Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Fertilizer efficiency and incorporation and soil dynamics in forest ecosystems of northern Vancouver.. 1996

You don't seem to have a PDF reader installed, try download the pdf

Item Metadata

Download

Media
ubc_1996-147339.pdf [ 11.77MB ]
Metadata
JSON: 1.0075277.json
JSON-LD: 1.0075277+ld.json
RDF/XML (Pretty): 1.0075277.xml
RDF/JSON: 1.0075277+rdf.json
Turtle: 1.0075277+rdf-turtle.txt
N-Triples: 1.0075277+rdf-ntriples.txt
Citation
1.0075277.ris

Full Text

FERTILIZER N EFFICIENCY AND INCORPORATION AND SOIL N DYNAMICS IN FOREST ECOSYSTEMS OF NORTHERN VANCOUVER ISLAND by Scott Xiaochuan Chang B. Agronomy, Zhejiang Agricultural University, Hangzhou, P.R. China M.Sc. (Soil Ecology), Institute of Soil Science, Chinese Academy of Sciences, P.R. China A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY IN THE FACULTY OF GRADUATE STUDIES Department of Forestry We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA April 1996 © Scott Xiaochuan Chang, 1996 In presenting this thesis in partial fulfillment of the requirements for an advanced degree at The University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purpose may be granted by the Head of my department or by his or her representative. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of Forest Sciences The University of British Columbia 193-2357 Main Mall Vancouver, B.C., Canada V6T 1Z4 Date April 18. 1996 11 A B S T R A C T To better understand the nutritional problems affecting conifer regeneration on western redcedar (Thujaplicata Donn ex D. Don) - western hemlock (Tsuga heterophylla (Raf.) Sarg.) (CH type) cutover sites, this thesis was aimed to: (1) investigate the effect of salal (Gaultheria shallon Pursh.) competition on tree growth and nutrition, particularly the fate of fertilizer N; (2) detennine the distribution and availability of residual fertilizer N (fertilizer N not lost from the soil after certain time period following application) in forest soils; and (3) study if soil microbial competition for nutrients was also a factor limiting early conifer growth on CH cutover sites. The study sites were located on Tree Farm License (TFL) 25 near Port McNeill, on northern Vancouver Island, which is part of the very wet maritime Coastal Western Hemlock biogeochmatic subzone. One experiment had single-tree plots (1 m radius) planted (in 1987) with western redcedar, western hemlock, and Sitka spruce (Picea sitchensis (Bong.) carr.), with understory (mainly salal) either removed or remaining. Understory removal was initiated just after planting and was maintained throughout the experiment by repeated clipping. Plots were fertilized with 200 kg N ha"1 in the form of (15NH4)2S04 in spring 1991 when the trees were four years old, and destructively sampled in fall 1992. Understory removal significantly increased total height, root collar diameter and biomass growth in all of the tree components (1- and 2-year-old foliage and branches, >3-year-old branches, and various sized roots); however, biomass allocation among the tree components was not altered. Better growth in the understory removal treatment resulted from reduced uptake and immobilization of nutrients by the competing vegetation. I l l Chemical analysis showed that the majority of the 1 5 N in the trees two growing seasons after fertilizer application was in the 1-year-old needles. Understory removal did not affect 1 5 N abundance (atom %) but significantly increased 1 5 N contents (mg plot"1) in the aboveground tree components. Total 1 5 N recovery in trees in the treated plots was 1.9, 8.9, and 2.1 times that in the control plots, for western redcedar, western hemlock, and Sitka spruce, respectively. Understory in the control (salal remaining) plots immobilized 14.8, 24.6, and 13.5% of the applied N for plots planted with western redcedar, western hemlock, and Sitka spruce, respectively. Total recoveries of 1 5 N in the soil-plant systems ranged from 57 to 87%; of the total amount recovered, 59 to 82% was in the soil compartments. Results clearly showed that trees competed poorly with the understory and soil microbial populations for the applied N. A second experiment examined the transformations and extractabilities of the residual N in forest floor sampled 24 hours, 7 months, and 31 months after N fertilizer application. Net mineralization of total and applied N in a 42-day aerobic incubation was greatest in the samples from the 24-hr treatment followed by those from the 31-month treatment (p<0.05), indicating that recently immobilized 1 5 N was more remineralizable. The percentage of applied N found in the total N mineralized (net) ranged from 76.6 to 87.4%, 13.1 to 42.0% and 10.6 to 14.0% in samples from the 24-hr and 7- and 31-month treatments, respectively, showing reduced relative availability of applied N with increased residence time. Both total and applied N in the extractable organic N fraction and in the N flushed after fumigation with chloroform had the following order: 24-hr > 7-month > 31-month treatment. The results confirmed that N fertilizer was being immobilized by the soil organic matter vwthin hours after application and that the immobilized N had a low mineralization potential one growing season after N application. iv The extractability of the residual 1 5 N was studied using 2 M KCL 0.5 M K2SO4, autoclaving, and acidic permanganate (of different strength) extractions and fiunigation-extraction methods. The incorporation of 1 5 N into the classical fuh/ic (FA), humic (HA) and humin fractions was also studied. Greater amounts of total and applied N were extracted from the 24-hr than from the 7- and 31-month treatments (p<0.05), with the difference between the last two nonsignificant. The extracted fractions were always enriched with 1 5 N relative to the bulk soil. A greater (p<0.05) percentage of the total recovered 1 5 N was in the 24-hr than in the other two treatments in the FA and the relationship for the humin fraction was reversed. The results agree with the mineralization studies and showed that the extractability of residual 1 5 N was quickly reduced with increased residence time due to its incorporation into the stable humin fraction of the soil organic matter (SOM); however, residual 1 5 N remained more extractable than the bulk soil N regardless of the length of residence time. A third experiment studied the dynamics of microbial biomass and N in old-growth CH forests, and in 3- and 10-yr-old western redcedar plantations. Three forest floor layers: F (partly decomposed litter material), woody F (Fw) and H (well-decomposed, amorphous organic matter) were sampled four times in May, July, August, and October of 1992. Microbial biomass C and N were relatively constant throughout the sampling period. Microbial C content was in the order: old-growth forests > 10-yr-old plantations > 3-yr-old plantations. Microbial N content was significantly greater in the old-growth forest than in the young plantations, for both F (p<0.001) and H (p<0.05), but was not different between the plantations. Therefore, the hypothesis that the microbial biomass acted as a net sink in the 10-yr-old plantations by immobilizing N into the microbial N pool was rejected. Microbial C/N ratios were greater (p<0.05) in the 10-yr-old plantations than in the old-growth forests and in the 3-yr-old plantations in H and on July 16 in F, indicating that microbial competition for N was probably a factor in the growth decline in the 10- yr-old plantations. Extractable C and N, and mineralizable N were generally higher in the old- growth forests than in the 3-yr-old plantations and higher in the 3-yr-old than in the 10-yr-old plantations. As a result of better nutritional conditions, tree and understory foliage in the 3-yr-old plantations had higher N concentrations and lower C/N ratios than in the 10-yr-old plantations. It was concluded that (1) salal was strongly competitive for the applied fertilizer N in the CH cutover sites; (2) salal competition greatly reduced crop tree growth; (3) N incorporation into SOM further reduced the availability of fertilizer N; and (4) microbial competition for N might have contributed to the N shortage problems in the CH cutover sites. TABLE OF CONTENTS Abstract 1 1 Table of contents vi List of Tables ix List of Figures x Acknowledgements xi Chapter 1. Introduction The setting 1 The objectives 5 Chapter 2. A literature review on biomass partitioning, fertilizer N fate, and the microbial role in soil N cycling I.) The allocation of photosynthate in trees 10 JX) Fate of fertilizer N in forest ecosystems 14 Distribution of 1 5 N in plant-soil systems 16 Movement of 5 N in soil profiles 18 Recovery of 1 5 N fertilizers 20 UJ.) Dynamics of N in forest soils 22 Immobilization of N by soil organic matter 23 Remineralization of nrmobilized N 25 Nitrogen incorporation and extractabilities 28 TV.) Microbial role in soil N cycling 32 Chapter 3. Understory competition effect on tree growth and biomass allocation on a coastal old-growth forest cutover site Introduction 36 Materials and methods 38 Results and discussion 43 Tree height and basal diameter growth 43 Tree biomass accumulation 52 Tree biomass allocation 55 Conclusions 57 Chapter 4. Effect of understory competition on distribution and recovery of 1 5 N applied to a western redcedar-western hemlock clearcut site Introduction 60 Materials and methods 61 Results 67 Nitrogen-15 distribution within trees 67 Nitrogen-15 distribution within understory 67 Nitrogen-15 distribution in the soil-plant system 69 Nitrogen-15 abundance and contents in aboveground components 70 Recovery of 1 5 N 70 Discussion 74 Nitrogen distribution within biomass .; 74 Competition of salal for N 76 Nitrogen budget 79 Conclusions 80 Chapter 5. Transformations of residual 1 5 N in a coniferous forest soil Introduction '. 81 Materials and methods 82 Results 87 Selected properties of humus materials 87 Aerobic incubation 87 Anaerobic incubation 94 Nitrogen flush from fumigation 94 Discussion 97 Chapter 6. Incorporation and extractability of residual 1 5 N in a coniferous forest soil Introduction 106 Materials and methods 107 Results 110 Effect of labelling duration on N extractability 110 Vlll Effect of extraction method on N extractability 113 Extraction with acidic permanganate 114 Organic matter fractionation 117 Discussion 118 Conclusions 125 Chapter 7. Soil microbial biomass and microbial and mineralizable N in a clearcut chronosequence Introduction 127 Materials and methods 129 Results 133 Water content and pH 133 Microbial biomass C and N 137 Extractable organic C and N 140 Mineralizable N 143 Foliar N 143 Discussion 144 Conclusions 153 Chapter 8. General discussion and conclusions Introduction ; 154 Understory competition 155 Residual fertilizer availabihties 159 Microbial competition for N 162 Future research 164 Conclusions 167 References 169 ix LIST OF TABLES 3.1. ANOVA table for the effect of understory competition on tree growth 44 3.2. ANOVA table for the effect of understory competition on biomass accumulation 49 3.3. Allocation (%) of biomass in various components of trees 58 4.1. Selected soil properties of the study site 64 4.2. Within-tree distribution of 5 N after two growing seasons 66 4.3. Distribution of 1 5 N in understory after two growing seasons 68 4.4. 1 5 N abundance and contents in aboveground tree components and ANOVA 72 4.5. Recovery of 1 5 N in the soil-plant system after two growing seasons 75 5.1. Selected properties of humus materials 88 5.2. ANOVA table for N and C mineralization in a 42-day aerobic incubation 89 5.3. Applied N as a % of total N in various measurements 95 6.1. Extractability ratios and N extracted by various extraction methods I l l 6.2. Extractability ratios and N extracted by KMn0 4 of various concentration 115 6.3. Flush of 1 5 N after fumigation and 1 5 N recovered in the humus fractions 124 7.1. ANOVA table for water content, extractable C and N, and microbial C and N 134 7.2. Mineralizable N and pH of forest floor samples collected on August 26 138 X LIST OF FIGURES 1.1. Representative stands of old-growth CH type forest and HA type forest 2 1.2. Location of the study site 7 3.1. Pictures of the 'Control' and 'Treated' plots planted with western redcedar 40 3.2. Height growth of three tree species in various periods 45 3.3. Root collar diameter growth in total and in summer 1992 47 3.4. Biomass accumulated in various aboveground tree components 50 3.5. Biomass accumulated in various belowground tree components 51 3.6. Understory and dead salal and litter mass at the final harvesting 53 4.1. Distribution of 1 5 N recovered in the soil-plant systems 71 5.1 Mineralization of N in samples with N added at different times 92 5.2 Extractable (KC1) organic N at day 0, 28 and 42 in the aerobic incubation 93 5.3 Carbon mineralization in samples with N added at different times 96 5.4 N pool sizes before and after anaerobic incubation in 1 5 N labeled samples 98 5.5 Flush of N after chloroform fumigation in 1 5 N labeled samples 99 6.1. Percentage distribution of applied N recovered in humus fractions 119 6.2. Extractability ratios of humus fractions 120 7.1. 3- and 10-yr-old western redcedar plantations on CH cutover sites 130 7.2. Water (%) in forest floor layers in old-growth and 3- and 10-yr-old forests 136 7.3. Microbial C and N in forest floor layers of CH forests and cutover sites 139 7.4. Microbial C/N ratios in forest floor layers of CH forests and cutover sites 141 7.5. Extractable C and N in forest floor layers of CH forests and cutover sites 142 7.6. Foliar N (%) and C/N ratios of western redcedar and salal 145 7.7. Precipitation in the study area in summer 1992 147 ACKNOWLEDGMENTS It has been an intense program since I started my Ph.D. study here in September 1991. I had to catch up on so much forestry material as well as to begin the thesis research soon after I arrived. It was the timely supervision from my supervisors Dr. Gordon Weetman and Dr. Caroline Preston which made my journey in Forest Sciences a smooth one. I am particularly grateful to Caroline for allowing me to use her laboratory space at the Pacific Forestry Centre, Natural Resources Canada and for looking after the thesis research. The encouragement and guidance from the other members (Drs. Tim Ballard, Hamish Kimmins, and Art Bomke) of my supervisory committee is also much appreciated. A special thank you to John Barker for his time when he was my industrial supervisor in 1994/95. The completion of this thesis would not have been possible without the help of the following people: Kevin McCullough, Tony Trofymow, Frank Portlock, Cindy Prescott, Rob HageL Gary Roke, Barbara Cade-Menun, Tuula Aarnio, Ann van Niekerk, Tom Bown, Richard Winder, and past co-op students working with Caroline. I also would like to thank Paul Bavis, and Cindy Fox and their staff at Western Forest Products Ltd. (WFP) in Port McNeill for help. C. Messier permitted the use of one of his study sites near Port McNeill. The friendship and logistic help from Barbara Kishchuk, Steve Mitchell, Brian Sieben, and Barry White are acknowledged. Financial support was provided by an NSERC (Natural Sciences and Engineering Research Council of Canada) grant for SCFURP (Salal-Cedar-Hemlock Integrated Research Program) with contributions from WFP, McMillan Bloedel Ltd. and Fletcher Challenge Ltd.. WFP also provided field equipment and accommodation for the duration of the study. Partial support was also received from British Columbia Science Council in the form of a G.R.E.A.T. (Graduate Research, Engineering And Technology) award with WFP as the industrial sponsor. Finally, I would like to thank my wife for help in some of the field trips and for her understanding and support of my study. And to my parents and parents-in-law for continued support during my "student career". 1 CHAPTER 1. INTRODUCTION The setting A widespread problem has been observed in the submontane very wet maritime variant of the Coastal Western Hemlock biogeoclimatic zone (CWHvml) (Green and Khnka 1994) on northern Vancouver Island, British Columbia. Plantations of Sitka spruce (Picea sitchensis (Bong.) carr.), and naturally regenerated western hemlock (Tsuga heterophylla (Raf.) Sarg.) and western redcedar (Thujaplicata Donn ex D. Donn), have experienced reduced leader growth and showed symptoms of nutrient deficiencies about 8-10 years after regeneration or planting (Weetman et al. 1989 a and b). The problem was only observed on cutovers of western redcedar - western hemlock (CH type) forests but not on adjacent western hemlock - amabilis fir (Abies amabilis (Dougl.) Forbes) (HA type) forest cutovers, which usually exist side by side with the CH type forests. The CH type old-growth forest (Figure 1. la) is a somewhat open stand dorninated by western redcedar and western hemlock, with a minor component of amabilis fir and a dense understory of salal (Gaultheria shallon Pursh). This type of very old forest is the climatic climax community and is less prone to windthrow and other kinds of catastrophic disturbances (Prescott and Weetman 1994). There is a vast amount of coarse woody debris accumulated on the forest floor. The HA type forest (Figure 1. lb) is densely stocked and dominated by western hemlock and amabilis fir. This type of forest appears to be even-aged and regenerated after a widespread windstorm in 1906. There is little understory growth under the canopy in this type of forest and significantly less coarse woody debris accumulated than in the CH type forests (Keenan et al. 2 3 1993). After clearcutting, HA regeneration is vigorous and does not display the growth stagnation seen on CH cutovers. Both the regeneration problems observed on the CH sites and the differences between the two types of forests attracted the attention of a group of researchers who formed and organized a coordinated research program called SCHIRP - the Salal-Cedar-Hemlock Integrated Research Program. This program involves participation by the forest industry, the provincial and federal governments and universities. The real impetus came from the forest industry, notably Western Forest Products Ltd., which saw the need to do applied research to address the problem of poor growth of regeneration on CH cutovers. As was summarized by Prescott and Weetman (1994), four hypotheses were developed to explain the differences in growth rates and nutrition between the CH and HA type forests: 1) the disturbance hypothesis, which states that the HA forests are the result of frequent and catastrophic windthrow events. Windthrow improves soil conditions and helps to produce dense stands which exclude salal through shading. In the long-term absence of windthrow, the CH forests open up and salal invades and dominates the site; 2) the western redcedar hypothesis, which suggests that western redcedar logs have a slow decay rate resulting in low rates of nutrient cycling and low nutrient availability. Western redcedar is a more suitable species on CH sites due to its greater ability to obtain N under N limited situations and compete with salal; 3) the site-difference hypothesis, that the CH and HA forests are two different plant associations deteraiined by topography; 4) the salal hypothesis, which proposes that salal either has a nutritional advantage over conifer trees or inhibits conifer growth through allelopathy. 4 Over the past ten years or so, studies and trials conducted under the SCHIRP umbrella include: forest fertilization (Germain 1985; McDonald et al. 1994; Thompson and Weetman 1992; Weetman et al. 1989a and b), N mineralization and nutrient availabilities (Messier and Kimmins 1990; Prescott et al. 1993b; Prescott and Preston 1995), soil biology (Battigelli et al. 1994; Xiao 1994), investigation of site factors (deMontigny et al. 1993; Cade-Menun 1995; Keenan et al. 1993), and salal competition (Fraser 1993; Messier 1993; Messier and Kimmins 1991). Prior to the formal establishment of the SCHIRP research group, Lewis (1982) investigated the salal-dominated sites on northern Vancouver Island and initially proposed some of the explanations for differences in tree growth between the CH and HA type forests. In a greenhouse study, Newsome (1985) evaluated the effects of fertilization with N and P on the growth of western hemlock seedling grown on humus material sampled from the low- productivity CH sites. The team effort yielded many results from the decade-long research in the SCHIRP program. Nitrogen and phosphorus were found to be deficient on the CH sites and fertilization with N and P substantially improved tree growth and alleviated nutrient deficiency problems (Weetman et al. 1989 a and b). Other silvicultural treatments such as burning, liming, herbicide application, and higher planting densities were not as effective as fertilization treatments in improving CH site conifer growth (Prescott and Weetman 1994). However, it is certain that declining nutrient availability, after an initial nutrient flush is consumed by growing biomass following clearcutting and slash-burning, is one of the major factors contributing to regeneration problems on CH cutover sites. A substantial amount of N in the forest ecosystem is immobilized by salal biomass. Aboveground biomass of salal was found to increase from 1058 kg ha"1 two 5 years after clearcutting and slash-burning, to 4078 kg ha"1 after 8 years (Messier and Kimmins 1991). Therefore the availability of nutrients was greatly reduced by salal competition. The CH forest floor had smaller concentrations of total and extractable N, and less N mineralized, than that of the HA forest floor, and lower amounts of total and extractable P were found in the CH than in the HA Utter layer (Prescott et al. 1993b). CH forests had slower nutrient cycling but higher nutrient use efficiency (Prescott and Weetman 1994). Mycological studies showed that salal may be able to use simple organic forms of N and thus has an advantage in surviving and expanding in a nutrient limited situation; also ericoid mycorrhizal fungi of salal can interfere with the estabhshment and function of the ectomycorrhizal fungi of hemlock (Xiao 1994). The objectives The causes of low N supply in old-growth CH forests have been hypothesized to be determined by three processes (Prescott and Weetman 1994): 1) accumulation of organic matter (mor humus type) in the cool moist climate in the absence of disturbance, the latter being due to the resistence of the somewhat open cedar stand to blowdown; 2) cedar Utter has a low nutrient content because of high N use efficiency of cedar (greater nutrient resorption during senescence); and 3) N mineralization in the forest floor is inhibited by high tannin contents in salal Utter. Those conclusions supported the "salal hypothesis". In spite of the work that has been done under the SCHTRP program, as has been summarized above, there are gaps in our knowledge regarding the mechanisms governing N availabiUty and tree nutrition on the CH cutover sites, where nutrient deficiency problems and poor early conifer growth have been widely observed. A fuU interpretation of the phenomena 6 observed in the field is still difficult. Detailed information was lacking on the fate of applied fertilizer N and the influence of salal competition on fertilizer N recovery, on the interactions of nutritional properties of soils and other biotic and abiotic factors. The overall objectives of this thesis were to investigate the mechanisms underlying the low nutrient availability problem in the forest floor of the CH type forests, and to quantify the effect of competition between the tree species and understory vegetation on tree growth and fertilizer N fate. Results from such studies are useful for understanding the pathways of N cycling and C sequestration in the CH forests and may aid the interpretation of field observations. Conducted on the SCHIRP sites (see the location map, Figure 1.2) on northern Vancouver Island, this study had the following specific objectives: a. ) To examine the effect of salal understory removal on tree height and diameter growth and biomass accumulation b. ) To investigate the effect of understory removal on the distribution of biomass among different year's growth of foliage, branch and roots of different sizes under field conditions c. ) To study the distribution and fate of labeled N in the plant-soil systems with, and without, understory competition d. ) To test if different tree species have different capacities to compete with understory vegetation for N supply e. ) To investigate the effect of labeling duration on net N mineralization, microbial N pool size and long-term fertilizer N availability Figure 1.2. Map of Vancouver Island and location of the study site 7 8 f. ) To determine the effect of labeling duration on the extractability of both total and applied N using various extraction methods g. ) To quantify the changes of microbial biomass C and N and mineralizable N in the forest floors of a CH clearcut chronosequence (uncut old-growth forests, 3 - and 10-year-old plantations) h. ) To test the hypothesis that microbial biomass acts as a net sink in the 10-year-old plantations by immobilizing more N into the microbial N pool than in the 3-year-old plantations Although similar knowledge gaps exist for the HA type forests as well, this thesis was focused on the CH cutover sites. The following hypotheses were proposed to addess the specific objectives outlined above: 1. ) Understory competition from salal will reduce the height, diameter and biomass growth of regenerating trees 2. ) Biomass partitioning to current year foliage and branch and fine roots will be increased in plots with understory compared to plots without understory as a strategy to survive the strong competition from salal 3 . ) Understory competition will greatly reduce the uptake of fertilizer N by crop trees thus reducing the fertilizer use efficiency 4. ) The mineralization potential of residual 1 5 N decreases with the increasing residence time 5. ) The extractability of residual 1 5 N decreases with the increasing residence time 6. ) Applied N is quickly incorporated into stable SOM fractions 9 7.) Microbial population also affects tree growth by competing for the available N in the soil; microbial biomass acts as a net sink in the 10-year-old plantations (where nutrient was deficient) by immobilizing more N into the microbial N pool than in the 3-year-old plantations (where no nutrient deficiency was observed) 10 CHAPTER 2. A LITERATURE REVIEW ON BIOMASS PARTITIONING, FERTILIZER N FATE, AND THE MICROBIAL ROLE IN SOIL N CYCLING I.) The allocation of photosynthate in trees In ecosystems under stress, biomass allocation strategies among different parts of a tree might be an effective way of alleviating the stress (Grime 1979). The description of biomass allocation (the proportion of total biomass stored in each organ/compartment) was first proposed by Harper and Ogden (1970) as a way to study those strategies. Our understanding of the effect of competition (stress) on the allocation of growth is very important for the construction of biological growth models (Nilsson and Albrektson 1993). Biomass allocation patterns have been the subject of many studies (Harper and Ogden 1970; Pearson et al. 1984; Nilsson and Hallgren 1993; Haggar and Ewel 1995). Nilsson and Albrektson (1993) summarized that the allocation of photo synthates in plants is prioritized in relation to the competitiveness of the environment. The order of priority is: 1) maintenance of respiration; 2) fine root and foliage production; 3) seed production; 4) length growth of stems, branches and roots; and 5) diameter growth. However, this order of priority may change depending on the degree of competitiveness. Biomass allocation is not to be confused with nutrient allocation or energy allocation (although energy allocation can generally be inferred from biomass allocation), because resource allocation patterns of biomass and those of energy and chemical elements may not be necessarily the same (Abrahamson and Caswell 1982). Their study on three semelparous Verbascum thapsus (common mullein) and five iteroparous solidago L. (Compositae) (goldenrod) species 11 revealed that the mineral elements examined were allocated differently than biomass (Abrahamson and Caswell 1982). Therefore, nutrient allocation usually cannot be inferred from biomass allocation measurements, and vice versa. They also argued that biomass is a reasonable variable by which to measure allocation patterns, because biomass integrates the entire suite of physiological processes of plants. In studying biomass allocation in trees of different classes (dominant, codominant, intermediate and suppressed) in 16-year-old Scots pine (Pinus sylvestris L.) stands planted at different initial densities, Nilsson and Albrektson (1993) reported more biomass being allocated to stem wood and less to branches and needles in the suppressed trees than in dominant trees, in contradiction to the order of priority listed above. They explained that the result reflects the trees' priority to survive overtopping in a dense stand. However, higher fine root biomass was observed in the high density stand than in the low density stand, in agreement with the order of priority. Barclay et al. (1986) also reported a case contradictory to the order of priority for growth allocation in which thinning (released stress) of 34-year-old Douglas-fir (Pseudotsuga menziesii (Mirb.) Franco) stands decreased the proportion of biomass allocated to wood and bark but increased the allocation to foliage and live branches. Tree biomass allocation is known to vary with stand structure and site characteristics. In six lodgepole pine (Pinus contorta spp. latifolia (Engelm. ex Wats.) Critchfield) forests in southeastern Wyoming, total biomass ranged from 123 to 180 Mg ha"1 and on average 61, 7, 6, 20, and 6% of the total biomass was in boles, branches, foliage, woody roots, and fine roots, respectively (Pearson et al. 1984). Stand density was found to be a more important factor than site characteristics in controlling biomass distribution in different organs, because biomass 12 distribution was similar in stands of similar density on contrasting sites, but different in stands of different densities on the same site. The root/shoot ratios in two dense stands were found to be greater than in four more open stands. Pearson et al. (1984) argued that high intraspecific competition may have an effect similar to low site potential. The effect of site characteristics on tree growth was also studied by Keyes and Grier (1981), and Kurz (1989a). Keyes and Grier (1981), in studying above- and belowground net primary production in 40-year-old Douglas-fir stands in Washington state, U.S.A., reported that aboveground net primary production was 13.7 and 7.3 t ha"1 on high and low productivity sites, respectively; the corresponding belowground dry matter production was 4.1 and 8.11 ha"1, respectively. Only 8% of the total stand dry matter production was in fine roots on the high productivity sites in contrast to greater than 36% on the low productivity sites. They suggested that the need for a greater investment in the fine roots on the low productivity sites resulted in the differences in biomass allocations. Kurz (1989a) also found that the allocation of the total production to belowground stand components was reduced with increasing site index. Similar to the effect of site characteristics on C allocation, fertilization and resource availabihty alterations change biomass partitioning in forest ecosystems (Proe and Millard 1994; Grier et al. 1984; Persson 1983; Axelsson and Axelsson 1986). Proe and Millard (1994) reported that, six weeks after receiving balanced nutrient solution, the partition of growth favored roots of plants in a low relative addition rate (RAR) treatment in comparison with a high-RAR treatment in a greenhouse incubation study involving 2-year-old Sitka spruce seedlings. Grier et al. (1984) found that fertilization increased leaf biomass per tree relative to control on sites having low N availability (the response decreased with increasing N availability) in Douglas-fir stands of varied 13 ages in Washington state; however, the proportionate allocation of biomass was not evaluated. Both site and fertilization effects on dry matter allocation have shown that reduced biomass allocation to fine roots in favor of aboveground production is one of the most important mechanisms for greater aboveground productivity following improved nutrition (Axelsson and Axelsson 1986). The availability of supplies (resources) to the growing organs, which determine plant growth and the growth of different structures, is partly a function of competition for such supplies. Environmental changes may alter the growth rates of different organs to different extents and consequently alter the overall pattern of dry matter distribution (Harper and Ogden 1970). Working on an annual plant, Senecio vulgaris L., Harper and Ogden (1970) found that the proportion of total biomass in roots increased with increasing stress (decreasing soil volume) while the proportion in stems and seeds declined. Traditionally, the allocation of biomass among various plant organs largely depends on the nature of the limiting factors (Abrahamson 1979). Abrahamson and Gadgil (1973) hypothesized that if light is the limiting resource, plants respond by 1) allocating more biomass to stems which results in taller plants to avoid shading; or 2) building more leaves to intercept more light. More recent work showed that root/shoot ratios change as a result of imbalances in carbon-nutrient (especially N) supply rather than the result of the plant's strategy to adjust to the stress (CE. Prescott 1995 pers. comm.). Resource allocation can be further affected by (Abrahamson 1975, 1979): 1) life cycle (annuals, biennials, and perennials); 2) seasonal cycles of species; 3) growth form of the plants; 4) the microclimate of the individual plant; 5) life span of 14 the plant; and 6) competition interactions. Differences in energy allocation patterns maybe important in the continued coexistence of species in a community (Hickman 1977). It must be pointed out that the majority of investigations of site productivity and sihicultural effects were concerned only with aboveground production (Kurz 1989b), due mainly to the difficulties in quantifying the belowground production, especially that of the fine roots (Persson 1983). Therefore, there is a paucity of knowledge concerning the effect of biotic and abiotic environmental factors and silvicultural treatments on belowground biomass production and allocation. Carbon allocation belowground can constitute 50% or more of the total C- budget (Agren et al. 1980); moreover, fine root production can account for about 40% of total net production (cf. Vogt et al. 1980). Vogt et al. (1980) showed that, during the winter months, mycorrhizal roots constitute the highest proportion of the weight of fine roots (29%), possibly a response of trees to higher levels of photosynthate translocation to the root systems during those months. Therefore, a re-evaluation of the existing models for C allocation may be necessary when more data on belowground production are available. H.) Fate of fertilizer N in forest ecosystems The study of N cycling in coniferous ecosystems has a long history, especially in North America and Europe. A thorough understanding of the processes of N cycling in forest ecosystems is important to the improvement of forest management and for the resolution of environmental issues of concern to human life. It has long been known that N is the nutrient element which most frequently limits the growth of forest trees. In the Scandinavian countries, the use of N fertilizers in practical forestry 15 had attained significant proportions as early as in the 1960's (Nommik and Moller 1981). In North America, their experience is essentially being followed. Forest fertilization is becoming an important silvicultural tool to increase forest productivity in British Columbia since commercially available timber from natural old-growth stands has been significantly reduced and the land base available for future commercial forest management has been greatly decreased. From both economic and environmental protection points of view, it is important that fertilization methods used be those that can maximize the assimilation of N applied to forest ecosystems. Therefore, one of the main objectives of forest fertilization research is to obtain information about the fate of fertilizer N applied to forest ecosystems (Nonunik; and Larsson 1989). There are two commonly used methods available to investigate the fate of fertilizer N added to forest ecosystems. The first one is the "difference method" which involves the determination of the quantity of N in soil and trees of both fertilized and control plots (Nommik and Popovic 1971). The excess of N in fertilized plots is considered to be derived from the N source added. The disadvantage of this method is that in reality some of the assumptions made in establishing the method may not always be met and that results obtained may be unreliable (Olson 1979). Soil N variability can be so high that statistical uncertainty about the difference between fertilized and control plots often enables no clear conclusion about most of the N applied. On the other hand, the second method, the 15N-tracer technique, has several advantages. Firstly, the 15N-tracer method may be considered a direct method. It permits direct identification of the labelled N as it goes through different processes. Secondly, the 5N-tracer technique 16 provides higher sensitivity and precision than the difference method (Nommik and Larsson 1989). Although the 1 5 N tracer method offers a range of advantages over the conventional method, only a limited number of studies have been conducted in forest fertilization researches which involve the use of ' 5 N isotope due partly to the fact that l 5 N is still a relatively expensive chemical and the analysis of ' 5 N requires expensive instruments. Distribution of15N in plant-soil systems The uptake of ' 5 N fertilizer by trees is generally very fast. In a study involving the use of calcium nitrate and ammonium nitrate in Scots pine in three growing seasons, 62% of the total uptake took place in the first growing season while 98% was taken up in the first two growing seasons (Nommik and Larsson 1989). Nommik and Larsson (1989) in studying N accumulation in young Scots pine stands (29- to 43-year-old) in Sweden found that when fertilizers were applied in spring or early in the summer, the proportion of total N that was labelled was generally markedly higher in current needles than in older ones. When the application was made at a later stage of current needle development, the distribution pattern of fertilizer-derived N in foliage was quite different. This indicated that the mobility of assimilated N may be influenced by the development stage of current tissues when fertilizer was added. By the second and third growing seasons after fertilizer application, the differences in 5 N distribution in foliage had been mostly leveled out. In their study, 50-54% of the accumulated N in the aboveground components of the trees was found to be distributed in the foliage. The proportion of labelled N in branches ranged from 21 17 to 24%, while in stems it averaged 25%. In the soil profile, 55% of the total recovery in soil was located in the L, F, and H horizons. The quantity of 1 5 N in foliage and in branch samples after the first growing season following fertilization usually increases from old to current growth (Pang 1985; Nommik 1966; Heilnaan et al. 1982b). However, Pang (1985) found that after the second growing season following fertilization, the current growth contained less 5 N than in the previous year's growth. This agrees with data obtained by Mead and Pritchett (1975b), who found that after the first growing season the highest proportion of added labelled N in the total N fraction of the tree occurred in parts which had the highest metabolic activity - i.e. foliage, buds, recent branches, recent wood and inner fine roots. In the second growing season, the current flush was found to have a slightly lower proportion of labelled N in the total N fraction than the 1-yr-old foliage. This and other evidence indicated that ' 5 N in the current foliage at the end of the second growing season appeared to be derived mainly from translocation of N from non-current foliage rather than from the residual 1 5 N in the soil (Mead and Pritchett 1975b; Hulm and Killham 1990). In a recent publication, Mead and Preston (1994) showed that from year 1 to 8 after N fertilization, 14%of ,5N was retranslocated from the lower crown to the upper crown. Interestingly, Nommik (1966) reported higher 1 ? N per cent in current needles than in older needles even after the third growing season following fertilization. He proposed that there was probably new uptake of residual labelled N from the soil or from the root tissue as internal 1 5 N re-distribution cannot maintain a higher I 5 N per cent in current needles than in older needles for so long. The distribution of 1 ? N percentage in needles and shoots showed a tendency to increase with height in the crown. In bark and wood, the corresponding figures increased from the base 18 to the top of the stem (Melin et al. 1983). Mead and Pritchett (1975b) found that the "N atom per cent in branches differed with crown position and treatment. At a low rate of N application (56 kg N ha"1), differences in 1 5 N atom per cent among categories of branches were not significant, but at high-N treatment (224 kg N ha"1) all categories, except recent branch-wood from the upper and lower crowns, differed one from another. An analysis of the stems of two sample trees which were felled about 15 months after the N fertilizer application showed that the 5 N per cent in the wood of different annual rings (from 2nd to 8th annual ring) decreased as the age of the rings increased (Nommik 1966). Movement of N in soil profiles In order to assess the transformation and vertical distribution of different sources of fertilizer N in forest soil under field conditions, Nommik and Popovic (1971) applied 5 N labelled fertilizer (100 kg N ha"1) to laterally isolated micro-plots on a hon-humus Podzol. After the experiment was started in May, 1968, samples were taken four times in the growing season in 1968 (July, August, September and November) and once in May 1969. The vertical distribution of labelled N in the soil profile on the first sampling occasion revealed that more than 47% of the added urea-N could be recovered in the L horizon. The corresponding figures for ammonium sulphate and calcium nitrate were 30% and 19%, respectively. Except for the calcium nitrate treatment, less than half of the labelled N recovered in the L horizon was in inorganic forms. In the H horizon and mineral soil (0-5, 5-10, and 10-20 cm) layers, the recovery of N was in the order: calcium nitrate > ammonium sulphate > urea. On the subsequent sampling dates, the total recovery for labelled N decreased markedly. The tendency towards accumulation of urea-N in 19 the L horizon was still very clear. The higher mobility of N from calcium nitrate and ammonium sulphate was evident from the relatively high levels of labelled inorganic N in the mineral soil. By the last sampling date, most of the N from calcium nitrate had been lost from the soil sampled to a depth of 20 cm. The N from urea and ammonium sulphate was more easily immobilized than that from calcium nitrate. By the use of the 1 5 N tracer technique, it was clearly demonstrated that the added ammonium-N was successively incorporated into the stable humic fraction of the soil organic matter. Pang (1985) showed that ammonium was more quickly immobilized into the organic N fractions of the soil than nitrate. The high retention rate of urea-N in the L horizon was to a large extent due to the intensive microbial immobilization under the increased pH produced by urea hydrolysis and to a lesser extent due to increased cation exchange capacity of the organic material and thus retaining more urea-N (Nommik and Popovic 1971). However, Foster et al. (1985) studied urea incubation with sterilized L and F materials or with natural L and F materials under controlled pH conditions. They concluded that hydrolysis of urea creates a chemical environment in forest floor conducive to a rapid chemical fixation of fertilizer N by organic matter and a slower microbial inmiobilization of fertilizer N over a longer period. They also reported a low recovery of calcium nitrate-N on the last sampling date and the relatively rapid downward displacement of ammonium sulphate N in the non-nitrifying forest soil, indicating a considerable risk of leaching loss. Mead and Pritchett (1975a) thought that the rapid depletion of labelled-N from the Utter + soil in the first 12 weeks after N appUcation was probably a result of leaching and gaseous losses, as weU as uptake by the trees. 