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 Cade, Barbara Jean 1995-12-31

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

Item Metadata

Download

Media
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
Original Record: 1.0088246 +original-record.json
Full Text
1.0088246.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\ ^o~-tiie required standard  THE UNIVERSITY OF BRITISH COLUMBIA October 1995 (c) Barbara Jean Cade-Menun, 1995  In  presenting  degree freely  this  at the available  copying  of  department publication  thesis  in  partial  fulfilment  University  of  British  Columbia,  for  this or of  reference  thesis by  this  for  his thesis  and study. scholarly  or for  her  of  (Signature)  Department  of  The University of British Columbia Vancouver, Canada  Date  DE-6 (2/88)  /I Oct  I9 9<r-  It  gain shall not  permission.  that  agree  may  representatives.  financial  requirements  I agree  I further  purposes  the  that  the  an  advanced  Library shall make  permission for  be  granted  is  understood be  for  by  the that  allowed without  it  extensive  head  of  my  copying  or  my  written  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. &  Williams  method  and  the Bowman  overestimated organic P.  Both the Saunders  & Moir extraction  procedure  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  31  P 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  differences  in  Al  and  Fe.  P levels between ii  There the  were  forest  no  types.  significant P-31  NMR  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 postharvest. P  and  By 10 years postburn, significant reductions in organic  organically  horizons.  bound  Fe  and  Al  were  revealed  in  mineral  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 Obj ectives Hypotheses Thesis Structure  CHAPTER 2  Literature Review Forest Soil Phosphorus Cycles Soil Phosphorus Methodology  6 25  CHAPTER 3  Study area Sampling design Study sites  32 37 38  CHAPTER 4  A Comparison of Methods for Total, Organic and Available P Introduction Methods Results Discussion Conclusions  42 46 47 53 57  CHAPTER 5  CHAPTER 6  1 3 4 5  A Comparison of Soil Extraction Procedures for NMR Spectroscopy Introduction Methods Results Discussion Conclusions A Comparison of P Forms on CH and HA Sites Introduction Methods Results Discussion Conclusions iv  31  P 58 61 65 81 84  86 87 90 108 118  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 Methods Results Discussion Conclusions  162 164 168 180 188  General Conclusions  18 9  CHAPTER 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  31  P 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  80  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:  95  CH-HA pyrophosphate-extracted Fe, Al and Mn  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  Table 6-10:  CH-HA P fractions  Table 6-11:  CH-HA organic and inorganic P  10 0 .101 101  Table 6-12 : CH-HA correlation matrix  103  Table 6-13:  106  CH-HA P forms from NMR specta  Table 7-1:  Postburn moisture content and pH  Table 7-2:  ANOVA table for postburn moisture, pH, C, N, C/N....125  vi  124  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:  130  Postburn pyrophosphate-extracted Fe, Al and Mn  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  Table 8-8:  Greenhouse expt. myc. effects on acid phosphatase.176  Table 8-9:  Greenhouse expt. alkaline phosphatase  178  Correlation matrix, greenhouse expt  179  Table 8-10:  vii  176  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  32  Figure 3-2:  A typical stand of the CH forest type  34  Figure 3-3:  A typical stand of the HA forest type  36  Figure 3-4:  Location of sampling  39  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  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  98  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, USDAARS, 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 inlaws, 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  Don.) and western hemlock (Tsuga  plicata  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  Germain  (1985) indicated  nitrogen  and phosphorus  that and  to grow well.  A study by  the CH sites were deficient responded  suggesting a nutrient cycling problem.  well  to  in  fertilization,  Preliminary research by  Western Forest Products Limited, the University of British Columbia and Natural Resources Canada's Pacific Forestry Centre led to the establishment Program  of  (SCHIRP),  the to  Salal-Cedar-Hemlock investigate  the  Integrated  many  aspects  Research 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  nutrient, especially in conifer plantations.  of  any  other  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. forests  of  The acidic soils and cool temperatures of the  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  concentrations of organic P in soils. used:  methods  to  measure  Two indirect methods are  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  31  P 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 were  developed  for  (Anderson 197 5) .  agricultural  appropriate for forested soils.  soils,  Most soil P methods  and  many  may  not  be  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. by Weetman et al.  Studies  (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  fertilization than do the other species.  respond  less  to  P  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. examine extraction procedures for  The second objective was to  31  P 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  31  P 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  spectroscopic analysis.  of  extraction  procedures  for  31  P NMR  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. conclusions of this research.  5  Chapter 9 summarizes the  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). in the management of soil P,  To aid  many attempts have been made to model  the soil P cycle. Conceptually,  this main cycle can be subdivided  further cycles (Dighton and Boddy 1988).  into two  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: (840,000  x  1012 kg), terrestrial  soils  dissolved inorganic P043~ in the ocean (80  marine  sediments  (96-160 x  1012 kg),  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 phosphate  and are not acidic.  are:  wavellite  Some secondary minerals of  [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. solution, the dissociation of phosphoric acid is: H3P04 <  > HP042" <  > H2P04" < 7  > P043"  In dilute  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 compounds.  (1986)  recognizes  six  major  groups  of  soil  P  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 semiarid 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) the  is in  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; presence  drainage;  of ligands  mineralization;  redox  potential;  that can replace phosphate when  the  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 1990b).  al.  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 adsorption.  influence  on Px levels  in  the  soil  is phosphate  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). clays decreases in the order:  Phosphate fixation to  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  (Violante et al.  soil materials 1991).  to reduce phosphate  adsorption  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). by  ligand  exchange,  while  the  slow  The fast reaction occurs 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 phospholipids nucleic acids phosphoproteins metabolic phosphates  2-50% 1-5% 0.2-2.5% trace 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 nucleic  acids.  include phospholipids  and  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, inositol (Fig. 2-2A).  commonly  called  Of a variety of possible esters, the most  common in soil is the hexaphosphate 11  ester (Fig. 2-2A), which  is  0 tl  OH /  R-O-P-0"  0=P-0-R1  1 0"  \  0-R 2  Monoester  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.  phosphates such as  1994) .  In addition  to hexaphosphate,  lower  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). as  ether,  benzene  or  They are soluble in fat solvents such  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). are  the glycerophosphatides,  lecithin,  phosphatidyl  phosphatidyl inositol. (Emsley and Niazi 1983).  Included in this group  such as phosphatidyl  serine,  phosphatidyl  choline or  ethanolamine  and  Phosphatidyl choline is the most abundant Most soil phospholipids are of microbial  (bacterial and fungal) origin.  13  /P  H2*>3 OH  0  OH  HO/H  H2P03-0/H-"  H\H  _S/o-P0 3 H 2  HN-LJ^OH OH  A)  03H2  O  0  /  H  H  PO3H2  H2P03  Phytic acid  myo-lnositol  0  I /°  H  ?  CH 2 0P-O 0 I CH 2 CH 2 H(CH 3 ) 3 OH  B)  R'COOCH  R'COOCH  CHOCR.'  -o-L'ecithin  +  H,C-O-?-0CH 2 CHSH 3 0" COO"  Phosphatidyl  serine  I ? HJC-O-P-OCHJCHJSHJ 0"  Phosphatidyl ethano l a m i n e  NH2  C) O" 1  -o—p=o I o I  ^CH  "O— P = 0 1  o  I "O—P=O  \  ^CH  I o I  D)  CH2  If f OH OH Adenosine triphosphate (AT?)  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  (deoxyribonucleic acid).  RNA  (ribonucleic  acid)  and  DNA  Each of these consists of a chain of  nucleotides, and each nucleotide contains a pentose sugar, a purine or  pyrimidine  base,  and  phosphoric  adjoining pentose units (Fig. 2-2C).  acid  residue  which  links  Anderson (1970) isolated two  nucleoside phosphates from NaOH extracts of soil.  One appeared to  contain thymine, the other uracil. Other P0 Compounds Other include:  P0  compounds  ATP  which  have  been  isolated  soils  (Fig. 2-2D) and several high molecular weight P0  compounds which have not yet been identified 1983) .  from  They were all in low abundance.  (Emsley and Niazi  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  The above-ground stand is estimated to contain about 2 5% of  "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  by harvesting.  upset  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 16  (1972) showed  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,  (Stevenson 1986).  especially  in  agricultural  soils  In forest soils, Schoenau and Bettany (1987)  observed leaching of labile P-rich compounds from surface horizons to lower horizons, where they became plants.  fixed and unavailable to  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  mineralization, while Pz is converted  is  converted  to P0 by  to  PT  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,  variations 1987) .  It  and may be  complicated  by  in soil microbial populations is  influenced  by  the  many  spatial  and  (Stewart and factors  which  temporal Tiessen 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  phospholipid followed  and  Tiessen  1987).  The  most  in bacteria is phosphatidyl ethanolamine  by  phosphoglycerol,  phosphatidic  common (> 50%) ,  acid,  and  phosphoinositols (Stewart and Tiessen 1987).  Teichoic acids may be  present  P may  in bacterial  bacterial  cells  walls, and  surplus  as polyphosphates  (Stewart  accumulate  and Tiessen  in  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,  population or by the plant itself  either  by  the  microbial  (Stewart and Tiessen  1987).  Organic acids, which may also be produced by the plant and/or the microbial  population,  phosphate  in  Phosphate  is  the  will  also  rhizosphere  released  by  increase  the  (Comerford  and  these 19  acids  through  availability Skinner  of  1989).  ligand-exchange  reactions, the dissolving of metal-oxide surfaces which adsorb P, or  the  complexing  precipitation  of  metals  in  of metal phosphates  solution  to  (Fox et al.  prevent  1990a).  the 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  phosphatases  through  the production  (Dighton and Boddy 1988).  of  organic  acids  and  Fungi may also obtain P  from parasitic associations on living organisms (Dighton and Boddy 1988). fungi  Phosphorus is concentrated in the mycelium, and when the are  alive,  extracellular  is  released  excretion,  of  (Swift  1977), and on their death by other fungi and organisms  (Stark  Systems  and grazing by  secretion animals  1972) .  enzymes, leaching  by  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).  also relocate P.  20  Spore dispersal will  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) . Pinaceae,  Many of the host plants belong to the families  Fagaceae,  Betulaceae  ascomycetes and basidiomycetes.  and  Myrtaceae;  the  fungi  are  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 vesiculararbuscular involving  (VA) mycorrhizae. members  of  the  Orchid mycorrhizae are symbioses Orchidaceae 21  and  basidiomycetes  or  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 The  fungi  of  ascomycetes.  ericalean  mycorrhizae  are  (Isaac 1992).  basidiomycetes  and  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 (Thuja  they are formed with some conifers such as western red cedar 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  Ectomycorrhizae may produce acid phosphatases (MacFall efc al. Lapeyrie 1991), protease, cellulase, phenol oxidase (Entry 1991) and phytase  (Dighton 1983).  Cromack  et al.  plant. 1991; 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 22  accumulate  polyphosphate  for  storage or hyphal P transport (Martin et al. 1992) .  Phosphatase production  has  1983; MacFall et  also been  demonstrated  al. 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 Jennings 1995).  (Bolan 1991;  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). transfer.  The central role of P in plants is in energy  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  therefore altering  (Pokarzherskii result the  and  Gordienko  in the release  relative  1985).  of nutrients,  contributions  of  Grazing  may  in addition to  different  fungal  and  bacterial species to the decomposition process (Dighton and Boddy 1989) .  After  decomposers,  death,  returning  soil P  to  fauna the  are  themselves  bacterial  and  attacked fungal  by  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  (Thomas et al.  1967;  1985; Schoenau and Bettany 1987; Xiao et al.  1991;  Reddy 1995); digestion with H2S04 and H202 Roberts et al. Compton  1994;  Fyles  and  Cote  1994; Hanafi  digestion with H2S04, H202 and HF  and  Syers  1994);  (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. (aqua  1993; Silver et al. 1994); and digestion with HC1 and HN03 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 soilcrop-climate  combination  (Thomas  and  Peaslee  1973).  