Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Phosphorus forms of podzolic soils of northern Vancouver Island and their use by western red cedar 1995

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

Item Metadata

Download

Media
ubc_1995-059286.pdf
ubc_1995-059286.pdf [ 10.51MB ]
Metadata
JSON: 1.0088246.json
JSON-LD: 1.0088246+ld.json
RDF/XML (Pretty): 1.0088246.xml
RDF/JSON: 1.0088246+rdf.json
Turtle: 1.0088246+rdf-turtle.txt
N-Triples: 1.0088246+rdf-ntriples.txt
Citation
1.0088246.ris

Full Text

PHOSPHORUS FORMS OF PODZOLIC SOILS OF NORTHERN VANCOUVER ISLAND AND THEIR USE BY WESTERN RED CEDAR by BARBARA JEAN CADE-MENUN B. Sc. (Hons.), Queen's University, 1986 M. Sc., The University of British Columbia, 1989 A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE STUDIES Department of Soil Science We accept this thesis as conforming r\ ô~-tiie required standard THE UNIVERSITY OF BRITISH COLUMBIA October 1995 (c) Barbara Jean Cade-Menun, 1995 In presenting this thesis in partial fulfi lment 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 purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. (Signature) Department of The University of British Columbia Vancouver, Canada Date /I Oct I9 9<r- DE-6 (2/88) ABSTRACT After clear-cutting and slashburning the hemlock-amabilis fir (HA) forest types of northern Vancouver Island support good growth, but the trees on the cedar-hemlock (CH) forest types suffer a growth check which can be overcome with N and P fertilization. This study focussed on soil phosphorus (P) fractions in CH and HA forests. Extraction methods were evaluated for total, organic and available P. The Parkinson & Allen digest was better than the Saunders & Williams ignition method for total P. Both the Saunders & Williams method and the Bowman & Moir extraction procedure overestimated organic P. The Bray PI and Mehlich 3 procedures were suitable for available P. Extracts of forest floor samples by NaOH, NaOH-EDTA and Chelex in both water and NaOH were analysed by 31P NMR spectroscopy. The NaOH-EDTA extracted the greatest portion of the total P and yielded spectra with a greater diversity of P compounds. However, this extractant also maintained other ions in solution which reduced the quality of the spectra. Evaluation of P status in relation to soil chemistry of mature CH and HA forests revealed that CH forests had higher pH values and C concentrations in the forest floor. The CH forests also exhibited higher loss on ignition, wider C/N and C/P ratios, and increased concentrations of extractable Ca in mineral horizons. The HA forests had higher C concentrations in mineral horizons and higher concentrations of N, extractable Mg, Al and Fe, and more organically complexed Al and Fe. There were no significant differences in P levels between the forest types. P-31 NMR ii spectroscopy showed a diversity of compounds, and organic forms throughout the profile. The persistence of labile diester phosphates and wide C/N and C/P ratios suggest slow decomposition. Comparison of the P status and soil chemistry of mature CH forests to those after burning revealed increases in pH, available P, inorganic P and extractable N, and decreased organic P post- harvest. By 10 years postburn, significant reductions in organic P and organically bound Fe and Al were revealed in mineral horizons. P-31 NMR showed a shift to orthophosphate after burning, but a return to organic forms within 10 years in surface horizons. The results suggest that the burning of organic matter temporarily disrupts illuviation and the P cycle. In a pot study, cedar grown with high (50 mg P/2) or low (10 mg P/1) levels of phytic acid, ATP, glycerophosphate, pyrophosphate or KH2P04 showed adequate growth with all P forms but phytic acid, and grew best with the high rate of the P compounds. The poor growth with phytic acid was attributed to its binding of Ca, Zn and Cu. Utilization of organic P compounds was facilitated by various phosphatases, produced by cedar plants, mycorrhizae and/or rhizosphere microbes. Phosphorus did not appear to play an important role in the growth check problem of CH sites. iii TABLE OF CONTENTS Abstract ii Table of Contents iv List of Tables vi List of Figures viii Acknowledgements ix CHAPTER 1 Introduction 1 Obj ectives 3 Hypotheses 4 Thesis Structure 5 CHAPTER 2 Literature Review Forest Soil Phosphorus Cycles 6 Soil Phosphorus Methodology 2 5 CHAPTER 3 Study area 32 Sampling design 37 Study sites 38 CHAPTER 4 A Comparison of Methods for Total, Organic and Available P Introduction 42 Methods 46 Results 47 Discussion 53 Conclusions 57 CHAPTER 5 A Comparison of Soil Extraction Procedures for 31P NMR Spectroscopy Introduction 58 Methods 61 Results 65 Discussion 81 Conclusions 84 CHAPTER 6 A Comparison of P Forms on CH and HA Sites Introduction 86 Methods 8 7 Results 90 Discussion 108 Conclusions 118 iv CHAPTER 7 A Comparison of P Forms on CH Sites after Burning Introduction 12 0 Methods 122 Results 123 Discussion 149 Conclusions 161 CHAPTER 8 The Use of Organic P by Western Red Cedar Introduction 162 Methods 164 Results 168 Discussion 180 Conclusions 188 CHAPTER 9 General Conclusions 18 9 Literature Cited 198 v LIST OF TABLES Table 3-1: Horizon thickness 38 Table 4-1: Total P methods comparison 49 Table 4-2: Organic P methods comparison 51 Table 4-3 : Available P methods comparison 52 Table 5-1: Chemistry of soils used in extractant trial 62 Table 5-2: Interpretation of 31P NMR spectra 71 Table 5-3 : Percentage of P in compound classes 72 Table 5-4 : P extracted by various methods 73 Table 5-5 : Orthophosphate diester/monoester ratios 73 Table 5-6: Metals in each solution following extraction .74 Table 5-7: Proportion of P in P compounds after P addition 8 0 Table 6-1: CH and HA moisture content and pH 91 Table 6-2: ANOVA for CH-HA moisture, pH, C, N, C/N, LOI, Ca...92 Table 6-3 : CH-HA C, N, LOI, C/N 93 Table 6-4 : CH-HA extractable Ca, Mg, Fe, Al 94 Table 6-5: CH-HA pyrophosphate-extracted Fe, Al and Mn 95 Table 6-6 : ANOVA table for CH-HA Fe, Al, Mn 95 Table 6-7: CH-HA Fe, AL and Mn extracted by CBD and AAO 96 Table 6-8: CH-HA available, total and organic P, C/P ratio 99 Table 6-9: ANOVA table for CH-HA P fraction 10 0 Table 6-10: CH-HA P fractions .101 Table 6-11: CH-HA organic and inorganic P 101 Table 6-12 : CH-HA correlation matrix 103 Table 6-13: CH-HA P forms from NMR specta 106 Table 7-1: Postburn moisture content and pH 124 Table 7-2: ANOVA table for postburn moisture, pH, C, N, C/N....125 vi Table 7-3 : Postburn C, LOI and C/N 12 7 Table 7-4 : Postburn extractable Ca, Mg, Fe and Al 128 Table 7-5: Postburn pyrophosphate-extracted Fe, Al and Mn 130 Table 7-6 : ANOVA for postburn Fe, AL and Mn 131 Table 7-7: Postburn Fe, Al and Mn extracted by CBD and AAO 132 Table 7-8: Postburn available, total and organic P, C/P .135 Table 7-9 : ANOVA for postburn P fraction 13 6 Table 7-10: Postburn P fractions 137 Table 7-11: Postburn organic and inorganic P 138 Table 7-12 : Postburn correlation matrix 13 9 Table 7-13 : Postburn P forms, from NMR spectra 142 Table 7-14: Postburn P forms, from NMR spectra 146 Table 8-1: Greenhouse experiment diam. change, dry weight 169 Table 8-2: ANOVA for greenhouse expt.: growth, nutrient 170 Table 8-3: Greenhouse expt. myc. effects on growth, Zn 170 Table 8-4 : Greenhouse expt. foliar Ca, Cu, Zn 171 Table 8-5: Greenhouse expt. foliar N, P 174 Table 8-6: Greenhouse expt. col., soil P, acid phosphatase... 175 Table 8-7: ANOVA for greenhouse expt. Col, Soil P 176 Table 8-8: Greenhouse expt. myc. effects on acid phosphatase.176 Table 8-9: Greenhouse expt. alkaline phosphatase 178 Table 8-10: Correlation matrix, greenhouse expt 179 vii LIST OF FIGURES Figure 2-1: Structure of monoesters and diesters 12 Figure 2-2: Structures of organic P compounds 14 Figure 3-1: Map of Vancouver Island 3 2 Figure 3-2: A typical stand of the CH forest type 34 Figure 3-3: A typical stand of the HA forest type 3 6 Figure 3-4: Location of sampling 3 9 Figure 3-5: Typical 10, 5 and 0 year postburn sites 41 Figure 5-1A: Spectra of HA forest floor, var. extractants 66 Figure 5-1B: Spectra of CH forest floor, var. extractants 67 Figure 5-1C: Spectra of 0-YR forest floor, var. extractants... 68 Figure 5-1D: Spectra of 5-YR forest floor, var. extractants... 69 Figure 5-1E: Spectra of 10-YR forest floor, var. extractants..70 Figure 5-2: Metals in solution after adsorption trial 76 Figure 5-3A: Spectra after P addition, NaOH-EDTA extraction...78 Figure 5-3B: Spectra after P addition, Chelex-NaOH extraction.79 Figure 6-1: Organic, amorphous and crystalline Fe and Al 98 Figure 6-2 : Spectra of HA soil profiles 104 Figure 6-3 : Spectra of CH soil profiles 105 Figure 7-1: Organic, amorphous and crystalline Fe and Al 133 Figure 7-2: Spectra of mature CH soil profiles 141 Figure 7-3: Spectra of recently burned CH soil profiles 144 Figure 7-4: Spectra of 5-year postburn CH soil profiles 145 Figure 7-5: Spectra of 10-year postburn CH soil profiles 147 viii ACKNOWLEDGEMENTS A project such as this could not have been completed without the help of many people. I am indebted to my committee for their patience: Shannon Berch, for funding through NSERC and the BC Ministry of Forests, and for convincing me that I could master soil chemistry, despite my views to the contrary; Les Lavkulich, for lab space, manpower and Saturday morning chats; Caroline Preston, for countless hours on the NMR, for allowing me to "play' with extractants, and for keeping me abreast of the literature during my pregnancy and after my move; Cindy Prescott, for coordinating the SCHIRP project and for providing considerable input during the write up of this thesis; and to Lawrence Lowe, for his assistance prior to his retirement. I wish to thank Dr. R. A. Bowman, USDA- ARS, Akron, CO, for sharing his NaOH-EDTA extraction technique with me during a 1991 telephone conversation. I am grateful to the staff of Western Forest Products, especially Paul Bavis and Cindy Fox, for providing accommodation and assistance with field work. My sample collection could not have been completed without Buffy, Sandy, Scott, Laura, Shanghou, Chuck and the staff at the Port Hardy Emergency Room. At UBC Laura, Bernie, Lauch, Jane, Amy, Pam, Kim and Amanda helped with analyses, while Kevin, Ann and a host of co-op students assisted me at PFC. I am grateful to Dawn, Jane, Marcia, Pam, Kim and Amanda for help with my trees. I must also thank the people who stocked the vending machine in the basement of McMillan, and the manufacturers of the double chocolate brownies, for helping me through many a long night of lab work. I wish to thank my parents, Jim and Marilyn Cade, for their love and support through the years (Yes Dad, I'm finally out of school! ) and my sister Linda Cade for a warm welcome and accommodation during my Victoria trips. I am grateful to my in- laws, Ted and Eileen Menun, for their support as well. To Jacob Christopher Menun, born during the final year of my thesis work, thank you for being a healthy, happy baby, and for sleeping occasionally so that Mommy could work. It is difficult to adequately thank my husband, Chuck Menun, for his assistance and support. He helped with sampling and statistics, comforted me during the difficult times and pushed when I got lazy. I love you, Chuck. And finally, to Oscar, the beloved companion of my graduate school years, Rest in Peace. I miss you. ix CHAPTER ONE General Introduction The forests of northern Vancouver Island can be divided into two distinct types: the CH type, which is composed of western red cedar {Thuja plicata Don.) and western hemlock (Tsuga heterophylla (Raf.) Sarg.) stands, and the HA type, made up of stands of western hemlock and amabilis fir (Abies amabilis Dougl.) (Lewis 1982). After logging in the late 1970's, these sites were replanted with Sitka spruce {Picea sitchensis (Bong.) Carr.). Initially, survival and growth were good, but within five years after planting, the trees on the CH sites suffered a growth check and became chlorotic, while the HA plantations continued to grow well. A study by Germain (1985) indicated that the CH sites were deficient in nitrogen and phosphorus and responded well to fertilization, suggesting a nutrient cycling problem. Preliminary research by Western Forest Products Limited, the University of British Columbia and Natural Resources Canada's Pacific Forestry Centre led to the establishment of the Salal-Cedar-Hemlock Integrated Research Program (SCHIRP), to investigate the many aspects of this regeneration problem (Prescott and Weetman 1994). Phosphorus (P) is an important element in forest ecosystems. Pritchett and Fisher (1987) suggest that there are more reports of P deficiencies in forests than of deficiencies of any other nutrient, especially in conifer plantations. Produced initially by the solubilization of the soil parent material, P is taken up in 1 inorganic forms by pioneering organisms and is returned to the soil in organic forms. The actions of bacteria, fungi and fauna allow mineralization and immobilization to occur, and P is cycled through these organisms and higher plants in a variety of organic and inorganic forms. The acidic soils and cool temperatures of the forests of northern Vancouver Island limit the rate of mineralization, so that the majority of soil P is in organic forms. Little is known about the chemistry of organic P in soils, and less than half of the organic P in most soils can be accounted for in known compounds (Stevenson 1982). Currently, there are no direct methods to measure concentrations of organic P in soils. Two indirect methods are used: ignition and extraction (Olsen and Sommers 1982) . In both, organic P is calculated by subtracting the inorganic P from the total P. To determine the forms of organic P in a soil, a number of extractions must be done, each specific to a particular organic P class. The development of 31P NMR spectroscopy as a tool in soil phosphorus research allows a simpler, more direct examination of the forms and amount of organic P in a soil. As a new method, however, refinement of the extraction procedure is still required, as well as testing on a wide range of soil types, because the results with any method for P determination appear to depend on soil type and environment (Anderson 197 5) . Most soil P methods were developed for agricultural soils, and many may not be appropriate for forested soils. There has been little testing of 2 soil P methodology to determine which procedures are most suitable for the podzolic soils of Coastal British Columbia. Not all tree species replanted onto CH sites exhibit as severe a growth check as was initially observed in Sitka spruce. Studies by Weetman et al. (1989a, b) suggest that western red cedar trees growing on clearcut and burned CH sites grow better than replanted Sitka spruce and western hemlock, and respond less to P fertilization than do the other species. Western red cedar is an unusual conifer in this region in that it forms vesicular-rbuscular (VA) mycorrhizae, rather than ectomycorrhizae. This symbiotic association may give this species some advantage on these sites, and may allow western red cedar to use nutrient sources unavailable to the other tree species, such as organic P. Objectives The general objective of this thesis work was to investigate the concentration and forms of phosphorus in Orthic Ferro-Humic Podzols of the CH and HA forest types on northern Vancouver Island. Specifically, the first objective of this study was to test several methods for total, organic and available P, to find the methods most suited to this type of soil. The second objective was to examine extraction procedures for 31P NMR analysis, to find the most suitable methods for soils of this type, and to further expand the use of this technique as a means by which soil phosphorus forms may be characterized. The third objective was to characterize the soil P forms of mature, uncut CH and HA stands, and to examine some 3 aspects of the chemistry of these soils which could influence P forms and levels. The fourth objective was to characterize the forms of P and related soil chemistry of CH stands 10 years, 5 years and immediately after clearcutting and slash-burning, comparing these to uncut CH stands, to determine if changes in P after cutting and burning could be producing the observed growth check. The final objective was to examine the use of organic forms of phosphorus by mycorrhizal and non-mycorrhizal western red cedar. Hypotheses The hypotheses tested within this thesis include: 1. Some procedures to determine organic, total and available P may be better suited to these forest soils than are other extraction procedures. 2 . Some soil extraction procedures for 31P may be better suited to these forest soils than are other procedures. 3 . The soils of the mature CH stands contain different P forms from the mature HA stands, as well as differences in other aspects of the soil chemistry. 4. The soils of the CH stands 10 years, 5 years and immediately after burning contain different P forms from one another and from the mature CH stands, as well as differences in other aspects of the soil chemistry. 5. Western red cedar is able to use organic P forms, as well as or instead of inorganic forms, and mycorrhizal trees use different P forms from non-mycorrhizal trees. 4 Thesis Structure This thesis consists of nine chapters, including this introduction (Chapter 1). Chapter 2 is a review of the literature relevant to this thesis. Chapter 3 describes the study area and sampling design. Chapter 4 comprises an investigation of the suitability of several methods for total, organic and available P. Chapter 5 is the study of extraction procedures for 31P NMR spectroscopic analysis. In Chapter 6, the study of P forms on mature CH and HA stands is presented, and Chapter 7 details the study of soil P forms on cut and burned CH sites. Chapter 8 describes the greenhouse study in which the ability of western red cedar to use organic P forms is tested. Chapter 9 summarizes the conclusions of this research. 5 CHAPTER TWO LITERATURE REVIEW Forest Soil Phosphorus Cycles In a broad sense, the soil phosphorus (P) cycle involves the uptake of P by plants and its return to the soil in plant and animal remains (Stevenson 1982). It is a cohesive, dynamic system, which is influenced by both long-term chemical transformations and short-term changes from plant uptake (Tiessen et al. 1984). To aid in the management of soil P, many attempts have been made to model the soil P cycle. Conceptually, this main cycle can be subdivided into two further cycles (Dighton and Boddy 1988). The first is the external (geological) cycle, which involves P inputs to the ecosystem from the atmosphere (including both natural sources and pollutants), from the weathering of soil parent material, and from fertilizer inputs. Losses of P from the external cycle result from leaching, burning and harvesting. The other cycle is the internal (biological) cycle, which involves P exchanges among the soil, plants, fauna and decomposers. In natural systems, the P cycle is virtually closed, and most plant P is recycled by microbial breakdown of litter and organic matter. In agricultural ecosystems, the P cycle tends to be more open, due to disturbances such as crop removal, ploughing and fertilization (Tate 1984). These are annual disturbances in agriculture, but may only occur in forestry once in fifty years, 6 and result in higher losses of P from agricultural ecosystems than from forest ecosystems. The External (Geological Cycle) The major reserves of P on earth are: marine sediments (840,000 x 1012 kg), terrestrial soils (96-160 x 1012 kg), dissolved inorganic P043~ in the ocean (80 x 1012 kg), crushed rocks such as apatite (19 x 1012 kg) and the biota or biomass (2.7 x 1012 kg) (Stevenson 1986) . A very small amount is circulated as dust. Apatite [Ca5 (F, Cl, OH) (P04)3] is the most commonly occurring phosphate mineral in rocks (Wild 1988). It is present as a primary mineral in the sand fraction of soils, especially if the soils are fairly young and are not acidic. Some secondary minerals of phosphate are: wavellite [Al3 (P04) 2 (OH) 3. 5H20] , vivanite [Fe3(P04)2.8H20] , dufrenite [FeP04 .Fe (OH) 3] , strengite [Fe (P04) .H20] and variscite [Al(P04) . 2H20] (Stevenson 1986; Wild 1988). The majority of P compounds have coordination numbers of 3 and 4, although coordination numbers of 1, 3, 4, 5 and 6 are possible (Stevenson 1986) . In soil, P is found mainly in its oxidized state, as orthophosphate, and as complexes with Ca, Fe and Al, and silicate minerals. The forms of the phosphate ion are pH-dependent. In dilute solution, the dissociation of phosphoric acid is: H3P04 < > H2P04" < > HP042" < > P043" 7 At the common soil pH range of 5-8, H2P04" and HP042" dominate, and the amounts of the other forms are negligible. The usual soil total P (PT) concentration is in the 500 to 800 ug/g range, on a dry weight basis (Stevenson 1986) . In the soil profile, PT is highest in the upper A horizon and lowest in the lower A and upper B, due to uptake by plants in these lower regions (Stevenson 1986). Stevenson (1986) recognizes six major groups of soil P compounds. These are: 1. Soluble inorganic and organic compounds in the soil solution. 2. Weakly adsorbed (labile) inorganic phosphates. 3. Insoluble phosphates a) of Ca in calcareous and alkaline soils of arid and semi- arid regions b) of Fe and Al in acidic soils 4. Phosphates strongly adsorbed and/or occluded by hydrous oxides of Fe and Al 5. Insoluble organic forms a) of microbial biomass b) in undecomposed plant and animal residues c) as part of the soil organic matter Inorganic Phosphorus Essentially all of the inorganic soil phosphorus (Px) is in the form of orthophosphate or a derivative of phosphoric acid (H3P04) . Only a small fraction occurs in water-soluble forms at any time (Stevenson 1986) . The distribution of Px into its various forms in soil is controlled by the activities of various ions, including Fe, Al, Mn, Ca and P itself. Among the many factors controlling the activities of these ions are: soil pH and its effects on the solubility of Fe, Al, Mn and various phosphates; Ca availability; drainage; mineralization; redox potential; the presence of ligands that can replace phosphate when it is in complexes with Ca, Fe, Al, and Mn; and soil weathering and age (Chang and Jackson 1958; Pritchett and Fisher 1987; Fox et al. 1990b). In relatively unweathered soils, Ca- and Al-phosphates are more likely to be formed than Fe-phosphates. Fe-phosphates are the least soluble, and with time, Ca- and Al-phosphates will change to Fe-phosphates (Chang and Jackson 1958). Another influence on Px levels in the soil is phosphate adsorption. This can be defined as any process in which phosphate ions in solution react with atoms on the surface of soil particles (Barrow 1978). It is a two-step process, with an initially rapid step followed by a slower step because adsorption increases the negative charge on the surface of the soil, making it more difficult for each additional increment of phosphate to adsorb (Barrow 1978) . Adsorption may occur onto clays and sesquioxides in the soil. In mull soils in Quebec, low Px levels were thought to be due to phosphate adsorption by Al and Fe sesquioxides in Ah horizons, as the Ah was higher in Al and Fe, and lower in Px, than H horizons (Pare and Bernier 1989). Fernandez and Struchtemeyer (1985) observed similar adsorption in B horizons of podzols under spruce-fir stands. In podzols of Vancouver Island, the Px of Bhf and Bf horizons is usually sequestered in amorphous sesquioxides 9 (Sanborn 1987; Yuan and Lavkulich 1994). Phosphate fixation to clays decreases in the order: amorphous hydrous oxides > goethite = gibbsite > kaolinite > montmorillonite, and is pH-dependent (Meuller-Harvey et al. 1985; Stevenson 1986) . Humic or fulvic acids and low molecular weight aliphatic or aromatic acids may block sites on soil materials to reduce phosphate adsorption (Violante et al. 1991). Phosphate which has been fixed or adsorbed in such a way that it is unavailable for desorption or removal is called occluded phosphate (Wild 1988). The reverse of adsorption is desorption, which also consists of fast and slow reactions (Wild 1988). The fast reaction occurs by ligand exchange, while the slow reaction depends on the dissolving of Ca-phosphates among other processes (Wild 1988). Organic Phosphorus The amount of organic P (P0) in soil is related to the organic matter content of the soil profile, and is quite variable (Wild 1988). The factors which influence soil P0 levels include: P supply, parent material, climate, drainage, cultivation, pH and soil depth. Soils derived from granite tend to have a lower P0 content than do soils from basalt or basic igneous rocks (Stevenson 1982). P0 contents of fine-textured soils are higher than coarse- textured ones, and acidic soils have more P0 than alkaline ones (Stevenson 1982). P0 decreases with depth in the soil profile (Stevenson 1986). 10 The P0 compounds commonly found in soil, and their approximate recoveries, are: (Stevenson 1982) inositol phosphate 2-50% phospholipids 1-5% nucleic acids 0.2-2.5% phosphoproteins trace metabolic phosphates trace phosphonic acid derivatives trace Structurally, most of these compounds can be grouped into: monoesters, which include inositol phosphates, sugar phosphates and mononucleotides; and diesters, which include phospholipids and nucleic acids. The monoesters contain one organic moiety per orthophosphate; the diesters two (Fig. 2-1) It would appear that soil P0 compounds are not simply the result of accumulation of decay-resistant Po compounds from plants, but instead are in many cases derived directly, or after biological transformations and synthesis, from organic matter (Wild 1988) . Little is known about the chemistry of P0 in soils, and less than half of the P0 in most soils can be accounted for in known compounds (Stevenson 1982). The types and relative quantities of P0 compounds in soil are related to environmental conditions and soil management. For example, phosphonate P occurs only in soils in which mineralization was constrained by cool, moist conditions (Condron et al. 1990b). Inositol Phosphate These are esters of hexahydrocyclohexane, commonly called inositol (Fig. 2-2A). Of a variety of possible esters, the most common in soil is the hexaphosphate ester (Fig. 2-2A), which is 11 0 tl R - O - P - 0 " 1 0" Monoester OH / 0 = P - 0 - R 1 \ 0-R2 Diester Figure 2-1: General structures of orthophosphate monoesters and diesters. R represents an organic moiety. primarily derived from soil microbes. Although inositol compounds are produced in low amounts by living organisms, they are the most abundant of the P0 compounds in soil due to stabilization via the formation of inositol complexes with metal ions and other organic substances (Wild 1988; Stewart and Tiessen 1987)and from adsorption of inositol phosphates onto surface hydroxyls of soil colloids (Ognalaga et al. 1994) . In addition to hexaphosphate, lower phosphates such as mono-, di-, tri-, tetra- and pentaphosphates may also be found in soil (Stevenson 1994). Inositol hexa- and pentaphosphates can comprise up to 60% of soil P0 (Tate 1984) . Nine positional stereoisomers of inositol are possible, depending on the arrangements of H and OH groups. These include seven optically inactive forms and one pair of optically active isomers. Of these, the best known is myo-inositol, which is widely found in 12 nature and from which phytic acid (myo-inositol hexaphosphate) is derived (Stevenson 1986). Other naturally occurring isomers are d- chiro-, 1-chiro-, scyllo-, and neo-inositol. These are limited in distribution, and are found in microbes but not in higher plants, The microbial synthesis of these other isomers could occur by: the cyclization of carbohydrates; the direct phosphorylation of free inositol from soil organic matter; or from the formation of epimers from myo-inositol or myo-inositol hexaphosphate (L'Annunziata 1975) . Phospholipids These are esters of fatty acids and alcohols, and contain a phosphate group (Fig. 2-2B). They are soluble in fat solvents such as ether, benzene or chloroform. Because they contain a hydrophobic (glyceride) group and a hydrophilic (phosphate) group in one molecule, they are not particularly stable (Baker 1975). They are the second most abundant group of soil P0 compounds, comprising up to 5% of P0 (Stevenson 1982). Included in this group are the glycerophosphatides, such as phosphatidyl choline or lecithin, phosphatidyl serine, phosphatidyl ethanolamine and phosphatidyl inositol. Phosphatidyl choline is the most abundant (Emsley and Niazi 1983). Most soil phospholipids are of microbial (bacterial and fungal) origin. 13 A) OH OH H O / H H \ H H N - L J ^ O H OH H myo-lnositol H2*>3 0 H 2 P 0 3 - 0 / H - " 0 / H 2P0 3 / P 0 3 H 2 O _S/o-P03H2 H  PO3H2 Phytic acid B) 0 CHOCR.' I / ° H CH20P-O 0 I R'COOCH R'COOCH ? + I ? H,C-O-?-0CH2CHSH3 H J C - O - P - O C H J C H J S H J 0 " COO" 0 " CH2CH2H(CH3)3OH - o - L ' e c i t h i n P h o s p h a t i d y l s e r i n e P h o s p h a t i d y l e thano l a m i n e NH2 C) D) O" 1 -o—p=o I o I "O— P = 0 1 o I " O — P = O I o I CH2 \ ^ C H If f OH OH Adenosine triphosphate (AT?) ^ C H Adenosine diphosphate (ADPi Figure 2-2: Structures of soil organic P compounds. A) myoinositol and phytic acid, the hexaphosphate ester of myoinositol B) the phospholipids lethicin, phosphatidyl serine and phosphatidyl ethanolamine C) a nucleotide, 3 '-AMP D) adenosine triphosphate (ATP) and adenosine diphosphate (ADP) . (A and B from Stevenson (1994); C and D from Suttie (1972) . 14 Nucleic Acids Nucleic acids are found in all living cells and are released in soil by the microbial decomposition of plant and animal remains. Approximately 3% of P0 in soils is in nucleic acids or derivatives, and they are easily decomposed by microorganisms (Stevenson 1986). The two known types are RNA (ribonucleic acid) and DNA (deoxyribonucleic acid). Each of these consists of a chain of nucleotides, and each nucleotide contains a pentose sugar, a purine or pyrimidine base, and phosphoric acid residue which links adjoining pentose units (Fig. 2-2C). Anderson (1970) isolated two nucleoside phosphates from NaOH extracts of soil. One appeared to contain thymine, the other uracil. Other P0 Compounds Other P0 compounds which have been isolated from soils include: ATP (Fig. 2-2D) and several high molecular weight P0 compounds which have not yet been identified (Emsley and Niazi 1983) . They were all in low abundance. Teichoic acid has also been identified in native Grey Luvisols under aspen forest by Condron et al. (1990b). Teichoic acid is an orthophosphate diester form of P0, consisting of sugar units linked by phosphate groups such as polyribitol phosphates. Naturally occurring teichoic acid is not a single compound but a complex, found only in bacterial cell walls, and its composition changes with environmental conditions and nutritional status of the bacterial population. Condron et al. (1990b) found it in natural but not cultivated 15 soils, and suggested that its occurrence in the L horizon of the moder which they studied indicated that the leaf litter in this horizon had undergone substantial bacterial decomposition. Losses from the External Cycle Burning, either from prescribed burning or forest fires, reduces litter and produces small short-term available P increases, but large long-term total P losses (DeBano and Klopatek 1988; Saa et al. 1993; Romanya et al. 1994). The heat of the fire, if intense enough, may kill beneficial fungi such as those forming mycorrhizae. The remaining P in the ash following a fire is susceptible to wind erosion and to fixation to sesquioxides (DeBano and Klopatek 1990). Harvesting removes P and other nutrients from the ecosystem but the net effect depends on the balance of P gains and losses and the forms in which P is retained in the soil (Turner and Lambert 1986). The above-ground stand is estimated to contain about 2 5% of the "available" P pool in a forest. Harvesting of logs, with branches and slash left behind, removes about 4% of the available pool, or 2% if the bark were also left behind (Turner and Lambert 1986). The equilibrium between Px and P0 in the soil is also upset by harvesting. The soils most affected by harvesting losses are those lowest in PT at the start. Tsubota (1959) claimed to have observed the volatilization of P through the microbial reduction of phosphate to phosphine (PH3) . However, extensive testing by Burford and Bremner (1972) showed 16 that any phosphine formed in soils would be adsorbed by soil constituents and thus would not be lost to the atmosphere. Although leaching of phosphates is generally low, it can occur following fertilization, especially in agricultural soils (Stevenson 1986). In forest soils, Schoenau and Bettany (1987) observed leaching of labile P-rich compounds from surface horizons to lower horizons, where they became fixed and unavailable to plants. Leaching is also thought to be common in some coastal lands and organic soils of the southeastern USA (Pritchett and Fisher 1987), and may be most common in more weathered soils (St. Arnaud et al. 1988) . The extent of P leaching, and subsequent fixation in lower horizons or loss to the ground water, depends on the nature of the mineral surfaces throughout the soil profile (Frossard et al. 1989). Wind erosion may also remove soil P (Stevenson 1986). Internal (Biological) Cycle The internal or biological cycle of soil phosphorus involves exchanges among plants, fauna and decomposers. The most important processes in this cycle are mineralization and immobilization. These are companion processes: P0 is converted to PT by mineralization, while Pz is converted to P0 by immobilization. Because immobilized nutrients cannot be taken up by plant roots, this can be considered a temporary form of P fixation in the soil (Baath and Soderstrom 1979). The importance of P0 mineralization 17 in providing plant-available Pz is well-established (Stevenson 1986) . Generally, the C: P0 ratio determines which process will occur. Stevenson (1982) suggests that when the C: P0 ratio is 300 or more, net immobilization will occur, and when the ratio is 200 or less, there will be net mineralization. Both of these processes are mediated by soil microbes, and thus are governed by the factors affecting microorganisms, such as temperature, moisture, aeration, pH and energy supply (Tate 1984). There do not appear to be any specific bacterial or fungal groups involved in either of these processes; they are mediated by a wide range of organisms instead. Some purely biochemical mineralization may also occur via the hydrolysis of esters by extracellular enzymes present in the soil or released by plants and microbes in response to low Pz availability in the soil solution (Stewart and Tiessen 1987) . Bacteria The soil bacteria represent a relatively labile P pool. Besides transforming P, soil microbes are an important source and sink for P (Van Veen et al. 1987) . Microbial biomass P is determined by the change in P0 and Px following chloroform treatment of the soil, and may be complicated by spatial and temporal variations in soil microbial populations (Stewart and Tiessen 1987) . It is influenced by the many factors which affect mineralization and immobilization, such as temperature, moisture, aeration, and C and M supply. The amount of P which is sequestered 18 in biomass may also be influenced by soil texture, with more in clay than in sand (Van Veen et al. 1987). Of microbial intracellular P, over 60% is in nucleic acids, 20% is in acid-soluble phosphate esters, and 5% is in lipids, with variations (Stewart and Tiessen 1987). The most common phospholipid in bacteria is phosphatidyl ethanolamine (> 50%) , followed by phosphoglycerol, phosphatidic acid, and phosphoinositols (Stewart and Tiessen 1987). Teichoic acids may be present in bacterial walls, and surplus P may accumulate in bacterial cells as polyphosphates (Stewart and Tiessen 1987). Biomass P0 can be taken up directly by predators or by saprophytes and incorporated into new consumer biomass, or may be released back into the soil solution by grazers such as protozoa and nematodes (Dighton and Boddy 1989). Mineralization of P is highest in the rhizosphere, where substances from root exudates, sloughed-off root cells, tissues and mucigels sustain a larger and more active microbial population than in the rest of the soil (Tate 1984). Phosphorus may be more plant- available in the rhizosphere (Gillespie and Pope 1990a, 1990b), due to the production of phosphatases, either by the microbial population or by the plant itself (Stewart and Tiessen 1987). Organic acids, which may also be produced by the plant and/or the microbial population, will also increase the availability of phosphate in the rhizosphere (Comerford and Skinner 1989). Phosphate is released by these acids through ligand-exchange 19 reactions, the dissolving of metal-oxide surfaces which adsorb P, or the complexing of metals in solution to prevent the precipitation of metal phosphates (Fox et al. 1990a). Acids involved include oxalic, formic, citric, malic and acetic, all of which are commonly found in soils (Fox et al. 1990b). Fungi Fungi play a crucial role in the cycling of P, and in mor humus of temperate coniferous forests may have the greatest biomass of all organisms (Baath and Soderstrom 1979) . Fungi, as saprophytes, are major decomposers of soil organic matter, and are responsible for the return of the P which is immobilized in dead plant, animal and microbial tissue to the soil P pool, solubilizing organic P forms through the production of organic acids and phosphatases (Dighton and Boddy 1988). Fungi may also obtain P from parasitic associations on living organisms (Dighton and Boddy 1988). Phosphorus is concentrated in the mycelium, and when the fungi are alive, is released by excretion, secretion of extracellular enzymes, leaching and grazing by animals (Swift 1977), and on their death by other fungi and organisms (Stark 1972) . Systems of hyphae may move P considerable distances, drawing on areas of high P concentration to assist in areas of low P concentration (Dighton and Boddy 19 89). Spore dispersal will also relocate P. 20 Plants Most of the movement of P to plant roots occurs by diffusion, and depends on the concentration gradient between the soil and the root surface, and on the diffusion coefficient (Bhadoria et al. 1991) . Mass flow is a very minor mechanism for P transport to roots. Phosphorus is absorbed by plant roots as the negatively charged primary and secondary orthophosphate ions (H2P04~ and HP04=) , which are present in the soil solution. The concentration of P at the root surface regulates the quantity of P absorbed by the root (Gillespie and Pope 1990b), and decreases as the buffering capacity of the soil for phosphate increases (Barrow 1978). Most of the P in soil is in forms which are not readily available to plants, such as organic P forms (Stevenson 1986). In response to a need for P, plants may produce extracellular phosphatases (Tate 1984). They also form mycorrhizae, which are symbiotic associations of higher plants and fungi (Harley and Smith 1983) . Ectomycorrhizae, in which the fungi do not enter the cortical cells, are the most common associations for forest trees (Isaac 1992) . Many of the host plants belong to the families Pinaceae, Fagaceae, Betulaceae and Myrtaceae; the fungi are ascomycetes and basidiomycetes. In other types of mycorrhizae, the fungi penetrate the cortex of the host plant. These types of mycorrhiza can be subdivided into orchid, ericalean and vesicular- arbuscular (VA) mycorrhizae. Orchid mycorrhizae are symbioses involving members of the Orchidaceae and basidiomycetes or 21 ascomycetes. Ericalean mycorrhizae are formed by plants of the order Ericales (ericoid, arbutoid and monotropoid), and are often found on acid soils of northern temperate forests (Isaac 1992). The fungi of ericalean mycorrhizae are basidiomycetes and ascomycetes. The VA mycorrhizae are the most widely occurring of all mycorrhizae, and are particularly common in crop plants, herbs and tropical trees. They are found in temperate coniferous forests less frequently than ectomycorrhizae and ericalean mycorrhizae, but they are formed with some conifers such as western red cedar (Thuja plicata Don). Many plant species form VA mycorrhizae; the fungi are all zygomycetes (Harley and Smith 1983). Mycorrhizae assist the plant in the uptake of nutrients, especially P (Harley and Smith 1983). They improve the uptake efficiency of plants by increasing the absorptive surface area in the soil, thus allowing the plant access to pools of P which were unavailable or less available to the non-mycorrhizal plant. Ectomycorrhizae may produce acid phosphatases (MacFall efc al. 1991; Lapeyrie 1991), protease, cellulase, phenol oxidase (Entry et al. 1991) and phytase (Dighton 1983). Cromack et al. (1979) and Alexander and Hardy (1981) reported the exudation of oxalic acid by ectomycorrhizal fungi in forest litter. These enzymes and acids allow the plant to use P forms not available to non-mycorrhizal plants, particularly organic P. The P content of the hyphae of ectomycorrhizal fungi often exceeds that of plant roots (Fogel and Hunt 1983), and mycorrhizae may accumulate polyphosphate for 22 storage or hyphal P transport (Martin et al. 1983; MacFall et al. 1992) . Phosphatase production has also been demonstrated in ericoid mycorrhizae (Mitchell and Read 1981; Straker and Mitchell 1986; Shaw and Read 1989; Dighton and Coleman 1992) and orchid mycorrhizae (Antibus and Lesica 1990). It is generally believed that VA mycorrhizae obtain their P from the same soil pool as non-mycorrhizal plants (Bolan 1991; Jennings 1995). However, recent work suggests that VA mycorrhizae may produce phosphatase to mineralize organic P forms (Jayachandran et al. 1992; McArthur and Knowles 1993; Thiagarajan and Ahmad 1994; Tarafdar and Marschner 1994). Within higher plants, the three main types of P compounds are: inositol hexaphosphates, particularly phytin; phospholipids; and nucleic acids (Ting 1982). Other plant P compounds include ATP, NAD, NADP and mono-, di- and triphosphates of inositol (Stevenson 1994; Ting 1982). The central role of P in plants is in energy transfer. In P-deficient plants, growth is stunted and maturity is delayed. Plants return P to the soil cycle in many forms. Foliage dominates the above-ground return of P in litter to the forest floor, while the contribution of woody litter to P inputs is minor (Fogel and Hunt 1983). Many deciduous species conserve P by translocating it into perennial parts before leaf drop (Morrison 1991). Decomposition of below-ground biomass may return as much as 80% of the total tree P in Douglas-fir to the soil (Fogel and Hunt 23 1983) . Pollen rain is an important P source in forest soils because it provides an input of nutrients during the summer when there is little leaching of nutrients out of the dry surface L layer, but when the F layer is still moist enough for considerable decomposition and biological activity at summer temperatures (Stark 1972) . Fauna Soil fauna affect P cycling by fragmenting litter, grazing on bacteria and fungi, and improving soil structure, which in turn improves conditions for microbes (Reichle 1977; Dighton and Boddy 1989) . Soil fauna may contain higher concentrations of P than plant litter, most of which was obtained from grazing on bacteria and fungi (Pokarzherskii and Gordienko 1985). Grazing may therefore result in the release of nutrients, in addition to altering the relative contributions of different fungal and bacterial species to the decomposition process (Dighton and Boddy 1989) . After death, soil fauna are themselves attacked by decomposers, returning P to the bacterial and fungal pool. Earthworms enrich surface soils with P, and increase the amount of readily exchangeable inorganic P as well as the rate of mineralization of organic P by the improvement of soil structure (MacKay et al. 1982). 