lable at ScienceDirect

Soil Biology & Biochemistry 40 (2008) 2736–2742

Contents lists avai

Soil Biology & Biochemistry

journal homepage: www.elsevier .com/locate/soi lb io

Nitrogen conservation strategies of cranberry plants and ericoid mycorrhizalfungi in an agroecosystem

Sarah M. Stackpoole a,*, Beth Ann A. Workmaster a, Randall D. Jackson b, Kevin R. Kosola a

a University of Wisconsin, Madison Horticulture Department, 1575 Linden Drive, Madison, WI 53706, USAb University of Wisconsin, Madison Agronomy Department, 1575 Linden Drive, Madison, WI 53706, USA

a r t i c l e i n f o

Article history:Received 18 January 2008Received in revised form 11 July 2008Accepted 15 July 2008Available online 15 August 2008

Keywords:Ericoid mycorrhizal symbiosisRhizoscyphus (syn. Hymenoscyphus) ericaeVaccinium macrocarpon (cranberry)Litter decompositionResorption15N natural abundanceNitrogen cycling

* Corresponding author.E-mail address: smstackpoole@gmail.com (S.M. St

0038-0717/$ – see front matter � 2008 Elsevier Ltd.doi:10.1016/j.soilbio.2008.07.017

a b s t r a c t

Ericoid mycorrhizal fungi (ERM) are ubiquitous in cultivated cranberry beds, but no prior studies haveused 15N natural abundance measurements to track ERM influence on nitrogen (N) cycling in theseagroecosystems. Cranberries (Vaccinium macrocarpon) evolved in low-nutrient peat bogs, and ERMsymbioses often occur in nutrient-poor environments where N is limiting. We investigated ERMsymbiosis, resorption rates, as well as leaf litter quality and decomposition to infer N conservationstrategies in cultivated cranberry plants. We expected moderate to high resorption rates and low leaflitter decomposition rates. We also expected to find evidence of ERM fungi mediating N uptake.Resorption rates (35.7� 3.5% [s.e.]) were moderate, and the resulting cranberry leaf litter had high C:Nratios (56.9� 0.02). Decomposition was slow with 57� 2% of initial leaf litter mass remaining at the endof 2.5 years. The negative d15N values of cranberry leaves (�2.61� 0.27&) and positive d15N values ofcranberry roots (1.04� 0.35&) suggest that N uptake was mediated by ERM fungi as 14N is preferred over15N during transfer from mycorrhizal fungi to host plant. We demonstrated that even in this intensivelymanaged and fertilized agroecosystem, cranberry plants retain N conservation strategies observed inwildland cranberry ecosystems, namely high resorption efficiencies, low-quality leaf litter, and activeERM symbiosis. This study has management implications, since current nutrient plans mainly focus oninorganic N fertilizer regulation and do not recognize the role of ERM fungi in mediating N uptake.

� 2008 Elsevier Ltd. All rights reserved.

1. Introduction

Native cranberries evolved in low-nutrient temperate bogs (Eck,1990), yet cultivated cranberries (Vaccinium macrocarpon cv. Ste-vens) grow in a fertilized agroecosystem and receive considerableadditions of inorganic N fertilizer (w40 kg N ha�1 yr�1). Cultivatedcranberry may still exhibit some nitrogen (N) conservation strate-gies that may reduce the need for expensive N fertilizer that isprone to aqueous runoff or atmospheric loss. Plant ecology theoryposits that plant species in nutrient-poor environments havehigher nutrient resorption efficiencies than species from nutrient-rich sites (Aerts, 1996). However, evergreen species, such as cran-berry can reduce N losses by producing leaves with a long life spanand low nutrient concentrations, thus reducing the importance ofhigh resorption efficiencies (Aerts, 1996; Eckstein et al., 1999).Evergreen plants also produce low-quality leaf litter with high C:Nratios, which decomposes slowly (Berendse, 1994; Chapman et al.,2006). This conserves nutrients proximal to the plant root, making

ackpoole).

All rights reserved.

them easily accessible for plant uptake. Finally, symbiotic associa-tion with mycorrhizal fungi is another N conservation strategyfound in plants growing in nutrient-poor systems (Aerts et al.,2000; Chapman et al., 2006). The symbiosis enables plants to accessmore nutrients, including complex organic N forms from decom-posing organic material (Bajwa and Read, 1985; Read and Perez-Moreno, 2003).

Cranberry, a low-growing evergreen vine, grows in soils that aredifferent from most agronomic and horticultural crops. The top 10–15 cm of cranberry soils at our research sites are characterized bydistinct alternating layers of organic matter (OM) and sand becauseWisconsin growers apply 1–2.5 cm of sand to the beds every 3–5years in the winter. Organic matter layers are made of mostly leaflitter, with some stem and root material. These layers are usuallyaerobic during the growing season, since it is common for growersin Wisconsin to flood only for fall harvest and to form ice foroverwinter frost protection.

