumigation–extraction method (Vance et al., 1987). Two fresh soil (<2 mm)samples were weighed into glass beakers. One of the soil sampleswas fumigated with ethanol-free CHCl3, while the other was notfumigated. After a period of 24 h, both samples were extracted ina 0.5 M K2SO4 solution with a 1:5 (w:v) soil:extractant ratio. Theorganic C of the extracts was analyzed using an automated Shima-dzu TOC-TN analyzer (Shimadzu, TOC-Vcph, Japan). The MBC con-centration was calculated as the difference in the K2SO4-extractedC between the fumigated and non-fumigated soils and using anextraction efficiency factor of 0.45 (Wu et al., 1990).

2.6. Statistical analyses

The data presented in this paper were the average of three rep-licates. All of the statistical analyses were performed using theSPSS software (SPSS 13.0 for Windows, SPSS Inc., Chicago, USA).Before performing the analysis of variance (ANOVA), the normalityof distribution and homogeneity of the variance were tested, andthe data were log-transformed if the assumption of homogeneitywas violated. A two-way ANOVA and an LSD test were used to testthe effects of understory management and fertilization on soilphysical and chemical properties, the annual mean of WSOC andMBC concentrations, cumulative GHG fluxes, and the time-inte-grated GWP of CO2, CH4 and N2O in the studied Chinese chestnutplantation. An alpha level of 0.05 was used in all statistical analy-ses, unless otherwise noted. Linear regression analyses were con-ducted to determine the relationship between soil GHG fluxesand environmental factors. To describe the relationship betweensoil CO2 efflux and soil temperature, the following exponentialfunction was used:

y ¼ a� ek�t ð3Þ

where y is soil CO2 flux (mg CO2 m�2 h�1), t is soil temperature, anda and k are constants.

The GWP from soil CO2, CH4, and N2O emissions in a 100-yeartime frame was calculated as follows:

Total GWP ðCO2 equivalents; Mg ha�1 yr�1Þ¼ FCO2 þ 25 FCH4 þ 298 FN2O ð4Þ

where FCO2 , FCH4 , and FN2O were the annual cumulative emissions(Mg ha�1 yr�1) of CO2, CH4, and N2O, respectively. The factors of

25 and 298 were the molecular GWPs of CH4 and N2O, respectively,over a 100-year time horizon, as compared with CO2 (Forster et al.,2007).

3. Results

3.1. Understory vegetation management and fertilization effects onsoil properties and microclimatic conditions

Understory replacement (P < 0.05), fertilization (P < 0.01), andtheir interaction (P < 0.05) markedly affected soil organic C (SOC)and NO3

�–N concentrations in the 0–20 cm soil layer (Table 1).However, soil pH, bulk density, and the available P and K concen-trations were not significantly affected by all treatments in thestudied plots (Table 1). In addition, the soil NH4

+–N and TN concen-trations in the F and MS + F treatments were higher than that in theControl and MS treatments (P < 0.01) (Table 1).

Understory replacement, fertilization and their interaction didnot affect the seasonal trends of soil temperature at a depth of5 cm and moisture content (Fig. 2a and b). In addition, there wasno significant difference in the annual mean soil temperature inall treatments (Fig. 2a), while the annual mean soil moisture con-tent was markedly higher in the F and MS + F treatments than inthe Control (P < 0.05) (Fig. 2b).

3.2. Understory vegetation management and fertilization effects onsoil WSOC and MBC concentrations

The highest WSOC value was observed over the period from Julyto September, and the lowest value was observed over the periodfrom January to February (Fig. 2c). However, soil MBC concentrationreached its highest in November and December and its lowest inFebruary (Fig. 2d). In addition, understory replacement (P < 0.05),fertilization (P < 0.01), and their interaction (P < 0.05) affected theannual mean of WSOC and MBC concentrations in the studied Chi-nese chestnut plantation (Table 2). Fertilization markedly increasedthe annual mean concentrations of WSOC and MBC (P < 0.01), bothwith and without seeding of understory with M. sativa L. during thestudy periods, while there was no significant difference in theWSOC and MBC concentrations between the MS treatment andthe Control in the chestnut plantation (Table 2).

3.3. Understory vegetation management and fertilization effects onsoil GHG emissions

Soil CO2 emissions in all plots were at the highest in July–August and the lowest in January–February (Fig. 3a). The meanannual CO2 fluxes increased by 26.9% and 44.9% in the F andMS + F treatments, respectively, in comparison with the Control(336 mg CO2 m�2 h�1) (P < 0.05), while no significant change wasobserved in the MS treatment. In addition, the soil CO2 fluxes inthe F and MS + F treatments were higher than that in the MS andControl treatments between June and November and betweenApril and July (P < 0.05), but there were no treatment effectsbetween December and February (Fig. 3a).

The soil had the highest CH4 uptake rates in early autumn andthe lowest in late winter and early spring (Fig. 3b). The annualmean uptake in the Control was 35.8 lg CH4 m�2 h�1, which washigher than that in the F and MS + F treatments (P < 0.05), butwas not different from that in the MS treatment (Fig. 3b).

Soil N2O fluxes exhibited strong seasonal variations, with themaximum observed in the summer and the minimum values inthe winter (Fig. 3c). Meanwhile, the annual mean fluxes in the Fand MS + F treatments were 68.0% and 94.2% higher, respectively,than that in the Control (25.5 lg N2O m�2 h�1) (P < 0.05). In

Table 1Effects of understory vegetation management and fertilization on selected chemical and physical properties of soils (0–20 cm) in a Chinese chestnut plantation (n = 3).

