当前位置：文档库 › Effects of carbon additions on iron reduction and phosphorus availability

Effects of carbon additions on iron reduction and phosphorus availability

Effects of carbon additions on iron reduction and phosphorus availability in a humid tropical forest soil

Daniel Liptzin *,Whendee L.Silver

Department of Environmental Science,Policy,&Management,University of California -Berkeley,137Mulford Hall #3114,Berkeley,CA 94720,USA

a r t i c l e i n f o

Article history:

Received 20February 2009Received in revised form 20May 2009

Accepted 25May 2009

Available online 17June 2009Keywords:Iron cycling

Phosphorus limitation Tropical soils Carbon cycling

a b s t r a c t

In the highly weathered soils of humid tropical forests,iron (Fe)plays a key role in ecosystem biogeo-chemical cycling through its interactions with carbon (C)and phosphorus (P).We used a laboratory study to explore the role of C quantity and quality in Fe reduction and associated P mobilization in tropical forest soils.Soils were incubated under an ambient atmosphere headspace (room air)with multiple levels of leaf litter leachate or acetate https://www.wendangku.net/doc/bb6246682.html, Fe reduction occurred in all the treatments and at every time point.The more complex mixture of organic compounds in leaf litter leachate stimulated Fe reduction as much acetate,an easily fermentable C source.At the end of the experiment,Fe reduction was generally greater with higher C additions than in the low C additions and controls.The microbial biomass P had increased signi?cantly suggesting rapid microbial uptake of P liberated from Fe.This occurred without increases in the available (NaHCO 3)P pool.The immobilization of P by microbes during the incubation provides a P conservation mechanism in these soils with ?uctuating redox potential,and may ultimately stimulate more C cycling in these highly productive ecosystems.Iron cycling appears to be an important source of P for the biota and can contribute signi?cantly to C oxidation in upland tropical forest soils.

1.Introduction

Tropical forests have among the highest rates of net primary productivity in the world,yet occur on some of the most phosphorus (P)-de?cient soils (Vitousek and Sanford,1986).Phosphorus fertilization studies provide evidence of P limitation to soil respiration (Cleveland and Townsend,2006),microbial activity (Cleveland et al.,2002),litter decomposition (Kaspari et al.,2008;Hobbie and Vitousek,2000)and net primary productivity (Herbert and Fownes,1995).Phosphorus limitation likely derives in part from the high concentrations of iron (Fe)and aluminum (Al)oxides and hydroxides in strongly weathered tropical soils (Vitousek and Sanford,1986).These oxidized secondary minerals can bind P making it temporarily unavailable for plants and microbes (Sanchez,1976;Cross and Schlesinger,1995).One major difference between Fe and Al is that Fe is redox active and the reductive dissolution of secondary Fe minerals tends to liberate P (Baldwin and Mitchell,2000).Thus,the redox dynamics of Fe may play an important role in P cycling and biological activity in these ecosystems.

Iron redox dynamics are well documented in saturated,largely anoxic environments like marine and freshwater sediments and submerged soils (Lovley,1991;Thamdrup,2000).In sediments,Fe reduction typically occurs at a speci?c depth related to the deple-tion of electron acceptors with higher redox potential (e.g.Can?eld et al.,1993).The spatial pattern of Fe cycling in wetlands often depends on the distribution of plants.Iron reduction is greater in vegetated soils,particularly in the rhizosphere (Weiss et al.,2004;Roden and Wetzel,1996).In these saturated environments,Fe cycling has been linked to P availability (Ponnamperuma,1972;Patrick and Khalid,1974;Szilas et al.,1998),nitrogen (N)availability (Sahrawat,2004),and carbon (C)oxidation (Dettling et al.,2006;Weiss et al.,2004;Roden and Wetzel,1996).

Upland tropical forest soils are also likely to support rapid rates of Fe cycling.The deeply weathered soils,typically oxisols or ulti-sols,tend to be ?nely textured and rich in secondary Fe oxide minerals (Vitousek and Sanford,1986).The high biological activity typical of humid tropical forests,coupled with warm temperatures and abundant rainfall,leads to periodic O 2depletion,where O 2is consumed faster than it can rediffuse into soils (Silver et al.,1999;Schuur and Matson,2001).This ?uctuation in redox potential,together with high net primary productivity and pulsed inputs of labile C drive the temporal patterns of microbial activity and biogeochemical cycling (Lodge et al.,1994).During periods of low

*Corresponding author.Tel.:t151********;fax:t151********.E-mail address:liptzin@https://www.wendangku.net/doc/bb6246682.html, (D.

Liptzin).Contents lists available at ScienceDirect

Soil Biology &Biochemistry

journal hom epa ge:

https://www.wendangku.net/doc/bb6246682.html,/locate/soilbio

Soil Biology &Biochemistry 41(2009)1696–1702

O2concentrations,Fe reduction has the potential to release Fe-bound P.There are several potential fates of the liberated P including readsorption on soil exchange sites,precipitation as vivianite,leaching,or immobilization by plants and microbes. When soils are re-oxygenated,solid phase Fe-oxides are formed again completing the Fe cycle(Thompson et al.,2006).The Fe and C cycles are also linked.Iron reduction is a source of CO2via heterotrophic respiration during anoxic periods(Weber et al., 2006).Thus,Fe cycling in soils with?uctuating redox has the potential to affect C cycling directly when it acts as a terminal electron acceptor for soil microbes,and indirectly by altering the availability of the nutrient potentially limiting to plants and microbes.