We established a study to investigate the role of three Nconservation strategies, resorption efficiency, leaf litter decompo-sition, and mycorrhizal symbiosis in the fertilized cranberry agro-ecosystem. Aerts (1996) found in a survey of evergreen species fromEurope and the USA that resorption efficiencies were less than 50%

NitrificationLeaching

Atmospheric loss

a

LEAF LITTERMINERAL N

PLANT

EricoidMycorrhizal

Fungib

c

SOIL ORGANIC MATTER

Fig. 1. Important N pathways in the cranberry agroecosystem. Solid arrows indicatethe standard pathway of N cycling, which is prone to N loss from the system. Nitrogenis resorbed from leaves prior to senescence and leaf fall (a). Fallen leaf litter decom-poses (b), becomes part of the soil OM, and then it is mineralized to ammonium, whichcan be taken up by the plants. The dotted arrows indicate the pathway where N fromfertilizer, mineralization, or even organic N from leaf litter may be taken up directly byplants via their ERM symbiosis (c).

S.M. Stackpoole et al. / Soil Biology & Biochemistry 40 (2008) 2736–2742 2737

of leaf N. Fertilization appears to influence resorption, as theresorption efficiency of the evergreen ericaceous species, Empetrumhermaphroditum, fell from about 55% in natural bog conditions toabout 40% when 100 kg N ha�1 of fertilizer was added annually forfour years (van Heerwaarden et al., 2003). Because cranberry plantsare also evergreen and grow in fertilized conditions, we expectedresorption efficiencies to be below 50% of leaf N (Fig. 1, pathway a).

We expected that low leaf nutrient concentrations accompaniedby moderate to high resorption efficiencies would contribute tohigh C:N ratios in leaf litter, which would slow decomposition andmineralization in the cranberry beds (Fig. 1, pathway b). Therefore,we expected that through their symbiosis with ericoid mycorrhizalfungi (ERM) cranberry plants would mobilize N from fertilizer orpossibly organic N directly from decomposing leaf litter (Fig. 1,pathway c). We considered the leaf litter a possible N source since ithas been demonstrated in lab studies that ERM fungi can accesscomplex forms of N (Kerley and Read, 1995). Ericaceous plantscommonly form associations with ERM fungi such as Rhizoscyphus(syn. Hymenoscyphus) ericae (D.J. Read, W.Y. Zhuang and Korf)(Read and Perez-Moreno, 2003; Read et al., 2004). Ericoid mycor-rhizal fungi have been found throughout cultivated cranberry bedsin Oregon and Wisconsin (Scagel, 2005; Kosola and Workmaster,2007). Evidence of both endo- and ectomycorrhizal fungi mobi-lizing N for plant uptake can be determined by comparing the d15Nvalues of plant leaves and mycorrhizal roots (Hogberg et al., 1996;Emmerton et al., 2001; Hobbie and Hobbie, 2006). These studiesfound 15N enrichment of fungal tissue relative to plant tissue,especially leaves. During the transfer of N from fungal hyphae to theplant, the mycorrhizae became enriched with the heavier isotope.Since cranberry roots contained fungal hyphae of ERM, we expec-ted cranberry root tissue to be enriched with 15N relative to cran-berry leaves.

The importance of ERM fungi mediating N uptake has beenrelatively well documented in unmanaged systems such as heath-lands (Read and Perez-Moreno, 2003), boreal forests (Lindahl et al.,2007), and the arctic tundra (Michelsen et al., 1998; Hobbie andHobbie, 2006). Here we described a field study designed to test thehypothesis that in a fertilized agroecosystem, cultivated cranberryplants display N conservation strategies, including ERM symbiosis.This may be important for plant management and production, sinceN fertilizer is expensive and is susceptible to runoff or atmosphericloss (Drinkwater, 2004). To test our hypothesis, we collected liveleaf material and compared it to recently senesced leaf litter toestimate resorption. We used in situ leaf litter bags to measure litterdecomposition. We measured d15N values from live cranberryleaves and roots to infer ERM mediation of N uptake. Since the

cranberry roots and ERM fungi colonize leaf litter, we alsomeasured d15N values from OM layers to infer whether ERMmobilized N throughout the rooting zone.

2. Materials and methods

2.1. Cranberry bed characteristics

Data were collected from five conventionally cultivated cran-berry beds on four different farms in Wood (44�300 N/90�00 W) andJuneau County (43�470 N/90�040 W) Wisconsin, USA. All bedscontained the variety ‘Stevens’ and were relatively young plantings(9–12 years old). Underneath the top 15 cm of Wisconsin cranberrysoil, which has alternating layers of OM and sand, is at least 30 cmof pure sand. Underneath the pure sand are substrata of either sandor peat. We selected two beds with peat substrata and three bedswith sand substrata to represent both types. A grid of 35 points, 5columns by 7 rows, was laid out in each bed, which were rectan-gular and between 1 and 2 ha. Columns were typically 61 m apart,and rows were 6 m apart. All samples were collected fromrandomly selected points on this grid (soil cores, live plant biomass,trap litter, and leaf litter bags, and ericoid mycorrhizal fungi).

