Temperate eucalypt forest decline is linked to altered ectomycorrhizal communities mediated by soil chemistry

Temperate eucalypt forest decline is linked to altered ectomycorrhizal communities mediated by soil chemistry

Forest Ecology and Management 302 (2013) 329–337 Contents lists available at SciVerse ScienceDirect Forest Ecology and Management journal homepage: ...

737KB Sizes 0 Downloads 53 Views

Forest Ecology and Management 302 (2013) 329–337

Contents lists available at SciVerse ScienceDirect

Forest Ecology and Management journal homepage: www.elsevier.com/locate/foreco

Temperate eucalypt forest decline is linked to altered ectomycorrhizal communities mediated by soil chemistry Bryony M. Horton a,b,c,d,⇑, Morag Glen b, Neil J. Davidson c,d,f, David Ratkowsky b, Dugald C. Close b,c, Tim J. Wardlaw d,e, Caroline Mohammed b a

School of Plant Science, University of Tasmania, Private Bag 55, Hobart, Tasmania 7001, Australia Tasmanian Institute of Agriculture, University of Tasmania, Private Bag 98, Hobart, Tasmania 7001, Australia Bushfire Cooperative Research Centre, Level 5, 340 Albert Street, East Melbourne, Victoria 3002, Australia d Cooperative Research Centre for Forestry, Private Bag 12, Hobart, Tasmania 7001, Australia e Forestry Tasmania, 79 Melville Street, Hobart, Tasmania 7000, Australia f Greening Australia, 30 Burnett St., North Hobart, Tasmania 7000, Australia b c

a r t i c l e

i n f o

Article history: Received 21 January 2013 Received in revised form 3 April 2013 Accepted 5 April 2013 Available online 30 April 2013 Keywords: Australia Cortinariaceae Fungi Nitrogen Phosphorus Russulaceae

a b s t r a c t Eucalypt forest decline has a complex aetiology often linked to altered soil chemistry caused by environmental disturbances. Forest decline has also been linked to alterations in ectomycorrhizal (ECM) fungal communities, which are imperative for nutrient transfer and affect ecosystem productivity and health. Our aim was to determine the influence of soil chemistry on ECM fungal communities and tree health in declining temperate eucalypt forests. We hypothesise that forests with changed soil chemistry, in particular altered nitrogen cycling associated with forest decline, supports unique ECM fungal communities. ECM communities from twelve Eucalyptus delegatensis forest plots were characterised by DNA sequencing of root tip and sporocarp samples. Tree health and nutrient concentrations from soil and foliage samples were quantified for each plot. Multivariate and regression analyses and t-tests were used to determine ECM fungal community differences between forest health classes, and identify which soil variables were important for defining these communities. Elevated available soil nitrogen and soil acidity were associated with severely declining forest. Soil pH, nitrate and organic carbon significantly explained the majority of variation in ECM fungal community composition and structure, which differed between moderately and severely declining forest. Russulaceae species richness was greatest in acidic soils (severely declining forest) while Cortinariaceae species richness was greatest in soils with lower concentrations of soil nitrate (moderately declining forest). Total ECM fungal richness was inversely related to available soil phosphorus and soil nitrate. Thus, altered soil chemistry associated with eucalypt forest decline mediates changes in the ECM fungal community. Forest management must consider the role of disturbance in maintaining suitable soil conditions for symbiotic fungi which are important for maintaining healthy eucalypt forest and restoring declining forest ecosystems. Ó 2013 Elsevier B.V. All rights reserved.

1. Introduction Tree decline has been observed in many different forests across the globe (Mueller-Dombois, 1988; Ciesla and Donaubauer, 1994; Wardle et al., 2004). These include European and North American coniferous and broadleaf forests (Chevone and Linzon, 1988; Innes, 1992). In Australia, Eucalyptus decline is widespread, affecting many different species across a large geographical range in rural landscapes and in native forest (Old et al., 1980; Heatwole and

⇑ Corresponding author. Address: 68a Mildura Street, Coffs Harbour, NSW 2450, Australia. Tel.: +61 403891397. E-mail address: [email protected] (B.M. Horton). 0378-1127/$ - see front matter Ó 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.foreco.2013.04.006

Lowman, 1986). Factors causing eucalypt declines in native forest are poorly understood. Altered fire regimes since European settlement have been proposed as one of the possible causes of eucalypt declines in temperate Australian forests (Jurskis and Turner, 2002; Jurskis, 2005; Close et al., 2009). In temperate eucalypt ecosystems that have adapted to a particular fire regime, the absence of fire is proposed to cause a cascade of changes with numerous feedbacks (Close et al., 2009). These feedbacks result in changes in vegetation dynamics, soil chemistry, and soil microbial communities, including mycorrhizal fungal communities. Altered soil chemistry is often proposed as causal factor of eucalypt decline in the absence of fire, due to increases in nitrogen (N) in the form of NO 3 and decreases in phosphorus (P) (Turner and Lambert, 2005; Close et al., 2009). These changes are consistent with those associated with

330

B.M. Horton et al. / Forest Ecology and Management 302 (2013) 329–337

declining forests throughout the world in the absence of disturbance, where there is often a shift from N limitation to P limitation (Wardle et al., 2004). Many temperate tree species including Eucalyptus spp. form ECM associations, which facilitate the uptake of P, N and immobile nutrients (Smith and Read, 1997). Ectomycorrhizal fungal communities have the potential to affect forest composition and nutrient cycling and therefore in turn impact plant productivity and health (Nantel and Neumann, 1992; Aerts, 2002; Debellis et al., 2006; Twieg et al., 2007; Tedersoo et al., 2008). Conversely, the diversity of ECM fungal communities can be shaped by the presence of different plant species (Twieg et al., 2007; Gates et al., 2011; Horton et al., unpublished) and the availability of various nutrients (Lilleskov et al., 2001; Peter et al., 2001a; Lilleskov et al., 2002; Avis et al., 2003; Ishida et al., 2007; Tedersoo et al., 2008; Twieg et al., 2009; Lilleskov et al., 2011). In many declining forests ECM fungal communities are altered in comparison to healthy forest. Reduced sporocarp diversity and production in European forests has been attributed to such causes as altered forest management, increased N deposition through air pollution, and reduced tree vitality (Arnolds, 1991). Fellner and Pešková (1995) reported the decline of both ECM sporocarp production and active mycorrhizal roots following visible damage to forests caused by air pollution, acidification, or fertilisation. Declining trees growing in Norway Spruce (Picea abies) forest, were characterised by fewer mycorrhizal roots than healthy trees (Estivalet et al., 1990; Perrin and Estivalet, 1990) and more generally, declining forests had a higher abundance of mycorrhizal morphotypes but with less complex ramification and a marked decrease in diversity with soil depth compared to healthy forest (Vinceti et al., 1998). Furthermore, ECM fungal species richness as sampled in root tips and as sporocarps, was reduced in declining forest compared to less severely affected forests (Peter et al., 2008). Similarly, reduced mycorrhizal fungal richness was observed in Douglas Fir (Pseudotsuga menziesii) forest in the Netherlands (Jansen, 1991) where tree vitality was inversely related to ECM fungal richness and mycorrhizal colonisation potentially as a result of changes soil conditions. In Australia, eucalypt crown health has been shown to be significantly negatively correlated to ECM density indicating that decline symptoms are associated with absence of ECM (Scott et al., 2012). The percentage of mycorrhizal roots on E. delegatensis seedlings grown in soils collected from declining and healthy forests differed in abundance and form (Ellis and Pennington, 1992). Seedlings grown in a mix of soil collected from healthy and declining forest formed prolific mycorrhizas with fungi of the Basidiomycota, whereas seedlings grown purely in soil from declining forests formed less abundant mycorrhizas with the Ascomycota (Ellis and Pennington, 1992). Considering the importance of ECM for plant productivity and the differences seen between ECM fungal communities of healthy and declining forests, the causal factors of eucalypt decline likely include feedback loops involving ECM associations (Close et al., 2009). This study explores the relationship between ECM fungal communities and eucalypt forest decline, mediated by soil and foliage nutrients that are typically associated with the decline process. The aims of this study were to (a) determine if soil and foliage chemistry, especially nitrogen and phosphorus, influence ECM fungal community composition, structure and richness; and (b) investigate the relationship between ECM fungal communities and eucalypt forest health, within Tasmanian E. delegatensis forest. Changes in the ECM fungal communities of these forests are hypothesised to be tightly linked to altered soil conditions, particularly altered nitrogen cycling, that occur as a part of the eucalypt decline process, leading to unique ECM fungal communities within forests of differing health status.

