INFLUENCE OF PLANT SPECIES AND SOIL CONDITIONS ON

PLANT-SOIL FEEDBACK IN THE JARRAH FOREST

Martha Orozco Aceves

Bacteriological Parasitological Chemist, National Polytechnical Institute,

Mexico

MSc. Alternative Agriculture with mention in Ecological Agriculture,

National University of Costa Rica, Costa Rica

Thesis presented for the degree of Doctor of Philosophy

The University of Western Australia

School of Earth and Environment

Faculty of Natural and Agricultural Sciences

2015

i

ABSTRACT

Plant-soil feedback (PSF) describes how changes to soil properties are caused by plants

(via conditioning), which in turn influence the plant performance (via soil feedback).

Plant-soil feedback is driven by soil conditions and plant species involved in the

interaction, however, the influence or ‘weight’ of each one of these two components on

PSF outcomes has been scarcely examined. Studying the influence of plant species and

soil conditions on PSF is especially important in highly endemic ecosystems such as the

jarrah forest. This ecosystem displays particular features that define its structure and

function, and general ecological theories often fall short to explain those particularities.

In addition, the jarrah forest has been subjected to a strong anthropogenic pressure by

mining activity. In this case, knowing the influence of plant species and soil conditions

on PSF may contribute to the design of effective restoration strategies.

My thesis focuses on the influence of plant species and soil conditions on PSF that

operates in the jarrah forest of Western Australia. Specifically, PSF was investigated in

the contexts of unmined soils and soils restored after bauxite mining. The research is

presented in three themes: 1) a detailed study of the influence of plant species and soil

contexts on individual PSF associated with representative plant species of the jarrah

forest, described in chapter 3; 2) a study of the influence of changing soil conditions

present in a chronosequence of restoration ages on plant performance and associated

individual PSF, described in chapter 4; and 3) the influence of long-term conditioned

soil by non-native plantation species on PSF associated to native jarrah plants, described

in chapter 5.

To accomplish the aims of the research, a series of glasshouse experiments were

designed using unmined and restored soils that were conditioned in the field and the

glasshouse. To examine individual effects of PSF, the standard method of comparing

ii

plant biomass produced in soil conditioned by the same plant species (‘home soil’) with

the plant biomass produced in soil conditioned by a different plant species (‘away soil’)

was consistently used in each of the experiments. Plants used for the experiments were

representative of the jarrah forest, these being Eucalyptus marginata, Bossiaea ornata

(both long-lived species), and Acacia pulchella (a short-lived species).

Results from the first glasshouse experiment designed to study the influence of plant

species and soil contexts on individual PSF are described in Chapter 3. The main

findings from this experiment were that individual PSF associated with B. ornata and A.

pulchella was influenced by soil context (unmined, restored), and species-specific

conditioning of soil caused by plant species, influenced the PSF only in restored soil.

In unmined soil, the three species E. marginata, B. ornata and A. pulchella conditioned

the properties via changes to the chemistry (as compared with unplanted soils), but had

little effect on the biological properties measured. Eucalyptus marginata and B. ornata

conditioned the chemical properties of soil in a similar manner, and A. pulchella

conditioned soils differently. In restored soil, altered conditioning of the chemical

properties was observed for E. marginata and B. ornata and no conditioning was

observed for A. pulchella. Again, the biological properties of restored soil were little

affected by conditioning. Overall, these findings indicate that individual PSF and

conditioning capacity associated with the studied plant species is driven by soil context,

which can be critical for the restoration of native forest post-bauxite mining given the

modified properties present in restored soil.

In chapter 4, I described the results from a second glasshouse experiment designed to

study the influence of soil conditions on PSF. For the study, field-conditioned soils from

a chronosequence of restoration ages were used. These soils displayed different

properties that approached those of unmined soil as restoration age increased. The soils

iii

exerted differential soil feedback on the biomass produced by B. ornata. As a result,

different amounts of biomass were produced by the plant species. Overall, the soil

feedback of soils from the chronosequence was driven by soil age (since restoration),

not by plant species from which soils were collected in the field. From individual

properties of soil, the number of bacterial-feeding nematodes and the rate of CO2

produced by microbial respiration of phenolic compounds from CLPP were identified as

key components of the soil feedback across all soils from the chronosequence. These

properties were correlated with B. ornata biomass production. The results reinforced the

hypothesis of a PSF influenced by soil conditions more than by plant species is

operating in restored jarrah forest.

Finally described in chapter 5, a third glasshouse experiment was conducted to

investigate changes in properties of jarrah forest soils after long-term conditioning by

the non-native plantation species pine (Pinus radiata) and blue gum (Eucalyptus

saligna), and native jarrah (E. marginata). Then, the influence of conditioned soils on

growth of jarrah seedlings was tested. Pines and blue gums similarly conditioned the

physico-chemical properties of soils, which differed from soil conditioning caused by

jarrah. The two eucalypt species similarly conditioned the biological properties of soil,

and these differed from conditioning caused by pines. Species-specific conditioning of

soil did not translate into differences in the amounts of biomass produced by jarrah

seedlings, and a neutral PSF was observed. The results suggest that jarrah displays a

plastic PSF that is not strongly influenced by soil context, instead PSF associated with

this plant species is driven by the plant itself. This behavior might contribute to the

persistence and dominance of E. marginata in the jarrah forest.

The studies included in this thesis have contributed to a better understanding of PSF

operating in the endemic jarrah forest ecosystem. Specifically, this thesis contributed to

iv

understand the influence of plant species and soil conditions on PSF. Moreover, the

influence of plant species and soil conditions was examined in the contexts of unmined

soil and restored soil post-bauxite mining, as mining is a common disturbance of the

jarrah forest. Restored soil post-bauxite mining displays modified physico-chemical and

biological properties that are likely to alter individual PSF, with implications for the

restoration of the ecosystem.

v

TABLE OF CONTENTS

Abstract i

Table of contents v

Thesis declaration viii

Acknowledgements ix

List of publications xi

Chapter 1. General Introduction 1

Plant-soil feedback theory 1

Influence of plant species on plant-soil feedback: soil conditioning 2

Root deposition 3

Litter deposition 6

Influence of soil conditions on plant-soil feedback 8

Soil feedback mediated by chemical properties 9

Soil feedback mediated by biological properties 10

Empirical evidence for importance of plant-soil feedback in

vegetation succession 12

Implications of the plant-soil feedback for restoration of ecosystems 16

The jarrah forest 17

Distribution and climate 18

Soils 19

Ecological aspects 19

Evidence of potential mechanisms of soil conditioning by jarrah forest plants 21

Mining activity and restoration of the jarrah forest 22

Thesis outline and research objectives 29

Chapter 2. General methods 30

Response variables 30

vi

Soil chemical properties 30

Water content and water holding capacity 30

Soil biology and associated properties 31

Microbial biomass-carbon (C) 31

Phospholipid-fatty acid (PLFA) profiles 31

Community level physicological profiles (CLPP) 32

Nematode trophic community 33

Chapter 3. Soil conditioning and plant-soil feedbacks associated with

jarrah forest plants are soil-context dependent 34

Introduction 34

Materials and methods 36

Sampling and experimental design 36

Response variables 40 Statistical analyses 49

Results 42

Changes in properties of soils conditioned during phase one 42

Unmined soil 42

Restored soil 44

Individual PSF 45

Discussion 47

Chapter 4. Plant-soil feedback unaffected by soil condition in a

post-mining restoration chronosequence 54

Introduction 54

Materials and methods 57

Sampling and experimental design 57

Response variables 58

Statistical analyses 58

vii

Results 62

Properties of field-conditioned soils across the chronosequence

and their trajectory of change 62

Soil feedback of soils from the chronosequence 69

Individual PSF with respect to soil age 72

Discussion 73

Chapter 5. Long-term conditioning of soil by eucalypts and pines does

not alter the plant-soil feedback associated with native jarrah 83

Introduction 83

Materials and methods 86

Sampling sites and experimental design 86

Response variables 87

Statistical analyses 87

Results 88

Properties of soils conditioned in the field by pines, jarrah, and blue gums 88

Individual plant-soil feedback 94

Discussion 94

Chapter 6. Concluding discussion 101

References 106

Appendices 123

Appendix 1: Carbon substrates used for CLPP 123

Appendix 2: Publications 124

viii

THESIS DECLARATION

The thesis was substantially completed during my period of enrolment in a PhD degree

in the School of Earth and Environment at the University of Western Australia and has

not been previously accepted for a degree at this or any other institution.

The thesis is my own composition, conducted under the supervision of Mark Tibbett

and Rachel J. Standish. Supervisors assisted with establishing the direction of the

research, experimental design and planning, offered technical advice and provided

editorial input to drafts. I designed the experiments, conducted field, glasshouse, and

laboratory work, as well as data analyses.

ix

ACKNOWLEDGEMENTS

I would like to express my sincere gratitude and appreciation to my supervisors Mark

Tibbett and Rachel Standish for their support, advice, and encouragement throughout

my candidature, and to teach me the meaning of being a ‘Scientist’.

I am very appreciative to the Consejo Nacional de Ciencia y Tecnología (National

Council on Science and Technology of Mexico) for funding my PhD studies. I also

acknowledge all the staff members of the Graduate Research School for their excellent

job and for financially support my stay in the South Australian Research and

Development Institute (SARDI) in Adelaide.

Thanks to all the staff in the School of Earth and Environment and Soil Science and

Plant Nutrition department, especially Tim Morald for his help during field and

laboratory work, Michael Smirk, Evonne Walker, and Darryl Roberts for his technical

assistance during laboratory work. Robert Creasy and staff for providing the necessary

materials and equipment at the plant growth facility during glasshouse work. Yoshi

Sawada, Natasha Banning, and Suman George for their advice on methodological

aspects of my research. Margaret Pryor, Gail Ware and Karen Newham for making all

the administration work easy.

Enormous thanks to John Koch from Alcoa World Alumina Australia for his assistance

and support for field work. Jackie Nobbs from SARDI to training me on identification

of trophic groups of nematodes, and Vivien Vanstone, Sarah Collins and staff from the

Department of Agriculture and Food of Western Australia (DAFWA) for help with

nematode extraction. I thank Ken Dodds and staff from the ChemCentre, and Richard

Hobbs from the School of Plant Biology (UWA) for their support to complete the PLFA

x

analyses. Deborah Lin, Khalil Kariman and Alonso Calvo Araya for their help during

critical phases of the experiments.

I would like to thank all members of the Soil Ecology group and the Ecosystem

Restoration and Intervention Ecology Research group (ERIE), and postgraduate

students at the Soil Science and Plant Nutrition department for your advice, support, and

friendship.

Finally I would like to express gratitude and appreciation to my family and friends for

your love, support, and motivation.

xi

LIST OF PUBLICATIONS

Publications arising from this thesis (Appendix 1):

Orozco-Aceves, M., Standish, R.J., Tibbett, M., 2015. Soil conditioning and plant-soil

feedbacks in a modified forest ecosystem are soil-context dependent. Plant Soil DOI

10.1007/s11104-015-2390-z

Orozco-Aceves, M., Standish, R.J., Tibbett, M., 2015. Long-term conditioning of soil

by plantation eucalypts and pines does not affect growth of the native jarrah tree. Forest

Ecol Manage 338:92-99.

Presentations in Scientific events

Orozco-Aceves, Martha; Tibbett, Mark; Standish, Rachel J. 2011. Niche construction by

plants in soils modified by bauxite mining in the jarrah forest in Western Australia.

Student Symposium of the Australasian Plant Pathology Society. South Perth, Western

Australia, 27 October.

Orozco-Aceves, Martha; Tibbett, Mark; Standish, Rachel J. 2011. Niche construction by

plants in soils modified by bauxite mining in the jarrah forest in Western Australia.

Rhizosphere 3 Conference, Perth, Western Australia, 25 – 30 September.

1

CHAPTER 1

General Introduction

Plant–soil feedback theory

Plant-soil feedback (PSF) theory describes how plants perceive and respond to the

edaphic environment by changing the physical, chemical and biological properties of

soil, that in turn affect plant performance (Bever 1994; Bever et al. 1997; Bever 2003).

Plant-soil feedback can be divided into two processes for study and understanding, these

are: plant conditioning (Harrison & Bardgett 2010; Pernilla et al. 2010; van de Voorde

et al. 2011) and, soil feedback (van der Putten et al. 2013; Figure 1). The two processes

occur simultaneously and the resulting effects (i.e. described by their magnitude and

direction) can be positive or negative. A positive PSF has occurred when there is an

increased plant biomass produced in soil conditioned by the same plant species (often

referred to as ‘home soil’) compared with the plant biomass produced in soil

conditioned by a different plant species (often referred to as ‘away soil’; Mills & Bever

1998; van der Heijden et al. 2008). Positive or negative PSF can ultimately shape the

structure and function of the entire community if the PSF is large relative to intrinsic

growth differences among plant species to affect the outcome of competitive

interactions (Wardle et al. 2004; Bardgett et al. 2005; van der Putten et al. 2009).

The mechanistic basis by which plant performance is regulated by PSF has been

described. A negative PSF occurs due to the accumulation of parasites, pathogens, and

root herbivores in the rhizosphere that can directly remove C and nutrients from plant

tissues and reduce root uptake capacity. Examples of soil pathogens that can act as

agents of negative PSF are the members of the genus Pythium (oomycetes), the fungus

Fusarium oxysporum, and plant-parasitic nematodes (van der Putten & Peters 1997;

Mills & Bever 1998; van der Stoel et al. 2002; Mangan et al. 2010; Liu et al. 2012).

2

Plants show variable degree of susceptibility to parasites and pathogens, and so the

multiple interactions between these groups of organisms and plant species tend to be

species-specific. Negative PSF can be also associated to plant-symbiont interactions that

confer nutritional advantage to specific plant species over other species (Mills & Bever

1998; van der Putten 2000; Mangan et al. 2010; Liu et al. 2012).

A positive PSF is driven by the accumulation of beneficial organisms in soil around

roots such as those within the soil food webs, and the establishment of mutualistic

symbionts such as N-fixing bacteria and arbuscular-mycorrhizal fungi (AMF; Ferris et

al. 2001; Reynolds et al. 2003; Bardgett et al. 2005; Compan et al. 2005; Ahmad et al.

2008; Bever et al. 2012). Soil food webs unlock nutrients present in organic matter,

which improves the rates of recycling and turnover, thus increasing nutrient availability

and plant productivity (Bever 1994; Bever et al. 1997; De Deyn et al. 2004; Wardle et

al. 2004; Xiaoyun et al. 2007; van der Heijden 2008). Arbuscular-mycorrhizal fungi

increase interplant transfer of C and nutrients via hyphal links, and enhance access to

limiting nutrients such as P, thus increasing plant growth (Bever 2002; Reynolds et al.

2003; van der Heijden 2008; Bever et al. 2010).

Influence of plant species on plant-soil feedback: soil conditioning

Plant species can condition the physical, chemical and biological properties of soil

(Kyle et al. 2007; Kulmatiski et al. 2008; van der Putten et al. 2013) by inputs of

chemical compounds and organic matter such as root and litter deposition (Marschner

1995; Neumann & Römheld 1999; Uren 2001; Bertin et al. 2003; Ehrenfeld et al. 2005;

Neumann 2007; Pang et al. 2010). The conditioning of soil can be direct or indirect.

The first involves a direct plant effect on soil properties, for example root deposition

can directly improve the structure of soil by addition of mucilage. The indirect

conditioning is mediated by other soil elements, for example, C released to soil via root

3

deposition promotes an increased microbial biomass, which in turn positively impact

plant growth. Mechanisms for conditioning of soil via root and litter deposition will be

discussed in more detail in the following sections.

Figure 1. Plant-soil feedback (PSF) can be divided in two: 1) plant conditioning describing the effect of plants on soil properties by influence of root exudation and root and litter deposition, and 2) soil feedback describing the effect of physical, chemical and biological properties of soil on growth, productivity, and reproductive success of plants.

Experimentally, it has been observed that soil conditioning is influenced by the

procedure used. There are two main approaches to obtain conditioned soil; these are the

‘natural experiment approach’ where plant-conditioned soils are collected from single

plants in the field, and the ‘cultivation experiment approach’ where soils are

conditioned under controlled conditions in the glasshouse (Kulmatiski & Kardol 2008;

Pernilla et al. 2010; van de Voorde et al. 2012).

Root deposition

Root deposition comprises exudates and dead roots (Uren 2001), and accounts for 30 to

60% of the C fixed by photosynthesis that is translocated to roots, from which up to

1. P

lant

con

diti

onin

g 2. Soil feedback

Plant-soil feedback

4

70% is released to soil (Marschner 1995; Brimecombe et al. 2001; Neumann &

Römheld 2001; Figure 2). Root deposition helps plants grow, especially under

conditions of environmental stress such as mechanical impedance, anaerobiosis,

drought, presence of pollutants, and nutrient deficiency (Marschner 1995; Uren 2001;

Nautiyal et al. 2008; Martin et al. 2014). Root deposition also varies with plant species,

cultivars, and plant developmental stage (Marschner 1995; Uren 2001). This

observation implies that root deposition to soil can vary within the life span of a plant

with implications for the conditioning of soil properties at a local level and subsequent

levels of community organisation through its effects on succession of the plant

community.

Root exudates directly influence availability of nutrients such as K, Ca, Mg, P, S, Fe,

and other micronutrients by modifying the soil pH (Jones 1998; Uren 2000; Bais et al.

2008; Richardson et al. 2009). Plants excrete inorganic ions (i.e. HCO3−, OH−, H+) and

organic acids (i.e. citric, malic, malonic) that solubilise or mobilize soil nutrients to

make them readily available for uptake by plant roots (Dakora & Phillips 2002;

Ehrenfeld et al. 2005; Richardson et al. 2009; Carvalhais et al. 2011). Plant-driven

changes of soil pH help plants to overcome toxic levels of elements present in soils (i.e.

Al; Jones 1998; Dakora & Phillips 2002). Also, by altering the pH of soil, plants have

major impact on the structure of microbial communities around the root (Badri &

Vivanco 2009; Richardson et al. 2009).

Plants release to soil chemical signals to attract beneficial and harmful organisms to

colonize the rhizosphere. Among beneficial organisms there are those that tend to act as

agents of positive PSF, for example symbionts such as N-fixing bacteria and AMF.

Symbiotic organisms mobilise and transfer nutrients (i.e. N, P) to plants (Hirsch et al.

2003; Atkinson 2006; Sbrana 2006). Root exudates also serve as sources of C for free-

5

living soil organisms (Bais et al. 2008), which in turn promote plant growth, mineralise

nutrients, detoxify pollutants, modify soil structure, and regulate the abundance of plant

pests and other opportunistic species (Neher 1999; Wardle 1999; Ferris et al. 2001).

Agents of negative PSF such as oomycetes, pathogenic fungi, and plant-parasitic

nematodes (Werner 2001; Nelson 2006; Bais et al. 2008; Badri & Vivanco 2009; Perry

& Tovar-Soto 2012) are also attracted to plant roots by chemical signals released by

plants in a species-specific manner (Hirsch et al. 2003).

Some plant species can release allelochemical compounds into soil that negatively

affect soil organisms and other plants. The main group of allelochemical compounds are

phenolic acids and their derivatives: terpenoids, steroids, coumarins, flavonoids,

alkaloids, cyanohydrins and tannins (Lavelle & Spain 2003). An important group are

the flavonoids, which can directly and indirectly influence chemical and biological

properties of soil by affecting the C, N, P, S cycles (Lavelle & Spain 2003; Cesco et al.

2012; Lotina-Hennsen et al. 2006; Hadacek 2008).

6

Figure 2. Classification, quantities and release mechanisms of root deposition. Reproduced from Neumann (2007).

Litter deposition

Litter chemistry (i.e. C:N ratio, lignin and phenolic compounds content) is a species-

specific trait of plants that directly influences the physical, chemical and biological

properties of soil (Hobbie 1992; Guo & Sims 1999; Berg & McClaugherty 2008;Witt &

Setälä 2009; Uselman et al. 2012). Litter decomposition releases nutrients that alter the

nutritional status of the soil that in turn affects plant nutrition (Lugo et al. 1990;

Ganjegunte et al. 2005; Berg & McClaugherty 2008; Uselman et al. 2012). Litter

decomposition produces humus that improves soil structure (Jastrow et al. 1998; Malik

& Scullion 1998). Functional aspects of soil are also influenced by litter deposition, for

example, plants producing litter with low C:N ratios (i.e. high-quality litter) promote a

soil biota dominated by bacteria and its predators (i.e. protozoa, bacterial-feeding

ROOT DEPOSITION

LYSATES

Passive release from damaged

and sloughed-off root cells

Root turn-over

ROOT BORDER CELLS

Controlled release of intact calyptra cells

EXUDATES

Release from intact root cells

DIFFUSATES SECRETIONS EXCRETIONS

Passive release by diffussion

Controlled release of

compounds with specific

functions

Controlled release of metabolic

waste productisLow-molecular-

weight-compounds

Allelochemicalcompounds

(i.e. phenolics, phenol,

flavonoids)

PhytosiderophoresCarboxylates

PhosphohydrolasesMucilage

Lactic acid

7

nematodes; Figure 3). As a consequence, the bacterial-based energy channel of organic

matter decomposition predominantly operates in soil. In contrast, plants producing litter

with high C:N ratios (i.e. low-quality litter) promote a soil biota dominated by fungi and

its predators (i.e. fungal-feeding nematodes, collembolans, oribatid mites), and then the

fungal-based energy channel of decomposition predominantly operates in soil (Moore &

Hunt 1988; Scheu et al. 2005). The bacterial-based energy channel is the fast track for

organic matter decomposition, in contrast to the fungal-based energy channel, which is

much slower (Moore & Hunt 1988; Figure 3).

Another mechanism by which plants can condition the soil properties via litter

deposition is the release of allelochemical compounds that are leached from fallen litter.

These allelochemical compounds produce alterations to the chemical and biological

properties of soil. For example, degradation products of litter lignin such as quinines

and phenols are involved in series of reactions that produce H2O2 (Inderjit & Malik

1997; Inderjit & Weiner 2001), which is toxic to microbes due to peroxidation of lipids

of cell membranes. Consequently, membrane permeability increases and destabilization

of cell processes occurs (Weir et al. 2004). Phenolic compounds also have inhibitory

effects on N-fixing microorganisms, symbiotic or free-living; monoterpenes negatively

affect nitrifying bacteria, and a wide variety of essential oils show antimicrobial activity

(Chalkos & Karamanoli 2006).

8

Figure 3. Simplified bacterial and fungal energy channels of a soil food web. The bacterial energy channel represents the ‘fast cycle’ due to the higher turnover rates of detritus by bacteria and their consumers relative to the fungi and their consumers. The fungal energy channel represents the ‘slow cycle’. Changes in the C:N ratio of the detritus, net primary productivity (NPP), or rates of N mineralization have been associated with shifts in the relative dominance of one channel to the other. Reproduced from Moore et al. (2007).

Influence of soil conditions on plant-soil feedback

Soil conditions comprise the physical, chemical and biological properties that are

shaped by the effects of natural phenomena (i.e. weathering, biogeochemical cycles)

and anthropogenic activities (i.e. agriculture, forestry, livestock, mining). Soil

conditions determine the growth, productivity and reproductive success of plants via

soil feedback (Pregitzer et al. 2010; van der Putten et al. 2013; Figure 4). The soil

properties are perceived by sensors and signal transducers of plants (Hirsch et al. 2003;

Kanesaki et al. 2010; Rock et al. 2010), which activate chemical pathways for gene

expression (Rock et al. 2010), thus affecting the plant phenotypes. Soil is a dynamic

Decrease NPPWiden C:N

Decrease N-mineralizationLabile

detritus

Bacteria

Bacterial feeders

Predators

Fungi

Fungal feeders

Recalcitrant detritus

Bacteria Fungi

Bacterial feeders

Fungal feeders

Predators

Increase NPPNarrow C:N

Increase N-mineralization

“Fast cycle” “Slow cycle”

9

entity and its properties experience changes over time, therefore plant species might be

subjected to differential soil feedback during ontogeny, with variable effects on

individual PSF (i.e. the PSF associated to individual plant species; Kardol et al. 2013;

van der Putten et al. 2013).

Research on soil feedback generally tests the simultaneous effects of multiple soil

properties on plant biomass production (Bonanomi & Mazzoleni 2005; Pregitzer et al.

2010; Brandt et al. 2013). However, it is likely that different soil properties act as key

drivers of the soil feedback during the life time of a plant, and so understanding which

of these drivers are operating at any one time will necessary involve detailed study of

PSF associated with plants of increasing age (Kardol et al. 2013). The soil feedback

mediated by physico-chemical properties is well documented and understood,

particularly for young plants, compared with understanding of those mediated by

biological properties and for mature plants. The knowledge of soil feedback mediated

by biological properties is rapidly developing (Miki 2012; Kardol et al. 2013; van der

Putten et al. 2013), but pieces of information are still missing. In addition, we lack

information regarding the relative contributions or ‘weight’ of physico-chemical and

biological properties on the soil feedback. In the following section soil feedback

mediated by chemical and biological properties are briefly described.

Soil feedback mediated by chemical properties

Chemical properties of soil vary according to the characteristics of the parent material

from which it originated (Buscot 2005). Through weathering process, ions are removed

from rock and mineral, given to soil particular chemical properties such as

concentrations of S, P, K, Mg, Ca, Si, Na, Fe, and a characteristic pH (Gorbushina &

Krumbein 2005). Concentrations of the nutrient elements are essential for plants to

complete their life cycle. Low concentrations of mineral elements in soil can be limiting

10

for plant growth (Foth 1990; Marschner 1995), whereas high concentration of mineral

elements can induce nutritional imbalances or toxic effects to plants (Bohn et al. 2001).

Therefore, soil nutrients and their interactions are critical components of the soil

feedback.

Soil pH directly affects the composition of plant communities, which show a clear

differentiation along gradients of pH. Soil pH is directly related to concentrations of

cations, and plant community composition is specially affected by extreme or unusual

concentrations of particular cations (Ehrenfeld et al. 2005). Soil pH also affects the

structure of the microbial community, which drives the organic matter decomposition

process and thus the availability of nutrients for plant uptake (Ehrenfeld et al. 2005;

Moore & Hunt 1988). Generally, soil pH below 5 favours the growth of fungi, whereas

neutral soil pH favours the growth of bacteria (Madigan et al. 2012). In turn, a

microbial community dominated by fungi promotes the functioning of the fungal-based

energy channel of organic matter decomposition. In contrast, a microbial community

dominated by bacteria promotes the functioning of the bacterial-based energy channel

of decomposition (Moore & Hunt 1988; Madigan et al. 2012). Thus, soil pH influences

the rates at which organic matter decomposes in soil. Moreover, changes in soil pH due

to human activities such as agriculture, forestry, and mining are likely to have

significant consequences for PSF. I have used soil pH as an example, yet there are other

chemical properties, that if modified through human activities, could lead to

modifications to PSF that usually operates in the ecosystem.

Soil feedback mediated by biological properties

Soil microbes are very abundant in soil and are the first organisms making contact with

plants, so it is probable plants are subjected to their influence during initial stages of

development through diverse mechanisms; for example, through growth mediated by

11

chemical substances such as carbon and nitrogen, solubilisation of mineral elements for

their nutrition, such as phophorus, mineralization of organic matter, soil aggregation,

regulation of abundance of pests and pathogens, N-fixation, (Neher 1999; Wardle 1999;

Altieri 1999; Buscot & Varma 2005). Plants are also subjected to the direct or indirect

influence of other soil microbes that are less abundant in soil (i.e. chemolithothrophs,

authotrophs. In addition to microbes, plants are subjected to the direct and indirect

influence of other groups of organisms within the soil food web, which are in large

numbers in soil such as protozoa, nematodes, mites, collembola, and insects (Gupta &

Yeates 1997; Griffiths et al. 2007; Yeates et al. 2009; Figure 4). For example, plants are

directly influenced by free-living nematodes (i.e. bacterial-, fungal-feeding) that

mobilize and mineralize a considerable amount of N, which is then available for plant

uptake (Freckman 1988; Postma-Blaauw et al. 2005; Yeates et al. 2009). Plants are

directly influenced by plant-parasitic nematodes that cause severe damage to plants,

specifically due to root damage (Perry & Tovar-Soto 2012). Opposite to ther organisms

in soil, nematodes are able to move in water films, for this reason nematodes are

capable to reach plant roots and influence plant development positively or negatively

(Lavelle & Spain 2003). Migration of nematodes occurs in soil and between plants only

when water is available or, soil texture adequate (Castillo & Vovlas, 2007). Protozoa

and collembola positively influence plant productivity due to ingestion of propagules of

fungal and bacterial pathogens (Lavelle & Spain 2003; Coleman et al. 2004). In

summary soil micro, meso and macrofauna influence plant growth by recycling

carbohydrates and nutrients through mineralisation, transferring energy in the detritus

food chain, providing reservoirs of nutrients, detoxifying pollutants, modifying soil

structure, and regulating the abundance of pests and other opportunistic species (Neher

1999; Ferris et al. 2001). Plants are strongly influenced by soil feedback mediated by

biological properties, hence soil modification by human activities that generally has

12

negative impacts on soil biota (Banning et al. 2008; Schimann et al. 2012; Menta et al.

2013) is expected to importantly affect plant performance with implications for the

restoration of ecosystems (Harris 2009).

Figure 4. The soil feedback operates through physical, chemical and biological compartments of soil. Arrows describe linkages between the components of each compartment. Reproduced from Ehrenfeld et al. (2005).

Empirical evidence for importance of plant-soil feedback in vegetation succession

Plant-soil feedback theory has been invoked to explain ecological phenomena such as

vegetation succession. According to traditional models, succession consists of an

orderly and predictable series of plant species replacements that occur after a

disturbance, so-called relay floristics (Cattelino et al. 1979). In this context, PSF theory

PLANT

Physical Biogeochemical Biotic

Structure

Water

Temperature

Organic matter

Nutrients and cations

Microbiota

Micro-Meso-Macro-fauna

13

has been suggested as a mechanism driving the serial species replacement through

regulation of the amounts of plant biomass produced (Bever et al. 1997; Bever 2003;

Kardol et al. 2006; Kulmatiski et al. 2008; van der Putten et al. 2009).

Negative PSF more than positive PSF, has been described as a key driver of structural

changes of plant communities during succession (van der Putten & Peters 1997; van der

Stoel et al. 2002; Bonanomi et al. 2005b; Mangan et al. 2010; Liu et al. 2012). Plant-

soils feedback tends to be negative in early-successional communities (van der Putten &

Peters 1997; Kardol et al. 2006). Early-successional plants show the greatest maximum

growth rate but the highest level of susceptibility to pathogens and parasites, which

creates adverse conditions to plants, opening gaps for the establishment of other species

(Kulmatiski et al. 2008). Therefore, negative PSF induces a greater species turnover

both in time and space, increasing in this way the rate of succession and contributing to

species coexistence (van der Putten & Peters 1997; Mills & Bever 1998; Bever 2003;

Kulmatiski et al. 2008; Figure 5).

The role of parasites, pathogens, and herbivores as agents of negative PSF has been

documented in a variety of ecosystems such as a tropical forest in Panama, a subtropical

monsoon forest in China, and a temperate black cherry forest in Europe (van der Putten

2000; Mangan et al. 2010; Liu et al. 2012). These findings suggest that negative PSF

operates across diverse ecosystems around the world, and in diverse contexts of soils

and plants (Bonanomi et al. 2005a; Bonanomi et al. 2005b; Mangan et al. 2010;

Reinhart 2012).

Positive PSF is likely to enhance plant dominance and so retard succession (De Deyn et

al. 2004). Plant-soil feedback tends to be positive in late-successional communities

(Bever 1994, Bever et al. 1997, Wardle et al. 2004), late-successional plants grow

slowly, and can establish mutualistic symbiosis (i.e. mycorrhiza, N-fixing bacteria),

14

which leads to favourable conditions for their growth, dominance, and persistence in

ecosystems (van der Heijden et al. 1998; Bever 2002; Ehrenfeld et al. 2005; Miki et al.

2010; Figure 5).

Some studies suggest that not only the successional stage of plant species influences

PSF during succession, but the interactive effects of both, plant species and soil

conditions shape PSF. For example, late-successional plants perform best in late-

successional soil (i.e. mature forest soil), and worst in early-successional soil (i.e.

cleared forest, abandoned agricultural fields). In contrast, early-successional plants can

perform bad in both early- and late-successional soils (Kardol et al. 2006). The

nutritional status of the soil can act as driver of PSF (Bezemer et al. 2006; Manning et

al. 2008; Meijer et al. 2011). Positive PSF plays a central role in early-successional

plant communities on nutrient poor soils. In these soils, positive effects of microbes on

plant productivity enhance the supply of limiting nutrients such as N and P to plants (De

Deyn et al. 2004). This microbe-mediated supply is critical in such ecosystems since up

to 90 % of P and N are provided by AMF and N-fixing bacteria (van der Heijden et al.

2008). In contrast, negative PSF is expected to contribute to species replacement and

diversification at more fertile sites. The relative importance of soil biota in suppressing

dominant plant species will increase with decreasing soil fertility (De Deyn et al. 2004).

Other hypothesis points out that plant-soil interactions are complex, during succession

this complexity enhances plant species diversity by delaying community convergence

(reduced beta diversity). Diversity enhancement occurs because the complex plant-soil

interactions lengthen the time during which local species composition change. Due to

the longer time for changes in species composition, a high level of beta (community)

and gamma (regional) diversity at early stages of succession is maintained for a long

time despite eventual community convergence (Fukami et al. 2013).

15

Figure 5. Above-belowground interactions between an early- (left) and late- (right) successional plants, their herbivores, symbionts, predators and top predators. Solid lines indicate direct interactions and dashed lines indicate indirect interactions. Thick lines indicate strong interactions and thin lines weak. Long arrows indicate strong effects on the rate of succession, and short arrows weak effects, the direction implies driving (arrow pointing to the right), or slowing down (arrow to the left) of succession. Long-dash lines relate to the early-successional species, short-dash lines relate to the successor plant, a long/short dash line relates to both plant species. The decomposer subsystem responds relatively slowly when compared with the response of herbivores and symbionts and will generally promote succession. Reproduced from van der Putten et al. (2009).

16

Implications of the plant-soil feedback for restoration of ecosystems

Plant-soil feedback has important implications for restoration of ecosystems, however,

studies on ecosystems restoration that have explicitly consider plant–soil linkages are

currently scarce (Harris 2009; Kardol & Wardle 2010; van der Putten et al. 2013). A

PSF approach considers that restoration of one of the ecosystems’ compartments (below

or aboveground) would be improved by inclusion or consideration of the other (Kardol

& Wardle 2010). For example, in temperate Australian grassy woodlands, the biomass

of exotic annual plant species (aboveground structure) was dramatically reduced by

adding to soil a C source (belowground manipulation) such as sucrose to deplete

seasonally available nitrate pool, which otherwise favours the growth of invasive

species (Prober et al. 2009).

On the other hand, successional-based restoration management is usually inefficient due

to strong feedbacks operating between above and belowground compartments, which

can make the ecosystem resilient to restorative change (Suding et al. 2004). A PSF

approach allows identifying undesirable PSF, and efficiently addressing the constraints

(Suding et al. 2004; Suding & Hobbs 2009). For example, in the Chihuahua desert,

abandoned agricultural fields have slow infiltration rates as consequence of agriculture

practice. The little rainfall that occurs is lost to runoff, creating a positive feedback

where drier conditions promote poor vegetation growth that in turn holds less water

(identification of undesirable PSF). Therefore, restoration treatments that scratch the

soil surface are advisable, once water is held grasses regenerate spontaneously, which in

turn catches more water, thereby reversing desertification (problem addressed by

breaking the undesirable PSF; Whisenant 1999).

In human-modified ecosystems the properties of soil can be importantly altered, this

implies that processes such as plant conditioning and soil feedback are likely to alter,

17

with unpredictable effects on individual PSF associated with specific plant species

(Manning et al. 2008; te Beest et al. 2009; Kardol & Wardle 2010). For example, in

many industrialized countries ex-arable soils display modified properties such as high

fertility, simplified soil communities, and seed limitation. Together, these aspects

produced negative feedbacks on native plant species, which constrained restoration of

the historic ecosystem (Kardol & Wardle 2010). In Florida, post-phosphate mining soils

have modified properties such as increased P concentration and clay, the latter lowers

air and water movement and can range from quite sticky when wet to brick-hard when

dry. These soil conditions do not promote successional change, but only the growth of

colonizing species in these areas, so that the plant community composition of mined

sites does not resemble that of the native ecosystem even after 30 years (Brown 2005).