2.2. Soil N pools

Soil N pools (ammonium, nitrate, and total dissolved nitrogen(TDN)) were measured from two (2.5 cm� 15 cm) soil cores takenfrom six randomly chosen points in the grid at three timesthroughout growing season, 6 May, 24 June, and 24 August 2006.The cores were transported on ice and stored at 4 �C until analyzed(within 48 h of collection). Net N mineralization in the field wasassayed with the buried bag technique (Eno, 1960). At eachsampling location, a soil core was placed in a sealed polyethylenebag, re-inserted back into the hole at the same time and the samelocation as the soil cores. Three different sets of buried bags wereincubated between five and seven weeks throughout the growingseason. The first set of bags was removed on 24 June 2006 (49 daysincubation). The second set of bags was removed on 3 August 2006(38 days incubation). The third set of bags was removed on 2October 2006 (62 days incubation). A 1:10 w/v soil to 1 M KClsolution was used to extract N from soil cores and buried bagsamples. Samples were shaken for 1 min, and extracted for 24 h atroom temperature. Samples were then filtered with Whatman A/E0.2 mm glass fiber filters. The extracts were frozen (�20 �C) untilanalysis (Robertson et al., 1999). Results of extractable N from theoriginal soil cores were subtracted from the results of the buriedbags to obtain a net N mineralization rate.

Ammonium measurements were determined colorimetricallyon a Labsystems Multiskan EX microplate reader (Thermo FischerScientific, Walthman, MA) at 650 nm absorbance using the Ber-thelot assay (Rhine et al., 1998). TDN measurements were madeusing this assay, following a persulfate digestion and a reductionwith DeVarda’s alloy (Sims et al., 1995). We substituted phenol for2-phenylphenol sodium salt tetrahydrate (PPS) which is a compo-nent of reagent #2 in the Berthelot assay. A precipitate consistentlyformed in the digested samples when PPS was used. The sampleswere clear when reagent #2 was made with fresh phenol, pHadjusted to 11. The absorbance values of undigested ammoniumstandards using phenol were identical to those using PPS (S.Stackpoole, unpublished data). Nitrate samples were reduced toammonium with DeVarda’s alloy and then analyzed. Dissolvedorganic nitrogen (DON) was determined by subtracting ammoniumand nitrate from the TDN values.

S.M. Stackpoole et al. / Soil Biology & Biochemistry 40 (2008) 2736–27422738

2.3. Live plant biomass and trap litter collections

To sample live plant material and leaf litter from OM layers, onebiomass core (15 cm deep� 8 cm diameter) was taken through theplant canopy from 5 to 6 randomly chosen points on the grid inAugust and November 2005. Cores were stored at 4 �C until pro-cessed. A sub-sample of August 2005 leaves was used in thecomparison of live fresh leaves to recently senesced trap litter (seeSection 2.4). Another sub-sample of August 2005 leaves was used tocompare the 15N natural abundance of cranberry leaves to cran-berry roots. The belowground portion of the core was carefullywashed to remove most of the sand. Leaves and roots were sortedfrom the rest of the plant material. Cranberry roots and ERM wereanalyzed together because it was impossible to separate themycorrhizal hyphae from the root tissue. The November 2005 soilcores were used to compare leaf litter sub-samples from the OMlayers of the soil profile. The core was split in half, and leaf litter wasextracted from each of the layers with tweezers. Age of the leaflitter was roughly determined from sanding reports from thegrowers. It was also impossible to separate the individual compo-nents of the leaf litter, which were cranberry roots, cranberry leaflitter, and fungal hyphae. Therefore, from this point on, it will bereferred to as the ‘leaf litter complex’, to represent all three of thecomponents. All of the samples were dried at 65 �C for 48 h,weighed, and ground in 2 ml centrifuge tubes with two stainlesssteel balls (0.3175 cm, McMaster Carr, Type 302, Grade 100).

Samples were analyzed for d15N and percent nitrogen (%N) ona Europa 20/20 continuous flow mass spectrometer in series witha Carlo-Erba CHN analyzer (University of Wisconsin–MadisonHorticulture Department). Stable isotope abundances are reportedas d15N in parts per mille (&) expressed as (Rsam/Rstd� 1)� 1000,where R is the 15N/14N ratio of either the sample (sam) or thereference standard (std), which was an in-house flour standard.Precision for replicated stable isotope samples was �0.3&.

To sample detached, senesced leaves, 12 litter traps were placedon the bed surface in May 2005 using the 35-point grid. The traps(6 cm outer diameter, 5.1 cm inner diameter) were held in placewith sod stakes. Leaf litter was collected from the traps inSeptember 2005. The majority of leaf litter fell in August andSeptember (S. Stackpoole, personal observation). All trap littersamples were also ground in 2 ml centrifuge tubes with stainlesssteel ball bearings. They were analyzed for %N and %C using drycombustion on a Carlo-Erba elemental CHN analyzer (Universityof Wisconsin–Madison Agronomy Department). Resorption effi-ciency (RE) was calculated as: RE ¼ 100� 1� fN content intrapleaf litter=N content live leavesg. The trap leaf litter was comparedto live leaf material collected in August 2005. Nitrogen concentra-tion of the trap leaf litter and lives leaves was expressed as the Nmass per leaf dry mass or milligrams N per gram of leaf litter (vanHeerwaarden et al., 2003).