Several factors may control the rates of Fe reduction in soils. First,Fe reduction rates may be affected by the concentration and phase of the Fe(III)minerals present.Poorly crystalline phases of Fe(III)like ferrihydrite are thought to be more reducible than more crystalline phases(Roden and Urrutia,2002).Although Fe reduc-tion rates were found to increase with initial Fe(III)concentrations in incubated sediments(Roden and Wetzel,2002),this relationship was not found in soils(Ku¨sel et al.,2002).Second,the presence of organic molecules that can shuttle electrons from the organisms to solid phase Fe minerals affects Fe reduction rates(Lovley et al., 1996).The addition of anthraquinone-2,6-disulfonate(AQDS), a humic acid analog,has been found to stimulate Fe reduction in soils(Chaco′n et al.,2006;Peretyazhko and Sposito,2005;Weiss et al.,2004).Further,the addition of these electron shuttles may increase the reducibility of more crystalline Fe phases(Lovley et al., 1998).As it is a heterotrophic process,Fe reduction appears to be stimulated by the addition of labile C compounds such as acetate or glucose(Chaco′n et al.,2006;Ku¨sel et al.,2002).

A growing number of studies have estimated a potential rate of Fe reduction in suspensions under anaerobic conditions(Teh et al., 2008;Dubinsky,2008;Thompson et al.,2006;Chaco′n et al.,2006; Peretyazhko and Sposito,2005;Ku¨sel et al.,2002).Several studies have also demonstrated increases in pH and soluble P in association with Fe(II)production with and without the addition of simple C substrates or synthetic electron shuttle compounds(Chaco′n et al., 2006;Thompson et al.,2006;Peretyazhko and Sposito,2005). However,no studies to date have quanti?ed Fe reduction potential under oxic atmospheric conditions in unsaturated soils.Tropical soils harbor a complexity of microsite conditions that cover a wide range in redox potential,labile C pools,and Fe species(Teh et al., 2008;Silver et al.,1999).To better understand the role of Fe reduction in C and P dynamics in soils we need to determine how these biogeochemical cycles are coupled under oxic and suboxic conditions.In this study we conducted a laboratory experiment to investigate the linkages among Fe,C,and P in an upland tropical forest soil.We examined changes in extractable Fe and P pools after the pulsed addition of acetate or leaf litter leachate with an oxic headspace.Our research tested the hypothesis that rates of Fe reduction are a function of labile C availability in weathered soils with high Fe(III).We also hypothesized that Fe reduction would not increase mineral P pools due to rapid microbial immobilization in these low P soils.

2.Methods

2.1.Study area

Soils were collected at a valley site within the Bisley Research Watersheds at an elevation of approximately400m asl(Silver et al., 1999).These watersheds are within the Luquillo Experimental Forest(LEF)and are part of the NSF-sponsored Long Term Ecolog-ical Research network.Precipitation is evenly distributed throughout the year with an annual total of3500mm of rainfall (Brown et al.,1983).The highly leached soils in the LEF range from inceptisols to oxisols and have developed on Cretaceous volcani-clastic sediments that were intruded by a quartz diorite pluton (Frizano et al.,2002).Within the Bisley watersheds the soils are generally ultisols in the Cristal/Humatus/Zarzal complex(Scatena, 1989).Topographic position(ridge,slope,valley)plays a large role in controlling ecosystem processes in this highly dissected land-scape(Silver et al.,1994).The valley sites,such as in the current study,are well drained but may contain both red and gray/white mottles(Scatena,1989).These sites have?uctuating O2concen-trations and average<10%O2at a depth of10cm;the presence of anoxic microsites in space and time is illustrated by net methane and nitrous oxide production in these soils(Silver et al.,1999).

2.2.Soil sampling and incubation

Soils(0–10cm depth)were collected from a200cm2area,and shipped immediately to UC Berkeley.Our goal was to use a rela-tively homogeneous soil sample that would allow us to replicate initial conditions among treatments.Soils were homogenized and large roots and rocks were removed by hand.Approximately40g of wet soil were pressed into47cm diameter Taral vials to approxi-mate?eld bulk density(0.65g cmà3).The vials were placed into pint size Mason jars and covered with a polyethylene bag to limit moisture loss and to prevent buildup of trace gases and depletion of O2in the headspace.The cores were allowed to equilibrate for24h prior to the start of the treatments.Five cores were harvested for initial conditions after this pre-incubation period.The remaining cores received one of six solutions:control(DI water),leaf litter leachate(50,100,150,or200mg C Là1),or acetate(200mg C Là1). In total,the soils received of0,25,50,75,or100m g C g dry soilà1. The leaf litter for the leachate was collected from the soil surface near the soil collection sites.The litter was pulverized in a blender, shaken overnight,and?ltered through Whatman41ashless?lter paper.The total C added was similar to the total amount added by Chaco′n et al.(2006)to simulate small pulses of C to soils that occur naturally at the site.The solutions were added,after removing the polyethylene bags,by pipetting1.5mL of the appropriate solution on the soil surface approximately twice weekly with a total volume of9mL added.While the polyethylene bags limited evaporative loss,the soils did lose some soil moisture between the additions of the solutions,typically<5%of the soil water.The initial gravimetric soil moisture was100%with the mean on other dates ranging from 96to103%.At selected times(2,6,12,and27days)three replicate cores per treatment were destructively harvested for soil properties.