In Italy, soils post-sand and gravel mining featured altered texture, decreased porosity,

pH and microbial population. Plant-soil feedback developing in these soils is

unfavourable for plant growth, consequently, mined sites are almost devoid of plant

cover (Menta et al. 2013). Examples like these show how soil modification can alter the

patterns of growth, dominance and abundance of plant species via PSF phenomena.

Ultimately, altered PSF can make an ecosystem resistant to restorative change, or can

promote an undesirable restoration trajectory (Suding et al. 2004; Suding & Hobbs

2008).

The jarrah forest

The jarrah (Eucalyptus marginata) forest is an ecosystem native to the south-western

corner of Western Australia. It occurs on the highly leached soils of the Darling Plateau,

which are characterized by strong limitations of nutrients such as P and N (Dell et al.

1989). To overcome nutritional limitations of soil, native plant species have developed

series of strategies to efficiently acquire nutrients (Lambers et al. 2007; Hopper 2009;

18

Sardans & Peñuelas 2013). This information suggests that strong links between plants

and soils operate in the jarrah forest (Abbott & Robson 1977; Malajczuk & McComb

1977; Malajczuk 1979; Malajczuk & McComb 1979; Grierson & Adams 2000;

Lambers et al. 2007; Bleby et al. 2009; Hopper 2009), which makes the ecosystem ideal

to study PSF phenomena. Moreover, the jarrah forest has been under strong

anthropogenic pressure due to activities such as mining for bauxite (alumina). After

mines close, the sites are restored to native forest (Koch 2007b), however, restored

jarrah forest soil displays modified properties (Ward 2000; Jasper 2007; George et al.

2010; Tibbett 2010) that are likely to alter PSF with important implications for the

restoration of the ecosystem. Restored jarrah forest offers a field laboratory to study

PSF phenomena, there is availability of restored soils of diverse types (i.e. types of

forest, types of restoration practices, types of conditioning) and ages (post-restoration)

to test the importance of soil contexts on PSF, and how PSF evolves as succession

proceeds. In the following sections relevant environmental aspects on the jarrah forest

are described.

Distribution and climate

The jarrah forest distribution extends from Toodyay in the north of Albany in the south

and its eastern limit is approximately the 750 mm rainfall isohyets (Figure 6). The

climate is Mediterranean, which is characterized by cool, wet winters and hot, dry

summers. Average summer monthly maximum temperature is 28 °C and average winter

minimum is 5 °C. Annual rainfall ranges from 750 to 1400 mm yr-1. Seasonal summer

drought may last from four to six or seven months (Churchward & Dimmock 1989;

Gentili 1989).

19

Soils

Five broad groupings of soils can be recognized in the jarrah forest, these are sandy

gravels, siliceous sands, earths, non-calcareous loams, and duplex soils (Churchward &

Dimmock 1989; Gentili 1989). Jarrah forest soils (< 2 mm fraction) have lower stores of

total and extractable forms of nutrient elements, mainly P and N. Part of the reason for

the low stores is that an average of about 75% of the whole soil consists of ferruginous

gravel (> 2 mm fraction). Although the gravel contains considerable amounts of some

nutrients, they are physically and chemically inaccessible to plant roots. Thus gravel

limits the amount of nutrients that plants can extract from a given volume of soil,

however, the soil is still the major nutrient store, followed by the aboveground biomass

and the litter (Hingston et al. 1989). For this reason soil feedback mediated by chemical

properties is likely to have strong influence on plant biomass production. In turn, plant

species of the jarrah forest might have efficient conditioning capacities to condition soil

properties in favour of their growth.

Ecological aspects of the jarrah forest

Paradoxically the jarrah forest, one of the most limiting ecosystems for plant growth in

the world, is a diversity hotspot due to its exceptional concentration of endemic species

and high biodiversity (Myers et al. 2000). Contrary to what one might think, the

environmental conditions prevalent in the ecosystem provide considerable scope for

diversity in the floristic composition. This diversity includes a vast adaptive

characteristics and structural features of the plant communities. Jarrah (E. marginata) is

the dominant tree in the forest, with long-lived resprouting species commonly

dominating the understorey, accompanied by short-lived obligatory reseeding species

which are stimulated to germinate by high soil temperatures in a fire, and a few species

which require dispersal from adjacent unburnt sites (Bell & Heddle 1989).

20

Fire, which is the main natural disturbance in the jarrah forest, has marked influence on

the vegetation composition, eliminating intolerant species and encouraging the spread of

fire-resistant species. In adapting to this environmental pressure, current members of the

flora show a variety of adaptive traits to maintain a presence in the post-fire community.

Two major fire-response syndromes are apparent in the jarrah forest: resprouters and

reseeders. Resprouting species are those species whose post-fire presence depends

predominantly on individual survival. Individuals of reseeding species die as a result of

the fire, and their continued presence in the post-fire community depends on

establishment of a new generation of seedlings from soil-stored or canopy-stored seeds

(Dell et al. 1989; Bell 2001; Keeley et al. 2012).

In the jarrah forest, plant communities contain a mix of resprouters and reseeders,

however, fire frequency influences the proportions of resprouters and reseeders in those

communities. Short intervals between fires promote the success of resprouters, which

recover quickly after fire and have the capacity to outgrow and outcompete other plant

species. Resprouter species also tend to be resilient to even highly frequent burning and,

therefore, do not have the risk of localized extinction as do reseeders (Bell 2001; Pekin

et al. 2012).

Long intervals between fires favor the reseeding strategy. With increasing time between

fires, fuel loads tend to accumulate, and the size of individuals increases with

progressive self-thinning. Fires in such long-unburned areas would be more intense,

resulting in a greater likelihood of the deaths of individual resprouters, as well as all

reseeder individuals. Also, the size of the post-fire openings may be expected to

increase with longer intervals between fires. In this situation, nutrient release following

fire may be great, and the rapid growth rate of recruits within the relatively larger

21

clearings may mean that reseeder species will eventually outcompete slower-growing

resprouter seedlings (Bell 2001; Pekin et al. 2012).

In the jarrah forest, the variability in the post-fire environment following different fires

constitutes a selection pressure that maximizes genotypic variation, plant species

diversity is greatest shortly after fires, when the vegetation simultaneously contains both

resprouting and reseeding species. An initial post-fire increase in community species

richness and diversity followed by a gradual decline is the typical pattern in most of the

world's fire-prone, Mediterranean-climate ecosystems. The decline in species richness

and diversity is predominantly a decline in living representatives (Bell 2001; Burrows

2008).

Succession in Western Australian plant communities generally follows the initial

floristic composition model (Cattelino et al. 1979; Grant 2006; Norman et al. 2006),

with a full complement of the flora represented as aboveground living individuals at the

beginning of the fire interval and a mix of live individuals and quiescent seeds at the

end of a fire cycle. Resprouter species populations in Australian communities are multi-

aged, with cohorts of seedlings surviving only if germinated early in the interfire

interval. Populations of reseeders, on the other hand, are nearly always even- and single-

aged. The high proportion of resprouters in the southwestern Western Australian plant

communities also means that post-fire plant cover and biomass recovery is very rapid,

with a leveling off of these community attributes after only 5-10 years (Dell et al. 1898;

Bell 2001).

Evidence of potential mechanisms of soil conditioning by jarrah forest plants

The native plants in the jarrah forest have the capacity to cope with low nutrient

availability by their slow growth, leaf adaptations, and recycling of nutrients within

individual plants (Hingston et al. 1989), and also due to processes occurring

22

belowground. Native plants can efficiently condition the soil properties, especially the

most limiting P and N. Conditioning of P by jarrah forest plants includes the formation

of specialized cluster roots that solubilise P by releasing root exudates, particularly

organic acids (Lambers et al. 2008; Hopper 2009; Lambers et al. 2012). Another

strategy to condition P is via formation of mycorrhizal roots to increase the uptake

efficiency of the nutrient from soil (Abbott & Robson 1977; Brundrett & Abbott 1995;

Solaiman & Abbott 2003). Conditioning of N requires the formation of symbiotic

associations between N-fixing bacteria and legumes from the jarrah forest (Lawrie

1981).

In order to establish symbioses with AMF and N-fixing bacteria, jarrah forest plants

must release chemical signals via root exudates, which is considered a conditioning

mechanism. For example, flavonoids are released to attract symbionts to the rhizosphere

(Hirsch et al. 2003; Amalesh et al. 2011; Cesco et al. 2012). Jarrah, the dominant tree in

the jarrah forest releases exudates containing a variety of sugars, amino acids and

organic acids (Malajczuk & McComb 1977), which might contribute to condition the

physical, chemical and biological properties of soil. As a result, large populations of

bacteria, actinomycetes, saprophytic fungi and AMF inhabit the rhizosphere and roots

of jarrah trees (Malajczuk 1979; Malajczuk & McComb 1979; Soilaiman & Abbott

2003).

Allelopathy comprises another mechanism by which jarrah forest plants can condition

the properties of soil (Willis 1999). Allelochemical compounds have inhibitory effects

on reproduction of soil biotic communities and seed germination, therefore the effects

of these substances can be indirectly observed as decreased microbial communities of

soil, and reduced rate of seed germination. Allelochemicals with antimicrobial action

have been detected in exudates and litter of eucalypt species and other Australian plants

23

such as Acacias (Deans 2002). For example, A. pulchella produces volatile (2- and 3-

methylbutanol, hexanol, pentanal, 2- and 3-methylbutanal, 4-methylacetophenone and

C-disulfide) and non-volatile allelochemicals (saponines; Alexander et al. 1978;

Whitfield 1981; D’Souza et al. 2004). Acacia dealbata releases allelochemical

compounds that negatively affect other plant species and soil microbes (Lorenzo et al.

2013). Acacia melanoxylon produces several phenolics and flavonoids with phytotoxic

activity (Hussain et al. 2011). Allelopathy is critical for plant persistence in ecosystems

such as the jarrah forest, where competition with other plants and soil organisms for

moisture and nutrients is intense, thus plants equipped with efficient inhibitory

chemicals raise their chances of survival (Willis 1999; Manoharachary & Mukerji

2006).

Mining activity and restoration of the jarrah forest

The jarrah forest has been mined for bauxite (alumina) since 1963. Mining is one of the

most destructive human activities to ecosystems, as a result, the physical, chemical and

biological properties of soil can be modified. Mining activity in the jarrah forest causes

alterations to soil conditions (Ward 2000; Jasper 2007; George et al. 2010; Tibbett

2010) that in turn might exert altered soil feedback on plants. In 1966 restoration

activities of the jarrah forest post-bauxite mining started. First restoration attempts

consisted in planting of fast-growing, exotic, softwoods, mainly pines (Pinus radiata)

and cypresses (Cupressus sp.), as well as eastern Australian hardwood eucalypts such as

Sydney blue gum (Eucalyptus saligna; Koch 2007a; Koch 2007b).

At the present time restoration process includes practices to improve the physical,

chemical, and biological properties of soil to facilitate vegetation succession,

encouraging the re-establishment of the original and particular characteristics of the

ecosystem (Koch 2007a; Koch 2007b). Approximately 550 ha of native forest are

24

restored per year (Bell & Hobbs 2007), however, to date there are still 3,600 ha of

plantations of non-native species in the jarrah forest that will be eventually harvested to

restore native forest for habitat preservation and exotic weed management

(Conservation Commission of Western Australia 2012).

The current mining and restoration process generally takes about 2–3 years. The

methods of restoration have changed and improved over the years as knowledge has

increased and as larger earthmoving machinery has become available. The significant

improvements have been mainly in the areas of landscaping after mining, soil return

methods, deep ripping to relieve compaction, selection of appropriate plant species for

restoration, plant propagation methods, and techniques to encourage return of fauna

(Koch 2007b).

The complete mining and restoration process includes the following steps (Koch

2007b):

Soil stripping: Soil is stripped in two layers. The process is called double

stripping. The top 15 cm (topsoil) contains the majority of the seeds, organic

material, plant nutrients, and microbial activity. It is important to return this

material on the surface during the restoration process. The second layer is from

15 cm down to the top of the lateritic duricrust layer and ranges from 10 to 80

cm deep, averaging 40 cm. This layer is called overburden and is stockpiled next

to the mined area to be returned to the pit floor during the restoration process.

Whenever possible, topsoil is direct returned. This process involves stripping

from one area that is about to be mined and immediately returning the soil to a

nearby area that is being restored. This ensures that the biological components of

the topsoil are not degraded by being stored in a stockpile.

25

Blasting: The lateritic duricrust is part of the bauxite ore deposit and averages 1

m deep. It is broken either by blasting or by ripping with a large remote control

bulldozer, a Komatsu 575. Ripping the duricrust is quieter than blasting and is

used in locations where blasting noise may impact on neighbours.

Mining: The broken duricrust and the underlying friable bauxite are loaded into

190-tonne trucks using large hydraulic excavators. The bauxite depth varies

from 2 to 10 m and averages about 4.5 m. The bauxite is trucked to a central

crusher that feeds onto a conveyor that supplies the refinery 20–25 km away.

Approximately 20 million tones of bauxite are mined each year from two mines

at Huntly and Willowdale (Figure 6).

Landscaping and pre-ripping: Following the completion of mining, bulldozers

push down the pit walls and shape the pits to a smooth rolling landscape,

ensuring that the topography created does not allow water to run from the mined

pits into adjoining unmined areas. Large rocks are buried but some are left on

the surface to provide fauna refuges. After landscaping, the pit floor is deep

ripped at 1.6-m spacing using a bulldozer fitted with a 1.5-m-long winged tine.

This breaks up the compacted pit floor, allowing the infiltration of water through

the soil profile and facilitates the growth of plant roots.

Overburden and topsoil return: Using scrapers, first the stockpiled overburden

and then the fresh topsoil (0–15 cm), usually sourced from a pit nearby that is

being prepared for mining, are spread on the area being restored. This ‘directly

returned’ topsoil contains most of the forest seed bank, nutrients, and soil

microbial activity, allowing a more rapid return of plant species and soil

processes. Where direct return is not possible, some fresh topsoil is screened

during the dry summer period to remove much of the inert gravel (50–60% of

26

the soil), and this fine fraction with its concentrated seed content is then spread

over pits, which have received stockpiled topsoil.

Return of logs and rocks for fauna habitat: Waste timber and rocks are placed in

the restored areas to provide habitat for both vertebrate and invertebrate fauna.

Without this activity, it would take hundreds of years for this type of habitat to

develop. Logs and rocks are used as shelter for foraging mammals, reptiles, and

a wide range of insects, spiders, and other invertebrates.

Contour ripping: The pre-ripping that is carried out after landscaping relieves

mining-related compaction and creates a friable rooting zone to 1.5 m depth.

Following topsoil and overburden return, the area is then ripped again to a depth

of 0.8 m on contour by a bulldozer using three tines. This final contour ripping

removes compaction caused by the soil return operation and produces contour

furrows, which allow water infiltration and provide rainfall erosion protection

until the vegetation develops

Seeding: A mixture of 78–113 plant species, including tree and understory

plants, is represented in the broadcast seed mix. Seed is collected within a

defined zone up to about 20 km from each mine to ensure that local genetic

material is used for restoration. The seeds are applied at a rate of about 1 kg/ha

via a purpose built air seeder attached to the ripping bulldozer. Seeds are sown

in the dry summer and autumn months. Appropriate germination treatments

(such as hot water or smoke) are applied to each species as required before

sowing.

Planting: Plant species that do not return from the soil seed bank or from the

broadcast seed are called ‘recalcitrant’ and are grown from cuttings, from scarce

seed quantities, or by tissue culture. Propagation is carried out at Alcoa’s

nursery, and the plants are hand planted during winter. Up to 28 species are

27

propagated. Up to 200,000 plants are grown and planted each year. Many of

these are dryland rushes and sedges that produce little viable seed. The grass-

like rushes and sedges are heavily grazed by kangaroos and need to be protected

while they are young. Small mesh bags are used to provide grazing protection.

Fertilizing: Diammonium phosphate fertilizer with potassium and micronutrients

is applied at 280 kg/ha by helicopter to newly restored areas in late winter each

year. This is a once-off application. The helicopter is used to prevent

recompaction of the ripped ground and damage to the contour furrows, which

protect against erosion. This is the last stage of the restoration operation.

Figure 6. Alcoa World Alumina Australia mining lease in the Darling Range of south-western Western Australia and the location of the bauxite mines at Jarrahdale, Huntly, and Willowdale. Reproduced from Standish et al. (2010). Soils for the experiment were collected from these three bauxite mines.

As a result of the mining and restoration process, restored soils tend to have reduced

biological activity, pH, and C, and increased P, N, K and gravel percentage (Ward 2000;

Jasper 2007; George et al. 2010; Tibbett 2010). However, many soil parameters showed

28

significant changes in the period 3.5–8.5 years after restoration. Levels of total nitrogen

in the top 10 cm of soil in restored areas increased from around 0.04–0.05% initially to

levels approaching those found in good-quality jarrah forest (0.10–0.30%) after 8.5

years. Soil pH declines in the surface layer after restoration. The rate of acidification

decreases in time, and the levels of other soil chemical parameters move towards the

range of values found in areas typically mined for bauxite (Ward 2000). Twenty-six

years post-restoration, organic matter quantity recovers to a level comparable to that of

unmined forest soil, soil N recovers faster than soil C, and recovery of microbial and

soluble organic C and N fractions is faster than total soil C and N (Banning et al. 2008;

George et al. 2010; Lin et al. 2011). Arbuscular mycorrhizal fungi recover to pre-

mining levels in bauxite restoration in five years. Ectomycorrhizal fungi are poorly

adapted to disturbance; however, they are able to reinvade through wind-blown spores.

The density of ectomycorrhizal fungi has been found to be equivalent in seven-year-old

restoration and adjacent forest, but both abundance and diversity are correlated with

development of a litter layer. Fortunately, rhizobia are known to be tolerant of soil

disturbance, and failure of N-fixation by legumes has not been reported in restoration

(Jasper 2007).

Despite a vast body of studies on restoration of jarrah forest after bauxite mining have

documented a recovery of properties of aboveground and belowground compartments of

the ecosystem, we lack information to predict how modified properties of soil will

affect the feedbacks between both compartments (i.e. PSF) since these studies have

been traditionally done on independent compartments of the ecosystem.

29

Thesis outline and research objectives

The overall aim of the studies described in this thesis was to determine the influence of

plant species (via conditioning process) and soil conditions (via soil feedback) on PSF

in the jarrah forest. Soil conditions comprise those of unmined soil and restored soil

post-bauxite mining. More specifically the research objectives were:

1. Determine the influence of plant species and soil context on PSF by examining

species-specific conditioning of soil properties by native plant species (i.e. Eucalyptus

marginata, Acacia pulchella, and Bossiaea ornata) in unmined and restored soils, as

well as individual PSF associated with these combinations of plants species and soil

contexts (Chapter 3).

2. Examine the influence of soil conditions on PSF by investigating the soil feedback

exerted by soils from a chronosequence of restoration ages (i.e. 1.5-, 7-, 22-year-old

soils post-bauxite mining, and an unmined soil) on individual PSF associated with B.

ornata plants (Chapter 4).

3. Investigate the influence of soil conditions caused by long-term conditioning (i.e. 35

years) of non-native (Pinus radiata and Eucalyptus saligna) and native species (E.

marginata) on individual PSF associated with E. marginata seedlings (Chapter 5).

30

CHAPTER 2

General methods

This chapter outlines the methods common to all experimental chapters only. Methods

specific to individual chapters, including experimental design and data analysis, are

described on a chapter by chapter basis.

Response variables

Soil chemical properties

Total N (%), total C (%), available (bicarbonate extractable) P (mg kg-1) and K

(mg kg-1), pH (CaCl2) and EC (dS m-1) were determined by the commercial laboratory

Cummings Smith British Petroleum (CSBP, Bibra Lake Western Australia). Total N

and total C were measured in a combustion analyser (LECO® USA), available P,

available K, pH, EC, were determined using standard methods described in Rayment &

Higginson (1992). Ammonium (NH4+ mg kg-1) and nitrate (NO3

- mg kg-1) analyses were

done simultaneously by the Skalar Autoanalyser (Skalar Analytical, The Netherlands) in

laboratories at the University of Western Australia.

Water content and water holding capacity of soils

To measure the water content of soil, the gravimetric method with oven drying was

used. This method consists in weighing a moist soil sample, which is then oven dried at

105 °C for 24 h and reweighed (Topp et al. 2008). To determine the water holding

capacity of soils, a known amount of soil was saturated with water in a cylinder, which

was then placed on an absorbent membrane until the excess water is drawn away by

gravity overnight. The water holding capacity was calculated based on the weight of the

water held in the sample vs. the sample dry weight (Rayment & Higginson 1992).

31

Soil biology and associated properties

Microbial biomass-carbon (C)

Microbial biomass-C was determined by chloroform-fumigation extraction technique

described in Coleman et al. (2004). Organic C in fumigated and unfumigated samples

was measured with a Total Organic-C Analyser (Shimadzu 5000A, Japan). Results from

unfumigated samples were subtracted from fumigated ones in order to obtain the

microbial biomass-C values. For calculations a kEC factor of 0.45 was used and data

were reported on a dry soil basis (µg C g-1 dry soil).

Phospholipid–fatty acid (PLFA) profiles

Phospholipid–fatty acid profiles reveal information about the community structure and relative

abundance of bacteria (Gram positive, Gram negative, actinomycetes), fungi (filamentous and

mycorrhizal), protozoa, other eukaryotic organisms including algae in soils (Cavigelli et al.

1995, Ranganayaki et al. 2001). Phospholipid-fatty acid profiles of soils were determined

as follows: total lipids were extracted from 8 g of freeze-dried soil using

chloroform:methanol:phosphate buffer at a ratio of 1:2:0.8 following the procedure

described in White et al. (1979). Then, solid phase extraction tubes (silica-bonded polar,

1 mL volume, Supelclean™) were used to separate phospholipids from neutral lipids

and glycolipids. Phospholipids were subjected to a mild alkaline hydrolysis to form

fatty acid methyl esters (FAMES) as described by Zelles & Bai (1993). Individual

PLFAs were detected using gas chromatograph with flame ionization detector (Agilent

7890A Dean Switch GCFID). Fifty μL (75μg mL-1) of methyl-nonadecanoate (Me-C19

Sigma) were added to FAMES extracts as internal standard. PLFAs were identified by

comparing their retention times with those of qualitative standards of methyl esters

(Supelco®). To calculate concentration of each FAME, the following formula was used:

32

Where PFAME and PISTD are the peak areas of fatty acid and the internal standard

respectively; μgStd is the concentration of the internal standard (μg μL-1 solvent); and W

is the dry weight of soil used (g). Finally, biomarkers for specific groups of soil

organisms such as bacteria (15:0i, 15:0a, 16:0i, 17:0i, 17:0cy), fungi (18:2ω6,9,

18:1ω9), and microeucaryotes (22:0, 24:0, 25:0; Zelles 1999; Treonis et al. 2004;

Bezemer et al. 2006) were determined.

Community level physiological profile (CLPP) of soils

To determine CLPPs of the soils the MicroResp™ methodology was used (Campbell et

al. 2003; Lalor et al. 2007). CLPP measures the functional diversity and activity of soil

microbes involved in the carbon cycle and is usually assessed by carbon substrate utilization.

The C subtrates used during the experiment were selected based on their known

occurrence in jarrah forest ecosystems, as substances produced during the thermal

oxidation of organic matter, or known plant root exudates (Banning et al. 2012). Types

of C substrates, as well as concentrations of substrates and reagents in cresol red

detector plates were the same used by Lalor et al. (2007; see Appendix 1). After a 6

hour incubation period at 25 °C, absorbance of detector plates (at 590 nm) was

determined using a microplate reader (ASYS Hitech, EXPERT 96) and headspace

%CO2 was calculated with the calibration curve equation experimentally obtained

utilizing the plate reader mentioned above: 0.6(exp(-9.95(x-0.42))+0.06. Finally the

CO2 rate (μg CO2–C g-1 soil h-1) was calculated as follows:

h

dwtsoilsoilfwt

Tvol

100%

)273(273

4412

2244

100CO2%

WPgPsoilgg

ISTD

StdFAMEacidfattyindividual

1

33

Where %CO2 is the absorbance value interpolated in the calibration curve; 44, 22, 12

and 273 are constants; T is the incubation temperature (°C); vol is the headspace volume

in the well (μl); soilfwt is the fresh weight of soil per well (g); h is the incubation time in

hours and soil%dwt is soil sample %dry weight.

Nematode trophic community

Nematodes are found in most marine, freshwater and terrestrial ecosystems. In soils, nematodes

are represented within several trophic levels of the soil food web (Yeates et al. 2009).

Nematodes can be directly influenced by plants (i.e. plant parasites), or indirectly by microbes

associated to roots (i.e. bacterial and fungal-feeders). As a result, plant conditioning is likely to

be reflected in nematode trophic community. Nematodes were extracted from 100 g of soil

(per sample) using a mistifier, where a fine mist of water at 25 ˚C is sprayed over the

soil sample for ten seconds at ten-minute intervals over a four-day period (Hooper et al.

2005). Aqueous extracts were collected and nematodes within them were sorted per1

mL extract using a counting cell (Sedgewick Rafter cell, Graticules England) observed

at 10x magnification in a compound microscope (Olympus Vanox). Nematodes of the

following trophic groups were identified and counted: bacterial feeders (BF; Rhabditida,

Enoplida), fungal feeders (FF; Tylenchida), omnivorous (Dorylaimida), predators

(Mononchida) and plant parasites (Tylenchida, Dorylaimida). Groups were assigned

based on morphological characteristics of the cephalic region, and esophageal structures

(Yeates et al. 1993). Data were expressed as the number of nematodes of each group per

100 g of dry soil. Additionally, the nematode channel ratio (FF/FF+BF) was calculated

as an indicator of the decomposition channel (bacterial or fungal) working in soils

(Yeates 2003).

34

CHAPTER 3

Soil conditioning and plant-soil feedbacks associated with jarrah forest plants

are soil-context dependent

INTRODUCTION

There is increasing interest in the restoration of ecosystems where soil has been

modified by anthropogenic activities (Harris 2009). Understanding how plant-soil

feedback (PSF) operates in soil can be a critical aspect of ecosystem restoration (van der

Putten et al. 2013). Therefore, it is timely to consider whether PSF operating in human-

modified soils function in the same way as in non-modified soils. An important

component of PSF is the conditioning of soil by plants, which is defined as changes of

soil properties that are driven in a plant species-specific manner (Harrison & Bardgett

2010; Pernilla et al. 2010; van de Voorde et al. 2011). Soil conditioning is driven by

root and litter deposition (Kulmatiski et al. 2008). Rhizodeposition and exudates can

modify the physical, chemical and biological properties of soil (Marschner 1995;

Neumann & Römheld 1999; Uren 2001; Bertin et al. 2003; Neumann 2007; Pang et al.

2010), and vary with plant species, cultivars, and plant developmental stage. As well,

rhizodeposition and exudates vary with environmental stress such as mechanical

impedance, anaerobiosis, drought, presence of pollutants, and nutrient deficiency

(Marschner 1995; Uren 2001; Nautiyal et al. 2008; Martin et al. 2014). Conditioning

might play a fundamental role in the restoration of ecosystems that have been modified

by anthropogenic activities because plant species that are capable of conditioning

modified soils may increase their chances of establishment and growth (Hobbie 1992;

Eviner 2004; Jordan et al. 2008).

35

In this chapter, I investigate the influence of plant species and soil context (i.e. unmined

and restored) on individual PSF associated with representative plant species of the

jarrah forest (i.e. Eucalyptus marginata, Acacia pulchella, and Bossiaea ornata). Native

plants have developed a variety of nutrient-acquisition strategies to allow them to

establish and persist in the impoverished soils present in the jarrah forest (Lambers et al.

2007; Hopper 2009; Sardans & Peñuelas 2013). This suggests the existence of strong

links between plants and soil, and the capability of native plants of conditioning the

physical, chemical and biological properties of soils (Abbott & Robson 1977;

Malajczuk & McComb 1977; Malajczuk 1979; Malajczuk & McComb 1979; Grierson

& Adams 2000; Lambers et al. 2007; Bleby et al. 2009; Hopper 2009).

Mining for bauxite occurs in the jarrah forest. After a mine closes, jarrah forest is

restored (Koch 2007b), however the properties of soils under restored forest (hereafter

restored soils) are altered compared with those of unmined soils. Specifically, restored

soils tend to have reduced biological activity, pH, and C, and increased P, N, K and

gravel percentage (Ward 2000; Jasper 2007; George et al. 2010; Tibbett 2010). It is

possible that soil conditioning by native plants is modified in restored soils compared

with unmined soils, which may ultimately affect the patterns of biomass production of

the restored plant community.

In the jarrah forest, plant community succession follows an initial floristic composition

model (Cattelino et al. 1979; Grant 2006). This model describes how the initial plant

community evolves only as changes in the abundance of the species present in the

landscape (Hobbs & Legg 1983; Norman et al. 2006) in response to disturbances such

as fire (Bell & Koch 1980). Therefore, it is likely that jarrah forest plants species

condition soil differently so as to secure their persistence in the community. This could

happen because the species that establish early in succession are those that are still

36

present later. The aims of this work were: 1) to examine in a detailed manner the soil

conditioning process of representative native jarrah forest plants species, and individual

PSF (i.e. magnitude and direction) associated with these plant species. 2) To determine

how conditioning process and individual PSF associated with the same plant species

differed in restored soil post-bauxite mining as compared with unmined soil.

The plant species used for this experiment were Eucalyptus marginata (i.e. long-lived

lignotuber-regenerating species), Acacia pulchella (i.e. short-lived, fire-induced species,

and Bossiaea ornata (i.e. long-lived, resprouter species). Specifically, it was

hypothesised that E. marginata, A. pulchella, and B. ornata would condition the

chemical and biological properties of unmined soil in a species-specific manner. The

second hypothesis was that modification of soils by bauxite mining and restoration

practice would differentially influence the patterns of conditioning of the plant species,

because conditioning process would differ in restored soil as compared to unmined soil.

The third hypothesis was that biomass production of jarrah forest plants and

consequently individual PSF would be influenced by plant species and soil conditions.

Positive PSF would be associated with long-lived B. ornata grown in unmined soil,

whereas negative PSF would be associated with short-lived A. pulchella in the same soil

type. However, individual PSF associated with the plant species might change in

magnitude and direction when grown in restored, compared with unmined soil.

MATERIALS AND METHODS

Sampling and experimental design

To test the hypotheses a pot experiment with two phases of plant growth (Figure 1) was

designed. Two different soil types were used, soil collected from jarrah forest with no

history of mining activity (i.e. unmined soils; 116°5' E 32°36' S) and a restored soil

collected from a site that had been cleared, mined and then restored to jarrah forest by

37

Alcoa World Alumina Australia in 2008 (i.e. restored soils; 153°21' E 27°13' S) using

practices described by Koch (2007b). The soils were collected in November 2009. Both

unmined and restored soils were gravelly sands with the following characteristics:

unmined soil presented a density of 1.05 g cm-3, 36.92% of gravel, soil fraction < 2 mm

was 89.9% sand, 5.62% silt, 5.09% clay, 0.08% and 3.05% of total N and total C

respectively, 60 and 4 mg kg-1 of available K and P respectively, pH (CaCl2) of 5.1 and

EC of 34.67 dS m-1. Restored soil presented a density of 1.3 g cm-3, soil fraction < 2

mm was 60% of gravel, 88.39% sand, 5.01% silt, 6.6% clay, 0.07% and 3.15% of total

N and total C respectively, 35 and 22 mg kg-1 of available K and P respectively, pH

(CaCl2) of 4.88 and EC of 23.3 dS m-1.

All soils were collected adjacent to roots to a depth of 15 cm (0.8 – 1.0 kg per plant).

Restored soils were sampled from beneath the plant species E. marginata (Myrtaceae),

A. pulchella (Fabaceae) and B. ornata (Fabaceae), whereas unmined soil was sampled

from the species mentioned above plus the legumes Acacia lateviticola, Gompholobium

polymorphum Bossiaea aquifolium, Labichea punctata, Hovea trisperma and Hovea

chorizemifolia since it was not possible to locate enough specimens of the three species

that were sampled at the restored site. Approximately 80 plants were sampled at each

site. All samples of unmined soils were thoroughly mixed and distributed into 48

bottom-sealed pots (1.2 L; diameter = 13.3 cm, height = 12.4 cm). This procedure was

repeated using restored soils. There were 96 pots in total and these were harvested in

two batches of 48 pots each after phases one and two respectively as described below.

The plant species grown in phase one were E. marginata A. pulchella and B. ornata.

These species were selected based on their characteristics and previous observation of

their growth behaviour in the field. Eucalyptus marginata (jarrah) is a lignotuber-

regenerating, long-lived tree (i.e. 300 years) that dominates the canopy of the forest

38

(Abbott 1984). Acacia pulchella is a fire-induced species whose population peaks after

forest burning but declines four years later (Monk et al. 1981). Bossiaea ornata is a

resprouter, long-lived bush (Bell 2001). Eucalyptus marginata is common in both

restored and unmined soils (Bell 2001), A. pulchella can be abundant in newly restored

sites but not as common in unmined jarrah forest (Monk et al. 1981), and B. ornata can

be common in both unmined and early restored mined sites but is not as prolific under

fertilisation practice typical in restoration (Bell & Heddle 1989, George et al. 2006).

The glasshouse experiment ran from November 2009 to August 2010. The 48 pots per

soil type were planted with seedlings as follows: 18 with E. marginata, 12 with A.

pulchella, 12 with B. ornata and 6 were left unplanted as controls. One seedling was

planted into each pot and dead seedlings were replaced. Within two weeks all pots had a

growing plant. To produce seedlings, seeds of A. pulchella and B. ornata were boiled

for 30 seconds to induce germination. Afterward, seeds of the three species were

surface-disinfected with ethanol (70%) and bleach (50%) to prevent pathogen growth

and germinated at 14 °C in darkness.

After phase one (4.5 months) plants from the first set of 48 pots were harvested (six

replicates × two soil types (unmined and restored soils) × three plant species (E.

marginata, A. pulchella and B. ornata) + unplanted controls) and potted soils were

sampled. After the growing phase the root system of the plants had spread throughout

the soil within the pots, and so it was assumed that a considerable amount of soil was to

some extent under the influence of roots, and thus all the potted soil was used for the

analyses. Soil from each pot was sieved (mesh = 2 mm) and stored at 4 °C pending

analyses. A soil subsample from each pot was frozen (-20 °C) and then subsequently

freeze-dried pending phospholipid-fatty acid analysis (PLFA).

39

The remaining set of 48 pots was used for phase two, shoots from plants growing in this

set of pots were removed leaving the soil intact and new seedlings of A. pulchella and

B. ornata were planted in soil previously conditioned by the same plant (home soil) or

in soil previously conditioned by E. marginata (away soil). Soil conditioned by E.

marginata was selected as ‘away soil’ because this is the dominant tree in the jarrah

forest, either in unmined or restored forest, therefore it is likely that a considerable

amount of soil in the forest has been influenced (conditioned) by this species.

Consequently, in the field other native species might grow in soil that was influenced by

jarrah at some stage. The design for phase two included six replicates × two soil types

(unmined and restored soils) × two plant species (A. pulchella and B. ornata) × two

conditioning types (home and away soils; Figure 1). Phase two ran for 4.5 months, after

which the plants were harvested; shoots and roots were oven-dried for one week at 60

°C and weighed.

Throughout the experiment pots were kept in two root cooling tanks filled with water

set to a constant temperature of 20 °C in order to maintain the roots and soil under

homogeneous conditions. Root cooling tanks were housed in a glasshouse for the

duration of the experiment. Temperatures inside the glasshouse ranged between 13 and

31 °C during phase one, and between 6 and 26 °C during phase two. Pots were rotated

between and within tanks every two weeks to negate a block effect and watered once a

week to 60% water-holding capacity.

40

Figure 1. The experiment had two phases (4.5 months each): phase one where the plants Eucalyptus marginata, Acacia pulchella, and Bossiaea ornata were planted into pots containing either unmined soils or soils from beneath jarrah forest restored after bauxite mining; and, phase two where only A. pulchella and B. ornata were planted in pots previously conditioned by the same plants (home soil) or by E. marginata (away soil). Unplanted pots were included as controls in phase one. After phase one soil properties were measured to estimate the effects of soil conditioning by plants (A). Plant biomass in home and away soils was measured after phase two to determine the magnitude and direction of individual PSF, in both unmined soils and restored soils (B).

Response variables

Refer to chapter two for methodologies used to determine response variables measured.