20 In a lysimeter study in a recently thinned radiata pine stand on a pumice soil in New Zealand, Worsnop and Will (1980) showed that fertilizer N had reached all three soil layers (0- 10, 10-20, and 20-30 cm, respectively) sampled by the end of the first week. Thus fertilizer movement in the soil was very quick after application. Of the N present in the L horizons 1 and 2 weeks after the fertilizer was applied, nearly 20% had its origin from the fertilizer. Particularly in the F and H horizons, fertilizer N was discovered to have become interchangeable with the original N present in the litter as was demonstrated by the fact that after week 2 when the total N content of these horizons returned to pretreatment levels there was still about 15% fertilizer derived N present. Recovery of Nfertilizers The amount of applied fertilizers that can be recovered in trees is usually quite small. In the study of Bjorkman et al. (1967), 20.8% and 33.3% of the added N were found to be recovered by vegetation (includes above- and belowground parts of both trees and ground vegetation) for ammonium sulphate and calcium nitrate, respectively, when quantified one growing season after application to a 15-year-old Scots pine stand. The corresponding recoveries in soil were 58.5% and 56.3%, respectively. A total recovery of 8-10% by trees was more common (Mead and Pritchett 1975b; Knowles and Lefebvre 1972), indicating that a very small amount of the added N fertilizers would contribute to the increased growth of crop trees. Even lower values were reported by Preston et al. (1990) from work on an 11-year-old lodgepole pine at Spillimacheen in the British Columbia interior where only 5.3 and 1.9% of applied fertilizer N were recovered after one growing season when N was applied as NH 4 N0 3 21 and NH 4 I 5 N0 3 (both at 100 kg N ha"1), respectively. In a subsequent study, it was found that little additional ' 5 N was taken up by plot trees, but there was continuing uptake by understory, when quantified 7 years after the first study (Preston and Mead 1994). The majority of applied N fertilizer is normally found to be immobilized in the soil profile, especially in the humus layers, as has already been discussed in the previous section. Interestingly, Heilman et al. (1982a) reported a rather low recovery of 1 5 N in the soil when fertilizer was applied as urea. By the end of second growing season, only 38±6% of the fertilizer N was detected in the soil. Recovery in soil averaged 44% after one year, which is well below the 78% recovery reported by Popovic and Nonimik (1972) in an iron-humus Podzol in Sweden. On the other hand, fertilizer N recovered in total tree appeared to be much higher (varying from 25 to 36% of the application) in the Pacific Northwest Douglas-fir stands (Heilman et al. 1982a) than in other stands elsewhere (Mead and Pritchett 1975b; Nommik 1966). The recovery of fertilizer N was often found to be affected by many factors. In the study carried out by Nommik and Larsson (1989), the effect of N sources, application rates, date of application and particle sizes of the fertilizer material on the recovery of fertilizer N in Scots pine was demonstrated. Results showed that recovery in trees was high for ammonium nitrate (39%) and calcium nitrate (37%) and low for urea (27%). The total recovery in both soil and trees varied between 46 and 84%. The highest recovery was in urea-treated plots primarily due to the high degree of immobilization of urea N in the organic L, F, and H horizons of the soil. Application rate had affected the recovery of labelled N in both the trees and soil. The greater the application rate, the smaller the percentage of N recovered. For treatments with calcium nitrate, the results confirmed that the total recovery dropped substantially when application was 22 postponed from June to late July. This result was believed to be affected by the precipitation pattern of the growing season. Granule size had no definite effect on the recovery of fertilizer N in different components. Variability in fertilizer recovery appeared to be related to differences in rainfall patterns shortly after fertilization (Heilman et al. 1982b). To some extent, the trees appeared to overcome the delayed availabihty of N from spring application (due to the delayed rainfall) by more rapid uptake of fertilizer N. Fertilizer accounted for in soils and trees at the end of two growing seasons averaged 68% of the application. Lowest recovery (52%) corresponded to the treatment which had lowest rainfall. The work of Melin and Nommik (1988) done on a 50-year-old mixed coniferous stand consisting of Scots pine and Norway spruce (Picea abies (L.) Karst.) showed that total recovery in the stands varied from 76 to 92% of the fertilizer N applied (150 kg N ha"1). Calcium nitrate, which showed the highest accumulation of labelled N in the stand (44%), had the lowest recovery in the soil (28%). Contrarily, the urea source of N, which showed the lowest accumulation in the stand (20%), had the highest recovery in the soil (62%). The highest recovery figure applied to treatment with a split dose of ammonium nitrate. For a low dose of ammonium nitrate (50 kg N ha"1) the recovery was close to 100%. This indicates that management prescriptions should consider to incorporate the best application methods to improve the efficiency of fertilizers in forest ecosystems. III.) Dynamics of N in forest soils 23 The cycling of N in forest soils is affected by many factors, such as the turnover of N through mineralization-immobilkation (He et al. 1988), decomposition of Utter and soil organic matter (Aber et al. 1990), relative amounts of N input (rainfall, biological N fixation, etc.) and output (demtrification, leaching, etc.), and biotic and abiotic factors affecting those processes (Mahendrappa et al. 1986). Nitrogen mineralization and immobilization are by far the most important and most studied processes affecting N availabiUty in forest soils (Adams and AttiwiU 1986; Carlyle and Malcolm 1986; Frazer et al. 1990; Zak et al. 1989). Immobilization ofNby soil organic matter In cold temperate and boreal coniferous forests, large quantities of organic matter are usuahy accumulated in the forest floor (Vogt et al. 1986). This organic layer (the forest floor) not only contains a significant proportion of the total ecosystem N (Cole 1981) but it can also act as a net sink for inorganic N (Hart and Firestone 1991). Although the forest floor supphes N (acting as a net source for inorganic N) through the mineralization of organic N (van Cleve et al. 1986), the forest floor in coniferous forests is generaUy a site for net immobilization due to its long residence time and high carbon-to-nitrogen ratios (Edmonds 1987). This immobilization and the subsequent tniavailabiUty for tree uptake of fertilizer N appUed to such forest ecosystems have been of major concern to forest managers and researchers. The relative recovery rate of 1 5 N in soil and vegetation is also affected by season of fertilizer apphcation (Heilman et al. 1982a), N status of the site before N addition (Heilman et al. 1982a), rainfall after fertilizer apphcation (Heilman etal. 1982a; Mead and Pritchett 1975a), 24 fertilizer application rate (Mead and Pritchett 1975b), and the complex interactions of physical, chemical and biological factors (Overrein 1972). In Europe, Melin et al. (1983) found that two years after a 120-140 year old pine stand was fertilized with 100 kg N ha"1 of ammonium nitrate, only 10% of fertilizer N was in aboveground tree biomass, with more than 46% recovered in the soil. In some 29- to 43-year- old Scots pine stands, Nommik and Larsson (1989) reported 1 5 N recoveries in soil ranging from 17 to 80% of the total applied and the highest immobilization rate was observed in stands treated with urea. Nommik and Popovic (1971) reported that 76% of applied urea-N was recovered from the forest floor of a mature Scots pine stand twelve months after fertilizer application. Of the three forms of fertilizer N applied (calcium nitrate, ammonium sulphate and urea), urea-N had the lowest mobility and tended to accumulate in the L and H horizons. In North America, studies showed that 1 5 N recovery in soils of urea-N (224 kg N ha"1) applied to Douglas-fir stands two growing seasons after application averaged 38% of the total applied (Heilman et al. 1982). Mead and Pritchett (1975b) documented a recovery of 1 5 N in the soil from 17.9 to 29.6% (including soil and fitter layer) while the total recovery of the ecosystem varied from 44.5 to 54.2% in a slash pine (Pinus elliottii Engelm. Var elliottii) forest. Foster et al. (1985) measured immobilization of 15N-labelled urea in a 45-year-old jack pine (Pinus banksiana Lamb.) forest in a laboratory experiment and found that 25% of the added N was immobilized by the forest floor three months after the fertilization. The authors found that the amount of N fixed chemically by the forest floor materials was greater than the amount of N immobilized in microbial biomass. In a field study on a 45-year-old jack pine stand, fertilizer N 25 irnmobilized in the soil (forest floor and mineral soil) was 36% of the applied amount (Morrison and Foster 1977). More information on the immobilization of fertilizer N by forest floor and mineral soils can be found in Preston et al. (1990) and Binkley and Hart (1989). Studies on other forest types not summarized here also showed that a substantial amount of fertilizer 1 5 N was immobilized in the forest soil (Worsnop and Will 1980; Hulm and Killham 1990). The low recovery of fertilizer derived N in the crop trees and the high rate of immobilization of fertilizer N suggests that the dynamics of mineralization and uptake of the bulk of the immobilized fertilizer N in the forest floor should be emphasized (Hulm and Killham 1990; Ledgard et al. 1992). Remineralization of immobilized N Nitrogen fertilization of temperate forests where N is limiting is becoming a common silvicultural practice to increase timber yield (Chappell et al. 1991). Apart from increasing tree growth, fertilization may also affect the activities of soil animals and microorganisms (Soderstrom et al. 1983) and thus affects the remineralization and long-term availability of immobilized N. Evidence regarding the effect of fertilization on microbial activities has been contradictory so far. Some studies found that fertilization increased microbial activity (Fessenden et al. 1971; Salonius and Mahendrappa 1975; van Cleve and Moor 1978). Others reported decreasing activities after fertilization (Kowalenko et al. 1978; Baath et al. 1981; Foster et al. 1980). The effects of fertilization on microbial activities are complex, depending on the forest ecosystem type, type of fertilizer applied, and rate of apphcation, among other factors. The result of 26 increased or decreased microbial activity may also have two possible effects on subsequent nutrient availabilities. On one hand, stimulation of soil microbial activity may reduce the availabihty of nutrients by immobilization in microbial tissue (van Cleve and Moore 1978). Alexander (1977) stated that increased biological activity is usually accompanied by greater immobilization of N as protoplasmic turnover is enhanced. On the other hand, stimulation of biological activity may result in more mineralization of organic matter and turnover of nutrients (van Cleve and Moore 1978). Thus, site specific investigations are required to support any decision making. Studies of the effect of N fertilization on N transformations in the soil have often shown increased N mineralization (Martikainen et al. 1989; Popovic 1977). This is often accompanied by decreased C mineralization. Consequently, the relationship between C and N mineralization after N fertilization generally appears to be poor (Johnson et al. 1980; Martikainen et al. 1989). Bosatta and Berendse (1984) modified a model developed by Bosatta and Staff (1982) to explain the dynamics of N after N addition (perturbation) to forest soils. They (Bosatta and Berendse 1984) attempted to explain the retention of fertilizer N in forest soils using the modified model. By defining energy (C)-deficient and nutrient (N)-deficient systems, Bosatta and Berendse (1984) predicted that N nmeralization-immobihzation and C mineralization dynamics in N-deficient system would oscillate around steady-state conditions after C or N additions. On the other hand, addition of C or N to C-deficient systems would lead the C and N mineralization rates to shift from the steady-state status before returning to the steady-state without oscillation. White et al. (1988b) evaluated the model in a field perturbation experiment in two Douglas-fir stands in New Mexico and found that their data supported the model's predictions. 27 Addition of inorganic N usually increases the mineralization and availabihty to plants of soil N, an effect sometimes called a ' jjriming" effect (Fried and Broeshart 1974), or "added nitrogen interaction", or ANI (Jenkinson et al. 1985). In agricultural soils, Hart et al. (1986) showed that in a pot experiment, fertilizer N increased the uptake of native soil N in samples from a fallow site with low C content (1.6%) and in a grassland soil with high C content (3.8%). The ANI in the fallow soil was smaller and was found to be the result of pool substitution (labelled fertilizer N standing proxy for unlabelled inorganic soil N) and N addition did not affect gross or net mineralization of soil N. In the grassland soil, in addition to pool substitution, N fertilization increased soil N mineralization. In a corresponding field experiment, ANI was not observed because the fertilizer and soil N did not mix and pool substitution did not occur. Fertilization with ammonium sulphate has been shown to increase the NH4 -N and NO3-N pools, and nitrification and N-mineralization potentials in two Douglas-fir stands in New Mexico (White et al. 1988a). However, Polglase et al. (1992) reported that specific N rnineralization (the amount of N mineralized as a percentage of the organic substrate present initially) was little affected by fertilization in a laboratory study (42-d aerobic incubation) of soil from young slash pine and loblolly pine (P. Taeda L.) plantations growing on Spodosols in Florida, but was significantly increased in the field measurements. In forest raw humus of boreal forests, laboratory incubation up to 90 days showed that there was no remineralization of immobilized N (non-extractable in 1 M KC1) at incubation temperatures of 4, 12, and 20 °C, when ammonium was added as the label (Overrein 1967). However, vvdthin 3 days of incubation, significant net remineralization of immobilized N occurred 28 at the highest temperature (20 °C) used when the isotope added was nitrate N. Therefore, mmeralization-immobilization behavior differs among N species. Addition of an energy source such as glucose can dramatically change the course of soil processes, i.e., from net mineralization to net immobilization (Jones and Richards 1978; Schimel et al. 1992; Johnson and Edwards 1979). Addition of a readily available energy source removes the C limitation to microbial growth leading to net immobilization of inorganic N (Schimel et al. 1992). Studies mainly conducted on agricultural soils showed that the availability of organically incorporated 1 5 N gradually decreased over time due to (cf. Chichester et al. 1975): 1) a greater susceptibility of most recently immobilized N to mineralization; 2) the reversion of the recently incorporated N to more stable organic forms; and 3) the accumulation of resistant fractions of microbial tissues when the more decomposable materials are consumed. Shen et al. (1989), also working on agricultural soils, reported that recently immobilized N was about 7 times more mineralizable than native soil organic N and explained that there was so much more native than labelled inorganic N after one growing season because the soil contained 100-200 times more native than labelled organic N. More labelled N (which was non-symbiotically fixed in soil) was recovered in hydrolysable forms which seemed to support that recently fixed N was more available for plant uptake (Azam et al. 1988). Nitrogen incorporation and extractabilities The incorporation of applied N into soil and SOM affects the form and distribution of immobilized N and its availability to plant uptake. Immobilization studies generally consider the 29 soil organic matter as a whole, while incorporation studies emphasize the distribution (over time) of N into different pools. Four pools of soil organic C and N are widely recognized (cf. Strickland et al. 1992): 1) a debris pool, which includes Utter and roots; 2) an active pool, which includes microbial cells and metabohtes; 3) a medium stable pool with turnover time of decades to several hundreds of years; and 4) a stable pool with turnover time of over several centuries. The relative distribution of N among different pools and the time course of incorporation are of great importance to N availability and therefore have been the subject of many studies (Strickland et al. 1992; He et al. 1988; Legg et al. 1971). However, the mechanisms for the incorporation of C and N into the slower-turnover pools remain poorly understood (Tiessen et al. 1984; Schimel and Firestone 1989a). The incorporation of C and N can be influenced by the clay and SOM contents of the soils, the form of C and N, and the C/N ratios of the materials added to the soil (Janzen et al. 1988). When glucose, ceUulose, wheat straw and corn stalks were aerobically incubated for 12 weeks in a Chernozem soil with 1 5 N labeUed ammonium sulphate, maximum incorporation of 1 5 N in humic compounds was obtained in the cellulose-amended soil (Azam et al. 1985). Continued cropping was found to increase the incorporation of fertilizer N into stable organic forms (Legg et al. 1971). Abiotic processes may also be important in N incorporation of SOM. Ammonium assimilation rates in sterilized forest floor samples were sUghtly less than 20% of that in the unsterilized samples (Schimel and Firestone 1989b), presumably due to reaction of NH3 with 30 activated phenol or quinone rings. However, the assimilation of N by SOM is strongly pH dependent (T. Ballard pers. comm.) and is also affected by temperature (Overrein 1970). The nature of incorporation of applied N in soils and the availability of immobilized N have been extensively studied using various fractionation and extraction methods, sometimes in search for an easily obtainable index (Stanford 1969; Keeney and Bremner 1966; Stanford and Smith 1978; Kelley and Stevenson 1985). Strickland et al. (1992) found that a substantial amount of 1 5 N was incorporated into a denser (heavy) fraction of SOM after 60 days incubation with I 5NH4C1. The heavy fraction represents the more stable pools of the SOM. In a short-term incubation (7 days) of a Mollisol with (15NH,)2S04, 46% of the 1 5 N recovered in the organic matter was in the humin fraction (He et al. 1988). The immobilized 1 5 N in the humin fraction was as extractable as that of the native soil N, and therefore is no more available to plants than the native soil N (He et al. 1988). However, Smith and Power (1985) reported that residual N (fertilizer N not lost from the soil) was about 3 to 10 times more susceptible to mineralization than the soil N. They suggested that the differences in comparison to other studies might be caused by the frequent cultivation (disturbance) of the soil. However, the relationship between the indices measured in the laboratory and field plant N uptake measurements has not been as apparent as hoped, and new extraction methods have been proposed (Gianello and Bremner 1986; Clay and Malzer 1993). Early work on agricultural soils by Stanford and coworkers found that the alkali- distillable N fraction of the extract obtained by autoclaving in 0.01 M CaCl2 can be used as a satisfactory chemical index of soil N availability (Smith and Stanford 1971; Stanford and DeMar 1969 and 1970). When the selectivity of a given extractant to remove organic 1 5 N from soil is 31 expressed as an extractability ratio (ER, which equals the 1 5 N excess of extracted N divided by the I 5 N excess of total soil N), the autoclave method yielded very low (closer to 1) ER values compared to other extraction methods (Legg et al. 1971; Azam et al. 1989b; Juma and Paul 1984), indicating that the autoclaving method extracted the maximum amount of native soil N and thus had a low selectivity. Stanford and Smith (1978) proposed the use of acid KMn0 4 as an extractant to estimate the potentially mineralizable soil N. They suggested that the extracted NH4-N was derived from oxidation of the soil organic matter fraction most susceptible to mineralization. A good correlation was obtained between NH4-N extracted by 0.01 M or 0.02 M KMn04 in 0.5 M H2SO4 and the potentially mineralizable N (No, derived through curve fitting of aerobic N mineralization data): the former being about one-third to one-half of the latter. However, extractability ratios for acid KMn0 4 extraction were near 1.0, indicating that microbial N and native soil N were released in very similar proportions and selectivity was poor (Kelly and Stevenson 1985; Juma and Paul 1984). Juma and Paul (1984) found that if the concentration of K M N O 4 was reduced by 20-fold (from 0.2 M to 0.01 M), the extraction was 2.5 to 4.0 times more selective for 1 5 N. Juma and Paul (1984) showed that the highest extractibility ratios were obtained by extraction of samples using KC1 after chloroform fumigation-incubation relative to autoclaving in dilute CaCk, KMn0 4 oxidation, or acid hydrolysis. They suggested that the chloroform fumigation method extracted a biologically meaningful fraction. Extractability ratios were similar in fumigated and unfumigated samples. Azam et al. (1989b) extracted samples right after 32 fumigation, using 0.5 M K 2 S0 4 , and found that chloroform fiimigation increased the extractability of non-biomass N (the difference between total soil N and microbial biomass N). In general, milder extractants (hot water, hot 10 mM CaCk, hot 5 mM NaHCC«3 and cold NaHC03, hot KC1) extract less total and labelled N than strong extractants (0.02 M KMn0 4 in H2SO4, anhydrous formic acid, autoclaving in dilute CaCh, HC1 hydrolysis) (Clay and Malzer 1993; Juma and Paul 1984; Kelly and Stevenson 1985). In addition, milder extractants are more selective in extracting the incorporated I 5 N in SOM (Kelly and Stevenson 1985). IV.) Microbial role in forest soil N cycling The soil microbial populations mediate processes of residue decomposition, organic matter turnover, and nutrient cycling (McGill et al. 1986; Holmes and Zak 1994). They play a major role in regulating nutrient availabihties in forest soils. Soil microbial biomass is both a source and sink for nutrients. The assimilation and release of nutrients by and from the biomass may constitute a significant determinant of plant nutrition (Martens 1990). Martikainen and Palojarvi (1990) reported that microbial biomass C in coniferous and deciduous forest soils averaged 673 and 859 pg cm"3, or 1.19 and 1.13% of the total C, respectively. Microbial biomass N averaged 97 and 180 pg cm"3, or 5.94 and 3.43% of the total N for the coniferous and deciduous forest soils, respectively. In most soils studied, the amount of N contained in microbial biomass does not exceed 3% of total N (Stevenson 1986). Gallardo and Schlesinger (1990) found 3.3% of the total N in microbial biomass when measured by the fumigation-extraction method. 33 In the humid temperate forest soils of Spain, microbial biomass C and N ranged from 280 to 1610 ug C g"1 soil and 40 to 240 ug N g"1 soil, respectively. The amount of P, K, and Ca contained in the microbial biomass averaged 86, 73, and 10 ug g"1 soil, respectively (Diaz-Ravina et al. 1993a). In the humid tropics, microbial biomass C (fumigation-extraction) ranged from 760 to 2380 ug C g"1 soil in two soils with 20-year-old secondary vegetation; in annually harvested sites and bare soil, microbial C ranged from 580 to 1350 ug C g"1 soil and 110 to 1110 pg C g"1 soil, respectively (Henrot and Robertson 1994). Data on a coastal sand planted to radiata pine in New Zealand showed that 1.3 to 2.3% of the total soil C and 4.9 to 9.5% of the total soil N could be found in the microbial C and N fractions (Ross and Sparling 1993). The generally higher percentages for microbial N reflected the fact that microbial biomass usually has a lower C/N ratio than the bulk soil. Microbial populations in soils are affected by a range of factors, for example, the availability of C source (C input, fertilizer apphcation and soil organic matter quality) (Martens 1990), forest management practices (timber harvesting, residue removal and prescribed burning) (Jurgensen et al. 1980), agricultural management practices (McGill et al. 1986), and soil moisture, temperature and other properties (Alexander 1977). In studying the effect of fire intensity on soil microbial properties, Pietikainen and Fritze (1993) found that the amount of microbial biomass C and N, and soil respiration did not return to their control levels three years after prescribed (intense) burning, while recovery to control levels were observed for a simulated forest fire (less intense). The effect of burning on soil microbial biomass is the sum of heat and ash deposition (Fritze et al. 1994). Entry et al. (1986) reported that clearcutting and residue-burning treatment significantly reduced soil microbial biomass 34 compared to the other treatments (clearcut and residue left; clearcut and residue removed; uncut control). There was no difference in microbial biomass between the clearcut and residue removed treatment and control. Due to large inputs of organic substrate and the insulating effect from residue, microbial biomass in summer and winter was significantly greater in the clearcut and residue left treatment than in other treatments. Nitrogen availabihty is thought to be directly controlled by microbial activity (Holmes and Zak 1994). However, attempts to relate microbial biomass (C and N) to soil N availabihties have frequently failed (Fenn et al. 1993; Holmes and Zak 1994). Holmes and Zak (1994) hypothesized that microbial biomass and net N mineralization are inversely related on a seasonal basis. However, their result revealed a marked seasonal variability for net N mineralization with relatively constant microbial biomass C and N throughout the sampling period. Thus, the hypothesis had to be rejected and they suggested that the turnover rate of microbial biomass controlled the availabihty of N while keeping microbial biomass size constant. Diaz-Ravina et al. (1993b) argued that it is the combined effect of seasonal changes of microbial biomass, nutrient uptake by vegetation, litter fall, and organic matter mineralization that induces seasonal nutrient availabilities. Although Holmes and Zak (1994) reported that microbial biomass appeared to decrease during the summer and increase in the fall, a statistical significance of these temporal changes was not found. Results of Diaz-Ravina (1993b) and Entry et al. (1986) showed similar trends: microbial population was high in spring and autumn and low in summer and winter. Seasonal changes are caused by variations in climatic condition and substrate availabihty. In situations where no seasonal changes are found, Patra et al. (1990) suggested that a relatively even 35 distribution of litter inputs with time can smooth out the seasonal fluctuation in microbial pools. Holmes and Zak (1994) argued that the long-term patterns of C input have a greater role in influencing soil microbial biomass pool sizes than the seasonal variation in C inputs. The review section II.) indicated that fertilizer N can be quickly immobilized into SOM, through a range of mechanisms, with microbial immobilization as one of them Microbial population is very competitive for available N with tree roots. Jamieson and Killham (1994) investigated the root/microbe competition for N in a forest soil by using biocides to selectively inhibit target microbial groups. They found that inhibition of microbial populations, especially fungi, significantly increased the uptake of fertilizer and total N by Sitka spruce seedlings. According to Schimel et al. (1989), microbial population in a grassland ecosystem was found to be a stronger competitor for NH4-N than plants because NIL* is relatively immobile in soils and the ubiquitous distribution of microbes facilitates the uptake of NH4-N by microbes; whereas plants are stronger competitors for NO3-N than microbes because NO3" is mobile and easily accessible by plant roots. This type of experiment clearly demonstrates the role of microbial biomass in regulating the fate of fertilizer N and in tree nutrition. More recent work showed that the microbial role in regulating N cycling may be more predominant in relatively fertile soils. In infertile soils, N cycling may be short-circuited by uptake of organic forms of N directly by plants (Chapin 1995). This confirms that plants are able to adapt to a specific environment by developing different nutrient utilization strategies and indicated the importance in differentiating the prevailing pathways with varying fertility. 36 CHAPTER 3. UNDERSTORY COMPETITION EFFECT ON TREE GROWTH AND BIOMASS ALLOCATION ON A COASTAL OLD-GROWTH FOREST CUTOVER SITE* Introduction Understory vegetation affects and is affected by overstory vegetation and the conditions of the surrounding physical environment. Effects of environmental conditions and overstory vegetation on understory are well documented (Beatty 1984; Kimmins 1987). Under certain circumstances, understory vegetation is desirable, such as for wildlife habitat (Armleder and Dawson 1992; Hoefer and Bratton 1988) and for soil erosion control in erosion prone areas (Stewart and Forsling 1931). Understory vegetation is often undesirable, especially in reforestation, because understory will compete for nutrients (Neary et al. 1990; Messier and Kimmins 1990), water (Flint and Childs 1987), and light (Flint and Childs 1987; Brand and Janas 1988) with overstory and seedlings. Sometimes understory vegetation may exert allelopathic interferences by chemicals released through above- or belowground organs (Del Moral and Cates 1971). Under those situations, understory vegetation control becomes a necessity for crop tree estabhshment in the early stages of reforestation. Regeneration failure caused by competing understory vegetation (mainly salal) on northern Vancouver Island (Chapter 1) is of deep concern to the forest industry in British Columbia, because its allowable cut is going to be affected by declining second growth productivity, and satisfactory regeneration is required by law. *: A modified version of this Chapter is in press in For. Ecol. and Manage. 