24 Soil Phosphorus Methodology Many methods are currently used to determine total (PT), organic (P0) and available (PA) phosphorus concentrations in soils. There does not appear to be an ideal method to determine any of these soil P fractions, and the results appear to depend on soil type and environment (Anderson 1975) . Most soil P tests have been developed for agricultural soils. Some methods may be better suited to forest soils than others. Total P Total P (PT) analysis requires the conversion of insoluble material to soluble forms, followed by colorimetric analysis (Olsen and Sommers 1982) . The most commonly used methods are fusion with Na2C03 and digestion with HC104 (Syers et al. 1967; Syers et al. 1968; Sommers and Nelson 1972; Dick and Tabatabai 1977; Olsen and Sommers 1982) . Other PT methods in use include: ignition of the soil followed by extraction with H2S04 (Syers et al. 1967; Olila and Reddy 1995); digestion with H2S04 and H202 (Thomas et al. 1967; Roberts et al. 1985; Schoenau and Bettany 1987; Xiao et al. 1991; Compton 1994; Fyles and Cote 1994; Hanafi and Syers 1994); digestion with H2S04, H202 and HF (Bowman 1988); digestion with H2S04, H202, Li2S04 and Se (Parkinson and Allen 197 5; Edmonds 1980; Rowland and Grimshaw 1985; Tiedemann and Klemmedson 1986; Prescott et al. 1993; Silver et al. 1994); and digestion with HC1 and HN03 (aqua regia) (Crosland et al. 1995). The Na2C03 fusion method gives the highest recovery of PT from soil, but it is tedious and 25 time consuming (Dick and Tabatabai 1977; Bowman 1988). The other PT procedures are simpler for routine laboratory analysis. Available P Labile phosphorus is related to the quantity of P available to plants growing in a soil - so-called "available" P (PA) (Thomas and Peaslee 1973). Soil testing procedures for PA, developed for agricultural soils, can provide an accurate "relative index" of the quantity of P that plants may utilize from the soil, and an index of the quantity of fertilizer P required for some range of soil- crop-climate combination (Thomas and Peaslee 1973). These procedures should be tested and calibrated for each soil, crop and environment. The main extractants used to determine PA are grouped by the soil pH range at which they are most effective. For alkaline soils, the Olsen et al. (1954) extraction using NaHC03 is most suitable. The most common PA methods for acid soils are the Bray Pi (Bray and Kurtz 1945) and the Mehlich 3 (Mehlich 1984) methods, which use fluoride as a complexing ion in dilute acid. Organic P A direct method to determine the total organic P (P0) concentration of soil has not yet been devised (Olsen and Sommers 1982) . Soil P0 is estimated by either of two indirect methods: ignition or extraction. Ignition methods utilize ashing at either low temperature (Legg and Black 1955) or high temperature (Saunders and Williams 1955) to oxidize the soil organic matter prior to acid extraction. An unignited sample is concurrently extracted with 26 acid, and the soil P0 concentration is the difference between the P contents of the ignited and unignited extracts, after colorimetric analysis (Olsen and Sommers 1982). Extraction methods involve treating soils with acids, bases, or both, followed by the determination of P in the extract before and after the oxidation of organic matter. The P0 content of the soil is the difference in the P content of the extract before and after oxidation, following colorimetric analysis (Olsen and Sommers 1982). The extraction method used most often is that of Mehta et al. (1954), which is a sequential treatment of soil with HC1 and NaOH. Extraction with NaOH and EDTA has also been proposed (Bowman and Moir 1993). The identification of specific P0 compounds in soil has traditionally been conducted using partition chromatography following extraction. For example, the inositol phosphates are extracted from soils with acid and alkali and are precipitated as insoluble Fe-salts from acid media or Ba-salts from alkaline media, prior to anion-exchange chromatography (Stevenson 1994). Phospholipids may be recovered from soil using sequential extraction with ethanol-benzene and methanol-chloroform (Baker 1975). These methods are time-consuming, and a different extraction procedure is required for each class of P0 compounds. The introduction of 31P nuclear magnetic resonance (NMR) spectroscopy as a soil science technique allows the qualitative and quantitative analysis of P0. The theory of NMR spectroscopy is based on the fact that when a sample containing suitable nuclei is 27 placed in a magnetic field, the spin energy levels of the nucleus are split (Wilson 1991) . The size of the splitting depends on the strength of the magnetic field "0 (corresponding to the energy gap) and characteristics of the nucleus. Transitions between the new energy levels can be brought about by electromagnetic radiation U0 in the radiofrequency range. The strength of the magnetic field and frequency of irradiation are related by where is the gyromagnetic ratio and is characteristic of the nucleus (Wilson 1991). Thus, different nuclei occur at characteristic radiofrequencies (the Larmour frequency) for a given field strength. In addition, the electrons around a nucleus shield the nucleus from the magnetic field so that nuclei of the same element with different electronic distributions resonate at different frequencies from the Larmour frequencies. This results in a range of 'chemical shifts', which can be used to detect different functional groups of an element (Wilson 1991). Not all nuclei undergo magnetic resonance. The number of energy states created in the presence of a magnetic field depends on the spin quantum number of the nucleus investigated. The most commonly studied nuclei in soils are 1H, 13C, 27A1, 29Si, 15N, and 31P. In order to detect the signal from nuclei such as 31P above the background noise, a large number of scans must be recorded and averaged so that the signal becomes proportionally larger than the background noise. As sweeping the spectrum takes many seconds, a more efficient way to collect spectral data is to pulse the sample 28 X with a radiofrequency pulse, causing all the nuclei to resonate at the same time (Wilson 1991). Fourier transformation then converts the data into the form of a continuous wave spectrum. Between pulses, a sufficient length of time must be left for nuclei to relax back to their original equilibrium distribution (Wilson 1991). The length of time for which the pulse is left on during an NMR experiment is the pulse width, also expressed as a pulse angle. For maximum sensitivity, short pulse angles of 45° or less are normally used in solution studies of soil organic matter (Wilson 1991). Many investigated nuclei are in close proximity to protons, the spins of which can be aligned with or against those of the nucleus under study. The process by which the interaction with protons is removed is called decoupling, and is essential to obtain a simple spectrum of the studied substance (Wilson 1991). Decoupling involves irradiating the protons in the sample, so that the protons undergo rapid transition, and the interactions with other nuclei are averaged out. The nuclei under study will also be affected by this irradiation, so that the signal intensity from the nuclei close to the protons will be greater than from nuclei remote from the protons. This is termed the nuclear Overhauser effect, and it can affect quantification (Wilson 1991). To reduce the enhancement effect, the decoupler may be turned off during the pulse delay and used only during the scan, in a process called inverse gated decoupling. 29 Both solutions and solids may be used for 31P NMR spectroscopy. However, the natural P levels in soils are usually low. As 31P NMR is relatively insensitive, requiring more than 100 ug P/ml for quantitative analysis (Adams and Byrne 1989), solution NMR is used for soil P, and extraction and concentration are required to produce clear spectra. Since the first use of 31P NMR for soil P determination by Newman and Tate (1980), a number of extractants have been tried. The extractants tested include: 0.5 M NaOH (Newman and Tate 1980; Tate and Newman 1982; Ogner 1983; Hawkes et al. 1984; Zech et al. 1985; Preston et al. 1986; Zech et al. 1987; Hinedi et al. 1989; Gil-Sotres et al. 1990; Forster and Zech 1993); 0.5 M NaOH with a citrate-dithionite-bicarbonate pretreatment (Ingall et al. 1990); combined HCl, HF and TiCl4 (Hinedi et al. 1989); sequential extraction with NaOH and acetylacetone (Condron et al. 1985); tetra-n-butyl ammonium hydroxide (Bu4NOH) (Emsley and Niazi 1983); trichloroacetic acid and KOH (Hinedi et al. 1989); and Chelex, a cation exchange resin (Adams and Byrne 1989; Adams 1990) . The soil P0 classes which 31P NMR spectroscopy shows are: phosphonates, orthophosphate, phosphate monoesters, phosphate diesters, pyrophosphate, ATP and polyphosphates. There is close agreement on quantities of P estimated by both conventional laboratory methods and 31P NMR spectroscopy (Tate and Newman 1982). P Fractionation Inorganic P bound to Ca, Fe or Al minerals in soil is usually estimated with various acid digests modified from the Chang and 30 Jackson (1957) method (Olsen and Sommers 1982; Cross and Schlesinger 1995) . Phosphorus extracted by NaOH is thought to be the non-occluded phosphate bound to the surfaces of Al or Fe hydrous oxides; P removed by citrate-bicarbonate-dithionite extraction is the P occluded within the matrices of Fe and Al oxides and hydrous oxides; and P removed with HCl is the extracted calcium phosphate of the nonoccluded apatite fraction (Williams et al. 1980; Olsen and Sommers 1982). Another fractionation scheme, based on the function of P compound classes in soil rather than on chemical compounds, was developed by Hedley et a.1. (1982) . This method is traditionally used to separate plant-available or labile forms of P from various refractory P pools, and was recently reviewed by Cross and Schlesinger (1995). 31 CHAPTER THREE Study Area The study site is located between Port Hardy and Port McNeill on northern Vancouver Island (Fig. 3-1), at 50°60' latitude and 127°3 5' longitude. This is a wet area, with mild winters and cool summers. The annual rainfall is approximately 1700 mm, 65% of which falls between October and February. There is less rainfall in summer than winter, and in most years there is no soil moisture deficit (Lewis 1982). This area has a long frost-free period (175 days) and receives a small amount of snow from December to February. The Figure 3-1: Map of Vancouver Island, with inset of British Columbia. 32 mean annual temperature is 7.0°C, with a mean daily temperature range from 3 .0°C in January/February to 13.7°C in July/August. Theaverage hours of sunshine range from a high of 6.4 h/day in July to a low of 1.5 h/day in December, reflecting the frequency of fog in summer and frontal clouds in winter. All weather data are from the Port Hardy Airport weather station (Messier 1991, Keenan 1993). The northern part of the study area sits in the Suquash basin, which is gently undulating, and seldom exceeds 3 00 m in elevation. Geologically, the surface material consists of deep unconsolidated morainal and fluvial sediments overlying three types of bedrock: gently dipping sedimentary rocks of the cretaceous Nanaimo formation, relatively soft volcanics of the Bonanza group, and a small area of harder Karmutsen Formation basalt (Lewis 1982) . The southern portion of the northern area is located in the Nawhitti Lowland. It is underlain by the altered basaltic rock of the Karmutsen formation. Here, surficial materials tend to be discontinuous and are interrupted frequently by rock outcrops, in contrast with the relatively continuous cover of surficial materials of the Suquash Basin (Lewis 1982). This locality is considered to be one of the first places along the British Columbia coast to become ice-free after the most recent (Fraser) glaciation, with vegetation establishing about 14,000 years ago (Hebda 1983). The study site is in the very wet maritime subzone of the Coastal Western Hemlock (CHW^) biogeoclimatic zone (Pojar et al. 33 1991) which occupies the lower and middle latitudes of Vancouver Island and the British Columbia coastal mainland. Lewis (1982) considered the ecosystem association to be the Thuja plicata - Tsuga heterophylla - Abies amabilis - Gaultheria shallon - Rhytidiadelphus loreus (the salal-moss SI) association. Due to the moist climate, wildfires are uncommon, and windstorms are the predominant source of disturbance. The forests in this area form two distinct types. The first, called CH, consists of old growth (more than 500 years old) stands of western red cedar (Thuja plicata Don.) and western hemlock (Tsugra heterophylla (Raf.) Sarg.) (Fig. 3-2). This forest type has an open canopy which allows light to penetrate, and thus there is a dense understory of salal Figure 3-2: A typical stand of the CH forest type. 34 (Gaultheria shallon Pursh.), and blueberry {Vaccinium parvifolium Howell and V. alaskaense Smith). The forest floor is occupied by the mosses Bhytidiadelphus loreus (Hedw.) Warnst., Kindbergia oregana (Sull.) Ochyra and Hylocomium splendens (Hedw.) B.S.C., with occasional ferns such as Blechnum spicant L. (Germain 1985; Prescott and Weetman 1994). The forest floors of the CH forest type are lignohumimors and humimors (Klinka et al. 1981) . Beneath these are moderately to somewhat imperfectly drained Duric or Orthic Ferro-Humic Podzdls. Following cutting and burning, the CH sites are quickly invaded by salal, which regenerates from rhizomes. Natural tree regeneration is slow and sparse, and consists mainly of western red cedar and western hemlock seedlings. Forests of the CH type are thought to not have been catastrophically disturbed for at least one thousand years, and are believed to represent the climatic climax association for this region (Lewis 1982). The second forest type in this area, the HA, is characterized by closed stands of second growth western hemlock and amabilis fir (Abies amabilis Dougl.) (Fig. 3-3). These stands appear to be even-aged and originated following a wide-spread windstorm in 1906 (Lewis 1982). The understory is sparse, with small patches of blueberry (Vaccinium alaskaense and V. parvifolium), deer fern (Blechnum spicant), and the mosses Kindbergia oregana, 35 Figure 3-3: A typical stand of the HA forest type. Rhytidiadelphus loreus and Hylocomium splendens (Germain 1985; Prescott and Weetman 1994). The forest floors of the HA forest type are humimors (Klinka et al. 1981), overlying well-drained Ferro-Humic Podzolic soils. After cutting and burning, the HA sites are quickly invaded by a dense cover of fireweed (Epilobium angustifolium L.) and salal. Natural tree regeneration is rapid and dense, and consists mainly of western hemlock. The transition between CH and HA is quite abrupt and there is no evidence of occupation of HA sites by cedar prior to the 1906 windstorm (Keenan 1993). In classifying the ecosystems of this region, Lewis (1982) could not distinguish between the CH and HA forest types on the basis of topography or mineral soil characteristics, and included them in the same ecosystem association. He further hypothesized that they were different stages of a successional sequence. 36 Sampling Design Soil samples were collected July 14, 15, 31 and August 1, 2, 3, 1992. A nested design was used in this study, with three locations for each age or forest type, and three soil pits per location. The pit locations were randomly selected using a soil probe to determine suitability, and then pits were dug of no more than 0.1 m3. Only Orthic Ferro-Humic Podzols lacking buried horizons or woody forest floor material were used. The soils from each pit were sampled by horizon, based on visible characteristics such as colour and texture (Agriculture Canada Expert Committee on Soil Survey 1987). The horizons used were LF, H, Bhf and Bf, which is a typical horizon sequence for podzols in this region (Lewis 1976). The L and F horizons of the forest floor were not separated, because the L horizons were very thin. On newly burned sites, this horizon was an ash layer, which was sampled and labelled as LF. Ae horizons are rare in the podzols of northern Vancouver Island (Lewis 1976). In the few pits where they were observed, they were thin and discontinuous, and were avoided during sampling. The mean thickness and ranges of thickness for the LF, H and Bhf horizons sampled for each forest type and age postburn are shown in Table 3-1. The total thickness of the Bf horizon was not measured. 37 Table 3-1: The mean horizon thickness and ranges of thickness for the LF, H and Bhf horizons sampled for old growth CH and HA sites, and CH sites 0, 5 and 10 years after cutting and burning. Horizon Thick. (cm) LF H Bhf mean range mean range mean range HA-OG 5.4 2-14 7.2 3-13 7.1 2-14 CH-OG 5.6 2-10 9.0 4-15 10.8 2-25 CH-0 3.7 1.5-8 4.6 1.5-10 6.2 2-10 CH-5 5.1 3-10 5.2 2-10 4.0 2-8 CH-10 2.8 1-6 4.3 2-9 4.8 1-8 Study Sites The sites from which samples were collected were all located within Block 4 of Tree Farm Licence 25, (Fig. 3-4) which is operated by Western Forest Products Limited (WFP). The study sites were selected with the assistance of Paul Bavis of WFP. The locations of uncut CH and HA stands were chosen in part for their proximity to other research areas of the SCHIRP project, to facilitate the comparison of data. The selection of five-year and ten-year postburn sites was restricted to unfertilized CH areas, which limited the choices. The 0-year postburn sites were very limited, as only a few CH sites had been burned in the spring prior to sample collection. 38 Figure 3-4: Location of sampling sites within Block 4, Tree Farm Licence 25. (Approx. scale is 1:190 000.) A description of the sample locations follows. 10-1: WFP code 412/81, located at Rupert 470. This site was logged in 1981, burned in the fall of 1982, scarified in the winter of 1985, and planted in 1986. The elevation is 150 m, with a flat aspect, a slope of 0-65% and a rolling landform, and covers 67.9 ha. There are some rocky ridges and glacial surficial deposits. It was thought to have a heavy brush potential and poor natural regeneration potential. 10-2: WFP code 424/80, located at Rupert 200. This was logged in 1980, burned in the fall of 1980, and planted in 1981. The elevation is 150 m, with a SE aspect, a slope of 0-40% and a hummocky landform, morainal veneers, and covers 5.8 ha. 39 10-3 : WFP code 451/80, located at Rupert 200. This was logged in 1980, burned in the fall of 1980, and planted in 1981. The elevation is 60 m, with a slope of 0-25% and a flat landform, on morainal blankets, and covers 23.1 ha. 5-1: WFP code 240/86, located at West 83A. This was logged in 1986, burned in the spring of 1987, and planted in 1988. There is 5% bedrock and glacial surficial deposits, with a slope of 0- 60%, and covers 68.4 ha. It was thought to have a medium brush potential and a poor natural regeneration potential. 5-2: WFP code 263/86, located on Misty 930. This was logged in 1986, burned in the spring of 1987, and planted in 1988. It has 10% bedrock and glacial surficial deposits, and covers 94.2 ha. It was thought to have heavy brush potential and poor natural regeneration potential. 5-3 : WFP code 264/86, located on Misty 900. This was logged in 1987, burned in the spring of 1987, and planted in 1988. It has glacial till, surficial deposits and covers 56.6 ha. It was thought to have a light brush potential and poor natural regeneration potential. The 0-year sites were all logged in 1991 and burned in the spring of 1992. 0-1 was located at Rupert 470, 0-2 at Rupert 450 and 0-3 at Rupert 440. Typical examples of 10-year, 5-year and 0-year sites can be seen in Figure 3-5. 40 Figure 3-5: Typical 10 year (A), 5 year (B) and 0 year (C) post- burn sites. 41 CHAPTER FOUR A COMPARISON OF METHODS FOR TOTAL, ORGANIC AND AVAILABLE PHOSPHORUS Introduction Before comparing phosphorus in different types of forests or assessing the effects of burning on the forest P cycle, the concentrations of total, organic and available P must be reliably measured. Most soil phosphorus determinations have two phases: the preparation of a solution containing the desired soil P fraction, such as total (PT) , organic (P0) or available (PA) ; and the quantitative determination of P in the solution (Olsen and Sommers 1982) . The second step usually involves colorimetric analysis, most often the molybdate blue method of Murphy and Riley (1962) (Olsen and Sommers 1982) . For the first step, many different methods are currently in use to determine PT, P0 and PA. There does not appear to be an ideal method to determine any of the soil P fractions, and the results with any method appear to depend on soil type and environment (Anderson 1975). Total P (PT) analysis requires the conversion of insoluble materials to soluble form. The most commonly used methods are fusion with Na2C03 and digestion with HC104 (Syers et al. 1967; Syers et al. 1968; Sommers and Nelson 1972; Dick and Tabatabai 1977; Olsen and Sommers 1982). The Na2C03 fusion method gives quantitative recovery of all P forms in soils, but it is tedious and time consuming, and is not suitable for large numbers of samples (Dick and Tabatabai 1977). The HC104 digestion method is 42 preferred because of its simplicity and adaptability for routine analysis (Dick and Tabatabai 1977). However, this method requires careful handling because of the unsafe feature of boiling HC104, and its use has been discontinued in many areas, such as the University of British Columbia campus. Digestion with HC104 may also give low P recoveries in strongly weathered soils (Syers et al. 1967; Syers et al. 1968). Other methods, which retain the ease of the HC104 digestion method but which lack its unsafe aspects, have been introduced. These include: extraction with H2S04 following ignition of the soil (Syers et al. 1967; Olila and Reddy 1995); the Thomas et al. (1967) method of hot H2S04 followed by H202 (Roberts et al. 1985; Schoenau and Bettany 1987; Xiao et al. 1991; Compton 1994; Fyles and Cote 1994; Hanafi and Syers 1994); digestion with combined H2S04, H202 and HF (Bowman 1988) ; digestion with aqua regia (HCl and HN03) (Crosland et al. 1995); and the Parkinson and Allen (1975) digest, which uses H2S04 and H202 as oxidants, with the addition of Li2S04 to elevate the digestion temperature and Se as a catalyst (Edmonds 1980; Rowland and Grimshaw 1985; Tiedemann and Klemmedson 1986; Prescott et al. 1993; Silver et al. 1994) . A direct method to determine the organic P (P0) content of soil has not yet been devised (Olsen and Sommers 1982). Soil P0 is estimated by either of two indirect methods: ignition or extraction (Saunders and Williams 1955; Legg and Black 1955; Hance 43 and Anderson 1962; Dormaar and Webster 1963; Enwezor and Moore 1966; Williams and Walker 1967; Williams et al. 1970; Steward and Oades 1972; Ipinmidun 1973; Anderson 1975; Olsen and Sommers 1982; Condron et al. 199 0b; Bowman and Moir 1993). For many soils, these methods give comparable values (Olsen and Sommers 1982) . Ignition methods utilize either low temperature (Legg and Black 1955) or high temperature (Saunders and Williams 1955) ashing to oxidize the soil organic matter prior to acid extraction. An unignited sample is concurrently extracted with acid, and the soil P0 is the difference between the P contents of the ignited and unignited extracts (Olsen and Sommers 1982). The ignition method can underestimate the P0 content of soils due to acid hydrolysis during treatment of unignited samples (Harrap 1963; Olsen and Sommers 1982) or by incomplete extraction of P released during ignition (Dormaar and Webster 1963; Williams et al. 1970; Anderson 1975; Olsen and Sommers 1982) . Phosphorus may also be lost by volatilization during ignition (Dormaar and Webster 1964; Williams et al. 1970; Anderson 1975), although these losses are minimal at ignition temperatures under 800°C (Saunders and Williams 1955; Anderson 1975) . After ignition, the solubility of Px compounds may be increased, resulting in erroneously high P0 values (Legg and Black 1955; Dormaar and Webster 1963; Harrap 1963; Williams and Walker 1967; Williams et al. 1970; Anderson 1975; Olsen and Sommers 1982; Soltanpour et al. 1987; Condron et al. 1990b). Extraction methods involve treating soils with acids, bases or 44 both, followed by the determination of P in the extract before and after the oxidation of organic matter. The P0 content of the soil is the difference in the P content of the extract before and after oxidation. The extraction method used most often is that of Mehta et a.1. (1954), which is a sequential treatment of soil with concentrated HCl, NaOH at room temperature and NaOH at 90°C (Olsen and Sommers 1982) . This is a laborious procedure, and may result in underestimating the P0 content of a soil by the incomplete extraction of P0 or by hydrolysis of P0 during extraction (Williams et al. 1970; Anderson 1975; Olsen and Sommers 1982; Condron et al. 1990b; Bowman and Moir 1993). A new extraction procedure to determine P0 was recently introduced by Bowman and Moir (1993). This is a one-step extraction using a combination of Na2EDTA and NaOH. The NaOH solubilizes the P0 associated with soil organic matter, while the EDTA chelates metal cations to increase the efficiency of the organic matter extraction. The main extractants in use to determine available P (PA) are grouped by the soil pH range at which they are most effective. For alkaline soils, the Olsen et al. (1954) extraction, which contains NaHC03, is most suitable, while for acid soils the Bray PI (Bray and Kurtz 1945) and the Mehlich 3 (Mehlich 1984) methods are most commonly used. These are both dilute acid methods using fluoride as a complexing ion to release P (Curran 1984) . In addition to NH4F, Bray PI contains HCl, while the Mehlich 3 extractant contains CH3COOH, NH4N03, HNO3 and EDTA. The Mehlich 3 extractant was 45 developed as a multielement extractant, and can be used to quantitatively determine K, Ca, Mg, Cu, Zn and Mn, in addition to P, thus making it more useful than Bray in many laboratory situations. Comparisons of the Bray PI and Mehlich 3 methods show that the P results are highly correlated for agricultural soils (Mehlich 1984; Wolf and Baker 1985; Michaelson et al. 1987; Tran et al. 1990; Cade 1989; Simard et al. 1994; Wendt 1995). All of the soil tests described above were developed for agricultural soils, and some methods may be better suited to forest soils than others. The Bowman and Moir (1993) extraction for P0 in particular is very new and has not yet been adequately tested for use with forest soils. The objective of the research in this chapter was to compare the Saunders and Williams (1955) and Parkinson and Allen (1975) methods to determine PT, the Saunders and Williams (1955) and Bowman and Moir (1993) methods to determine P0 and the Bray PI and Mehlich methods to determine PA, to determine the suitability of these procedures for use with the Orthic Ferro- Humic Podzolic soils of northern Vancouver Island. Materials and Methods For PT, P0 and PA, the P content of all solutions was read colorimetrically using the molybdate blue method (Murphy and Riley 1962) on a Lachat Flow Injection Analyzer (FIA). The soils used are those described and analyzed in Chapters 5, 6 and 7. 46 Total P The two PT methods used were the Parkinson and Allen (1975) digest and the first step of the Saunders and Williams (1955) ignition method for P0 (Olsen and Sommers 1982). Parkinson and Allen Soil samples were oven-dried overnight at 60°C prior to analysis. A 1 g sample was weighed into a 100 ml digestion tube, 5 ml of concentrated H2S04 was added and the tube contents were mixed with a vortex mixer. A 1 ml aliquot of the digestion mix, containing 7.0 g Li2S04, 0.21 g Se powder and 175 ml of 30% H202, was added. After the foaming reaction had ceased, three more 1 ml aliquots were added. The sample was then digested at 360°C on a block digester. After 1.5 h, the sample was cooled briefly and 0.5 ml of 30% H202 was added. After a further 30 min. of digestion, the sample was again briefly cooled and a second 0.5 ml aliquot of H202 was added. After a total digestion of 2.5 h, the samples were cooled and made to volume, and were analyzed colorimetrically. Saunders and Williams Soil samples were oven-dried overnight at 60°C prior to analysis. A 1 g sample was weighed into a porcelain crucible, and the crucible was placed in a cool muffle furnace. The temperature was raised to 550°C over a 2 hour period, and maintained at 550°C for 1 hour. After cooling, the ignited sample was transferred to a 10 0 ml centrifuge tube, and 5 0 ml of 0.5 M H2S04 was added. The sample was shaken overnight, centrifuged, and then the extract was read colorimetrically. 47 Organic P The two P0 methods used were the Saunders and Williams (1955) ignition method (Olsen and Sommers 1982) and the Bowman and Moir (1993) extraction method. Saunders and Williams In addition to the procedure described above for PT, a 1 g sample of unignited soil was shaken overnight in 50 ml of 0.5 M H2S04. After centrifugation, the samples were filtered with Whatman 41 filter paper to remove the floating organic matter. The extract was read colorimetrically, and the difference between the P contents in the ignited and unignited samples was calculated to determine the P0 content. Bowman and Moir A 5 g sample of air-dry soil was extracted in 100 mL of a 1:1 mix of 0.5 M NaOH and 0.1 M EDTA in a 125 mL erlenmeyer flask at room temperature overnight, with occasional stirring. After filtering through Whatman 41 filter paper with a Buchner funnel, a 2 mL subsample was digested with persulphate (Bowman 198 9). The P in solution after digestion was determined using the Watanabe and Olsen (1965) procedure. This method was also used as an extraction procedure for 31P NMR spectroscopy (Chapter 5). Available P The two PA methods used were the Bray PI method, as described by Olsen and Sommers (1982) and the Mehlich 3 method, as described by Mehlich (1984). For the Bray PI method, 2 g of air-dried soil in 2 0 ml of extractant and a 5 min. extraction time were used, 48 while for the Mehlich 3 method, 5 g of air-dried soil in 50 ml of extractant and a 5 min. extraction time were used. Results The results obtained for the total P analyses are shown in Table 4-1. When all horizons were combined, the PT results from Table 4-1: Means and (standard deviations) of the results for total P analyses, in mg/kg. P&A is the Parkinson & Allen (1975) digest, while S&W is the Saunders & Williams (1955) ignition method. The correlations between the two methods for all horizons combined and for each horizon separately are shown at the bottom. HORIZON LF H Bhf Bf METHOD P&A S&W P&A S&W P&A S&W P&A S&W HA-OG 714.0 (67.5) 729.0 (75.6) 561.4 (98.3) 546.9 (108.2) 524 .2 (110.5) 459.8 (139.1) 361.0 (105.5) 316.9 (69.7) CH-OG 585.8 {66.6) 611.2 (78.6) 482.8 (152.1) 511.7 (172.6) 352.1 (89.6) 330.3 (97.1) 289.2 (115.6) 267.8 (115.6) CH-0 713 .9 (181.4) 728.0 (211.0) 524 . 0 (219.9) 558.0 (195.6) 349.4 (199.5) 341.2 (211.7) 230.8 (41.1) 336.3 (237.3) CH-5 675.8 (141.9) 725.6 (134.7) 569.5 (150.6) 618.7 (93.9) 468.6 (170.6) 445.1 (188.2) 235.5 (102.2) 212 .8 (70.0) CH-10 485.3 (163.1) 540.8 (121.8) 515.4 (150.9) 528.6 (159.0) 354.5 (66.1) 353.3 (147.5) 211.8 (70.2) 193 .2 (70.2) CORR. (r) All 0.917 LF 0.869 H 0.917 Bhf 0.907 Bf 0.561 49 the Parkinson and Allen digests were highly correlated with those of the Saunders and Williams ignition method. When the horizons were analyzed individually, the results for the LF, H and Bhf were highly correlated. The results for the Bf were also correlated, but not as highly as for the other horizons. Although the mean PT values for all ages and horizons were very similar, the Saunders and Williams method produced higher PT concentrations in the highly organic surface horizons, and more PT was extracted from the mineral horizons when the Parkinson and Allen digest was used. Table 4-2 displays the results of the organic P analysis. When all of the results were examined together, those obtained by the Bowman and Moir extraction method were highly correlated with the ones from the Saunders and Williams ignition method. Separately, the methods were correlated for each horizon, but not highly. Generally, the P0 concentrations obtained by ignition were higher than those from extraction for all but the LF and H horizons of the recent burn (CH-0), particularly in the Bf horizon. The concentrations of available P are shown in Table 4-3. The Bray PI extract was highly correlated with the Mehlich 3 extract for all horizons combined and for the LF, H and Bhf horizons when analyzed separately. There was no correlation between the results of the two methods in the Bf horizon. In the LF and H, the Mehlich 3 method consistently extracted more PA. In the Bhf, the results are varied, while Bray PI extracted more PA in the Bf. Virtually no PA was extracted from the Bf horizon with the Mehlich 3 extractant. 