IMPROVING REFORESTATION SUCCESS OF HIGH-VALUE AND KEY FOREST SPECIES BY DIRECT SEEDING IN SOUTHEAST ASIA AND WESTERN AUSTRALIA

By Thea So BSc. (Sylviculture), MSc. (Tropical Forest Resources Management)

This thesis is submitted in fulfillment of the requirement for the degree of Doctor of Philosophy

School of Biological Sciences and Biotechnology Faculty of Science, Engineering and Sustainability Murdoch University, Perth, Western Australia November 2011

Declaration

DECLARATION

I declare that all the work described in this thesis is my own account of my research which was undertaken while I was enrolled as a full time research student for the degree of Doctor of Philosophy at Murdoch University, Western Australia, from 2007 to 2010. This work has not previously been submitted for a degree at any tertiary education institution.

Thea So November 2011

Murdoch University 2011 ii

Acknowledgement

ACKNOWLEDGMENTS

First of all, I would like to extend my sincere gratitude to my principal supervisor, Professor Bernard Dell, for his tirelessness and inspiration in supervising my study. I appreciate his patience in reading and correcting all my work. My sincere thankfulness to Dr Katinka Ruthrof and Dr Lambert Braü for their guidance, constructive advice and supports in many aspects of this thesis.

I would like to thank Mr Chairat Aramsri, Professor Bunvong Thaiutsa and Ms Montathip Sommeechai for their coordination in the establishment of the field trial in Thailand. A/Professor Mike Calver and Professor Ananchai Khuantham are thanked for their guidance on experimental design. Thanks are due to Mr Jason Foster Ralph Sarich for allowing me to use his land for the field trial in Western Australia. Mycorrhizal inocula were provided by Zadco For Quality Gro Pty Ltd and Dr Phakpen Poomipan. The Forest Industry Organization (FIO) of Thailand is thanked for the excellent collaboration in establishment of the trial on its land. Mr Chaiya Junsawang and all FIO staff at the Sakeaw Forest Plantation provided excellent hospitality and support. Dr Jumnian Wongmo is thanked for her assistance in the field trial establishment in Western Australia and procurement of AM inocula in Thailand. Dr Yvette Hill provided instructions on working with N2-fixing bacteria and in setting up glasshouse experiments. Liza Parkinson provided assistance in molecular work. Mr Dy Sophy is thanked for his assistance in monitoring the research trial in Cambodia. Many people provided assistance during my field work in Cambodia, especially Ry Sam El, Suo Hai, So Than, Long Boung, Moy Rotha, Uon Sam Ol, Huot Ainun, Lim Sopheap, Thai Seila Tina, Nuon Pov Ratana, Ken Phan, Lao Sethaphal, Sreang Meng Srun and Preap Sam. I am indebted to colleagues and fellow students at Murdoch University for their support, encouragement and assistance, especially Harry, Tan, Endah and Lily. Special thanks are due to Peter Scott for his support and encouragement.

The Australian Leadership Awards (ALA) is gratefully acknowledged for granting a scholarship for this study. The field trials in Cambodia and Thailand were partially financed by Murdoch University and AusAID. A big gratefulness is due to my parents, my wife and son, Veasna, sisters and brother for their encouragement, love and support. Last but not least, I would like to express my sincere thanks to my bosses, H. E. Chheng Kim Sun, and H. E. Chea Sam Ang for their encouragement and support of my study.

Murdoch University 2011 iii

Abbreviation

ABREVIATIONS

AM arbuscular mycorrhiza/l ANR assisted natural regeneration ANOVA analysis of variance a.s.l. above sea level BSO breeding seed orchard cm centimetre CITES Convention on International Trade in Endangered Species CRD completely randomized design CSBP Cumming Smith British Petroleum CSO clonal seed orchard CTSP Cambodia Tree Seed Project DBH diameter at breast heigh d.f. degree of freedom DI water distilled deionized water DNA deoxyribonucleic acid DTPA diethylene triamine pentaacetic acid ECM ectomycorrhiza/l EDTA ethylenediaminetetraacetic acid e.g. for example (Latin exemli gratia) EN endangered F (statistic) A value based on a standard statistical test used in ANOVA/MANOVA FA Forestry Administration FAO Food and Agriculture Organization of the United Nations FLD Forest & Landscape Denmark FORRU Forest Restoration Research Unit GoC Government of Cambodia h hour ha hectare i.e. that is (Latin id est) IFSR Independent Forest Sector Review INVAM International Culture Collection of (Vesicular) Arbuscular Mycorrhizal Fungi