These  procedures should be tested and calibrated for each soil, crop and environment. by  The main extractants used to determine PA are grouped  the soil pH range at which they are most effective.  alkaline soils, the Olsen et al. most suitable.  For  (1954) extraction using NaHC03 is  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  colorimetric analysis (Olsen and Sommers 1982).  extracts,  after  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).  method used most often is that of Mehta et al.  The extraction  (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  traditionally  been  of  specific  conducted  following extraction.  P0 compounds  using  partition  in  soil  has  chromatography  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  Phospholipids  may  be  chromatography  recovered  extraction with  ethanol-benzene  1975).  methods  These  are  from  (Stevenson  soil  using  sequential  and methanol-chloroform  time-consuming,  and  1994).  a  (Baker  different  extraction procedure is required for each class of P0 compounds. The  introduction  of  31  P 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 U 0 in the radiofrequency range.  The strength of the magnetic field  and frequency of irradiation are related by  X  where  is the gyromagnetic ratio and is characteristic of the nucleus  (Wilson 1991).  Thus, different nuclei occur at characteristic  radiofrequencies strength.  (the  Larmour  frequency)  for  a  given  field  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  frequencies from the Larmour frequencies. of  'chemical  shifts', which  can be  functional groups of an element undergo magnetic resonance.  resonate  at  different  This results in a range  used  to  (Wilson 1991).  detect  different  Not all nuclei  The number of energy states created in  the presence of a magnetic field depends on the spin quantum number of the nucleus investigated. soils are 1 H,  13  C, 27A1, 29Si,  The most commonly studied nuclei in  15  N, and 31P.  In order to detect the signal from nuclei such as  31  P 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  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 1991).  to  their  original  equilibrium  distribution  (Wilson  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  31  P NMR spectroscopy.  However, the natural P levels in soils are usually low.  As  31  P 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  produce clear spectra.  and concentration  Since the first use of  are required  to  31  P 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.  Hinedi et al. 0.5  M  NaOH  1985; Preston et al.  1989; Gil-Sotres et al. with  (Ingall et al.  a  1986; Zech et al. 1987;  1990; Forster and Zech 1993);  citrate-dithionite-bicarbonate  pretreatment  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  phosphonates,  classes  which  31  P  orthophosphate,  NMR  spectroscopy  phosphate  monoesters,  diesters, pyrophosphate, ATP and polyphosphates. agreement  on  quantities  laboratory methods and  of  P  estimated  shows  by  are:  phosphate  There is close  both  conventional  31  P 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  Schlesinger 1995) .  (Olsen  and  Sommers  1982;  Cross  and  Phosphorus extracted by NaOH is thought to be  the non-occluded phosphate bound  to the surfaces of Al or Fe  hydrous  citrate-bicarbonate-dithionite  oxides;  P  removed  by  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 reviewed by Cross and Schlesinger  31  P pools, and was  (1995).  recently  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. annual  rainfall  is approximately  between October and February.  1700 mm,  65% of which  The falls  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.  Figure  3-1:  Map of Vancouver Columbia. 32  Island, with  inset  of  The  British  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  materials of the Suquash Basin  continuous  cover  (Lewis 1982).  of  surficial  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  33  al.  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  shallon  -  heterophylla  Rhytidiadelphus  loreus  -  Abies  amabilis  -  Gaultheria  (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 plicata  Don.) and western hemlock (Tsugra heterophylla  (Thuja  (Raf.) Sarg.)  (Fig. 3-2). This forest type has an open canopy which allows light to penetrate,  Figure 3-2:  and  thus  there  is  a  dense understory of salal  A typical stand of the CH forest type. 34  (Gaultheria  shallon  Pursh.), and blueberry {Vaccinium  Howell and V. alaskaense the mosses oregana  Smith).  Bhytidiadelphus  parvifolium  The forest floor is occupied by  loreus  Kindbergia  (Hedw.) Warnst.,  (Sull.) Ochyra and Hylocomium  with occasional ferns such as Blechnum  splendens spicant  (Hedw.) B.S.C., L. (Germain 1985;  Prescott and Weetman 1994). The forest floors of the and humimors (Klinka et al. somewhat  imperfectly  CH forest  type are  lignohumimors  1981) . Beneath these are moderately to  drained  Duric  or  Orthic  Ferro-Humic  Podzdls. Following  cutting  and burning,  the CH  sites  invaded by salal, which regenerates from rhizomes.  are  quickly  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,  blueberry  (Vaccinium  (Blechnum  spicant),  alaskaense and  with small patches of  and V. parvifolium), deer fern the  35  mosses  Kindbergia  oregana,  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 (Epilobium  dense cover of fireweed  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  characteristics, association.  the and  basis  of  included  topography them  in  the  or  mineral same  soil  ecosystem  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 Canada Expert Committee on Soil Survey 1987).  (Agriculture  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. of the Bf horizon was not measured.  37  The total thickness  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.  LF  H  Bhf  Horizon  Thick.  (cm)  HA-OG  CH-OG  CH-0  CH-5  CH-10  mean  5.4  5.6  3.7  5.1  2.8  range  2-14  2-10  1.5-8  3-10  1-6  mean  7.2  9.0  4.6  5.2  4.3  range  3-13  4-15  1.5-10  2-10  2-9  mean  7.1  10.8  6.2  4.0  4.8  range  2-14  2-25  2-10  2-8  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)  operated by Western Forest Products Limited (WFP). were  which is  The study sites  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  facilitate the comparison of data.  of  the  SCHIRP project,  to  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, 39  and covers 5.8 ha.  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 060%, 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. was  thought  to  have  heavy  brush  potential  and  poor  It  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. glacial thought  till, surficial deposits and covers to  have  a  light  brush  potential  56.6 ha. and  poor  It has It was 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) postburn 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. the  Most soil phosphorus determinations have two phases:  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 Sommers  1982) .  The  second  step usually  involves  (Olsen and  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 1977;  al.  (Syers et  al.  1967;  1968; Sommers and Nelson 1972; Dick and Tabatabai  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  samples (Dick and Tabatabai 1977). 42  for large numbers of  The HC104 digestion method is  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  following ignition of the soil (Syers et al. 1995); the Thomas et al.  with H2S04  1967; Olila and Reddy  (1967) method of hot H2S04 followed by H202  (Roberts et al.  1985; Schoenau and Bettany 1987; Xiao et al.  Compton  Fyles  1994;  and  Cote  1994; Hanafi  and  Syers  1991; 1994);  digestion with combined H2S04, H202 and HF (Bowman 1988) ; digestion with aqua  regia  (HCl and HN03)  Parkinson and Allen  (Crosland et al.  1995); and the  (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  Grimshaw 1985; Tiedemann and Klemmedson 1986; Prescott et al. Silver et al.  and  1993;  1994) .  A direct method to determine the organic P (P0) content of soil has not yet been devised (Olsen and Sommers 1982). estimated  by  either  of  two  indirect  methods:  Soil P0 is 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. unignited  An  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. 1975;  Olsen and Sommers 1982) .  1970; Anderson  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. 1982; Soltanpour et  al.  1970; Anderson 1975; Olsen and Sommers  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. et a.1.  The extraction method used most often is that of Mehta  (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  1990b; Bowman and Moir 1993).  A new extraction procedure  determine P0 was recently introduced by Bowman and Moir  al. to  (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. alkaline soils, the Olsen  et al.  For  (1954) extraction, which contains  NaHC0 3 , 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.  45  The Mehlich  3 extractant  was  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  situations.  it  more  useful  than  Bray  in many  laboratory  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. al.  1990; Cade 1989; Simard et All  al.  1987; Tran  et  1994; Wendt 1995).  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  the  was  to  compare  Saunders  Parkinson and Allen (1975) methods  and  Williams  (1955)  and  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 FerroHumic 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  analysis.  were  oven-dried  overnight  at  60°C prior  to  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. block digester.  The sample was then digested at 360°C on a  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 analysis.  samples  were  oven-dried  overnight  at  60°C prior  to  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. was  read  colorimetrically,  and  the  difference  The extract  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. procedure for  This method was also used as an extraction  31  P 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  METHOD  CH-OG  CH-0  CH-5  CH-10  714.0 (67.5)  585.8 {66.6)  713 .9 (181.4)  675.8 (141.9)  485.3 (163.1)  S&W  729.0 (75.6)  611.2 (78.6)  728.0 (211.0)  725.6 (134.7)  540.8 (121.8)  P&A  561.4 (98.3)  482.8 (152.1)  524 . 0 (219.9)  569.5 (150.6)  515.4 (150.9)  S&W  546.9 (108.2)  511.7 (172.6)  558.0 (195.6)  618.7 (93.9)  528.6 (159.0)  P&A  524 .2 (110.5)  352.1 (89.6)  349.4 (199.5)  468.6 (170.6)  354.5 (66.1)  S&W  459.8 (139.1)  330.3 (97.1)  341.2 (211.7)  445.1 (188.2)  353.3 (147.5)  P&A  361.0 (105.5)  289.2 (115.6)  230.8 (41.1)  235.5 (102.2)  211.8 (70.2)  S&W  316.9 (69.7)  267.8 (115.6)  336.3 (237.3)  212 .8 (70.0)  193 .2 (70.2)  (r)  All  P&A LF  H  Bhf  Bf  CORR.  HA-OG  0.917  LF  H  Bhf  0.869  0.917  0.907  49  Bf 0.561  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. no  PA was  extracted  from  the  Bf  extractant. 50  horizon  with  the  Virtually Mehlich 3  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  CH-0  CH-5  CH-10  492 .0 (98.8)  662.3 (272.4)  551.2 (177.6)  391.1 (124.2)  580.2 (49.8)  471.2 (75.2)  358.2 (129.7)  578.7 (107.8)  420.3 (109.8)  B&M  283 .8 (133.9)  308.7 (123.2)  360.4 (243.9)  401.0 (127.5)  306.2 (112.2)  S&W  442 .3 (90.1)  410.7 (156.0)  349.7 (93.4)  469.7 (101.1)  417.0 (133.9)  B&M  227.8 (110.3)  145.1 (89.2)  174 .2 (97.2)  220.2 (111.0)  198.7 (93.1)  S&W  367.2 (119.9)  276.2 (86.0)  234 .3 (148.6)  365.0 (168.1)  293.9 (134.6)  B&M  99.6 (35.7)  64.8 (36.4)  41.3 (23.0)  63 .5 (36.8)  21.6 (24.8)  S&W  188.8 (51.3)  157.7 (85.4)  205.9 (251.6)  109.7 (70.8)  65.4 (31.5)  (r)  All  Bf  METHOD  HA-OG  B&M  527.8 (162.0)  S&W  LF  H  Bhf  Bf  CORR.  0.712  CH-OG  LF  H  Bhf  0.375  0.443  0.562  51  0.403  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  METHOD  CH-OG  CH-0  CH-5  CH-10  BRAY  38 .2 (6.2)  30.5 (10.0)  89.0 (31.4)  27.2 (9.4)  MEHLICH  57.6 (8.5)  54.9 (28.3)  110.5 (32.5)  56.3 (28.3)  33.9 (16.3)  BRAY  20 .6 (6.8)  20.6 (9.7)  44 .1 (24.9)  39.1 (21.4)  12 .6 (5.1)  28.9 (13.7)  42 .8 (29.4)  73.6 (31.4)  68.9 (45.9)  36.3 (18.2)  LF  H MEHLICH BRAY Bhf MEHLICH  10.2 (7.0)  6.06 (2.92)  8.84 (8.12)  7.75 (15.2)  6.90 (10.21)  7.29 (12.47)  9.24 (6.54) 9.77 (17.4)  17.3 (7.15)  8.59 (1.78) 5.44 (5.89)  BRAY  4.80 (0.89)  6.31 (1.02)  6.43 (1.12)  6.06 (1.05)  6.60 (0.51)  MEHLICH  0.41 (0.88)  0.03 (0.10)  0 .36 (0.74)  0.0 (0.0)  0.21 (0.63)  Bf  CORR.  HA-OG  (r)  All 0.874  LF  H  Bhf  Bf  0.889  0.698  0.777  -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  HC10 4 is believed to extract the majority of PT from most soils (Sherrell and Saunders 1966; Syers et al. but  its  hazardous  discontinue its use. method  aspects  have  1967; Syers et  caused  many  1968),  laboratories  The Saunders and Williams  and the Parkinson and Allen  al.  to  (1955) ignition  (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  et al.  1970;  Anderson 1975; Olsen and Sommers 1982; Soltanpour et al.  1987;  Harrap  1963; Williams  Condron et al.  and Walker  1967; Williams  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 slightly  and Allen method higher  PT  is better  concentrations  in  for mineral the  soils.  organic  The  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 procedures  P0  is  estimated  by  either  (Olsen and Sommers 1982).  ignition  or  extraction  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. difference  between  the  results  from  ignition  and  The  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. extracts  This method requires that Pz be determined  prior  to digestion;  the  P0 content  in the  is calculated  by  subtracting the pre-digestion Pz content from the post-digestion PT content. could  In this study, the Px content of the undigested extracts  not be determined  because the extracts were  coloured for reliable colorimetric analysis.  