2.4. Surface leaf litter decomposition

One-year-old leaf litter was sampled from the soil surface of fourcranberry beds in April 2004. The litter was returned to the lab,sorted and cleaned, and dried at 65 �C. Leaf litter bags (10 cm2)were made of noseeum fabric (0.3 mm mesh size); roots andhyphae were not excluded. Bags were filled with approximately 1 gdry weight of leaf litter. Thirty-five leaf litter bags, containing litteroriginally from that bed, were placed in the grid pattern asdescribed above. Leaf litter bags were sampled from four randomlocations in the grid at seven different times: November 2004; May,June, August, and November 2005; and May and November 2006.Roots that had invaded the bags were separated from leaf litter; leaflitter and roots were dried and weighed separately. Sand was alsocarefully removed from litter. All leaf litter samples were ground in

2 ml centrifuge tubes as described above. They were analyzed for%N and %C using dry combustion on a Carlo-Erba elemental CHNanalyzer (University of Wisconsin–Madison Agronomy Depart-ment). Selected samples (November 2004, November 2005, andNovember 2006) were analyzed for d15N values as described above.

2.5. Ericoid mycorrhizal colonization

To measure ericoid mycorrhizal fungi colonization, sub-samplesof roots were taken from the top layer (1–2 cm) of each soil coreand stored in 50% ethanol. Samples were taken in 2004 and 2005 at4–5 times through the growing season. Samples were cleared andstained using the methods described in Ref. (Kosola and Work-master, 2007). Root length colonization (percent root length con-taining ERM structures) was determined with a dissectingmicroscope (40–90� magnification), with a line intercept method(Giovannetti and Mosse, 1980).

2.6. 15N natural abundance of soil nitrogen

To measure 15N natural abundance of N in the soil, resin bagswere deployed for six weeks (26 June 2006–3 August 2006). Bags,made of nylon stocking material, were filled with 2.5 ml (w½ tsp)of anion (Dowex 1, 8–50 mesh size, Cl� form) and 2.5 ml of cationresin (Dowex C-211, 16–50 mesh size, Hþ form). The bags wereinstalled 15 cm below the soil surface. Three bags were installedrandomly on the grid within three beds. After the bags wereremoved from the field, the samples from each bed were pooled toget a composite sample. They were extracted two times for 1 h eachwith 2 M KCl (1:10 v/v). To get a total d15N signature we used anacid trapping procedure described in Ref. (Khan et al., 1998).Ammonium and nitrate were analyzed together because ammo-nium values were so low in some samples as to be undetectable. Weused DeVarda’s alloy to reduce NO3

� to NH4þ, and MgO was used to

free NH4þ as gaseous NH3. The N in the vapor was captured on two

acidified disks (Whatman 41, 6.4 mm), following a five to eight hourincubation period. Samples of ammonium based fertilizer (5–10–30) were also collected from growers. Samples were analyzed ford15N using an elemental analyzer coupled with a mass spectrom-eter as described above.

2.7. Statistical analyses

All means and standard errors were calculated with SAS PROCMEAN (SAS 9.1 Statistical Software, Cary, NC). Data were tested fornormality and homogeneity. PROC ANOVA was used to comparemeans of surface leaf litter d15N values from 2004, 2005 and 2006and to compare mean C:N ratios of leaf litter in OM layers.Regression analyses were carried out in CoStat (CoHort Software,Monterey, CA).

3. Results

3.1. Soil and plant N pools in cultivated cranberry bogs

Extractable soil ammonium (0.97� 0.11 kg NH4þ ha�1) and

nitrate (0.19� 0.05 kg NO3� ha�1) levels averaged over three sample

periods in the 2006 growing season were low compared to DONpools (13.6�1.2 kg DON ha�1). The average pH was acidic,5.3� 0.12. Although net N mineralization throughout the growingseason was low (4�10�2� 4�10�3 kg ha�1 day�1), there was noindication of net N immobilization in these soils. On average16.5�1.8 kg N ha�1 was removed with berry harvest, which isnearly 40% of total N added in ammonium based fertilizer annually.

S.M. Stackpoole et al. / Soil Biology & Biochemistry 40 (2008) 2736–2742 2739

3.2. Resorption and surface leaf litter decomposition

Live cranberry leaf N concentration was 0.95� 0.02% N. Recentlysenesced cranberry leaf concentration was 0.60� 0.03%. Resorptionefficiency of leaves was 35.7� 3.49%. The mass remaining (% ofinitial) of surface leaf litter decreased through time (P¼ 0.02)(Fig. 2a). There was also a decrease in the C:N ratio (P¼ 0.05)(Fig. 2b). For these two parameters, most of the changes occurredwithin the first year of the leaf litter bags being in the field. Therewas no significant difference in absolute N through time (P¼ 0.56).Colonization of leaf litter by cranberry roots and ERM fungi wasdetected in May 2006, which was approximately two years afterthe leaf litter bags were placed on the soil surface. There was nosignificant difference between the d15N values of the surface leaflitter from November 2004, November 2005, and November 2006(P¼ 0.75) (Fig. 2d).