Murdoch University 2011 iv

Abbreviation

IUCN International Union for Conservation of Nature L litre MAFF Ministry of Agriculture, Forestry and Fisheries MANOVA multivariate analysis of variance min minute mL millilitre mm millimetre n.a. not available NFP National Forest Programme NGO non-governmental organization nM nanomole nom. ined unpublished name (Latin nomen ineditum) OM organic matter p probability PCR polymerase chain reaction RCBD randomized complete block design rpm revolution per minute rRNA ribosomal ribonucleic acid s second S.E. standard error Sig. significance TAE Tris, acetic acid and EDTA TY-medium tryptone peptone yeast extract-medium UNEP United Nations Environment Programme UTM Universal Transverse Mercator V volt VU vulnerable v:v volume:volume WA Western Australia WCMC World Conservation Monitoring Centre w:v weight:volume w:w weight:weight µL microlitre μm micrometre

Murdoch University 2011 v

Definition of terms

DEFINITION OF TERMS

Buffer (solution) An aqueous solution containing a weak acid in its conjugate base (Chapter 6). Effective Significant improvement in establishment, survival and growth (height, diameter or dry weight), brought about by a treatment (as compared to the control) inoculated to seeds or seedlings under field or glasshouse conditions. Taq DNA A thermostable emzyme wich replicate DNA at 74°C. The ability of polymerase the enzyme to survive multiple rounds of temperature cycling make it extremely useful in PCR and cycle sequencing (Fisher Biotech Australia, www.fisherbiotech.com). Survival The state of seedlings that have survived after trial establishment (seedling) until the time of monitoring.

Treatment A single or a set of materials that were applied or inoculated to seeds or seedlings under field or glasshouse conditions in an expectation that they promote establishment, survival and growth of the seedlings. In the experimental Chapters 3 and 5-7, the term treatment is frequently abbreviated by the letter T followed by a number (e.g., T1, T2, …,T8). Note: The species specific epithet is used when a species is cited the first time. However, for those species used in experiments, the full names are given in each Chapter the first time they are cited.

Murdoch University 2011 vi

Abstract

ABSTRACT

Worldwide, natural forests have been decreasing in area at an alarming rate. In Cambodia, the annual deforestation rate was 127 000 ha year-1 from 2005 to 2010 and this seriously threatens biodiversity and the livelihoods of rural communities. Therefore, there is an urgent need to reforest or establish forest plantations to meet two main objectives: economic development and biodiversity conservation. This thesis concerns the promotion of early survival and growth of planted or direct-seeded seedlings to overcome the harsh conditions of reforestation sites in tropical and mediterranean-type ecosystems, with special attention given to threatened high-value timber species of Southeast Asia.

Worldwide, there is an increasing effort to reforest degraded forests and old agricultural lands. However, reforestation of degraded lands is often difficult and is usually expensive; therefore, direct seeding is an alternative to conventional tree planting. Some of the major constraints facing reforestation efforts were reviewed, and opportunities that could be useful for promotion of early establishment and growth of seedlings were explored. These were then used to define reforestation field studies in Australia, Thailand and Cambodia.

In a harsh mediterranean-type ecosystem in Western Australia, effects of microorganisms

(mycorrhizal fungi and N2-fixing bacteria) and planting material (seed and seedling) on survival and growth of two key post-disturbance colonizing species, Eucalyptus gomphocephala and Acacia saligna, were investigated. For E. gomphocephala, survival at 13 months was higher for out-planted seedlings (81%) than from direct seeding (7.5%). Inoculation with ectomycorrhizal fungal spores was not beneficial. For A. saligna, survival at 13 months was also higher for seedlings (84%) than for seeding (42.5%). Nitrogen-fixing bacteria from crushed root nodules of A. saligna did not promote survival or growth of the species nor did a mixed commercial mycorrhizal inoculum.

In Southeast Asia, the leguminaceous rosewoods, Afzelia xylocarpa and Dalbergia cochinchinensis, are threatened throughout their range by habitat loss and over exploitation for their extremely highly-prized timber. The species have been promoted for reforestation in Cambodia for economic development and genetic conservation. The

Murdoch University 2011 vii

Abstract current conservation status of A. xylocarpa and D. cochinchinensis in Cambodia was examined, and information on silviculture, trade and current conservation measures applied in that country was drawn together. Some important steps in the development of domestication strategies, including testing and improving silvicultural practices and increasing the supply of genetically superior seeds from seed production areas and seed orchards, were outlined. This information was then used to help select species for the trials in Thailand and Cambodia.

