TRANSLOCATION AND ACCUMULATION OF ORGANIC AND INORGANIC NITROGEN IN WOOD RESOURCES COLONIZED BY THE MYCELIAL CORD SYSTEMS OF THE DECAY FUNGUS HYPHOLOMA FASCICULARE by Timothy James Philpott B.Sc (Honours), Queen’s University, 2009 A THESIS SUBMITTED IN PARTIAL FULLFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE in The Faculty of Graduate Studies (Forestry) THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver) April 2012 © Timothy James Philpott, 2012 ii ABSTRACT Translocation of nitrogen (N) through mycelial cords of wood decay fungi is thought to be the mechanism responsible for the observed increase in absolute N content in woody debris over time. This research evaluates the ability of the mycelial cords of the wood decay fungus Hypholoma fasciculare to translocate and accumulate labeled organic ( 15 N-glycine, 15 N Douglas- fir litter) and inorganic N ( 15 NH4 + , 15 NO3 - ) in its wood substrate. Each N form was supplied separately to the growing fronts of mycelial cords established over 67 days from wood blocks (Douglas-fir) in soil microcosms. Three sampling occasions (days 6, 18 and 30 after N addition) were used to identify trends in 15 N transfer and total N accumulation. Wood blocks inoculated with Hypholoma fasciculare assimilated significantly more 15 N than uninoculated blocks for all 15 N treatments on at least one sampling occasion. After 73 days of incubation (day 6 sampling occasion), inoculated wood blocks increased in absolute N content by 211% relative to uninoculated control blocks, but 80% of this accumulated N was lost after 97 days of incubation (day 30 sampling occasion). The small amount of 15 N that was transferred contrasted with the large increase in total N, suggesting that the site of N transfer was largely from the soil underneath wood blocks rather than at the site of 15 N injection. The precipitous decline in absolute N content was attributed to visible indications of mycelial senescence. This research demonstrates that the mycelial cords of Hypoloma fasciculare are capable of translocating 15 N into a wood substrate and can also greatly increase the absolute N content of wood blocks. The results are discussed in the context of fungal ecology as well as woody debris management. iii TABLE OF CONTENTS ABSTRACT ..................................................................................................................................................... ii LIST OF TABLES ............................................................................................................................................. v LIST OF FIGURES .......................................................................................................................................... vi LIST OF ABBREVIATIONS............................................................................................................................. vii ACKNOWLEDGEMENTS .............................................................................................................................. viii 1 RATIONALE ........................................................................................................................................... 1 2 LITERATURE REVIEW ............................................................................................................................ 3 2.1 Coarse woody debris in northern forests ....................................................................................... 3 2.2 Decay and nitrogen dynamics of CWD ........................................................................................ 5 2.2.1 Physical aspects of coarse woody debris decomposition ...................................................... 5 2.2.2 Nitrogen dynamics of CWD through decay .......................................................................... 6 2.3 The role of wood decay fungi in CWD decomposition .............................................................. 10 2.3.1 Structure and foraging strategies of mycelial networks ...................................................... 10 2.3.2 Fungal decomposition of coarse woody debris ................................................................... 12 2.3.3 Enzymatic decomposition of wood ..................................................................................... 15 2.3.4 Nutrient uptake, translocation, and release ......................................................................... 17 2.4 Effects of woody debris removal on future forest productivity .................................................. 20 2.4.1 Nitrogen leaching and woody debris removal .................................................................... 21 2.4.2 Woody debris removal and site fertility .............................................................................. 22 2.5 Summary ..................................................................................................................................... 24 3 HYPOTHESES ....................................................................................................................................... 26 4 METHODS ............................................................................................................................................ 28 4.1 Overview ..................................................................................................................................... 28 4.2 Sample collection ........................................................................................................................ 28 4.3 Preparation and inoculation of wood blocks ............................................................................... 30 4.4 Microcosm development ............................................................................................................. 30 4.5 Preliminary experiment: fungal biomass measurement and 15 N movement through soil ........... 33 4.6 Main experiment ......................................................................................................................... 35 4.7 Douglas-fir litter experiment ....................................................................................................... 36 4.8 Experiment: transfer of 15 NO3 - to wood blocks ........................................................................... 37 4.9 Elemental and isotope analysis ................................................................................................... 38 4.10 Phospholipid Fatty Acid (PLFA) Analysis ................................................................................. 39 4.11 Calculations ................................................................................................................................. 42 iv 4.12 Statistics ...................................................................................................................................... 43 5 RESULTS .............................................................................................................................................. 45 5.1 Hypothesis 1: 15 N transfer via mycelial cords ............................................................................. 45 5.1.1 Preliminary experiment ....................................................................................................... 45 5.1.2 Main experiment ................................................................................................................. 47 5.2 Hypothesis 2: preferential use of N forms .................................................................................. 54 5.3 Hypothesis 3: absolute N accumulation ...................................................................................... 55 5.4 Question 1: transfer of litter derived 15 N ..................................................................................... 55 5.5 Question 2: temporal patterns of N transfer and accumulation ................................................... 56 5.6 Transfer of 15 NO3 - to uninoculated wood blocks ........................................................................ 58 6 DISCUSSION ......................................................................................................................................... 63 6.1 Hypothesis 1: 15 N transfer via mycelial cords ............................................................................. 63 6.2 Hypothesis 2: preferential use of N forms .................................................................................. 65 6.3 Hypothesis 3: absolute N accumulation ...................................................................................... 66 6.4 Question 1: transfer of litter derived 15 N ..................................................................................... 66 6.5 Question 2: temporal patterns of N transfer and accumulation ................................................... 67 6.6 15 NO3 - transfer to uninoculated wood blocks .............................................................................. 69 6.7 Implications for fungal ecology and woody debris management ............................................... 71 7 CONCLUSIONS ..................................................................................................................................... 74 8 RECOMMENDATIONS AND FUTURE RESEARCH ................................................................................... 75 REFERENCES ............................................................................................................................................... 77 v LIST OF TABLES Table 4-1 Inorganic N (NO3 - , NH4 + ), total soluble N, soluble organic N and amino acid N concentrations (mg·kg -1 ) from collected soil. ............................................................................... 30 Table 4-2 Yielded %N and δ15N values for standards used in wood sample calibrations ............ 39 Table 4-3 PLFAs identified from PLFA extraction procedure ..................................................... 41 vi LIST OF FIGURES Figure 4-1 Preliminary microcosm design .................................................................................... 31 Figure 4-2 Final microcosm design .............................................................................................. 33 Figure 5-1 Abundance of the PLFA biomarker 18:2ω6,9 (relative to total extractable PLFAs) for four soil portions in inoculated and uninoculated microcosms .................................................... 46 Figure 5-2 δ15N value (‰) of each soil portion in inoculated and uninoculated microcosms. .. 47 Figure 5-3 Soil δ15N values in inoculated and uninoculated soil portions for each set of 15N treated microcosms; A) 15 N-glycine B) 15 NH4 + C) 15 NO3 - .. ......................................................... 50 Figure 5-4 15 N assimilated (A), absolute N content (B) and relative density (C) of inoculated and uninoculated wood blocks for 15 NH4 + treated microcosms at three sampling times (Days 6, 18 and 30).. ........................................................................................................................................ 51 Figure 5-5 15 N assimilated (A), absolute N content (B) and relative density (C) of inoculated and uninoculated wood blocks for 15 NO3 - treated microcosms at three sampling times (Days 6, 18 and 30). ......................................................................................................................................... 52 Figure 5-6 15 N assimilated (A), absolute N content (B) and relative density (C) of inoculated and uninoculated wood blocks for 15 N-glycine treated microcosms at three sampling times (Days 6, 18 and 30) ..................................................................................................................................... 53 Figure 5-7 15 N assimilated in inoculated wood blocks for each treatment ( 15 N-glycine, 15 NH4 + , 15 NO3 - ) and sampling time. ........................................................................................................... 54 Figure 5-8 15 N assimilated (A), absolute N content (B) and relative density (C) of inoculated and uninoculated wood blocks for microcosms treated with 15 N labelled Douglas-fir litter after 30 days of litter application................................................................................................................ 59 Figure 5-9 Absolute N content of inoculated and uninoculated wood blocks pooled across 15 N treatments at three sampling times (Days 73, 85 and 97) ............................................................. 60 Figure 5-10 Mass loss of inoculated wood blocks pooled across 15 N treatments at three sampling times (Days 73, 85 and 97). .......................................................................................................... 60 Figure 5-11 C/N ratio of uninoculated blocks and inoculated blocks at each sampling occasion (73, 85 and 97). Different letters indicate significant differences between sampling times and block type ...................................................................................................................................... 61 Figure 5-12 Relationship between final water content of wood blocks and assimilated 15 N after 15 NO3 - was added to the microcosms. ........................................................................................... 62 vii LIST OF ABBREVIATIONS 14 C 14 C isotope of carbon 15 N 15 N isotope of nitrogen AIB aminoisobutyric acid NH4 + ammonium NH4Cl ammonium chloride C carbon CWD coarse woody debris CH conventional harvesting DBH diameter at breast height FAME fatty acid methyl ester IAEA international atomic energy agency IRMS isotope ratio mass spectrometry NO3 - nitrate N nitrogen PLFA phospholipid fatty acid KNO3 potassium nitrate S.E. standard error USGS United States geological survey WTH whole tree harvesting viii ACKNOWLEDGEMENTS Some gracious thanks are in order for those who have guided me through this project. A special thanks to Drs. Cindy Prescott and Sue Grayston for giving me the freedom to design my own project and for the moral and academic support throughout this endeavour. As well, a much deserved thanks to Dr. Bill Chapman for tromping around the Alex Fraser Research Forest in search of Hypholoma fasciculare, and for his assistance with culturing. Dr. Tony Kozak was an invaluable resource without whom I would still be floundering in a sea garbled SAS code. A big thanks to Kate Del Bel and Alice Chang, who were instrumental in the BEG labs, pun very intended. As well, Patricia Micks from the University of Michigan School of Natural Resources and Environment, who still managed to analyse 15 N in my wood samples despite two laboratory floods. Of course, all the student members of BEG, who served as shoulders to cry on and friends to share a pint with. Again, a very special thank you to Cindy Prescott as well as Gayle Kosh for their tireless support of the Forestry Graduate Student Association; my involvement with this organisation will always bring back fond memories of the Faculty of Forestry. Not least, a warm thank you to the friends I made throughout this degree. This thesis would not have happened without your support. Thanks to NSERC, the Faculty of Forestry Internal Awards program, and the Faculty of Forestry Graduate Award program for generous personal and research support. And finally, Neil Young, Gordon Lightfoot, and Stan Rogers, it is a lonely ride up to Williams Lake, and you three, among others, made it entirely enjoyable. ix To Mom and Dad 1 1 RATIONALE Concern over rising carbon emissions is driving intensive forest biomass harvesting for use as feedstock in the bioenergy sector. The rationale behind this trend is simple: using forest residue as a fuel source can provide society with energy while dramatically reducing carbon emissions, potentially mitigating the effects of climate change. Northern European countries are already intensively harvesting forest residues for bioenergy production; Finland has seen a 22- fold increase in woody residue harvesting from 1995 – 2003 (Walmsley et al., 2009). Bioenery production and use in Canada is largely within the forestry sector, accounting for 6% of Canada’s energy needs (Bradley, 2006). With a 16.3 million ha mountain pine beetle outbreak, British Columbia is also poised to drastically increase its woody residue harvesting (Ministry of Forests and Range, 2010). However, a reduction in carbon emissions may be negated if removal of forest residue reduces long-term carbon gains in aboveground biomass via loss of site and soil fertility associated with residue removal. Coarse woody debris (CWD) is a major component of this residue and has been the focus of many studies. Coarse woody debris, in some instances, increases in absolute nitrogen (N) content over time (Laiho and Prescott, 2004), an observation that has fueled speculation that removal of this material will result in export of N off-site and cause reduced site fertility. Nutrient transfer by wood decay fungi is widely hypothesized as the dominant mechanism responsible for this increase (Boddy and Watkinson, 1995). The effects on site fertility of logging residue removal (including coarse woody debris) have been studied extensively since the 1970s. Lacking from the literature, however, is demonstration of the N transfer mechanism to woody debris and application of this mechanism 2 to estimate N immobilization in the CWD pool. This research will investigate capacity of the wood decay fungus Hypholoma fasciculare to translocate four forms of isotopically labeled N ( 15 NH4 + , 15 NO3 - , 15 N-glycine and litter derived 15 N) from soil to wood inoculated wood blocks. Additionally, the rate of 15 N transfer to wood inocula will be established over a time course. The capacity of the fungus to increase the absolute N content of a wood substrate will also be evaluated. 