3.3. Ericoid mycorrhizal colonization and d15N values

Ericoid mycorrhizal colonization, measured as percent total rootlength, was present in all beds with an average of 42�13% of the

Fig. 2. Decomposition of leaf litter bags placed on the soil surface, means� 1 S.E. St(y¼ 0.72� 0.11� ln(x); r2¼ 0.7; n¼ 25) and (b) C:N ratio (y¼ 37.54� 5.39� ln(x); r2¼ 0.58;The d15N values represent three collection times (November 2004, 2005, and 2006), and th

total root length colonized by ERM fungi. Live cranberry leaves hadnegative d15N values (�2.61�0.27). Cranberry roots, which wecould not physically separate from ERM fungi, had positive d15Nvalues (1.04� 0.35). The d15N values of inorganic N extracted fromsoil resin were negative (�11.36�1.26) reflecting the naturalabundance level of both ammonium and nitrate. Results from oneammonium based fertilizer sample (5–10–30) confirmed that itsd15N value was close to atmospheric levels (0.10). The relationshipbetween the vertical position of leaf litter layers and the d15Nnatural abundance was significant and positive (P¼ 0.004) (Fig. 3a).We found no significant difference between the C:N ratios of theleaf litter layers (P¼ 0.25) (Fig. 3b).

4. Discussion

4.1. Comparison of cultivated cranberry farms to wildland bogs

Despite being a fertilized and managed system, soil N pools, pH,and net N mineralization rates demonstrated that cultivated cran-berry farms were similar to low nutrient, acidic wildland bogs.Therefore, these data supported our hypothesis that cultivated

andard errors are shown where larger than size of symbol. (a) Mass remainingn¼ 25). (c) There was no significant change in absolute amount of N through time. (d)ere was no significant difference among the means.

Fig. 3. Changes in d15N values and leaf litter chemistry with vertical position of the leaflitter layers. Means� SE are plotted for three beds. (a) d15N of leaf litter and ERM(y¼� 3.60þ 1.72� x, r2¼ 0.99, n¼ 9) and (b) C:N ratio of leaf litter and ERM(y¼ 36.35þ 0.469� x, r2¼ 0.03, n¼ 9).

S.M. Stackpoole et al. / Soil Biology & Biochemistry 40 (2008) 2736–27422740

cranberries grow in an environment where nutrient conservationstrategies should be beneficial. Dissolved organic nitrogen domi-nated extractable soil N pools. The pH was low, and the rate of Nmineralization was five to ten times slower in cranberry beds thanpeat bogs in North America and Europe (Verhoeven et al., 1990;Bridgham et al., 1998). The low net N mineralization rate in cran-berry beds may be due to the recalcitrant leaf litter, which had a C:Nratio of around 80. The C:N ratio for peat in the bog studies variedbetween 26 and 50 (Verhoeven et al., 1990; Bridgham et al., 1998).Fertilizer inputs (w40 kg N ha�1) delivered between June and Julyannually, provided a flush of inorganic nutrients to the cranberryagroecosystem. Approximately 40% of the N fertilizer appliedannually was removed with harvested berries. The rest was storedin plant and soil pools, or lost to the environment.

4.2. Moderate to high resorption rates and slow litterdecomposition

The N concentration of senesced leaves (0.60%) indicated thatcranberry plants went through complete resorption (Killingbeck,1996). The resorption values (between 40% and 50%) determinedfrom van Heerwaarden’s (van Heerwaarden et al., 2003) fertilizednative bog study using the evergreen ericaceous species E.

hermaphroditum, were comparable to our cranberry resorptionestimate (36.5%), although ours was lower. The slight decrease inresorption efficiency, despite higher fertilization rates, indicatesthat nutrient status is not the only factor that can influenceresorption efficiencies. Low water availability (del Arco et al., 1991)and shade (Chapin and Moilanen, 1991) can also reduce resorptionefficiency below potential rates. To our knowledge, the influence ofmycorrhizal status of plants on resorption rates has not yet beeninvestigated (Aerts, 1996; Killingbeck, 1996). Since fungi are able toaccess nutrients that are sometimes unavailable to the plant, whenconsidering the influence of soil fertility on resorption efficiencies,it may be useful to also consider DON soil pools, OM build-up, andthe mycorrhizal status of plants.

The decomposition rates of cranberry more closely resembledericaceous tundra plants grown in unfertilized conditions, ratherthan an agronomic crop such as corn, where sizeable additions of Nfrom plant residue (40–80 kg N ha�1) are added to the agro-ecosystem annually. For example, decomposition rates for cran-berry leaf litter were higher (57% initial mass remained) than theevergreen ericaceous species Vaccinium vitis-idaea (roughly 80%initial mass remained), in an unfertilized incubation experiment at4 �C and 10 �C (Hobbie, 1996). This was presumably because therewere higher initial N values in cranberry leaves (0.95% vs. 0.70%)and higher surface temperatures in Wisconsin cranberry bogscompared to the temperatures of the incubation experiment.However, compared to a highly managed agronomic crop such ascorn, cranberry decomposition was much slower. After two years inthe field, only 20% of corn leaves in surface leaf litter bags remained(Burgess et al., 2002).

There was a slight increase in absolute N values during the firsthalf of the study, followed by a slight decrease. However, there wasno statistically significant net import or export of N from thesurface leaf litter. This suggests that the decrease in the C:N ratio ofsurface leaf litter through time is due changes in carbon, most likelycarbon use by microbial communities (Lindahl et al., 2007). C:Nratios of less than 25 (dry weight plant material) have been set asthresholds for net N mineralization from decomposing litter (Lutzand Chandler, 1947; Berg and Mc Claugherty, 2003). Using thisvalue as an indicator, after 2.5 years in the field, it was likely that Nwas still being immobilized by microbes decomposing the surfaceleaf litter.