Many high-value timber species of continental Southeast Asia, including some rosewoods, have been promoted in reforestation programmes. However, the slow- growing habit at the early stage of development is a challenge for promoting these species in tree plantings. Therefore, effects of beneficial microorganisms and fertilizer on establishment and growth of direct-seeded seedlings of D. cochinchinensis and Xylia xylocarpa were investigated in a trial on former agroforestry land in Thailand and compared to Acacia mangium, an exotic fast-growing plantation species. After 20 months, a mixed inoculum of arbuscular mycorrhizal (AM) + ectomycorrhizal (ECM) fungi and a mixed inoculum of AM fungi + N2-fixing bacteria (crushed root nodule) improved survival of D. cochinchinensis by 15 and 17%, respectively. The co-inoculation of AM with ECM also improved diameter growth of the same species by 43%. A second field trial explored the effects a water retention polymer and fertilizer on direct seeding of A. mangium, Afzelia xylocarpa, D. cochinchinensis, Eucalyptus camaldulensis, Sindora cochinchinensis and X. xylocarpa. The combination of polymer and fertilizer increased height growth of all six tree species by 40%. The effect of the polymer and fertilizer was further investigated in Cambodia with direct seeding of the same indigenous species. There, the combination of the polymer with fertilizer increased seedling establishment only by 7%.

As selected strains of compatible N2-fixing bacteria were not available, crushed root nodules were used in some of the field trials. In order to improve the technology in the future, three strains of N2-fixing bacteria were isolated from root nodules of D. cochinchinensis grown in Cambodia and then tested under glasshouse conditions and a seed coating technique was employed as a means to deliver bacteria to seeds along with broth culture. After 16 weeks, one of the three isolates increased total dry weight of D. cochinchinensis seedlings by ca. 150% over the uninoculated control that was not fed

Murdoch University 2011 viii

Abstract inorganic nitrogen. The effective strain was identified as Bradyrhizobium elkanii after the DNA was amplified by polymerase chain reaction using RPO1 primer (5'-AAT TTT CAA GCG TCG TGC CA-3') and then partial 16S rRNA nucleotide gene sequences were compared with the Gene Bank database. Both methods of delivering of bacteria (seed coating and broth culture) were equally effective. The effectiveness of B. elkanii was explored under field conditions in Cambodia with direct seeding of D. cochinchinensis, but no effect on seedling establishment or growth was obtained after six months of trial establishment. Competition from indigenous bacteria was suggested as one of the reasons for the ineffectiveness of the introduced strain.

The main finding of this thesis was the suitability of the four high-value timber species in reforestation by direct seeding in tropical regions, in former agricultural land where proper site preparation and intensive weeding were provided, as well as on land previously under degraded forest where minimal site preparation was undertaken. Also, the application of a water retention polymer promoted establishment and growth of seedlings under tropical conditions. This should be evaluated further in a wider range of reforestation sites including sandy soils. More work should be undertaken to identify effective symbionts for the tropical rosewoods. These and other symbionts should be evaluated in reforestation trials in which more attention is paid to site characteristics and populations of indigenous beneficial organisms.

Murdoch University 2011 ix

References

TABLE OF CONTENTS

DECLARATION ...... ii ACKNOWLEDGMENTS ...... iii ABREVIATIONS ...... iv DEFINITION OF TERMS ...... vi ABSTRACT ...... vii TABLE OF CONTENTS ...... x

CHAPTER 1: Introduction ...... 1

CHAPTER 2: Literature review: Issues and limitations of reforestation and strategies for improvement of reforestation ...... 5 2.1 Reforestation ...... 6 2.2 Constraints for reforestation programmes ...... 7 2.3 Reforestation by direct seeding ...... 9 2.4 Research priorities in reforestation ...... 10 2.4.1 Species selection ...... 10 2.4.2 Seed procurement and storage ...... 11 2.4.3 Direct seeding ...... 11 2.4.4 Social issues ...... 13 2.5 Some strategies to enhance reforestation success ...... 13 2.5.1 Application of fertilizer and water retention polymers ...... 13 2.5.2 The potential benefits of microorganisms in reforestation ...... 15 2.5.3 Principles and techniques for selection of microorganisms for reforestation ...... 16 2.5.3.1 Methods for screening effective mycorrhizal fungi ...... 16 2.5.3.2 Methods for screening effective rhizobial bacteria for legumes...... 22 2.5.4 Seed coating/pelleting ...... 23 2.5.4.1 Seed coating material and method ...... 24 2.5.5 Biochar―an emerging technology for stimulating plant growth ...... 26 2.5.5.1 Effects of biochar on plant growth ...... 27