Statistical analysis

As evidence of soil conditioning (Figure 1A) differences in the properties of soils grown

with the experimental plants with respect to the properties of the unplanted controls

were analysed. To do this, permutational analyses of variance (PERMANOVA) of the

sets of chemical and biological properties, and of individual properties were performed.

Although PERMANOVA is a multivariate method, it can be used for univariate

analyses; to perform PERMANOVA using one response variable and Euclidean

distance yields Fisher’s traditional univariate F-statistic (Anderson 2010). The analyses

were performed independently for unmined and restored soils.

41

As well, differences in the patterns of conditioning of soil properties among the

experimental plants were determined. To do this we used PERMANOVA followed by

pair-wise tests on the chemical and biological soil properties. The total set of PLFAs

detected, the respiration rates of C-substrates from CLPPs and the total trophic groups

of nematodes were independently analysed using PERMANOVA. To visualize the

results of PERMANOVAs, scaling plots were built using principal co-ordinate analysis

(PCO; Anderson 2001). To build PCOs, variables were transformed when necessary and

normalized, and Euclidean distances were used to calculate the resemblance matrices.

Additionally, a vector showing the properties causing the major effects on the

separation of clusters along the main PCOs (i.e. PCO1 and PCO2) was included in each

plot. The vector shows the properties whose Pearson correlation coefficient with the

main PCOs was > 0.8.

To determine individual PSF (i.e. magnitude and direction; Figure 1B), univariate

PERMANOVA with plant biomass as the response variable was used. A significant

increase of the plant biomass production in home with respect to away soil was

considered a positive PSF. A significant decrease of the plant biomass production in

home soil with respect to away soil was considered a negative PSF. No difference in

biomass production in home with respect to away soil was considered a neutral PSF.

For all the statistical analyses PRIMER v6 +PERMANOVA (PRIMER-E Ltd) was

used. For the multivariate analyses, permutation of residuals under a reduced model

was selected as permutation method as well as type III (partial) sum of squares. The

unrestricted permutation of raw data as permutation method and type III (partial) sum

of squares were used for the univariate analyses. In all cases data were transformed

when necessary and Euclidean distance was used to build resemblance matrices, except

for the analyses of the trophic groups of the nematode community and CLPPs when

42

Bray-Curtis coefficient and Manhattan distance were respectively utilized to build the

similarity matrices.

RESULTS

Changes in properties of soils conditioned during phase one

Unmined soil

The chemical properties of unmined soils were conditioned by E. marginata, A.

pulchella, and B. ornata compared with the chemical properties of the unplanted

controls (t = 2.94, P = 0.002; t = 1.79, P = 0.015; t = 3.86, P = 0.003 for E. marginata,

A. pulchella, and B. ornata, respectively; Figure 2a). Also, during phase one the

chemical properties of soils grown with A. pulchella differed from the chemical

properties of soils conditioned by E. marginata (t = 2.81, P = 0.002) and B. ornata (t =

3.57, P = 0.003; Figure 2a). Eucalyptus marginata and B. ornata similarly changed the

chemical properties of soils (i.e. chemical soil properties were not significantly

different; Figure 2a). Statistical analyses of individual properties showed that E.

marginata modified the C:N ratio (t = 3.56, P = 0.002), available P (t = 17.81, P =

0.005), EC (t = 3.63, P = 0.005), pH (t = 9.48, P = 0.004) and total C (t = 4.36, P =

0.006), these properties differed from those of controls (Table 1). Acacia pulchella

altered the C:N ratio (t = 3.29, P = 0.011) and total N (t = 3.41, P = 0.022), whereas B.

ornata changed the C:N ratio (t = 4.45, P = 0.004), available P (t = 19.4, P = 0.001), EC

(t = 4.5, P = 0.009), pH (t = 9.48, P = 0.004), and total C (t = 5.05, P = 0.003; Table 1).

Significant decreases in concentrations of NH4+ in soils grown with E. marginata and B.

ornata were interpreted as an effect of plant uptake (Table 1).

The biological properties of soils planted with E. marginata, A. pulchella, and B. ornata

did not differ from the biological properties of the unplanted controls, thus there was no

43

evidence for biological conditioning of soils (Figure 2b). However, statistical analyses

of individual properties revealed that E. marginata positively affected the numbers of

plant-parasitic nematodes compared with control soils (t = 3.23, P = 0.015; Table 1).

Total PLFAs (i.e. sixty different PLFAs, 10% unknown; F3, 20.5 = 0.79, P = 0.83),

biomarkers for bacteria (F3, 1.73 = 0.80, P = 0.75), fungi (F3, 2.36 = 0.91, P = 0.53),

and microeucaryotes (F3, 1.46 = 0.49, P = 0.78), and CLPPs (F3, 3.02 = 0.53, P = 0.82)

were all similar among planted soils and unplanted controls (i.e. not significantly

different).

Figure 2. Scaling plots from principal co-ordinate analysis of a) chemical properties: total C and total N (%), C:N ratio, available P and K (mg kg-1), EC (dS m-1), pH (CaCl2), NO3

- and NH4

+ (mg kg-1), and; b) biological properties: microbial biomass-C, bacterial-, fungal-feeding, plant-parasitic, predatory, omnivorous nematodes (all per 100 g-1 soil), and nematode channel ratio of unmined soil conditioned by Eucalyptus marginata, Acacia pulchella, and Bossiaea ornata after phase one, unplanted controls are included. Vector showing the main properties driving the separation of clusters along PCOs are overlayed on plots.

44

Table 1. Values for properties of unmined soils (mean ± SE; n = 6) after conditioning by three plant species (Eucalyptus marginata, Acacia pulchella, and Bossiaea ornata) compared with the unconditioned control soils. Only the properties that differed in conditioned soils with respect to controls are shown in the table. Values in bold font indicate significant differences between conditioned and unconditioned soils (P ≤ 0.02)

Property A. pulchella E. marginata B. ornata Control

C:N 28.99 (1) 41.15 (2.2) 39.92 (1.3) 33.23 (0.85)

Available P (mg kg-1) 4.00 (0.26) 15.67 (0.61) 16.17 (0.54) 3.67 (0.33)

EC (dS m-1) 0.035 (0.002) 0.03 (0.003) 0.03 (0.002) 0.04 (0.002)

pH 5.43 (0.02) 5.15 (0.02) 5.15 (0.02) 5.45 (0.02)

Total N (%) 0.10 (0.004) 0.08 (0.005) 0.08 (0.002) 0.08 (0.003)

Total C (%) 2.76 (0.18) 3.25 (0.14) 3.26 (0.11) 2.54 (0.09)

NH4+ (mg kg-1) 5.52 (1.14) 3.57 (1.38) 3.75 (0.88) 4.99 (0.84)

Plant-parasitic nematodes (per 100 g-1 soil)

0 39 (12) 0 4 (3)

Restored soil

The chemical properties of restored soil were modified by E. marginata and B. ornata

only, and differed from the chemical properties of controls (t = 2.63, P = 0.002; t = 3.04,

P = 0.002 for E. marginata and B. ornata respectively; Figure 3a). In contrast, the

chemical properties of soil grown with A. pulchella did not differ from controls (t =

1.36, P = 0.089). Also, the chemical properties of soil grown with the three plant

species were significantly different to each other indicating a level of interspecific

conditioning (E. marginata versus A. pulchella t = 2.85, P = 0.002; E. marginata versus

B. ornata t = 2.22, P = 0.0017; A. pulchella versus B. ornata t = 3.53, P = 0.0027;

Figure 3a). Statistical analyses of individual properties showed that A. pulchella did not

change any chemical property significantly. Eucalyptus marginata modified available P

(t = 22.6, P = 0.002), and pH (t = 6.99, P = 0.004), these properties differed from those

45

of controls. Bossiaea ornata altered available P (t = 28.36, P = 0.004), and pH (t = 12.8,

P = 0.002; Table 2). Decreased concentrations of available K in soil conditioned by B.

ornata were considered a plant uptake effect (Table 2).

The biological properties of restored soil grown with each of the three plant species did

not differ from the biological properties of controls. Also, the biological properties of

soils grown with the three plant species were all similar to each other (i.e. no significant

differences; Figure 3b). However, analyses of individual properties indicated that E.

marginata positively affected the number of fungal-feeding nematodes (t = 2.97, P =

0.01) compared with the unplanted controls (Table 2). Analyses of the total PLFAs (F3,

23.2 = 0.94, P = 0.54), biomarkers for bacteria (F3, 5.15 = 0.82, P = 0.75), fungi (F3, 2.35 =

1.28, P = 0.26), and microeucaryotes (F3, 2.78 = 1.39, P = 0.24) detected in the soils

revealed no statistical differences between planted soils and unplanted controls. The

same results were found for CLPPs that were also similar among planted soils and

controls (F3, 2.75 = 0.44, P = 0.89).

Individual PSF

Individual PSF associated with the two plant species differed in both magnitude and

direction according to soil type. In unmined soils, A. pulchella and B. ornata produced

similar amounts of biomass in home and away soils (F1, 0.01 = 2.66, P = 0.139 and F1, 0.04

= 0.43, P = 0.535 for A. pulchella and B. ornata respectively; Figure 4), thus a neutral

PSF was observed in both cases. In restored soils, both A. pulchella and B. ornata

produced less biomass in home compared with away soils (F1, 1.4 = 13.314, P = 0.004;

F1, 5.65 = 10.1, P = 0.01 for A. pulchella and B. ornata respectively; Figure 4) so a

negative PSF was observed. Also, A. pulchella produced up to three times more

biomass in restored soil than that produced in unmined soil.

46

Figure 3. Scaling plots from principal co-ordinate analysis of a) chemical properties: total C and total N (%), C:N ratio, available P and K (mg kg-1), EC (dS m-1), pH (CaCl2), NO3

- and NH4

+ (mg kg-1), and; b) biological properties: microbial biomass-C, bacterial-, fungal-feeding, plant-parasitic, predatory, and omnivorous nematodes (all per 100 g-1 soil), and nematode channel ratio of restored soils conditioned by Eucalyptus marginata, Acacia pulchella, and Bossiaea ornata, unplanted controls are included. Vector showing the main properties driving the separation of clusters along PCOs are overlayed on plots.

Table 2. Values for properties of restored soils (mean ± SE; n = 6) that were statistically different between soils conditioned by three plant species (as for Table 1) and unconditioned controls. Significant pair-wise comparisons are indicated in bold (P < 0.01)

Property A. pulchella E. marginata B. ornata Control

Available P (mg kg-1) 3.00 (0.26) 16.00 (0.68) 15.00 (0.45) 3.17 (0.17)

Available K (mg kg-1) 36.00 (2.1) 36.83 (1.25) 21.83 (1.11) 33.67 (1.71)

pH 5.37 (0.02) 5.15 (0.03) 5.12 (0.02) 5.42 (0.02)

Fungal-feeding nematodes

(in 100 g-1 soil)

530 (163) 869 (209) 209 (37) 307 (75)

47

Figure 4. Total (shoot + root) biomass (mean ± SE; n = 6) produced by the plant species Acacia pulchella (ACA) and Bossiaea ornata (BOS) grown in unmined and restored soils conditioned by the same plant species (Ho = home soil) or by Eucalyptus marginata (Aw = away soil). Pairs of means were compared in home vs. away for each plant species and soil type. Means with the same letters were not significantly different (P < 0.05).

DISCUSSION

This study provides empirical data describing the conditioning process and individual

PSF associated with representative plants of the jarrah forest, and how modification of

soil by mining and restoration cause alteration of the conditioning process and

individual PSF. Conditioning of unmined soil by long-lived plants E. marginata and B.

ornata was similar to each other but different from that of short-lived A. pulchella.

Especially relevant was the conditioning (increase) in available P by E. marginata and

B. ornata, which might contribute to their persistence in the forest given the low content

of P in soil. In contrast, lack of soil P conditioning by A. pulchella might contribute to

its being short-lived. Patterns of soil conditioning were altered when plants grew in

restored soil, for example the altered C:N ratios. None of the experimental plants

conditioned the microbiological properties of unmined or restored soils, which

aa a a

a

b

ab

0

0.2

0.4

0.6

0.8

1

1.2

1.4

1.6

1.8

2

ACA-Ho

ACA-Aw

BOS-Ho

BOS-Aw

ACA-Ho

ACA-Aw

BOS-Ho

BOS-Aw

Unmined soil Restored soil

Tota

l bio

mas

s (g

)

48

contradicts the general finding of positive conditioning of microbial communities in the

root affected soil. Other groups of the soil biota such as the nematodes were little

affected by plant conditioning. The results indicated that individual PSF associated with

both, A. pulchella and B. ornata was affected by soil context, a neutral PSF was

observed in restored soils and a negative PSF was produced in unmined soil.

It was predicted that chemical as well as biological properties of unmined soil would be

differentially conditioned by the three plant species, however in unmined soil, species-

specific conditioning primarily occurred through changes in the chemical properties.

Eucalyptus marginata and B. ornata, both long-lived species similarly conditioned the

chemical properties of unmined soils, which differed from conditioning by short-lived

A. pulchella. The total C concentration of soil increased under the influence of the first

two plant species, likely as a result of root deposition (Marschner 1995; Brimecombe et

al. 2001; Neumann & Römheld 2001). Acacia pulchella did not affect the total C

concentration of soil, instead this species promoted an increase in the total N

concentration, which can be attributed to exudation of N-compounds such as amino

acids (Uren 2001; Bertin et al. 2003). As a result, the C:N ratios of conditioned

(unmined) soils differed from those of the unconditioned controls. Eucalyptus

marginata and B. ornata caused an increase in the C:N ratio, whereas A. pulchella

caused a decrease in the C:N ratio. The C:N ratio controls microbial-mediated soil

processes such as the decomposition of the organic matter (Adl 2003; Lavelle & Spain

2003; Coleman et al. 2004). According to the results, A. pulchella might be enhancing a

faster turnover of organic matter (i.e. by decreasing the C:N ratio), whereas E.

marginata and B. ornata may cause slower decomposition of organic matter (i.e. by

increasing the C:N ratio). Thus, the species-specific effects on C:N ratios may, in turn,

cause additional impacts on soil processes in the longer-term.

49

Root exudates play an important role in P mobilization in these P deficient (unmined)

soils and specifically, the organic acids that are released by roots to solubilise mineral

bound P, which is often accompanied by a decrease in soil pH (Jones 1998; Neumann &

Römheld 1999; Dakora & Phillips 2002; Carvalhais et al. 2011). In this study, E.

marginata and B. ornata plants appeared to actively mobilize P in unmined soil, with a

four-fold increase in the available P fraction as a result of soil conditioning by these two

species. Mobilization was accompanied by a decrease in soil pH, which suggests that

carboxylic acids or acid phosphatase enzymes could be playing an important role in this

process. Unmined soil conditioned by A. pulchella presented available P concentrations

and pH that did not differ from those of controls. Therefore mobilisation of P by this

plant might not be a critical component of the conditioning process as it appeared to be

for the other two plant species. Differential conditioning of P that is the main limiting

soil mineral, might contribute to explain the persistence of the three plant species in the

forest. Eucalyptus marginata and B. ornata are long-lived species (Abbott 1984; Bell

2001), which can be attributed to their efficient capacity to condition P. Acacia

pulchella is a short-lived species (Monk et al. 1981), a trait that could be explained by

its poor capacity to condition soil P.

There was limited evidence of conditioning of soil biological properties by the plant

species. Conditioning of the microbial community (i.e. bacteria, fungi) was not

observed, and these results contradicted the general finding of positive conditioning of

soil microbes that is mediated by root exudation (Reynolds et al. 2003; Kardol et al.

2007; Bever et al. 2010; Bever et al. 2012; Miki 2012). It is likely that plant species

used in the experiment require longer time spans to condition their own microbial

community, as previously reported for other plant species (Kulmatiski & Beard 2011).

Of the biological properties that I measured, plant-parasitic nematodes was the only

property of unmined soil to be conditioned, and just by one plant species, E. marginata.

50

Interactions between plant-parasitic nematodes and their host plants are mediated by

root exudates (Hirsch et al. 2003; Zavaleta-Mejía & Kaloshian 2012). Therefore, it is

likely E. marginata produced chemical signals to attract plant-parasitic nematodes to

the rhizosphere. Plant parasites and pathogens are agents of negative PSF, which

contribute to shaping the structure of plant communities during succession (Mills &

Bever 1998; van der Putten 2000; Bonanomi et al. 2005a; Bonanomi et al. 2005a;

Mangan et al. 2010). Information on the ecological role of nematodes in jarrah forest is

scarce, however, these organisms might contribute to counteract the dominance of

plants such as E. marginata and thus contribute towards species coexistence.

In restored soil, the patterns of conditioning of chemical properties differed from those

observed in unmined soil as predicted, but similar to unmined soils, conditioning of the

biological properties was scarcely evident. The results suggest that modified properties

initially perceived by plants induced changes to the production of rhizodeposition and

exudates, which in turn affected soil conditioning (Marschner 1995; Uren 2001;

Nautiyal et al. 2008). In contrast to the results for unmined soils, the chemical

properties of restored soils were not conditioned by A. pulchella, and chemical

properties were similarly conditioned by E. marginata and B. ornata, except by the

concentration of available K that decreased in soils grown with B. ornata, possibly due

to an increased uptake by this plant species. In contrast to unmined soils E. marginata

and B. ornata did not condition the C concentration and A. pulchella did not condition

the N concentration of restored soils. These altered patterns of conditioning are likely to

affect the C and N cycles as previously suggested by other studies (Banning et al.

2008), with important implications for the restoration of the ecosystem function.

Soil conditioning by E. marginata, A. pulchella, and B. ornata produced no change in

the C:N ratio of restored soils as compared to controls (as observed for unmined soil).

51

No change in the C:N ratios suggests that organic matter decomposition process is

differentially operating in restored soil as compared to unmined soil when these were

under the influence of the experimental plants. In contrast, conditioning of available P

was not affected by soil modification, since E. marginata and B. ornata, but not A.

pulchella, conditioned the concentration of available P of restored soil as observed in

unmined soil. This finding suggests that conditioning of P is a preserved plant trait that

might help maintaining the flux of nutrients in the ecosystem (Eviner & Hawkes 2008).

Therefore, restoration of long-lived species such as E. marginata and B. ornata might

be critical for the re-establishment of ecosystem functioning for the entire developing

forest system.

In restored soil, the numbers of fungal-feeding nematodes increased as a result of soil

conditioning by E. marginata. It is likely that rhizodeposition and exudates released to

soil by E. marginata (Malajczuk & McComb 1977) produced a positive effect on the

fungal microbial biomass. Fungi was not meaured directly but was possibly seen

indirectly in the PSF response. In restored soil, plant-parasitic nematodes were not

conditioned by E. marginata as observed in unmined soil. This result suggests that the

patterns of root exudation of the plant differed, and chemical attractants for plant-

parasitic nematodes were scarcely/ not produced. Another possibility is that plant-

parasitic nematodes were so scarce in restored soil that they were not detected during

the screening. The absence or low abundance of negative PSF agents such as plant-

parasitic nematodes could have implications on the structure and evolution of the plant

community during restoration (De Rooij-van der Goes 1995; van der Putten & Peters

1997; van der Stoel et al. 2002).

In unmined soil a neutral PSF was observed for both A. pulchella and B. ornata. These

findings contrasted with the prediction of a negative PSF associated with A. pulchella,

52

and positive PSF associated with B. ornata. The results suggested that individual PSF

was not influenced by soil properties previously conditioned by either the same (i.e.

home soil) or other plant species (i.e. away soil). In restored soil, individual PSF was

altered with respect to unmined soil, and negative PSF was observed for both A.

pulchella and B. ornata. The results indicated that individual PSF associated with the

two plant species was influenced by the modification to the soil. However, opposite to

unmined soil, in the context of restored soil, the legacy of previous plant growth (i.e. E.

marginata or B. ornata growth) seemed to influence the amounts of biomass produced

by the experimental plants (i.e. PSF that is soil-context dependent; De Deyn et al.

2004).

The notable increased biomass produced by A. pulchella in restored soil suggests a soil

feedback effect that induced a species-specific response in A. pulchella, but not in B.

ornata. Species-specific responses have been reported previously for a diversity of

couplings of soils and plants (Bret-Harte et al. 2001; Manning et al. 2008; te Beest et al.

2009; Baeten et al. 2010). These types of responses can alter the patterns of plant

abundance and dominance, and ultimately the successional trajectory of the plant

community.

In conclusion, conditioning of soil by the representative jarrah forest plants used in this

study was species-specific, the mechanism might have implications for the persistence

of plant species in the ecosystem. Modification of soil by mining and restoration

practice induced patterns of plant conditioning that differed with respect to unmined

soil. Individual PSF associated with the studied jarrah forest plants operated in a soil

context-dependent basis, as a result individual PSF observed in restored soil differed as

compared with that observed in unmined soil for both A. pulchella and B. ornata.

Alteration to conditioning process and individual PSF operating in restored soils can

53

have implications for ecosystem recovery by affecting the successional trajectory of the

plant community.

54

CHAPTER 4

Plant-soil feedback unaffected by soil condition in a post-mining

restoration chronosequence

INTRODUCTION

Ecological restoration of ecosystems that have been damaged or destroyed can result in

changes to the properties of both above and belowground compartments such that

recovery is evident over time (Ward 2000; Mummey et al. 2002; Norman et al. 2006;

Jasper 2007; Banning et al. 2008; Chodak et al. 2009). Restoration can include practices

such as ripping, seeding, planting, fertilising, and inoculation of beneficial

microorganisms to improve the physico-chemical and biological properties of soil to

accelerate successional change (Suding et al. 2004; Grant 2006; Koch 2007b; Kardol &

Wardle 2010). As a result, the concentrations of soil nutrients change due to

biogeochemical processes, C concentration increases through litter and root deposition,

the size and diversity of biotic communities increases, and the soil food chain length

grows (Schipper et al. 2001; Ferris & Matute 2003; Wardle et al. 2004; Bardgett et al.

2005; Wardle 2006; Holtkamp et al. 2008; Berg 2010; Eisenhauer & Reich 2012), all of

which ultimately contribute to ecosystem recovery. Nevertheless, there is a lack of

information as to whether the restoration of soil results in the recovery of the soil

feedback in such a way as to re-establish the plant-soil feedbacks (PSF) that are

characteristic of the undisturbed ecosystem in their magnitude and direction.

Soil feedback is the second component of PSF (after plant conditioning), and it is

important because the overall effect of physico-chemical and biological properties of

soil determines the growth, productivity, and reproductive success of individual plants

in ecosystems (van der Putten et al. 2013). In human-modified ecosystems soil

55

properties can be altered and likely their associated soil feedback, which can ultimately

modify the patterns of plant growth in the short term via PSF phenomena (Bret-Harte et

al. 2001; Baeten et al. 2010; but also see Chapter 3). In the long-term, the development

of soil properties during restoration suggests that the soil feedback and associated PSF

might experience changes over time. Empirical data indicating whether or not the soil

feedback of restored ecosystems moves toward to a soil feedback approaching that of

unmodified ecosystem are limited (Maluf & Ferreira 2004; Ribeiro et al. 2004; Shono et

al. 2005; Paul et al. 2010; Gong et al. 2013).

In the jarrah (Eucalyptus marginata) forest of south-western Western Australia, patches

of forest have been cleared for bauxite mining since 1963, and restored after mining

since 1966. However, restored soil displays modified properties such as reduced

biological activity, pH, and C, and increased P, N, K and gravel percentage (Ward 2000;

Jasper 2007; George et al. 2010; Tibbett 2010). Increased concentrations of P, N, and K

are consequence of fertiliser addition as part of the restoration process (Koch 2007b).

Increased amounts of these nutrients are expected to be significant components of the

soil feedback that in turn influence plant growth in recently-restored soil. Recently

restored jarrah forest soil also shows lower C concentration and biological activity than

native forest, but after 18 and 8 years respectively, each one of these soil properties

recovers to levels comparable to those of unmined forest soil (Jasper 2007; Banning et

al. 2008; George et al. 2010; Lin et al. 2011). Therefore, it is likely that C concentration

and biological communities of soil strongly influence plant growth via soil feedback in

the mid-, long-term, and not in the short-term. The relevance of these properties in the

soil feedback might become apparent with increasing restoration age given the

dependence of native plants on biologically-supplied nutrients such as N and P (Lawrie

1981; Brundrett & Abbott 1991; Lambers et al. 2007), and decreasing concentrations of

nutrients from fertilisers. The changing soil feedback components of restored jarrah

56

forest soil differentially influence plant biomass production (i.e. the individual PSF), as

described in the previous chapter (Chapter 3), where negative PSF was associated with

plant species grown in recently-restored soils (i.e. 1.5-year-old soils), and neutral PSF

was associated with the same plant species grown in unmined jarrah forest soils.

This investigation aimed to elucidate the influence of soil conditions on individual PSF

associated with Bossiaea ornata plants. For this purpose a glasshouse experiment was

designed to study the soil feedback produced in soils collected from a chronosequence

of restoration ages (1.5-, 7-, and 22-year-old soils) and an unmined soil that is the

benchmark for restoration. Specifically, a first aim was to determine the factors (i.e.

soil age and plant species) influencing the physico-chemical and biological properties

(i.e. the soil feedback) of soils from the chronosequence. It was predicted that properties

of soils from the chronosequence would be significantly influenced by both; age since

restoration and plant species grown on soils. Also, the properties of soil from the

chronosequence would be more similar to those of unmined soils as restoration age

increases (Jasper 2007; Banning et al. 2008; George et al. 2010; Lin et al. 2011).

There is potential for jarrah forest soils of diverse restoration age to exert differential

soil feedback on plant growth in response to variable soil properties (Ward 2000;

Banning et al. 2008; George et al. 2010; Lin et al. 2011). Therefore, the second aim of

this investigation was to examine whether or not the differential soil feedback

associated with soils from the restoration chronosequence differentially influence the

biomass produced by native B. ornata seedlings. This plant species was selected

because displays variable growth behaviour in restored as compared with unmined soils.

Bossiaea ornata is a resprouter, long-lived bush (Bell 2001) that is abundant in unmined

soils, but is not as prolific under fertilisation practice typical in restoration (Bell &

Heddle 1989; George et al. 2006). One hypothesis that could help explaining this

57

behaviour is that soil feedback exerts a strong influence on the establishment and

growth of this plant species. Therefore, variable biomass production by B. ornata was

expected in response to differential soil feedback associated to soils of variable

restoration age. Specifically, the biomass produced by B. ornata would be significantly

influenced by the effects of soil age and plant species previously grown on soils. It is

also predicted that physico-chemical properties would be significant components of the

soil feedback in recently-restored soils, whereas biological properties would be

significant components of the soil feedback in late-restored soils (Bardgett et al. 2005;

Wardle 2006; Holtkamp et al. 2008; Berg 2010; Eisenhauer & Reich 2012).

Finally, the third aim was to investigate if variable soil feedback of soils from the

chronosequence differentially influence individual PSF (i.e. its magnitude and direction)

associated with native B. ornata. A negative PSF associated with B. ornata grown in

recently-restored soil (i.e. 1.5-year-old soil) was expected, and then a transition to

neutral PSF associated to the same plant species grown in late-restored soil was

expected according to results from Chapter 3 (Kardol et al. 2006).

MATERIAL AND METHODS

Sampling and experimental design

Sites of increasing age since restoration after bauxite mining (1.5-, 7-, and 22-year-old

soils) and an unmined site that was burnt under controlled conditions in 2008 (hereafter

referred to collectively as ‘the chronosequence’) were selected within the mining lease

of Alcoa World Alumina Australia. Soils for PSF experiment were collected in June

2011. At each site, six B. ornata (Fabaceae) plants and six E. marginata (Myrtaceae)

plants were located in an area of approximately 1 km2, and sampled at random to take

two intact soil cores (0.785 L; diameter = 10 cm, depth = 10 cm) from near to the root

area of each plant. Sampled plants were some meters apart from each other. One half of

58

soil cores were used to measure the physico-chemical and biological properties of soils

to determine effects of field-based soil conditioning. Soil subsamples were sieved (2

mm-mesh), frozen (-20 °C) and then subsequently freeze-dried pending phospholipid-

fatty acid analysis (PLFA). The second half of soil cores were placed in bottom-sealed

pots (1.2 L; diameter = 13.3 cm, depth = 12.4 cm) ready for use in the pot experiment.

The experimental design included 6 replicates × four soil ages (1.5-, 7-, 22-year-old,

unmined soil) × two plant species (B. ornata or E. marginata) = 48 pots (Figure 1).

The pot experiment ran from June to October 2011. Pots containing the intact soil cores

(hereafter the field-conditioned soils) were placed into two root cooling tanks that were

filled with water set to a constant temperature of 20 °C in order to maintain the area of

contact between roots and soil under homogeneous conditions. Root cooling tanks were

housed in a glasshouse for the duration of the experiment. Temperatures inside the

glasshouse ranged between 4.5 and 27.8 °C. For the experiment seedlings of B. ornata

were used. In each pot several seedlings of B. ornata were planted, once a plant

established, the rest of seedlings were removed leaving only one growing plant per pot.

To produce the seedlings, seeds were submerged in boiling water for 30 seconds to

break dormancy, then surface-disinfected with ethanol (70%) and bleach (50%) to

prevent pathogen growth and germinated at 14 °C in darkness. Pots planted with

seedlings were rotated between and within tanks every two weeks to negate a block

effect and watered once a week to 60% water-holding capacity. After four months

plants were harvested; shoots and roots were removed, separated, oven-dried for one

week at 60 °C and weighed.

59

Figure 1. Experimental design where intact soil cores were collected from a chronosequence of restoration ages (1.5-, 7-, 22-year-old soils and unmined soils). At each site soil cores were taken from area near roots of Bossiaea ornata and Eucalyptus marginata. One half of cores were used to determine the physico-chemical and biological properties (A) to determine soil feedback components (B). The second half of the cores was placed in pots where B. ornata seedlings were planted to have B. ornata seedlings grown in soils conditioned in the field by B. ornata plants (i.e. home soils), and B. ornata seedlings grown in soils conditioned in the field by E. marginata plants (i.e. away soils). After four months all plants were harvested and plant biomass produced in each restoration age was compared in home vs. away soils (C).

Response variables

Refer to chapter two for methodologies used to determine response variables measured.

Statistical analysis

The influence of the factors soil age (with four levels; 1.5-, 7-, 22-year-old, unmined

soils) and plant species (with two levels; E. marginata and B. ornata; Figure 1A) on soil

properties was tested. To do this, a two-way permutational analysis of variance

(PERMANOVA; Anderson 2001) on the sets of physico-chemical and biological

properties of the field-conditioned soils was used. To visualize the results of the

PERMANOVA, scaling plots using principal co-ordinate analysis (PCO; Anderson

2001) were built. PCO plots showing the distances from centroids of all soil ages were

also built to determine the trajectory of multivariate change through time followed by

Properties of field-

conditioned soils

C

Intact soil cores conditioned by plant species from a chronosequence of

restoration ages

Bossiaea ornata

Eucalyptus marginata

Bossiaea ornata

grown in “home soil”

Total biomass production in

home vs. away soils

Individual Plant-soil feedback

A

Bossiaeaornata

grown in “away soil”

60

the physico-chemical and biological properties of soils from the chronosequence. The

plots allowed determining if the properties of field-conditioned soils from the

chronosequence were approaching those of unmined soils. As well, PERMANOVA

analyses were performed to determined significant effects of soil age and plant species

on total PLFAs, PLFA biomarkers for total bacteria, Gram positive bacteria, Gram

negative bacteria, fungi, and microeucaryotes, respiration rates of total C-substrates

from CLPPs, and groups of C-substrates of similar chemical nature (i.e. carboxylic

acids, aminoacids, phenolics, carbohydrates, etc. See appendix 1), and the trophic

groups of nematodes of soils within the chronosequence.

To determine differential soil feedback effect of soils from the chronosequence on

biomass produced by B. ornata seedlings (Figure 1B), two types of analyses were

performed; a univariate two-way PERMANOVA for the factors soil age (with four

levels; 1.5-, 7-, 22-year-old, unmined soils) and plant species (from which soils were

collected in the field; with two levels; E. marginata and B. ornata) on the response

variable plant biomass. Although PERMANOVA is a multivariate method, it can be

used for univariate analyses. To perform PERMANOVA using one response variable

and Euclidean distance yields Fisher’s traditional univariate F-statistic (Anderson

2010). The analysis allowed to determine significant differences in biomass produced

by B. ornata in soils from the chronosequence in response to the factors mentioned

above. The other type of analysis used was an analysis of covariance (ANCOVA) that

consisted in univariate two-way PERMANOVAs for the factors soil age and plant

species on the response variable plant biomass, but individual soil properties were

included in the model as covariables. These analyses allowed determining significant

and strong effects of the covariables on the biomass produced by B. ornata (Anderson et

al. 2008). Tests for the interactions between covariables and factors (i.e. soil age and

61

plant species) were also included in the PERMANOVA design to evidence differential

effects of the covariables across soil ages (Anderson et al. 2008).

Finally, individual PSF (i.e. magnitude and direction; Figure 1C) was determined

through univariate PERMANOVA with plant biomass as the response variable. The

effect of the factor conditioning type (with two levels; home soils - conditioned in the

field by B. ornata, and away soils – conditioned in the field by E. marginata) was

determined for each soil age. These results were interpreted as follows: significantly

increased plant biomass production in home with respect to away soils was considered

evidence of positive PSF. Significantly decreased plant biomass production in home

soils with respect to away soils was considered evidence of negative PSF. No difference

in biomass production in home with respect to away soils was considered neutral PSF.

The software PRIMER v6 +PERMANOVA (PRIMER-E Ltd) was used for all statistical

analyses. Permutation of residuals under a reduced model and type III (partial) sum of

squares were selected to perform multivariate analyses, whereas unrestricted

permutation of raw data and type III (partial) sum of squares were selected for

univariate analyses. In all cases data were transformed when necessary and normalised,

and Euclidean distance was used to build resemblance matrices, except for the analyses

of the trophic groups of the nematode community when Bray-Curtis coefficient was

utilized because individuals were counted, and for the analysis of CLPPs when the

Manhattan distance was used because it has been pointed out as an appropriate distance

for the analysis (Lalor et al. 2007; Gonzalez-Quiñones et al. 2009; Banning et al. 2012).

62

RESULTS

Properties of field-conditioned soils across the chronosequence and their trajectory

of change

The physico-chemical and biological properties of field-conditioned soils (i.e. the soil

feedback) were significantly influenced by soil age (i.e. 1.5-, 7-, 22-year-old, unmined

soils; F3, 79.5 = 13.48, P = 0.001 for physico-chemical properties, and F3, 41.7 = 6.19, P =

0.001 for biological properties) but not by plant species (i.e. E. marginata, B. ornata;

Table 1; Figure 2). Post-PERMANOVA pair-wise tests revealed that physico-chemical

properties of field-conditioned soils were all significantly different to each other

according to soil age (Table 2). As well, the biological properties of field-conditioned

soils were all significantly different to each other according to soil age, except by the

1.5-year-old as compared with the 7-year-old soils, and the 7-year-old as compared with

the 22-year-old soils (Table 2).

Figure 2. Scaling plots from principal co-ordinate analysis (PCO) of a) physico-chemical: density (g cm-3), gravel (%), available P and K (mg kg-1), EC (dS m-1), pH, total N and C (%), NO3

- and NH4+, (mg kg-1), and C:N ratio, and b) biological properties: microbial biomass-C (μg-

C g-1 soil), bacterial-feeding, fungal-feeding, omnivorous, predatory, plant-parasitic nematodes (per 100 g-1 soil), total CO2 rates produced from CLPPs (µg CO2 g-1soil h-1), and nematode channel ratio of field-conditioned soils from the chronosequence of restoration ages (1.5-, 7-, and 22-year-old, unmined soil). PCOs showed a significant effect of soil age, but not of plant species on the physico-chemical and biological properties of soils as revealed by two-way PERMANOVAs.

63

Total PLFA of soils (50 different PLFA), were not significantly influenced by any of

the factors soil age and plant species, but analyses of specific PLFA biomarkers

revealed a significant interaction soil age × plant species for Gram positive bacteria

(F3, 2 = 2.05, P = 0.03). Further analyses revealed significant differences in PLFA

biomarkers for Gram positive bacteria only for the 1.5-year-old soils (F1, 2.05 = 2.68, P =

0.033), and for unmined soils (F1, 0.99 = 2.61, P = 0.042; Table 1; Figure 3a). PLFA

biomarkers for microeucaryotes were significantly influenced by soil age (F3, 1.58 = 2.28,

P = 0.029), pair-wise analyses indicated that the 7-year-old soils were different from

unmined soils (t = 2.35, P = 0.007; Table 1). The rest of pair comparisons were not

significantly different (Figure 3b). PLFA biomarkers for total bacteria, Gram negative

bacteria, and fungi were not significantly influenced for the factors soil age or plant

species.