4.3. d15N values indicate ERM fungi access N from rooting zone

Our data supported the hypothesis that ERM fungi mediated Nuptake in cultivated cranberry beds, since cranberry leaves weredepleted in 15N and the cranberry roots and ERM were enriched.We were also able to link the plant isotope data with the leaf littercomplex isotope data to trace the d15N values through variousstages of cranberry leaves’ life cycle (Fig. 4). We began by focusingon the 15N depleted live cranberry leaves (Fig. 4a). During theprocess of resorption, there was no discrimination against 15N (Kolband Evans, 2002), so when the 15N depleted cranberry leaves sen-esced and fell (Fig. 4b), a layer of depleted leaf litter formed on thesurface of the cranberry bed. Root colonization of the leaf litter wasdetected at the beginning of the second year of the study, but therewere no changes in the surface leaf litter d15N values within the first2.5 years. However, we found higher d15N values in the layerdirectly beneath the soil surface, and we used sanding dates fromthe growers to estimate that ERM fungi began accessing N 2.5–6years after the leaf litter initially fell. We were not able to determinefrom the leaf litter isotope values the source of N that was mobi-lized by ERM. Mostly likely it was fertilizer N, but there is alsoa possibility that organic N was accessed directly from the leaf litter.A tracer study, which includes 15N enriched leaf litter, would need

a) Cranberry leaves senesce and fall tobed surface.

b) Roots and ERM fungi colonizeleaf litter.

(-) 15N

(-)

OM Layer 1

OM Layer 2

OM Layer 3

Sand

Sand

c) ERM fungi mobilize N. Leaf littercomplex becomes enriched with 15N.

d) Layers get buried becauseof sanding.

*OM = organic matter

15N

(-) 15N

(+) 15N

(+) 15N

Fig. 4. Tracking the changes in d15N values in cranberry leaves’ life cycle. The 15N depleted cranberry leaves senesced and fell (a) which produced a layer of depleted leaf litter on thesurface of the cranberry bed. Eventually cranberry roots and ERM began to colonize the leaf litter (b). The ERM fungi mobilized and transferred N to the plant, and the heavierisotope remained in the fungal hyphae. Since the ERM was so closely associated with the leaf litter tissue, the 15N level of this material was also positive (c). With time, these leaflitter layers got buried because cranberry growers applied a layer of sand to the bed surface every 3–5 years (d).

S.M. Stackpoole et al. / Soil Biology & Biochemistry 40 (2008) 2736–2742 2741

to be implemented to identify the contribution of N from these twosources.

Our mycorrhizal colonization rates were similar to those foundby Kosola and Workmaster (2007), and they also measured highermycorrhizal colonization rates in the shallowest OM layers. Webelieve that there was more fungal activity near the surface, andwith time the leaf litter complex was buried by additional appli-cations of sand every 3–5 years (Fig. 4d), and that is where wefound the highest d15N values. The deepest soil layers we examinedwere 9–12 years old. This isotope pattern could also be explained byplant uptake of N that is depleted in 15N relative to soil total N,followed by deposition of isotopically light N on soil surface(Nadelhoffer and Fry, 1994). These factors may also contribute tothe 15N natural abundance patterns in the cranberry agro-ecosystem, but our data also strongly suggest that ERM fungiinfluenced the distinct d15N values of cranberry roots and leaves, aswell as the d15N values with increasing depth in the OM layers.

The C:N ratios of the leaf litter complex were not what weexpected and did not contribute much to our overall understandingof decomposition processes in the OM layers. Other field-basedstudies found an increase in C:N ratios in leaf litter that was locatedin deeper layers of the soil profile (Kosola and Workmaster, 2007;Lindahl et al., 2007). There was a slight decrease in C:N ratiosbetween Layer 1 and Layer 2, followed by a slight increase betweenLayer 2 and Layer 3. The trend was not statistically significant, andtherefore we were unable to infer net immobilization or netmineralization of N from these older leaf litter layers. This could bedue to the shorter time frame of our cores, which were 11 years old.In Lindahl et al. (2007) study the cores were 45 years old, and thecores were 15 years old in Kosola and Workmaster (2007) study.

4.4. d15N of inputs and microbial processing of OM

We also measured the 15N natural abundance of the main inputsto the cranberry agroecosystem, since it has been documented thatd15N values of N sources can also influence d15N values of plantleaves (Schulze et al., 1994; Michelsen et al., 1998). Ammoniumbased fertilizer, applied in June and July accounted for 60–70% of N