3 2 LITERATURE REVIEW 2.1 Coarse woody debris in northern forests Input of coarse woody debris (CWD) into a forest ecosystem is complex, often unpredictable, and is controlled primarily by factors affecting tree mortality. The proximate cause of tree death is often the end result of a continuum of processes that weaken tree vigor over its lifetime. Franklin et al. (1987) describe a mortality spiral that reflects the cumulative effect of multiple stresses that may eventually lead to tree death. For example, competition may cause susceptibility to defoliators, which may further weaken the tree and lower its resistance to bark beetles (Franklin et al., 1987). The beetles carry with them blue stain fungus, a disease to which the tree eventually succumbs. While the proximate cause of death is the fungus, the ultimate cause is a combination of many factors. The tree can also recover from its stresses and be released from the mortality spiral. The complexity of CWD input is thus partially derived from the inherent complexity of tree mortality. Some of the natural agents of CWD production (tree mortality) have already been listed, and are further summarized by Harmon et al. (1986). These include wind, fire, insects, disease, and suppression and competition. The dominant cause of tree mortality varies greatly between ecosystems and is often stochastic. Wind produces CWD by snapping and/or uprooting trees. Windthrow varies spatially and temporally, and its intensity and distribution is dependent on many biotic and abiotic factors including, species, stem condition, stand condition, storm intensity, topography, soil characteristics etc. (Everham and Brokaw, 1996). Fire, the main cause of tree death in British Columbia, produces CWD directly by girdling stems, scorching crowns or burning root systems, and indirectly by weakening the tree and making it susceptible to disease 4 organisms or windthrow (Harmon et al., 1986). As with windthrow, the amount of coarse woody debris produced by fire is dependent on many factors, such as, species, fire type and intensity, and forest structure (Harmon et al., 1986). Inputs of CWD after severe fires can be as high as 1000 Mg/ha in the coastal forests of the Pacific Northwest, which compares to 1200-2200 Mg/ha of CWD input over 500 years in these undisturbed forests (Harmon and Hua, 1991). Insect outbreaks can produce large quantities of woody debris; this is currently the case in British Columbia where a 16.3 million ha mountain pine beetle infestation has produced millions of standing dead stems (Ministry of Forests and Range, 2010). Non-disturbance input also occurs in stand regeneration from suppression, competition and through self-thinning (Laiho and Prescott, 2004). Patterns of CWD production have been described as ‘U-shaped’ with initially high CWD volume generated from the disturbance. This is followed by decay of this initial large input and eventual recovery of CWD volumes associated with input from self-thinning and small scale disturbances (e.g. windthrow) as the stand matures (Sturtevant et al., 1997). In managed stands, CWD volume is known to decline after successive rotations and is between 2-30% of the quantity in natural stands, primarily due to removal of stems, the source of CWD input (Fridman and Walheim, 2000). One of the challenges of studying the ecological role of CWD is trying to arrive at a standard definition. Generally, CWD consists of decaying logs and stumps. However, disagreement arrives when identifying minimum diameters and the number of decay classes used. Minimum debris diameters range from 1 to 15 cm in some studies (Feller, 2003), and number of decay classes also range from 2 to 5 in some studies (Harmon et al., 1986). The B.C. 5 Ministry of Forests and Range (Stevens, 1997) addresses this variability and arrives at the following definition for CWD: “Sound and rotting logs and stumps, and coarse roots in all stages of decay, that provide habitat for plants, animals and insects and a source of nutrients for soil structure and development. Material generally greater than 7.5 cm in diameter.” 2.2 Decay and nitrogen dynamics of CWD 2.2.1 Physical aspects of coarse woody debris decomposition From a physical perspective, decomposition of coarse woody debris as well as other litter types is primarily controlled by i) substrate quality, and ii) environment (aeration, moisture, and temperature) (Laiho and Prescott, 2004). While the microbial and faunal community carry out the brunt of the decomposition process, the rate of decay is dependent on these factors. Generally, gymnosperms decompose more slowly relative to angiosperms. This is mainly due to differences in lignin and hemicellulose composition and distribution between these two plant groups, and due to the presence of non-structured phenolics and aromatics (Weedon et al., 2009). Gymnosperm genera, such as Abies and Picea decay faster relative to Thuja and Tsuga (Yin, 1999), although variation also occurs within species in a genus (Laiho and Prescott, 2004). Environmental extremes impede decomposition by reducing the efficacy of the decomposer community. Rates of decomposition of CWD in northern coniferous forests, which are commonly estimated using an exponential decay model (Olson, 1963), are highly variable (Laiho and 6 Prescott, 2004). Species common in the boreal and sub-boreal forests of British Columbia include interior spruce (hybridized Picea glauca and Picea engelmannii), subalpine fir (Abies lasiocarpa), and lodgepole pine (Pinus contorta), with associated decay rates (k, year -1 ) of 0.0271, 0.0054, 0.0286, 0.0507, respectively (Laiho and Prescott, 1999; Johnson and Greene, 1991). The mean and median decay rate of several northern coniferous species falls around 0.02 year -1 (Laiho and Prescott, 2004). Hermann and Prescott (2008), through a 21-year log survey, predicted that the time to 95% decay of Pinus contorta, Picea glauca and Abies lasiocarpa logs was 42, 123 and 58 years, respectively. A previous 14-year survey of the same logs predicted that 95% decay would be reached over a shorter time period for all species, suggesting that the decay rate slows over time, also highlighting problems associated with predicting long-term decay from short term data (Hermann and Prescott, 2008). 2.2.2 Nitrogen dynamics of CWD through decay It is widely reported that nitrogen in CWD increases in concentration through the decomposition process (Alban and Pastor, 1993; Arthur and Fahey, 1990; Krankina et al., 1990; Laiho and Prescott, 2004; Palviainen et al., 2008). It is important to note the distinction between concentration and content, the latter being an absolute amount. Increases in N concentration in woody debris through time do not necessarily imply that a net increase in N content has been observed. Change in N concentration occurs concomitantly with a decrease in C/N ratio, the implication being that microbial respiration of CO2 has increased N concentration. That being said, increases in absolute N content in woody debris through time have been reported (Foster and Lang, 1982; Graham and Cromack, 1982; Sollins et al., 1987; Arthur and Fahey, 1989; 7 Alban and Pastor, 1993; Busse, 1994; Laiho and Prescott, 1999; Laiho and Prescott, 2004). Krankina et al. (1999) suggest that these reported increases in woody debris N content overestimate N accumulation due to improper or non-existent assessment of volume loss through time. Holub et al. (2001) developed a volume adjusted method and applied this to existing chronosequence studies of nutrient concentrations in CWD, concluding that N content increased through decomposition with a range of -5-50% change in decay class four, and 20-150% change in decay class 5. Proposed mechanisms for this increase include throughfall (Harmon et al., 1986), N fixation (Hendrickson, 1991) and import via wood decay fungi (Boddy and Watkinson, 1995). Nitrogen fixation in logs could be as high as 3 μg N·(g CWD)-1·day-1 (Holub et al., 2001), but this is still too low to explain observed N accumulation. It is possible that N translocation by cord-forming wood decay fungi is largely responsible for N accumulation in woody debris (Boddy and Watkinson, 1995). Evidence, as above, suggests that CWD has the capacity to accumulate N over time; CWD in some cases acts as an N sink. Release of this N into the forest ecosystem is thus of interest for forest N budgeting. In some cases, N release is reported relatively quickly in the decay process; Palviainen et al. (2008) report that Decay Class 5 logs of Scots pine, Norway spruce and silver birch had released 53%, 73% and 59% of their initial N content, respectively. In this case, between 40 and 50% of initial N was released by these species after 30 years (Palviainen et al. 2008). After 14 years of decay, Laiho and Prescott (1999) demonstrated that Pinus logs gained N, Picea logs released some N and Abies logs released almost 30% of their original N content. In this study, the original N content of the pine logs was lowest relative to the spruce logs, suggesting that CWD with low initial N content retains more N through decomposition compared to logs with high initial N content (Laiho and Prescott, 1999). Busse 8 (1994) reported N accumulation in Pinus contorta logs through decomposition, but N content reduced from Decay Class 4 to 5, indicating mineralization after a minimum of 38 years. In another study, N release from Scots pine, Norway Spruce and birch logs averaged 6.1 kg·ha - 1 ·year -1 10 years after disturbance and declined rapidly to 0.13 kg N·ha -1 ·year -1 by age 50 (Krankina et al., 1999). Additionally, these three species lost 45%, 39% and 60% of their N stores, respectively, from Decay Class 1 through 4 (Krankina et al., 1999). The timing of N release from logs remains variable; in some cases N release begins early in the decomposition process and in others it occurs later. What is clear from the literature is that patterns of N storage and release from decaying CWD are difficult to establish due to significant methodological challenges. The long time span of CWD decomposition requires a chronosequence approach be used to estimate N dynamics. Logs are classified in qualitative decay classes and the original volume and mass are rarely known. Interpreting changes in N content is difficult as these changes are often based on an estimate of volume and mass at time of death. Some studies do take these considerations into account (eg. Krankina et al.,1999, and Holub et al., 2001), but long-term study of logs with known original volume, mass, and nutrient concentrations followed through the decomposition process are better positioned to accurately document CWD N dynamics. Laiho and Prescott (1999), their findings reported above, is one such study. While the results from many studies indicate significant variability, both N accumulation and release occurs in coniferous CWD, with N accumulation more likely in CWD with lower initial nutrient concentrations (Laiho and Prescott, 1999). More pertinent, however, is the relative importance of decomposing CWD to overall forest N budgets; is CWD a significant N pool? In Yellowstone National Park, dominated by 9 Pinus contorta, Smithwick et al. (2009) found downed trees to account for 10.3% of total ecosystem N with an additional 2% in standing dead wood. This constitutes a substantial amount of N (11.6 g N·m -2 ), the second largest pool of detrital N after the forest floor (15.3 g N·m -2 ), but these pools are small relative to soil N where 57% of total ecosystem N is stored (Smithwick et al. 2009). However, in a Pinus contorta stand in central Oregon, Busse (1994) concluded that downed boles contributed only 2% of the total N found within mineral soil (to 60 cm), organic horizon and downed bole wood. The absolute amount of N in coarse woody debris is estimated to be 100-244 kg·ha -1 for old-growth forests in coastal Oregon, 200 times less than what is stored in the soil (Harmon et al. 1986). Whether this N is available for plant uptake is of concern; Laiho and Prescott (1999) found that CWD contributed less than 2% of N release in Pinus and Picea stands and 5% in Abies stands. However, this study assumed that N was lost at the same rate as mass, effectively ignoring net accumulation of N in CWD over time. Hart (1999) concluded that net mineralization of N in CWD provides between 4-6% of total N uptake by plants, lower than other internal fluxes of N and input from the atmosphere. It is likely that relative to CWD, soil N is a more abundant and readily available N source in coniferous forest ecosystems and in most cases the amount of N contained within woody debris is small. This, of course, is a contentious statement in the context of biomass harvesting; the function of CWD for forest productivity will be discussed in greater detail. 10 2.3 The role of wood decay fungi in CWD decomposition 2.3.1 Structure and foraging strategies of mycelial networks Higher fungi, mostly within the phylum Basidiomycota, function as primary decomposers of plant detritus in forest ecosystems. Dead organic matter, the energy and nutrient supply for saprotrophic fungi, is distributed heterogeneously and discontinuously in space and time. Fungi have evolved strategies for utilizing these heterogeneous resources, and can be broadly grouped as either unit-restricted or non-unit-restricted with respect to their resource supply (Boddy, 1999). While both groups produce spores to colonize new resources, non-unit restricted fungi extend from a resource with exploratory mycelium and can form rhizomorphs and mycelial cords. Literally meaning ‘root-like,’ rhizomorphs are thick melanized aggregations of fungal hyphae that grow apically. Mycelial cords are similar structures, but develop behind a growing margin of diffuse hyphae (Boddy et al., 2009). Fungi that use spatially discrete resources separated by centimetres or even metres require specialized foraging strategies. Fricker et al. (2008) describe these strategies as, 1) active growth and search for new resources 2) a ‘sit and wait’ strategy, where mycelia wait for resources to be delivered (e.g. branch-fall) followed by colonization and 3) a combination of both these strategies. Active growth of a mycelial network in search of new resources is controlled by both abiotic and biotic factors that continually remodel network architecture. Different species will exhibit varying network patterns, which are further controlled by microclimate, nutrient status of the mycelial network and of the host resource, nutrient status of the unexploited environment, 11 competitive interaction between mycelia of different species, invertebrate grazing, and time, size and age of the network (Boddy et al., 2009). While these factors will influence the foraging pattern of mycelia, colonies will generally form two separate zones: 1) a zone of exploratory hyphae at the growing front, and 2) a zone consisting of high-capacity cords that result in economized transport pathways (Boddy et al., 2009). Over time, mycelial cords connect wood resources, and this results in more open networks (lower mass fractal dimension, Dm) (Fricker et al., 2008). The ‘sit and wait’ strategy involves development of large (m2 to ha2), persistent mycelial networks that can colonize litter fall or dead roots upon delivery (Fricker et al., 2008). Arrival of new resources can change mycelial network architecture through redirection of biomass and changes in nutrient partitioning (Boddy and Jones, 2007). Encounter of a resource larger in comparison to the original resource causes thick cord formation, a decrease in radial extension from the original resource, and eventual autolysis and nutrient recycling of mycelium from the original resource (Boddy et al. 1999, Boddy and Watkinson, 1995). Colonization of new resources can cause system-wide reorganisation; this was demonstrated by Downson et al. (1986) where Hypholoma fasciculare inoculated on a wood block was allowed to grow radially outward toward a ring of uncolonized wood blocks. Upon contact, further outgrowth was halted as the new wood blocks were colonized, and the network behind the blocks was reorganized through regression of exploratory hyphae and aggregation of cords connecting each wood resource. 12 2.3.2 Fungal decomposition of coarse woody debris Coarse woody debris (CWD) serves as the primary energy substrate for many saprotrophic basidiomycetes, and these fungi are the central agents of wood decomposition (Boddy and Watkinson, 1995). Globally, of the 360 Pg C contained in forest plant biomass (Dixon et al., 1994), 10-20% is present as CWD, however; this may vary widely with ecosystem type and disturbance regime (Cornwell et al., 2009). Rates of input are also variable, ranging from 0.12 Mg ha -1 year -1 in a Quercus nigra forest to 30 Mg ha -1 year -1 in a Pseudotsuga-Tsuga forest (Harmon et al., 1986). Additionally, the proportion of total aboveground tree litter inputs that consists of CWD in northern coniferous forests ranges from 3-73% (various sources, as summarized by Laiho and Prescott, 2004). The amount of CWD available in managed stands is generally lower relative to natural stands as harvesting removes boles which are the dominant supply for CWD. Thus, in many forest ecosystems, CWD is a relatively abundant, but variable, carbon source for fungi. Decomposition of CWD is typically classified in three categories: soft, white and brown rot, determined by their relative degradation of wood biopolymers (cellulose, lignin, hemicellulose) (Rayner and Boddy, 1988). These biopolymers differ in their inherent decomposability, but all are relatively recalcitrant when compared to simple carbon compounds. Cellulose, for example, is most easily utilized, but is usually embedded within a lignin matrix. Soft rot fungi decompose cellulose, and hemicellulose, but lignin is not appreciably decomposed and decay is often localized. Brown rot is similar to soft rot, but lignin is only modified and hemicelluloses and cellulose are selectively removed, leaving behind a brown residue consisting of oxidized lignin (Rayner and Boddy 1988). White rot fungi are able to decompose all 13 components of wood and are the most common wood decay fungi found in nature (Boddy and Watkinson, 1995). Brown rot fungi are commonly associated with gymnosperm decomposition, whereas white rot fungi are better able to decompose angiosperm wood (Rayner and Boddy, 1988). Decomposition of naturally produced woody debris starts long before a log or branch reaches the forest floor. Fungal community succession dictates the decay process and is controlled primarily by three determinants: incidence of competitors, stress (tolerance of abiotic extremes, e.g. temperature, pH, water content), and disturbance. These determinants have led to the development of three categories of life-history strategies for fungi; i) competitive (C- selective), ii) stress tolerant (S-selected) and iii) ruderal (R-selected) (Boddy and Heilmann- Clausen, 2008). Generally, a succession of fungi inhabit woody debris through the decay process, usually starting with S- and R-selected fungi attacking a tree wound, followed by replacement through combative interactions primarily involving C-selected fungi. Pioneer fungi attacking living trees usually enter through wounds, are delivered though passive means (e.g. air or water currents, animal vectors etc.), and arrive as spores or vegetative mycelium (Boddy, 2001). At early stages of decay in living trees, heartwood is preferentially attacked despite there being increased amounts of inhibitory allelopathic chemicals and an unfavourable gaseous environment (Blanchette and Biggs, 1992). Functional sapwood in living trees is a low-oxygen environment and is water saturated, making it difficult to colonize. Thus, pioneer heartrot fungi have S and R selected characteristics and are relatively successful at colonization in early decay stages. 