37 Belowground interference imposed by salal, which spreads vegetatively by means of rhizomes, either by competition for nutrients or by allelopathy has generally been regarded as the cause of the growth stagnation (Messier 1991; DeMontigny 1992). While the exact mechanisms are still not clear, studies carried out on those sites and elsewhere do show that controlling the understory growth improved tree diameter and height (Messier 1993) and volume growth (Deyoe and Dunsworth 1988). Messier (1992) investigated the effects of neutral shade and growing media on growth, biomass allocation and competitive ability of salal in a pot experiment; however, no study has been done on the salal-dominated sites to investigate the effect of understory competition on the biomass accumulation and allocation in various components, such as in foliage, branches, and various sized roots, of trees growing under field conditions. Biomass allocation information might be very useful in explaining the mechanisms by which understory affects crop tree growth. Nilsson and Albrektson (1993) indicated that the allocation of carbon to stem wood production had high priority for trees under high competitive stress. In a study on planted black spruce (Picea mariana (Mill.) B.S.P.) in the Ontario Clay Belt, Munson and Timmer (1990) found similar trends in that seedlings responded to site nutrient stress by allocating proportionally more biomass to the stem and roots. Newton and Jolliffe (1993) reported a reversed trend for second growth black spruce stands in that bark and foliar mass proportions increased while stem and branch mass proportions declined with increasing density stress. Barclay et al. (1986) reported that thinning of 34-year-old Douglas-fir stands decreased the proportion of biomass allocated to wood and bark but increased the allocation to foliage and live branches. Therefore biomass allocation priorities are often not absolute and overlapping (Oliver and Larson 1990). In 38 this study, I hypothesized that biomass partitioning to current year foliage and branch and fine roots in western redcedar, western hemlock, and Sitka spruce would be increased in the control plots as a strategy to survive the strong competition from salal. The objectives of this study were to examine the effect of understory removal on height and diameter growth and biomass accumulation and allocation among different years' growth of foliage and branch and roots of different sizes under field conditions. This work was a continuation of part of Messier's (1993) salal removal experiment. Materials and methods Study site The study site is located on Block 4 of Tree Farm License (TFL) 25 near Port McNeill, on northern Vancouver Island, British Columbia, Canada (50°36'N, 127°15'W). The ecosystem at the study site was classified as the submontane very wet maritime variant of the Coastal Western Hemlock biogeoclimatic zone (Green and Klinka 1994) which comprises 98% of the Block (Lewis 1982). The old-growth western redcedar-western hemlock forest (CH type) is the climatic climax community consisting of a somewhat open western redcedar-western hemlock stand with a minor Pacific silver fir (Abies amabilis) component (Lewis 1982). The site used for this study was clearcut and burned before planting. After logging, the site was quickly and vigorously occupied by salal in response to the extra light. Other species found on the site include Vaccinium spp., fireweed (Epilobium angustifolium), and mosses. The CH phase ecosystem is situated on a gently undulating topography. The soil is a Ferro-humic Podzol (Germain 1985). A typical soil profile has the following horizons: L 39 horizon, usually very thin (1-3 cm); F horizon, usually 8-20 cm in thickness (in the burned cutover sites, this horizon is often reduced to less than 5 cm); a thick, mostly greater than 45 cm, H horizon; a thin Ae horizon; Bhf, about 10-20 cm in thickness; Bf, about 20-25 cm in thickness; Bfgj, about 20-30 cm in thickness; and followed by a BC or C horizon. Parent material is unconsolidated morainal and fluvial outwash material (Lewis 1982). Climatic condition is as follows: annual precipitation 1730 mm, with most of it occurring in the winter months as rain. May, June, July and August are the driest months; mean daily temperature varies from 2.4 °C in January to 13.8 °C in August. Field trial setup After clearcutting and slash burning of a CH phase ecosystem site, nursery-grown 'plug type1 1-0 seedlings of western redcedar, western hemlock, and Sitka spruce were planted at the start of the growing season in 1987. One set of seedlings was planted without the understory vegetation (treated). This was achieved by periodically removing all the aboveground understory in the surrounding area with a radius of 1 meter from the seedling stem; belowground competition from adjacent vegetation was eliminated by periodically cutting to a depth of 40 cm in a circle of 1 m radius (Messier 1993). Another set of seedlings was grown under natural conditions, i.e., with competing understory vegetation (control). In the fall of 1990, in conjunction with a micro-plot level fertilization trial, four seedlings of each species from both treated and control treatments were trenched in a circle of 1 m radius and plastic barriers installed to a depth of 50 cm In April 1991, fertilizer was apphed as ammonium sulphate ( 1 5N labeled) at 4 0 41 a rate of200 kg N ha"1 to all of these trenched microplots. For my experiment, the unfertilized plots of Messier (1993) were used. Aboveground understory removal was carried out throughout the experiment from half of the microplots by capping the shoots and leaving them on the microplot surface. Plots planted with cedar for both 'Control' and 'Treated' are shown in Figure 3.1. For the measurements described below, basal diameter and height growth were detennined on four trees for each species. Biomass measurements for trees, understory and the L horizon were based on two. replicates. Field measurements and sampling Basal diameter (root collar diameter) and height of each tree were measured in early June, 1991 right after the fertilizer application, on May 21, and on September 30, 1992, before the final sampling. Two measurements were taken for basal diameter at two directions for each tree. The incremental height growth for the years 1992, 1991, and 1990 was measured before the destructive sampling. Height growth from planting to 1989 reported in this Chapter was obtained by subtracting the growth from 1990 to 1992 from the total height growth. Half of the microplots (two replicates for each species by treatment combination) were destructively sampled in late October 1992. For each plot, the aboveground tree was cut at the root collar and put into a plastic bag. Tree roots were collected by excavating as much as possible of the root system. Understory vegetation was cut at ground level and separated into salal and non-salal aboveground components. From each microplot, two 25 x 25 cm subplots were excavated by 10-cm increments to a depth of 50 cm for tree fine root and understory root biomass quantification. A segregation of understory root mass into salal and non-salal components was 42 not possible, therefore only the total understory root mass was reported. Standing dead understory and L horizon were collected from each microplot. Laboratory analysis After the samples were transported to the laboratory, aboveground trees were separated into 1- and 2-yr old foliage and branches, and 3-yr and older components. Tree roots were washed free of soil and separated into four groups: stump, roots >1 cm (coarse roots), roots 0.25-1 cm (medium roots), and roots <0.25 cm (fine roots). Stump mass excludes the obvious roots which were cut off. Fine roots recovered from the two soil pits (25 x 25 x 50 cm) were used to correct the tree root biomass for each microplot. Understory roots were recovered from the soil samples without further separation into roots of different understory species. Because of the small quantities of roots obtained in the layers deeper than 10 cm, the understory roots were grouped into two samples for each profile: 0-20 cm and 20-50 cm All the samples were then dried at 65 °C. Western redcedar, an ^determinate species, exhibits morphology quite distinct from its coniferous associates in the Pinaceae, which makes it difficult to differentiate different years' growth using major branch whorls. I therefore identified different years' growth for redcedar following the method proposed by Parker and Johnson (1987). Statistical analysis Homogeneity of variance and normality of distribution were checked before any further statistical analysis. Logarithmic (10) or square root transformations were performed, when needed, to homogenize the variance and to normalize the distribution of the data set. Analysis of 43 variance was performed on all experimental variables using the General Linear Models (GLM) procedure of the SAS package (SAS Institute, Inc., 1989). Multiple comparisons (LSD) were used to test the differences between means within each species and treatment. Results and discussion Tree height and basal diameter growth Data for tree height growth in various periods and the total height on October 1 are presented in Figure 3.2 for the three species studied. Statistical analysis (Table 3.1) showed that treatments had significant effect on height growth in 1992 (P<0.1); however, no significant differences were found for height growth in the summer 1992, and in 1991 and 1990. Height growth between 1987 and 1989, and total height measured in early October 1992 were significantly greater in the treated plots than in the control plots, with P=0.0481 and 0.0096, respectively (Figure 3.2 and Table 3.1). No treatment by species interaction was observed for any of the parameters measured (Table 3.1), which means that treatment had a rather uniform effect on tree height growth for all the species studied. In a salal grubbing experiment reported by Weetman et al. (1989a), western redcedar plots that had salal removed tended to produce more height growth after 3 years of treatment. The salal grubbing effect on tree height growth was less obvious on western hemlock. The grubbing treatment increased relative foliar N concentrations for both cedar and hemlock in the first two years after treatment. In a similar study, the response in leader growth of plantation Sitka spruce was found to be immediate in fertilized salal-dominated plots (Weetman et al. 1989b). The annual height growth recovered matched the spruce growing on the salal-free sites. 44 Table 3.1. Analysis of variance* for the effect of understory competition and tree species on tree height and RCD@ growth Height RCD Height growth Height Height Height Total Total growth growth summer growth growth growth height RCD summer Variable df 1992 1992 1991 1990 87-89 growth growth 1992 Treatment 1 + ns ns ns * ** *** * Species 2 + + *** *** ns * ns + T x S 2 ns ns ns ns ns ns ns ns *: The difference between means was significant at + P<0.1; * P<0.05; ** P<0.01; ***P<0.001; ns - non-significant @: RCD - root collar diameter Fig. 3.2. Height growth of western redcedar, western hemlock and Sitka spruce in various periods with ('Treated') and without ('Control') understory removal. Vertical error bars represent standard deviations. 1992 summer 1992 summer 1992 summer 1992 1992 1992 Cedar Cedar Hemlock Hemlock Spruce Spruce 1987-89 total 1987-89 total 1987-89 total 46 It is therefore reasonable to conclude that the improved height growth in the plots with salal removal was the result of improved N availabihty. The tree height growth rate increases obtained in 1991 and 1992 reflected the effect of fertilization in April 1991, although climatic conditions in different years may affect the height growth to some extent. As can be seen from Figure 3.2a, most of the height growth was occurring in the summer months as was measured in 1992. When understory was removed from the plots, 72.1 and 75.4% of the annual growth occurred during the summer as compared to 85.3 and 78.9% when understory was present, for plots planted with hemlock and spruce, respectively. For plots planted with redcedar, comparable values are 69.8% vs. 53.1%. This indicates that for hemlock and spruce, tree height growth tended to be spread over the year more evenly when the understory competition was removed, i.e., when competition exists, tree height growth was less severely impeded in the summer than in the other seasons, probably indicating more severe nutritional limitations in the fall to spring seasons. This may be partly explained by the fact that nutrients are more scarce in the winter because of slow mineralization under low temperature (Theodorou and Bowen 1983) and high possibility of leaching loss under high precipitation (Zakharchenko 1974; Kowalenko 1989) in the winter. The reason for the reversed trend with redcedar was unknown. One of the redcedar trees in the treated plots had a height of 80 cm when measured on October 1, 1992. This was a rather low value for redcedar tree height in the treated plots in relation to other trees. The height growth in the summer of 1992 was zero. This was obviously an abnormal tree and was excluded from the subsequent calculations. The height growth calculations might be somewhat affected by this exclusion. 47 Fig. 3.3. Root collar diameter (RCD) growth in total and in summer 1992, for western redcedar, western hemlock and Sitka spruce. 'Treated' and 'Control' refer to treatments with and without understory removal in the plots. Vertical error bars represent standard deviations. M Treated • Control 8 -r 7 1 Cedar Cedar Hemlock Hemlock Spruce Spruce total summer total summer total summer 1992 1992 1992 48 Redcedar and hemlock are ^determinate species (Weetman et al. 1989a); therefore height growth was benefited immediately after fertilization in 1991 (Figure 3.2b). Sitka spruce is a determinate species, its current year's growth potential is determined by the growth condition in the year before. This was reflected by the increase in height growth for Sitka spruce in 1992, one year after the fertilizer application (Figure 3.2a). Understory removal also increased root collar diameter growth (Figure 3.3). The total root collar diameter, when measured in early October 1992, was 38 (P<0.1), 88 (P<0.05), and 65% (P<0.05) greater in the treated plots than in the control plots, for redcedar, hemlock and spruce, respectively. The growth of root collar diameter in the summer 1992 was 111% (P<0.05) greater in the treated plots than in the control plots for redcedar, but was not significantly different for hemlock and spruce. The differences between species were generally not significant and no treatment by species interactions were found (Table 3.1). Positive growth responses following understory vegetation removal were also found for Douglas-fir on southern Vancouver Island (Green 1990), and for Sitka spruce in heather (Calluna vulgaris, L.) dominated sites in Scotland (Dickson and Savill 1974). Messier's (1993) study showed that understory vegetation removal had significantly increased total height and diameter increments in comparison with the control treatment in the first 3 years of field experiment. Since this study used Messier's (1993) field plots, it was interesting to note that increased height and diameter growth in the treated plots were sustained 6 years after seedling outplanting. Using the resin bag method, Messier (1993) discovered that removal of understory vegetation increased resin ammonium N availabihty by 20-40% and resin phosphorus availabihty by 15-32% and estimated that 30-45% of the potentially available N was taken up annually by 49 Table 3.2. Analysis of variance* for the effect of understory competition and tree species on biomass accumulation of trees, understory and non-living components 1-year 1-year 2-year 2-year >3-year AG@ Roots Variable df fohage branch foliage branch branch tree Slump >1 cm Treatment 1 * * * Species 2 ns ns T x S 2 ns ns Roots Roots USR@ USR 0.25- <0.25 0-20 20-50 Non- Dead Variable df 1 cm cm cm cm salal Salal Litter Salal *** * * * * * * * ** ns + ns ns ns ** ns + + ns ns Treatment 1 ** ** *** * * *** *** ** Species 2 * * * ns * *** * ns T x S 2 * * + + * *** * ns *: The difference between means was significant at + P<0.1; * P<0.05; * * P O . 0 ' 1 ; * * * p<o.001; ns - non-significant @: A G - aboveground; USR - understory roots 5 0 Fig. 3 .4 . Biomass accumulated at the final harvesting in various aboveground tree components. 'Treated' and 'Control' refer to treatments with and without understory removal in the plots. Vertical error bars represent standard deviations. o m Cedar Cedar Hemlock Hemlock Spruce Spruce 1-year 1-year 1-year 1-year 1-year 1-year Foliage Branch Foliage Branch Foliage Branch o Q. O e> (0 CO CO E o m 800 700 600 500 400 300 200 100 (b) I ^ W Treated • Control Cedar Cedar Hemlock Hemlock Spruce Spruce 9-uaar 2-year 9-w»flr 7-\K*ar ?-u^ar ?-\jpar 2-ye  2 year 2 ye r 2 ye  2 ye Foliage Branch Foliage Branch Foliage Branch o a o i _ O (0 (0 (0 E o ffl 5000 4500 4000 3500 3000 2500 2000 1500 1000 500 0 Cedar Cedar 3-yr and above- older ground Hemlock Hemlock Spruce Spruce 3-yr and above- 3-yr and above- older ground older ground 51 Fig. 3.5. Biomass accumulated at the final harvesting in various belowground tree components. 'Treated' and 'Control' refer to treatments with and without understory removal in the plots. Vertical error bars represent standard deviations. Cedar Cedar Hemlock Hemlock Spruce Spruce stump roots stump roots stump roots >1 cm >1 cm >1 cm ^Trea ted • Control Cedar Cedar Hemlock Hemlock Spruce Spruce roots roots roots roots roots roots 0.25-1 cm <0.25cm 0.25-1 cm <0.25cm 0.25-1 cm <0.25cm 52 competing vegetation on CH sites. The reduced uptake and immobilization of N and other nutrients by the biomass of salal and other competing species in the treated plots apparently were factors contributing to the improved tree height and diameter growth. Tree biomass accumulation The comparison of tree biomass growth between treated and control plots was plotted in Figures 3.4 and 3.5. A prolonged exclusion of understory competing vegetation had resulted in a significant increase in biomass of all the components that were studied (Table 3.2). Compared with the results in the height and root collar diameter growth, biomass seems to be a more reliable and sensitive measure of the effect of understory competition on tree growth, as more significant (lifferences were detected. In effect, biomass growth reflects the growth along the entire stem (Brand and Magnussen 1988) and the integration of the entire suite of physiological processes (Abrahamson and Caswell 1982). Some significant interactions were found between treatments and species (Table 3.2). For example, the interaction for the 2-yr fohage (P<0.01) was caused by the extremely low 2-yr old fohage biomass in the control hemlock plots. The current year biomass growth (Figure 3.4a) showed that treatment had increased biomass growth from 1.6 (spruce branch) to 7.6 times (hemlock fohage) over control. The greatest increase was attained with hemlock, the least with spruce. Treatment effect on the biomass accumulation in 2- yr old fohage and branches and 3-yr and older components was similar as was the magnitude of increases for the species (Figure 3.4b and 3.4c). For all the root components studied, the "stump" and ">1 cm roots" had the highest amounts of biomass in the treated plots. The differences in biomass accumulation between different root components were much smaller in the 53 Fig. 3.6. Understory and dead salal and litter mass at the final harvesting in plots with (Treated) and without (Control) understory removal. Vertical bars are SDs. (a) Understory aboveground; (b) Understory belowground; USR-understory roots; (c) Understory standing dead and litter. 2500 o 2000 Q. 1 5 0 0 CO CO E o m 500 o 5000 T 4500 - 2 4000 - 5- 3500 - S 3000 - M 2500 - |g 2000 - | 1500 - O 1000 - g) 500 (a) - -- T —IE— H I—1—! —x^. H Cedar Cedar Hemlock Hemlock Spruce Spruce non- salal non- salal non- salal salal salal salal (b) m Treated • Control Cedar Cedar Hemlock Hemlock Spruce Spruce USR USR USR USR USR USR 0-20 cm 20-50 cm 0-20 cm 20-50 cm 0-20 cm 20-50 cm o Q. CO (0 (0 03 E o m 1800 1600 1400 1200 1000 800 600 400 200 0 X (c) Cedar litter Cedar dead salal Hemlock Hemlock Spruce litter dead litter salal Spruce dead salal 54 control plots (Figure 3.5a and 3.5b). Salal removal gave an overall root biomass increase of 2.3 (P=0.4725), 7.6 (P=0.0065), and 2.9 (P-0.0713) times over control, for plots planted with redcedar, hemlock and spruce, respectively. Figure 3.6 shows the above- and belowground biomass of the understory vegetation and standing dead salal and litter mass at the time of the final harvest. The aboveground understory biomass in the treated plots was kept at a minimum by repeated clippings. The amount of aboveground understory vegetation harvested in the treated plots was the growth between the last cutting (July 16, 1992) and the final harvest (September 30, 1992). The aboveground understory biomass was as high as 1554, 2971, and 1373 g plot"1, or 4947, 9457, and 4370 kg ha"1 for control plots planted with redcedar, hemlock and spruce, respectively (Figure 3.6a). These data are in accordance with what Messier and Kimmins (1991) obtained on similar sites eight years after clearcutting and planting, while the belowground understory biomass data were somewhat lower. Interestingly, the belowground biomass in the treated plots was very high (Figure 3.6b) compared with what was on the aboveground giving root/shoot ratios of 9.9, 98.5, and 10.8 for plots planted with redcedar, hemlock, and spruce, respectively, whereas the root/shoot ratios for the corresponding control plots were only 1.3, 1.4, and 1.6. This indicated that belowground understory was quite persistent to surviving even after a prolonged period (6 years) of aboveground understory vegetation removal. In a previously mentioned study on a similar site, aboveground salal biomass was 73% of total aboveground biomass eight years after logging and burning (Messier and Kimmins 1991). In this study, the aboveground salal biomass was 63, 62, and 52% of the total aboveground biomass in the control plots planted with redcedar, hemlock and spruce, respectively, indicating 55 that fertilization may had improved the conifer seedling growth and had thus reduced the proportion of the aboveground salal biomass in the total aboveground biomass. Calculations showed that belowground salal biomass was 91, 71 and 79% of the total belowground biomass for the respective control plots. The greater proportion of salal biomass in the total biomass belowground than aboveground showed that the competition from salal for the limited resources is mainly from belowground. One of the consequences of uncurbed understory growth was the accumulation of Utter mass in the control plots (Figure 3.6c). The Utter mass coUected from the treated plots was primarily from the understory put back into the plots when it was chpped. Noticeable amounts of standing dead salal mass were also accumulated in the control plots, whilst there was none in the treated plots. The impUcation from this is that abundant understory growth wiU alter the resource allocation within the ecosystem and indirectly affect conifer growth, because nutrients immobilized in the Utter and standing dead biomass wiU only become available through decomposition, which is a slow process (Fahey 1983). Tree Biomass Allocation Although most of the traits of tree growth (e.g., height, diameter, and biomass in various fractions) were changed by the treatment, no statistically significant difference was found for the aUocation of tree biomass among the various components, except for the hemlock 2-yr branch and spruce <0.25 cm roots, when biomass components were expressed as a percentage of the total tree biomass, nor were shoot/root ratios significantly different between treatments (Table 3.3). Therefore, the hypothesis proposed in the introduction is rejected based on these results. 56 The results obtained here are different from those of Newton and Jolliffe (1993, decrease in stem and branch and increase in foliar and bark mass proportion on increased stand density) and Nilsson and Albrektson (1993, increased proportion of stem mass under high competitive stress). When the competition is mainly for belowground resources (nutrients), the resource depletion model of competition states that competition acts to reduce the relative growth rates of all individuals by the same proportion which results in an unchanged or lowered size inequality at higher densities after a given period of growth (Weiner and Thomas 1986; Newton and Jolliffe 1993). It would be reasonable to postulate that in order to have the size inequality unchanged or lowered at higher densities, the partitioning of dry matter in different components of a tree will be stable at different competition levels. If at higher densities the relative growth rates of all the mdrviduals are not reduced by the same proportion, it would be more likely that plants will adopt different resource allocation strategies. In other words, when belowground resources are the hmiting factors, sometimes biomass allocation may be unchanged to arrive at proportional reductions in relative growth rate of various sized trees. In most reported cases biomass allocation strategies were somewhat altered by the competitive stresses imposed (Lieffers and Titus 1989; Barclay et al. 1986; Munson and Timmer 1990; Nilsson and Albrektson 1993). In this study, the concurrent increases in height and root collar diameter in the treated plots, as was discussed earlier, also suggest that the relative proportion of biomass allocation might be unchanged. Table 3.3 indicates some statistically non-significant differences between treatments in biomass partitioning for the three species: higher in current year foliar and branch components, lower in previous year foliar components (significant for hemlock), higher in the 0.25-1 cm roots 57 (significant for spruce) in plots under salal competition than in plots without understory competition. Shoot/root ratios were also non-significantly lower in cedar and hemlock plots with salal. In the other components, the partitioning patterns were more irregular. The lower amount of biomass partitioned in the 2-yr foliar component in plots with salal might be a result of greater transfer of nutrients to the current year components which resulted in lower photosynthetic efficiencies. Thus, another possibility is that the hypothesis is acceptable but was falsely concealed by the fertilizer apphcation in April 1991. The fertilization treatment may have alleviated the nutrient shortage problem in the salal dominated plots in the short term, leading to non-significant differences in biomass allocation in different tree components. Conclusions Plants with high competitive ability may achieve a high rate of acquisition of resources in productive, crowded vegetation. They have two important competitive characteristics (Grime 1979): 1) the potential to produce a dense canopy of leaves and a large root surface area when growth condition is favorable; and 2) the capacity to rapidly adjust morphogenetically both in the apportionment of photosynthate between root and shoot and in the size, morphology, and distribution of individual leaves and roots. Salal has all of those characteristics. It grows rapidly from extensive interconnected rhizomes, which persist for a long time even with the aboveground biomass ehminated (Bunnell 1990; this study). It is capable of enhanced uptake of nutrients and water assisted by mycorrhizas. Salal competition poses considerable threat to the regeneration of coastal British Columbia forests. It was evident from this study that salal competition reduces coniferous tree height and diameter growth and biomass accumulation. As is shown in this study, 6 years after planting, <+4 ~ ° 'd co © 8 ts § I 1 1 <+•! id o ci , . CB Co © O (N "is +- O O O CS C/3 co O o </-> cs o V CN a t - l id CS o CD d a CS £ CD re (50 <D cet > i i CN cS (-1 ,d CQ o <U es >> c3 i—1 | H <u CS &0 <D c« >> 1 H 8 ts H CO VO VO CS CS in cs CS CS o CS ON 00 cs O © cn cn cs cn c s CS cs * cs o 00 cn I In CS o o CS CD CS ! cs cs ! cs es I—1 i r- ! o O I r-- ! vo CS <rl | cn cn j c s cn CS ! cs CS ! cs es ON vo ! •<*• ON ! o ! c s O i—J ! cn cn ! cn CS CS ! cs CS ! cs X> cn • 00 c s i cn m ! cn | o o cn ! <n vd ! v~> vd I cs CS ! ^ CS es ON 1 ON VO j c s cn vq ! ° i n 1 i—J r-cn ! vo' m' | i - H CS ! « CS ! «a es vo 1 ON VO ! vo i— i cn | o OO | m vd ! vo vd ' 0 0 CS ! cs CS i CS CS cn ! <=> cn ! c s VO <n ! vo cn I r — | CS 00 1 ON © ! m' >n c s ! <"n ! cn cn CS m ! « es ! « es ! 00 ! vo mi © ! °̂ ! 1—1 ! ON 00 vd CS cs - o es CS o i 00 cn >n ON cn m i— i ON i—i •*' ON i n i n CS cs CS es es ON o <n m> cn i n © CS <n m vd vd ON CS cs CS es es <n o ON o © ON cn vd i n 00 o cn i— i i—i c s o o •a o 1 1 CD o s o a d O o O CD +-» es <D \-i m cs ©' s es es ' t3 <D 8 6 a? -9 I cs ts o o CO co O • i—i r P 13 O 8 o o o CS o co •3 CD co O 'HH -d ts CD CD .a i co _3 PH o o 0 <D .a co 1 d o o o & 2 59 salal competition was still severely reducing tree growth. An extended study of the salal exclusion effect on tree growth is recommended to see whether the growth rates of young coniferous trees in natural conditions (with salal competition) can catch up with that in the treated plots. No effective ways have been found yet to eliminate or control salal growth on an operational scale. Recently, Prescott et al. (1993a) reported that repeated N fertilization to two salal dominated coastal Douglas-fir forests had reduced or completely ehminated aboveground salal growth. A 10-year trial to test the influence of fertilization on salal development on the salal dominated CH phase forests recently indicated that salal growth was greatly diminished after crown closure of the trees (G.F. Weetman pers. comm.). If this method can be used on an operational scale to control salal growth on the salal dominated sites, there are two benefits to forest management: 1) salal competition could be reduced (at least to some extent) or even wiped out; and 2) tree nutrition and growth could be improved by repeated fertilization. 60 CHAPTER 4. EFFECT OF UNDERSTORY COMPETITION ON DISTRIBUTION AND RECOVERY OF 1 S N APPLIED TO A WESTERN REDCEDAR-WESTERN H E M L O C K CLEARCUT SITE* Introduction Nitrogen is often the factor limiting growth of temperate and boreal coniferous forests (Wollum and Davey 1975). Although increased research activities have led to an improved understanding of N dynamics in forest ecosystems, interpretation and evaluation of N status in forest soils is still difficult because mechanisms of N cycling and availability in forest soils remain poorly understood and site specific (Binkley and Hart 1989). Nutritional deficiencies (mainly N) have been identified as one of the factors leading to young conifer tree growth stagnation on the CH sites (Messier 1993; Prescott et al. 1993b). The rapid regrowth of salal (both above- and belowground) after clearcutting and slash-burning (Messier and Kimmins 1991) and its vigorous competition for scarce nutrients (Messier 1993) exerts further adverse effects on tree growth performance. According to Messier (1991), 30-45% of the available N on CH clearcut sites could be explained by uptake of the non-crop vegetation. Fertilization is the most commonly used method to overcome nutrient deficiency problems in forestry. However, N apphcation may also encourage abundant growth of salal and other understory vegetation before crown closure, thus increasing understory competition for nutrients and other resources. If a large proportion of the applied fertilizer N is immobilized by non-crop understory vegetation, then the efficiency of fertilization in releasing growth stagnation is *: A modified version of this Chapter is published in the Can. J. of For. Res., 26: 313-321(1996). 61 reduced, which could lead to the failure of fertilization programs in reaching the desired silvicultural objectives. Fertilizer N recovery has been studied using the 1 5 N tracer technique in Sitka spruce (Hulm and Killham 1990), Scots pine and Norway spruce (Melin and Nommik 1988), slash pine (Mead and Pritchett 1975a and b), lodgepole pine (Preston et al. 1990) and Douglas-fir (Heilman et al. 1982a; Preston et al. 1990), and some other tree species. These studies indicated that recovery of N in trees was generally low (10-25%). However, no reports were found on western hemlock and western redcedar forests. The objectives of the present study were: 1) to study the distribution of applied N in the plant-soil system; 2) to determine the effect of salal competition on the uptake of fertilizer N by trees; and 3) to test if different tree species have different capacities to compete with understory vegetation for N supply. Although the 1 5 N method employed here to track the fate of fertilizer N has higher sensitivity and precision over indirect methods, the result can be complicated by the biological isotope exchange processes occurring in the soil as was discussed in Chapter 2. Therefore, the 1 5 N method may underestimate 1 5 N recovery (Nommik and Larson 1989). I was aware of the trade off between the 1 5 N method and other methods. It was also assumed that plants do not selectively take up the various N species in the soil, since the differences in chemical properties of 1 4 N and 1 5 N are negligible for most tracer studies (Hauck and Bremner 1976). Materials and methods Study area A description of the study area is found in Chapter 3. Selected soil properties for the various depths sampled are given in Table 4.1. 62 Field experiment This experiment was established as a completely randomized factorial design, with three species (western redcedar, western hemlock, and Sitka spruce) and two treatments (no understory removal - "control", and aboveground understory removal - "treated"). Each treatment and species combination was replicated four times. Single-tree microplots (1 m radius) were established in the fall of 1990 when the trees were 4 years old and they were trenched and separated from the bulk soil using plastic sheeting to a depth of 50 cm, which was beyond the rooting depth. A total of 24 microplots were established (3 species x 2 treatments x 4 replicates). Nitrogen-15 labeled (NH 4) 2S0 4 (3.38044 atom% 1 ? N enrichment) was applied on April 16, 1991 at a rate of 200 kg N ha"1 to all of the 24 microplots. For each plot, the fertilizer was evenly applied from a large watering can in a total of 4 L water. As the top soil layer (0-10 cm) has a water holding capacity of over 3 kg kg"1, the penetration of added water should not have exceeded 1-2 cm at the time of apphcation. There was some regrowth of salal and other understory species on the treated plots; in particular salal continued to resprout from its network of rhizomes despite several years of suppression. Removal of understory vegetation was maintained in the treated plots by periodic clipping and the small amounts of material were placed on the surface of the plot. Destructive sampling took place in October 1992, two growing seasons after fertilizer apphcation. Half of the 24 microplots were randomly selected and destructively sampled (3 species x 2 treatments x 2 rephcates). Half of the plots were left to grow for another a few years for a longer-term 5 N uptake study. For the destructive sampling, first the aboveground understory vegetation was cut at ground level. Next, the L layer and standing dead biomass 63 (mostly salal) were collected. Then the aboveground tree was removed from the plot. Two randomly located soil pits (25 by 25 cm) were excavated to a depth of 50 cm in intervals of 10- cm. After the soil pits were finished, the tree roots in the microplots were carefully excavated using hand tools to recover visible roots as much as possible from the entire plot. Laboratory analyses Soil samples in each plastic bag (from a 10 cm thick layer of a pit) were weighed in the laboratory. Visible roots were removed from the soil samples. Soil water content was measured to determine the bulk density of each soil layer, based on a soil volume of 0.00625 m3 (0.25 x 0.25 x 0.1 m). A proportionate (by weight) subsample was taken from each of the two samples from the same depth of a microplot, combined and air-dried for total N and 1 5 N analysis. After the biomass samples were brought into the laboratory, the aboveground tree biomass was separated into: 1) needles: 1-yr-old and 2-yr-old; 2) branches: 1-yr-old, 2-yr-old, and >3-yr- old. There were few >3-yr-old needles and therefore they were not separated from the >3-yr-old branches. The belowground tree was separated into: stump, small roots (< 0.25 cm in diameter), medium roots (0.25 - 1.0 cm), and large roots ( > 1.0 cm). Stump mass excludes the obvious roots which were cut off. The aboveground understory vegetation was separated into salal and non-salal. The non-salal component comprised mainly fireweed, deer fern (Blechnum spicant), blueberry (Vaccinium spp.) and bunchberry (Comus canadensis). The belowground understory roots were divided into two layers: 0-20 and 20-50 cm. Roots picked from the soil samples were separated into tree roots and understory roots, and were converted to per plot basis. Tree roots picked from the soil samples were used to correct the tree root biomass for each plot. A further separation of the understory roots into salal and non-salal was not possible. Plant materials were Table 4.1. Selected soil properties of the study site Bulk Depth Organic C Nitrogen C/N density (cm) PH* (g kg1) gkg 1 kg ha 1 ratio (Mg m"3) 0-10 3.00 464.6 7.30 1291 63.7 0.177 10-20 3.11 376.5 5.31 1167 70.9 0.220 20-30 3.40 291.0 3.79 1301 76.7 0.343 30-40 3.92 114.3 2.01 1167 57.0 0.582 40-50 4.30 62.5 1.32 958 47.5 0.728 Measured in 1:1 (v/v) 0.01 M CaCl2 65 dried at 65 °C. All the plant components were weighed (oven dry basis), coarsely ground and then subsampled for fine grmding. After the plant samples were ground to pass a No. 40 sieve (0.425 mm), they were analyzed for total N by the serriimicro-Kjeldahl method described by Bremner and Mulvaney (1982) except that mercuric oxide was used as the catalyst. The distillates, collected in boric acid-ethanol, were dried at 70 °C. The ammonium N was converted to dinitrogen gas using the Rittenberg reaction with alkaline hthium hypobromite, and analyzed for ' 5 N enrichment using a Vacuum Generators Sira 9 mass spectrometer (Preston et al. 1990). Field-moist soil samples were air dried and ground to pass a #10 sieve (2 mm), and were then oven-dried at 65 °C and ground to 50 um in a Siebteknik mill. Total N was analyzed by the semimicro-Kjeldahl method and 1 5 N enrichment by mass spectrometer, following the methods used for plant sample analysis. The pH for soil samples was measured in 1:1 (v/v) 0.01 M CaCk suspensions. Total C concentrations of soil samples were detennined using a LECO CR-12 C analyzer (Model 781-600, Leco Corporation 1981). Statistical analyses Homogeneity of variances and normality of distributions of data sets were checked before any statistical analysis. Data that were not homogeneous (recovery of 1 ? N in understory roots) were SIN transformed prior to analysis. However, means were reported on untransformed data. Analyses of variance were performed on all experimental variables using the General Linear Models (GLM) procedure of the SAS package (SAS Institute Inc. 1989). Group means of independent variables were compared between treatments for each species by Scheffe's multiple range test, for each component considered. 66 15 Table 4.2. Within-tree distribution (whole tree as 100%) of applied N by tree component after two growing seasons. Values in parentheses are SEs of means Cedar Hemlock Spruce Component Control Treated Control Treated Control Treated Needles 1-yr-old 66.55 64.27 A W 32.10 30.70 B 37.29 35.19B (0.80) (1.03) (4.48) (0.41) (5.44) (0.78) 2-yr-old 7.89 8.74 A ns 17.18 22.35 B 13.62 15.11 C (0.04) (0.61) (1.19) (2.42) (0.72) (2.19) Branches 1-yr-old 5.37 5.58 Ans 7.09 5.47 A 6.56 6.50 A (0.32) (0.58) (0.77) (1.49) (2.78) (0.38) 2-yr-old 4.80 5.31 Ans 6.99 7.34 A 5.99 5.03 A (0.16) (1.24) (0.88) (2.02) (0.02) (0.40) >3-yr-old 8.08 10.07 Ans 25.02 21.34 B 23.33 22.56 B (0.42) (0.64) (4.58) (0.76) (5.10) (2.57) Stump 1.98 1.54 Ans 2.49 1.78 AB 3.72 3.16B (0.16) (0.07) (0.37) (0.33) (1.20) (0.26) Roots > 1 cm 1.14 1.57 Ans 2.71 2.35 AB 2.02 4.09 B (0.12) (0.17) (0.33) (0.19) (1.11) (0.49) 0.25-1 cm 1.74 1.05 Ans 4.02 3.74 B 3.98 3.51 B (0.33) (0.01) (0.62) (0.56) (0.23) (0.30) < 0.25 cm 2.49 1.89 Ans 2.43 4.96 A 3.50 4.87 A (0.04) (0.96) (0.96) (0.15) (2.10) (0.90) \ Same uppercase letters (letters A, B, and C are reserved for this comparison) indicate no species effect (at p=0.05 level) when there was no treatment x species interaction. #: ns - nonsignificant; *: p<0.05. Compares treatment effect when there was no treatment x species interaction. 