inputs to the system (S. Stackpoole, unpublished data), and the d15Nvalue for that N source was close to zero. The d15N value of theinorganic N (both ammonium and nitrate) from soil resin extractswas negative (�11). Precipitation carries negative d15N value(Russell et al., 1998), and this value probably reflects that the resinbags captured ammonium and nitrate from rainfall that percolatedthrough sand and OM layers in the bed. This negative value mayalso be an integration of multiple N transformation processes,which may include nitrification and possibly dissimilatory nitratereduction to ammonium (DNRA) (Ostrom et al., 2002). The low 15Nsignature of the inorganic soil solution relative to the plant indi-cates that N mineralization from litter contributes small amounts ofN to the cycle. Therefore, the negative cranberry leaf d15N wasa result of total N inputs including fertilizer N (zero d15N), soil Npools (strong negative d15N), and possibly leaf litter N (negatived15N). Cranberry plants may draw upon the latter two nutrientsources later in the growing season when fertilizer applicationshave ceased at the end of July and fruit growth is at a maximumin August (Hart et al., 2000). The resin bags had a similar meanvalue for ammonium (1.87 mg l�1�1.46) as the soil extracts(1.08 mg l�1�0.04), but the resin bags had a higher mean value fornitrate (2.17 mg l�1�1.09) than the soil extracts (0.16 mg l�1�0.04). We assumed this was because the resin tightly binds thenitrate, while nitrate in the sandy soil was more prone to leaching.

5. Conclusions

We found that even with N additions through fertilizer, cran-berries exhibited traits of plants growing in nutrient limitedconditions, including litter with high C:N ratios, and slow decom-position. Additionally, the distinct soil OM layers in cultivatedcranberry beds provided a unique system to study ERM effects on Ncycling. Using 15N natural abundance measurements, we were ableto provide evidence that ERM fungi mobilize N from the rootingzone. Until now, the role of ERM fungi in tightly regulating N cyclingwas not recognized, and most cranberry growers did not accountfor ERM influences on N uptake. This may have importantenvironmental and management implications, since inorganic

S.M. Stackpoole et al. / Soil Biology & Biochemistry 40 (2008) 2736–27422742

N fertilization is susceptible to runoff and atmospheric loss. Futurework should quantify the N flux through the ERM pathway todetermine if and how much organic N is mobilized by ERM fungifrom organic matter relative to fertilizer inputs.

Acknowledgments

We thank Amanda Zimmerman for careful preparation of tissuesamples, and Armand Krueger for running samples on the massspectrometer. We thank grower cooperators in Wisconsin forpermitting us to sample in their beds. This study was partiallysupported by funding from the Robert and Helen Gottschalkendowment, USDA–CREES project, and SARE grants.

References

Aerts, R., 1996. Nutrient resorption from senescing leaves of perennials: are theregeneral trends? Journal of Ecology 84, 597–608.

Aerts, R., Chapin, F.S., Fitter, A.H., Faffaelli, D.G., 2000. Mineral Nutrition of WildPlants Revisited: A Re-evaluation of Processes and Patterns, pp. 2–55. In:Advances in Ecological Research, vol. 30. Academic Press, San Diego.

Bajwa, R., Read, D.J., 1985. The biology of mycorrhiza in the Ericaceae. IX. Peptides asnitrogen sources for the ericoid endophyte and for mycorrhizal and non-mycorrhizal plants. New Phytologist 101, 459–467.

Berendse, F., 1994. Litter decomposability – a neglected component of plant fitness.Journal of Ecology 82, 187–190.

Berg, B., Mc Claugherty, C., 2003. Plant Litter Decomposition, Humus Formation,Carbon Sequestration. Springer-Verlag, Berlin, Heidelberg.

Bridgham, S.D., Undergraff, K., Pastor, J., 1998. Carbon, nitrogen and phosphorousmineralization in northern wetlands. Ecology 79, 1545–1561.

Burgess, M.S., Mehuys, G.R., Madramootoo, C.A., 2002. Nitrogen dynamics ofdecomposing corn residue components under three tillage systems. SoilScience Society of America 66, 1350–1358.

Chapin, F.S., Moilanen, L., 1991. Nutritional controls over nitrogen and phosphorousresorption from Alaskan birch leaves. Ecology 72, 709–715.

Chapman, S.K., Langley, J.A., Hart, S.C., Koch, G.W., 2006. Plants actively control nitrogencycling: uncorking the microbial bottleneck. New Phytologist 169, 27–34.

del Arco, J.M., Escudero, A., Garrida, M.V., 1991. Effects of site characteristics onnitrogen retranslocation from senescing leaves. Ecology 72, 701–708.

Drinkwater, L.E., 2004. Improving Fertilizer Nitrogen use Efficiency Through anEcosystem-based Approach, pp. 93–102. In: Mosier, A.R. (Ed.), Agriculture andthe Nitrogen Cycle, vol. 65. Island Press, Washington.

Eck, P., 1990. American Cranberry Rutgers University Press, New Brunswick.Eckstein, R.L., Karlsson, P.S., Weih, M., 1999. Leaf life span and nutrient resorption as

determinants of plant nutrient conservation in temperate-arctic regions. NewPhytologist 143, 177–189.

Emmerton, K.S., Callaghan, T.V., Jones, H.E., Leake, J.R., Michelson, A., Read, D.J.,2001. Assimilation and isotopic fractionation of nitrogen by mycorrhizal andnonmycorrhizal subarctic plants. New Phytologist 151, 513–524.

Eno, C.F., 1960. Nitrate production in the field by incubating the soil in polyethylenebags. Soil Science Society of America 24, 277–279.

Giovannetti, M., Mosse, B., 1980. An evaluation of techniques for measuring vesic-ular arbuscular mycorrhizal infection in roots. New Phytologist 84, 489–500.