2. Materials and methods 2.1. Study sites A total of 12 plots were set up as six pairs across two sites in E. delegatensis dominated forest in northern Tasmania. Paired plots were randomly located within 40–100 m of one another within the same vegetation type, at the same altitude (800–900 m above sea level) and had been exposed to the same known management history. Moderately and severely declining forest plots were interspersed throughout the study area. All plots were on flat or gently undulating terrain and had similar soil characteristics typical of the soil type of Tasmanian montane forests (Grant et al., 1994). Eight of the 12 plots (plots 1–8) were established in the highlands of north-east Tasmania along Ben Ridge Road, 60 km east northeast of Launceston (41.35°S, 147.67°E and 41.37°S, 147.61°E) and had been included in an earlier study of eucalypt dieback (Ellis et al., 1980). Soils were brown dermasols derived from Devonian granodiorite (Grant et al., 1994). All eight plots had a mean annual rainfall of 1320 mm and mean annual temperature of 9.1 °C (ESOCLIM module of ANUCLIM 5.2 Houlder and Hutchinson, 2000). Plots 1–4 (41.35°S, 147.67°E) supported an E. delegatensis forest with a rainforest mid-storey dominated by Nothofagus cunninghamii and fern ground cover. In contrast, plots 5–8 (41.37°S, 147.61°E) supported a patchy mid-storey of Acacia melanoxylon and dry sclerophyll shrubby to grassy understorey and were located 6 km east of plots 1–4. Plots 9–12 were located 81 km west southwest of Launceston, Tasmania (41.58°S, 146.17°E) along Gads Hill Road. These plots have a mean annual rainfall of 1564 mm and mean annual temperature of 9.3 °C (ESOCLIM module of ANUCLIM 5.2 Houlder and Hutchinson, 2000). Soils were dark brown ferrosols derived from Tertiary Basalt (Grant et al., 1994). Plots 9 and 10 had a rainforest mid-storey dominated by N. cunninghamii and sparse understorey of mixed wet sclerophyll and rainforest shrubs and ferns. Plots 11 and 12 were co-dominated by E. delegatensis and E. dalrympleana and had a thick dry sclerophyll understorey with a ground layer of ferns.

2.2. Nutrient sampling and analysis At each plot, four surface soil samples (approximately 5  5  5 cm) were taken in May 2007 from five randomly chosen individual trees, at approximately 2 m from the trunk and at equal spacing (90° apart) on a circle around each tree. Soil samples were pooled for the study plot. Pre-washed resin bags made of stocking nylon containing either 5 g of a cation- or anion-exchange resin (Amberlite, Sigma–Aldrich) were used to sample available nitrogen and phosphorus within the soil solution (Giblin et al., 1994). Four sets of resin bags were buried in pairs at 5 cm depth approximately 1 m apart under the drip-line of three randomly selected trees per site, and were collected approximately 6 months later. Eucalypt foliage was collected by Forestry Tasmania during August 2007. For each individual tree, whole branches were shot down from the outer and lower part of the crown on the sunlit side of the tree. These trees were the same trees sampled for soil nutrients (five per plot). Foliage samples consisted of 30–40 young fully expanded undamaged leaves, from each whole branch. Foliage samples were dried at 40 °C before sub-sampling for nitrogen (foliar N) and phosphorus (foliar P) analysis. Soil samples were analysed for total soil nitrogen (N), soil  ammonium (NH 4 ), soil nitrate (NO3 ), soil organic carbon (SOC), to tal soil phosphorus (P), soil pH, available soil NO 3 and NH4 , available soil P. All nutrient analyses were conducted by a private sector

B.M. Horton et al. / Forest Ecology and Management 302 (2013) 329–337

laboratory services provider, CSBP Limited (Perth, Australia) during 2008–2009. Standardisation was performed by evenly distributing quality control standards throughout the experimental samples, as per policy of the commercial laboratory that undertook soil analyses. In addition a sub-set of samples were analysed in duplicate to ensure precision and blanks (with no soil) were analysed to check for contamination. All results of the quality control standards must fall within 10% of their expected result. 2.3. Crown health assessments Primary crown dieback (PCD), defined as the proportion of primary branches that have died back from the terminal shoot (Wardlaw, 1989), was used as an indicator of overall tree health. PCD was assessed on a scale of 0–1, with 1 representing full health. This method has proven to be an effective measure of eucalypt crown dieback (Horton et al., 2011). The total number of trees assessed per plot is shown in Fig. 1. Crown health scores were averaged per study plot to provide an overall eucalypt health score for each plot. Plots were allocated to one of three health categories based on Podger et al. (1980); a score >0.8 represented healthy plots while a health score of 60–80% (0.6–0.8) indicated moderate decline (a level of decline from which trees can recover). Finally severe decline was defined by a health score of <0.6. Eucalypt trees with dieback affecting >0.4 of their crown (health score <0.6) exhibited reduced growth in girth, from which they were unable to recover (Wardlaw 1989). 2.4. Vegetation survey Vegetation surveys were conducted in November 2007 by scoring vascular plant species cover within four randomly located 5 m  5 m quadrats within each of the study plots. Plant species cover was assessed using the Domin scale for all species present in the quadrat. Domin categories were converted to means for analysis. Domin cover categories (means in brackets) were: <1 (0.5), 1–4 (2.5), 5–9 (7), 10–24 (17), 25–32 (28.5), 33–49 (41), 50–74 (62), 75–94 (84.5), 95–100 (97.5). Plant species richness