14 Studies highlighting changes in fungal community composition through decomposition of woody debris from standing tree to the forest floor are limited. However, changes in fungal community composition through decomposition of felled wood are better understood (e.g. Coates and Rayner 1985 a,b and c, Chapela et al., 1988). Once sapwood cells die, C-selected fungi, largely Basidiomycetes, are more competitive than many S-selected fungi and eventually replace these species through combative interactions (Boddy, 1993). Cord-forming fungi, such as Hypholoma fasciculare, Megacollybia platyphylla, Phallus impudicus, Phanerochaete velutina and Armillaria gallica, establish within six months, replacing earlier colonizers, and become the dominant Basidomycota after 4.5 years of decay in felled Beech logs (Boddy and Heilmann- Clausen, 2008). Late-stage decay community composition is not well understood, but sporocarp surveys indicate that agarics (e.g. Mycena, Pluteus spp.) are increasingly dominant as are ectomycorrhizal species and soil/litter saprotrophs (Heilmann-Clausen, 2001). A recent molecular based study analysing fungal rRNA (metabolically active fungi) from a decaying log survey noted that species richness increased dramatically with decay class and that soft- and white-rot fungi dominate at early stages, followed by white- and brown-rot fungi (depending on the tree species) in moderately decayed logs, and ectomycorrhizal fungi in late stage decay (Rajala et al., 2011). Community development through decomposition responds largely to environmental factors; stress (either alleviation or aggravation), disturbance, and combative interactions (Boddy and Heilmann-Clausen, 2008). These factors determine community development; for example, under a worsening water regime, fungi tolerant to this stress will dominate, whereas stress alleviation allows for C-selected fungi to dominate. Additionally, wood is physically changed through the decomposition process, cellulose, lignin and hemicelluloses are utilized and nutrients 15 become increasingly bound in living fungi, bacteria and fauna. Thus, abiotic or biotic antecedent conditions will also influence subsequent community development over the decay process. 2.3.3 Enzymatic decomposition of wood As previously mentioned, cord-forming fungi, such as Hypholoma fasciculare and Phanerochaete velutina, are the dominant Basidiomycetes present through mid and late-stage decay. These fungi are largely C-selected and use their cord-forming abilities to colonize heterogeneously distributed woody resources by emerging from the forest floor. Hyphae diffuse into fan-shaped mycelia and spread over the wood surface to increase the colonized surface area. Individual hyphae then enter tracheid (wood) cells through pits, which are water and solute conduits, or through direct bore holes, and begin enzymatic degradation (Blanchette 1991). Upon entry, hyphae first encounter the thin, but heavily lignified, ‘S3’ layer of the secondary wood cell wall. The secondary cell wall layers ‘S2’ and ‘S1’ are then sequentially degraded with a battery of enzymes (Carlile et al. 2001). Enzymatic degradation of wood biopolymers is a complex biochemical process, and at least for lignin degradation, is not well understood. This is partly due to the structure of these wood polymers. Cellulose is a highly ordered polymer consisting of glucose units linked by β- 1,4-glycosidic bonds to form D-glucan chains. D-glucan units are crosslinked through hydrogen bonding and van der Waals forces to form a crystalline-like structure – the microfibril (Martinez et al. 2005). Cellulose biodegradation occurs via an enzymatic system involving endoglucanases, exoglucanases and β-glucosidases. Endoglucanases begins the process by disrupting the crystalline structure through hydrolysis of D-glucan units into shorter cellulose chains. Following this, exo-enzymes, collectively termed cellobiohydrolases, hydrolyse β-1,4 bonds to produce 16 cellobiose and glucose, which can then be taken up by hyphae. Cellobiose and other smaller glucose molecules produced by cellobiohydrolases are further broken down by cell wall bound or intracellular β–glucosidases (Baldrian, 2008). Biochemically, cellobiose dehydrogenase, a β– glucosidase, oxidises cellobiose (and other polysaccharides) and reduces a variety of molecules to produce hydrogen peroxide, a constituent in lignin decomposition (Baldrian, 2008). Cellulose dehydrogenase thus enables the combination of lignin and cellulose decomposition. The exact biochemical pathway for lignin degradation is elusive, largely due to the amorphous nature of this compound. Lignin confers mechanical resistance in woody plants and is found in the secondary cell wall and the middle lamella of tracheid cells (Blanchette, 1991). Lignin, with hemicellulose, forms an amorphous matrix where cellulose microfibrils are embedded, protecting cells against biodegradation (Martinez et al., 2005). Structurally, lignin is a branched polymer of three aromatic alcohols – coumaryl alcohol and its methoxyolated derivatives coniferyl and sinapyl alcohol, joined together by carbon-carbon and ether bonds (Baldrian, 2008). These aromatic alcohols are precursors to lignin biosynthesis and form three types of lignin units – p-hydroxyphenyl-, guaiacyl- and syringyl units. Gymnosperms have the highest lignin content which consists primarily of guaiacyl lignin units, whereas angiosperms contain higher amounts of syringyl units. Mineralization of lignin requires two key processes – breakdown of the lignin polymer and ring cleavage of aromatic nuclei (Tuor et al., 1995). This primarily oxidative process requires oxidases, peroxidases and hydrogen peroxide producing enzymes. White rot fungi primarily use, but are not limited to, laccase, lignin peroxidase and manganese oxidase in their enzyme arsenal. Comprising less than 10% of the total lignin polymer, phenolic lignin units are oxidized with laccase using molecular oxygen (Baldrian, 2008; Marteniz et al., 2005). Lignin 17 peroxidase degrades non-phenolic lignin units (up to 90% of the polymer) and is mediated with hydrogen-peroxide; an electron is removed from aromatic nuclei, which produces radical species that further undergo non-enzymatic radical reactions, resulting in polymer cleavages (Marteniz et al., 200;, Blanchette, 1991). Manganese peroxidase oxidizes Mn 2+ , producing Mn 3+ , which further acts as a diffusible oxidizer on several substrates within the lignin complex (Baldrian, 2008; Martinez et al., 2005). While these enzymes are important for lignin biodegradation, they are too large penetrate intact lignin. It has been suggested that smaller initiating molecules, such as hydrogen peroxide, veratryl alcohol, oxylate, and manganese, are required to enlarge pores in wood cells before enzymatic degradation begins (Evans et al., 1994; Perez et al., 2002). As outlined above, acquisition of energy (carbon) from wood biopolymers during decomposition by saprotrophic fungi requires the production and release of extracellular enzymes. Woody debris provides ample carbon supplies for energy and growth, but lacks nitrogen, having a C:N ratio in the range of 200-1200:1 (Cornwall et al., 2009). Production of extracellular enzymes by fungi is nutrient intensive, but the growth substrate for saprotrophic fungi (wood) is relatively nutrient deficient (Allison and Vitousek, 2005). Thus, saprotrophic fungi must augment the nutrient demand of their enzymes by exploring the forest floor for more readily available nutrients required for enzyme synthesis. 2.3.4 Nutrient uptake, translocation, and release Nitrogen and phosphorous (P) are the most limiting nutrients for saprotrophic fungi. Nitrogen is an integral element for protein synthesis, and thus is critical for the production and function of wood decay enzymes, while P, in the form of ATP, is commonly given the moniker of ‘energy currency’ of the cell. Fungi are able to utilize most forms of nitrogen and can readily 18 immobilize inorganic N (NH4 + , NO3 - ) and organic N (Carlile et al., 2001). Fungal metabolism of N has largely been studied in three species; Aspergillus nidulans, Neurospora crassa, and Saccharomyces cerevisiae. These fungi are able to use a diversity of N sources (ammonium, nitrate, amino acids, acetamide, purines and proteins), but NH4+, glutamate and glutamine are favoured (Jennings, 1988). Use of other less desired N sources (nitrate, acetamide, purines etc.) is repressed by nitrogen metabolite repression and repression must be lifted before enzymes catabolising less desired N sources can be activated (Caddick, 2002). Preferential use of N sources is less well studied in saprotrophic organisms. New lines of evidence are increasingly pointing to organic N as the primary N currency in N-limited environments. In these environments, which include boreal ecosystems, saprotrophic fungi as well as other soil microbes likely utilize organic N and control its availability to other organisms through depolymerisation reactions (Schimel and Bennett, 2004). In most soils, N and P are held within complex polyphenolic organic molecules (Yavitt and Fahey, 1986), and require specialized enzymes (polyphenoloxidases, peroxidases) for degradation and nutrient release (Rayner and Boddy, 1988). Saprotrophic fungi, with their enzymatic capabilities, are well suited to foraging for organically bound nutrients from the forest floor. With their carbon and nutrient sources often being spatially separate, saprotrophic fungi are well known to translocate carbohydrates and nutrients throughout their mycelium (Olsson, 1999). Translocation allows fungi to link areas of supply, such as CWD, to areas of demand, such as growth over nutrient poor substrate to new areas of nutrient supply (Watkinson et al., 2006). In experimental microcosms, 32 P has been shown to move throughout colonized resources to areas of demand (Hughes and Boddy, 1994), and is known to move bi-directionally (Lindahl 19 et al., 2001). Nitrogen exhibits similar translocation patterns; 15 N-labelled fertilizer applied to the carbon source of S. lacrymans was transferred to a newly encountered straw bait (Watkinson 1984). More recently, organic N translocation has been shown in a wood decay fungus in real time using a radioactive non-metabolized amino acid analogue ([ 14C]α-Aminoisobutyric acid, [ 14 C]AIB) (Tlalka et al., 2008). This study is the first to-date to show network reallocation of nutrient resources in response to discovery of freshly colonized wood, but the behaviour of non- metabolized amino acids is a tenuous comparison to amino acids found in the forest floor. Nutrient uptake from foraging fronts is also opportunistic; 32 P has been shown to be translocated back to the inoculum even if an equivalent amount of P was added to the inoculum in compensation (Wells et al. 1990). This is an important result, as it indicates that some saprotrophic fungi hoard nutrients above their nutritional requirements. The inherent nutrient ‘greediness’ of saprotrophic fungi has important consequences for nutrient cycling (Boddy and Watkinson, 1995). The conventional model for N cycling, based on agricultural systems, indicates that N is mineralized in soil when the C/N ratio of a substrate falls below a critical level (Swift et al., 1979). In saprotrophic fungi, nutrients are thought to be released into the soil when the C/nutrient ratio of the wood is the same as that of the mycelium; however, Watkinson et al. (1981) have shown that mycelium of Serpula lacrymans extending away from a woody resource of 0.4-0.7 mg N g -1 dry weight contained a N concentration of 16- 24 mg N g -1 . Thus, the model proposed by Swift et al. (1979) may be too simplified to apply to forest ecosystems (Boddy and Watkinson, 1995). Lindahl et al. (2002) have proposed a model for boreal (N-limited) ecosystems whereby N from plant litter is not significantly mineralized, but rather it is retained in fungal mycelia and translocated to N-poor woody substrates. Evidence of reduced mineralization of N due to fungal translocation is reported by Boberg et al. (2010). 20 These authors demonstrate that N mineralisation was reduced when a litter decomposing fungus has access to both C-rich litter and N-rich glycine medium, whereas N mineralisation was significantly higher in microcosms that only contained the N-rich medium. Thus, the fungus was able to overcome C-deficiency in the N-rich medium by translocating carbon from the litter to facilitate growth and nutrient uptake. The nutrient content of fungal biomass can represent a significant nutrient pool, containing 15-20% of N and up to 18% of organic P in the F/H horizon of boreal forests (Bååth and Soderstrom, 1979). If these nutrients are not conventionally released as proposed by Lindahl et al. (2002), other mechanisms must be responsible for their release. Boddy and Watkinson (1995) propose that nutrients are released from fungi during interaction with other mycelia or bacteria, through invertebrate grazing, through autolysis associated with biomass redistribution, or ultimately as mycelia die and decompose. Combative interactions between mycelia of different species have been shown to result in 32 P transfer between species at the interaction zone and loss of 32 P to the soil (Woodward and Boddy, 2008). Anderson et al. (1983) found increased ammonium released from leaf litter colonized by Basidiomycetes in the presence of soil fauna, however there are limited studies investigating nutrient mineralization via invertebrate grazing. 2.4 Effects of woody debris removal on future forest productivity Since the 1970s, biomass removal and subsequent nutrient export related to harvesting practices has been of concern for forest ecologists. A pulse of research investigating the effects of logging residue removal on several measures of future site productivity occurred in the 1970s as a result of advances in mechanized harvesting practices and a push for greater use of 21 unmerchantable wood (Thiffault et al. 2010). Currently, demand from the bioenergy sector is driving a renewed pulse of research investigating the effects of biomass harvesting on future forest productivity. A dominant hypothesis among forest ecologists is that whole tree harvesting (WTH) or woody debris removal will result in increased N losses associated with reduced immobilization capacity due to removal of woody logging residues. 2.4.1 Nitrogen leaching and woody debris removal Immobilization of N in low-nutrient litter and subsequent reduction in plant-available N is a well-established concept in agricultural systems. Through decomposition, microbes eventually exhaust their carbon source and nutrients are released back to the environment. This concept has carried over to forest N-cycling research; removal of woody debris will result in increased leaching losses of N without the nutrient ‘sponge’ capacity of woody debris. However, a competing theory suggests that without organic matter inputs, microbial populations are C- limited, reducing competition for N and resulting in increased short-term plant-available N (Strahm et al., 2005). Due to a decline in plant uptake and increased net mineralisation, N leaching increases after clearcutting, usually in the form of nitrate (Prescott, 2002), but the impacts of logging residue removal on leaching losses remain variable (Slesak et al., 2009); residue retention can lead to increased leaching losses of N (Strahm et al., 2005), decreased losses (Carlyle et al. 1998), or have no detectable effects (Mann et al. 1988). Results are often site-specific; Carlyle et al. (1998) found a 13% reduction in leaching loss of total N after residue retention on a site with sandy soil and low nutrient status. Strahm et al. (2005) found the 22 opposite result on a fertile site with high N status; 75 kg N·ha -1 ·year -1 was lost from residue retention plots, compared to 29 kg N·ha -1 ·year -1 lost from whole-tree-harvested sites. Increased leaching loss of N associated with residue retention is not an unexpected result. In one study, whole-tree-harvesting represented a 234% increase in N export off site via relative to stem-only harvesting (Fahey et al., 1991). Strahm et al. (2005) estimated that 433 kg N·ha -1 was retained on residue retention plots, and leaching losses of N were proportional to the forest floor and coarse woody debris N capital. Thus, sites with residue retained are inherently more N rich, and may leach more N, even if the amount of N retained is negligible relative to the much larger soil N pool. It is important to note the distinction between logging residues and coarse woody debris. The former consists of slash produced from tops and delimbing and includes both coarse woody debris (tops and large branches), as well as fine woody debris and green needles. Fahey et al. (1991) report that foliage in Sitka spruce logging residue contained of 56% of the total residue N. Reports of either N loss or retention after residue removal do not necessarily imply that woody debris is responsible for the effect; the foliage may play a much larger role (Laiho and Prescott, 2004). 2.4.2 Woody debris removal and site fertility Changes in soil C and N content are often cited as master variables related to site fertility and are widely used to characterize changes in site fertility related to harvesting practices. Most research investigating effects of residue removal on soil C and N content compare whole-tree- harvesting (residue removal, WTH) to conventional harvesting (residue retention, CH). Reported 23 results are highly variable, but in a meta- analysis, Johnson and Curtis (2001) demonstrate that WTH reduces soil C and N by 6%, while CH harvesting increases these variables by 18%. These results contrast with those of Olsson et al. (1996), where no difference in soil C and N content between WTH and CH treatments were detected 15-16 years after clearfelling. In the latter study, site specific treatment differences were detected; WTH and woody debris removal (needles left on site) treatments at a northern spruce site in Sweden caused greater reductions of N pools in humus compared to the CH treatment. Other studies report that residue removal has no effect on soil C or N content in the short term (e.g. Carter et al. 2002, Sanchez et al. 2006, Powers et al. 2005). In a longer-term study (30 years after clearfelling), Wall and Hytönen (2011) report that WTH (with needles retained) in a Norway spruce stand did not significantly reduce soil C and N content. This result is supported by Vaneguelova et al. (2010), who report unchanged soil C and N content with residue retention. The latter study demonstrated that CH plots had lower soil C and N concentrations, attributed to increased mineralisation rate in these plots. While long-term impacts of residue removal on soil C and N content cannot be ruled out, it is likely that early and mid-rotation effects of residue removal are largely site-specific. Measures of site productivity (tree height, stand basal area, DBH) can offer direct evidence of the impact of residue removal on site productivity. Relative to conventional harvesting, whole-tree-harvesting at the Beddgelert Forest experiment 23 years into a Sitka spruce second rotation resulted in a 10.3% reduction in DBH, and an 8.2% reduction in mean tree height (Walmsley et al., 2009). Stand basal area of the WTH treatment in this study was also reduced by 6.8 m 2 ·ha -1 . Similarly, tree height was reduced in WTH plots relative to CH plots by 1 m in a 20-year-old second-growth Sitka spruce stand in the Kielder forest, UK (Vanguelova et al., 2010). However, no difference in tree height was observed in the same stand after 28 years of 24 growth. Tree DBH in this study was higher in CH plots relative to WTH after 28 years of growth (Vanguelova et al., 2010). Other studies, however, report no significant effects of residue removal on stand growth characteristics, but a tendency is often observed toward a reduction of growth through WTH (Fleming et al., 2006; Powers et al., 2005; Sanchez et al., 2006; Wall and Hytönen, 2011). It is clear from a number of studies that dramatic reductions in site productivity associated with residue removal have not been observed in the short-term. Whether long-term effects of residue removal on productivity will occur through a rotation and into future rotations remains an important research directive. However, soil properties and regenerating species, are good indicators of site sensitivity to residue removal. Thiffault et al. (2010) recommend that management practices consider site characteristics; e.g. high-risk such as those with coarse- textured soils, and low nutrient capital, should not be subjected to residue removal. 