too darkly  This is a common  problem when soils with high organic matter contents are analyzed for P0 using extraction procedures (Olsen and Sommers 1982). extracts from the Bowman and Moir procedure were analyzed by  The  31  P NMR  spectroscopy (Chapters 4, 5 and 6 ) , which revealed that 20% to 80% of  the  P  subtracting  in  solution the  Px  was  in  orthophosphate.  the  extract  Consequently,  results  in  a  not  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 55  procedures  for  P0  is  hydrolysis of P0 compounds during extraction. P forms revealed by  The diversity of the  31  P 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  two  reported  high  correlations  between  the  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. horizon  In this study, virtually no PA was removed from the Bf 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. et  al.  Tran  (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  both  (Chapters  4  and  5) 56  shows  that  organic  and  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 method.  (1955)  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  orthophosphate,  of  in  soils,  polyphosphate,  orthophosphate  monoesters  orthophosphate  diesters  analytically  P  less  phosphonate,  such such  complex  including  as as  than  inositol  pyrophosphate, phosphate,  phospholipids, the  inorganic  and  detailed  it  and is  partition  chromotography techniques otherwise required to identify specific organic P compounds. usually low.  As  However, the natural P levels in soils are  31  P 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 involving  31  P NMR studies have employed a rapid extraction technique ultrasonic  dispersion  in  0.5  M NaOH, which  extracts less than 50% of the total phosphorus  (Newman and Tate  1980; Tate and Newman 1982; Ogner 1983; Hawkes et al. al.  1985; Zech et al.  1987; Hinedi et al.  1990; Forster and Zech 1993).  usually  1984; Zech et  1989; Gil-Sotres et  al.  Others have used 0.5 M NaOH without  58  sonication  (Preston et al.  1986) or with a citrate-dithionite-  (Ingall et  bicarbonate pretreatment  al.  1990) .  also removed less than 50% of the total P.  These treatments  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. al.  Condron et  (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. ignition  Up to 80% of the total organic P as determined by  (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. of the total organic P.  This too removed about 80%  Emsley and Niazi  (1983) tried tetra-n-  butyl ammonium hydroxide (Bu4NOH), hoping to utilize the salting-in effect of the large organic cation. effect  did  not  occur,  and  they  However, this salting-in  concluded  effective, but no more so, as NaOH or KOH.  that  Bu4NOH  Hinedi et al.  was  as  (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 1984; Hutson et al.  1992).  31  P NMR spectra  (Hawkes et  al.  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:  Adams and Byrne (1989)].  Fe>Al>>Ca>>>Na;  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  EDTA  is  able  The NaOH can solubilize the organic P, while to  chelate  efficiency of P extraction.  metal  cations  to  increase  the  This method extracted as much as twice  the amount of organic P as NaOH alone (Bowman and Moir 1993). has not yet been tested as an extractant for  It  31  P NMR spectroscopy.  With any soil extraction procedure, there is a danger that soil  P  compounds  extraction.  will  be  chemically  altered  during  or  after  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  paramagnetic  form  ions.  and  relatively  Results  with  determination are site specific 60  high any  levels  method  for  of  other soil  P  (Anderson 1975). Therefore, one  objective of this research was to compare several soil extraction procedures, to determine the one most suitable for of  forest  floor  samples  Vancouver Island.  CH  and  HA  P NMR analysis  forests  of  northern  The second objective was to examine the method  of Bowman and Moir extractant for  from  31  (1993) to determine its effectiveness as an  31  P 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 (CH10) 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  HA-OG  CH-OG  CH-0  CH-5  CH-10  pH  3.1  3.5  5.0  3.4  4 .2  49.9  37.2  23 .2  46.7  49.6  LOI  3500  695  213  1742  1006  Total N (%)  1.159  0.851  0.879  0.807  0.995  C/N  43 . 1  43 . 7  26.4  57.9  49.8  Avail Ca (mg/kg)  1906.1  3300.2  7395.0  3639.1  7188.8  Avail Mg (mg/kg)  393.4  255.6  640.0  498.0  673.5  Avail Fe (mg/kg)  139.6  182.6  220.0  118.3  112.2  Avail Al (mg/kg)  380.7  649.1  720.0  364.9  193.9  Avail P (mg/kg)  40 . 04  51.5  68.4  23 .47  21.16  Total P (mg/kg)  674.0  582.0  796.0  713 .0  653.0  Inorg P (mg/kg)  129.0  162 .0  396.0  140.0  124 . 0  Org P 1 (mg/kg)  545.0  420 .0  400 .0  513 .0  589.0  C  (%)  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  62  procedures,  samples  were  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. vortexed  for 2 min.  Samples were  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. were  left  to stand  supernatants  were  All of the samples  for 2 h, and then were centrifuged. transferred  into  10 ml  NMR  tubes  and  The 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). 63  Peak areas were  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 control.  A blank containing 3 0 ml of water was used as  The soil sample used was CH-OG.  NaOH was not used in  this trial as it caused the metals to precipitate. were placed in stoppered 100 ml centrifuge tubes.  The samples  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  compounds,  the effect  5 g samples of  of the extractants  CH-OG had  0.05  on various P  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, 64  Grade  II,  Sigma  A-3377);  glycerophosphate polyphosphate These  were  (disodium  pentahydrate,  Sigma  G6504);  (formed by fusion of KH2P04 as per Kulaev  then  prepared  for NMR  analysis  as was  or  K-  (1979) ) .  previously  described.  Results NMR Spectra Figure 5-1, A-E, displays the NMR spectra generated from these extractant  trials,  while  Table  interpretation of the peaks.  5-2  shows  a  guide  for  the  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) .  the  extractant  NaOH  monoester P overlap.  + EDTA  In some of the spectra for  the peaks  for  orthophosphate  and  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. extraction.  The broadest peaks were produced by the NaOH + EDTA 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 PHOS  IS  10 s ORTHO MONO  ~T~ 0 PPH DIEST  "T" PYRO  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  C H O 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  It 3 ORTHO MONO DIEST  -If  PYRO  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<  -Jt  5  ORTHO MONO  OIEST  PYRO  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  COMPOUNDS  15-20  phosphonates  6-8  orthophosphate  3-6  phosphate monoesters -inositol phosphates -sugar phosphates -mononucleotides phosphate diesters -phospholipids -RNA, DNA  l-(-D (-3)-(-6) (-20)  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. 51A) .  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  Phos  Orth  Mono  Dies  Pyro  Poly  Chelex + NaOH  0  26  44  15  15  0  Chelex + H 2 0  0  62  16  22  0  0  NaOH + EDTA  0  21  49  15  7  7  NaOH  0  36  51  4  9  0  Chelex + 2 NaOH  0  58  30  5  7  0  Chelex + NaOH  0  36  40  14  10  0  Chelex + H 2 0  0  54  18  23  5  0  NaOH + EDTA  0  17  33  39  0  11  NaOH  0  51  43  0  6  0  Chelex + NaOH  0  74  18  6  2  0  Chelex + H 2 0  0  73  13  13  1  0  NaOH + EDTA  0  51  33  11  6  0  NaOH  0  55  45  0  0  0  Chelex + 2 NaOH  3  34  38  8  17  0  Chelex + NaOH  2  31  39  9  19  0  Chelex + H 2 0  3  49  22  18  8  0  NaOH + EDTA  0  23  40  18  7  12  NaOH  0  49  33  0  18  0  Chelex + 2 NaOH  0  34  42  8  16  0  Chelex + NaOH  1  15  57  9  19  0  Chelex + H 2 0  0  68  14  10  4  4  NaOH + EDTA  0  17  27  32  10  14  NaOH  0  49  41  0  10  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  HA-OG  CH-OG  CH-0  CH-5  CH-10  210.1  210.5  166.8  182.6  183 .7  % Total P  31.2  36.2  21. 0  25.6  28.1  (mg/kg)  77.4  66.2  82.8  55. 0  71.6  % Total P  11.4  11.4  10.4  526.4  424.8  643.5  451.6  643 .1  78.1  73.0  80.8  63.3  98.5  205.0  198.8  159.6  198.8  207.6  30.4  34 .2  20.1  27.9  31.8  (mg/kg)  NaOH+Chelex  H 2 0+Chelex  (mg/kg)  NaOH+EDTA  % Total P (mg/kg)  NaOH  % Total P  7.71  11.0  Table 5-5: Orthophosphate diester/monoester ratios, calculated from Table 5-3. HA-OG  CH-OG  CH-0  CH-5  CH-10  Chelex/2 NaOH  N/A  0.30  N/A  0.21  0.19  Chelex/NaOH  0.34  0.35  0.33  0 .17  0.16  Chelex/H 2 0  1.38  1.27  1.00  0.82  0.71  NaOH/EDTA  0.31  1. 18  0.33  0.45  1.18  NaOH  0.08  0  0  0  0  the fewest types of P compounds, and  monoesters  in  all  showing peaks for orthophosphate  samples,  but  pyrophosphate in only a few samples.  peaks  for  diesters  There were no peaks for  phosphonates or polyphosphates with this extraction procedure.  73  and  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  NaOH  NaOH + EDTA  H20 + Chelex  NaOH + Chelex  Fe  390  210  300  90  Mn  60  240  0  0  Cu  30  36  21  18  Zn  12  21  0  0  Fe  1080  870  870  450  Mn  150  720  60  0  Cu  24  27  9  9  Zn  0  15  0  0  Fe  900  1890  780  450  Mn  150  960  90  60  Cu  24  33  24  3  Zn  0  27  0  0  Fe  240  120  120  0  Mn  120  900  60  0  Cu  36  24  21  12  Zn  21  45  0  0  Fe  300  180  300  120  Mn  210  1530  60  0  Cu  9  27  21  21  Zn  9  51  12  0  74  ug/1) .  The  Fe  levels  in  the  NaOH  +  EDTA  intermediate, except for CH-0, the recent burn. solutions contained the lowest levels of Fe.  solutions  were  NaOH + Chelex  NaOH + EDTA extracted  the greatest concentration of Mn, especially in the three postburn samples.  The  lowest Mn  levels were  found  extracts.  The NaOH and NaOH + EDTA extracts contained similar  levels of Cu; NaOH + Chelex had the least Cu.  in the two  Chelex  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) 120 100  A)  30  20  40  100  Fe Added (mg/!) Water + Chelex ^ E D T A Mn in Solution (mg/l) 120  B)  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  extracts are shown in Figure 5-3, A and B.  to the NMR  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 77  the same percentages of  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  WrAysvVj  w^A^Ay^vvvH  + ATP  ^  Vvww v ^ j ^WU^AMV-*  ^^vA^J^VvOV,  GLYCEROPHOSPHATE ^ v ^^M^Avv^AMv^^^^^  ORIGINAL EXTRACT  T 20  PHOS  —i— 15  —I -IS  — i —  o  10  ORTHO  PPH  MONO  DIEST  PYRO  -25  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. P in Soil mg/kg  ortho phos .  NaOH+EDTA + polyphos.  6500  11.7  NaOH+Chelex + polyphos.  3800  NaOH+EDTA + ATP  3600  NaOH+Chelex + ATP  Sample  Monoester %  Diester  6.3  4.3  12.8  40.3  4.9  16.1  n/a  38.7  14 .5  10.2  7.2  24.7 alpha  21.7 beta  21.7 gam.  3500  2.8  17.8  4.7  28 .0 alpha  27.1 beta  19.6 gam.  NaOH+EDTA +glycerophos.  5900  26.4  63.9  9.7  0  n/a  0  NaOH+Chelex +glycerophos.  3000  3 .8  89.7  6.5  0  n/a  0  NaOH+EDTA orig. extract  424 .8  33 .3  16.7  38.9  0  n/a  NaOH+Chelex orig. extract  210.5  36.1  40.3  13 .9  9.7  n/a  o  0  o  Pyrophos . %  (-10) peak %  poly phos %  n/a  64.9  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 1989) or from sequential extraction of soil (Condron et  al.  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  and Zech 1993) .  al.  1990; Gil-Sotres et  al.  1990; Forster  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. peaks  were  clearly  separated  concentrations of Fe and Mn. cause most of the peak overlap:  The one sample in which the  (HA-OG)  contained  the  lowest  High concentrations of Mn seem to 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  been observed elsewhere (Hutson et al.  1992).  31  P NMR spectra has  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. the soil, however.  EDTA alone did not extract much Fe from  More Fe was released when EDTA was combined  with NaOH, probably due to the solubilization of organic matter (Stevenson 1994). and  signal-to-noise  extractions. to  The best spectra, in terms of peak separation ratio, were produced by the Chelex + NaOH  These were further improved by adjusting the pH prior  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.  al.  1990).  1989; Condron et al.  1990a; Gil-Sotres et  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. (1963)  reported  1990), and Hance and Anderson  that phospholipids were  alkaline solution.  readily  hydrolysed  The diester/monoester ratios for NaOH are also  comparable to reported values of 0.18-0.46 (Zech et al. Sotres et al.  in  1990; Forster and Zech 1993).  1985; Gil-  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. should  be  noted  that  the  lack  of  clear  separation  of  It 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. Byrne  1989),  and  are  usually  at  very  low  1987; Adams and  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. other reagents (Subbarao et al.  They may be hydrolysed by  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. artifacts,  as  the  NaOH  + EDTA  It is unlikely that they are  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.  84  The major  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 possible with a variety of reagents.  31  P NMR analysis is  The forms of P which dissolve  depend on the reagent; the complex nature of soil organic P may make  it  impossible  compounds.  for a single extractant  to dissolve  all P  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  31  P  NMR  analysis.  It  extracted  a  higher  concentration of P and a greater diversity of P compounds that any other  extractant  compounds.  tested,  with  less  apparent  hydrolysis  of  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  P  for  samples  with  high  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  31  P 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. that  trees  phosphorus  on and  the  CH  sites  responded  were  well  to  (1989a, b) indicated  deficient  in  nitrogen  fertilization,  and  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). studied.  