Factors pH Bulk density (g cm�3) Organic C Total N NH4+–N NO3

�–N Available P Available K

Understory management Fertilization (g kg�1) (mg kg�1)

0 0 4.63 a Ab 1.17 a A 15.40 a A 1.89 b A 2.39 b A 2.10 b A 10.99 a A 98.53 a A0 F 4.61 a A 1.16 a A 16.51 a B 2.09 a B 2.48 a B 2.46 a B 11.42 a A 102.87 a AMSa 0 4.58 a A 1.18 a A 15.75 b A 1.91 b A 2.40 b A 2.13 b A 11.02 a A 99.91 a AMS F 4.60 a A 1.16 a A 18.53 a A 2.23 a A 2.56 a A 2.65 a A 11.51 a A 101.77 a A

F valueMS 1.67 0.31 12.4* 4.39 3.6 31.20** 0.04 0.41F 0.01 4.93 33.5** 48.9** 29.12** 319.96** 2.68 0.05MS � F 0.89 1.23 6.2* 2.16 2.57 6.12* 0.01 0.25

a MS – seeding of Medicago sativa L. as an understory vegetation, F – fertilization.b Different lower- and uppercase letters within a column indicate significant differences between the different fertilization practices within the same understory man-

agement treatment and different understory management practices within the same fertilization treatment, respectively, at the P = 0.05 level, based on the least significantdifference (LSD) test.

* Significant at P = 0.05.** Significant at P = 0.01.

0

10

20

30

40

150

200

250

300

350

50

100

150

200

2011

/6/1

2011

/7/1

2011

/8/1

2011

/9/1

2011

/10/

1

2011

/11/

1

2011

/12/

1

2012

/1/1

2012

/2/1

2012

/3/1

2012

/4/1

2012

/5/1

2012

/6/1

2012

/7/1

100

200

300

400

500

Soil

tem

pera

ture

( o C

)

Control MS F MS+F

(a)

Soil

moi

stur

e (g

kg-1

)

(b)

WSO

C

(mg

kg-1

)

(c)

MB

C

(mg

kg-1

)

Date (year/month/day)

(d)

Fig. 2. Seasonal changes in (a) soil temperature at a depth of 5 cm, (b) soil moisturecontent in the 0–20 cm soil layer, (c) soil water soluble organic C (WSOC)concentration in the 0–20 cm soil layer, and (d) soil microbial biomass C (MBC)concentration in the 0–20 cm soil layer in a Chinese chestnut plantation from June2011 to June 2012 under four different management practices, i.e., control, seedingof understory with Medicago sativa L. (MS), fertilization (F), and MS + F. Error barsrepresent standard deviations (n = 3).

Table 2Effects of understory vegetation management and fertilization on soil water solubleorganic C (WSOC) and microbial biomass C (MBC) concentrations (mg kg�1) in aChinese chestnut plantation (n = 3).

Factors WSOC MBC

Understory management Fertilization

0 0 73 b Ab 254 b A0 F 111 a B 296 a BMSa 0 78 b A 259 b AMS F 138 a A 351 a A

F valueMS 14.6** 13.7*

F 135.5** 66.7**

MS � F 7.2* 9.5*

a MS – seeding of Medicago sativa L. as an understory vegetation, F – fertilization.b Different lower- and uppercase letters within a column indicate significant

differences between the different fertilization practices within the same understorymanagement treatment and different understory management practices within thesame fertilization treatment, respectively, at the P = 0.05 level, based on the leastsignificant difference (LSD) test.

* Significant at P = 0.05.** Significant at P = 0.01.

130 J. Zhang et al. / Forest Ecology and Management 337 (2015) 126–134

addition, the soil N2O fluxes in the F and MS + F treatments werehigher than that in the Control and MS treatments between Juneand January (P < 0.05), but not different between February andJune (Fig. 3c).

Understory replacement (P < 0.05), fertilization (P < 0.01), andtheir interaction (P < 0.05) affected the annual cumulative soilCO2, CH4 and N2O fluxes and total GWP of GHG emissions in thestudied forest (Table 3). Fertilization markedly increased the soilCO2, CH4 and N2O fluxes, and GWP (P < 0.01 for CO2, CH4, N2O,and GWP), in the plots of both with and without seeding ofunderstory with M. sativa L., while understory replacement signif-

icantly increased the GHG fluxes and total GWP in the F and F + MStreatments, but not in the Control and MS treatments (Table 3).

In comparison with that in the control, the annual cumulativesoil CO2, CH4, and N2O fluxes increased by 25.0%, 11.7%, and67.4%, respectively, in the F treatment (P < 0.05 for CO2, CH4, andN2O), and by 45.3%, 26.1%, and 101.3%, respectively, in the MS + Ftreatment (P < 0.05 for CO2, CH4, and N2O) (Table 3). Irrespectiveof the treatment, the CO2 had the largest contribution (97%) tothe total GWP, while N2O and CH4 fluxes contributed less than3%. The total GWP in the F and MS + F treatments were 41.9 and36.0 Mg CO2-e ha�1 yr�1, respectively, and these values werehigher than those in the Control and MS treatments (P < 0.05)(Table 3).