67 Results Nitrogen-15 distribution within trees The distribution of 1 5 N in tree components is shown in Table 4.2. Within the trees, the largest proportions of 1 ? N recovered were found in the 1-yr-old needles of the three species irrespective of treatment. The second largest proportion of the N recovery was in the >3-yr-old branches, except for 1 ? N in the treated western hemlock plots, followed by 2-yr-old needles. From 49 to 74% of the 1 5 N recovered in the trees was found in the needles. Between 84 and 94% of the ' 5 N in the trees was found in the aboveground tree components, while 6 to 16% was in belowground tree components including the stump. Based on analysis of variance results (data not shown), no significant treatment by species interactions were found on 5 N distribution for any of the components studied. No significant difference was induced by salal-removal treatment for 1 5 N distribution in all of the tree components studied (Table 4.2). Tree species significantly affected the distribution of N in 1-yr- old needles (F=77.99, p=0.0001), 2-yr-old needles (F=30.39, p=0.0007), >3-yr-old branches (F=14.30, p=0.0052), stumps (F=5.28, p=0.0476), and coarse (F=5.53, p=0.043) and medium roots (F=24.86, p=0.0012). The distribution of 1 5 N in various components was quite similar for hemlock and spruce under the same treatment (except for the 2-yr-old needles), while the data for western redcedar were quite different from those of hemlock and spruce, for the tree components with significant treatment x species interaction. Nitrogen-15 distribution within understory When 5 N contents in various understory components were expressed as a percentage of 68 Table 4.3. Distribution (as % of total N in understory) of applied N in understory components after two growing seasons. Based on untransformed data. Values in parentheses are SEs of means Cedar Hemlock Spruce Component Control Treated Control Treated Control Treated Aboveground, 57.37 6.74 A 1 ** 41.65 0.17 A 46.03 12.97 A salal (0.56) (0.66) (16.65) (0.17) (7.67) (10.66) Aboveground, 1.19D* 6.97 dJ ns§ 7.67 D 0.64 d * § 3.61 D 6.04 a ns non-salal (0.86) (3.45) (0.69) (0.64) (2.39) (1.83) Root 0-20 cm 39.42 73.80 A * 44.65 93.56 A 48.20 67.96 A (2.32) (0.64) (14.27) (5.01) (3.96) (1.82) Root 20-50 cm 2.03 12.49 A ns 6.04 5.64 A 2.17 13.03 A (0.91) (4.75) (3.08) (4.20) (1.32) (10.65) Total understory 100 100 100 100 100 100 ^ d * : See Table 4.2. *: Same uppercase or lowercase letters (letters D, d, E, e, F, and f are reserved for this comparison) indicate no species effect (at p=0.05 level) for the "Control" or "Treated" treatments when there was significant treatment x species interaction. §: ns: nonsignificant. *: p<0.05. Compares treatment effect for the tree species "Cedar", "Hemlock", or "Spruce" when there was significant treatment x species interaction. 69 total understory 1 5 N, there were dramatic differences in the distribution of N in understory components between treatments with and without understory removal (Table 4.3). Analysis of variance results (data not shown) showed that there were no treatment by species interactions for any of the components studied, except for the aboveground non-salal component. Significantly larger proportions of 1 5 N were in the aboveground salal biomass in the control plots than in the treated plots. This was expected since in the control plots there was abundant understory growth, while in the treated plots there was very limited understory (aboveground) regrowth after repeated removal. Treatment by species interaction was significant (F=5.87, p=0.0386) for 1 5 N distribution in the aboveground non-salal understory component. Nitrogen-15 distribution in the aboveground non-salal understory components was variable, with a larger proportion in the treated plots than in the control plots for cedar (p<0.05), larger proportion in the control plots than in the treated plots for hemlock (p<0.05), and no difference between treatments for spruce. A much greater proportion of the 5 N was found in the understory roots in the treated plots than in the control plots. No effect of tree species on 1 5 N distribution in the understory components was found. Nitrogen-15 distribution within the soil-plant system When the 1 5 N content in the soil-plant system was expressed as a percentage of total 1 5 N recovered, Fig. 4.1 shows that significantly greater proportions of 5 N were in the tree components in the treated plots than in the control plots without understory removal (p<0.05); the relationship was reversed for the proportion of N in the understory (p<0.05). Greater proportions of N were found in the litter and standing dead components in the control plots than 70 in the treated plots (p<0.05), while the reverse was found for ' 5 N distribution in the soil (p<0.05). Neither species effect nor treatment by species interaction was found to be significant for N distribution in the soil-plant system (data not shown). Nitrogen-15 abundance and contents in aboveground tree components Nitrogen-15 abundance and contents in the aboveground tree components and then ANOVA table are presented in Table 4.4. Nitrogen-15 abundance is expressed as the atom percentage of 1 5 N in total N and 1 5 N content is expressed on per gram plant dry mass basis. Salal- removal treatment apparently reduced (non-significantly) 5 N abundance in the aboveground tree components, regardless of species (Table 4.4), with the exception of spruce 1-yr-old branches. This obviously was a dilution effect where the greater 1 5 N uptake could not compensate for the greater biomass accumulation in the treated plots. Species exhibited significant differences in N abundance in some of the components (1-yr-old and 2-yr-old fohage, and 2-yr-old branches) but treatment effects were not significant in any of the components. When 5 N was expressed in mg plot"1, salal removal significantly increased the storage of 5 N in all of the aboveground tree components (Table 4.4). The treatment by species interaction was significant for I 5 N content in 2-yr-old needles and >3-yr-old branches. For those two components, treatment effects were found to be significant (p<0.05) only for western hemlock. Differences between species in N storage (mg plot"1) were found for 2-yr-old needles and >3-yr-old branches. Recovery of N From 7.7 to 17.8% of the applied fertilizer was recovered in the trees in the treated plots, Fig. 4.1. Distribution of in the soil-plant systems expressed as a % of total applied recoveries. SD, standing dead. Litter is all plant Utter (tree + understory). B T r e e g U n d e r s t o r y • L i t t e r ^ S D n S o i l C o n t r o l T r e a t e d C o n t r o l T r e a t e d C o n t r o l T r e a t e d C e d a r H e m l o c k S p r u c e 72 Table 4.4. Effect of understory removal and tree species on N abundance and contents in aboveground tree components and analysis of variance Treat- 1 5 N 1-yr-old 1-yr-old 2-yr-old 2-yr-old >3-yr-old Species ment as foliage branch fohage branch branch Cedar Control (atom %) 1.36 (mg plot"1) 70.33 Treated (atom %) 1.31 (mg plot"1) 130.29 Hemlock Control (atom %) 1.61 (mg plot"1) 16.29 Treated (atom %) 1.43 (mg plot"1) 139.67 Spruce Control (atom %) 1.91 (mg plot"1) 45.52 Treated (atom %) 1.61 (mg plot"1) 88.26 1.45 1.40 1.29 1.28 5.57 8.28 5.19 8.84 1.36 1.28 1.24 1.20 10.91 18.23 11.53 21.59 1.63 1.71 1.65 1.51 3.53 7.97 3.49 11.47 1.46 1.48 1.47 1.33 25.07 99.64 33.59 99.23 1.49 1.90 1.87 1.66 9.22 15.43 6.97 26.53 1.59 1.76 1.63 1.50 16.22 38.80 12.43 57.05 ANOVA F value# Species (S) (atom %) 6.72* 0.96 5.76* 7.25* 1.29 (mg plot"1) 1.99 0.96 16.16** 2.25 26.67*** Treatment (atom %) 3.47 0.34 1.92 2.25 0.63 (mg plot"1) 29.20** 9.23* 49.50*** 10.78** 91.55*** S x T (atom %) 0.58 0.80 0.08 0.30 0.03 (mg plot"1) 3.09 1.92 18.26** 3.60 24.59** # : The difference between means is significant at *p<0.05; **p<0.01; ***p<0.001. Degrees of freedom for Treatment (1), Species (2), and S x T (2). 73 compared to only 2 to 5% in the control plots (Table 4.4). Treatment x species interaction was significant for 1 5 N recoveries in aboveground trees, tree roots, and total trees. Treatment effects on 1 5 N recoveries in above- and belowground tree and total tree components were significant for hemlock (p<0.05). Differences in N recovery among species were observed only between cedar and hemlock for the understory removal treatment. Treatment by species interaction was also significant for 5 N recovery in aboveground understory (F=l 1.68, p=0.0085). Understory removal reduced 5 N recovery in the aboveground understory compartment for every species studied. However, differences between species existed between hemlock and spruce only for the control treatment. There were no interactions between treatment and species for I 5 N recovery in understory root and in the htter+standing dead compartments. Understory removal reduced the 15 15 amount of N recovered in those compartments (p<0.05). The significantly higher amount of N recovered in the above- and belowground understory and litter + standing dead in the control compared to the treated, regardless of the tree species, was due to the manipulation of the understory in the experiment. The greatest total recovery of applied 5 N in the soil-plant system was in the treated hemlock plots (87.3%), while the least was in the treated spruce plots (56.8%, Table 4.5). Those differences are primarily caused by the varied capacities of the tree species to take up N. After two growing seasons, most of the recovered N was in the soil component. Total recovery of 5 N in the soil was as high as 66.8% of added N in the treated hemlock plots. There was no significant effect from treatment, species and treatment by species interaction for total recovery and recovery in the soil. 74 Discussion Nitrogen distribution within biomass Understory competition did not affect the proportionate distribution ( 1 5N in tree components/15N in whole tree) of 1 5 N within the trees. Values for I 5 N abundance in above- (Table 4.4) and belowground tree components (data not given) were also not significantly affected by the salal removal. The detailed analysis of biomass changes at this site (Chapter 3) showed that understory removal resulted in biomass increases in all tree components, without any effect on their distribution within trees. Tree height and basal diameter also responded positively to understory removal. Therefore I conclude that the pattern of 1 $ N distribution was determined by the distribution pattern of biomass in various tree components, or the resource depletion model (Newton and Jolliffe 1993). This model states that competition acts to reduce the relative growth rates of all individuals by the same proportion if the competition is mainly for belowground resources (nutrients). Thus, not only N distribution within trees, but also plant N content and 1 5 N recovery were largely determined by biomass changes following understory removal treatment. The distribution of ' 5 N (and also total N, data not shown) in the tree components was comparable to that reported by Preston et al. (1990) in studies on 11-yr-old lodgepole pine stands and 13-yr-old Douglas-fir stands in British Columbia. The distribution of 5 N followed that of total N very closely, indicating either a proportionate uptake of applied N and native N by tree components or a rapid turnover and redistribution of 5 N within the tree (Mead and Preston 1994). Internal N cycling from old tissues to new growth was suggested by Mead and Pritchett (1975a) and Nonnnik and Larsson (1989). Using a greenhouse sand culture incubation of 2-year- old Sitka spruce seedlings, Proe and Millard (1994) found that N taken up in the current season and partitioned to preexisting shoots or roots was not remobilized and translocated late in the Table 4.5. Recovery of fertilizer N in the soil-plant system after two growing seasons Cedar Hemlock Spruce Component Control Treated Control Treated Control Treated Aboveground 3.78 D* 7.2 d* ns§ 1.72 D 15.54 d* § 4.27 D 8.60 dns tree (0.50) (1.31) (0.62) (1.28) (1.68) (1.99) Tree root 0.30 D 0.45 dns 0.23 D 2.29 e * 0.60 D 1.65 dens (0.05) (0.01) (0.08) (0.21) (0.12) (0.60) Total tree 4.08 D 7.65 dns 1.95 D 17.83 d * 4.86 D 10.25 dns (0.55) (1.32) (0.70) (1.49) (1.80) (2.59) Aboveground 8.66 DE 0.41 d* 10.82 D 0.25 d * 6.72 E 0.55 d* understory (0.27) (0.08) (0.56) (0.25) (0.88) (0.16) Understory root 6.15 2.65 A 1 ** 13.75 2.34 A 6.78 0.84 A (0.55) (0.42) (8.06) (0.54) (0.55) (0.18) Litter + 4.79 1.07 A * 6.45 0.08 A 5.86 0.25 A standing dead (0.33) (0.86) (0.15) (0.05) (2.32) (0.13) Soil 50.68 56.20 Ans 46.85 66.81 A 40.69 44.88 A (0.85) (10.72) (0.33) (18.91) (7.07) (13.11) Total 74.35 67.97 Ans 79.81 87.30 A 64.90 56.76 A (0.79) (8.89) (6.98) (16.60) (7.91) (10.67) \ # : See Table 4.2;*,§: See Table 4.3. 76 season to support growth of new shoots. They found, however, that in a treatment with high relative addition rate of N, 1 5 N from current shoots formed earher in the season was remobilized to support the growth of a second flush later in the season. They further pointed out that budget studies have then limitations in quantifying internal cycling of N. The finding that the majority of the 1 5 N in the trees was in the 1-yr-old needles is consistent with other reports (Melin et al. 1983; Nambiar and Bowen 1986; Melin and Nommik 1988). After several years of understory removal, one would expect the aboveground understory biomass and hence 5 N content to be zero in the treated plots. However, salal is a very persistent plant (Prescott et al. 1993a) which resprouts quickly after the aboveground biomass is chpped. Therefore, some understory biomass and N were recovered in the treated plots at the time of the final harvesting. The understory removal treatment greatly altered the distribution of 1 5 N in the understory components. For example, a greater amount of N accumulated in the understory, standing dead biomass, and Utter layer in the control than in the treated plots because of greater standing dead and litter mass produced in the former than in the latter. As for the trees, a similar 15 15 distribution of N and total N was found in the understory, as the distribution of N and total N in understory was also largely determined by biomass distribution. Competition of salal for N Preston et al. (1990) studied the fate of ' 5 N labeled fertilizer applied on snow at two forest sites and found that understory vegetation reduced uptake of fertilizer N by crop trees. In a lodgepole pine stand, tree uptake ranged from 1.9 to 10.1% while understory uptake from 2.4 to 3.4%. However, in a Douglas-fir stand, understory took up more fertilizer N (10.8%) than the 77 tree (5.5%). In the present study, total ' 5 N recovery by trees was 4.1, 2.0 and 4.9% in control plots planted with western redcedar, western hemlock and Sitka spruce, respectively. The corresponding values for the treated plots were 7.7, 17.8 and 10.3%. These generally low values reveal that fertilization of those stands has low efficiency which is not unique to this site. Nitrogen-15 recovery in understory was far greater than that in the trees in the control plots. Most of the litter and all of the standing dead biomass originated from understory biomass. If the 1 5 N recovered in the litter and standing dead biomass is added to that of the understory, it reveals that 1 5 N labeled fertilizer immobilized by understory vegetation was 4.8, 15.9 and 3.3 times as much as that taken up by trees in the control plots. The effect of understory competition on the ' N recovery in trees was the focus of two recent pubhcations (Clinton and Mead 1994a and b). They studied N uptake by trees under simulated grazing, complete removal of pasture and rank ryegrass-cocksfoot-clover pasture and found that i.) removing pasture competition doubled tree 5 N; ii.) pasture was a major competitor for N due to its greater root biomass and root density; and hi.) there were no significant treatment differences in 1 5 N recovery in the 0-20 cm depth of soil. Some of the findings from the present study were similar: i) removal of salal competition doubled tree ' 5 N uptake by cedar and spruce, and increased 5 N uptake by hemlock more than 8 times; ii) salal was a persistent competitor for "5N, i.e., six years after understory removal, from 1.4 to 3.1% of the applied ' 5 N was recovered in the understory of treated plots; iii) removal of salal understory increased N incorporation into soil (0-50 cm, non-significant); iv) total recovery tended to be higher in the plots with understory present, because the understory took up more applied N shortly after application, thus reducing N loss from the system by leaching or other mechanisms; and v) the accumulation of 5 N in the 78 understory biomass, litter and standing dead material greatly reduced the availability of fertilizer N to crop trees. There have been no studies to indicate whether N tied up in the salal biomass/litter would be more easily available to trees, through mineralization, than that immobilized in the soil organic matter; however, in the short term, incorporation of N in salal biomass reduces its availability to crop trees. It is unlikely that N immobilized by salal would be turned over very quickly because of the persistent survival of salal even under adverse conditions and because of low rates of salal litter decomposition. For example, six years after aboveground salal removal was initiated recovery of understory root biomass was from 523 to 800 g plot"1 (Chapter 3). Standing live biomass 1 5 N in understory vegetation (excluding litter and standing dead biomass) was 14.8, 24.6 and 13.5% of the total applied, for control plots planted with cedar, hemlock and spruce, respectively. Those data were in the upper range of N usually found in understory (Preston et al. 1990). In the plots where salal was repeatedly removed, from 80 to 99% of the 5 N recovered in understory was in the understory roots, because of the strong regrowth capacity of the salal rhizomes. Results clearly showed that trees competed poorly with the understory vegetation for the fertilizer N applied. The large amount of 1 $ N stored in the soil (mostly in organic form) which was immobilized largely through microbial processes demonstrated that trees also competed poorly with the soil microbial biomass for the apphed N. The retention and recovery of 5 N in the soil profile is affected by a combination of physical, chemical and biological factors (Overrein 1972). It appears that biological factors were the main ones affecting the final ' 5 N recovery in the soil, because the physical and chemical factors were held constant in the experimental design. The presence of abundant understory vegetation in the 79 control plots perhaps resulted in more 1 5 N being taken up initially by plants in the plot and might also encourage re-mineralization (because plants can increase substrate availabihty to soil microorganisms, Fisher and Gosz 1986) and uptake by plants of the ' 5 N hrunobilized by the soil organic matter, thus reducing the amount of 5 N in the soil. Nbudget There have been few studies of forest ecosystems that provided complete balance sheets for applied fertilizer N, whether using conventional methods or 5 N tracer technique (Melin and Nommik 1988). The total recovery in the soil-plant system in this study ranged from 57 to 87% of the total applied. The part that was not accounted for was presumably lost from the system. The most probable pathway for this loss might be leaching. Other mechanisms for N loss include transport of N out of the plots by roots extending beyond the plot boundaries, litter falling outside of the plots or carried by wind (Heilman et al. 1982a). The heterogeneous nature of the soil causes sampling errors which may lead to under- or over-estimation of N recovery in the soil- plant systems. Nitrification, and thus denitrification and volatilization loss, were probably rninimal as this soil has very low pH (Bjorkman et al. 1967; Melin et al. 1983). In a comprehensive study of the effect of clearcutting on forest floor N dynamics on some mesic sites on southwestern Vancouver Island of the wetter maritime CWH biogeoclimatic zone, Martin (1985) found that nitrification was substantial during in situ incubations using the buried- bag method, although the forest floor was strongly acidic (pH in 0.01 M CaCh ranged from 3.2 to 3.7). The forest floor samples incubated were sieved before returning to the field and this might have caused some artificial effect for the in situ incubation; leaching loss of nitrate in the first 10 years post-clearcutting was found to be about 20 kg N ha"1 yr"1; high potentials for denitrification were found especially in the recent clearcuts; a negative correlation between initial nitrate-N concentration and nitrification rates suggested that denitrification might be rapid. These observations illustrated the need for further investigations on N budget of the salal-dominated sites on northern Vancouver Island. Conclusions Western hemlock and Sitka spruce seem to be more responsive to removal of competing vegetation than western redcedar in terms of applied 5 N uptake by above- and belowground tree components. Based on the limited data obtained in this study, I propose that if N fertilization after understory control is possible, western hemlock and Sitka spruce stands should have the priority. However, if fertilization is to be carried out without pretreatment for controlling understory vegetation, western redcedar stands should have the priority. In the control plots, understory took up more 1 5 N than trees indicating that salal was a persistent competitor for N and N availability was reduced by uptake by understory biomass. Fertilization on those CH clearcut sites brings a practical problem: on the one hand, young plantations around 8-10 years old experience nutrient deficiencies and growth stagnation and are in need of external nutrient additions; on the other hand, fertilization on those young plantations has a low efficiency. Despite low fertilizer recovery efficiencies, fertilization of those sites has been shown to be effective in improving plantation growth. 81 CHAPTER 5. TRANSFORMATIONS OF RESIDUAL 1 5 N IN A CONIFEROUS FOREST SOH, ON NORTHERN VANCOUVER ISLAND Introduction Nitrogen applied to forest ecosystems is immobilized quickly by forest floor material and soil organic matter (Nornmik and Larsson 1989; Preston et al. 1990). This process leads to the accumulation of N in the upper horizons of the soil profile, with up to 76% of the apphed N found in the soil (Heilman et al. 1982a; Nommik and Larsson 1989; Nommik and Popovic 1971). Despite the recent report showing the availability of dissolved organic N in very acidic and infertile forest sites (Northup et al. 1995), the availability of apphed N to crop trees is low once it is immobilized in the soil organic matter (SOM) since most of the immobilized N is found in the recalcitrant organic matter fractions shortly after fertilizer is apphed to the forest. For example, Preston and Mead (1994) showed that 8 years after 15N-labeled fertilizer apphcation to a lodgepole pine stand in the interior of British Columbia little additional 1 5 N had been taken up by plot trees as compared to that recovered 8 months (one growing season) after the apphcation. The availability of immobilized N in forest soils is affected by many factors, both biotic and abiotic. One of the major controls on N cycling and availability in forest ecosystems is the transformation of N in the soil profile (Schimel and Firestone 1989a and b). Past studies on N mineralization and availability have been focused on the seasonal changes of mineralization (Harris and Riha 1991; Nadelhoffer and Aber 1984), mineralization rates in different ecosystems (Adams and Attiwill 1982; White and Gosz 1987), and the effect of management practices on N mineralization (Polglase et al. 1992). On the other hand, most fertilization studies focus on the effects of fertilizer apphcation on soil and fohar nutrient status, while the effects of fertilization on 82 nutrient recycling are often neglected (Aarnio and Martikainen 1994). Martikainen et al. (1989) studied the mineralization of C and N in soil samples taken from three pine stands where fertilizer had been applied seven years before the sampling. However, few studies have reported the differences in N transformation in forest soils with fertilizer N applied at different times. Furthermore, the mechanism governing the retention of N in the soil organic matter following N- fertilization of forest soils is not clear (Bosatta and Berendse 1984). In the Salal-Cedar-Hemlock Integrated Research Program (SCHIRP) there have had field experiments to study the fate of fertilizer N in CH cutover sites (Chapter 4). The experimental arrangement provided plots with 15N-labeled fertilizer applied at different times. The objective of this study was to investigate the effect of labeling duration on N mineralization (both total and labeled N), microbial biomass N pool size, and long-term fertilizer N availabihty. The aim was to provide information on N transformations in those samples to explain the nature of low availabihties of the residual 1 5 N after one growing season. Materials and methods The site A general description of the study site can be found in Chapter 3. The site used for this study is the same for Chapters 3 and 4. A description of soil properties is presented in Chapter 4 (Table 4.1). Nitrogen-15 labeling and sampling Nitrogen-15 labeled ( N F L t ^ S C ^ (200 kg N ha"1) was applied to microplots on April 16, 1991, April 24, 1993, and December 1, 1993. Samples were collected from those microplots on 83 December 2, 1993, to give material which had been labeled with l 5 N for 24 hours, 7 months, and 31 months. For the 31-month treatment, samples were collected from 1 5 N microplots estabhshed for a long-term fertilization study (Chapter 4) on the same site. The microplots with understory removal were used. Nitrogen-15 enrichment in the fertilizer solution apphed was 3.38044%. For the 7-month and 24-hr treatments, the aboveground understory vegetation was removed before applying the prepared fertilizer solution which had an 1 5 N enrichment of 2.37753%. In all situations, fertilizer was made into solution in 2 L water for each plot and apphed using a watering can with a long neck. The fertilization treatments were rephcated three times. The top 10 cm of the H horizon material was collected from the whole plot in the field. Approximately 12 kg of sample was collected from each plot and was treated as one sample. After the samples were brought into the laboratory, they were picked free of visible roots, sieved through an 8 mm sieve, thoroughly mixed using a Monarch (Winnipeg, Canada) cement mixer, and subsampled (approximately 1 kg) for mineralization and chemical extraction studies in the laboratory. Laboratory methods Water holding capacity (WHC) of the samples was measured essentially following the method described by Jenkinson and Powlson (1976). Briefly, fresh soil (50 g) was placed in a conical filter funnel fitted with a small glass wool plug. The bottom of the funnel was closed by closing (using a chp) the short piece of rubber tubing attached to the stem of the funnel. Then 50- mL of distilled water was added to the funnel and the chp was opened after 30 minutes to measure the excess water drained. The difference between the volume of water added and that of drained, plus the water contained in the fresh soil was used as the soil's water holding capacity. 84 Soil pH was measured using a pH meter with 1:2 (v/v) soil:0.01 M CaCh or water. Total soil N was measured by the Kjeldahl digestion and steam distillation method (Bremner, 1965). The distillates were collected in boric acid and titrated. For laboratory incubations where fresh samples were used, incubation commenced within a week after sampling. For the aerobic incubation experiment, an equivalent of 6 g (oven-dry basis) fresh sample was weighed into a 100-mL plastic container; each sample was duplicated. The soil moisture content of the samples was adjusted to 60% of the WHC before the incubation commenced. The five san l̂ings were at day 4, 10, 18, 28, and 42 plus a baseline sampling at day 0. The plastic containers containing a sample, or without a sample (blanks), were placed into 1-L Mason jars (previously aerated using compressed air). They were wetted with a few drops of water in the bottom of the jar and loosely capped to minimize water loss but allowing gas exchange between the jar and the environment. For the jars to be sampled on the 4th day of the incubation (randomly picked from the pool of jars of the whole incubation), glass bottles (40-mL in volume) containing 20-mL 1 M NaOH were included in the jars to absorb C O 2 released during the incubation; the jars were tightly sealed and incubated in the dark in incubators at 22 °C. The jars were randomly placed in the incubator. Soil respiration was only measured (by the NaOH absorption method) for the jars to be sampled on the 4th day at this stage, because soil respiration was measured for each incubation interval (e.g., for days 0-4, 4-10, 10-18, 18-28, and 28-42). The cumulative C mineralization is the sum of C mineralized in each incubation period. A set of baseline samples (day 0) was extracted for initial mineral N content by adding about 50-mL 2 M KC1 (the exact amount in grams recorded and converted into volume according to its volume:weight ratio at room temperature) to the plastic containers containing the weighed organic material at the time the incubation commenced. To control the solution volume added to 85 about 50-mL is to maintain a stable solution volume:sample weight ratio. The containers were then shaken for 1-hr. The extracts were filtered through Whatman #42 filters on a vacuum filtration system The extracts were stored frozen until analysis for then inorganic and organic N contents. This set of data was used as a baseline to calculate the net N mineralization or immobilization during the incubation. Nitrate and nitrite N was measured on selected samples after incubation, but no significant nitrate and nitrite N was found in those samples before or after incubation. On each scheduled sampling date, the respective jars were removed from the incubator. The N mineralized after the incubation was extracted as described above. Net N mineralization was calculated as: inorganic N extracted at a sampling date - initial inorganic N content before incubation. The amount of CO2-C absorbed in the NaOH solutions during the 4-day incubation was analyzed using back titration. A portion (1-mL) of the 1 M NaOH solution was transferred into a 50-mL beaker and 20-mL de-ionized water was added. The solution was titrated to pH 10 using 0.1 MHC1. On the same date, Mason jars for the next sampling were aerated using compressed air and set up with NaOH traps. This was to ensure that the initial ah composition in the jars was comparable for each incubation. Following the incubation period they were sampled and analyzed for C and N mineralization using the same procedure described above. Anaerobic incubation followed the method proposed by Powers (1980). An equivalent of 2 grams (oven-dry basis) of fresh organic material was weighed into glass vials and 20-mL de- ionized water added. There was about 20-mL head space left in the glass vials after the distilled water was added. The vial was inverted a few times to mix the contents and incubated at 30 °C 86 for two weeks. At the end of the incubation, the samples were transferred to 200-mL plastic bottles and 20-mL 4 M KC1 was added to bring the final KC1 concentration to 2 M. After shaking for 1 hr, the samples were filtered through Whatman #42 filters. The flush of N extracted using 0.5 M K 2 S0 4 after chloroform fumigation was used to study the transformation of inorganic N into microbiaUy immobilized form The fumigation-extraction method was modified from Brookes et al. (1985b) as is described in Chapter 7. The extracts from both the aerobic and anaerobic incubation experiments were analyzed for NH4-N by steam distillation with MgO and titration (Keeney and Nelson, 1982). The mixture after steam (hstillation for samples from the time-0, day 28 and day 42 samplings was analyzed for organic N by the semimicro-Kjeldahl digestion and steam disthlation method (Bremner, 1965). The extracts from the chloroform firmigation-extraction experiment were analyzed for total N (inorganic + organic N) also by the semi micro-Kjeldahl digestion and steam distillation method (Bremner, 1965). The distUlates, collected in boric acid-ethanoL were dried at 70 °C, and the ammonium N was converted to dinitrogen gas using the Rittenberg reaction with alkaline hthium hypobromite, and analyzed for I 5 N abundance using a Vacuum Generators Sira 9 mass spectrometer (Preston et al. 1990). Due to the low inorganic N content in some of the samples (31- and 7-month treatments), spiking with 0.600 mg N was used to bring the final N content to greater than 0.600 mg N per sample before 1 5 N analysis. Data analysis Applied N in the extracted fractions was calculated using the following formula (He et al. 1988): Applied N = A(X - NA)/(Y-NA) where A is the total N in each fraction and X is the 1 5 N abundance of that fraction; Y is the 1 5 N abundance of the apphed fertilizer and NA is the natural abundance of the sample. Note that in this study "total N" refers to the sum of native and apphed N in any analysis. Statistical analyses were performed using the commercial Statistical Analysis System software (SAS Institute Inc., 1989). Means and standard errors of the means for each treatment and sampling date were calculated using the PROC MEANS statement and analysis of variance on independent variables was accomplished by using the General Linear Models (GLM) procedure due to missing values. In some cases, group means of independent variables were compared between treatments for each sampling date using Scheffe's multiple, range test. For N pool sizes before and after the anaerobic incubation and the flush of N after fumigation, group means of independent variables were also compared between treatments by Scheffe's multiple range test. Results Selected properties of humus materials The selected properties of humus material from different plots are presented in Table 5.1. Soil reaction was acidic for the humus layers. Fertilizer N apphcation somewhat increased humus N content in the 24-hr treatment above the level found in the other two treatments because there should be very little of the apphed N lost in such a short period and most of the apphed N was retained in the system. There seemed to be differences in WHC and water contents of the humus materials among the treatments. However, when water content was expressed as a percentage of WHC, they were not much different from each other. Aerobic incubation Table 5.1. Selected properties of humus materials pH pH Soil Water Treatment water CaCl2 N WHC water as % of (1:2 v/v) (1:2 v/v) (gkg1) (kg kg1) (kg kg1) WHC 24-hr 3.69 3.43 12.20 5.04 3.64 72.1 7-month 3.88 3.13 9.78 5.68 3.93 69.2 31-month 3.94 3.16 9.63 4.85 3.12 64.3 Table 5.2. Analysis of variance for the effect of treatment and sampling date on total and applied N and carbon mineralization in a 42-day aerobic incubation Total N fug g"1) Applied Nfug g1) CO?-C (mg g') Variable df F P F P F P Treatment 2 447.34 0.0001 96.17 0.0001 9.06 0.0008 Date 4 7.29 0.0003 2.05 0.1137 66.50 0.0001 T x D 8 5.26 0.0004 2.19 0.0588 2.49 0.0335 90 Fig. 5.1 presents the total and apphed N mineralized during the 42-day aerobic incubation. Total N here refers to the sum of N mineralized which includes the apphed N and native N. These data are net N mineralizations obtained by subtracting the initial inorganic N from that in the incubated samples. Measurements were based on ammonium-N as nitrate-N was always below detection limit. After incubation for four days, samples from ah of the treatments produced a net mineralization of total N ranging from 2.05 ug g"1 in the 7-month treatment to 188.83 ug g"1 in the 24-hr treatment (Fig. 5. la). For the 7- and 31-month treatments, from day 4 to day 10 and from day 10 to day 18, net immobilization of total N occurred (negative slope of the respective lines). For these two treatments, net mineralization of total N occurred from day 18 to day 42. For the 24-hr treatment, net mineralization of total N occurred until day 28 and then net immobilization (negative slope) occurred in the last incubation period. Statistical analysis showed that there were significant differences between treatments and sampling dates (Table 5.2). Significant interactions between sampling date and treatments were obtained due to the changes from net mineralization to net immobilization, or vice versa, during the sampling period. The process of net mineralization for N derived from the apphed fertilizer followed quite closely the pattern of net mineralization of total N, especially for the 7- and 31-month treatments (Fig. 5. lb). For the 24-hr treatment, net mnnobilization of 1 5 N was observed from day 4 to day 10, instead of net mineralization in the case of total N. The magnitude of difference in the amount of apphed N being mineralized/immobilized was much greater than that of total N. Taking the inorganic N pool size before incubation as the reference point, net mineralizations were observed in the samples taken 24 hours and 31 months after fertilization throughout the incubation period (6 weeks); however, in the samples taken 7 months after fertilization, net immobilization was observed for both total and apphed N (Fig. 5.1). The percentage of apphed N recovered in the 91 total N mineralized ranged from 76.6 to 87.4%, 13.1 to 42.0% and 10.6 to 14.0% for the 24-hr, 7- and 31-month treatments, respectively. The organic N extracted by 2 M KC1 can be viewed as an intermediate phase between inorganic N (highly available) and insoluble organic N (unavailable). Microorganisms can take up inorganic N (immobilization) and low-molecular-weight soluble nitrogenous organic compounds (Molina et al. 1983; Hadas et al. 1987). Only after the organic N assimilated by the microbial biomass is released (mineralization) will the N be rendered available. There is not enough evidence to suggest that low-molecular-weight soluble nitrogenous organic compounds are available to all higher plants. Both total and applied N in the extracted organic fraction showed an increase from day 0 to day 42 (Fig. 5.2), resulting in non-significant treatment x sampling date interactions (ANOVA data not shown). These results showed that longer incubation encouraged solubilization of organic N. However, the sharp increase for the total and apphed N in the soluble organic fraction of the 24-hr treatment at day 42 seemed to have come from the irnmobilization of inorganic N. There were significant differences in total and apphed N extracted in the soluble organic fraction among the samples with 1 5 N labeled fertilizer apphed at different times. The percentage of apphed N recovered in the extracted organic N fraction was from 30.8 to 37.2%, 15.7 to 18.5% and 9.6 to 12.2% for the 24-hr, 7- and 31-month treatments, respectively. Fig. 5.3 shows that the 7-month treatment always had the highest organic carbon mineralization rates and the 31-month treatment always had the lowest rates. However, for the 24-hr treatment, the carbon mineralization rate went from high to low (relative to the other two treatments) during the incubation. Thus, significant treatment x date interaction was observed (Table 5.2). Fig. 5.3 also shows that the amount of carbon mineralized from all the treatments was constantly rising from one incubation period to the next (upward trend of the slope for the 92 Fig. 5.1. Net mineralization (NFJ/-N) of total and applied mineral N in samples with N added at different times in an aerobic incubation experiment. Vertical bars are standard errors. c © >_ 3 O • j 45 40 U) c 35 30 w 4 25 +•» c 20 o 15 E • 10 co 73 5 C CO - 0 1 I*. - -5 10 18 28 42 93 Fig. 5.2. Total and applied N found as KCl-extractable organic N at day 0, 28 and 42 in the aerobic incubation. Vertical bars are standard errors. 