50 Table 4-2: Means and (standard deviations) of the results for organic P analyses, in mg/kg. B&M is the Bowman & Moir (1993) extraction, while S&W is the Saunders & Williams (1955) ignition method. The correlations between the two methods for all horizons combined and for each horizon separately are shown at the bottom. HORIZON LF H Bhf Bf METHOD B&M S&W B&M S&W B&M S&W B&M S&W HA-OG 527.8 (162.0) 580.2 (49.8) 283 .8 (133.9) 442 .3 (90.1) 227.8 (110.3) 367.2 (119.9) 99.6 (35.7) 188.8 (51.3) CH-OG 492 .0 (98.8) 471.2 (75.2) 308.7 (123.2) 410.7 (156.0) 145.1 (89.2) 276.2 (86.0) 64.8 (36.4) 157.7 (85.4) CH-0 662.3 (272.4) 358.2 (129.7) 360.4 (243.9) 349.7 (93.4) 174 .2 (97.2) 234 .3 (148.6) 41.3 (23.0) 205.9 (251.6) CH-5 551.2 (177.6) 578.7 (107.8) 401.0 (127.5) 469.7 (101.1) 220.2 (111.0) 365.0 (168.1) 63 .5 (36.8) 109.7 (70.8) CH-10 391.1 (124.2) 420.3 (109.8) 306.2 (112.2) 417.0 (133.9) 198.7 (93.1) 293.9 (134.6) 21.6 (24.8) 65.4 (31.5) CORR. (r) All 0.712 LF 0.375 H 0.443 Bhf 0.562 Bf 0.403 51 Table 4-3: Means and (standard deviations) of the results for available P analyses, in mg/kg. Bray is the Bray PI extraction (Olsen and Sommers 1982), while Mehlich is the Mehlich 3 extraction (Mehlich 1984). The correlations between the two methods for all horizons combined and for each horizon separately are shown at the bottom. HORIZON LF H Bhf Bf METHOD BRAY MEHLICH BRAY MEHLICH BRAY MEHLICH BRAY MEHLICH HA-OG 38 .2 (6.2) 57.6 (8.5) 20 .6 (6.8) 28.9 (13.7) 10.2 (7.0) 7.75 (15.2) 4.80 (0.89) 0.41 (0.88) CH-OG 30.5 (10.0) 54.9 (28.3) 20.6 (9.7) 42 .8 (29.4) 6.06 (2.92) 6.90 (10.21) 6.31 (1.02) 0.03 (0.10) CH-0 89.0 (31.4) 110.5 (32.5) 44 .1 (24.9) 73.6 (31.4) 8.84 (8.12) 7.29 (12.47) 6.43 (1.12) 0 .36 (0.74) CH-5 27.2 (9.4) 56.3 (28.3) 39.1 (21.4) 68.9 (45.9) 9.24 (6.54) 9.77 (17.4) 6.06 (1.05) 0.0 (0.0) CH-10 17.3 (7.15) 33.9 (16.3) 12 .6 (5.1) 36.3 (18.2) 8.59 (1.78) 5.44 (5.89) 6.60 (0.51) 0.21 (0.63) CORR. (r) All 0.874 LF 0.889 H 0.698 Bhf 0.777 Bf -0.106 52 Discussion As was mentioned in the introduction, the traditional methods for PT determination are Na2C03 fusion and HC104 digestion. Fusion with Na2C03 is thought to be the best method, recovering all forms of P from most soils (Syers et al. 1967; Dick and Tabatabai 1977), but it is difficult to use for routine analysis. Digestion with HC104 is believed to extract the majority of PT from most soils (Sherrell and Saunders 1966; Syers et al. 1967; Syers et al. 1968), but its hazardous aspects have caused many laboratories to discontinue its use. The Saunders and Williams (1955) ignition method and the Parkinson and Allen (1975) digestion remove PT concentrations which are comparable to those of HC104 (Syers et al. 1967; Sommers and Nelson 1972; Rowland and Grimshaw 1985) and these methods are simple enough for routine analysis. The Parkinson and Allen digest was initially developed for plant material, although Parkinson and Allen (1975) felt that it would also be suitable for soils. After extensive testing on a range of British soils, Rowland and Grimshaw (1985) demonstrated that it was a comparable method to HC104 digestion for the determination of PT, even for samples high in apatite, which are known to be a problem in PT analysis (Syers et al. 1967). The Saunders and Williams method was developed to determine P0 in soils, as will be discussed below. Although the ignition step is intended only to oxidize soil organic matter prior to extraction, it may also increase the solubility of many Px compounds (Legg and Black 1955; Dormaar and Webster 1963; 53 Harrap 1963; Williams and Walker 1967; Williams et al. 1970; Anderson 1975; Olsen and Sommers 1982; Soltanpour et al. 1987; Condron et al. 1990b) , allowing it to be used to determine PT in some soils (Syers et al. 1967). For the orthic ferro-humic podzols of northern Vancouver Island, the high overall correlation suggests that either method would be suitable for PT determinations. The lower PT values in the Bf, and the lower r, indicate that the Parkinson and Allen method is better for mineral soils. The slightly higher PT concentrations in the organic matter-rich surface horizons when the Saunders and Williams method is used may indicate volatilization losses during the Parkinson and Allen digest, or interferences during colorimetric analysis (Olsen and Sommers 1982). Soil P0 is estimated by either ignition or extraction procedures (Olsen and Sommers 1982). Although it was used to determine PT in the preceding section, the Saunders and Williams (1955) ignition method was developed for P0 analysis. To yield meaningful results, ignition must quantitatively mineralize the P0 in a sample without altering the acid solubility of the native Pz, the mineralized P0 should be completely extractable in acid, and no P must be lost through volatilization (Anderson 1975). Additionally, P0 must not be hydrolysed from the unignited sample during acid extraction (Harrap 1963). Extraction procedures, such as that of Bowman and Moir (1993) must ensure that all of the P0 is removed from a sample without hydrolysis of any of the P0 forms. 54 Ignition usually produces higher P0 concentrations than extraction (Dormaar and Webster 1963; Enwezor and Moore 1966; Ipinmidun 1973; Anderson 1975; Soltanpour et al. 1987; Condron et al . 1990b), as was observed in almost every horizon and age in this study. The difference between the results from ignition and extraction procedures increased with depth in the soil profile, suggesting that the ignition method is overestimating the soil P0 due to changes in the Px during ashing, as was discussed in the section on PT analysis. Any errors due to incomplete mineralization or volatilization losses are probably minimal. Problems also exist with the Bowman and Moir (1993) extraction procedure. This method requires that Pz be determined in the extracts prior to digestion; the P0 content is calculated by subtracting the pre-digestion Pz content from the post-digestion PT content. In this study, the Px content of the undigested extracts could not be determined because the extracts were too darkly coloured for reliable colorimetric analysis. This is a common problem when soils with high organic matter contents are analyzed for P0 using extraction procedures (Olsen and Sommers 1982). The extracts from the Bowman and Moir procedure were analyzed by 31P NMR spectroscopy (Chapters 4, 5 and 6), which revealed that 20% to 80% of the P in solution was orthophosphate. Consequently, not subtracting the Px in the extract results in a serious overestimation of the soil P0 content, especially in the LF horizon of the recently burned (CH-0) sites and in the mineral horizons. Another problem inherent in extraction procedures for P0 is 55 hydrolysis of P0 compounds during extraction. The diversity of the P forms revealed by 31P NMR spectroscopy suggests that hydrolysis is minimal with the Bowman and Moir extractant. Thus, if some way were found to read the Px in the extract prior to digestion, this would be an excellent method for P0 determination in soils. There was a high overall correlation between the PA results obtained by the Bray PI and Mehlich 3 methods, and good correlation in all horizons but the Bf. The Mehlich 3 concentrations were higher than those from the Bray PI method in the LF and H horizons, while the Bray PI results were higher in the Bf. Other researchers have reported high correlations between the two methods for agricultural soils (Mehlich 1984; Wolf and Baker 1985; Michaelson et al. 1987; Cade 1989; Tran et al. 1990; Wendt 1995) and although Wolf and Baker (1985) found that the two methods extracted similar P concentrations, the others report that the Mehlich 3 method extracted more PA from soils than was extracted by the Bray PI method. In this study, virtually no PA was removed from the Bf horizon samples with the Mehlich 3 extractant, which suggests either that the Bray PI method is overestimating the PA in this horizon, or that the Mehlich 3 method is underestimating it. Tran et al. (1990) obtained similar results for a very acidic spodosol with a high P sorption capacity. They felt that the Mehlich 3 extractant was more reliable on these soils, as the higher NH4F concentration in the Bray extractant was removing strongly fixed Al-P and thus overestimating the available P. Phosphorus-31 NMR spectroscopy (Chapters 4 and 5) shows that both organic and 56 inorganic P forms are present in most of the Bf horizons of these sites, and some of this P may be plant-available. Thus, although the Bray PI method may be removing some of the fixed P, the Mehlich 3 method may not be extracting all of the available P, and the true PA value probably lies somewhere between the results of the two procedures. A measure of PA in this horizon may be immaterial, however, as feeder roots are rarely found as far down the soil profile as the Bf horizon. Conclusions For the Orthic Ferro-Humic Podzols of northern Vancouver Island, the Parkinson and Allen (1975) digest appears to be a good method for total P determinations, and to extract more completely the PT from mineral soils than the Saunders and Williams (1955) method. For organic P analysis, there are problems with both the Saunders and Williams (1955) ignition method and the Bowman and Moir (1993) extraction procedure. If some way were found to read the solution Pz after extraction and prior to digestion, despite the dark colour of the solutions, then the Bowman and Moir method would be suitable for P0 analysis for these soils. Both the Bray PI and the Mehlich 3 methods are suitable for measuring available P in the soils of this study. The Mehlich 3 procedure can be used for multielement analysis, so it may be preferred over the Bray PI method in some circumstances. 57 CHAPTER FIVE A COMPARISON OF SOIL EXTRACTION PROCEDURES FOR 31-P NMR SPECTROSCOPY Introduction Phosphorus-31 nuclear magnetic resonance (NMR) spectroscopy can be used to obtain both qualitative and quantitative estimates of the various forms of P in soils, including inorganic orthophosphate, polyphosphate, phosphonate, pyrophosphate, orthophosphate monoesters such as inositol phosphate, and orthophosphate diesters such as phospholipids, and it is analytically less complex than the detailed partition chromotography techniques otherwise required to identify specific organic P compounds. However, the natural P levels in soils are usually low. As 31P NMR is relatively insensitive, requiring more than 100 ug P ml"1 for quantitative analysis (Adams and Byrne 1989) , solution NMR is used for P, and extraction and concentration are required to produce clear spectra. An ideal extractant should remove virtually all of the P from a soil sample, without altering in any way the forms of P found in the soil. Most 31P NMR studies have employed a rapid extraction technique involving ultrasonic dispersion in 0.5 M NaOH, which usually extracts less than 50% of the total phosphorus (Newman and Tate 1980; Tate and Newman 1982; Ogner 1983; Hawkes et al. 1984; Zech et al. 1985; Zech et al. 1987; Hinedi et al. 1989; Gil-Sotres et al. 1990; Forster and Zech 1993). Others have used 0.5 M NaOH without 58 sonication (Preston et al. 1986) or with a citrate-dithionite- bicarbonate pretreatment (Ingall et al. 1990) . These treatments also removed less than 50% of the total P. Hinedi et al. (1989) used water, ice cold HC104 and a combination of HCl/HF/TiCl4, which only showed orthophosphate peaks on the NMR spectra. Condron et al. (1985) used a sequential extraction procedure of 0.1 M NaOH, 0.2 M aqueous acetylacetone and 0.5 M NaOH, washing between with 0.5 M HC1. Up to 80% of the total organic P as determined by ignition (Saunders and Williams 1955) was removed, mainly by the initial 0.1 M NaOH step of the extraction. Condron et al. (1990a) also used a sequential extraction procedure, with 0.5 M NaOH, 1 N HC1 and 0.5 M NaOH, washing with water. This too removed about 80% of the total organic P. Emsley and Niazi (1983) tried tetra-n- butyl ammonium hydroxide (Bu4NOH), hoping to utilize the salting-in effect of the large organic cation. However, this salting-in effect did not occur, and they concluded that Bu4NOH was as effective, but no more so, as NaOH or KOH. Hinedi et al. (1989) tried a sequential treatment of trichloroacetic acid (TCA) and KOH, and found that it extracted 86-99% of the total P from sewage sludge. One drawback to these methods is that, in addition to P, they extract other paramagnetic ions, such as Mn and Fe, which cause line broadening and distortion of 31P NMR spectra (Hawkes et al. 1984; Hutson et al. 1992). Adams and Byrne (1989) and Adams (1990) 59 utilized Chelex, a cation exchange resin, as an extractant in an attempt to remove these interfering ions. Chelex ™ (Bio-Rad Laboratories) is a chelating cation exchange resin which shows a high preference for Fe and other polyvalent metal ions over cations such as Na or K. [The order of preference is: Fe>Al>>Ca>>>Na; Adams and Byrne (1989)]. In the Na form, Chelex is alkaline (pH 11-12) and so can also solubilize organic P from the soil sample. This method extracted approximately the same amount of total P as the NaOH method (Adams and Byrne 198 9). Recently, Bowman and Moir (1993) have proposed the use of a mixture of NaOH and EDTA as a one-step extractant to determine total soil organic P. The NaOH can solubilize the organic P, while the EDTA is able to chelate metal cations to increase the efficiency of P extraction. This method extracted as much as twice the amount of organic P as NaOH alone (Bowman and Moir 1993). It has not yet been tested as an extractant for 31P NMR spectroscopy. With any soil extraction procedure, there is a danger that soil P compounds will be chemically altered during or after extraction. Hydrolysis, especially of orthophosphate diester to monoester, is thought to be a problem with NaOH (Tate and Newman 1982; Hawkes et al. 1984) and with Chelex (Adams and Byrne 1989). A detailed comparison of extraction methods has never been conducted on forest floor samples, which contain low levels of P in mainly organic form and relatively high levels of other paramagnetic ions. Results with any method for soil P determination are site specific (Anderson 1975). Therefore, one 60 objective of this research was to compare several soil extraction procedures, to determine the one most suitable for 31P NMR analysis of forest floor samples from CH and HA forests of northern Vancouver Island. The second objective was to examine the method of Bowman and Moir (1993) to determine its effectiveness as an extractant for 31P NMR spectroscopy. Materials and Methods Soil Samples Five forest floor samples from the sites described in Chapter 3 were used for this extraction study. One was from under an old- growth stand of hemlock-amabilis fir (HA-OG) , and one was from under an old-growth stand of cedar-hemlock (CH-OG) . The other three were from cedar-hemlock sites 0 (CH-0), 5 (CH-5) and 10 (CH- 10) years after logging and slash burning. These samples were all relatively high in total P, and had a range of other soil chemical properties (Table 5-1). These forest floor samples were air-dried and ground to pass through a 2 mm sieve. Extractants Four different extractants were used. These were: 1. 0.5 M NaOH + 0.1 M EDTA (1:1) (Bowman and Moir 19 93) 2. 0.25 M NaOH 3. 1:6 soil:Chelex (weight basis) in deionized water (Adams and Byrne 198 9) 4. 1:6 soil:Chelex (wt. basis) in 0.25 M NaOH 61 Table 5-1: Chemistry of the soils used for the extraction trials. Note: pH measured in CaCl2; C measured with Leco; total N measured via modified Kjeldahl digest; available Ca, Mg, Fe and Al measured with Mehlich extraction; available P measured with Bray PI; total P, inorganic P and organic P 1 measured with Saunders & Williams ashing. Sample pH C (%) LOI Total N (%) C/N Avail Ca (mg/kg) Avail Mg (mg/kg) Avail Fe (mg/kg) Avail Al (mg/kg) Avail P (mg/kg) Total P (mg/kg) Inorg P (mg/kg) Org P 1 (mg/kg) HA-OG 3.1 49.9 3500 1.159 43 . 1 1906.1 393.4 139.6 380.7 40 . 04 674.0 129.0 545.0 CH-OG 3.5 37.2 695 0.851 43 . 7 3300.2 255.6 182.6 649.1 51.5 582.0 162 .0 420 .0 CH-0 5.0 23 .2 213 0.879 26.4 7395.0 640.0 220.0 720.0 68.4 796.0 396.0 400 .0 CH-5 3.4 46.7 1742 0.807 57.9 3639.1 498.0 118.3 364.9 23 .47 713 .0 140.0 513 .0 CH-10 4 .2 49.6 1006 0.995 49.8 7188.8 673.5 112.2 193.9 21.16 653.0 124 . 0 589.0 For each extractant, 5 g of air-dry soil and 100 ml of liquid were used. The NaOH-EDTA and NaOH samples were extracted in 125 ml erlenmeyer flasks at room temperature overnight with occasional stirring. For both of the Chelex procedures, samples were 62 extracted in 250 ml plastic bottles overnight at room temperature on a reciprocal shaker. All samples were then filtered with Buchner funnels and Whatman 41 filter paper. A subsample of each was digested with persulphate (Bowman 1989) and was read using the Watanabe and Olsen (1965) method to determine the total amount of phosphorus which had been extracted. The remainder of each sample was freeze-dried. Preparation of NMR Samples Approximately 1 g of the freeze-dried extract was weighed into 50 ml plastic centrifuge tubes, with 2.5 ml of D20. Samples were vortexed for 2 min. A few of the Chelex + NaOH samples were prepared in duplicate, and to one of each pair was added 1 pellet (approx. 0.5 g) of NaOH prior to vortexing. All of the samples were left to stand for 2 h, and then were centrifuged. The supernatants were transferred into 10 ml NMR tubes and were refrigerated until used for NMR spectroscopy. NMR Analysis Phosphorus-31 NMR spectra were obtained at 101.27 MHz on a Bruker WM 250 high resolution NMR spectrometer using a 45° pulse with a 1.5 s delay and an acquisition time of 0.508 s. The P spectra were proton decoupled using an inverse-gated pulse sequence to overcome the nuclear Overhauser enhancement in order to achieve quantitative results. Accumulation times ranged from 24 to 48 h, and were dependent on the length of time necessary to achieve a strong signal-to-noise ratio. The assignment of peaks was based on Newman and Tate (1980) and Adams and Byrne (1989). Peak areas were 63 determined by integration. Metals Analysis To assess the effect of the chelators (EDTA and Chelex) on interfering paramagnetic ions, the concentrations of Fe, Mn, Cu and Zn in the solutions following extraction were measured by atomic absorption spectroscopy, prior to freeze-drying. Although Zn and Cu are not paramagnetic ions, as divalent cations they should be affected by EDTA and Chelex, and their levels in these forest floor samples were high enough for reliable measurement. In addition, an adsorption trial was conducted. Iron at concentrations of 0, 20, 30, 40, 60 and 100 mg/1 or Mn at 0, 15, 20, 30, 40 and 80 mg/1 were added to samples containing either 30 ml of 0.05M EDTA or 5 g Chelex in 30 ml of water, with or without 1 g of air-dry soil. A blank containing 3 0 ml of water was used as a control. The soil sample used was CH-OG. NaOH was not used in this trial as it caused the metals to precipitate. The samples were placed in stoppered 100 ml centrifuge tubes. After shaking on a reciprocal shaker for 1 hour, the samples were filtered through Whatman 42 filter paper, and then metal concentrations were determined using atomic absorption spectroscopy. Added P Compounds To determine the effect of the extractants on various P compounds, 5 g samples of CH-OG had 0.05 g of the following compounds added prior to extraction by 0.2 5 M NaOH + 0.05 M EDTA or by 1:6 soil:Chelex (wt. basis) in 0.25 M NaOH: ATP (Adenosine 5'- Triphosphate, disodium salt, Grade II, Sigma A-3377); 64 glycerophosphate (disodium pentahydrate, Sigma G6504); or K- polyphosphate (formed by fusion of KH2P04 as per Kulaev (1979) ) . These were then prepared for NMR analysis as was previously described. Results NMR Spectra Figure 5-1, A-E, displays the NMR spectra generated from these extractant trials, while Table 5-2 shows a guide for the interpretation of the peaks. It should be noted that the lower pH in the Chelex + water extraction causes a peak shift, reversing the orthophosphate and the monoester peak positions relative to the other extraction procedures. This was also observed by O'Neill et al. (1980) and Adams and Byrne (1989) . In some of the spectra for the extractant NaOH + EDTA the peaks for orthophosphate and monoester P overlap. When there was clear peak separation with this extractant, there was a valley between the peaks at 6 ppm, and this therefore was used as the dividing line where overlapping occurred. The sharpest peaks, with the best separation, were produced by the Chelex + NaOH extraction. These were further improved when extra NaOH was added to adjust the pH prior to NMR analysis. The broadest peaks were produced by the NaOH + EDTA extraction. These spectra also had the poorest separation of the orthophosphate and monoester peaks. The trends from sample to sample were quite consistent for each extraction method, but the 65 HA OLD GROWTH LF 20 IS P H O S 10 s O R T H O M O N O ~T~ "T" 0 PPH PYRO DIEST 1 -20 POLY Figure 5-1A: Phosphorus-31 NMR spectra for HA old growth forest floor extracted with: Chelex + NaOH; Chelex + Water; NaOH + EDTA; and NaOH. Phos is phosphonate, ortho is orthophosphate, mono is monoester P, diest is diester P, pyro is pyrophosphate, and poly is polyphosphate. 66 CH OLD GROWTH Figure 5-IB: Phosphorus-31 NMR spectra for CH old growth forest floor extracted with: Chelex + NaOH; Chelex + Water; NaOH + EDTA; and NaOH. Phos is phosphonate, ortho is orthophosphate, mono is monoester P, diest is diester P, pyro is pyrophosphate, and poly is polyphosphate. 67 CHO YEAR Figure 5-1C: Phosphorus-31 NMR spectra for CH 0-year forest floor extracted with: Chelex + NaOH; Chelex + Water; NaOH + EDTA; and NaOH. Phos is phosphonate, ortho is orthophosphate, mono is monoester P, diest is diester P, pyro is pyrophosphate, and poly is polyphosphate. 68 C H 5 YEAH CHELEX + NaOH + NaOH 21 PHOS I t 3 ORTHO MONO DIEST PYRO - I f POLY Figure 5-1D: Phosphorus-31 NMR spectra for CH 5-year forest floor extracted with: Chelex + NaOH; Chelex + Water; NaOH + EDTA; and NaOH. Phos is phosphonate, ortho is orthophosphate, mono is monoester P, diest is diester P, pyro is pyrophosphate and poly is polyphosphate. 69 CH i O YEAR CHELEX + NaOH «• NaOH V^AvVrv^v^v/vv^wV^^A/n^^ p«os 1< 5 ORTHO MONO OIEST PYRO - J t POLY Figure 5-1E: Phosphorus-31 NMR spectra for CH 10-year forest floor extracted with: Chelex + NaOH; Chelex + Water; NaOH + EDTA; and NaOH. Phos is phosphonate, ortho is orthophosphate, mono is monoester P, diest is diester P, pyro is pyrophosphate and poly is polyphosphate. 70 TABLE 5-2: Interpretation of 31P NMR spectra, showing the ppm range at which the various P compound classes are located. PPM 15-20 6-8 3-6 l-(-D (-3)-(-6) (-20) COMPOUNDS phosphonates orthophosphate phosphate monoesters -inositol phosphates -sugar phosphates -mononucleotides phosphate diesters -phospholipids -RNA, DNA pyropho sphat e polyphosphate ATP, ADP results from each extractant were very different within each forest floor sample. NaOH + EDTA was the only method to show polyphosphate peaks for three of the samples, while diester peaks in the NaOH extracts were seen only with the HA-OG samples (Fig. 5- 1A) . Phosphonate was only unambiguously detected in sample CH-5 (Fig. 5-ID) using the three extractions involving Chelex. Table 5-3 shows the percentage of P found within each class of compounds, calculated from the spectra by integration. The NaOH + EDTA extraction seems to have produced the greatest range of compounds. It also extracted the most P of all the extractants: 63.3-98.5% of PT, compared with 20.1-34.2% by NaOH, 7.7-11.4% by Chelex + water and 21.0-36.2% by Chelex + NaOH (Table 5-4) . The NaOH + EDTA also extracted more diester P than monoester P (Table 5-5), producing higher diester/monoester ratios than any other method except the Chelex + water extraction. The NaOH extraction method extracted 71 Table 5-3: The percentage of total P in solution found in the various P compound classes, as calculated from 31P NMR spectra by integration. Phos is phosphonate, orth is orthophosphate, mono is monoester P, dies is diester P, pyro is pyrophosphate and poly is polyphosphate. Sample HA-OG CH-OG CH-0 CH-5 CH-10 Extractant Chelex + NaOH Chelex + H20 NaOH + EDTA NaOH Chelex + 2 NaOH Chelex + NaOH Chelex + H20 NaOH + EDTA NaOH Chelex + NaOH Chelex + H20 NaOH + EDTA NaOH Chelex + 2 NaOH Chelex + NaOH Chelex + H20 NaOH + EDTA NaOH Chelex + 2 NaOH Chelex + NaOH Chelex + H20 NaOH + EDTA NaOH Phos 0 0 0 0 0 0 0 0 0 0 0 0 0 3 2 3 0 0 0 1 0 0 0 Orth 26 62 21 36 58 36 54 17 51 74 73 51 55 34 31 49 23 49 34 15 68 17 49 Mono 44 16 49 51 30 40 18 33 43 18 13 33 45 38 39 22 40 33 42 57 14 27 41 Dies 15 22 15 4 5 14 23 39 0 6 13 11 0 8 9 18 18 0 8 9 10 32 0 Pyro 15 0 7 9 7 10 5 0 6 2 1 6 0 17 19 8 7 18 16 19 4 10 10 Poly 0 0 7 0 0 0 0 11 0 0 0 0 0 0 0 0 12 0 0 0 4 14 0 72 Table 5-4: P extracted by various methods. The percent of total P was calculated using the Total P values shown in Table 5-1. Extractant NaOH+Chelex H20+Chelex NaOH+EDTA NaOH (mg/kg) % Total P (mg/kg) % Total P (mg/kg) % Total P (mg/kg) % Total P HA-OG 210.1 31.2 77.4 11.4 526.4 78.1 205.0 30.4 CH-OG 210.5 36.2 66.2 11.4 424.8 73.0 198.8 34 .2 CH-0 166.8 21. 0 82.8 10.4 643.5 80.8 159.6 20.1 CH-5 182.6 25.6 55. 0 7.71 451.6 63.3 198.8 27.9 CH-10 183 .7 28.1 71.6 11.0 643 .1 98.5 207.6 31.8 Table 5-5: Orthophosphate diester/monoester ratios, calculated from Table 5-3. Chelex/2 NaOH Chelex/NaOH Chelex/H20 NaOH/EDTA NaOH HA-OG N/A 0.34 1.38 0.31 0.08 CH-OG 0.30 0.35 1.27 1. 18 0 CH-0 N/A 0.33 1.00 0.33 0 CH-5 0.21 0 .17 0.82 0.45 0 CH-10 0.19 0.16 0.71 1.18 0 the fewest types of P compounds, showing peaks for orthophosphate and monoesters in all samples, but peaks for diesters and pyrophosphate in only a few samples. There were no peaks for phosphonates or polyphosphates with this extraction procedure. 73 Metals Analysis Table 5-6 displays the analysis of metals within each solution following extraction. The Fe concentrations extracted by NaOH alone and by H20 + Chelex were comparable (240-1080 versus 120-870 Table 5-6: Metals measured in each solution following extraction, in mg/kg. Sample HA-OG CH-OG CH-0 CH-5 CH-10 Metal Fe Mn Cu Zn Fe Mn Cu Zn Fe Mn Cu Zn Fe Mn Cu Zn Fe Mn Cu Zn NaOH 390 60 30 12 1080 150 24 0 900 150 24 0 240 120 36 21 300 210 9 9 NaOH + EDTA 210 240 36 21 870 720 27 15 1890 960 33 27 120 900 24 45 180 1530 27 51 H20 + Chelex 300 0 21 0 870 60 9 0 780 90 24 0 120 60 21 0 300 60 21 12 NaOH + Chelex 90 0 18 0 450 0 9 0 450 60 3 0 0 0 12 0 120 0 21 0 74 ug/1) . The Fe levels in the NaOH + EDTA solutions were intermediate, except for CH-0, the recent burn. NaOH + Chelex solutions contained the lowest levels of Fe. NaOH + EDTA extracted the greatest concentration of Mn, especially in the three postburn samples. The lowest Mn levels were found in the two Chelex extracts. The NaOH and NaOH + EDTA extracts contained similar levels of Cu; NaOH + Chelex had the least Cu. NaOH + EDTA extracts had the most Zn; the other methods generally.contained very little if any Zn. Figure 5-2 displays the results from the adsorption trial, using soil. For the samples without soil, the EDTA and water extracts contained the same levels of Fe or Mn that had been added to the samples, while the Chelex removed all of the Fe and Mn, reading 0 mg/1 at all levels. In Figure 5-2 (A) , the EDTA kept all of the added Fe in solution, and also extracted additional Fe from the soil. The Chelex removed almost all of the added Fe from solution. With water alone, much of the Fe was adsorbed onto the soil surfaces, as the Fe in solution at all levels was lower than that which had been added to the sample. For Mn (Fig. 5-2 (B)), the Chelex removed all of the added Mn from solution. The EDTA kept what had been added to the sample in solution, but did not appear to extract any additonal Mn from the soil. At the highest level of Mn addition (80 mg/1) there appeared to be some adsorption onto the soil. In water alone, all of the Mn added appeared to stay in solution. 75 Fe in Solution (mg/l) A) 120 100 20 30 40 Fe Added (mg/!) Water + Chelex ^ E D T A 100 Mn in Solution (mg/l) B) 120 20 30 Mn Added (mg/l) Figure 5-2: Metals in solution following the adsorption trial. The chelators used, Chelex and EDTA, were in water. CH-OG was the soil sample. The results with Fe are shown in (A), while those for Mn appear in (B). 76 P Addition The results from the addition of P compounds to the NMR extracts are shown in Figure 5-3, A and B. The P concentrations in each extract and the proportion of P in each compound class are found in Table 5-7. The Chelex + NaOH extraction produced much sharper peaks than those by NaOH + EDTA, but peaks are seen at the same positions in the spectra regardless of extractant (Fig. 5-3 A and B). The added P compounds dominate each spectrum, changing the spectra from those of the original extract. When added polyphosphate was extracted with NaOH + EDTA (Fig. 5-3A) , there was a large, broad peak at -20 ppm, the polyphosphate position. Pyrophosphate, diester, monoester and orthophosphate peaks also appeared, but they were small and broad relative to the polyphosphate peak. In the Chelex + NaOH extract (Fig. 5-3B) , there was a sharp polyphosphate peak, as well as sharp monoester and pyrophosphate peaks. The main difference between these two methods when polyphosphate was added was the amount of P extracted: 6500 mg/kg for NaOH + EDTA versus 3800 mg/kg by Chelex + NaOH (Table 5-7) . NaOH + EDTA extracted more of the total P as orthophosphate, and less as monoester. When ATP was added to the sample, orthophosphate, monoester and diester peaks appeared (Fig. 5-3, A and B). There were also three approximately equal peaks at -5, -10 and -20 ppm. These represent the alpha, beta and gamma phosphates in the ATP molecule. Both extraction methods yielded the same concentration of P in solution (Table 5-7) and approximately the same percentages of 77 NaOH + EDTA POLYPHOSPHATE + ATP MONO DIEST PYRO POLY Figure 5-3A: Phosphorus-31 NMR spectra in soil and after the addition of polyphosphate, ATP and glycerophosphate, extracted with NaOH + EDTA. 78 CHELEX + NaOH • POLYPHOSPHATE + ATP WrAysvVj w^A^Aŷ vvvH ^ Vvww v ^ j ^WU^AMV-* ^ ^ v A ^ J ^ V v O V , GLYCEROPHOSPHATE ^ v ^^M^Avv^AMv^^^^^ ORIGINAL EXTRACT T 20 PHOS —i— 15 10 ORTHO — i — o PPH — I - I S -25 MONO DIEST PYRO POLY Figure 5-3B: Phosphorus-31 NMR spectra in soil and after the addition of polyphosphate, ATP and glycerophosphate, extracted with Chelex + NaOH. 79 Table 5-7: Total P content, and the proportions of P in the various soil P compound classes, after the addition of polyphosphate, ATP or glycerophosphate to soil, and after extraction with either NaOH + EDTA or Chelex + NaOH. For ATP, the peaks at -5, -10 and -20 are the alpha, beta and gamma phosphates in the ATP molecule. Sample NaOH+EDTA + polyphos. NaOH+Chelex + polyphos. NaOH+EDTA + ATP NaOH+Chelex + ATP NaOH+EDTA +glycerophos. NaOH+Chelex +glycerophos. NaOH+EDTA orig. extract NaOH+Chelex orig. extract P in Soil mg/kg 6500 3800 3600 3500 5900 3000 424 .8 210.5 ortho phos . o 11.7 0 14 .5 2.8 26.4 3 .8 33 .3 36.1 Mono- ester % 6.3 40.3 10.2 17.8 63.9 89.7 16.7 40.3 Di- ester o 4.3 4.9 7.2 4.7 9.7 6.5 38.9 13 .9 Pyro- phos . % 12.8 16.1 24.7 alpha 28 .0 alpha 0 0 0 9.7 (-10) peak % n/a n/a 21.7 beta 27.1 beta n/a n/a n/a n/a poly phos % 64.9 38.7 21.7 gam. 19.6 gam. 0 0 11.1 0 total P in the various compound classes, although NaOH + EDTA had slightly more orthophosphate and less monoester P than did Chelex + NaOH. Glycerophosphate, a monoester formed in soil after hydrolysis of glycerophosphatides (Hance and Anderson 1963), appeared at the monoester position with both extraction methods (Fig. 5-3, A and B) . The NaOH + EDTA solution (Fig. 5-3A) contained nearly twice as much P as the Chelex + NaOH solution (Table 5-7), and had more P as orthophosphate and less as monoester. 80 Discussion Of the reagents used in this study, NaOH + EDTA extracted the most P from each forest floor sample, with results comparable to those from sewage sludge extracted with TCA and KOH (Hinedi et al. 1989) or from sequential extraction of soil (Condron et al. 1990a) . This agrees with the findings of Bowman and Moir (1993), that it is a good extractant of organic P. The amount of P extracted by NaOH (20-34%) is comparable to, or slightly lower than that reported in the literature (Newman and Tate 1980; Tate and Newman 1980; Hawkes et al. 1984; Ingall et al. 1990; Gil-Sotres et al. 1990; Forster and Zech 1993) . The levels of P extracted by Chelex with both water and NaOH are lower than those obtained by Adams and Byrne (1989) and Adams (1990), which were comparable with NaOH levels, as was the amount extracted by Bu4NOH (Emsley and Niazi 1983) . There were also lower P levels in the Chelex + NaOH samples after P compounds such as polyphosphate and glycerophosphate were added prior to extraction. The quality of the spectra produced by NaOH + EDTA was poor, however, relative to the other extractants used in this trial, with poor separation of the orthophosphate and monoester P peaks. This seems to be caused by the complexing by EDTA of paramagnetic ions other than P, particularly Fe and Mn. The one sample in which the peaks were clearly separated (HA-OG) contained the lowest concentrations of Fe and Mn. High concentrations of Mn seem to cause most of the peak overlap: spectra containing less than 200 81 ug/g of Mn have good separation between the orthophosphate and monoester P peaks. Peak broadening by Mn in 31P NMR spectra has been observed elsewhere (Hutson et al. 1992). This is one drawback to the NaOH-EDTA method - it removed cations from P compounds to allow more P to be extracted, but did not remove the metals from solution as did Chelex. EDTA alone did not extract much Fe from the soil, however. More Fe was released when EDTA was combined with NaOH, probably due to the solubilization of organic matter (Stevenson 1994). The best spectra, in terms of peak separation and signal-to-noise ratio, were produced by the Chelex + NaOH extractions. These were further improved by adjusting the pH prior to analysis. However, overlapping of the orthophosphate and monoester P peaks was seen in all of the Chelex + H20 spectra, and in the specta for the CH-10 sample extracted with Chelex + NaOH. The quality of spectra reported in the literature are quite variable. NaOH often produces poorly resolved resonance in the orthophosphate monoester region, with monoesters appearing as shoulders on the orthophosphate peak (Newman and Tate 198 0; Zech et al. 1987; Hinedi et al. 1989; Condron et al. 1990a; Gil-Sotres et al. 1990). This has also been reported in Chelex extracts (Adams and Byrne 1989). The TCA and KOH extractions (Hinedi et al. 1989) produced clear, sharp spectra, as did the sequential extraction procedure of Condron et al. (1985). However, Hinedi et al. 1989 examined sewage sludge, which may not be comparable to forest floor. It is difficult to judge the quality of Bu4N0H as an 82 extractant, as spectra were not published, and results for orthophosphate and monoesters were reported as one peak (Emsley and Niazi 1983). There were considerable differences in the diversity of compounds extracted by the different reagents in this study. NaOH + EDTA extracted the most, while NaOH extracted the least. This may in part be due to the amount of P extracted by each reagent, but it may also be due to the nature of the reagent. The diester/monoester ratios suggest that NaOH caused hydrolysis of orthophosphate diesters, which did not occur with NaOH + EDTA. Hydrolysis has been reported by other NMR researchers (Ogner 1983; Adams and Byrne 1989; Ingall et al. 1990), and Hance and Anderson (1963) reported that phospholipids were readily hydrolysed in alkaline solution. The diester/monoester ratios for NaOH are also comparable to reported values of 0.18-0.46 (Zech et al. 1985; Gil- Sotres et al. 1990; Forster and Zech 1993). The lack of diesters in NaOH samples in this particular study is probably due to the length of extraction - overnight in this case, as opposed to only a few minutes with sonication, used by other researchers. It should be noted that the lack of clear separation of the orthophosphate and monoester P peaks using NaOH + EDTA probably underestimates the proportion of P in either or both of these compound classes, and thus may widen the diester/monoester ratio. The lower compound diversity in the Chelex extracts compared with NaOH + EDTA may be due to a loss of P compounds when the 83 Chelex is removed from the extracting solution. When P compounds such as polyphosphate and glycerophosphate were added to the soil prior to extraction, the NaOH + EDTA extracted more P than Chelex + NaOH. This would suggest that P is removed along with the Chelex during filtration, possibly via cation linkages. The K+ from the added K-polyphosphate may be exchanged with divalent cations such as Mn++ or Fe++, which could then link the polyphosphate to the Chelex (Levesque and Schnitzer 1967). Polyphosphates are rarely seen in reported NMR spectra (Tate and Newman 1980; Emsley and Niazi 1983; Zech et al. 1987; Adams and Byrne 1989), and are usually at very low levels. This is surprising as they are widely distributed in nature, especially in forest soils (Kulaev 1979; Martin et al. 1985). The high amounts extracted by NaOH + EDTA in this study may be due to the greater amount of total P extracted by this method, as they are not seen in the spectra of the other extractants. They may be hydrolysed by other reagents (Subbarao et al. 1977), they may be part of the more than 50% of the total P which was not extracted, or they may be lost during the extraction procedure. It is unlikely that they are artifacts, as the NaOH + EDTA extracting conditions are not conducive to polymerization (Kulaev 1979). Conclusions Phosphorus-31 NMR spectroscopy is a valuable tool for the direct identification of P compounds in soil extracts. The major 84 advantage is that it is analytically less complex than the detailed partition chromatography techniques otherwise required for identifying specific organic P compounds. Extraction of P compounds from soil for 31P NMR analysis is possible with a variety of reagents. The forms of P which dissolve depend on the reagent; the complex nature of soil organic P may make it impossible for a single extractant to dissolve all P compounds. It also may not be possible to produce high quality spectra without some alteration of P compounds by hydrolysis. It appears from this study that the NaOH-EDTA extraction procedure for organic P (Bowman and Moir 1993) could be used as an extractant for 31P NMR analysis. It extracted a higher concentration of P and a greater diversity of P compounds that any other extractant tested, with less apparent hydrolysis of compounds. However, it also maintained other paramagnetic ions in solution, causing line broadening and overlapping peaks, thus reducing the quality of spectra. Consequently, it would be most suitable for samples with high P levels and low levels of interfering ions, unless some way could be found to remove the EDTA-metal complexes after extraction without altering the P compounds in solution. This study also demonstrated that the extractant used will greatly affect the results of 31P NMR analysis of soil samples, making it difficult to compare results from studies using different extractants. 