Some  research  Phosphorus has been less thoroughly 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. et  al.  1994),  but  no  detailed  investigation  1993; Prescott 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 1993) .  aerobic  incubations  in the laboratory  (Prescott  et  al.  Also indicative of less complete decomposition was the  slightly higher ratio of carbohydrate to lignin moieties in the CH forest, as revealed by  13  C 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  31  P 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. 87  In the  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. subsequently  stored  at  room  temperature  in  All samples were 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  (Lavkulich 19 81) .  procedure  to  convert  the  N  to  ammonium  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  The new  analysis  on the Lachat FIA.  extraction  procedure of Bowman and Moir (1993) was utilized to extract P0 from air-dried samples, followed by persulphate digestion (Bowman 1989) and  colorimetric  comparison  of  analysis  these methods  on  a  Technicon  to others  Autoanalyzer.  for PA, PT, and  A  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  31  P NMR spectroscopy.  The samples selected  were high in PT, and had chemical characteristics close to the mean values for the forest type they represented. 89  LF, H, Bhf and Bf  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.  CONSTANT  + TYPE  The model statement used was:  + HORIZON  + LOCATION{TYPE}  VARIABLE =  + 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. be significant. indicates  that  Thus the HORIZON*LOCATION{TYPE} term should not A significant result for the TYPE*HORIZON term  the  CH  and HA  forest  different for a particular horizon.  types  are  significantly  Pearson pairwise correlations  and Tukey' s HSD tests were also conducted with the Systat program. Homogeneity against  of  variance  estimates.  Log  was and  determined log  (n+1)  by  plotting  residuals  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  differences between moisture  in the Bf.  the CH and HA  There were no forest  types.  significant The air-dry  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. methods was in the Bf horizon.  The highest pH with both  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 CaCl 2 . A * indicates a statistically significant difference between the HA and CH forest types for a horizon at p<0.05. (n=9) HOR.  FOR. TYPE  FIELD MOISTURE %  AIR-DRY MOISTURE %  pH IN WATER  pH IN CaCl2  LF  HA  226 (120)  12.65 (0.84)  3.6(0.19)*  3.2(0.14)*  CH  203 (122)  12.95 (1.18)  3.9(0.22)  3.6(0.27)  HA  230 (76)  13.39 (3.74)  3.4(0.11)  2.9(0.10)*  CH  314 (149)  16.17 (6.06)  3.6(0.13)  3.1(0.19)  HA  206 (109)  12.67 (10.17)  3.8(0.32)  3.2(0.32)  CH  249 (152)  HA CH  H  Bhf  Bf  8.70  (4.69)  3.8(0.23)  3.2(0.17)  97 (48)  13.53  (8.70)  4.6(0.31)  4.0(0.31)  84 (27)  10.44 (10.19)  4.5(0.32)  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  HORIZON*LOCATION{TYPE}  horizons  than  term was also  the  CH.  However,  significant  the  (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. TYPE  HORIZON  LOCATION {TYPE}  TYPE* HORIZON  HORIZON* LOCATION {TYPE}  MULT. R2  Field Moisture  0.551  0.000  0.000  0.349  0.010  0.796  Air Dry Moisture  0.111  0.001  0.002  0.085  0.191  0.577  pH in Water  0.064  0.000  0.313  0.018*  0.052  0.833  pH in CaCl2  0.003  0.000  0.005  0.003*  0.226  0.848  C  0.016  0.000  0.180  0.000*  0.006*  0.963  N  0.000  0.000  0.118  0.121  0.013  0.924  C/N  0.001  0.000  0.078  0.306  0.973  0.687  LOI  0.880  0.000  0.455  0.011*  0.055  0.946  Extr. Ca  0.666  0.000  0.003  0.890  0.243  0.795  Extr. Mg  0.001  0.000  0.000  0.135  0.013  0.871  Extr. Al  0.000  0.000  0.099  0.184  0.555  0.883  Extr. Fe  0.000  0.000  0.489  0.103  0.103  0.731  92  C, and design.  thus did not  fit the assumptions  for the experimental  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  horizons, and had  a  had  lower  higher  C/N  ratio  N  concentrations than  the  CH  in all 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.  FOR. TYPE  C %  TOTAL N %  C/N  LOI  LF  HA  46.58 (5.10)  1.06 (0.18)  44.54 (4.80)  1975.4 (907.7)  CH  48.05 (4.31)  0.92 (0.11)  52.56 (6.20)  2000.5 (612.2)  HA  40.85 (7.9)  1.01 (0.10)  40.53 (7.66)  1168.0* (921.3)  CH  47.32 (6.2)  0.91 (0.16)  53.40 (11.59)  2014.4 (780.6)  HA  20.03 (4.67)  0.77 (0.14)  26.09 (4.48)  142.1 (104.7)  CH  14.32 (4.74)  0.54 (0.20)  29.78 (9.55)  87.8 (64.5)  HA  8.81 (2.55)  0.24 (0.06)  37.04 (4.99)  25.54 (6.41)  CH  6.74 (1.21)  0.17 (0.04)  40.99 (8.25)  21.53 (4.53)  H  Bhf  Bf  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.  FOR. TYPE  EXTR Ca (mg/kg)  EXTR Mg (mg/kg)  EXTR Al (mg/kg)  EXTR Fe (mg/kg)  LF  HA  1853.0 (307.3)  419.5 (78.6)  604.6 (155.4)  216.8 (73.7)  CH  3354.3 (616.5)  372.7 (99.9)  274.4 (209.2)  112.4 (29.6)  HA  1660.9 (762.1)  532.9 (117.2)  956.5 (274.1)  309.0 (99.8)  CH  2720.9 (1027.4)  441.3 (150.5)  355.8 (318.9)  127.3 (88.0)  HA  945.0 (1045.2)  334.7 (207.5)  1336.9 (301.4)  442.5 (110.5)  CH  1272.8 (1325.5)  194.6 (169.8)  1043.6 (406.4)  347.7 (166.8)  HA  39.4 (11.1)  55.9 (52.8)  2118.5 (515.4)  187.3 (117.0)  CH  222.3 (286.1)  40.5 (39.9)  1917.7 (222.3)  156.3 (107.2)  H  Bhf  Bf  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.  FOR. TYPE  Fe-PYRO %  Al-PYRO %  Mn-PYRO %  LF  HA  0.041 (0.044)  0.089 (0.036)  0.030 (0.018)  CH  0.010 (0.031)  0.051 (0.040)  0.063 (0.036)  HA  0.207 (0.219)  0.219 (0.157)  0.009 (0.012)  CH  0.069 (0.184)  0.081 (0.157)  0.009 (0.012)  HA  1.528 (0.662)  0.573 (0.149)  0.004 (0.003)  CH  1.395 (0.934)  0.577 (0.234)  0.000 (0.000)  HA  1.442 (0.476)  1.559 (0.297)  0.002 (0.004)  CH  0.910 (0.401)  1.188 (0.364)  0.000 (0.000)  H  Bhf  Bf  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. TYPE  HORIZON  LOCATION {TYPE}  TYPE* HORIZON  HORIZON* LOCATION {TYPE}  MULT. R2  Fe AAO  0.206  0.000  0.200  0.104  0.292  0.634  Fe CBD  0.561  0.001  0.041  0.238  0.378  0.570  Fe Pyro  0.080  0.000  0.154  0.853  0.080  0.850  Al AAO  0.645  0.000  0.026  0.968  0.269  0.876  Al CBD  0.078  0.000  0.123  0.203  0.765  0.849  Al Pyro  0.014  0.000  0.552  0.331  0.831  0.909  Mn CBD  0.000  0.736  0.015  0.748  0.075  0.690  Mn Pyro  0.074  0.001  0.285  0.074  0.847  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 citratebicarbonate-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  Fe-CBD %  Al-CBD  Mn-CBD  Fe-AAO  Al-AAO  %  %  %  %  HA  1.802 (0.950)  0.525 (0.186)  0.009 (0.003)  1.235 (0.524)  0.560 (0.203)  CH  2.013 (1.296)  0.477 (0.175)  0.001 (0.001)  1.283 (0.691)  0.519 (0.182)  HA  3.276 (0.437)  1.802 (0.442)  0.012 (0.013)  2.413 (0.342)  2.272 (0.576)  CH  2.636 (1.111)  1.465 (0.385)  0.001 (0.001)  1.780 (0.479)  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  matter.  with  organic  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. 6-1.  The results of these calculations are shown in Figure  There were no significant differences between the CH and HA  forests in amorphous or crystalline Fe (Fig. 6-1A) or Al (Fig. 61B).  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. either  Crystalline Al was not found in the Bhf or Bf horizons of 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  CH Bf  Horizon lOrganic E d A m o r p h o u s ED Crystalline % Al  B)  HA Bhf  CH Bhf  HA Bf  CH Bf  Horizon 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. highest in the LF, and decreased with depth.  Both P0 and PT were 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'  an  ^  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.  FOR. TYPE  Avail. P (mg/kg)  Total P (mg/kg)  Org. P (mg/kg)  C/P  LF  HA  38.17 (6.19)  714.0 (67.45)  527.8 (162.0)  659 (102)  CH  30.46 (10.01)  585.8 (66.59)  492.0 (98.8)  815 (135)  HA  20.55 (6.75)  561.4 (98.29)  283.8 (133.9)  736 (135)  CH  20.62 (9.65)  482.8 (152.1)  308.7 (123 .2)  1055 (389)  HA  10.19 (7.02)  524.2 (110.5)  227.8 (110.3)  384 (67)  CH  6.06 (2.92)  352.1 (89.55)  145.1 (89.21)  392 (152)  HA  4.80 (0.89)  3 61.0 (105.5)  99.59 (35.74)  256 (82)  CH  6.31 (1.02)  289.2 (115.6)  64.78 (36.39)  252 (90)  H  Bhf  Bf  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 611.  These are means calculated from each sample, and so are not  completely additive. P  HCI/  P  NaoH an<3  procedure.  P  CBD  The Px was calculated from the sum of  extracted by the Chang and Jackson fractionation  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. TYPE  HORIZON  LOCATION {TYPE}  TYPE* HORIZON  HORIZON* LOCATION {TYPE}  MULT. R2  Avail P  0.171  0.000  0.028  0.242  0.198  0.872  Org P  0.293  0.000  0.763  0.641  0.564  0.779  Tot P  0.001  0.000  0.313  0.505  0.637  0.727  P-HCl  0.934  0.000  0.012  0.762  0.013  0.688  P-NaOH  0.580  0.000  0.000  0.116  0.000  0.858  P-CBD  0.012  0.000  0.814  0.339  0.589  0.823  C/P  0.021  0.000  0.931  0.056  0.891  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.  FOR. TYPE  P-HC1 (mg/kg)  P-NaOH (mg/kg)  P-CBD (mg/kg)  Bhf  HA  7 (0.5)  44 (15.4)  30 (9.8)  CH  7 (0.5)  37 (28.0)  22 (17.0)  HA  16 (7.7)  23 (8.4)  96 (15.9)  CH  19 (13.6)  25 (12.4)  74 (23.6)  Bf  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.  FOR. TYPE  ORG. P  INORG. P  % RECOVERY  LF  HA  73.4 (20.0)  N/A  73.4 (20.0)  CH  83.3 (12.7)  N/A  83.3 (12.7)  HA  51.4 (23.3)  N/A  51.4 (23.2)  CH  65.5 (28.3)  N/A  65.5 (28.3)  HA  42.7 (17.4)  17 (6.1)  59 (15.7)  CH  39.1 (17.9)  19 (10.7)  58 (22.1)  HA  27.5 (3.4)  39 (7.8)  67 (8.7)  CH  23.9 (12.0)  44 (15.6)  68 (17.3)  H  Bhf  Bf  101  percentage recovery ranged from 51.4% to 83.3%, but the variability was high. Table 6-12 displays the correlation matrix. and Al extracted by CBD, AAO and pyrophosphate positive correlations with one another.  Extractable Al, all had high  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. relationship  between  the values  There was a positive  for the two methods  for pH.  Positive correlations were also seen with extractable Mg to Ca, and with Mn extracted by pyrophosphate to CBD. positively with C, extractable Mg, and P0.  Total N correlated There was a negative  relationship between C and pH in CaCl2, and a positive one with PT to P0. The  31  P 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?  rtdooddoooooooo  I  BS^SSBBSS^SM  & o  -Jdoddddodododotp"  "~  § o •* N vo o •* o N W tn £ N N N ~;  §3  z  ••  HdoddodddoodoooD  f  rtDododddddooDOOoo o N N N •£ vo mfcOD •-> N o N  <JJ  q; m y  § 8s8SR2?S3!?ieSSSI8SB  I  m  d.  -??  § — roin  S  |  ^ooooooooooooooooo  c3  ! N S 3 C5 N •* •* vfi 8 o $ tn f\i (n -i it) o• o  L  •HOOOOOOOO  H  ooo  o  ?°°??T-T  -Jddd £  i D.  •Hoddd  m  UJ  I  ~6ao6666ooaooa66o6a6  a  §sifeisai§s§ss§888sass  a.  •loaaaa  (M  ••••••• Hoaooao  8SB$8Si3B28S3$IBiBS58SB -ddddddddddddddddddddd  s I cc  BsBSS^BBB ^oaaaaada  -dddddddddddddddddddddddd  IBiBBsBBBiBsBSBSSBSSIglsgSill Sddddddddddddddddddddddddd  SBSBBSMSfiSBSisSiSSSBSBSBB OODOOOOaOOOOOQOODOOOOOD  o a o r p. — • .• . n  T > «(£  CE  loaoaoaa  -ododddo^dddddddipoooodd  §BI8is?SSi5SBB5f2SI!§SfisSS^  ft  !*8BS*gB  SB§883gSg§sg£SS§g§BBgg8  zmmm  B85Sga2SiSSSBS§85BB«Sa»5  I i  J  ??  -i X a :  •? •? 1  Lain  x y J( x  UJ UJ UJ UJ U. Li. t  103  ~6666aa6aa  3  §B3J8SS2$i$ -oddooododd  5 BISSSIilSiBSSS x  ~6o66aaaaa66  immmm* ^~66dd6aaaa66a  HEMLOCK-AMABAUS FIR OLD GROWTH A  H E M L O C K - A M A B A U S FIR OLD GROWTH  I  If HORIZON  /  8 HORIZON  Bbf HORIZON  /A^Ayv^^ Bf HORIZON  -Jl ORTMO M O N O  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  ORTHOMOHO  PHOS  r  J*^  pYRO  -if  -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  HA  CH  CH  Hor.  A  B  A  B  Phos  Orth  Mono  Dies  Pyro  Poly  LF  3  10  24  39  8  16  H  0  20  20  44  0  16  Bhf  4  17  29  50  0  0  Bf  0  60  25  10  5  0  LF  0  33  45  17  5  0  H  9  30  30  26  5  0  Bhf  0  22  54  15  0  0  Bf  0  31  51  18  0  0  LF  0  17  33  39  0  11  H  5  7  38  36  5  7  Bhf  4  38  36  17  5  0  Bf  0  23  61  16  0  0  LF  0  31  38  18  13  0  H  0  39  32  20  9  0  Bhf  0  35  34  24  7  0  Bf  0  29  32  24  15  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  monoester P were similar in the LF, H, and Bhf.  and  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  31  the extractant used in this study for  P NMR spectroscopy is that  the orthophosphate and monoester peaks overlap in some spectra. a  valley  between peaks  occurs  at  6 ppm when  there  is  As  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  different from profile A.  (Fig. 6-2B,  H  horizon,  orthophosphate.  where  6-13)  was  quite  The orthophosphate and monoester peaks  were distinct in all horizons. the  Table  it  Monoester P dominated in all but  was  proportionally  comparable  to  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  H  phosphonate  and  pyrophosphate  peaks  in  the  and  horizons, and polyphosphate peaks in the LF and H horizons. profile B  Bhf  In CH  (Fig. 6-3B; Table 6-13), most of the P was found as  orthophosphate  or  as  monoesters, 107  at  comparable  percentages.  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 1994).  (eg:  de Montigny 1992; Keenan 1993; Prescott et  al.  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  13  C 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. and C.  There is a high positive correlation between LOI  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  polysaccharides (de Montigny 1992; de Montigny et al.  labile  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. increased  This higher C concentration may also be due to  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. Prescott  et al.  1995).  1993;  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  presence of fixed NH4 in clays.  to mineral horizons to the  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). more  Ca  in  differences  de Montigny (1992) found significantly  the F horizon of CH for Mg.  Keenan  than HA, but no  significant  (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. species  (Krajina  1969;  associated tree species.  Cedar is known to be a Ca accumulator Ballard  and  Carter  1986)  relative  to  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  those  researchers  were  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  the  of  organic-associated  Fe  and  Al  reflects  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: species differences  in Al content; differences  tree  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 Keenan  (Sanborn 1987).  