331

for a plot was defined as the total number of plant species from all four quadrats surveyed within the plot. 2.5. Ectomycorrizal fungi sampling and identification Ten soil cores (5 cm  5 cm  10 cm deep) were collected from each of the 12 plots, in March, May, June, November and December of 2007; and February, March, May and June of 2008. Each soil core was collected from within the canopy drip zone (generally <3 m) of the E. delegatensis tree (DBH > 90 cm) nearest to a randomly generated co-ordinate within each plot. Soil cores were stored at 4 °C until processed within 4 weeks of collection (Glen et al., 2008). The soil from a single core was shaken through a series of graded sieves (500, 1000, and 2360 lm). Fine roots were extracted, cleaned and examined under a Zeiss semi 2000-C dissecting microscope with Zeiss KL500 electronic light. ECM root tips found within each soil core were grouped into morphotypes based on appearance such as colour, size, branching, mantle surface, emanating hyphae and rhizomorphs (Agerer, 1991). Roots that were positively identified by morphology as non-eucalypt (i.e. Acacia, Nothofagus, grasses) were not sampled. Dark tips lacking turgidity were assumed dead. At least one sample of each morphotype within each soil core (1–5 single root tips or a pyramidal, monopodially branching cluster of 3–5 root tips) was frozen at 80 °C for subsequent DNA analyses. Sporocarp surveys were conducted on ten occasions during autumn and early winter of 2007 and 2008, totalling 15.5 personhours/plot, split evenly between surveying for epigeous and hypogeous sporocarps. With the exception of known saprotrophic or parasitic fungi, epigeous surveys were conducted by searching 5 m wide strips for sporocarps during the allocated time. Surveys of hypogeous fungi were concentrated within the drip zone of E. delegatensis trees. The upper soil layer (5 cm in depth) surrounding randomly selected trees were searched for 5–10 min using hand cultivators to remove the top litter layer averaging 5–10 trees per survey. All observed epigeous and hypogeous sporocarps were collected and stored in labelled paper bags until processed. In the laboratory (within 3 days of collection), macroscopic (cap, stipe, gills) and other distinctive characteristics were noted for each collection, and where possible, were used to identify the specimen to the level of genus. Small sections of the sporocarp cap and gills (or gleba) were excised for DNA extraction and stored in 1.5 mL tubes at 80 °C until processed. Collections were dried and frozen as reference collections, and were deposited in the National Herbarium of Victoria (collector numbers E702-E703 and T451-T1302). 2.6. Molecular analysis

Fig. 1. Mean crown health (PCD) of eucalypts for each plot. Bars indicate standard error. The black line indicates the boundary between moderate decline (above 0.6) and severe decline (below 0.6). Unless otherwise indicated, n = 10 and represents the number of eucalypt trees assessed in each plot. Plots with a rainforest understorey are filled and plots with a dry sclerophyll understorey are open. Plots 1–8 are located along Ben Ridge Road in north-east Tasmania. Plots 9–12 are located along Gads Hills Road, north west-Tasmania.

DNA was extracted from all root tips and sporocarps, and purified by a silica-binding method (Glen et al., 2002). DNA was eluted in 25 lL TE buffer and aliquots diluted to 1:40, 1:20 or 1:10 in TE buffer (Sambrook et al., 1989) then stored at 80 °C. The ITS region was amplified by PCR in 50 lL reactions containing 1  NH4 reaction buffer (Bioline, London, UK), 2 mM MgCl2 (Fisher Biotec, Wembley, Australia), 0.2 mg/mL BSA (Fisher Biotec, Wembley, Australia), 200 lM dNTP (Bioline, London, UK), 0.25 lM of primers ITS1-F (Gardes et al., 1991) and ITS4 (White et al., 1990) (Geneworks, Adelaide, Australia), 0.04 U/lL of MangoTaq DNA Polymerase (Bioline, London, UK) and 5 lL of DNA template. The thermocycler program was; 95 °C for 2 min, then 35 cycles of 95 °C for 30 s, 55 °C for 30 s and 72 °C for 30 s, with a final extension at 72 °C for 7 min then cooled to 14 °C for 1 min. When DNA amplification failed, alternative DNA dilutions were used as the PCR template until DNA amplification was optimised. Sequencing was performed in a single direction, unless quality was poor, in which case the reverse direction was also sequenced. The majority

332

B.M. Horton et al. / Forest Ecology and Management 302 (2013) 329–337

of sequencing was performed by the University of California Berkeley DNA sequencing facility (USA) (http://mcb.berkeley.edu/barker/dnaseq/index.html) or Macrogen Pty., Ltd., (Korea) (www.macrogen.com). The remaining sequences were obtained according to the following procedure. PCR products were cleaned using 0.5 lL ExoSAP IT (USB Corp, Cleveland, OH, USA) in 3.5 lL of PCR product, and incubated at 37 °C for 45 min, followed by 80 °C for 15 min in a 2720 Thermal Cycler (Applied Biosystems, Foster City, CA, USA). Sequence reactions were carried out using an Applied Biosystems BigDye Terminator v3.1 Cycle Sequencing Kit according to the manufactures’ instructions. Sequencing was performed using an ABI 3100 Genetic Analyzer (Applied Biosystems, Foster City, CA, USA). 2.7. Sequence analysis and identification Chromatograms were edited using DNAStar SeqMan II v 5.05. Searches of public DNA databases (GenBank, EMBL, DDBJ, UNITE), were carried out with the nucleotide–nucleotide (blastn) BLAST searches (Altschul et al., 1997), through the Australian National Genome Information Service BioManager online database (www.angis.org.au). Sequences were grouped according to BLAST search results and aligned using Clustal W (Thompson et al., 1994). Species were initially defined as having P98% sequence similarity over the whole of the ITS region, then refined by phylogenetic analysis of related sequences. Phylogenetic trees were constructed using DNAML of the Phylip package (Felsenstein, 1989; Abarenkov et al., 2010) with default settings and rooted by outgroup. In some cases where PCR or DNA sequencing failed, some sporocarp samples were allocated to an species based solely on their morphological similarity. Taxonomic nomenclature followed Crous et al. (2004) (www.mycobank.org) apart from the genus Thaxterogaster and Zygomycota, which followed Peintner et al. (2002) and Hibbett et al. (2007), respectively. DNA sequences were submitted to GenBank (accession numbers for speciess recorded in this study are JF960600-JF960854 and JQ513390). 2.8. Statistical analysis Differences in ECM species richness (total, and for the most species rich families), community and structure (relative species richness of each ECM family within each plot) were tested between forests of different health category, using the combined presence-absence data from the root tip and sporocarp surveys. Structure data were converted to proportional data and angular (arcsine) transformed prior to all statistical analyses. All independent soil and foliage nutrient concentrations were log (x + 1) transformed and vegetation cover was arcsine transformed prior to inclusion in statistical analyses. Least squares multiple linear regression was used to model ECM species richness on the basis of soil and foliage nutrients (soil pH, total soil N, total soil P, SOC,    soil NO 3 , soil NH4 , available NO3 , available NH4 , available P, foliage P and foliage N) and was performed using PROC REG of the statistical program SAS (SAS Institute Inc.). Regression analysis was also used to test for correlations between ECM families and particular soil variables. t-tests were used to determine differences in soil and foliage N:P and the relative richness of the Cortinariaceae, Russulaceae and Thelephoraceae between forest health levels, as well as differences in crown health between understorey type. Two-tailed t-tests assuming unequal variance and were performed in GenStat v7.1 (VSN International Ltd., 2003). All other analyses were performed using the program PRIMER v6 with PERMANOVA + (Anderson et al., 2008). Bray–Curtis similarity measure was used to construct resemblance matrices using the presence/absence data set obtained from sporocarp and root tip sampling and only included species that were recorded in more