250 T3 <D -g _ 200 o • .2 o)100 50 0 • 24-hour • 7-month M 31 -month Day 0 Day 28 I (5.2a) Day 42 100 • O 9 0 £ -o 80 8 1 70 § a 60 © « 50 40 30 .2 O ) §) 3 20 <5 10 0 (5.2b) Day 0 Day 28 Day 42 Days of incubation 94 lines connecting adjacent two points), but the daily mineralization rate actually had a decreasing trend from day 1 to day 42 due to the increased sampling intervals. Both treatment and incubation period (sampling date) had significant effects on carbon mineralization (Table 5.2). Correlation analysis showed that mineralization of C and net mineralization of N were poorly correlated with each other (r=-0.0201, p=0.8960). Anaerobic incubation In the fresh samples before incubation, there were significantly greater amounts of total N extracted by 2 M KC1 in the 24-hr than in the 7- or 31-month treatments (Fig. 5.4a, p<0.05). The extremely large initial N pool size in the 24-hr treatment was caused by the addition of fertilizer N in the field, as most of the total N was composed of apphed N. For the apphed N extracted, the difference was significant only between the 24-hr and the other two treatments (p<0.05). After two weeks of anaerobic incubation at 30 °C, the N pool sizes measured were significantly different among the treatments for the total N but differed significantly only between the 24-hr and the other two treatments for the apphed N mineralized. The result shows a net mineralization of total and apphed N for the 7- and 31-month treatments. For the 24-hr treatment, there was a net immobilization of 132.38 and 238.81 ug"1 g for the total and apphed N, respectively. The higher net immobilization for apphed N than for total N shows that there was actually net mineralization for native soil N during the anaerobic incubation from the samples with fertilizer N labeled for 24 hours. Nflush from fumigation 95 Table 5.3. Applied N as a % of total N in various measurements Measurement 24-hr 7-month 31-month Initial N 87.4 20.5 12.2 N flush after fumigation 43.4 12.9 11.8 Net mineralization at day 42, A l 1 22.9 16.5 7.3 Net SON1 increases at day 42, A l 39.8 18.1 8.6 Net mineralization, ANI1 (180.4)2 28.9 22.4 l: A l - aerobic incubation; SON - soluble organic N; ANI - anaerobic incubation 2: The numbers in parenthesis represent net immobilization; the number is greater than 100% because more 1 5 N was immobilized than total N. 96 Fig. 5.3. Cumulative C mineralization in samples with N added at different time in an aerobic incubation experiment. Vertical bars are standard errors. 97 Fumigation of the fresh samples with chloroform for 24 hours resulted in a flush of both total and apphed N in the extracts (0.5 M K2SO4) (Fig. 5.5). The flush of N, which represents the size of N pool in the microbial biomass was in the order of 24-hr > 7-month > 31-month treatments for both total and apphed N. Differences were significant between the 24-hr and the other two treatments for the flush of total and apphed N after chloroform fumigation (p<0.05). Discussion Samples with fertilizer N incorporated for differing durations exhibited a wide range in net mineralization. This was true for total and labeled N, accumulation of total and labeled organic N during aerobic incubation and the immobilization of labeled and native N by soil microbial biomass. However, nitrate concentration or nitrification activities in fresh samples from the site were usually under detection limit, when measured in this and some earher studies. Therefore, it is not surprising that nitrate concentrations were not significantly increased during aerobic incubation of samples from the 24-hr, 7-month, and 31-month treatments, since nitrification can be suppressed by low pH and low ammonium levels (Alexander 1977; White and Gosz 1987; Vitousek and Matson 1988), or by inhibitory compounds (Donaldson and Henderson 1990b). It may also take longer time for the nitrifiers to build up and for nitrification to show up (Binkley and Hart 1989). Although Martin (1985) found nitrification in strongly acidic forest floors on southwestern Vancouver Island, lack of nitrification in this study after a prolonged incubation indicated that nitrification was inhibited in the samples I studied. The reasons for the differences are the subject of further studies. In the aerobic incubation study, the large amount of total and apphed N mineralized in the 24-hr treatment appears to be due to the following reasons. First, the short labeling duration in 98 Fig. 5.4. N pool sizes (total and applied) before and after anaerobic incubation in samples with N added at different time. Vertical bars are standard errors. 1200 Total N Applied N 1200 140 Total N Applied N Type of N 99 Fig. 5 .5 . Flush of N (total and apphed) after chloroform fumigation in samples with N added at different times. Vertical bars are standard errors. 700 O) 600 o c 500 "F 400 g 300 t 200 E 100 • 24-hour • 7-month M 31 -month Total N Appl ied N Type of nitrogen 100 this treatment means that the apphed N was less stably bonded to the organic matter than in either of the other two treatments. This is supported by the higher amount of apphed N recovered in the soluble organic N fraction of the 24-hr treatment. Apparently the foUowing balance exists in the soil system (Hadas et al. 1992; Khanna and Ulrich 1984; Duxbury and Nkambule 1994): mineralization solubilization Inorganic N / = = = = = = = / Extractable organic N /========= / Stable organic N immobilization (and microbial biomass) humification The extractable organic N is the active soil organic N in Duxbury and Nkambule (1994). Asmar et al. (1994), in studying the effect of extracellular-enzyme activities on the solubilization of soil organic N, based then experiment on a similar model to investigate the relationships among inorganic N, soluble N (0.5 M K2SO4 extraction), and the recalcitrant fractions of SOM. The higher amounts of total and apphed organic N extracted in the 24-hr treatment than in the other treatments indicated that there was a bigger pool of mineralizable organic N for potential mineralization in the samples from the 24-hr treatment. Second, with little of the recently apphed N (at 200 kg N ha"1) lost to volatilization or leaching in the 24-hr field labeling period, the C/N ratio of the studied material from the 24-hr treatment was much less than those from the other two treatments; thus, a moderate turnover (carbon mineralization) rate resulted in a high net N mineralization. Lastly, the greater amount of native soil N mineralized in the 24-hr treatment demonstrates the 'priming effect" resulting from the apphcation of the inorganic N (Jenkinson et al. 1985). Initial mineral N content and net N mineralization rate were lower in the 7-month than in the 31-month treatment (Fig. 5.4a and 5.1) but the percentage of apphed N in the initial mineral N and net N mineralized was higher in the former than in the latter (Table 5.3). The foUowing 101 scenario is proposed for the observed results. Seven months after fertilizer N apphcation, the apphed N was undergoing strong immobilization after the initial priming effect from the N apphcation. Therefore, little N could be extracted by 2 M KC1 as NFV-N or mineralized from the 7-month treatment. Evidence from the microbial biomass N measurement indicated that more N was immobilized in the 7-month than in the 31-month treatment. Asmar et al. (1994) also found that decreases in total soluble N (NO3-N + NFL-N + soluble organic N) and NO3-N coincided with increases in total soluble N in the microbial biomass, due to immobilization by the soil microbial biomass. TJhrty-one months after fertilization, mineralization/immobilization was more balanced and most of the I 5 N went into the more recalcitrant fractions and was diluted after prolonged mmeralization-immobilization exchange (Stams et al. 1990). Therefore, more total N but a smaller proportion of the apphed N can be mineralized from the 31-month than from the 7- month treatment. The pattern of net mineralization over incubation time for the apphed N of the 24-hr treatment was different from that of the total N. However, it was roughly similar for the 7- and 31-month treatments. This shows that for the 7- and 31-month treatments, apphed N was further equilibrated with the native N and the proportion of apphed N in the total amount of N mineralized or immobilized was rather constant during the aerobic incubation. The higher (p<0.05) amounts of the total and apphed N contained in the extracted organic fraction (Fig. 5.2) in the 7-month than in the 31-month treatment demonstrated that apphed N was less stabilized in the former than in the latter. Systems having substrates with C/N ratios less than a critical C/N ratio (which equals the biomass C/N ratio of the microbial decomposers divided by the assimilation efficiency) can be defined as energy (C)-deficient systems (mineralization predominates over hmnobilization) 102 (Bosatta and Berendse 1984). Assimilation efficiency is the ratio of microbial production to assimilation. Similarly, systems having substrates with C/N ratios greater than the critical C/N ratio are defined as N-deficient systems (immobilization predominates over mineralization). Their model predicted that, in response to either N or C additions, N mmeralization-immobilization in N-deficient systems would oscillate between increased N mineralization and net immobilization in their return to steady-state conditions. The samples for this study were taken from a N deficient site. Therefore the cyclic nature of N nrmerahzation-inmiobilization in the samples with N apphed for various periods supported the predictions of the model proposed by Bosatta and Berendse (1984). Stams et al. (1990) reported that 43 to 65% of the apphed 1 5 N was incorporated into the organic N fraction 98 days after an aerobic incubation in two acidic forest soils (about 25% remaining as mineral N and the rest unaccounted). Then work demonstrated that the incorporation was the result of biological isotope exchange through simultaneous mineralization and immobilization or through a passive exchange reaction of enzymes involved in microbial metabolism, while incorporation through chemical exchange and autotrophic nitrification- mediated nitrosation was negligible. Incorporation of apphed 1 5 N into the organic N fraction in my study was much faster than that reported by Stams et al. (1990), i.e., the percentage of apphed 1 5 N incorporated into organic fractions in 24 hours was calculated to be about 47%. During the aerobic incubation, higher carbon mineralization (7-month vs. 31-month treatment) did not yield higher net N mineralization, and similar C mineralization rates (24-hr vs. 7 and 31-month treatments) were associated with a wide variation in net N mineralization. This might be caused by the changes in the C/N ratios of the microbial population (Kelly and Henderson 1978), due to fertilization, and might have reflected the poor measure of gross N 103 mineralization by the net N mineralization (Davidson et al. 1992). This gave a poor correlation between C and N mineralization (r=-0.0201, p=0.8960). Poor correlation between C and N mineralization after N fertilization was also observed by Johnson et al. (1980) and Martikainen et al. (1989). The dynamics of microbial respiration in those samples can be readily explained by the general model for mineralkation-inmobilization of Bosatta and Berendse (1984). Since the samples were N-limited, osculation of C mineralization around steady-state was expected. In the literature, N mineralized from anaerobic incubations has been found to correlate well with tree growth response (Keeney 1980) and thus may be a rehable biological index for assessing N availability. The anaerobic incubation results for the 24-hr treatment were somewhat different from those of the aerobic incubation for the same samples. A comparison of Fig. 5.4a and 5.4b shows that a net immobilization occurred in the 24-hr treatment while net mineralization occurred in the other two treatments. The net immobilization in the 24-hr treatment suggests that anaerobic incubation encouraged immobilization of inorganic N when there was a large pool of mineral N in the soil. I have not seen any report studying the dynamics of mineralization- immobilization during anaerobic incubations, therefore, I do not have any evidence to suggest that the net immobilization observed is a random effect, i.e., N mineralization under anaerobic condition has a cychc nature. Adams and Attiwhl (1986) noted that N mineralized during anaerobic incubation was always greater than that during in situ incubations. They and Myrold (1987) concluded that anaerobic incubations are similar to soil fumigation methods where microbial biomass N is measured when N is released through decomposing dead microbial cells. The fumigation-extraction results for the 24-hr treatment showed a positive flush of total and apphed N while the anaerobic incubation showed net immobilization. Therefore, I propose that the conclusion made by the above authors 104 is not appropriate for samples containing high amounts of inorganic N. Griffiths et al. (1990) also suggested that the anaerobic N mineralization and the biomass C measured with the chloroform fumigation method did not measure the same sources of organic N and C for samples of ectomycorrhizal mat and nonmat soils of Douglas-fir forests. The flush of N after chloroform fumigation was the result of N released from microbial cells killed by the fumigant (Jenkinson and Powlson 1976; Brookes et al. 1985a). The fumigation- extraction result shows that apphed N had been immobilized quickly by the microbial population in the soils. Assuming that little apphed N was lost to volatilization and inorganic N was not leached beyond the 10 cm sampling depth in 24 hours, about 14.5% of the apphed N was immobilized by microbial biomass in about 24 hours, while the majority of the apphed N stayed in inorganic form in the soil solution. I used 0.177 Mg m"3 as the bulk density value for the top 10 cm H layer. The quick immobilization of inorganic N reduced the possibility of fertilizer N loss through leaching or volatilization (Adams and Attiwill 1986; Vitousek and Matson 1985) which is of very practical importance. Data summarized in Table 5.3 show that there was always a greater percentage of apphed N in the total N extracted in the 24-hr treatment than in the other two, for all of the measurements made. The difference between the 7- and 31-month treatments was generally not significant. This again illustrates that the relative availabihty of apphed N was reduced from the 24-hr to the 7- and 31-month labeling. In summary, immobilization of inorganic N by microbial biomass was quick and was the major sink and transformation pathway for the apphed N. Immobilized N in the microbial biomass was gradually transformed into more stable organic N form through humification or transformed into inorganic forms through mineralization. Newly immobilized N is highly 105 mineralizable. However, when the soil samples were incubated under anaerobic conditions, net immobilization was encouraged when an excess amount of inorganic N was present in the soil solution, pushing the balance from mineralization to immobilization. The big drop in the available mineral N (both total and apphed ), net N mineralization, soluble organic N and microbial N, from the short-term (24-hr treatment) to the long-term (7- and 31-month treatments) indicates that the availabihty of apphed N declined to a minimal level after one growing season. 106 CHAPTER 6. INCORPORATION AND EXTRACTABILITY OF RESIDUAL 1 5 N IN A CONIFEROUS FOREST SOIL ON NORTHERN VANCOUVER ISLAND Introduction The incorporation of fertilizer N into different soil organic matter (SOM) fractions is one of the major controls on N cycling (Schimel and Firestone 1989a and b). The longer since the N has been apphed, the more apphed N is immobilized and stabilized through the repeated mineralization-irnmobilization process (Jansson and Persson 1982). However, the mechanisms for apphed N stabilization are largely unclear. Therefore, knowledge about factors affecting variations in the availabihty and extractability of apphed N immobilized by SOM may improve our understanding of the N immobilkation-mineralization process (Legg et al. 1971). In trying to find the best extraction method to measure the availabihty of N in soils, researchers have developed various protocols; for example, extraction with HF (Stevenson et al. 1967), HC1 hydrolysis (Stewart et al. 1963; Porter et al. 1964), extraction with salts, e.g., 0.01 M NaHC03(MacLean 1964), 0.5 M K 2 S0 4 (Brookes et al. 1985a), and 2 M KC1 (Bremner 1965), autoclaving in water or 0.01 M CaCl2 (Stanford and DeMar 1969), and organic matter fractionation (He et al. 1988; Azam et al. 1989a). Extraction with salts (2 M KC1 and 0.5 M K 2S0 4) and autoclaving with 0.01 M CaCl2have been the most widely used methods that give a measure of the available N (Keeney 1982). While studies on N incorporation with labeled N apphed for a prolonged period (McGill and Paul 1976; Smith and Power 1985) and for a very short periods, e.g., from a few hours to a few days (He et al. 1988; Schimel and Firestone 1989a) are available, studies are needed 107 encompassing samples with N apphed for both short and long terms. A study on the sequential changes in residual N extractability in samples with N apphed at different times may yield useful information on the characteristics of N incorporation into different SOM fractions. In this way, the incorporation of apphed N into SOM fractions is viewed as a dynamic process. The set-up of the SCHIRP program provided plots with 15N-labeled fertilizer apphed at different times, thus aUowing us to examine the effect of labeling duration on the characteristics of residual fertilizer N in forest soils. Residual N is defined as the total of apphed N not lost from the soil system The objectives of this study were (1) to determine the effect of labeling duration on the extractability of both total and apphed N using various extraction methods; and (2) to investigate the mechanism of stabilization of apphed N into the recalcitrant fractions of organic matter. Materials and methods The site, field sampling and sample preparation A detailed description of the study site can be found in Chapter 3 and 4. Field labeling and sample preparation is presented in Chapter 5, as this Chapter used the same samples as Chapter 5. Laboratory and statistical analyses Fresh samples were used for the extraction studies. Approximately 50 g (wet weight) sample was weighed into a 200-mL plastic bottle and about 50-mL of 2 M KC1 was added. The mixture was shaken for 1 hour and filtered through Whatman #42 filters on a vacuum filtration system. The extracts were kept in a freezer until analysis for total and 1 5 N. A similar procedure was used for extraction with 0.5 M K 2 S0 4 , except that the mixture was only shaken for 30 108 minutes. The extraction using 0.5 M K 2 S0 4 is identical to that of extracting the initial N in the fumigation-extraction method for quantifying microbial N. The autoclaving procedure used was as follows: 3 g of soil sample (air-dry) was weighed into a 50-mL polyethylene test tube which can resist the heat and pressure of the autoclaving. Samples in the test tube were amended with 25-mL 0.01 M CaCl2 and autoclaved for 16 hours (121 °C and 1.04xl05 Pa). After the samples were cooled, they were centrifuged and filtered. Approximately 25-mL of 0.01 M CaCl2 was used to wash the residue, which was again centrifuged and filtered. The combined extracts were made to 100-mL and frozen until further chemical analysis. For the firmigation-extraction experiment, fresh samples of approximately 25 g (wet mass) were fumigated with ethanol-free chloroform for 24-hr at room temperature (20 °C). After removal of the chloroform, the fumigated samples were extracted with 0.5 M K 2 S0 4 on an end- over-end shaker (150 rev. min."1) for 0.5 hr, followed by vacuum filtration through Whatman #42 filters (Brookes et al. 1985b; Chapter 7). Samples were extracted with 0.5 M H 2S0 4 , and with 0.01 M, 0.02 M and 0.05 M KMn0 4 (in 0.5 M H 2S0 4) for 1 hr. The oxidative release of NFL-N (data reported in Table 6.2) from SOM was obtained by subtracting NFL-N extracted by 0.5 M H 2 S0 4 from that extracted by the acidic permanganate. Organic matter fractionation followed the method of Schnitzer (1982). Briefly, a portion (15 g) of the ah-dried sample was weighed into a centrifuge bottle, amended with 100-mL of 0.5 M K 2 S0 4 and shaken for 1 hr. After centrifugation (at 1000 g for 30 minutes), the supernatant was discarded by decantation. This step was used to remove the K 2 S0 4 extractable N in the samples. The residue was then amended with 150-mL 0.1 M NaOH (1:10 soikNaOH solution) 109 and shaken for 24 hours at low speed. After centrifugation, the supernatant was recovered by decantation. The residue (the humin fraction) from this step was washed twice with about 40-mL deionized water. The recovered supernatant solutions were combined and acidified to pH 1 using 3 M H 2S0 4. After being left to settle for 24 hours, the mixture was centrifuged and separated into fulvic (FA) and humic (HA) acids. The HA was washed with deionized water twice and the solution recovered after centrifugation was added to the FA. The FA fraction was made to 200- mL and the HA fraction was redisoh/ed in 0.1 M NaOH and made into 100-mL. The humin fraction was transferred to a paper bag and air-dried. Extracts from the extractions and the autoclaving experiment were analyzed for NH4-N by steam distillation with MgO and titration (Keeney and Nelson 1982). The FA and HA solutions were measured for total N by the Kjeldahl digestion and steam distillation method (Bremner 1965). The distillates from the above inorganic or total N measurements, collected in boric acid- ethanol, were dried at 70 °C and analyzed for 1 5 N abundance as described in Chapter 5. Spiking with a standard was used for some of the samples before 1 5 N analysis also as described in Chapter 5. According to Azam et al. (1989b), the extractability ratio first proposed by Legg et al. (1971) can be simplified as follows: Extractability ratio = (Atom% 1 5 N of the extracted N)/(Atom% 1 5 N of the total soil N) If the incorporated fertilizer N has the same chemical extractability as that of the total soil, the extractability ratio will be one (Legg et al. 1971). The calculation of apphed N in each extracted fraction followed He et al. (1988). Therefore, "total N" in this Chapter refers to the sum of native and apphed N in any analysis. 110 Statistical analysis was performed using the commercial Statistical Analysis System (SAS) software (SAS Institute Inc. 1989). Group means of independent variables were compared between treatments for each extraction method (or permanganate strength) or between methods (or concentrations) of each treatment using Scheffe's multiple range test if interactions were significant. Otherwise, multiple comparisons were performed for treatment and method (or concentration ) means. Results Effect of labeling duration on N extractability Analysis of variance results showed that there were non-significant treatment x extraction method interactions for the total N extracted (ANOVA not shown). Multiple comparison analysis showed that there was no significant difference for the total amount of N extracted between the 7- and 31-month treatments; however, a greater amount was extracted from the 24-hr treatment than from the other two treatments (p<0.05, Table 6.1). The total N extracted represented from 8.45 to 13.38%, 0.03 to 3.54%, and 0.15 to 4.51% of the total soil N, for the 24-hr, 7- and 31-month treatments, respectively, depending on the extraction methods used. Significant treatment x extraction method interactions (ANOVA not shown) for the percentage of total soil N extracted, the apphed N extracted, the apphed N extracted as a percentage of total N extracted, and the extractability ratios, forced me to perform the multiple comparison on treatments based on each extraction method, or compare extraction methods for each treatment. It was clear that, regardless of the extraction method used, a greater percentage of the total soil N was extracted from the 24-hr than from the 7- and 31-month treatments (p<0.05). However, there was no Table 6.1. Extractability ratios and total and applied N extracted by various extraction methods in samples with 1 5 N labeled for different durations. Treat- Extraction method ment 2MKC1 0.5 M K 2 S0 4 Autoclave Fumigation Average Total N extracted (|ug g"1 soil) 31-month 20.66 22.87 634.19 191.57 217.32 a1 7-month 4.93 18.31 605.39 243.72 218.09 a 24-hour 1086.09 1032.06 1616.44 1632.84 1304.36 b Average 370.56 A 2 357.75 A 952.00 B 639.38 C Total N extracted as a % of total soil N 31-month 0.15 a A 0.16 a A 4.51 a B 1.37aA 7-month 0.03 a A O. l laA 3.54aB 1.43 a AB 24-hour 8.89 b A 8.45 b A 13.26 bB 13.38 bB Apphed N extracted (ug g"1 soil) 31-month 2.52 a A 2.21 a A 29.76 a A 22.24 a A 7-month 1.01 a A 1.96 a A 32.74 a A 45.86 a A 24-hour 949.25 b A 907.83 b A 866.14 b A 1298.84 b B Table 6.1, continued next page 112 Table 6.1, continued Applied N extracted as a % of total N extracted 31-month 11.98 a A 9.09 a A 4.67 a A 11.48 a A 7-month 22.06 a A 10.70 a AB 5.43 a B 18.74 a A 24-hour 87.35b A 87.68b A 53.45 bB 79.36b A Extractability ratios 31-month 3.29 a A 2.52 a AB 1.16 a B 3.16 a A 7-month 5.09 ab A 2.48 a BC 1.17 a C 4.30 ab AB 24-hour 5.99 b A 6.00 b A 3.55 bB 5.43 b AB The same lowercase letters indicate that there was no significant (p=0.05) difference between the treatments for each extraction method. 2: The same uppercase letters indicate that there was no significant (p=0.05) difference between the extraction methods at each treatment level. 113 difference in the percent of total soil N extracted between the 7- and 31-month treatments (Table 6.1). The amount of apphed N extracted followed fairly closely that of the total N extracted. A greater amount of apphed N extracted by the KC1 and K 2 S0 4 methods in the 31- than in the 7- month treatment Vs a greater amount of apphed N extracted in the 7- than in the 31-month treatment resulted in a significant treatment x extraction method interaction. However, regardless of the extraction method used, the only difference was between the 24-hr and the other two treatments (p<0.05, Table 6.1). There was consistently a greater percentage of apphed N in the total N extracted for the 7- than for the 31-month treatment, although the difference was non- significant. The greatest percentage of apphed N extracted was in the 24-hr treatment. The extractability ratio increased from samples with 1 5 N labeled for longer periods to samples with recent 1 5 N labeling. Extractability ratios were all greater than 1, indicating that the extracted fractions were always more enriched with 1 5 N than the bulk soil. The significant treatment x extraction method interaction for the extractability ratio was probably caused by the outstandingly high number in the 7-month treatment of the KC1 extraction and the apparently low number in the 24-hr treatment of the autoclaving method. Effect of extraction method on N extractability On average, of the four extraction methods used, the autoclaving extracted the most amount of total N and the KC1 and K2SO4 methods extracted the least. (Table 6.1). There were no significant differences between the 2 M KC1 and 0.5 M K2SO4 extraction methods, except for the extractability ratio of the 7-month treatment. The autoclaving method extracted a greater percentage of the total soil N than the KC1 and K2SO4 methods (p<0.05), regardless of the 114 treatment (Table 6.1). Fumigation of the samples resulted in an intermediate percentage of the total N being extracted for the 7- and 31-month treatments, while a percentage similar to the autoclaving method was found for the 24-hr treatment. For the apphed N extracted, the only significant difference among the extraction methods was between fumigation-extraction and the other three methods for the 24-hr treatment. The greater amount of apphed N recovered in the fumigation-extraction was due to the release of the extra N immobilized in the microbial biomass. The amounts of apphed N recovered by the autoclaving and fumigation methods were higher than that recovered by the KC1 and K 2 S0 4 extraction methods for the 7- and 31-month treatments. However, the differences were not statistically different due to high variations in the data set. Comparing the apphed N and the total N extracted, it was obvious that the autoclaving method extracted a lower percent of apphed N in the total N extracted than the other three methods. The extractability ratios, which combine information on the 1 5 N abundance both in the extracted fractions and the bulk soil, were far more diverse than the other measurements. The main difference was a smaller ER value for the autoclaving than for the other three methods. The similarity in ER values among the fumigation, KC1 and K2SO4 extraction methods reveals that there was similar 1 5 N enrichment in the microbial biomass and in the extractable inorganic N. Extraction with acidic permanganate There was no treatment x concentration (of acidic permanganate) interactions for any of the parameters examined (ANOVA not shown), therefore, multiple comparison analysis was performed for treatment or concentration means (Table 6.2). 115 Table 6.2. Extractability ratios and total and applied N extracted by KMn0 4 in samples with 1 5 N labeled for different durations Treat- ment KMnC>4 concentration 0.01 M 0.02 M 0.05 M Multiple comparison 7-month M-C 1 31-month 7-month 24-hour M-C 31-month 7-month Total N extracted (pg g"1 soil) 31-month 31.24 20.52 24-hour 1480.28 44.14 32.34 1520.27 A 52.18 39.43 1505.54 A Total N extracted as a % of total soil N 0.22 0.12 12.12 0.31 0.19 12.45 0.37 0.23 12.33 A A A Apphed N extracted (Lig g"1 soil) 2.82 3.59 24-hour 1295.42 3.66 3.74 1302.28 4.26 4.18 1219.89 a b M-C Table 6.2, continued next page 116 Table 6.2, continued Applied N extracted as a % of total N extracted 31-month 9.39 8.03 7.90 a 7-month 17.82 11.67 10.62 b 24-hour 87.41 85.40 80.75 c M-C A AB Extractabihtv ratio B 31-month 2.67 2.21 2.15 a 7-month 4.07 2.66 2.49 a 24-hour 5.97 5.83 5.52 b M-C A AB B M-C: multiple comparison. The same lowercase (uppercase) letters indicate that there was no significant (p=0.05) difference between the treatments (concentrations). 117 Similar statistical significance results were observed for the total N extracted, apphed N extracted and total N extracted as a percentage of total soil N, i.e., there were no significant differences among the different KMn04 concentrations used and between the 7- and 31-month treatments; but greater amounts (or percentages) were obtained in the 24-hr treatment than in the other two treatments (p<0.05). However, the percentage of apphed N recovered in the total N extracted increased from the 31- to the 7-month, and from the 7-month to the 24-hr treatment (p<0.05). The percentage decreased as the strength of KMn0 4 increased, with the difference between the 0.01 M and 0.05 M KMn0 4 extraction significantly different (Table 6.2). The ER values also decreased as the strength of KMn0 4 increased corresponding to changes in the percentage of apphed N in the total soil N extracted. The extractability ratios were significantly greater in the 24-hr than in the other two treatments. Organic matter fractionation The percentage of apphed N recovered in the classical organic matter fractions is presented in Fig. 6.1. The material (humus) type and treatment interaction was significant (ANOVA not shown) due to the changes in relative 1 5 N distribution among the humic fractions. In the FA fraction, a significantly greater (p<0.05) percentage of the total recovered 1 5 N was in the 24-hr treatment than in the other two treatments. However, there was no difference in 1 5 N distribution among the treatments for HA. In the humin fraction, a significantly lower (p<0.05) percentage of apphed N was recovered in the 24-hr treatment than in the other two treatments which was opposite to the 1 5 N distribution in the FA fraction. Most of the recovered I 5 N was in the humin fractions for the 7- and 31-month treatments, while for the 24-hr treatment, most of the recovered 118 1 5 N was almost evenly distributed in the FA and humin fractions, with little in the FIA fraction (Table 6.3). The ER values, presented in Fig. 6.2, decreased with the increase in labeling duration for the FA fraction, while the opposite was observed for the HA and humin fractions. The ER values also showed that N contained in the FA fraction was always enriched with 1 5 N relative to the bulk soil for each treatment, while the opposite was true for the humin fraction. In the HA fraction, relative to the bulk soil, 1 5 N was enriched for the 7- and 31-month treatment while depleted for the 24-hr treatment. Discussion It was quite clear that samples with 1 5 N labeled for 1 and 3 growing seasons had very similar properties in terms of N extractability and incorporation in SOM, and that apphed N became more difficult to extract and thus much less available for tree uptake just one growing season after apphcation. However, in the samples with N apphed for 24 hours, apphed N was more readily extractable. Clay and Mafzer (1993) suggested that those differences may mean compositional changes in soil organic N pools, thus influencing organic N extractability and mineralization. Extractability ratios were greater than unity regardless of treatments and extraction methods used (Table 6.1), indicating that the residual 1 5 N was more extractable than the native soil N (He et al. 1988). The ratios decreased with increased labeling duration, indicating that the longer the residence time, the less extractable the immobilized 1 5 N became relative to the native soil N. The percentage of apphed N relative to the total N extracted also confirmed this trend which showed gradually decreasing values from the samples with increasing 1 5 N residence times. These results g. 6.1. Percentage distribution of applied N recovered in humus fractions. 100 FA HA Humin Humus type Fig. 6.2. Extractability ratios of humus fractions. FA HA Humus type Humin 121 could partly explain field observations that indicate little additional N being taken up by trees one year after fertilizer N apphcation (Preston and Mead 1994). The lowest extractability ratios were obtained by the autoclaving method at each treatment level. Other authors have also concluded that the autoclaving method gives low ER values and has poor selectivity for immobilized 1 5 N (Legg et al. 1971; Juma and Paul 1984; Kelly and Stevenson 1985; Azam et al. 1989b). Legg et al. (1971) argued that this could be due to the dispersion of protective soil aggregates by the autoclaving treatment which makes essentially all of the soil N accessible during extraction. In this study, total apphed N extracted by the autoclaving method was less than that of the KC1 extractable apphed N in inorganic form, with total N extracted higher in the autoclaving method. This shows that (1) the autoclaving process had more even access to both immobilized 1 5 N and native N as suggested by Legg et al. (1971); (2) there was chemical nnneralization-immobilization going on during autoclaving which exchanged native soil N in the structural organic matter for 1 5 N, because microbial mineralization- immobilization would be impossible under the autoclaving condition; and (3) autoclaving is a rigorous extraction method which may attack and decompose the humin fraction, because for the 7- and 31-month treatments, only the humin fraction had ER values less than 1. Only if large amounts of N enter into the extracts from the humin fraction will the ER values for the autoclaving extraction become closer to 1. Both total and apphed N extracted by 0.5 M K2SO4 after the samples were fumigated with chloroform increased from the non-fumigated samples which gave ER values of the fumigation- extraction similar to the 2 M KC1 extraction method. The difference between the N extracted after and before the fumigation represents the size of N stored in the soil microbial biomass (Brookes et al. 1985b). The fiunigation-extraction and 2 M KC1 extraction methods were highly 122 selective for extracting newly immobilized N at each treatment level and reflected the treatment effects. Thus they are more desirable extraction methods in obtaining more biologically meaningful N fractions. Differences in the amount of total and apphed N extracted and in extractability ratios seemed to demonstrate that those extraction methods rendered N from different pools (Stockdale and Rees 1994). The ER values and the amount of total and 1 5 N extracted using acidic permanganate for the 24-hr treatment resembled that of the fumigation-extraction method, regardless of the permanganate strength used. However, for the 7- and 31-month treatments, the result was similar to that of the 0.5 M K 2 S O 4 or 2 M KC1 extraction. In studying the oxidative release of potentially mineralizable soil N by acidic permanganate extraction, Stanford and Smith (1978) found that the extracted NH4-N was derived from oxidation of the soil organic N fraction most readily susceptible to decomposition by microorganisms. Perhaps the most recently immobilized 1 5 N (24- hr treatment) was more susceptible to oxidative reaction with the acidic permanganate than that which had been immobilized for a longer period. The recently immobilized 1 5 N can be converted to highly stable humin forms as was indicated by the higher ER values for the HA and humin fractions in the 7- and 31-month treatments than in the 24-hr treatment. Although both total and apphed N extracted by the acidic permanganate increased with increasing permanganate strength (except for the 24-hr treatment using 0.05 M KMn04), the increase of extracted apphed N lagged behind that of the total N, thus giving decreasing ER values foUowing increased permanganate strength at each treatment level. This shows that for the best selectivity of extracting immobilized I 5 N, 0.01 M is the optimal KMnCv concentration for this soil type and the range of KMn04 concentrations tested. 