3.4. Relationships between soil GHG emissions and environmentalfactors

Soil CO2 flux increased exponentially with soil temperature(5 cm below the soil surface) and was linearly correlated with soilWSOC concentration (P < 0.01) but not with soil moisture contentor MBC concentration (Table 4). Furthermore, the Q10 value wassignificantly higher in the F (1.80) treatment than that in theControl (1.49), MS (1.54), and MS + F (1.65) treatments. Soil CH4

flux was negatively correlated with both soil temperature andWSOC (P < 0.05) but not with soil moisture content or MBC

0

200

400

600

800

1000

-100

-80

-60

-40

-20

0

2011

/6/1

2011

/7/1

2011

/8/1

2011

/9/1

2011

/10/

1

2011

/11/

1

2011

/12/

1

2012

/1/1

2012

/2/1

2012

/3/1

2012

/4/1

2012

/5/1

2012

/6/1

0

25

50

75

100

CO

2 flux

(mg

m-2

h-1

) Control MS F MS+F

(a)

CH

4 flux

(µg

m-2

h-1

)

(b)

N2O

flux

g m

-2 h

-1)

Date (year/month/day)

(c)

(µ

Fig. 3. Seasonal changes in soil (a) CO2, (b) CH4, and (c) N2O fluxes in a Chinesechestnut plantation from June 2011 to June 2012 under four different managementpractices, i.e., control, seeding of understory with Medicago sativa L. (MS),fertilization (F), and MS + F. Error bars represent standard deviations (n = 3).

J. Zhang et al. / Forest Ecology and Management 337 (2015) 126–134 131

(Table 4). Soil N2O flux was positively correlated with both soiltemperature and WSOC (P < 0.05) but not with soil MBC (Table 4).In addition, significant relationships between soil moisture contentand soil N2O flux were observed in the Control (Y = 3.12X � 56.03,R2 = 0.44, P < 0.01) and MS (Y = 2.46X � 40.98, R2 = 0.31, P < 0.05)treatments, but not in the F and MS + F treatments.

4. Discussion

4.1. Understory vegetation management and fertilization effects onsoil GHG emissions

Fertilization and understory vegetation management affect soilGHG emissions in forest plantations (Li et al., 2009; Wang et al.,

Table 3Effects of understory vegetation management and fertilization on the annual cumulativechestnut plantation (n = 3).

Factors CO2 flux(Mg CO2 ha�1 yr�1)

N2

(kgUnderstorymanagement

Fertilization

0 0 28.1 b Ab 1.90 F 35.1 a B 3.2MSa 0 29.2 b A 2.0MS F 40.8 a A 3.9

F valueMS 24.1** 12F 182.6** 22MS � F 11.0* 7.2

a MS – seeding of Medicago sativa L. as an understory vegetation, F – fertilization.b Different lower- and uppercase letters within a column indicate significant differen

agement treatment and different understory management practices within the same ferdifference (LSD) test.

* Significant at P = 0.05.** Significant at P = 0.01.

2011; Tu et al., 2013; Zhang et al., 2014). In this study, fertilizationmarkedly increased soil CO2 and N2O emissions in the Chinesechestnut plantation over the one-year field study (Fig. 3a and c,Table 3), consistent with Zhang et al. (2008a), where they reportedthat N addition significantly increased soil N2O emission in amature tropical plantation forest. Tu et al. (2013) similarly foundthat N addition significantly increased soil respiration in a subtrop-ical bamboo ecosystem. In addition, the application of fertilizerdecreased soil CH4 uptake in our study, similar to the results fromplantations in Zhang et al. (2011) and Fender et al. (2012). The pos-sible explanations for these results would be: (1) inorganic N fertil-ization enhanced microbial and plant root activities (increased CO2

fluxes) and increased N availability in the soil which increased N2Oemissions through nitrification and denitrification (Zhang et al.,2008a; Jassal et al., 2010; Li et al., 2012); (2) fertilization increasedsoil NH4

+–N and NO3�–N concentrations (Table 1) and the high avail-

ability of soil N (especially ammonium) might have competitivelyinhibited the oxidation of CH4 (Zhang et al., 2008b; Jassal et al.,2011; Krause et al., 2013).

Understory replacement with selected legume species signifi-cantly enhanced soil CO2 and N2O emissions and decreased CH4

uptake in E. urophylla and A. mangium plantations (Li et al., 2009;Li, 2010; Wang et al., 2011), which was consistent with our resultsthat understory replacement with M. sativa L. increased CO2 andN2O fluxes and decreased CH4 uptake in the fertilized plots, butwas different from the lack of effect of seeding with M. sativa L.on GHG emissions in the no fertilization plots. A possible explana-tion for these interesting results would be that fertilization alteredthe short-term effects of seeding with M. sativa L. on the availabil-ity of substrates (including SOC, WSOC and MBC) for soil microor-ganism in the studied plantation (Tables 1–3), and consequentlyaffected the emission of soil CO2, CH4 and N2O (Zhang et al.,2008a; Wang et al., 2010; Liu et al., 2013; Zhang et al., 2014).

