APP下载

Air-drying changes the distribution of Hedley phosphorus pools in forest soils

2020-04-01JipengWANGYanhongWUJunZHOUHaijianBINGHongyangSUNJiLUOandShengyanPU2KeyLaboratoryofMountainSurfaceProcessesandEcologicalRegulationInstituteofMountainHazardsandEnvironmentChineseAcademyofSciencesChengdu6004China

Pedosphere 2020年2期

Jipeng WANG,Yanhong WU,Jun ZHOU,Haijian BING,Hongyang SUN,Ji LUO and Shengyan PU2Key Laboratory of Mountain Surface Processes and Ecological Regulation,Institute of Mountain Hazards and Environment,Chinese Academy of Sciences,Chengdu 6004(China)

2State Key Laboratory of Geohazard Prevention and Geoenvironment Protection,Chengdu University of Technology,Chengdu 610059(China)

ABSTRACTHedley labile phosphorus(P)pools in soil tend to be several times larger than annual forest requirements,even in highly weathered soils characterized by P limitation.The discrepancy between plant and soil P status could be partly attributable to the frequently adopted air-drying pretreatment that tends to increase soil P solubility.In this study,the effects of air-drying on the distribution of Hedley P fractions were examined using soils collected under 4 forest types at Gongga Mountain,southwestern China.The results showed that the microbial biomass P(Pmic)in the organic horizon decreased markedly after air-drying.The concentrations of Hedley labile P in the air-dried samples were 31%-73%more than those in the field-moist samples.Consequently,the air-drying-induced increments of Hedley labile P pools in the surface soil horizons were 0.8-3.8 times the annual plant P requirements.The organic horizon was more susceptible to the air-drying-induced increases in Hedley labile P than the mineral horizon,probably because of the stronger release of Pmic and disruption of soil organic matter.The quality of P,indexed by the ratio of Hedley labile P to slowly cycling P,shifted in favor of the Hedley labile fractions after air-drying,further revealing that air-drying changed the distribution of Hedley P pools in forest soils.These indicated that the effects of air-drying could not be ignored when interpreting the discrepancy between the P status of plants and the Hedley labile P pools in forest soils.To more efficiently evaluate the P status in forest soils via the Hedley fractionation procedure,the use of field-moist soils is recommended.

Key Words: Hedley fractionation,labile phosphorus,microbial biomass phosphorus,phosphorous limitation,slowly cycling phosphorus,soil pretreatment

INTRODUCTION

The Hedley sequential phosphorus (P) fractionation procedure has been widely used in assessing the status of P availability in terrestrial ecosystems (Negassa and Leinweber,2009).Generally,in Hedley fractionation,soil samples are treated with increasingly stronger extractants(Tiessen and Moir, 1993), and the P fractions including resin-extractable inorganic P and the 0.5 mol L-1NaHCO3-extractable inorganic and organic P are regarded as plant available(Johnson et al.,2003).However,there are discrepancies between the plant-available P extracted by Hedley fractionation (Hedley labile P) and plant growth. Recent studies suggested that Hedley labile P pools may be several times larger than the annual plant P requirement in forests,even in highly weathered soils characterized by P limitation(Johnson et al., 2003; Yang and Post, 2011). In contrast,Elser et al.(2007)noted,based on fertilization experiments,that P limitation prevailed in terrestrial ecosystems worldwide. Thus, the plant-available P in forest soils could be overestimated by the Hedley labile P pools.