The finding of  (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 1987),  and also reflects  (Sanborn  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  al.  probably  reflects  methodological differences, such as length of extraction.  Prescott  et al.  et  (1995)  for  the  LF.  This  (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. P was not measured by other SCHIRP researchers. 112  Organic  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  cycling.  are not  However,  inherently differences  different may  be  in  P  masked  contents by  the  and high  variability in all measured P concentrations, and as shown by the 31  P 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. the LF, virtually all of the soil phosphorus is found as P0.  In 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  litterfall.  contents  of  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  reflects the influence of litter  (Compton 1994).  31  P-NMR also  Diester P in  plant tissues occurs as phospholipids in cell walls, and as nucleic acids.  Plant  mononucleotides  monoesters  consist  and  phosphates,  sugar  intermediaries (Bieleski 1973). transfer, but are not seen in  of  inositol which  are  phosphates, metabolic  ADP and ATP are involved in energy 31  P-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  (Condron et al. store  1985).  excess  P  ectomycorrhizal  during  extraction  for NMR  analysis  Fungi and other microbes have been shown to  in  polyphosphates  fungi are thought  (Bieleski  to transport  hyphae as polyphosphate (MacFall et al.  1992).  1973),  while  P within their  The polyphosphates  of the LF and H horizons are most likely from the mycorrhizae associated with trees on these sites. Phosphonates (Hilderbrand  are  1983),  formed  and  are  by  a  variety  thought  of  soil  to accumulate  microbes  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  for  of pyrophosphate  is also  thought to be a marker  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  (Dighton and Boddy 1988), and may be responsible 114  bacterial  for the wide  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) . by  In the H horizon, the P forms shown  31  P-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 found  in  the  spectroscopy. soil  forest  floor  of  these  forests  (1992) also  using  13  C-NMR  Lipids are usually considered to be quite labile in  (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'  P  CBD  an<  3  P  HCI  •  P  NaoH 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 1980;  Olsen and Sommers  1982) .  In the Bhf horizon of  al.  these  forests, PNa0H is higher than PCBD and PHCi/ suggesting that Px is nonoccluded, 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). also present in some Bhf and Bf horizons.  Pyrophosphate is  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. that apatite  This is somewhat unusual in  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  Orthophosphate revealed  by  and  31  P-NMR  of,  or  in  monoester  addition  to,  phosphate  are  spectroscopy.  Fares  calcium the al.  et  phosphate.  main  P  (1974)  forms 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  31  P-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  31  P-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. have  been  obtained  by  In retrospect, better results might  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  31  P-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 the soil types present on these sites.  of  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. measured  However, the high variability in all  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  31  P-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. forms  In the H horizon there is more humification, and P  associated  with  soil  organisms  are  seen.  inorganic phosphorus is predominantly non-occluded.  In  the Bhf,  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 1994) .  (Prescott and Weetman  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 1975).  (Grier  After a fire is out, nutrients may be redistributed by wind  and water erosion or by leaching of the ash layer and soil. frequently observed nutrient effects include:  Some  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  120  al.  1983;  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 Brockley et al.  1992).  (Feller 1982;  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  (DeBano and Klopatek 1988; Saa et al.  forest  productivity  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 121  digestion  procedures,  while  fractionation and were  used  31  P nuclear magnetic resonance (NMR) spectroscopy  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  spectroscopy. Chapter  6.  sites  The The  procedures for  31  were  criteria  chosen  for  for sample  extraction,  sample  analysis  selection  preparation  31  by  are and  P  NMR  listed  in  analytical  P 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 each  a nested design forest  statement  used  LOCATION{AGE} pairwise  type,  as  was: +  (Hicks 1982), with location nested was  VARIABLE  TYPE*HORIZON  =  in  Chapter  CONSTANT  +  4.  AGE  The +  model  HORIZON  +  + HORIZON*LOCATION{AGE}.  Pearson  correlations and Tukey's HSD tests were also  conducted  with the Systat program. plotting  described  within  residuals  Homogeneity of variance was tested by  against  estimates.  Log  and  log  (n+1)  transformations were performed where necessary.  Results The moisture,  results pH  from  in water  the  analysis  and pH  for  in CaCl 2  field moisture, are  shown  air-dry  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  HOR.  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] AGE  3.60 (0.27) b  0-YR  67 (45) b  12.57 (2.58) a  4.96 (0.45) a  4.58 (0.40) a  5-YR  106 (132) b  12.95 (0.77) a  4.27 (0.32) b  3.81 (0.35) b  80 (94) b  11.97 (1.24) a  4.16 (0.42) b  3.66 (0.39) b  314 (149) a  16.17 (6.06) a  3.59 (0.13) b  3.12 (0.19) b  0-YR  275 (85) a  24.17 (12.46) a  4.36 (0.61) a  3.76 (0.65) a  5-YR  249 (87) a  17. 04 (6.30) a  3. 96 (0.32) a  3.32 (0.31) ab  10-YR  188 (116) a  17.62 (10.29) a  3.90 (0.36) ab  3.41 (0.23) ab  OG  249 (152) a  8.70 (4.69) a  3 .75 (0.23) a  3.23 (0.17) a  0-YR  276 (93) a  9.25 (5.03) a  4.23 (0.49) a  3 .67 (0.41) a  5-YR  183 (99) a  10.90 (4.06) a  4.09 (0.24) a  3.36 (0.30) a  207 (211) a  8.64 (5.05) a  3.79 (0.24) a  3.26 (0.16) a  OG  84 (27) a  10.44 (10.19) a  4.45 (0.32) a  3.96 (0.29) a  0-YR  86 (35) a  15.56 (7.09) a  4.81 (0.30) ab  4.31 (0.23) a  5-YR  120 (99) a  16.35 (8.78) a  4.93 (0.20) ab  4.28 (0.35) a  10-YR  53 (10) a  9.39 (3.96) a  4. 94 (0.21) b  4 .61 (0.21) a  10-YR  Bf  pH IN CaCl 2  3.90 (0.22) b  OG  Bhf  pH IN WATER  12.95 (1.18) a  10-YR  H  AIR DRY MOISTURE %  203 (122) a  OG  LF  FIELD MOISTURE %  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. AGE  HORIZON  LOCATION {AGE}  AGE* HORIZON  HORIZON* LOCATION  MULT. R2  {AGE} Air-Dry Moisture  0.015  0.000  0.349  0.152  0.645  0.513  Field Moisture  0.000  0.000  0.000  0.012*  0.007*  0 .781  pH in Water  0. 000  0.000  0.000  0 . 002*  0.977  0 .769  pH in CaCl 2  0. 000  0.000  0 .000  0.000*  0.959  0.805  C  0.034  0.000  0.218  0.192  0.362  0.905  N  0. 023  0.000  0.000  0.696  0.145  0.879  C/N  0. 075  0.010  0.073  0.511  0.893  0.376  LOI  0.000  0.000  0.003  0.055  0.119  0.924  Extr. Ca  0.010  0.000  0.001  0.097  0.324  0.827  Extr. Mg  0.000  0.000  0.000  0.007*  0.001*  0.884  Extr. Al  0.000  0.000  0.338  0.652  0.987  0.482  Extr. Fe  0.238  0.000  0.864  0 .400  0 .411  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. year sites, the pH the other ages. year  postburn  of the LF was significantly higher than that of  The pH was also higher in the LF of the 5- and 10sites  than  in the old  difference was not significant. H  horizon,  At the 0-  with  increased  growth  samples, but  the  This same pattern was seen in the  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. the pattern changed.  In the Bf,  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  AGE OG  LF  0-YR 5-YR 10-YR OG  H  0-YR 5-YR 10-YR OG 0-YR  Bhf 5-YR 10-YR OG  Bf  0-YR 5-YR 10-YR  C %  TOTAL N %  48.05 (4.31 ) a  0.92 (0.11 ) a  C/N  LOI  52.56 (6.20 ) a  2001 (612 ) a  46.34 (16.28  a  694 (588  a  a  46.38 (7.69  a  1381 (567  a  0.82 (0.20  a  48.46 (5.49  a  774 (597  a  a  0.91 (0.16  a  53.40 (11.59  a  2014 (780  a  41.55 (9.70  a  0.93 (0.14  a  45.95 (15.11  a  1512 (1845  a  40.98 (13.90  a  0.90 (0.22  a  47.16 (19.44  a  1549 (851  a  41.46 (4.69  a  0.84 (0.21  a  51.00 (10.05  a  823 (540  a  14.32 (4.74  a  0.54 (0.20  a  29.78 (9.55  a  88 (65  a  15.07 (7.01  a  0.51 (0.35  a  34.17 (9.53  a  88 (79  a  19.41 (13.43] a  0.57 (0.22  a  34.87 (21.40  a  114 (79  a  16.67 (6.31] a  0 .41 (0.17  a  53.29 (50.78  a  99 (75  a  6.74 (1.21] a  0.17 (0.04  a  40.99 (8.25  a  22 (5  a  6.19 (2.71  a  0.14 (0.07  a  44.19 (10.47  a  21 (9  a  5.14 (2.56] a  0.13 (0.08  a  40.62 (10.30  a  18 (9  a  4.08 (2.15] a  0.08 (0.04  a  53.69 (22.37  a  13 (3  a  35.50 (11.48  a  0.79 (0.17 ) a  43.85 (5.49  a  0.96 (0.15  39.05 (8.30  a  47.32 (6.20  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 :Eor a horizon at p<0.05. OG is old growth. HOR.  AGE  H  Bhf  Bf  EXTR Mg (mg/kg)  EXTR Al (mg/kg)  EXTR Fe  (mg/kg)  3354.3 (616.5) a  372.7 (99.9) a  274.4 (209.2) a  112.4 (29.6) a  0-YR  4366.4 (2145.5) a  662.4 (64.5) b  621.0 (318.7) a  200.4 (64.9) a  5-YR  4631.8 (1342.8) a  586.7 (70.2) b  455.9 (274.8) a  159.3 (61.4) a  10-YR  3331.7 (1746.1) a  574.4 (101.1) b  690.9 (439.8) a  239.7 (106.1) a  OG  2720.9 (1027.4) a  441.3 (150.5) a  355.8 (318.9) a  127.3 (88.0) a  0-YR  2814.7 (1448.3) a  652 .8 (85.9) b  742 .1 (265.4) a  189.7 (91.0) a  5-YR  2550.0 (934.6) a  552 .8 (12 9.2) ab  696.4 (473.4) a  275.1 (381.7) a  OG LF  EXTR Ca (mg/kg)  (n=9)  10-YR  2189.5 (1236.9) a  599.4 (137.2) b  813 .2 (358.4) a  267.6 (106.1) a  OG  1272.8 (1325.5) a  194.6 (169.8) a  1043.6 (406.4) a  347.7 (166.8) a  0-YR  1303.6 (1243.8) a  301.2 (188.5) a  1198.6 (306.7) a  421.3 (81.3) a  5-YR  847.7 (1253.4) a  266.7 (171.3) a  1162.8 (455.7) a  366.0 (139.1) a  10-YR  348.2 (377.9) a  242.6 (308.0) a  1552.4 (328.8) a  387.5 (135.6) a  OG  222.3 (286.1) a  40.5 (39.9) a  1917.7 (222.3) a  156.3 (107.2) a  0-YR  159.1 (87.0) a  50.2 (28.8) a  2102.5 (219.4) a  150 .0 (60.7) a  5-YR  88.5 (132.0) a  26.4 (23.3) a  2236.6 (229.8) a  144 .2 (65.4) a  57.4 (64.5) a  16.0 (5.3) a  2130.8 (111.2) a  101.4 (15.0) a  10-YR  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  variability was high.  horizons  (Tables 7-2, 7-4),  and  the  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. effect  (Table  However, there was also a significant location  7-6).  There  pyrophosphate-extracted  were  Mn  in  significant  the  LF,  differences  with  the  in  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 p<0.05 HOR.  LF  H  Bhf  Bf  OG is old growth. (n=9)  AGE  Fe-PYRO %  Al-PYRO %  OG  0.010  0.031  a  0.051  0.040  a  0.063  0.036)a  0-YR  0.121  0.155  a  0.128  0.086  a  0.170  0.178)b  5-YR  0.013  0.029  a  0.071  0.038  a  0.126 (C).113)ab  10-YR  0.111  0.113  a  0.176  0.128  a  0.035  0.034)a  OG  0.069  0.184  a  0.081  0.157  a  0.009  [0.012)a  0-YR  0.121  0.162  a  0.127  0.077  a  0.044  [0.056)a  5-YR  0.123  0.189  a  0.135  0.147  a  0.013  [0.010)a  10-YR  0.167  0 .202  a  0.169  0.124  a  0.014  0.012)a  OG  1.395  0.934 ab  0.577  0.234  a  0.000  0.000)a  0-YR  1.527  0.642  a  0.470  0.253  a  0. 011  0.023)a  5-YR  1.174  0.633 ab  0.374  0.166  a  0.001  0.002)a  10-YR  0 .772  0.511  b  0.410  0.162  a  0.002  0.004)a  OG  0.910  0.401  a  1.188  0.364  a  0.000  0.000)a  0-YR  0.887  0.528  a  1.119  0.481  a  0.004  0.009)a  5-YR  0.886  0.698  a  1.030  0.533  a  0.001  0.002)a  10-YR  0.294  0.118  b  0.637  0.170  b  0.000  O.OODa  130  Mn-PYRO %  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 10year postburn sites.  In the Bf, the 5-year sites had the most Fe-  CBD, while the 10-year  sites  contained the least.  concentrations were similar among the sites in  The Al-CBD  the Bhf horizon,  Table 7-6: Analysis of variance (ANOVA) table for pyrophosphateextracted 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. AGE  HORIZON  LOCATION {AGE}  AGE* HORIZON  HORIZON* LOCATION  MULT. R2  {AGE} Fe AAO  0.197  0.000  0.038  0.489  0.357  0.612  Fe CBD  0.103  0.000  0.008  0.370  0.689  0.595  Fe Pyro  0.006  0.000  0.015  0.006*  Al AAO  0.763  0.000  0.005  0.659  0.086  0.878  Al CBD  0 .141  0.000  0.001  0.268  0.208  0.863  Al Pyro  0 . 071 0. 000  0.003  0.000*  0 .023*  0.841  Mn CBD  0.157  0.627  0.190  0.879  0.936  0.311  Mn Pyro  0. 003  0. 000  0.706  0. 013*  8.582  0.616  131  0.14 0  0.800  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. all ages.  In the Bf, the concentrations were very similar for 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 citratebicarbonate-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  AGE  Fe-CBD %  Al-CBD %  Mn-CBD %  Fe-AAO %  Al-AAO %  OG  2.013 (1.296)a  0.477 (0.175)a  0.001 (0.001)a  1.283 (0.691)a  0 .519 (0.182)a  0-YR  2.304 (1.124)a  0.469 (0.217)a  0.014 (0.027)a  1.289 (0.450)a  0.446 (0.164)a  5-YR  2.126 (1.127)a  0.488 (0.204)a  0.005 (0 .006)a  1.285 (0.624)a  0.483 (0.235)a  10-YR  1.362 (0.662)a  0.451 (0.195)a  0.003 (0.004)a  0.791 (0.536)a  0.439 (0.199)a  OG  2.636 (1.111)a  1.465 (0.385)a  0.001 (O.OOl)a  1.780 (0.479)a  2 .390 (1.097)a  0-YR  3.456 (1.195)a  1.535 (0.324)a  0.016 (0.034)a  1.914 (0.560)a  2 .312 (0.702)a  5-YR  3.825 (1.492)a  1.593 (0.583)a  0.004 (0.004)a  2.033 (0.571)a  2.326 (0.777)a  10-YR  3 .197 (0.723)a  1.231 (0.186)a  0.011 (0.023)a  1.813 (0.296)a  2.663 (0.427)a  Bhf  Bf  132  % Fe  Bhf-OGBhf-0 B h f - 5 B h f - 1 0 B f - 0 G  Bf-0  Bf-5  Bf-10  Horizon A)  •  Organic I/2 Amorphous Gi3 Crystalline  % Al  Bhf-OGBhf-0 Bhf-5Bhf-10Bf-OG B  )  Bf-0  Bf-5 Bf-10  Horizon  Figure 7 - 1 : The mean p e r c e n t a g e s of o r g a n i c , a m o r p h o u s a n d c r y s t a l l i n e i r o n (A) a n d a l u m i n i u m (B) i n t h e Bhf a n d Bf h o r i z o n s of CH o l d g r o w t h (OG) and 0, 5 a n d 10 y e a r s p o s t b u r n sites. 