In this study, the contributions of soil N2O and CH4 emissions tothe total GWP were very low (<3%), and emissions of CO2

accounted for more than 97% of the total GWP in the chestnutplantation (Table 3). These results were similar to the findings ofWang et al. (2010, 2013) in the subtropical plantations, and ofChristiansen et al. (2012) in a Danish temperate forest. Therefore,in comparison with soil N2O and CH4 emissions, much more atten-tion should have been paid to soil CO2 emission in terms of its con-tribution to GWP. Moreover, the responses of different soilrespiration components, i.e., autotrophic and heterotrophic respi-rations, should be investigated, due to their different responsesto management practices or changes in environmental conditions(Tu et al., 2013; Kukumagi et al., 2014).

soil fluxes and global warming potential (GWP) of CO2, N2O and CH4 in a Chinese

O fluxN2O ha�1 yr�1)

CH4 flux(kg CH4 ha�1 yr�1)

Total of GWP(Mg CO2-e ha�1 yr�1)

6 b A �3.19 b A 28.59 b A8 a B �2.83 a B 36.01 a B5 b A �3.11 b A 29.71 b A5 a A �2.35 a A 41.93 a A

.1* 7.9* 25.0**

1.5** 31.5** 194.1**

* 4.0* 11.5*

ces between the different fertilization practices within the same understory man-tilization treatment, respectively, at the P = 0.05 level, based on the least significant

Table 4Relationships between soil CO2, N2O, and CH4 fluxes and environmental factors under the understory vegetation management and fertilization in a Chinese chestnut plantation(n = 13).

Factors Soil temperature (�C) WSOC (mg kg�1)

Understory management Fertilization

CO2 flux (mg m�2 h�1)0 0 Y = 154.55 e0.040X (R2 = 0.90, P < 0.01) Y = 8.45X � 283.74 (R2 = 0.66, P < 0.01)0 F Y = 131.34 e0.050X (R2 = 0.94, P < 0.01) Y = 7.56X � 410.96 (R2 = 0.74, P < 0.01)MSa 0 Y = 147.56 e0.043X (R2 = 0.87, P < 0.01) Y = 9.15X � 366.81 (R2 = 0.75, P < 0.01)MS F Y = 178.48 e0.050X (R2 = 0.85, P < 0.01) Y = 6.40X � 397.89 (R2 = 0.68, P < 0.01)

CH4 flux (lg m�2 h�1)0 0 Y = �1.40X � 10.29 (R2 = 0.32, P < 0.05) Y = �1.14X + 48.12 (R2 = 0.29, P < 0.05)0 F Y = �1.19X � 9.90 (R2 = 0.28, P < 0.05) Y = �0.53X + 27.51 (R2 = 0.37, P < 0.05)MS 0 Y = �1.41X � 9.16 (R2 = 0.32, P < 0.05) Y = �1.27X + 64.33 (R2 = 0.43, P < 0.01)MS F Y = �0.96X � 8.99 (R2 = 0.29, P < 0.05) Y = �0.42X + 31.91 (R2 = 0.49, P < 0.01)

N2O flux (lg m�2 h�1)0 0 Y = 1.42X � 0.42 (R2 = 0.80, P < 0.01) Y = 0.90X � 40.08 (R2 = 0.42, P < 0.01)0 F Y = 3.01X � 11.91 (R2 = 0.75, P < 0.01) Y = 1.01X � 69.25 (R2 = 0.56, P < 0.01)MS 0 Y = 1.46X + 0.09 (R2 = 0.80, P < 0.01) Y = 0.94X � 46.96 (R2 = 0.55, P < 0.01)MS F Y = 2.92X � 3.56 (R2 = 0.70, P < 0.01) Y = 0.81X � 62.28 (R2 = 0.47, P < 0.01)

a MS – seeding of Medicago sativa L. as an understory vegetation, F – fertilization.

132 J. Zhang et al. / Forest Ecology and Management 337 (2015) 126–134

It is worth mentioning here that this study covered a relativelyshort time span after the management practices were applied, andthus the response of soil GHG fluxes to different management prac-tices should be regarded as the short-term effect of managementpractices, since other studies reported that there were significantdifferences between short- and long-term effects of managementpractices on soil GHG emissions (Jassal et al., 2010; Wei et al.,2014). Therefore, the long-term effects of management practiceson soil carbon pool or GHG emissions in plantations need to beaddressed in the future.

4.2. Understory vegetation management and fertilization effects onsoil C and N pools

In general, the application of organic or inorganic fertilizerswould markedly affect the storage and sequestration of soil Cand N in managed ecosystems (Vogel et al., 2011; Fröberg et al.,2013). Our results also showed that concentrations of SOC, WSOC,MBC, TN, NO3

�–N, and NH4+–N markedly increased in the fertilized

plots in the chestnut forest (Tables 1 and 2; Fig. 3), suggesting thata moderate rate of fertilization, considered an effective practice,could increase the soil C and N stocks in the studied forestplantation.

Moreover, understory vegetation replacement alone had littleimpact on soil C and N concentrations, but the combination ofunderstory replacement and fertilization significantly increasedsoil C and N storages in this study (Tables 1 and 2). Similarly, Liet al. (2009, 2010) found that understory removal with adding N-fixing species (after half a year) had no effects on SOC or NH4

+–Nconcentration but increased the MBC and NO3

�–N concentrations;however, understory removal without adding N-fixing speciesreduced SOC but had little impact on MBC, NO3

�–N and NH4+–N con-

centrations in the subtropical plantations in China. These small dif-ferences could be attributed to the effects of understorymanagement that depended on soil microbial communities, soilC and N status, forest types, sampling duration, and other combina-tions of management practices in forest plantations (Xiong et al.,2008; Li et al., 2009; Wang et al., 2010; Wang et al., 2014; Zhaoet al., 2013).