The competition from microbes was regarded as the major reason why plants suffered from P limitation even when Hedley labile P was abundant(Yang and Post,2011).Additionally,we noticed that air-dried or oven-dried soils were used in the majority (76%) of the studies cited in the Hedley fractionation reviews (Cross and Schlesinger,1995;Johnson et al.,2003;Negassa and Leinweber,2009;Yang and Post, 2011). However, moist soils were used in the original Hedley fractionation (Hedley et al., 1982).Numerous studies focusing on the effect of air-drying in cropland and grassland soils indicated that air-drying could increase the solubility of labile P.On one hand,considerable proportions of microbial cells are killed during air-drying(Sparling et al., 1985; Bünemann et al., 2013), and the remaining cells can be further damaged by osmotic shock in the subsequent extraction process(Turner and Haygarth,2001; Turner et al., 2003a). These killing effects release labile P in inorganic and organic forms.On the other hand,the destabilization of soil aggregates and organic matter coatings on particle surfaces after air-drying and rewetting also increases the extractability of labile P (Bartlett and James, 1980; Bünemann et al., 2013). Forest soils are characterized by an organic horizon with high microbial biomass and soil organic matter(SOM)(Wood et al.,1984)and thus,samples from the organic horizon,especially those lacking dry-rewetting cycles in the field, may be highly susceptible to the effects of air-drying. Therefore, the airdrying procedure may partly explain the discrepancy between the P status of plants and the Hedley-available P pools in forest soils.

Most of the studies on the effects of air-drying were conducted with grassland or cropland soils(Sparling et al.,1985; Turner and Haygarth, 2001; Turner et al., 2003b;Styles and Coxon,2006;Soinne et al.,2010;Bünemann et al.,2013).In addition,little attention has been paid to the changes in slowly cycling P after air-drying(Soinne et al.,2010;Xu et al.,2011).As proposed by Richter et al.(2006),the slowly cycling Hedley P can be markedly bioavailable on decadal time scales. In this study, the effects of airdrying on P pools were examined using soils collected from the Hailuogou chronosequence where a primary vegetation succession had developed,from shrubs to broadleaf forests and then coniferous forests.The changes in Hedley labile and slowly cycling P following air-drying were determined and were related to soil properties.The aim of the study was to test how air-drying affected the distribution of P fractions in forest soils using a modified Hedley fractionation procedure.We hypothesized that 1) the lysis of soil microbial cells induced by air-drying would result in an increase in Hedley labile P;2)the air-drying effects would be more pronounced in the organic horizon with a higher microbial biomass;and 3)the distribution of Hedley P fractions would shift in favor of the labile P.

MATERIALS AND METHODS

Study area

Gongga Mountain(29◦20′-30◦20′N,101◦30′-102◦15′E; summit: 7 556 m above sea level (a.s.l.)) is located in the transition zone of the Tibetan Plateau and the Sichuan Basin,China.The Hailuogou Glacier is a monsoonal glacier stretching down the eastern slope of Gongga Mountain,and it has gradually retreated since the end of the Little Ice Age.In the glacier retreat area,a chronosequence developed from the terminal moraine and extended to the northeast for approximately 2 km.Three sites(Sites 1-3,45-125 years after glacier retreat)were selected in the retreat area and one reference site(Site 4,about 1 400 years after glacier retreat)was selected about 0.6 km further down the Hailuogou valley(Fig.1).The climate in the study area is mainly controlled by the Asian monsoon(Bing et al.,2014).The mean annual temperature is 4.2◦C and the mean annual precipitation is 1 947 mm in the Gongga Mountain Alpine Ecosystem Observation Station, Chinese Academy of Sciences (29◦34′34.69′′N,101◦59′55.08′′E;3 000 m a.s.l.)(Wu et al.,2013).The vegetation is broadleaf forest(dominated by Populus purdomii Rehder)at Site 1,conifer forest at Sites 2 and 3(dominated by Abies fabri(Mast.)Craib and Picea brachytyla(Franch.)E.Pritz, respectively), and mixed conifer-broadleaf forest(dominated by P.brachytyla(Franch.)E.Pritz.and Fargesia spathacea Franch)at Site 4.The soils in the retreat area and the reference site are classified as Regosols and Podzols,respectively,according to the World Reference Base for Soil Classification(IUSS Working Group WRB,2006).The soil moisture content is above 20%throughout the year in the glacier retreat area(Luo et al.,2012).