2.5 Summary Effective management of woody debris in British Columbia demands application of robust scientific data to inform policy decisions. Many forest ecologists are concerned that removal of woody debris and logging residue from managed forests will lead to long-term declines in forest productivity. Results showing increases in absolute N content in woody debris through time have fuelled this debate; will removal of this material contribute to increased leaching loss of N as a result of a loss of the functional capacity of woody debris to act as an N ‘sponge’? Most short term studies (<30 years) have not shown dramatic negative effects on soil nutrients and stand characteristics. Variation and methodological challenges prevent generalization of the effects of residue removal, but current knowledge underscores the prudence 25 of conserving woody debris and logging residue on sensitive sites, such as those with low nutrient capital and coarse-textured soils. Long-term data concerning the effects of woody debris removal may eventually inform forest managers on a more general basis, but in the meanwhile, site specific management could be carried out. Lacking from much of the research concerning nutrient dynamics of woody debris is a thorough understanding of mechanisms responsible for observed increases in absolute N content. Many authors hypothesize that nutrient transfer by cord-forming wood decay fungi is the central mechanism responsible for this observation, but evidence, especially for N transfer, is lacking. Wood decay fungi are known to translocate phosphorus in a supply-demand fashion from the forest floor (supply) to woody debris (demand), but similar N dynamics have yet to be shown. Woody debris constitutes a significant C source, but in order to access it, fungi must invest N for enzymatic decomposition, which is likely imported from the surrounding forest floor and soil. Many aspects of the ecology of fungal wood decomposition support this hypothesis; wood decay fungi are known to conserve nutrients within their mycelia above their nutritional requirements, they secrete a battery of N-rich enzymes during wood decomposition and cord-forming fungi capable of nutrient translocation dominate the fungal community in mid and late-stage wood decomposition. Demonstration of N-transfer from soil to woody debris via mycelial cords of wood decay fungi would provide direct evidence of the fungal mechanism involved in N accumulation in woody debris. This information is increasingly relevant in the context of biomass harvesting for bioenergy production in British Columbia; does harvest of woody debris eliminate the ability of this material to retain significant amounts of N on site via translocation of N by cord-forming wood decay fungi? 26 3 HYPOTHESES This thesis will examine the role that wood decay fungi play in facilitating translocation of inorganic and organic nitrogen from the forest floor to woody debris. Specifically, the following hypotheses will be tested: H1: Several studies suggest that wood decay fungi are capable of mobilising N from areas of supply (e.g. the forest floor) to areas of demand (a woody substrate) via their mycelial cords. Thus, I hypothesize that inorganic ( 15 NH4 + , 15 NO3 - ) and organic ( 15 N-glycine) nitrogen will be transferred via mycelial cords of the wood decay fungus Hypholoma fasciculare from soil to inoculated wood blocks. H2: Laboratory studies of fungal nitrogen metabolism suggest that fungi can use diverse sources of nitrogen, however, ammonium, glutamate and glutamine tend to be favoured. Thus, I hypothesize that the mycelial cords of Hypholoma fasciculare growing from an inoculated wood block will preferentially use 15 NH4 + over 15 NO3 - and 15 N-glycine. H3: The absolute N content of CWD is observed in several studies to increase over time and it is often suggested that wood decay fungi are responsible for this increase. However, direct evidence supporting this hypothesis is lacking. With my experimental design, I can directly test the hypothesis that the absolute N content of wood blocks inoculated with Hypholoma fasciculare will become higher relative to uninoculated blocks (H3). In addition to the above hypotheses, I explored the following research questions: Q1: Are the mycelial cords of Hypholoma fasciculare capable of mobilising 15 N from labelled Douglas-fir needle litter and translocating it to inoculated wood blocks? 27 Q2: What is the temporal pattern of 15 N transferred from soil to inoculated wood blocks by the mycelial cords of Hypholoma fasciculare? Similarly, what is the temporal pattern for absolute N content? 28 4 METHODS 4.1 Overview The overall approach I used to test my hypotheses involved a laboratory study where the wood decay fungus Hypholoma fasciculare was inoculated onto small Douglas-fir (Pseudotsuga menziesii) blocks and allowed to grow across the length of a soil-filled microcosm. After an incubation period, isotopically labelled nitrogen ( 15 N) was applied to the soil near the mycelial growing front of the fungus. With this method, I compared the movement of four forms of 15 N ( 15 NO3 - , 15 NH4 + , 15 N-glycine, and 15 N-labelled Douglas-fir litter) from the soil into uninoculated blocks and inoculated blocks. In order to establish a rate of 15 N transfer, I used a sequence of microcosms to test the movement of the four N forms after 6, 18 and 30 days of tracer application. I then compared my data to other previously published results to estimate the significance of fungal transport of N into wood blocks relative to other N pools in forest ecosystems. 4.2 Sample collection Hypholoma fasciculare (family: Strophariaceae) is a widely distributed wood decay fungus that appears in dense clusters on decaying wood and is abundant in the fall and winter (Arora, 1986). Sporocarps were collected from the Alex Fraser Research Forest (Williams Lake, British Columbia, Canada), the Malcolm Knapp Research Forest (Maple Ridge, British Columbia, Canada), and Pacific Spirit Park (Vancouver, British Columbia, Canada) during the late spring (Alex Fraser) and fall of 2010. Sporocarps were collected from decomposing Douglas-fir logs. A mushroom identification field guide (Arora, 1986) assisted with positive visual identification 29 though distinct morphological characteristics: dense clustering, cap size (2-5 cm), shape (broadly conical at centre, then becoming convex) and colour (orange-tan at centre becoming bright sulphur-yellow elsewhere), stalk size (5-12 cm) and colour (yellow but developing rust- or brown-coloured stains at base), gill colour (greenish-yellow, darkening to black with age due to spores), and spore print (purple-brown to deep purple-grey). Hypholoma fasciculare was distinguished from other similar species (Hypholoma capnoides and Pholiota spp.) by its growth habit, yellow to greenish-yellow gills, purple-brown spore print, and its yellow flesh. Within 1-2 hours of collection, samples were cultured onto Potato Dextrose Agar plates using sterile technique under a laminar flow hood. Cultures were then stored at room temperature (24°C) in the dark and re-cultured as necessary. Soil was collected from one area in a Douglas-fir second growth stand at the Malcolm Knapp Research Forest in November 2010. The Research Forest is in the Coastal Western Hemlock (CWH) vm1 (very wet maritime) subzone (Pojar et al. 1986). The soils are Orthic and Sombric Humo-Ferric Podzols of gravelly loamy sand over a glaciofluvial blanket over glacial marine deposits (Carter and Lowe 1986). The forest floor was removed and A-horizon material was collected. Soil was transported back to the laboratory in a cooler and was then sieved to 2 mm and refrigerated at 4°C until required. Soil water content was also determined at this time. Dissolved inorganic (NH4 + , NO3 - ), dissolved organic and amino acid N (mg·kg -1 ) was determined (Table 4-1) at the Analytical Laboratory of the B.C. Ministry of Forests, Lands, and Natural Resource Operations (Victoria, BC). 30 Table 4-1 Inorganic N (NO3 - , NH4 + ), total soluble N, soluble organic N and amino acid N concentrations (mg·kg -1 ) from collected soil. Values in parentheses represent standard error of the mean. Nitrogen form Concentration (mg·kg -1 ) NO3 - 1.51 (0.30) NH4 + 15.81 (0.16) Total soluble N 30.9 (0.28) Soluble organic N 13.6 (0.22) Amino acid N 8.88 (0.97) 4.3 Preparation and inoculation of wood blocks Douglas-fir wood was chosen as the substrate for inoculation as this is a dominant species in coastal and interior forests of British Columbia and I have observed Hypholoma fasciculare growing on this species. A freshly-felled Douglas-fir log (Pseudotsuga menziesii) was collected from the Malcolm Knapp Research Forest and was cut into 1 x 2 x 2 cm blocks. The blocks were frozen at -18°C until required. Only sapwood blocks were used for inoculation; heartwood and sapwood blocks were differentiated visually and by density separation using water. Before use, blocks were first thawed overnight in distilled water, and then autoclaved twice for 30 minutes (wet cycle) at 121°C over a 24-hour interval. Blocks were added to the growing front of 2-week-old cultures of Hypholoma fasciculare growing on PDA in standard petri dishes. These dishes were then incubated in the dark at 24°C for an additional 15 days until the blocks were covered with mycelium. 4.4 Microcosm development Preliminary microcosms were used to determine ideal growing conditions for Hypholoma fasciculare. I constructed fifty microcosms using clear Plexiglas (3 mm thick), with inside 31 dimensions (length x width x height) 17 x 3.5 x 2.5 cm. A 4.5 x 3.5 x 2.1 cm labelling well was created by inserting a piece of Plexiglas 4.5 cm from one end of the microcosm (Figure 4-1). Seams were with sealed acrylic glue and waterproof silicon sealant. Trays were tested for leaks by filling each tray with water, inspecting for leaks, and re-sealing if necessary. Figure 4-1 Preliminary microcosm design, a) Plan view, b) Side view c) Microcosm at 27 days of incubation. The dotted line represents the approximate 15 N addition point. Each tray was filled with approximately 55 g (wet mass) of sieved soil, which was previously frozen for 24 hours to reduce soil fauna and hence mycelial grazing. Soil was slightly compressed to leave approximately 1 mm of space between the soil surface and the lip of the tray. An inoculated wood block was removed from a petri dish, adhering agar was scraped off, 17 cm 3.5 cm 4.5 cm 2.5 cm 2.1 cm A B Soil Wood block C 32 and its mass was recorded. The block was then placed on the soil surface at the end opposite the labelling well. The microcosms were placed in sealed polyethylene bags that allowed airflow through a 1 cm x 1cm opening covered with glass fibre filter paper (Whatman GF/A 1.6 μm). The total mass of each microcosm (tray + soil + wood block) was recorded and water loss was compensated gravimetrically on a weekly basis using tap water and a spray bottle. Additionally, the inside of the bag was sprayed weekly with tap water to maintain high humidity. Microcosms were then incubated in the dark at 17°C until the growing front of all microcosms had progressed to the labeling well (70 days). After incubation of the preliminary microcosms was complete, it was evident that several modifications were required. An additional 60 microcosms were constructed and maintained in the same manner as the preliminary microcosms, but with the following differences (Figure 4-2). It was noted that mycelia grew preferentially on the bottom of the microcosm, not across the soil surface as was expected. As such, mycelia were unable to grow over the labelling well partition and for this reason the partition was not added in the new microcosm design. To encourage mycelia to explore for additional resources, the mass of soil added was reduced to 14.5 g (wet mass) and the size of the microcosms was reduced to 11.5 x 3.5 x 1 cm (length x width x height). Additionally, half of the inoculated wood block was buried beneath the soil surface to ensure greater surface area contact between mycelia and soil. It was expected that mycelia would grow to the end of the microcosm, but slower than expected growth and evidence of maturing cord systems suggested that mycelial progression of at least 5 cm from the wood block was sufficient (67 days). 33 Figure 4-2 Final microcosm design, a) Plan view, b) Side view. The dotted line represents the approximate 15 N addition point. 4.5 Preliminary experiment: fungal biomass measurement and 15N movement through soil Poor fungal growth in the initial microcosms allowed only 9 of 50 microcosms to be used in a preliminary experiment; thus, only one of three N-addition treatments was employed. After incubation, microcosms were randomly divided into three sampling times (Day 3, 15 and 30) and one N-addition treatment ( 15 NH4 + ). Three sampling times were chosen to identify timing of maximum N uptake. To isolate the fungus as the factor responsible for N transfer, control microcosms were prepared in the same manner as inoculated microcosms, but wood blocks were not inoculated. This design resulted in 18 microcosms (2 treatments, inoculated and uninoculated, x 3 sampling times x 3 replicates). To enable tracing of fungal uptake of N, a 15 N-enriched solution was prepared by dissolving powdered 15 NH4Cl (Cambridge Isotope Labs, USA) in distilled water. In this solution, the labelled 15 NH4Cl (99 atom %) was combined in equal proportion with unlabelled KNO3 and 11.5 cm 3.5 cm 1 cm A B Soil Wood block 34 glycine. The total N concentration of the final solution was 45 μg N·mL-1. Each microcosm received 2.5 mL of solution, added 2 cm from the visible growing front with a 5-mL syringe. After incubating for their respective treatment time (3, 15 or 30 days), each microcosm was harvested. First, the wood block was removed, adhering soil was brushed off, and the block mass was recorded. Wood blocks were then oven dried at 105 ° C for 48 hours and the state of decay of each wood block was measured using relative density (g·cm -3 , oven dried weight/fresh volume), and the water content was determined. Soil from each microcosm was divided into four 3 cm portions: under, proximal, middle, and distal (Figure 4-3). Tools were cleaned using an Alconox- water solution and 95% ethanol between collection of each soil portion. Soil was then dried using a freeze dryer (Labconco). This experiment was conducted largely to avoid any unforeseen methodological challenges in future experiments. This experiment was designed to provide evidence that fungal mycelia were responsible for transporting inorganic 15 N from soil to an inoculated wood block (Hypothesis 1). The movement of 15 N through the soil was compared between uninoculated and inoculated microcosms by analyzing a subsample from each soil portion (under, proximal, middle and distal) for 15 N content. Additionally, the relative abundance of the fungal biomarker Under Proximal Middle Distal Figure 4-3 Preliminary microcosm showing portion divisions 35 18:2ω6,9, a proxy for fungal biomass, was determined for each soil portion in uninoculated and inoculated microcosms using PLFA analysis. This was to test for any correlation between fungal biomass and the mass of 15 N transported through the soil. 4.6 Main experiment Improved growing conditions in redesigned microcosms produced enough viable microcosms to allow for 3 treatments ( 15 NH4 + , 15 NO3 - 15 N-Glycine) and 3 times (Days 6, 18 and 30). Four replicates were used for each treatment-time combination. Microcosms with uninoculated blocks were prepared in the same manner as in the preliminary experiment. This design resulted in 72 microcosms (3 treatments x 3 times x 4 replicates, and a fully-crossed and replicated uninoculated set). Three 15 N-enriched solutions were prepared by dissolving powdered K 15 NO3 , 15 NH4Cl, and 15 N-Glycine (Cambridge Isotope Labs, USA) in distilled water. In each solution, one labelled N form was combined in equal proportions with the remaining two unlabelled N forms, i.e. 15 NH4Cl (99 atom%) was combined with unlabeled KNO3 and glycine, K 15 NO3 (99 atom%) was combined with NH4Cl and glycine, and 15 N-Glycine (98 atom %) was combined with KNO3 and NH4Cl. The N concentration of each solution was 72 μg N·mL -1, with one third (24 μg) of the added N as 15 N. The estimated N concentration (KCl extractable NO3 - , NH4 + , and organic N) from this soil was 30.9±0.28 μg N·g soil-1 (see Table 3-1). Each microcosm received 1 mL of labelled solution that was added 2 cm from the growing front with a 5-mL syringe. The addition point used for control microcosms was 2 cm from an average of the maximum visible mycelial extent for each treatment (5.1 cm, 5 cm and 5.5 cm from the wood block for 15 NH4 + , 15 NO3 - , 15 N- Glycine treatments, respectively). 36 After incubating for 6, 18 or 30 days, microcosms were harvested using the same procedure as in the preliminary experiment. Soil from each microcosm was divided into three 4-cm portions: ‘under,’ ‘proximal,’ and ‘distal.’ Tools were cleaned using an Alconox-water solution and 95% ethanol between collection of each soil portion. Soil was then dried using a freeze-dryer (Labconco, USA). Soil portions and wood blocks from all replicates were subsampled and analysed for δ15N (‰), %N and %C. Relative density, water content, absolute N content, mass loss (%), and the C/N ratio were determined for each wood block. The amount of 15 N derived from the tracer that was found in each wood block (assimilated 15 N) was determined using Equation 1 (see section 4.11, Calculations). Comparison of assimilated 15 N results from each treatment ( 15 NH4 + , 15 NO3 - , 15 N-Glycine) were used to test Hypothesis 1 and 2. Comparison of absolute N content and relative density of wood blocks was used to test Hypothesis 3. Comparison of assimilated 15 N, absolute N, relative density, C/N ratio, and mass loss results were used to provide evidence to address Question 2. 