123 Other studies have reported that labeled N apphed to soils is quickly immobilized by soil microbial biomass and organic matter. Azam et al. (1989b) reported that the time required for complete immobilization of apphed N ranged from as little as 24 hours with low N apphcation rates (66 and 133 pg g"1 soil) to 106 hours with the highest rate used (333 pg g"1 soil) when the soil was incubated at 30 °C with glucose added as a highly available C source. Therefore, in field situations under natural condition the length of time required for complete immobilization might be quite variable. Smith and Power (1985) reported that complete immobilization of added N did not occur even after five growing seasons in a silt loam soil with an established crested wheatgrass (Agropyron desertorum, var Mandan) stand. In my experiment, 24 hours after 1 5 N apphcation, 47.70% of the recovered apphed N was found in the insoluble fraction of the SOM; after 7 and 31 months, the corresponding numbers were 98.50 and 98.87%. This confirms that immobilization of added N by organic material was relatively fast under the field conditions of this study. Labeled N immobilized into the humin fraction is highly insoluble (He et al. 1988). He et al. (1988) found that after a seven-day incubation, much of the 1 5 N in the humin fraction could not be solubilized by sequential extraction with a variety of inorganic and organic extractants. More 1 5 N was incorporated into the humin fraction in the 7- and 31-month treatments than in the 24-hr treatment, indicating that more immobilized 1 5 N became recalcitrant in the transformation process. Azam et al. (1989a) also demonstrated that the percentage of 1 5 N recovered in the humin fraction increased with time (from 0 to 112 days of laboratory incubation) for two Pakistani soils and that recovery of 15Ninthe humin fraction was very high (to 89%) for all the three soils studied. I removed the 0.5 M K2SO4 extractable N before fractionating the organic matter into the classical fractions, therefore, the apphed N recovered in the three humus fractions included 1 5 N in 124 Table 6.3. Flush of apphed N (ug g"1 soil) after chloroform fumigation and apphed N recovered in the humus fractions. Numbers in the parenthesis are standard errors. Treat- Humus fractions ment N flush FA HA Humin 31-month 20.02 (3.24) 7-month 29.26 (4.09) 24-hour 260.69(16.65) 10.48 (1.28) 24.64 (3.97) 441.98 (60.95) 11.17(2.26) 8.86 (2.85) 12.16(2.38) 502.29 (83.28) 720.66(16.07) 394.73 (43.64) the microbial biomass. Table 6.3 shows the net change (N flush) of recovered N before and after chloroform fumigation and apphed N recovered in the humus fractions. It shows that apphed N recovered in the FA + HA fraction was only shghtly greater than apphed N flushed after fumigation-extraction in the 7- and 31-month treatments. Since there must have been some 1 5 N chemically immobilized in the FA and HA fractions, the only reasonable explanation is that most of the microbially immobilized 1 5 N was in the recalcitrant humin fractions. However, in the 24-hr treatment, a greater percentage of the newly immobilized microbial 1 5 N is presumably in the FA fraction as indicated by the apparently large amount of 1 5 N recovered in that fraction. The fact that 24 hours after 15N-labeled fertilizer was apphed to the soil a quantity of the apphed N larger than that flushed from fumigation was recovered in the humus fractions can be explained by (1) the fast turnover of microbial biomass in the soil. Microorganisms may be able to immobilize 1 5 N in a quantity greater than the microbial biomass in a short period. He et al. (1988) found that much of the newly immobilized N was in the insoluble components of microbial cells; and (2) the immobilization of some of the recovered 1 5 N by soil organic matter through chemical fixation. Hart et al. (1993) reported that as much as 50% of the I 5 N recovered in the SOM pool may have been abioticahy fixed, in a laboratory incubation study of sterilized soil samples. Conclusions While it was found that the incorporation of N into soil biomass and organic matter is affected by many factors, such as substrate addition and quality (Bremer and van Kessel 1990; Janzen et al. 1988), drying and rewetting cycles (Stockdale and Rees 1994), soil type (Stockdale and Rees 1994) and presence of plants (Wheatley et al. 1990), this study confirmed that the extractability of apphed N relative to native N was reduced with increasing residence time, because much immobilized N was incorporated into the stable humus forms. One growing season after 1 5 N apphcation, most of the recovered N in the SOM had been immobilized by the humin fraction, leading to nonsignificant differences in N extractabihties from those with N added for three growing seasons. The results were consistent with N mineralization studies of the same samples in which immobilized N in the 7- and 31-month treatments had low mineralization potentials (Chapter 5). Different extraction methods attack different N pools and thus yield a range of apphed and native N being extracted. 127 CHAPTER 7. SOH, MICROBIAL BIOMASS AND MICROBIAL AND MINERALIZABLE N IN A C L E A R C L T CHRONOSEQUENCE* Introduction Plants and microorganisms depend on each other for N (Woods et al. 1982). Microbial biomass, considered a labile fraction of the soil organic matter, is known to be the sink and source of plant available N in soil (Jenkinson and Ladd 1981; McGill and Myers 1987). Nitrogen can be immobilized by the microbial biomass present in soil and thus microbial biomass can be considered as a sink for soil N. The N content of microbial biomass constitutes a significant amount of the mineralizable N that is a potentially available source to plants (Marumoto et al. 1982). It is the net balance between immobilization and mineralization which determines the availabihty of N for plant uptake. Microbial activities are easily changed by management practices such as clearcutting (Vitousek 1981; Smethurst and Nambiar 1990a), litter management (Weber et al. 1985; Smethurst and Nambiar 1990b), forest fire (White 1986; Weber 1987; Bell and Binkley 1989), and season and successional stages (White et al. 1988a; Vitousek et al. 1989), thereby affecting plant growth in the field. By altering the amount and type of organic matter, and soil temperature, moisture and pH, forest harvesting can cause long-lasting impacts on microbial activity (Harvey et al. 1980) and nutrient availabihty (Entry et al. 1986). In areas where reforestation failures have been encountered after clearcutting of old-growth forests, particular attention has been paid to the role of the soil microbial population and activities in the fertility decline. Niemela and Sundman (1977) and Sundman et al. (1978) studied the changes in soil bacterial population in areas 0, 4, 7 * : A modified version of this Chapter is published in the Can. J. of For. Res., 25: 1595-1607 (1995). 128 and 13 years after clearcutting in coniferous forests in northern Finland where reforestation has frequently failed. They found that clearcutting causes increases in bacterial biomass and changes in population structure. Entry et al. (1986) studied microbial biomass changes in a northern Rocky Mountain forest soil subjected to clearcutting, residue removal and burning treatments and found that in the clearcut and residue burned treatment microbial biomass was lower than in the other treatments for most of the studied period. However, little study has been done to evaluate the effect of clearcutting and slash-burning on microbial biomass dynamics and its relationship to N cycling and tree nutrition. Growth stagnation of planted or naturally regenerated Sitka spruce, western redcedar and western hemlock on cutovers of old-growth CH forests was observed in 10-year-old stands in the wetter Coastal Western Hemlock biogeoclimatic zone (CWHb) of British Columbia, in association with the invasion of an ericaceous evergreen shrub, salal (Weetman et al. 1989a and b). Messier (1993) found that conifer see<ilings planted on CH sites 8 years after clearcutting and burning were growing slower than seedlings planted on 2-year-old CH sites with and without competing vegetation, although removal of competing vegetation helped seedling growth on the same site. Therefore, competing vegetation is not the sole factor in declining growth in the old CH cutover sites. Using solid-state ^ C NMR, deMontigny et al. (1993) found evidence for tannin in F horizons in the CH old-growth and suggested that increasing inputs of salal tannin in CH cutovers might contribute to less effective litter decomposition; however, no direct evidence was provided. The present work was conducted to see if microbial biomass and microbially mediated processes such as N nhheralization/immobilization are contributing to the declining N availabihty in those CH cutover sites 8-10 years after clearcutting and planting. 129 The objectives of this study were: 1.) to quantify the changes of microbial biomass C and N, and mineralizable N in the forest floors of a CH clearcut chronosequence (uncut old-growth forests, 3- and 10-yr-old plantations); and 2.) to test the hypothesis that microbial biomass acted as a net sink in the 10-yr-old plantations by immobilizing more N into the microbial N pool than in the 3-yr-old plantations. Materials and methods Study area and stand selection The study was conducted in the western redcedar-western hemlock ecosystems. The general description of the study area can be found in Chapter 3. Three old-growth CH forests were selected that had similar stand structure and slope position. Three 3-yr-old and three 10-yr-old plantations of western redcedar on cutovers of CH forests were also selected. These had similar stand structure (based on cedar stumps) and slope position to the three old-growth forests. The 9 sites selected are within 5 km of each other. The cutovers had been slash-burned after logging and then planted with western redcedar. The slash- burning had little effect on the humus layer. In the 3-yr-old plantations, salal was dominant but it was short (about 30 cm) and not very dense. In the 10-yr-old plantations, salal was much denser and taller (50-80 cm). A representative old-growth CH stand is shown in Figure 1.1. Representative 3- and 10-yr-old stands are shown in Figure 7.1. For convenience, the different aged stands are referred to as treatments. Field sampling Forest floors were sampled on May 23, July 16, August 26 and October 18, 1992. One Figure 7.1. Western red cedar plantations on C H cutover sites: (a) 3-yr-old; (b) 10-yr-old (b) A 10-year-old site 131 composite sample each of F, woody F (Fw) and H materials was taken from the forest floor in each selected stand. Each composite sample was a mixture of the same material collected at 6-7 randomly selected points in each site. The F layer in the old-growth forests consisted of fine twigs and leaf litter, and was about 5 cm thick. The F layer in the 3- and 10-yr-old plantations was 1-2 cm thick and was the residue from burning. Fw was partially decomposed, but the woody structure held when rubbed between the fingers. H was more than 80% amorphous, with a greasy texture and a dark color (deMontigny 1992). H, sometimes thicker than 1 m, is the main component of the forest rooting zone. Samples were kept in a cooler on ice for three days while being transported to the laboratory where they were stored at 4°C before use. AU samples were analyzed within one week of sampling. On the same four dates, current-year fohage of western redcedar was coUected from the upper branches of at least 15 trees in each plantations. In the old-growth stands, western redcedar fohage was obtained by shooting off branches of the upper crown with a shotgun. Current-year salal fohage was coUected in the plantations and the old-growth forests. Laboratory analyses Moisture content was measured on fresh unsieved samples by the gravimetric method at 105°C for 24 hours. The rest of the sample was sieved through an 8 mm screen and visible roots were removed. Microbial biomass C and N were measured by the chloroform fumigation-extraction method (Brookes et al. 1985b; Vance et al. 1987; Wu et al. 1990). Appropriate weight portions (3g dry basis for F and Fw, and 5g dry basis for H) of sieved F, Fw and H samples were fumigated with 132 ethanol-free chloroform for 24 h at room temperature (20°C). After removal of the chloroform, fumigated samples were extracted with 0.5 M K ŜCJj on an end-over-end shaker (150 rev./min.) for 0.5 h, followed by vacuum filtration through Whatman #42 filters. Another set of unfumigated samples used as controls was extracted in the same way at the time fumigation commenced. The filtrate was analyzed for total dissolved organic carbon by the wet oxidation diffusion method of Snyder and Trofymow (1984) and total N by Kjeldahl digestion and disthlation (Bremner 1965). The total extractable C and N measured from the control samples are presented in the results and discussion sections as extractable organic C and N. Microbial biomass C (Be) and N (Bn) were calculated by equations Bc=Fc/Kc, and Bn=Fn/Kn, respectively, where Fc and Fn were the differences in total dissolved organic carbon and total N, respectively in fumigated and unfumigated samples, and Kc is 0.379 (Vance et al. 1987), and Kn is 0.54 (Brookes et al. 1985b). Microbial C/N ratio was calculated as Bc/Bn. Mineralizable N and pH were determined on ah-dried and sieved (2 mm screen) samples of F, Fw and H from the August 26 collection. pH was measured in 1:2 (v/v) 0.01 M CaCl2 solutions. Mineralizable N was measured after a 14-d anaerobic incubation at 30°C (Waring and Bremner 1964), followed by Kjeldahl digestion and steam disthlation of pre- and post-fumigation samples (Keeney and Nelson 1982). Fohar samples were dried at 65°C overnight and ground to 20 mesh in a Wiley mill. Total N concentrations were determined by Kjeldahl analysis with steam distillation (Bremner 1965). Total C concentrations were detennined using a LECO CR-12 C analyzer (Model 781-600, LECO Corporation 1981). Statistical analyses 133 Homogeneity of variances and normality of distributions of data sets were checked before any further statistical analysis. Data that were not homogeneous (extractable C of un&migated Fw samples, microbial biomass C/N ratios of Fw samples and total N content in salal) were logarithmic(lO) transformed prior to analysis. Analyses of variance were performed on all experimental variables using the General Linear Models (GLM) procedure of the SAS package (SAS Institute Inc. 1989). Group means of independent variables were compared between treatments (age of stands) at each sampling date by using Scheffe's test when there was a significant treatment x date interaction, for each type of material sampled. When treatment x date was non-significant, Scheffe's test was used to examine treatment means across the dates, for each type of material sampled. Results Water content and pH Significant treatment x date interaction for water content in F resulted from the change in water content from higher (on May 23, July 16 and October 18) to lower (on August 26) in the old-growth forests than in the plantations. Water content in the F layer was significantly greater (p<0.05) in the old-growth forests than in the 3- and 10-yr-old plantations on July 16 (Fig. 7.2). Treatment x date interactions were not significant for water content in Fw and H. Water content in Fw, which is also in the uppermost horizon of the forest floor, was slightly greater in the old- growth forests than in the plantations (p<0.10, Table 7.1). There were no differences in water content of H among the treatments. Measurements showed that F and Fw had higher water holding capacities in the old-growth forest than in the young plantations (data not shown). The effect of sampling date on water content was significant for F (p<0.001), Fw (p<0.01) and H Table 7.1. Mean square and level of significance0 for water content, extractable C and N, and microbial C and N Source Measure- of Forest floor layer̂ ment variation F Fw H Water Treatment (T) 54598*** content Date (D) 43252*** T x D 8625** N Error Microbial T C D T x D Error Microbial T D T x D Error 1979 11918+ 27442** 827 3959 1310444 1962430 1437381 535819*** 11582 8151 10380 1244007+ 255611 462660 4459 655 8679+ 621 3382 90315715*** 884712 9894344*** 14691 7455 61789 1071386 1043986 67376* 35294** 18866 9786 12044 Table 7.1, continued next page Table 7.1, continued Microbial C/N ratio D organic C organic N T x D Error Extractable T D T x D Error Extractable T D T x D 18.94** 26.83*** 16.31*** 2.45 2927978*** 252575** 6669 49994 5818 *** 0.043 0.089* 0.082* 0.024 0.07* 0.02 137.5 5134*** 461.1 64.36*** 9.52 1.94 5.00 0.23*** 278707*** 0.36** 46710* 233 394.0 34538+ 14113 543.1+ 3117.9*** 285.6 Error 223 400.0 202.3 a The difference between means is significant at +p<0.10; *p<0.05; **p<0.01; ***p<0.001. Degrees of freedom for Treatment (2), Date (3), and T x D (6). b F= partially decomposed litter horizon, Fw= woody F, and H= humus. 136 Fig. 7.2. Water content (%) in forest floor layers in the old-growth forests and 3- and 10-yr-old plantations. Abbreviations are shown in Table 1. Old-growth B 3-yr old r j l O - y r o l d 500 450 - s 400 - 350 c 300 0> -t— c 250 - o 200 -o i— CD 150 - 100 - ? 50 - 2 3 - M a y 16-Ju l 26-Aug 18-Oct (Fw) 2 3 - M a y 16-Ju l 26 -Aug 18-Oct 137 (p<0.10) (Table 7.1). The trend for pH was: old-growth forests > 10-yr-old plantations > 3-yr- old plantations (Table 7.2). There were no differences in pH of F among the treatments. Fw and H had significantly greater pH values in the old-growth forests and the 10-yr-old plantations than in the 3-yr-old plantations (p<0.05). Microbial biomass C and N Microbial biomass C in F (p<0.001), Fw (non-significant) and H (P<0.001) was in the following order: old-growth forests > 10-yr-old plantations > 3-yr-old plantations (Fig. 7.3 and Table 7.1). However, the multiple comparison test showed that the significant treatment effect in F hes in the differences between the old-growth forests and the two young plantations(p<0.05); there was no significant difference between the two plantations. For H, microbial biomass was significantly higher in the old-growth forest and in the 10-yr-old plantations than in the 3-yr-old plantations. The greatest amount of microbial C was in F (11.7 mg g"') and the least amount was in Fw (1.3 mg g"1). Microbial C in F was 225 and 55% greater than that in Fw and H, respectively. The amount of microbial biomass C in the three types of humus materials was fairly constant throughout the sampling period, resulting in nonsignificant sampling date effects and nonsignificant treatment x date interactions (Table 7.1). Microbial N was significantly different among stand types for F (p<0.001) and H (p<0.05) (Table 7.1). Microbial N in Fw was not different among stand types. Similar to the seasonal changes in microbial biomass, microbial N was not affected by sampling dates (except in Fw) and no treatment x date interactions were found (Table 7.1). Multiple comparison test showed that microbial biomass N was not significantly different between the 3- and 10-yr-old plantations in F (434.04 vs. 443.37 pg g1), Fw (162.86 vs. 199.92 pg g1) and H (437.80 vs. 401.23 pg g1), but Table 7.2. pH and mineralizable N (ug g"1) of forest floor layers in old-growth forests and 3- and 10-yr-old plantations0 for samples coUected on August 26 (n=3) Measure- ment Treat- ment Forest floor layer F Fw H pH Old-growth 4.16 a* 3.31 a 3.16 a 3-yr old 3.75 a 3.04 b 2.85 b 10-yr old 3.76 a 3.20 a 3.11 a Minera- Old-growth 394.1 a 40.33 a 148.32 a lizable 3-yr old 159.2 b 25.59 a 78.08 b N 10-yr old 140.5 b 31.38 a 90.24 b * Same lowercase letter indicates no significant difference (p=0.05 level) between treatment means for the same measurement and forest floor material type. a Abbreviations are described in Table 7.1. 139 Fig. 7.3. Microbial C and N in forest floor layers in the old-growth forests and 3- and 10-yr-old plantations. Abbreviations are shown in Table 1. | O l d - g r o w t h B 3 - y r o l d • 1 0 - y r o l d £ 10 o "53 'n 2 o 23- 16- 26- 18- May Jul Aug Oct | O l d - g r o w t h a 3 - y r o l d • 1 0 - y r o l d .2 0 23- 16- 26- 18- May Jul Aug Oct _ 14 E 0 1 o o 10 8 6 4 2 0 (Fw) 23- May 16- Jul 26- Aug 18- Oct T 14 cn 1 2 ? 10 — 8 4 O (H) to 4 o 23- May 16- Jul 26- Aug 18- Oct S a m p l i n g date T 1-2 1 £ 0.8 Z 0.6 0-4 •§ 0 2 o o (H) lAfcofc 23- 16- 26- 18- May Jul Aug Oct S a m p l i n g date 140 was significantly greater in the old-growth forests than in the 3- and 10-yr-old plantations in F and significantly greater in the old-growth forests than in the 10-yr-old plantations in H (p<0.05). Microbial biomass C and N were both greatest in F and least in Fw. Then ratios were significantly affected by stand type in F and H and by sampling date in F and Fw (Table 7.1). There was no treatment effect on microbial C/N ratio of Fw. The coefficient of variation for microbial C/N ratio was 12, 47 and 22% for F, Fw and H, respectively, indicating that the data set for Fw had greater deviation from the mean which led to the lack of significance of treatment effects. The rise in microbial C/N ratio in the old-growth forest relative to the 10-yr-old plantations from the July 16th to October 18th sampling in F and from May 23rd and August 26th to July 16th and October 18th sampling in Fw (Fig. 7.4) resulted in significant treatment x date interactions in these two materials. Multiple comparison revealed that the microbial C/N ratio in F was greater in the 10-yr-old plantations than in the 3-yr-old plantations and the old-growth forest only on the July 16th sampling (p<0.05); in H, microbial C/N ratio in the 10-yr-old plantations was greater than those in the 3-yr-old plantations and the old-growth forests (p<0.05). Extractable organic C and N The amount of organic C extracted by 0.5 M K 2 S0 4 from the unfumigated samples across the sampling dates was 159 and 107% greater (p<0.05) in the old-growth forests than in the 3-yr- old plantations in F and FL respectively (Fig. 7.5, Table 7.1), since there were no treatment x date interactions. Although the overall treatment effect was significant for Fw, there was no difference among the stand types at any particular sampling date due to the significant interaction. Extractable C was generally higher in the 3-yr-old than in the 10-yr-old plantations, for F and H 141 Fig. 7.4. Microbial C:N ratio in forest floor layers in the old-growth forests and 3- and 10-yr-old plantations. Abbreviations are shown in Table 1. O l d - g r o w t h a 3 - y r o l d r j l O - y r o l d 40 § 35 4 2 30 Z 25 4 O 20 .2 15 £1 O 10 .2 5 (F) 2 3 - M a y 16-Ju l 26 -Aug 18-Oct 2 3 - M a y 16-Ju l 26-Aug 18-Oct 40 Lp 35 L - 30 * 25 O 20 .S3 15 O 10 O 5 = 0 (H) 2 3 - M a y 16-Jul 26-Aug 18-Oct S a m p l i n g d a t e 142 Fig. 7.5. Extractable C and N (0.5 M K 2 S O 4 ) in forest floor layers in the old-growth forests and 3- and 10-yr-old plantations. Abbreviations are shown in Table 1. I Old-growth a3-yr old rj10-yr old I Old-growth >3-yr old rjlO-yrold 1800 01 1600 Di 1400 w 1200 o 1000 JS 800 0 6 0 0 Tj 400 £ 200 fi LU 1800 » p 1600 0 1 1400 3 1200 o 1000 £ 800 May Aug (Fw) 23- 16- 26- 18- May Jul Aug Oct r-^ 1800 'DI 1600 DI 1400 w 1200 O 1000 OJ 800 (0 TJ cd fi LU 600 400 200 0 ^T" 160 'DI 140 D] 120 ^ 100 g 20 LU 160 7 140 Di Di 120 W 100 •Q 60 0 TJ 40 S LU ^ 160 'DI 140 May Aug (Fw) May Jul Aug Oct (H) Di 120 100 3 60 cd 20 fi 0 LU Sampling date 23- 16- 26- 18- May Jul Aug Oct Sampling date 143 (both non-significant). On average, F contained the greatest amounts of extractable C, and H contained the least. There was a significant sampling date effect on extractable C in all three humus layers (Table 7.1). The greatest extractable organic C was from the July samples in F and H and in the August samples in Fw. No treatment x date interactions were found for extractable organic N in ah of the materials studied. Extractable organic N in the F material was 51% greater in the old-growth forests than in the 3-yr-old plantations (p<0.05), and 24% greater in the latter than in the 10-yr-old plantations (non-significant). There was no treatment effect or sampling date effect on extractable N in Fw material (Fig. 7.5, Table 7.1). The coefficient of variation for extractable N was 22, 43 and 23% for F, Fw and FL respectively. In H, extractable N showed a trend to decrease from the 3-yr-old plantations to the old-growth forests and to the 10-yr-old plantations. The greatest amounts of extractable N were in the July samples, except for Fw in the 10-yr-old plantations. Substantially less N was extracted from Fw than from F and H materials (Fig. 7.5). Mineralizable N The amount of N mineralized during the 14-d anaerobic incubation was greater in the old- growth forests than in the 3- and 10-yr-old plantations in F and H (p<0.05, Table 7.2). Differences between the 3- and 10-yr-old plantations were not significant. F contained the highest amount of mineralizable N, followed by H and Fw. Foliar N Fohar N concentrations and C/N ratios were fairly constant among treatments and sampling dates resulting in nonsignificant treatment x date interactions for cedar and salal. Nitrogen 144 concentrations in cedar foliage were consistently higher (p<0.05) in the 3-yr-old plantations than in the old-growth forests and the 10-yr-old plantations (Fig. 7.6) on all four dates. Nitrogen concentrations in salal fohage were significantly higher in the 3-yr-old plantations than in the 10- yr-old plantations (p<0.05), but were non-significantly higher in the former than in the old-growth forest and in the old-growth forest than in the 10-yr-old plantations. Fohar C/N ratios of cedar and salal were significantly lower in the 3-yr-old than in the 10-yr-old plantations and old-growth forests. Discussion Humus type is one of the most important factors affecting the amounts, forms, and minerahzation/immobilization processes of N in forest soils (Richards et al. 1985). Woody debris is an obvious component of old-growth CH forests in coastal British Columbia (Keenan et al. 1993). Humus materials derived from coarse woody debris (large stems) and those from fine debris (small twigs and leaf litter) have different chemical and physical properties. Therefore, I differentiated the forest floor partially decomposed litter (F) and humus (H) layers into F, woody F (Fw), H and woody H (Hw) types (Prescott et al. 1993b). Because Hw type of humus was frequently missed during field sampling due to its discontinuous distribution in the subhorizons in the soil profile, results for Hw are not reported here. The higher soil moisture content in F and Fw in the old-growth forest than in the plantations (Fig. 7.2) reflected the changes in physical properties of F and Fw after slash-burning. The water holding capacity and consequently the water content of F and Fw were reduced by changes in the structure of the forest floor after burning. Differences in near-surface microclimatic conditions induced by changes in vegetation cover may also have affected moisture contents in F and Fw. 145 Fig. 7.6. Foliar N concentrations and C:N ratios of western redcedar and salal in old-growth cedar-hemlock forests and 3- and 10-yr-old cedar plantations. I Old-growth B 3-yr old • 10-yr old I Old-growth B 3-yr old • 10-yr old (Cedar) 1.6 ^ 1.4 o 1 2 S 0.8 § 0.6 O 0.4 O 0.2 O (Salal) 23- May 16- Jul 26- Aug 18- Oct 70 ° 60 2 50 Z 40 6 30 ra 20 o 10 (Cedar) 7 0 (Salal) ^ © 60 k m 11II 23- May 16- Jul 26- Aug 18- Oct 23- May 16- Jul 26- Aug 18- Oct Sampling date Sampling date 146 The moisture content of H was not significantly different between treatments because this layer was less affected by slash-burning. The low water content in F in the summer relative to that in early fall was related to the amount of rainfall received in the period shortly before sampling (Fig. 7.7). According to Viro (1974), and Pietikainen and Fritze (1993), forest fires normally cause increases in soil pH through the release of basic cations. This effect is usually only a temporarily phenomenon. In this study, however, the pH of F in the 3- and 10-yr-old plantations that had been slash-burned was less than that in the old-growth forests (non-significant, Table 7.2). In Fw and H, pH was lower in the 3-yr-old than in the 10-yr-old plantations and the old-growth forests (p<0.05). Decreasing pH in the 3-yr-old plantations may have been caused by increased release of acidic organic matter from microbial decomposition and exudation after logging and slash- burning. Niemela and Sundman (1977) reported a short-term increase of organisms producing acid from sucrose after clearcutting a spruce forest. The abundant regrowth of salal on the clearcut sites may also contribute to the decrease of pH through increased root exudates. However, the exact reason for the decrease of pH three years after logging and burning cannot be determined from the available data. Higher pH in the 10-yr-old than in the 3-yr-old plantations reflected a slow recovery process in balancing pH to the level in the old-growth forests. The change of pH was -0.41, -0.27, and -0.31 units for F, Fw and H, respectively, from old-growth forests to 3-yr-old plantations and -0.40, -0.11 and -0.05 units for F, Fw and H, respectively, from old-growth forests to 10-yr-old plantations. In general, the change of pH in each type of humus materials was probably too narrow to cause significant changes in microbial community by treatments. Fig. 7.7. Daily precipitation in the study area in summer 1992. Climate data was obtained from the Port Hardy Airport 15 km away from the study site. o 30 + 10 -I i l l . X ILELLJ 1-May 31-May 30-Jun 30-Jul Date + 29-Aug 28-Sep 28-Oct 148 Clearcutting alone does not reduce microbial biomass in forest soils, in fact, microbial biomass was higher in a clearcut and residue-left treatment than in an uncut treatment in a northern Rocky Mountain forest soil (Entry et al. 1986), because of increased decomposible organic C input into the system and other beneficial effects from disposed slash. Lundgren (1982) showed that bacterial biomass increased initiaUy after clearcutting a Scots pine forest, but decreased compared to a reference stand during the third year after clearcutting and onwards. Sundman et al. (1978) also showed increases in bacterial counts after clearcutting a spruce forest. However, if clearcutting is followed by slash-burning, microbial biomass would be drastically decreased (Entry et al. 1986; Pietikainen and Fritze 1993). Results from this study showed that microbial biomass was lower in the 3- and 10-yr-old plantations that were clearcut and slash- burned than in the uncut and unburned old-growth forests. Higher microbial C in the 10-yr-old than in the 3-yr-old plantations indicated that microbial population was gradually recovering from clearcutting and slash-burning between years 3 and 10 which is in good agreement with Fritze et al. (1993). The changes in microbial biomass may be affected by other biotic and abiotic factors. One of them is the availabihty of soluble carbon compounds (Wheatley et al. 1990). Greater microbial C in the old-growth forests was associated with higher extractable C (Fig. 7.5). A simple regression analysis showed a relationship between microbial biomass and extractable C for the old-growth forests and young plantations: Microbial-C (mg g"1) = 2522 + 4.34*Extractable-C (mg g"1), r=0.64 (p=0.0001). The amount of extractable C was reduced by clearcutting and slash- burning. Pietikainen and Fritze (1993) reported that fire treatment increased the amount of extractable C and N in humus layers, but these decreased to control levels within 3 years. Then study sampled a maximum of 800 days after the fire treatment which makes it difficult to do a 149 complete comparison with the fmdings of this paper. Correlation analysis in this study showed that microbial biomass had a weak relationship with soil moisture content (r=-0.16, p=0.0963). This indicates that soil moisture content offers little help in explaining the dynamics of microbial biomass over the sampling period, perhaps because soil moisture content was never at a level limiting microbial biomass growth. Microbial N showed the same patterns of changes as microbial C in the old-growth forests, 3- and 10-yr-old plantations. Measured on the August 26 sampling, the percentage of total soil N found in microbial biomass was found to be greater in the old-growth forests (8.21%) than in the 3- (4.93%) and 10-yr-old (4.95%) plantations for F (p<0.05). The percentage of the total N found in microbial biomass was 11.84, 6.50 and 6.32% in Fw and 5.02, 4.58 and 4.76% in H, for the old-growth forests, and the 3- and 10-yr-old plantations, respectively. The differences among stand types were not significant for Fw and H. This result indicated that a greater percentage of the total N was potentially available in the old-growth forests. The seasonal variation in microbial biomass C and N was small which resulted in non- significant sampling date effects for F and H. This resembles results obtained in hardwood forests by Holmes and Zak (1994). The similar microbial N pool sizes in the 3- and 10-yr-old plantations led us to reject the hypothesis that the microbial biomass acted as a net sink in the 10-yr-old plantations by immobilizing more N into the microbial N pool. Holmes and Zak (1994) suggested that in maintaining a relatively constant pool it is the turnover rate of microbial biomass which controls N availabihty. Therefore, further studies are needed to test if microbial biomass and N in the 10-yr-old plantations have a slower turnover rate and thus reduced N availabihty than in the 3- yr-old plantations. 150 In this study, the microbial C/N ratios in most samples were between 6.5 and 16. A few comparatively high C/N ratio values were found, i.e., Fw in the old-growth forest in the July (22.1) and October (23.6) sampling, and Fw in the 3-yr-old plantations in the July sampling (29.5). These values were higher than typical microbial C/N ratios found in coniferous forest soils (7-13, Fenn et al. 1993). Both Collins et al. (1992) and Dalai and Mayer (1987) reported microbial C/N ratios (about 20) close to those of bulk soils and proposed that the ratios represented that of "stabilized" microbial biomass. Microbial C/N ratios may be affected by the Kc and Kn values used in calculating the microbial C and N, but comparisons between treatments within the same experiment are probably vahd. A change in microbial population structure is a readily available explanation for the shift in microbial C/N ratios (Wheatley et al. 1990). Anderson and Domsch (1980) and Ross (1988) reported that fungi generally have higher C/N ratios than bacteria in culture incubations. However, there has been little discussion on the implications of microbial C/N ratio changes, whether or not accompanied by population structural changes, on N availabihty in soils. Microbial C/N ratios may provide an indication of the availabihty of microbial N for mineralization. Whether microbial biomass acts as a net source or sink of available N depends, in part, on the C/N ratio of the microbial substrates, the rate of microbial biomass growth and the microbial demand for N (Fenn et al. 1993; Edmonds 1987). High microbial C/N ratios may be associated with the low N availabihty in forest floors (Edmonds and Chappell 1994; Ross et al. 1995). In general, low C/N ratios in microbial biomass are associated with net N mineralization, and high C/N ratios with net immobilization of N in the microbial biomass (Paul and Clark 1989; Edmonds 1987). The microbial C/N ratios in F (p<0.05 on one of the sampling dates) and H (p<0.05) were greatest in the 10-yr-old plantations and least in the 3-yr-old plantations. This 151 suggests that microbial biomass became N-stressed from year 3 to year 10 after clearcutting and slash-burning and that it is more likely to result in a net mineralization in the 3-yr-old plantations while a net immobilization in the 10-yr-old plantations during decomposition of organic materials. This has practical implications for N availabihty in those ecosystems, because the difference between microbial release of N and immobilization of N into microbial biomass largely determines the amount of N available for plant uptake (Binkley and Hart 1989; Paul and Clark 1989). Since microbial biomass tended to increase from the 3-yr-old to the 10-yr-old plantations for the three material types studied, increases in available C and N in the soil in the 10-yr-old plantations may result in further increases in microbial biomass and assimilation of the available N into microbial biomass. Therefore, even if the microbial community maintains its current C/N ratio level, microbes would be more severely competing with trees for N in the 10-yr-old than in the 3-yr-old plantations to meet then maintenance requirements and growth potential. The N-stressed situation in the 10-yr-old plantations was also indicated by the extractable N and fohar N analysis results. Extractable N may serve as an indication of available N in soils. The higher (statistically non-significant) amount of extractable N in the 3-yr-old than in the 10-yr-old plantations (Fig. 7.5) indicated that the N supply tended to be better in the 3-yr-old plantations which was consistent with the results of fohar N analysis. Better N supply on the younger cutover sites has also been demonstrated by Messier (1993). Mineralizable N is the amount of N potentially available for plant uptake upon mineralization (Powers 1980). This parameter, measured only on the August 26 sampling, is not very helpful in explaining differences between the two young plantations, because similar amounts of mineralizable N were obtained from the 3- and 10-yr-old plantations for the three humus types sampled. However, there were reports showing that N mineralized during the 14-day anaerobic 152 incubation at 30 °C could be related to field growth performance (Shumway 1978; Powers 1980). Others found that anaerobic incubations are similar to soil fumigation methods where microbial N is measured when N is released through decomposing dead microbial cells (Adams and Attiwill 1986; Myrold 1987). Therefore, the anaerobic incubation results were not surprising since microbial N was not different in the two plantations. In this study, the amount of N mineralized from the anaerobic incubation never exceeded 50% of the microbial N. The significantly greater amount of mineralizable N in the old-growth forests than in the plantations illustrated again that clearcutting and slash-burning had greatly changed the properties of the forest floor. The overall nutrient supply problems on CH cutovers have been suggested to be partially related to poor organic matter quality (Prescott et al. 1993b; Prescott and Preston 1994). Differences in microbial activities and soil nutrient dynamics were ultimately displayed in the growth performance of trees in the field. Old-growth forests consist of mature trees which may require less N per capita for efficient photosynthesis and maintaining then growth because of more developed internal cycling mechanisms (van den Driessche 1984). Also, as trees grow older, the proportion of nutrients in the fohage tends to become less while that in the bole and bark becomes more (Wright and Will 1958). As a result of better nutritional conditions, western redcedar in the 3-yr-old stands had a higher N content and lower C/N ratio in current year fohar samples. Those in the 10-yr-old stands had fohar N content and C/N ratios in the ranges of the old-growth trees, and showed chlorosis symptoms and poor growth in the field. It has been found that mycorrhizal fimgi, which form ericoid mycorrhizae with salal roots, can use some simple organic N sources, such as amino acid, peptide and protein (G. Xiao pers. comm.); therefore, salal may be more efficient in utilizing limited available N supply in the soil. However, in the 10-yr-old stands, salal fohage also had lower fohar N contents and higher C/N ratios, further indicating severe nutrient deficiencies in those stands. Conclusions The clearcutting + slash-burning treatment had substantially reduced microbial biomass and activities and altered environmental conditions. Water content, extractable C and N, microbial C and N, and mineralizable N were ah reduced by harvesting and burning which made the forest floor of the old-growth forests distinctively different from those of the young plantations. Greater extractable N and mineralizable N in the old-growth forests than in the young plantations suggested that the old-growth forests may be in a much better nutritional condition than the young plantations. Between the 3- and 10-yr-old plantations, slower growth in the 10-yr-old plantations was associated with the lower amount of available N in the forest floor, greater competition for available N from the microbial community and more vigorous growth of competing understory salal vegetation. Microbial activities appeared to play an important role in regulating N cycling and availabihty in the underground ecosystem However, not enough evidence was obtained to conclude that microbial biomass was a net sink for available N in the older plantations. Further studies are needed to clarify if it is the reduced fluxes of N in the forest floor in the old plantations that are contributing to the N-stressed situation by using in situ mineralization methods; particularly useful would be methods that quantify gross N mineralizations (Hart et al. 1994). 