Understory removal alone did not affect ecosystem N retentionand fertilization increased soil N pools, but the combination ofunderstory removal and fertilization significantly decreased eco-system N retention in slash and loblolly pine forests (Vogel et al.,2011). Our results showed a significant positive interaction

between understory replacement and fertilization on SOC, WSOC,MBC and NO3

�–N concentrations in the chestnut forest, suggestingthat understory replacement with N-fixing species combined witha moderate rate of fertilization could be beneficial for maintainingsoil fertility and the sustainable development of forest plantations.

4.3. Effects of environmental factors on soil GHG emissions

Soil CO2, N2O and CH4 fluxes were strongly correlated with tem-perature in our study (P < 0.05) (Table 4), consistent with previousstudies (Mo et al. 2007; Li et al. 2010; Fang et al., 2010). In addition,the Q10 value was used to assess the effects of forest managementpractices on soil respiration (Pang et al., 2013). In this study, theQ10 values ranged from 1.49 to 1.80, in accordance with previousstudies in subtropical and tropical forests (Deng et al., 2010;Wang et al., 2011). A higher Q10 value was found in the F andMS + F treatments, likely due to the increased substrate availabilityfor microbial respiration as a result of fertilization (Deng et al.,2010; Zhang et al., 2011).

In this study, soil N2O fluxes correlated with soil moisture con-tent in the no fertilization plots but not in the fertilization treat-ment. Lin et al. (2012) similarly found N2O fluxes related withsoil moisture content in a upland but not in an orchard, due tothe differences on moisture content and other biophysical condi-tions that affected the N2O fluxes under intensive managementpractices (Iqbal et al., 2008; Lin et al., 2012).

Soil CO2, CH4 and N2O fluxes were significantly correlated withWSOC (P < 0.05) but not with MBC in this study (Table 4), similar tothe findings in subtropical or tropical China (Iqbal et al., 2008; Liet al., 2010; Lin et al., 2012). In addition, we also found that the cor-relation between soil CH4 flux and WSOC was better in the MS andMS + F treatments than in the Control and F treatments (Table 4).Therefore, we conclude that soil GHG fluxes from the studied Chi-nese chestnut forest were primarily controlled by soil temperatureand WSOC concentration, which likely masked the effects of MBCon GHG emissions.

5. Conclusions

Soil GHG fluxes in the studied Chinese chestnut plantation hada strong seasonal pattern, regardless of the management practice.Understory replacement and fertilization changed soil GHG fluxesand labile organic C pools in the Chinese chestnut plantation. Interms of soil GHG fluxes, understory replacement likely is the

J. Zhang et al. / Forest Ecology and Management 337 (2015) 126–134 133

optimum management technique as it caused little effect on GHGemissions, while understory replacement plus inorganic fertiliza-tion is undesirable as it substantially increased GHG fluxes. How-ever, fertilization can enhance the effects of understoryreplacement on increasing soil organic C and nutrient availability.Therefore, based on this one-year study, the combination of under-story replacement with legume species and a moderate rate of fer-tilization would be an alternative management regime, which isbeneficial for increasing C sequestration and maintaining soil fer-tility in Chinese chestnut plantations. In addition, this study wasconducted in the first year after such treatments; thus it merelyrepresented the first transitional effects including the immediateeffects of disturbance caused by the management applied. Long-term effects of these treatments are likely different from whathas been found in this study. To gain a better understanding ofthe effects of such management practices on soil C dynamics inChinese chestnut plantations, long-term field experiments shouldbe conducted in the future.

Acknowledgements

We thank Zhenming Shen for his help in establishing the fieldexperiment, and Zhanlei Wang, Xueshuang Chen, Xiaojie Zhang,Qiying Huang and Wei Xu for their help with the soil and gas sam-pling. This work was financially supported by the National NaturalScience Foundation of China (Nos. 31170576, 30911090), NaturalScience Foundation of Zhejiang Province (No. LY14C160007), Zhe-jiang Province Key Science and Technology Innovation Team (No.2010R50030-10), and Training Program for the Top Young Talentof Zhejiang A & F University (No. 2034070003).

References

Alves, B.J.R., Smith, K.A., Flores, R.A., Cardoso, A.S., Oliveira, W.R.D., Jantalia, C.P.,Urquiaga, S., Boddey, R.M., 2012. Selection of the most suitable sampling timefor static chambers for the estimation of daily mean N2O flux from soils. SoilBiol. Biochem. 46, 129–135.

Bray, R.H., Kurtz, L.T., 1945. Determination of total, organic, and available forms ofphosphorus in soils. Soil Sci. 59, 39–46.

Chen, D.M., Zhang, C.L., Wu, J.P., Zhou, L.X., Lin, Y.B., Fu, S.L., 2011. Subtropicalplantations are large carbon sinks: evidence from two monoculture plantationsin South China. Agr. Forest Meteorol. 151, 1214–1225.

Christiansen, J.R., Gundersen, P., Frederiksen, P., Vesterdal, L., 2012. Influence ofhydromorphic soil conditions on greenhouse gas emissions and soil carbonstocks in a Danish temperate forest. Forest Ecol. Manage. 284, 185–195.