Figure 3. Scaling plot from principal co-ordinate analysis (PCO) of a) PLFA biomarkers for Gram positive bacteria, which were conditioned by Eucalyptus marginata and Bossiaea ornata in a species-specific manner only in the 1.5-year-old soils and in unmined soils, (significant interaction soil age × plant species) and b) PLFA biomarkers for microeucaryotes that were significantly influenced by soil age.

64

Total CLPPs were significantly influenced by soil age (F3, 305 = 13.32, P = 0.0001) but

not by plant species (Figure 4). Pair-wise tests evidenced that CLPPs of soils were all

different, except by the 1.5- and the 7-year-old soils whose CLPPs were similar (Table

2). Also, the respiration rates (from CLPP) of the groups of C-substrates: carboxylic

acids (F3, 17 = 24.17, P = 0.001), aminoacids (F3, 2.44 = 15.2, P = 0.001), phenolic

compounds (F3, 0.16 = 7.81, P = 0.001), aromatic compounds (F3, 1.43 = 23.12, P = 0.001),

carbohydrates (F3, 1.36 = 14.7, P = 0.001), alcohols (F3, 0.37 = 13.54, P = 0.001), and

phenolic acids (F3, 0.53 = 24.28, P = 0.001), were significantly influenced by soil age

(and not by plant species; Table 1). Respiration rates of vitamins were not influenced by

soil age or plant species.

Figure 4. Scaling plot from principal co-ordinate analysis (PCO) of 31 C-substrates used to determine CLPPs (µg CO2 g-1soil h-1) of field-conditioned soils (i.e. conditioned by Eucalyptus marginata or Bossiaea ornata) from a chronosequence of restoration ages (1.5-, 7-, and 22-year-old, unmined soil).

Finally, the nematode trophic community was significantly influenced by soil age (F3,

5007 = 7.29, P = 0.0001), not by plant species. Post-PERMANOVA pair-wise tests

indicated that the trophic community of nematodes was different across all experimental

soils (Table 2). The groups of nematodes: bacterial-feeders (F3, 300000 = 5.93, P =

0.0003), fungal-feeders (F3, 1723 = 11.24, P = 0.0003), and plant-parasites (F3, 104730 =

65

11.96, P = 0.0001) were significantly influenced by soil age, and not by plant species.

Whereas predatory and omnivorous nematodes were not influenced by any of the

factors tested (Table 1). The nematode channel ratio of soils from the chronosequence

was significantly influenced by soil age (F3, 0.15 = 3.28, P = 0.026), not by plant species

(Table 1; Figure 5).

Figure 5. Scaling plot from principal co-ordinate analysis (PCO) of nematode trophic community (specimens per 100 g soil) of field-conditioned soils (i.e. conditioned by Eucalyptus marginata or Bossiaea ornata) from a chronosequence of restoration ages (1.5-, 7-, and 22-year-old, unmined soil).

66

Table 1. Mean values for the physico-chemical and biological properties (± SE; n = 6) of field-conditioned soils from the chronosequence (1.5-, 7-, and 22-year-old, unmined soils).

Soil age

Property 1.5-year-old 7-year-old 22-year-old Unmined

Eucalyptus marginata

Bossiaea ornata

Eucalyptus marginata

Bossiaea ornata

Eucalyptus marginata

Bossiaea ornata

Eucalyptus marginata

Bossiaea ornata

Density (g cm-3)*** 1.68 (0.16)

1.8 (0.12)

1.71 (0.11)

1.95 (0.10)

1.16 (0.09)

1.41 (0.16)

1.02 (0.08)

1.35 (0.12)

Gravel (%)** 66.98 (3.40)

76.85 (1.47)

74.33 (3.40)

74.15 (1.47)

65.71 (2.0)

64.97 (2.41)

56.38 (3.32)

69.04 (3.99)

Available P (mg kg-1)*** 7.0 (1.15)

8.4 (1.69)

4.0 (0.77)

4.6 (1.33)

2.5 (0.34)

3.83 (0.79)

6.8 (1.06)

4.83 (0.73)

Available K (mg kg-1)*** 43.17 (2.75)

35.2 (6.14)

36.0 (7.28)

48.2 (3.97)

53.17 (6.15)

64.17 (11.49)

136.6 (16.04)

102.5 (8.46)

EC (dS m-1)** 0.03 (0.001)

0.026 (0.003)

0.022 (0.004)

0.025 (0.003)

0.051 (0.003)

0.051 (0.005)

0.097 (0.009)

0.082 (0.006)

pH*** 5.15 (0.05)

5.0 (0.05)

5.07 (0.13)

4.9 (0.10)

4.68 (0.07)

4.4 (0.10)

3.8 (0.15)

4.5 (0.07)

Total N (%)*** 0.083 (0.01)

0.084 (0.01)

0.048 (0.01)

0.062 (0.01)

0.095 (0.004)

0.11 (0.02)

0.29 (0.03)

0.23 (0.04)

Total C (%)*** 2.44 (0.30)

2.52 (0.89)

1.09 (0.13)

1.77 (0.34)

2.28 (0.26)

2.57 (0.50)

9.4 (1.38)

6.18 (1.04)

NO3- (mg kg-1) 0.075

(0.08) 0.17 (0.14)

0.19 (0.10)

0.11 (0.09)

0.17 (0.13)

0.23 (0.16)

0.29 (0.12)

0.18 (0.12)

NH4+ (mg kg-1)*** 1.29

(0.31) 1.23 (0.26)

0.57 (0.27)

0.51 (0.06)

0.92 (0.36)

1.28 (0.46)

6.43 (3.07)

3.52 (0.83)

C:N ratio 28.93 (1.86)

29.67 (3.25)

23.41 (1.13)

27.93 (2.86)

24.57 (3.46)

23.11 (0.52)

31.47 (2.86)

27.45 (0.74)

Microbial biomass-C (μg-C g-1 soil)**

368.73 (95.37)

270.18 (86.32)

291.58 (82.41)

468.71 (108.69)

570.56 (33.57)

616.27 (195.32)

2084.01 (507.05)

826.37 (270.78)

Total PLFAs (μg g-1 soil)

87.14 (67.66)

56.84 (7.77)

61.38 (5.89)

122.37 (104.27)

91.59 (101.47)

77.7 (55.94)

61.44 (4.08)

59.37 (8.08)

Total bacterial PLFAs (μg g-1 soil)

37.41 (12.02)

21.82 (0.99)

25.34 (0.66)

49.85 (16.67)

35.52 (16.22)

32.65 (9.57)

22.78 (1.09)

19.91 (1.01)

Gram+ PLFAs (μg g-1 soil)

4.01 (0.95)

1.95 (0.28)

2.46 (0.45)

6.88 (3.20)

3,61 (1.81)

2.20 (0.55)

2.50 (0.33)

1.76 (0.19)

Gram- PLFAs (μg g-1 soil)

5.32 (0.80)

5.75 (0.30)

4.45 (0.61)

6.80 (1.39)

4.39 (0.87)

5.09 (0.92)

5.14 (0.45)

5.08 (0.57)

Fungal PLFAs (μg g-1 soil)

20.80 (5.82)

16.01 (1.21)

13.68 (0.95)

29.90 (10.40)

23.27 (9.13)

20.44 (5.71)

17.05 (1.20)

18.26 (1.16)

Microeucaryote PLFAs* (μg g-1 soil)

2.92 (0.51)

1.94 (0.21)

2.84 (0.25)

5.53 (2.44)

3.53 (2.02)

2.01 (0.82)

2.29 (0.26)

1.79 (0.24)

Total CO2 from CLPP***

(μg CO2–C g-1 soil h-1)

5.43 (0.75)

2.95 (1.04)

6.31 (3.04)

5.58 (1.94)

9.58 (1.96)

11.91 (4.10)

25.08 (3.92)

25.35 (2.34)

Total CO2 produced from carboxylic acids ***

(μg CO2–C g-1 soil h-1)

0.3 (0.53)

0.17 (0.75)

0.35 (2.12)

0.30 (1.43)

0.52 (1.59)

0.63 (2.97)

1.43 (2.48)

1.42 (1.76)

Total CO2 produced from aminoacids***

(μg CO2–C g-1 soil h-1)

0.13 (0.14)

0.06 (0.14)

0.14 (0.46)

0.12 (0.23)

0.19 (0.19)

0.27 (0.74)

0.48 (0.59)

0.47 (0.29)

Total CO2 produced from phenolics***

(μg CO2–C g-1 soil h-1)

0.02 (0.01)

0.01 (0.01)

0.02 (0.01)

0.03 (0.05)

0.06 (0.03)

0.06 (0.07)

0.08 (0.05)

0.05 (0.04)

Total CO2 produced from aromatics***

(μg CO2–C g-1 soil h-1)

0.08 (0.03)

0.04 (0.05)

0.08 (0.12)

0.09 (0.10)

0.11 (0.08)

0.13 (0.14)

0.38 (0.27)

0.53 (0.25)

Total CO2 produced from carbohydrates***

(μg CO2–C g-1 soil h-1)

0.14 (0.08)

0.05 (0.07)

0.17 (0.27)

0.15 (0.18)

0.19 (0.13)

0.20 (0.21)

0.53 (0.40)

0.56 (0.22)

Total CO2 produced from alcohols***

(μg CO2–C g-1 soil h-1)

0.05 (0.01)

0.03 (0.02)

0.1 (0.06)

0.08 (0.03)

0.11 (0.03)

0.18 (0.06)

0.39 (0.10)

0.38 (0.08)

Total CO2 produced from phenolic acids***

(μg CO2–C g-1 soil h-1)

0.003 (0.002)

0.005 (0.003)

0.022 (0.01)

0.046 (0.02)

0.13 (0.04)

0.08 (0.04)

0.30 (0.11)

0.33 (0.09)

67

Total CO2 produced from vitamins

(μg CO2–C g-1 soil h-1)

0.003 (0.003)

0.03 (0.02)

0.003 (0.003)

0.02 (0.02)

0.06 (0.02)

0.01 (0.01)

0.02 (0.01)

0.001 (0.001)

Bacterial-feeding nematodes*** (per 100 g-1 soil)

286 (97)

257 (40)

303 (142)

413 (147)

1827 (678)

858 (310)

441 (210)

487 (98)

Fungal-feeding nematodes)***

(per 100 g-1 soil)

198 (77)

242 (53)

695 (392)

385 (132)

1569 (339)

652 (213)

1982 (628)

1634 (424)

Omnivorous nematodes (per 100 g-1 soil)

12 (8)

16 (7)

47 (17)

44 (15)

41 (14)

61 (17)

22 (11)

90 (33)

Predatory nematodes (per 100 g-1 soil)

0 0 15 (5)

18 (9)

5 (5)

10 (10)

0 10 (10)

Plant-parasitic nematodes*** (per 100 g-1 soil)

0 0 0 0 0 0 271 (100)

102 (40)

Nematode channel ratio* 0.49 (0.11)

0.47 (0.09)

0.52 (0.13)

0.50 (0.12)

0.51 (0.04)

0.49 (0.12)

0.85 (0.04)

0.74 (0.05)

* P < 0.05, ** P < 0.01 *** P < 0.001

Table 2. Results for pair-wise comparisons post-PERMANOVA of soil properties present in the chronosequence of restoration ages

Soil age comparison

Property 1.5-year-old

vs. 7-year-old

1.5-year-old

vs. 22-year-old

1.5-year-old

vs. unmined

7-year-old vs.

22-year-old

7-year-old vs.

unmined

22-year-old

vs. unmined

Physico-

chemical

t = 1.86 t = 3.06 t = 4.36 t = 2.74 t = 5.08 t = 3.32

P = 0.008 P = 0.001 P = 0.001 P = 0.001 P = 0.001 P = 0.001

Biological t = 1.31 t = 2.41 t = 3.98 t = 1.35 t = 2.84 t = 2.49

P = 0.13 P = 0.001 P = 0.001 P = 0.13 P = 0.001 P = 0.002

CLPP t = 0.98 t = 2.97 t = 6.75 t = 1.89 t = 4.83 t = 3.54

P = 0.35 P = 0.0015 P = 0.0001 P = 0.049 P = 0.0003 P = 0.0001

Nematode

trophic

community

t = 1.94 t = 3.21 t = 3.61 t = 2.23 t = 2.8 t = 2.27

P = 0.016 P = 0.0002 P = 0.0001 P = 0.012 P = 0.001 P = 0.0007

The physico-chemical and biological properties of field-conditioned soils within the

chronosequence did not display linear trajectories of change in ordination space (Figure

6). The distances among centroids for physico-chemical and biological properties per

plant species (i.e. E. marginata and B. ornata) tended to be smaller with respect to

unmined soils as restoration age increases, except by the distances between the 7-year-

68

old soils and unmined soils for the physico-chemical and biological properties

conditioned in the field by E. marginata, and the physico-chemical properties

conditioned in the field by B. ornata. The distances among centroids for the physico-

chemical properties conditioned in the field by E. marginata are: 1.5-year-old to

unmined = 5.01; 7-year-old to unmined = 6.13; 22-year-old to unmined = 4.35. The

distances among centroids for the biological properties conditioned in the field by E.

marginata are: 1.5-year-old soil to unmined = 3.65; 7-year-old to unmined = 3.75; 22-

year-old to unmined = 3.2. The distances among centroids for the physico-chemical

properties conditioned by B. ornata in the field are: 1.5-year-old soil to unmined = 4.29;

7-year-old to unmined = 4.52; 22-year-old to unmined = 3.03. The distances among

centroids for the biological properties conditioned in the field by B. ornata are: 1.5-

year-old soil to unmined = 3.7; 7-year-old to unmined = 3.4; 22-year-old to unmined =

2.6.

69

Figure 6. Trajectory followed by a) the physico-chemical and b) biological properties of soils from a chronosequence of restoration ages (1.5-, 7-, and 22-year-old, unmined soil) that were conditioned in the field by Eucalyptus marginata. And, the trajectory followed by c) the physico-chemical and d) biological properties of soils from the chronosequence that were conditioned in the field by Bossiaea ornata. Values for the distances among centroids are indicated.

Soil feedback of soils from the chronosequence

The biomass produced by B. ornata grown in soils from the chronosequence was

significantly influenced by soil age (F3, 0.06 = 3.27, P = 0.018), but not by plant species

from which soils were collected in the field (i.e. E. marginata and B. ornata; Figure 7).

The smallest plants were produced in unmined soils, whereas the biggest plants were

produced in the 22-year-old soils, and this difference was statistically significant (t =

2.39, P = 0.024).

70

Figure 7. Total biomass (± SE; n = 6) produced by Bossiaea ornata grown in soils collected from a chronosequence of restoration ages (1.5-, 7-, and 22-year, and unmined soils). Means with the same letters were not significantly different (P < 0.05).

From the individual soil properties tested, the number of bacterial-feeding nematodes

and the rate of CO2 produced by microbial respiration of phenolic compounds from

CLPP had a significant effect on the total biomass produced by B. ornata grown in soils

from the chronosequence (Table 3). This result indicated a significant relationship

between these two properties and the biomass produced by B. ornata, therefore only

these two properties were considered significant components of the soil feedback in all

soils. The nature of the relationship between the biomass produced by B. ornata and the

number of bacterial-feeding nematodes and the rate of CO2 produced by microbial

respiration of phenolic compounds did not differ across soil ages or plant species

previously grown on soils, as evidenced by non-significant interactions between the

variables bacterial-feeding nematodes and total CO2 produced from phenolics and the

factors soil age and plant species included in the ANCOVA test (F3, 0.02 = 1.4, P =

0.253).

a

a

b

a

0

0.05

0.1

0.15

0.2

0.25

0.3

0.35

1.5-year-old 7-year-old 22-year-old Unmined

Tota

l bio

mas

s (g

)

71

Table 3. A summary of ANCOVA results to explain the factors and properties significantly influencing biomass produced by B. ornata. Statistics for the factors soil age and plant species and individual soil properties as covariables are shown.

Covariables df MS Pseudo-F P-value

Bacterial-feeding nematodes(per 100 g-1 soil) 1 0.27 17.36 0.003 Soil age 3 7.23x10-2 1.53 0.22 Plant species 1 3.58x10-2 2.28 0.14

Phenolics-CLPP (μg g-1 soil) Soil age Plant species

1 3 3

0.13 6.02x10-2

1.68x10-2

7.00 3.21 0.90

0.018 0.032 0.36

Vitamins-CLPP (μg g-1 soil) Soil age Plant species

1 3 1

7.06x10-2

5.45x10-2

2.72 x10-2

4.14 3.19 1.59

0.054 0.029 0.21

Gram Positive PLFA (μg g-1 soil) Soil age Plant species

1 3 1

4.07x10-2

7.64x10-2

1.06x102

2.16 4.05 0.56

0.18 0.015 0.47

Omnivorous nematodes (per 100 g-1 soil) 1 7.59x10-2 3.53 0.06 Soil age 3 6.87x10-2 3.20 0.04 Plant species 1 1.35x10-2 0.63 0.41

Nematode channel ratio Soil age Plant species

1 3 1

5.98x10-2 6.49x10-2 1.91x10-2

3.64 3.95 1.16

0.07 0.02 0.31

Total bacterial PLFA (μg g-1 soil) Soil age Plant species

1 3 1

6.53x10-2

6.77x10-2

8.82 x10-3

3.28 3.40 0.44

0.08 0.04 0.50

C:N ratio Soil age Plant species

1 3 1

7.58x10-2 5.36x10-2 2.62x10-2

3.64 2.57 1.26

0.08 0.07 0.27

Density (g cm-3) 1 6.05x10-2 2.82 0.11 Soil age 3 6.98x10-2 3.26 0.04 Plant species 1 5.88 x10-2 2.74 0.10

Gravel (%) 1 3.9x10-2 1.84 0.19 Soil age 3 7.3x10-2 3.43 0.03 Plant species 1 4.84x10-2 2.28 0.12

Available P (mg kg-1) 1 3.46x10-2 1.53 0.23 Soil age 3 6.27x10-2 2.77 0.08 Plant species 1 2.49x10-2 1.10 0.28

pH Soil age Plant species

1 3 1

2.24x10-2 8,52x10-2 2.77x10-2

0.10 3.78 1.23

0.32 0.02 0.31

Total C (%) 1 1.93x10-2 0.96 0.32 Soil age Plant species

3 1

6.86x10-2 2.69x10-2

3.40 1.33

0.03 0.27

Total PLFA (μg g-1 soil) Soil age Plant species

1 3 1

2.22x10-2

1.13x10-2

0.92

1.02 3.25 0.02

0.32 0.03 0.36

Predatory nematodes (per 100 g-1 soil) 1 2.37x10-2 0.97 0.34 Soil age 3 7.22x10-2 2.94 0.04 Plant species 1 2.05x10-2 0.83 0.37

Fungal PLFA (μg g-1 soil) Soil age Plant species

1 3 1

2.15x10-2

7.41x10-2

1.32x10-2

0.90 3.10 0.55

0.34 0.04 0.46

NH4+(mg kg-1) 1 1.36x10-2 0.74 0.40

Soil age 3 6.98x10-2 3.79 0.02 Plant species 1 2.64x10-2 1.43 0.23

Plant-parasitic nematodes (per 100 g-1 soil) 1 1.39x10-2 0.64 0.43 Soil age 3 7.55x10-2 3.45 0.03 Plant species 1 4.93x10-2 2.26 0.16

Microeucaryote PLFA (μg g-1 soil) Soil age Plant species

1 3 1

1.6x10-2

7.58x10-2

1.66x10-2

0.61 2.91 0.64

0.43 0.034 0.41

Microbial biomass-C (μg-C g-1 soil) 1 1.14x10-2 0.49 0.50 Soil age 3 8.05x10-2 3.47 0.03 Plant species 1 3.33x10-2 1.43 0.26

NO3- (mg kg-1) 1 7.9x10-3 0.42 0.50

Soil age 3 7.45x10-2 3.94 0.02 Plant species 1 2.88x10-2 1.52 0.25

Carboxylic acids-CLPP (μg CO2–C g-1 soil h-1) 1 8.1x10-3 0.30 0.58

72

Soil age 3 7.42x10-2 2.73 0.051 Plant species 1 2.1x10-2 0.78 0.42

Total N (%) 1 4.94x10-3 0.20 0.65 Soil age 3 7.72x10-2 3.21 0.04 Plant species 1 2.68x10-2 1.11 0.30

EC (dS m-1) 1 4.1x10-3 0.19 0.66 Soil age 3 9.68x10-2 4.56 0.01 Plant species 1 3.63x10-2 1.71 0.19

Total C-CLPP (μg CO2–C g-1 soil h-1) 1 3.28x10-3 0.13 0.71 Soil age 3 7.33x10-2 2.87 0.055 Plant species 1 2.56x10-2 1.00 0.33

Aromatics-CLPP (μg g-1 soil) Soil age Plant species

1 3 1

2.1 x10-3

7.81x10-2 1.81x10-2

0.10 3.72 0.86

0.74 0.018 0.36

Gram Negative PLFA (μg g-1 soil) Soil age Plant species

1 3 1

2.9 x10-3

7.37x10-2

2.40x10-2

0.11 2.81 0.91

0.74 0.06 0.31

Phenolic acids-CLPP (μg CO2–C g-1 soil h-1) Soil age Plant species

1 3 1

2.15x10-3

7.74x10-2

1.77x10-2

8.23x10-2

2.96 0.680

0.78 0.04 0.44

Available K (mg kg-1) 1 1.2x10-3 6.06x10-2 0.79 Soil age 3 7.98x10-2 4.02 0.02 Plant species 1 2.95x10-2 1.44 0.25

Alcohols-CLPP (μg CO2–C g-1 soil h-1) Soil age Plant species

1 3 1

1.62x10-3

8.6x10-2

1.9x10-2

7.32x10-2

3.89 0.86

0.79 0.02 0.36

Aminoacids-CLPP (μg CO2–C g-1 soil h-1) Soil age Plant species

1 3 1

1.33x10-3

8.05x10-2

2.21x10-2

5.07x10-2

3.07 0.84

0.82 0.024 0.37

Carbohydrates-CLPP (μg CO2–C g-1 soil h-1) Soil age Plant species

1 3 1

1.40x10-3

9.52x10-2

1.32x10-2

5.92x10-2

4.04 0.56

0.83 0.019 0.45

Fungal-feeding nematodes (per 100 g-1 soil) 1 4.11x10-4 1.95x10-2 0.90 Soil age 3 7.39x10-2 3.51 0.03 Plant species 1 2.85x10-2 1.35 0.27

Individual PSF with respect to soil age

No statistical differences were found in the total biomass produced by B. ornata in

home (i.e. soils conditioned in the field by B. ornata) with respect to away soils (i.e.

soils conditioned in the field by E. marginata), therefore a neutral PSF was observed in

all soils from the chronosequence (Figure 8).

73

Figure 8. Total biomass (mean ± SE; n = 6) produced by Bossiaea ornata grown in soils previously conditioned in the field by B. ornata (home soil = Ho) or by E. marginata (away soil = Aw). Soils were collected from a chronosequence of restoration ages (1.5-, 7-, and 22-year-old, unmined soils). The biomass produced by B. ornata in home as compared to away soils did not differ in any of the experimental soils.

DISCUSSION

Physico-chemical and biological properties, and thus the soil feedback of soils from the

chronosequence were all significantly different to each other. These differences were

influenced by soil age (i.e. 1.5-, 7-, 22-year-old, unmined soils), but not by plant species

(i.e. E. marginata vs. B. ornata). The trajectory of change followed by the physico-

chemical and biological properties of field-conditioned soils indicated that soils from

the chronosequence are approaching the conditions of unmined soils. Differential soil

feedback of soils from the chronosequence translated into variable amounts of biomass

produced by B. ornata. The biggest plants were produced in the 22-year-old soils,

whereas the smallest plants were produced in unmined soils. The amount of biomass

produced by B. ornata seedling was significantly influenced by soil age, not by plant

species from which soils were collected in the field. From individual properties, the

0.0

0.1

0.1

0.2

0.2

0.3

0.3

0.4

Ho Aw Ho Aw Ho Aw Ho Aw

1.5-year-old 7-year-old 22-year-old Unmined

Tota

l bio

mas

s (g

)

74

number of bacterial-feeding nematodes and the rate of CO2 produced by microbial

respiration of phenolic compounds from CLPP were evidenced as significant

components of the soil feedback across soils from the chronosequence. These properties

had a strong and significant effect on the biomass produced by B. ornata in all soils.

Differential soil feedback of soils from the chronosequence did not translate into

variable individual PSF, and neutral PSF was associated with B. ornata across all soils.

The physico-chemical and biological properties of field-conditioned soils from the

chronosequence were all different to each other, and significantly influenced by soil age

(i.e. 1.5-, 7-, 22-year-old, unmined soils) as predicted. The time-dependent changing

nature of soils responds to biogeochemical and biological processes (Hobbie 1992;

Eviner 2004; Viketoft et al. 2005; Bezemer et al. 2006; De Bellis et al. 2007; Berg &

Smalla 2009). However, the results agreed (partially) with the initial prediction, since a

significant effect of plant species on soil properties was also expected. Particularly the

physico-chemical properties of the 1.5-year-old soils were not significantly influenced

by plant species as observed in the experiment described in Chapter 3. The discrepancy

in the results observed for this and the previous experiment for the 1.5-year-old soil can

be the result of both E. marginata and B. ornata causing similar changes to soils. This

behaviour was observed previously but only in unmined soils, where similar

conditioning processes of the chemical properties were found by E. marginata and B.

ornata (Chapter 3). Another possibility is that in the field, species-specific conditioning

of soil properties was constrained by strong soil age effects, since soil is a dynamic

evolving entity due to biogeochemical processes, climate influence, and changes

associated to biotic communities (Foth 1990; Lavelle & Spain 2003; Buscot & Varma

2005).

75

Experimentally, the discrepancy observed in case of the previous and this research

might be attributed to differences in soil conditioning processes present in each case.

Soil conditioning is influenced by the experimental procedure used, and for this

research, plant-conditioned soils were collected from single plants in the field, which is

known as the ‘natural experiment approach’ (Kulmatiski & Kardol 2008; Pernilla et al.

2010; van de Voorde et al. 2012), whereas for the previous work soils were collected

from root area of E. marginata, B. ornata, and Acacia pulchella, and from other

legumes, all soils were mixed, distributed into pots and conditioned under controlled

conditions in the glasshouse, this procedure is referred to as the ‘cultivation experiment

approach’ (Kulmatiski & Kardol 2008; Pernilla et al. 2010; van de Voorde et al. 2012).

These findings suggest that conditioning process is context-dependent and shows some

level of plasticity.

Properties such as density, gravel content, and pH decreased as restoration age

increased, probably in response to development of soil structure, stabilization of

aggregates, and biogeochemical processes (Jastrow et al. 1998; Malik & Scullion 1998;

Ward 2000; Banning et al. 2008). The concentrations of available K and EC increased

with increasing soil age, while available P, total N and total C decreased in the 7-year-

old soils as compared to the 1.5-year-old soils, but increased again in the 22-year-old

soils. Restoration of the jarrah forest includes N-P-K fertilisation early in the process

(Koch 2007b), and for this reason these nutrients are in high concentrations at early

stages of restoration as observed in the 1.5-year-old soils. With time, N-P-K

concentrations decrease due to plant uptake and leaching. As plant biomass production

increases and the process of organic matter recycling recovers, the concentrations of

nutrients and C in soil increase (Ward 2000; Banning et al. 2008; George et al. 2010;

Lin et al. 2011) as observed in the 22-year-old soils whose physico-chemical and

biological properties are close to those of unmined soils.

76

Microbial biomass-C increased with restoration age, whereby the oldest soils in the

chronosequence had values that were similar to that of unmined soils, a result which is

consistent with other studies (Banning et al. 2008). From PLFA biomarkers, only those

for Gram positive bacteria and microeucaryotes statistically differed in experimental

soils. Gram positive bacteria were conditioned by E. marginata and B. ornata in a

species-specific manner but only in the 1.5-year-old soils and in unmined soils. In these

two soil ages a higher concentration of PLFA biomarkers for Gram positive bacteria

were found under the influence of E. marginata as compared with B. ornata. Gram

positive bacteria usually form endospores, and their population is large in soils,

consequently, this bacterial group can survive under adverse conditions (Lavelle &

Spain 2003; Büdel 2005; Barriuso et al. 2008), such as those present in soil restored

post-bauxite mining. In degraded soils, Gram positive bacteria recover quickly because

these bacteria are very responsive to C inputs released to soil via root exudates (Jasper

2007; Caesar-TonThat et al. 2008). Bacterial populations developed under early-

successional stages of ecosystems are more sensitive to C inputs from plant species

(Jasper 2007; Papatheodorou 2008), as was observed for the 1.5-year-old soils, and not

for the other soils within the chronosequence. However, since root exudates are

produced in a species-specific (Marschner 1995; Uren 2001), I expected that

populations of Gram positive bacteria would be conditioned according to plant species

as was found during the experiment. Previous works have evidenced the capacity of E.

marginata to condition biotic communities of soil likely via root exudates (Malajczuk &

McComb 1977; Malajczuk 1979; Malajczuk & McComb 1979). In unmined soils, plant

conditioning of soil microbial populations (i.e. Gram positive bacteria) was expected

due to species-specific rhizodeposition released to soil for long time spans, a

phenomenon that has been evidenced by other works (Malajczuk & McComb 1979; De

Bellis et al. 2007; Zechmeister-Boltenstern et al. 2011; Prescott & Graystone 2013).

77

PLFA biomarkers for microeucaryotes were in higher concentration in the 7-year-old

soils but only when this soil age was compared with unmined soils. Since the rest of

comparisons for microeucaryote PLFAs among soils within the chronosequence were

not significantly different, I suggest this is a particular soil property of the 7-year-old

soils (i.e. an outlier).

Functional aspects of the microbial community (CLPP) differentiated only at late stages

of restoration, a result which is consistent with what has been previously observed

(Lalor et al. 2007). Populations of nematodes from diverse trophic groups augmented

with increasing restoration age, maximum numbers were observed in unmined soils

(except for bacterial-feeding nematodes). Plant-parasitic nematodes were exclusively

observed in the latter soils. According to the nematode channel ratio values, both the

bacterial-based as well as the fungal-based energy channel of decomposition were

simultaneously operating in soils across the chronosequence (i.e. nematode channel

ratio close or equal to 0.5). In contrast, in unmined soils a fungal-based energy channel

of organic matter decomposition operated (i.e. nematode channel ratio > 0.5; Moore &

Hunt 1988; Yeates 2003). Individual biological properties of soils from the

chronosequence tended to increase or ‘develop’ with increasing restoration age,

however, even after 22 years post-restoration, there are functional aspects of soil (i.e.

the nematode channel ratio) that are dissimilar from unmined soils.

Individual properties of field-conditioned soils of increasing restoration age tended to

reach the values of the properties of unmined soils as initially predicted. However, the

overlaid trajectory of multivariate change based on the physico-chemical and biological

properties of soils conditioned in the field by E. marginata and B. ornata was not linear.

Hypothetically, a maximum distinction between 1.5-year-old and unmined soils with

regard to physico-chemical and biological properties should be observed, and this

78

distinction should lessen as restoration age increases. Nonetheless distances between

centroids of the 7-year-old soils and the unmined soils were even longer than the

distances between centroids of 1.5-year-old soils and the unmined soils (except by the

biological properties conditioned by B. ornata). These results suggest that the properties

and thus the soil feedback of restored soils post-bauxite are recovering with increasing

time, however soils from the 7-year-old site selected for this experiment might be

following an alternative trajectory of change, and statistically this site can be considered

as an outlier (Suding et al. 2004; Grant 2006).

There was a differential soil feedback effect on the biomass produced by B. ornata

grown in soils from the chronosequence. Plant biomass was significantly influenced by

soil age as predicted, however any significant effect of plant species was observed.

These findings differed from the results observed in the experiment described in the

previous chapter, where plant species previously grown in soils had a significant effect

on the biomass produced by B. ornata in the 1.5-year-old soils (see Chapter 3).

However, as explained above, the experimental approach used in each one of the

experiments might have influenced the results (i.e. cultivation experiment approach

used in experiment described in Chapter 3 vs. natural experiment approach used in this

experiment; Kulmatiski & Kardol 2008; Pernilla et al. 2010; van de Voorde et al. 2012).

Differences in biomass produced by B. ornata in soils from the chronosequence suggest

that biomass production was influenced by changing contexts of soils (Manning et al.

2008; te Beest et al. 2009; Meijer et al. 2011). The 22-year-old soils exerted a positive

feedback on B. ornata that produced up to two times more biomass with respect to the

other soils. The values for the properties of the 22-year-old soils were increased, and

approached those of unmined soils, therefore, similar amounts of biomass were

expected to be produced by B. ornata in the 22-year-old and in unmined soils. However,

79

plants grown in the 22-year-old soils were significantly bigger that those grown in

unmined soils. The more marked differences between the 22-year-old soils and unmined

soils were the presence of plant-parasitic nematodes in the latter, and increased numbers

of bacterial-feeding and predatory nematodes in the 22-year-old soils. Therefore, the

increased biomass produced by B. ornata in the 22-year-old soils might respond to the

absence of plant-parasitic nematodes, and faster organic matter decomposition and

turnover due to the presence of bacterial-feeding and predatory nematodes. Plant-

parasitic nematodes can cause severe damage to plants due to root damage (Perry &

Tovar-Soto 2012), and are drivers of negative PSF in ecosystems (De Rooij-van der

Goes 1995; van der Putten & Peters 1997; van der Stoel et al. 2002). Bacterial-feeding

nematodes can mineralize a substantial amount of nitrogen that is available for plant

uptake (Freckman 1988), therefore can be considered drivers of positive PSF.

The number of bacterial-feeding nematodes and the rate of CO2 produced by microbial

respiration of phenolic compounds from CLPP were identified as significant

components of the soil feedback of soils from the chronosequence. These properties

showed a strong and significant influence on the biomass produced by B. ornata across

soils from the chronosequence. This finding supports the results of the increased

biomass produced by B. ornata in the 22-year-old soils, which might be a response of

increased bacterial-feeding nematodes in soil. Bacterial-feeding nematodes display

characteristics that make them an important group of soil organisms. They directly

influence plant productivity by releasing ammonia (i.e. the major excretory product of

nematodes) that is available for plant uptake (Freckman 1988; Yeates 2003; Postma-

Blaauw et al. 2005). Bacterial-feeding nematodes are diverse in ecosystems and

characteristics such as a short generation time and high fecundity allow these organisms

to quickly recover after soil disturbances (Freckman 1988; Bongers & Bongers 1998;

Yeates 2003; Holtkamp et al. 2008), including restoration post-bauxite mining.

80

Consequently their influence on plant productivity might start at early stages of plant

development and might persist at late stages of restoration. In the mid- and long-term,

bacterial feeding nematodes indirectly affect plant growth by dispersing microbes,

feeding on pathogenic bacteria, and being source of nutrients for other organisms within

the soil food web (Freckman 1988; Wardle 1999).

Three fenolic compounds were tested as substrates for CLPP, these were catechin,

coumarin and quercetin. These phenolic compounds are exuded by roots, and exhibit

antimicrobial and allelopathic properties (Lavelle & Spain 2003; Singh & Mukerji

2006; Bais et al. 2008; Hadacek 2008; Badri & Vivanco 2009). Chemicals with

antimicrobial and allelopathic action have been detected in exudates and litter of

Australian plants (Whitfield 1981; Deans 2002; D’Souza et al. 2004; Hussain et al.

2011; Lorenzo et al. 2013). Phenolic compounds released to soil via exudates and litter

are degraded (respired) by several microorganisms, for example pseudomonad (Inderjit

2005; Reuben et al. 2008; Zhang et al. 2009). Microbial degradation of C compounds is

substrate-dependent, my results indicated that phenolic compounds have been present in

soils and consequently degraded by soil microbes as evidenced by CLPP results. This

finding is expected because in jarrah forest, production of allelochemicals is critical for

plant persistence, since competition with other plants and soil organisms for moisture

and nutrients is intense, thus plants equipped with efficient inhibitory chemicals raise

their chances of survival (Willis 1999; Manoharachary & Mukerji 2006). I suggest that

respiration of phenolic compounds was evidenced as a key component of the soil

feedback because this is an indirect way to observe allelopathic properties of jarrah

forest plants, which is a key phenomenon for plant establishment and development (i.e.

biomass production) in jarrah forest.