85 CHAPTER SIX A COMPARISON OF PHOSPHORUS FORMS ON CH AND HA SITES Introduction As discussed in the general introduction, the forests examined on northern Vancouver Island exist as two phases. These are: the CH type, composed of western red cedar and western hemlock; and the HA type, composed of western hemlock and amabilis fir. After logging and planting, the trees on the CH sites suffer a growth check and become chlorotic, while the HA plantations grow well. Studies by Germain (1985) and Weetman et al. (1989a, b) indicated that trees on the CH sites were deficient in nitrogen and phosphorus and responded well to fertilization, suggesting a nutrient cycling problem. Carbon and N forms on these sites have been examined in some detail (for example, see de Montigny 1992; Keenan 1993; Prescott and Weetman 1994; Chang 1995). Phosphorus has been less thoroughly studied. Some research suggests that there are higher concentrations of available and total P on the HA sites (Messier 1991; de Montigny 1992; Keenan 1993; Prescott et al. 1993; Prescott et al. 1994), but no detailed investigation of P has been conducted. There appears to be less effective decomposition on the CH sites. All layers of the CH forest floor had smaller concentrations of total and extractable N, and mineralized less N 86 during aerobic incubations in the laboratory (Prescott et al. 1993) . Also indicative of less complete decomposition was the slightly higher ratio of carbohydrate to lignin moieties in the CH forest, as revealed by 13C NMR spectroscopy. Nitrogen alone does not appear to control the rate of litter decomposition (Prescott 1995); other nutrients such as P may play an important role. Differences in decomposition may result in different concentrations and forms of organic P between the CH and HA forest types. The objective of the research in this chapter was to examine the soil phosphorus of mature, uncut CH and HA stands. Concentrations of total, organic and available P were determined by extraction and digestion procedures, and fractionation and 31P nuclear magnetic resonance (NMR) spectroscopy were used to characterize the P forms. Some aspects of the chemistry of these soils which could influence P forms and levels were also examined. Materials and Methods Sample Collection The study sites and sampling design were described in Chapter 3. Sampling was conducted in late July and August, 1992, using three locations for each of the CH and HA forest types and three pits per location, for a total of nine samples per horizon per forest type. Samples from non-woody horizons were collected after pits were dug, using a metal trowel. They were placed in plastic bags in coolers, and later were frozen for transportation. In the 87 laboratory, the samples were thawed, air-dried at 25°C, and sieved to less than 2 mm. The forest floor samples were ground with a stainless steel coffee grinder prior to sieving. All samples were subsequently stored at room temperature in airtight plastic containers. General Chemical Analysis Gravimetric moisture content was determined by oven-drying at 105°C for 16 hours, using thawed samples prior to air-drying. Air- dried moisture contents were obtained the same way, but using dried and sieved material (Lavkulich 1981). Soil pH was measured in both water and 0.01 M CaCl2, using a 1:2 (w/v) soil:liquid ratio for mineral horizons and a 1:5 (w/v) ratio for forest floor material (Lavkulich 1981; McLean 1982). To determine loss on ignition (LOI), oven-dried samples were burned in a muffle furnace for 1 hour at 3 7 5°C and for 16 hours at 5 5 0°C. The LOI was calculated from the weight difference between the oven-dried and the ashed sample. Total C was estimated by the use of a Leco Induction Furnace (Bremner and Tabatabai 1971; Lavkulich 1981), using oven-dried material. Total N was measured on oven-dried samples by using a semimicro Kjeldahl procedure to convert the N to ammonium (Lavkulich 19 81) . The ammonium in solution was then determined colorimetrically with a Lachat Flow Injection Analyzer (FIA). Calcium, Mg, Fe and Al were extracted from air-dried samples by the Mehlich method III (Mehlich 1984) and were read using atomic 88 absorption spectroscopy (AAS) . Iron, Al and Mn were determined on oven-dried material using sodium pyrophosphate extraction, acid ammonium oxalate extraction and citrate bicarbonate dithionite extraction (Lavkulich 1981), followed by AAS. P Methodology Available P (PA) was extracted from air-dried material by the Bray Pi method (Bray and Kurtz 1945; Olsen and Sommers 1982). Solution P was then measured colorimetrically (Watanabe and Olsen 1965) using the Lachat FIA. Total P (PT) was determined on oven- dried material by the Parkinson and Allen (1975) digest, with colorimetric analysis on the Lachat FIA. The new extraction procedure of Bowman and Moir (1993) was utilized to extract P0 from air-dried samples, followed by persulphate digestion (Bowman 1989) and colorimetric analysis on a Technicon Autoanalyzer. A comparison of these methods to others for PA, PT, and P0 was discussed in Chapter 4. The Chang and Jackson (1957) procedure, as described by Olsen and Sommers (1982) was utilized to fractionate P into NaOH (PNa0H) , citrate-bicarbonate (PCB) / citrate-bicarbonate-dithionite (PCBD) an<3 HC1 (PHCi) extractable forms. NMR Spectroscopy Two soil profiles for each of the CH and HA forest types were chosen for analysis by 31P NMR spectroscopy. The samples selected were high in PT, and had chemical characteristics close to the mean values for the forest type they represented. LF, H, Bhf and Bf 89 horizons were analyzed for each profile. Air-dried soils were extracted using 100 ml of a 1:1 mixture of 0.25 M NaOH and 0.05 M EDTA (Bowman and Moir 1993), as was described in detail in Chapter 5. The NMR sample preparation and analytical procedure was described in Chapter 5. Statistical Analysis Statistical analyses were conducted using the Systat program (Wilkinson 1990) to perform analysis of variance test at p<0.05, using a nested design (Hicks 1982), with location nested within each forest type. The model statement used was: VARIABLE = CONSTANT + TYPE + HORIZON + LOCATION{TYPE} + TYPE*HORIZON + HORIZON*LOCATION{TYPE}, with VARIABLE the measured parameters such as available P. The assumption with this model is that the three locations used for each forest type are not significantly different from one another. Thus the HORIZON*LOCATION{TYPE} term should not be significant. A significant result for the TYPE*HORIZON term indicates that the CH and HA forest types are significantly different for a particular horizon. Pearson pairwise correlations and Tukey' s HSD tests were also conducted with the Systat program. Homogeneity of variance was determined by plotting residuals against estimates. Log and log (n+1) transformations were performed where necessary. Results The results from the analysis for field moisture, air dry moisture, pH in water and pH in CaCl2 are shown in Table 6-1. 90 Table 6-2 displays the analysis of variance results for these data. The field moisture content was relatively uniform for the surface horizons, but dropped in the Bf. There were no significant differences between the CH and HA forest types. The air-dry moisture content was similar for all horizons, and there were no significant differences between the two forest types. Measuring the pH in water and in CaCl2 produced similar trends, but the pH values were lower in CaCl2 than in water. The highest pH with both methods was in the Bf horizon. The pH of the surface horizons of the CH forests was significantly higher than that of the HA in both water and CaCl2. The pH in CaCl2 of the H horizon of the CH was also significantly higher than that of the HA. Table 6-1: The mean values and (standard deviations) of field moisture content, air-dry moisture content and pH in water and CaCl2. A * indicates a statistically significant difference between the HA and CH forest types for a horizon at p<0.05. (n=9) HOR. LF H Bhf Bf FOR. TYPE HA CH HA CH HA CH HA CH FIELD MOISTURE % 226 (120) 203 (122) 230 (76) 314 (149) 206 (109) 249 (152) 97 (48) 84 (27) AIR-DRY MOISTURE % 12.65 (0.84) 12.95 (1.18) 13.39 (3.74) 16.17 (6.06) 12.67 (10.17) 8.70 (4.69) 13.53 (8.70) 10.44 (10.19) pH IN WATER 3.6(0.19)* 3.9(0.22) 3.4(0.11) 3.6(0.13) 3.8(0.32) 3.8(0.23) 4.6(0.31) 4.5(0.32) pH IN CaCl2 3.2(0.14)* 3.6(0.27) 2.9(0.10)* 3.1(0.19) 3.2(0.32) 3.2(0.17) 4.0(0.31) 4.0(0.29) 91 The highest carbon concentrations were found in the organic horizons (Table 6-3). The HA forests had higher C concentrations in the Bhf and Bf horizons than the CH. However, the HORIZON*LOCATION{TYPE} term was also significant (Table 6-2), indicating that the locations used had significant differences in Table 6-2: Analysis of variance table for field moisture, air-dry moisture, pH in water, pH in CaCl2, total C, total N, C/N ratio, loss on ignition (LOI), and exchangeable Ca, Mg, Fe and Al. As a nested experimental design was used, the terms for the ANOVA were TYPE, HORIZON, LOCATION{TYPE}, which is location nested within forest type, TYPE*HORIZON, HORIZON*LOCATION{TYPE}, and multiple R2. Shown here are the probabilities as calculated by the Systat statistical program. A * indicates statistical significance at p<0.05. Field Moisture Air Dry Moisture pH in Water pH in CaCl2 C N C/N LOI Extr. Ca Extr. Mg Extr. Al Extr. Fe TYPE 0.551 0.111 0.064 0.003 0.016 0.000 0.001 0.880 0.666 0.001 0.000 0.000 HORIZON 0.000 0.001 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 LOCATION {TYPE} 0.000 0.002 0.313 0.005 0.180 0.118 0.078 0.455 0.003 0.000 0.099 0.489 TYPE* HORIZON 0.349 0.085 0.018* 0.003* 0.000* 0.121 0.306 0.011* 0.890 0.135 0.184 0.103 HORIZON* LOCATION {TYPE} 0.010 0.191 0.052 0.226 0.006* 0.013 0.973 0.055 0.243 0.013 0.555 0.103 MULT. R2 0.796 0.577 0.833 0.848 0.963 0.924 0.687 0.946 0.795 0.871 0.883 0.731 92 C, and thus did not fit the assumptions for the experimental design. Loss on ignition (LOI) was significantly higher in the Hhorizon of the CH forests than the HA, and was also higher, but not significantly so, in the LF horizons in the CH forests. In the Bhf and Bf horizons, LOI was generally higher on the HA sites. There were no significant differences between the HA and CH in total N concentration or C/N ratio (Tables 6-2 and 6-3) . However, the HA forests generally had higher N concentrations in all horizons, and had a lower C/N ratio than the CH forests. Table 6-3: The mean values and (standard deviations) of total carbon (%), total nitrogen (%), loss on ignition (LOI) and the C/N ratio. A * indicates a statistically significant difference between the HA and CH forest types for a horizon at p<0.05. (n=9) HOR. LF H Bhf Bf FOR. TYPE HA CH HA CH HA CH HA CH C % 46.58 (5.10) 48.05 (4.31) 40.85 (7.9) 47.32 (6.2) 20.03 (4.67) 14.32 (4.74) 8.81 (2.55) 6.74 (1.21) TOTAL N % 1.06 (0.18) 0.92 (0.11) 1.01 (0.10) 0.91 (0.16) 0.77 (0.14) 0.54 (0.20) 0.24 (0.06) 0.17 (0.04) C/N 44.54 (4.80) 52.56 (6.20) 40.53 (7.66) 53.40 (11.59) 26.09 (4.48) 29.78 (9.55) 37.04 (4.99) 40.99 (8.25) LOI 1975.4 (907.7) 2000.5 (612.2) 1168.0* (921.3) 2014.4 (780.6) 142.1 (104.7) 87.8 (64.5) 25.54 (6.41) 21.53 (4.53) 93 The total N concentrations were highest in the surface horizons and lowest in the Bf. The C/N ratio was lowest in the Bhf. The concentrations of extractable Ca, Mg, Al and Fe are shown in Table 6-4, and the analysis of variance results may be found in Table 6-2. Generally, the CH contained more extractable Ca, while the HA contained more extractable Mg, Al and Fe. However, the variability was high and the differences were not significant. Pyrophosphate-extracted Fe (Table 6-5) was generally higher in the HA than CH, but the results were not significant (Table 6-6). Table 6-4: The mean values and (standard deviations) of extractable Ca, Mg, Fe and Al, all in mg/kg. A * indicates a statistically significant difference between the HA and CH forest types for a horizon at p<0.05. (n=9) HOR. LF H Bhf Bf FOR. TYPE HA CH HA CH HA CH HA CH EXTR Ca (mg/kg) 1853.0 (307.3) 3354.3 (616.5) 1660.9 (762.1) 2720.9 (1027.4) 945.0 (1045.2) 1272.8 (1325.5) 39.4 (11.1) 222.3 (286.1) EXTR Mg (mg/kg) 419.5 (78.6) 372.7 (99.9) 532.9 (117.2) 441.3 (150.5) 334.7 (207.5) 194.6 (169.8) 55.9 (52.8) 40.5 (39.9) EXTR Al (mg/kg) 604.6 (155.4) 274.4 (209.2) 956.5 (274.1) 355.8 (318.9) 1336.9 (301.4) 1043.6 (406.4) 2118.5 (515.4) 1917.7 (222.3) EXTR Fe (mg/kg) 216.8 (73.7) 112.4 (29.6) 309.0 (99.8) 127.3 (88.0) 442.5 (110.5) 347.7 (166.8) 187.3 (117.0) 156.3 (107.2) 94 Table 6-5: The mean values and (standard deviations) of pyrophosphate-extracted Fe, Al and Mn. A * indicates a statistically significant difference between the HA and CH forest types for a horizon at p<0.05. (n-9) HOR. LF H Bhf Bf FOR. TYPE HA CH HA CH HA CH HA CH Fe-PYRO % 0.041 (0.044) 0.010 (0.031) 0.207 (0.219) 0.069 (0.184) 1.528 (0.662) 1.395 (0.934) 1.442 (0.476) 0.910 (0.401) Al-PYRO % 0.089 (0.036) 0.051 (0.040) 0.219 (0.157) 0.081 (0.157) 0.573 (0.149) 0.577 (0.234) 1.559 (0.297) 1.188 (0.364) Mn-PYRO % 0.030 (0.018) 0.063 (0.036) 0.009 (0.012) 0.009 (0.012) 0.004 (0.003) 0.000 (0.000) 0.002 (0.004) 0.000 (0.000) Table 6-6: Analysis of variance table for pyrophosphate-extracted Fe, Al and Mn; citrate-dithionite-bicarbonate (CBD)-extracted Fe, Al and Mn; and acid ammonium oxalate (AAO)-extracted Fe and Al. As a nested experimental design was used, the terms for the ANOVA were TYPE, HORIZON, LOCATION{TYPE} (which is location nested within forest type), TYPE*HORIZON, HORIZON*LOCATION{TYPE}, and multiple R2. Shown here are the probabilities as calculated by the Systat statistical program. A * indicates statistical significance at p<0.05. Fe AAO Fe CBD Fe Pyro Al AAO Al CBD Al Pyro Mn CBD Mn Pyro TYPE 0.206 0.561 0.080 0.645 0.078 0.014 0.000 0.074 HORIZON 0.000 0.001 0.000 0.000 0.000 0.000 0.736 0.001 LOCATION {TYPE} 0.200 0.041 0.154 0.026 0.123 0.552 0.015 0.285 TYPE* HORIZON 0.104 0.238 0.853 0.968 0.203 0.331 0.748 0.074 HORIZON* LOCATION {TYPE} 0.292 0.378 0.080 0.269 0.765 0.831 0.075 0.847 MULT. R2 0.634 0.570 0.850 0.876 0.849 0.909 0.690 0.466 95 There was an increase in pyrophosphate-Fe and -Al with depth. Pyrophosphate-Al was higher in the HA forests in the LF, H and Bf horizons. Pyrophosphate-extracted Mn had high variability, and was highest in the forest floor. There were no significant differences in pyrophosphate-Mn between the CH and HA forests. There were no significant differences between the HA and CH forests in citrate bicarbonate dithionite (CBD)-extracted Fe, Al or Mn (Table 6-7) . The concentrations of Fe-CBD and Al-CBD were higher in the Bf horizon than the Bhf. The Mn-CBD results were highly variable. The percentages of Fe and Al extracted by acid ammonium oxalate (AAO) were also higher in the Bf than the Bhf (Table 6-7). The Bf horizons of the HA sites contained more Fe-AAO than the CH, but the difference was not significant at p<0.05 Table 6-7: The mean values and (standard deviations) for citrate- bicarbonate-dithionite (CBD)-extracted Fe, Al and Mn, and for acid ammonium oxalate (AAO)-extracted Fe and Al. The LF and H horizons were not extracted. A * indicates a statistically significant difference between the HA and CH forest types for a horizon at p<0.05. (n=9) HOR. Bhf Bf FOR. TYPE HA CH HA CH Fe-CBD % 1.802 (0.950) 2.013 (1.296) 3.276 (0.437) 2.636 (1.111) Al-CBD % 0.525 (0.186) 0.477 (0.175) 1.802 (0.442) 1.465 (0.385) Mn-CBD % 0.009 (0.003) 0.001 (0.001) 0.012 (0.013) 0.001 (0.001) Fe-AAO % 1.235 (0.524) 1.283 (0.691) 2.413 (0.342) 1.780 (0.479) Al-AAO % 0.560 (0.203) 0.519 (0.182) 2.272 (0.576) 2.390 (1.097) 96 (Tables 6-6 and 6-7). There were no significant differences between CH and HA soils in Al-AAO. Pyrophosphate, AAO and CBD each extract a different fraction of the soil Fe and Al. Pyrophosphate extracted the Fe and Al associated with organic matter. Acid ammonium oxalate (AAO) extracted both organic Al and Fe, and the amorphous Fe and Al oxides and hydroxides associated with allophane and imogolite. The CBD procedure extracted crystalline, amorphous and organic forms. A measure of the amorphous component was obtained by subtracting the pyrophosphate results from the AAO results. Subtracting the AAO concentrations from those obtained by CBD gave the crystalline Al and Fe. The results of these calculations are shown in Figure 6-1. There were no significant differences between the CH and HA forests in amorphous or crystalline Fe (Fig. 6-1A) or Al (Fig. 6- 1B). There were, however, differences between the Bh and Bhf horizons. Iron associated with organic matter dominated both the Bhf and Bf horizons. There was no amorphous Fe in the Bhf, and relatively equal concentrations of crystalline and amorphous Fe in the Bf. Crystalline Al was not found in the Bhf or Bf horizons of either forest type. Only organic Al was present in the Bhf horizons. The results from the analysis for PA, PT, P0 and the C/P ratio are displayed in Table 6-8. There were no significant differences between the two forest types for any of the horizons (Table 6-9). There was a decrease in PA with depth in the soil profile, from a 97 % Fe A) ! - HA Bhf CH Bhf HA Bf H o r i z o n CH Bf lOrganic EdAmorphous ED Crystall ine % Al B) HA Bhf CH Bhf HA Bf H o r i z o n CH Bf Figure 6-1: The mean percentages of organic, amorphous and crystalline iron (A) and aluminium (B) in the Bhf and Bf horizons of CH and HA forest types. 98 high of 3 8.2 mg/kg in the LF to 4.8 in the Bf. Both P0 and PT were highest in the LF, and decreased with depth. In the LF and H horizons, the C/P was higher in the CH forests than the HA, but the difference was not significant. The C/P generally was higher in the surface horizons, and lower in the mineral soil. The Chang and Jackson fractionations produced PHC1, PNa0H' a n^ PCBD (Table 6-10) . No P was extracted by citrate bicarbonate (PCB) . The PHC1 concentrations were much lower than those for PNaOH and PCBD- There were no significant differences between the CH and HA Table 6-8: The mean values and (standard deviations) of available P (Bray-extracted), total P (Parkinson & Allen digest), organic P (NaOH-EDTA extraction) and C/P ratio. A * indicates a statistically significant difference between the HA and CH forest types for a horizon at p<0.05. (n=9 HOR. LF H Bhf Bf FOR. TYPE HA CH HA CH HA CH HA CH Avail. P (mg/kg) 38.17 (6.19) 30.46 (10.01) 20.55 (6.75) 20.62 (9.65) 10.19 (7.02) 6.06 (2.92) 4.80 (0.89) 6.31 (1.02) Total P (mg/kg) 714.0 (67.45) 585.8 (66.59) 561.4 (98.29) 482.8 (152.1) 524.2 (110.5) 352.1 (89.55) 3 61.0 (105.5) 289.2 (115.6) Org. P (mg/kg) 527.8 (162.0) 492.0 (98.8) 283.8 (133.9) 308.7 (123 .2) 227.8 (110.3) 145.1 (89.21) 99.59 (35.74) 64.78 (36.39) C/P 659 (102) 815 (135) 736 (135) 1055 (389) 384 (67) 392 (152) 256 (82) 252 (90) forests. The PHC1 and PCBD concentrations are highest in the Bf, while the PNaOH is highest in the Bhf. The percentage of PT found as P0 and Px, and the percent recovery of the PT for each horizon and forest type are in Table 6- 11. These are means calculated from each sample, and so are not completely additive. The Px was calculated from the sum of PHCI/ PNaoH an<3 PCBD extracted by the Chang and Jackson fractionation procedure. Inorganic P was not determined in the LF and H horizons, as a precipitate formed in highly organic samples. There was a decrease in P0, and an increase in Px, with soil depth. The Table 6-9: Analysis of variance table for available P (Bray), organic P (NaOH-EDTA extraction), total P (Parkinson and Allen digest), P-HCl, P-NaOH and P-CBD (Chang & Jackson), and the C/P ratio. Because a nested experimental design was used, the terms for the ANOVA were TYPE, HORIZON, LOCATION{TYPE} (which is location nested within forest type), TYPE*HORIZON, HORIZON*LOCATION{TYPE} , and multiple R2. Shown here are the probabilities as calculated by the Systat statistical program. A * indicates statistical significance at p<0.05. Avail P Org P Tot P P-HCl P-NaOH P-CBD C/P TYPE 0.171 0.293 0.001 0.934 0.580 0.012 0.021 HORIZON 0.000 0.000 0.000 0.000 0.000 0.000 0.000 LOCATION {TYPE} 0.028 0.763 0.313 0.012 0.000 0.814 0.931 TYPE* HORIZON 0.242 0.641 0.505 0.762 0.116 0.339 0.056 HORIZON* LOCATION {TYPE} 0.198 0.564 0.637 0.013 0.000 0.589 0.891 MULT. R2 0.872 0.779 0.727 0.688 0.858 0.823 0.808 100 Table 6-10: The mean values and (standard deviations) of P extracted by HCl, NaOH and citrate bicarbonate dithionite (CBD), during the Chang & Jackson fractionation procedure. The LF and H horizons were not extracted. A * indicates a statistically significant difference between the HA and CH forest types for a horizon at p<0.05. (n=9) HOR. Bhf Bf FOR. TYPE HA CH HA CH P-HC1 (mg/kg) 7 (0.5) 7 (0.5) 16 (7.7) 19 (13.6) P-NaOH (mg/kg) 44 (15.4) 37 (28.0) 23 (8.4) 25 (12.4) P-CBD (mg/kg) 30 (9.8) 22 (17.0) 96 (15.9) 74 (23.6) Table 6-11: The means and (standard deviations) of total P found as organic and inorganic P, and the percentage of total P recovered. Organic P was determined with NaOH-EDTA extraction and total P by Parkinson & Allen digest. Inorganic P is the sum of the fractions determine by Chang & Jackson fractionation. N/A indicates not analyzed. (n=9) HOR. LF H Bhf Bf FOR. TYPE HA CH HA CH HA CH HA CH ORG. P 73.4 (20.0) 83.3 (12.7) 51.4 (23.3) 65.5 (28.3) 42.7 (17.4) 39.1 (17.9) 27.5 (3.4) 23.9 (12.0) INORG. P N/A N/A N/A N/A 17 (6.1) 19 (10.7) 39 (7.8) 44 (15.6) % RECOVERY 73.4 (20.0) 83.3 (12.7) 51.4 (23.2) 65.5 (28.3) 59 (15.7) 58 (22.1) 67 (8.7) 68 (17.3) 101 percentage recovery ranged from 51.4% to 83.3%, but the variability was high. Table 6-12 displays the correlation matrix. Extractable Al, and Al extracted by CBD, AAO and pyrophosphate all had high positive correlations with one another. Acid ammonium oxalate (AAO)-Al was also positively correlated with PHC1 and pH in water and CaCl2, and was negatively correlated with extractable Fe. Aluminium extracted by CBD was positively correlated with PCBD and both pH methods. There was a positive relationship between Fe extracted with CBD and AAO, and between extractable Al and PCBD. Extractable Fe had a negative correlation with pH determined by either method. Loss on ignition was positively correlated with extractable Mg, C, total N and available P. There was a positive relationship between the values for the two methods for pH. Positive correlations were also seen with extractable Mg to Ca, and with Mn extracted by pyrophosphate to CBD. Total N correlated positively with C, extractable Mg, and P0. There was a negative relationship between C and pH in CaCl2, and a positive one with PT to P0. The 31P NMR spectra for the two HA soil profiles are in Figure 6-2 (A and B); those for the CH profiles are in Figure 6-3 (A and B). The percentage of P found within each class of compounds, calculated from the spectra by integration, is in Table 6-13. A guide for the interpretation of NMR spectra can be found in Chapter 5 (Table 5-2) . 102 Table 6-12: Correlation matrix. Al- and Fe-AAO were extracted by acid ammonium oxalate, A1-, Fe-, and Mn-CBD by citrate bicarbonate dithionite. A1-, Fe-, and Mn-pyro by pyrophosphate. Avail P was extracted by Bray. Extr-Al, -Ca, -Fe, and -Mg were extracted by Mehlich. H2 0A is air dry moisture content; H20F is field moisture content. Org P is by NaOH-EDTA extraction and Tot P is by Parkinson & Allen digest. i? I f I c3 L H UJ I a. I i s I cc & o rtdooddoooooooo BS^SSBBSS^SM - J d o d d d d o d o d o d o t p " "~ § o •* N vo o •* o N W tn £ N N N ~; z H d o d d o d d d o o d o o o D rtDododddddooDOOoo d. § o N N N •£ vo m fc OD •-> N o N <JJ q; m y S 8s8SR2?S3!?ieSSSI8SB | ^ o o o o o o o o o o o o o o o o o ! N S 3 C5 N •* •* vfi 8 o $ tn f\i (n -i it) • o £ • H O O O O O O O O ?°°??T-T o o o o o ~6ao6666ooaooa66o6a6 §sifeisai§s§ss§888sass 8SB$8Si3B28S3$IBiBS58SB -ddddddddddddddddddddd SB§883gSg§sg£SS§g§BBgg8 - o d o d d d o ^ d d d d d d d i p o o o o d d B85Sga2SiSSSBS§85BB«Sa»5 §BI8is?SSi5SBB5f2SI!§SfisSS^ -dddddddddddddddddddddddd IBiBBsBBBiBsBSBSSBSSIglsgSill Sddddddddddddddddddddddddd SBSBBSMSfiSBSisSiSSSBSBSBB i D. a (M J §3 • • m -?? § — roin - J d d d • H o d d d m •loaaaa • • • • • • • H o a o o a o !*8BS*gB l o a o a o a a O O D O O O O a O O O O O Q O O D O O O O O D ?? BsBSS^BBB ^oaaaaada zmmm ~6666aa6aa 3 §B3J8SS2$i$ - o d d o o o d o d d 5 BISSSIilSiBSSS x  ~6o66aaaaa66 immmm* ^~66dd6aaaa66a o a o r p. — • .• . ft n •? •? 1 T > x y J( x « ( £ UJ UJ UJ UJ U. Li. CE t - i X a : L a i n 1 0 3 HEMLOCK-AMABAUS FIR OLD GROWTH A HEMLOCK-AMABAUS FIR OLD GROWTH I If HORIZON / 8 HORIZON Bbf HORIZON /A^Ayv^^ Bf HORIZON ORTMO M O N O - J l POLY Figure 6-2: 31P NMR spectra for two soil profiles from mature HA sites, extracted with NaOH-EDTA. Phos is phosphonate, ortho is orthophosphate, mono is monoester phosphate, diest is diester phosphate, pyro is pyrophosphate and poly is polyphosphate. 104 CEDAR-HEMLOCK OLD GROWTH A IF HORIZON M HORIZON A Bhf HORIZON i r » PHOS ORTHOMOHO J * ^ pYRO - i f -M POLY CEDAR-HEMLOCK OLD GROWTH B U HORIZON PHOS ORTHO MONO p j ^ - PYRO POLY Figure 6-3: 31P NMR spectra for two soil profiles from mature CH sites, extracted with NaOH-EDTA. Phos is phosphonate, ortho is orthophosphate, mono is monoester phosphate, diest is diester phosphate, pyro is pyrophosphate and poly is polyphosphate. 105 Table 6-13: The percentage of total P in solution found within each P form revealed by 31P MMR spectroscopy. Phos is phosphonate, orth is orthophosphate, mono is monoester phosphate, dies is diester phosphate, pyro is pyrophosphate and poly is polyphosphate. For. Type HA A HA B CH A CH B Hor. LF H Bhf Bf LF H Bhf Bf LF H Bhf Bf LF H Bhf Bf Phos 3 0 4 0 0 9 0 0 0 5 4 0 0 0 0 0 Orth 10 20 17 60 33 30 22 31 17 7 38 23 31 39 35 29 Mono 24 20 29 25 45 30 54 51 33 38 36 61 38 32 34 32 Dies 39 44 50 10 17 26 15 18 39 36 17 16 18 20 24 24 Pyro 8 0 0 5 5 5 0 0 0 5 5 0 13 9 7 15 Poly 16 16 0 0 0 0 0 0 11 7 0 0 0 0 0 0 In HA profile A (Fig. 6-2A; Table 6-13), the predominant P class in the LF, H and Bhf horizons was diester, and orthophosphate dominated in the Bf. The percentages of orthophosphate and monoester P were similar in the LF, H, and Bhf. There were small pyrophosphate peaks in the LF and Bf horizons, and very small phosphonate peaks in the LF and Bhf. Sharp polyphosphate peaks appeared in the LF and H. The orthophosphate and monoester peaks 106 were separated completely in only the Bf horizon. One problem with the extractant used in this study for 31P NMR spectroscopy is that the orthophosphate and monoester peaks overlap in some spectra. As a valley between peaks occurs at 6 ppm when there is clear separation, this was chosen as the dividing line where overlapping occurred. This is discussed in more detail in Chapter 5. Profile B for the HA (Fig. 6-2B, Table 6-13) was quite different from profile A. The orthophosphate and monoester peaks were distinct in all horizons. Monoester P dominated in all but the H horizon, where it was proportionally comparable to orthophosphate. Small pyrophosphate peaks occurred in the LF and H, and very small phosphonate peaks were found in the H and Bhf. There was no detectable polyphosphate in this profile. The two CH soil profiles (Figs. 6-3A and 6-3B) were also quite different from one another. The orthophosphate and monoester peaks were well separated in only the Bf horizon in profile A, but were distinct in all horizons of profile B. In A, (Fig. 6-3A; Table 6- 13), the percentages of monoester and diester P were comparable in the LF and H, and were nearly double that of orthophosphate P in these horizons. Orthophosphate and monoester were the dominant P classes in the Bhf; monoesters dominated in the Bf. There were small phosphonate and pyrophosphate peaks in the H and Bhf horizons, and polyphosphate peaks in the LF and H horizons. In CH profile B (Fig. 6-3B; Table 6-13), most of the P was found as orthophosphate or as monoesters, at comparable percentages. 107 Diester and pyrophosphate peaks also occurred, but there was no phosphonate or polyphosphate in any horizon. Discussion The pH results are comparable to other measurements in CH and HA forests (eg: de Montigny 1992; Keenan 1993; Prescott et al. 1994). The pH was significantly higher in the LF and H horizons of the CH forests than in the HA. Higher pH in soils under cedar than under other conifers has been noted in the literature (Stone 197 5; Keenan 1993), and is thought to be due to the nature of cedar foliage. The high negative correlation of C to pH in CaCl2 indicates that the acidity of these soils is due to organic matter. Schnitzer (1977) has suggested that much of the total acidity of cool, temperate acid soils is due to oxygen-containing functional groups associated with organic matter, and 13C NMR spectra show that COOH groups are present in the non-woody organic horizons of the CH and HA forests (deMontigny et al. 1993). The increased pH in the Bf horizons is typical of podzols in this region (Lewis 1976) and reflects in part the lower organic matter content of these horizons. Carbon and loss on ignition (LOI) both decrease with depth in the soil profile. There is a high positive correlation between LOI and C. Loss on ignition is a measure of soil organic matter content, although it may include C from carbonates, water and hydroxyl groups from clays, and other volatiles in the soil (Kalra 108 and Maynard 1991). The C values are comparable to those obtained by de Montigny (1992), Keenan (1993) and Prescott et al. (1994). The increased C in the LF and H horizons of CH forests supports the theory that decomposition is less effective on these sites (de Montigny et al. 1993; Prescott et al. 1995). Carbon-13 NMR spectroscopy revealed a higher ratio of carbohydrate to lignin C in CH samples, and higher levels of lipids and total and labile polysaccharides (de Montigny 1992; de Montigny et al. 1993) . The change in the Bhf and Bf horizons, with the C content significantly higher in the HA forests, suggests increased mixing of organic and mineral material. The HA forests are known to maintain a higher abundance and biomass of soil fauna than do CH forests (Battigelli et al. 1994), although the abundance of fauna which would mix organic and mineral soils, such as earthworms, is low in both forest types. This higher C concentration may also be due to increased illuviation of organic compounds through the soil profile. The high positive correlation of total N to C and to LOI indicates that much of the nitrogen on these sites is associated with organic matter. In general, there is more total N in the HA for all horizons, but the differences were not significant. These results agree with those found by others in the SCHIRP project (Germain 1985; de Montigny 1992; Keenan 1993; Prescott et al. 1993; Prescott et al. 1995). Cedar had consistently lower foliar N concentrations and a significantly higher rate of N resorption than 109 western hemlock or amabilis fir (Keenan 1993) which in turn was expressed as lower N concentrations in the soil organic matter of CH forests. The C/N ratio was wider in the CH than HA forests for all horizons, but the difference was not significant. Keenan (1993) also reported a drop in the C/N ratio in the Bhf horizon, relative to the other horizons. The percentage of carbon dropped by nearly half in the Bhf horizons of both CH and HA forests, but the change in nitrogen is not nearly as precipitous. The C levels dropped again in the Bf, but the N concentrations do also, which raises the C/N ratio. Jenkinson (1988) attributes the drop in the C/N ratio when moving from organic to mineral horizons to the presence of fixed NH4 in clays. There appears to be more NH4 fixation at the interface between the organic and mineral layers in the podzols of the SCHIRP site than lower in the profile at the Bf horizon. Extractable Ca and Mg were measured by a different extraction procedure from the ammonium acetate procedure used by de Montigny (1992) and Keenan (1993). de Montigny (1992) found significantly more Ca in the F horizon of CH than HA, but no significant differences for Mg. Keenan (1993) found increased Ca in the mineral horizons of some, but not all, of the CH sites which he examined, and the differences were not statistically significant. He did not measure Mg. Cedar is known to be a Ca accumulator species (Krajina 1969; Ballard and Carter 1986) relative to associated tree species. Consequently, the higher Ca on the CH 110 sites is due to the incorporation of cedar foliage into the soil organic matter. The high positive correlations of extractable Mg to LOI and total N suggest that it is also a component of soil organic matter. Iron and Al were quantified thoroughly because of their known role in P retention and cycling in podzolic soils (Sanborn 1987; Yuan and Lavkulich 1994). The only Fe and Al results reported by SCHIRP researchers were those of Keenan (1993), who found significantly higher pyrophosphate Fe in the Bhf of HA soils than CH, and no significant differences in pyrophosphate Al. His reported values are comparable to those found in this study. The predominance of organic-associated Fe and Al reflects the characteristic illuviation of organic matter and organo-metallic complexes in podzolic soils (Oades 1989). The presence of crystalline Fe but not Al is typical of podzols on Vancouver Island (Lewis 1976; Sanborn 1987). The amorphous Fe and Al in the Bf horizons reflects the volcanic parent material of this region. The positive correlation of AAO-Al to pH and C indicates that the partitioning of Al between organically complexed and allophanic forms is controlled by pH and soil organic matter content, an observation also made by Sanborn (1987) . The higher organic Al in the HA Bf than in the CH Bf may reflect a number of factors: tree species differences in Al content; differences in the organic ligands transporting the Al to the Bf horizon (Keenan 1993); or faunal activity, which is higher in the HA forests, and which may 111 alter the illuviation patterns (Sanborn 1987). The finding of Keenan (1993) of significantly more pyrophosphate-Fe on the HA sites, but no difference in the Al, whereas this study found the reverse is probably due to the high spatial variability in these forests. The correlation of all of the Al extraction procedures with one another suggests that these methods are less selective for particular Al fractions than they are for Fe fractions (Sanborn 1987), and also reflects the more limited number of Al forms relative to Fe forms in these soils. The Mn results were highly variable, as noted by de Montigny (1992), and were highest in the LF and H horizons. There were no significant differences between the CH and HA forest types for any of the P measurements. The HA forests were generally higher in available P (PA), total P (PT) and organic P (P0) . The PA results are comparable to those of de Montigny (1992), but are much higher than those of Keenan (1993) for the Bhf, and Prescott et al. (1995) for the LF. This probably reflects methodological differences, such as length of extraction. Prescott et al. (1995) found significantly lower concentrations of PA in the L and H layers of CH forests; de Montigny (1992) in the H of CH forests. The PT results are comparable to those of Keenan (1993) and Prescott et al. (1995), although Prescott et al. (1995) found significantly more PT in the L horizon of the HA forests. Organic P was not measured by other SCHIRP researchers. 