than one plot. The resemblance matrix for ECM fungal community structure was based on the transformed proportions of each ECM family within each plot. Canonical analyses of principal co-ordinates (CAP) (Anderson and Willis, 2003) were used to assess differences in fungal community composition between health categories (moderately or severely declining) and were tested using 9999 permutations. Spearman’s rank correlations were used for the vector overlays. Canonical correlations analysis (CCorA) was used to determine correlations between ECM fungal community similarities (composition or structure) and eucalypt crown health, soil ammonium or soil pH. Distance-based multivariate multiple linear regressions (DISTLM) were used to determine which predictor variables (soil pH, total soil N, total soil P, SOC, soil NO 3 , soil   NH , available NO , available NH , available P, foliage P and foliage 4 4 3 N, crown health, % cover of A. dealbata, A. melanoxylon, E. delegatensis, E. dalrympleana, L. lanigerum, M. squarrosa, N. cunninghamii and P. juniperina) were important for explaining variation in ECM fungal community composition and structure. Variables were added step-wise and all models were run using 9999 permutations, using Akaike’s Information Criterion (AIC) as the goodness-of-fit measure. The models reported here include only those variables that were significant in predicting ECM fungal communities. Distancebased redundancy analysis (dbRDA) was used for ordination of the fitted values from the models. For all analyses, significance was indicated by p < 0.05.

3. Results 3.1. Eucalypt crown health Eucalypt crown health varied greatly among plots (Fig. 1). Five plots had an average tree crown health score within the range of moderate decline (0.6–0.8 crown health score), and the remaining seven plots fell within the range of severe decline (<0.6 crown score) (Fig. 1). The mean crown health score across plots with moderately declining eucalypts was 0.71 ± 0.04 SE (n = 5), for severely declining eucalypts was 0.36 ± 0.06 SE (n = 7), and the median crown health score across all plots was 0.52 (n = 12). Crown health was not significantly lower in plots with rainforest understorey as compared with dry sclerophyll understorey (0.38 v. 0.63, t(7) = 2.89, p = 0.06).

3.2. Soil and foliage nutrients Mean concentrations of nutrients for moderately and severely declining forest were compared against the overall median, which allowed a comparison of nutrient values from severely and moderately declining forest plots within the skewed data set (Table 1). All but one of the plots with moderately declining trees had available  mineral N contents (combined levels of NO 3 and NH4 ) < 500 mg/L, whereas plots with severe decline had much higher levels of mineral N (Table 1). Plots with severe decline had higher soil N:P than plots with moderate decline (mean 2.57 v. 1.61). Foliar N:P ratios were higher for severely declining forest than moderately declining forest (mean 13.12 v. 8.91). Four of the seven severely declining plots had foliar N:P > 14, while four of the five moderately declining plots had foliar N:P < 14 (Table 1). However, t-tests indicated that none of the soil and foliar nutrients significantly differed between moderately and severely declining forest. There were statistically significant differences in soil N:P and foliar N:P between rainforest and dry sclerophyll forest (soil N:P 0.13 v. 0.07, t(8) = 4.76, p < 0.01, and foliar N:P 9.70 v. 5.37, t(10) = 3.82, p < 0.01).

333

B.M. Horton et al. / Forest Ecology and Management 302 (2013) 329–337

Table 1 Summary of plot attributes. Data shown is raw untransformed data. Italicized columns are plots with severe decline and clear columns are plots with moderate decline. Rf refers to rainforest and Sc to dry sclerophyll. Means show ± standard error. Attribute

Plot

Plot

1

2

3

4

5

6

7

8

9

10

11

12

Understorey Crown health ECM richness Soil pH Soil NO 3 (mg/kg) Soil NH 4 (mg/kg) Total soil N (%) Total soil P (mg/kg) Soil N:P SOC (%) Soil C:N Foliage N (%) Foliage P (%) Foliage N:P Available NH 4 (mg/L) Available NO 3 (mg/L) Available N (mg/L) Available P (ppm)

Rf 0.36 32 4.5 4 111 0.62 202 3.07 10 16.13 1.47 0.08 17.92 1085 769.1 1853.60 1.3

Rf 0.52 32 4.2 19 143 1.34 499 2.69 10 7.46 1.51 0.1 15.4 3315 1020 4335.47 2.9

Rf 0.08 35 4.3 1 12 1.06 254 4.17 10 9.43 1.52 0.09 17.74 590.2 124.4 714.53 1.8

Rf 0.28 53 4.4 1 141 1.04 349 2.98 10.00 9.62 1.4 0.09 16.03 1444 253.6 1697.55 0.75

Sc 0.48 38 5.5 1 108 0.58 712 0.82 6.92 11.93 1.54 0.26 6.05 505.1 21.15 526.21 11.3

Sc 0.52 33 5.4 1 89 0.46 540 0.85 9.96 21.65 1.35 0.16 8.38 298.5 230.2 528.61 1.85

Sc 0.60 22 5.1 1 66 0.49 672 0.73 5.28 10.78 1.41 0.24 5.91 354.7 163.7 518.37 20.2

Sc 0.78 29 5.1 1 93 0.63 743 0.85 9.61 15.25 1.34 0.3 4.46 359.4 85.33 444.72 9.2

Rf 0.80 62 3.7 1 49 1.17 436 2.68 5.14 4.39 1.12 0.1 11.13 282.2 14.08 296.27 1.1

Rf 0.26 44 3.7 3 114 1.8 525 3.43 5.11 2.84 1.09 0.11 10.36 712 155.3 867.33 1.75

Sc 0.76 57 4.3 1 120 0.82 402 2.04 10 12.2 1.58 0.11 14.75 404.7 2.56 407.25 1.05

Sc 0.66 51 4.3 1 102 0.6 340 1.77 8.11 13.52 1.31 0.16 8.32 173.5 2.96 176.45 8.5

3.3. Ectomycorrhizal fungal communities of moderately and severely declining forest The Cortinariaceae tended to have higher species richness in moderately declining forest, while Russulaceae had higher species richness in severely declining forest. Of the species recorded in multiple plots, 34% of Cortinariaceae species were unique to moderately declining plots while 18% of Russulaceae species were only recorded in severely declining plots. Canonical analysis of principal coordinates indicated that the ECM fungal community composition and structure of moderately and severely declining eucalypt plots were significantly different (CAP composition p = 0.04 and mis-classification =25%, CAP structure p = 0.08 misclassification =25%). A higher proportion of Cortinariaceae was found within moderately declining forest plots compared to plots with severe decline (Table 2). ECM fungal community composition was moderately correlated to crown health in the CCorA (Fig. 2) indicating that plots with similar levels of crown health tended to exhibit similar ECM fungal community compositions. 3.4. Relationship of ectomycorrhizal fungal communities to nutrients The variation in ECM richness was explained by a linear model that included two significant predictors, available P (p < 0.01) and 2 available NO 3 (p < 0.01) (R = 77.3):