81

The fact that biological properties were found to be components of the soil feedback of

soils from the chronosequence disagree with my initial prediction, I expected to observe

mainly physico-chemical properties significantly influencing the biomass produced by

B. ornata via soil feedback in recently-restored soils. Then, as restoration age increased,

I expected to observe mainly biological properties significantly influencing the biomass

produced by B. ornata via soil feedback at late-restored soils. This prediction was

formulated given that recently-restored soils display chemical properties such as

increased concentrations of P, N, and K that are consequence of fertiliser addition as

part of the restoration process (Koch 2007b). Biological properties of soil are likely to

influence plant growth via soil feedback in the mid-, long-term, and not in the short-

term due to the dependence of native plants on biologically-supplied nutrients such as N

and P (Lawrie 1981; Brundrett & Abbott 1991; Lambers et al. 2007), and decreasing

concentrations of nutrients from fertilisers.

Differential soil feedback of soils from the chronosequence did not translate into

differential individual PSF (i.e. magnitude and direction), that is, a neutral PSF was

associated with B. ornata. The results for the 22-year-old and unmined soils are

consistent with the prediction of neutral PSF associated with B. ornata. However, PSF

associated with B. ornata grown in the 1.5-year-old soils disagreed with the prediction

of negative PSF based on results from a previous experiment (see Chapter 3). In both,

this and the previous experiment recently-restored soils were the same age (i.e. 1.5-

year-old soils), however the results suggested that conditioning process is influenced by

soil contexts (Kulmatiski & Kardol 2008; Pernilla et al. 2010; van de Voorde et al.

2012). In the previous experiment soils were collected from different plants species,

mixed, and conditioned in the glasshouse, whereas for this research soils were collected

from single plants in the field, thus soils were conditioned under field conditions.

82

In conclusion, the results from the characterization of soils from the chronosequence

indicated that soil feedback of restored soils post-bauxite mining is recovering, but

alternative trajectories can occur as observed for the 7-year-old soil. However, these

results did not translate into altered patterns of biomass produced by B. ornata in the 7-

year-old soil. Instead, an increased biomass production was observed in the 22-year-old

soils, which suggested this soil age exerted a differential and positive soil feedback.

Despite this, individual PSF associated with B. ornata was not affected, and neutral PSF

was observed in all soils from the chronosequence.

83

CHAPTER 5

Long-term conditioning of soil by eucalypts and pines does not alter

the plant-soil feedback associated with native jarrah

INTRODUCTION

Plant-soil feedbacks can influence the restoration of native forests where soil has been

conditioned by non-native species for long periods of time (i.e. agricultural, forestry

activities, invader species; Kulmatiski et al. 2006; Jordan et al. 2008; Kulmatiski &

Beard 2008). Specifically, non-native species can condition soil properties in ways that

lead to reduced subsequent growth of native species (van Grunsven et al. 2007; Perkins

& Nowak 2012; Suding et al. 2013), due to alterations of nutrient cycling processes

(Bezemer et al. 2006; Kulmatiski et al. 2006; Kardol & Wardle 2010; Weidenhamer &

Callaway 2010), imbalance of microbial populations (i.e. free-living, plant-pathogens

and root symbionts; Indejit & van der Putten 2010), and negative effects of toxic

allelochemicals (Bertin et al. 2003; Badri & Vivanco 2009; Indejit & van der Putten

2010; Weidenhamer & Callaway 2010). Plant-conditioned properties of soil can persist

even after the removal of the non-native species (Kardol & Wardle 2010).

Soil conditioning operates through two mechanisms: 1) leaf litter deposition, and 2) root

deposition. In particular, litter quality is a species-specific plant trait (i.e. C:N ratio,

lignin and phenolic compounds content) that directly influences the chemical and

biological properties of soil. Elements released from litter during decomposition alter

the concentrations of soil macro and micronutrients in a species-specific manner

(Hobbie 1992; Guo & Sims 1999; Berg & McClaugherty 2008; Witt & Setälä 2010;

Uselman et al. 2012). As well, litter deposition drives changes in the microbial and

microfauna (i.e. nematode) community, including its size, composition, and

physiological properties (Berg & McClaugherty 2008). Generally, plant species that

84

produce labile litter (i.e. low C:N ratio) such as legumes, promote a microbial

community dominated by bacteria and their consumers (i.e. protozoa, bacterial-feeding

nematodes). In contrast, plant species that produce recalcitrant litter (i.e. high C:N ratio)

such as pines and eucalypts promote a microbial community dominated by fungi and

their consumers (i.e. fungal-feeding nematodes, collembolans, oribatid mites; Moore &

Hunt 1988; Scheu et al. 2005; Moore et al. 2007; Holtkamp et al. 2008; Witt & Setälä

2010).

In the jarrah forest of south-west Western Australia, the first restoration attempts post-

bauxite mining consisted of commercial plantations of fast-growing, exotic softwoods,

mainly pines (Pinus radiata) and cypresses (Cupressus sp.), as well as eastern

Australian hardwood eucalypts such as Sydney blue gum (Eucalyptus saligna; Koch

2007a; Koch 2007b). To date there are 3,600 ha of plantations of non-native species

within the jarrah forest ecoregion that will be harvested for timber in the near future and

then restored to native jarrah forest (Conservation Commission of Western Australia

2012). At these sites non-native trees have been conditioning the soil properties for

decades. The general aim of my research was to determine what differences (if any)

occurred in the soils after 35 years of conditioning by non-native species, and to

determine the effect of conditioning on the subsequent growth of jarrah (E. marginata)

seedlings (i.e. individual PSF).

Specifically, the research aimed to determine changes in properties of mined soils that

have been conditioned in the long-term by pines (P. radiata), Sydney blue gums (E.

saligna; both non-native, plantation trees), and jarrah. It was expected that conditioning

process of soil by the three plant species would be primarily driven by litter deposition,

thus influenced by litter characteristics of each species in particular. Pines and eucalypts

both produce recalcitrant litter, nonetheless, differences in conditioning processes occur

85

between the two genera associated with changes in the soil chemistry and biology

owing to differences in litter quality. Conditioning of soil by pines via litter deposition

includes acidification of soil, alteration of concentrations such as N, K, Mg, and Ca, and

low microbial activity (Feller 1983; Lavelle & Spain 2003; Ehrenfeld et al. 2005). In

contrast, allelopathy has been described as a key mechanism driving soil conditioning

by eucalypts. Litter extracts from eucalypts inhibit the germination of a diversity of

plant species as well as the growth of microorganisms such as nitrifying bacteria

(Bezuidenhout & Laing 2006; Kohli et al. 2006). For this investigation it was predicted

that the plant species (pine, jarrah and blue gum) would condition the soil in a species-

specific manner (Aponte et al. 2013), however, differences in soil conditioning would

be more marked after pine than eucalypt growth since the former presents litter with

composition that is different from eucalypts. In contrast, differences in conditioning

between blue gums and jarrah would be less distinct due to similar litter quality given

that both belong to the same family.

There is the potential for pines and eucalypts to condition soils differently but few tests

of the influence of the conditioned soils on plant growth, and in particular how

conditioning of forest soil by non-native species will influence the growth of native

species such as jarrah via soil feedbacks. This issue was addressed in the current study.

Jarrah displays a high degree of growth plasticity in response to resource availability

and the soil environment (Bleby et al. 2009). This plasticity suggests that E. marginata

is capable of producing constant biomass even under variable soil conditions. Therefore,

the second specific aim was to test PSF associated with jarrah plants grown in soils

conditioned by non-native plants by comparing their growth in soils conditioned by the

same species. I predicted that species-specific conditioning of soil by non-native plants

would little affect the biomass produced by jarrah seedlings (i.e. a neutral PSF would be

observed).

86

MATERIALS AND METHODS

Sampling site and experimental design

Soils for the experiment were collected from a site that had been cleared of jarrah forest,

then mined for bauxite, and then planted with diverse native and non-native eucalypt

species, pines and cypresses to create an arboretum in 1976 (i.e. 35 years prior to

sampling). Soil samples were collected in June 2011. At the site, six pines (P. radiata)

(Pinaceae), six jarrah (E. marginata) and six blue gums (E. saligna; both Myrtaceae)

were selected at random and four intact soil cores (0.785 L; diameter = 10 cm, depth =

10 cm) were taken from beneath the canopy area of each tree where litter is deposited.

Two cores were used to measure the physico-chemical and biological properties of soils

to determine species-specific conditioning (Figure 1A). These soils were sieved (2-mm

mesh) and stored at 4 °C pending analyses, and a subsample of soil was frozen (−20 °C)

and then subsequently freeze-dried pending phospholipid-fatty acid analysis (PLFA).

The two remaining cores were placed in bottom-sealed pots (1.2 L; diameter = 13.3 cm,

height = 12.4 cm) ready to use in the pot experiment.

Pots containing the intact soil cores were placed into a root cooling tank that was

housed in a glasshouse with temperatures ranging between 4.5 and 27.8 °C for the

duration of the experiment. To produce seedlings, seeds were surface-disinfected with

ethanol (70%) and bleach (50%) to prevent growth of pathogens and germinated at 14

°C in darkness. Several jarrah seedlings were planted in each pot, and once a plant

established, the superfluous seedlings were removed leaving only one plant per pot. As

a result there were jarrah seedlings growing in soil cores collected in the field from

jarrah trees (i.e. home soils) and in soil cores collected from pines or blue gums (i.e.

away soils). Pots were rotated within the tank every two weeks to negate a block effect

and watered once a week to 60% water-holding capacity. After four months growth

87

(from June to October 2011), plants were harvested; shoots and roots were removed,

oven-dried for one week at 60 °C and weighed. Plant biomass produced in home and

away soils was compared to determine individual PSF (magnitude and direction; Figure

1B).

Figure 1. Experimental design where intact soil cores were collected from beneath canopy of pines, jarrah and blue gums. The sampling site was mined for bauxite and planted with the pines, jarrah and blue gums 35 years earlier. One half of the cores were used to determine the properties conditioned in the field by the three tree species for 35 years by measuring their physico-chemical and biological soil properties (A). The second half of the soil cores were planted with jarrah seedlings in a home (i.e. conditioned by jarrah) and away (i.e. conditioned by pines or blue gums) arrangement. After four months, plants were harvested; the total biomass produced was measured and compared in home with respect to away soils to determine individual PSF (B).

Response variables

Refer to chapter two for methodologies used to determine response variables measured.

Statistical analyses

To determine the evidence for species-specific conditioning of field soils (Figure 1A)

the physico-chemical and biological properties (including PLFAs, CLPPs and nematode

trophic groups) of soils were compared using multivariate and univariate permutational

analyses of variance (PERMANOVA) followed by pair-wise tests. Although

PERMANOVA is a multivariate method, it can be used for univariate analyses. To

B

Intact soil cores conditioned by plant species in the field

Pines Plant-soil feedback

Biomass production home vs.

away soils

Jarrah seedlings

Jarrah Blue gums

Properties of field-conditioned

soils

A

Pine soil cores

Jarrah soilcores

Blue gum soil cores

88

perform PERMANOVA using one response variable and Euclidean distance yields

Fisher’s traditional univariate F-statistic (Anderson 2010). Scaling plots from principal

co-ordinate analysis (PCO; Anderson 2001) were built to visualise the results of

PERMANOVA analyses.

To determine individual PSF (i.e. magnitude and direction; Figure 1B), I used univariate

PERMANOVAs with total, shoot and root biomass produced in home vs. away soil as

response variables. A significant increase of the plant biomass production in home with

respect to away soil was considered a positive PSF. A significant decrease of the plant

biomass production in home soil with respect to away soil was considered a negative

PSF. No difference in biomass production in home with respect to away soil was

considered a neutral PSF.

For all the statistical analyses PRIMER v6 +PERMANOVA (PRIMER-E Ltd) was

used. For the multivariate analyses, permutation of residuals under a reduced model

was selected as permutation method as well as type III (partial) sum of squares. The

unrestricted permutation of raw data as permutation method and type III (partial) sum

of squares were used for the univariate analyses. In all cases data were transformed

when necessary and Euclidean distance was used to build resemblance matrices, except

for the analyses of CLPPs and the trophic groups of the nematode community when

Bray-Curtis coefficient and Manhattan distance were respectively utilized to build the

similarity matrices.

RESULTS

Properties of soils conditioned in the field by pines, jarrah, and blue gums

The physico-chemical properties of soils conditioned in the field were significantly

different according to plant species (F2, 26 = 2.83, P = 0.003; Figure 2a). Pair-wise

89

comparisons showed that the physico-chemical properties of soils conditioned by jarrah

differed from those conditioned by blue gums (t = 1.74, P = 0.031), and pines (t = 1.91,

P = 0.034), blue gums and pines were not significantly different (t = 1.36, P = 0.06;

Figure 2a). The biological properties of field soils were differentially conditioned

according to plant species (F2, 16 = 2.24, P = 0.007). Pair-wise comparisons showed that

the biological properties of soils conditioned by pines differed from those conditioned

by jarrah (t = 1.72, P = 0.01) and by blue gums (t = 1.73, P = 0.013), and that soils

conditioned by jarrah and blue gums were not significantly different (t = 1.005, P =

0.45; Figure 2b).

Figure 2. Scaling plots from principal co-ordinate analysis (PCO) of a) physico-chemical properties: bulk density (g cm-3), total C and total N (%), C:N ratio, available P and K (mg kg-1), EC (dS m-1), pH (CaCl2), NO3

- and NH4+ (mg kg-1) and, b) biological properties: microbial

biomass-C (µg g-1 soil), total CO2 produced from CLPPs (µg CO2 g-1soil h-1), bacterial-, fungal-feeding, plant-parasitic, predatory, and omnivorous nematodes (all per 100 g-1 soil), and nematode channel ratio, of soils conditioned in the field by pines, jarrah, and blue gums. The plots show that jarrah differently conditioned the physical-chemical properties of soils, whereas pines differentially conditioned the biological properties of soils.

Statistical analyses of individual physico-chemical properties revealed differences in

species-specific conditioning of the C:N ratio (F2, 2.34 = 21.11, P = 0.0002), pH (F2, 0.04 =

90

3.88, P = 0.042) and K concentration (F2, 37.7 = 18.2, P = 0.001). The C:N ratio of soils

conditioned by the three species were all significantly different (jarrah versus pine t =

6.02, P = 0.0016; jarrah versus blue gum t = 3.52, P = 0.0081; pine versus blue gum t =

3.24, P = 0.0113; Figure 3a). Soil pH conditioned by blue gums statistically differed

from pH of soils conditioned by pines (t = 2.78, P = 0.024; Figure 3b), and K

concentration of soils conditioned by the three species were statistically different from

each other (jarrah versus pine t = 2.63, P = 0.042; jarrah versus blue gum t = 5.14, P =

0.004; pine versus blue gum t = 3.93, P = 0 004; Figure 3c).

Statistical analyses of individual biological properties revealed species-specific

conditioning of the numbers of fungal-feeding nematodes (F2, 1904 = 3.77, P = 0.03),

omnivorous nematodes (F2, 5219 = 4.11, P = 0.039), and the nematode-channel ratio (F2,

0.06 = 4.6, P = 0.02). The numbers of fungal-feeding nematodes significantly differed in

soils conditioned by pines as compared with soils conditioned by jarrah (t = 2.41, P =

0.029; Figure 3d) and by blue gums (t = 2.21, P = 0.036). The numbers of fungal-

feeding nematodes present in soils conditioned by the two eucalypt species were not

significantly different from one another (t = 0.49, P = 0.76). The number of omnivorous

nematodes significantly differed in soils conditioned by pines with respect to soils

conditioned by jarrah (t = 3.05, P = 0.015; Figure 3e), but not with respect to soils

conditioned by blue gums (t = 1.91, P = 0.07). The numbers of omnivorous nematodes

present in soils conditioned by the two eucalypt species were not significantly different

(t = 0.69, P = 0.47). The nematode channel ratio conditioned by pines was different to

that conditioned by jarrah (t = 2.78, P = 0.033; Figure 3f) and by blue gums (t = 2.54, P

= 0.0188 respectively). Soil conditioned by pines presented the highest nematode

channel ratio.

91

Figure 3. Properties of soils (mean ± SE; n = 6) that were differently conditioned (P < 0.05) by pines (Pinus radiata), jarrah (Eucalyptus marginata), and blue gums (Eucalyptus saligna) in the field. Means with the same letters were not significantly different. Numbers of fungal-feeding and omnivorous nematodes are reported per 100 g of dry soil. NCR = nematode channel ratio.

No significant differences were observed in the total PLFAs (i.e. thirty nine different

PLFAs; F2, 5.9 = 0.73, P = 0.61), in biomarkers for total bacteria (F2, 4.8 = 0.76, P =

0.57), Gram positive bacteria (F2, 0.7 = 0.86, P = 0.5), Gram negative bacteria (F2, 0.4 =

0.51, P = 0.72), total fungi (F2, 1.82 = 0.57, P = 0.69), or microeucaryotes (F2, 0.28 = 0.33,

P = 0.89) of the three field-conditioned soils, thus there was no evidence of plant

92

conditioning of the structure of the microbial community. In contrast, species-specific

conditioning was observed for CLPPs (F2, 94 = 3.58, P = 0.0284; Figure 4), pair-wise

tests revealed that CLPPs conditioned by pines significantly differed from CLPPs

conditioned by blue gums (t = 2.17, P = 0.043), but CLPPs conditioned by pines did not

differ from CLPPs conditioned by jarrah (t = 1.85, P = 0.086). CLPPs conditioned by

jarrah and blue gums were similar (t = 0.96, P = 0.4; Figure 4). In general, the microbial

community conditioned by pines metabolized the C-substrates from CLPP at lower rates

as compared with the other two plant species. Refer to figure 5 for a summary of the

effects of pines, jarrah and blue gums on the physico-chemical and biological properties

of the jarrah-forest soils.

Figure 4. Scaling plot from principal co-ordinate analysis (PCO) of CLPPs of soils conditioned in the field by pines, jarrah, and blue gums. Thirty-one different C substrates were tested and used for the analyses. The plot shows significant differences between CLPPs of soils from pines and CLPPs of soils from blue gums (P < 0.05). The rest of comparisons between pairs of treatments were not significantly different.

93

Figure 5. Schematic diagram of soil conditioning via litter deposition by a) pine, b) jarrah, and c) blue gum. It was assumed that conditioning process of field soils by plant species was mostly due to litter deposition due to the location of the sampling but this assumption needs testing. Boxes represent soil properties that were measured: light gray for chemical and darker shades of gray indicate higher trophic levels. Ovals represent soil processes (inferred rather than measured). Arrows are drawn to represent the likely relationships among the soil properties. Plus sign (+) indicates a positive feedback regulates the interaction between properties, a minus sign (-) indicates a negative feedback regulates the interaction between properties.

94

Individual PSF

Jarrah seedlings grown in soils conditioned in the field by pines, blue gums (i.e. away

soils), and jarrah (i.e. home soils) produced amounts of biomass that were not

significantly different from each other (Figure 6). Total biomass (F2, 0.2 = 0.91, P =

0.418), shoot biomass (F2, 0.02 = 1.99, P = 0.17), and root biomass (F2, 0.003 = 1.33, P =

0.28) of jarrah seedlings did not differ in soils in soils conditioned by the three plant

species, thus neutral PSF was associated with jarrah plants.

Figure 6. Shoot and root biomass (means ± SE; n = 6) produced by jarrah seedlings grown in away soils (Aw = conditioned by pines and blue gums) and home soils (Ho = conditioned by jarrah).

DISCUSSION

Field-based conditioning of soil properties was observed in soils collected beneath

pines, jarrah and blue gums. A first prediction was that both eucalypt species would

similarly condition the properties of soil and that these species would condition soils

differently from pines. The results supported the prediction for the biological properties

(i.e. CLPPs, numbers of fungal-feeding nematodes, omnivorous nematodes, and

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

Pine (Aw) Jarrah (Ho) Blue gum (Aw)

Tota

l bio

mas

s (g

)

Root biomass

Shoot biomass

95

nematode channel ratio), these soils were similarly conditioned by both eucalypt species

and conditioning by these trees differed from conditioning of soil by pines.

Nevertheless, the initial prediction was not supported by results of the physico-chemical

properties because these were differentially conditioned by the two eucalypt species. In

contrast, blue gums and pines similarly conditioned the physico-chemical properties of

soils. Especially relevant were the differences in conditioning of the properties C:N

ratio, pH, and available K. Despite the differences, species-specific conditioning of soil

did not translate into differences in the amounts of biomass produced by jarrah

seedlings and neutral PSF was observed as predicted. This result suggests that restoring

native jarrah trees to soils that have been conditioned by non-native species for decades

may not be as problematic as envisaged.

It was assumed that differences in properties of field-conditioned soils grown with the

three plant species were mainly the result of conditioning via leaf litter deposition rather

than root deposition. Jarrah trees produce a tap root system, from which lateral roots

and sinker roots emerge. Soil samples were taken from old tress (35-years-old), whose

roots are well developed (Szota et al. 2007). Because root exudation occurs mainly in

root hairs from younger roots (Bertin et al. 2003), it was assumed that in the first 10 cm

deep of soil, from which samples were collected, the influence of root deposition was

minimal. Moreover, soil samples were collected from beneath tree canopy where litter

deposition has occurred for 35 years. consequently, the main mechanism of

conditioning observed for these trees in the first 10 cm of soil was via leaf litter.

Conditioning of soil via litter deposition is directly related to the chemical composition

of litter (Hobbie 1992; Guo & Sims 1999; Berg & McClaugherty 2008; Witt & Setälä

2010; Uselman et al. 2012). For example, litter with high C:N ratio causes an increase

in the C:N ratio of soil and vice versa. Soils collected from beneath pines can have

higher C:N ratios (i.e. 28 – 36; O’Brien et al. 2003; Hopmans & Elms 2009) than soils

96

collected from beneath eucalypts (i.e. 14–23; Adams & Attiwill 1986; O’Brien et al.

2003) in response to differences in litter composition. These results are in keeping with

reports for soils previously collected from pines (i.e. C:N of 30), whereas soils collected

from jarrah and blue gums sites presented a C:N ratio within the range of 20–25 that

was slightly higher than the value reported in the literature.

Generally, litter with C:N ratios between 20 and 30 supply the necessary nutritional N-

requirements for soil microbes avoiding immobilization by microbial biomass, which in

turn, can limit plant growth (Foth 1990; Lavelle & Spain 2003; Johnson 2009; Hopkins

et al. 2010). Since soils collected from pines had the highest mean C:N ratio (i.e. 30) of

all soils, it was expected for the microorganisms inhabiting these soils to be more N-

limited as compared with those inhabiting soils conditioned by eucalypts. The results

supported this assertion; soils conditioned by pines presented the lowest microbial

respiration rates (from CLPPs) of the three field-conditioned soils, which indicated a

low microbial activity. Chemical analyses of soil revealed that the amounts of total N,

NO3–, and NH4

+ did not significantly differ in soils conditioned by pines from soils

conditioned by jarrah or blue gums. Therefore, N limitations in pine soils may be a

result of inaccessibility to N rather than low concentration of the nutrient. Pines can

release polyphenols that bind to organic N, then pines can utilize this N that is

inaccessible to other plants and microbes (Ehrenfeld et al. 2005). In soils conditioned by

eucalypt species, lower soil C:N ratios translated into higher rates of respired-C, which

indicated that soil microbes were less limited by N resources and were metabolically

more active than microbes in soils conditioned by pines.

The addition of leaf and woody litter to soil can cause soil pH to change due to release

of base cations, organic anions and N during litter decomposition (Bhupinderpal-Singh

& Rengel 2007; Berg & McClaugherty 2008). Due to species-specific litter chemistry,

97

changes in soil pH driven by litter deposition are also species-specific. The three plant

species used in the experiment produced changes in soil pH of different magnitude.

Soils collected from beneath pines presented the lowest pH of all soils (i.e. 4.0), which

is consistent with information reporting acidification of soil by this species (Feller 1983;

Laskowski et al. 1995; Lavelle & Spain 2003). The results also indicated that pH of

soils conditioned by jarrah and blue gums were less acidic than soils conditioned by

pines. Both eucalypt species conditioned soil pHs that were significantly different to

each other (i.e. pH = 4.5 for jarrah, and pH = 5 for blue gums) but that fall within the

range reported for eucalypt species (i.e. pH range of 3.8 – 6; Feller 1983; Adams &

Attiwill 1986; O’Brien et al. 2003; Turner & Lambert 2008; Spain et al. 2014). Species-

specific conditioning of soil pH by plants can, in turn, drive differential chemical and

biological properties of soil (Jones 1998; Uren 2001; Lavelle & Spain 2003; Bais et al.

2008; Richardson et al. 2009; Madigan et al. 2012). Overall, differences in litter quality

may ultimately impact structural and functional aspects of soil grown with pines, jarrah

and blue gums.

The structure of the microbial communities did not differ among the field-conditioned

soils. However, these findings could be a result of differential populations of microbial

predators, as the numbers of fungal-feeding and omnivorous nematodes were

significantly higher in soils conditioned by pines than in soils conditioned by the

eucalypt species. Increased populations of fungal-feeding and omnivorous nematodes

indirectly indicated an increased fungal biomass produced in soils conditioned by pines.

Soils conditioned by the two eucalypt species presented lower numbers of fungal-

feeding nematodes that were associated with higher pHs and lower C:N ratios as

compared with soil conditioned by pines. Understanding the interactions among the soil

food web is critical to being able to interpret these data.

98

The results for the nematode channel ratio indicated that an energy channel of organic

matter decomposition driven by the fungal compartment was operating in soils

conditioned by pines (i.e. nematode channel ratio = 0.88), which probably reflects the

recalcitrance of pine litter (Moore & Hunt 1988; Moore et al. 2003; Condron et al.

2010; Witt & Setälä 2010). Autochthonous fungi possess the unique ability to

decompose recalcitrant organic matter in a variety of ecosystems at slow rates (Lavelle

& Spain 2003; de Boer et al. 2005; Berg & McClaugherty 2008; Condron et al. 2010).

Both the bacterial- and fungal-based energy channels were operating in soils

conditioned by blue gums (i.e. nematode channel ratio = 0.57). Since the bacterial-based

energy channel is faster than the fungal one, quicker rates of organic matter

decomposition were occurring in soils conditioned by blue gums, probably in response

to the labile litter (Moore & Hunt 1988; Moore et al. 2003; Yeates 2003; Condron et al.

2010; Witt & Setälä 2010). Finally, the fungal-based energy channel was favoured over

the bacterial-based channel in soils conditioned by jarrah (i.e. nematode channel ratio =

0.7), which indicated a rate of organic matter decomposition that is intermediate

between pines and blue gums. Differences in organic matter decomposition rates

occurring in soils conditioned by the three plant species may have implications on soil

processes and ultimately on plant community structure and evolution (Hobbie 1992;

Eviner 2004; Weidenhamer & Callaway 2010).

Potassium is a cation that plays key roles in plant metabolism such as enzyme

activation, protein synthesis, photosynthesis, osmoregulation, cell extension, stomatal

movement, and phloem transport (Marschner 1995). Potassium is deposited in soil by

leaching and/or catabolic activity of microorganisms during decomposition of litter

(Ganjegunte et al. 2005; Berg &McClaugherty 2008). Amounts of K in soils are

expected to vary for two main reasons: the K concentrations present in fresh litter and

the rate of litter decomposition. Soils conditioned by blue gums, pines and jarrah,

99

presented the highest, intermediate, and the lowest concentrations of K respectively.

Litter from blue gums is reported to contain medium to high concentrations of K (i.e.

4.39–7.35 g kg-1; Lugo et al. 1990; Binkley & Ryan 1998), which might be quickly

released to soil due to an increased rate of litter decomposition (i.e. as suggested by the

lowest nematode channel ratio value of the three plant species). Litter from jarrah is

reported to contain lower K concentrations (i.e. 2.6 – 5.7 g kg-1; O’Connell et al. 1978;

Hingston et al. 1981), and according to the nematode channel ratio, litter from jarrah

might decompose at lower rates. These two aspects translated into low concentrations of

K in soils conditioned by jarrah plants. Litter from pines is reported to contain high K

concentrations (i.e. 7.31 g kg-1; Hopmans & Elms 2009), but the decomposition rate of

this litter type is slow (i.e. as suggested by the highest nematode channel ratio). As a

result, K concentrations in soil grown with pines were not as high as those conditioned

by blue gums.

The differences in the properties of soils conditioned in the field did not affect

individual PSF (i.e. magnitude or direction) and neutral PSF was observed for jarrah

seedlings during the pot experiment. This finding agreed with the initial prediction of

neutral PSF associated with jarrah. Previous work demonstrated that jarrah shows a high

aboveground architectural plasticity in response to varying environmental conditions

(Bleby et al. 2009). The results of this research suggest this plasticity is not exclusive to

the aboveground compartment, but to the belowground compartment as well, and is not

exclusive to morphological aspects but also to functional aspects. For that reason, jarrah

plants are able to maintain a stable biomass production despite the changing soil

properties, even those conditioned by different plant species. This ability is evident in

the field where jarrah establishes and grows well in non-disturbed but also in soil post-

bauxite mining (Koch & Samsa 2007), which presents modified properties due to

mining and restoration processes (i.e. reduced biological activity, pH, and C, and

100

increased P, N, K and gravel percentage; Ward 2000; Jasper 2007; George et al. 2010;

Tibbett 2010). Plant species displaying plastic growth responses including neutral PSFs

such as jarrah are ideal to use during restoration strategies. This trait allows restored

plants to establish, grow, and persist in restored sites.

In conclusion, non-native plant species can condition soils in ways that differ to native

species, which has implications for ecosystem functioning. This is critical for the

restoration of human-modified ecosystems, since the ultimate goal of most restoration

activities is a complete functional ecosystem. In restored jarrah forest post-bauxite

mining, despite alterations to soil properties are caused by non-native species, jarrah

seedlings were able to establish in these conditioned soils, without it affecting their

biomass production. This suggests that the jarrah–soil system displays a plastic PSF, a

trait of jarrah that might contribute to its persistence for long time spans, and to adapt to

modified soil conditions present in restored soil where jarrah establishes and grows

well.

101

CHAPTER 6

Concluding discussion

This thesis contributed to understand the influence of plant species (Chapter 3 & 5) and

soil conditions (Chapters 4 & 5) on PSF operating in jarrah forest. Plant-soil feedback

was examined in the contexts of unmined soil and restored soil post-bauxite mining.

Plant species of the jarrah forest contribute to PSF by conditioning the properties of soil

in species-specific manner (Chapter 3). The three plant species tested - Eucalyptus

marginata, Acacia pulchella, and Bossiaea ornata, conditioned the soil via changes to

chemical properties (Chapter 3; Figures 2 & 3, Chapter 4; Figure 2). Species-specific

conditioning of biological properties was evidenced only in soils from the field (Chapter

4; Figures 3 & 4), not in soils conditioned in the glasshouse. This finding indicated that

when studying PSF phenomena, the experimental approach used (‘natural experiment

approach’ vs. ‘cultivation experiment approach’; Kulmatiski & Kardol 2008; Pernilla et

al. 2010; van de Voorde et al. 2012) is a critical aspect of the research.

How jarrah forest plants condition the soil is likely to influence their establishment and

persistence in the field. For example, plant species such as E. marginata and B. ornata

conditioned the concentration of P efficiently when grown in unmined soil (Chapter 3;

Table 1). The conditioning capacity of soil P might be related to the long persistence of

both plant species (Abbott 1984; Bell 2001) given the low concentration of this nutrient

in jarrah forest soil (Hingston et al. 1989). Conditioning capacity of P was not altered

when the plants grew in restored soil (Chapter 3; Table 2), which suggests that

conditioning of P is a preserved plant trait that might help maintaining the flux of

nutrients in the ecosystem (Eviner & Hawkes 2008). Therefore after mining, restoration

of long-lived species such as E. marginata and B. ornata might be critical for the re-

establishment of the whole ecosystem functioning.

102

Acacia pulchella showed an inefficient capacity to condition P of soil, either unmined

or restored (Chapter 3; Tables 1 & 2). This trait might contribute to this species being

short-lived, A. pulchella populations decline four years after a disturbance such as fire

(Monk et al. 1981). Acacia pulchella conditioned the total N of unmined soil, through

nitrogen-fixation which is a trait generally recognised for its improvement of the soil

fertility. However, this conditioning capacity was affected by modified soil conditions

present in restored soil (Chapter 3; Tables 1 & 2). Hence it is likely soil functions

related to the N cycle could be affected, as previously suggested by studies that have

determined a conservative N cycle is developing in restored jarrah forest (i.e. 26 years

post-mining; Banning et al. 2008). A conservative N cycle developing in the ecosystem

implies that critical functions are recovering slowly.

Species-specific conditioning of soil by jarrah forest plants is likely to have major

influence on soil functioning. In unmined soil, E. marginata and B. ornata conditioned

higher C:N ratios (as compared with unplanted controls), whereas A. pulchella

conditioned lower C:N ratios (Chapter 3; Table 1). The results indicated that organic

matter decomposition is occurring at slower rates in soils influenced by E. marginata

and B. ornata, whereas faster decomposition rates occur in soils influenced by A.

pulchella. These patterns were not observed in modified soil (Chapter 3; Table 2), the

C:N ratios of restored soils grown with the three plant species did not differ from those

of unplanted controls, which suggests that altered patterns of conditioning can have

additional impacts on soil functioning.

Individual PSF associated with the plants used in this study was soil-context dependent

(De Deyn et al. 2004; Chapter 3; Figure 4). The results suggest that PSF is a

phenomenon that exhibits some level of plasticity. Field observations support my

findings, in that a neutral PSF associated with B. ornata might contribute to its

103

abundance and long-persistence in unmined soil. In contrast, this plant species shows

variable abundance in restored soil (Pang et al. 2010), which could be due, in part, to

the expression of negative PSF. Acacia pulchella was very responsive to changing soil

conditions (via soil feedback), and specifically to fertilisation that is part of restoration

practice. Despite the negative PSF observed in the experiment, A. pulchella produced up

to three times more biomass in restored soil as compared with unmined soil, which

agrees with the dominance of this plant species observed in early-restored sites. These

results have implications for the restoration of the jarrah forest, since altered PSF might

produce altered patterns of plant abundance and dominance.

Results from Chapter 4, supported the finding of soil conditions (the soil feedback)

shaping individual PSF associated to B. ornata seedlings. The biomass produced by this

plant species grown in soils from the chronosequence of restoration ages was

significantly influenced by soil age (time since restoration), but not by plant species

from which soils were collected in the field. Properties of soils from the

chronosequence approached those of unmined soil as restoration age increased, which

suggested a recovery of the soil feedback (Chapter 4; Figure 4). However, a differential

soil feedback effect was observed as variable amounts of biomass produced by B.

ornata. Particularly important was the positive soil feedback associated with the 22-

year-old soil (Chapter 4; Figure 6), where the biggest plants were produced as compared

with unmined soil. The increased biomass produced in the 22-year old soil might

respond to the absence of plant-parasitic nematodes (as compared with unmined soils),

and the increased numbers of bacterial-feeding nematodes (Chapter 4; Table 1). These

results suggest that decreased or increased numbers of key soil organisms contribute to

alter the patterns of plant dominance and abundance in restored jarrah forest. The

differential soil feedback observed in soils from the chronosequence of restoration ages

did not affect individual PSF associated with B. ornata, and neutral PSF was observed

104

in all soils (Chapter 4; Figure 8). The fact that B. ornata produced similar amounts of

biomass in home soils as compared with away soils reinforce the hypothesis of a PSF

influenced by soil conditions, and not by plant species conditioning of the soil.

Furthermore, the results described in Chapters 3 & 4, suggest that individual PSF

associated with representative jarrah forest plant species is driven by soil conditions, not

by plant species. However, PSF associated with jarrah (E. marginata) was the

exception. Different soil properties conditioned in the field by non-native species

(different contexts of soil; Chapter 5; Figures 2 & 3) did not affect individual PSF, and

neutral PSF was observed for jarrah seedlings in the pot experiment (Chapter 5; Figure

6). This finding suggested that PSF associated with jarrah is mainly driven by plant

species, and exhibits some level of plasticity (Bleby et al. 2009), which allows the plant

to maintain a stable biomass production despite changing contexts of soil. Field

observations support this assertion, jarrah establishes and grows well in unmined but

also in soil post-bauxite mining (Koch & Samsa 2007). A plastic PSF (neutral PSF) and

efficient conditioning capacity might contribute to the long persistence of jarrah in the

forest, and moreover, these traits might contribute to its dominance in the ecosystem. In

the context of restoration, plant species displaying plastic growth responses including

neutral PSF such as jarrah are ideal to use during restoration strategies because this trait

allows restored plants to establish, grow, and persist in restored sites.