112 The lack of significant differences in any kind of phosphorus between the CH and HA forests for any horizon suggests that the two forest types are not inherently different in P contents and cycling. However, differences may be masked by the high variability in all measured P concentrations, and as shown by the 31P NMR spectroscopy. This variability might have been reduced by more intensive sampling. The cycling of P in these forests occurs mainly within the internal (biological) cycle (Dighton and Boddy 1988). The external (geological) cycle is only of any importance in the Bf horizon. In the LF, virtually all of the soil phosphorus is found as P0. The PA is highest in this horizon, indicating a high level of labile, easily leached P. This horizon is most influenced by the P forms and contents of litterfall. The HA has slightly higher concentrations of P0 and PT, which may be explained by the greater resorption of P by cedar than by western hemlock or amabilis fir (Keenan 1993). The diversity of P forms, as shown by 31P-NMR also reflects the influence of litter (Compton 1994). Diester P in plant tissues occurs as phospholipids in cell walls, and as nucleic acids. Plant monoesters consist of inositol phosphates, mononucleotides and sugar phosphates, which are metabolic intermediaries (Bieleski 1973). ADP and ATP are involved in energy transfer, but are not seen in 31P-NMR spectra of extracted soil samples, probably because they are found in such low levels in these soils. It is unlikely that they were destroyed during the 113 extraction process, because added ATP was readily recovered during the extraction trial (Chapter 6). Some higher plants may store excess P as polyphosphates (Bieleski 1973). These compounds are also found in soil organisms, which also produce P compounds not found in plants. Pyrophosphate is believed to be involved in biological P cycling in the soil, and may be present as an organic ester which is hydrolysed during extraction for NMR analysis (Condron et al. 1985). Fungi and other microbes have been shown to store excess P in polyphosphates (Bieleski 1973), while ectomycorrhizal fungi are thought to transport P within their hyphae as polyphosphate (MacFall et al. 1992). The polyphosphates of the LF and H horizons are most likely from the mycorrhizae associated with trees on these sites. Phosphonates are formed by a variety of soil microbes (Hilderbrand 1983), and are thought to accumulate under acid conditions, where bacteria containing phosphonatase enzymes are low in number (Hawkes et al. 1984). Thus, they are characteristic of cool moist acidic soils, such as those in CH and HA forests. The presence of pyrophosphate is also thought to be a marker for restricted biological decomposition (Preston et al. 1986) . Acidic conditions generally suppress bacteria and actinomycetes, so that decomposition is primarily fungal (Harris 1988). Fungal mats are common in the F layers of these forests (de Montigny 1992; personal observation). Fungal decomposition is slower than bacterial (Dighton and Boddy 1988), and may be responsible for the wide 114 variety of P forms seen in the LF and H horizons of these sites. The relatively high C/N and C/P ratios are also indicative of slower decomposition. Although John et al. (1965) did not find a direct relationship between C and P0 when examining P0 in a range of BC soils (agricultural), C is correlated with P0 in these forests, which also suggests reduced decomposition. The lower PA concentrations in the H horizon, relative to the LF, are due to increased leaching of labile P compounds from the H horizon. The increased humification of this horizon relative to the LF is shown by the lower P0 concentration and the reduced percent recovery. Differences in the methods used to determine P0 and PT also contribute to the lower recovery rate: the higher temperature of the PT digest releases P from aromatic groups and humic substances, which would not occur in the lower temperature P0 extraction (Kristensen 1990) . In the H horizon, the P forms shown by 31P-NMR spectroscopy are similar to those seen in the LF. There was more phosphonate and pyrophosphate in this horizon in some of the profiles, showing that P in this horizon is more influenced by soil microbes, and less by the kind of litterfall. The diesters in this horizon are probably lipids, which de Montigny (1992) also found in the forest floor of these forests using 13C-NMR spectroscopy. Lipids are usually considered to be quite labile in soil (Stevenson 1986); their accumulation is indicative of slow decomposition, as is the high C/P ratio. Decomposition is usually lower in the H than the LF, as the more labile materials are no 115 longer available. However, more nutrients are mineralized per C atom in the H than in the LF (Hart et al. 1994) . In the Bhf horizon, the external P cycle begins to influence P forms. The Chang and Jackson fractionation shows the presence of inorganic P forms: PNaoH' PCBD an<3 PHCI • PNaoH i-s thought to be the non-occluded phosphate bound to the surfaces of Al or Fe hydrous oxides (Olsen and Sommers 1982) . The PCBD fraction is comprised of P occluded within the matrices of Fe and Al oxides and hydrous oxides, while the PHC1 is thought to be the extracted calcium phosphates of the non-occluded apatite fraction (Williams et al. 1980; Olsen and Sommers 1982) . In the Bhf horizon of these forests, PNa0H is higher than PCBD and PHCi/ suggesting that Px is non- occluded, most likely with organically associated Fe, as suggested by the positive correlation. The diversity of P forms is reduced relative to the forest floor, with orthophosphate and monoester phosphate predominating. The accumulation of monoesters in the Bhf and Bf is due to the adsorption of inositol phosphate onto surface hydroxyls of soil colloids by ligand exchange mechanisms (Ognalaga et al. 1994), which hampers their biodegradation. The diesters and other monoesters do not appear to adsorb to soil colloids, and thus are lost by leaching and degradation. Some diester phosphates are retained in the Bhf and Bf of the CH and HA forests, perhaps by Al or Fe bridges to humic substances (Gerke 1992). Pyrophosphate is also present in some Bhf and Bf horizons. The high affinity of pyrophosphate for organically-bound Fe and Al has led to its use as 116 an extractant. Therefore, it is possible that, in these horizons, pyrophosphate is linked to Al- and/or Fe-organic matter complexes. In the Bf horizon, PCBD is the largest inorganic P fraction, and it correlates strongly to amorphous Al and Fe. This suggests that much of the P in this horizon is occluded, and is sequestered in amorphous sesquioxides. This is typical of the podzolic Bf horizon (Sanborn 1987; Yuan and Lavkulich 1994). It would appear from the increased PHC1 fractions in the Bf relative to the Bhf that apatite is present in this horizon. This is somewhat unusual in that apatite should weather under acidic conditions. Lindsay (1979) reports that brushite (CaHP04.2H20) and monetite (CaHP04) are readily formed under acid conditions. Thus the PHC1 fraction attributed by most authors to apatite may be either brushite or monetite. In addition, the strong correlation of PHC1 to AAO- extracted Al suggests that P sorbed to amorphous Al may have been extracted instead of, or in addition to, calcium phosphate. Orthophosphate and monoester phosphate are the main P forms revealed by 31P-NMR spectroscopy. Fares et al. (1974) have suggested that P0 does not actually occur in B horizons, but that the P0 determined by chemical procedures is instead inorganic P bridged to humic substances. The results from 31P-NMR spectroscopy show that this is likely for some of the P, as orthophosphate is the predominant P form in the Bf, but organic P compounds are also present in these horizons. The 31P-NMR results reveal that P forms are quite specific to 117 a particular microsite, as no obvious trends for the CH or HA forest types were observed. In retrospect, better results might have been obtained by using composite samples. Although a different extractant was used in this study (see Chapter 6), the results are comparable to others examining P forms in forest soils with 31P-NMR spectroscopy. The low pH of forest soils has been shown to favour a wider variety of P species than are seen in agricultural soils (Preston et al. 1986), with the persistence and accumulation of relatively labile compounds such as diesters (Tate and Newman 1982; Condron et al. 1990a; Gil-Sotres et al. 1990). Conclusions Although this study sampled only Orthic Ferro-Humic Podzols, the results from the general chemical analyses are very similar to those obtained by other SCHIRP researchers, who did not restrict their sampling to only Orthic Ferro-Humic Podzols, sampling all of the soil types present on these sites. No significant differences in phosphorus forms or concentrations were found between the CH and HA forest types, suggesting that P-related nutrient cycling problems after logging and slash burning are not due to inherent site differences in soil P. However, the high variability in all measured P forms may have masked statistical differences in P between the forest types. More intensive sampling may be required to reveal differences P on these sites, or these soils may not be inherently different in soil P concentrations or levels. 118 The diversity of P forms as revealed by 31P-NMR spectroscopy was typical of cool, moist acidic forests, as was the persistence of the very labile diester phosphates throughout the soil profile, albeit at very low levels in the mineral horizons. This, together with the relatively high C/P ratio, indicates that decomposition is slow. Most of the P in the LF is in organic forms typical of litterfall. In the H horizon there is more humification, and P forms associated with soil organisms are seen. In the Bhf, inorganic phosphorus is predominantly non-occluded. Organic P is present, mainly as monoester phosphates which are probably adsorbed on soil colloids. In the Bf, most of the P is occluded in amorphous sesquioxides, and there are low levels of organic P, mainly as monoester phosphates. 119 CHAPTER SEVEN A COMPARISON OF PHOSPHORUS FORMS ON CH SITES AFTER BURNING Introduction The lack of significant differences in phosphorus forms or concentrations between the CH and HA forest types (Chapter 6) suggests that the P-related growth check observed in trees on the CH sites five to eight years after cutting and replanting is not due to inherent differences in P between the CH and HA forest types. To achieve more rapid regeneration on the CH and HA sites and to allow planter access, slash-burning is used to reduce slash accumulations and to control the heavy cover of the ericaceous shrub salal, especially in the CH forests (Prescott and Weetman 1994) . Burning is known to affect forest soils, and the general effects of fire have been summarized in several reviews (Ahlgren and Ahlgren 1960; Kozlowski and Ahlgren 1974; Feller 1982). During a fire, the nutrients incorporated in vegetation, litter and soil can potentially be volatilized during pyrolysis or combustion, mineralized during oxidation, or lost by ash convection (Grier 1975). After a fire is out, nutrients may be redistributed by wind and water erosion or by leaching of the ash layer and soil. Some frequently observed nutrient effects include: increases in soil pH; increases in the availability of P, Ca and Mg; and decreases in total N and S (Ahlgren and Ahlgren 1960; Kozlowski and Ahlgren 1974; Feller 1982; Ellis and Graley 1983; Feller et al. 1983; 120 Khanna and Raison 1986; Macadam 1987; Tomkins et al. 1991; Brockley et al. 1992; Mangas et al. 1992; Rice 1993; Romanya et al. 1994). These nutrient changes generally occur in the forest floor and surface soil; the changes at depth in the soil profile are less and occur more slowly as nutrients are leached down (Feller 1982; Brockley et al. 1992). The magnitude of these effects will depend on fire severity, site and soil characteristics, and fire intensity and duration (Brockley et al. 1992). In forests, P is tightly conserved. The P cycle is virtually closed, with most plant P recycled by microbial breakdown of litter and organic matter. Fire is one of the few sources of P loss in forests, during prescribed burning or wildfires. Many researchers have reported large increases in available P in surface horizons immediately after fire, but these are short term increases that can produce long term losses which may reduce forest productivity (DeBano and Klopatek 1988; Saa et al. 1993; Romanya et al. 1994). Most studies of burning have investigated changes in available P; little is known about the effects of fire on other P forms, or the changes in P levels and forms which may occur with time after burning. The objective of the research in this chapter was to compare the soil phosphorus and related soil chemistry of cut CH stands 10 years, 5 years and immediately after burning to old growth CH stands. Concentrations of total, organic and available P were determined by extraction and digestion procedures, while 121 fractionation and 31P nuclear magnetic resonance (NMR) spectroscopy were used to characterize the P forms. Some aspects of the chemistry of these soils which could influence P forms and levels were also examined. Materials and Methods Sample Collection The study sites and sampling design were described in Chapter 3, and the procedures for sample collection and preparation were described in Chapter 6. The CH old growth (OG) samples used in this chapter are the same ones which were used in Chapter 6. The 0-year sites were sampled within one month of burning, with little rainfall between the times of burning and sampling. The 5-year and 10-year sites were sampled 5 and 10 years postburn, respectively. The LF, H, Bhf and Bf horizons were collected from each pit, except from the 0-year sites, where an ash layer was collected for the LF horizon. Three locations per age were sampled, with 3 pits per location, for a total of 9 samples per horizon per age. The samples were collected in late July and early August, 1992. General Chemical Analysis The analytical procedures used were described in Chapter 6. P Methodology The analytical procedures used were described in Chapter 6. A comparison of the methods used in this chapter to other soil P methods is found in Chapter 4. 122 NMR Spectroscopy Two soil profiles for each of the old growth and 0-, 5- and 10-year postburn sites were chosen for analysis by 31P NMR spectroscopy. The criteria for sample selection are listed in Chapter 6. The extraction, sample preparation and analytical procedures for 31P NMR spectroscopy are found in Chapter 5. Statistical Analysis Statistical analyses were conducted using the Systat program (Wilkinson 1990) to perform analysis of variance test at p<0.05, using a nested design (Hicks 1982), with location nested within each forest type, as was described in Chapter 4. The model statement used was: VARIABLE = CONSTANT + AGE + HORIZON + LOCATION{AGE} + TYPE*HORIZON + HORIZON*LOCATION{AGE}. Pearson pairwise correlations and Tukey's HSD tests were also conducted with the Systat program. Homogeneity of variance was tested by plotting residuals against estimates. Log and log (n+1) transformations were performed where necessary. Results The results from the analysis for field moisture, air-dry moisture, pH in water and pH in CaCl2 are shown in Table 7-1. Table 7-2 displays the analysis of variance results for these data. The field moisture content was significantly lower in the LF of all of the postburn sites relative to the old growth sites. As noted in Chapter 6, the field moisture content was relatively uniform in 123 Table 7-1: Mean values and (standard deviations) for field moisture content, air-dry moisture content and pH in water and CaCl2 • Different letters indicate statistical significance among the ages for a horizon at p<0.05. OG is old growth (n=9] HOR. LF H Bhf Bf AGE OG 0-YR 5-YR 10-YR OG 0-YR 5-YR 10-YR OG 0-YR 5-YR 10-YR OG 0-YR 5-YR 10-YR FIELD MOISTURE % 203 (122) a 67 (45) b 106 (132) b 80 (94) b 314 (149) a 275 (85) a 249 (87) a 188 (116) a 249 (152) a 276 (93) a 183 (99) a 207 (211) a 84 (27) a 86 (35) a 120 (99) a 53 (10) a AIR DRY MOISTURE % 12.95 (1.18) a 12.57 (2.58) a 12.95 (0.77) a 11.97 (1.24) a 16.17 (6.06) a 24.17 (12.46) a 17. 04 (6.30) a 17.62 (10.29) a 8.70 (4.69) a 9.25 (5.03) a 10.90 (4.06) a 8.64 (5.05) a 10.44 (10.19) a 15.56 (7.09) a 16.35 (8.78) a 9.39 (3.96) a pH IN WATER 3.90 (0.22) b 4.96 (0.45) a 4.27 (0.32) b 4.16 (0.42) b 3.59 (0.13) b 4.36 (0.61) a 3. 96 (0.32) a 3.90 (0.36) ab 3 .75 (0.23) a 4.23 (0.49) a 4.09 (0.24) a 3.79 (0.24) a 4.45 (0.32) a 4.81 (0.30) ab 4.93 (0.20) ab 4. 94 (0.21) b pH IN CaCl2 3.60 (0.27) b 4.58 (0.40) a 3.81 (0.35) b 3.66 (0.39) b 3.12 (0.19) b 3.76 (0.65) a 3.32 (0.31) ab 3.41 (0.23) ab 3.23 (0.17) a 3 .67 (0.41) a 3.36 (0.30) a 3.26 (0.16) a 3.96 (0.29) a 4.31 (0.23) a 4.28 (0.35) a 4 .61 (0.21) a 124 Table 7-2: Analysis of variance (ANOVA) table for field moisture, air-dry moisture, pH in water, pH in CaCl2, total C, total N, C/N ratio, loss on ignition (LOI), and extractable Ca, Mg, Fe and Al, showing the probabilities as calculated by the Systat statistical program. As a nested experimental design was used, the terms for the ANOVA were AGE, HORIZON, LOCATION{AGE},which is location nested within forest type, AGE*HORIZON, HORIZON*LOCATION{AGE}, and multiple R2. A * indicates statistical significance at p<0.05. Air-Dry Moisture Field Moisture pH in Water pH in CaCl2 C N C/N LOI Extr. Ca Extr. Mg Extr. Al Extr. Fe AGE 0.015 0.000 0. 000 0. 000 0.034 0. 023 0. 075 0.000 0.010 0.000 0.000 0.238 HORIZON 0.000 0.000 0.000 0.000 0.000 0.000 0.010 0.000 0.000 0.000 0.000 0.000 LOCATION {AGE} 0.349 0.000 0.000 0 .000 0.218 0.000 0.073 0.003 0.001 0.000 0.338 0.864 AGE* HORIZON 0.152 0.012* 0 . 002* 0.000* 0.192 0.696 0.511 0.055 0.097 0.007* 0.652 0 .400 HORIZON* LOCATION {AGE} 0.645 0.007* 0.977 0.959 0.362 0.145 0.893 0.119 0.324 0.001* 0.987 0 .411 MULT. R2 0.513 0 .781 0 .769 0.805 0.905 0.879 0.376 0.924 0.827 0.884 0.482 0.523 the H and Bhf horizons, but dropped in the Bf. There were no significant differences in moisture content among the sites in these horizons. The air-dry moisture content was relatively uniform for all horizons, with no significant differences among the sites. 125 There were significant differences in pH among the sites, with similar trends when pH was measured in water or CaCl2. At the 0- year sites, the pH of the LF was significantly higher than that of the other ages. The pH was also higher in the LF of the 5- and 10- year postburn sites than in the old growth samples, but the difference was not significant. This same pattern was seen in the H horizon, with increased pH values in all postburn samples relative to the old growth, but with only the values of the 0-year samples significantly different. There were no significant differences in pH in the Bhf, but the values were still higher in the postburn soils than in those from the old growth. In the Bf, the pattern changed. The highest pH values were found 10 years after burning, and these were significantly different from those of the old growth when pH was measured in water. There were no significant differences among the ages in any horizon for total carbon, total nitrogen or the C/N ratio (Tables 7-2, 7-3) . Generally, the C content was lower in the 0-year LF than in the LF of the other ages. The C concentration was lower in all of the postburn samples than in the old growth samples in the LF and H horizons, but was higher in the postburn samples in the Bhf. In the Bf, the lowest C concentration was found in the 10- year postburn sites. Total N concentration was lowest on the 0- year sites in the LF horizon, but for the H, Bhf and Bf horizons it was lowest in the 10-year postburn samples. The C/N ratio was widest in the old growth stand in the LF and H horizons. In the Bhf and Bf horizons, it was widest on the 10-year postburn sites. 126 Table 7-3: The mean values and (standard deviations) for total carbon (%) , total nitrogen (%), loss on ignition (LOI) and the C/N ratio. Different letters indicate statistically significant differences among the ages for a horizon at p<0.05. OG is old growth. (n=9) HOR LF H Bhf Bf AGE OG 0-YR 5-YR 10-YR OG 0-YR 5-YR 10-YR OG 0-YR 5-YR 10-YR OG 0-YR 5-YR 10-YR C % 48.05 (4.31 35.50 (11.48 43.85 (5.49 39.05 (8.30 47.32 (6.20 41.55 (9.70 40.98 (13.90 41.46 (4.69 14.32 (4.74 15.07 (7.01 19.41 (13.43] 16.67 (6.31] 6.74 (1.21] 6.19 (2.71 5.14 (2.56] 4.08 (2.15] ) a a a a a a a a a a a a a a a a TOTAL 0.92 (0.11 0.79 (0.17 0.96 (0.15 0.82 (0.20 0.91 (0.16 0.93 (0.14 0.90 (0.22 0.84 (0.21 0.54 (0.20 0.51 (0.35 0.57 (0.22 0 .41 (0.17 0.17 (0.04 0.14 (0.07 0.13 (0.08 0.08 (0.04 N % ) a ) a a a a a a a a a a a a a a a C/N 52.56 (6.20 46.34 (16.28 46.38 (7.69 48.46 (5.49 53.40 (11.59 45.95 (15.11 47.16 (19.44 51.00 (10.05 29.78 (9.55 34.17 (9.53 34.87 (21.40 53.29 (50.78 40.99 (8.25 44.19 (10.47 40.62 (10.30 53.69 (22.37 ) a a a a a a a a a a a a a a a a LOI 2001 (612 694 (588 1381 (567 774 (597 2014 (780 1512 (1845 1549 (851 823 (540 88 (65 88 (79 114 (79 99 (75 22 (5 21 (9 18 (9 13 (3 ) a a a a a a a a a a a a a a a a 127 Table 7-4: The mean values and (standard deviations) for extractable Ca, Mg, Fe and Al, all in mg/kg. Different letters indicate statistically significant differences among the ages : HOR. LF H Bhf Bf AGE OG 0-YR 5-YR 10-YR OG 0-YR 5-YR 10-YR OG 0-YR 5-YR 10-YR OG 0-YR 5-YR 10-YR Eor a horizon EXTR Ca (mg/kg) 3354.3 (616.5) a 4366.4 (2145.5) a 4631.8 (1342.8) a 3331.7 (1746.1) a 2720.9 (1027.4) a 2814.7 (1448.3) a 2550.0 (934.6) a 2189.5 (1236.9) a 1272.8 (1325.5) a 1303.6 (1243.8) a 847.7 (1253.4) a 348.2 (377.9) a 222.3 (286.1) a 159.1 (87.0) a 88.5 (132.0) a 57.4 (64.5) a at p<0.05. OG is old growth. (n=9) EXTR Mg (mg/kg) 372.7 (99.9) a 662.4 (64.5) b 586.7 (70.2) b 574.4 (101.1) b 441.3 (150.5) a 652 .8 (85.9) b 552 .8 (12 9.2) ab 599.4 (137.2) b 194.6 (169.8) a 301.2 (188.5) a 266.7 (171.3) a 242.6 (308.0) a 40.5 (39.9) a 50.2 (28.8) a 26.4 (23.3) a 16.0 (5.3) a EXTR Al (mg/kg) 274.4 (209.2) a 621.0 (318.7) a 455.9 (274.8) a 690.9 (439.8) a 355.8 (318.9) a 742 .1 (265.4) a 696.4 (473.4) a 813 .2 (358.4) a 1043.6 (406.4) a 1198.6 (306.7) a 1162.8 (455.7) a 1552.4 (328.8) a 1917.7 (222.3) a 2102.5 (219.4) a 2236.6 (229.8) a 2130.8 (111.2) a EXTR Fe (mg/kg) 112.4 (29.6) a 200.4 (64.9) a 159.3 (61.4) a 239.7 (106.1) a 127.3 (88.0) a 189.7 (91.0) a 275.1 (381.7) a 267.6 (106.1) a 347.7 (166.8) a 421.3 (81.3) a 366.0 (139.1) a 387.5 (135.6) a 156.3 (107.2) a 150 .0 (60.7) a 144 .2 (65.4) a 101.4 (15.0) a 128 There were no significant differences among the ages for loss on ignition (LOI) in any horizon. Generally, LOI was lower than the old growth on the 0- and 10-year postburn sites for the LF horizon, and on the 10-year sites in the H horizon, and decreased with depth through the soil profile. The concentrations of extractable Ca were similar for all of the ages in the LF and H horizons (Tables 7-2, 7-4), and the variability was high. In the Bhf and Bf, extractable Ca was lower in the 5- and 10-year postburn sites, but the differences were not significant. There were significant differences among the sites for extractable Mg in the LF and H horizons, but there was also a significant location effect (Table 7-2). More extractable Mg was found on the postburn sites than on the old growth sites in the LF, H and Bhf horizons. There were no significant differences in extractable Al or Fe among the sites for any horizon. Generally, the postburn sites contained more extractable Al for all horizons, and more extractable Fe in the LF, H and Bhf horizons. In the Bf, the lowest extractable Fe levels were found in the 10-year postburn samples. There were no significant differences in pyrophosphate- extracted Fe among the sites in the LF and H horizons (Tables 7-5, 7-6) . In the Bhf, the 10-year postburn sites were significantly lower in pyrophosphate-Fe than the recently burned sites. Pyrophosphate-Fe was also significantly lower than in soils from the other ages in the Bf horizons of the 10-year sites. There were no significant differences in pyrophosphate-Al in the LF, H and Bhf 129 horizons. In the Bf, the 10-year sites were significantly lower in pyrophosphate-Al. However, there was also a significant location effect (Table 7-6). There were significant differences in pyrophosphate-extracted Mn in the LF, with the highest concentration in the recent burn and the lowest in the 10-year postburn and old growth sites. There were no significant Table 7-5: The mean values and (standard deviations) for pyrophosphate-extracted Fe, Al and Mn. Different letters indicate statistically significant differences among the ages for a horizon at HOR. LF H Bhf Bf AGE OG 0-YR 5-YR 10-YR OG 0-YR 5-YR 10-YR OG 0-YR 5-YR 10-YR OG 0-YR 5-YR 10-YR p<0.05 OG is old growth. Fe-PYRO % 0.010 0.121 0.013 0.111 0.069 0.121 0.123 0.167 1.395 1.527 1.174 0 .772 0.910 0.887 0.886 0.294 0.031 0.155 0.029 0.113 0.184 0.162 0.189 0 .202 0.934 0.642 0.633 0.511 0.401 0.528 0.698 0.118 a a a a a a a a ab a ab b a a a b Al-PYRO % 0.051 0.128 0.071 0.176 0.081 0.127 0.135 0.169 0.577 0.470 0.374 0.410 1.188 1.119 1.030 0.637 0.040 0.086 0.038 0.128 0.157 0.077 0.147 0.124 0.234 0.253 0.166 0.162 0.364 0.481 0.533 0.170 a a a a a a a a a a a a a a a b (n=9) Mn-PYRO % 0.063 0.170 0.126 (C 0.035 0.009 0.044 0.013 0.014 0.000 0. 011 0.001 0.002 0.000 0.004 0.001 0.000 0.036)a 0.178)b ).113)ab 0.034)a [0.012)a [0.056)a [0.010)a 0.012)a 0.000)a 0.023)a 0.002)a 0.004)a 0.000)a 0.009)a 0.002)a O.OODa 130 differences in the other horizons, and the values were low. There were no significant differences among the sites in any horizon for citrate bicarbonate dithionite (CBD)-extracted Fe, Al or Mn (Table 7-7) , or for Al and Fe extracted by acid ammonium oxalate (AAO) (Table 7-7) . In the Bhf horizon, the concentrations of Fe-CBD were highest in the recent burn, and lowest on the 10- year postburn sites. In the Bf, the 5-year sites had the most Fe- CBD, while the 10-year sites contained the least. The Al-CBD concentrations were similar among the sites in the Bhf horizon, Table 7-6: Analysis of variance (ANOVA) table for pyrophosphate- extracted Fe, Al and Mn; citrate-dithionite-bicarbonate (CBD)- extracted Fe, Al and Mn; and acid ammonium oxalate (AAO) - extracted Fe and Al, showing the probabilities as calculated by the Systat statistical program. As a nested experimental design was used, the terms for the ANOVA were TYPE, HORIZON, LOCATION{TYPE} (which is location nested within forest type), TYPE*HORIZON, HORIZON*LOCATION{TYPE}, and multiple R2. A * indicates significance at p<0.05. Fe AAO Fe CBD Fe Pyro Al AAO Al CBD Al Pyro Mn CBD Mn Pyro AGE 0.197 0.103 0.006 0.763 0 .141 0 . 071 0.157 0. 003 HORIZON 0.000 0.000 0.000 0.000 0.000 0. 000 0.627 0. 000 LOCATION {AGE} 0.038 0.008 0.015 0.005 0.001 0.003 0.190 0.706 AGE* HORIZON 0.489 0.370 0.006* 0.659 0.268 0.000* 0.879 0. 013* HORIZON* LOCATION {AGE} 0.357 0.689 0.14 0 0.086 0.208 0 .023* 0.936 8.582 MULT. R2 0.612 0.595 0.800 0.878 0.863 0.841 0.311 0.616 131 and were lowest in the 10-year sites for the Bf horizon.. Acid ammonium oxalate (AAO)-extracted Fe was lowest on the 10-year sites in the Bhf. In the Bf, the concentrations were very similar for all ages. The values of Al-AAO were also very similar for all sites in the Bhf and Bf. Amorphous Fe was found only in the 5-year postburn sites in the Bhf (Figure 7-1A). The Fe of this horizon was predominantly in organic form, and the 10-year postburn sites contain less organic Fe than the other sites. In the Bf, Fe is found in amorphous, crystalline and organic forms. The 10-year postburn sites contain Table 7-7: The mean values and (standard deviations) for citrate- bicarbonate-dithionite (CBD)-extracted Fe, Al and Mn, and for acid ammonium oxalate (AAO)-extracted Fe and Al. The LF and H horizons were not extracted. Different letters indicate statistically significant differences among the ages for a horizon at p<0.05. OG is old growth. (n=9) HOR Bhf Bf AGE OG 0-YR 5-YR 10-YR OG 0-YR 5-YR 10-YR Fe-CBD % 2.013 (1.296)a 2.304 (1.124)a 2.126 (1.127)a 1.362 (0.662)a 2.636 (1.111)a 3.456 (1.195)a 3.825 (1.492)a 3 .197 (0.723)a Al-CBD % 0.477 (0.175)a 0.469 (0.217)a 0.488 (0.204)a 0.451 (0.195)a 1.465 (0.385)a 1.535 (0.324)a 1.593 (0.583)a 1.231 (0.186)a Mn-CBD % 0.001 (0.001)a 0.014 (0.027)a 0.005 (0 .006)a 0.003 (0.004)a 0.001 (O.OOl)a 0.016 (0.034)a 0.004 (0.004)a 0.011 (0.023)a Fe-AAO % 1.283 (0.691)a 1.289 (0.450)a 1.285 (0.624)a 0.791 (0.536)a 1.780 (0.479)a 1.914 (0.560)a 2.033 (0.571)a 1.813 (0.296)a Al-AAO % 0 .519 (0.182)a 0.446 (0.164)a 0.483 (0.235)a 0.439 (0.199)a 2 .390 (1.097)a 2 .312 (0.702)a 2.326 (0.777)a 2.663 (0.427)a 132 % Fe A) Bhf-OGBhf-0 B h f - 5 B h f - 1 0 B f - 0 G B f -0 B f - 5 B f - 1 0 Hor izon • Organic I/2 Amorphous Gi3 Crystalline % Al B h f - O G B h f - 0 B h f - 5 B h f - 1 0 B f - O G B f - 0 B f - 5 B f - 1 0 B ) H o r i z o n Figure 7 - 1 : The mean p e r c e n t a g e s of o r g a n i c , amorphous and c r y s t a l l i n e i r o n (A) and a luminium (B) i n t h e Bhf and Bf h o r i z o n s of CH o l d growth (OG) and 0, 5 and 10 y e a r s p o s t b u r n s i t e s . 133 less organic Fe, and more amorphous Fe. The old growth sites contain the least crystalline Fe. These differences, however, were not significant. Aluminium was almost entirely in organic form in the Bhf horizon of all sites (Figure 7-1B). In the Bf, the 10-year postburn samples contained significantly more amorphous Al. Crystalline Al was not present in these soils. The results from the analysis for PA, PT, P0 and the C/P ratio are displayed in Tables 7-8 and 7-9 . Available P (Pa) was significantly higher on the recently burned sites in the LF horizon, and on the 0- and 5-year sites in the H horizon. There were no significant differences in PA in the Bhf and Bf. There were no differences among the sites in P0 in the LF, H and Bhf horizons. However, in the Bf, P0 was significantly lower on the 10-year sites. There were no significant differences for any horizon among the sites for PT or the C/P ratio, and the variability was high. The Chang and Jackson fractionation procedure produced PHCi/ PNaoH and PCBD- As was noted in Chapter 6, PNaOH is thought to be the non-occluded phosphate bound to the surfaces of Al or Fe hydrous oxides (Olsen and Sommers 1982) . The PCBD fraction is comprised of P occluded within the matrices of Fe and Al oxides and hydrous oxides, while the PHC1 is thought to be the extracted calcium phosphates of the non-occluded apatite fraction (Williams et al. 1980; Olsen and Sommers 1982). There were no significant differences among the sites for PHC1 or PCBD in either the Bhf or Bf 134 Table 7-8: The mean values and (standard deviations) for available P, total P, organic P and C/P ratio. Different letters indicate statistically significant differences among the ages for a horizon at p<0.05. OG is old growth. (n=9) HOR LF H Bhf Bf AGE OG 0-YR 5-YR 10-YR OG 0-YR 5-YR 10-YR OG 0-YR 5-YR 10-YR OG 0-YR 5-YR 10-YR Avail• (mg/kg 30.46 (10.01 88.97 (31.25 27.16 (9.38 17.25 (7.15 20.62 (9.65 44 .11 (24.89 39.05 (21.37 12.62 (5.08 6.06 (2.92) 8.84 (8.12) 9.24 (6.54) 8.59 (1.78) 6.31 (1.02) 6.43 (1.12) 6.09 (1.05) 6.60 (0.51) P b a b b b a a b a a a a a a a a Total P (mg/kg) 585.8 (66.6) a 713.9 (181.4) a 675.8 (141.9) a 485.3 (163.1) a 482.8 (152.1) a 524 .0 (219.9) a 569.5 (150.6) a 515.4 (150.9) a 352.1 (89.6) a 349.4 (199.5) a 468.6 (170.6) a 354 .5 (66.1) a 289.2 (115.6) a 230.8 (41.1) a 235.5 (102.2) a 211. 8 (70.2) a Org. P (mg/kg) 492.0 (98.8 622.3 (272.4) 551.2 (177.6) 391.1 (124.2) 308.7 (123.2) 360.4 (243 .9) 401. 0 (127.5) 306.2 (112.2) 145.1 (89.2) 174.2 (97.7) 220.2 (111.0) 198.7 (93.1) 64 .8 (36.4) 41.3 (23.0) 63 .5 (36.8) 21.6 (24.8) a a a a a a a a a a a a a a a b C/P 815 (135) 539 (242) 673 (162) 878 (295) 1055 (389) 958 (536) 826 (368) 866 (258) 392 (152) 450 (111) 343 (112) 468 (151) 252 (90) 268 (104) 221 (65) 193 (71) a a a a a a a a a a a a a a a a 135 Table 7-9: Analysis of variance (ANOVA) table for available P, organic P, total P, P-HCl, P-NaOH and P-CBD (Chang & Jackson), and the C/P ratio, showing the probabilities as calculated by the SYSTAT statistical program. Because a nested experimental design was used, the terms for the ANOVA were AGE, HORIZON, LOCATION{AGE} (which is location nested within forest type), AGE*HORIZON, HORIZON*LOCATION{AGE} , and multiple R2. A * indicates significance at p<0.05. Avail P Org P Tot P P-HCl P-NaOH P-CBD C/P AGE 0.000 0.002 0.011 0.481 0. 011 0.049 0.095 HORIZON 0.000 0.000 0. 000 0.000 0.000 0.000 0.000 LOCATION {AGE} 0.128 0.318 0.000 0.175 0.135 0.008 0.035 AGE* HORIZON 0.000* 0. 0 04* 0.093 0.496 0.027* 0.228 0.065 HORIZON* LOCATION {AGE} 0.572 0.324 0.984 0.175 0.000* 0.032 0.922 MULT. R2 0.863 0.785 0.724 0.601 0.685 0.732 0.783 horizons. PNaoH w a s significantly higher in the Bhf horizon of the recently burned sites than the old growth or 10-year sites, and was lowest in the 10-year sites, but there was also a location effect. There were no significant differences in the Bf horizon. The percentage of total P (PT) found as organic P (P0) and inorganic P (Px) , and the percent recovery of the PT for each horizon and age can be found in Table 7-11. As these means were calculated from each sample, they are not completely additive. There were no significant differences among the ages in any horizon for P0 or percent recovery. Significantly more of the PT was found as Px in the Bhf of the recent burn than on the 5- and 10-year 136 Table 7-10: The mean values and (standard deviations) for P extracted by HCl, NaOH and citrate bicarbonate dithionite (CBD) , during the Chang & Jackson fractionation procedure. The LF and H horizons were not extracted. Different letters indicate statistically significant differences among the ages for a horizon. OG is old growth. (n=9l HOR. Bhf Bf AGE OG 0 YR 5 YR 10 YR OG 0 YR 5 YR 10 YR P-HC1 (mg/kg) 7 8 7 7 19 24 22 30 (0.5) a (1.3) a (1.0) a (0.7) a (13.6) a (12.1) a (24.0) a (14.5) a P-NaOH (mg/kg) 37 58 42 27 25 23 17 18 (28.0) b (34.8) a (13.6) ab (16.9) b (12.4) a (12.0) a (10.8) a (6.8) a P-CBD (mg/kg) 22 34 26 15 74 59 68 50 (17.0) a (20.5) a (19.3) a (13.9) a (23.6) a (24.8) a (35.8) a (14.7) a postburn sites. There were no significant differences in Pj. in the Bf. Table 7-12 displays the correlation matrix for the results of this chapter. Extractable Al, and Al extracted by CBD, AAO and pyrophosphate all had high correlations with one another. Acid ammonium oxalate (AAO) -Al was also positively correlated with PHC1 and pH in water and in CaCl2, and was negatively correlated with extractable Fe, total N and the C/P ratio. Aluminium extracted by CBD was positively correlated with PHC1, Fe-AAO, Fe-CBD and both pH methods. Pyrophosphate-Al correlated positively with PCBD and negatively with extractable Mg and total N. There was a positive relationship between Fe extracted by CBD and AAO, and with pyrophosphate-extracted Mn to Mn-CBD and organic P. Extractable Al 137 Table 7-11: The means and (standard deviations) of total P found as organic and inorganic P, and the percentage of total P recovered. Inorganic P is the sum of the fractions determined by Chang & Jackson fractionation. N/E indicates not extracted. Different letters indicate statistically significant differences between ages for a horizon at p<0.05. HOR. LF H Bhf Bf AGE OG 0 YR 5 YR 10 YR OG 0 YR 5 YR 10 YR OG 0 YR 5 YR 10 YR OG 0 YR 5 YR 10 YR ORG. P 83.3 85.6 80.0 83 .5 65.5 65.7 73 .4 62.7 39.1 50.4 49.1 57.4 23.9 17.8 28.1 10.2 112.7 16.9 10. 8 17. 9 28.3 16.7 28.6 25.1 17.9 12.9 20.1 24.8 12.0 10.2] 12 .5] 10.9] ) a ) a a a a a a a a a a a a a a a INORG. P N/E N/E N/E N/E N/E . N/E N/E N/E 19 (10.7) 33 (14.6) 17 (6.1) 13 (7.5) 44 (15.6) 46 (9.7) 48 (14.8) 50 (15.6) ab a b b a a a a % RECOVERY 83.3 85.6 80.0 83 .5 65.5 65.7 73 .4 62.7 58 .0 83 .2 65.9 70.8 68.1 ( 64.2 ( 65.9 ( 60.3 ( 112.7 16.9 10.8 17. 9 28.3 16.7 28.6 25.1 22.1 18.7 21.5 25.7 17.3 18.7] 21.5] 14 .5] ) a a a a a a a a a a a a a a a a had a negative correlation with extractable Fe and Mg, and with C. Loss on ignition (LOI) was positively correlated with total N, C and the C/P ratio. There was a positive relationship between the two pH methods. Positive correlations were also seen for: 138 Table 7-12: Correlation matrix. Al- and Fe-AAO were extracted by acid ammonium oxalate, A1-, Fe-, and Mn-CBD by citrate bicarbonate dithionite. Al-, Fe-, and Mn-pyro by pyrophosphate. Avail P is extracted by Bray. Extr-Al, -Ca, Fe, and -Mg were extracted by Mehlich. H20A is air-dry moisture content; H20F is field moisture content. Org P is by NaOH-EDTA extraction and Tot P is by Parkinson & Allen digest. | -iddddddddddddd 4, -Idddddddddddddd 0 o ro ro % in 05 •* co ro N er> \o cr> \o-* •>«• x 1 -iddddddddddddddd CP o N in •«• co en -< (\l o N -H vo N m m N i*. g L S - * ro *C4 -Hinrgm N N ~I ro •* rvico \ i o "x *idddddddd dd d d ddd d a. UJ T T T t i l l Jj ON-H*criC3ni\0iJ)!QiP'-i'-ir>JCn(rif^PJ z ^ - d ? ? d v * d d d d d d d d d d d c i 1 "x ^ 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 a "5 rt d d o d d d o d d d d o d d d d d o o o. •H ddcpd d o d d oocpo o d d odcp d o S u §g§8nl§iS!Slg§g§S^S§Sg — d d d d d d d d d d d d d d d d d d d d d o Q-* i^ — (sir^N-'iO\D m N to i^ m en o (T> •• w o — m o B N en vo \o N « Nfo N as a o -*m tncKcvi p -* i^coro L w d d d d d d d d d d d d d d d d d d d d d d i S §!M^8^i^H8^o^s § A) I > -i o o o o o o o o o o o o o o o o o o o o o o o fc L S f J p « •»• en to 3 N P B N ve ro N t 3 &i N co ro « * - * i^ in a 9i 3 •«• K « in s i n o s o •* co in — •-< in a co In <\i ro in ro\o \c •* _i T - d d d d d d d d d d d d d d d d d d d d d d d d tr § o ro (M •» in + m in in —-H « \o -H •*•>«• -* -• i^ co t en in CM CO ro it 5 in — •* u3 cr>55 in S vo ro •* i^ ro sfi en o *co co 01 O N K - \O O S o co ro in -< f N •<*• \o Co \o \i o m to •* a-Hin in in 3 •<*• . inm Q i o • • X _ « o o o o p o o p o o o a o o p p o o o o o o o o cc I I T I T I T I T T I I i ti § g N -H ro o \o so en co vo ro in ro (\i -• in •* \o ** en in ro s. ro ro r>. ro L o 55 N N ID CM m s f N Inin * ro r\i •* in a -• J K 3 s in in \D ro cj ^ ^ d d d p ' d d d c i o o d d p ' d a c i c i d d d d d o D O D o o o a u x a - H m o/ oio a o i. u. " a o a w p -< I • z: a T T C L I » ^ J ^ S i J J T t c I p x T a o i i i i z T i - i - cc cc — 4 uliDiiJuJLi.ii . ix £ fc x a cc u. r a. - - d -?? SSRfc • » • » - « o o o -cpdocp Oin •-«*mo ^ d d d d d t8B2?S8 'idddddo §S81HS2 • • • • » • » • - • o o a o o o a -. d d o o d d d o !8^§s§ias' O N i -Id dd d d d d d d •Idddddddddd §88?sB5l8i«iB H o d c i d d d d d d d d - • d o d d d d d o p a o o 139 available P with organic P and total P; organic P with total N, total P, Ca, Mg and C; and total P with total N and Ca. Carbon correlated positively with C/P, total N and Ca. Extractable Ca was positively correlated with extractable Mg and total N. The 31P NMR spectra for the two old growth profiles are found in Figure 7-2 (A and B) . The percentage of P found within each class of compounds, calculated from the spectra by integration, are shown in Table 7-13 for the old growth and recently burned profiles, and in Table 7-14 for the 5- and 10-year sites. Figure 7-3 shows the spectra from the recently burned sites, while Figure 7-4 displays spectra from sites 5 years after burning, and Figure 7-5 from sites 10 years after burning. A guide for the interpretation of NMR spectra can be found in Chapter 5 (Table 5- 2) . As discussed in Chapter 6, the spectra for the two mature CH forests were quite different from one another. The orthophosphate and monoester peaks were separated well in only the Bf horizon in profile A, but were distinct in all horizons of profile B. This problem with peak separation is discussed in Chapters 5 and 6. In A (Fig. 7-2A; Table 7-14) the proportionss of monoester and diester P were comparable in the LF and H, and were nearly double those of orthophosphate P in these horizons. Orthophosphate and monoester P were the dominant P classes in the Bhf; monoesters dominated in the Bf. There were small phosphonate peaks in the LF and H horizons. In profile B (Fig. 7-2B; Table 7-13), most of the P was found as orthophosphate or as monoesters, at comparable 140 A) CEDAR-HEMLOCK OLD GROWTH A If HORIIOH PHOS 1 1 1 I 1 — i* t > -S -10 O R T M O M O N O  ( ™ ^ T PYRO 1 -28 POLY CEDAR-HEMLOCK OLD GROWTH B If HORIZON PHOS ORTHO MONO DIE3T POLY F i g u r e 7-2 lP NMR spectra for two soil profiles from mature CH sites, extracted with NaOH-EDTA. 141 Figure 7-13: The proportion of P found within each P form, as calculated from the 31P-NMR specta by integration, for the old growth (OG) and 0-year postburn profiles. Phos is phosphonate, orth is orthophosphate, mono is monoester P, diest is diester P, pyro is pyrophosphate and poly is polyphosphate. AGE OG A OG B 0-YR A 0-YR B HOR LF H Bhf Bf LF H Bhf Bf LF H Bhf Bf LF H Bhf Bf Phos o, 0 0 5 4 0 0 0 0 0 0 0 0 0 0 5 10 0 Orth 17 7 38 23 31 39 35 29 51 12 25 85 24 23 14 13 Mono o, o 33 38 36 61 38 32 34 32 32 52 43 15 41 34 47 62 Dies o, o 39 36 17 16 18 20 24 24 11 23 24 0 29 33 24 25 Pyro o, 0 5 5 0 13 9 7 15 6 12 8 0 6 0 5 0 Poly % 11 7 0 0 0 0 0 0 0 0 0 0 0 5 0 0 percentages. Diester and pyrophosphate peaks also occurred, but there were no phosphonate or polyphosphate peaks in any horizon. On the recently burned sites (Fig. 7-3 A and B; Table 7-13), there were differences in the spectra for the two profiles. The orthophosphate and monoester peaks were distinctly separated in the 142 Bhf and Bf horizons of profile A, but not in the LF and H horizons, and they were not clearly separated in any horizon of profile B. Orthophosphate was the predominant P form in the LF of profile A. Monoesters were also present, as well as traces of diester P and pyrophosphate. In the H horizon, the percentages of monoester, diester and pyrophosphate P increased, while orthophosphate decreased. In the Bhf horizon, orthophosphate increased, while diester and pyrophosphate were similar to the H horizon. In the Bf, orthophosphate was the predominant P form, with a small amount of monoester. The LF and H horizons of profile B had orthophosphate and diester at equal levels, and a higher percentage of monoester. In the Bhf, phosphonate and pyrophosphate were present, in addition to orthophosphate, monoester and diester P. The P of the Bf horizon occurred mainly as monoesters, with some orthophosphate and diester P. Five years after burning (Fig. 7-4 A and B; Table 7-14), monoester phosphate was the predominant P form in all horizons of profile A. The LF and H also contained diester, orthophosphate and polyphosphate peaks. There was a phosphonate peak in the H horizon, and pyrophosphate peaks in the LF and Bhf. In the Bf, the main P form was monoester, with some orthophosphate and diester P. In profile B, peaks were seen only for orthophosphate, monoester and diester P. Monoesters were the dominant P form in all but the Bf, where diester P was highest. 