ECM richness ¼ 95:28  24:13ðavailable soil PÞ  21:22ðavailable soil NO3 Þ: None of the other soil or foliage nutrient variables that were  available for model selection (soil pH, soil NO 3 , soil NH4 , total soil  N, total soil P, foliage N, foliage P, and available NH4 ) were significant in predicting ECM species richness and were not included in the final model. Ectomycorrhizal species richness was greatest at low levels of available P and/or NO 3 (Fig. 3a and b) and was lowest in plots where trees were in severe decline (and had highest levels of available NO 3 ). Seventy-two percent of the variation in ECM fungal community composition was explained with five axes in a distance based linear model. Soil pH (p < 0.01) and available NO 3 (p < 0.01) were highly significant in the model while soil P (p = 0.04), soil NO 3 (p = 0.04) and crown health (p = 0.04) were also significant in pre-

Mean

Median

n/a 0.51 41 4.54 ± 0.17 2.92 ± 1.49 95.67 ± 10.93 0.88 ± 0.12 472.83 ± 50.63 2.17 ± 0.34 8.34 ± 0.62 11.27 ± 1.49 1.39 ± 0.05 0.15 ± 0.02 11.37 ± 1.4 793.65 ± 252.58 236.88 ± 93.37 1030.53 ± 336.43 5.14 ± 1.74

n/a 0.52 37 4.4 1 105 0.73 468 2.36 9.79 11.35 1.41 0.11 10.75 454.88 139.85 527.41 1.83

dicting ECM fungal community composition. Plots clustered on the basis of understorey type and region (Fig. 4a). When woody plant species abundances were trialed in the model, none of the vegetation variables were significant. The final model selected only soil pH (p < 0.01), available NO 3 (p = 0.01) and soil P (p = 0.04) as predictors (Fig. 4a). A separate model exploring predictors of ECM fungal community structure rather than composition, indicated that soil pH (p < 0.01) and SOC (p = 0.02) were significant variables (Fig. 4b). Again, plots clustered according to understorey type and region, but the relative position of rainforest and sclerophyll plots on axis 2 (SOC) switched between the two regions. Cortinariaceae richness significantly decreased as available NO 3 concentrations increased (p = 0.01, Fig. 5a) and also increased marginally as soil pH increased (p = 0.06, Fig. 5b). The richness of the Russulaceae showed no relationship to available NO 3 (Fig. 5c) but was significantly positively correlated to soil pH (p < 0.01, Fig. 5d). 4. Discussion We found evidence to link ECM fungal communities to eucalypt decline in E. delegatensis forest, mediated through soil chemistry. Soil conditions shift from N limitation (foliar N:P < 14) to P limitation (foliar N:P > 16) as these forests become increasingly unhealthy as indicated by higher foliar N:P in severely declining forest (mostly due to reduced soil and foliage P, suggesting high P demand). These changes in the soil environment associated with eucalypt forest decline (i.e. increases in soil N, notably available NO 3 and decreases in soil P) influence ECM fungal communities of E. delegatensis forests as soil mineral N, soil P and soil pH are important determinants of ECM fungal community diversity (richness, composition and structure). Furthermore, tree health predicted ECM fungal community compositional similarities of the plots, and moderately and severely declining forest supported unique ECM fungal community compositions. These differences included shifts in the relative dominance of ECM families, as well as the assemblage of species present. The Cortinariaceae had greater richness in healthier forest than severely declining forest, whilst the Russulaceae were richer in the severely declining forest, with both groups showing strong correlations to particular soil nutrients. The strong correlation between edaphic variables and ECM fungal communities of E. delegatensis forest indicates that the differences in ECM fungal communities in forest of varying health are related to levels of soil chemistry, in particular available N,

334

B.M. Horton et al. / Forest Ecology and Management 302 (2013) 329–337

Table 2 The proportion of three ectomycorrhizal families in plots with eucalypts showing moderate or severe decline. Significant p values are in bold. Family

Severe decline (mean)

Moderate decline (mean)

d.f.

T-statistic

p-Value

Cortinariaceae Russulaceae Thelephoraceae

0.30 0.19 0.05

0.46 0.14 0.02

9 10 8

2.38 0.60 1.60

0.04 0.56 0.15

Fig. 2. Canonical correlations analysis of ectomycorrhizal community composition and crown health. Filled markers represent forest plots with trees in severe decline and open markers represent plots with trees in moderate decline, squares represent forest with a rainforest understorey and circles a dry sclerophyll understorey, northeast plots are black and northwest plots grey. Ectomycorrhizal composition similarities of plots 1 – 12, depicted as proximity of plots to one another along CAP1 axis, is moderately correlated to the average crown health of each plot (y = 0.75x + 0.5, R2 = 0.71, p = 0.2).

per se. If the decline is very severe then the reduction in tree health is likely to have a stronger effect on the fungal community due to loss of C source, unless effectively replaced by alternative hosts. Available NO 3 significantly predicted ECM fungal community composition as well as ECM richness, which varied inversely with available NO 3 . Soil N is likely a key driver of changes in the ECM fungal communities of E. delegatensis forest as larger changes in concentrations were observed in declining forests compared to other soil nutrients. ECM sporocarp richness of northern hemisphere boreal forests was strongly correlated to foliar N concentra-

tions and soil N availability (Kranabetter et al., 2009a,b). The relationship reported by Kranabetter et al. (2009b) differed to those reported for artificial N gradients where N addition has been shown to decrease detected sporocarp richness (Lilleskov et al., 2001; Peter et al., 2001b; Lilleskov et al., 2002; Avis et al., 2003). In boreal forests, Inocybe, Lactarius and Russula were found to favour soils with elevated inorganic N whilst Cortinarius species richness peaked on moderately fertile soils with strong organic N components (Kranabetter et al., 2009b). Genera such as Cortinarius, Russula, Tricholoma, and Lactarius decline in species richness or abundance with increasing mineral N and these genera are described as ‘nitrophobic’ (Lilleskov et al., 2001). Consistent with this, we found that species richness of Cortinarius decreased with increases in available NO 3 and like Kranabetter et al. (2009a) the converse was observed for species richness of Russula. ECM species that respond to differences in N availability and type are likely to differ in functional traits such as hydrophobicity, exploration type and protein use (Lilleskov et al., 2011). Our results showed a negative relationship between ECM species richness and available soil P. Ectomycorrhizal fungi are known to be important for plant P uptake, especially in situations of low P availability (Bolan, 1991; Richardson, Hocking et al., 2009). Like mineral N, P was significant in predicting ECM species richness and fungal community composition indicating that P is also an important determinant of the ECM fungal community. The relationship between ECM richness and soil P in this study is consistent with trends reported by Twieg et al. (2009) who found that levels of mineral soil P were negatively correlated to ECM species richness in a Douglas-fir forest. Soils in E. delegatensis forest that have lower concentrations of soil P may therefore have higher ECM richness, although the influence of P on the ECM fungal community would be masked by any increases in soil NO 3 , resulting in an overall decrease of ECM species richness. ECM species differ in their preference for substrate pH (Hung and Trappe, 1983; McAfee and Fortin, 1987; Sundari and Adholeya, 2003) and soil pH has been found to predict both ECM abundance and richness (Rühling and Tyler, 1990; Lu et al., 1999; Lilleskov et al., 2002). Our study strongly supports these findings. Soil pH significantly predicted the structure of ECM fungal communities in E. delegatensis forest, explaining more variation than other soil variables. Species in the Cortinariaceae made up a greater percentage of the ECM fungal community under more acidic soil conditions in E. delegatensis plots, whereas species in the Russulaceae made up a greater percentage of the community under less acidic soil conditions. These differences potentially indicate inherent adaptations within these taxonomic groups to different soil pH conditions.