Neutral PSF associated to the experimental plants used (i.e. long-lived and short-lived

plant species) during the three experiments (Chapters 3, 4 & 5) suggests that biomass

production of jarrah forest plants might be little affected by changing properties of soils.

This finding contradincts the accepted PSF theory, which indicates that negative PSF is

associated to plants grown in early-successional soils, whereas positive PSF is

associated to late-successional soils (Kardol et al. 2006). It is important to highlight that

105

most of the ecosystems studied from a PSF perspective follow a relay floristics model

of succession, (i.e. an orderly and predictable series of plant species replacements that

occur after a disturbance; Cattelino et al. 1979). The traditional PSF theory fits very

well within this model of succession, since PSF phenomena help explaining the orderly

series of plant replacement.

In the jarrah forest, plant succession follows an initial floristic model of succession that

describes how the initial plant community evolves only as changes in the abundance of

the species present in the landscape (Hobbs & Legg 1983; Norman et al. 2006) in

response to disturbances such as fire (Bell & Koch 1980). It is likely that PSF operating

in jarrah forest occurs differently as compared with other ecosystems so as to secure

plant persistence in the community. This could happen because the species that establish

early in succession are those that are still present later. The findings of my three

experiments fit very well within the initial floristic model of succession, since in order

to persist in the ecosystem, jarrah forest plants must condition properties of soil

efficiently, a plant trait that was observed during the experiments. Moreover, soil is an

evolving entity, soil properties change through time and these influence the patterns of

abundance and dominance of plant species. Therefore, a neutral PSF associated to jarrah

forest plants help explaining their persistence in the field, because jarrah forest plants

are able to maintain a stable biomass production despite changing soil properties.

Individual PSF associated to plant species was altered in restored soil, so it is likely the

patterns of dominance and abundance of plant species are modified in restored forest as

compared with unmined forest, with implications on the structure of the plant

community composition of restored forest.

106

REFERENCES

Abbott, L.K., Robson, A.D., 1977. The distribution and abundance of vesicular arbuscular endophytes in some western Australian soils. Aust J Bot 25:515-522.

Abbott, I. 1984, Comparisons of spatial pattern, structure, and tree composition between virgin and cut-over jarrah forest in Western Australia. For Ecol Manage 9:101-126.

Adams, M.A., Attiwill, P.M., 1986. Nutrient cycling and nitrogen mineralization in eucalypt forests of south eastern Australia II. Indices of nitrogen mineralization. Plant Soil 92:341-362.

Adl, S.M., 2003. The ecology of soil decomposition. CABI Publishing, UK.

Ahmad, I., Pichtel, J., Hayat, S. 2008. Plant-bacteria interactions. Wiley-VCH Verlag GmbH & Co. KGaA, Germany.

Alexander, R., Croft, K.D., Kagi, R.I., Shea, S., 1978. Saponins of Acacia pulchella. Aust J Chem 31: 2741-2744.

Altieri, M.A., 1999. The ecological role of biodiversity in agroecosystems. Agr Ecosyst Environ 74:19-31.

Amalesh, S., Das, G., Das, S.K., 2011. Roles of flavonoids in plants. Int J Pharm Sci Tech 6(1): 12-35.

Anderson, M.J., 2001. A new method for non-parametric multivariate analysis of variance. Austral Ecol 26:32–46.

Anderson, M.J., Gorley, R.N., Clarke, K.R., 2008. PERMANOVA+ for PRIMER: Guide to software and statistical methods. Institute of Information and Mathematical Sciences (IIMS) Massey University, Auckland.

Anderson, M.J., 2010. Detecting multivariate changes in biological assemblages: experimental design and data analysis. Multivariate analysis of complex experimental designs using PERMANOVA+ for PRIMER. Institute of Information and Mathematical Sciences (IIMS) Massey University, Auckland.

Aponte, C., García, L.V., Marañón, T., 2013. Tree species effects on nutrient cycling and soil biota: a feedback mechanism favouring species coexistence. For Ecol Manage http://dx.doi.org/10.1016/j.foreco.2013.05.035

Atkinson, D., 2006. Arbuscular mycorrhizal fungi and the form and functioning of the root system. In: Mukerji, K.G., Manoharachary, C., Singh, J. (eds) Microbial activity in the rhizosphere. Springer, Germany.

Badri, D.V., Vivanco, J.M., 2009. Regulation and function of root exudates. Plant Cell Environ 32:666-681.

Baeten, L., Vanhellemont, M., De Frenne, P., De Schrijver, A., Hermy, M., Verheyen, K., 2010. Plasticity in response to phosphorus and light availability in four forest herbs. Oecologia 163:1021-1032.

Bais, H.P., Broeckling, C.D., Vivanco, J.M., 2008. Root exudates modulate plant–microbe interactions in the rhizosphere. In: Karlovsky, P. (ed) Secondary metabolites in soil ecology. Springer, Berlin.

107

Banning, N.C., Grant, C.D., Jones, D.L., Murphy, D.V., 2008. Recovery of soil organic matter, organic matter turnover and nitrogen cycling in a post-mining forest rehabilitation chronosequence. Soil Biol Biochem 40: 2021–2031.

Banning, N.C., Lalor, B.M., Cookson, W.R., Grigg, A.H., Murphy, D.V. 2012. Analysis of soil microbial community level physicological profiles in native and post-mining rehabilitation forest: Which subtrates discriminate? Appl Soil Ecol 56:27-34.

Bardgett, D.R., Bowman, D.W., Kaufmann, R., Schmidt, K.S., 2005. A temporal approach to linking abovegroung and belowground ecology. Trends Ecol Evol 20(11):634-641.

Barriuso, J., Ramos, S.B., Lucas, J.A., Probanza, L.A., Garcia-Villaraco, A., Gutierrez, M.F.J 2008. Ecology, genetic diversity and screening strategies of plant growth promoting rhizobacteria. In: Ahmad, I., Pichtel, J.,Hayat, Shamsul (eds). Plant-bacteria interactions. Wiley-VCH Verlag GmbH & Co. KGaA, Germany.

Bell, D.T., Koch, J.M., 1980. Post-fire succession in the northern jarrah forest of Western Australia. Aust J Ecol 5:9-14.

Bell, D.T., Heddle, E.M., 1989. Floristic, morphologic and vegetational diversity. In: Dell, B., Havel, J.J., Malajczuk, N. (eds) The jarrah forest: a complex Mediterranean ecosystem. Kluwer Academic Publishers, The Netherlands.

Bell, D.T., 2001. Ecological response syndromes in the flora of southwestern Western Australia: fire resprouters versus reseeders. Bot Rev 67(4):417-440.

Bell, T.D., Hobbs, J.R., 2007. Jarrah forest ecosystem restoration: a foreword. Restor Ecol 15(4):S1-S2.

Berg, B., McClaugherty, C., 2008. Plant Litter: decomposition, humus formation, carbon sequestration. 2nd edition. Springer, Germany.

Berg, G., Smalla, K., 2009. Plant species and soil type cooperatively shape the structure and function of microbial communities in the rhizosphere. FEMS Microbiol Ecol 68:1-13.

Berg, M.P., 2010. Spatio-temporal structure in soil communities and ecosystem processes. In: Verhoef, H.A., Morin, P.J. (eds) Community Ecology: processes, models and applications. Oxford University Press, Great Britain.

Bertin, C., Yang, X., Weston, L.A., 2003. The role of root exudates and allelochemicals in the rhizosphere. Plant Soil 256:67-83.

Bever, D.J., 1994. Feedback between plants and their soil communities in an old field community. Ecology 75(7):1965-1977.

Bever, J.D., Westover, K.M., Antonovics, J., 1997. Incorporating the soil community into plant population dynamics: the utility of the feedback approach. J Ecol 85:561-573.

Bever, J.D., 2002. Host-specificity of AM fungal population growth rates can generate feedback on plant growth. Plant Soil 244:281-290.

Bever, J.D., 2003. Soil community feedback and the coexistence of competitors: conceptual frameworks and empirical tests. New Phytol 257:465-473.

Bever, J.D., Dickie, I.A., Facelli, E., Facelli, J.M., Klironomos, J., Moora, M., Rillig, M.C., Stock, W.D., Tibbett, M., Zobel, M., 2010. Rooting theories of plant community ecology in microbial interactions. Trends Ecol Evol 25:468-478.

Bever, J.D., Platt, T.G., Morton, E.R., 2012. Microbial population and community dynamics on plant roots and their feedbacks on plant communities. Annu Rev Microbiol 66:265-283.

108

Bezemer, T.M., Lawson, C.S., Hedlund, K., Edwards, A.R., Brook, A.J., Igual, J.M., Mortimer, S.M., van der Putten, W.H., 2006. Plant species and functional group effects on abiotic and microbial soil properties and plant–soil feedback responses in two grasslands. J Ecol 94:893-904.

Bezuidenhout, S.R., Laing, M., 2006. Allelopathy and its influence in soil systems. In: Uphoff, N., Ball, A.S., Fernandes, E., Herren, H., Husson, O., Laing, M., Palm, C., Pretty, J., Sanchez, P., Sanginga, N., Thies, J. (eds.) Biological approaches to sustainable soil systems. CRC Taylor & Francis Group, USA.

Bhupinderpal-Singh, Rengel, Z., 2007. The role of crop residues in improving soil fertility. In: Marschner, P., Rengel, Z. (eds.) Nutrient cycling in terrestrial ecosystems. Springer-Verlag, Berlin.

Binkley, D., Ryan, M.G., 1998. Net primary production and nutrient cycling in replicated stands of Eucalyptus saligna and Albizia facaltaria. For Ecol Manage 112:79-85.

Bleby, T.M., Colquhoun, I.J., Adams, M.A., 2009. Architectural plasticity in young Eucalyptus marginata on restored bauxite mines and adjacent natural forest in south-western Australia. Tree Physiol doi:10.1093/treephys/tpp044.

Bohn, H.L., McNeal, B.L., O’Connor, G.A. 2001. Soil Chemistry 3rd edition. John Wiley & Sons, Inc., USA.

Bonanomi, G., Mazzoleni, S., 2005. Soil history affects plant growth and competitive ability in herbaceous species. Community Ecol 6(1):23-28.

Bonanomi, G., Rietkerk, M., Dekker, S.C., Mazzoleni, S., 2005a. Negative plant–soil feedback and positive species interaction in an herbaceous plant community. Plant Ecol 181:269–278.

Bonanomi, G., Giannino, F., Mazzoleni, S., 2005b. Negative plant-soil feedback and species coexistence. Oikos 111:311-321.

Bongers, T., Bongers, M., 1998. Functional diversity of nematodes. Appl Soil Ecol 10:239-251.

Brandt, A.J., de Kroon, H., Reynolds, H.L., Burns, J.H., 2013. Soil heterogeneity generated by plant-soil feedbacks has implications for species recruitment and coexistence. J Ecol 101:277-286.

Bret-Harte, M.S., Shaver, G.R., Zoerner, J.P., Johnstone, J.F., Wagner, J.L., Chavez, A.S., Gunkelman, R.F., Lippert, S.C., Laundre, J.A., 2001. Developmental plasticity allows Betula nana to dominate tundra subjected to an altered environment. Ecology 82(1):18-32.

Brimecombe, M.J., De Leij, F.A.A.M., Lynch, J.M., 2001. The effect of root exudates on rhizosphere microbial populations. In: Pinton, R., Varanini, Z., Nannipieri, P. (eds) The rhizosphere: biochemistry and organic substances at the plant-soil interface. Marcel Dekker Editorial, USA.

Brown, M.T., 2005. Landscape restoration following phosphate mining: 30 years of co-evolution of science, industry and regulation. Ecol Eng 24:309-329.

Brundrett, M.C., Abbott, L.K., 1991. Roots of jarrah forest plants. I. Mycorrhizal associations. Aust J Bot 39:445–457.

Brundrett, M.C., Abbott, L.K., 1995. Mycorrhizal fungus propagules in the jarrah forest II. Spatial variability in inoculum levels. New Phytol 131:461-469.

109

Büdel, B., 2005. Microorganisms of biological crusts on soil surfaces. In: Buscot, F., Varma, A. (eds), 2005. Microorganisms in soil: roles in genesis and functions. Springer-Verlag, Berlin.

Burrows, N.D., 2008. Linking fire ecology and fire management in south-west Australian forest landscapes. Forest Ecol Mange 255(7):2394-2406.

Buscot, F., Varma, A. (eds), 2005. Microorganisms in soil: roles in genesis and functions. Springer-Verlag, Berlin.

Buscot, F., 2005. What are soils? In: Buscot, F., Varma, A., (eds.) Microorganisms in soil: roles in genesis and functions. Springer-Verlag, Berlin.

Caesar-TonThat, T.C., Caesar, A.J., Gaskin, J.F., Sainju, U.M., Stevens, W.B. 2008. Ecology and function of culturable microbes in soil aggregation. In: Liu, T-X (ed) Soil ecology research developments. Nova Science Publishers, New York.

Campbell, C.D., Chapman, S.J., Cameron, C.M., Davidson, M.S., Potts, J.M., 2003. A rapid microtiter plate method to measure carbon dioxide evolved from carbon substrate amendments so as to determine the physiological profiles of soil microbial communities by using whole soil. Appl Environ Microb 69:3593–3599.

Carvalhais, L.C., Dennis, P.G., Fedoseyenko, D., Hajirezaei, M.R., Borriss, R., von Wirén, N., 2011. Root exudation of sugars, amino acids, and organic acids by maize as affected by nitrogen, phosphorus, potassium, and iron deficiency. J Plant Nutr Soil Sc 174:3–11.

Castillo, P. & Vovlas, N. (2007). Pratylenchus (Nematoda: Pratylenchidae): diagnosis, biology, pathogenicityand management. Nematology Monographs & Perspectives, 6 (Series Editors: Hunt, D.J. & Perry,R.N.). Leiden, The Netherlands, Brill Academic Publishers, 555 pp.

Cattelino, P.J., Noble, I.R., Slatyer, R.O., Kessell, S.R., 1979. Predicting the multiple pathways of plant succession. Environ Manage 3(1):41-50.

Cavigelli, M.A., Robertson, P., Klug, M.J., 1995. Fatty acid methyl ester (FAME) profiles as measures of soil microbial community structure. Plant Soil 179:90-113

Cesco, S., Mimmo, T., Tonon, G., Tomasi, N., Pinton, R., Terzano, R., Neumann, G., Weisskopf, L., Renella, G., Landi, L., Nannipieri, P., 2012. Plant-borne flavonoids released into the rhizosphere: impact on soil bio-activities related to plant nutrition. A review. Biol Fertil Soils 48:123-149.

Chalkos, V.D., Karamanolo, K., 2006. Microorganisms and allelopathy: a one-sided approach. In: Reigosa, M.J., Pedrol, N., González, L. (eds) Allelopathy: A physiological process with ecological implications. Springer, The Netherlands.

Chapman, S.J., Campbell, C.D., Artz, R.R.E., 2007. Assessing CLPPs using MicroResp™, a comparison with Biolog and multi-SIR. J Soils Sediments 7(6):406–410.

Chodak, M., Pietrzykowski, M., Niklinska, M., 2009. Development of microbial properties in a chronosequence of sandy mine soils. Appl Soil Ecol 41:259-268.

Churchward, H.M., Dimmock, G.M., 1989. The soils and landforms of the northern jarrah forest. In: Dell, B., Havel, J.J., Malajczuk, N. (eds) The jarrah forest: a complex Mediterranean ecosystem. Kluwer Academic Publishers, The Netherlands.

Coleman, D.A., Crossley, D.A., Hendrix, P.F., 2004. Fundamentals of soil ecology 2nd edition. Elsevier Academic Press, Amsterdam.

110

Compan, S., Duffy, B., Nowak, J.,Clément, C., Ait Barka, E., 2005. Use of plant growth-promoting bacteria for biocontrol of plant diseases: principles, mechanisms of action, and future prospects. Appl Environ Microbiol 71(9):4951-4959.

Condron, L., Stark, C., O’Callaghan, M., Clinton, P., Huang, Z., 2010. The role of microbial communities in the formation and decomposition of soil organic matter. In: Dixon, G.R., Tilston, E.L. (eds) Soil microbiology and sustainable crop production. Springer, New York.

Conservation Commission of Western Australia, 2012. Draft forest management plan 2014 – 2023. Conservation Commission of Western Australia, Perth.

D’Souza, N.K., Colquhoun, I.J., Shearer, B.L., Hardy, G.E.S.J., 2004. The potential of five Western Australian native species for biological control of Phytophtora cinnamomi. Aust J Bot 52:267-279.

Dakora, F.D., Phillips, D.A., 2002. Root exudates as mediators of mineral acquisition in low-nutrient environments. Plant Soil 245:35-47.

De Bellis, T., Kernaghan, G., Widden, P., 2007. Plant community influences on soil microfungal assemblages in boreal mixed-wood forests. Mycologia 99(3):356-367.

de Boer, W., Folman, L.B., Summerbell, R.C., Boddy, L., 2005. Living in a fungal world: impact of fungi on soil bacterial niche development. FEMS Microbiol Rev 29:795–811.

De Deyn, B.G., Raaijmakers, E.C., van der Putten, H.W., 2004. Plant community development is affected by nutrients and soil biota. J Ecol 92:824-834.

De Rooij-van der Goes, P.C.E.M., 1995. The role of plant-parasitic nematodes and soil-borne fungi in the decline of Ammophila arenaria (L.) Link. New Phytol 129:661-669.

Deans, S.G., 2002. Antimicrobial activity of eucalyptus oil. In: Coppen, J.J.W. (ed) Eucalyptus: the genus Eucalyptus.Taylor & Francis, London.

Dell, B., Havel, J.J., Malajczuk N., 1989. The jarrah forest: a complex Mediterranean ecosystem. Kluwer Academic Publishers, The Netherlands.

Ehrenfeld, J.G., Ravit, B., Elgersma, K., 2005. Feedback in the plant-soil system. Annu Rev Environ Resour 30:75–115.

Eisenhauer, N., Reich, P.B., 2012. Above- and below-ground plant inputs both fuel soil food webs. Soil Biol Biochem 45:156-160.

Eviner, V.T., 2004. Plant traits that influence ecosystem processes vary independently among species. Ecology 85(8):2215-2229.

Eviner, V.T., Hawkes, C.V., 2008. Embracing variability in the application of plant–soil interactions to the restoration of communities and ecosystems. Restor Ecol 16(4):713-729.

Farrar, J., Hawes, M., Jones, D., Lindow, S., 2003. How roots control the flux of C to the rhizosphere. Ecology 84(4): 827-837.

Feller, M.C., 1983. Effects of an exotic conifer (Pinus radiata) plantation on forest nutrient cycling in Southeastern Australia. For Ecol Manage 7:77-102.

Ferris, H., Bongers, T., Goede, R.G.M., 2001. A framework for soil food web diagnostics: extension of the nematode faunal analysis concept. Appl Soil Ecol 18:13-29.

111

Ferris, H., Matute, M.M., 2003. Structural and functional succession in the nematode fauna of a soil food web. Appl Soil Ecol 23:93-110.

Foth, H.D., 1990. Fundamentals of soil science 8th edition. John Wiley & Sons USA.

Freckman, D.W., 1988. Bacterivorous nematodes and organic-matter decomposition. Agr Ecosyst Environ 24:195-217.

Frostegård, A., Bååth, E., 1996. The use of phospholipid fatty acid analysis to estimate bacterial and fungal biomass in soil. Biol Fertil Soils 22(1):59-65.

Fukami, T., Nakajima, M., 2013. Complex plant-soil interactions enhance plant species diversity by delaying community convergence. J Ecol 101:316-324.

Ganjegunte, G.K., Condron, L.M., Clinton, P.W., Davis, M.R., 2005. Effects of mixing radiata pine needles and understory litters on decomposition and nutrients release. Biol Fertil Soils 41:310-319.

Gardner, J.H., Bell, D.T., 2007. Bauxite mining restoration by Alcoa World Alumina Australia in Western Australia: social, political, historical, and environmental contexts. Restor Ecol 15(4):S3-S10.

Gentili, J., 1989. Climate of the jarrah forest. In: Dell, B., Havel, J.J., Malajczuk, N. (eds) The jarrah forest: a complex Mediterranean ecosystem. Kluwer Academic Publishers, the Netherlands.

George, S.J., Kelly, R.N., Greenwood, P.F., Tibbett, M., 2010. Soil carbon and litter development along a reconstructed biodiverse forest chronosequence of south-western Australia. Biogeochemistry 101:197–209.

George, S.J., Tibbett, M., Braimbridge, M.F., Davis, S.G., Vlahos, S., Ryan, M., 2006. Phosphorus fertiliser placement and seedling success in Australian jarrah forest. In: Fourie, A., Tibbett, M. (eds) Mine Closure. Australian Centre for Geomechanics, Perth.

Gong, X., Liu, Y., Li, Q., Wei, X., Guo, X., Niu, D., Zhang, W., Zhang, J., Zhang. L., 2013. Sub-tropic degraded red soil restoration: Is soil organic carbon build-up limited by nutrients supply. For Ecol Manage 300:77-87.

Gonzalez-Quiñones, V., Banning, N.C., Jimenez, B.R., Murphy, D.V. 2009. Influence of cold storage on soil microbial community level physicological profiles and implications for soil quality monitoring. Soil Biol Biochem 41:1574-1576.

Gorbushina, A.A., Krumbein, W.E., 2005. Role of microorganisms in wear down of rocks and minerals. In: Buscot, F., Varma, A. (eds) Microorganisms in soil: roles in genesis and functions. Springer-Verlag, Berlin.

Grant, C.D., 2006. State-and-transition successional model for bauxite mining rehabilitation in the jarrah forest of Western Australia. Restor Ecol 14(1):28-37.

Grierson, P.F., Adams, M.A., 2000. Plant species affect acid phosphatase, ergosterol and microbial P in a Jarrah (Eucalyptus marginata Donn ex Sm.) forest in south-western Australia. Soil Biol Biochem 32:1817-1827.

Griffiths, B.S., Christensen, S., Bonkowski, M., 2007. Microfaunal interactions in the rhizosphere, how nematodes and protozoa link above- and belowground processes. In: Cardon, Z.G., Whitbeck, J.L. (eds) The rizosphere: an ecological perspective. Elsevier, USA.

112

Guo, L.B., Sims, R.E.H., 1999. Litter decomposition and nutrient release via litter decomposition in New Zealand eucalypt short rotation forests. Agr Ecosyst Environ 75:133–140.

Gupta, V.V.S.R., Yeates, G.W., 1997. Soil microfauna as bioindicators of soil health. In: Pankhurst, C., Doube, B.M., Gupta, V.V.S.R. (eds) Biological indicators of soil health. CAB International, UK.

Hadacek, F., 2008. Constitutive secondary metabolites and soil fungi: defence against or facilitation of diversity. In: Karlovsky, P. (ed) Secondary metabolites in soil ecology. Springer-Verlag, Berlin.

Harris, J., 2009. Soil microbial communities and restoration ecology: facilitators or followers. Science 325:572-574.

Harrison, K.A., Bardgett, R.D., 2010. Influence of plant species and soil conditions on plant-soil feedback in mixed grassland communities. J Ecol 98:384-395.

Hingston, F.J., Dimmock, G.M., Turton, A.G., 1981. Nutrient distribution in a jarrah (Eucalyptus marginata Donn Ex Sm.) ecosystem in south-west Western Australia. For Ecol Manage 3:183-207.

Hingston, F.J., O’Connell, A.M., Grove, T.S., 1989. Nutrient cycling in jarrah forest. In: Dell, B., Havel, J.J., Malajczuk, N. (eds) The jarrah forest: a complex Mediterranean ecosystem. Kluwer Academic Publishers, The Netherlands.

Hirsch, A.M., Bauer, W.D., Bird, D.M., Cullimore, J., Tyler, B., Yoder, J.I., 2003. Molecular signals and receptors: controlling rhizosphere interactions between plants and other organisms. Ecology 84(4):858-868.

Hobbie, S.E., 1992. Effects of plant species on nutrient cycling. Trends Ecol Evol 7(10):336-339.

Hobbs, R.J., Legg, C.J., 1983. Markov models and initial floristic composition in heathland vegetation dynamics. Vegetatio 56:31–43.

Holtkamp, R., Kardol, P., van der Wal, A., Dekker, S.C., van der Putten, W.H., de Ruiter, P.C., 2008. Soil food web structure during ecosystem development after land abandonment. Appl Soil Ecol 39:23–34.

Hooper, D.J., Hallmann, J., Subbotin, S.A., 2005. Methods for extraction, processing and detection of plant and soil nematodes. In: Luc, M., Sikora, R.A., Bridge, J. (eds) Plant parasitic nematodes in subtropical and tropical agriculture 2nd edition. CAB Publishing, UK.

Hooper, D.U., Bignell, D.E., Brown, K.V., Brussard, L., Dangerfield, J.M., Wall, D.H., Wardle, D.A., Coleman, D.C., Giller, K.E., Lavelle, P., van der Putten, W.H., de Ruiter, P.C., Rusek, J., Whendee, L.S., Tiedje, J.M., Wolters, V., 2000. Interactions between aboveground and belowground biodiversity in terrestrial ecosystems: patterns, mechanisms, and feedbacks. BioScience 50(12):1049-1061.

Hopkins, D.W., Dungait, J.A.J., 2010. Soil microbiology and nutrient cycling. In: Dixon, G.R., Tilston, E.L. (eds) Soil microbiology and sustainable crop production. Springer, New York.

Hopmans, P., Elms, S.R., (2009) Changes in total carbon and nutrients in soil profiles and accumulation in biomass after a 30-year rotation of Pinus radiata on podzolized sands: impacts of intensive harvesting on soil resources. For Ecol Manage 258:2183-2193.

113

Hopper, S.D., 2009. OCBIL theory: towards an integrated understanding of the evolution, ecology and conservation of biodiversity on old, climatically buffered, infertile landscapes. Plant Soil 322:49-86.

Hussain, M.I., González, L., Souto, C., Reigosa, M.J., 2011. Ecophysicological responses of three native herbs to phytotoxic potential of invasive Acacia melanoxylon R. Br. Agroforest Syst 83:149-166.

Inderjit, Mallik, A.U., 1997. Effect of phenolic compounds on selected soil properties. Forest Ecol Manag 92:11-18.

Inderjit, 2005. Soil microorganisms: An important determinant of allelopathic activity. Plant Soil 274:227-236.

Inderjit, Weiner, J., 2001. Plant allelochemical interference or soil chemical ecology? Perspect Plant Ecol 4(1):3-12.

Inderjit, van der Putten, W.H., 2010. Impacts of soil microbial communities on exotic plant invasions. Trends Ecol Evol 25:512-519.

Insam, H., Domsch, H.K., 1988. Relationship between soil organic carbon and microbial biomass on chronosequences of reclamation sites. Microbial Ecol 15(2): 177-188.

Jasper, A.D., 2007. Beneficial soil microorganisms of the and their recovery in bauxite mine restoration in South Western Australia. Restor Ecol 15(4):S74-S84.

Jastrow, J.D., Miller, R.M., Lussenhop, J., 1998. Contributions of interacting biological mechanisms to soil aggregate stabilization in restored prairie. Soil Biol Biochem 30(7):905-916.

Johnson, C., 2009. Biology of soil science. Oxford Book Company, Delhi.

Jones, D.L., 1998. Organic acids in the rhizosphere – a critical review. Plant Soil 205:25-44.

Jordan, N.R., Larson, D.L., Huerd, S.C., 2008. Soil modification by invasive plants: effects on native and invasive species of mixed-grass prairies. Biol Invasions 10:177-190.

Kanesaki, Y., Los, D.A., Suzuki, I., Murata, N., 2010. Sensors and signal transducers of environmental stress in Cyanobacteria. In: Pareek, A., Sopory, S.K., Bohnert, H.J., Govindjee (eds) Abiotic stress adaptation in plants: physiological, molecular and genomic foundation. Springer, The Netherlands.

Kapoor, R., Mukerji, K.G., 2006. Rhizosphere microbial community dynamics. In: Mukerji, K.G., Manoharachary, C., Singh, J. (eds) Microbial activity in the rhizosphere. Springer, Germany.

Kardol, P., Bezemer, T.M., van der Putten, W.H., 2006. Temporal variation in plant–soil feedback controls succession. Ecol Lett 9: 1080–1088.

Kardol, P., Cornips, N.J., van Kempen, M.M.L., Bakx-Schotman, J.M.T., van der Putten, W.H., 2007. Microbe-mediated plant-soil feedback causes historical contingency effects in plant community assembly. Ecol Monogr 77:147-162.

Kardol, P., Wardle, D.A., 2010. How understanding aboveground-belowground linkages can assist restoration ecology. Trends Ecol Evol 25(11):670-679.

Kardol, P., De Deyn, G.B., Laliberté, E., Mariotte, P., Hawkes, C.V., 2013. Biotic plant-soil feedbacks across temporal scales. J Ecol 101:309-315.

114

Keeley, J.E., Bond, W.J., Bradstock, R.A., Pausas, J.G.,Rundel, P.W., 2012. Fire in Mediterranean ecosystems: Ecology, evolution and management. Cambridge University Press, USA.

Koch, J.M., 2007a. Bauxite mining restoration by Alcoa World Alumina Australia in Western Australia: social, political, historical, and environmental contexts. Restor Ecol 15(4):S3-S10.

Koch, J.M., 2007b. Alcoa’s mining and restoration process in south Western Australia. Restor Ecol 15(4):S11-S16.

Koch, J.M., Samsa, G., 2007. Restoring jarrah forest trees after bauxite mining in Western Australia. Restor Ecol 15(4):S17-S25.

Kohli, R.K., Batish, D.R., Singh, H.P., 2006. Allelopathic interactions in agroecosystems. In: Reigosa, M.J., Pedrol, N., González, L. (eds) Allelopathy: a physiological process with ecological implications. Springer, The Netherlands.

Kulmatiski A, Beard KH, Stark JM (2006) Soil history as a primary control on plant invasion in abandoned agricultural fields. J Appl Ecol 43:868-876.

Kulmatiski, A., Beard, H.K., Stevens, R.J., Cobbold, M.S., 2008. Plant-soil feedbacks: a meta-analytical review. Ecol Lett 11:980-992.

Kulmatiski, A., Beard, H.K., 2008. Decoupling plant-growth from land-use legacies in soil microbial communities. Soil Biol Biochem 40:1059-1068.

Kulmatiski, A., Kardol, P., 2008. Getting plant-soil feedbacks out of the greenhouse: experimental and conceptual approaches. Prog Bot 69:449-472.

Kulmatiski, A., Beard, H.K., 2011. Long-term plant growth legacies overwhelm short-term plant growth effects on soil microbial community structure. Soil Biol Biochem 43:823-830.

Kylafis, G., Loreau, M., 2008. Ecological and evolutionary consequences of niche construction for its agent. Ecol Lett 11:1072-1081.

Kyle, G.P., Beard, K.H., Kulmatiski, A., 2007. Reduced soil compaction enhances growth of non-native species. Plant Ecol 193(2):223-232.

Ladygina, N., Hedlund, K., 2010. Plant species influence microbial diversity and carbon allocation in the rhizosphere. Soil Biol Biochem 42:162-168.

Lalor, E.L., Cookson, W.R., Murphy, D.V., 2007. Comparison of two methods that assess soil community level physiological profiles in a forest ecosystem. Soil Biol Biochem 39:454–462.

Lambers, H., Raven, J.A., Shaver, G.R., Smith, S.E., 2007. Plant nutrient-acquisition strategies change with soil age. Trends Ecol Evol 23(2):95-103.

Lambers, H., Bishop, J.G., Hopper, S.D., Laliberté, E., Zuñiga-Feest, A., 2012. Phosphorus-mobilization ecosystem engineering: the roles of cluster roots and carboxylate exudation in young P-limited ecosystems. Ann Bot-London 110:329-348.

Laskowski, R., Niklinska, M., Maryanski, M., 1995. The dynamics of chemical elements in forest litter. Ecology 76(5):1393-1406.

Lavelle, P., Spain, A.V., 2003. Soil Ecology. Kluwer Academic Publishers, USA

Lawrie, A.C., 1981. Nitrogen fixation by native Australian legumes. Aust J Bot 29:143-157.

115

Lin, D.S., Greenwood, P.F., George, S.J., Somerfield P.J., Tibbett, M., 2011. The development of soil organic matter in restored biodiverse jarrah forests of South-Western Australia as determined by ASE and GCMS. Environ Sci Pollut R 18:1070–1078.

Liu, Y., Yu, S., Xie, Z., Staehelin, C., 2012. Analysis of a negative plant–soil feedback in a subtropical monsoon forest. J Ecol 100:1019-1028.

Lorenzo, P., Pereira, C.S., Rodríguez-Echeverría, S., 2013. Differential impact on soil microbes of allelopathic compounds released by the invasive Acacia dealbata Link. Soil Biol Biochem 57:156-163.

Lotina-Hennsen, B., King-Díaz, B., Aguilar, M.I., Hernandez-Terrones, M.G., 2006. Plant secondary metabolites. Targets and mechanisms of allelopathy. In: Reigosa, M.J., Pedrol, N., González, L. (eds) Allelopathy: A physiological process with ecological implications. Springer, The Netherlands.

Lugo, A.E., Cuevas, E., Sánchez, M.J., 1990. Nutrients and mass in litter and top soil of ten tropical tree plantations. Plant Soil 125:263-280.

Madigan, M.T., Martinko, J.M., Stahl, D.A., Clark, D.P., 2012. Brock Biology of Microorganisms 13th edition. Benjamin Cummings Editorial, USA.

Malajczuk, N., McComb, A.J., 1977. Root exudates from Eucalyptus calophylla R.Br. and Eucalyptus marginata Donn. ex Sm. seedlings and their effect on Phytophthora cinnamomi Rands. Aust J Bot 25:501-514.

Malajczuk, N., McComb, A.J., 1979. The microflora of unsuberized roots of Eucalyptus calophylla R.Br. and Eucalyptus marginata Donn ex Sm. seedlings grown in soil suppressive and conducive to Phytophthora cinnamomi Rands. I Rhizosphere bacteria, actinomycetes and fungi. Aust J Bot 27:235-254.

Malajczuk, N., 1979. The microflora of unsuberized roots of Eucalyptus calophylla R. Br. and Eucalyptus marginata Donn ex Sm. seedlings grown in soils suppressive and conducive to Phytophthora cinnamomi Rands. II Mycorrhizal roots and associated microflora. Aust J Bot 27:255-272.

Malik, A., Scullion, J., 1998. Soil development on restored opencast coal sites with particular reference to organic matter and aggregate stability. Soil Use Manage 14:234-239.

Maluf, de S.F., Ferreira, B.J.L., 2004. Restoration of seasonal semideciduous forests in Brazil: influence of age and restoration design on forest structure. For Ecol Manage 191:185-200.

Mangan, S.A., Schnitzer, S.A., Herre, E.A., Mack, K.M.L., Valencia, M.C., Sanchez, E.I., Bever, J.D., 2010. Negative plant–soil feedback predicts tree-species relative abundance in a tropical forest. Nature 466 doi: 10.1038/nature09273.

Manning, P., Morrison, S.A., Bonkowski, M., Bardgett, R.D., 2008. Nitrogen enrichment modifies plant community structure via changes to plant-soil feedback. Oecologia, 157(4): 661-673.

Manoharachary, C., Mukerji, K.G., 2006. Rhizosphere biology – an overview. In: Mukerji, K.G., Manoharachary, C., Singh, J. (eds) Microbial activity in the rhizosphere. Springer, Germany.

Marschner, H., 1995. Mineral nutrition of higher plants 2nd edition. Academic Press, Great Britain.

116

Martin, B.C., George, S.J., Price, C.A., Ryan, M.H., Tibbett, M., 2014. The role of root exuded low molecular weight organic anions in facilitating petroleum hydrocarbon degradation: current knowledge and future directions. Sci Total Environ 472:642-653.

Meijer, S.S., Holmgren, M., van der Putten, W.H., 2011. Effects of plant–soil feedback on tree seedling growth under arid conditions. J Plant Ecol doi: 10.1093/jpe/rtr011

Menta, C., Conti, F.D., Pinto, S., Leoni, A., Lozano-Fondón, C., 2013. Monitoring soil restoration in an open-pit mine in northern Italy. Appl Soil Ecol doi.org/10.1016/j.apsoil.2013.07.013

Miki, T., Ushio, M., Fukui, S., Kondoh, M., 2010. Functional diversity of microbial decomposers facilitates plant coexistence in a plant–microbe–soil feedback model. PNAS 107(32):14251-14256.