154 CHAPTER 8 GENERAL DISCUSSION AND CONCLUSIONS Introduction This study set out to expand our understanding of the regeneration problems in the early stage on the CH cutover sites. Although this issue has been extensively studied (cf. Prescott and Weetman 1994), some of the hypotheses need to be re-tested and new hypotheses need to be developed because the fundamental cause of the nutritional problem on CH sites is not yet understood. SalaL an ericaceous evergreen shrub which grows densely from the panhandle of Alaska along the entire coast of British Columbia to south California (Fraser et al. 1993), resprouts rapidly from rhizomes foUowing clearcutting and slash-burning on the CH sites. As a persistent and pervasive woody perennial (Prescott et al. 1993a), salal constitutes a serious competitor with conifer species particularly on the recently regenerated CH sites. Salal has been found to be resistant and resilient to many herbicides (Fraser et al. 1993), to mechanical weed control, and to shading (G.F. Weetman pers. comm.). Therefore, the role of salal in the regenerated CH sites with respect to competition for the limited resources, especially soil nutrients, has been one of the foci of the SCHIRP project. Through the re-examination of the effect of understory (salal) removal on conifer growth and ability to respond to fertilization with N, this study was specifically aimed at clarifying the role of salal in competing with conifers for fertilizer N and thus its effect on the fate of N. Further, to address the question of why fertilizer N quickly becomes unavailable for tree uptake after being apphed to the soil, the mineralization and extractabihties of residual fertilizer N were studied. Just as important as salal competition on the CH cutover sites is the competition for N from the microbial organisms in the soil. Therefore, the possible role of microbial competition on tree nutrition on the CH cutover sites was investigated. The foUowing is a general discussion and summary of the fmdings from the present study relating to the objectives set forth before I commenced this thesis research. Understory competition The first hypothesis (Chapter 1) is accepted, i.e., understory competition significantly reduced conifer height, diameter and biomass growth. Root coUar diameter was found to be a more sensitive parameter than tree height for quantifying competition effect on tree growth. Significant differences in height growth between the treated (understory removal) and control (understory remaining) treatments were only found when measured after several growing seasons, whereas significant differences in RCD (root coUar diameter) growth between treatments can be detected after just one growing season's growth. However, biomass growth was even more sensitive than RCD in quantifying treatment effects, reflecting the fact that biomass integrates the vertical as weU as the horizontal growth of trees (Abrahamson and CasweU 1982; Brand and Magnussen 1988). The competition from the understory on the CH cutover sites is essentiaUy a belowground competition, since conifer saplings have already established a canopy above the salal about ten years after planting or regeneration, when growth stagnation problems start to be noticed. At this stage, radiation is no longer a limiting factor for tree growth. Although salal was suspected to have aUelopathetic effects on conifer growth (deMontigny 1992), competition for nutrients had been found to be the main factor contributing to conifer growth stagnation (Weetman et al. 1989a and b; Messier 1993). Even though this study supports the finding that salal survives under complete shoot removal by preserving belowground biomass, the total salal biomass and the nutrients immobilized therein were greatly reduced by the salal shoot removal treatment, thus alleviating nutritional stress on conifers. Grime (1979) suggested that relatively large amounts of biomass impose high rates of demand on soil resources when belowground competition is intense. Reduced belowground understory biomass growth in the treated plots effectively increased the 'biological space' (Ross and Harper 1972) available for conifer trees for development of roots. Physical restriction on root growth can lead to reduced plant growth irrespective of water or nutrient levels (McConnaughay and Bazzaz 1991). Hypothesis number two had to be rejected because no significant effect of understory competition on biomass allocation was found among various tree components (Chapter 3). Biomass partitioning among different parts of a tree may be an effective way of alleviating the stresses (Grime 1979) and the study of biomass partitioning can broaden our understanding of the competition processes (Newton and Jolliffe 1993). However, fmdings on the effect of competition on biomass partitioning have been inconsistent (McConnaughay and Bazzaz 1991; Newton and Jolliffe 1993; Nilsson and Albrektson 1993). The possibility of an age-dependence of the observed results could not be tested, but it may affect the interpretation of the biomass allocation patterns (J.P. Kimmins pers. comm.). The prioritized order for the allocation of photosynthates in plants is not always followed. This may be partly because field situations are complex, and reaction to competition/stress can be species specific (McConnaughay and Bazzaz 1991) or site specific, i.e., trees may have different priorities in allocating photosynthates under different situations and with interaction with various factors other than competition/stress. McConnaughay and Bazzaz (1991) further argued that the 157 inconsistencies in results may be partly related to differences in the duration of restriction treatments. The possible underestimation of fine root and mycorrhizas biomass is not evaluated in this thesis as the measurement of those components is rather time consuming (Kurz 1989b; Vogt et al. 1980; Chapter 2). Therefore, the lack of significant differences between treatments in biomass allocation in various tree components found in this study can be the result of three possibilities: (1) biomass allocation may be regulated by the limited nutrient availabihties, as implicated by the resource depletion hypothesis (Chapter 3). Tree components might have responded similarly to belowground nutrient availability and competition stress, by incurring equivalent declines in relative growth rates (Newton and Jolliffe 1993); (2) the biomass allocation pattern was altered by the fertilizer apphcation two growing seasons before sampling. The N addition might have substantially alleviated the nutritional stress both for the control and treated treatments; and (3) duration of the understory removal should be longer to observe significant changes. The fact that there were substantial amounts of understory root biomass and there were salal resprouts in the control plots indicated there was still competition from belowground in the treated plots. This might be confounded by the N addition effect. Timmer and Munson (1991) and Timmer et al. (1991) reported that containerized black spruce seedlings raised at high fertilizer regimes (nutrient loaded) or treated with steady-state nutrient preconditioning (maintaining stable tissue N, P, and K concentrations during the exponential growth phase) during greenhouse culture had greater height growth and dry matter production, especially on more N-deficient sites, than those cultured by conventional fertilization in the first growing season after outplanting. Similar results were observed for red pine (Pinus resinosa Ait.) seedlings (Miller and Timmer 1994). However, the life span of such an effect was 158 not examined since their experiments only looked at the outplanting performance of the seedlings in the first growing season. I suspect that the nutrient reserves built up during the nursery culftuing would not last for very long, e.g., more than one growing season, for the effects on seedling growth to be significant. Therefore, nursery nutrient loading should not be a factor confounding biomass allocations in my study. Furthermore, the seedlings used for this study were uniformly treated in the nursery and should not have confounded the treatment effect. The , 5 N study illustrated that there was a strong competition from the understory vegetation for nutrients. Nitrogen-15 labeled fertilizer immobilized by understory vegetation was 3.3 to 16 times as much as that taken up by trees in the control plots and understory removal increased 1 5 N uptake of trees by 2 to 9 times. Those results supported the third hypothesis in Chapter 1; that understory competition reduces the uptake of fertilizer N by crop trees. The recovery numbers show that fertilization efficiency is greatly reduced by understory competition. Therefore, some form of understory control (e.g., scarification) is recommended before applying fertilizer N. No operationally applicable means of understory control have been found. The apphcabihty of repeated fertilization to reduce and eliminate salal growth (Prescott et al. 1993a) on the CWH zone is being tested in a separate study. Maxinhzing the uptake of N fertilizer in the first growing season should be an important issue, as other studies indicate that there is little additional fertilizer N taken up beyond the first growing season (Preston and Mead 1994). As an evergreen shrub, salal is a persistent, pervasive woody perennial and is a strong competitor with coniferous species (Fraser et al. 1993). It may not be possible to change the pattern of 1 5 N distribution/uptake between the understory and trees by altering the timing of apphcation and choice of fertilizer forms. This again illustrates the importance of understory vegetation control to increase fertilizer use efficiency on those sites. 159 The 1 5 N study showed that western hemlock and Sitka spruce were more responsive to understory removal than western redcedar in terms of 1 5 N uptake by above- and belowground tree components. This seems to agree with the results of Messier (1991) in which western hemlock and Sitka spruce were found to respond more than did western redcedar to all treatments or site differences that affected or were related to nutrient availabihty. Fraser et al. (1995) also found little evidence of a competitive effect of salal on cedar. But salal was found to compete with hemlock. One of the problems with N fate studies is that apphcation of N induces added N interaction, or ANI (Jenkinson et al. 1985). Addition of inorganic N increases the mineralization and availabihty to plants of native soil N, with the added N entering into the SOM through pool substitution (Hart et al. 1986) and other mechanisms (Jenkinson et al. 1985). This ANI apparently decreased the 1 5 N recovery percentage regardless of treatments in this study, because the mineralized native soil N substituted for the apphed N in some way. There are no studies to show how to correct the ANI effect in 1 5 N uptake studies. The difference method may solve some of the problems, i.e., it recovers the native N substituted by the 1 5 N; however, it introduces another error, i.e., ANI effect may increase native N mineralization. One of the other possibilities is to use very small quantities of N addition (tracer level). However, whether ANI disappears or not with tracer level N apphcation still needs to be examined. Residual fertilizer availabilities Since apphed N is quickly immobilized by the SOM and becomes relatively unavailable, there is a need to investigate the transformations and availabihty of the residual N in forest soils. There is a general lack of such information on forest soils. In the agricultural science field, there 160 is more information on residual N availability, due to the fact that N fertilization in agricultural production has a much longer history and a much wider apphcation. Therefore, historically there has been a greater economic impetus to study residual N availabilities in agricultural soils. The aerobic incubation study for the samples with different 1 5 N residence times showed that there were periods of net immobilization and net mineralization (Chapter 5), for both apphed and native soil N. Therefore, 1 5 N and native soil N are always in a dynamic system. However, the present data set could not explain why the dynamic process always makes the apphed N more stably immobilized into the SOM and less available for plant uptake. Green and Blackmer (1995) found that nonlabeled N was mineralized at rates five to seven times greater than labeled N in a 42-week aerobic incubation after soils were treated with 15N-labeled NO3" and with crop residues. The real question is, if the recently immobilized N is always more mineralizable and available, why is the mmeralization-immobilkation exchange not limited to the inorganic N and the more recently immobilized N pool. In other words, how did N in the stable SOM fractions become exchanged with 1 5 N and become inorganic forms if they are somewhat protected from being mineralized and if they have long turnover times. Green and Blackmer (1995) suggested that the period of net mineralization could be attributed primarily to mineralization of native N from SOM. In a related paper, Blackmer and Green (1995) discussed, in much detail, N turnover by sequential immobilization and mineralization during residue decomposition in agricultural soils and pointed out that sequential mineralization and immobilization is a potential error in tracer studies because the assumption that mineralization and immobilization occur simultaneously is violated. The results in Chapter 5 supported the fourth hypothesis that samples with longer 1 5 N residence time have lower mineralization potentials. However, the contradictory results for the 161 24-hr treatment between the aerobic and anaerobic incubations showed that the anaerobic incubation method may be unsuitable for evaluating N mineralization potentials when there is an excessive amount of inorganic N in the samples. Anaerobic incubation encouraged irnmobilization of inorganic N in such a situation. This also violates the hypothesis of Adams and Attiwill (1986) and Myrold (1987) that anaerobic incubation measures microbial biomass N similar to the fumigation incubation method. The extraction study showed that greater percentages of total soil N and apphed N were extracted from samples of the 24-hr treatment than from that of the 7- and 31-month treatments, regardless of the extraction method used (Chapter 6). The extractability ratios were highest in samples with recent 1 5 N labeling. In this experiment, 47.7% of the recovered 1 5 N was found in the insoluble fractions of the SOM for the 24-hr treatment; and more 1 5 N was incorporated into the humin fraction in samples of the 7- and 31-month treatments than that of the 24-hr treatment. Thus, hypotheses number 5 and 6 (Chapter 1) are accepted. Strickland et al. (1992) reported that the majority of the 1 5 N incorporated after a 60-day exposure to 1 5 N was in a non-active pool. They suggested that rapid incorporation of 1 5 N into stable organic fractions may be through incorporation of N into existing SOM heteropolycychcs. He et al. (1988) explained that the stabilization of fertilizer N is mainly due to reversion of immobilized N into less available forms through repeated mineralization and immobilization cycles and the accumulation of recalcitrant components of microbial origin. Hart et al. (1993) found reciprocal relationships in recovery of 1 5 N in microbial and SOM pools over time in a forest soil and suggested N was being transferred between the two fractions. However, the mechanisms for the incorporation of N into the slower turnover pools remain poorly understood (Tiessen et al. 1984; Schimel and Firestone 1989a; Strickland et al. 1992). The benefits of studying residual N 162 fertilizer dynamics are to provide information on its availabilities and to explain why there is little N fertilizer uptake beyond the first growing season after fertilizer apphcation. Since the incorporation of C and N in forest soils is affected by many factors, in fiiture studies the effect of management practices, such as residue management, on N incorporation should be investigated. Microbial competition for N The research on microbial biomass and N dynamics was conducted in an effort to provide new hypotheses regarding the regeneration failures on the CH cutover sites. Microbial growth in soils is usually limited by C availabihty (Zak et al. 1994). Site preparation techniques routinely used in British Columbia, such as slash-burning after clearcutting, reduce the availability of organic C. Microbial biomass in humus and mineral soil has been shown to be reduced by clearcutting due to decreased levels of root growth and exudation; slash-burning also reduces soil microbial biomass (cf. Pietikainen and Fritze 1995). However, with the growth of trees in the second rotation, more organic C input into the soil system is expected, through increased root exudates and increased litter (both above- and belowground) input. The growing root has been found to be a significant C source for microbial populations (cf. Wheatley et al. 1990). Therefore, there is the potential for microbial populations to assimilate more N in the process to maintain then increased growth with greater available energy sources. Soil microorganisms competing for nutrients may greatly limit then availabihty for root acquisition (Hendrickson and Richardson 1993). However, based on the data obtained in the present study (Chapter 7), the hypothesis (hypothesis number 7) that microbial biomass acted as a net sink in the 10-year-old plantation, where nutrients were deficient, by immobilizing more N 163 into the microbial N pool than in the 3-year-old plantations, where no nutrient deficiency was observed, had to be rejected. Although the role of microbial populations in regulating soil N dynamics has been emphasized (Wheatley et al. 1990; Fenn et al. 1993; Holmes and Zak 1994; Joergensen et al. 1995) , a definitive relationship between the microbial biomass N and the rates of soil N mineralization is generally not found (Fenn et al. 1993; Hossain et al. 1995). Results in Chapter 5 and in the study on microbial biomass in the clearcut chronosequence (Chang and Trofymow 1996) revealed that there is very poor correlation between C and net N mineralization. This might be a problem with pool-size based estimates which measure the sum of competing consumptive and productive processes (Hart et al. 1994). Hart et al. (1994) showed that gross rates and net rates of N processes are sometimes uncorrelated; rates of CO2 evolution and gross N mineralization and immobilization were significantly correlated but not for CO2 evolution and net N mineralization rates. Therefore, studies on gross rates of N transformations should be conducted to assess the relationships among C and N cycling processes in forest ecosystems. Similar to results obtained in Chapter 7, Ross et al. (1995) also reported that, corresponding to the lowest soil microbial N and mineralizable N, the highest microbial C/N ratios were found in litter and mineral soil of a treatment with whole trees and original forest floor removed. The causes for differences in microbial C/N ratios have been extensively discussed (Wheatley et al. 1990; Pietikainen and Fritze 1993; Henrot and Robertson 1994; Hossian et al. 1995; Joergensen et al. 1995); however, there has been little discussion on the implications of C/N ratio changes for N processes. The factors affecting microbial C/N ratios and then implications for N fertility is an area for further research. 164 Future research Dissolved organic N (DON) Recent pubhcations in Nature (Northup et al. 1995; Chapin 1995) have stirred up new interests in looking at the availabihty of DON for plant uptake, although direct and indirect evidences have long suggested the possibility of DON uptake by plants directly or through the help of ectomycorrhizal fungi (Mori et al. 1979; van Cleve et al. 1986; Abuzinadah and Read 1989; Finlay 1992; Chapin et al. 1993). The emphasis is that this mechanism may be important in forest sites having low N contents, high hgnin or tannin contents and low rates of N mineralization and nitrification or which have strong acidity (van Cleve et al. 1986; Northup et al. 1995). On the SCHIRP sites, ericoid mycorrhizae associated with salal have been found to be able to use simple organic N (Xiao 1994). Van Cleve and White (1980) observed the turnover of soluble organic N as a major pathway in the forest floor of paper birch (Betula papyrifera) stands in Alaska and found evidence that trees can utilize soluble organic N as a major source of N. It would be interesting to investigate if ectomycorrhizae associated with conifers are also able to use simple organic N, such as DON, on the CH cutover sites. Although conifers have a low capacity for competing with salal for limited N supply, it is highly possible for conifers to take up DON, especially since there is generally no nitrate present in those forest floors and mineral soils and ammonium content is usually very low and therefore cannot meet the requirement for ecosystem nutrient uptake. A comparison on the role of DON in tree nutrition between the CH and HA sites should be conducted, as the CH sites can be viewed as poor sites and the HA sites are relatively richer nutritionally. 165 Dissolved organic N is important in the mineralization of SOM (He et al. 1988; Strickland et al. 1992; Asmar et al. 1994), but the mechanisms for N transformations between inorganic N and DON and between DON and more stable N forms in SOM need to be investigated. Field N mineralization and nitrification studies More detailed studies of in situ N mineralization would help elucidate the relative supply and demand of N in the SCHTRP sites. This would also help in evaluating the possible uptake of DON by those forests. Of particular interest is that NO3-N is normally absent from the SCHTRP sites (Chapter 5) and many other strongly acidic coniferous forest soils (Donaldson and Henderson 1990a and b). Donaldson and Henderson (1990a) summarized the hypotheses regarding low NO3 levels in some forest soils as follows: (1) low soil pH inhibits nitrification; (2) autotrophic nitrifiers have low ability to compete for NH4 with heterotrophic organisms and plants, thus low NH4-N contents limit NO3-N production; (3) low levels of other nutrients may limit nitrification; and (4) allelopathic chemicals in forest soils limit nitrification. Better understanding of the mechanisms limiting nitrification should provide useful insights into the functions of those CH type forest ecosystems. Also of interest, relating to nitrification, is denitrification measurement in the CH forests. I will further discuss denitrification in the next section. Nitrogen budget Chapter 4 showed that only 57 to 87% of the fertilizer N was recovered in the CH forests. I argued that besides sampling errors and N transfer out of the plot boundaries, leaching may be the most probable pathway of loss. Denitrification loss may be minimal because of low N0 3-N levels 166 in the forest floor throughout the year. However, direct evidence was lacking. In view of the thick humus layers and high CEC levels of the humus layers, leaching loss may also be very small. Then what are the pathways for the lost N in those forests? Would the following argument be vahd? There is so little NO3-N present in the forests because denitrification or leaching of NO3-N is so fast that as soon as NO3 is formed it is lost from the ecosystem? As was discussed in Chapter 4, Martin (1985) reported that denitrification was the main pathway for the unaccounted N of some recent clearcut mesic sites in the Coastal Western Hemlock zone on southwestern Vancouver Island. However, the vahdity of this observation needs to be tested for the CH cutover sites. Or is ammonia volatilization also a possible pathway for N loss from these ecosystems? Although ammonia volatilization loss is usually found under forest situations when urea is apphed (Nason et al. 1988; Craig and Wollum II 1982), little volatilization loss was found when inorganic N other than urea was apphed. Also if denitrification, leaching, and volatilization do occur, what are the seasonal dynamics of those processes? Microbial turnover and Nfluxes The question raised in Chapter 7 needs to be addressed. Investigation on the N fluxes through the microbial biomass may further explain whether microbial competition for N is an important factor in nutrient deficiency problems observed in 8- to 10-year-old plantations on CH cutover sites. Microbial biomass turnover and N fluxes through microbial biomass are central to understanding the functioning of soil nutrient cycles. Soil faunal populations rely on and regulate microbial populations (Elliott et al. 1984; Clarholm 1984). There have been studies on the dynamics of soil microbial biomass (this study) and soil faunal populations (Battigelh et al. 1994) in the SCHIRP sites. Future studies should be 167 directed to study the interactions between soil microbial and faunal population dynamics and then controls on N availabihties. In other words, integration of some of the past work on SCHJRP sites would yield more useful information on the functioning of those forest ecosystems. Conclusions 1. The salal dominated understory in the CH cutover sites is a strong competitor for fertilizer N apphed to those ecosystems by immobilizing 3 to 12 times more 1 5 N than the crop trees in the understory-remaining treatment. 2. Abundant understory growth changed the 1 5 N distribution in the studied soil-plant system. Greater proportions of the recovered 1 5 N were found in the litter and standing dead components in the control plots than in the treated plots, while the reverse was true for 1 5 N distribution in the soil. The relative availabihties of the residual fertilizer N was thus altered. 3. Understory removal treatment unproved tree height, diameter and biomass growth due to increased nutrient availabihties; however, understory removal did not seem to affect the distribution patterns of biomass in the different tree components. Biomass and diameter growth were better parameters for quantifying changes induced by treatments. 4. Fertilizer N was being immobilized by the SOM within hours after apphcation. Residual fertilizer N was found to be more mineralizable in the samples after a short period in the soil than after a longer period, showing reduced availabihties of apphed N with increased residence time in the soil. 5. The extractabilities of the residual fertilizer N decreased with increasing fertilizer N residence time. The decreased mineralization potentials and extractabilities with increasing I 5 N residence time were attributed to the incorporation of 1 5 N into stable SOM fractions. 6. Microbial biomass N constituted a significant proportions of the total soil N; however, microbial biomass N pool size is a poor indicator for N availabihty. Microbial population may be a strong competitor for the limited amount of available N in the CH clearcut sites experiencing nutrient deficiency problems. 169 REFERENCES Aarnio, T., and Martikainen, P. J. 1994. Mineralization of carbon and nitrogen in acid forest soil treated with fast and slow-release nutrients. Plant and Soil 164, 187-193. Aber, J.D., Melillo, J.M., and McClaugherty, C A . 1990. Predicting long-term patterns of mass loss, nitrogen dynamics, and soil organic matter formation from initial fine fitter chemistry in temperate forest ecosystems. Can. J. Bot. 68:2201-2208. Abrahamson, W.G. 1975. Reproductive strategies in dewberries. Ecology 56: 721-726. Abrahamson, W.G. 1979. Patterns of resource allocation in wildflower populations of fields and woods. Amer. J. Bot. 66: 71-79. Abrahamson, W.G., and Caswell, H. 1982. On the comparative allocation of biomass, energy, and nutrients in plants. Ecology. 63:982-991. Abrahamson, W.G., and Gadgil, M.D. 1973. Growth form and reproductive effort in goldenrods (Solidago, Compositae). Am Nat. 107:651-661. Abuzinadah, R.A., and Read, D.J. 1989. The role of proteins in the nitrogen nutrition of ectomycorrhizal plants. IV. The utilization of peptides by birth (Betula pendula L.) infected with different mycorrhizal fungi. NewPhytol. 112:55-60. Adams, M. A., and AttiwilL P. M. 1986. Nutrient cycling and nitrogen mineralization in eucalypt forests of south-eastern Australia. Plant and Soil 92, 341-362. Adams, M.A., and AttiwilL P.M. 1982. Nitrogen mineralization and nitrate reduction in forests. Soil Biol. Biochem 14:197-202. Agren, G.I., Axelsson, B, Flower-Ellis, J.G.K., Linder, S., Persson, H., Staaf, H., and Troeng, E. 1980. Annual carbon budget for a young Scots pine. In: Persson, T. (ed.) Structure and function of a northern coniferous forest - an ecosystem study. Ecol. Bull. (Stockholm) 32: 307-313. Alexander, M. 1977. Introduction to Soil Microbiology. 2nd ed. John Wiley and Sons, New York. Anderson, J.P.E., and Domsch, K.H. 1980. Quantities of plant nutrients in the microbial biomass of selected soils. J. Soil Sci. 130:211-216. Armleder, H.M., and Dawson, R. J., 1992. Logging on mule deer winter range: an integrated management approach. For. Chron. 68:132-137. 170 Asmar, F., Eiland, F., and Nielsen, N. E. 1994. Effect of extracellular-enzyme activities on solubilization rate of soil organic nitrogen. Biol. Fertil. Soils. 17:32-38. Axelsson, E., and Axelsson, B. 1986. Changes in carbon allocation patterns in spruce and pine trees foUowing irrigation and fertilization. Tree Physiology, 2: 189-204. Azam, F., Haider, K , and Malik, K 1985. Transformations of 1 4 C labeled plant components in soil in relation to immobilization and reminerahzation of 1 5 N fertilizer. Plant Soil 86: 15-25. Azam, F., Mulvaney, R. L., and Stevenson, F. J. 1989a. Chemical distribution and transformations of non-symbioticaUy fixed 1 5 N in three soils. Soil Biol. Biochem. 21, 849-855. Azam, F., Mulvaney, R.L., and Stevenson, F.J. 1988. Cnxantification and potential avahabihty of non-symbioticaUy fixed 1 5 N in son. Biol. Fertil. Sou's. 7: 32-38. Azam, F., Stevenson, F. J., and Mulvaney, R. L. 1989b. Chemical extraction of newly immobilized 1 5 N and native soh N as influenced by substrate addition rate and soh treatments. Soh Biol. Biochem 21, 715-722. Baath, E., Lundgren, B., and Soderstrom, B. 1981. Effects of nitrogen fertilization on the activity and biomass of fungi and bacteria in a podzohc soil. Zentralbl. Bakteriol. Parasitenkd. Infektionskr. Hyg. Abt. 1: Orig. Reihe C, 2: 90-98. Barclay, H. J., Pang, P.C., and PoUard, D.F.W. 1986. Aboveground biomass distribution within trees and stands in thinned and fertilized Douglas-fir. Can. J. For. Res. 16:438-442. BattigeUi, J.P., Berch, S.M., and MarshaU, V.G. 1994. Soil fauna communities in two distinct but adjacent forest types on northern Vancouver Island, British Columbia. Can. J. For. Res. 24: 1557-1566. Beatty, S.W. 1984. Influence of microtopography and canopy species on spatial patterns of forest understory plants. Ecology, 65:1406-1419. BeU, K L . , and Binkley, D. 1989. Soh nitrogen mineralization and immobilization in response to periodic prescribed fire in a lobloUy pine plantation. Can. J. For. Res. 19:816-820. Binkley, D., and Hart, S. 1989. Components of nitrogen availabUity assessments in forest soils. Adv. Soil Sci. 10:57-116. Bjorkman, E., Lundeberg, G., and Nommik, H. 1967. Distribution and balance of ^ N labeled fertilizer nitrogen appUed to young pine trees (Pinus sylvestris L.). Studia ForestaUa Suecica. 48:1-15 Blackmer, A.M., and Green, C.J. 1995. Nitrogen turnover by sequential immobiUzation and mineralization during residue decomposition in soils. Soh Sci. Soc. Am. J. 59: 1052-1058. 171 Bosatta, E . , and Berendse, F. 1984. Energy or nutrient regulation of decomposition: implications for the minerahzation-immobilization response to perturbations. Soil Biol. Biochem 16: 63- 67. Bosatta, E . , and Staaf, H. 1982. The control of nitrogen turnover in forest litter. Oikos 39: 143- 151. Brand, D.G., and Janas, P.S. 1988. Growth and acclimation of planted white pine and white spruce seedlings in response to environmental conditions. Can. J. For. Res., 18:320-329. Brand, D.G., and Magnussen, S. 1988. Asymmetric, two-sided competition in even-aged monocultures of red pine. Can. J. For. Res. 18:901-910. Bremer, E . , and van KesseL C. 1990. Extractability of microbial 1 4 C and 1 5 N following addition of variable rates of labeled glucose and (NFL^SC^ to soil. Soil Biol. Biochem 22:707-713. Bremner, J.M. 1965. Total nitrogen. In Methods of Soil Analysis, Part 2. Edited by C A . Black, D.D. Evans, J.L. White, L . E . Enseninger and F.E. Clark. Agron. #9, A m Soc. Agron., Inc., Wisconsin, pp. 1149-1178. Bremner, J.M., and Muh/aney, C S . 1982. Nitrogen-total. In Methods of soil analysis. Part 2. Chemical and microbiological properties. Edited by A . L . Page. A m Soc. Agron. and Soil Sci. Soc. A m , Madison, WI. pp.595-624. Brookes, P. C , Kragt, J. F., Powlson, D. S., and Jenkinson, D. S. 1985a. Chloroform fumigation and the release of soil nitrogen: the effects of fumigation time and temperature. Soil Biol. Biochem 17,831-835. Brookes, P . C , Landman, A. , Pruden, G., and Jenkinson, D.S. 1985b. Chloroform fumigation and the release of soil nitrogen: a rapid direct extraction method to measure microbial biomass nitrogen in soil. Soil Biol. Biochem 17:837-842. Bunnell, F.L. 1990. Reproduction of salal (Gaultheria shallori) under forest canopy. Can. J. For. Res. 20:91-100. Cade-Menun, B. 1995. Phosphorus forms of podzolic soils of northern Vancouver Island and then use by western red cedar. Ph.D. Thesis, Dept. Of Soil Science, Univ. Of British Columbia, Vancouver, B.C. 216 p. Carlyle, J . C , and Malcolm, D.C. 1986. Nitrogen availabihty beneath pure spruce and mixed larch+spruce stands growing on a deep peat. I. Net N mineralization measured by field and laboratory incubations. Plant Soil, 93: 95-113. 172 Chang, S.X., and Trofymow, J.A. 1996. Microbial respiration and biomass (SIR) in soils of old- growth and regenerating forests on northern Vancouver Island, British Columbia. Biol. Fertil. Soils. (In press) Chapin, F.S., HL 1995. New cog in the nitrogen cycle. Nature 377: 199-200. Chapin, F.S., HI. Moilanen, L., and Kielland, K. 1993. Preferential use of organic nitrogen for growth by a non-mycorrhizal arctic sedge. Nature 361: 150-153. ChappelL H.N., Cole, D.W., GesseL S.P., and Walker, R.B. 1991. Forest fertilization research and practice in the Pacific Northwest. Fertilizer Research 27:129-140. Chichester, F.W., Legg, J.O., and Stanford, G. 1975. Relative mineralization rates of indigenous and recently incorporated 15N-labeled nitrogen. Soil Sci. 120: 455-460. Clarholm, M. 1984. Heterotrophic, free-living protozoa: neglected microorganisms with an important task in regulating bacterial populations. In: Klug, M. J., and Reddy, C.A. (eds.) Microbial ecology - congresses. Am. Soc. Microbiol. Washington, D.C. pp.321-326. Clay, D.E., and Malzer, G.L. 1993. Comparison of two chemical methods for extracting residual N fertilizer. Biol. Fertil. Soils. 15:179-184. Clinton, P.W., and Mead, D.J. 1994a. Competition for nitrogen between Pinus radiata and pasture. I. Recovery of X^N after one growing season. Can. J. For. Res. 24: 882-888. Clinton, P.W., and Mead, D.J. 1994b. Competition for nitrogen between Pinus radiata and pasture, n. Trends in plant and soil processes. Can. J. For. Res. 24. 889-896. Cole, D.W. 1981. Nitrogen uptake and translocation by forest ecosystems. Ecol. Bull. (Stockholm) 33: 219-232. Collins, H.P., Rasmussen, P.E., and Douglas, C.L., Jr. 1992. Crop rotation and residue management effects on soil carbon and microbial dynamics. Soil Sci. Soc. Am. J. 56:783- 788. Craig, J.R., and Wohum II, A.G. 1982. Ammonia volatilization and soil nitrogen changes after urea and ammonium nitrate fertilization of Pinus taeda L. Soil Sci. Soc. Am. J. 46: 409- 414. Dalai, R.C., and Mayer, R. J. 1987. Long-term trends in fertility of soils under continuous cultivation and cereal cropping in southern Queensland. VU. Dynamics of nitrogen mineralization potentials and microbial biomass. Aust. J. Soil Res. 25:461-472. Davidson, E. A., Hart, S. C , and Firestone, M. K. 1992. Internal cycling of nitrate in soils of a mature coniferous forest. Ecology 73, 1148-1156. 173 Del Moral, R., and Cates, R.G. 1971. Allelopathic potential of the dominant vegetation of western Washington. Ecology, 52:1030-1037. deMontigny, L. 1992. An investigation into the factors contributing to the growth-check of conifer regeneration on northern Vancouver Island. Ph.D. thesis. Faculty of Forestry, University of British Columbia. Vancouver. 191pp. deMontigny, L.E., Preston, C M , Hatcher, P.G., and Kogel-Knaber, I. 1993. Comparison of humus horizons from two ecosystem phases on northern Vancouver Island using l^C NMR spectroscopy and CuO oxidation. Can. J. Soil Sci. 73:9-25. Deyoe, D.R., and Dunsworth, B.G. 1988. Duration and intensity of vegetation control: estabhshing seedling growth potential. FRDA rep. No. 026. Ministry of Forests and Lands, Victoria, B.C., Canada. 10pp. Diaz-Ravina, M., Acea, M. J., and Carballas, T. 1993a. Microbial biomass and its contribution to nutrient concentrations in forest soils. Soil Biol. Biochem 25: 25-31. Diaz-Ravina, M., Acea, M.J., and Carballas, T. 1993b. Seasonal fluctuations in microbial populations and available nutrients in forest soils. Biol. Fertil. Soils 16: 205-210. Dickson, D.A., and SaviU, P.S. 1974. Early growth of Picea sitchensis (Borg.) Carr. on deep oligotrophic peat in northern Ireland. Forestry, 47:57-88. Donaldson, J. M., and Henderson, G. S. 1990a. Nitrification potential of second-succession upland oak forests: I. Mineralization and nitrification during laboratory incubations. Soil Sci. Soc. Am J. 54: 892-897. Donaldson, J. M., and Henderson, G. S. 1990b. Nitrification potential of second-succession upland oak forests: n. Regulation of ammonium-oxidizing bacteria populations. Soil Sci. Soc. Am J. 54,898-902. Duxbury, J.M., andNkambule, S.V. 1994. Assessment and significance of biologically active soil organic nitrogen. In: Defining soil quality for a sustainable envhonment. SSSA Spec. Publ. No. 35. SSSA, 677 S. Segoe Rd, Madison, WI 53711. pp. 125-146. Edmonds, R.L. 1987. Decomposition rates and nutrient dynamics in small-diameter woody litter in four forest ecosystems in Washington, U.S.A. Can. J. For. Res. 17:499-509. Edmonds, K L . , and Chappell, H.N. 1994. Relationship between soil organic matter and forest productivity in western Oregon and Washington. Can. J. For. Res. 24:1101-1106. Elliott, E.T., Coleman, D.C, Ingham, R.E., and Trofymow, J.A. 1984. Carbon and energy flow through microflora and microfauna in the soil subsystem of terrestrial ecosystems. In: Klug, M.J., and reddy, C A . (eds.) Microbial ecology - congresses. Am Soc. Microbiol. Washington, D.C. pp.424-433. 