Soil sampling

The soil samples were collected in May 2016.During the field sampling, 3 soil profiles were hand-dug at each site and field-moist samples(approximately 50 g)of the Oa(humified litter layer) and A horizons (top 1.5 cm of the mineral soils with a dark brown color and humus enrichment)were collected in each profile. The sampling locations at each site were randomly selected and were at least 10 m apart from each other.The samples were transported in coolers(with ice inside)to the laboratory.

Fig.1 Location of the study area and sampling sites at Gongga Mountain,southwestern China.Numbers in the parentheses are years after glacier retreat.

Chemical analysis

The samples were passed through a 2-mm sieve. One subsample was air-dried at 25◦C for 14 d,and the remaining portion (field-moist) was stored in polyethylene bags in a 4◦C refrigerator for 14 d before Hedley fractionation.Soil properties and P fractions,except the total C and N,were measured using both field-moist and equivalent air-dried samples.Subsamples of the air-dried materials were milled and analyzed for total C and N using an element analyzer(Elementar vario MACRO cube,Elementar,Germany).The pH was measured in suspensions at ratios of 1:10 and 1:2.5(air-dry weight:volume)for the Oa and A horizons,respectively.The oxalate-extractable iron(Feox),aluminum(Alox),and P (Pox) were extracted with NH4-oxalate according to Courchesne and Turmel (2006), and the Fe, Al, and P concentrations were determined using an inductively coupled plasma-optical emission spectroscopy(Optima 8300,PerkinElmer,USA).The degree of P saturation(DPS,%),which was related to the P sorption capacity of soils,was calculated as DPS = Pox/[0.5 × (Feox+ Alox)] × 100, in which the unit for Pox, Feox, and Aloxwas mol g-1soil(Beauchemin and Simard,1999).

The microbial biomass P (Pmic) was measured as described by Wu et al. (2000), with slight modifications.Field-moist samples(2 g)or their equivalent air-dried samples were weighed into 100-mL centrifuge tubes and exposed to chloroform(CHCl3)vapor for 24 h in a vacuum desiccator at room temperature.After the residual CHCl3was removed,20 mL Bray-1 extractants(0.03 mol L-1NH4F+0.025 mol L-1HCl) were added to the tubes, and the mixture was rotary shaken for 30 min.Suspensions were centrifuged at 6 000×g for 10 min,and the orthophosphate concentrations in the supernatants were determined using the method of Murphy and Riley(1962).Equivalent portions of unfumigated samples were also extracted with Bray-1 extractants.The Pmicwas calculated by subtracting the unfumigated P concentration from the fumigated P concentration and then corrected using the conversion factor(KP=0.4)(Jenkinson et al.,2004).No correction was conducted for the incomplete recovery of the Pmicusing added inorganic P according to Wu et al.(2000).

Hedley P fractionation was conducted according to Tiessen and Moir(1993),with slight modifications(Fig.S1,see Supplementary Material for Fig. S1). The field-moist samples (1 g) or their equivalent air-dried samples were successively extracted with two resin strips in deionized water, 0.5 mol L-1NaHCO3, 0.1 mol L-1NaOH,1 mol L-1(diluted)HCl,and hot concentrated HCl.The extracted P fractions were designated as resin-P,NaHCO3-P,NaOH-P,HCldil-P,and HClconc-P,respectively.For the extraction of residual P,the remaining samples were oven-dried at 60◦C,of which approximately 0.1 g were digested with concentrated HNO3(6 mL)and HF(2 mL)in a microwave oven(Mars 6,CEM,USA).Except for residual P,the concentrations of inorganic P(Pi)were determined according to Murphy and Riley(1962),and those of the total P(Pt)were determined using the same method after digestion of the extract with(NH4)2S2O8under acidic conditions.The concentrations of organic P(Po)were calculated as the differences between those of the Ptand Pi. For the determination of the residual P,the P concentrations in the solutions after digestion were measured using an inductively coupled plasma-mass spectrometry(NexION 300X,Perkin Elmer,USA).

All the soil properties above were determined with three replicates and expressed based on air-dry weight.