2.3.Soil analyses

To quantify net Fe(II)production,we measured acid-extractable Fe(II)using0.5M HCl(Lovley and Phillips,1987).Ferrous iron was quanti?ed using a modi?ed Ferrozine method because of the high concentration of Fe(III)in the extracts(Viollier et al.,2000;Stookey, 1970).Brie?y,100uL of the extract is added to2mL of color reagent (1g Là1Ferrozine in50mM HEPES buffer pH8)and the absorbance at562nm was measured.To minimize Ferrozine-catalyzed reduc-tion of Fe(III),the absorbance was read immediately on a spectro-photometer(Milton Roy,USA)with a?nal solution pH of7(Pullin and Cabaniss,2003).Because Fe(III)also binds with Ferrozine, a second solution is made consisting of the extract and color reagent mixed with100uL of10%hydroxylamine hydrochloride. The reduction of Fe occurs essentially instantaneously and the absorbance at562nm is read again.Following Viollier et al.(2000), the concentrations of Fe(II)and Fe(III)were calculated as

D.Liptzin,W.L.Silver/Soil Biology&Biochemistry41(2009)1696–17021697

C FeeIIT?

A13FeeIITlàA23FeeIIITl 3FeeIITl

3FeeIITlà3FeeIIITl

C FeeIIIT?

A2àA1

3FeeIITlà3FeeIIITl

Where A1is the measured absorbance before reduction,A2is the measured absorbance after reduction,3Fe(II)and3Fe(III)are the measured molar absorption coef?cients for Ferrozine for Fe(II)and Fe(III),and l is the path length.

Poorly crystalline Fe was quanti?ed with a citrate/ascorbate(C/A) extraction(Reyes and Torrent,1997).This extraction is preferable to acid ammonium oxalate extraction because oxalate has been found to extract more crystalline Fe(III)phases as Fe(II)concentra-tions increase(Schwertmann,1991).Iron concentrations in C/A extracts were measured on an inductively coupled plasma atomic emission spectrometer(Perkin–Elmer,USA).Phosphorus concentra-tions were determined with0.5M sodium bicarbonate(NaHCO3-P) to measure labile P and a separate extraction with0.1M sodium hydroxide(NaOH)to quantify more recalcitrant Al-and Fe-bound P (Tiessen and Moir,1993).We report NaOH P as the difference in the two extractions following Tiessen and Moir(1993).Microbial biomass P was determined as the difference in P extracted from chloroform fumigated soils and the unfumigated NaHCO3extrac-tions following Hedley and Stewart(1982).A K p factor of0.4was used;the resin pretreatment was not performed because of the low concentrations of P in the soils(Hedley and Stewart,1982; McGroddy and Silver,2000).The procedure was modi?ed such that chloroform vapor in a dessicator was used instead of adding liquid chloroform directly to the soils.All P extracts were analyzed for PO4 according to Murphy and Riley(1962)after digestion in acidic ammonium persulfate(Tiessen and Moir,1993).Soil pH was obtained in a1:1soil to water slurry.

2.4.Statistical analyses

The effects of treatment and time on each variable were evalu-ated with analysis of variance(ANOVA)and analysis of covariance (ANCOVA)models.The overall effects were?rst evaluated with ANCOVA models with sampling date(Day2,6,12,27)as a contin-uous variable and C treatment as a categorical variable.With the exception of pH,neither treatment nor the treatment?date inter-action were signi?cant predictors in the overall model.Differences among dates(also including Day0)were then tested with one-way ANOVAs with date as a?xed factor.When appropriate,signi?cant differences among dates were determined with the Student-New-man-Keuls multiple comparisons test.The treatment effects at the end of the incubation(Day27),were evaluated with two separate tests:1)To test the effect of C quantity the variables were regressed on total C added excluding the acetate treatment,and2),the effects of C quality were evaluated with t-tests comparing the means of the acetate and leaf litter leachate treatment that both received a total of100ug g soilà1.The Pearson product–moment correlation coef?-cients were calculated to assess the linear associations among the soil variables.The Fe(II)production rate for each treatment was estimated by calculating the slope of the relationship between Fe(II) concentration and day of sampling.The amount of CO2evolved associated with the Fe reduction was calculated based on the stoichiometry of Fe(III)reductive dissolution:4moles Fe(III)reduced per mole CO2produced(Roden and Wetzel,1996).Data were log transformed as necessary to meet the assumptions of homogeneity of variance and normality.All statistical analyses were performed in SAS v9.1(SAS Institute,USA).3.Results

There were statistically signi?cant changes in Fe and P pools, and pH over time(Table1).The HCl extractable Fe(II)concentra-tions,averaged across treatments,increased from440m g Fe gà1soil to769m g Fe gà1soil over the27days of the incubation period (Fig.1a).The concentrations were signi?cantly higher on days12 and27than on the?rst three sampling dates.The HCl extractable Fe(III)also increased signi?cantly over time reaching a maximum on day27(Fig.1a).In contrast,the poorly crystalline(C/A extract-able)Fe was signi?cantly lower on day27(Fig.1a).

The microbial biomass P concentrations increased over time reaching100m g P gà1soil on day27,signi?cantly higher than the initial concentrations(Fig.1b).The microbial biomass P was signif-icantly positively correlated with Fe(II)concentrations(r?0.45, p<0.0001,Table2).The NaHCO3-P pool was relatively small,varying from8to18m g P gà1soil(Fig.1b).The NaOH-P concentrations were 10-fold higher than the NaHCO3-P and decreased signi?cantly after the initiation of the treatments(Fig.1b).The soil pH varied from5.3 to5.5with the maximum occurring on day6.Soil pH was positively correlated with both Fe(II)(r?0.41,p<0.001)and with microbial biomass P(r?0.27,p<0.05).