143 A) CEDAR-HEMLOCK 0 YEAR A IF HORIZON H HORIZON Ay/,VVWv/ to s « -a ORTHO. MONO p j ^ j pYHO B) CEDAR-HEMLOCK OYEAR B IF HORIZON H HORIZON Bhf HORIZON Bf HORIZON POLY F i g u r e 7 - 3 : 31 P NMR spectra for two soil profiles from recently burned CH sites, extracted with NaOH-EDTA. 144 CEDAR-HEMLOCK FIVE YEAR A 23 I« PHOS ORTHO MONO pjgg.,. PYBO A) CEDAR-HEMLOCK FIVE YEAR B IF HORIIOH J^yviA^ H HORIZON Bhf HORIZON 4^v^>^^ Bf HORIZON j IV^Kvy B) F i g u r e 7-4 24 PHOS ORTHO MONO £"!„ PYRO POLY LP NMR spectra for two soil profiles from CH sites 5 years after burning, extracted with NaOH-EDTA. 145 Figure 7-14: The proportion of P found within each P form, as calculated from the 31P-NMR spectra by integration, for the 5- year and 10-year postburn profiles. Phos is phosphonate, orth is orthophosphate, mono is monoester P, diest is diester P, pyro is pyrophosphate and poly is polyphosphate. AGE 5-YR A 5-YR B 10-YR A 10-YR B HOR LF H Bhf Bf LF H Bhf Bf LF H Bhf LF H Bhf Bf Phos % 0 4 0 0 0 0 0 0 0 4 0 0 0 0 0 Orth % 23 15 12 30 33 23 23 18 17 14 31 36 8 12 65 Mono % 40 36 55 60 44 59 51 37 27 39 51 53 69 36 25 Dies % 18 25 29 10 23 18 26 45 32 24 14 11 18 42 0 Pyro % 7 0 4 0 0 0 0 0 10 6 4 0 5 10 0 Poly % 12 20 0 0 0 0 0 0 14 13 0 0 0 0 10 The spectra for the two profiles from the 10-year sites were very different (Fig. 7-5 A and B; Table 7-14). In profile A, the orthophosphate and monoester peaks overlapped in the LF and H horizons, but were separate in the Bhf. A spectrum could not be obtained for the Bf horizon, due to the low P concentration in this soil, and the high concentration of interfering paramagnetic ions 146 CEDAR-HEMLOCK 10 YEAR A IF HORIZON Ko^A^Uv/^YNMAvfy^^ Bf HORIZON A spectrum could not be obtained ORTHO MONO DIEST PYRO POLY CEDAR-HEMLOCK TEN YEAR B 20 PHOS IS 10 5 0 ORTHO MONO «•« DIEST B) Figure 7-5: 31P NMR spectra for two soil profiles from CH sites 10 years after burning, extracted with NaOH-EDTA. 147 such as Mn. There was a range of P forms in the LF horizon of profile A. Diester P was highest, followed by monoester P. Orthophosphate and pyrophosphate were also present, as well as a sharp polyphosphate peak. The spectrum for the H horizon was similar to that of the LF, but with a smaller polyphosphate peak. A small phosphonate peak was also present in the H horizon. In the Bhf, most of the P occurred as monoesters. The percentages of orthophosphate were greater than in the LF and H, while diester P was less. There was also a small amount of pyrophosphate. In profile B, the orthophosphate and monoester peaks were separated only in the Bf horizon. In the LF, the P was predominantly found as monoester phosphate, with smaller quantities of diester and monoester P. In the H horizon, the proportion of orthophosphate dropped, while the monoester and diester levels increased. Pyrophosphate was also present. The levels of monoester and diester P were almost equal in the Bhf, with lower levels of orthophosphate and pyrophosphate. The Bf horizon contained mainly orthophosphate, with low levels of monoester P and polyphosphate. The signal-to-noise ratio was low for this horizon, and it was difficult to distinguish peaks from the background noise. This was due to the low P concentration in this sample, and the high concentration of interfering paramagnetic ions. 148 Discussion Prior to discussing the results of this research, it should be noted that a true chronosequence was not used. Although only CH forest types were sampled, differences between locations may have developed after burning which were independent of the burning effects. In addition, the 10-year sites were burned in the fall, while the 0- and 5-year sites were burned in the spring. The slash and litter were drier in the fall, resulting in a more intense burn. This was reflected in the depth to the Bf horizon, which was less on most of the 10-year sites than it was on the old growth, 0- or 5-year sites (personal observation). Fire intensity and duration, reflected in the quantity of slash and forest floor material consumed by the fire, are believed to determine the absolute amounts of nutrients lost via burning, through increased volatilization and convection losses (Boyle 1973; Brockley et al. 1992) . The choice of 10-year sites available for sampling was very limited, as most 10-year postburn sites had received N and P fertilizers to overcome the growth check problem. The locations used in this study were considered to be the "better" sites, with fewer growth problems, and thus had not been fertilized (P. Bavis, per. commun.). As well, the 10-year sites were replanted with different species from the 5-year sites. Sitka spruce (Picea sitchensis (Bong) Carr.) and Douglas-fir {Pseudotsuga menziesii (Mirb.) Franco) were found on the 10-year sites in addition to 149 western hemlock and western red cedar, which grew on the 5-year sites. There was a greater choice of 5-year sites than 10-year sites, because none of the 5-year sites had been fertilized. Site selection was based to some extent on ease of sampling: on many sites the salal was so dense that it made sample collection very difficult. The 5-year locations all had some salal, but the most densely covered locations were not used. Generally, there was less salal on the 10-year sites (personal observation). There was no choice of 0-year sites, as only three locations had been burned at the time of sample collection in the vicinity of the other postburn locations. The availability of sampling locations largely determined the experimental design. A nested design was used, with locations nested within each age, which should take location effects into account. Despite many apparent trends in the data, some of the differences were not statistically significant. This may be due to insufficient sampling: the high variability of some of the elements investigated probably necessitates more than nine samples per treatment. This variability is due both to natural distribution patterns and to uneven heating and burning of slash and litter during each fire. There was a significant pH increase in the surface horizon of the recently burned sites, as others have documented (Ahlgren and Ahlgren 1960; Kozlowski and Ahlgren 1974; Feller 1982; Ellis and Graley 1983; Macadam 1987; Tomkins et al. 1991; Brockley et al. 1992; Romanya et al. 1994). This is part of the so-called ashbed 150 effect of fire on soils (Humphreys and Lambert 1965; David 1987). The significant pH increase in the H horizon of the recent burn may reflect contamination of the samples with the ash layer during sampling. For the 5-year samples, the significant pH increase in the H horizon reflects movement of alkaline salts and ash down the horizon (Tomkins et al. 1991), which also accounts for the significantly elevated pH in the Bf of the 10-year postburn sites. Ellis and Graley (1983) and Tomkins et al. (1991) also reported pH increases at depth with time after burning. There was a decrease in loss on ignition (LOI) in the LF horizon of the 0-year sites, as well as a lower C concentration, and a decrease in LOI in the H horizon 10 years after burning. As was discussed in Chapter 6, LOI is a measure of soil organic matter. Fire is expected to reduce organic matter content in surface horizons (Macadam 1987; Tomkins et al. 1991), especially in mor humus forms , where organic matter is not incorporated into soil, but accumulates on the mineral surface (Feller 1982). Litter inputs seem to bring the organic matter content of the LF horizons back to close to the old growth levels by 5 and 10 years after burning. However, LOI and C were lowest in the Bf horizon 10 years after burning, suggesting that burning the surface organic matter will affect the movement of organic matter through the profile. Macadam (1987) also reported a decrease in C concentration with time in mineral soil after fire. The lowest total N concentrations in the LF were on the recent 151 burn; for the other horizons, total N was lowest on the 10-year sites. Those differences, however, were not significant. Reports in the literature on total N levels after fire are varied. Some researchers report decreases (Beaton 1957; Grier 1975; St. John and Rundel 1976; Ellis and Graley 1983; Khanna and Raison 1986; Macadam 1987; Brockley et al. 1991); others report increases (Vlamis et al. 1955; Mangas et al. 1992). Nitrogen is readily volatilized during fire (Mroz et al. 1980), and the amount of N lost is relative to the intensity of the burn. The correlation of total N to LOI and to other components of soil organic matter (such as organic P) suggests that the changes seen in this study reflect changes in organic matter, rather than volatilization losses of N. However, the lower N concentrations on the 10-year sites in the H, Bhf and Bf horizons may indicate some volatilization losses with the higher intensity burns used on these locations. The results for Ca were highly variable, with no significant differences and no obvious trends, except in the LF horizon, where the Ca concentrations were slightly higher on the 0- and 5-year sites. Other researchers have reported significant increases in Ca concentration after fire (Vlamis et al. 1955; Beaton 1957; Boyle 1973; Grier 1975; St. John and Rundel 1976; Ellis and Graley 1983; Feller et al. 1983; Khanna and Raison 1986; Macadam 1987; Tomkins et al. 1991; Brockley et al. 1992; Rice 1993), with Ca increases proportional to fire intensity (Grier 1975; Brockley et al. 1992; Rice 1993). 152 Magnesium has also been shown to increase after burning (Grier 1975; Ellis and Graley 1983; Feller et al. 1983; Khanna and Raison 1986; Macadam 1987; Tomkins et al. 1991; Brockley et al. 1992; Mangas et al.1992; Rice 1993). In this study, Mg concentrations were significantly increased in the LF and H horizons of all of the postburn sites relative to the old growth. Although statistically there was a location effect, it occurred in the old growth data, as one location was higher in Mg in the LF and H horizons that the other two location used for this age. Becauses the Mg concentrations of the postburn sites were higher than even the highest old growth location, this apparent effect of fire on Mg is probably valid. The positive correlation of Mg to LOI, total N and organic P suggests that this increase was due to the release of Mg when organic matter was destroyed. The high Mg levels in the surface horizons 5 and 10 years after burning, with no increases at depth, indicate that it is not very mobile in these soils. Significant differences in pyrophosphate-extractable Al and Fe were seen only in the mineral soil. Pyrophosphate-Fe was significantly lower in the 10-year sites in the Bhf and Bf horizons, while pyrophosphate-Al was significantly lower in the 10- year Bf. There were no significant differences in citrate bicarbonate dithionite (CBD)-extracted Fe and Al, nor in acid ammonium oxalate (AAO)-extracted Fe and Al. As discussed in Chapter 6, pyrophosphate extracts the organic-associated Fe and Al. In natural, unmodified podzols, Fe and Al in the Bhf and Bf 153 horizons are predominantly in organic form, reflecting the characteristic illuviation of organic matter and organo-metallic complexes in podzolic soils (Oades 1989). The presence of crystalline Fe but not Al is typical of podzols of Vancouver Island; the amorphous Fe and Al in the Bf horizons reflects the volcanic parent material of this region (Lewis 1976; Sanborn 1987). In the Bhf horizon, the 10-year sites had less organic Fe than the other ages. In the Bf, the 10-year locations had less organic Fe and more amorphous Fe, and significantly more amorphous Al. Amorphous Al and Fe concentrations were determined by subtracting the pyrophosphate extraction results from the AAO extraction values. The significantly lower pyrophosphate results may falsely elevate the amorphous Al and Fe concentrations. In light of the lack of differences in AAO-extracted Fe and Al, the apparent increase in amorphous Fe and Al is probably not a true effect of burning, unlike the change in organic Fe and Al. The intense burning of the 10-year sites appears to have altered the illuviation of organic matter and organo-metallic complexes. The same changes do not appear to have occurred on the 5-year sites, either because there has been insufficient time since burning for changes to manifest, or because the less intense burn on the 5-year sites destroyed less organic matter, leaving the illuviation processes unaffected. These changes in the Bf are probably only temporary, although they may persist for several years. The regrowth of vegetation on these sites will increase litter inputs to the surface, which should increase the cycling of organic matter 154 throughout the soil profile. There were no reports in the literature of changes in pyrophosphate-extracted Fe and Al. However, Humphreys and Lambert (1965) observed a decrease in oxalate-soluble Al in mineral soils nine years after burning, while Kwari and Batey (1991) demonstrated that heating soils in laboratory ovens could increase CBD-extracted Al and Fe, which they considered to be a measure of free Fe and Al oxides. There was a significant increase in pyrophosphate-extracted Mn in the LF immediately after burning, as Mn was released from organic matter. This appears to have been a short-term effect, however, as the Mn concentration of the 5-year sites was only slightly increased, and there were no significant changes in any other horizon. A significant increase in available P (PA) after burning has been reported by other researchers, and is one of the most consistent effects of fire on soils (Ahlgren and Ahlgren 1960; Humphreys and Lambert 1965; Kozlowski and Ahlgren 1974; Feller 1982; Feller et al. 1983; Macadam 1987; DeBano and Klopatek 1988; Tomkins et al. 1991; Brockley et al. 1992; Mangas et al. 1992; Saa et al. 1993; Romanya et al. 1994) . This increase in PA is relative to the severity of the burn, and is of short duration, decreasing with time (Macadam 1987; DeBano and Klopatek 1988; Tomkins et al. 1991; Romanya et al. 1994) . On these sites, PA in all horizons had returned to old growth levels in the 10-year postburn sites. As was observed with soil pH, significant increases in PA in the H 155 horizon of the recent burn probably reflect contamination with ash during sampling. There were no significant differences among the ages for total P (PT) , and the variability was high. Generally, PT was highest on the recently burned sites in the LF horizon. In the H horizon, all of the burned sites had elevated PT levels relative to the old growth. In the Bf, PT was lowest 10 years postburn. Beaton (1957), Ellis and Graley (1983), Tomkins et al. (1991) and Romanya et al. (1994) also report increased PT in surface horizons after burning, while the results of Saa et al. (1993) were variable, showing no clear pattern. Organic P was elevated relative to the old growth on the 0- and 5-year sites in the LF and H horizons. Previous studies report decreased P0 after fire (Kwari and Batey 1991; Saa et al. 1993; Romanya et al. 1994), which is believed to be due to the combustion of organic matter, and to the solubilization and transport of P0 down the profile, caused by the pH increase in the upper horizons (Romanya et al. 1994) . The apparent increase in P0 found in this study is probably an artifact of the method used to determine P0. In this procedure, organic matter is oxidized to PI# and a subsequent colorimetric test measures the Pj. in solution. This is then related back to the P0 content, without accounting for the Pz which may already be present. In the recently burned samples, where fire has oxidized the organic matter, releasing PI# this would cause an overestimation of P0. When the Saunders and 156 Williams (1955) ignition method which measured both P0 and Px was used, it showed a significant increase in Pz and a significant decrease in P0 in the LF horizon of the 0-year sites relative to the other ages. This is supported by the NMR results, which showed mainly inorganic orthophosphate in the LF of recently burned soils. The methods used in this study are discussed in more detail in Chapter 4 . The significant decrease in P0 in the Bf horizon of the 10-year sites reflects the altered illuviation patterns in these stands. The 31P NMR spectra show changes in P forms after fire. As was noted in Chapter 6, the NMR spectra are specific to each profile examined, making generalizations difficult. Better results might have been obtained by using composite samples. In the old growth samples, the LF and H horizons show spectra typical of wet areas with slow decomposition: the diversity of P forms is high, and relatively labile compounds such as diesters have persisted (Tate and Newman 1982; Condron et al. 1990a; Gil-Sotres et al. 1990) . In the Bhf and Bf horizons, the diversity of P forms is reduced, and orthophosphate predominates, although organic P forms are found, even in the Bf. Immediately after burning, most of the P in the surface horizons was converted to orthophosphate, although some organic forms are still present. The lower horizons appear to be unchanged from those of the old growth, with the same general trends. Five years after burning, orthophosphate is the predominant P form in the LF and H horizons. Polyphosphate peaks are also present for one profile. Khanna and Raison (1986) suggest that the 157 P which is mineralized by soil heating may be deposited in ash in slowly soluble forms such as polyphosphates. Polyphosphates are easily formed in the laboratory by heating orthophosphate (Kulaev 1979), and their presence in these horizons may be an effect of burning. However, as discussed in Chapter 6, they are also storage products of ectomycorrhizae and soil microbes, and thus their origin may be biological. The spectra for the Bhf and Bf horizons are not unlike those of the old growth and recent burn profiles, except for the Bf of the 5-year A, which produced a poor quality spectrum due to its low P concentration. Litter deposits from reestablished vegetation resulted in spectra for the LF and H horizons of 10-year profile A which were comparable to those for the old growth. In profile B, though, monoester phosphate was the predominant P form in all but the Bf horizon. The most noticable difference of the 10-year sites, relative to the other ages, was the poor quality of the spectra obtained for the Bf horizon. In fact, a spectrum could not be obtained for the Bf of profile A. This was due to low P concentrations in these horizons, reflecting lower organic P levels. Romanya et al. (1994) have suggested that the increased pH of the surface horizons after fire may cause the solubilization and transport of P0 compounds down the soil profile. These NMR results negate that, however, as fewer P0 forms are found at depth with time after fire. There are no published studies showing 31P NMR spectra of forest soils following burning. Zech et al. (1987) state that cutting and burning of Pinus mugo and establishment of pasture did 158 not significantly influence the patterns of 31P NMR spectra of the surface horizons, but they give no information on the length of time since burning, or the intensity of the fire, and they only looked at one sample. There were no significant differences among the sites for PHC1, which measured calcium phosphates, or PCBD/ which measured P occluded within Fe and Al oxides and hydroxides. In the Bhf, PNa0H (non-occluded) was significantly higher in the samples from the recent burn than from the old growth or 10-year sites. There was also a significant location effect, as one of the recent burn locations had very high concentrations of PNa0H/ while one of the old growth locations had very low concentrations, and the variability was high. It is unlikely that this increased PNaOH is a n effect of fire, because only the surface layers were burned, and there had not been sufficient time or rainfall for burning effects to be experienced in the lower horizons. In the Bf, the 10-year postburn samples appeared to have increased levels of PHC1, and reduced PNa0H and PCBD relative to the other ages. The formation of calcium phosphates in surface soils after fire has been reported elsewhere (Ellis and Graley 1983; Kwari and Batey 1991; Saa et al. 1993) . It is possible that the higher intensity fires on the 10-year sites, together with the elevated pH, caused the Ca and P released from organic matter by burning to recombine into calcium phosphates. These subsequently moved down the soil profile to the Bf horizon, where they have persisted because the pH is still significantly elevated over old growth levels. However, the positive correlation 159 of PHC1 to Al extracted by acid ammonium oxalate and citrate bicarbonate dithionite suggests that the extraction procedure for PHC1 may be measuring P associated with Al rather than Ca. Increased aluminium phosphate concentrations have been reported after fire (Humphreys and Lambert 1965; Khanna and Raison 1986; Kwari and Batey 1991), as well as increased P sorption capacity, which is thought to be due to Al released from organic matter after burning (St. John and Rundel 1976; Kwari and Batey 1991; Romanya et al. 1994) . Although phosphate sorption capacity was not measured in this study, Yuan and Lavkulich (19 94) have demonstrated that it can be predicted from the concentrations of acid ammonium oxalate- extracted Fe and Al. Because these were not significantly different among the ages, it is unlikely that fire has significantly altered the P sorption capacity of these soils. One major drawback to this research is that the effects of harvesting and burning could not be separated, because unburned, harvested sites for each age were not available. However, for the 10-year sites, one location was scarified and the other two locations were not. This did not show up as a location effect during statistical analysis, and the standard deviations of the data on the 10-year sites were low. The observed changes were also consistent with burning effects reported by other researchers. This suggests that the burning effects had more impact on the measured data than did harvesting. However, the distinction between harvesting and burning effects were not addressed by the scope of this research, and warrants further research. 160 Conclusions After clear-cutting and burning, the soils of the CH forest types experience an ashbed effect, with increased pH and higher concentrations of available P, Ca, Mg and Mn in the surface horizons. These increases are only temporary, and most of these factors return to preburn levels within ten years. The combustion of organic matter is responsible for much of the ashbed effect. Destruction of organic matter in the more intense fall burnings appears to disrupt illuviation processes throughout the soil profile, and may produce longterm changes in lower mineral horizons. Although total P levels were not changed, there was a shift from organic P forms to inorganic P forms, and changes in P forms with time at depth in the profile. These changes in the P cycle may contribute to the growth check observed on these sites. 161 CHAPTER EIGHT THE USE OF ORGANIC PHOSPHORUS BY WESTERN RED CEDAR Introduction As discussed in previous chapters, a growth check which may be overcome with N and P fertilization occurs in Sitka spruce and western hemlock trees replanted onto clearcut and slash-burned cedar-hemlock (CH) sites. Studies by Weetman et al. (1989a, b) suggested that western red cedar were performing better than Sitka spruce and western hemlock when replanted on the CH sites, and showed less response to P fertilization than the other species. Pot seedling bioassays confirmed that cedar was relatively insensitive to nutrient availability (Messier 1993); apparently, cedar was able to access nutrients of low availability in CH soils. Western red cedar is unusual in that it forms vesicular-arbuscular (VA) mycorrhizae rather than ectomycorrhizae (Curran and Dunsworth 1988) . This symbiotic association may give this species some advantage on CH sites. Previous research (Chapters 6 and 7) showed that much of the P in the soils of the CH sites is in organic form. Organic P (P0) must be hydrolysed to orthophosphate before it can be used by plants (Tate 1984). This mineralization is catalysed by phosphatases, which are grouped into phosphoric monoester hydrolases (EC 3.1.3), phosphoric diester hydrolases (EC 3.1.4) or acid anhydride hydrolases (EC 3.6.1) (Speir and Ross 1978; Tabatabai 1982). The monoester hydrolases include: acid and 162 alkaline phosphatases, characterized by the pH at which they are most active and acting on compounds such as inositol phosphates, sugar phosphates and glycerophosphate; and enzymes such as phytase, which catalyses the removal of phosphate from inositol hexaphosphate (phytic acid) (Speir and Ross 1978; Tabatabai 1982) . The diester hydrolases include enzymes which act on nucleotides and phospholipids. The acid anhydride hydrolases act on phosphoryl- containing anhydrides such as ATP and pyrophosphate (Speir and Ross 1978; Tabatabai 1982). Enzymes which can hydrolyse soil P0 compounds have been demonstrated in soil microorganisms, including fungi and bacteria (Szember 1962; Ko and Hora 1970; Greenwood and Lewis 1977; Dick and Tabatabai 1978, 1984; Browman and Tabatabai 1978; Beever and Burns 1980; Helal and Dressier 1989; Fox and Comerford 1992); in higher plants (Saxena 1964; Hasegowa et al. 1976; McLachlan 1980a, b; Basha 1984; Tarafdar and Claasen 1988; Tarafdar and Jungk 1987; Helal 1980; Garci and Ascencio 1992; Adams and Pate 1992; Dinkelaker and Marschner 1992; Barrett-Lennard et al. 1993; Tadano et al. 1993; Fernandez and Ascencio 1994; Pant et al. 1994); in ectomycorrhizae (Bartlett and Lewis 1973; Ho and Zak 1979; Alexander and Hardy 1981; Dighton 1983; Antibus et al. 1986; 1992; Kroehler et al. 1988; Ho 1989; Bae and Barton 1989; Haussling and Marschner 1989; Kropp 1990; Meysselle et al. 1991; Pasqualini et al. 1992; Tarn and Griffiths 1993; McElhinney and Mitchell 1993); in ericoid mycorrhizae (Straker and Mitchell 1986; Shaw and Read 1989; Read 1991; Dighton and Coleman 1992) and in orchid 163 mycorrhizae (Antibus and Lesica 1990). Although phosphatase production has been demonstrated in VA mycorrhizae (Gianinazzi- Pearson and Gianinazzi 1976; MacDonald and Lewis 1978; Gianinazzi et al. 1979,-Kapoor et al. 1988; Dodd et al. 1987; Jayachandran et al. 1992; Thiagarajan and Ahmad 1994; Khalil et al. 1994; Tarafdar and Marschner 1994), it is believed that VA mycorrhizal and non- mycorrhizal roots obtain P from the same soil P pools (Bolan 1991; Jennings 1995) . The objective of this research was to investigate the ability of mycorrhizal and non-mycorrhizal western red cedar seedlings to grow when supplied with organic P compounds such as phytic acid, glycerophosphate and ATP, and with inorganic pyrophosphate and KH2P04. In addition to measuring changes in growth parameters and foliar nutrient concentrations, the activities of several phosphatase enzymes were also determined. Materials and Methods Plants Two-year-old western red cedar seedlings, in styroblock containers, were purchased from Koksilah Nursery, Duncan BC, in March 1992. In June, 1992, the seedlings were transplanted into a sand-peat mixture, with four seedings in each 5 L pot. The seedlings were grown in a greenhouse with supplemental fluorescent and incandescent lighting, at 15-25°C. The seedlings were watered with tap water, and were fed weekly with modified Long Ashton 164 solution (Hewitt 1966), with P at 1/4 strength. In Feb. 1993, the seedlings were transplanted into open pots containing a 3:1 mixture of medium sand and Turface, a calcined montmorillonite clay (Applied Industrial Materials Corporation, Deerfield, IL), with one tree per 5 L pot. They were irrigated with tap water, and were fed weekly with modified Long Ashton nutrient solution. Feeding Experiment As all of the seedlings were colonized by VA mycorrhizal fungi (species unknown), a systemic fungicide was applied to half of the pots to suppress the mycorrhizal fungi, to produce a 'non- mycorrhizal' treatment. Benomyl [(1-butyl-carbamoyl)-2- (benzimidazole) carbamic acid, methyl ester; (C14H18N403) ; sold as 'Benlate' (50 WP) by Later's Chemicals Ltd., Richmond, BC] was applied at a rate of 30 L/pot of a 5 g/L solution, every 9 days beginning April 20, 1993. The trees were given 3 00 mL of tap water every third day, plus a light overhead misting to increase the humidity. Beginning May 14, 1993, the watering every ninth day was replaced with 200 ml of modified Long Ashton solution, containing KN03 (400 mg/L), K2S04 (350 mg/L), Ca (N03) 2. 5H20 (900 mg/L), MgS04. 7H20 (500 mg/L), MnS04.4H20 (2.25 mg/L), CuS04. 5H20 (0.25 mg/L), ZnS04. 7H20 (0.30 mg/L), H3BO3 (3.0 mg/L), NaCl (5.0 mg/L) and one of eleven P treatments. These treatments were: 1 . No P 2 . p h y t i c a c i d h i g h ( i n o s i t o l h e x a p h o s p h o r i c a c i d , dodecasod ium s a l t ; C6H6024PsNa12; Sigma P-3168) , 50 mg P/L 165 3. phytic acid low, 10 mg P/L 4. ATP high (Adenosine 5'-Triphosphate, disodium salt, Grade II; C10H14N5O13P3Na2; Sigma A-3377) , 5 0 mg P/L 5. ATP low, 10 mg P/L 6. glycerophosphate high (disodium salt; C3H706PNa. 5H20; Sigma G-6501) , 5 0 mg P/L 7. glycerophosphate low, 10 mg P/L 8. pyrophosphate high (Na4P207.10H2O; ACS Fisher S390) 50 mg P/L 9. pyrophosphate low, 10 mg P/L 10. orthophosphate high (KH2P04, BDH ACS 657) 50 mg P/L 11. orthophosphate low, 10 mg P/L At the time of feeding, the benomyl-treated plants were given 3 0 mL of a 5 g/L solution of benomyl and 7 0 mL of tap waster, while the mycorrhizal plants were given 100 mL of tap water. In the winter months, the seedlings were watered with only 200 mL of tap water, and the extra water was eliminated from the benomyl-treated plants at feeding time, The mycorrhizal plants were given only 3 0 mL of extra tap water, to match the liquid from the fungicide which the benomyl-treated plants received. There were 5 mycorrhizal and 5 benomyl-treated seedlings for each P treatment; for a total of 110 seedlings. The experiment was arranged in the greenhouse in a completely randomized design, and the pots were rearranged periodically to maintain randomness of greenhouse effects. 166 Laboratory Analyses The trees were harvested May 9-20, 1994, after 52-53 weeks of growth. Initially, 2 trees from each treatment were harvested. Immediately after harvest, approximately half of the roots and 50 g soil were assayed for acid and alkaline phosphomonoesterase, phosphodiesterase and pyrophosphatase activity, as described in Tabatabai (1982) . Three replicates and one control sample were used per assay and the results for each plant and soil were averaged. For all plants in the experiment, a portion of randomly sampled roots was frozen for later analysis and the soil was air- dried. The above-ground biomass was oven-dried at 60°C for 48 hours, after which the dry weight was measured. The foliage was removed from the branches, ground with a stainless steel coffee grinder, and digested in glass tubes using the Parkinson and Allen (19 75) method. The N in the digests was determined colorimetrically using a LaChat Flow Injection Analyzer, while P, Ca, Mg, Cu, Fe, Al and Zn were read using an Inductively Coupled Argon Plasma (ICAP) spectrophotometer (Fisher Scientific Co.). The soils were analyzed for available P using the Bray PI method (Olsen and Sommers 19 82), and for pH in water (1:1, w/v) on one sample per treatment (McLean 1982). Thawed roots were cleared and stained in trypan blue with a modified version of the Phillips and Hayman (1970) method. Roots were cut into 1-2 cm pieces, spread evenly in destaining solution over the bottom of a Petri plate, and counted as per the gridline-intersect method of Giovanetti and Mosse (1980). The tree heights from the root plug, and diameters at 167 2 5 cm above the root plug, were measured at the start and end of the experiment. Statistical Analysis Statistical analyses were done using the Systat program (Wilkinson 1990) to perform analysis of variance and Tukey's tests. Homogeneity of variance was tested by plotting residuals against estimates. Where necessary, log and log (n+1) transformations were performed. Results The diameter changes and mean dry weights at the end of the feeding experiment are shown in Table 8-1, and the results from the analysis of variance can be found in Table 8-2. In Table 8-1 and in subsequent tables where the TREATMENT*MYC0RRHIZAL interaction was not significant, but the treatment differences were significant, both the mycorrhizal and benomyl-treated plants were used for each treatment to calculate the means. There were no significant differences among the treatments in tree height or diameter at the start of the feeding experiment (Table 8-2). One year later, the trees receiving the high rate of ATP had the greatest diameter increase while those receiving no P increased the least. The only other treatment in which the diameter increase was significantly different from the no P control was the high orthophosphate treatment. The highest dry weights were found in the treatments receiving high levels of ATP and orthophosphate, while the lowest dry weights were measured in the treatments fed no 168 Table 8-1: The means and (standard deviations) for diameter change and dry weight of above ground material after the feeding experiment. Ortho is orthophosphate, phyt is phytic acid, glyc is glycerophosphate and pyro is pyrophosphate. L is 10 mg P/l, H is 50 mg P/l. Different letters in a column indicate statistically significant treatment effects at p<0.05. (n=10) Treatment No P Ortho L Ortho H ATP L ATP H Phyt L Phyt H Glyc L Glyc H Pyro L Pyro H Diameter < (mm 3.0 4.6 5.4 4 .4 5.6 3 .2 3.4 3.6 4.9 3.6 4.2 [1.05 [1.52 1.26 1.26 1.51 0.79 1. 71 1.78 1.29 0.97 1.48 Change c abc ab abc a be be be abc abc abc Dry Weight (g) 59.5 (18.6) c 68.6 (10.0) be 98.3 (14.1) a 65.6 (18.0) be 91.5 (7.88) a 50.2 (10.3) c 56.7 (15.8) c 60.7 (21.0) c 86.6 (15.0) ab 65.1 (6.41) be 81.9 (20.0) ab P, low glycerophosphate, and both rates of phytic acid. Significant mycorrhizal effects were observed for height change, dry weight, and diameter change (Tables 8-2, 8-3) . In all cases, the measurements for the benomyl-treated plants were greater than those for the mycorrhizal plants. There were no significant differences in concentrations of foliar Al, Mg or Fe at the end of the feeding experiment (Table 8- 2) . The mean foliar Al concentration was 94.4 ug/g, the mean foliar Mg concentration was 2430.2 ug/g, and the mean foliar Fe 169 Table 8-2: Analysis of variance results for height and diameter at the start of the feeding study, height and diameter change, dry weight at harvest, and foliar P, Ca, Mg, Cu, Fe, Al, Zn and N. A * indicates significance at p<0.05. Ht Start Ht Change Diam Start Diam Change Dry Weight Foliar P Foliar Ca Foliar Mg Foliar Cu Foliar Fe Foliar Al Foliar Zn Foliar N Treat 0.062 0.503 0.086 0.000* 0.000* 0.000 0.000* 0.132 0.002* 0.280 0.725 0.000* 0.000 Myc 0. 065 0.031* 0.287 0.027* 0.010* 0.000 0.227 0.340 0.473 0.120 0.228 0. 049* 0.000 Treat*Myc 0 . 749 0.176 0.618 0.201 0.315 0.046* 0 .372 0.759 0.103 0.466 0.448 0.718 0.007* Mult R2 0.246 0.245 0.232 0.377 0.608 0.807 0 .415 0.208 0.352 0.219 0.173 0.357 0.625 N 110 110 110 110 110 110 110 110 110 110 110 110 110 Table 8-3: The means and (standard deviations) for mycorrhizal and benomyl-treated effects on height change, dry weight, diameter change and foliar Zn after the feeding experiment. Different letters in a column indicate statistically significant differences at p<0.05. (n=55) Mycorrhizal Treatment Mycorrhizal Benomyl-Treated Height Change (cm) 29.7 (12.10) b 34 .6 (11.67) a Dry Weight (g) 67. 7 (22.03) b 75.0 (19.01) a Diameter Change (mm) 3 .93 (1.78) b 4.42 (1.24) a Foliar Zn (ug/g) 19.31 (6.56) b 21.39 (5.49) a 170 Table 8-4: The means and (standard deviations) for concentrations of Ca, Cu and Zn in foliage after the feeding experiment. Ortho is orthophosphate, phyt is phytic acid, glyc is glycerophosphate and pyro is pyrophosphate. L is 10 mg P/1, H is 50 mg P/1. Different letters in a column indicate statistically significant treatment effects at p<0.05. (n=10) Treatment No P Ortho L Ortho H ATP L ATP H Phyt L Phyt H Glyc L Glyc H Pyro L Pyro H Foliar ( (ug/g) 9774.8 (871.9 10950.4 (5650.3 10768.8 (1898.4 9533.9 (1625.4 9207.0 (1502.9 8988.6 (1348.9 9323.6 (1665.8 14585.3 (4484 . 8 11778.5 (1678.7 11694.9 (2214.8 12413.5 (3438.3 2a. b b b b b c b a ab ab ab Foliar Cu (ug/g) 6.7 (1.3) be 8.2 (4.4) abc 8.9 (2.7) abc 6.7 (1.4) be 8 .1 (1.1) abc 6.5 (1.2) c 7.7 (1.6) abc 9.8 (3.3) a 9.3 (2.1) ab 8.1 (1.3) abc 9.1 (3.1) ab Foliar Zn (ug/g) 17.2 (3.3) b 18.5 (6.7) ab 18.2 (4.3) ab 17.6 (2.8) b 21.5 (5.3) ab 17.6 (2.1) b 17.9 (4.9) b 26 .2 (10.3) a 20.5 (5.9) ab 22 .6 (2.8) ab 26.1 (6.5) a concentration was 118.0 ug/g. Significant treatment effects occurred for foliar Ca, Cu and Zn (Tables 8-2, 8-4). The highest foliar Ca concentrations were seen with both levels of 171 glycerophosphate and pyrophosphate. The plants fed the low level of phytic acid had the lowest foliar Ca concentrations. Foliar Cu was highest when glycerophosphate, at either level, was the P source, and lowest when the trees received the low rates of phytic acid and ATP, or no P. Foliar Zn was highest with the low glycerophosphate and high pyrophosphate treatments, and lowest when the plants were fed no P, low ATP and both levels of phytic acid. There was also a significant mycorrhizal effect for foliar Zn (Tables 8-2, 8-3). Higher concentrations of Zn were found in the foliage of benomyl-treated plants than mycorrhizal plants. There were significant TREATMENT*MYCORRHIZA interactions for foliar N and foliar P (Tables 8-2, 8-5) . The benomyl-treated plants had higher foliar N concentrations for every treatment except the high rate of ATP. The benomyl-treated plants receiving the low rate of phytic acid had the highest N concentrations; the mycorrhizal trees fed high levels of orthophosphate had the lowest foliar N concentrations. Generally, there were no significant differences in foliar P between mycorrhizal and benomyl-treated plants for each P treatment. The exception was the low rate of orthophosphate: the mycorrhizal trees had higher P concentrations than the benomyl-treated seedlings. Foliar P concentrations were higher when the trees were given the higher rate of the P source for all but phytic acid. The low rates of ATP, phytic acid and pyrophosphate resulted in foliar P concentrations which were not significantly different from the controls receiving no P. Significant TREATMENT*MYCORRHIZA interactions were also found 172 for mycorrhizal colonization, soil P and root acid phosphatase activity (Tables 8-6, 8-7). Colonization was generally higher in the mycorrhizal plants than in the benomyl-treated plants for each treatment, although significant differences were seen only for the high rates of glycerophosphate and pyrophosphate. The highest level of colonization was in the mycorrhizal treatment without P; the lowest in the benomyl-treated trees fed high rates of orthophosphate, glycerophosphate and pyrophosphate. The concentration of available soil P at the end of the feeding experiment was highest when the plants were fed the high rate of glycerophosphate, pyrophosphate or orthophosphate. Available P was lowest in the soil receiving the low rate of phytic acid (mycorrhizal and benomyl-treated). The mycorrhizal low phytic acid treatment was significantly lower in available soil P than the no P mycorrhizal treatment. The mean soil pH in the pots at the end of the feeding experiment was 5.56. Root acid phosphatase activity, in ug p-nitrophenol/g of roots/hour, was highest in the mycorrhizal plants receiving the low rate of orthophosphate and the high rate of ATP, and in the benomyl-treated seedlings fed the low rate of glycerophosphate and the high rate of orthophosphate (Tables 8-6, 8-7) . The other treatments were not significantly different from one another. Soil acid phosphatase activity was significantly higher in the pots containing benomyl-treated plants than in the pots of mycorrhizal trees (Tables 8-7, 8-8). A significant treatment effect was also observed for root alkaline phosphatase activity (Tables 8-7, 8-9). 