Miki, T., 2012. Microbe-mediated plant-soil feedback and its roles in a changing world. Ecol Res 27:509-520.

Mills, E.K., Bever, D.J., 1998. Maintenance of diversity within plant communities: soil pathogens as agents of negative feedback. Ecology 79(5):1595-1601.

Monk, D., Pate, J.S., Loneragan, W.A., 1981. Biology of Acacia pulchella R.Br. with special reference to symbiotic nitrogen fixation. Aust J Bot 29: 579–592.

Moore, J.C., Hunt, H.W., 1988. Resource compartmentation and the stability of real ecosystems. Nature 33:261-263.

Moore, J.C., McCann, K., Setäla, H., de Ruiter, P.C., 2003. Top-down is bottom-up: does predation in the rhizosphere regulate aboveground dynamics? Ecology 84:846-857.

Moore, J.C., McCann, K., de Ruiter, P.C., 2007. Soil rhizosphere food webs, their stability, and implications for soil processes in ecosystems. In: Cardon, Z.G., Whitbeck, J.L. (eds) The rhizosphere: an ecological perspective. Elsevier, USA.

Mummey, D.L., Stahl, P.D., Buyer, J.S., 2002. Soil microbiological properties 20 years after surface mine reclamation: spatial analysis of reclaimed and undisturbed sites. Soil Biol Biochem 34:1717-1725.

Myers, N., Mittermeier, R.A., Mittermeier, C.G., da Fonseca, G.A.B., Kent, J., 2000. Biodiversity hotspots for conservation priorities. Nature 403:853-858.

Nautiyal, C.S., Srivastava, S., Chauhan, P.S., 2008. Rhizosphere colonization: molecular determinants from plant-microbe coexistence perspective. In: Nautiyal, C.S., Dion, P. (eds) Molecular mechanisms of plant and microbe coexistence. Springer, Germany.

Neher, D.A., 1999. Soil community composition and ecosystem processes: comparing agricultural ecosystems with natural ecosystems. In: Lefroy, E.C., Hobbs, R.J., O’Connor, M.H., Pate, J.S. (eds) Agriculture as a mimic of natural ecosystems. Kluwer Academic Publishers, The Netherlands.

Nelson, E.B., 2006. Rhizosphere regulation of preinfection behaviour of oomycete plant pathogens. In: Mukerji, K.G., Manoharachary, C., Singh, J. (eds) Microbial activity in the rhizosphere. Springer, Germany.

Neumann, G., Römheld, V., 1999. Root excretion of carboxylic acids and protons in phosphorus-deficient plants. Plant Soil 211:121-130.

Neumann, G., Römheld, V., 2001. The release of root exudates as affected by the plant’s physicological status. In: Pinton, R., Varanini, Z., Nannipieri, P. (eds) The rhizosphere:

117

biochemistry and organic substances at the plant-soil interface. Marcel Dekker Editorial, USA.

Neumann, G., 2007. Root exudates and nutrient cycling. In: Marschner, P., Rengel, Z. (eds) Nutrient cycling in terrestrial ecosystems. Springer-Verlag, Berlin.

Norman, M.A., Koch, J.M., Grant, C.D., Morald, T.K., Ward, S.C., 2006. Vegetation succession after bauxite mining in Western Australia. Restor Ecol 14(2):278-288.

O’Brien, N.D., Attiwill, P.M., Weston, C.J., 2003. Stability of soil organic matter in Eucalyptus regnans forests and Pinus radiata plantations in south eastern Australia. For Ecol Manage 185:249-261.

O’Connell, A.M., Grove, T.S., Dimmock, G.M., 1978. Nutrients in the litter of jarrah forest soils. Aust J Ecol 3:253-260.

Pang, J., Ryan, M.H., Tibbett, M., Cawthray, G.R., Siddique, K.H.M., Bolland, M.D.A., Denton, M.D., Lambers, H., 2010. Variation in morphological and physiological parameters in herbaceous perennial legumes in response to phosphorus supply. Plant Soil 331:241-255.

Papatheodorou, E.M., 2008. Responses of soil microbial communities to climatic and human impacts in Mediterranean regions. In: Liu, T-X (ed) Soil ecology research developments. Nova Science Publishers, New York.

Paul, M., Catterall, C.P., Pollard, P.C., Kanowski, J., 2010. Recovery of soil properties and functions in different rainforest restoration pathways. For Ecol Manage 259:2083-2092.

Pekin, B.K., Wittkuhn, R.S., Boer, M.M., MacFarlane, C., Grierson, P., 2012. Response of plant species and life form diversity to variable fire histories and biomass in the jarrah forest of south-west Australia. Aust Ecol 37:330-338.

Perkins, L.B., Nowak, R.S., 2012. Soil conditioning and plant-soil feedbacks affect competitive relationships between native and invasive grasses. Plant Ecol 213:1337-1344.

Pernilla, B.E., van der Putten, W.H., Bakker, E.J., Verhoeven, K.J.F., 2010. Plant–soil feedback: experimental approaches, statistical analyses and ecological interpretations. J Ecol 98:1063-1073.

Perry, R.N., Tovar-Soto, A., 2012. Biology and host-parasite interactions. In: Manzanilla-López, R.H., Marbán-Mendoza, N. (eds) Practical plant Nematology. Biblioteca Básica de Agricultura, Mexico.

Philippot, L., Germon, J.C., 2005. Contribution of bacteria to initial input and cycling of nitrogen in soil. In: Buscot, F., Varma, A. (eds) Microorganisms in soil: roles in genesis and functions. Springer-Verlag, Berlin.

Ponge, J.F., 2013. Plant-soil feedbacks mediated by humus forms: A review. Soil Biol Biochem 57:1048-1060.

Postma-Blaauw, M.B., de Vries, F.T., de Goede, R.G.M., Bloem, J., Faber, J.H., Brussaard, L., 2005. Within-trophic group interactions of bacterivorous nematode species and their effects on the bacterial community and nitrogen mineralization. Oecologia 142:428-439.

Pregitzer, K.S., Azk, D.R., Loya, W.M., Karberg, N.J., King, J.S., Burton, A.J., 2007. The contribution of root – rhizosphere interactions to biogeochemical cycles in a changing world. In: Cardon, Z.G., Whitbeck, J.L. (eds) The rhizosphere: an ecological perspective. Elsevier, USA.

118

Pregitzer, C.C., Bailey, J.K., Hart, S.C., Schweitzer, J.A., 2010. Soils as agents of selection: feedbacks between plants and soils alter seedling survival and performance. Evol Ecol 24:1045–1059.

Prescott, C.E., Graystone, S.J. 2013. Tree species influence on microbial communities in litter and soil: current knowledge and research needs. Forest Ecol Manage 309:19-27.

Prober, S.M., Lunt, I.D., Morgan, J.W. 2009. Rapid internal plant-soil feedbacks lead to alternative stable states in temperate Australian grassy woodlands. In: Hoobs, R.J., Suding, K.N. (eds) New models for ecosystem dynamics and restoration. Island Press, USA.

Ranganayaki, N., Tilak, K.V.B.R., Manoharachary, C., Mukerji, K.G., 2006. Methods and techniques for isolation, enumeration and characterization of rhizosphere microorganisms In: Mukerji, K.G., Manoharachary, C., Singh, J. (eds) Microbial activity in the rhizosphere. Springer, Germany.

Rayment, G.E., Higginson, F.R., 1992. Australian laboratory handbook of soil and water chemical methods: Australian soil and land survey handbook Vol 3. Inkata Press, Australia.

Reuben, S., Bhinu, V.S., Swarup, S. 2008. Rhizosphere metabolomics:methods and applications. In: Karlovsky, P. (ed) Secondary metabolites in soil ecology. Springer, Berlin.

Reynolds, H.L., Packer, A., Bever, J.D., Clay, K., 2003. Grassroots ecology: plant–microbe–soil interactions as drivers of plant community structure and dynamics. Ecology 84(9):2281-2291.

Ribeiro, R.R., Venancio, M.S., de Barros, L.C., 2004. Tropical Rain Forest regeneration in an area degraded by mining in Mato Grosso State, Brazil. For Ecol Manage 190:323-333.

Richardson, A.E., Barea, J.M., McNeill, A.M., Prigent-Combaret, C., 2009. Acquisition of phosphorus and nitrogen in the rhizosphere and plant growth promotion by microorganisms. Plant Soil 321:305–339.

Rock, C.D., Sakata, Y., Quatrano, R.S., 2010. Stress signalling I: The role of abscisic acid (ABA) In: Pareek, A., Sopory, S.K., Bohnert, H.J., Govindjee (eds) Abiotic stress adaptation in plants: physiological, molecular and genomic foundation. Springer, The Netherlands.

Sardans, J., Peñuelas, J., 2013. Plant-soil interactions in Mediterranean forest and shrublands: impacts of climatic change. Plant Soil 365:1-33.

Sbrana, C., 2006. Fungal recognition responses to host derived signals by arbuscular mycorrhizal fungi. In: Mukerji, K.G., Manoharachary, C., Singh, J. (eds) Microbial activity in the rhizosphere. Springer, Germany.

Scheu, S., Ruess, L., Bonkowski, M., 2005. Interactions between microorganisms and soil micro- and mesofauna. In: Buscot, F., Varma, A. (eds) Microorganisms in soils: roles in genesis and functions. Springer-Verlag, Berlin.

Schimann, H., Petit-Jean, C., Guitet, S., Reis, T., Domenach, A.M., Roggy, J.-C. 2012. Microbial bioindicators of soil functioning after disturbance: The case of gold mining in tropical rainforests of French Guiana. Ecol Indic 20:34-41.

Schipper, L.A., Degens, B.P., Sparling, G.P., Duncan, L.C., 2001. Changes in microbial heterotrophic diversity along five plant successional sequences. Soil Biol Biochem 33:2093-2103.

119

Scholefield, D., Jewkes, E., Bol, Roland., 2007. Nutrient cycling budgets in managed pastures. In: Marschner, P., Rengel, Z. (eds) Nutrient cycling in terrestrial ecosystems. Springer-Verlag, Berlin.

Shannon, S., Flory, S.L., Reynolds, H., 2012. Competitive context alters plant-soil feedback in an experimental woodland community. Oecologia 169:235-243.

Shono, K., Davies, S.J., Kheng, C.Y., 2006. Regeneration of native plant species in restored forests on degraded lands in Singapore. For Ecol Mange 237:574-582.

Singh, G., Mukerji, K.G., 2006. Root exudates as determinant of rhizospheric microbial diversity. In: Mukerji, K.G., Manoharachary, C., Singh, J. (eds) Microbial activity in the rhizosphere. Springer, Germany.

Solaiman, M.Z., Abbott, L.K., 2003. Phosphorus uptake by a community of arbuscular mycorrhizal fungi in jarrah forest. Plant Soil 248:313-320.

Spain, A.V., Tibbett, M., Hinz, D.A., Ludwig, J.A., Tongway, D.J., 2014. The mining-restoration system and ecosystem development following bauxite mining in a biodiverse environment of the seasonally dry tropics, Northern Territory, Australia. In: Tibbett, M. (ed) Mining in Ecologically Sensitive Landscapes. CSIRO Press, In Press

Spring, S., Schulze, R., Overmann, J., Schleifer, K-H., 2000. Identification and characterization of ecologically significant prokaryotes in the sediment of freshwater lakes: molecular and cultivation studies. FEMS Microbiol Rev 24(5):573-590.

Standish, R.J., Tibbett, M., Vlahos, S., Stokes, B.A., Hobbs, R.J., 2010. Effect of fertiliser on early-successional jarrah forest restored after bauxite mining in south-western Australia. In: Fourie, A., Tibbett, M., Wiertz, J. (eds) Mine Closure. Australian Centre for Geomechanics, Perth.

Suding, K.N., Gross, K.L., Houseman, G.R. (2004) Alternative states and positive feedbacks in restoration ecology. Trends Ecol Evol 19(1):46-53.

Suding, K.N., Hobbs, R.J., 2008. Threshold models in restoration and conservation: a developing framework. Trends Ecol Evol 24:271-279.

Suding, K.N., Harpole, W.S., Fukami, T., Kulmatiski, A., MacDougall, A.S., Stein, C., van der Putten, W.H., 2013. Consequences of plant-soil feedbacks in invasion. J Ecol 101:298-308.

te Beest, M., Stevens, N., Olff, N., van der Putten, W.H., 2009. Plant–soil feedback induces shifts in biomass allocation in the invasive plant Chromolaena odorata. J Ecol 97:1281–1290.

Tibbett, M., 2010. Large-scale mine site restoration of Australian eucalypt forests after bauxite mining: soil management and ecosystem development. In: Batty, L.C., Hallberg, K. (eds) Ecology of industrial pollution. Cambridge University Press, UK.

Topp, G.C., Parkin, G.W., Ferré, T.P.A. Soil water content In: Carter, M.R., Gregorich, E.G. (eds) 2008. Soil sampling and methods of analysis 2nd edition. CRC Press, USA.

Treonis, A.M, Ostle, N.J., Stott, A.W., Primrose, R., Grayston, S.J., Ineson, P., 2004. Identification of groups of metabolically-active rhizosphere microorganisms by stable isotope probing of PLFAs. Soil Biol Biochem 36:533-537.

Turner, J., Lambert, M.J., 2008. Nutrient cycling in age sequences of two Eucalyptus plantation species. For Ecol Manage 255:1701-1712.

120

Uren, N.C., 2001. Types, amounts, and possible functions of compounds released into the rhizosphere by soil-grown plants. In: Pinton, R., Varanini, Z., Nannipieri, P. (eds) The Rhizosphere: Biochemistry and organic substances at the soil-plant interface. Marcel Dekker Editorial, USA.

Uselman, S.M., Qualls, R.G., Lilienfein, J., 2012. Quality of soluble organic C, N, and P produced by different types and species of litter: root litter versus leaf litter. Soil Biol Biochem 54:57-67.

van de Voorde, T.F.J., van der Putten, W.H., Bezemer, T.M., 2011. Intra- and interspecific plant–soil interactions, soil legacies and priority effects during old-field succession J Ecol doi: 10.1111/j.1365-2745.2011.01815.

van der Heijden, M.G.A., Klironomos, J.N., Ursic, M., Moutoglis, P., Streitwolf-Engel, R., Boller, T., Wiemken, A., Sanders, I.R., 1998. Mycorrhizal fungal diversity determines plant biodiversity, ecosystem variability and productivity. Nature 396: 69–72.

van der Heijden, M.G.A., Bardgett, R.D., van Straalen, N.M., 2008. The unseen majority: soil microbes as drivers of plant diversity and productivity in terrestrial ecosystems. Ecol Lett 11:296-310.

van der Putten, W.H., Peters, B.A.M., 1997. How soil-borne pathogens may affect plant competition. Ecology 78(6):1785-1795.

van der Putten, W.H., 2000. Pathogen-driven forest diversity. Nature 404:232-233.

van der Putten, W.H., Bardgett, R.D., de Ruiter, P.C., Hol, W.H.G., Meyer, K.M., Bezemer, T.M., Bradford, M.A., Christensen, S., Eppinga, M.B., Fukami, T., Hemerik, L., Molofsky, J., Schädler, M., Scherber, C., Strauss, S.Y., Vos, M., Wardle, D.A., 2009. Empirical and theoretical challenges in aboveground–belowground ecology. Oecologia 161:1–14.

van der Putten, W.H., Bardgett, R.D., Bever, J.D., Bezemer, T.M., Casper, B.B., Fukami, T., Kardol, P., Klironomos, J.N., Kulmatiski, A., Schweitzer, J.A., Suding, K.N., van de Voorde, T.F.J., Wardle, D.A., 2013. Plant-soil feedbacks: the past, present and future challenges. J Ecol 101(2):265-276.

van der Stoel, C.D., van der Putten, W.H., Duyts, H., 2002. Development of a negative plant-soil feedback in the extension zone of the clonal grass Ammophila arenaria following root formation and nematode colonization. J Ecol 90:978-988.

van Grunsven, R.H.A., van der Putten, W.H., Bezemer, T.M., Tamis, W.L.M., Berendse, F., Veenendaal, E.M., 2007. Reduced plant-soil feedback of plant species expanding their range as compared to natives. J Ecol 95:1050-1057.

Viketoft, M., Palmborg, C., Sohlenius, B., Huss-Danell, K., Bengtsson, J., 2005. Plant species effects on soil nematode communities in experimental grasslands. Appl Soil Ecol 30:90-103.

Ward, S.C., 2000. Soil development on restored bauxite mines in south-west Australia. Aust J Soil Res 38(2):453.

Wardle, D.A., 1999. How soil food webs make plants grow. Trends Ecol Evol 14 (11):418-420.

Wardle, D.A., Bardgett, D.R., Klironomos, N.J., Setäla, H., van der Putten, H.W., Wall, H.D. 2004. Ecological linkages between aboveground and belowground biota. Science 304:1629-1633.

Wardle, D.A., 2006. The influence of biotic interactions on soil biodiversity. Ecol Lett 9:870-886.

121

Wardle, D.A., Bardgett, R.D., Klironomos, J.N., Setälä, H., van der Putten, W.H., Wall, D.H., 2004. Ecological linkages between aboveground and belowground biota. Science 304(5677):1629-1633.

Weidenhamer, J.D., Callaway, R.M., 2010. Direct and indirect effects of invasive plants on soil chemistry and ecosystem function. J Chem Ecol 36:59-69.

Weir, T.L., Park, S-W., Vivanco, J.M., 2004. Biochemical and physiological mechanisms mediated by allelochemicals. Curr Opin Plant Biol 7:472-479.

Werner, D., 2001. Organic signals between plants and microorganisms. In: Pinton, R., Varanini, Z., Nannipieri, P. (eds) The Rhizosphere: Biochemistry and organic substances at the soil-plant interface. Marcel Dekker Editorial, USA.

Whisenant, S.G. 1999, Repairing damaged wildlands: A process-oriented, landscape-scale Approach. Cambridge University Press.

White, D.C., Davies, W.M., Nickels, J.S., Kings, J.D., Bobbie, R.J., 1979. Determination of the sedimentary microbial biomass by extractible lipid phosphate. Oecologia 40: 51-62.

Whitfield, F.B., Shea, S.T., Gillen, K.J., Shaw, K.J., 1981. Volatile components from the roots of Acacia pulchella R.Br. and their effect on Phytophtora cinnamomi Rands. Aust J Bot 29:195-208.

Willis, R., 1999. Australian studies on allelopathy in Eucalyptus: A review. In: Inderjit, Dakshini, K.M.M., Foy, L.F. (eds) Principles and practices in plant ecology: allelochemical interactions. CRC Press, USA.

Witt, C., Setälä, H., 2010. Do plant species of different resource qualities form dissimilar energy channels below-ground? Appl Soil Ecol 44:270-278.

Xiaoyun, C., Manqiang, L., Feng, H., Xiaofang, M., Huixin, L., 2007. Contributions of soil micro-fauna (protozoa and nematodes) to rhizosphere ecological functions. Acta Ecol Sin 27(8):3132-3143.

Yeates, G.W., Bongers, T., M de Goede, R.G., Freckman, D.W., Georgieva, S.S., 1993. Feeding habits in soil nematode families and genera – An outline for soil ecologists. J Nematol 25(3):315-331.

Yeates, G.W., 2003. Nematodes as soil indicators: functional and biodiversity aspects. Biol Fertil Soils 37:199–210.

Yeates, G.W., Ferris, H., Moens, T., van der Putten, W.H., 2009. The role of nematodes in ecosystems. In: Wilson MJ, Kakouli-Duarte T (eds) Nematodes as environmental indicators. CABI, UK.

Zavaleta-Mejía, E., Kaloshian, I., 2012. Plant-nematode interactions. In: Manzanilla-López, R.H., Marbán-Mendoza, N. (eds) Practical plant Nematology. Biblioteca básica de Agricultura, México.

Zechmeister-Boltenstern, S., Michel, K., Pfeffer, M. 2011. Soil microbial community structure in European forests in relation to forest type and atmospheric nitrogen deposition. Plant Soil 343(1):37-50.

Zelles, L., Bai, Q.Y., 1993. Fractionation of fatty acids derived from soil lipids by soil phase extraction and their quantitative analysis by GC–MS. Soil Biol Biochem 25:495-507.

Zelles, L., 1999. Fatty acid patterns of phospholipids and lipopolysaccharides in the characterisation of microbial communities in soil: a review. Biol Fetil Soils 29:111-129.

122

Zhang, Z-Y., Pan, L-P., Li, H-H., 2009. Isolation, identification and characterization of soil microbes which degrade phenolic allelochemicals. J Appl Microbiol 108:1839-1849.

123

APPENDIX 1: Carbon substrates (31 in total) used for community-level physiological profiles of soils (CLPPs) Carbohydrates Carboxylic

acids Amino acids

Phenolic acids

Aromatics Various phenolics

Vitamins Alcohols

L-rhamnose Formic L-alanine Cinnamic acid

Urocanic acid

Catechin Thiamine D-mannitol

D-galactose L-ascorbic L-asparagine

Imidazole Coumarin

a-ketobutyric L-methionine

Syringic acid

Quercetin

a-ketoglutaric L-valine Fumaric L-

threonine

L-tartaric L-glutamic acid

Glycolic Succinic Malic Citric D-galacturonic Quinic Lactic

124

APPENDIX 2: PUBLICATIONS

NOTE: Journal article removed due to copyright restrictions. Orozco-Aceves, M., Standish, R. J., & Tibbett, M. (2015). Soil conditioning and plant-soil feedbacks in a modified forest ecosystem are soil-context dependent. Plant and Soil, 390(1-2), 183-194. 10.1007/s11104-015-2390-z

Forest Ecology and Management 338 (2015) 92–99

Contents lists available at ScienceDirect

Forest Ecology and Management

journal homepage: www.elsevier .com/ locate/ foreco

Long-term conditioning of soil by plantation eucalypts and pines doesnot affect growth of the native jarrah tree

http://dx.doi.org/10.1016/j.foreco.2014.11.0070378-1127/� 2014 Elsevier B.V. All rights reserved.

Abbreviations: PSF, plant–soil feedback; PLFA, phospholipid-fatty acid profile;CLPP, community level physiological profile; PCO, principal co-ordinate analysis.⇑ Corresponding author at: Cranfield Soil and Agrifood Institute, Cranfield

University, Cranfield, Bedfordshire MK43 0AL, UK. Tel./fax: +44 (0) 1234750111x2991.

E-mail address: Mark.tibbett@cranfield.ac.uk (M. Tibbett).

Martha Orozco-Aceves a, Rachel J. Standish b, Mark Tibbett a,c,⇑a School of Earth and Environment, University of Western Australia, M087, 35 Stirling Hwy, Crawley, WA 6009, Australiab School of Plant Biology, University of Western Australia, M090, 35 Stirling Hwy, Crawley, WA 6009, Australiac Cranfield Soil and Agrifood Institute, Cranfield University, Cranfield, Bedfordshire MK43 0AL, UK

a r t i c l e i n f o

Article history:Received 3 September 2014Received in revised form 3 November 2014Accepted 11 November 2014

Keywords:Bauxite miningPlant–soil feedbackSoil biologySoil chemistrySoil feedbackRestoration

a b s t r a c t

Plant species can condition the physico-chemical and biological properties of soil in ways that modifyplant growth via plant–soil feedback (PSF). Plant growth can be positively affected, negatively affectedor neutrally affected by soil conditioning by the same or other plant species. Soil conditioning by otherplant species has particular relevance to ecological restoration of historic ecosystems because sites setaside for restoration are often conditioned by other, potentially non-native, plant species. We investigatedchanges in properties of jarrah forest soils after long-term (35 years) conditioning by pines (Pinus radiata),Sydney blue gums (Eucalyptus saligna), both non-native, plantation trees, and jarrah (Eucalyptus margin-ata; dominant native tree). Then, we tested the influence of the conditioned soils on the growth of jarrahseedlings. Blue gums and pines similarly conditioned the physico-chemical properties of soils, which dif-fered from soil conditioning caused by jarrah. Especially important were the differences in conditioning ofthe properties C:N ratio, pH, and available K. The two eucalypt species similarly conditioned the biologicalproperties of soil (i.e. community level physiological profile, numbers of fungal-feeding nematodes,omnivorous nematodes, and nematode channel ratio), and these differed from conditioning caused bypines. Species-specific conditioning of soil did not translate into differences in the amounts of biomassproduced by jarrah seedlings and a neutral PSF was observed. In summary, we found that decades of soilconditioning by non-native plantation trees did not influence the growth of jarrah seedlings and willtherefore not limit restoration of jarrah following the removal of the plantation trees.

� 2014 Elsevier B.V. All rights reserved.

1. Introduction

Plant–soil feedbacks (PSF) can influence the restoration ofnative forests where soil has been conditioned by non-native spe-cies for long periods of time (i.e. agricultural, forestry activities,invader species; Kulmatiski et al., 2006; Jordan et al., 2008;Kulmatiski and Beard, 2008). Specifically, non-native species cancondition soil properties in ways that lead to reduced growth ofnative species (van Grunsven et al., 2007; Perkins and Nowak,2012; Suding et al., 2013), due to alterations of nutrient cyclingprocesses (Bezemer et al., 2006; Kulmatiski et al., 2006; Kardoland Wardle, 2010; Weidenhamer and Callaway, 2010), imbalance

of microbial communities (i.e. free-living organisms, plant-patho-gens and root symbionts; Inderjit and van der Putten (2010)),and negative effects of toxic allelochemicals (Bertin et al., 2003;Badri and Vivanco, 2009; Inderjit and van der Putten, 2010;Weidenhamer and Callaway, 2010). Plant-conditioned propertiesof soil can persist for long periods of time even after the removalof the non-native species (Kardol and Wardle, 2010).

Soil conditioning operates through two mechanisms: (1) leaflitter deposition, and (2) root deposition. In particular, litter qual-ity is a species-specific plant trait (i.e. C:N ratio, lignin and pheno-lic content) that directly influence the chemical and biologicalproperties of soil. Elements released from litter during decompo-sition alter the concentrations of soil macro and micronutrients ina species-specific manner (Hobbie, 1992; Guo and Sims, 1999;Berg and McClaugherty, 2008; Witt and Setälä, 2010; Uselmanet al., 2012). As well, litter deposition drives changes in themicrobial and microfauna (i.e. nematode) community, includingits biomass, composition, and physiological properties (Berg and

M. Orozco-Aceves et al. / Forest Ecology and Management 338 (2015) 92–99 93

McClaugherty, 2008). Generally, plant species that produce labilelitter (i.e. low C:N ratio) such as legumes, promote a microbialcommunity dominated by bacteria and their consumers (i.e. pro-tozoa, bacterial-feeding nematodes). In contrast, plant speciesthat produce recalcitrant litter (i.e. high C:N ratio) such as pinesand eucalypts promote a microbial community dominated byfungi and their consumers (i.e. fungal-feeding nematodes, collem-bolans, oribatid mites; Moore and Hunt, 1988; Scheu et al., 2005;Moore et al., 2007; Holtkamp et al., 2008; Witt and Setälä, 2010).Thus, in this manner soil conditioning occurs via differences inleaf litter quality.

Bauxite has been mined in the jarrah forest of south-west Wes-tern Australia since 1964. Initial attempts to restore mine sitespost-bauxite mining consisted of commercial plantations of fast-growing, exotic softwoods, mainly pines (Pinus radiata) andcypresses (Cupressus sp.), as well as eastern Australian hardwoodeucalypts such as Sydney blue gum (Eucalyptus saligna; Koch,2007a,b). To date there are 3600 ha of plantations of non-nativespecies within the jarrah forest ecoregion that will be harvestedfor timber in the near future and then restored to native jarrah for-est (Conservation Commission of Western Australia, 2012). Atthese sites non-native trees have been conditioning the soil for35 years, causing species-specific changes to soil properties. Here,we tested species-specific conditioning of soil by pines (P. radiata),Sydney blue gums (E. saligna; both non-native, plantation trees),and jarrah (Eucalyptus marginata; dominant native tree). Wasexpected that the conditioning process of soil by the three plantspecies would be primarily driven by litter deposition, thus influ-enced by litter characteristics of each species in particular. Pinesand eucalypts both produce recalcitrant litter; nonetheless, differ-ences in conditioning processes occur between the two generaassociated with changes in the soil chemistry and biology owingto differences in litter quality.

Conditioning of soil by pines via litter deposition includes acid-ification of soil, alteration of mineral concentrations such as N, K,Mg, and Ca, and low microbial activity (Feller, 1983; Lavelle andSpain, 2003; Ehrenfeld et al., 2005). In contrast, allelopathy hasbeen described as a key mechanism driving soil conditioning byeucalypts. Litter extracts from eucalypts inhibit the germinationof a diversity of plant species as well as the growth of microorgan-isms such as nitrifying bacteria (Bezuidenhout and Laing, 2006;Kohli et al., 2006). There is the potential for pines and eucalyptsto condition soils differently but no tests of the influence of theconditioned soils on plant growth, and in particular how condition-ing of forest soil by non-native species will influence the growth ofnative species such as jarrah via PSF are available. Jarrah displays ahigh degree of growth plasticity in response to resource availabilityand the soil environment (Bleby et al., 2009). This plasticity sug-gests that jarrah is capable of producing constant biomass evenunder variable soil conditions.

The aims of this research were to determine what differences (ifany) occurred in the soils after 35 years of conditioning by non-native species, and to determine the effect of conditioning on thesubsequent growth of jarrah (E. marginata) seedlings (i.e. the neteffect of the PSF). We predicted that the plant species (pine, jarrahand blue gum) would condition the soil in a species-specific man-ner (Aponte et al., 2013), however, differences in soil conditioningwould be more marked after pine than eucalypt growth since theformer presents litter with composition that is different from thatof the eucalypts. In contrast, differences in conditioning betweenblue gums and jarrah would be less distinct due to similar litterquality given that both belong to the same genus. The second spe-cific aim was to test the net PSF associated with jarrah plantsgrown in soils conditioned by non-native plants by comparingtheir growth in soils conditioned by jarrah. We predicted that spe-cies-specific conditioning of soil by non-native plants would not

affect the biomass produced by jarrah seedlings (i.e. a neutralPSF would be observed).

2. Materials and methods

2.1. Sampling site and experimental design

Soils for the experiment were collected from a site that hadbeen cleared of jarrah forest, mined for bauxite, and then plantedwith diverse native and non-native eucalypt species, pines andcypresses to create an arboretum in 1976 (i.e. 35 years prior tosampling). Soil samples were collected in June 2011. At the site,six pines (P. radiata: Pinaceae), six jarrah (E. marginata: Myrtaceae)and six blue gums (E. saligna: Myrtaceae) were selected at randomand four intact soil cores (0.785 L; diameter = 10 cm,depth = 10 cm) were taken from beneath the canopy area of eachtree where litter is deposited. Two cores were used to measurethe physico-chemical and biological properties of soils to deter-mine species-specific conditioning. These soils were sieved (2-mm mesh) and stored at 4 �C pending analyses, and a subsampleof soil was frozen (�20 �C) and then subsequently freeze-driedpending phospholipid-fatty acid analysis (PLFA). The two remain-ing cores were placed in bottom-sealed pots (1.2 L; diame-ter = 13.3 cm, height = 12.4 cm) ready to use in the pot experiment.

Pots containing the intact soil cores were placed into a rootcooling tank that was housed in a glasshouse with temperaturesranging between 4.5 and 27.8 �C for the duration of the experi-ment. To produce seedlings, seeds were surface-disinfected withethanol (70%) and bleach (50%) to prevent growth of pathogensand germinated at 14 �C in darkness. Several jarrah seedlings wereplanted in each pot, and once a plant established, the superfluousseedlings were removed leaving only one plant per pot. As a resultthere were jarrah seedlings growing in soil cores collected in thefield from jarrah trees (i.e. home soils) and in soil cores collectedfrom pines or blue gums (i.e. away soils). Pots were rotated withinthe tank every two weeks to negate a block effect and wateredonce a week to 60% water-holding capacity. After four monthsgrowth (from June to October 2011), plants were harvested; shootsand roots were removed, oven-dried for one week at 60 �C andweighed.

2.2. Response variables

2.2.1. Soil chemical propertiesTotal N (%), total C (%), available (bicarbonate extractable) P

(mg kg�1) and K (mg kg�1), pH (CaCl2) and EC (dS m�1) were deter-mined by the commercial laboratory Cummings Smith BritishPetroleum (CSBP, Bibra Lake Western Australia). Total N and totalC were measured in a combustion analyser (LECO� USA), availableP, available K, pH, EC, were determined using standard methodsdescribed in Rayment and Higginson (1992). Ammonium (NH4

+

mg kg�1) and nitrate (NO3� mg kg�1) analyses were done simulta-

neously by the Skalar Autoanalyser (Skalar Analytical, The Nether-lands) in laboratories at the University of Western Australia.

2.2.2. Soil biology and associated properties2.2.2.1. Microbial biomass-carbon (C). Microbial biomass-C wasextracted by chloroform-fumigation technique described inColeman et al. (2004), and then measured with a Total Organic-CAnalyser (Shimadzu 5000A, Japan). For calculations a kEC factor of0.45 was used and data were reported on a dry soil basis (lg C g�1

dry soil).

2.2.2.2. Phospholipid-fatty acid (PLFA) profiles. Phospholipid-fattyacid profiles of soils were determined following the procedure

94 M. Orozco-Aceves et al. / Forest Ecology and Management 338 (2015) 92–99

described in White et al. (1979). Solid phase extraction tubes (sil-ica-bonded polar, 1 mL volume, Supelclean™) were used to sepa-rate phospholipids from neutral lipids and glycolipids.Phospholipids were subjected to a mild alkaline hydrolysis to formfatty acid methyl esters (FAMEs) as described by Zelles and Bai(1993). Individual PLFAs were detected using gas chromatographwith flame ionization detector (Agilent 7890A Dean Switch GCFID).Fifty lL (75 lg mL�1) of methyl-nonadecanoate (Me-C19 Sigma)were added to FAMEs extracts as internal standard. PLFAs wereidentified by comparing their retention times with those ofqualitative standards of methyl esters (Supelco�). To calculateconcentrations of FAMEs, the following formula was used:

lgindividual fatty acid g�1 soil ¼ PFAME � lgStd

PISTD �W

where PFAME and PISTD are the peak areas of fatty acid and the inter-nal standard respectively; lgStd is the concentration of the internalstandard (lg lL�1 solvent), and W is the dry weight of soil used (g).Finally, biomarkers for specific groups of soil organisms such as bac-teria (15:0i, 15:0a, 16:0i, 17:0i, 17:0cy), fungi (18:2x6,9, 18:1x9),and microeucaryotes (22:0, 24:0, 25:0; Zelles, 1999; Treonis et al.,2004; Bezemer et al., 2006) were determined.

2.2.2.3. Community level physiological profile (CLPP) of soils. Todetermine CLPPs of the soils the MicroResp™ methodology wasused (Campbell et al., 2003; Lalor et al., 2007). Types of C sub-strates, as well as concentrations of substrates and reagents in cre-sol red detector plates were the same used by Lalor et al. (2007).After a 6 h incubation period at 25 �C, absorbance of detector plates(at 590 nm) was determined using a microplate plate reader (ASYSHitech, EXPERT 96) and headspace %CO2 was calculated with thecalibration curve equation experimentally obtained utilizing theplate reader mentioned above: 0.6(exp(�9.95(x � 0.42)) + 0.06.Finally the CO2 rate (lg CO2–C g�1 soil h�1) was calculated asfollows:

%CO2100 � vol� 44

22� 1244� 273

ð273þTÞsoilfwt�soil%dwt

100h

where %CO2 is the absorbance value interpolated in the calibrationcurve; 44, 22, 12 and 273 are constants; T is the incubation temper-ature (�C), vol is the headspace volume in microplate wells (ll), soil-fwt is the fresh weight of soil per well (g), h is the incubation time inhours and soil%dwt is soil sample %dry weight.

2.2.2.4. Nematode trophic community. Nematodes were extractedfrom 100 g of fresh soil (per sample) using a mistifier as describedin Hooper et al. (2005). Aqueous extracts were collected and nem-atodes within them were sorted per1 mL extract using a countingcell (Sedgewick Rafter cell, Graticules England) observed at 100�magnification (eyepiece = 10�; lens = 10�) in a compound micro-scope (Olympus Vanox). Nematodes of the following trophicgroups were identified and counted according to morphologicalstructures described in Yeates et al. (1993): bacterial feeders (BF;Rhabditida, Enoplida), fungal feeders (FF; Tylenchida), omnivorous(Dorylaimida), predators (Mononchida) and plant parasites(Tylenchida, Dorylaimida). Data were expressed as the number ofnematodes of each group per 100 g of dry soil. Additionally, thenematode channel ratio (FF/FF + BF) was calculated as an indicatorof the decomposition channel (i.e. bacterial or fungal) working insoils (Yeates, 2003).