Hart, J., Davenport, J., DeMoranville, C., Roper, T., 2000. Nitrogen for BearingCranberries in North America. Oregon State University Extension Service.

Hobbie, J.E., Hobbie, E.A., 2006. 15N in symbiotic fungi and plants estimatesnitrogen and carbon flux rates in arctic tundra. Ecology 87, 816–822.

Hobbie, S., 1996. Temperature and plant species control over litter decomposition inAlaskan tundra. Ecological Monographs 66, 503–522.

Hogberg, P., Hogbom, L., Schinkel, H., Hogberg, M., Johannisson, C., Wallmark, H.,1996. 15N abundance of surface soils, roots, and mycorrhizas in profiles ofEuropean forest soils. Oecologia 108, 207–214.

Kerley, S.J., Read, D.J., 1995. The biology of mycorrhiza in the Ericaceae. XVIII. Chitindegradation by Hymenoscyphus ericae and transfer of chitin–nitrogen to thehost plant. New Phytologist 131, 369–375.

Khan, S.A., Mulvaney, R.L., Brooks, P.D., 1998. Diffusion methods for automatednitrogen-15 analysis using acidified disks. Soil Science Society of America 62,406–412.

Killingbeck, K.T., 1996. Nutrients in senesced leaves: keys to the search for potentialresorption and resorption proficiency. Ecology 77, 1716–1727.

Kolb, K.J., Evans, R.D., 2002. Implications of leaf nitrogen recycling on the nitrogenisotope composition of deciduous plant tissues. New Phytologist 156, 57–64.

Kosola, K.R., Workmaster, B.A.A., 2007. Mycorrhizal colonization of cranberry:effects of cultivar, soil type, and leaf litter composition. Journal of the AmericanSociety of Horticultural Science 132, 134–141.

Lindahl, B.D., Ihrmark, K., Boberg, J., Trumbore, S.E., Hogberg, P., Stenlid, J.,Finlay, R.D., 2007. Spatial separation of litter decomposition and mycorrhizalnitrogen uptake in a boreal forest. New Phytologist 173, 611–620.

Lutz, H.J., Chandler, R.F., 1947. Forest Soils. John Wiley and Sons, New York.Michelsen, A., Quarmby, C., Sleep, D., Johnson, L., 1998. Vascular plant 15N natural

abundance in heath and forest tundra ecosystems is closely correlated withpresence and type of mycorrhizal fungi in roots. Oecologia 115, 406–418.

Nadelhoffer, K., Fry, B., 1994. Nitrogen Isotope Studies in Forest Ecosystems.pp. 22–44. In: Lajtha, K., Michener, R.H. (Eds.), Stable Isotopes in Ecology andEnvironmental Science. Blackwell Scientific Publishers, Oxford.

Ostrom, N.E., Hedin, L.O., Fischer, J.C., Robertson, G.P., 2002. Nitrogen trans-formations and NO3

� removal at a soil–stream interface: a stable isotopeapproach. Ecological Applications 12, 1027–1043.

Read, D.J., Perez-Moreno, J., 2003. Mycorrhizas and nutrient cycling in ecosystems –a journey toward relevance? New Phytologist 157, 475–492.

Read, D.J., Leake, J.R., Perez-Moreno, J., 2004. Mycorrhizal fungi as drivers ofecosystem processes in heathland and boreal forest biomes. Canadian Journal ofBotany 82, 1243–1263.

Rhine, E.D., Sims, G.K., Mulvaney, R.L., Pratt, E.J., 1998. Improving the berthelotreaction for determining ammonium in soil extracts and water. Soil ScienceSociety of America Journal 62, 473–480.

Robertson, G.P., Sollins, P., Ellis, B.G., Lajtha, K., 1999. Exchangeable Ions, pH, andCation Exchange Capacity, pp. 462. In: Robertson, G.P., Coleman, D.C.,Bledsoe, C.S., Sollins, P. (Eds.), Standard Soil Methods for Long-Term EcologicalResearch, vol. 2. Oxford University Press, Oxford.

Russell, K.M., Galloway, J.N., Macko, S.A., Moody, J.L., Scudlark, J.R., 1998. Sourcesof nitrogen in wet deposition to the Chesapeake Bay region. AtmosphericEnvironment 14–15, 2453–2465.

Scagel, C., 2005. Mycorrhizal status of sand-based cranberry (Vaccinium macro-carpon) bogs in Southern Oregon. Small Fruits Review 2, 31–41.

Schulze, E.D., Chapin, F.S., Gebauer, G., 1994. Nitrogen nutrition and isotopedifferences among life forms at the northern treeline of Alaska. Oecologia 100,406–412.

Sims, G.K., Ellsworth, T.R., Mulvaney, R.L., 1995. Microscale determination of inor-ganic nitrogen in water and soil extracts. Communications in Soil Science andPlant Analysis 26, 303–316.

van Heerwaarden, L.M., Toet, S., Aerts, R., 2003. Nitrogen and phosphorousresorption efficiency and proficiency in six sub-arctic bog species after 4 year ofnitrogen fertilization. Journal of Ecology 91, 1060–1070.

Verhoeven, J.T.A., Maltby, E., Schmitz, M.B., 1990. Nitrogen and phosphorousmineralization in fens and bogs. Journal of Ecology 78, 713–726.