133  less organic Fe, and more amorphous Fe. contain the least crystalline Fe. not significant.  These differences, however, were  Aluminium was almost entirely in organic form in  the Bhf horizon of all sites (Figure 7-1B). postburn  The old growth sites  samples  contained  significantly  In the Bf, the 10-year 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  significantly  Tables  higher  on  7-8 the  and  7-9 .  recently  Available  burned  sites  P  (Pa) was  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  horizons.  However, in the Bf, P0 was significantly lower on the  10-year sites. horizon  among the sites in P0 in the LF, H and Bhf  among  There were no significant differences for any the  sites  for  PT  or  the  C/P  ratio,  and  the  variability was high. The Chang and Jackson fractionation procedure produced PHCi/ P  NaoH a n d  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 1980;  Olsen  and  Sommers  1982).  There  were  no  al.  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  AGE  Avail• P (mg/kg  OG  30.46 (10.01  b  585.8 (66.6) a  88.97 (31.25  a  27.16 (9.38  713.9 (181.4) a  622.3 (272.4) a  539 (242) a  b  675.8 (141.9) a  551.2 (177.6) a  673 (162) a  17.25 (7.15  b  485.3 (163.1) a  391.1 (124.2) a  878 (295) a  20.62 (9.65  b  482.8 (152.1) a  308.7 (123.2) a  1055 (389) a  44 .11 (24.89  a  524 .0 (219.9) a  360.4 (243 .9) a  958 (536) a  39.05 (21.37  a  569.5 (150.6) a  401. 0 (127.5) a  826 (368) a  b  515.4 (150.9) a  306.2 (112.2) a  866 (258) a  OG  6.06 (2.92) a  352.1 (89.6) a  145.1 (89.2) a  392 (152) a  0-YR  8.84 (8.12) a  349.4 (199.5) a  174.2 (97.7) a  450 (111) a  5-YR  9.24 (6.54) a  468.6 (170.6) a  220.2 (111.0) a  343 (112) a  10-YR  8.59 (1.78) a  354 .5 (66.1) a  198.7 (93.1) a  468 (151) a  OG  6.31 (1.02) a  289.2 (115.6) a  64 .8 (36.4) a  252 (90) a  0-YR  6.43 (1.12) a  230.8 (41.1) a  41.3 (23.0) a  268 (104) a  5-YR  6.09 (1.05) a  235.5 (102.2) a  63 .5 (36.8) a  221 (65) a  10-YR  6.60 (0.51) a  211. 8 (70.2) a  21.6 (24.8) b  193 (71) a  0-YR  OG 0-YR 5-YR 10-YR  Bf  12.62 (5.08  135  492.0 (98.8  C/P  815 (135) a  10-YR  Bhf  Org. P (mg/kg) a  5-YR  H  Total P (mg/kg)  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. AGE  HORIZON  LOCATION  AGE* HORIZON  HORIZON*  {AGE}  LOCATION {AGE}  MULT. R2  Avail P  0.000  0.000  0.128  0.000*  0.572  0.863  Org P  0.002  0.000  0.318  0. 0 04*  0.324  0.785  Tot P  0.011  0. 000  0.000  0.093  0.984  0.724  P-HCl  0.481  0.000  0.175  0.496  0.175  0.601  P-NaOH  0. 011  0.000  0.135  0.027*  0.000*  0.685  P-CBD  0.049  0.000  0.008  0.228  0.032  0.732  C/P  0.095  0.000  0.035  0.065  0.922  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. calculated  As these means were  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.  AGE  P-HC1  P-NaOH (mg/kg)  P-CBD  (mg/kg)  OG  7  (0.5) a  37 (28.0) b  22  (17.0) a  0 YR  8  (1.3) a  58 (34.8) a  34  (20.5) a  5 YR  7  (1.0) a  42 (13.6) ab  26  (19.3) a  10 YR  7  (0.7) a  27 (16.9) b  15 (13.9) a  OG  19  (13.6) a  25 (12.4) a  74 (23.6) a  0 YR  24  (12.1) a  23  59  5 YR  22  (24.0) a  17 (10.8) a  68 (35.8) a  10 YR  30  (14.5) a  18  50 (14.7) a  Bhf  Bf  postburn sites. Bf.  (mg/kg)  (12.0) a  (6.8) a  (24.8) a  There were no significant differences in Pj. in the  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  negatively with extractable Mg and total N. relationship  between  Fe  extracted  by  CBD  PCBD and  There was a positive and  pyrophosphate-extracted Mn to Mn-CBD and organic P.  137  with  AAO,  and  with  Extractable Al  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  ORG. P  INORG. P  % RECOVERY  OG  83.3  112.7 ) a  N/E  83.3  0 YR  85.6  16.9 ) a  N/E  85.6  16.9  a  5 YR  80.0  10. 8  a  N/E  80.0  10.8  a  10 YR  83 .5  17. 9  a  N/E  83 .5  17. 9  a  OG  65.5  28.3  a  N/E .  65.5  28.3  a  0 YR  65.7  16.7  a  N/E  65.7  16.7  a  5 YR  73 .4  28.6  a  N/E  73 .4  28.6  a  10 YR  62.7  25.1  a  N/E  62.7  25.1  a  OG  39.1  17.9  a  19  (10.7) ab  58 .0  22.1  a  0 YR  50.4  12.9  a  33  (14.6) a  83 .2  18.7  a  5 YR  49.1  20.1  a  17  (6.1) b  65.9  21.5  a  10 YR  57.4  24.8  a  13  (7.5) b  70.8  25.7  a  OG  23.9  12.0  a  44  (15.6) a  68.1 ( 17.3  a  0 YR  17.8  10.2] a  46  (9.7) a  64.2 ( 18.7] a  5 YR  28.1  12 .5] a  48  (14.8) a  65.9 ( 21.5] a  10 YR  10.2  10.9] a  50  (15.6) a  60.3 ( 14 .5] a  112.7 ) 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. two  pH  methods.  There was a positive relationship between the Positive  correlations  138  were  also  seen  for:  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 •* coroN er> \o cr> \o-* •>«•  1  -iddddddddddddddd  CP L "x UJ Jj  o N in •«• co en -< (\l o N -H vo N m m N i*. S - * ro *C4 -Hinrgm N N ~I ro •* rvico \i *idddddddd dd d d ddd d TTT till ON-H*criC3ni\0iJ)!QiP'-i'-ir>JCn(rif^PJ d  ^  d  - ?? v*  "x  dddddddddddci  x  -d g o a. z  -?? SSRfc  1  • »• » -«ooo  -cpdocp  ^ 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0  a  ddodddoddddodddddoo  o.  ^ddddd  •H d d c p d d o d d o o c p o o d d odcp d o  S  'idddddo  Oin •-«*mo  "5  rt  u  t8B2?S8  §g§8nl§iS!Slg§g§S^S§Sg  §S81HS2 • • • • » • » •  —ddddddddddddddddddddd o  S  o L  wdddddddddddddddddddddd  i  !8^§s§ias'  O Ni  -Id dd d d d d d d  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  L  9i  3 •«• K « in s i n o s o •* co in — •-< in a co In <\i ro in ro\o \c •*  T  -dddddddddddddddddddddddd  §  -. d d o o d d d o  §!M^8^i^H8^o^s §  A)  tr  -•ooaoooa  Q-* i^ — (sir^N-'iO\D m N to i^ m en o (T> •• w o — m B N en vo \o N « Nfo N as a o -*m tncKcvi p -* i^coro  S f J p « •»• en to 3 N P B N ve ro N t 3 &i N coro« * - * i^ in  o ro (M •» in + m in in —-H « \o -H •*•>«• -* -• i^ co t en in CM CO ro  5 in — •* u3 cr>55 in S vo ro •* i^ ro sfi en o *co co 01 O N K - \O O o co ro in -< f N •<*• \o Co \o \i o m to •* a-Hin in in 3 •<*• . i n m i • _ « o oI oI oT p Io o p o o o aT oI oT pTp Io Io o o o o i o o •to cc T I i g N -H ro o \o so en co vo ro in ro (\i -• in •* \o ** en in ro s. ro ro r>. ro  fc a _i  •Idddddddddd it S Q X  §88?sB5l8i«iB  §  L Hodcidddddddd  o 55 N N ID CM m s f N Inin * ro r\i •* in a -• J K 3 s in in \D ro ^^dddp'dddciooddp'dacicidddddoDOD o o o a u x a - H m o/ oio a o i. u. " a o a w p -< I • z: a TTCLI» ^J^SiJJTtcIpx T a o i i i i zTi-i-  cj  cc cc — 4  cc  uliDiiJuJLi.ii.ix  u.  £ fc x  r  a. -  139  a  -•dodddddopaoo  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. correlated positively with C/P, total N and Ca.  Carbon  Extractable Ca was  positively correlated with extractable Mg and total N. The  31  P 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 52) . 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. problem with peak separation is discussed in Chapters 5 and 6.  This 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  horizons.  In profile B (Fig. 7-2B; Table 7-13), most of the P was  found  orthophosphate  as  small phosphonate peaks  or  as 140  monoesters,  in the LF and H  at  comparable  CEDAR-HEMLOCK OLD GROWTH A  If HORIIOH  A) 1 PHOS  1  i*  1  t  I -S  >  ORTMOMONO  (  ™ ^  T  PYRO  1 — -10  1 -28 POLY  CEDAR-HEMLOCK OLD GROWTH B  If HORIZON  PHOS  ORTHO  MONO D I E 3 T  POLY  l P NMR spectra for two soil profiles from mature CH F i g u r e 7-2 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  HOR  Phos  Orth  o, 0  OG A  OG B  0-YR A  0-YR B  Dies  Pyro  o, o  o, o  o,  Poly %  LF  0  17  33  39  0  11  H  5  7  38  36  5  7  Bhf  4  38  36  17  5  0  Bf  0  23  61  16  0  0  LF  0  31  38  18  13  0  H  0  39  32  20  9  0  Bhf  0  35  34  24  7  0  Bf  0  29  32  24  15  0  LF  0  51  32  11  6  0  H  0  12  52  23  12  0  Bhf  0  25  43  24  8  0  Bf  0  85  15  0  0  0  LF  0  24  41  29  6  0  H  5  23  34  33  0  5  10  14  47  24  5  0  0  13  62  25  0  0  Bhf Bf  percentages.  Mono  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. diester  and  decreased.  In the H horizon, the percentages of monoester,  pyrophosphate  P  increased,  while  orthophosphate  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  horizon, and pyrophosphate peaks in the LF and Bhf.  peak  in the H  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  CEDAR-HEMLOCK 0 YEAR A  IF HORIZON  H HORIZON  Ay/,VVWv/  A)  to s ORTHO. MONO  « pj^j  -a pYHO  CEDAR-HEMLOCK OYEAR  B  IF HORIZON  H HORIZON  Bhf HORIZON  Bf HORIZON  B) POLY  31 P NMR spectra for two soil profiles from recently Figure 7-3: burned CH sites, extracted with NaOH-EDTA.  144  CEDAR-HEMLOCK FIVE Y E A R A  23  I«  ORTHO MONO  PHOS  pjgg.,.  PYBO  A)  CEDAR-HEMLOCK FIVE YEAR  B  IF HORIIOH  J^yviA^ H HORIZON  Bhf HORIZON  4^v^>^^ Bf HORIZON  jIV^Kvy B)  24  PHOS  ORTHO MONO £"!„  PYRO  POLY  L P NMR spectra for two soil profiles from CH sites 5 F i g u r e 7-4 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 5year 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  Phos %  Orth %  Mono %  Dies %  Pyro %  Poly %  LF  0  23  40  18  7  12  H  4  15  36  25  0  20  Bhf  0  12  55  29  4  0  Bf  0  30  60  10  0  0  LF  0  33  44  23  0  0  H  0  23  59  18  0  0  Bhf  0  23  51  26  0  0  Bf  0  18  37  45  0  0  LF  0  17  27  32  10  14  H  4  14  39  24  6  13  Bhf  0  31  51  14  4  0  LF  0  36  53  11  0  0  H  0  8  69  18  5  0  Bhf  0  12  36  42  10  0  Bf  0  65  25  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  in the LF and H  and monoester peaks overlapped  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 b e obtained ORTHO  MONO  DIEST  PYRO  POLY  CEDAR-HEMLOCK TEN YEAR B  B) 20  PHOS  IS  10  5  0  ORTHO MONO  «•« DIEST  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. profile  A.  There was a range of P forms in the LF horizon of 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. Bhf, most of the P occurred as monoesters.  In the  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. and  The signal-to-noise ratio was low for this horizon,  it was difficult  noise.  to distinguish peaks from the background  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, 0or  5-year  sites  (personal  duration, reflected material  observation).  Fire  intensity  in the quantity of slash and forest  consumed by the fire, are believed  and  floor  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.). different sitchensis  As well, the 10-year sites were replanted with  species from the 5-year sites. (Bong) Carr.) and Douglas-fir  Sitka spruce {Pseudotsuga  (Picea 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. selection was based to some extent on ease of sampling:  Site  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. insufficient  sampling:  the  high variability  This may be due to 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. 1992; Romanya et al.  1994).  1991; Brockley et  al.  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. was  discussed  matter.  As  in Chapter 6, LOI is a measure of soil organic  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 1955; Mangas et al. fire (Mroz et al.  1992).  al.  Nitrogen is readily volatilized during  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. et al.  1983; Khanna and Raison 1986; Macadam 1987; Tomkins  1991; Brockley et al.  1992; Rice 1993), with Ca increases  proportional to fire intensity (Grier 1975; Brockley et al. Rice 1993). 152  1992;  Magnesium has also been shown to increase after burning (Grier 1975; Ellis and Graley 1983; Feller et al. 1986; Macadam 1987; Tomkins et al. Mangas et al.1992;  Rice 1993).  1983; Khanna and Raison  1991; Brockley et al.  1992;  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  significantly  in  lower  the in  mineral the  soil.  10-year  Pyrophosphate-Fe  sites  in  the  Bhf  and  was Bf  horizons, while pyrophosphate-Al was significantly lower in the 10year  Bf.  There  bicarbonate ammonium  were  dithionite  oxalate  no  significant  (CBD)-extracted  (AAO)-extracted  differences  in  citrate  Fe and Al, nor in acid  Fe and Al.  As  discussed  in  Chapter 6, pyrophosphate extracts the organic-associated Fe and Al. In natural, unmodified podzols, Fe and Al  153  in the Bhf and Bf  horizons  are  predominantly  in  organic  form,  reflecting  the  characteristic illuviation of organic matter and organo-metallic complexes  in  crystalline  podzolic  soils  Fe but not Al  (Oades  1989).  The  presence  of  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. and  more  In the Bf, the 10-year locations had less organic Fe  amorphous  Fe,  and  significantly  more  amorphous Al.  Amorphous Al and Fe concentrations were determined by subtracting the  pyrophosphate  values.  extraction  results  from  the  extraction  The significantly lower pyrophosphate results may falsely  elevate the amorphous Al and Fe concentrations. lack  AAO  of  differences  in AAO-extracted  In light of the  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. burning  of  the  10-year  sites  appears  to  have  The intense 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  processes unaffected. temporary,  although  organic  matter,  leaving  the  illuviation  These changes in the Bf are probably only 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  literature  of  However,  soil  profile.  changes  Humphreys  in  There  were  no  reports  pyrophosphate-extracted  and Lambert  (1965)  observed  Fe  in  the  and  Al.  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  organic matter.  from  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,  consistent effects of fire on soils  and  is  one  of  the  most  (Ahlgren and Ahlgren 1960;  Humphreys and Lambert 1965; Kozlowski and Ahlgren 1974; Feller 1982; Feller et al. Tomkins et al. et al.  1983; Macadam 1987; DeBano and Klopatek 1988;  1991; Brockley et al.  1993; Romanya et al.  1992; Mangas et al.  1992; Saa  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 1991; Romanya et al.  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.  (1957), Ellis and Graley (1983), Tomkins et al. et al.  Beaton  (1991) and Romanya  (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 0and 5-year sites in the LF and H horizons. decreased P0 after fire Romanya et al.  