Murdoch University 2011 x

References

2.5.5.2 Effects of biochar on microorganisms ...... 28 2.5.6 Artificial seeds ...... 29 2.6 Concluding remarks ...... 29

CHAPTER 3: Reforestation of a degraded Mediterranean-type Eucalyptus gomphocephala A. DC. woodland in Western Australia ...... 31 3.1 Introduction ...... 32 3.2 Materials and methods ...... 33 3.2.1 Site conditions and preparation ...... 33 3.2.2 Procurement of materials ...... 35 3.2.2.1 Seeds and seedlings ...... 35 3.2.2.2 Fertilizer ...... 36 3.2.2.3 Mycorrhizal fungi ...... 36 3.2.3 Experimental design and treatment ...... 37 3.2.4 Trial establishment ...... 38 3.2.5 Data collection and analysis ...... 39 3.3 Results ...... 40 3.3.1 Experiment 1: Eucalyptus gomphocephala ...... 40 3.3.1.1 Seedling survival ...... 40 3.3.1.2 Seedling growth ...... 41 3.3.2 Experiment 2: Acacia saligna ...... 42 3.3.2.1 Seedling survival ...... 42 3.3.2.2 Seedling growth ...... 43 3.3.2.3 Phillode nutrient concentration ...... 43 3.4 Discussion ...... 45 3.4.1 Effects of treatments on survival and growth ...... 45 3.4.2 Seedling survival and growth―differences between plant material ...... 47 3.4.3 Survival from direct seeding―comparison of the two species ...... 48 3.5 Concluding remarks ...... 50

Murdoch University 2011 xi

References

CHAPTER 4: Conservation and utilization of threatened hardwood species through reforestation―an example of Afzelia xylocarpa (Kruz.) Craib and Dalbergia cochinchinensis Pierre in Cambodia ...... 51 4.1 Introduction ...... 52 4.2 Commercial status of Afzelia xylocarpa and Dalbergia cochinchinensis ...... 53 4.2.1 Commercial attributes of the species ...... 53 4.2.2 Utilization and trade ...... 54 4.3 Conservation of the species ...... 56 4.3.1 Threats and conservation ...... 56 4.3.2 Tree domestication ...... 58 4.4 Plantation establishment and community restoration ...... 60 4.4.1 Suitability for reforestation ...... 60 4.4.2 Plantations in Cambodia and the region ...... 60 4.4.3 Plantation technologies: a case study in the Kbal Chhay Watershed area 61 4.4.3.1 Use of nurse trees ...... 62 4.4.3.2 Plantation in open areas ...... 63 4.4.4 Silvicultural treatment of natural (degraded) forests ...... 66 4.4.5 Relationships between tree growth, tree form, site and stand management67 4.5 Challenges and opportunities ...... 69 4.5.1 Importance of reforestation to the wellbeing of local communities ...... 69 4.5.2 Seed supply strategy ...... 69 4.5.3 Ecological and biological characteristics―research needs ...... 71 4.5.4 Development of a tree improvement programme ...... 71 4.6 Concluding remarks ...... 73

CHAPTER 5: Improving reforestation success of high-value and key forest tree species by direct seeding: field trials in Thailand ...... 75 5.1 Introduction ...... 76 5.2 Materials and methods ...... 78 5.2.1 Study site ...... 78 5.2.2 Procurement of materials ...... 80 5.2.2.1 Tree species selection and seed procurement ...... 80