Deng, Q., Zhou, G., Liu, J., Liu, S., Duan, H., Zhang, D., 2010. Responses of soilrespiration to elevated carbon dioxide and nitrogen addition in youngsubtropical forest ecosystems in China. Biogeosciences 7, 315–328.

Elser, J.J., Bracken, M.E., Cleland, E.E., Gruner, D.S., Harpole, W.S., Hillebrand, H.,Ngai, J.T., Seabloom, E.W., Shurin, J.B., Smith, J.E., 2007. Global analysis ofnitrogen and phosphorus limitation of primary producers in freshwater, marineand terrestrial ecosystems. Ecol. Lett. 10, 1135–1142.

Fang, H.J., Yu, G.R., Cheng, S.L., Zhu, T.H., Wang, Y.S., Yan, J.H., Wang, M., Cao, M.,Zhou, M., 2010. Effects of multiple environmental factors on CO2 emission andCH4 uptake from old-growth forest soils. Biogeosciences 7, 395–407.

Fender, A., Pfeiffer, B., Gansert, D., Leuschner, C., Daniel, R., Jungkunst, H.F., 2012.The inhibiting effect of nitrate fertilisation on methane uptake of a temperateforest soil is influenced by labile carbon. Biol. Fert. Soils 48, 621–631.

Forster, P., Ramaswamy, V., Artaxo, P., Berntsen, T., Betts, R., Fahey, D.W., Haywood,J., Lean, J., Lowe, D.C., Myhre, G., Nganga, J., Prinn, R., Raga, G., Schulz, M.,Dorland, R.V., 2007. Changes in atmospheric constituents and in radiativeforcing. In: Solomon, S. et al. (Eds.), Climate Change 2007: The Physical ScienceBasis. Contribution of Working Group I to the Fourth Assessment Report of theIntergovernmental Panel on Climate Change. Cambridge, UK, pp 130–234.

Fröberg, M., Grip, H., Tipping, E., Svensson, M., Strömgren, M., Kleja, D.B., 2013.Long-term effects of experimental fertilization and soil warming on dissolvedorganic matter leaching from a spruce forest in Northern Sweden. Geoderma200, 172–179.

Gao, Q., Hasselquist, N.J., Palmroth, S., Zheng, Z.M., You, W.H., 2014. Short-termresponse of soil respiration to nitrogen fertilization in a subtropical evergreenforest. Soil Biol. Biochem. 76, 297–300.

Iqbal, J., Hu, R.G., Du, L.J., Lan, L., Shan, L., Tao, C., Ruan, L.L., 2008. Differences in soilCO2 flux between different land use types in mid-subtropical China. Soil Biol.Biochem. 40, 2324–2333.

Iqbal, J., Lin, S., Hu, R.G., Feng, M.L., 2009. Temporal variability of soil-atmosphericCO2 and CH4 fluxes from different land uses in mid-subtropical China. Atmos.Environ. 43, 5865–5875.

Janssens, I.A., Dieleman, W., Luyssaert, S., Subke, J.A., Reichstein, M., Ceulemans, R.,Ciais, P., Dolman, A.J., Grace, J., Matteucci, G., Papale, D., Piao, S.L., Schulze, E.D.,Tang, J., Law, B.E., 2010. Reduction of forest soil respiration in response tonitrogen deposition. Nat. Geosci. 3, 315–322.

Jassal, R.S., Black, T.A., Roy, R., Ethier, G., 2011. Effect of nitrogen fertilization on soilCH4 and N2O fluxes, and soil and bole respiration. Geoderma 162, 182–186.

Jassal, R.S., Black, T.A., Trofymow, J.A., Roy, R., Nesic, Z., 2010. Soil CO2 and N2O fluxdynamics in a nitrogen-fertilized Pacific Northwest Douglas-fir stand.Geoderma 157, 118–125.

Kirk, P.L., 1950. Kjeldahl method for total nitrogen. Anal. Chem. 22, 354–358.Kukumagi, M., Ostonen, I., Kupper, P., Truu, M., Tulva, I., Varik, M., Aosaar, J., Sober,

J., Lohmus, K., 2014. The effects of elevated atmospheric humidity on soilrespiration components in a young silver birch forest. Agr. Forest Meteorol. 194,167–174.

Krause, K., Niklaus, P.A., Schleppi, P., 2013. Soil-atmosphere fluxes of thegreenhouse gases CO2, CH4 and N2O in a mountain spruce forest subjected tolong-term N addition and to tree girdling. Agr. Forest Meteorol. 181, 61–68.

Lavoie, M., Kellman, L., Risk, D., 2013. The effects of clear-cutting on soil CO2, CH4,and N2O flux, storage and concentration in two Atlantic temperate forests inNova Scotia, Canada. Forest Ecol. Manage. 304, 355–369.

Li, C.F., Zhou, D.N., Kou, Z.K., Zhang, Z.S., Wang, J.P., Cai, M.L., Cao, C.G., 2012. Effectsof tillage and nitrogen fertilizers on CH4 and CO2 emissions and soil organiccarbon in paddy fields of Central China. PLoS ONE 7, e34642.

Li, H.F., 2010. Soil CH4 fluxes response to understory removal and N-fixing speciesaddition in four forest plantations in Southern China. J. Forest Res. 21, 301–310.