Statistical analysis

A paired two-sample t-test was used to compare the P concentrations between the field-moist and air-dried samples and compare sample properties between soil horizons. A simple linear regression and Spearman correlation were applied to establish the relationships among the changes in P fractions, and between changes in P fractions and soil properties.Differences in soil properties among sampling sites were analyzed using a one-way analysis of variance(ANOVA)and post-hoc Ducan’s multiple range test.Values of P <0.05 were considered significant. Analyses were performed using SAS 9.1 software(SAS Institute,USA)and Origin 8.0 software package(Origin Lab,USA).

RESULTS

Soil properties

Soil properties in the study area are presented in Table I.Soil total C(3-321 g kg-1),total N(0.3-22.3 g kg-1),and Pmic(field-moist, 3-376 mg kg-1)were higher in the Oa horizon than in the A horizon;at different succession stages,they were higher at Sites 1 and 2 compared to Sites 3 and 4 in the Oa horizon, and were the highest at Site 4 in the A horizon.Soil pH(field-moist,4.5-6.3)of both horizons displayed the same trend with forest succession and was higher at Sites 1 and 3 than at Sites 2 and 4. Soil Feox(field-moist,1.70-6.30 g kg-1),Alox(field-moist,0.37-2.56 g kg-1), and DPS (field-moist, 6%-30%) were generally the highest at Site 4, whereas they showed no consistent variations between soil horizons.

Some of the soil properties changed after air-drying.Soil pH was significantly lower(paired t-test,P <0.000 1)in air-dried (mean 5.2 and 5.6 in the Oa and A horizons,respectively) than field-moist soils (mean 5.6 and 5.9 in the Oa and A horizons,respectively)at Sites 1-3,whereas an opposite trend was observed at Site 4 (Table I). ThePmicof the Oa horizon decreased markedly(paired t-test,P <0.000 1) after air-drying, from mean value of 216 mg kg-1in the field-moist soils to 44 mg kg-1in the airdried soils(mean decrease by 80%).This decreasing trend also occurred in the A horizon,but it was not statistically significant(Table I).Soil Feoxdid not change after air-drying,whereas Aloxslightly decreased(paired t-test,P <0.01)in both the Oa(from mean of 1.34 g kg-1in the field-moist soils to 1.26 g kg-1in the air-dried soils)and A horizons(from mean of 0.99 g kg-1in the field-moist soils to 0.94 g kg-1in the air-dried soils) (Table I). Soil DPS of the Oa horizon displayed no consistent change after air-drying;however,soil DPS of the A horizon decreased(paired t-test,P <0.001) from a mean value of 19% in the field-moist soils to 17%in the air-dried soils.

TABLE ISoil propertiesa) of field-moist and air-dried soils in the Oa and A horizons at four sampling sites(Sites 1-3 in the Hailuogou Glacier retreat area and Site 4 down the Hailuogou valley)

Changes in P fractions after air-drying

Generally,no significant change was observed in total P,total Pi,and total Poafter air-drying(Fig.2).Fractions of the Hedley labile P changed significantly after air-drying except for the NaHCO3-Piof the A horizon(Fig.3).Average resin-Piconcentration of the Oa horizon increased markedly from 4 to 37 mg kg-1(mean increase by 782%)after airdrying,and that of the A horizon increased from 4 to 8 mg kg-1(mean increase by 116%)(Fig.3).Average NaHCO3-Piconcentration of the Oa horizon decreased from 38 to 25 mg kg-1(mean decrease by 33%)after air-drying,whereas that of the A horizon did not change significantly(Fig.3).Average NaHCO3-Poconcentration of the Oa horizon increased significantly from 36 to 53 mg kg-1(mean increase by 49%)after air-drying,and that of the A horizon increased from 13 to 21 mg kg-1(mean increase by 58%)(Fig.3).Fractions of the slowly cycling P showed no consistent changes after air-drying(Fig.4).Average NaOH-Piconcentration of the Oa horizon decreased slightly from 72 to 63 mg kg-1after air-drying(mean decrease by 12%),whereas that of the A horizon did not change significantly (Fig. 4). Changes in NaOH-Po,HCldil-Pi,and HClconc-Piof both horizons were not significant after air-drying(Fig.4).Average HClconc-Poconcentration of the Oa horizon decreased slightly from 98 to 89 mg kg-1after air-drying(mean decrease by 9%),and that of the A horizon decreased from 44 to 37 mg kg-1(mean decrease by 15%)(Fig.4).Average residual P concentration of the Oa horizon was not affected by air-drying;however,that of the A horizon increased from 144 to 152 mg kg-1(mean increase by 6%)(Fig.4).