Carbon quantity had a signi?cant effect on Fe reduction.On day 27,soil pH,microbial P,and HCl extractable Fe(II)all signi?cantly increased with increasing C additions(Fig.2).Soil pH increased slightly but signi?cantly from 5.2in the control to 5.3in the 100m g C g dry soilà1litter leachate treatment.(R2?0.23,p<0.04). Microbial P increased from85m g P g dry soilà1in the control to 110m g P g dry soilà1in the highest litter leachate treatment (R2?0.36,p<0.009).The HCl extractable Fe(II)increased from less than600m g Fe(II)$g dry soilà1in the control,25and 50m g C g dry soilà1treatments to over800m g Fe(II)$g dry soilà1 (R2?0.38,p<0.006)in the two highest litter leachate treatments (t?7.3,p<0.002).Unlike C quantity,the only variable affected by the type of C added was soil pH,which was signi?cantly higher in the acetate treatment(Fig.2).

Net Fe reduction occurred in all treatments and at all time points (Fig.3).The maximum concentrations of Fe(II)for the control and two lower litter leachate treatments were observed on day12and for the two higher litter leachate and acetate treatments on day27. The C oxidation over the course of the incubation based on net Fe reduction ranged from0.1to1.4ug C g soilà1dayà1(Table3).The C oxidation associated with Fe reduction for the acetate treatment represented40%of the C added over the course of the incubation.

4.Discussion

4.1.Iron dynamics

Iron reduction has been well documented in anaerobic envi-ronments,but its importance in upland soils with heterogeneous redox environments has been largely ignored.This may be

related Table1

Log transformed data analyzed.

b Two outliers removed from analysis.

D.Liptzin,W.L.Silver/Soil Biology&Biochemistry41(2009)1696–1702 1698

to methodological dif?culties.While the signi?cance of other common terminal electron acceptors such as nitrate and sulfate can be studied with isotopic tracer experiments,this is not possible for Fe reduction due to complex isotopic exchange during Fe redox transformations (Roden and Lovley,1993).Nor is there a unique product of Fe reduction as occurs during methanogenesis.Typically,the net change in Fe(II)concentrations over time is used as the metric of Fe reduction.Similarly,C oxidation under anaerobic conditions associated with Fe reduction can be estimated by the net Fe(II)production,or by difference once the other terminal electron acceptors are accounted for (e.g.Roden and Wetzel,1996).

The net rates of Fe reduction during this experiment increased with the amount of C added to the soils.In the absence of O 2,Fe-reducing bacteria can outcompete sulfur reducing bacteria and methanogens for acetate (Teh et al.,2008;Lovley and Phillips,1987).While simple C compounds,like acetate,are well docu-mented to stimulate microbial Fe reduction (Teh et al.,2008;

Chaco

′n et al.,2006),it appears that litter leachate can support as much Fe reduction as pure acetate.In addition to the direct supply of C as an energy source for dissimilatory Fe-reducing bacteria,leaf

litter leachate may promote abiotic reductive dissolution of Fe(III)oxides because it contains organic reductants.It may also contain electron shuttling compounds,like humic acids with quinone moieties.The native reducing capacity of the leaf litter leachate was found to be 3.2mmol c mol à1C,similar to values measured for puri?ed humic acids suggesting the potential for electron shuttling (Peretyazhko and Sposito,2006).Few studies have examined how the addition natural humic substances affects Fe reduction in slurried soils and sediments (Rakshit et al.,2008;Nevin and Lovley,2000),but synthetic quinone-containing compounds (e.g.AQDS)

have been well documented to stimulate Fe reduction (Chaco

′n et al.,2006;Peretyazhko and Sposito,2005;Fredrickson et al.,1998;Lovley et al.,1998,1996).The leaching of surface leaf litter is likely to be an important source of electron shuttles,together with rhizodeposition which contributes a mixture of organic compounds to soils.The latter process has been hypothesized to be a major source of mineral dissolution and translocation resulting in soil weathering (Fimmen et al.,2008).

100

150

200

Microbial P

NaOH P

P h o s p h o r u s (μg P ·g s o i l -1)

Day 0Day 2Day 6Day 12

Day 27

Day 0

Day 2Day 6Day 12Day 27

c c

a bc

b ab ab a

a c c c

NaHCO 3 P

2000400060008000

10000

F e (μg F e ·g s o i l -1)

Fig.1.Extractable iron (a)and phosphorus (b)concentrations (Mean ?1SE)by date.Signi?cant letters indicate differences among

dates.

Table 2

Pearson product moment correlation coef?cients for the measured variables across all time points and treatments.Values in bold indicate signi?cant correlations 0

2505007501000125015000

100

H C l e x t r a c t a b l e F e (I I ) (μg F e ·g s o i l -1)

Total C added (μg C·g soil -1)

607080901001101200

100

M i c r o b i a l P (μg P ·g s o i l -1)

Total C added (μg C·g soil -1)

5.25

5.5

5.75

100

p H

Total C added (μg C·g soil -1)

Fig.2.Treatment effects on Day 27for a)HCl extractable Fe(II),b)Microbial biomass P,c)pH.Values are means ?1SE.Letters indicate signi?cant differences among treatments.The reported R 2values are for the regression of the variables on C quantity in the leachate treatments (i.e.excluding acetate).