4.7 Douglas-fir litter experiment A total of 14 microcosms prepared in the same manner as in the main experiment were used to investigate the capacity of fungal mycelia to mobilize 15 N from labelled Douglas-fir needle litter. Ground 15 N-labelled Douglas-fir litter was obtained from the University of Alberta in Edmonton, AB, which was originally collected as part of a 15 N-fertilizer (urea) application study carried out near Nanaimo, British Columbia (see Nason, 1989). The litter had a δ15N (‰) value of 2437 ± 119 (S.E.) and a %N content of 0.78 ± 0.02 (S.E.) (analysis was carried out at the University of British Columbia Forest Sciences Stable Isotope Facility). Each microcosm 37 received approximately 0.2 g of labelled litter added 1 cm from the growing front. The same amount of litter was added to uninoculated control microcosms (n=4) 1 cm from the average maximum mycelia extent of inoculated microcosms (3.6 cm). All microcosms were incubated for a total of 67 days before the addition of 15 N labelled needle litter and incubated afterward for an additional 91 days to allow for decomposition and mineralization of 15 N by Hypholoma fasciculare and other soil microorganisms. Soil portions and wood blocks from all replicates were subsampled and analysed for δ15N (‰) and %N. Relative density, water content, and absolute N content was determined for each wood block. The amount of 15 N derived from the tracer that was found in each wood block (assimilated 15 N) was determined using Equation 1 (see section 4.11 Calculations). Comparison of results from this experiment was used to provide evidence to address Question 1. 4.8 Experiment: transfer of 15NO3 - to wood blocks Some results indicated that the water content of wood blocks might have affected 15 NO3 - transfer to uninoculated wood blocks. To further investigate this effect, a simple experiment was designed by replicating control microcosms and varying the water content of wood blocks. Uninoculated wood blocks were prepared in the same manner as in previous experiments, but water content was varied by submerging blocks in distilled water for 24 hours (high water content), by air drying at room temperature (24°C) for 48 hours (low water content), or by using the ambient water content after autoclaving (medium water content). A total of 9 microcosms were constructed (3 treatments x 3 replicates per treatment). Microcosms were prepared and maintained in the same manner as previous experiments; 1 mL of the 15 NO3 - solution was added to the soil 5 cm from the wood block and microcosms were incubated in sealed polyethylene 38 bags in the dark at 17°C for 6 days. After this time, wood blocks were collected and analysed for δ15N (‰) and %N, relative density, water content, absolute N content, and assimilated 15N. 4.9 Elemental and isotope analysis To ensure efficient combustion, wood samples were ground into a fine powder using a planetary ball mill (Fritsch, Germany). Soil was homogenized by manually grinding using a ceramic mortar and pestle. To prevent isotope contamination, tools and grinding vessels were thoroughly cleaned between samples using Alconox and 95% ethanol. Elemental (C, N) and isotopic analysis ( 15 N) of soil was done at the UBC Department of Forest Sciences Stable Isotope Facility. Samples (9-10 mg) were weighed into tin capsules and analyzed using an Elementar (Germany) Vario EL Cube elemental analyzer (EA) interfaced via continuous flow to an Isoprime (Germany) isotope ratio mass spectrometer (IRMS). The EA was operated in N mode such that a CO2 scrubber (sodium hydroxide) was employed to remove excess CO2 before N2 passed through to the thermoconductivity detector. Moisture in the samples was removed by two water traps (phosphorus pentoxide). For elemental concentrations, results are reported as wt. %N. Repeated analyses of the marine sediment standard MESS-3 and PACS-2 (Natural Research Council of Canada) yielded %N values of 0.21 ± 0.02 and 0.32 ± 0.01 (1-sigma), respectively. For isotope determinations, results are reported in the standard δ- notation relative to air (δ15N = 0‰). Measured values are corrected using a 2-point calibration (Coplen et al. 2006) for USGS 40 and USGS 41 (L-glutamic acid). Repeated analysis of a quality control standard (IAEA N2, ammonium sulfate) yielded values of +19.84 ± 0.31 ‰ vs. air. 39 Elemental (C, N) and isotopic analysis ( 15 N) of wood samples was completed at the University of Michigan Terrestrial Ecology Stable Isotope Laboratory. Samples (16-18 mg) were weighed into tin capsules and analyzed using a Carlo Erba NC2500 EA (Italy) interfaced via a Conflo II (Thermo Finnagin, USA) to a Thermo Detla Plus continuous flow IRMS (Thermo Finnagin, USA). To ensure complete combustion of high C%, low N% wood samples, excess oxygen was introduced to the combustion stream using a 25 mL oxygen loop. Standards used for %N calculation included acetanilide, ammonium sulfate (IAEA-N1, IAEA-N2 and USGS 25), and potassium nitrate (IAEA-N3). For isotope determinations IAEA-N1, IAEA-N2, IAEA-N3 and USGS 25 were used as standards in a 4-point calibration. Additionally, uncertified lab reference materials (Rhinelander stem 804, NBS tomato leaves, LECO corn flour) were run for comparison purposes, but were not used in the calibration. Results from all standards used in calibrations for wood samples are reported in Table 4-2. Table 4-2 Yielded %N and δ15N values for standards used in wood sample calibrations. Values in parentheses represent standard error of the mean Standard Yielded %N Yielded δ15N Acetanilide 10.3 (0.16) n/a Ammonium sulfate (IAEA N1) 21.8 (1.4) 0.1 (0.6) Ammonium sulfate (IAEA N2) 21.7 (0.9) 20.4 (0.3) Ammonium sulfate (USGS 25) 21.1 (0.9) -30.2 (0.3) Potassium nitrate (IAEA N3) 14.3 (1.1) 4.4 (0.5) Rhinelander stem 804 0.47 (0.01) 14.1 (1.1) NBS tomato leaves 5.05 (0.10) 3.6 (0.4) LECO corn flour 1.09 (0.09) 0.8 (0.4) 4.10 Phospholipid Fatty Acid (PLFA) Analysis Fungal biomass in soil from the preliminary experiment was approximated using Phospholipid Fatty Acid (PLFA) analysis, as described by Frostegård and Bååth (1996). Briefly, 40 PLFAs are cell membrane components used in microbial ecology as chemotaxonomic biomarkers to identify groups of bacteria and fungi. Additionally, this method can be used to estimate total fungal biomass using a fungi-specific PLFA and a conversion factor. The PLFA 18:2ω6,9 has been suggested as a biomarker for total fungi as it is common in fungi but absent or rare in bacteria (Klamer and Bååth, 2004). This PLFA does exist in the cell membranes of other eukaryotes (such as plants), but high correlation to the fungal biomarker ergosterol suggests this PLFA is a good fungal indicator (Kaiser et al. 2010). Soil used in these experiments was sieved to 2 mm, removing most root fractions, thus there is likely little contribution of plant-based 18:2ω6,9 PLFA. The nomenclature used to describe PLFAs is found in Tunlid and White (1992). The procedure was adapted from Dewi, 2009. Briefly, phospholipid fatty acids were extracted from a 1-g subsample of sieved and freeze-dried mineral soil with a mixture of chloroform, methanol, and citrate buffer (1:2:0.8, by volume). The extract was split into two phases with equal portions of chloroform and citrate buffer, and the lipid containing phase was recovered and dried down under a constant stream of N2. The dried-down lipids were then dissolved in chloroform and fractionated through a silica-based solid phase extraction cartridge (SampliQ Silica, Agilent Technologies, USA) with subsequent additions of chloroform, acetone and methanol. An internal standard (methyl nonadecanoate, Sigma N5377, USA) was added to the methanol fraction and samples were again dried down using N2. Fatty acid methyl esters (FAMEs) were obtained from each sample by subjecting the phospholipids to a mild alkaline methanolysis, and were then analysed with a gas chromatograph (Agilent 6890N, Agilent Technologies, USA) equipped with a mass selective detector (Agilent 5973N, Agilent Technologies, USA). Fatty acids were identified from the resultant chromatograph by comparing 41 the retention time of each peak to a bacterial acid methyl ester (BAME) standard mix (Supeleco 47080-U, USA) and by using a template PLFA chromatograph from Knief et al. (2003). The abundance of identified PLFAs was determined by calculating the area under the corresponding PLFA peak from the chromatogram. Results were first reported as the concentration of extracted PLFA per gram of freeze dried soil (nmol·g soil -1 ). The extraction efficiency varies from sample to sample, and as such, results were reported as the abundance of the PLFA of interest relative to the total amount of extracted and identified PLFAs less the internal standard (PLFA 19:0). Only total fungal PLFA abundance was of interest, so all PLFA results reflect the relative abundance of the 18:2ω6,9 biomarker (nmol·g soil-1/nmol total PLFAs·g soil -1 ). Extracted and identified PLFAs are summarized in Table 4-3. Table 4-3 PLFAs identified from PLFA extraction procedure Phospholipid fatty acid Representative organisms Reference i16:1ω7c Gram-netative bacteria Zogg et al. 1997 16:1ω7c Gram-netative bacteria Zelles 1997 16:0 Common i17:1ω8c Gram-netative bacteria Zogg et al. 1997 16:0ω6m/10Me 17:0 Actinobacteria Federle 1986, Frostegård et al. 1993 a17:0 Gram-positive bacteria Bååth et al. 1992 17:0 Total bacteria Tunlid et al. 1989 18:2ω6,9 Fungi Zelles 1999 18:1ω9c Total microbial biomass Lindahl et al. 1997, O'Leary and Wilkinson 1988 18:1ω7c Gram-netative bacteria Wilkinson 1988 18:0 Total bacteria cy19:0 Gram-netative bacteria Lindahl et al. 1997 16:1ω9c Gram-netative bacteria Fritze et al. 2000 42 4.11 Calculations The amount of 15 N assimilated by the fungus was calculated using 1) F = T (AS - AB)/AF Where F is the total weight of N derived from the treatment, T is the total weight of N in the wood sample, AS is the atom % excess 15 N in the wood sample, AB is the atom % excess 15 N in untreated wood blocks (background enrichment), and AF is the atom % excess 15 N of the treatment solution (Powlson and Barraclough 1993). Repeat measures of wood blocks (n=5) determined AB (background enrichment) as δ 15 N = -5.48 ± 0.17 ‰. Results reported in δ15N were converted to atom % 15N using equations 2 and 3 2) R = (δ15N/1000 + 1) x 0.0036765 3) atom % 15N = R/(R+1) x 100 Water content (θdw) of soil and wood was calculated on a dry weight basis: 4) θdw = (wet weight of sample – dry weight of sample) / dry weight of sample Absolute N content in wood blocks was calculated as follows: 5) Absolute N content = %N x dry weight of wood (µg N) The state of decay of each wood block was calculated by comparing the final mass of the wood block to the initial volume (4 cm 3 ): 6) Relative density = final mass of wood block / 4 cm3 (g·cm-3) 43 The mass loss of wood blocks was calculated for each time period as follows: 7) Mass loss = [(Mean mass of uninoculated blocks – mean mass of inoculated blocks at time t) / Mean mass of uninoculated blocks] x 100 (%) The C/N ratio of wood blocks was calculated as follows: 8) C/N ratio = Mass of C in wood blocks / Mass of N in wood blocks 4.12 Statistics All data analysis was performed using SAS Institute Inc, Version 9.2 (Cary, NC, USA). The residuals of the data were first tested for normality and equal variances using Kolmogorov- Smirnov and Bartlett’s tests, respectively. Several data sets failed these tests and transformations were required to normalize the residuals and equalize their variances. In many cases, a log transformation normalized the residuals, but was only able to marginally improve the test statistic for equal variances. In cases where the Kolmogorov-Smirnov and Bartlett test statistics improved with log transformation, further statistical analysis was completed using log- transformed data. Test statistics after log-transformation are reported in the results for individual analyses. All results are presented as back-transformed data. Differences were tested between uninoculated and inoculated wood blocks within the same treatment [e.g. NH4 + uninoculated (Day 6) vs. NH4 + inoculated (Day 6)], between treatments [e.g. 15 NH4 + inoculated (Day 6) vs. 15 NO3 - inoculated (Day 6)], and through each sampling time [e.g. NH4 + inoculated (Day 6) vs. NH4 + inoculated (Day 30)], for all variables using a two-way Analysis of Variance (ANOVA) with 0.05 level of significance. Similarly, differences were tested between uninoculated and inoculated soil portions (under, proximal, middle and distal) for 44 fungal biomass and δ15N. Bonferroni’s multiple comparison tests were used to compare the means. In most cases, 36 comparisons were made, reducing the alpha level from α=0.05 to 0.0014. It should be noted that failure to meet the assumptions of normality and equal variances weakens the sensitivity of ANOVA; therefore, significant differences found under these conditions are highly likely to be significantly different. 45 5 RESULTS 5.1 Hypothesis 1: 15N transfer via mycelial cords 5.1.1 Preliminary experiment In a preliminary experiment, relative abundance of the PLFA 18:2ω6,9 and δ15N (‰) values were measured for the soil portions distal, middle, proximal and under, in microcosms with inoculated or uninoculated wood blocks 30 days after application of dissolved 15NH4 +. The relative abundance of the PLFA 18:2ω6,9 was not significantly different in soil portions when comparing inoculated and uninoculated microcosms within the same treatment or between treatments. Comparisons were made using log-transformed data; the transformation normalized the residuals (Kolmogorov-Smirnov, p=0.1370) but was only able to improve residual variances (Bartlett, p=0.0382). Overall, the relative abundance of the PLFA 18:2ω6,9 tended to be higher in inoculated microcosms (Figure 5-1). For both inoculated and uninoculated microcosms, soil δ15N values were highest near the site of application (‘distal’ soil portion) and gradually became less enriched toward the wood block (‘under’ soil portion) (Figure 5-2). The δ15N values of all soil portions differed significantly when comparing within inoculated or uninoculated treatments. However, only the proximal soil portion was significantly different (p<0.0001) when comparing between inoculated and uninoculated microcosms. After 30 days, all soil portions in both inoculated and uninoculated microcosms had δ15N values above the δ15N natural abundance of the soil (δ15N = 2.92 ‰ ± 0.01 S.E.). Comparisons were made using log-transformed data, but the transformation was only able to improve the test statistics for normality and equal variance. I intended to 46 correlate soil portion δ15N values and fungal biomarker results in order to determine the correlation between these two variables, but the data were highly variable and 15 N was transferred under wood blocks in both uninoculated and inoculated microcosms. 0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 Under Proximal Middle Distal R el a ti v e P L F A a b u n d a n ce (n m o l· g s o il -1 / n m o l to ta lP L F A s· g s o il -1 ) Soil position 15 NH4 + Inoculated 15 NH4 + Uninoculated Figure 5-1 Abundance of the PLFA biomarker 18:2ω6,9 (relative to total extractable PLFAs) for four soil portions (under, proximal, middle and distal to the wood block) in inoculated and uninoculated microcosms 30 days after application of dissolved 15 NH4 + (p<0.05, n=2-4, mean ± S.E.). 47 5.1.2 Main experiment Soil δ15N values for distal, proximal and under soil portions is presented in Figure 5-3. Soil δ15N data in uninoculated microcosms is only available for the third sampling occasion (day 97, 30 days after N addition), thus comparisons were only made between uninoculated and inoculated microcosms on this occasion. Generally, δ15N values in soil were elevated above the background in all soil portions regardless of treatment. As in preliminary microcosms, δ15N values were highest in ‘distal’ portions of the microcosms and became less enriched in ‘under’ soil portions. Across all 15N treatments, there were no significant differences in δ15N of all soil portions when comparing between inoculated and uninoculated microcosms. Distal portions within uninoculated microcosms across treatments had significantly higher δ15N values relative to ‘under’ and ‘proximal’ soil portions (p<0.0001). No significant differences in δ15N values were found between ‘under’ and ‘proximal’ soil portions across treatments in uninoculated 0 50 100 150 200 250 300 350 Under Proximal Middle Distal δ 1 5 N ( ‰ ) Soil position 15 NH 4 + inoculated 15 NH 4 + uninoculated Figure 5-2 δ15N value (‰) of each soil portion (under, proximal, middle and distal to the wood block) in inoculated and uninoculated microcosms. An asterisk indicates a significant difference between inoculated and uninoculated treatments (p<0.05, n=4, mean ± S.E.). * 48 microcosms. Similarly, in inoculated microcosms, ‘distal’ soil portions across treatments had significantly higher δ15N values compared to ‘under’ and ‘proximal’ soil portions (p<0.0001). However, in inoculated 15 NO3 - microcosms, there was no difference in δ15N values between distal and proximal soil portions. No significant differences were found when comparing δ15N values in ‘proximal’ and ‘under’ soil portion in15NH4 + and 15 N-Glycine treated microcosms, but δ15N values in proximal soil portions of 15NO3 - treated microcosms were higher than in ‘under’ soil portions (p=0.0002). For all 15 N treatments ( 15 NH4 + , 15 NO3 - , and 15 N-glycine), significantly more 15 N was assimilated into inoculated wood blocks relative to uninoculated blocks on at least one sampling occasion (Figures 5-4,-5,-6). For 15 NH4 + microcosms, significantly more 15 N was assimilated into the inoculated wood block at day 6 (p<0.0001), but not at days 18 or 30, although the latter sampling occasions approached significance (p=0.0015 and p=0.0016 for days 18 and 30, respectively, α=0.0014). In 15N-glycine microcosms, significantly more 15N was assimilated into the inoculated wood block at day 6 (p=0.0002) and at day 18 (p<0.0001), but not at day 30. The only significant difference found in 15 NO3 - microcosms was at day 18 (p=0.0008). In 15 NH4 + and 15 N-glycine microcosms, no significant differences were found when comparing 15 N assimilated by uninoculated blocks between sampling times. However, in 15 NO3 - microcosms, significantly more 15 N was assimilated into uninoculated wood blocks at day 30 compared to uninoculated wood blocks at day 6 (p<0.0001) and day 18 (p<0.0001). 