Fig.2 Concentrations of total P,total inorganic P(Pi),and total organic P(Po)in the Oa and A horizons of the field-moist and air-dried soils at four sampling sites(Sites 1-3 in the Hailuogou Glacier retreat area and Site 4 down the Hailuogou valley).Total P is the sum of all P fractions obtained in Hedley sequential extraction;total Pi is the sum of resin-,NaHCO3-,NaOH-,diluted HCl-,and concentrated HCl-extractable Pi;and Total Po is the sum of NaHCO3-,NaOH-,and concentrated HCl-extractable Po.Asterisks indicate significant differences between field-moist and air-dried soils at a site(n=3,paired t-test).Values of P <0.05 indicate significant differences between field-moist and air-dried soils when the samples from Sites 1-4 are pooled together(n=12,paired t-test).

DISCUSSION

Effects of air-drying on soil properties

Although widely adopted for research convenience,airdried soil samples go through changes in properties that may cause the redistribution of P fractions compared with field-moist soils. Soil pH is affected by a range of soil properties(e.g.,activities of Ca,Mg,Al,OH-,P,CO2,and organic associated protons),which change during air-drying(Bartlett and James,1980).The inconsistent changes in soil pH between Sites 1-3 and Site 4 after air-drying suggested the variable effects of air-drying on pH(Table I).In grassland soils,microbial biomass C(Cmic)decreased by 11%-68%after air-drying(Sparling et al.,1985),and a stronger dryinginduced reduction in Pmic(by 69%) than Cmic(by 39%)was reported by Bünemann et al. (2013); in addition, the osmotic shock during rewetting of the air-dried soils may lead to further lysis of the remaining microbial cells(Turner et al.,2003a).In the forest soils of this study,a marked and a weak decrease,respectively,in the Pmicof the Oa horizon(by 80%)and of the A horizon(by 13%)were observed after air-drying (Table I), which were apparently related to the concentrations of total C (r = -0.90, P <0.000 1) and Pmic(field-moist, r = -0.99, P <0.000 1). As reported by Haynes and Swift(1985), soil Feoxand Aloxwere not or slightly affected by short-term(14 d)air-drying in this study(Table I).However,an increasing trend may occur over time.For example,Peltovuori and Soinne(2005)reported significant increases in Feoxand Aloxafter long-term(22 months)air-drying.The slight decrease in Alox(amorphous Al with high P sorption ability)in this study may be caused by the increased crystallinity of Al oxides after air-drying(McLaughlin et al.,1981;Chepkwony et al.,2001).Despite the decrease in Alox, P sorption capacity of soils did not change in the Oa horizon or increase in the A horizon as indicated by the DPS values (Table I). This supported previous findings (Bartlett and James, 1980; Haynes and Swift,1985;Peltovuori and Soinne,2005)and was probably attributable to the exposure of previously protected surfaces(Turner et al.,2003b).

Fig.3 Concentrations of Hedley labile P in the Oa and A horizons of the field-moist and air-dried soils at four sampling sites(Sites 1-3 in the Hailuogou Glacier retreat area and Site 4 down the Hailuogou valley).Hedley labile P includes resin-and NaHCO3-extractable inorganic P(resin-and NaHCO3-Pi,respectively)and NaHCO3-extractable organic P(NaHCO3-Po).Asterisks indicate significant differences between field-moist and air-dried soils at a site(n = 3,paired t-test).Values of P <0.05 indicate significant differences between field-moist and air-dried soils when the samples from Sites 1-4 are pooled together(n=12,paired t-test).