Fig. 3. Relationship between available nitrate and phosphorus, and ectomycorrhizal fungal species richness. (a) ECM richness and available soil NO 3 . (b) ECM richness and available soil P. Soil variables were transformed prior to analysis and are presented as log (x + 1). Filled markers represent observed values; crosses show predicted values  from the model (not illustrated): ECM richness = 95.28  24.13(available soil P) – 21.22(available soil NO3 ).

B.M. Horton et al. / Forest Ecology and Management 302 (2013) 329–337

335

Fig. 4. Distance-based multivariate multiple linear regression model of ectomycorrhizal community composition and structure using Bray-Curtis similarity matrix and all foliage and soil nutrient concentrations, plus crown health, as predictor variables. Spearman’s Rho correlations of variables with axes 1 and 2 are shown as vectors for values >0.75. Filled markers represent forest plots with trees in severe decline and open markers represent plots with trees in moderate decline, squares represent forest with a rainforest understorey and circles a dry sclerophyll understorey, northeast plots are black and northwest plots grey. (a) ECM community composition AIC = 90.95 and R2 = 0.52. Plots in the positive region of axis 1 have more acidic soils and higher concentrations of soil P, while those in the negative region of axis 2 have higher 2 concentrations of available soil NO 3 (b) ECM community structure AIC = 76.65 and R = 0.44. Plots located in the positive region of axis 1 had higher soil pH while plots located in the positive area of axis 2 had greater concentrations of SOC.

Fig. 5. Linear regressions between two soil variables and richness of two ectomycorrhizal fungal families. Soil variables were transformed prior to analysis and are presented as log (x + 1). Filled markers represent forest plots with trees in severe decline and open markers represent plots with trees in moderate decline, squares represent forest with a rainforest understorey and circles a dry sclerophyll understorey, northeast plots are black and northwest plots grey. Plot numbers are labeled above markers. (a) 2 2 Cortinariaceae = 4.88  available NO 3 + 57.80, R = 0.47, p = 0.01. (b) Cortinariaceae = 69.34  soil pH + 154.82, R = 0.32, p = 0.06 c. Russulaceae = 0.74  available 2 2 NO + 13.55, R = 0.02, p = 0.68 d. Russulaceae = 88.32  soil pH – 133.98, R = 0.84, p < 0.01. 3

Soil organic carbon was also significant in predicting ECM fungal community structure in our declining E. delegatensis forest. This relationship was not due to the influence of understorey type as patterns in SOC of rainforest and sclerophyll plots differed between regions. Many species of ECM fungi occupy the organic soil and litter layers at the top of the soil profiles, and the level of organic C may be important for habitat selection by ECM fungi. Levels of SOC are cor-

related with soil N (Attiwill and Leeper, 1987) and may be a reflection of other soil conditions that are influencing the presence of ECM taxa. In the absence of fire, altered C:N ratios have been linked with eucalypt forest decline (Polglase et al., 1986; Turner et al., 2008) resulting in higher mineralisation and accelerated nutrient cycling (Adams and Attiwill, 1982) which may impact on ECM fungal communities of declining E. delegatensis forests.

336

B.M. Horton et al. / Forest Ecology and Management 302 (2013) 329–337

The results of this study further allude to the intimate and determinative relationship between understorey vegetation and ECM fungal communities. Vegetation community attributes such as tree species, species dominance and species composition are known to influence the diversity and structure of ECM fungal communities (Nantel and Neumann, 1992; Aerts, 2002; Debellis et al., 2006; Twieg et al., 2007). Distinct forest types tended to support similar assemblages of ECM species, suggesting that the vegetation community also plays a role in shaping the ECM fungal community. Although important our results suggest that vegetation is the not the driving force is determining ECM fungal communities of these declining forests. It is likely that understorey vegetation in these forests is being driven by the soil environment (especially pH) ((Barbier et al., 2008; Horton, 2011), and like fungal communities, vegetation responds to these changed conditions in declining forests. Changed climate such as altered rainfall patterns and temperatures may also contribute or exacerbate eucalypt dieback and decline (Allen et al., 2010) and potentially would also alter the ECM fungal community dynamics (Swaty et al., 2004) which would further contribute to a positive feedback cycle of declining ecosystem health.

5. Conclusions This study of eucalypt forest decline is the first to investigate the link between ECM fungal community attributes and health of the forest trees. In the absence of disturbance such as fire, N accumulates to a level where nitrification occurs, and pH decreases. P availability is reduced along with a substantial reduction in P uptake, reflected in the higher foliar N:P ratios. Under these conditions (a shift to P limitation from N limitation) eucalypts may lose their competitive advantage leading to altered vegetation community composition, explaining the tendency for forest with a rainforest understory to exhibit more severe decline symptoms. The ECM fungal communities of E. delegatensis forests play an important role in this eucalypt forest decline process, responding to changes in soil chemistry (increased N and declining P and pH). Shifts in ECM fungal community composition and structure as a result of altered soil chemistry under conditions of forest decline include a reduction in Cortinariaceae diversity, increases in Russulaceae species and a decrease in overall ECM species richness. These changes impact on ecosystem function and tree health, and ultimately affect plant performance. Tree nutrient acquisition could be reduced (i.e. P uptake) due to these shifts in ECM fungal communities and associated altered functions. Particular species and assemblages of ECM species may be better adapted to cope with ecosystem changes associated with eucalypt decline (e.g. the Cortinariaceae or Russulaceae which were related to different level of forest decline and soil chemistry) and could be important for the recovery of these forests. The use of disturbance, such as fire, should be considered an important tool for the management of healthy eucalypt forest in order to maintain diverse ECM fungal communities along soil chemistry gradients. Similar changes in soil chemistry associated with decline (increases in N, decreasing C:N and increased acidity) have been observed with increasing time since fire (Turner et al., 2008). Fire can act to change the availability and form of soil nutrients in the environment (Raison, 1980; Certini, 2005) which would then influence the composition, structure and function of the ECM fungal and vegetation communities. Management actions should be implemented prior to obvious symptoms of forest decline in order to promote functionally appropriate ECM fungal communities for the maintenance of healthy forest ecosystems according to the underlying edaphic conditions. Soil chemistry indicators may

be useful in recognising a possible decline in forest health before the onset of canopy dieback.