15 N was assimilated into all uninoculated wood blocks across treatments and sampling times, however, the amount assimilated was near zero, except for uninoculated blocks in 15 NO3 - microcosms. Values below zero of 15 N assimilated in uninoculated wood blocks were found in 15 NH4 + microcosms. It should be noted that in the equation (1) used to calculate 15 N assimilated in wood 49 blocks, the term ‘AB’ (atom % excess of untreated blocks) is derived from a δ 15 N value averaged from untreated wood blocks (n=6). When the amount of 15 N assimilated is close to nil, the variability of δ15N values for wood blocks may result in erroneously positive or negative values reported for assimilated 15 N. The ANOVA for 15 N assimilation data was run using log-transformed data. The transformation normalized the residuals (Kolmogorov-Smirnov, p=0.0628), but was only able to improve their variances (Bartlett’s, p=0.0006). 50 0 50 100 150 200 250 300 350 400 0 50 100 150 200 250 300 350 400 0 50 100 150 200 250 300 350 400 Under Proximal Distal δ 1 5 N ( ‰ ) Inoculated Uninoculated Figure 5-3 Soil δ15N values in inoculated and uninoculated soil portions (under, proximal, distal to the wood block) for each set of 15 N treated microcosms; A) 15 N-glycine B) 15 NH4 + C) 15 NO3 - . An asterisk indicates a significant difference between uninoculated and inoculated soil portions for a given soil portion (p<0.05, n=4, mean ± S.E.). A B C Soil position 51 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 1 5 N a ss im il a te d ( µ g N ) A b so lu te N c o n te n t (µ g N ) R el a ti v e d en si ty ( g ·c m -3 ) NH 4 inoculated NH 4 uninoculated 0 200 400 600 800 1000 1200 1400 1600 1800 2000 0 0.1 0.2 0.3 0.4 0.5 0.6 Day 6 Day 18 Day 30 Time * Figure 5-4 15 N assimilated (A), absolute N content (B) and relative density (C) of inoculated and uninoculated wood blocks for 15 NH4 + treated microcosms at three sampling times (Days 6, 18 and 30). An asterisk indicates a significant difference between uninoculated and inoculated wood blocks for a given variable and sampling time (p<0.05, n=4, mean ± S.E.). A B C * 52 0 0.1 0.2 0.3 0.4 0.5 0.6 Day 6 Day 18 Day 30 Time 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0 200 400 600 800 1000 1200 1400 1600 1800 2000 1 5 N a ss im il a te d ( µ g N ) A b so lu te N c o n te n t (µ g N ) R el a ti v e d en si ty ( g ·c m -3 ) NO 3 inoculated NO 3 uninoculated Figure 5-5 15 N assimilated (A), absolute N content (B) and relative density (C) of inoculated and uninoculated wood blocks for 15 NO3 - treated microcosms at three sampling times (Days 6, 18 and 30). An asterisk indicates a significant difference between uninoculated and inoculated wood blocks for a given variable and sampling time (p<0.05, n=4, mean ± S.E.). A B C * * * 53 0 0.1 0.2 0.3 0.4 0.5 0.6 Day 6 Day 18 Day 30 Time 0 200 400 600 800 1000 1200 1400 1600 1800 2000 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 1 5 N a ss im il a te d ( µ g N ) A b so lu te N c o n te n t (µ g N ) R el a ti v e d en si ty ( g ·c m -3 ) Glycine inoculated Glycine uninoculated * * Figure 5-6 15 N assimilated (A), absolute N content (B) and relative density (C) of inoculated and uninoculated wood blocks for 15 N-glycine treated microcosms at three sampling times (Days 6, 18 and 30). An asterisk indicates a significant difference between uninoculated and inoculated wood blocks for a given variable and sampling time (p<0.05, n=4, mean ± S.E.). A B C * 54 5.2 Hypothesis 2: preferential use of N forms The amount of 15 N assimilated by inoculated wood blocks did not differ significantly between 15 N treatments when compared within or between sampling times (Figure 5-7). The mean amount of 15 N (µg) assimilated by inoculated blocks for each treatment across all sampling times was 0.29 ± 0.12, 0.20 ± 0.03, and 0.19 ± 0.04 (S.E.) in the 15 NH4 + , 15 NO3 - and 15 N-glycine treatments, respectively. The amount of 15 N assimilated was variable across treatments, ranging from 0.013 to 1.36 µg 15 N. The average amount of 15 N (µg ) assimilated across all treatments was 0.22 ± 0.04 (S.E.). This represents an average of 0.91 ± 0.17 (S.E.) % of 15 N assimilated in inoculated wood blocks relative to the 24 µg of 15 N applied to each microcosm. 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 Day 6 Day 18 Day 30 1 5 N a ss im il a te d ( µ g N ) Time Glycine NH4 NO3 Glycine NH 4 NO 3 Figure 5-7 15 N assimilated in inoculated wood blocks for each treatment ( 15 N- glycine, 15 NH4 + , 15 NO3 - ) and sampling time (n=4, mean ± S.E.). 55 5.3 Hypothesis 3: absolute N accumulation For all N treatments, the absolute N content of inoculated wood blocks was significantly higher relative to uninoculated blocks on at least one sampling occasion (Figure 5-4,-5,-6 pane B). For 15 NH4 + microcosms, the absolute N content of inoculated blocks was significantly higher at Day 6 (p<0.0001), and approached significance at Day 30 (p=0.0014). Significantly more N was found in inoculated blocks at Day 6 (p<0.0001) for 15 N-glycine treatments. For 15 NO3 - microcosms, N content in inoculated microcosms was significantly higher at Days 6 and 18 (p<0.0001). The N content of inoculated blocks did not differ significantly when comparing treatments within and between sampling times. Additionally, no significant differences were found when comparing the N content of uninoculated blocks between treatments and across sampling times. The ANOVA for absolute N content data was run using log-transformed data. The transformation normalized the residuals (Kolmogorov-Smirnov, p=0.0557), but was only able to improve their variances (Bartlett’s, p=0.0013). 5.4 Question 1: transfer of litter derived 15N Significantly more 15 N derived from needle litter was found in inoculated blocks when compared to uninoculated blocks (p=0.0068, Figure 5-8). The N content of inoculated blocks was significantly higher (p=0.017), and the relative density of inoculated blocks was significantly lower (p<0.0001) then uninoculated blocks (Figure 5-8 B,C). The ANOVA for all three variables was run using untransformed data. The variances of the residuals were equal for absolute N and relative density data (Bartlett’s, p=0.1857, p=0.7364, 56 respectively), but not for assimilated 15N data (Bartlett’s test, p=0.0151). The residuals for absolute N data were normally distributed (Kolmogorov-Smirnov, p>0.1500), but not for assimilated 15 N or relative density (Kolmogorov-Smirnov, p>0.1500, p=0.0391, respectively). Despite this, significant differences were found for all three variables and it was not necessary to transform the data. 5.5 Question 2: temporal patterns of N transfer and accumulation As was noted in Section 5.3, there was no difference in absolute N content when comparing between treatments in each sampling time (e.g. sampling time 1, 15 NO3 - microcosms vs. 15 NH4 + microcosms), therefore, results from each treatment were pooled for each sampling occasion when analyzing trends through time. Furthermore, days since inoculation were used instead of days since application of tracer (hence, day 6 = day 73, day 18 = day 85, day 30 = day 97). Generally, in all treatments, the absolute N content of inoculated wood blocks declined from d73 to d97. The absolute N content of uninoculated blocks remained constant from d73 to d85, and declined from Day 85 to Day 97 (Figure 5-9). For inoculated wood blocks, the decline in N content was significant between each sampling time (Day 73 vs. Day 85, p=0.0014; Day 97 vs. Day 73, Day 85, p<0.0001; α=0.0014). For uninoculated blocks, the absolute N content at Day 97 was significantly lower when compared to blocks at Days 73 (p=0.0023) and 85 (p=0.0031). On each sampling occasion, inoculated blocks contained significantly more absolute N relative to the uninoculated blocks (p<0.0001 for each comparison); the absolute N content of inoculated wood blocks was 211±12, 171±19, and 144±10 % higher than uninocualted blocks at Days 73, 85 and 97, respectively. By Day 97, inoculated blocks had lost 80% of accumulated N. No clear 57 temporal trend was visible for assimilated 15 N data across treatments; the amount of 15 N assimilated was relatively constant through time in inoculated blocks for each treatment. The declining N content trend through time is also visible for relative density. Generally, the relative density of inoculated blocks across treatments declined from Day 6 to 30, while the relative density of uninoculated blocks remained constant. The relative density of inoculated blocks at Day 97 was significantly lower compared to Day 73 for 15 NO3 - and 15 N-glycine treatments (p<0.0001, p=0.0012, respectively; α=0.0014), and approached significance for the 15 NH4 + treatment (p=0.0072; α=0.0014). Mass loss (%) of wood blocks presented in Figure 5-10 and is only available for inoculated blocks. The final mass of uninoculated wood blocks, pooled across treatments, was not significantly different between each sampling time. Since the initial mass was not recorded, the average mass of uninoculated wood blocks pooled across treatments and sampling occasions (n=36) was used as the initial mass (1.50 ± 0.01 g) for inoculated wood blocks. For this reason, mass loss results are interpreted with caution. After 73 days of incubation, mass loss of inoculated wood blocks was 12.2 ± 2.7 %, but there was no significant difference in mass loss by day 85. However, from Day 85 to Day 97, mass loss (26.9±1.0%) was significantly higher when compared to mass loss at Days 73 (p=0.0003) and 85 (p=0.001). The change in the C/N ratio over time for both uninoculated and inoculated wood blocks is presented in Figure 5-11. There was no significant change in the C/N ratio of uninoculated blocks from Day 73 to 85, but a significant increase occurred from Day 85 to 97 (p=0.0001). The C/N ratio also increased between Days 73 and 97 in inoculated blocks, but the difference was 58 only significant when comparing Day 73 to Day 97 (p=0.0002). Inoculated blocks had significantly lower C/N ratios than uninoculated blocks on all sampling occasions (p<0.0001). 5.6 Transfer of 15 NO3 - to uninoculated wood blocks The highest average assimilated 15 N (0.55 ± 0.12 µg 15 N) was recorded in uninoculated wood blocks at Day 30. This unexpected result warranted further investigation and it was hypothesized that the water content of wood blocks was a controlling variable. An experiment was designed to test this hypothesis (see Methods, Section 3.8), but the treatments used to control the water content of wood blocks did not produce distinct groups (high, medium and low water content) for use in an ANOVA. Instead, I produced a simple regression comparing water content (θdw) to assimilated 15 N (Figure 5-12). A significant negative correlation (R 2 =0.6817, p=0.0116) was found between assimilated 15 N and the final water content of uninoculated wood blocks in 15 NO3 - microcosms. One outlier, the wood block with both the lowest water content and the lowest assimilated 15 N, was removed from the regression. 59 0 0.005 0.01 0.015 0.02 0.025 0.03 1 5 N a ss im il a te d ( µ g N ) 0 200 400 600 800 1000 1200 1400 1600 1800 2000 A b so lu te N c o n te n t (µ g N ) 0 0.1 0.2 0.3 0.4 0.5 0.6 Day 30 Litter inoculated Litter uninoculated R el a ti v e d en si ty ( g ·c m -3 ) * * * Figure 5-8 15 N assimilated (A), absolute N content (B) and relative density (C) of inoculated and uninoculated wood blocks for microcosms treated with 15N labelled Douglas-fir litter after 30 days of litter application. An asterisk indicates a significant difference between uninoculated and inoculated wood blocks for a given variable (p<0.05, n=14, inoculated and 4, uninoculated, mean ± S.E.). A B C 60 Figure 5-9 Absolute N content of inoculated and uninoculated wood blocks pooled across 15 N treatments at three sampling times (Days 73, 85 and 97). An asterisk indicates a significant difference between uninoculated and inoculated wood blocks at each sampling time (p<0.0001, n=12, mean ± S.E.). Figure 5-10 Mass loss of inoculated wood blocks pooled across 15 N treatments at three sampling times (Days 73, 85 and 97). Different letters indicate significant differences between sampling occasions (p<0.05). 0 200 400 600 800 1000 1200 1400 1600 1800 Day 73 Day 85 Day 97 A b so lu te N c o n te n t (µ g N ) 0 5 10 15 20 25 30 70 75 80 85 90 95 100 M a ss l o ss ( % ) Time (days) Inoculated Uninoculated * * * a a b 61 Figure 5-11 C/N ratio of uninoculated blocks (open circles,○) and inoculated blocks (closed circles,●) at each sampling occasion (73, 85 and 97). Different letters indicate significant differences between sampling times and block type (p<0.05). 0 200 400 600 800 1000 1200 1400 70 75 80 85 90 95 100 C :N Time (days) a a b c c d 62 Figure 5-12 Relationship between final water content of wood blocks and assimilated 15 N after 15 NO3 - was added to the microcosms and microcosms were incubated for 6 days. 0 0.05 0.1 0.15 0.2 0.25 0.3 0.8 1.3 1.8 2.3 1 5 N A ss im il a te d ( µ g N ) Water content (g/g) y=-0.1101x+0.2741 R2=0.6817 p=0.0116 63 6 DISCUSSION 6.1 Hypothesis 1: 15N transfer via mycelial cords The findings from this experiment support hypothesis one; the wood decay fungus Hypholoma fasciculare translocated added 15 N from soil to colonized wood blocks (Figures 5-3,- 4,-5, A). While more 15 N was assimilated into inoculated wood blocks for all 15 N treatments, it is unclear from which portion of the soil that 15 N was translocated. From observation, soil directly under the wood blocks was caked together with fungal mycelia, whereas soil in proximal and distal portions of the microcosm was only loosely held together. This suggests that hyphae were concentrated directly under the wood block, an observation that is loosely supported by the relative abundance of the fungal biomarker 18:2ω6,9; in preliminary microcosms, relative abundance was highest under wood blocks, but not significantly. After 73 days of incubation, the absolute N content of wood blocks increased across all treatments by 211±12% relative to uninoculated blocks, but of the 15 N that was added, less than one percent was found in inoculated wood blocks, averaged across all treatments. Thus, while inoculated wood blocks were a strong sink for N, they were poor scavengers of the 15 N added at the distal end of the microcosm. Given that inoculated wood blocks were such a strong sink for N, it is not clear why only a small amount of the added 15 N was translocated to the wood blocks. It is possible that the site of 15 N uptake was directly under the wood blocks where the mycelial density was highest rather than at the distal portion of the microcosms where 15N was added. From soil δ15N results (Figure 5-3), 15 N was found in all soil portions elevated above natural abundance in both uninoculated and inoculated microcosms; 15 N diffused through the soil regardless of the presence of the fungus. Microcosms were watered on a weekly basis, and 15 N movement would have been 64 facilitated through soil water. As 15 N moved through soil water toward the wood block, the dense mycelia in close proximity to the wood block transferred the 15 N. This could explain why only a small amount of the added 15 N was found in the wood blocks; the amount transferred depended on how much arrived by transport through soil water. Also, the amount of 15 N that was transferred into wood blocks was highly variable, possibly due to the uncontrolled movement of 15 N through soil water. Ultimately, the design of the microcosms precludes unequivocal evidence demonstrating from where within the microcosms 15 N was transferred. A labelling well was built into preliminary microcosms and this would have isolated the site of 15 N uptake at the distal end of the microcosms. Unfortunately, mycelia did not grow into the labelling well, so it was removed in all other experiments. Regardless, it is clear that 15 N was only transferred to inoculated wood blocks, so the presence of the fungus was responsible for 15 N transfer. An anomalous amount of 15 NO3 - was found in uninoculated wood blocks; this result is discussed is Section 6.6. Also, 15 N- glycine may have been mineralized by other soil organisms prior to being translocated into the wood blocks. Thus, it is not certain that the amino acid was transferred intact. The exact site of N uptake can reveal important aspects of N translocation in wood decay fungi. If the site of N uptake in this experiment was largely under the wood blocks where mycelial density was greatest, then the role of exploratory hyphae at the growing front may not necessarily be for nutrient acquisition. While N is often thought of as a limiting nutrient, carbon, being the energy source for saprotrophic organisms, must be prioritized first to facilitate growth, followed by N scavenging for protein synthesis (Watkinson et al. 2006). That the fungus did not transfer very much of the added 15 N may indicate that the fungus had fulfilled its nitrogen requirements through N transferred close to the wood block. Further evidence to support this 65 conjecture is supplied in Section 5.5, and if this is the case, then hyphae at the growing front may have been primarily foraging for a new C supply rather than actively scavenging for nitrogen. This work supports claims that wood decay fungi are capable of N translocation, and complements similar work carried out demonstrating phosphorus translocation from soil to inoculated wood resources (e.g. Wells and Boddy 1995). Additionally, this work reinforces fungal 15 N translocation that has been demonstrated from agricultural soils into straw litter (Frey et al. 2000). 6.2 Hypothesis 2: preferential use of N forms Results from this work do not support the hypothesis that Hypholoma fasciculare preferentially used any form of added 15 N; no differences in 15 N assimilated were found between treatments (Figure 5-7). However, as has been previously mentioned, it is not clear where 15 N uptake occurred within the microcosm. The absolute N content of inoculated wood blocks more than doubled after 73 days of incubation, but less than 1% of the added 15 N was transferred to the wood blocks. Movement though soil water may have been largely responsible for transporting 15 N to areas of high mycelial density, at which point the 15 N was translocated to the wood blocks. Both the design of the experiment and the observation that hyphae at the growing front did not translocate large amounts of 15 N make it difficult to conclude that Hypholoma fasciculare has no preference for any of the 15 N forms added. If 15 N were added underneath the wood blocks where mycelial density was highest and where I suspect the greatest amount of N translocation occurred, the fungus may have shown a preference, possibly for 15 NH4 + . This would be in accordance with molecular-based research demonstrating fungal preference for ammonium over nitrate and other amino acids as its N source (Caddick 2002). 