Effects of air-drying on total P

Soil total P is considered to be unaffected by the way of soil pretreatment(e.g.,field-moist,air-drying,and freezing)(Xu et al.,2011).Our results also demonstrated that air-drying did not change total P (Fig. 2). Therefore, the differences in P fractions between field-moist and air-dried samples probably originated from the redistribution of Hedley P pools, instead of the changes in the P recovery rate caused by soil pretreatment.

Effects of air-drying on Hedley labile P

TABLE IIHedley labile P pools in the Oa and A horizons of the field-moist and air-dried soils at four sampling sites(Sites 1-3 in the Hailuogou Glacier retreat area and Site 4 down the Hailuogou valley),and annual plant requirement of P

Despite the marked differences in soil properties between soil horizons and among sites(Table I),the Hedley labile P clearly increased after air-drying(Fig.3, Table II).Based on Hedley labile P concentrations, the air-drying effects were more pronounced in the organic than mineral horizons(Table II).Our results supported previous findings that labile P in organic matter-rich soils was highly susceptible to airdrying(Sparling et al.,1985).For example,Xu et al.(2011)found a stronger response of Hedley labile P to air-drying in forest soils (53.1 g C kg-1soil) when comparing with grassland sites (2.7 g C kg-1soil); using organic matterrich soils from the grassland of western Ireland,Styles and Coxon(2006)observed large increases in P solubility after air-drying. Based on Hedley labile P pools of the surface horizons,the increments induced by air-drying were 0.8 to 3.8 times the annual plant requirement(Table II).Considering that 76% of Hedley fractionation studies were conducted with dry soils,the air-drying effects may partly explain the findings of Johnson et al.(2003)and Yang and Post(2011)that the pools of Hedley labile P in soil were several times larger than annual plant requirements.

Increases in resin-Piafter air-drying were more pronounced than changes in other Hedley labile fractions(Fig.3).This was consistent with previous findings of Styles and Coxon(2006)and Xu et al.(2011)that increases in desorbable P(similar to the resin-Piin this study)or resin-Picontributed to the majority of the air-drying-induced changes in labile P.The release of Pmicduring air-drying and rewetting is thought to be responsible for the high labile P extracted from air-dried soils(Turner and Haygarth,2001;Bünemann et al.,2013).In this study,the air-drying-induced changes in resin-Piand Pmicwere strongly negatively correlated(Fig.5,Tables III and IV), suggesting the microbial origin of the increased resin-Pi. The released Pmicoccurs in forms of orthophosphate and labile Po(Oehl et al.,2001;Richardson and Simpson,2011).Wang et al.(2016)reported high phosphomonoesterase activities in the Oa horizon of our study area,which may lead to quick mineralization of the released microbial Poduring air-drying.Regression analysis(Fig.5)indicated that approximately 45%of the released Pmicwas recovered in resin-Pi(∆resin-Pi= -0.18∆Pmic+ 0.53,assuming a conversion factor of 0.4 between chloroformextractable P and Pmic)(Jenkinson et al.,2004),which was comparable to the recovery rate(38%)reported by Bünemann et al.(2013).The partial recovery of Pmicby resin extraction and the significant relationships between the increases in resin-Piand Feoxor DPS(Tables III and IV)suggested the sorption of the released Pmicduring air-drying(Chepkwony et al.,2001;Styles and Coxon,2006).This is supported by the phenomena that the amount of the released Pmicextracted as orthophosphate differs depending on the strength of the extractants used. For example, only trace amounts of the orthophosphate in air-dried soils were extracted by water(Turner and Haygarth,2001;Blackwell et al.,2009),whereas almost all the released Pmicwas extracted as orthophosphate by 0.5 mol L-1NaHCO3(Sparling et al.,1985).