174 Entry, J.A., Stark, N.M., and Loewenstein, H. 1986. Effects of timber harvesting on microbial biomass fluxes in a northern Rocky Mountain forest soil. Can. J. For. Res. 16:1076-1081. Fahey, T.J. 1983. Nutrient dynamics of above-ground detritus in lodgepole pine (Pinus contorta ssp. Latifolia) ecosystems, southeastern Wyoming. Ecol. Monographs. 53:51-72. Fenn, M.E., Poth, M.A., Dunn, P.H., and Barro, S.C. 1993. Microbial N and biomass, respiration and N mineralization in soils beneath two chaparral species along a fire-induced age gradient. Soil Biol. Biochem. 25:457-466. Fessenden, R.J., Calvert, R.F., and Armson, K.A. 1971. Effect of some fertilizers and simazine on the activity of the microorganisms in jack pine humus. The Forestry Chronicle 47: 227-228. Finlay, R.D., Frostegard, A., and Sonnerfeldt, A.M. 1992. Utilization of organic and inorganic nitrogen sources by ectomycorrhizal fungi in pure culture and in symbiosis with Pinus contorta Dougl. ex Loud. NewPhytol. 120:105-115. Fisher, F.M., and Gosz, J.R. 1986. Effects of plants on net mineralization of nitrogen in forest soil microcosms. Biol. Fertil. Soils. 2:43-50. Flint, L.E., and Childs, S.W. 1987. Effect of shading, mulching and vegetation control on Douglas-fir seedling growth and soil water supply. For. Ecol. Manage., 18:189-203. Foster, N.W., Beauchamp, E.G., and Corke, C.T. 1980. Microbial activity in a Pinus banksiana Lamb, forest floor amended with nitrogen and carbon. Can. J. Soil Sci. 60: 199-209. Foster, N.W., Beauchamp, E.G., and Corke, C.T. 1985. Immobilization of nitrogen-15-labeled urea in a jack pine forest floor. Soil Sci. Soc. Am. J. 49:448-452. Fraser, L. 1993. The influence of salal on planted hemlock and cedar saplings on northern Vancouver Island. M.Sc. Thesis, Dept. Of Botany, Univ. Of British Columbia, Vancouver, B.C. 116 p. Fraser, L., Chanway, CP., and Turkinton, R. 1995. The competitive role ofGaultheria shallon on planted western hemlock and western red cedar saplings on northern Vancouver Island. For. Ecol. Manage. 75: 27-39. Fraser, L., Turkinton, R., and Chanway, CP. 1993. The biology of Canadian weeds. 102. Gaultheria shallon Pursh. Can. J. Plant Sci. 73: 1233-1247. Frazer, D.W., McColL J.G., and Powers, R.F. 1990. Soil nitrogen mineralization in a clearcutting chronosequence in a northern California conifer forest. Soil Sci. Soc. Am J. 54: 1145- 1152. 175 Fried, M., and Broeshart, H. 1974. Priming effect of nitrogen fertilizers on soil nitrogen. Soil Sci. Soc. Am. P. 38: 859. Fritze, H., Pennanen, T., and Pietikainen, J. 1993. Recovery of soil microbial biomass and activity from prescribed burning. Can. J. For. Res. 23:1286-1290. Fritze, H., Smolander, A , Levula, T., Kitunen, V., and Malkonen, E. 1994. Wood-ash fertilization and fire treatments in a Scots pine forest stand: effects on the organic layer, microbial biomass, and microbial activity. Biol. Fertil. Soils. 17:57-63. Gallardo, A., and Schlesinger, W.H. 1990. Estimating microbial biomass nitrogen using the fumigation-incubation and fumigation-extraction methods in a warm temperate forest soil. Soil Biol. Biochem 22: 927-932. Germain, A. 1985. Fertilization of stagnated Sitka spruce plantations on northern Vancouver Island. M.Sc. Thesis, Faculty of Forestry, University of British Columbia, Vancouver, B.C., Canada. Gianello, C , and Bremner, J.M. 1986. Comparison of chemical methods of assessing potentially available organic nitrogen in soil. Commun. Soil Sci. Plant Anal. 17: 215-236. Green, R.N. 1990. Response of Douglas-fir seedlings to salal control. FRDA 2.17.1., Ministry of Forests, Victoria, B.C., Canada. 21pp. Green, R.N., and Khnka, K. 1994. A field guide to site identification and interpretation for the Vancouver Forest Region. Land Management Handbook 28. Province of British Columbia, Ministry of Forests. 1994. 285 pp. Green, C. J., and Blackmer, A.M. 1995. Residual decomposition effects on nitrogen availabihty to corn following corn or soybean. Soil Sci. Soc. Am J. 59: 1065-1070. Grier, C.C., Lee, K M . , and Archibald, K M . 1984. Effect of urea fertilization on allometric relations in young Douglas-fir trees. Can. J. For. Res. 14: 900-904. Griffith, KP., Caldwell, B.A., Cromack, K , Jr., and Morita, K Y . 1990. Douglas-fir forest soils colonized by ectomycorrhizal mats. I. Seasonal variation in nitrogen chemistry and nitrogen cycle transformation rates. Can. J. For. Res. 20: 211-218. Grime, J.P. 1979. Plant Strategies and Vegetation Processes. John Wiley & Sons. 222pp. Hadas, A , Molina, J.A.E., Feigenbaum, S., Clapp, C.E. 1987. Simulation of nitrogen-15 immobilization by the model of NCSOIL. Soil Sci. Soc. Am. J. 51: 102-106. Hadas, A., Sofer, M., Molina, J. A. E., Barak, P., and Clapp, C. E. 1992. Assimilation of nitrogen by soil microbial population: NH4 versus organic N. Soil Biol. Biochem. 24:137- 143. 176 Haggar, J.P., and EweL J.J. 1995. Establishment, resource acquisition, and early productivity as determined by biomass aUocation patterns of three tropical tree species. For. Sci. 41: 689- 708. Harper, J.L., and Ogden, J. 1970. The reproduction strategy of higher plants. I. The concept of strategy with reference to Senecio vulgaris L. J. Ecol. 58: 681-698. Harris, M. M., and Riha, S. J. 1991. Carbon and nitrogen dynamics in forest floor during short- term laboratory incubations. Soil Biol. Biochem. 23,1035-1041. Hart, P.B.S., Rayner, J.H., and Jenkinson, D.S. 1986. Influence of pool substitution on the interpretation of fertilizer experiments with 1 5 N. J. Soil Sci. 37: 389-403. Hart, S. C , Firestone, M. K., Paul E. A., and Smith, J. L. 1993. Flow and fate of soil nitrogen in an annual grassland and a young mixed-conifer forest. Soil Biol. Biochem. 25, 431-442. Hart, S.C., and Firestone, M.K 1991. Forest floor-mineral soil interactions in the internal nitrogen cycles of an old-growth forest. Biogeochem 12: 103-127. Hart, S.C., Nason, G.E., Myrold, D.D., and Perry, D.A. 1994. Dynamics of gross nitrogen transformations in an old-growth forest: the carbon connection. Ecology 75: 880-891. Harvey, A.E., Jurgensen, M.F., and Larsen, M.J. 1980. Ecology of ecotomycorrhizae in northern Rocky Mountain forests. In Environmental consequences of timber harvesting in Rocky Mountain coniferous forests: symposium proceedings. U.S. Dep. Agric. For. Serv. Gen. Tech. Rep. Int-90. pp. 188-210. Hauck, R.D., and Bremner, J.M. 1976. Use of tracers for soil and fertilizer nitrogen research. Adv. Agron. 28: 219-266. He, X.-T., Stevenson, F. J., Mulvaney, R. L., and Kelley, K. R. 1988. Incorporation of newly immobilized 1 5 N into stable organic forms in soil. Soil Biol. Biochem 20,75-81. Heilman, P.E., Dao, T.M., Cheng, H.H., Webster, S.R., and Christensen, L. 1982a. Comparison of fall and spring apphcation of 15N-labeled urea to Douglas-fir. U. Fertilizer recovery in trees and soil after 2 years. Soil Sci. Soc. Am J. 46:1300-1304. Heilman, P.E., Dao, T.M., Cheng, H.H., Webster, S.R., and Harper, S.S. 1982b. Comparison of fall and spring apphcation of 15N-labeled urea to Douglas-fir. I. Growth response and nitrogen levels in fohage and soil. Soil Sci. Soc. Am J. 46:1293-1299. Hendrickson, O.Q., and Richardson, J. 1993. Nested forest nutrient cycles: implications for plantation management. The Forestry Chronicle 69: 694-698. 177 Henrot, J., and Robertson, G.P. 1994. Vegetation removal in two soils of the humid tropics: effects on microbial biomass. Soil Biol. Biochem 26: 111-116. Hickman, J.C. 1977. Energy allocation and niche differentiation in four co-existing annual species of polygonum in western north America. J. Ecol. 65:317-326. Hoefer, P., and Bratton, G.F. 1988. The role of trees and shrubs as economic enterprises and wildlife habitat development in the Great Plains. USDA Forest Services General Tech. Rep., RM-158. p. 109-112. Holmes, W.E., and Zak, D.R. 1994. Soil microbial biomass dynamics and net nitrogen mineralization in northern hardwood ecosystems. Soil Sci. Soc. Am. J. 58:238-243. Hossain, A.K.M.A., Raison, KJ. , and Khanna, P.K 1995. Effects of fertilizer apphcation and fire regime on soil microbial biomass carbon and nitrogen, and nitrogen mineralization in an Australian subalpine eucalypt forest. Biol. Fertil. Soils 19: 246-252. Hulm, S.C., and K. Killham. 1990. Response over two growing seasons of a Sitka spruce stand to 15N-urea fertilizer. Plant and Soil. 124:65-72. Jamieson, N., and Killham, K. 1994. Biocide manipulation of N flow to investigate root/microbe competition in forest soil. Plant Soil 159: 283-290. Jansson, S. L., and Persson, J. 1982. Mineralization and immobilization of soil nitrogen. In Nitrogen in Agricultural Soil, ed. F J Stevenson, p.p. 229-252. American Society of Agronomy, Madison WL Janzen, K A., Shaykewich, C. F., and Goh, T. B. 1988. Stabilization of residual C and N in soil. Can. J. Soil Sci. 68, 733-745. Jenkinson, D. S., and Powlson, D. S. 1976. The effects of biocidal treatments on metabohsm in soil. V. A method for measuring soil biomass. Soil Biol. Biochem 8, 209-213. Jenkinson, D. S., Fox, K H , and Rayner, J.H. 1985. Interactions between fertilizer nitrogen and soil nitrogen-the so-called 'priming' effect. J. Soil Sci. 36, 425-444. Jenkinson, D.S., and Ladd, J.N. 1981. Microbial biomass in soil: measurements and turnover. In Soil biochemistry. Edited by E.A. Paul and J.N. Ladd. Vol. 5, pp.415-471. Dekker, New York. Joergensen, R.G., Anderson, T.H., and Wolters, V. 1995. Carbon and nitrogen relationships in the microbial biomass of soils in beech (Fagus sylvatica L.) forests. Biol. Fertil. Soils 19: 141-147. 178 Johnson, D. W., Edwards, N. T., and Todd, D. E. 1980. Nitrogen mineralization, immobilization and nitrification following urea fertilization of a forest soil under field and laboratory conditions. Soil Sci. Soc. Am J. 44: 610-616. Johnson, D.W., and Edwards, N.T. 1979. The effects of stem ghdling on biogeochemical cycles within a mixed deciduous forest in eastern Tennesssee n. Soil nitrogen mineralization and nitrification rates. Oecologia 40: 259-271. Jones, J.M., and Richards, B.N. 1978. Fungal development and the transformation of 1 5 N labeled amino- and ammonium-nitrogen in forest soils under several management regimes. Soil Biol. Biochem. 10: 161-168. Juma, N. G., and Paul E. A. 1984. Mineralizable soil nitrogen: Amounts and extractability ratios. Soil Sci. Soc. Am. J. 48, 76-80. Jurgensen, M.F., Larsen, M. J., and Harvey, A.E. 1980. Microbial processes associated with nitrogen cycling in northern Rocky Mountain forest soils. In: Environmental Consequences of Timber Harvesting in Rocky Mountain Coniferous Forests. P. 175-188. USDA For. Serv. Gen. Tech. Rep. INT-90. 526p. Keenan, R. J., Prescott, C.E., and Kimmins, J.P. 1993. Mass and nutrient content of woody debris and forest floor in western red cedar and western hemlock forests on northern Vancouver Island. Can. J. For. Res. 23:1052-1059. Keeney, D. R. 1980. Prediction of soil N availabihty in forest ecosystems: A literature review. For. Sci. 26:159-171. Keeney, D. R. 1982. Nitrogen-availability indices. In Methods of Soil Analysis. Agron. #9. Part 2. 2nd ed. Ed. A L Page, R H Miller and D R Keeney. pp. 711-734. Am Soc. Agron. Inc., Wisconsin. Keeney, DR., and Bremner, J.M. 1966. A chemical index of soil nitrogen availabihty. Nature 211: 892-893. Keeney, D.R., and Nelson, D.W. 1982. Nitrogen-inorganic forms. In Methods of Soil Analysis, Part 2. 2nd ed. Edited by A.L. Page, R.H. Miller and D.R. Keeney. Agron! #9, Am Soc. Agron., Inc. Wisconsin, pp.643-698. Kelley, K.R., and Stevenson, F.J. 1985. Characterization and extractability of immobilized 1 5 N from the soil microbial biomass. Soil Biol. Biochem. 17: 517-523. Kelly, J. M., and Henderson, G. S. 1978. Effects of nitrogen and phosphorus additions on deciduous litter decomposition. Soil Sci. Soc. Am. J. 42, 972-976. Keyes, M.R., and Grier, C.C. 1981. Above- and below-ground net production in 40-year-old Douglas-fir stands on low and high productivity sites. Can. J. For. Res. 11: 599-605. 179 Khanna, P. H., and Ulrich, B. 1984. Soil characteristics influencing nutrient supply in forest soils. In Nutrition of plantation forests. Ed. G D Bowen and E K S Nambiar. pp. 79-117. Academic Press Inc. (London) Ltd. Kinnnins, J.P. 1987. Forest Ecology. MacMillan, New York. 531pp. Knowles, R., and Lefebvre, J. 1972. Field, greenhouse and laboratory studies on the transformation and translocation of 15N-labeled urea in a boreal forest black spruce system, p.349-358. In: Isotopes and radiation in soil-plant relationships including forestry. IAEA, Vienna. Kowalenko, C.G., Ivarson, K.C., and Cameron, DR. 1978. Effect of moisture content, temperature and nitrogen fertilization on carbon dioxide evolution from field soils. Soil Biol. Biochem 19: 417-423. Kowalenko, C.G. 1989. The fate of apphed nitrogen in a Fraser Valley soil using 1 5 N in field microplots. Can. J. Soil Sci. 69: 825-833. Kurz, W. A. 1989a. Net primary production, production allocation, and fohage efficiency in second growth Douglas-fir stands with differing site quality. Ph.D. thesis, Faculty of Forestry, The University of British Columbia, Vancouver, B.C., Canada. 224 pp. Kurz, W.A. 1989b. Significance of shifts in carbon allocation patterns for long-term site productivity research. Pp. 149-164. In: Dyck, W.J., and Mees, C A . (eds.) Research strategies for long-term site productivity. Proceedings, ffiA/BE A3 Workshop, Seattle, WA, August 1988. D5A/BE A3 Report No. 8. Forest Research Institute, New Zealand, Bulletin 152. Ledgard, S.F., Smith, G.S., and Sprosen, M.S. 1992. Fate of 1 5 N labeled nitrogen fertilizer apphed to Iriwifruit (Actinidia deliciosd) vines. I. Recovery in plant and soil. Plant Soil 147: 49-57. Legg, J. O., Chichester, F. W., Stanford, G., and DeMar, W. H .1971. Incorporation of 1 5 N- tagged mineral nitrogen into stable forms of soil organic nitrogen. Soil Sci. Soc. Am P. 35, 273-276. Lewis, T. 1982. Ecosystems of the Port McNeill Block 4 of Tree Farm License 25. Internal Rep., western Forest Products Ltd., Vancouver, B.C., Canada. 55pp. Lieffers, V.J., and Titus, S.J. 1989. The effects of stem density and nutrient status on size inequality and resource allocation in lodgepole pine and white spruce seedlings. Can. J. Bot. 67:2900-2903. Lundgren, B. 1982. Bacteria in a pine forest soil as affected by clear-cutting. Soil Biol. Biochem. 14:537-542. 180 MacLean, A. A. 1964. Measurement of nitrogen supplying power of soils by extraction with sodium bicarbonate. Nature (London). 3, 1307-1308. Mahendrappa, M.K., Foster, N.W., Weetman, G.F., and Krause, H.H. 1986. Nutrient cycling and availability in forest soils. Can. J. Soil Sci. 66: 547-572. Martens, K 1990. Contribution of rhizodeposits to the maintenance and growth of soil microbial biomass. Soil Biol. Biochem 22: 141-147. Martikainen, P. J., Aarnio, T., Taavitsainen, V.-M., Paivinen, L., and Salonen, K 1989. Mineralization of carbon and nitrogen in soil samples taken from three fertilized pine stands: long-term effects. Plant and Soil 114,99-106. Martikainen, P. J., and Palojarvi, A. 1990. Evaluation of the fumigation method for the determination of microbial C and N in a range of forest soils. Soil Biol. Biochem 22: 797- 802. Martin, W.L. 1985. Post-clearcutting forest floor nitrogen dynamics and regeneration response in the coastal western hemlock wet subzone. Ph.D. thesis, Faculty of Forestry, The University of British Columbia, Vancouver, B.C., Canada. 350 pp. Marumoto, T., Anderson, J.P.E., and Domsch, K.H. 1982. Mineralization of nutrients from soil microbial biomass. Soil Biol. Biochem 14:469-475. McConnaughay, K.D.M., and Bazzaz, F.A. 1991. Is physical space a soil resource? Ecology 72: 94-103. McDonald, M.A., Hawkins, B.J., Prescott, C.E., and Kimmins, J.P. 1994. Growth and fohar nutrition of western red cedar fertilized with sewage sludge, pulp sludge, fish silage and wood ash on northern Vancouver Island. Can. J. For. Res. 24: 297-301. McGill, W. B., and Paul, E. A. 1976. Fractionation of soil and 1 5 N nitrogen to separate the organic and clay interactions of immobilized N. Can. J. Soil Sci. 56, 203-212. McGill, W.B., and Myers, RJ.K. 1987. Controls on dynamics of soil and fertilizer nitrogen. In Soil fertility and organic matter as critical components of production systems. SSSA Spec. Pub. No. 19, pp.73-99. McGill, W.B., Cannon, K.R., Robertson, J.A., and Cook, F.D. 1986. Dynamics of soil microbial biomass and water-soluble organic C in Breton L after 50 years of cropping to two rotations. Can. J. Soil Sci. 66: 1-19. Mead, D.J., and Preston, C M . 1994. Distribution and retranslocation of ^ N in lodgepole pine over eight growing seasons. Tree Physiology. 14: 389-402. 181 Mead, D.J., and Pritchett, W.L. 1975a. Fertilizer movement in a Slash pine ecosystem. I. Uptake of N and P and N movement in the soil. Plant & Soil. 43:451-465. Mead, D.J., and Pritchett, W.L. 1975b. Fertilizer movement in a Slash pine ecosystem. II. N distribution after two growing seasons. Plant & Soil. 43:467-478. Melin, J., and Nommik, H. 1988. Fertilizer nitrogen distribution in a Pinus sylvestrisl Picea abies ecosystem, central Sweden. Scand. J. For. Res. 3:3-15. Melin, J., Noirrmik, H., Lohm, U., and Flower-Ellis, J. 1983. Fertilizer nitrogen budget in a Scots pine ecosystem attained by using root-isolated plots and A^N tracer technique. Plant Soil, 74:249-263. Messier, C. 1991. Factors limiting early conifer growth in salal-dominated cutovers in northern Vancouver Island, British Columbia. Ph.D. thesis, University of British Columbia, Faculty of Forestry, Vancouver. Messier, C. 1992. Effects of neutral shade and growing media on growth, biomass allocation, and competitive ability of Gaultheria shallon. Can. J. Bot. 70:2271-2276. Messier, C. 1993. Factors limiting early growth of western red cedar, western hemlock and Sitka spruce seedlings on ericaceous-dominated clearcut sites in coastal British Columbia. For. Ecol. Manage., 60:181-206. Messier, C , and Kimmins, J.P. 1990. Nutritional stress in Picea sitchensis plantations in coastal British Columbia: the effect of Gaultheria shallon and declining site fertility. Water, Ah, SoilPollut. 54:257-267. Messier, C , and Kimmins, J.P. 1991. Above- and below-ground vegetation recovery in recently clearcut and burned sites dominated by Gaultheria shallon in coastal British Columbia. For. Ecol. Manage. 46:275-294. Miller, B.D., and Timmer, V.R. 1994. Steady-state nutrition of Pinus resinosa seedlings: response to nutrient loading, irrigation and gardening regimes. Tree Physiology, 14: 1327-1338. Molina, J.A.E., Clapp, C.E., Shaffer, M.J., Chichester, F.W., Larson, W.E. 1983. NCSOIL, a model of nitrogen and carbon transformation in the soil: description, cahbration, and behavior. Soil Sci. Soc. Am. J. 47: 71-90. Mori, S., Nishimura, Y., and Uchino, H. 1979. Nitrogen absorption by plant root from the culture medium where organic and inorganic nitrogen coexist. I. Effect of pretreatment nitrogen on the absorption of treatment nitrogen. Soil Sci. Plant Nutr. (Tokyo) 25: 39-50. Morrison, I.K., and Foster, N.W. 1977. Fate of urea fertilizer added to a boreal forest Pinus banksianahamb. stand. Soil Sci. Soc. Am. J. 41: 441-448. 182 Munson, A.D., and Timmer, V.R. 1990. Site-specific growth and nutrition of planted Picea mariana in the Ontario Clay Belt. III. Biomass and nutrient allocation. Can. J. For. Res. 20:1165-1171. Myrold, D.D. 1987. Relationship between microbial biomass nitrogen and nitrogen availabihty index. Soil Sci. Soc. Am. J. 51:1047-1049. Nadelhoffer, K J., and Aber, J. D. 1984. Seasonal patterns of ammonium and nitrate uptake in nine temperate forest ecosystems. Plant and Soil 80, 321-335. Nambiar, E.K.S., and Bowen, G.D. 1986. Uptake, distribution and retranslocation of nitrogen by Pinus radiata from ^N-labeled fertilizer apphed to podzolized sandy soil. For. Ecol. Manage. 15:269-284. Nason, G.E., Pluth, D.J., and McGill, W.B. 1988. Volatilization and fohar recapture of ammonia foUowing spring and faU appUcation of nitrogen-15 urea to a Douglas-fir ecosystem Soil Sci. Soc. Am J. 52: 821-828. Neary, D.G., Rockwood, D.L., Comerford, N.B., Swindel, B.F., and Cooksey, T.E. 1990. Importance of weed control, fertilization, irrigation, and genetics in slash and lobloUy pine early growth on poorly drained Spodsols. For. Ecol. Manage., 30:271-281. Newsome, T.H. 1985. Fertilization of western hemlock with nitrogen, phosphorus, and lime: a greenhouse study. M.Sc. thesis, University of British Columbia, Department of Soil Science. 129 pp. Newton, P.F., and JoUiffe, P.A. 1993. Above-ground dry matter partitioning, size variation, and competitive processes within second-growth black spruce stands. Can. J. For. Res. 23:1917-1929. Niemela, S., and Sundman, V. 1977. Effects of clear-cutting on the decomposition of bacterial populations of northern spruce forest soh. Can. J. Microbiol. 23:131-138. Nilsson, U., and Albrektson, A. 1993. Productivity of needles and aUocation of growth in young Scots pine trees of different competitive status. For. Ecol. Manage. 62:173-187. Nilsson, U., and HaUgren, J.-E. 1993. Changes in growth aUocation owing to competition for light in young fertilized Norway spruce trees. For. Ecol. Manage. 62: 157-172. Nommik, H. 1966. The uptake and translocation of fertilizer 1 5 N in young trees of Scots pine and Norway spruce. Studia Forestalia Suecica. Nr 35, p. 1-18. Nommik, HL, and MoUer, G. 1981. Nitrogen recovery in soil and needle biomass after fertilization of a Scots pine stand, and growth responses obtained. Studia Forestalia Suecica. Nr 159. p. 1-37. 183 Nommik, H., and Larsson, K. 1989. Assessment of fertilizer nitrogen accumulation in Pinus sylvestris trees and retention in soil by A^N recovery technique. Scand. J. For. Res. 4:427- 442. Nonnnik, H., and Popovic, B. 1971. Recovery and vertical distribution of ^N-labeled fertilizer nitrogen in forest soil. Studia Forestalia Suecica 92, 1-19. Northup, K R., Yu, Z., Dahlgren, R. A., and Vogt, K. A. 1995. Polyphenol control of nitrogen release from pine litter. Nature 377: 227-229. Oliver, CD. , and Larson. B.C. 1990. Forest Stand Dynamics. Biological Resource Management Series. McGraw-HhT, New York, 467pp. Olson, R.A. 1979. Isotope studies on soil and fertilizer nitrogen. In: Isotope and radiation in research on soil-plant relationships. IAEA, Vienna, 1979. p.3-32. Overrein, L.N. 1967. Immobilization and mineralization of tracer nitrogen in forest raw humus. Plant Soil, 27: 1-19. Overrein, L.N. 1970. Tracer studies on nitrogen immobilkation-mineralization relationships in forest raw humus. Plant Soil, 32: 478-500. Overrein, L.N. 1972. Isotope studies on nitrogen in forest soil. EL Distribution and recovery of l̂ N-labeled fertilizer nitrogen in a 40-month lysimeter investigation. Meddr Norske SkogforsVes, 122: 308-324. Norwegian Forest Research Institute, As, Norway, pp.308- 324. Pang, P.C.K 1985. Distribution and recovery of 1 5 N after fertilization of Douglas-fir saplings with different nitrogen sources. Plant and Soil. 84:167-174. Parker, T., and Johnson, D. 1987. Branching and terminal growth of western red cedar. Northwest Science. 61:7-12. Patra, D.D., Brookes, P.C, Coleman, K , and Jenkhison, D.S. 1990. Seasonal changes of soil microbial biomass in an arable and a grassland soil which have been under uniform management for many years. Soil Biol. Biochem 22:739-742. Paul, E.A., and Clark, F.E. 1989. Soil Microbiology and Biochemistry. Academic Press Inc. 1989. p. 133. Pearson, J.A., Fahey, T.J., and Knight, D.H. 1984. Biomass and leaf area in contrasting lodgepole pine forests. Can. J. For. Res. 14: 259-265. Persson, H. 1983. The distribution and productivity of fine roots in boreal forests. Plant and Soil, 71: 87-101. 184 Pietikainen, J., and Fritze, H. 1993. Microbial biomass and activity in the humus layer foUowing burning: short-term effects of two different fires. Can. J. For. Res. 23:1275-1285. Pietikainen, J., and Fritze, H. 1995. Clear-cutting and prescribed buniing in coniferous forest: comparison of effects on soh fungal and total microbial biomass, respiration activity and nitrification. SoU Biol. Biochem 27: 101-109. Polglase, P. J., Comerford, N. B., and Jokela, E. J. 1992. Mineralization of nitrogen and phosphorus from soil organic matter in southern pine plantations. Soil Sci. Soc. Am. J. 56, 921-927. Popovic, B. 1977. Effects of ammonium nitrate and urea fertilizers on nitrogen minerahzation, especiaUy nitrification, in a forest soU. Royal CoUege of Forestry, Stockholm, Department of Forest Ecology and Forest Sohs, Research Notes 30: 1-33. Popovic, B., and Nommik, H. 1972. Use of the 1 5 N technique for studying fertilizer nitrogen transformation and recovery in forest soh. In: Isotopes and radiation in plant-soil relationships including forestry, IAEA, Vienna, 359-368. Porter, L. K., Stewart, B. A., and Haas, H. J.1964. Effects of long-time cropping on hydrolyzable organic nitrogen fractions in some Great Plains soils. Soil Sci. Soc. Am Proc. 28, 368- 370. Powers, R.F. 1980. Mineralizable nitrogen as an index of nitrogen availabihty to forest trees. Soil Sci. Soc. Am. J. 44:1314-1320. Prescott, C.E., and Preston, C M . 1994. Nitrogen mineralization and decomposition in forest floors in adjacent plantations of western red cedar, western hemlock and Douglas-fir. Can. J. For. Res. 24:2424-2431. Prescott, C.E., and Weetman, G.F. 1994. Salal cedar hemlock integrated research program: a synthesis. Faculty of Forestry, University of British Columbia, Vancouver, B.C. 85 p. Prescott, C.E., Coward, L.P., Weetman, G.F., and Gessel, S.P. 1993a. Effects of repeated nitrogen fertilization on the ericaceous shrub, salal {Gaultheria shallon), in two coastal Douglas-fir forests. For. Ecol. Manage. 61:45-60. Prescott, C.E., McDonald, M.A., and Weetman, G.F. 1993b. Avahabihty of N and P in the forest floor of adjacent stands of western red cedar-western hemlock and western hemlock- amabihs fir on northern Vancouver Island. Can. J. For. Res. 23:605-610. Preston, C. M., and Mead, D. J. 1994. Growth-response and recovery of 15N-fertilizer one and eight growing seasons after apphcation to lodgepole pine in British Columbia. For. Ecol. Manage. 65,219-229. 185 Preston, C M . , Marshall, V.G., McCullough, K., and Mead, D.J. 1990. Fate of 1 5 N -labeled fertilizer applied to snow at two forest sites in British Columbia. Can. J. For. Res. 20:1583-1592. Proe, M.F., and Millard, P. 1994. Relationships between nutrient supply, nitrogen partitioning and growth in young Sitka spruce (Picea sitchensis). Tree Physiology, 14: 75-88. Richards, B.N., Smith, J.E.N., White, G.J., and Charley, J.L. 1985. Mineralization of soil nitrogen in three forest communities from the New England region of New South Wales. Aust. J. Ecol. 10:429-441. Ross, D. J., and Sparling, G.P. 1993. Comparison of methods to estimate microbial C and N in htter and soil under Pinus radiata on a coastal sand. Soil Biol. Biochem 25: 1591-1599. Ross, D. J., Sparling, G.P., Burke, C M . , and Smith, C T . 1995. Microbial biomass C and N, and mineralizable-N, in htter and mineral soil under Pinus radiata on a coastal stand: influence of stand age and harvest management. Plant Soil 175: 167-177. Ross, D. J.Y. 1988. Soil microbial biomass estimated by the fumigation-incubation procedure: seasonal fluctuations and influence of soil moisture content. Soil Biol. Biochem 19:397- 404. Ross, M.A., and Harper, J.L. 1972. Occupation of biological space during seedling estabhshment. J. Ecol. 60: 77-88. Salonius, P.O., and Mahendrappa, M.K. 1975. Microbial respiration and exchangeable ammonium in podzol organic horizon materials treated with urea. Can. J. For. Res. 5: 731-734. SAS Institute Inc. 1989. SAS/STAT User's Guide. Version 6, 4th edition. Volume 1 (943p.) and Volume 2 (846p.). Cary, NCSAS Institute Inc., 1989 Schimel, J. P., and Firestone, M. K. 1989a. Nitrogen incorporation and flow through a coniferous forest soil profile. Soil Sci. Soc. Am J. 53, 779-784. SchimeL J. P., and Firestone, M. K 1989b. Inorganic N incorporation by coniferous forest floor material. Soil Biol. Biochem 21,41-46. SchimeL J.P, Jackson, L.E., and Firestone, M.K 1989. Spatial and temporal effects on plant- microbial competition for inorganic nitrogen in a California annual grassland. Soil Biol. Biochem 21: 1059-1066. Schimel, J.P., Heifer, S., and Alexander, I.J. 1992. Effects of starch addition on N turnover in Sitka spruce forest floor. Plant Soil 139: 139-143. 186 Schnitzer, M. 1982. Organic matter characterization. In Methods of Soil Analysis. Agron. #9. Part 2. 2nd ed. Ed. A L Page, R H Miller and D R Keeney. pp. 643-698. Am. Soc. Agron. Inc., Wisconsin. Shen, S.M., Hart, P.B.S., Powlson, D.S., and Jenkinson, D.S. 1989. The nitrogen cycle in the Broadbalk wheat experiment: 15N-labeled fertilizer residues in the soil and in the soil microbial biomass. Soil Biol. Biochem. 21:529-533. Shumway, J.S. 1978. Predicting nitrogen fertilizer response in unthinned stands of Douglas-fir. Agron. Abstr. 1978. p. 193. Smethurst, P.J., and Nambiar, E.K.S. 1990a. Distribution of carbon and nutrients and fluxes of mineral nitrogen after clear-felhng a Pinus radiata plantation. Can. J. For. Res. 20:1490- 1497. Smethurst, P.J., and Nambiar, E.K.S. 1990b. Effects of slash and litter management on fluxes of nitrogen and tree growth in a young Pinus radiata plantation. Can. J. For. Res. 20:1498- 1507. Smith, S. J., and Power, J. F. 1985. Residual forms of fertilizer nitrogen in a grassland soil. Soil Sci. 140,362-367. Smith, S.J., and Stanford, G. 1971. Evaluation of a chemical index of soil nitrogen availabihty. Soil Sci. 111:228-232. Snyder, J.D., and Trofymow, J.A. 1984. A rapid accurate wet oxidation diffusion procedure for determining organic and inorganic carbon in plant and soil samples. Comm. Soil Sci. Plant Anal. 15:587-597. Soderstrom, B., Baath, E., and Lundgren, B. 1983. Decrease in soil microbial activity and biomass owing to nitrogen amendments. Can. J. Microbiol. 29: 1500-1506. Stams, A. J. M., Lutke, I. J., Marnette, E. C. L., Beemsterboer, B., and Woittiez, J. R. W. 1990. Conversion of 15N-ammonium in forest soils. Plant and Soil 125, 129-134. Stanford, G. 1969. Extraction of soil organic nitrogen by autoclaving in water. EL A kinetic approach to estimating the NaOH-distiUable fraction. Soil Sci. 107: 323-328. Stanford, G., and DeMar, W. H. 1969. Extraction of soil organic nitrogen by autoclaving in water: I. The NaOH-distillable fraction as an index of nitrogen availabihty in soils. Soil Sci. 107, 203-205. Stanford, G., and DeMar, W.H. 1970. Extraction of soil organic nitrogen by autoclaving in water. 3. Diffusible ammonia, an index of soil nitrogen availabihty. Soil Sci. 109: 190-196. 187 Stanford, G., and Smith, S. J. 1978. Oxidative release of potentially mineralizable soil nitrogen by acid permanganate extraction. Soil Sci. 126, 210-218. Stevenson, F.J. 1986. Cycles of the soil. Wiley, New York. Stevenson, F.J., Kidder, G., and Tilo, S.N. 1967. Extraction of organic nitrogen and ammonium from soil with hydrofluoric acid. Soil Sci. Soc. Am. Proc. 31: 71-75. Stewart, B. A., Porter, L. K., and Johnson, D. D. 1963. Lnmobihzation and mineralization of nitrogen in several organic fractions of soil. Soil Sci. Soc. Am. Proc. 27, 302-304. Stewart, G., and Forsling, C L . 1931. Surface runoff and erosion in relation to soil and plant cover on high grazing lands of central Utah. J. Am. Soc. Agron. 23:815-832. Stockdale, E. A., and Rees, R. M. 1994. Relationships between biomass nitrogen and nitrogen extracted by other nitrogen availabihty methods. Soil Biol. Biochem. 26, 1213-1220. Strickland, T .C , Sollins, P., Rudd, N., and SchimeL D.S. 1992. Rapid stabilization and immobilization of 1 5 N in forest and range soils. Soil Biol. Biochem. 24: 849-855. Sundman, V., Huhta, V., and Niemela, S. 1978. Biological changes in northern spruce forest soil after clear-cutting. Soil Biol. Biochem., 10:393-397. Theodorou, C , and Bowen, G.D. 1983. Effects of temperature, moisture and htter on nitrogen mineralization in Pinus radiata forest soils. Aust. For. Res. 13: 113-119. Thompson, W.A., and Weetman, G.F. 1992. Operational fertilization of young stands of cedar, hemlock and spruce on northern Vancouver Island: five year results and projected economic impact. Report to the B.C. Ministry of Forests. 51 p. Tiessen, H., Stewart, J.W.B., and Hunt, H.W. 1984. Concepts of soil organic matter transformations in relation to organo-mineral particle size fractions. Plant Soil 76: 287-295. Timmer, V.R., Armstrong, G., and Miller, B.D. 1991. Steady-state nutrient preconditioning and early outplanting performance of containerized black spruce seedlings. Can. J. For. Res. 21: 585-594. Timmer, V.R., and Munson, A.D. 1991. Site-specific growth and nutrition of planted Picea mariana in the Ontario Clay Belt. IV. Nitrogen loading response. Can. J. For. Res. 21: 1058-1065. van Cleve, K , and Moore, T.A. 1978. Cumulative effects of nitrogen, phosphorus and potassium fertilizer additions on soil respiration, pH and organic matter content. Soil Sci. Soc. Am. J. 42: 121-124. 188 van Cleve, K., HeaL O.W., and Roberts, D. 1986. Bioassay of forest floor nitrogen supply for plant growth. Can. J. For. Res. 16: 1320-1326. van Cleve, K , and White, R. 1980. Forest floor nitrogen dynamics in a 60-year-old paper birch ecosystem in interior Alaska. Plant and Soil, 54: 359-381. van den Driessche, R. 1984. Nutrient storage, retranslocation and relationship of stress to nutrition. In Nutrition of Plantation Forests. Edited by G.D. Bowen and E.K.S. Nambiar. pp. 181-209. Academic Press Inc. (London) Ltd. Vance, E.D., Brookes, P.C., and Jenkinson, D.S. 1987. An extraction method for measuring soil microbial biomass C. Soil Biol. Biochem 19:703-707. Viro, P.J. 1974. Effects of fibre on soil. In Fire and Ecosystems. Edited by T.T. Kozlowski and C.E. Ahlgen. Academic Press, New York. pp.7-45. Vitousek, P. M., and Matson, P. A. 1985. Distribution, nitrogen availabihty, and nitrogen losses in an intensively managed loblolly pine plantation. Ecology 66, 1360-1376. Vitousek, P. M., and Matson, P. A. 1988. Nitrogen transformations in a range of tropical forest soils. Soil Biol. Biochem 20,361-367. Vitousek, P.M. 1981. Clear-cutting and the nitrogen cycle. In Terrestrial Nitrogen Cycles. Ecol. Bull. (Stockholm). Edited by F.E. Clark and T. Rosswall. 33:631-642. Vitousek, P.M., Matson, P.A., and Van Cleve, K. 1989. Nitrogen availabihty and nitrification during succession: primary, secondary, and old-field seres. In Ecology of Arable Land. Edited by M. Clarhohn and L. Bergstrom. pp.161-171. Vogt, K.A., Edmonds, K L . , Grier, C.C., and Piper, S.K 1980. Seasonal changes in mycorrhizal and fibrous-textured root biomass in 23- and 180-year-old Pacific silver fir stands in western Washington. Can. J. For. Res. 10: 523-529. Vogt, K.A., Grier, C.C., and Vogt, D.J. 1986. Production, turnover, and nutrient dynamics of above- and belowground detritus of world forests. Adv. In Ecological Res. 15: 303-377. Waring, S.A., and Bremner, J.M. 1964. Ammonium production in soil under waterlogged conditions as an index of nitrogen availabihty. Nature 201:951-952. Weber, M.G. 1987. Decomposition, htter fall, and forest floor nutrient dynamics in relation to fire in eastern Ontario jack pine ecosystems. Can. J. For. Res. 17:1496-1506. Weber, M.G., Methven, I.K, and van Wagner, C.E. 1985. The effect of forest floor manipulation on nitrogen status and tree growth in an eastern Ontario jack pine ecosystem. Can. J. For. Res. 15:313-318. 189 Weetman, G.F., Fournier, K , Barker, J., and Schnorbus-Panozzo, E. 1989a. Foliar analysis and response of fertilized chlorotic western hemlock and western red cedar reproduction on salal-dominated cedar-hemlock cutovers on Vancouver Island. Can. J. For. Res. 19:1512- 1520. Weetman, G.F., Fournier, K , Barker, J., Schnorbus-Panozzo, E., and Germain, A. 1989b. Fohar analysis and response of fertilized chlorotic Sitka spruce plantations on salal-dominated cedar-hemlock cutovers on Vancouver Island. Can. J. For. Res. 19:1501-1511. Weiner, J., and Thomas, S.C. 1986. Size variability and competition in plant monocultures. Oikos, 47:211-222. Wheatley, K , Ritz, K., and Griffiths, B. 1990. Microbial biomass and mineral N transformations in soil planted with barley, ryegrass, pea or turnip. Plant Soil 127:157-167. White, C. S., and Gosz, J. K 1987. Factors controlling nitrogen mineralization and nitrification in forest ecosystems in New Mexico. Biol. Fertil. Soils 5, 195-202. White, C.S. 1986. Effects of prescribed fire on rates of decomposition and nitrogen mineralization in a ponderosa pine ecosystem. Biol. Fertil. Soils 2:87-95. White, C.S., Gosz, J K , Horner, J.D., and Moore, D.I. 1988a. Seasonal, annual, and treatment- induced variation in available nitrogen pools and nitrogen cycling processes in soils of two Douglas-fir stands. Biol. Fert. Soils 6:93-99. White, C.S., Moore, D.I., Horner, J.D., and Gosz, J.K 1988b. Nitrogen mineralization- immobilization response to field N or C perturbations: an evaluation of a theoretical model. Soil Biol. Biochem 20: 101-105. Wollum, A.G., and Davey, C.B. 1975. Nitrogen accumulation, transformation and transport in forest soils. In Forest soils and forest land management. Proceedings of the Fourth North American Forest Soils Conference, p.p. 67-106. Edited by B. Bernier and C H . Winget. Laval Univ. Press, Quebec. Woods, L.E., Cole, C.V., Elliott, E.T., and Anderson, R.V. 1982. Nitrogen transformations in soil as affected by microbial-microfaunal interactions. Soil Biol. Biochem 14:93-98. Worsnop, G., and Will, G.M. 1980. Fate of 1 5 N urea fertilizer apphed to a recently thinned radiata pine stand on a pumice soil. New Zealand J. of Forestry Sci! 10:381-394. Wright, T.W., and Will, G.M. 1958. The nutrient content of Scots and Corsian pines growing on sand dunes. Forestry 31:13-25. Wu, J., Joergensen, K G . , Pommerening, B., and Chaussod, R. 1990. Measurement of soil microbial biomass C by fumigation-extraction: an automated procedure. Soil Biol. Biochem. 22:1167-1169. 190 Xiao, G. 1994. The role of root-associated fungi in the dominance ofGautheria shallon. Ph.D. Thesis, Dept. Of Botany, Univ. Of British Columbia, Vancouver, B.C. 137 p. Zak, D.R., Host, G.E., and Pregitzer, K.S. 1989. Regional variability in nitrogen mineralization, nitrification, and overstory biomass in northern Lower Michigan. Can. J. For. Res. 19: 1521-1526. Zak, D.R., Tilman, D., Parmeuter, R.R., Rice, C.W., Fisher, F.M., Vose, J., Milchunas, D., and Martin, C.W. 1994. Plant production and soil microorganisms in late-successional ecosystems: a continental-scale study. Ecology 75: 2333-2347. Zakharchenko, I.G. 1974. Supply of nitrogen with atmospheric precipitation and its losses during soil leaching in the Ukrainian Poless'ye and forest-steppe. Soviet Soil Sci. 6:63-67

Cite

Citation Scheme:

    

Usage Statistics

Country Views Downloads
United States 8 0
Japan 4 0
France 2 0
China 1 0
Germany 1 5
City Views Downloads
Ashburn 7 0
Tokyo 4 0
Unknown 3 5
Beijing 1 0
Sunnyvale 1 0

{[{ mDataHeader[type] }]} {[{ month[type] }]} {[{ tData[type] }]}

Share

Share to:

Comment

Related Items