173 Table 8-5: The means and (standard deviations) for foliar N (%) , and foliar P (ug/g) after the feeding experiment. Ortho is orthophosphate, phyt is phytic acid, glyc is glycerophosphate and pyro is pyrophosphate. L is 10 mg P/1, H is 50 mg P/I. M is mycorrhizal, NM is benomyl-treated. Different letters for a parameter (eg foliar N) , including M and NM indicate statistically significant treatment effects at p<0.05. (n=5) Treat- ment No P Orth L Orth H ATP L ATP H Phyt L Phyt H Glyc L Glyc H Pyro L Pyro H Foliar N j Foliar N M j NM 2.37 I 2.43 (0.37) be j (0.33) b 1.72 i 2.59 (0.37) ij (0.25) a 1.67 I 1.87 (0.20) j ! (0.29) hi 2.12 i 2.61 (0.14)def j (0.31) a 2.11 i 2.21 (O.lO)defg (0.13) de 2.06 i 2.59 (0.19) fg (0.15) a 2.06 i 2.40 (0.08) fg | (0.18) b 2.11 ! 2.4 0 (0.14)defg j (0.19) b 1.90 ! 2.08 (0.19) h j (0.21) efg 2.01 i 2 . 67 (0.34) fg j (0.28) a 1.96 ! 2.25 (0.16) gh j (0.37) cd Foliar P j Foliar P M | NM 581.8 i 534.6 (73.6) fg ! (86.0) g 991.9 ! 556.1 (385.8) cdef (121.3) g 1554.5 i 1346.2 (288.6) abc j (264.1)abcd 694.1 | 727.7 (132.4) efg j (109.7) efg 1401.0 ! 1139.4 (326.4) abed (367.1) bede 581.5 i 606.9 (86.5) fg j (50.6) fg 1138.4 i 861.9 (180.2) bede ; (158.1) defg 1167.1 ! 743.7 (410.7) bede \ (142.9) efg 2034.1 i 1555.7 (422.2) a (309.2) abc 874.6 i 680.6 (90.0) defg j (166.1) efg 1615.1 i 1743.4 (360.4) abc j (497.7) ab 174 Table 8-6: Means and (std. deviations) for colonization (%) , soil P (mg/g) and root acid phosphatase (Rt Ac Pase) activity after the feeding experiment. Orth=orthophosphate, phyt=phytic acid, glyc=glycerophosphate, pyro=pyrophosphate. L=10 mg P/1, H=50 mg P/I. M is mycorrhizal, NM is non-mycorrhizal. Different letters for a parameter, including M and NM indicate statistical significance at p<0.05. (n=5) Treatm ent No P Orth L Orth H ATP L ATP H Phyt L Phyt H Glyc L Glyc H Pyro L Pyro H Col M 1 38.6 (10.9) a 24.8 (7.9) abed 17.0 (7.0) abedef 26.8 (10.2) abc 8.4 (3.6) efgh 15.4 (5.0) bedef 28 .2 (14.2) ] abc .! 35.8 ! (13.1) i ab 20.0 ! (7.9) J abede ! 27.2 (13.0) abc 33 .4 (13.4) abc Col NM 18 .4 (7.5) abede 10.6 (4.7) defgh 6.8 (2.2) fgh 28.0 (7.1) abc 15.9 (6.1) bedef 17.0 (5.5) abedef 14.2 (5.3) cdefg 17.0 (3.9) abedef 5.8 (2.2) gh 14.6 (5.6) cde 5.2 (1.3) h Soil P M 16.2 (2.0) cd 18.6 (5.1) be 31.0 (5.5) a 11.2 (1.8) de 26.8 (6.3) ab 9.7 (3.0) e 14.2 (2.3) cde 15.7 (2.1) cd 28.3 (3.5) ab 13.7 (1.6) cde 29.2 (3.6) a Soil P NM 13.3 (1.5) cde 11.6 (0.9) cde 30.1 (7.8) a 12.0 (1.9) cde 26.2 (4.7) ab 10.3 (0.8) de 10.1 (2.6) e 13 . 7 (2.2) cde 30.9 (8.8) a 12.1 (1.3) cde 26.6 (2.8) ab Rt Ac Pase M 538 (122) cde 1192 (160) a 592 (116) cde 586 (374) cde 933 (35) abc 492 (188) e 722 (349) bede 376 (17) e 538 (111) e 535 (192) e 575 (416) de Rt Ac PaseNM 520 (67) e 380 (61) e 540 (156) e 628 (353) bede 427 (239) e 412 (26) e 701 (73) bede 954 (135) ab 541 (110) e 581 (173) cde 906 (612) abed 175 Table 8-7: Analysis of variance results for percent colonization, soil P at the end of the feeding trial, root and soil alkaline phosphatase (Pase), root and soil acid phosphatase, root and soil diesterase, and root and soil pyrophosphatase. A * indicates significance at p<0.05. Colonization Soil P Root AlkPase Soil AlkPase Root AcidPase Soil AcidPase Root DiesPase Soil DiesPase Root PyroPase Soil PyroPase Treat 0.000 0.000 0.000* 0.078 0.664 0.060 0.659 0. 963 0.063 0.602 Myc 0 .000 0.005 0.803 0.319 0.491 0.000* 0.662 0.387 0.893 0.477 Treat*Myc 0.000* 0.036* 0.425 0.998 0 . 034* 0.301 0.548 0.933 0.578 0.268 Mult R2 0.720 0.851 0.735 0.498 0.589 0.801 0.424 0.226 0.568 0.491 N 110 110 44 44 44 44 44 44 44 44 Table 8-8: The means and (standard deviations) for mycorrhizal and benomyl-treated effects on soil acid phosphatase activity (ug p-nitrophenol/g/h) after the feeding experiment. Different letters in a column indicate statistically significant differences at p<0.05. (n=55) Mycorrhizal Treatment Mycorrhizal Benomyl-Treated Soil Acid Phosphatase 14.3 (9.3) b 106.0 (140.5) a The highest root alkaline phosphatase activity was associated with plants fed the high rate of phytic acid, both rates of glycerophosphate, and the low rate of pyrophosphate. Root alkaline phosphatase activity was lowest in the no P, low orthophosphate and low ATP treatments. There were no significant differences among 176 the treatments in the activities of soil alkaline phosphatase, root and soil diesterase, and root and soil pyrophosphatase. The mean soil alkaline phosphatase activity was 20.1 ug p-nitrophenol/g/h. The mean soil diesterase activity was 7.9 ug p-nitrophenol/g/h, and that of roots was 303.0 ug p-nitrophenol/g/h. The mean soil pyrophosphatase activity was 3.7 ug P/g/h, and the mean root pyrophosphatase activity was 29.7 ug P/g/h. The variability was high within and among the P treatments for these enzyme assays. Table 8-10 shows the correlation matrix for the results from this greenhouse study. Root acid phosphatase activity was positively correlated with root diesterase activity and with diameter change. The activities of soil and root alkaline phosphatase were positively correlated, and there was a positive relationship between soil alkaline phosphatase and soil pyrophosphatase. The diameter and height at the start of the feeding experiment were positively correlated, as were the changes in diameter and height at the end. Dry weight correlated positively with diameter change, foliar P and soil P. Foliar Ca, Cu, Mg and Zn were positively correlated with one another, and foliar P was positively correlated with soil P and with foliar Ca, Cu and Zn. There was a high positive correlation of foliar Al to foliar Fe. Foliar N was negatively correlated with soil P and foliar P. 177 Table 8-9: The means and (standard deviations) for root alkaline phosphatase activity (ug p-nitrophenol/g/h) after the feeding experiment. Ortho is orthophosphate, phyt is phytic acid, glyc is glycerophosphate and pyro is pyrophosphate. L is 10 mg P/l, H is 50 mg P/2. Different letters in a column indicate statistically significant treatment effects at p<0.05. (n=10) Treatment No P Ortho L Ortho H ATP L ATP H Phyt L Phyt H Glyc L Glyc H Pyro L Pyro H Root Alkaline Phosphatase 82.2 (39.3) b 97.7 (2.6) b 141.8 (21.2) ab 120.6 (17.0) b 197.1 (69.7) ab 159.1 (64.6) ab 217.6 (45.1) a 273.8 (57.8) a 257.5 (122.9) a 259.5 (20.5) a 155.1 (79.3) ab 178 Table 8-10: Correlation matrix. Acidrt and acidsl are root and soil acid phosphatase activity; alkrt and alksl are root and soil alkaline phosphatase activity; col is colonization; diamch is change in diameter; diamst is diameter at start; diesrt and diessl are root and soil diesterase activity; drywt is dry weight; folal, folca, folcu, folfe, folmg, foln, folp and folzn are foliar Al, Ca, Cu, Fe, Mg, N, P and Zn; htchg is height change; htst is height at start; pyrort and pyrosl are root and soil pyrophosphatase activity; and soilP is the soil P C o n t e n t . ,, g^ocnincninfviNin-N u § CTi CO CO CO Q — vO CO T-CTi CO <*• - ' O o o o o ' d d o ' d d o T i i i i I X o € O ~ H 0 ^ O O O ~ « ~ * O O O O — o o o o o o o o o d d d < I I I 1 I I o~HC^(M(T)(T)cr>QrJtotr}r^crv — g — j2cofnocpv2y5mr\j^(\i — — o o o o o o o o o d d d d I I I O ! D f V i m - T O C 0 ( \ ; N - « D \ i L 0 O j D ' > t J ( T ) C r . N < D N i n i O O I ^ i n — 0 0 « 0 - H O O O O O O O ( \ I - H O - o o o o o o o d d d d d d d < I I I II g"̂ «oryococriCD-H-H'<finî -.o - o o o o o o o o o o ' d d d d d I i I I I i I i 7 P,£"}S<T)ffi!S.£:C32l0<1,>fcn-«-~o-r Sfcu,^02iQffiC','rtfm-«(T)3cD O O - W H O O O O O O O O W N Q O - • o o o o o o o o d d d d d d d d S v S ( Bs! i r t ' H l ' : ' ISoa3vDocDQo-<cn o i 2 f S S S S S 0 S S l n S r j « B N ! a N O H N O N O O n O O - O l H t O H O r a — o o o d o o ' o ' d d d c i d d d d d d O \D O C O O O N -*- O — CB o m o O \D \ f l ^c Q C 1 N N O O f\l o - o d d i O N I M C O f a vD — •*• <\l O — o - H — — d d d o § — — \X> CD \n en — N en o o o o o o — -< d d d d d o m — - « m o & o o o O « C N o - r y K i - o o o o o o ' d d d d d d d d d d d d 1  i l l i i i i i i i § f\ l C "«- CO C M ^ C* <\! N N ~H CO n a o n o o — o d d d d d o o o j o - < - > - . o o c n o - - I o - 5 o o o o V - . or\jocnco(\i'>tir) C - V O — — S — -V — o o d d d d d O - C W O O ( \ J O - — m r o o S o c M i y S o K i k J B O L O ( T ) ' ^ - ( \ J 5 D ' ^ - C O O — c o o o o o d d " , ? ? ? ? ° 0 ' ° ° ? ? ° ° o ° o ' o d d d d d ' ' ' ' I i i i i ( i i l i t C d c t D i p a o t m - - o e d d d d d d d *§ „ . ~ ^ ™ ° ^ * ^ ~ < : w . ° S F 5 2 2 o , j ; g a g g 2 - 8 C J D 2 . § S S O S ! 2 R ! o « o o o o o o d d d d d d d d d d d d d d d d ^ IS si l ts " l * 8 3 * ^ ^ " " - ^ o o « « .2.2 * * i ,?.9.P .P .9 «••«- .9 -£x L £ o •'—cooooooooo * <5 _ , S ^ ! ^ T3 <*•«*-<*- U. « . X TJ T3 T3 O -p X L L 0 a a 3 * en c a C cn-n 41 -H a U < - 6 - - N i l l L . « - * X ' ; ' ^ 0 9 " H l J - t l o b — 0 0 0 < ~ l t O - U X L L O a a 179 Discussion Benomyl, a systemic fungicide, was used to produce a 'non- mycorrhizal' control, and colonization rates were generally higher in mycorrhizal plants than benomyl-treated ones. Colonization rates were also influenced by P source, with the highest rates of colonization in mycorrhizal plants not fed P, and the lowest levels in plants receiving high rates of orthophosphate, glycerophosphate and pyrophosphate. Benomyl is known to decrease VA mycorrhizal colonization in many plant species (Bailey and Safir 1978; Rhodes and Larson 1981; Hale and Saunders 1982; Verkade and Hamilton 1983; Fitter 1986; Fitter and Nichols 1988; Sukarno et al. 1993), and is thought to be well-suited to generating non-mycorrhizal control plants (Fitter 198 6; Sukarno et al. 1993) . It reduces the number of living internal hyphae, arbuscules and living external hyphae, although the fungus is still able to colonize the root to some extent (Sukarno et al. 1993) . This was seen in the benomyl-treated seedlings of this study, which still had low levels of mycorrhizal colonization. A side-effect of benomyl treatment, which in retrospect should have been anticipated, was the apparent N enrichment of non-mycorrhizal plants. Benomyl contains a benzimidazole group, and N comprised 19.3% of the benomyl molecule. This appears to be supplying the non-mycorrhizal plants with extra N, at a calculated rate of 0.029 g N/pot at each feeding. While there has been no overt discussion of the effects of benomyl on 180 plant nutrition in the VA mycorrhizal literature, Verkade and Hamilton (1983) and Fitter (1986) both report improved plant growth, with increased foliar N concentrations, in benomyl-treated plants. Because benomyl was applied to the seedlings in this study for more than one year, the effects of benomyl on plant N many be more pronounced than in an experiment of shorter duration. The negative correlation of foliar N to foliar P and soil P probably also reflects the N-enrichment of benomyl-treated plants. The benomyl-treated plants were generally larger than the mycorrhizal trees, with greater increases in dry weight and diameter. This may be due to the N from benomyl, but may also reflect the cost to the plant for the benefits of mycorrhizal colonization: Marschner and Dell (1994) estimate that 10-20% of net photosynthates are required for formation, maintenance and function of mycorrhizal structures. Shoot and root dry weights are usually increased by inoculation with VA mycorrhizal fungi (Harley and Smith 1983; Tarafdar and Marschner 1994), but this is usually relative to completely non-mycorrhizal plants, which may be severely P deficient. Since the benomyl-treated trees were not completely uncolonized, P stress in these plants would be lessened relative to tree roots with no colonization. The increased growth of the benomyl-treated plants suggests that the cedar trees used in this experiment were N-limited. The source of phosphorus had a major influence on plant growth and nutrition in this study, as did colonization, but to a lesser extent. The only significant difference between mycorrhizal and 181 benomyl-treated seedlings for foliar P occurred with the low orthophosphate treatment: the mycorrhizal plants had higher foliar P concentrations than benomyl-treated plants. Generally, foliar P was higher with higher rates of each P source, and the low rates of ATP, phytic acid and pyrophosphate resulted in foliar P concentrations which were not significantly different from the no P control. There appears to be some P available to No P plants as well, because plant height and dry weight increased with this treatment. Although available P was not measured in the growth medium prior to the start of the experiment, Jayachandran et al. (1992) indicate that the available P concentration of a similar growth medium was 14 mg/kg. This, combined with translocation, prevented these trees from showing severe P deficiency symptoms. The poorest tree growth resulted from the phytic acid treatments, especially at the low application rate. These trees had the lowest foliar P concentrations, as well as low levels of foliar Ca, Zn and Cu. In animal nutrition, phytic acid is known to form complexes with essential metals such as Ca, Zn, Fe, Mg, Mn and Cu, causing deficiencies (Cosgrove 1980; O'Neill et al. 1980). This also appears to be occuring with the cedar in this experiment. As a calcium accumulator (Weetman et al. 1988), cedar is especially sensitive to Ca availability, and the symptoms which appeared in the trees receiving the low phytic acid treatment were consistent with Ca deficiency (Weetman et al. 1988) . In the literature, some researchers report that plants and mycorrhizae can use phytic acid 182 as a P source (Saxena 1964; Helal 1990; Mitchell and Read 1981; Adams and Pate 1992; Antibus et al. 1992; Pasqualini et al. 1992; Jayachandran et al. 1992; Tam and Griffiths 1993; Tarafdar and Marschner 1994), but others report that the P of phytic acid is not available to plants and mycorrhizae (Thomas et al. 1982; Adams and Pate 1992; Barrett-Lennard et al. 1993) . There were no reports of other nutrient deficiencies with phytic acid as a P source, possibly because of the increased sensitivity of cedar to Ca deficiencies, or the shorter duration of other experiments. Interestingly, cedar grew better with the high rate of phytic acid than the low rate, which is the reverse of what would be expected if growth problems were caused by the complexing of other nutrients. Findenegg and Nelemans (1993) also report better growth with high levels of phytic acid relative to low levels. This may be related to enzyme induction, which will be discussed in more detail below. Although phytic acid and other inositol hexaphosphates are relatively abundant in soils (Stevenson 1986), they are adsorbed to soil surface and are probably not readily available to plants. In this experiment, Na-phytate was used, which is soluble, while in soil the insoluble Fe- and Al-phytates are more common (Read 1991), making inferences about phytic acid as a P source in soil difficult from this study. The greatest plant growth was achieved with the high rate of ATP. The plants in this treatment also contained foliar N concentrations than other treatments, suggesting that the improved 183 growth was due to N as much or more than P from the ATP. The trees were also able to use glycerophosphate and inorganic pyrophosphate, which are known to provide a P source to plants and mycorrhizae (Bartlett and Lewis 1973; McKercher and Tollefson 1978; Beever and Burns 1980; Thomas et al. 1982; Tarafdar and Classen 1988; Pasqualini et al. 1992; Jayachandran et al. 1992; Barrett- Lennard et al. 1993; Tarn and Griffiths 1993; McElhinney and Mitchell 1993). Growth was improved with all P sources when they were supplied at the higher rate. The use of organic P compounds by western red cedar appears to be facilitated by phosphatase enzymes, as shown by the correlation of root acid phosphatase to the change in plant height and diameter. Enzyme activities were higher on roots than in bulk soil, suggesting that these enzymes are produced by roots, mycorrhizae or rhizosphere microorganisms. It should be noted that the "root" phosphatases discussed here are more accurately "rhizosphere" phosphatases, as the roots were not washed prior to enzyme assay. Because this experiment was not conducted under sterile conditions, these root phosphatase assays reflect enzyme activities of plant roots, mycorrhizae and rhizosphere microorganisms. In an attempt to separate rhizosphere and root phosphatase activity, a duplicate portion of the roots was shaken in an antibiotic solution prior to the enzyme assays. Unfortunately, this produced results which were highly variable, with no obvious patterns, and so these data were not included in 184 this report. "Soil" phosphatase activities were measured in the bulk soil, and were all very low. Rhizosphere phosphatase activities tend to be higher than activities remote from plant roots, due to greater microbial numbers in the rhizosphere and possibly because of higher phosphatase activities of rhizosphere organisms and the excretion of plant root enzymes (Spier and Ross 1978) . Phosphatases are inducible by phosphate depletion (Goldstein 1992) . They are high in soils with high organic C and organic P contents (Appiah and Thomas 1982; Rojo et al. 1990; Baligar et al. 1991), are sensitive to soil moisture levels (Spier and Cowling 1991; Baligar et al. 1991), and are strongly inhibited by inorganic phosphate (Spier and Ross 1978). Higher plants do not produce alkaline phosphatases (Speir and Ross 1978; Tabatabai 1982). In this study, enzyme differences between mycorrhizal and benomyl- treated seedlings were found only for root acid phosphatases, and these were influenced by P source. The P source also affected the activities of root alkaline phosphatase, while root diesterases and pyrophosphatases were present but were not significantly influenced by P treatment. There are varying reports in the literature as to the production of alkaline phosphatases by mycorrhizae. All types of mycorrhizae have been shown to produce acid phosphatase (eg Read 1991; Dodd et al. 1987), but alkaline phosphatase activity has been reported by only a few researchers (Gianinazzi-Pearson and Gianinazzi 1976, 1978; Ho and Zak 1979; Krishna et al. 1983; 185 Kapoor et al. 1988; Tarafdar and Classen 1988; Bae and Barton 1989; Read 1991; Pasqualini et al. 1992; McElhinney and Mitchell 1993; Thiagarajan and Ahmed 1994; Tarafdar and Marschner 1994). In VA mycorrhizae, alkaline phosphatases specific to the VA mycorrhizae and of suspected fungal origin have been reported (Gianinazzi- Pearson and Gianinazzi 1978). Thiagarajan and Ahmed (1994) found extra peaks for mycorrhizal but not non-mycorrhizal alkaline and acid phosphatases by column chromatography. There are also reports of ATPases in plants and mycorrhizae (Tikhaya et al. 1990; McArthur and Knowles 1993), which may be responsible for the hydrolysis of ATP in the present study. Both P source and rate of application influence phosphatase activity in this study. Acid phosphatase is higher in mycorrhizal plants, and in mycorrhizal plants is stimulated by low orthophosphate and high ATP, while in benomyl-treated seedlings it is stimulated by low glycerophosphate and high pyrophosphate. Alkaline phosphatase activity was increased by glycerophosphate, low levels of pyrophosphate and high levels of phytic acid. The hydrolysis of phytic acid by alkaline phosphatases at high levels may account for increased growth despite nutrient deficiencies, as previously discussed. The enzyme may be inducible only when there is sufficient substrate present, and hydrolysis may release some of the Ca, Zn and Cu, in addition to P. The variability in enzyme activity with the different levels of P substrates suggests that a number of enzymes may be operating as acid and alkaline 186 phosphatases, each induced with different rates of substrate. In a study of phosphatase activities in soils of Douglas-fir and hemlock stands of southern Vancouver Island, Pang and Kolenko (1986) reported that phosphomonoesterase activity was highest in the forest floor, and decreased with depth and with fertilization. Neutral phosphatases were present on the sites, as well as alkaline phosphatase activity under Douglas-fir. In the current study, phosphatase activity was not measured in the soils of the research sites on northern Vancouver Island. It is apparent from these results that cedar can access P0 compounds, with or without mycorrhizae. Plants grown under sterile conditions, without mycorrhizae,can use organic P forms such as phytic acid if the P form is supplied at a sufficiently high rate (Findenegg and Nelemans 1993). Lupins (Lupinus spp.), which never form mycorrhizae, can also obtain P from organic P sources (Adams and Pate 1992). As noted by Jayachandran et al. (1992), the P sources which reduce colonization of mycorrhizal plants tend to stimulate growth and P uptake in non-mycorrhizal plants. Apparently, mycorrhizal plants benefit from access to P0, but a P0 source which can be used directly by the plant may reduce reliance on the symbiosis. As cedars in forests are always colonized by VA mycorrhizae (Curran and Dunsworth 1988), it may be that soil levels of nutrients, both inorganic and organic, are too low for the plant to access directly. One drawback to the current study is that it is impossible to separate the effects of rhizosphere microorganisms from those of plants or mycorrhizae. However, in the forest, cedar 187 would also have associated rhizosphere organisms, so the results from this study may better reflect field conditions than a study conducted under aseptic conditions. In forests, synergistic interactions between bacteria and VA mycorrhizae may be important in facilitating P mobilization (Read 1991). Some of the apparent differences in this study between mycorrhizal and benomyl-treated plants, such as elevated Zn concentrations in benomyl-treated trees, may be due to the improved N nutrition of the benomyl-treated trees from the benomyl treatments. Unfortunately, this may have also affected some differences in plant P. Future studies using benomyl to produce non-mycorrhizal controls would be well-advised to compensate for the increased N from benomyl by supplementing the mycorrhizal plants. Conclusions Western red cedar mediated the hydrolysis of the organic P compounds glycerophosphate and ATP, and inorganic pyrophosphate, particularly when these compounds were supplied at high rates. The trees grew very poorly with phytic acid, which may have complexed Ca, Zn and Cu from the nutrient solutions, inducing Ca deficiencies. Although phophatases were produced, it was impossible to distinguish those of the plant and mycorrhizae from the enzymes of rhizosphere microorganisms. Mycorrhizal colonization improved foliar P content, but other mycorrhizal effects may have been masked by N enrichment from benomyl, the fungicide used to produce non-mycorrhizal control trees. 188 CHAPTER NINE GENERAL CONCLUSIONS This study was initiated to investigate the role of phosphorus in the growth check problem of trees replanted onto cedar-hemlock (CH) stands after logging and slash-burning, and to investigate the use of organic P forms by western red cedar, which did not experience a growth check on these siges to the same degree as other species such as western hemlock and Sitka spruce. Prior to the onset of this thesis research, it was hypothesized that some procedures to determine organic, total and available P might be better suited to the Orthic Ferro-Humic Podzols of the CH and HA forests of northern Vancouver Island than would other procedures. The results of this study indicate that the Parkinson and Allen digestion procedure removes more total P from soil samples than the Saunders and Williams ignition and extraction method. Both the Saunders and Williams method and the Bowman and Moir extraction procedure appeared to overestimate organic P in these soils. This was established by a comparison of the P-extraction techniques to the P forms revealed by 31P NMR spectroscopy. Either of the Bray PI or Mehlich 3 procedures were suitable to determine available P. A second hypothesis was that some extraction procedures for 31P NMR spectroscopy might be better suited to these soils than other extraction procedures. It would appear from this reasearch than an ideal soil extractant for 31P NMR does not yet 189 exist. The new extraction procedure for organic P developed by Bowman and Moir, which had not been used as an extractant for 31P NMR spectroscopy prior to this study, shows great promise because it extracts more of the total organic P than do the convential extractants NaOH and Chelex. However, this extractant also maintains paramagnetic ions such as Mn in solution, causing line broadening and reducing the quality of the spectra. Thus a compromise must be made to obtain reasonably interpretable spectra and representativeness of the organic phosphorus forms. A third hypothesis of this thesis was that the soils of the mature CH forests contained different P forms from the mature HA stands, as well as differences in other aspects of the soil chemistry. This study revealed that, although significant differences exist between the two forest types for pH, C concentration, loss on ignition, C/N and C/P ratios and extractable LCa concentration, significant differences in P concentration and forms did not exist. The fourth hypothesis of this research was that the soils of the CH stands 10 years, 5 years and immediately after burning contained different P forms and concentrations from one another and from the mature stands, as well as differences in other aspects of the soil chemistry. Immediately after clear-cutting and burning, the soils of the CH forest types experienced an ashbed effect, which temporarily increased the pH and the concentrations of available P, Ca, Mg, and Mn in the surface 190 horizons. By 10 years after cutting and burning, the concentrations of these nutrients had returned to preburn levels. However, significant decreases in organically-bound Fe and Al, and organic P, were observed in mineral horizons 10 years postburn. These decreases were attributed to altered illuviation patterns in these podzolic soils from the removal of organic matter by burning. Although total P concentrations were unchanged, there was a shift from organic P forms to inorganic P forms, and changes in P forms with time at depth in the profile. This may have a significance in altering the movement of sesquioxides through the soil profile and the relationship of organic P to inorganic P at lower depths in the soil. This would indicate that more orthophosphate may accumulate in B horizons after burning. Over time, however, the effect of the burn becomes less pronounced. The final hypothesis of this research was that western red cedar were able to use organic P forms in addition to, or instead of, inorganic forms, and that mycorrhizal trees used different P forms from non-mycorrhizal trees. Although N enrichment of the 'non-mycorrhizal' seedlings from benomyl interfered wtih the results of this study, the trees were able to grow when supplied with the organic compounds glycerophosphate and ATP, and the inorganic compounds KH2P04 and pyrophosphate, especially when these compounds were supplied at high rates. Phosphatases were produced, but it was impossible to distinguish those of the plant and mycorrhizae from the 191 enzymes of rhizosphere microorganisms. This study suggests a number of research directions. Phosphorus-31 NMR spectroscopy should be tied into conventional fractionation and partitioning techniques, to further our understanding of soil organic matter. To further develop extraction techniques for 31P-NMR, other combinations of extractants and resins could be tried, in an attempt to achieve the high percentage of total P extracted by the NaOH-EDTA procedure, but without the concommitant extraction of paramagnetic ions. It would be very interesting to resample the 5-year and 10-year postburn sites used in this study in 5 or 10 years, to determine if the 5-year sites would eventually resemble the 10-year sites in soil chemistry, and to determine how long-term the effects of burning are. The results of the greenhouse study also warrant further investigation, in a simpler study with better controls on rhizosphere organisms, and perhaps with a different fungicide to control mycorrhizal colonization. 192 LITERATURE CITED Adams, M. A. 1990. 31P-NMR identification of phosphorus compounds in neutral extracts of mountain ash (Eucalyptus regnans F. Muell.) soils. Soil Biol. Biochem. 22: 419-421. Adams, M. A., and L. T. Byrne. 1989. 31P-NMR analysis of phosphorus compounds in extracts of surface soils from selected karri (Eucalyptus diversicolor F. Muell.) forests. Soil Biol. Biochem. 21: 523-528. Adams, M. A., and J. S. Pate. 1992. Availability of organic and inorganic forms of phosphorus to lupins (Lupinus spp.). Plant Soil 145: 107-113. Agriculture Canada Expert Committee on Soil Survey. 1987. The Canadian System of Soil Classification. Second edition. Agric. Can. Publ. 1646. 164 pp. Ahlgren, I. F., and C. E. Ahlgren, eds. 1960. Ecological effects of forest fire. Bot. Rev. 26: 483-533. Alexander, I. J., and K. Hardy. 1981. Surface phosphatase activity of Sitka spruce mycorrhizas from a serpentine site. Soil Biol. Biochem. 13: 301-305. Anderson, G. 1975. Other organic phosphorus compounds. pp. SOS- SSI. In: Gieseking, J. E., ed. Soil Components. Vol. 1: Organic Components. Springer-Verlag. New York. Anderson, G. 1970. The isolation of nucleoside diphosphates from alkaline extracts of soil. J. Soil Sci. 21: 96-104. Antibus, R. K. , C. J. Kroehler and A. E. Linkins. 1988. The effects of external pH, temperature and substrate concentration on acid phosphatase activity of ectomycorrhizal fungi. Can. J. Bot. 64: 2383-2387. Antibus, R. K. , R. L. Sinsabaugh and A. E. Linkins. 1992. Phosphatase activities and phosphorus uptake from inositol phosphate by ectomycorrhizal fungi. Can. J. Bot. 70: 794-801. Antibus, R. K. , and P. Lesica. 1990. Root surface acid phosphatase activities of vascular epiphytes of a Costa Rican rain forest. Plant Soil 128: 233-240. Appiah, M. R., and R. L. Thomas. 1982. Inositol phosphate and organic phosphorus contents and phosphatase activity of some Canadian and Ghanaian soils. Can. J. Soil Sci. 137: 7-15. 193 Baath, E., and B. Soderstrom. 1979. Fungal biomass and fungal immobilization of plant nutrients in Swedish coniferous forest soils. Rev. Ecol. Biol. Sol. 16: 477-489. Bae, K.-S., and L. L. Barton. 1989. Alkaline phosphatase and other hydrolases produced by Cenococcum grant forme, an ectomycorrhizal fungus. Appl. Envir. Microb. 55: 2511-2516. Bailey, J. E., and G. R. Safir. 1978. Effect of benomyl on soybean endomycorrhizae. Phytopathology 68: 1810-1812. Baker, R. T. 1975. A new method for estimating the phospholipid content of soils. J. Soil Sci. 26: 432-436. Baligar, V. C , R. J. Wright and M. D. Smedley. 1991. Enzyme activities in Appalachian soils. 3. Pyrophosphatase. Commun. Soil Sci. Plant Anal. 22: 1537-1545. Ballard, T. M., and R. E. Carter. 1986. Evaluating forest stand nutrient status. BC Ministry of Forests Land Management Report No. 20. 6 0 pp. Barrett-Lennard, E. G., M. Dracup and H., Greenway., 1993. Role of extracellular phosphatases in the phosphorus nutrition of clover. J. Exp. Bot. 44: 1595-1600. Barrow, N. , J. 1978. The description of phosphate adsorption curves. J. Soil Sci. 29: 477-462. Bartlett, E. M., and D. H. Lewis. 1973. Surface phosphatase activity of mycorrhizal roots of beech. Soil Biol Biochem. 5: 249-257. Basha, S. M. 1984. Purification and characterization of an acid phosphatase from peanut (Arachis hypogaea) seed. Can. J. Bot. 62: 385-391. Battigelli, J. P., S. M. Berch and V. G. Marshall. 1994. Soil fauna communities in two distinct but adjacent forest types on northern Vancouver Island, British Columbia. Can. J. For. Res. 24: 1557-1566. Beaton, J. D. 1957. The influence of burning on the soil in the timber range area of Lac Le Jeune, British Columbia. II. Chemical properties. Can. J. Soil Sci. 39: 6-11. Beever, R. E., and D. J. W. Burns. 1980. Phosphorus uptake, storage and utilization by fungi. Adv. Bot. Res. 8: 128-219. 194 Bhadoria, P. B. S., J. Kaselowsky, N. Claasen and A. Jungk. 1991. Soil phosphate diffusion coeffiecients: their dependence on phosphorus concentration and buffer power. Soil Sci. Soc. Am. J. 55: 56-60. Bieleski, R. L. 1973. Phosphate pools, phosphate transport and phosphate availability. Ann. Rev. Plant Physiol. 24: 225- 252. Bolan, N. S. 1991. A critical review on the role of mycorrhizal fungi in the uptake of phosphorus by plants. Plant Soil 134: 189-207. Bowman, R. A. 1989. A sequential extraction procedure with concentrated sulfuric acid and dilute base for soil organic phosphorus. Soil Sci. Soc. Am. J. 53: 326-366. Bowman, R. A. 1988. A rapid method to determine total phosphorus in soils. Soil Sci. Soc. Am. J. 52: 1301-1304. Bowman, R. A., and J. O. Moir. 1993. Basic EDTA as an extractant for soil organic phosphorus. Soil Sci. Soc. Am. J. 57: 1516- 1518. Boyle, J. R. 1973. Forest soil chemical changes following fire. Commun. Soil Sci. Plant Anal. 4: 369-374. Bray, R. H., and L. T. Kurtz. 1945. Determination of total, organic and available forms of phosphorus in soils. Soil Sci. 59: 39-42. Bremner, J., and M. A. Tabatabai. 1971. Use of automated combustion techniques for total carbon, total nitrogen and total sulfur analysis of soils. pp. 1-16. In: Walsk, L. M., ed. Instrumental Methods for Analysis of Soils and Plant Tissue. Soil Sci. Soc. Am., Inc. Madison, WI. Brockley, R. P., R. L. Trowbridge, T. M. Ballard and A. M. Macadam. 1992. Nutrient management in interior forest types. pp. 43- 54. In: Chappell, H. N., G. F. Weetman and R. E. Miller, eds. Forest Fertilization: Sustaining and improving nutrition and growth of western forests. Institute of Forest Resources Contribution No. 73. University of Washington. Seattle. Browman, M. G., and M. A. Tabatabai. 1984. Phosphatase activity of soils. Soil Sci. Soc. Am. J. 42: 284-290. Burford, J. R. , and J. M. Bremner. 1972. Is phosphate reduced to phosphine in waterlogged soils? Soil Biol. Biochem. 4: 489- 495. 195 Cade, B. J. 1989. Vesicular-arbuscular mycorrhizal fungi in winter wheat in south coastal British Columbia. M. Sc. Thesis. The University of British Columbia. Vancouver. Chang, S. C. , and M. L. Jackson. 1958. Soil phosphorus fractions in some representative soils. J. Soil Sci. 9: 109-118. Chang, S. C , and M. L. Jackson. 1957. Fractionation of soil phosphorus. Soil Sci. 84: 133-144. Chang, S. X. 1995. Soil nitrogen dynamics and fertilizer nitrogen efficiency and incorporation in western red cedar-western hemlock forest ecosystems. Ph. D. Thesis. University of British Columbia. In progress. Comerford, N. B., and M. F. Skinner. 1989. Residual phosphorus solubility for an acid, clayey forested soil in the presence of oxalate and citrate. Can. J. Soil Sci. 69: 111-117. Compton, J. E. 1994. Phosphorus dynamics in red alder and Douglas-fir Stands. Ph. D. Thesis. University of Washington. Seattle. 119 pp. Condron, L. M., E. Frossard, H. Tiessen, R. H. Newman, and J. W. B. Stewart. 1990a. Chemical nature of organic phosphorus in cultivated and uncultivated soils under different environmental conditions. J. Soil Sci. 41: 41-50. Condron, L. M. , J. 0. Moir, H. Tiessen and J. W. B. Stewart. 1990b. Critical evaluation of methods for determining total organic phosphorus in tropical soils. Soil Sci. Soc. Am. J. 54: 1261-1266. Condron, L. M., K. M. Goh and R. H. Newman. 1985. Nature and distribution of soil phosphorus as revealed by a sequential extraction method followed by 31P nuclear magnetic resonance analysis. J. Soil Sci. 36: 199-207. Cosgrove, D. J. 1980. Inositol Phosphates: Their Chemistry, Biochemistry and Physiology. Elsevier Scientific Publishing Co. New York. 191 pp. Cromack, K., Jr., P. Sollins, W. C. Graustein, K. Speidel, A. J. Todd, G. Spycher, C. Y. Li and R. L. Todd. 1979. Calcium oxalate accumulation and soil weathering in mats of the hypogeous fungus Hysterangium crassum. Soil Biol. Biochem. 11: 463-468. 196 Crosland, A. P., F. J. Zhao, S. P. McGrath and P. W. Lane. 1995. Comparison of aqua regia digestion with sodium carbonate fusion for the determination of total phosphorus in soils by inductively coupled plasma atomic emission spectroscopy (ICP). Commun. Soil Sci. Plant Anal. 26: 1357-1368. Cross, A. F., and W. H. Schlesinger. 1995. A literature review and evaluation of the Hedley fractionation: Applications to the biogeochemical cycle of soil phosphorus in natural ecosystems. Geoderma 64: 197-214. Curran, M. P. 1984. Soil testing for phosphorus availability to some conifers in British Columbia. M. Sc. Thesis. The University of British Columbia. Vancouver. 113 pp. Curran, M. P., and B. G. Dunsworth. 1988. Coastal western red cedar regeneration: problems and potentials. pp. 2 0-32. In: Smith, N. J. , ed. Western Red Cedar - Does it Have a Future? Conference Proceedings. University of British Columbia. Faculty of Forestry. David, C. A. 1987. Post-harvest forest floor changes and nitrogen mobilization in an Engelmann spruce-subalpine fir forest. Ph. D. Thesis. University of British Columbia. Vancouver. 325 pp. DeBano, L. F., and J. M. Klopatek. 1988. Phosphorus dynamics of Pinyon-Juniper soil following simulated burning. Soil Sci. Soc. Am. J. 52: 271-277. de Montigny, L. 1992. An investigation into the factors contributing to the growth check of conifer regeneration on northern Vancouver Island. Ph. D. Thesis. The University of British Columbia. Vancouver. 191 pp. de Montigny, L. E., C. M. Preston, P. G. Hatcher and I. Kogel- Knabner. 1993. Comparison of humus horizons from two ecosystem phases on northern Vancouver Island using 13C CPMAS NMR spectroscopy and CuO oxidation. Can. J. Soil Sci. 73: 9- 25. Dick, W. A., and M. A. Tabatabai. 