Previous studies report  (Kwari and Batey 1991; Saa et al.  1993;  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#  subsequent colorimetric test measures the Pj. in solution.  and  a  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# would  cause  an overestimation  of 156  P0.  When  the  Saunders  this and  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  31  P 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. unchanged  The lower horizons appear to be  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. present for one profile.  Polyphosphate peaks are also  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  origin may be biological.  and soil microbes, and thus their  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 forest soils following burning. cutting and burning of Pinus  Zech et  al.  31  P NMR spectra of (1987) state that  mugo and establishment of pasture did 158  not significantly influence the patterns of  31  P 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  occluded within Fe and Al oxides and hydroxides.  measured  P  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. also a significant  There was  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  an  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 oxalateextracted different  Fe  and  among  Al. the  Because ages,  it  these is  were  not  unlikely  significantly  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. 10-year  sites, one  locations were not.  location  was  scarified  and  However, for the the  other  two  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 horizons.  of  available  P,  Ca, Mg  and Mn  in the  surface  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 profile, horizons.  to  disrupt  and  may  illuviation produce  processes  longterm  throughout  changes  in  lower  Although total P levels were not changed,  the  soil  mineral  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  Studies by Weetman et  (CH) sites.  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 Pot  seedling  bioassays  confirmed  insensitive to nutrient availability  that  cedar  was  species. relatively  (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  phosphatases,  1984). which  This are  mineralization  grouped  into  is  catalysed  phosphoric  by  monoester  hydrolases (EC 3.1.3), phosphoric diester hydrolases (EC 3.1.4) or acid  anhydride  Tabatabai  1982).  hydrolases  (EC  3.6.1)  (Speir  and  The monoester hydrolases include: 162  Ross 1978; acid and  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.  al.  1994); in ectomycorrhizae (Bartlett and Lewis 1973; Ho and Zak  1993; Fernandez and Ascencio 1994; Pant et  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. et al.  1991; Pasqualini  1992; Tarn and Griffiths 1993; McElhinney and Mitchell 1993);  in ericoid mycorrhizae (Straker and Mitchell 1986; Shaw and Read 1989;  Read  1991;  Dighton  and 163  Coleman  1992)  and  in  orchid  mycorrhizae  (Antibus  and  Lesica  1990).  Although  production has been demonstrated in VA mycorrhizae  phosphatase (Gianinazzi-  Pearson and Gianinazzi 1976; MacDonald and Lewis 1978; Gianinazzi et al. al.  1979,-Kapoor et al.  1988; Dodd et al.  1987; Jayachandran et  1992; Thiagarajan and Ahmad 1994; Khalil et al.  1994; Tarafdar  and Marschner 1994), it is believed that VA mycorrhizal and nonmycorrhizal 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. sand-peat  In June, 1992, the seedlings were transplanted into a  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  mycorrhizal'  the  mycorrhizal  treatment.  fungi,  Benomyl  to  produce  a  'non-  [(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 (350  mg/L),  MnS04.4H20 mg/L),  (2.25 mg/L),  H3BO3  treatments. 1.  Ca (N03) 2. 5H20  (3.0 mg/L),  (900  mg/L),  CuS04. 5H20 NaCl  (400 mg/L), K2S04  MgS04. 7H20  (0.25 mg/L),  (500  mg/L),  ZnS04. 7H20  (0.30  (5.0 mg/L) and one of eleven P  These treatments were:  No P  2. p h y t i c a c i d high ( i n o s i t o l hexaphosphoric a c i d , d o d e c a s o d i u m 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 used  (1982) .  per assay  averaged.  and  Three replicates and one control sample were the results  for each plant  and  soil were  For all plants in the experiment, a portion of randomly  sampled roots was frozen for later analysis and the soil was airdried.  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  (1980).  the  gridline-intersect  The  method  of  Giovanetti  and  Mosse  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). year  One  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  orthophosphate treatment.  from  the  no  P  control  was  the  high  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  Diameter <Change (mm  Dry Weight (g)  No P  3.0  [1.05  c  59.5  (18.6) c  Ortho L  4.6  [1.52  abc  68.6  (10.0) be  Ortho H  5.4  1.26  ab  98.3  (14.1) a  ATP L  4 .4  1.26  abc  65.6  (18.0) be  ATP H  5.6  1.51  a  91.5  (7.88) a  Phyt L  3 .2  0.79  be  50.2  (10.3) c  Phyt H  3.4  1. 71  be  56.7  (15.8) c  Glyc L  3.6  1.78  be  60.7  (21.0) c  Glyc H  4.9  1.29  abc  86.6  (15.0) ab  Pyro L  3.6  0.97  abc  65.1  (6.41) be  Pyro H  4.2  1.48  abc  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 82) .  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. Treat  Myc  Ht Start  0.062  0. 065  Ht Change  0.503  Diam Start  Treat*Myc  Mult R2  N  0 . 749  0.246  110  0.031*  0.176  0.245  110  0.086  0.287  0.618  0.232  110  Diam Change  0.000*  0.027*  0.201  0.377  110  Dry Weight  0.000*  0.010*  0.315  0.608  110  Foliar P  0.000  0.000  0.046*  0.807  110  Foliar Ca  0.000*  0.227  0 .372  0 .415  110  Foliar Mg  0.132  0.340  0.759  0.208  110  Foliar Cu  0.002*  0.473  0.103  0.352  110  Foliar Fe  0.280  0.120  0.466  0.219  110  Foliar Al  0.725  0.228  0.448  0.173  110  Foliar Zn  0.000*  0. 049*  0.718  0.357  110  Foliar N  0.000  0.000  0.007*  0.625  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) Dry Weight (g)  Diameter Change (mm)  Foliar Zn (ug/g)  Mycorrhizal Treatment  Height Change (cm)  Mycorrhizal  29.7 (12.10) b  67. 7 (22.03) b  3 .93 (1.78) b  19.31 (6.56) b  Benomyl-Treated  34 .6 (11.67) a  75.0 (19.01) a  4.42 (1.24) a  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) Foliar (2a. (ug/g)  Treatment  No P Ortho L Ortho H ATP L ATP H Phyt L Phyt H  Foliar Cu (ug/g)  Foliar Zn (ug/g)  9774.8 (871.9  b  6.7 (1.3) be  17.2 (3.3) b  10950.4 (5650.3  b  8.2 (4.4) abc  18.5 (6.7) ab  10768.8 (1898.4  b  8.9 (2.7) abc  18.2 (4.3) ab  9533.9 (1625.4  b  6.7 (1.4) be  17.6 (2.8) b  9207.0 (1502.9  b  8 .1 (1.1) abc  21.5 (5.3) ab  8988.6 (1348.9  c  6.5 (1.2) c  17.6 (2.1) b  9323.6 (1665.8  b  7.7 (1.6) abc  17.9 (4.9) b  Glyc L  14585.3 (4484 . 8 a  9.8 (3.3) a  Glyc H  11778.5 (1678.7  ab  9.3 (2.1) ab  20.5 (5.9) ab  11694.9 (2214.8  ab  8.1 (1.3) abc  22 .6 (2.8) ab  12413.5 (3438.3  ab  9.1 (3.1) ab  26.1 (6.5) a  Pyro L Pyro H  concentration  was  118.0  ug/g.  26 .2 (10.3) a  Significant  treatment  occurred for foliar Ca, Cu and Zn (Tables 8-2, 8-4). foliar  Ca  concentrations  were 171  seen  with  both  effects  The highest levels  of  glycerophosphate and pyrophosphate.  The plants fed the low level  of phytic acid had the lowest foliar Ca concentrations. was  highest  when  glycerophosphate,  Foliar Cu  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 (Tables 8-2, 8-3).  effect  for foliar Zn  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 plants had higher  (Tables 8-2, 8-5) .  foliar N concentrations  except the high rate of ATP.  The benomyl-treated for every  treatment  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. orthophosphate:  The exception was the low rate of  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  orthophosphate, concentration  of  the  benomyl-treated  glycerophosphate available  soil  trees  and P at  fed  high  rates  pyrophosphate. the  end  of  of The  the  feeding  experiment was highest when the plants were fed the high rate of glycerophosphate, pyrophosphate or orthophosphate. was  lowest  Available P  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) Treatment  Foliar N M  j j  Foliar N NM  Foliar P M  j |  Foliar P NM  No P  2.37 (0.37) be  I 2.43 j (0.33) b  Orth L  1.72 (0.37) ij  i 2.59 (0.25) a  991.9 ! 556.1 (385.8) cdef (121.3) g  Orth H  1.67 (0.20) j  I 1.87 ! (0.29) hi  1554.5 (288.6) abc  i 1346.2 j (264.1)abcd  ATP L  2.12 (0.14)def  i 2.61 j (0.31) a  694.1 (132.4) efg  | 727.7 j (109.7) efg  ATP H  2.11 i 2.21 (O.lO)defg (0.13) de  Phyt L  2.06 (0.19) fg  i 2.59 (0.15) a  Phyt H  2.06 (0.08) fg  i 2.40 | (0.18) b  Glyc L  2.11 ! 2.4 0 (0.14)defg j (0.19) b  1167.1 ! 743.7 (410.7) bede \ (142.9) efg  Glyc H  1.90 (0.19) h  ! 2.08 j (0.21) efg  2034.1 (422.2) a  i 1555.7 (309.2) abc  Pyro L  2.01 (0.34) fg  i 2 . 67 j (0.28) a  874.6 (90.0) defg  i 680.6 j (166.1) efg  Pyro H  1.96 (0.16) gh  ! 2.25 j (0.37) cd  1615.1 (360.4) abc  i 1743.4 j (497.7) ab  581.8 (73.6) fg  i 534.6 ! (86.0) g  1401.0 ! 1139.4 (326.4) abed (367.1) bede 581.5 (86.5) fg  i 606.9 j (50.6) fg  1138.4 i 861.9 (180.2) bede ; (158.1) defg  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  Col M  Col NM  Soil P M  Soil P NM  Rt Ac Pase M  Rt Ac PaseNM  1  No P  38.6 (10.9) a  18 .4 (7.5) abede  16.2 (2.0) cd  13.3 (1.5) cde  538 (122) cde  520 (67) e  Orth L  24.8 (7.9) abed  10.6 (4.7) defgh  18.6 (5.1) be  11.6 (0.9) cde  1192 (160) a  380 (61) e  Orth H  17.0 (7.0) abedef  6.8 (2.2) fgh  31.0 (5.5) a  30.1 (7.8) a  592 (116) cde  540 (156) e  ATP L  26.8 (10.2) abc  28.0 (7.1) abc  11.2 (1.8) de  12.0 (1.9) cde  586 (374) cde  628 (353) bede  ATP H  8.4 (3.6) efgh  15.9 (6.1) bedef  26.8 (6.3) ab  26.2 (4.7) ab  933 (35) abc  427 (239) e  Phyt L  15.4 (5.0) bedef  17.0 (5.5) abedef  9.7 (3.0) e  10.3 (0.8) de  492 (188) e  412 (26) e  Phyt H  28 .2 (14.2) abc  14.2 ] (5.3) .! cdefg  14.2 (2.3) cde  10.1 (2.6) e  722 (349) bede  701 (73) bede  Glyc L  35.8 (13.1) ab  ! 17.0 i (3.9) abedef  15.7 (2.1) cd  13 . 7 (2.2) cde  376 (17) e  954 (135) ab  20.0  !  (7.9)  J  5.8 (2.2) ! gh  28.3 (3.5) ab  30.9 (8.8) a  538 (111) e  541 (110) e  Glyc H  abede Pyro L  27.2 (13.0) abc  14.6 (5.6) cde  13.7 (1.6) cde  12.1 (1.3) cde  535 (192) e  581 (173) cde  Pyro H  33 .4 (13.4) abc  5.2 (1.3) h  29.2 (3.6) a  26.6 (2.8) ab  575 (416) de  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. Treat  Myc  Treat*Myc  Mult R2  N  Colonization  0.000  0 .000  0.000*  0.720  110  Soil P  0.000  0.005  0.036*  0.851  110  Root AlkPase  0.000*  0.803  0.425  0.735  44  Soil AlkPase  0.078  0.319  0.998  0.498  Root AcidPase  0.664  0.491  0 . 034*  0.589  44  Soil AcidPase  0.060  0.000*  0.301  0.801  44  Root DiesPase  0.659  0.662  0.548  0.424  44  Soil DiesPase  0. 963  0.387  0.933  0.226  44  Root PyroPase  0.063  0.893  0.578  0.568  44  Soil PyroPase  0.602  0.477  0.268  0.491  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  Soil Acid Phosphatase  Mycorrhizal Benomyl-Treated  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  glycerophosphate, and the low rate of pyrophosphate.  rates  of  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 pyrophosphatase  ug p-nitrophenol/g/h.  activity was 3.7  ug P/g/h,  pyrophosphatase activity was 29.7 ug P/g/h.  The mean soil  and  the mean root  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  positively  of  soil  correlated,  and  and  root  there  alkaline  was  phosphatase  a positive  relationship  between soil alkaline phosphatase and soil pyrophosphatase. diameter and height at the start of the feeding  were  experiment  The 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. negatively correlated with soil P and foliar P.  177  Foliar N was  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  Root Alkaline Phosphatase  No P  82.2  (39.3)  b  Ortho L  97.7  (2.6)  b  Ortho H  141.8  (21.2)  ab  ATP L  120.6  (17.0)  b  ATP H  197.1  (69.7)  ab  Phyt L  159.1  (64.6)  ab  Phyt H  217.6  (45.1)  a  Glyc L  273.8  (57.8)  a  Glyc H  257.5  (122.9)  a  Pyro L  259.5  (20.5)  a  Pyro H  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  Content.  ,,  g^ocnincninfviNin-N  u  § CTi CO CO CO Q — vO CO T-CTi CO  <*•  -'Ooooo'ddo'ddo T  i  i i i  O ~ H 0 ^ O O O ~ « ~ * O O O O  — o o< o o oI o o oIoId1dId I o~HC^(M(T)(T)cr>QrJtotr}r^crv — g — j2cofnocpv2y5mr\j^(\i —  I  O \D O C O O  — oI o o o o oI o o o d d dI d O ! D f V i m - T O C 0 ( \ ; N - « D \ i L 0 OjD'>tJ(T)Cr.N<DNiniOOI^in —  O N -*O — CB  o mo  0 0 « 0 - H O O O O O O O ( \ I - H O  - o o< oI Io Io o o d dIdI d d d d g"^«oryococriCD-H-H'<fini^-.o  O \D \ f l ^c Q C 1 N N O O f\l o  - o o oI o o o io Io o o ' d d d d d7 I I i I i  - o d d i  <T) l0 1, P,£"}S u , 0 ffi!S.£:C32 , r< t >fcn-«-~o-r  O O - W H O O O O O O O O W N Q O  O N I M C O f a vD — •*• <\l O — o -H —  -•oooooooodddddddd  —dd do  Sfc ^ 2iQffiC' ' fm-«(T)3cD  SvS(Bs!irt'Hl':'ISoa3vDocDQo-<cn oi2fSSSSS0SSlnSrj«BN!aN  X  o €  — \X> CD \n §— en — N en o  o oooo—  O H N O N O O n O O - O l H t O H O r a  -< d d d d d  —ooodoo'o'dddcidddddd  §n a o n o o  f\l C "«- CO C M ^ C* <\! N N ~H CO  om — - « m o & o o o O « C N o - r y K i - o o o o i ol ol ' d d di d di di di di di d d di  —odddd d  1  or\jocnco(\i'>tir)  ooojo-<->-.oocno--Io-5ooooV-.  C - V O  — — S — -V  — o o d d d d d OLO(T)'^-(\J5D'^-COO  O - C W O O ( \ J O - — mrooSocMiySoKikJB  —c o o o o o d d CdctDipaotm-  ",? ? ? ? ° 0 ' ° ° ?I ?i °i ° io i° o '( oi di d d d d ' ' ' ' l i t  - o e d d d d d d d  *§ „ .~^™°^*^~<:w.°SF522o,j;gagg2  -8CJD2.§SSOS!2R!  o « o o o o o o d d d d d d d d d d d d d d d d  •'—cooooooooo  ^ I S silts " l * 8 3 * ^ ^ " " - ^ o o«« * <5  O --p£Xx LL L£ o 0 .2.2 * * i ,?.9.P .P .9 «••«- .9 a a  _ , S ^ ! ^ T3 <*•«*-<*- U . « . X TJ T3 T3  179  3 * en c a C cn-n 41 -H a U<-6--NillL.«*X';'^09"HlJ-tlob — 0  0 0 < ~ l t O - U X L L O  a a  Discussion Benomyl, a systemic fungicide, was used to produce a 'nonmycorrhizal' 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 198 6; Sukarno  et al.  1993) .  (Fitter  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  retrospect  should  enrichment  of  side-effect have  been  of  benomyl  anticipated,  non-mycorrhizal  plants.  treatment, was  the  Benomyl  which  in  apparent  N  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  Hamilton  (1983)  in the VA mycorrhizal and  Fitter  literature, Verkade and  (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 diameter.  