Acknowledgements This research was made possible by funding from the Bushfire Co-operative Research Centre, the Holsworth Wildlife Trust, Forestry Tasmania, the Maxwell Jacobs Trust, the Jill Landsberg Trust and the University of Tasmania. We wish to thank the many people who gave technical assistance and statistical advice, especially David Page, Dr. Genevieve Gates, Chris Ware, Luci Augustini and the Bruns Laboratory, UC Berkeley. Many thanks to Dr. Alieta Eyles for her constructive feedback on the manuscript.

References Abarenkov, K., Henrik Nilsson, R., et al., 2010. The unite database for molecular identification of fungi – recent updates and future perspectives. New Phytologist 186 (2), 281–285. Adams, M.A., Attiwill, P.M., 1982. Nitrogen mineralization and nitrate reduction in forests. Soil Biology & Biochemistry 14 (3), 197–202. Aerts, R., 2002. The role of various types of mycorrhizal fungi in nutrient cycling and plant competition. In: van der Heijden, M.G.A., Sanders, I.R. (Eds.), Mycorrhizal Ecology. Springer-Verlag, Berlin. Agerer, R., 1991. Characterization of ectomycorrhiza. In: Norris, J.R., Read, D.J., Varma, A.K. (Eds.), Methods in Microbiology, vol. 23. Academic press, London, San Diego, pp. 25–73. Allen, C.D., Macalady, A.K., et al., 2010. A global overview of drought and heatinduced tree mortality reveals emerging climate change risks for forests. Forest Ecology and Management 259 (4), 660–684. Altschul, S.F., Madden, T.L., et al., 1997. Gapped blast and psi-blast: a new generation of protein database search programs. Nucleic Acids Research 25, 3389–3402. Anderson, M.J., Willis, T.J., 2003. Canonical analysis of principal coordinates: a useful method of constrained ordination for ecology. Ecology 84 (2), 511–525. Anderson, M.J., Gorley, R.N., et al., 2008. Permanova+ for primer: guide to software and statistical methods. PRIMER-E, Plymouth. Arnolds, E., 1991. Decline of ectomycorrhizal fungi in Europe. Agriculture Ecosystems & Environment 35 (2–3), 209–244. Attiwill, P.M., Leeper, G.W., 1987. Forest Soils and Nutrient Cycles. Melbourne University Press, Melbourne. Avis, P.G., McLaughlin, D.J., et al., 2003. Long-term increase in nitrogen supply alters above- and below-ground ectomycorrhizal communities and increases the dominance of russula spp. In a temperate oak savanna. New Phytologist 160 (1), 239–253. Barbier, S., Gosselin, F., et al., 2008. Influence of tree species on understory vegetation diversity and mechanisms involved – a critical review for temperate and boreal forests. Forest Ecology and Management 254, 1–15. Bolan, N.S., 1991. A critical review on the role of mycorrhizal fungi in the uptake of phosphorus by plants. Plant and Soil 134 (2), 189–207. Certini, G., 2005. Effects of fire on properties of forest soils: a review. Oecologia 143, 1–10. Chevone, B.I., Linzon, S.N., 1988. Tree decline in North America. Environmental Pollution 50 (1–2), 87–99. Ciesla, W.M, Donaubauer, E., 1994. Decline and dieback or trees and forests: a global overview. FAO forestry paper 120. Food and Agriculture Division of the United Nations. Close, D.C., Davidson, N.J., et al., 2009. Premature decline of eucalyptus and altered ecosystem processes in the absence of fire in some australian forests. Botanical Review 75, 191–202. Crous, P.W., Gams, W., et al., 2004. Mycobank: an online initiative to launch mycology into the 21st century. Studies in Mycology 50, 19–22. Debellis, T., Kernaghan, G., et al., 2006. Relationships between stand composition and ectomycorrhizal community structure in boreal mixed-wood forests. Microbial Ecology 52, 114–126. Ellis, R.C., Pennington, P.I., 1992. Factors affecting the growth of eucalyptus delegatensis seedlings in inhibitory forest and grassland soils. Plant and Soil 145 (1), 93–105. Ellis, R.C., Mount, A.B., et al., 1980. Recovery of eucalyptus delegatensis from high altitude dieback after felling and burning the understorey. Australian Forestry 43 (1), 29–35. Estivalet, D., Perrin, R., et al., 1990. Nutritional and microbiological aspects of decline in the vosges forest area (france). Forest Ecology and Management 37 (4), 233–248. Fellner, R., Pešková, V., 1995. Effects of industrial pollutants on ectomycorrhizal relationships in temperate forests. Canadian Journal of Botany 73 (Suppl. 1), S1310–S1315. Felsenstein, J., 1989. Phylip – phylogeny inference package (version 3.2). Cladistics 5, 164–166.

B.M. Horton et al. / Forest Ecology and Management 302 (2013) 329–337 Gardes, M., White, T.J., et al., 1991. Identification of indigenous and introduced symbiotic fungi in ectomycorrhizae by amplification of nuclear and mitochondrial ribosomal DNA. Canadian Journal of Botany 69 (1), 180–190. Gates, G.M., Mohammed, C., et al., 2011. Diversity and ecology of epigeous ectomycorrhizal macrofungal assemblages in a native wet eucalypt forest in tasmania, australia. Fungal Ecology 4 (4), 290–298. Giblin, A.E., Laundre, J.A., et al., 1994. Measuring nutrient availability in arctic soils using ion-exchange resins – a field-test. Soil Science Society of America Journal 58, 1154–1162. Glen, M., Tommerup, I.C., et al., 2002. Are sebacinaceae common and widespread ectomycorrhizal associates of eucalyptus species in australia? Mycorrhiza 12, 243–247. Glen, M., Bougher, N.L., et al., 2008. Ectomycorrhizal fungal communities of rehabilitated bauxite mines and adjacent, natural jarrah forest in western australia. Forest Ecology and Management 255, 214–225. Grant, J.C., Laffan, M.D., et al., 1994. Forest soil of tasmania: a handbook for identification and management. Forestry Tasmania, Hobart. Heatwole, H., Lowman, M., 1986. Dieback Death of an Australian Landscape. NSW, Reed Books, Sydney. Hibbett, D.S., Binder, M., et al., 2007. A higher-level phylogenetic classification of the fungi. Mycological Research 111 (5), 509–547. Horton, B.M., Close, D.C., et al., 2011. Crown condition assessment: An accurate, precise and efficient method with broad applicability to Eucalyptus. Austral Ecology 36, 709–721. Horton, B. M. Eucalypt decline and ectomycorrhizal fungal community ecology of Eucalyptus delegatensis forest, Tasmania, Australia. Unpublished PhD thesis, University of Tasmania, 2011. Houlder, D., M. Hutchinson, et al., 2000. Anuclim version 5.2. Canberra, Centre for Resource and Environmental Studies, Australian National University. Hung, L.L., Trappe, J.M., 1983. Growth variation between and within species of ectomycorrhizal fungi in response to ph in vitro. Mycologia 75, 234–241. Innes, J.L., 1992. Forest decline. Progress in Physical Geography 16 (1), 1–64. Ishida, T.A., Nara, K., et al., 2007. Host effects on ectomycorrhizal fungal communities: insight from eight host species in mixed conifer-broadleaf forests. New Phytologist 174, 430–440. Jansen, A.E., 1991. The mycorrhizal status of douglas fir in the netherlands: its relation with stand age, regional factors, atmospheric pollutants and tree vitality. Agriculture, Ecosystems & Environment 35, 191–208. Jurskis, V., 2005. Decline of eucalypt forests as a consequence of unnatural fire regimes. Australian Forestry 68 (4), 257–262. Jurskis, V., Turner, J., 2002. Eucalypt dieback in eastern australia: a simple model. Australian Forestry 65, 87–98. Kranabetter, J., Durall, D., et al., 2009a. Diversity and species distribution of ectomycorrhizal fungi along productivity gradients of a southern boreal forest. Mycorrhiza 19 (2), 99–111. Kranabetter, J., Friesen, J., et al., 2009b. Epigeous fruiting bodies of ectomycorrhizal fungi as indicators of soil fertility and associated nitrogen status of boreal forests. Mycorrhiza 19 (8), 535–548. Lilleskov, E.A., Fahey, T.J., et al., 2001. Ectomycorrhizal fungal aboveground community change over an atmospheric nitrogen deposition gradient. Ecological Applications 11 (2), 397–410. Lilleskov, E.A., Fahey, T.J., et al., 2002. Belowground ectomycorrhizal fungal community change over a nitrogen deposition gradient in alaska. Ecology 83 (1), 104–115. Lilleskov, E.A., Hobbie, E.A., et al., 2011. Conservation of ectomycorrhizal fungi: exploring the linkages between functional and taxonomic responses to anthropogenic n deposition. Fungal Ecology 4 (2), 174–183. Lu, X., Malajczuk, N., et al., 1999. Fruiting of putative ectomycorrhizal fungi under blue gum (eucalyptus globulus) plantations of different ages in western australia. Mycorrhiza 8 (5), 255–261. McAfee, B., Fortin, J., 1987. The influence of ph on the competitive interactions of ectomycorrhizal mycobionts under field conditions. Canadian Journal of Botany 17 (8), 859–864. Mueller-Dombois, D., 1988. Forest decline and dieback – a global ecological problem. Trends in Ecology & Evolution 3 (11), 310–312. Nantel, P., Neumann, P., 1992. The ecology of ectomycorrhizal-basidiomycete communities on a local vegetation gradient. Ecology 73 (1), 99–117.