D.Liptzin,W.L.Silver /Soil Biology &Biochemistry 41(2009)1696–17021699

In this study we documented net Fe(II)production even though the soils were incubated with an oxic headspace.This is due to the presence of anoxic microsites in these ?nely textured soils.Because portions of the soil are also aerobic,net Fe(II)production likely underestimates the true rate of Fe reduction as biotic or abiotic Fe oxidation can be occurring simultaneously within different portions of the same soil core.The difference in the time of maximum Fe(II)between the low C and high C addition treatments suggests that the balance between oxidation and reduction depended in part on the amount of C added.The observed decrease in C/A extractable Fe suggests that Fe cycling is much greater than the estimate of net Fe reduction.While the HCl extractable Fe(II)concentration had increased by 330ug Fe g soil à1averaged across the treatments on day 27,the HCl extractable Fe(III)concentration increased by 2100ug Fe g soil à1.In contrast,the C/A extractable Fe concentration had decreased by 3600ug Fe g soil à1by day 27.It appears that Fe is being transformed from a C/A extractable phase to an HCl extractable phase during Fe cycling related to the pulsed addition of the treatment solutions.These extractions differ in the mechanisms of Fe oxide dissolution:HCl promotes Fe oxide dissolution with protonation while the C/A extraction uses complexation and reduction (Schwertmann,1991).While the decrease in Fe extracted with C/A was not completely accounted for by the increase in HCl extractable Fe,the directional shifts in Fe in both extractions suggest a change in Fe mineralogy over time.This decrease in the C/A extractable Fe is consistent with the results of Thompson et al.(2006)suggesting that redox cycling increases the crystallinity of Fe minerals.4.2.P dynamics

We measured detectable changes in all P pools over the course of the incubation.The most consistent pattern over time was the signi?cant increase in microbial biomass P.By the last sampling

date there was also a trend,though not signi?cant,of increasing microbial biomass P with C addition.Microbial biomass P was signi?cantly correlated with Fe(II)concentrations,consistent with the theory that P bound to Fe(III)oxides solubilizes during Fe reduction and becomes available to organisms.While the NaOH extractable P is thought to include Fe-bound P,the size of this pool was not correlated to Fe(II)or microbial biomass P concentrations overall.However,the increase in microbial biomass P during the course of the incubation was approximately equal to the overall decrease in NaOH P,while the variability in the NaHCO 3-P pool was relatively small.We did not observe a relationship between NaHCO 3-P and Fe(II)suggesting that microbes rapidly immobilize P liberated during reductive dissolution of Fe minerals.Further,the NaHCO 3-P exhibited the smallest magnitude change in any of the P pools.Phosphorus release associated with Fe reduction appears to provide a transient supply of P to P-limited tropical ecosystems that would be missed with infrequent soil sampling.Presumably P liberated via Fe reduction would also be available to plant roots under ?eld conditions.

4.3.Ecosystem implications

Our results suggest that the Fe,P,and C cycles are tightly linked through oxidation and reduction dynamics in upland wet tropical ecosystems.Redox conditions ?uctuate over days to weeks in tropical forest soils (Silver et al.,1999).We propose a conceptual model (Fig.4)that links Fe,P,and C in these soils based on the links between redox,iron cycling,P availability and C oxidation described in other environments (Ponnamperuma,1972;Campbell and Torgersen,1980;Weber et al.,2006).Periods of low redox promote dissimilatory Fe reduction as a respiratory pathway and the concomitant release of Fe-bound P,followed by rapid P immobilization by the biota.Once the soils are suf?ciently re-oxygenated,days to weeks later,the reduced Fe is oxidized and can bind any P released by mineralization.The balance of P supply and P demand in tropical forests has been debated (Johnson et al.,2003),but in general,P is thought ecosystem processes in tropical rain forests (e.g.Herbert and Fownes,1995;Cleveland and Townsend,2006;Kaspari et al.,2008).The release of Fe-associated P during reduction and the rapid biotic immobilization may not be accounted for in periodic soil assays of P availability,particularly those that aerate the soil during laboratory extraction procedures.The periodic release and rapid immobilization of P released with Fe reduction may help maintain the high net primary productivity in these wet tropical forests.

Iron reduction can also be a signi?cant source of C oxidation.Roden and Wetzel (1996)reported that Fe reduction accounted for approximately 65%of the C oxidation in anaerobic wetland sediments.In an incubation with upland tropical forest soils,C oxidation was similar with an oxic and anoxic headspace suggest-ing that alternative electron acceptors can maintain high rates of respiration (K.DeAngelis,unpublished observations).In the current

20040060080010001200

Day 2Day 6Day 12

Day 27

e t F e r e d u c t i o n (μg F e ·g s o i l -1)

025*******Acetate

Fig. https://www.wendangku.net/doc/bb6246682.html, iron reduction over time

by treatment.The treatments are the total quantity of C added during the incubation in m g C g soil à1with the acetate treatment also receiving 100m g C g soil à1.Values are means ?1SE.