Murdoch University 2011 xii

References

5.2.2.2 Root nodule bacteria ...... 82 5.2.2.3 Mycorrhizal fungi ...... 82 5.2.2.4 Fertilizer ...... 83 5.2.2.5 Water retention polymer ...... 83 5.2.3 Experimental design and treatment ...... 83 5.2.4 Site preparation and trial establishment ...... 84 5.2.5 Maintenance of the field trial ...... 87 5.2.6 Data collection and analysis ...... 88 5.3 Results ...... 89 5.3.1 Experiment 1 ...... 89 5.3.1.1 Seedling survival ...... 89 5.3.1.2 Height and diameter growth ...... 89 5.3.2 Experiment 2 ...... 91 5.3.2.1 Seedling survival ...... 91 5.3.2.2 Height and diameter growth ...... 92 5.4 Discussion ...... 95 5.4.1 Can direct seeding of high-value timber species be an option for reforestation of former agricultural lands in tropical regions? ...... 95 5.4.2 Do microorganisms enhance reforestation by direct seeding? ...... 97 5.4.3 Do inorganic fertilizer and water-retention polymers enhance reforestation by direct seeding? ...... 99 5.5 Concluding remarks ...... 101

CHAPTER 6: Selection of effective N2-fixing bacteria for Dalbergia cochinchinensis Pierre ...... 103 6.1 Introduction ...... 104 6.2 Materials and methods ...... 104 6.2.1 Environmental conditions of the study site ...... 104 6.2.2 Seed procurement and germination ...... 104

6.2.3 Isolation of N2-fixing bacteria and production of broth culture ...... 105 6.2.4 Seed coating technique and counting of bacteria on coated seeds ...... 106 6.2.5 Potting, maintenance and harvest of plants ...... 107

Murdoch University 2011 xiii

References

6.2.6 Experimental design, data collection and analysis ...... 109 6.2.7 DNA amplification, isolation and sequencing ...... 109 6.2.7.1 Polymerase chain reaction-fingerprinting ...... 109 6.2.7.2 Partial 16S rRNA sequencing ...... 110 6.3 Results ...... 111 6.3.1 Effective bacteria for Dalbergia cochinchinensis ...... 111 6.3.2 Molecular characterisation and identification of isolates ...... 113 6.4 Discussion ...... 113 6.5 Conclusions ...... 117

CHAPTER 7: Improving reforestation success by direct seeding of high-value timber species on degraded forest land: field trials in Cambodia ...... 118 7.1 Introduction ...... 119 7.2 Materials and methods ...... 120 7.2.1 Study site ...... 120 7.2.2 Procurement of materials ...... 123 7.2.2.1 Seeds ...... 123

7.2.2.2 N2-fixing bacteria ...... 123 7.2.2.3 Fertilizer and water retention polymer...... 125 7.2.3 Experimental design and treatments ...... 125 7.2.4 Site preparation and trial establishment ...... 128 7.2.5 Maintenance of the field trial ...... 129 7.2.6 Data collection and analysis ...... 130 7.3 Results ...... 130 7.3.1 Experiment 1 ...... 130 7.3.2 Experiment 2 ...... 131 7.4 Discussion ...... 134

7.4.1 Seedling survival and growth in Experiment 1: effects of N2-fixing bacteria ...... 134 7.4.2 Seedling survival and growth in Experiments 1 and 2: effect of fertilizer and polymer ...... 137 7.5 Concluding remarks ...... 140

Murdoch University 2011 xiv

References

CHAPTER 8: General discusion ...... 141 8.1 Major findings ...... 142 8.2 Issues arising from the research ...... 145 8.2.1 Transfer of knowledge and technology to Cambodia ...... 145 8.2.2 Seed procurement and supply ...... 146 8.2.3 Coping with weed problems in reforestation ...... 147 8.2.4 Inoculation with mycorrhizal fungi ...... 148 8.2.5 Enrichment planting/seeding of degraded forests ...... 149 8.3 Further research ...... 150 8.3.1 Utilization of water retention polymers for improving reforestation success ...... 151

8.3.2 Screening effective N2-fixing bacteria ...... 152 8.3.3 Understanding the mycorrhizal dependency of reforestation species ...... 153

APPENDIX I: Relevant publications ...... 155 APPENDIX II: Soil chemical and physical properties of the trial site in Western Australia 156 APPENDIX III: Effects of treatment and plant material on survival and growth of Eucalyptus gomphocephala and Acacia saligna at 40 months ...... 157 APPENDIX IV: Species information of sindora cochinchinensis and Xylia xylocarpa159 APPENDIX V: Protocols for identification of root nodule bacteria of Dalbergia cochinchinensis ...... 162 APPENDIX VI: Partial 16S rRNA nucleotide gene sequences ...... 167 APPENDIX VII: Seed germination of Dalbergia cochinchinensis ...... 169 APPENDIX VIII: Sketch map of the field trial in Cambodia ...... 170