337

Old, K.M., Kile, G.A., et al., 1980. Eucalypt Dieback in Forests and Woodlands. CSIRO, Canberra. Peintner, U., Moser, M., et al., 2002. Thaxterogaster is a taxonomic synonym of cortinarius: new combinations and new names. Mycotaxon 81, 177–184. Perrin, R., Estivalet, D., 1990. Mycorrhizal association and forest decline (yellowing of spruce). Agriculture Ecosystems & Environment 28 (1–4), 381–387. Peter, M., Ayer, F., et al., 2001a. Above- and below-ground community structure of ectomycorrhizal fungi in three norway spruce (Picea abies) stands in switzerland. Canadian Journal of Botany 79 (10), 1134–1151. Peter, M., Ayer, F.O., et al., 2001b. Nitrogen addition in a norway spruce stand altered macromycete sporocarp production and below-ground ectomycorrhizal species composition. New Phytologist 149 (2), 311–325. Peter, M., Ayer, F., et al., 2008. Belowground ectomycorrhizal communities in three norway spruce stands with different degrees of decline in the Czech Republic. Mycorrhiza 18, 157–169. Podger, F.D., Kile, G.A., et al., 1980. An unexplained decline in some forests of Eucalyptus obliqua and eucalyptus regnans in southern Tasmania. Australian Forest Research 10 (1), 53–70. Polglase, P.J., Attiwill, P.M., et al., 1986. Immobilization of soil nitrogen following wildfire in two eucalypt forests of southeastern Australia. Oecologia Plantarum 7 (3), 261–271. Raison, R.J., 1980. A review of the role of fire in nutrient cycling in Australian native forests, and of methodology for studying the fire-nutrient interaction. Austral Ecology 5 (1), 15–21. Rühling, Å., Tyler, G., 1990. Soil factors influencing the distribution of macrofungi in oak forests of southern Sweden. Holarctic Ecology, 11–18. Sambrook, J., Fritsch, E.F., et al., 1989. Cold Spring Laboratory Press, Cold Spring Harbour, NY. Scott, P.M., Shearer, B.L., et al., 2012. Fine root and ectomycorrhizae density of declining Eucalyptus gomphocephala’’. Australasian Plant Pathology, 1–11. Smith, S.E., Read, D.J., 1997. Mycorrhizal Symbiosis. Academic Press, San Diego. Sundari, S.K., Adholeya, A., 2003. Growth profile of ectomycorrhizal fungal mycelium; emphasis on substrate pH influence. Antonie van Leeuwenhoek 83, 209–214. Swaty, R.L., Deckert, R.J., et al., 2004. Ectomycorrhizal abundance and community composition shifts with drought: predictions from tree rings. Ecology 85 (4), 1072–1084. Tedersoo, L., Jairus, T., et al., 2008. Strong host preference of ectomycorrhizal fungi in a Tasmanian wet sclerophyll forest as revealed by DNA barcoding and taxonspecific primers. New Phytologist 180 (2), 479–490. Thompson, J.D., Higgins, D.G., et al., 1994. CLUSTAL W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position-specific gap penalties and weight matrix choice. Nucleic Acids Research 22 (22), 4673–4680. Turner, J., Lambert, M., 2005. Soil and nutrient processes related to eucalypt forest dieback. Australian Forestry 68 (4), 251–256. Turner, J., Lambert, M., et al., 2008. Long term accumulation of nitrogen in soils of dry mixed eucalypt forest in the absence of fire. Forest Ecology and Management 256, 1133–1142. Twieg, B.D., Durall, D.M., et al., 2007. Ectomycorrhizal fungal succession in mixed temperate forests. New Phytologist 176 (2), 437–447. Twieg, B.D., Durall, D.M., et al., 2009. Influence of soil nutrients on ectomycorrhizal communities in a chronosequence of mixed temperate forests. Mycorrhiza 19, 305–316. Vinceti, B., Paoletti, E., et al., 1998. Analysis of soil, roots and mycorrhizae in a Norway spruce declining forest. Chemosphere 36 (4–5), 937–942. VSN International Ltd., 2003. Genstat v7.1. Wardlaw, T.J., 1989. Management of Tasmanian forests affected by regrowth dieback. New Zealand Journal of Forest Science 19, 265–276. Wardle, D.A., Walker, L.R., et al., 2004. Ecosystem properties and forest decline in contrasting long-term chronosequences. Science 305 (5683), 509–513. White, T.J., Bruns, T., et al., 1990. Amplification and direct sequencing of fungal ribosomal rna genes for phylogenetics. In: Innis, M.A., Gelfand, D.H., Sninsky, J.J., White, T.J. (Eds.), PCR protocols: A Guide to Methods and Applications. San Diego Academic Press, pp. 315–322.