trees,  with  greater  increases  in  dry  weight  and  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  severely P deficient.  non-mycorrhizal  plants,  which  may  be  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  orthophosphate treatment:  for  foliar  P occurred  with  the  low  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 treatment.  increased with this  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  treatments, especially at the low application rate.  acid  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. Jayachandran et al.  1992; Pasqualini et al. 1992;  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. Pate 1992; Barrett-Lennard et al. other  nutrient  possibly  deficiencies  because of  deficiencies,  or  1993) . There were no reports of  with  the increased  the  shorter  1982; Adams and  phytic  acid  sensitivity  duration  of  as  a  P source,  of cedar  other  to Ca  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  nutrients.  problems  were  caused  by  the  complexing  of  other  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 trees  were  also  able  to  use  P from the ATP.  glycerophosphate  and  The  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 Pasqualini et  1988;  et  Lennard  al.  al.  1992;  1982; Tarafdar and Classen  Jayachandran  1993; Tarn and  Mitchell 1993).  al.  et al. 1992; Barrett-  Griffiths  1993; McElhinney  and  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  diameter. soil,  phosphatase  to  the  change  plant  height  and  Enzyme activities were higher on roots than in bulk  suggesting  that  these  enzymes  are  mycorrhizae or rhizosphere microorganisms. the  in  "root"  phosphatases  discussed  here  produced  by  roots,  It should be noted that 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  microorganisms.  plant  roots,  mycorrhizae  and  rhizosphere  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. bulk  soil,  "Soil" phosphatase activities were measured in the 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  1991), are sensitive to soil moisture levels 1991; Baligar et al. phosphate  al.  (Spier and Cowling  1991), and are strongly inhibited by inorganic  (Spier and Ross 1978).  alkaline phosphatases  Higher plants do not produce  (Speir and Ross 1978; Tabatabai 1982).  In  this study, enzyme differences between mycorrhizal and benomyltreated 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.  reported  only  by  Gianinazzi  1987), but alkaline phosphatase activity has been a  few  1976, 1978;  researchers  (Gianinazzi-Pearson  Ho and Zak 1979; Krishna et al.  185  and 1983;  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 Pearson and Gianinazzi 1978).  (Gianinazzi-  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. plants,  and  in  Acid phosphatase is higher in mycorrhizal  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  186  as  acid  and  alkaline  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  Apparently, mycorrhizal plants  in  non-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. 187  However, in the forest, cedar  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  mycorrhizal  the and  apparent  differences  benomyl-treated  in  plants,  this  such  study  as  between  elevated  Zn  concentrations in benomyl-treated trees, may be due to the improved N  nutrition  treatments.  of  the  benomyl-treated  Unfortunately,  differences in plant P.  this may  trees have  from also  the  benomyl  affected  some  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  deficiencies.  from  the  Although  nutrient  phophatases  solutions, were  inducing  produced,  it  Ca was  impossible to distinguish those of the plant and mycorrhizae from the  enzymes  colonization  of  rhizosphere  improved  foliar  microorganisms.  P content, but  other  Mycorrhizal 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  ignition and extraction method.  the  Saunders  and  Williams  Both the Saunders and Williams  method and the Bowman and Moir extraction procedure appeared to overestimate organic P in these soils. a  comparison of the P-extraction  revealed by  This was established by  techniques  31  P NMR spectroscopy.  to the P forms  Either of the Bray PI or  Mehlich 3 procedures were suitable to determine available P. A second hypothesis was that some extraction procedures for 31  P NMR spectroscopy might be better suited to these soils than  other  extraction  procedures.  It  would  reasearch than an ideal soil extractant for 189  appear 31  from  this  P NMR does not yet  exist.  The new extraction procedure for organic P developed by 31  Bowman and Moir, which had not been used as an extractant for NMR  spectroscopy  prior  to  this  study,  shows  great  P  promise  because it extracts more of the total organic P than do the convential extractant  extractants also  NaOH  maintains  and  Chelex.  paramagnetic  However,  ions  such  this  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. differences  This study revealed that, although exist  concentration, extractable  between  loss  on  the  two  ignition,  LCa concentration,  forest C/N  and  significant  types C/P  significant for  pH,  ratios  differences  C  and 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. and  Immediately after clear-cutting  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  concentrations levels.  10 of  years these  after  cutting  nutrients  had  and  burning,  returned  to  the  preburn  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  in B  indicate that more orthophosphate may accumulate  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  enrichment  forms  of  the  from  non-mycorrhizal  'non-mycorrhizal'  trees.  seedlings  Although from  N  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 191  from the  enzymes of rhizosphere microorganisms. This  study  suggests  a  number  of  research  directions.  Phosphorus-31 NMR spectroscopy should be tied into conventional fractionation  and  understanding  of  extraction  partitioning soil  techniques,  organic  techniques  matter.  31  for  P-NMR,  To  other  to  further  further  our  develop  combinations  of  extractants and resins could be tried, in an attempt to achieve the  high percentage  procedure,  but  paramagnetic ions.  of  total  without  P extracted  the  by  concommitant  the  NaOH-EDTA  extraction  of  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. greenhouse  study  also  warrant  further  The results of the investigation,  in  a  simpler study with better controls on rhizosphere organisms, and perhaps  with  a  different  fungicide  colonization.  192  to  control  mycorrhizal  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. phosphorus compounds in extracts of selected karri (Eucalyptus diversicolor Soil Biol. Biochem. 21: 523-528.  31  P-NMR analysis of surface soils from F. Muell.) forests.  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. of forest fire. Bot. Rev. 26: 483-533.  Ecological effects  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. SOSSSI. 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 soybean endomycorrhizae. Phytopathology 68: 1810-1812.  on  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. activities in Appalachian soils. 3. Pyrophosphatase. Soil Sci. Plant Anal. 22: 1537-1545.  Enzyme Commun.  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 curves. J. Soil Sci. 29: 477-462.  of phosphate  Bartlett, E. M., and D. H. Lewis. 1973. activity of mycorrhizal roots of beech. 249-257.  adsorption  Surface phosphatase Soil Biol Biochem. 5:  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. storage and utilization by fungi.  194  1980. Phosphorus uptake, Adv. Bot. Res. 8: 128-219.  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: 225252. 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: 15161518. 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, W I . Brockley, R. P., R. L. Trowbridge, T. M. Ballard and A. M. Macadam. 1992. Nutrient management in interior forest types. pp. 4354. 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: 489495. 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. phosphorus. Soil Sci. 84: 133-144.  Fractionation of soil  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. Douglas-fir Stands. Seattle. 119 pp.  Phosphorus dynamics in red alder and Ph. D. Thesis. University of Washington.  Condron, L. M., E. Frossard, H. Tiessen, Stewart. 1990a. Chemical nature cultivated and uncultivated environmental conditions. J. Soil  R. H. Newman, and J. W. B. of organic phosphorus in soils under different 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. KogelKnabner. 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: 925. 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. Plant Soil 71:  Phosphatase production by mycorrhizal fungi. 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. Phosphatase of plants fungi. New  C. Burton, R. G. Burns and P. Jeffries. 1987. activity associated with the roots and rhizosphere infected with vesicular-arbuscular mycorrhizal 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. products of soil phospholipids.  Identification of hydrolysis 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. Academic Press. New York.  1983.  Mycorrhizal  Symbiosis.  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. soil phosphorus in the Bodenk. 152: 175-180.  1989. Mobilization and turnover of rhizosphere. Z. Pflanzenernahr.  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 ectomycorrhiizal fungi. Can. J. Bot. 57: 1203-1205.  six  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 31 P 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. New York. pp. 267-298.  Chapman and Hall.  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 Cambridge University Press. 622 pp. John,  of  Fungal  Nutrition.  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. soil microorganisms.  1970. Production of phospholipases by Soil Sci. 110: 355-358.  Kozlowski, T. T. , and C. E. Ahlgren, eds. Ecosystems. Academic Press. New York.  1974.  Fire  and  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: 551553 . 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. Phosphatesolubilizing activity of ectomycorrhizal fungi in vitro. Can. J. Bot. 69: 342-346. Lavkulich, L. M. 1981. Methods Manual, Pedology Laboratory. Dept. of Soil Science. Vancouver. 211 pp. Legg,  UBC  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. Interscience Publ.  Chemical Equilibria in Soil. John Wiley and Sons. New York.  Wiley-  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: 147160. 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. C, N, S, and P after forest clearance mineralization rates and spatial variability. 36: 585-591.  Soil organic in Nigeria: J. Soil Sci.  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 W I . 1244 pp. Ognalaga, M., E. Frossard and F. Thomas. 1994. Glucose-1phosphate and myo-inositol hexaphosphate adsorption mechanisms on goethite. Soil Sci. Soc. Am. J. 58: 332-337. Ogner, G. 1983. 31 P-NMR spectra of humic acids. A comparison of four different raw humus types in Norway. Geoderma 29: 215219. Olila, 0. G., and K. R. Reddy. 1995. Influence phosphorus retention in oxidized lake sediments. Soc. Am. J. 59: 946-959.  of pH on Soil Sci.  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 in forest soils. Soil Biol. Biochem. 18: 35-40.  activity  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 Douglasfir 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. 376-391.  Mycorrhizas in ecosystems.  Experientia 47:  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. Bot. 15:  1964. Phytase activity of plant roots. 654-658.  J. Exp.  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: 529533 . 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. York.  1986.  Cycles of Soil.  John Wiley & Sons.  New  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. change.  1975. Effect of species on nutrient cycles and soil 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. and Wilson, Inc. New York. 365 pp.  Holt, Rinehart  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: 193202 . Syers, J. K. , J. D. H. Williams and T. W. Walker. determination of total phosphorus in soils materials. N. Z. J. Agric. Res. 11: 757-762. 212  1968. The and parent  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: 9598. 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. Co. Don Mills, ON. 642 pp.  Addison-Wesley Publishing  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: 209215. Tsubota, G. 1959. Phosphate reduction in the paddy field. Soil PI. Fd. Tokyo 5: 10-15.  I.  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 32Plabelled bacterial cells. Soil Biol. Biochem. 19: 559-565. Verkade, S. D., and D. F. Hamilton. growth of Liriodendron tulipifera  the  vesicular-arbuscular  Scientia Hortic.  21:  1983. Effects of benomyl on seedlings inoculated with  fungus,  Glomus  fasciculatus.  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 NaHCl0 3 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 cedarhemlock 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 salaldominated 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. 4760. 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. 695741. 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. Inc. Evanston, IL. 677 pp.  SYSTAT,  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: 457459. 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,  Zech,  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.  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.  31 Zech, W., H. G. Alt, A. Zucker and I. Kogel. 1985. P-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:

        

Citations by CSL (citeproc-js)

Usage Statistics

Country Views Downloads
Germany 12 18
Canada 11 1
United States 11 8
China 10 13
France 9 0
Japan 4 0
Sri Lanka 2 0
Uganda 1 0
Iran 1 0
Spain 1 0
Romania 1 0
City Views Downloads
Unknown 15 34
Heidelberg 12 0
Ashburn 7 0
Beijing 5 7
Tokyo 4 0
Shenzhen 3 6
Shanghai 2 0
Kamloops 2 1
Colombo 2 0
Paris 2 0
Surrey 1 0
Ottawa 1 0
Sidney 1 0

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

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0088246/manifest

Comment

Related Items