REFERENCES ...... 171

Murdoch University 2011 xv

Chapter 1: Introduction

CHAPTER 1

INTRODUCTION

A forest resources assessment by the Food and Agriculture Organization (FAO) of the United Nations (2010a) indicates that worldwide forests cover an area of more than 4 billion ha. However, deforestation and forest degradation occurs at an alarming rate with about 13 million ha of forested area has lost every year during the period 2000 to 2010 (FAO, 2010a). During the 1990s, Asia was estimated to have an annual net loss of forested area of 800 000 ha. From 1990 to 2005, forest cover in the Southeast Asian region has declined from 245 to 203 million ha. The annual rate of deforestation was 2.8 million ha from 2000 to 2005 (FAO, 2007). Sodhi et al. (2004) projected that 75% of the original forested lands in Southeast Asia could be deforested by 2100. In Cambodia, the annual deforestation rate was 127 000 ha year-1 from 2005 to 2010 (FAO, 2010b). The forested area has declined from 13 million ha in 1990 to ca. 10 million ha or 57% of the total land area in 2010 (FAO, 2010b). The main threats to the forests in Cambodia are illegal logging, forest land encroachment, land conversion for agricultural purposes, fuelwood collection and resettlement (Strange et al., 2007). These threats to the future of Cambodian forests are expected to continue and intensify given that population growth and stakeholders are more interested in forested lands rather than the forest itself (IFSR, 2004).

In the world, many countries have been undertaking reforestation to meet two main objectives: economic development and environmental service (FAO, 2010a). In Australia, the annual deforestation rate accounted for 562 000 ha during the period 2000-2010, whereas only 55 000 ha of forest plantation has been established annually from 2005- 2010 (FAO, 2010b). Reforestation is urgently needed to counter the problems associated with deforestation and to recover the productivity of degraded lands. In Western Australia, for example, there is an urgent need of reforestation programmes to rehabilitate degraded lands left over from mining activities (Doronila and Fox, 2000), restore degraded farmlands (Piggott et al., 2002; Wade et al., 2008) and restore the structures and functions of the declining forests and woodlands (Yates and Hobbs, 1997b).

In Cambodia, forest is not only a source of economic development and environmental services, but it contributes significantly to the livelihoods of both the rural and urban

Murdoch University 2011 1

Chapter 1: Introduction population through consumption and income generation (IFSR, 2004; Kim et al., 2008). The majority of the rural people collect wood products for fuel and construction materials and non-timber forest products such as resin, rattan, edible fruits and vegetables, meat and medicinal plants for their livelihoods (IFSR, 2004; Kim et al., 2008). Deforestation and forest degradation in Cambodia not only lead to the loss of biodiversity and other services (IFSR, 2004), but diminish wood and non-wood forest products (MAFF, 2006; Kim et al., 2008) and thus threaten the livelihoods of rural communities. In 2010, the government endorsed the National Forest Programme acknowledging the urgent need of reforestation and rehabilitation of degraded forests, including establishment of multi-purpose tree plantations, as a means to biodiversity conservation, economic development and poverty alleviation (GoC, 2010).

Reforestation of degraded lands is difficult and is usually expensive, especially on sites that are difficult to reach (Poopathy et al., 2005). These programmes usually involve production and planting of nursery-raised seedlings which requires substantial labor and capital inputs (Hardwick et al., 2000). Therefore, direct seeding, a low cost method for forest restoration (Engel and Parrotta, 2001; Cole et al., 2011), can be a suitable alternative, especially where multi-species plantations are to be established for biodiversity and environmental concerns (Schmidt, 2008). Promotion of survival and early growth of direct-seeded stock is a challenge. In his review, Schmidt (2008) concludes that direct seeding is a viable method in reforestation under conditions where seed technology is applied to enhance seed germination and seedling survival, and competition from other plants or seed predation is controlled. Efforts have been undertaken to promote the early growth of direct-seeded seedlings by, for examples, improving the microsite conditions of seeding spots (Doust et al., 2006), testing seeding time (Turner et al., 2006; Doust et al., 2008) and coating seeds with microorganisms (Thrall et al., 2005). However, low survival rates, predominantly below 50%, are common in the first one to two years (e.g., Woods and Elliott 2004; Doust et al., 2006; Turner et al., 2006; Doust et al., 2008). These studies also noted that only a small number of tree species is suitable for direct seeding (mainly species with orthodox seeds). Inoculation of effective strains of N2-fixing bacteria to direct seeding of legume trees can enhance the success of forest restoration (e.g., Mandal and Nielsen, 2004; Thrall et al., 2005). However, the effects of these beneficial soil organisms on species indigenous to continental Southeast Asia, particularly high-value timber species, have