2.3. Statistical analyses

To determine the evidence for species-specific conditioning offield soils the physico-chemical and biological properties (includ-

ing total PLFAs and specific biomarkers for groups of organisms,CLPPs and nematode trophic groups) of soils were compared usingmultivariate and univariate permutational analyses of variance(PERMANOVA) followed by pair-wise tests. Although PERMANOVAis a multivariate method, it can be used for univariate analyses.Performing PERMANOVA using one response variable and Euclid-ean distance yields Fisher’s traditional univariate F-statistic(Anderson, 2010). Scaling plots from principal co-ordinate analysis(PCO; Anderson, 2001) were built to visualize the results of PER-MANOVA analyses. Vectors showing the properties causing themajor effects on the separation of clusters along the main PCOs(i.e. Pearson correlation coefficient with PCO1 and PCO2 > 0.7)were included in some of the plots.

To determine the net effects of the PSF (i.e. magnitude anddirection), we used univariate PERMANOVAs with total, shootand root biomass produced in home vs. away soil as response vari-ables. A significant increase of the plant biomass production inhome with respect to away soil was considered a net positivePSF. A significant decrease of the plant biomass production inhome soil with respect to away soil was considered a net negativePSF. No difference in biomass production in home with respect toaway soil was considered a neutral PSF.

For all the statistical analyses PRIMER v6 + PERMANOVA (PRI-MER-E Ltd) was used. For the multivariate analyses, permutationof residuals under a reduced model was selected as permutationmethod as well as type III (partial) sum of squares. The unrestrictedpermutation of raw data as permutation method and type III (par-tial) sum of squares were used for the univariate analyses. In allcases data were transformed when necessary and Euclidean dis-tance was used to build resemblance matrices, except for the anal-yses of CLPPs and the trophic groups of the nematode communitywhen Bray–Curtis coefficient and Manhattan distance were respec-tively utilized to build the similarity matrices.

3. Results

3.1. Properties of soils conditioned in the field by pines, jarrah, andblue gums

The physico-chemical properties of soils conditioned in the fieldwere significantly different according to plant species (F2,26 = 2.83,P = 0.003; Fig. 1a). Pair-wise comparisons showed that the physico-chemical properties of soils conditioned by jarrah differed fromthose conditioned by blue gums (t = 1.74, P = 0.031), and pines(t = 1.91, P = 0.034), and that soils conditioned by blue gums andpines were not significantly different from each other (t = 1.36,P = 0.06). The biological properties of field soils were differentiallyconditioned according to plant species (F2,16 = 2.24, P = 0.007;Fig. 1b). Pair-wise comparisons showed that the biological proper-ties of soils conditioned by pines differed from those conditionedby jarrah (t = 1.72, P = 0.01) and soils conditioned by blue gums(t = 1.73, P = 0.013), and that soils conditioned by jarrah and bluegums were not significantly different from each other (t = 1.005,P = 0.45).

Statistical analyses of individual physico-chemical properties(Table 1) revealed differences in species-specific conditioning ofthe C:N ratio (F2,2.34 = 21.11, P = 0.0002), pH (F2,0.04 = 3.88,P = 0.042) and K concentration (F2,37.7 = 18.2, P = 0.001). The C:Nratio of soils conditioned by the three species were all significantlydifferent (jarrah versus pine t = 6.02, P = 0.0016; jarrah versus bluegum t = 3.52, P = 0.0081; pine versus blue gum t = 3.24, P = 0.0113).Soil pH conditioned by blue gums statistically differed from pH ofsoils conditioned by pines (t = 2.78, P = 0.024), and K concentrationof soils conditioned by the three species were statistically differentfrom each other (jarrah versus pine t = 2.63, P = 0.042; jarrah

(a) (b)

Fig. 1. Scaling plots from principal co-ordinate analysis (PCO) of (a) physico-chemical properties: bulk density (g cm�3), total C and total N (%), C:N ratio, available P and K(mg kg�1), EC (dS m�1), pH (CaCl2), NO3

� and NH4+ (mg kg�1) and, (b) biological properties: microbial biomass-C (lg g�1 soil), bacterial-, fungal-feeding, plant-parasitic,

predatory, and omnivorous nematodes (all per 100 g soil), and the nematode channel ratio, of soils conditioned in the field by pines, jarrah, and blue gums. The plots showthat jarrah differentially conditioned the physical–chemical properties of soils compared with the other two species, whereas pines differentially conditioned the biologicalproperties of soils compared with the other two species.

Table 1Mean values for the physico-chemical and biological properties (± SE; n = 6) of soils conditioned by pines, jarrah and blue gums. From all properties, only available K, pH, C:N ratio,fungal-feeding nematodes, omnivorous nematodes and the nematode channel ratio significantly differed among soils. Values in bold font indicate significant differences amongconditioned soils, means with the same letters were not significantly different. Mean values of respiration rates from the 31 substrates used in CLPPs are shown per type ofsubstrate (i.e. thirteen carboxylic acids, six amino acids, three phenolic compounds, three aromatic compounds, three carbohydrates, one alcohol, one phenolic acid, and onevitamin). PLFA biomarkers for total bacteria, Gram + bacteria, Gram � bacteria, total fungi, and microeucaryotes are reported.

Plant species

Soil property Pine Jarrah Blue gum

Density (g cm�3) 1.18 (0.22) 1.52 (0.08) 1.24 (0.22)Gravel (%) 66.85 (1.98) 57.01 (7.48) 59.03 (3.94)Available P (mg kg�1) 5.8 (1.2) 3.50 (0.76) 7.4 (1.32)Available K (mg kg�1)** 85.0 (5.67)b 55.17 (10.59)a 152.67 (17.97)cEC (dS m�1) 0.09 (0.01) 0.06 (0.01) 0.08 (0.02)pH* 4.08 (0.21)a 4.40 (0.18)ab 4.77 (0.12)bTotal N (%) 0.17 (0.02) 0.15 (0.06) 0.20 (0.04)Total C (%) 5.14 (0.71) 2.91 (1.40) 4.85 (1.04)NO3� (mg kg�1) 4.83 (1.22) 2.83 (1.02) 8.06 (5.21)

NH4+ (mg kg�1) 0.26 (0.12) 0.04 (0.03) 0.08 (0.08)

C:N ratio*** 30.12 (1.44)b 18.05 (1.34)a 24.26 (1.08)cMicrobial biomass-C (lg-C g�1 soil) 543.37 (221.83) 362.58 (162.15) 508.59 (221.78)Bacterial-feeding nematodes (in 100 g soil) 344 (86) 573 (299) 413 (114)Fungal-feeding nematodes (in 100 g soil)* 4191 (1400)b 1474 (785)a 1034 (398)aOmnivorous nematodes (in 100 g soil)* 79 (22)a 5 (5)ab 32 (26)bcPredatory nematodes (in 100 g soil) 0 3 (3) 0Plant-parasitic nematodes (in 100 g soil) 13 (9)a 18 (18)b 18 (18)bNematode channel ratio* 0.88 (0.05) 0.70 (0.03) 0.57 (0.10)Total CO2 produced from carboxylic acids (lg CO2–C g�1 soil h�1) 8.70 (3.70) 16.51 (1.99) 16.91 (1.93)Total CO2 produced from aminoacids (lg CO2–C g�1 soil h�1) 1.49 (0.60) 2.94 (1.06) 3.21 (0.55)Total CO2 produced from phenolics (lg CO2–C g�1 soil h�1) 0.12 (0.04) 0.22 (0.09) 0.25 (0.08)Total CO2 produced from aromatics (lg CO2–C g�1 soil h�1) 0.30 (0.12) 0.83 (0.17) 1.41 (0.30)Total CO2 produced from carbohydrates (lg CO2–C g�1 soil h�1) 0.87 (0.32) 1.46 (0.19) 1.83 (0.33)Total CO2 produced from alcohols (lg CO2–C g�1 soil h�1) 0.17 (0.07) 0.36 (0.08) 0.35 (0.06)Total CO2 produced from phenolic acids (lg CO2–C g�1 soil h�1) 0.05 (0.02) 0.19 (0.03) 0.20 (0.07)Total CO2 produced from vitamins (lg CO2–C g�1 soil h�1) 0.004 (0.004) 0.001 (0.0004) 0.023 (0.023)Total bacterial fatty acids (lg g�1 soil) 17.68 (5.85) 17.09 (4.73) 12.45 (1.56)Fatty acids from Gram + bacteria (lg g�1 soil) 3.39 (1.49) 2.84 (1.32) 2.0 (0.39)Fatty acids from Gram � bacteria(lg g�1 soil) 4.59 (0.79) 4.59 (0.50) 4.19 (0.49)Total fungal fatty acids (lg g�1 soil) 23.26 (8.48) 26.07 (9.82) 16.89 (1.87)Fatty acids from microeucaryotes (lg g�1 soil) 2.59 (1.0) 1.64 (0.86) 1.54 (0.24)

* P < 0.05.** P < 0.01.

*** P < 0.001.

M. Orozco-Aceves et al. / Forest Ecology and Management 338 (2015) 92–99 95

versus blue gum t = 5.14, P = 0.004; pine versus blue gum t = 3.93,P = 0.004).

Statistical analyses of individual biological properties (Table 1)revealed species-specific conditioning of the numbers of fungal-feeding nematodes (F2,1904 = 3.77, P = 0.03), omnivorous nema-

todes (F2,5219 = 4.11, P = 0.039), and the nematode-channel ratio(F2,0.06 = 4.6, P = 0.02). The numbers of fungal-feeding nematodessignificantly differed in soils conditioned by pines as comparedwith soils conditioned by jarrah (t = 2.41, P = 0.029) and soils con-ditioned by blue gums (t = 2.21, P = 0.036). The numbers of fungal-

96 M. Orozco-Aceves et al. / Forest Ecology and Management 338 (2015) 92–99

feeding nematodes present in soils conditioned by the two euca-lypt species were not significantly different from one another(t = 0.49, P = 0.76). The number of omnivorous nematodes signifi-cantly differed in soils conditioned by pines with respect to soilsconditioned by jarrah (t = 3.05, P = 0.015), but not with respect tosoils conditioned by blue gums (t = 1.91, P = 0.07). The numbersof omnivorous nematodes present in soils conditioned by thetwo eucalypt species were not significantly different from eachother (t = 0.69, P = 0.47). The nematode channel ratio conditionedby pines was different to that conditioned by jarrah (t = 2.78,P = 0.033) and by blue gums (t = 2.54, P = 0.0188 respectively). Soilconditioned by pines presented the highest nematode channelratio.

No significant differences were observed in the total PLFAs (i.e.thirty-nine different PLFAs; F2,5.9 = 0.73, P = 0.61), in biomarkers fortotal bacteria (F2,4.8 = 0.76, P = 0.57), Gram + bacteria (F2,0.7 = 0.86,P = 0.5), Gram � bacteria (F2,0.4 = 0.51, P = 0.72), total fungi(F2,1.82 = 0.57, P = 0.69), or microeucaryotes (F2,0.28 = 0.33, P = 0.89)of the three field-conditioned soils, thus there was no evidenceof plant conditioning of the structure of the microbial community.In contrast, species-specific conditioning was observed for CLPPs(F2,94 = 3.58, P = 0.0284; Fig. 2), pair-wise tests revealed that CLPPsin soils conditioned by pines significantly differed from CLPPs insoils conditioned by blue gums (t = 2.17, P = 0.043), but CLPPs insoils conditioned by pines did not differ from CLPPs in soils condi-tioned by jarrah (t = 1.85, P = 0.086). CLPPs in soils conditioned byjarrah and blue gums were similar (t = 0.96, P = 0.4; Fig. 2). In gen-eral, the microbial community conditioned by pines metabolizedthe C-substrates from CLPPs at lower rates as compared with theother two plant species (Table 1). Fig. 3 is a summary of the effectsof pines, jarrah and blue gums on the physico-chemical and biolog-ical properties of the jarrah forest soils.

3.2. The net effect of the PSF

Jarrah seedlings grown in soils conditioned by pines, blue gums(i.e. away soils), and jarrah (i.e. home soils) produced similaramounts of biomass (Fig. 4). Specifically, total biomass(F2,0.2 = 0.91, P = 0.418), shoot biomass (F2,0.02 = 1.99, P = 0.17),and root biomass (F2,0.003 = 1.33, P = 0.28) of jarrah seedlings didnot differ in soils conditioned by the three plant species, and thusneutral PSF was associated with jarrah plants.

4. Discussion

Conditioning of soil properties was observed in field soils col-lected beneath pines, jarrah and blue gums. A first prediction

Fig. 2. Scaling plot from principal co-ordinate analysis (PCO) of CLPPs of soilsconditioned in the field by pines, jarrah, and blue gums. Thirty-one different Csubstrates were tested and used for the analyses. The plot shows significantdifferences between CLPPs of soils conditioned by pines and CLPPs of soilsconditioned by blue gums (P < 0.05). The other comparisons between pairs oftreatments were not significantly different.

was that both eucalypt species would similarly condition the prop-erties of soil and that these species would condition soils differ-ently from pines. The results supported the prediction for thebiological properties (i.e. CLPPs, numbers of fungal-feeding nema-todes, omnivorous nematodes, and nematode channel ratio), thesesoils were similarly conditioned by both eucalypt species and con-ditioning by these trees differed from conditioning of soil by pines.Nevertheless, the initial prediction was not supported by results ofthe physico-chemical properties because these were differentiallyconditioned by the two eucalypt species. In contrast, blue gumsand pines similarly conditioned the physico-chemical propertiesof soils. Especially important were the differences in conditioningof the properties C:N ratio, pH, and available K. Despite the differ-ences, species-specific conditioning of soil did not translate intodifferences in the amounts of biomass produced by jarrah seed-lings and neutral PSF was observed as predicted. This result sug-gests that restoring native jarrah trees to soils that have beenconditioned by non-native species for decades may not be as prob-lematic as envisaged.

It was assumed that differences in properties of field-condi-tioned soils grown with the three plant species were mainly theresult of conditioning via leaf litter deposition rather than rootdeposition. This assumption was based on the fact that soils forthe experiment were collected from the upper-horizon (0–10 cm),and from beneath tree canopy were litter deposition has occurredfor 35 years. Conditioning of soil via litter deposition is directlyrelated to the chemical composition of litter (Hobbie, 1992; Guoand Sims, 1999; Berg and McClaugherty, 2008; Witt and Setälä,2010; Uselman et al., 2012). For example, litter with high C:N ratiocauses an increase in the C:N ratio of soil and vice versa. Soils col-lected from beneath pines can have higher C:N ratios (i.e. 28–36;O’Brien et al., 2003; Hopmans and Elms, 2009) than soils collectedfrom beneath eucalypts (i.e. 14–23; Adams and Attiwill, 1986;O’Brien et al., 2003) in response to differences in litter composition.These results are in keeping with reports for soils previously col-lected from pines (i.e. C:N of 30), whereas soils collected from jarrahand blue gums sites presented a C:N ratio within the range of 20–25that was slightly higher than the value reported in the literature(Adams and Attiwill, 1986; O’Brien et al., 2003).

Generally, litter with C:N ratios between 20 and 30 supply thenecessary nutritional N-requirements for soil microbes avoidingimmobilization by microbial biomass, which in turn, can limitplant growth (Foth, 1990; Lavelle and Spain, 2003; Johnson,2009; Hopkins and Dungait, 2010). Since soils collected from pineshad the highest mean C:N ratio (i.e. 30) of all soils, it was expectedfor the microorganisms inhabiting these soils to be more N-limitedas compared with those inhabiting soils conditioned by eucalypts.The results supported this assertion; soils conditioned by pinespresented the lowest microbial respiration rates (from CLPPs) ofthe three field-conditioned soils, which indicated a lower microbialactivity. Chemical analyses of soil revealed that the amounts oftotal N, NO3

�, and NH4+ did not significantly differ in soils condi-

tioned by pines from soils conditioned by jarrah or blue gums.Therefore, N limitations in pine soils may be a result of N beingunavailable rather than low concentration of the nutrient, likelyin response to an increased amount of soil C proportionally. And/or to other mechanisms such as releasing of polyphenols by pinesthat bind to organic N, pines then utilize this form of N that isunavailable to other plants and to microbes (Ehrenfeld et al.,2005). In soils conditioned by eucalypt species, lower soil C:Nratios translated into higher rates of respired-C, which indicatedthat soil microbes were less limited by N resources and were met-abolically more active than microbes in soils conditioned by pines.

The addition of leaf and woody litter to soil can cause soil pH tochange due to release of base cations, organic anions and N duringlitter decomposition (Bhupinderpal-Singh and Rengel, 2007; Berg

Fig. 3. Schematic diagram of soil conditioning via litter deposition by (a) pine, (b) jarrah, and (c) blue gum. It was assumed that conditioning process of field soils by plantspecies was mostly due to litter deposition due to the location of the sampling but this assumption needs testing. Boxes represent soil properties that were measured: lightgray for chemical and darker shades of gray indicate higher trophic levels. Ovals represent soil processes (inferred rather than measured). Arrows are drawn to represent thelikely relationships among the soil properties. Plus sign (+) indicates a positive feedback regulates the interaction between properties, a minus sign (�) indicates a negativefeedback regulates the interaction between properties.

M. Orozco-Aceves et al. / Forest Ecology and Management 338 (2015) 92–99 97

and McClaugherty, 2008). Due to species-specific litter chemistry,changes in soil pH driven by litter deposition are also species-spe-cific. The three plant species used in the experiment producedchanges in soil pH of different magnitude. Soils collected frombeneath pines presented the lowest pH of all soils (i.e. 4.0), whichis consistent with information reporting acidification of soil by thisspecies (Feller, 1983; Laskowski et al., 1995; Lavelle and Spain,2003). The results also indicated that pH of soils conditioned byjarrah and blue gums were less acidic than soils conditioned bypines. Both eucalypt species conditioned soil pHs that were signif-icantly different to each other (i.e. pH = 4.5 for jarrah, and pH = 5.0for blue gums) but that fall within the range reported for eucalyptspecies (i.e. pH range of 3.8–6.0; Feller, 1983; Adams and Attiwill,1986; O’Brien et al., 2003; Turner and Lambert, 2008; Spain et al.,2014). Species-specific conditioning of soil pH by plants can, inturn, drive differential chemical and biological properties of soil(Jones, 1998; Uren, 2001; Lavelle and Spain, 2003; Bais et al.,2008; Richardson et al., 2009; Madigan et al., 2012). Overall, differ-ences in litter quality may ultimately impact structural and func-tional aspects of soil grown with pines, jarrah and blue gums.

The structure of the microbial communities did not differamong the field-conditioned soils, therefore evidence of species-specific conditioning of this biological property was limited. How-ever, these findings could be a result of differential populations ofmicrobial predators, as the numbers of fungal-feeding and omniv-orous nematodes were significantly higher in soils conditioned bypines than in soils conditioned by the eucalypt species. Increasedpopulations of fungal-feeding and omnivorous nematodes indi-rectly indicated an increased fungal biomass produced in soils con-ditioned by pines. Soils conditioned by the two eucalypt speciespresented lower numbers of fungal-feeding nematodes that wereassociated with higher pH and lower C:N ratios as compared withsoil conditioned by pines. Understanding the interactions amongthe soil food web is critical to being able to interpret these data.

The results for the nematode channel ratio indicated that anenergy channel of organic matter decomposition driven by the fun-gal compartment was operating in soils conditioned by pines (i.e.nematode channel ratio = 0.88), which probably reflects the recal-citrance of pine litter (Moore and Hunt, 1988; Moore et al., 2003;Condron et al., 2010; Witt and Setälä, 2010). Autochthonous fungi

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

Pine (Aw) Jarrah (Ho) Blue gum (Aw)

Dry

biom

ass (

g)

Shoot biomass

Root biomass

Fig. 4. Shoot and root biomass (means ± SE; n = 6) produced by jarrah seedlingsgrown in away soils (Aw = conditioned by pines and blue gums) and home soils(Ho = conditioned by jarrah).

98 M. Orozco-Aceves et al. / Forest Ecology and Management 338 (2015) 92–99

possess the unique ability to decompose recalcitrant organic mat-ter in a variety of ecosystems at slow rates (Lavelle and Spain,2003; de Boer et al., 2005; Berg and McClaugherty, 2008;Condron et al., 2010). Both the bacterial- and fungal-based energychannels were operating in soils conditioned by blue gums (i.e.nematode channel ratio = 0.57). Since the bacterial-based energychannel is faster than the fungal one, quicker rates of organic mat-ter decomposition were occurring in soils conditioned by bluegums, probably in response to the labile litter (Moore and Hunt,1988; Moore et al., 2003; Yeates, 2003; Condron et al., 2010;Witt and Setälä, 2010). Finally, the fungal-based energy channelwas favored over the bacterial-based channel in soils conditionedby jarrah (i.e. nematode channel ratio = 0.7), which indicated a rateof organic matter decomposition that is intermediate betweenpines and blue gums. Differences in organic matter decompositionrates occurring in soils conditioned by the three plant species mayhave implications on soil processes and ultimately on plant com-munity structure and evolution (Hobbie, 1992; Eviner, 2004;Weidenhamer and Callaway, 2010).

The differences in the properties of soils conditioned in the fielddid not affect the net PSF (i.e. magnitude or direction) and neutralPSF was observed for jarrah seedlings during the pot experiment.Previous work demonstrated that jarrah shows a high above-ground architectural plasticity in response to varying environmen-tal conditions (Koch and Samsa, 2007; Bleby et al., 2009). Theresults of this research suggest this plasticity is not exclusive tothe aboveground compartment, but to the belowground compart-ment as well, and is not exclusive to morphological aspects butalso to functional aspects. For that reason, jarrah plants are ableto maintain a stable biomass production despite the changing soilproperties, even those conditioned by different plant species.

5. Conclusion

Non-native plant species can condition soils in ways that differto native species, which has important implications for ecosystemfunctioning. This is critical for the restoration of human-modifiedecosystems, since the ultimate goal of most restoration activitiesis a complete functional ecosystem. In restored jarrah forestpost-bauxite mining, despite alterations to soil properties arecaused by non-native species, jarrah seedlings were able to estab-lish in these conditioned soils, without it affecting their biomassproduction. This suggests that the jarrah-soil system displays a

plastic PSF, a trait of jarrah that might contribute to its persistencefor long time spans, and to adapt to modified soil conditions pres-ent in restored soil where jarrah establishes and grows well.

Acknowledgements

We thank The Mexican National Council for Science and Tech-nology (CONACYT), John Koch (ALCOA World Alumina Australia)and Tim Morald (University of Western Australia – UWA) for theirsupport and help with field work, Michael Smirke (UWA) for histechnical support during laboratory work, Deborah Lin, Khalil Kari-man and Alonso Calvo Araya (UWA) for their help during theexperiment, Vivien Vanstone, Sarah Collins and staff (DAFWA) forhelp with nematode extraction as well as Jackie Nobbs (SARDI)for training on the identification of nematode trophic groups.Finally, we thank Richard Hobbs (UWA) and Ken Dodds (ChemCen-tre – WA) for their support to complete the PLFA analyses.

References

Adams, M.A., Attiwill, P.M., 1986. Nutrient cycling and nitrogen mineralization ineucalypt forests of south eastern Australia II. Indices of nitrogen mineralization.Plant Soil 92, 341–362.

Anderson, M.J., 2001. A new method for non-parametric multivariate analysis ofvariance. Austral Ecol. 26, 32–46.

Anderson, M.J., 2010. Detecting multivariate changes in biological assemblages:experimental design and data analysis. Multivariate analysis of complexexperimental designs using PERMANOVA+ for PRIMER. Institute ofInformation and Mathematical Sciences (IIMS) Massey University, Auckland.

Aponte, C., García, L.V., Marañón, T., 2013. Tree species effects on nutrient cyclingand soil biota: a feedback mechanism favouring species coexistence. For. Ecol.Manage. 309, 36–46.

Badri, D.V., Vivanco, J.M., 2009. Regulation and function of root exudates. Plant, CellEnviron. 32, 666–681.

Bais, H.P., Broeckling, C.D., Vivanco, J.M., 2008. Root exudates modulate plant–microbe interactions in the rhizosphere. In: Karlovsky, P. (Ed.), SecondaryMetabolites in Soil Ecology. Springer, Berlin.

Berg, B., McClaugherty, C., 2008. Plant Litter: Decomposition, Humus Formation,Carbon Sequestration, second ed. Springer, Germany.

Bertin, C., Yang, X., Weston, L.A., 2003. The role of root exudates and allelochemicalsin the rhizosphere. Plant Soil 256, 67–83.

Bezemer, T.M., Lawson, C.S., Hedlund, K., Edwards, A.R., Brook, A.J., Igual, J.M.,Mortimer, S.M., van der Putten, W.H., 2006. Plant species and functional groupeffects on abiotic and microbial soil properties and plant–soil feedbackresponses in two grasslands. J. Ecol. 94, 893–904.

Bezuidenhout, S.R., Laing, M., 2006. Allelopathy and its influence in soil systems. In:Uphoff, N., Ball, A.S., Fernandes, E., Herren, H., Husson, O., Laing, M., Palm, C.,Pretty, J., Sanchez, P., Sanginga, N., Thies, J. (Eds.), Biological Approaches toSustainable Soil Systems. CRC Taylor & Francis Group, USA.

Bhupinderpal-Singh, Rengel, Z., 2007. The role of crop residues in improving soilfertility. In: Marschner, P., Rengel, Z. (Eds.), Nutrient Cycling In TerrestrialEcosystems. Springer-Verlag, Berlin.

Bleby, T.M., Colquhoun, I.A., Adams, M.A., 2009. Architectural plasticity in youngEucalyptus marginata on restored bauxite mines and adjacent natural forest insouth-western Australia. Tree Physiol. 29, 1033–1045.

Campbell, C.D., Chapman, S.J., Cameron, C.M., Davidson, M.S., Potts, J.M., 2003. Arapid microtiter plate method to measure carbon dioxide evolved from carbonsubstrate amendments so as to determine the physiological profiles of soilmicrobial communities by using whole soil. Appl Environ Microbiol 69, 3593–3599.

Coleman, D.A., Crossley, D.A., Hendrix, P.F., 2004. Fundamentals of Soil Ecology,second ed. Elsevier Academic Press, Amsterdam.

Condron, L., Stark, C., O’Callaghan, M., Clinton, P., Huang, Z., 2010. The role ofmicrobial communities in the formation and decomposition of soil organicmatter. In: Dixon, G.R., Tilston, E.L. (Eds.), Soil Microbiology and SustainableCrop Production. Springer, New York.

Conservation Commission of Western Australia, 2012. Draft Forest ManagementPlan 2014–2023. Conservation Commission of Western Australia, Perth.

de Boer, W., Folman, L.B., Summerbell, R.C., Boddy, L., 2005. Living in a fungal world:impact of fungi on soil bacterial niche development. FEMS Microbiol. Rev. 29,795–811.

Ehrenfeld, J.G., Ravit, B., Elgersma, K., 2005. Feedback in the plant–soil system. AnnuRev Environ Resour 30, 75–115.

Eviner, V.T., 2004. Plant traits that influence ecosystem processes varyindependently among species. Ecology 85 (8), 2215–2229.

Feller, M.C., 1983. Effects of an exotic conifer (Pinus radiata) plantation on forestnutrient cycling in Southeastern Australia. For. Ecol. Manage. 7, 77–102.

Foth, H.D., 1990. Fundamentals of Soil Science, Eigth ed. John Wiley & Sons, USA.

M. Orozco-Aceves et al. / Forest Ecology and Management 338 (2015) 92–99 99

Guo, L.B., Sims, R.E.H., 1999. Litter decomposition and nutrient release via litterdecomposition in New Zealand eucalypt short rotation forests. Agric EcosystEnviron 75, 133–140.

Hobbie, S.E., 1992. Effects of plant species on nutrient cycling. Trends Ecol. Evol. 7(10), 336–339.

Holtkamp, R., Kardol, P., van der Wal, A., Dekker, S.C., van der Putten, W.H., deRuiter, P.C., 2008. Soil food web structure during ecosystem development afterland abandonment. Appl Soil Ecol 39, 23–34.

Hooper, D.J., Hallmann, J., Subbotin, S.A., 2005. Methods for extraction, processingand detection of plant and soil nematodes. In: Luc, M., Sikora, R.A., Bridge, J.(Eds.), Plant Parasitic Nematodes in Subtropical and Tropical Agriculture,second ed. CAB Publishing, UK.

Hopkins, D.W., Dungait, J.A.J., 2010. Soil microbiology and nutrient cycling. In:Dixon, G.R., Tilston, E.L. (Eds.), Soil Microbiology and Sustainable CropProduction. Springer, New York.

Hopmans, P., Elms, S.R., 2009. Changes in total carbon and nutrients in soil profilesand accumulation in biomass after a 30-year rotation of Pinus radiata onpodzolized sands: impacts of intensive harvesting on soil resources. For. Ecol.Manage. 258, 2183–2193.

Inderjit, van der Putten, W.H., 2010. Impacts of soil microbial communities onexotic plant invasions. Trends Ecol. Evol. 25, 512–519.

Johnson, C., 2009. Biology of Soil Science. Oxford Book Company, Delhi.Jones, D.L., 1998. Organic acids in the rhizosphere – a critical review. Plant Soil 205,

25–44.Jordan, N.R., Larson, D.L., Huerd, S.C., 2008. Soil modification by invasive plants:

effects on native and invasive species of mixed-grass prairies. Biol. Invasions 10,177–190.

Kardol, P., Wardle, D.A., 2010. How understanding aboveground–belowgroundlinkages can assist restoration ecology. Trends Ecol. Evol. 25 (11), 670–679.

Koch, J.M., 2007a. Bauxite mining restoration by Alcoa World Alumina Australia inWestern Australia: social, political, historical, and environmental contexts.Restor. Ecol. 15 (4), S3–S10.

Koch, J.M., 2007b. Alcoa’s mining and restoration process in south WesternAustralia. Restor. Ecol. 15 (4), S11–S16.

Koch, J.M., Samsa, G., 2007. Restoring jarrah forest trees after bauxite mining inWestern Australia. Restor. Ecol. 15 (4), S17–S25.

Kohli, R.K., Batish, D.R., Singh, H.P., 2006. Allelopathic interactions inagroecosystems. In: Reigosa, M.J., Pedrol, N., González, L. (Eds.), Allelopathy: APhysiological Process With Ecological Implications. Springer, The Netherlands.

Kulmatiski, A., Beard, K.H., Stark, J.M., 2006. Soil history as a primary control onplant invasion in abandoned agricultural fields. J. Appl. Ecol. 43, 868–876.

Kulmatiski, A., Beard, H.K., 2008. Decoupling plant-growth from land-use legacies insoil microbial communities. Soil Biol. Biochem. 40, 1059–1068.

Lalor, E.L., Cookson, W.R., Murphy, D.V., 2007. Comparison of two methods thatassess soil community level physiological profiles in a forest ecosystem. SoilBiol. Biochem. 39, 454–462.

Laskowski, R., Niklinska, M., Maryanski, M., 1995. The dynamics of chemicalelements in forest litter. Ecology 76 (5), 1393–1406.

Lavelle, P., Spain, A.V., 2003. Soil Ecology. Kluwer Academic Publishers, USA.Madigan, M.T., Martinko, J.M., Stahl, D.A., Clark, D.P., 2012. Brock Biology of

Microorganisms, Thirteenth ed. Benjamin Cummings Editorial, USA.Moore, J.C., Hunt, H.W., 1988. Resource compartmentation and the stability of real

ecosystems. Nature 33, 261–263.Moore, J.C., McCann, K., Setäla, H., de Ruiter, P.C., 2003. Top-down is bottom-up:

does predation in the rhizosphere regulate aboveground dynamics? Ecology 84,846–857.

Moore, J.C., McCann, K., de Ruiter, P.C., 2007. Soil rhizosphere food webs, theirstability, and implications for soil processes in ecosystems. In: Cardon, Z.G.,Whitbeck, J.L. (Eds.), The Rhizosphere: An Ecological Perspective. Elsevier, USA.

O’Brien, N.D., Attiwill, P.M., Weston, C.J., 2003. Stability of soil organic matter inEucalyptus regnans forests and Pinus radiata plantations in south easternAustralia. For. Ecol. Manage. 185, 249–261.

Perkins, L.B., Nowak, R.S., 2012. Soil conditioning and plant–soil feedbacks affectcompetitive relationships between native and invasive grasses. Plant Ecol. 213,1337–1344.

Rayment, G.E., Higginson, F.R., 1992. Australian Laboratory Handbook of Soil andWater Chemical Methods: Australian Soil and Land Survey Handbook, vol. 3.Inkata Press, Australia.

Richardson, A.E., Barea, J.M., McNeill, A.M., Prigent-Combaret, C., 2009. Acquisitionof phosphorus and nitrogen in the rhizosphere and plant growth promotion bymicroorganisms. Plant Soil 321, 305–339.

Scheu, S., Ruess, L., Bonkowski, M., 2005. Interactions between microorganisms andsoil micro- and mesofauna. In: Buscot, F., Varma, A. (Eds.), Microorganisms inSoils: Roles in Genesis and Functions. Springer-Verlag, Berlin.

Spain, A.V., Tibbett, M., Hinz, D.A., Ludwig, J.A., Tongway, D.J., 2014. The mining-restoration system and ecosystem development following bauxite mining in abiodiverse environment of the seasonally dry tropics, Northern Territory,Australia. In: Tibbett, M. (Ed.), Mining in Ecologically Sensitive Landscapes.CSIRO Press (in press).

Suding, K.N., Harpole, W.S., Fukami, T., Kulmatiski, A., MacDougall, A.S., Stein, C., vander Putten, W.H., 2013. Consequences of plant–soil feedbacks in invasion. J.Ecol. 101, 298–308.

Treonis, A.M., Ostle, N.J., Stott, A.W., Primrose, R., Grayston, S.J., Ineson, P., 2004.Identification of groups of metabolically-active rhizosphere microorganisms bystable isotope probing of PLFAs. Soil Biol. Biochem. 36, 533–537.

Turner, J., Lambert, M.J., 2008. Nutrient cycling in age sequences of two Eucalyptusplantation species. For. Ecol. Manage. 255, 1701–1712.

Uren, N.C., 2001. Types, amounts, and possible functions of compounds releasedinto the rhizosphere by soil-grown plants. In: Pinton, R., Varanini, Z., Nannipieri,P. (Eds.), The Rhizosphere: Biochemistry and Organic Substances at the Soil–Plant Interface. Marcel Dekker Editorial, USA.

Uselman, S.M., Qualls, R.G., Lilienfein, J., 2012. Quality of soluble organic C, N, and Pproduced by different types and species of litter: root litter versus leaf litter. SoilBiol. Biochem. 54, 57–67.

van Grunsven, R.H.A., van der Putten, W.H., Bezemer, T.M., Tamis, W.L.M., Berendse,F., Veenendaal, E.M., 2007. Reduced plant–soil feedback of plant speciesexpanding their range as compared to natives. J. Ecol. 95, 1050–1057.

Weidenhamer, J.D., Callaway, R.M., 2010. Direct and indirect effects of invasiveplants on soil chemistry and ecosystem function. J. Chem. Ecol. 36, 59–69.

White, D.C., Davies, W.M., Nickels, J.S., Kings, J.D., Bobbie, R.J., 1979. Determinationof the sedimentary microbial biomass by extractible lipid phosphate. Oecologia40, 51–62.

Witt, C., Setälä, H., 2010. Do plant species of different resource qualities formdissimilar energy channels below-ground? Appl Soil Ecol 44, 270–278.

Yeates, G.W., Bongers, T., de Goede, R.G.M., Freckman, D.W., Georgieva, S.S., 1993.Feeding habits in soil nematode families and genera – an outline for soilecologists. J Nematol 25 (3), 315–331.

Yeates, G.W., 2003. Nematodes as soil indicators: functional and biodiversityaspects. Biol. Fertil. Soils 37, 199–210.

Zelles, L., Bai, Q.Y., 1993. Fractionation of fatty acids derived from soil lipids by soilphase extraction and their quantitative analysis by GC–MS. Soil Biol. Biochem.25, 495–507.

Zelles, L., 1999. Fatty acid patterns of phospholipids and lipopolysaccharides in thecharacterisation of microbial communities in soil: a review. Biol. Fertil. Soils 29,111–129.