Table 3

Mean Fe reduction rates by treatment and the associated C oxidation based on the stoichiometric relationship –4moles of Fe reduced:1mole of organic C oxidized to CO 2D.Liptzin,W.L.Silver /Soil Biology &Biochemistry 41(2009)1696–1702

1700

study the overall net rate of Fe(II)production ranged from2.7to 26m g Fe(II)g soilà1dayà1in the control and acetate treatments https://www.wendangku.net/doc/bb6246682.html,ing a bulk density of0.65g cmà3for the top10cm (Silver et al.,1994),these rates of Fe reduction would result in an annual rate of3–34g C mà2respired.These?uxes are of similar magnitude of the C added to these soils in throughfall(13g C mà2, Heartsill-Scalley et al.,2007),and up to8%of annual litterfall C inputs(434g C mà2,Scatena et al.,1996).These values are conservatives estimates based on net Fe reduction under a well aerated headspace.Because Fe oxidation was likely occurring on the soil surfaces exposed to O2,the total amount of Fe reduction may have been signi?cantly higher.Based on the change in HCl and C/A Fe pools this rate may be as high as78or133m g Fe(II)g soilà1dayà1respectively or100–170g C mà2yrà1.These rates are of similar magnitude to Fe reduction measured under completely anaerobic conditions(Dubinsky,2008;Chaco′n et al., 2006;Thompson et al.,2006;Peretyazhko and Sposito,2005).

The importance of Fe reduction as a source of C oxidation in tropical soils likely depends on the frequency and duration of reducing conditions in the soil.In consistently dry or wet soils,Fe reduction may be less important because the redox conditions would not favor a high degree of Fe cycling.Humid tropical forest soils?uctuate between oxic and anoxic conditions;at the?eld site where the soils for this incubation were collected this?uctuation occurs with a periodicity of w2weeks(D.Liptzin,unpublished observations).Changes in the frequency and intensity of precipi-tation events due to climate change,even without changes in total precipitation,would alter the periodicity of redox?uctuations.This may lead in turn to changes in C cycling directly by altering the terminal electron acceptor of heterotrophic respiration.Less frequent reducing events could also lead to decreased P availability in these soils.

5.Conclusions

The importance of Fe redox cycling has been overlooked in unsaturated upland soils of humid environments.This study has demonstrated that Fe reduction occurs even without strict anaer-obic conditions and without slurrying soils.Further,this study is among the?rst to document that a more complex mixture of organic compounds,such as found in leaf litter leachate,stimulated Fe reduction as much as a simple C compounds.The increase in net Fe reduction during the incubation was associated with a parallel increase in microbial biomass P but not available(NaHCO3)P.Fe cycling appears to be an important source of P for microbes and results in signi?cant C oxidation in upland tropical forest soils. Acknowledgments

This research was supported by NSF grants DEB0543558to W.Silver and DEB0620910to the Institute for Tropical Ecosystem Studies,University of Puerto Rico,and to the International Institute of Tropical Forestry USDA Forest Service,as part of the Long-Term Ecological Research Program in the Luquillo Experimental Forest. The U.S.Forest Service(Dept.of Agriculture)and the University of Puerto Rico gave additional support.Additional support was provided by Agricultural Experiment Station funds to W.Silver and from the International Institute of Tropical Forestry,USDA Forest Service.We would like to thank B.Quintero,T.Wood,and C.Torrens with help in the?eld,and A Thompson,W.Yang,S.Rubol, E.Dubinsky,and S.Rakshit for help with the incubations and sample analysis.We thank two anonymous reviewers for helpful comments on the manuscript.

References

Baldwin,D.S.,Mitchell,A.M.,2000.The effects of drying and re-?ooding on the sediment and soil nutrient dynamics of lowland river–?oodplain systems:

a synthesis.Regulated Rivers Research&Management16,457–467.

Brown,S.,Lugo,A.E.,Silander,S.,Liegel,L.,1983.Research History and Opportunities in the Luquillo Experimental Forest.U.S.D.A.Forest Service Southern Forest Experimental Station,New Orleans,LA.General Technical Report SO-44. Campbell,P.,Torgersen,T.,1980.Maintenance of iron meromixis by iron redepo-sition in a rapidly?ushed monimolimnion.Canadian Journal of Fisheries and Aquatic Science37,1303–1313.

Can?eld, D.E.,Thamdrup, B.,Hansen,J.W.,1993.The anaerobic degradation of organic-matter in Danish coastal sediments–iron reduction,manganese reduction,and sulfate reduction.Geochimica Et Cosmochimica Acta57, 3867–3883.

Chaco′n,N.,Silver,W.L.,Dubinsky,E.A.,Cusack,D.F.,2006.Iron reduction and soil phosphorus solubilization in humid tropical forests soils:the roles of labile carbon pools and an electron shuttle compound.Biogeochemistry78,67–84. Cleveland, C.C.,Townsend, A.R.,Schmidt,S.K.,2002.Phosphorus limitation of microbial processes in moist tropical forests:evidence from short-term labo-ratory incubations and?eld studies.Ecosystems5,680–691.

Cleveland,C.C.,Townsend,A.R.,2006.Nitrogen and phosphorus additions cause substantial losses of soil carbon from a lowland tropical rain forest.Proceedings of the National Academy of Sciences of the United States of America103, 10316–10321.

Cross,A.F.,Schlesinger,W.H.,1995.A literature review and evaluation of the Hedley fractionation:applications to the biogeochemical cycle of soil phosphorus in natural ecosystems.Geoderma64,197–214.

Dettling,M.D.,Yavitt,J.B.,Zinder,S.H.,2006.Control of organic carbon minerali-zation by alternative electron acceptors in four peatlands,central New York state,USA.Wetlands26,917–927.

Dubinsky,E.A.,2008.Environmental Control on Microbial Community Structure and Iron Redox Dynamics in Upland Soils.Ph.D.Dissertation,University of California,Berkeley.