66 6.3 Hypothesis 3: absolute N accumulation There is strong evidence to support hypothesis three; after 73 days of incubation, absolute N content in inoculated wood blocks increased by 211±12% relative to uninoculated control blocks. This trend also continued through time, but the effect was most pronounced on the first sampling occasion. Uninoculated blocks showed relatively minor changes in absolute N content through time, providing further evidence that the presence of the fungus was the single factor responsible for the observed increase in absolute N content. These results suggest that during the incubation, inoculated wood blocks were a strong sink for soil-derived N. It is important to note that some of the N found within the wood blocks could have been imported from the culture medium (Potato Dextrose Agar) during the 15 day incubation prior to the blocks being transferred to the microcosms. The majority of the translocated N probably came from the soil as the length of incubation in the microcosms was much longer than that within the culture media. Wood decay fungi have long been suspected as responsible for observed increases in the absolute N content of woody debris over time. Although it is difficult to compare coarse woody debris to a 4 cm 3 wood block, these results provide further evidence that wood decay fungi import N into decomposing woody resources with broad C/N ratios. 6.4 Question 1: transfer of litter derived 15N Although significantly more 15 N from the Douglas-fir needle litter was assimilated into inoculated wood blocks vs. uninoculated wood blocks, the mechanism that released 15 N from the labelled Douglas-fir litter remains obscure. Generally, the amount assimilated from the needle litter was much smaller compared to soluble inorganic and organic 15 N. There was no visible 67 evidence from the soil surface of colonization of the added needle litter by the growing hyphae. Thus, it is not clear how the fungus was able to translocate the added 15 N. One possibility is that soil organisms in the unsterile soil mineralized the 15 N contained within the litter and the fungal mycelia scavenged this 15 N, translocating it to the wood substrate. Another possibility is that during watering of the microcosms, 15 N was leached from the litter and captured by fungal hyphae. The fungus was only able to assimilate 0.10±0.07% of the total 15 N available in the labelled litter, however, 139% more total N accumulated within inoculated wood blocks relative to uninoculated control blocks during 158 days of incubation. These results agree with results from the main experiment; inoculated wood blocks were a strong sink for N but hyphae were only able to mobilise a small amount of the 15 N from the labelled litter located at the distal end of the microcosms. 6.5 Question 2: temporal patterns of N transfer and accumulation While the absolute N content of inoculated wood blocks was significantly higher compared to uninoculated blocks at all sampling times, 80% of the N accumulated in inoculated blocks was lost by day 97. These results highlight a precipitous decline over time in the accumulated absolute N content of the inoculated wood blocks. However, the N content of the inoculated wood blocks did not drop below the initial N content of the uninoculated wood blocks, suggesting that at day 97, some of the accumulated N was still retained within the wood blocks. A similar, but less pronounced trend occurred in uninoculated blocks; by day 97, N was released from the uninoculated wood blocks when the N content was compared to that on days 73 and 85. 68 As N was lost from inoculated wood blocks, mass loss (%) increased (Figure 5-10) and relative density decreased. This indicates that the fungus and any other decay organisms present were respiring carbon from the wood block over time. Generally, for most litter types, a decrease in the C/N ratio occurs as decomposition proceeds and carbon is respired. However, the C/N ratio of inoculated wood blocks increased significantly from day 73 to 97 (Figure 5-11). This result is attributable to the precipitous loss of N from the blocks; while both C and N were lost, N was lost at a faster rate, resulting in an increase in the C/N ratio with time. The same trend occurred in uninoculated blocks, but the increase in the C/N ratio only occurred from day 85 to 97. The decline in N content as well as the increase in the C/N ratio of inoculated blocks may be related to visual indications that the fungus was senescing. Mature cord systems of Hypholoma fasciculare become yellow with age and in these experiments, mycelial cords on the soil surface and abutted against the microcosm walls changed from white to yellow during the incubation. Yellowing was especially pronounced at the time the tracer was added (day 67), and the decision to apply the tracer at this time was made because hyphal extension had effectively ceased. This observational evidence suggests that the fungus was senescing and that loss of N from the inoculated wood blocks could be attributed to hyphal autolysis and subsequent N mineralization or even denitrification by soil bacteria. It is not clear why the fungus began senescing, but it may be related to the cultural conditions. The microcosm system used for incubation was an open system and any number of cultural conditions may have contributed to the senescence. The wood substrate may have been too wet, allowing for soil bacteria, parasitic fungi, or soil fauna to overtake the fungus. Alternatively, the fungus could have been nutrient-limited, but it is not likely that N was limiting. 69 Nearly all of the 15 N added to the microcosms remained in the soil, suggesting that the fungus did not access much of this (presumably available) 15 N pool. The fungus was also able to greatly increase the absolute N content of the wood blocks, and given that wood decay fungi are known to hoard nutrients above their nutritional requirements and are capable of efficient nutrient recycling, the fungus should have been able to utilize the N it had already translocated (Boddy 1999). However, limitation by phosphorus or other micronutrients could have resulted in the observed senescence. Fungal senescence would also explain why there was no observed trend in 15 N translocation. In all treatments, the amount of 15 N assimilated remained relatively constant through time, but because the blocks were destructively harvested (not repeated measures), it is possible that all of the 15 N was translocated within a few days of tracer application, before the precipitous decline in absolute N occurred. With a higher-quality carbon source, the amount of 15 N translocated may have increased with time as the fungus mobilized carbon and nutrient resources to facilitate growth. 6.6 15NO3 - transfer to uninoculated wood blocks The largest amount of 15 N that was assimilated by wood blocks in any of the treatments occurred in uninoculated blocks in 15 NO3 - treated microcosms (Figure 5-5). It is clear that 15 NO3 - was highly mobile relative to 15 NH4 + and 15 N-glycine; soil δ15N results (Figure 5-3 C) show elevated but highly variable 15 N values in proximal and under soil portions. High mobility relative to ammonium and glycine is not an unexpected result given the negative charge on the nitrate ion, but 15 N assimilation into uninoculated wood blocks was unexpected. While biotic 15 NO3 - transfer cannot be ruled out, it is probable that was that this transfer occurred largely 70 abiotically; the blocks were autoclaved and there was no visible evidence of mycelial colonization of the wood blocks from the unsterile soil. Bacteria were probably present on the surface and within the inoculated wood blocks, but the capacity for nutrient translocation by bacteria is limited due to their single-cell nature. One possible explanation for this apparent abiotic 15 NO3 - uptake is that nitrate entered the wood block along with absorbed soil water. Nitrate, being highly mobile, could have been assimilated into the wood blocks as the water content of the blocks equilibrated to that of the underlying soil; 15 NO3 - in soil solution was assimilated as wood blocks absorbed soil water. Wood is a hydroscopic material and it equilibrates its water content to that of the environment through surface tension forces (absorption, capillary action) or adsorption to hydrogen-bonding sites within lignin, hemicellulose and cellulose (Haygreen and Bowyer 1982). Dry wood has a greater capacity to absorb water along with compounds in solution ( 15 NO3 - , for example). A simple linear regression shows a negative correlation (R 2 = 0.68) between 15 N assimilated and water content of the wood blocks. The sample size of the regression is low (n=9), but it does provide some evidence to support this hypothesis; the drier the wood block, the more water it absorbs and with that water 15 NO3 - is also assimilated. There was no evidence of a combined effect of absorption of 15 NO3 - through soil water and mycelial uptake. Fungal hyphae are hydrophobic in nature and it is probable that 15 NO3 - was assimilated via inducible cell membrane bound transport (Burstaller 1997). Additionally, the hyphae likely maintained a high moisture level within the wood block, limiting any potential hydroscopic uptake of 15 NO3 - . 71 Further study is required to assess the ecological significance of hydroscopic absorption of nitrate. Homyak et al. (2008) demonstrated that wood chips retained 30-40% of the nitrate pulse associated with clear felling and attributed this result to fungal translocation. However, it is not clear how much nitrate is actively translocated by fungal mycelium and how much is potentially absorbed through hydroscopic forces. 6.7 Implications for fungal ecology and woody debris management This work adds to the body of research demonstrating nutrient translocation by wood decay fungi from areas of supply (soil) to areas of demand (a woody substrate). Most of the work dealing with fungal nutrient transfer has focused on phosphorus. Early work on phosphorus translocation by Phanerochaete velutina and Phallus impudicus indicated that 90% of phosphorus added to the growing fronts of these wood decay fungi was translocated via mycelial cords either to freshly colonized wood baits or to the original inocula (Wells et al. 1990). Limited work has been conducted demonstrating fungal nitrogen transfer. Frey et al. (2000) showed that fungicide application reduced 15 N influx from labelled soil into wheat straw by 59- 78%. Using a 14 C-labelled non-metabolized amino acid analogue (aminoisobutyric acid, AIB) and specialized imaging techniques, Tlalka et al. (2007) were able to demonstrate system-wide preferential reallocation of the amino acid to a newly discovered wood resource. The work presented here provides further evidence demonstrating 15 N translocation to woody substrates via mycelial cords of the wood decay fungus Hypholoma fasciculare. This work also highlights the importance of C resources for retention of accumulated nitrogen in woody debris. Fungi have long been suspected as responsible for the increased absolute N content in woody debris over time, and this work demonstrates that the presence of 72 the fungus more than doubled the absolute N content of the wood block after 73 days of incubation. However, 80% of the accumulated N was lost by the end of the experiment, which was attributed to fungal senescence. If the fungi were growing from a woody substrate on the forest floor and then encountered a new C source before it exhausted the high-quality C within the inocula, it is possible that the accumulated N would be remobilized to the new C resource rather than being mineralised. Wells and Boddy (1990) supplied wood blocks inoculated with P. velutina at three different decay stages with 32 P and found that the amount of 32 P translocated from the inocula to a newly colonized wood bait increased with the age of the inocula. Additionally, mycelium of P. velutina growing from more decayed inocula received proportionally more 32 P, suggesting that as the inocula became increasingly decayed, the fungus invested more of its nutrients into its foraging mycelia. In a more recent study, the litter decay fungi Marasmius androsaceus and Mycena epipterygia, growing on a low C substrate (glycine, C:N = 2), mineralised NH4 + from glycine, but mineralisation ceased when the fungi were allowed to colonize adjacent needle litter (Boberg et al. 2010). Thus, accumulation of nutrients within mycelia growing on C substrates is dependent on the continued supply and colonization of new C resources as well as the ability to translocate nutrients to these new resources. In the absence of new C resources, N mineralisation will occur (as in inoculated blocks). While care should be taken when extrapolating results from a laboratory microcosm to field conditions, this research underscores some important questions about N release in decomposing woody debris. Many studies demonstrate significant loss of initial N capital from woody debris within 30 years of decomposition (Busse 1994; Laiho and Prescott 1999; Palviainen et al. 2008). However, these studies rarely discuss the fate of N after it is lost from the decomposing logs. Increasingly, new models of N cycling in N-limited ecosystems suggest that 73 N is tightly cycled and largely held within mycelium of decomposer organisms (Lindahl et al. 2002). Nitrogen and other nutrients are released largely as a result of competitive interactions between saprotrophic and mycorrhizal organisms, or via grazing by soil invertebrates (Lindahl et al. 2002, Lindahl et al. 2007, Crowther et al. 2011). Thus, while N release from decomposing woody substrates has been reported here and in other studies, it is not clear where that N goes following its release from the wood. Molecular evidence suggests that ectomycorrhizal fungi dominate in highly decayed woody debris (Rajala et al. 2011) and it has been shown in laboratory microcosms that ectomycorrhizal fungi can overtake saprotrophic mycelium and even capture their nutrients when saprotrophic fungi are restricted by limited carbon supply (Lindahl et al. 2001). When the carbon supply of woody substrates becomes unfavourable for saptrophic growth, as is the case with advancing decay, mycorrhizal fungi may scavenge these resources for the largely organic N remaining. If decomposer organisms in N- limited ecosystems tightly hold on to N and other nutrients, provided C is in ample supply, what are the consequences for N dynamics if significant amounts of C-rich residues are removed after harvest? It follows that N will be lost from sites as plant demand for N declines, mycorrhizal species senesce, and the abundance of saprotrophic decomposers is reduced. Studies evaluating leaching losses of N as a result of residue removal are variable, with reported increases, decreases or no observed changes in N leaching (Strahm et al. 2005; Carlyle et al. 1998; Mann et al. 1988). Further study is required not only to test the tight N-cycling model, but also to determine how residue removal changes decomposer communities in relation to N dynamics. 74 7 CONCLUSIONS  Mycelial cords of the wood decay fungus Hypholoma fasciculare translocated a significant amount of 15 NO3 - and 15 NH4 + from soil to inoculated wood blocks for at least one sampling time. The fungus was also able to translocate significant amounts of 15 N- glycine, but it is not clear if the glycine was translocated intact.  Over the course of the experiment, Hypholoma fasciculare showed no preference for any of the added forms of 15 N, but variability associated with movement of 15 N through soil water and evidence of mycelial senescence confounded these results.  After 73 days of incubation, inoculated wood blocks increased in absolute N content by 211% relative to uninoculated control blocks.  Significantly more litter-derived 15N was found in inoculated wood blocks relative to uninoculated blocks.  After 97 days of incubation, 80% of the N accumulated in inoculated wood blocks was released. This precipitous decline was attributed to mycelial senescence.  Uninoculated wood blocks in microcosms treated with 15NO3 - absorbed the largest amount of 15 NO3 - , more so than in inoculated microcosms. This was attributed to hydroscopic ab/adsorption of soil water as well as the high mobility of nitrate. 75 8 RECOMMENDATIONS AND FUTURE RESEARCH Current regulations regarding woody debris management in British Columbia require retention of a minimum of 4 logs per hectare (>5m in length and 30 cm diameter on the Coast, and >2 m in length and 7.5 cm diameter in the Interior) (Forest and Range Practices Act, 2010). While a recent report by British Columbia’s Chief Forester provides guidance for woody debris management, binding regulations are limited to those outlined above (Snetsinger, 2010). Management guidelines for slash and smaller woody debris are non-existent and this material is often piled and burned or left to decompose during site preparation. Results from this work indicate that wood decay fungi are capable of mobilising N from the underlying forest floor and accumulating it within a wood substrate. If these results are extrapolated to a cut-block, it follows that woody debris on the forest floor that becomes colonized by wood decay fungi would accumulate N, at least in early stages of decay, and this process may immobilise N after harvest. If retention of N on site after harvest is a management objective, site preparation that maximizes accumulation of N in woody debris by saprotrophic fungi should be employed. This would involve homogenously distributing woody debris on site in close contact with the forest floor in order to promote fungal colonization and N accumulation. What remains unclear is the fate of the N accumulated by wood decay fungi. Further work is required to follow N accumulated in woody debris though the decay process. This might involve a field experiment where branches previously labelled with 15 N are left to decompose on the forest floor and the 15 N is traced to other N pools in the surrounding area. Of particular interest would be to determine if this 15 N is released from the branch as decay proceeds, and to determine how this 15 N is partitioned within surrounding N pools. How much (if any) of this 15 N is 76 measured in leachate, in adjacent woody debris, in various parts of the forest floor and mineral soil, within surrounding vegetation (mycorrhizal uptake), or within the soil microbial community would serve to evaluate the fate of N that accumulates within woody debris. 