Murdoch University 2011 2

Chapter 1: Introduction never been tested. Therefore, this thesis aims to promote the early growth of direct- seeded or nursery-raised seedlings using beneficial organisms, water retention polymer and inorganic fertilizer. The main focus of the thesis is continental Southeast Asia, with a strong emphasis on Cambodia. The specific objectives of the thesis are as follows: 1) To explore the issues and limitations of reforestation by direct seeding and the effects of soil microorganisms on direct seeding (Chapter 2). 2) To investigate the effects of microorganisms on direct seeding and nursery-raised seedlings under field conditions (Chapter 3). A field trial will be established in Western Australia using two indigenous tree species to the region: Acacia saligna and Eucalyptus gomphocephala. 3) To examine the current status of high-value timber species, Afzelia xylocarpa and Dalbergia cochinchinensis, in their natural habitats, as well as opportunities for promoting reforestation that address conservation, economic development and poverty reduction in Cambodia (Chapter 4). 4) To investigate the effects of microorganisms, a water retention polymer and fertilizer on survival and growth of direct-seeded seedlings of high-value timber species indigenous to continental Southeast Asia. The field work will be undertaken in Thailand (Chapter 5).

5) To select effective N2-fixing bacteria for Dalbergia cochinchinensis (Chapter 6).

6) To investigate the effect of N2-fixing bacteria on direct seeding of Dalbergia cochinchinensis (Chapter 7). A field trial will be established in Cambodia.

The research projects involved ca. 60% laboratory and field work in Western Australia (WA) and ca. 40% fieldwork in continental Southeast Asia (Cambodia and Thailand). These two locations represent two contrasting environmental conditions, mediterranean and tropic. The mediterranean-type environment in WA is characterized by a low annual rainfall, ca. 700 mm, and low soil fertility. The two sites in continental Southeast Asia share similar environmental conditions except that the trial site in Cambodia receives more annual rainfall than that of the site in Thailand. By spreading the trial sites across different ecological regions, it is possible to generalize about the effects of the treatments across environments conditions. It is believed that the research project will be of benefit to the present needs of technologies in promoting the success in reforestation programmes. This thesis includes eight chapters, and their linkages are shown in Figure 1.1.

Murdoch University 2011 3

Chapter 1: Introduction

Improving reforestation success of high-value and key forest species by direct seeding in Southeast Asia and Western Australia

Introduction and thesis aims

(Chapter 1)

Issues and limitations of reforestation and strategies for improvement of reforestation

(Chapter 2)

Conservation and Inoculation with soil microorganism s and reforestation success by direct utilization of seeding and nursery -raised seedlings. Field trial in Australia

threatened (Chapter 3) hardwood species through reforestation―an Effects of soil microorganism s, fertilizer and a water retention polymer example of on reforestation success by direct seeding of high-value timber species. Afzelia xylocarpa Field trial in Thailand (Kruz.) Craib and (Chapter 5) Dalbergia

cochinchinensis Pierre in Effective N2-fixing bacteria for Dalbergia cochinchinensis. Glasshouse

Cambodia experiment in Australia (Chapter 4) (Chapter 6)

Delivery of N2-fixing bacteria on and off seeds. Effects of fertilizer and a water retention polymer on reforestation success by direct seeding of high-value timber species. Field trial in Cambodia (Chapter 7)

General discussion (Chapter 8)

Figure 1. 1 Diagram presenting the linkages between chapters of the thesis.

Murdoch University 2011 4

Chapter 2: Literature review

CHAPTER 2

LITERATURE REVIEW: ISSUES AND LIMITATIONS OF REFORESTATION AND STRATEGIES FOR IMPROVEMENT OF REFORESTATION

a b

Reforestation sites in Cambodia: (a), degraded land in Kampong Speu province in which top soil has been removed by erosion; and (b), a grassland in Siem Reap province, a typical reforestation site in Cambodia. The two common invasive grasses, Imperata cylindrica and Saccharum spontaneum, are present in the background. Photos courtesy of Cambodia Tree Seed Project, 2004 (a) and 2006 (b), respectively.