Fimmen,R.L.,Richter Jr., D.D.,Vasudevan, D.,Williams,M.A.,West,L.T.,2008.

Reductive dissolution and oxidative precipitation of iron in Ultisol B-horizons: evidence for iron cycling and the evolution of soil microsites in upland forest soils.Biogeochemistry87,127–141.

Fredrickson,J.K.,Zachara,J.M.,Kennedy,D.W.,Dong,H.,Onstott,T.C.,Hinman,N.W., Li,S.,1998.Biogenic iron mineralization accompanying the dissimilatory reduction of hydrous ferric oxide by a groundwater bacterium.Geochimica Et Cosmochimica Acta62,3239–3257.

Frizano,J.,Johnson,A.H.,Vann,D.R.,Scatena,F.N.,2002.Soil phosphorus fraction-ation during forest development on landslide scars in the Luquillo Mountains, Puerto Rico.Biotropica34,17–26.

Hedley,M.J.,Stewart,J.W.B.,1982.Method to measure microbial phosphate in soils.

Soil Biology and Biochemistry14,377–385.

Heartsill-Scalley,T.,Scatena,F.N.,Estrada,C.,McDowell,W.H.,Lugo,A.E.,2007.

Disturbance and long-term patterns of rainfall and throughfall nutrient?uxes in a subtropical wet forest in Puerto Rico.Journal of Hydrology333,472–485. Herbert,D.A.,Fownes,J.H.,1995.Phosphorus limitation of forest leaf-area and net primary production on a highly weathered soil.Biogeochemistry29,223–235. Hobbie,S.E.,Vitousek,P.M.,2000.Nutrient limitation of decomposition in Hawaiian forests.Ecology81,1867–1877.

Johnson,A.H.,Frizano,J.,Vann,D.R.,2003.Biogeochemical implications of labile phosphorus in forest soils determined by the Hedley fractionation procedure.

Oecologia135,487–499.

Anoxic

Oxic

Fig.4.Conceptual model of Fe reduction linked to P availability and C oxidation.

During more anoxic periods of time(e.g.following precipitation events)or in anoxic

microsites,dissimilatory iron reducing bacteria reduce solid Fe(III)oxides to soluble

Fe(II),oxidizing organic C compounds to CO2.Concomitantly,Fe-bound inorganic P(P i)

is released during the reductive dissolution of Fe and is quickly immobilized by roots or

microbes.During more oxic periods of time or in oxic microsites,the Fe(II)is oxidized

to Fe(III)either biotically or abiotically.The Fe(III)oxides are once again available to

bind any P i released by mineralization of organic phospohorus.

D.Liptzin,W.L.Silver/Soil Biology&Biochemistry41(2009)1696–17021701

Kaspari,M.,Wright,J.,Yavitt,J.,Harms,K.,Garcia,M.,Santana,M.,2008.Multiple nutrients limit litterfall and decomposition in a tropical forest.Ecology Letters 11,35–43.

Ku¨sel,K.C.,Wagner,T.,Trinkwalter,A.S.,Gobner,A.S.,Baumler,R.,Drake,H.L.,2002.

Microbial reduction of Fe(III)and turnover of acetate in Hawaiian soils.FEMS Microbiology Ecology40,73–81.

Lodge, D.J.,McDowell,W.H.,McSwiney, C.P.,1994.The importance of nutrient pulses in tropical forests.Trends in Ecology and Evolution9,384–387. Lovley, D.R.,1991.Dissimilatory Fe(III)and Mn(IV)reduction.Microbiological Reviews55,259–287.

Lovley,D.R.,Phillips,E.J.P.,https://www.wendangku.net/doc/bb6246682.html,petitive mechanisms for inhibition of sulfate reduction and methane production in the zone of ferric iron reduction in sediments.Applied and Environmental Microbiology53,2636–2641.

Lovley, D.R.,Coates,J.D.,Blunt-Harris, E.L.,Phillips, E.J.P.,Woodward,J.C.,1996.

Humic substances as electron acceptors for microbial respiration.Nature382, 445–448.

Lovley,D.R.,Fraga,J.L.,Blunt-Harris,E.L.,Hayes,L.A.,Phillips,E.J.P.,Coates,J.D.,1998.

Humic substances as a mediator for microbially catalyzed metal reduction.Acta Hydrochimica et Hydrobiologica26,152–157.

McGroddy,M.,Silver,W.L.,2000.Variations in belowground carbon storage and soils CO2?ux rates along a wet tropical climate gradient.Biotropica32, 614–624.

Murphy,J.,Riley,J.P.,1962.A modi?ed single solution method for determination of phosphate in natural waters.Analytica Chimica Acta26,31–36.

Nevin,K.P.,Lovley,D.R.,2000.Potential for nonenzymatic reduction of Fe(III)via electron shuttling in subsurface sediments.Environmental Science and Technology34,2472–2478.

Patrick,W.,Khalid,R.,1974.Phosphate release and sorption by soils and sediments: effect of aerobic and anaerobic conditions.Science186,53–55. Peretyazhko,T.,Sposito,G.,2005.Iron(III)reduction and phosphorous solubilization in humid tropical forest soils.Geochimica Et Cosmochimica Acta69, 3643–3652.

Peretyazhko,T.,Sposito,G.,2006.Reducing capacity of terrestrial humic acids.

Geoderma137,140–146.

Ponnamperuma, F.N.,1972.The chemistry of submerged soils.Advances in Agronomy26,29–96.