TABLE IIISpearman correlation coefficients for relationships between soil properties and changes in P fractions following air-drying in soil Oa horizon

TABLE IVSpearman correlation coefficients for relationships between soil properties and changes in P fractions following air-drying in soil A horizon

The air-drying effects led to different relationships between the Pmicand resin-Piin the field-moist and air-dried samples (Fig. 6). No linear relationships existed between the Pmicand resin-Piin the field-moist samples.However,possibly as an effect of Pmicrelease,positive relationships occurred in the air-dried soils, which might lead to questionable inferences that Pmicwas tightly associated with the amount of resin-Piin the field.Similarly,the relationships between the total C and resin-Piwere different between the field-moist and air-dried soils (Fig. 6). Other studies also reported positive relationships between the C content and resin-Piin the air-dried soils(e.g., R2=0.21, P <0.01,Tiessen et al.,1984; R2=0.87, P <0.000 1,Thomas et al.,1999;R2=0.61,P <0.000 1,Cassagne et al.,2000),and these relationships should be interpreted with caution due to the air-drying effects.

Fig.6 Relationships between microbial biomass P(Pmic)and resin-extractable inorganic P(resin-Pi)and between total C and resin-Pi in the Oa and A horizons of the field-moist and air-dried soils.R2adj =adjusted coefficient of determination.

The decrease in NaHCO3-Piin the Oa horizon after air-drying(Fig.3)was unexpected,considering the overall promotion effect of air-drying on P solubility reported in the literature.In this study,the air-drying-induced changes in NaHCO3-Piwas not comparable to those reported in Sparling et al.(1985),Turner et al.(2003b),and Soinne et al.(2010),since no resin extraction was conducted prior to the NaHCO3extraction in these studies.Xu et al.(2011),using the same extraction procedure as in this study,detected no significant difference in NaHCO3-Pibetween the air-dried and fieldmoist soils.One explanation for a lower NaHCO3-Piin the air-dried organic soils may be the increased crystallinity and reduced P sorption capacity of Al oxides (Table I)(McLaughlin et al.,1981;Chepkwony et al.,2001).Another explanation may be that a portion of the Pmicrecovered by the NaHCO3extraction of field-moist soils was released during air-drying and recovered by the resin extraction. In moist soils,NaHCO3-Picontains up to 15%of the Pmic(Hedley et al.,1982).In addition,changes in NaHCO3-Piand in Pmicafter air-drying were positively related(Table III),suggesting the role of Pmicrelease in the decrease of NaHCO3-Pi.The released Pmicmay be recovered by the resin extraction as reflected in the negative relationship between changes in NaHCO3-Piand in resin-Pi(Table III).

The significant increase in NaHCO3-Poobserved in this study (Fig. 3) agreed with the studies in which grassland soils were used (Turner et al., 2003b; Xu et al., 2011).Two processes may account for the increased solubility of Pofollowing air-drying, one being the lysis of microbial cells, and the other being the physical disruptions of soil aggregates and organic matter coatings on particle surfaces(Turner et al.,2003b; Bünemann et al.,2013).Consistent with the findings of Turner et al.(2003b),our data showed no significant relationship between changes in NaHCO3-Poand the soil properties measured,including the changes in Pmic(Tables III and IV).Similarly,Blackwell et al.(2009)found no direct link between the molybdate unreactive P in the soil leachate and drying-induced decreases in microbial biomass.Therefore,it is reasonable to infer that the physical disruption of soils during air-drying and rewetting was important for the increase in NaHCO3-Poin this study. This inference was supported by the source of NaHCO3-Poin the airdried soils.For example,Zhang et al.(1999)reported that orthophosphate monoesters (likely originated from dead organic matter),instead of orthophosphate diesters(the main Poforms in microbial cells),dominated the NaHCO3-Poin air-dried Canadian arable soils.