Li, H.F., Fu, S.L., Zhao, H.T., Xia, H.P., 2010. Effects of understory removal and N-fixingspecies seeding on soil N2O fluxes in four forest plantations in southern China.Soil Sci. Plant Nutr. 56, 541–551.

Li, H.F., Xia, H.P., Fu, S.L., Zhang, X.F., 2009. Emissions of soil greenhouse gases inresponse to understory removal and Cassia alata addition in an eucalyptusurophylla plantation in Guangdong province, China. Chin. J. Plant Ecol. 33 (6),1015–1022 (in Chinese).

Li, Y.F., Zhang, J.J., Chang, S.X., Jiang, P.K., Zhou, G.M., Shen, Z.M., Wu, J.S., Lin, L.,Wang, Z.S., Shen, M.C., 2014. Converting native shrub forests to Chinesechestnut plantations and subsequent intensive management affected soil C andN pools. Forest Ecol. Manage. 312, 161–169.

Lin, S., Iqbal, J., Hu, R.G., Ruan, L.L., Wu, J.S., Zhao, J.S., Wang, P.J., 2012. Differences innitrous oxide fluxes from red soil under different land uses in mid-subtropicalChina. Agr. Ecosyst. Environ. 146, 168–178.

Liu, J., Jiang, P.K., Wang, H.L., Zhou, G.M., Wu, J.S., Yang, F., Qian, X.B., 2011. Seasonalsoil CO2 efflux dynamics after land use change from a natural forest to Mosobamboo plantations in subtropical China. Forest Ecol. Manage. 262, 1131–1137.

Liu, X., Chen, C.R., Wang, W.J., Hughes, J.M., Lewis, T., Hou, E.Q., Shen, J., 2013. Soilenvironmental factors rather than denitrification gene abundance control N2Ofluxes in a wet sclerophyll forest with different burning frequency. Soil Biol.Biochem. 57, 292–300.

Manne, A.S., Richels, R.G., 2001. An alternative approach to establishing trade-offsamong greenhouse gases. Nature 410, 675–677.

Mo, J.M., Zhang, W., Zhu, W.X., Fang, Y.T., Li, D.J., Zhao, P., 2007. Response of soilrespiration to simulated N deposition in a disturbed and a rehabilitated tropicalforest in southern China. Plant Soil 296, 125–135.

Mo, J., Zhang, W., Zhu, W., Gundersen, P., Fang, Y., Li, D., Wang, H., 2008. Nitrogenaddition reduces soil respiration in a mature tropical forest in southern China.Global Change Biol. 14, 403–412.

Ni, K., Ding, W.X., Cai, Z.C., Wang, Y.F., Zhang, X.L., Zhou, B.K., 2012. Soil carbondioxide emission from intensively cultivated black soil in Northeast China:nitrogen fertilization effect. J. Soil Sediment. 12, 1007–1018.

Pang, X.Y., Bao, W.K., Zhu, B., Cheng, W.X., 2013. Responses of soil respiration and itstemperature sensitivity to thinning in a pine plantation. Agr. Forest Meteorol.171, 57–64.

Peng, Y.Y., Thomas, S.C., Tian, D.L., 2008. Forest management and soil respiration:implications for carbon sequestration. Environ. Rev. 16, 93–111.

Piao, S.L., Fang, J.Y., Ciais, P., Peylin, P., Huang, Y., Sitch, S., Wang, T., 2009. The carbonbalance of terrestrial ecosystems in China. Nature 458, 1009–1013.

Powers, R.F., Busse, M.D., McFarlane, K.J., Zhang, J., Young, D.H., 2013. Long-termeffects of silviculture on soil carbon storage: does vegetation control make adifference? Forestry 86, 47–58.

Song, X.Z., Yuan, H.Y., Kimberley, M.O., Jiang, H., Zhou, G.M., Wang, H.L., 2013. SoilCO2 flux dynamics in the two main plantation forest types in subtropical China.Sci. Total Environ. 444, 363–368.

Stanford, G., English, L., 1949. Use of the flame photometer in rapid soil tests for Kand Ca. Agron. J. 41, 446–447.

Tang, X.L., Liu, S.G., Zhou, G.Y., Zhang, D.Q., Zhou, C.Y., 2006. Soil-atmosphericexchange of CO2, CH4, and N2O in three subtropical forest ecosystems insouthern China. Global Change Biol. 12, 546–560.

Tu, L.H., Hu, T.X., Zhang, J., Li, X.W., Hu, H.L., Liu, L., Xiao, Y.L., 2013. Nitrogenaddition stimulates different components of soil respiration in a subtropicalbamboo ecosystem. Soil Biol. Biochem. 58, 255–264.

Vance, E.D., Brookes, P.C., Jenkinson, D.S., 1987. An extraction method for measuringsoil microbial biomass C. Soil Biol. Biochem. 19, 703–707.

Vogel, J.G., Suau, L.J., Martin, T.A., Jokela, E.J., 2011. Long-term effects of weedcontrol and fertilization on the carbon and nitrogen pools of a slash and loblollypine forest in north-central Florida. Can. J. Forest Res. 41, 552–567.

134 J. Zhang et al. / Forest Ecology and Management 337 (2015) 126–134

Walkley, A., Black, L.A., 1934. An examination of the Dgtjareff method fordetermining soil organic matter, and a proposed modification of the chromicacid titration method. Soil Sci. 37, 29–38.