Pullin,M.J.,Cabaniss,S.E.,2003.The effects of pH,ionic strength and iron-fulvic acid interactions on the kinetics of nonphotochemical iron transformations II.The kinetics of thermal reduction.Geochimica Et Cosmochimica Acta67, 4079–4089.

Rakshit,S.,Uchimiya,M.,Sposito,G.,2008.Iron(III)bioreduction in soil in the presence of added humic substances.Soil Science Society of America Journal73, 65–71.

Reyes,I.,Torrent,J.,1997.Citrate-ascorbate as a highly selective extractant for poorly crystalline iron oxides.Soil Science Society of America Journal61, 1647–1654.

Roden,E.E.,Lovley,D.R.,1993.Evaluation of55Fe as a tracer of Fe(III)reduction in aquatic sediments.Geomicrobiology Journal11,49–56.

Roden,E.E.,Urrutia,M.M.,2002.In?uence of biogenic Fe(II)on bacterial crystalline Fe(III)oxide reduction.Geomicrobiology Journal19,209–251.

Roden, E.E.,Wetzel,R.G.,https://www.wendangku.net/doc/bb6246682.html,anic carbon oxidation and suppression of methane production by microbial Fe(III)oxide reduction in vegetated and unvegetated freshwater wetland sediments.Limnology and Oceanography41, 1733–1748.Roden, E.E.,Wetzel,R.G.,2002.Kinetics of microbial Fe(III)oxide reduction in freshwater wetland sediments.Limnology and Oceanography47,198–211. Sahrawat,K.L.,2004.Ammonium production in submerged soils and sediments: the role of reducible https://www.wendangku.net/doc/bb6246682.html,munications in Soil Science and Plant Analysis 35,399–411.

Sanchez,P.A.,1976.Properties and Management of Soils in the Tropics.Wiley,New York,NY,pp.259–260.

Scatena,F.N.,1989.An Introduction to the Physiography and History of the Bisley Experimental Watersheds in the Luquillo Mountains of Puerto Rico.U.S.D.A.

Forest Service Southern Forest Experimental Station,New Orleans,LA.General Technical Report SO-72.

Scatena,F.N.,Moya,S.,Estrada,C.,Chinea,J.D.,1996.The?rst?ve years in the reorganization of aboveground biomass and nutrient use following Hurricane Hugo in the Bisley experimental watersheds,Luquillo experimental forest, Puerto Rico.Biotropica28,424–440.

Schuur, E.A.G.,Matson,P.A.,2001.Aboveground net primary productivity and nutrient cycling across a mesic to wet precipitation gradient in Hawaiian montane forest.Oecologia128,431–442.

Schwertmann,U.,1991.Solubility and dissolution of iron oxides.Plant and Soil130, 1–25.

Silver,W.L.,Scatena,F.N.,Johnson,A.H.,Siccama,T.G.,Sanchez,M.J.,1994.Nutrient availability in a montane wet tropical forest-Spatial patterns and methodo-logical considerations.Plant and Soil164,129–145.

Silver,W.L.,Lugo,A.E.,Keller,M.,1999.Soil oxygen availability and biogeochemistry along rainfall and topographic gradients in upland wet tropical forest soils.

Biogeochemistry44,301–328.

Stookey,L.L.,1970.Ferrozine–a new spectrophotometric reagent for iron.

Analytical Chemistry42,779–781.

Szilas, C.P.,Borgaard,O.K.,Hansen,H.C.B.,1998.Potential iron and phosphate mobilization during?ooding of soil material.Water,Air,and Soil Pollution106, 97–109.

Teh,Y.A.,Dubinsky,E.A.,Silver,W.L.,Carlson,C.M.,2008.Suppression of meth-anogenesis by dissimilatory Fe(III)-reducing bacteria in tropical rain forest soils: implications for ecosystem methane?ux.Global Change Biology14,413–422.

doi:10.1111/j.1365-2486.2007.01487.x.

Thamdrup,B.,2000.Bacterial manganese and iron reduction in aquatic sediments.

Advances in Microbial Ecology16,41–84.

Thompson,A.,Chadwick,O.A.,Rancourt,D.G.,Chorover,J.,2006.Iron-oxide crys-tallinity increases during soil redox oscillations.Geochimica Et Cosmochimica Acta70,1710–1727.

Tiessen,H.,Moir,J.O.,1993.Characterization of available phosphorus by sequential extraction.In:Carter,M.R.(Ed.),Soil Sampling and Methods of Analysis.

Canadian Society of Soil Science.Lewis,Boca Raton,FL,pp.75–86.

Viollier,E.,Inglett,P.W.,Hunter,K.,Roychoudhury,A.N.,Van Cappellen,P.,2000.The Ferrozine method revisited:Fe(II)/Fe(III)determination in natural waters.

Applied Geochemistry15,785–790.

Vitousek,P.M.,Sanford,R.L.,1986.Nutrient cycling in moist tropical forest.Annual Review of Ecology and Systematics17,137–167.

Weber,K.A.,Achenbach,L.A.,Coates,J.D.,2006.Microorganisms pumping iron: anaerobic microbial iron oxidation and reduction.Nature Reviews Microbiology 4,752–764.

Weiss,J.V.,Emerson,D.,Megonigal,J.P.,2004.Geochemical control of microbial Fe(III)reduction potential in wetlands:comparison of the rhizosphere to non-rhizosphere soil.FEMS Microbiology Ecology48,89–100.

D.Liptzin,W.L.Silver/Soil Biology&Biochemistry41(2009)1696–1702 1702