Effects of air-drying on slowly cycling P

Richter et al.(2006)reported marked bioavailability of slowly cycling Piand Poassociated with Fe and Al oxides and Ca minerals on decadal time scales.Nevertheless,few studies have been devoted to the changes in the slowly cycling P after air-drying.In grassland and cultivated soils,the amount of 0.1 mol L-1NaOH-extractable P did not change after air-drying,whereas the size distribution shifted in a way that small-sized P increased at the expense of the large-sized P (Soinne et al., 2010). In forest and organic matter-poor grassland soils, air-drying induced increases in slowly cycling Poand decreases in slowly cycling Pi(Xu et al.,2011).Our data showed inconsistent changes in slowly cycling P fractions after air-drying, i.e., NaOH-Piand HClconc-Podecreased,whereas other fractions did not change or slightly increased(Fig.4).The transformations between slowly cycling and Hedley labile P fractions were inconclusive based on our data,with only correlation analyses(Tables III and IV).However,the quality of P,indexed by the ratio of Hedley labile P to slowly cycling P(Richter et al.,2006),shifted in favor of the Hedley labile fractions after air-drying(Fig.7).Thus,air-drying changed the distribution of Hedley P pools and increased the portion of the Hedley labile P at the expense of the slowly cycling P.

Fig.7 Ratios of Hedley labile inorganic P(Pi)and organic P(Po)relative to their corresponding slowly cycling fractions in the Oa and A horizons of the field-moist and air-dried soils at four sampling sites(Sites 1-3 in the Hailuogou Glacier retreat area and Site 4 down the Hailuogou valley).Hedley labile Pi includes resin-and NaHCO3-extractable Pi,and Hedley labile Po includes NaHCO3-extractable Po;slowly cycling Pi includes NaOH-,diluted HCl-,and concentrated HCl-extractable Pi,and slowly cycling Po includes NaOH-and concentrated HCl-extractable Po.Asterisks indicate significant differences between field-moist and air-dried soils at a site(n=3,paired t-test).Values of P <0.05 indicate significant differences between field-moist and air-dried soils when samples from Sites 1-4 are pooled together(n=12,paired t-test).

The data obtained in this study may call for a reconsideration of the concepts of labile and occluded P depending on the pretreatment of soils and soil moisture conditions in the field. On one hand, the labile P in soils lack of moisture stress may be overestimated by air-drying treatment. For example, the resin-Piin air-dried soils may be derived from Pmic(Fig.5).Instead of serving as a source of readily available P,microbes can compete with plants for dissolved orthophosphate at the late stage of forest development(Turner et al., 2013). On the other hand, some Hedley P fractions commonly regarded as highly occluded may be prone to air-drying and rewetting events.For example,part of the HClconc-Poin the field-moist soils may partly come from particulate organic matter(Tiessen and Moir, 1993)or from microbial biomass (Table IV, as reflected by the positive relationship between changes in HClconc-Poand in Pmic)(Hedley et al.,1982),and significantly decreased after air-drying(Fig.4).Under climate conditions where soil undergoes extreme fluctuations in water contents,such as Mediterranean climate and future climate scenarios(Butterly et al.,2009; Bünemann et al.,2013),the occluded P may undergo fast transformation and become available for plants.

CONCLUSIONS

The use of air-dried soils in Hedley fractionation distorts the distribution of the Hedley P pools in forest soils,especially in the organic horizon.The marked increase in Hedley labile P after air-drying was attributable to the release of Pmicand the disruptions of soil aggregates and organic matter coatings on particle surfaces.Thus,in addition to the competition from microbes in the field,the air-drying treatment of samples may be another reason why plants suffer from P limitations even when Hedley labile P is abundant.Based on the results of this study,the use of field-moist soils is recommended when Hedley P fractionation or other wet chemical P extraction methods are used to assess the P status in forest soils.Furthermore,a reconsideration of the concepts of labile and occluded P is needed depending on the pretreatment of soils and soil moisture conditions in the field.

ACKNOWLEDGEMENT

This research was supported by the National Natural Science Foundation of China (Nos.41630751, 41401253 and 41471416).

SUPPLEMENTARY MATERIAL

Supplementary material for this article can be found in the online version.


登录APP查看全文