Wang, F.M., Zou, B., Li, H.F., Li, Z.A., 2014. The effect of understory removal onmicroclimate and soil properties in two subtropical lumber plantations. J. ForestRes. 19, 238–243.

Wang, H., Liu, S.R., Mo, J.M., Zhang, T., 2010. Soil-atmosphere exchange ofgreenhouse gases in subtropical plantations of indigenous tree species. PlantSoil 335, 213–227.

Wang, H., Liu, S.R., Wang, J.X., Shi, Z.M., Lu, L.H., Zeng, J., Ming, A.G., Tang, J.X., Yu,H.L., 2013. Effects of tree species mixture on soil organic carbon stocks andgreenhouse gas fluxes in subtropical plantations in China. Forest Ecol. Manage.300, 4–13.

Wang, X.L., Zhao, J., Wu, J.P., Chen, H., Lin, Y.B., Zhou, L.X., Fu, S.L., 2011. Impacts ofunderstory species removal and/or addition on soil respiration in a mixed forestplantation with native species in southern China. Forest Ecol. Manage. 261,1053–1060.

Wei, D., Xu, R., Liu, Y.W., Wang, Y.H., Wang, Y.S., 2014. Three-year study of CO2

efflux and CH4/N2O fluxes at an alpine steppe site on the central Tibetan Plateauand their responses to simulated N deposition. Geoderma 232, 88–96.

World Reference Base for Soil Resources (WRB), 2006. A Framework forInternational Classification, Correlation and Communication. Food andAgriculture Organization of the United Nations, Rome.

Wu, J.S., Jiang, P.K., Chang, S.X., Xu, Q.F., Lin, Y., 2010. Dissolved soil organic carbonand nitrogen were affected by conversion of native forests to plantations insubtropical China. Can. J. Soil Sci. 90, 27–36.

Wu, J., Joergensen, R.G., Pommerening, B., Chaussod, R., Brookes, P.C., 1990.Measurement of soil microbial biomass C by fumigation-extraction-anautomated procedure. Soil Biol. Biochem. 22, 1167–1169.

Wu, J.P., Liu, Z.F., Chen, D.M., Huang, G.M., Zhou, L.X., Fu, S.L., 2011a. Understoryplants can make substantial contributions to soil respiration: evidence fromtwo subtropical plantations. Soil Biol. Biochem. 43, 2355–2357.

Wu, J.P., Liu, Z.F., Wang, X.L., Sun, Y.X., Zhou, L.X., Lin, Y.B., Fu, S.L., 2011b. Effects ofunderstory removal and tree girdling on soil microbial community composition

and litter decomposition in two Eucalyptus plantations in South China. Funct.Ecol. 25, 921–931.

Xiong, Y.M., Xia, H.X., Li, Z.A., Cai, X.A., Fu, S.L., 2008. Impacts of litter and understoryremoval on soil properties in a subtropical Acacia mangium plantation in China.Plant Soil 304, 179–188.

Zhang, J.J., Li, Y.F., Chang, S.X., Jiang, P.K., Zhou, G.M., Liu, J., Wu, J.S., Shen, Z.M., 2014.Understory vegetation management affected greenhouse gas emissions andlabile organic carbon pools in an intensively managed Chinese chestnutplantation. Plant Soil 376, 363–375.

Zhang, T., Li, Y.F., Chang, S.X., Jiang, P.K., Zhou, G.M., Liu, J., Lin, L., 2013. Convertingpaddy fields to Lei bamboo (Phyllostachys praecox) stands affected soil nutrientconcentrations, labile organic carbon pools, and organic carbon chemicalcompositions. Plant Soil 367, 249–261.

Zhang, T., Zhu, W., Mo, J., Liu, L., Dong, S., 2011. Increased phosphorus availabilitymitigates the inhibition of nitrogen deposition on CH4 uptake in an old-growthtropical forest, southern China. Biogeosciences 8, 2805–2813.

Zhang, W., Mo, J.M., Yu, G.R., Fang, Y.T., Li, D.J., Lu, X.K., Wang, H., 2008a. Emissionsof nitrous oxide from three tropical forests in Southern China in response tosimulated nitrogen deposition. Plant Soil 306, 221–236.

Zhang, W., Mo, J.M., Zhou, G.Y., Gundersen, P., Fang, Y.T., Lu, X.K., Zhang, T., Dong,S.F., 2008b. Methane uptake responses to nitrogen deposition in three tropicalforests in southern China. J. Geophys. Res. 113. http://dx.doi.org/10.1029/2007JD009195.

Zhao, J., Wan, S.Z., Fu, S.L., Wang, X.L., Wang, M., Liang, C.F., Chen, Y.Q., Zhu, X.L.,2013. Effects of understory removal and nitrogen fertilization on soilmicrobial communities in Eucalyptus plantations. Forest Ecol. Manage. 310,80–86.

Zhao, J., Wan, S.Z., Li, Z.A., Shao, Y.H., Xu, G.L., Liu, Z.F., Zhou, L.X., Fu, S.L., 2012.Dicranopteris-dominated understory as major driver of intensive forestecosystem in humid subtropical and tropical region. Soil Biol. Biochem. 49,78–87.

Zhao, J., Wang, X.L., Shao, Y.H., Xu, G.L., Fu, S.L., 2011. Effects of vegetationremoval on soil properties and decomposer organisms. Soil Biol. Biochem. 43,954–960.