77 REFERENCES Alban DH and Pastor J. 1993. Decomposition of aspen, spruce, and pine boles on two sites in Minnesota. Can.J.for.Res 23(9):1744-9. Allison SD and Vitousek PM. 2005. Responses of extracellular enzymes to simple and complex nutrient inputs. Soil Biol Biochem 37(5):937-44. Anderson J, Ineson P, Huish S. 1983. Nitrogen and cation mobilization by soil fauna feeding on leaf litter and soil organic matter from deciduous woodlands. Soil Biol Biochem 15(4):463-7. Arora D. 1986. Mushrooms demystified: a comprehensive guide to the fleshy fungi. Berkley: Ten Speed Press. p. 959. Arthur MA and Fahey TJ. 1990. Mass and nutrient content of decaying boles in an Engelmann spruce– subalpine fir forest, Rocky Mountain National Park, Colorado. Can.J.for.Res 20(6):730-7. Bååth E and Söderström B. 1979. Fungal biomass and fungal immobilization of plant nutrients in Swedish coniferous forest soils. Rev.Ecol.Biol.Sol 16(4):477-89. Bååth E, Frostegård Å, Fritze H. 1992. Soil bacterial biomass, activity, phospholipid fatty acid pattern, and pH tolerance in an area polluted with alkaline dust deposition. Appl Environ Microbiol 58(12):4026-31. Baldrian P. 2008. Enzymes of saprotrophic basidiomycetes. In: Boddy L, Frankland JC, Van West P, editors. Ecology of Saprotrophic Basidiomycetes. San Francisco: Elsevier. p 211-238.. Bardgett RD, Streeter TC, Bol R. 2003. Soil microbes compete effectively with plants for organic- nitrogen inputs to temperate grasslands. Ecology 84(5):1277-87. Blanchette RA. 1995. Degradation of the lignocellulose complex in wood. Canadian Journal of Botany 73(S1):999-1010. Blanchette RA. 1991. Delignification by wood-decay fungi. Annu Rev Phytopathol 29(1):381-403. Blanchette RA and Biggs AR. 1992. Defense mechanisms of woody plants against fungi. Berlin: Springer-Verlag. p. 458 Boberg JB, Finlay RD, Stenlid J, Lindahl BD. 2010. Fungal C translocation restricts N‐mineralization in heterogeneous environments. Funct Ecol 24(2):454-9. Boddy L. 2001. Fungal community ecology and wood decomposition processes in angiosperms: From standing tree to complete decay of coarse woody debris. Ecological Bulletins (49):43-56. Boddy L. 1999. Saprotrophic cord-forming fungi: Meeting the challenge of heterogeneous environments. Mycologia 91(1):13-32. 78 Boddy L. 1993. Saprotrophic cord-forming fungi: Warfare strategies and other ecological aspects. Mycol Res 97(6):641-55. Boddy L. and Heilmann-Clausen J. 2008. Basidiomycete community development in temperate angiosperm wood. In: Boddy L, Frankland JC, Van West P, editors. Ecology of Saprotrophic Basidiomycetes. San Francisco: Elsevier. p 211-238. Boddy L and Jones T. 2007. Mycelial responses in heterogeneous environments: Parallels with macroorganisms. In: Gadd GM, Watkinson SC, and Dyer PS (editors). Fungi in the Environment New York: Cambridge University Press. p 112-158. Boddy L and Watkinson SC. 1995. Wood decomposition, higher fungi, and their role in nutrient redistribution. Canadian Journal of Botany 73(S1):1377-83. Boddy L, Hynes J, Bebber DP, Fricker MD. 2009. Saprotrophic cord systems: Dispersal mechanisms in space and time. Mycoscience 50(1):9-19. Haygreen JG and Bowyer JL. 1982. Forest products and wood science: An introduction. Ames: The Iowa State University Press. 489 p. Bradley D. 2006. Canada biomass-bioenergy report. Climate change solutions, Ottawa, Ont. 20 p. Available at www.climatechangesolutions.net/pdf/canada_country2006.pdf Caddick MX. 2002. What’s for dinner, what shall I choose? Genes Funct 1:37-49. Carlile MJ, Watkinson SC, Gooday G. 2001. The fungi. San Francisco: Academic Press. 588 p. Carlyle JC, Bligh MW, Nambiar EKS. 1998. Woody residue management to reduce nitrogen and phosphorus leaching from sandy soil after clear-felling Pinus radiata plantations. Can J For Res 28(8):1222-32. Carter MC, Dean TJ, Zhou M, Messina MG, Wang Z. 2002. Short-term changes in soil C, N, and biota following harvesting and regeneration of loblolly pine (Pinus taeda L.). For Ecol Manage 164(1- 3):67-88. Carter RE and Lowe RE. 1986. Lateral variability of forest floor properties under second-growth Douglas-fir stand and the usefulness of compiosite sampling techniques. Can J For Res 16:1128- 1132. Chapela I, Boddy L, Rayner A. 1988. Structure and development of fungal communities in beech logs four and a half years after felling. FEMS Microbiol Lett 53(2):59-70. Coates D and Rayner A. 1985a. Fungal population and community development in cut beech logs. I. establishment via the aerial cut surface. New Phytol 101(1):153-71. Coates D and Rayner A. 1985b. Fungal population and community development in cut beech logs. II. establishment via the buried cut surface. New Phytol 101(1):173-81. 79 Coates D and Rayner A. 1985c. Fungal population and community development in cut beech logs. III. spatial dynamics, interactions and strategies. New Phytol 101(1):183-98. Coplen TB, Brand WA, Gehre M, Gröning M, Meijer HAJ, Toman B, Verkouteren RM. 2006. New guidelines for δ13C measurements. Anal Chem 78(7):2439-41. Cornwell WK, Cornelissen JHC, Allison SD, Bauthus J, Eggleton P, Preston CM, Scarff F, Weedon JT, Wirth C, Zanne AMYE. 2009. Plant traits and wood fates across the globe-rotted, burned, or consumed? Global Change Biol 15(10):2431-49. Crowther TW, Jones TH, Boddy L. 2011. Species-specific effects of grazing invertebrates on mycelial emergence and growth from woody resources into soil. Fungal Ecology 4(5):333-341. Dewi M. 2009. Soil microbial community responses to green-tree retention harvesting in coastal British Columbia. Master’s thesis. The University of British Columbia. Dixon R, Brown S, Houghton R, Solomon A, Trexler M, Wisniewski J. 1994. Carbon pools and flux of global forest ecosystems. Science 263(5144):185-9. Dowson CG, Rayner ADM, Boddy L. 1986. Outgrowth patterns of mycelial cord-forming basidiomycetes from and between woody resource units in soil. Microbiology 132(1):203. Dowson C, Rayner A, Boddy L. 1988. The form and outcome of mycelial interactions involving cord- forming decomposer basidiomycetes in homogeneous and heterogeneous environments. New Phytol 109(4):423-32. Evans CS, Dutton MV, Guillen F, Veness RG. 1994. Enzymes and small molecular mass agents involved with lignocellulose degradation. FEMS Microbiol Rev 13(2-3):235-9. Everham EM and Brokaw NVL. 1996. Forest damage and recovery from catastrophic wind. The Botanical Review 62(2):113-85. Fahey T, Stevens P, Hornung M, Rowland P. 1991. Decomposition and nutrient release from logging residue following conventional harvest of Sitka spruce in north wales. Forestry 64(3):289. Federle T. 1986. Microbial distribution in soil—new techniques. In Megusar F and Gantar M (editors). Perspectives in Microbial Ecology. Ljubljana: Slovene Society for Microbiology. p 493-498 Feller M. 2003. Coarse woody debris in the old-growth forests of British Columbia. Env Rev 11(S1):135- 57. Fleming RL, Powers RF, Foster NW, Kranabetter JM, Scott DA, Ponder Jr F, Berch S, Chapman WK, Kabzems RD, Ludovici KH. 2006. Effects of organic matter removal, soil compaction, and vegetation control on 5-year seedling performance: A regional comparison of long-term soil productivity sites. Can J For Res 36(3):529-50. Foster JR and Lang GE. 1982. Decomposition of red spruce and balsam fir boles in the white mountains of New Hampshire. Can.J.for.Res 12(3):617-26. 80 Forest and Range Practices Act. 2010 Oct. Forest planning and practices regulation. < http://www.bclaws.ca/EPLibraries/bclaws_new/document/ID/freeside/12_14_2004> Accessed 2012 April 12. Franklin JF, Shugart H, Harmon ME. 1987. Tree death as an ecological process. Bioscience 37(8):550-6. Frey S, Elliott E, Paustian K, Peterson G. 2000. Fungal translocation as a mechanism for soil nitrogen inputs to surface residue decomposition in a no-tillage agroecosystem. Soil Biol Biochem 32(5):689- 98. Fricker M, Lee J, Bebber D, Tlalka M, Hynes J, Darrah P, Watkinson S, Boddy L. 2008. Imaging complex nutrient dynamics in mycelial networks. J Microsc 231(2):317-31. Fridman J and Walheim M. 2000. Amount, structure, and dynamics of dead wood on managed forestland in Sweden. For Ecol Manage 131(1-3):23-36. Fritze H, Pietikäinen J, Pennanen T. 2000. Distribution of microbial biomass and phospholipid fatty acids in podzol profiles under coniferous forest. Eur J Soil Sci 51(4):565-73. Frostegard A, Bååth E, Tunlid A. 1993. Shifts in the structure of soil microbial communities in limed forests as revealed by phospholipid fatty acid analysis. Soil Biol Biochem 25(6):723-30. Frostegård Å and Bååth E. 1996. The use of phospholipid fatty acid analysis to estimate bacterial and fungal biomass in soil. Biol Fertility Soils 22(1):59-65. Graham RL and Cromack Jr K. 1982. Mass, nutrient content, and decay rate of dead boles in rain forests of Olympic National Park. Can.J.for.Res 12(3):511-21 Harmon ME and Hua C. 1991. Coarse woody debris dynamics in two old-growth ecosystems. Bioscience 41(9):604-10. Harmon ME, Franklin JF, Swanson FJ, Sollins P, Gregory S, Lattin J, Anderson N, Cline S, Aumen N, Sedell J. 1986. Ecology of coarse woody debris in temperate ecosystems. Adv Ecol Res 15(3):133- 302. Hart SC. 1999. Nitrogen transformations in fallen tree boles and mineral soil of an old-growth forest. Ecology 80(4):1385-94. Heilmann-Clausen J. 2001. A gradient analysis of communities of macrofungi and slime moulds on decaying beech logs. Mycol Res 105(5):575-96. Herrmann S and Prescott CE. 2008. Mass loss and nutrient dynamics of coarse woody debris in three rocky mountain coniferous forests: 21 year results. Can J For Res 38(1):125-32. Holub SM, Spears JDH, Lajtha K. 2001. A reanalysis of nutrient dynamics in coniferous coarse woody debris. Can J For Res 31(11):1894-902. 81 Homyak PM, Yanai RD, Burns DA, Briggs RD, Germain RH. 2008. Nitrogen immobilization by wood- chip application: Protecting water quality in a northern hardwood forest. For Ecol Manage 255(7):2589-601. Hughes C and Boddy L. 1994. Translocation of 32P between wood resources recently colonised by mycelial cord systems of Phanerochaete velutina. FEMS Microbiol Ecol 14(3):201-12. Jennings D. 1988. Some perspectives on nitrogen and phosphorus metabolism in fungi. In Boddy L, Marchant R, Read DJ (editors). Nitrogen, Phosphorus and Sulphur utilization by fungi. New York: Cambridge University Press. p 1-32. Johnson DW and Curtis PS. 2001. Effects of forest management on soil C and N storage: Meta analysis. For Ecol Manage 140(2-3):227-38. Johnson EA and Greene D. 1991. A method for studying dead bole dynamics in Pinus contorta var. latifolia‐Picea engelmannii forests. Journal of Vegetation Science 2(4):523-30. Klamer M and Bååth E. 2004. Estimation of conversion factors for fungal biomass determination in compost using ergosterol and PLFA 18: 2 [omega] 6, 9. Soil Biol Biochem 36(1):57-65. Knief C, Lipski A, Dunfield PF. 2003. Diversity and activity of methanotrophic bacteria in different upland soils. Appl Environ Microbiol 69(11):6703-14. Krankina ON, Harmon ME, Griazkin AV. 1999. Nutrient stores and dynamics of woody detritus in a boreal forest: Modeling potential implications at the stand level. . Laiho R and Prescott CE. 2004. Decay and nutrient dynamics of coarse woody debris in northern coniferous forests: A synthesis. Can J For Res 34(4):763-77. Laiho R and Prescott CE. 1999. The contribution of coarse woody debris to carbon, nitrogen, and phosphorus cycles in three rocky mountain coniferous forests. Can J For Res 29(10):1592-603. Lindahl B, Stenlid J, Finlay R. 2001. Effects of resource availability on mycelial interactions and 32P transfer between a saprotrophic and an ectomycorrhizal fungus in soil microcosms. FEMS Microbiol Ecol 38(1):43-52. Lindahl BD, Ihrmark K, Boberg J, Trumbore SE, Högberg P, Stenlid J, Finlay RD. 2007. Spatial separation of litter decomposition and mycorrhizal nitrogen uptake in a boreal forest. New Phytol 173(3):611-20. Lindahl BO, Taylor AFS, Finlay RD. 2002. Defining nutritional constraints on carbon cycling in boreal forests–towards a less phytocentric perspective. Plant Soil 242(1):123-35. Nason, GE. 1989. Dynamics of fertilizer and native nitrogen in a Douglas-fir ecosystem. PhD thesis. The University of Alberta. Zogg GP, MacDonald NW, Zak DR, Ringelberg DB, Pregitzer KS, White DC. 1997. Compositional and functional shifts in microbial communities due to soil warming. Soil Sci Soc Am J 61(2):475-81. 82 Mann L, Johnson D, West D, Cole D, Hornbeck J, Martin C, Riekerk H, Smith C, Swank W, Tritton L. 1988. Effects of whole-tree and stem-only clearcutting on postharvest hydrologic losses, nutrient capital, and regrowth. For Sci 34(2):412-28. Martinez A, Speranza M, Ruiz-Dueñas FJ, Ferreira P, Camarero S, Guillén F, Martínez MJ, Gutiérrez A, Del Río JC. 2005. Biodegradation of lignocellulosics: Microbial, chemical, and enzymatic aspects of the fungal attack of lignin. International Microbiology 8(3):195-204. O’leary W and Wilkinson S. 1988. Gram-positive bacteria. In Ratledge C, Wilkinson SG (editors). Microbial Lipids vol. 1. London: Academic Press:117-201. Olson JS. 1963. Energy storage and the balance of producers and decomposers in ecological systems. Ecology 44(2):322-31. Olsson BA, Staaf H, Lundkvist H, Bengtsson J. 1996. Carbon and nitrogen in coniferous forest soils after clear-felling and harvests of different intensity. For Ecol Manage 82(1-3):19-32. Olsson PA. 1999. Signature fatty acids provide tools for determination of the distribution and interactions of mycorrhizal fungi in soil. FEMS Microbiol Ecol 29(4):303-10. Palviainen M, Laiho R, Makinen H, Finer L. 2008. Do decomposing scots pine, norway spruce, and silver birch stems retain nitrogen? Can J For Res 38(12):3047-55. Pandey K and Pitman A. 2003. FTIR studies of the changes in wood chemistry following decay by brown-rot and white-rot fungi. Int Biodeterior Biodegrad 52(3):151-60. Perez-Garcia J, Lippke B, Comnick J, Manriquez C. 2005. An assessment of carbon pools, storage, and wood products market substitution using life-cycle analysis results. Wood Fiber Sci 37(5):140-8. Pojar J, Klinka K, Meidinger DV. 1987. Biogeoclimatic ecosystem classification in British Columbia. For Ecol Manage 22(1): 119-154. Powers RF, Andrew Scott D, Sanchez FG, Voldseth RA, Page-Dumroese D, Elioff JD, Stone DM. 2005. The North American long-term soil productivity experiment: Findings from the first decade of research. For Ecol Manage 220(1-3):31-50. Powlson D and Barraclough D. 1993. Mineralization and assimilation in soil-plant systems. In Knowles R and Blackburn TH (editors). Nitrogen Isotope Techniques. San Diego: Academic Press. p. 209-242. Prescott CE. 2002. The influence of the forest canopy on nutrient cycling. Tree Physiol 22(15-16):1193- 1200. Rajala T, Peltoniemi M, Hantula J, Mäkipää R, Pennanen T. 2011. RNA reveals a succession of active fungi during the decay of norway spruce logs. Fungal Ecology 4(6):437-448. Rayner ADM and Boddy L. 1988. Fungal decomposition of wood: its biology and ecology. Chirchester: Wiley. p 587 83 Sanchez FG, Tiarks AE, Kranabetter JM, Page-Dumroese DS, Powers RF, Sanborn PT, Chapman WK. 2006. Effects of organic matter removal and soil compaction on fifth-year mineral soil carbon and nitrogen contents for sites across the United States and Canada. Can J For Res 36(3):565-76. Schimel JP and Bennett J. 2004. Nitrogen mineralization: Challenges of a changing paradigm. Ecology 85(3):591-602. Slesak RA, Schoenholtz SH, Harrington TB, Strahm BD. 2009. Dissolved carbon and nitrogen leaching following variable logging-debris retention and competing-vegetation control in Douglas-fir plantations of western Oregon and Washington. Can J For Res 39(8):1484-97. Smithwick EAH, Kashian DM, Ryan MG, Turner MG. 2009. Long-term nitrogen storage and soil nitrogen availability in post-fire lodgepole pine ecosystems. Ecosystems 12(5):792-806. Snetsinger J. 2010. Chief forester’s guidance on coarse woody debris management. Ministry of Forests, Lands, and Natural Resource Operations. Guidance and information from the resource stewardship division. Accessed 2011 November 4. Stevens V. 1997. The ecological role of coarse woody debris: An overview of the ecological importance of CWD in BC forests. Res.Br., BC Min.for., Victoria, BC Work.Pap 30 Strahm BD, Harrison RB, Terry TA, Flaming BL, Licata CW, Petersen KS. 2005. Soil solution nitrogen concentrations and leaching rates as influenced by organic matter retention on a highly productive Douglas-fir site. For Ecol Manage 218(1-3):74-88. Sturtevant BR, Bissonette JA, Long JN, Roberts DW. 1997. Coarse woody debris as a function of age, stand structure, and disturbance in boreal Newfoundland. Ecol Appl 7(2):702-12. Swift MJ, Heal OW, Anderson JM. 1979. Decomposition in terrestrial ecosystems. Oxford: Blackwell Scientific Publications. p 372. Thiffault E, Paré D, Brais S, Titus BD. 2010. Intensive biomass removals and site productivity in Canada: A review of relevant issues. For Chron 86(1):36-42. Tlalka M, Bebber D, Darrah P, Watkinson S, Fricker M. 2008. Quantifying dynamic resource allocation illuminates foraging strategy in phanerochaete velutina. Fungal Genetics and Biology 45(7):1111-21. Tunlid A and White D. 1992. Biochemical analysis of biomass, community structure, nutritional status, and metabolic activity of microbial communities in soil. Soil Biochemistry 7:229-62. Tunlid A, Hoitink H, Low C, White D. 1989. Characterization of bacteria that suppress Rhizoctonia damping-off in bark compost media by analysis of fatty acid biomarkers. Appl Environ Microbiol 55(6):1368-74. Tuor U, Winterhalter K, Fiechter A. 1995. Enzymes of white-rot fungi involved in lignin degradation and ecological determinants for wood decay. J Biotechnol 41(1):1-17. 84 Vanguelova E, Pitman R, Luiro J, Helmisaari HS. 2010. Long term effects of whole tree harvesting on soil carbon and nutrient sustainability in the UK. Biogeochemistry 101(1-3):43-59. Wall A and Hytönen J. 2011. The long-term effects of logging residue removal on forest floor nutrient capital, foliar chemistry and growth of a Norway spruce stand. Biomass Bioenergy 35(8): 3328- 3334. Walmsley J, Jones D, Reynolds B, Price M, Healey J. 2009. Whole tree harvesting can reduce second rotation forest productivity. For Ecol Manage 257(3):1104-11. Watkinson S, Davison E, Bramah J. 1981. The effect of nitrogen availability on growth and cellulolysis by Serpula lacrimans. New Phytol 89(2):295-305. Watkinson S, Bebber D, Darrah P, Fricker M, Tlalka M, Boddy L. 2006. The role of wood decay fungi in the carbon and nitrogen dynamics of the forest floor. In Gadd GM (editor). Fungi in Biogeochemical Cycles. New York: Cambridge University Press. p. 151-81. Weedon JT, Cornwell WK, Cornelissen JHC, Zanne AE, Wirth C, Coomes DA. 2009. Global meta‐ analysis of wood decomposition rates: A role for trait variation among tree species? Ecol Lett 12(1):45-56. Wells J and Boddy L. 1995. Phosphorus translocation by saprotrophic basidiomycete mycelial cord systems on the floor of a mixed deciduous woodland. Mycol Res 99(8):977-80. Wells J and Boddy L. 1990. Wood decay, and phosphorus and fungal biomass allocation, in mycelial cord systems. New Phytol 116(2):285-95. Wells J, Hughes C, Boddy L. 1990. The fate of soil-derived phosphorus in mycelial cord systems of Phanerochaete velutina and Phallus impudicus. New Phytol 114(4):595-606. Wilkinson S. 1988. Gram-negative bacteria. . In Ratledge C, Wilkinson SG (editors). Microbial Lipids vol. 1. London: Academic Press: p. 299-488 Woodward S. and Boddy L. 2008. Interactions between saprotrophic fungi. In: Boddy L, Frankland JC, Van West P (editors). Ecology of Saprotrophic Basidiomycetes. San Francisco: Elsevier. p. 125-141. Yavitt JB and Fahey TJ. 1986. Litter decay and leaching from the forest floor in Pinus contorta (lodgepole pine) ecosystems. J Ecol 74(2):525-45. Yin X. 1999. The decay of forest woody debris: Numerical modeling and implications based on some 300 data cases from North America. Oecologia 121(1):81-98. Zelles L. 1997. Phospholipid fatty acid profiles in selected members of soil microbial communities. Chemosphere 35(1-2):275-94.