1984. Kinetic parameters of phosphatases in soils and organic waste materials. Soil Sci. 137: 7-15. Dick, W. A., and M. A. Tabatabai. 1978. Inorganic pyrophosphatase activity of soils. Soil Biol. Biochem. 10: 59-65. Dick, W. A., and M. A. Tabatabai. 1977. An alkaline oxidation method for determination of total phosphorus in soils. Soil Sci. Soc. Am. J. 4: 511-514. 197 Dighton, J. 1983. Phosphatase production by mycorrhizal fungi. Plant Soil 71: 455-462. Dighton, J., and L. Boddy. 1988. Role of fungi in nitrogen, phosphorus and sulphur cycling in temperate forest ecosystems. pp. 269-298. In: Boddy, L., R. Marchant and D. J. Read, eds. Nitrogen, Phosphorus and Sulphur Utilization by Fungi. Cambridge University Press. Cambridge, UK. Dighton, J., and D. C. Coleman. 1992. Phosphorus relations of roots and mycorrhizas of Rhododendron maximum L. in the Southern Applachians, North Carolina. Mycorrhiza 1: 175-184. Dinkelaker, B., and H. Marschner. 1992. In vivo demonstration of acid phosphatase activity in the rhizosphere of soil-grown plants. Plant Soil 144: 199-205. Dodd, J. C , C. C. Burton, R. G. Burns and P. Jeffries. 1987. Phosphatase activity associated with the roots and rhizosphere of plants infected with vesicular-arbuscular mycorrhizal fungi. New Phytol. 107: 163-172. Dormaar, J. F., and G. R. Webster. 1964. Loss inherent in ignition procedures for determining total organic phosphorus. Can. J. Soil Sci. 44: 1-6. Dormaar, J. F., and G. R. Webster. 1963. Determination of total organic phosphorus in soils by extraction methods. Can. J. Soil Sci. 43: 35-43. Edmonds, R. L. 1980. Litter decomposition and nutrient release in Douglas-fir, red alder, western hemlock and Pacific silver fir ecosystems in western Washington. Can. J. For. Res. 10: 327-337. Ellis, R. C , and A. M. Graley. 1983. Gains and losses in soil nutrients associated with harvesting and burning eucalypt rainforests. Plant Soil 74: 437-450. Emsley, J, and S. Niazi. 1983. The analysis of soil phosphorus by ICP and 31P NMR spectroscopy. Phosphorus and Sulfur 16: 303-312. Entry, J. A., C. L. Rose and K. Cromack Jr. 1991. Litter decomposition and nutrient release in ectomycorrhizal mat soils of a Douglas fir ecosystem. Soil Biol. Biochem. 23: 285-290. 198 Enwezer, W. O. , and A. W. Moore. 1966. Comparison of two methods for determining organic phosphorus in some Nigerian soils. Soil Sci. 102: 284-285. Fares, F., J. C. Fardeau et F. Jacquin. 1974. Etude quantitative du phosphore organic dans differents types de sols. Phosphore et Agriculture 63: 25-40. Feller, M. C. 1982. The ecological effects of slashburning with particular reference to British Columbia: A literature review. BC Ministry of Forests Land Manage. Rep. No. 13. Victoria, BC. Feller, M. C , J. P. Kimmins and K. M. Tsze. 1983. Nutrient losses to the atmosphere during slashburns in southwestern British Columbia, pp. 128-135. In: Seventh Conference on Fire and Forest Meteorology. American Meteorological Society. Boston. Fernandez, D. S., and J. Ascencio. 1994. Acid phosphatase activity in bean and cowpea plants grown under phosphorus stress. J. Plant Nutr. 17: 229-241. Fernandez, I. J., and R. A. Struchtemeyer. 1985. Chemical characteristics of soils under spruce-fir forests in eastern Maine. Can. J. Soil Sci. 65: 61-69. Findenegg, G. R., and J. A. Nelemans. 1993. The effect of phytase on the availability of P from myo-inositol hexaphosphate (phytate) for maize roots. Plant Soil 154: 189-196. Fitter, A. H. 1986. Effect of benomyl on leaf phosphorus concentration in alpine grasslands: a test of mycorrhizal benefit. New Phytol. 103: 767-776. Fitter, A. H., and R. Nichols. 1988. The use of benomyl to control infection by vesicular-arbuscular mycorrhizal fungi. New Phytol. 110: 201-206. Fogel, R., and G. Hunt. 1983. Contribution of mycorrhizae and soil fungi to nutrient cycling in a Douglas fir ecosystem. Can. J. For. Res. 13: 219-232. Forster, J. C , and W. Zech. 1993. Phosphorus status of a soil catena under Liberian evergreen rain forest: results of 31P NMR spectroscopy and phosphorus adsorption experiments. Z. Pflanzenernahr. Bodenk. 156: 61-66. Fox, T. R. , and N. B. Commerford. 1992. Rhizosphere phosphatase activity and phosphatase hydrolysable organic phosphorus in two forested spodosols. Soil Biol. Biochem. 24: 579-583. 199 Fox, T. R., N. B. Commerford and W. W. McFee. 1990a. Kinetics of phosphorus release from spodosols: effects of oxalate and formate. Soil Sci. Soc. Am. J. 54: 1441-1447. Fox, T. R., N. B. Commerford and W. W. McFee. 1990b. Phosphorus and aluminum release from a spodic horizon mediated by organic acids. Soil Sci. Soc. Am. J. 54: 1763-1767. Frossard, E., J. W. B. Stewart and R. J. St. Arnaud. 1989. Distribution and mobility of phosphorus in grassland and forest soils of Saskatchewan. Can. J. Soil Sci. 69: 401-416. Furlani, A. M. C , R. B. Clark, J. W. Maranville and W. M. Ross. 1987. Organic and inorganic sources of phosphorus on growth and phosphorus uptake in Sorghum genotypes. J. Plant Nutr. 10: 163-186. Fyles, J. W., and B. Cote. 1994. Forest floor and soil nutrient status under Norway spruce and red pine in a plantation in southern Quebec. Can. J. Soil Sci. 74: 387-392. Garcia, M., and J. Ascencio. 1992. Root morphology and acid phosphatase activity in tomato plants during development of and recovery from phosphorus stress. J. Plant Nutr. 15: 2491-2503 . Gerke, J. 1992. Orthophosphate and organic phosphate in the soil solution of four sandy soils in relation to pH - evidence for humic-Fe (Al)-phosphate complexes. Commun. Soil Sci. Plant Anal. 23: 601-612. Germain, A. Y. 1985. Fertilization of stagnated Stika spruce plantations on northern Vancouver Island. M. F. Thesis. The University of British Columbia. Vancouver. 185 pp. Gianinazzi-Pearson, V., and S. Gianinazzi. 1976. Enzymatic studies on the metabolism of vesicular-arbuscular mycorrhiza. I. Effect of mycorrhiza formation and phosphorus nutrition on soluble phosphatase activity on onion roots. Physiol. Veg. 44: 833-841. Gianinazzi, S., V. Gianinazzi-Pearson and J. Dexheimer. 1979. Enzymatic studies on the metabolism of vesicular-arbuscular mycorrhiza. III. Ultrastructural localization of acid and alkaline phosphatase in onion roots infected by Glomus mosseae (Nichol. & Gerd). New Phytol. 82: 127-132. Gil-Sotres, F., W. Zech and H. G. Alt. 1990. Characterization of phosphorus fractions in surface horizons of soils from Galicia (N. W. Spain) by 31P NMR spectroscopy. Soil Biol. Biochem. 22: 75-79. 200 Giovanetti, M., and B. Mosse. 1980. An evaluation of techniques for measuring vesicular arbuscular mycorrhizal infection in roots. New Phytol. 84: 469-500. Goldstein, A. H. 1992. Phosphate starvation inducible enzymes and proteins in higher plants. In: Wray, J. L., ed. Inducible Plant Proteins. Cambridge University Press. Greenwood, a. J., and D. H. Lewis. 1977. Phosphatases and the utilization of inositol hexaphosphate by soil yeasts of the genus Cryptococcus. Soil Biol Biochem. 9: 161-166. Grier, C. C. 1975. Wildfire effects on nutrient distribution and leaching in a coniferous ecosystem. Can. J. For. Res. 5: 599-607. Hale, K. A., and F. E. Sanders. 1982. Effects of benomyl on vesicular-arbuscular mycorrhizal infection of red clover (Trifolium pratense L.) and consequences for phosphorus inflow. J. Plant Nutr. 5: 1355-1367. Hanafi, M. M., and J. K. Syers. 1994. Plant availability of two phosphate rock materials in acid Malaysian soils. Commun. Soil Sci. Plant Anal. 25: 3171-3189. Hance, R. J., and G. Anderson. 1963. Identification of hydrolysis products of soil phospholipids. Soil Sci. 96: 157-161. Hance, R. J., and G. Anderson. 1962. A comparative study of methods of estimating soil organic phosphate. J. Soil Sci. 13: 225-230. Harley, J. L. , and S. E. Smith. 1983. Mycorrhizal Symbiosis. Academic Press. New York. Harrap, F. E. G. 1963. Use of sodium EDTA in the determination of soil organic phosphorus. J. Soil Sci. 4: 82-87. Harris, P. J. 1988. The microbial population of the soil. pp. 449-471. In: Wild, A., ed. Russell's Soil Conditions and Plant Growth. Eleventh edition. John Wiley & Sons. New York. Hasegawa, Y., K. R. Lynn and W. J. Brockbank. 1976. Isolation and partial characterizaion of cytoplasmic and wall-bound acid phosphatases from wheat roots. Can. J. Bot. 54: 1163-1169. Haussling, M., and H. Marschner. 1989. Organic and inorganic soil phosphates and acid phosphatase activity in the rhizosphere of 8 0-year-old Norway spruce [Picea abies (L.) Karst.] trees. Biol. Fert. Soils 8: 128-133. 201 Hawkes, G. E., D. S. Powlson, E. W. Randall and K. R. Tate. 1984. A 31P nuclear magnetic resonance study of the phosphorus species in alkali extracts of soils from long-term field experiments. J. Soil Sci. 35: 35-45. Hebda, R. J. 1983. Late-glacial and post-glacial vegetation history at Bear Cove Bog, northeast Vancouver Island, British Columbia. Can. J. Bot. 61: 3172-3192. Hedley, M. J., J. W. B. Stewart and B. S. Chauhan. 198 2. Changes in inorganic and organic phosphorus fractions by cultivation practices and by laboratory incubations. Soil Sci. Soc. Am. J. 46: 970-976. Helal, H. M. 1980. Varietal differences in root phosphatase activity as related to the utilization of organic phosphates. Plant Soil 123: 161-163. Helal, H. M., and A. Dressier. 1989. Mobilization and turnover of soil phosphorus in the rhizosphere. Z. Pflanzenernahr. Bodenk. 152: 175-180. Hewitt, E. J. 1966. Sand and water culture methods used in the study of plant nutrition. Technical Communication No. 22 (2nd Ed., Revised). Commonwealth Agricultural Bureau, London. 547 pp. Hicks, C. R. 1982. Fundamental Concepts in the Design of Experiments. Third Edition. Saunders College Publishing. Fort Worth. 42 5 pp. Hilderbrand, R. L., ed. 1983. The Role of Phosphonates in Living Systems. CRC Press, Ltd. Boca Raton, FL. Hinedi, Z. R. , A. C. Chang and R. W. K. Lee. 1989. Characterization of phosphorus in sludge extracts using phosphorus-31 nuclear magnetic resonance spectroscopy. J. Environ. Qual. 18: 323-329. Ho, I. 1989. Acid phosphatase, alkaline phosphatase and nitrate reductase activity of selected ectomycorrhizal fungi. Can. J. Bot. 67: 750-753. Ho, I., and B. Zak. 1979. Acid phosphatase activity of six ectomycorrhiizal fungi. Can. J. Bot. 57: 1203-1205. Hubel, F. and E. Beck. 1993. In-situ determination of the P relations around the primary root of maize with respect to inorganic and phytate-P. Plant Soil 157: 1-9. 202 Humphreys, F. R., and M. J. Lambert. 1965. An examination of a forest site which has exhibited the ash-bed effect. Aust. J. Soil Res. 3: 81-94. Hutson, S. M., G. D. Williams, D. A. Berkich, K. F. LaNoue and R. W. Briggs. 1992. A 31P NMR study of mitochondrial inorganic phosphate visibility: Effects of Ca2+, . Mn2+ and the pH gradient. Biochemistry 31: 1322-1330. Ingall, E. D., P. A. Schroeder and R. A. Berner. 1990. The nature of organic phosphorus in marine sediments: new insights from 31P NMR. Geochim. Cosmochim. Acta 54: 2617-2620. Ipinmidun, W. B. 1973. Comparison of some methods for determining organic phosphorus in some Nigerian soils. Soil Sci. 115: 324-325. Isaac, S. 1992. Fungal-Plant Interactions. Chapman and Hall. New York. pp. 267-298. Jayachandran, K. , A. P. Schwab and B. A. D. Hetrick. 1992. Mineralization of organic phosphorus by vesicular-arbuscular mycorrhizal fungi. Soil Biol. Biochem. 24: 897-903. Jenkinson, D. S. 1988. Soil organic matter and its dynamics. pp. 564-607. In: Wild, A., ed. Russell's Soil Conditions and Plant Growth. John Wiley & Sons. New York. Jennings, D. H. 1995. The Physiology of Fungal Nutrition. Cambridge University Press. 622 pp. John, M. K., P. N. Sprout and C. C. Kelley. 1965. The distribution of organic phosphorus in British Columbia soils and its relationship to soil characteristics. Can. J. Soil Sci. 45: 87-95. Kalra, Y. P. and D. G. Maynard. 1991. Methods manual for Forest Soil and Plant Analysis. For. Can., Northwest Reg., North. For. Cent. Edmonton, AB. Inf. Rep. NOR-X-319. Kapoor, A., V. P. Singh and K. G. Mukeiji. 1988. Studies on the phosphatases of mycorrhizal and non-mycorrhizal Trigonella roots. pp. 125-127. In: Mahadevan, A., N. Raman, and K. Natarajan, eds. Mycorrhizae for Green Asia. University of Madras. Madras, India. Keenan, R. 1993. Structure and function of western red cedar and western hemlock forests on northern Vancouver Island. Ph. D. Thesis. The University of British Columbia. Vancouver. 167 pp. 203 Khalil, S., T. E. Loynachan and M. A. Tabatabai. 1994. Mycorrhizal dependency and nutrient uptake by improved and unimproved corn and soybean cultivars. Agron. J. 86: 949-958. Khanna, P. K., and R. J. Raison. 1986. Effect of fire intensity on solution chemistry of surface soil under a Eucalyptus pauclflora forest. Aust. J. Soil Res. 24: 423-434. Klinka, K. , R. N. Green, R. L. Trowbridge and L. E. Lowe. 1981. Taxonomic classification of humus forms in ecosystems of British Columbia. BC Ministry of Forests, Victoria, BC. Ko, W.-H., and F. K. Hora. 1970. Production of phospholipases by soil microorganisms. Soil Sci. 110: 355-358. Kozlowski, T. T. , and C. E. Ahlgren, eds. 1974. Fire and Ecosystems. Academic Press. New York. Krajina, V. J. 1969. Ecology of forest trees in British Columbia. Ecology of Western North America 2: 1-146. Krishna, K. R. , D. J. Bagyaraj and K. G. Papavinasasundarum. 1983. Acid and alkaline phosphatase activities in mycorrhizal and uninfected roots of Arachis hypogaea L. Ann. Bot. 51: 551- 553 . Kristensen, E. 1990. Characterization of biogenic organic matter by stepwise thermogravity. Biogeochemistry 9: 135-159. Kroehler, C. J., R. K. Antibus and A. E. Linkins. 1988. The effects of organic and inorganic phosphorus concentration on the acid phosphatase activity of ectomycorrhizal fungi. Can. J. Bot. 66: 750-756. Kropp, B. R. 1990. Variation in acid phosphatase among progeny from controlled crosses in the ectomycorrhizal fungus Laccaria bicolor. Can. J. Bot. 68: 864-866. Kulaev, I. S. 1979. The Biochemistry of Inorganic Polyphosphates. John Wiley & Sons. Toronto. Kwari, J. D., and T. Batey. 1991. Effect of heating on phosphate sorption and availability in some north-east Nigerian soils. J. Soil Sci. 42: 381-388. L'Annuziata, M. F. 1975. The origin and transformations of the soil inositol phosphate isomers. Soil Sci. Soc. Am. Proc. 39: 377-379. 204 Lapeyrie, F., J. Ranger and D. Vairelles. 1991. Phosphate- solubilizing activity of ectomycorrhizal fungi in vitro. Can. J. Bot. 69: 342-346. Lavkulich, L. M. 1981. Methods Manual, Pedology Laboratory. UBC Dept. of Soil Science. Vancouver. 211 pp. Legg, J. 0., and C. A. Black. 1955. Determination of organic phosphorus in soils. II. Ignition method. Soil Sci. Soc. Am. Proc. 19: 139-143. Levesque, M., and M. Schnitzer. 1967. Organo-metallic interactions in soil: 6. Preparation and properties of fulvic acid metal phosphates. Soil Sci. 103: 183-185. Lewis, T. 1982. Ecosystems of Block 4, TFL 25. Internal Report. Available from Western Forest Products Ltd., 1140 West Pender St., Vancouver, BC. 75 pp. Lewis, T. 1976. The till-derived podzols of Vancouver Island. Ph. D. Thesis. The University of British Columbia. Vancouver. 158 pp. Lindsay, W. L. 1979. Chemical Equilibria in Soil. Wiley- Interscience Publ. John Wiley and Sons. New York. Macadam, A. M. 1987. Effects of broadcast slashburning on fuels and soil chemical properties in the sub-boreal spruce zone of central British Columbia. Can. J. For. Res. 17: 1577-1584. MacDonald, R. M., and M. Lewis. 1978. The occurrence of some acid phosphatases and dehydrogenases in the vesicular-arbuscular mycorrhizal fungus Glomus mosseae. New Phytol. 80: 135-141. MacFall, J., S. A. Slack and J. Iyer. 1991. Effects of Hebeloma arenosa and phosphorus fertility on root acid phosphatase activity of red pine (Pinus resinosa) seedlings. Can. J. Bot. 69: 380-383. MacFall, J. S., S. A. Slack and S. Wehrli. 1992. Phosphorus distribution in red pine roots and the ectomycorrhizal fungus Hebeloma arenosa. Plant Physiol. 100: 713-717. MacKay, A. D., J. K. Syers, J. A. Springett and P. E. H. Gregg. 1982. Plant availability of phosphorus in superphosphate and rock phosphate as influenced by earthworms. Soil Biol. Biochem. 14: 281-287. 205 Mangas, V. J., J. R. Sanchez and C. Ortiz. 1992. Effects of fire on runoff and erosion on mediterranean forest soils in S. E. Spain. Pirineos 140: 37-51. Marschner, H., and B. Dell. 1994. Nutrient uptake in mycorrhizal symbiosis. Plant Soil 159: 89-102. Martin, F., D. Canet, D. Rolin, J. P Marchal and F. Larher. 1983. Phosphorus-31 nuclear magnetic resonance study of polyphosphate metabolism in intact ectomycorrhizal fungi. Plant Soil 71: 469-476. Martin, F., J.-P. Marchal, A. Timinska and D. Canet. 1985. The metabolism and physical state of polyphosphates in ectomycorrhizal fungi. A 31P nuclear magnetic resonance study. New Phytol. 101-2 75-290. McArthur, D. A. J., and N. R. Knowles. 1993. Influence of vesicular-arbuscular mycorrhizal fungi on the response of potato to phosphorus deficiency. Plant. Physiol. 101: 147- 160. McElhinney, C , and D. T. Mitchell, 1993. Phosphatase activity of four ectomycorrhzal fungi found in a Sitka spruce-Japanese larch plantation in Ireland. Mycol. Res. 97: 725-732. McKercher, R. B., and T. S. Tollefson. 1978. Barley response to phosphorus from phospholipids and nucleic acids. Can. J. Soil Sci. 58: 103-105. McLachlan, K. D. 1980 a. Acid phosphatase activity of intact roots and phosphorus nutrition on plants. I. Assay conditions and phosphatase activity. Aust. J. Agric. Res. 31: 429-440. McLachlan, K. D. 1980b. Acid phosphatase activity of intact roots and phosphorus nutrition in plants. II. Variations among wheat roots. Aust. J. Agric. Res. 31: 441-448. McLean, E. O . 1982. Soi pH and lime requirement. pp. 199-224. In: Page, A. L., R. H. Miller and D. R. Keeney, eds. Methods of Soil Analysis. Second Edition. Agronomy Series No. 9., Part 2. Soil Sci. Soc. Am., Inc. Madison, Wis. Mehlich, A. 1984. Mehlich 3 soil test extractant: a modification of Mehlich 2 extractant. Commun. Soil Sci. Plant Anal. 15: 1409-1416. Mehta, N. C , J. O. Legg, C. A. I. Goring and C. A. Black. 1954. Determination of organic phosphorus in soils. I. Extraction methods. Soil Sci. Soc. Am. Proc. 18: 443-449. 206 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. Managem. 60: 181-206. Messier, C. 1991. Factors limiting early growth of Thuja plicata, Tsuga heterophylla and Picea sitchensis seedlings on Gaultheria shallon dominated cutovers in coastal British Columbia. Ph. D. Thesis. The University of British Columbia. Vancouver. 16 0 pp. Meuller-Harvey, I., A. S. R. Juo and A. Wild. 1985. Soil organic C, N, S, and P after forest clearance in Nigeria: mineralization rates and spatial variability. J. Soil Sci. 36: 585-591. Meysselle, J. P., G. Gay and J. C. Debaud. 1991. Intraspecific genetic variation of acid phosphatase activity in monokaryotic and dikaryotic populations of the ectomycorrhizal fungus Hebeloma cylindrosporum. Can. J. Bot. 69: 808-813. Michaelson, G. J., C. L. Ping and G. A. Mitchell. 1987. Correlation of Mehlich 3, Bray PI and ammonium acetate extract P, K, Ca and Mg from Alaska agricultural soils. Commun. Soil Sci . Plant Anal. 18: 1003-1015. Mitchell, D. T., and D. J. Read. 1981. Utilization of inorganic and organic phosphates by the mycorrhizal endopytes of Vaccinium macrocarpon and Rhododendron ponticum. Trans. Br. mycol. Soc. 76: 255-260. Morrison, I. K. 1991. Addition of organic matter and elements to the forest floor of an old-growth Acer saccharum forest in the annual litter fall. Can. J. For. Res. 21: 462-468. Mroz, G. D., M. F. Jurgensen, A. E. Harvey and M. J. Larsen. 1980. Effects of fire on nitrogen in forest floor horizons. Soil Sci. Soc. Am. J. 44: 395-400. Murphy, J., and J. P. Riley. 1962. A modified single solution method for determination of phosphate in natural waters. Anal. Chim. Acta 27: 31-36. Newman, R. H., and K. R. Tate. 198 0. Soil phosphorus characterization by 31P nuclear magnetic resonance. Commun. Soil Sci. Plant Anal. 11: 835-842. 207 Oades, J. M. 1989. An introduction to organic matter in mineral soils. pp. 89-159. In: Dixon, J. B., and S. B. Weed, eds. Minerals in Soil Environments. 2nd edition. Soil Sci. Soc. Am., Madison WI. 1244 pp. Ognalaga, M., E. Frossard and F. Thomas. 1994. Glucose-1- phosphate and myo-inositol hexaphosphate adsorption mechanisms on goethite. Soil Sci. Soc. Am. J. 58: 332-337. Ogner, G. 1983. 31P-NMR spectra of humic acids. A comparison of four different raw humus types in Norway. Geoderma 29: 215- 219. Olila, 0. G., and K. R. Reddy. 1995. Influence of pH on phosphorus retention in oxidized lake sediments. Soil Sci. Soc. Am. J. 59: 946-959. 01sen, S. R., and L. E. Sommers. 1982. Phosphorus. pp. 403-430. In: Page, A. L., R. H. Miller and D. R. Keeney, eds. Methods of Soil Analysis. Second edition. Agronomy Series No. 9, Part 2. Soil Sci. Soc. Am., Inc. Madison, Wis. Olsen, S. R., C. V. Cole, F. S. Watanabe and L. A. Dean. 1954. Estimation of available phosphorus in soils by extraction with sodium bicarbonate. US. Dep. of Agric. Circ. 93 9. O'Neill, I. K., M. Sargent and M. L. Trimble. 1980. Determination of phytate in foods by phosphorus-31 Fourier transform nuclear magnetic resonance spectroscopy. Anal. Chem. 52: 12 88-12 91. Pang, P. C , and H. Kolenko. 1986. Phosphomonoesterase activity in forest soils. Soil Biol. Biochem. 18: 35-40. Pant, H. K. , D. Vaughan, and A. C. Edwards. 1994. Molecular size distribution and enzymatic degradation of organic phosphorus in root exudates of spring barley. Biol. Fertil. Soil 18: 285-290. Pare, D., and B. Bernier. 1989. Phosphorus-fixing potential of Ah and H horizons subjected to acidification. Can. J. For. Res. 19: 132-134. Parke, J. L., R. G. Linderman and J. M. Trappe. 1983. Effects of forest litter on mycorrhiza development and growth of Douglas- fir and western red cedar seedlings. Can. J. For. Res. 13: 666-671. Parkinson, J. A., and S. E. Allen. 1975. A wet oxidation procedure suitable for the determination of nitrogen and mineral nutrients in biological material. Commun. Soil Sci. Plant Anal. 6: 1-11. 208 Pasqualini, S., F. Panara and M. Antonielli. 1992. Acid phosphatase activity in Pinus pinea-Tuber albidum ectomycorrhizal associations. Can. J. Bot. 70: 1377-1383. Phillips, J. M., and D. S. Hayman. 1970. Improved procedures for clearing roots and staining parasitic and vesicular-arbuscular mycorrhizal fungi for rapid assessment of infection. Trans. Br. Mycol. Soc. 55: 58-160. Pojar, J., K. Klinka and D. A. Demarchi. 1991. Coastal western hemlock zone. In: Meidinger, D., and J. Pojar, eds. Ecosystems of British Columbia. BC Ministry of Forests, Victoria. 75 pp. Pokarzhevskii, A. D., and S. A. Gordienko. 1985. Soil animals in the biogenic migration of phosphorus in forest-steppe ecosystems. Soviet J. Ecol. 15: 130-133. Prescott, C. E. 1995. Does nitrogen availability control rates of litter decomposition in forest? Plant Soil 168-169: 83-88. Prescott, C. E., and G. F. Weetman. 1994. Salal Cedar Hemlock Integrated Research Program: A Synthesis. Faculty of Forestry, University of British Columbia, Vancouver, BC. Prescott, C. E., M. A. McDonald and G. F. Weetman. 1993. Availability of N and P in the forest floors of adjacent stands of western red cedar - western hemlock - amabalis fir on northern Vancouver Island. Can. J. For. Res. 23: 605-610. Prescott, C. E., L. F. de Montigny, C. M. Preston, R. J. Keenan and G. F. Weetman. 1995. Carbon chemistry and nutrient supply in forest floors under western red cedar - western hemlock, and western hemlock - amabalis fir on northern Vancouver Island. pp. 377-396. In: Kelly, J. M., ed. Proceedings of 8th North American Forest Soils Conference. May 1993. Gainesville, FL. Preston, C. M., J. A. Ripmeester, S. P. Mathur and M. Levesque. 1986. Application of solution and solid-state multinuclear NMR to a peat-based composting system for fish and crab scrap. Can. J. Spectrosc. 31: 63-69. Pritchett, W. L. , and R. F. Fisher. 1987. Properties and Management of Forest Soils. Second Edition. John Wiley & Sons. Toronto. 4 94 pp. Raison, R. J., P. K. Khanna and P. V. Woods. 1984. Mechanisms of element transfer to the atmosphere during vegetation fires. Can. J. For. Res. 15: 132-140. 209 Reichle, D. E. 1977. The role of soil invertebrates in nutrient cycling. Ecol. Bull. (Stockholm) 25: 145-156. Read, D. J. 1991. Mycorrhizas in ecosystems. Experientia 47: 376-391. Rhodes, L.H., and P. 0. Larsen. 1981. Effects of fungicides on mycorrhizal development of creeping bentgrass. Plant Disease 65: 145-147. Rice, S. K. 1993. Vegetation establishment in post-fire Adenostoma chaparrel in relation to fine-scale pattern in fire-intensity and soil nutrients. J. Veg. Sci. 4: 115-124. Roberts, T. L., J. W. B. Stewart and J. R. Bettany. 1985. The influence of topography on the distribution of organic and inorganic soil phosphorus across a narrow environmental gradient. Can. J. Soil Sci. 65: 651-665. Rojo, M. J., S. G. Carcedo and M. P. Mateos. 1990. Distribution and characterization of phosphatase and organic phosphorus in soil fractions. Soil Biol. Biochem. 22: 169-174. Romanya, J., P. K. Khanna and R. J. Raison. 1994. Effects of slash burning on soil phosphorus fractions and sorption and desorption of phosphorus. For. Ecol. Management 65: 89-103. Rowland, A. P., and H. M. Grimshaw. 1985. A wet oxidation procedure suitable for total nitrogen and phosphorus in soil. Commun. Soil Sci. Plant Anal. 16: 551-560. Saa, A. C. Trasar-Cepeda, F. Gil-Sotres and T. Carballas. 1993. Changes in soil phosphorus and acid phosphatase activity immediately following forest fires. Soil Biol. Biochem. 25: 1223-1230. St. Arnaud, R. J., J. W. B. Stewart and E. Frossard. 1988. Application of the 'Pedogenic Index' to soil fertility studies, Saskatchewan. Geoderma 43: 21-32. St. John, T. V., and P. W. Rundel. 1976. The role of fire as a mineralizing agent in a Sierran coniferous forest. Oecologia 25: 35-45. Sanborn, P. T. 1987. Ferro-humic podzols of coastal British Columbia. Aspects of Genesis and Chemistry. Ph. D. Thesis. University of British Columbia. Vancouver. 349 pp. 210 Saunders, W. M. H., and E. G. Williams. 1955. Observations on the determination of total organic phosphorus in soils. J. Soil Sci. 6: 254-267. Saxena, S. N. 1964. Phytase activity of plant roots. J. Exp. Bot. 15: 654-658. Schnitzer, M. 1977. Recent findings on the characterization of humic substances extracted from widely differing climatic zones. pp. 117-131. In: Proc. Symposium on Soil Organic Matter Studies. Braunschweig International Atomic Energy Agency. Vienna. Schoenau, J. J., and J. R. Bettany. 1987. Organic matter leaching as a component of carbon, nitrogen, phosphorus and sulfur cycles in a forest, grassland and gleyed soil. Soil Sci. Soc. Am. J. 51: 646-651. Shaw, G., and D. J. Read. 1989. The biology of mycorrhiza in the Ericaceae. XIV. Effects of iron and aluminium on the activity of acid phosphatase in the ericoid endophyte Hymenoscyphus ericae Korf. and Kernan. New Phytol. 113: 529- 533 . Sherrell, C. G., and W. H. M. Saunders. 1966. An evaluation of methods for the determination of total phosphorus in soils. N. Z. J. Agric. Res. 9: 972-979. Silver, W. L., F. W. Scatena, A. H. Johnson, T. G. Siccama and M. J. Sanchez. 1994. Nutrient availability in a montane wet tropical forest: Spatial patterns and methodological considerations. Plant Soil 164: 129-145. Simard, R. R., D. Cluis, G. Gangbazo and A. R. Pesant. 1994. Phosphorus sorption and desorption indices in soil. Commun. Soil Sci. Plant Anal. 25: 1483-1494. Soltanpour, P. N. , R. L. Fox and R. C. Jones. 1987. A quick method to extract organic phosphorus from soils. Soil Sci. Soc. Am. J. 51: 255-256. Sommers, L. E., and D. W. Nelson. 1972. Determination of total phosphorus in soils: a rapid perchloric acid digestion procedure. Soil Sci. Soc. Am. Proc. 36: 902-904. Speir, T. W. , and D. J. Ross. 1978. Soil phosphatases and sulphatases. pp. 1997-249. In: Burns, R. G., ed. Soil Enzymes. Academic Press. New York. 380 pp. 211 Speir, T. W., and J. C. Cowling. 1991. Phosphatase activities of pasture plants and soils: Relationship with plant productivity and soil P fertility indices. Biol. Fert. soils 12: 189-194. Stark, N. 1972. Nutrient cycling pathways and litter fungi. Bioscience 22: 355-360. Stevenson, F. J. 1994. Humus Chemistry: Genesis, Composition, Reactions. 2nd Edition. John Wiley & Sons, Inc. Toronto. 552 pp. Stevenson, F. J. 1986. Cycles of Soil. John Wiley & Sons. New York. Stevenson, F. J. 1982. Humus Chemistry: Genesis, Composition, Reactions. John Wiley & Sons. New York. 443 pp. Steward, J. H., and J. M. Oades. 1972. The determination of organic phosphorus in soils. J. Soil Sci. 23: 38-49. Stewart, J. W. B., and H. Tiessen. 1987. Dynamics of soil organic phosphorus. Biogeochemistry 4: 41-60. Stone, E. L. 1975. Effect of species on nutrient cycles and soil change. Phil. Trans. R. Soc. Lond. B. 271: 149-162. Straker, C. J., and D. T. Mitchell. 1986. The activity and characterization of acid phosphatases in endomycorrhizal fungi of the ericaceae. New Phytol. 104: 243-256. Subbarao, Y. V., R. Ellis, Jr., G. M. Paulson and J. V. Paukstelis. 1977. Kinetics of pyro- and tripolyphosphate hydrolyses in the presence of corn and soybean roots as determined by NMR spectroscopy. Soil Sci. Soc. Am. J. 41: 316-318. Sukarno, N. , S. E. Smith and I. S. Scott. 1993. The effect of fungicides on vesicular-arbuscular mycorrhizal symbiosis. I. The effects on vesicular-arbuscular fungi and plant growth. New Phytol. 25: 139-147. Suttie, J. W. 1972. Introduction to Biochemistry. Holt, Rinehart and Wilson, Inc. New York. 365 pp. Swift, M. J. 1977. The roles of fungi and animals in the immobilisation and release of nutrient elements from decomposing branch-wood. Ecol. Bull. (Stockholm) 25: 193- 202 . Syers, J. K. , J. D. H. Williams and T. W. Walker. 1968. The determination of total phosphorus in soils and parent materials. N. Z. J. Agric. Res. 11: 757-762. 212 Syers, J. K., J. D. H. Williams, A. S. Campbell and T. W. Walker. 1967. The significance of apatite inclusions in soil phosphorus studies. Soil Sci. Soc. Am. Proc. 31: 752-756. Szember, A. 1962. Influence on plant growth of the breakdown of organic phosphorus compounds by microorganisms. Plant Soil 13: 147-158. Tabatabai, M. A. 1982. Soil enzymes. pp. 902-947. In: Page, A. L., R. H. Miller and D. R. Keeney, eds. Methods of Soil Analysis. Part 2, Second Edition. American Society of Agronomy, Inc. Madison, WI. Tabatabai, M. A., and W. A. Dick. 1979. Distribution and stability of pyrophosphatase in soils. Soil Biol Biochem. 11: 655-659. Tadano, T., K. Ozama, H. Sakai, M. Osaki and H. Matusi. 1993. Secretions of acid phosphatase by the roots of corp plants under phosphorus-deficient conditions and some properties of the enzyume secreted by lupin roots. Plant Soil 155/156: 95- 98. Tarn, P. C. F. and D. A. Griffiths. 1993. Mycorrhizal associations in Hong Kong Fagaceae. IV. The mobilization of organic and poorly soluble phosphates by the ectomycorrhizal fungus Pisolithus tinctorius. Mycorrhiza 2: 133-139. Tarafdar, J. C , and N. Claasen. 1988. Organic phosphorus compounds as a phosphorus source for higher plants through the activity of phosphatases produced by plant roots and microorganisms. Biol. Fert. Soils 5: 308-312. Tarafdar, J. C , and A. Jungk. 198 7. Phosphatase activity in the rhizosphere and its relation to the depletion of soil organic phosphorus. Biol. Fert. Soils 3: 199-204. Tarafdar, J. C , and H. Marschner. 1994. Phosphatase activity in the rhizosphere and hydrosphere of VA mycorrhizal wheat supplied with inorganic and organic phosphorus. Soil Biol. Biochem. 26: 387-395. Tarafdar, J. C , A. V. Rao and Praveen-Kumar. 1992. Effects of different phosphatase-producing fungi on growth and nutrition of mung beans [Vigna radiata(L.) Wilczek] in an arid soil. Biol. Fert. Soils 13: 35-38. Tate, K. R. 1984. The biological transformations of P in soil. Plant Soil 76: 245-256. 213 Tate, K. R., and R. H. Newman. 1982. Phosphorus fractions of a climosequence of soils in New Zealand tussock grassland. Soil Biol. Biochem. 14: 191-196. Theodorou, C. 1971. The phytase activity of the mycorrhizal fungus Rhizopogan roseolus. Soil Biol. Biochem. 10: 33-37. Thiagarajan, T. R., and M. H. Ahmad. 1994. Phosphatase activity and cytokinin content in cowpea (Vigna unguiculata) inoculated with a vesicular-arbuscular mycorrhizal fungus. Biol. Fertil. Soils 17: 51-56. Thomas, G. W. , and D. E. Peaslee. 1973. Testing soils for phosphorus. pp. 115-132. In: Walsh, L. M., and J. D. Beaton, eds. Soil Testing and Plant Analysis. Soil Sci. Soc. Am., Inc. Madison, WI. 491 pp. Thomas, G. W. , C. A. Clarke, B. Mosse and R. M. Jackson. 1982. Source of phosphate taken up from two soils by mycorrhizal (Thelephora terrestris) and non-mycorrhizal Picea sitchensis seedlings. Soil Biol. Biochem. 14: 73-75. Thomas, R. L., R. W. Sheard and J. R. Moyer. 1967. Comparison of conventional and automated procedures for nitrogen, phosphorus and potassium analysis of plant material using a single digestion. Agron. J. 59: 240-243. Tiedemann, A. R., and J. 0. Klemmedson. 1986. Long term effects of mesquite removal on soil characteristics. I. Nutrients and bulk density. Soil Sci. Soc. Am. J. 50: 472-475. Tiessen, H., J. W. B. Stewart and C. V. Cole. 1984. Pathways of phosphorus transformations in soils of differing pedogenesis. Soil Sci. Soc. Am. J. 48: 853-858. Tikhaya, N. I., T. N. Stekhanova, M. D. Fedorovskaya and D. B. Vakhmistov. 1990. Properties of ecto Ca-ATPase isolated from barley {Hordeum vulgare) roots. pp. 193-194. In: Van Beusichem, M. L., ed. Plant Nutrition - Physiology and Applications. Kluwer Academic Publishers. Amsterdam. Ting, I. P. 1982. Plant Physiology. Addison-Wesley Publishing Co. Don Mills, ON. 642 pp. Tomkins, I. B., J. D. Kellas, K. G. Tolhurst and D. A. Oswin. 1991. Effects of fire intensity on soil chemistry in a eucalypt forest. Aust. J. Soil Res. 29: 25-47. Tran, T. Sen, M. Giroux, J. Guilbeault and P. Audesse. 1990. Evaluation of Mehlich III extractant to evaluate the available P in Quebec soils. Commun. Soil Sci. Plant Anal. 21: 1-28. 214 Trasar-Cepeda, M. C , T. Carballas, F. Gil-Sotres and E. deBlas. 1991. Liming and the phosphatase activity and mineralization of phosphorus in an andic soil. Soil Biol. Biochem. 23: 209- 215. Tsubota, G. 1959. Phosphate reduction in the paddy field. I. Soil PI. Fd. Tokyo 5: 10-15. Turner, J., and M. J. Lambert. 1986. Effects of forest harvesting nutrient removals on soil nutrient reserves. Oecologia 70: 140-148 . Turner, J., and M. J. Lambert. 1985. Soil phosphorus forms and related tree growth in a long term Pinus xa.dia.ta. phosphate fertilizer trial. Comm. Soil Sci. Pi. Anal. 16: 275-288. Van Veen, J. A., J. N. Ladd, J. K. Martin and M. Amato. 1987. Turnover of carbon, nitrogen and phosphorus through the microbial biomass in soils incubated with 14C-, 14N- and 32P- labelled bacterial cells. Soil Biol. Biochem. 19: 559-565. Verkade, S. D., and D. F. Hamilton. 1983. Effects of benomyl on growth of Liriodendron tulipifera seedlings inoculated with the vesicular-arbuscular fungus, Glomus fasciculatus. Scientia Hortic. 21: 253-260. Violante, A., C. Colombo and A. Buondonno. 1991. Competitive adsorption of phosphate and oxalate by aluminum oxides. Soil Sci. Soc. Am. J. 55: 65-70. Vlamis, J., H. H. Biswell and A. M. Schultz. 1955. Effects of prescribed burning on soil fertility in second growth ponderosa pine. J. Forestry 53: 905-909. Watanabe, F. S., and S. R. Olsen. 1965. Test of an ascorbic acid method for determining P in water and NaHCl03 extracts from soil. Soil Sci. Soc. Am. Proc. 29: 677-678. Weetman, G. F., R. M. Fournier, E. Schnorbus-Panozzo and J. Barker. 1990. Post-burn nitrogen and phosphorus availability in coastal British Columbia cedar/hemlock forests and the use of fertilization and salal eradication to restore productivity. In: Gessel, S. P., ed. Sustained Productivity of Forest Lands. Proceedings of the Seventh North American Forest Soils Conference. Friesen Press. Vancouver. Weetman, G. F., R. Fournier, J. Barker, E. Schnorbus-Panozzo and A. Germain. 1989a. Foliar analysis and response of fertilized chlorotic Sitka spruce plantations on salal-dominated cedar- hemlock cutovers on Vancouver Island. Can. J. For. Res. 12: 1501-1511. 215 Weetman, G. F., R. Fournier, J. Barker, and E. Schnorbus-Panozzo. 1989b. 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. 12: 1512-1520. Weetman, G. F., M. A. Radwan, J. Kumi and E. Schnorbus. 1988. Nutirition and fertilization of western red cedar. pp. 47- 60. In: Smith, N. J., ed. Western Red Cedar. Does it Have a Future? Conference Proceedings, University of British Columbia, Faculty of Forestry. Vancouver. 177pp. Wendt, J. W. 1995. Evaluation of the Mehlich 3 soil extractant for upland Malawi soils. Commun. Soil Sci. Plant Anal. 26: 687-702. Wild, A. 1988. Plant nutrients in soil: phosphate. pp. 695- 741. In: Wild, A., ed. Russell's Soil Conditions and Plant Growth. Eleventh Edition. John Wiley & Sons. New York. Wilkinson, L. 1990. SYSTAT. The System for Statistics. SYSTAT, Inc. Evanston, IL. 677 pp. Williams, J. D. H. , and T. W. Walker. 1967. Comparison of "ignition" and "extraction" methods for the determination of organic phosphate in rocks and soils. Plant Soil 27: 457- 459. Williams, J. D. H., T. Mayer and J. O Nriagu. 198 0. Extractability of phosphorus from phosphate minerals common in soils and sediments. Soil Sci. Soc. Am. J. 44: 462-465. Williams, J. D. H., J. K. Syers, T. W. Walker and R. W. Rex. 1970. A comparison of methods for the determination of soil organic phosphorus. Soil Sci. 110: 13-18. Wilson, M. A. 1991. Analysis of functional groups in soil by nuclear magnetic resonance spectroscopy. pp. 601-645. In: Smith, K. A., ed. Soil Analysis: Modern Instumental Techniques. 2nd Edition. Marcel Dekker, Inc. New York. Wolf, A. M., and D. E. Baker. 1985. Comparisons of soil test phosphorus by Olsen, Bray PI, Mehlich I and Mehlich III methods. Commun. Soil Sci. Plant Anal. 16: 467-484. Xiao, X. J., D. W. Anderson and J. R. Bettany. 1991. The effect of pedogenic processes on the distribution of phosphorus, calcium and magnesium in Gray Luvisols. Can. J. Soil Sci. 71: 397-410. 216 Yuan, G., and L. M. Lavkulich. 1994. Phosphate sorption in relation to extractable iron and aluminium in spodosols. Soil Sci. Soc. Am. J. 58: 343-346. Zech, W., H. G. Alt, L. Haumaier and R. Blasek. 1987. Characterization of phosphorus fractions in mountain soils of the Bavarian Alps by 31P NMR spectroscopy. Z. Pflanzenernahr. Bodenk. 150: 119-123. Zech, W., H. G. Alt, A. Zucker and I. Kogel. 1985. 31P-NMR spectroscopic investigations of NaOH extracts from soils with different land use in Yucatan (Mexico). Z. Pflanzenernahr. Bodenk. 148: 626-632. 217

Cite

Citation Scheme:

    

Usage Statistics

Country Views Downloads
Germany 12 2
Japan 4 0
China 2 5
Spain 1 0
Iran 1 0
United States 1 5
City Views Downloads
Heidelberg 12 0
Tokyo 4 0
Shenzhen 1 5
Tiran 1 0
Beijing 1 0
Unknown 1 2
Ashburn 1 0

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

Share

Share to:

Comment

Related Items