Murdoch University 2011 5

Chapter 2: Literature review

2.1 Reforestation

In response to worldwide deforestation and forest degradation, attempts have been made to slow down the loss of the remaining forest as well as to carry out reforestation programmes on degraded lands (Lamb, 2002). Many countries have introduced a variety of measures to reverse the process of deforestation such as controls on land clearance, logging bans and imposition of reforestation fees (Appanah, 2002). Objectives of reforestation vary depending on location and desired end products. Worldwide, forest plantations account for 264 million ha of which more than 120 million ha, the highest proportion, is located in Asia (FAO, 2010a).

In Southeast Asia, increasing efforts in reforestation of degraded forests are also being made. In this region, forest plantations increased from 10 million ha in 1990 to 12 million ha in 2005 (FAO, 2007). The plantations serve two main objectives: productive plantations, managed primarily for supplying wood and non-wood forest products, and protective plantations, managed for soil, water and biodiversity conservation. Both native (75%) and introduced tree species (25%) are being used in the reforestation programmes (FAO, 2010a). The benefits of using local tree species in reforestation programmes is that they are adapted to local environmental conditions, are well recognized by local communities who can utilize them, and contribute to the conservation of biodiversity (FAO et al., 2001). Some exotic species such as Eucalyptus camaldulensis Dehnh., Acacia auriculiformis A. Cunn. ex Benth., Acacia mangium A. Cunn. ex Benth. and Melaleuca leucadendra (L.) L. are widely planted for firewood, poles, lumber and shelter. They are also important plantation species for commercial wood produce such as pulpwood and woodchips for the export market (Midgley et al., 1996).

In Cambodia, reforestation programmes have mainly been carried out by government agencies, such as the central and local Forestry Administration, the Royal Armed Forces (IFSR, 2004) and private companies. By 2010, planted forests covered an area of about 69 000 ha (FAO, 2010b). This area is minuscule compared to the ca. six million ha of degraded forests (IFSR, 2004), that possibly need to be rehabilitated. In addition to the government tree planting programmes, the Government has organized an annual tree planting ceremony to encourage the public to participate in tree plantings (FA, 2007). A variety of tree species have been used including exotic species, such as Acacia spp., E.

Murdoch University 2011 6

Chapter 2: Literature review camaldulensis and Tectona grandis L.f.; and local species, such as Afzelia xylocarpa (Kruz.) Craib, Anisoptera costata Korth., Aquilaria crasna Pierre, Dalbergia cochinchinensis Pierre, Dipterocarpus alatus Roxb., Hopea odorata Roxb., Pinus merkusii Jungh et de Vriese and Tarrietia javanica Bl. (FA, 2007). However, in terms of planting area, the use of indigenous species in reforestation programmes is very limited (FA, 2007). The slow growth habit of some indigenous species at the early stage of development (Tsai and Faridah-Hanum, 1992; FAO, 2005) is the main limitation for promoting indigenous species in reforestation. Chapter 4 discusses in detail the use of high-value timber species, A. xylocarpa and D. cochinchinensis in reforestation in Cambodia as a means to conserve their genetic resources as well as for economic development and poverty reduction.

2.2 Constraints for reforestation programmes

Although the world has experienced an increase in forest plantations, mainly since the turn of the century (FAO, 2010a), reforestation is widely recognized as a difficult task (Setiadi, 2000; Poopathy et al., 2005). Miller and Hobbs (2007) have identified three types of major constraints to reforestation: ecological, economic and social. The ecological constraints concern what is possible in a given location and include climate, soils, landscape and biotic factors. The economic and social factors are closely inter-related and determine what is achievable in a given habitat.

Degraded sites which are available for reforestation are mainly characterized by shallow soils, low fertility and low pH (Cole et al., 1996; Poopathy et al., 2005; Macedo et al.,

2008). In addition, beneficial microorganisms, such as mycorrhizal fungi and N2-fixing bacteria, may be absent in degraded lands, or present at insufficient population levels to form effective symbiosis with roots of the planted seedlings (OTA, 1983). Degraded areas in the tropics are generally dominated by herbs and grasses. In Southeast Asia, the most common weed dominating deforested areas is a perennial grass, Imperata cylindrica (L.) P.Beauv. (Otsamo et al., 1995; Otsamo et al., 1997; Otsamo, 1998b; Friday et al., 1999). It grows quickly on both fertile and infertile soils (Bryson et al., 2010), and has a high competitive ability for soil moisture, nutrients and space (Daneshgar and Jose, 2009) leaving few resources for planted seedlings. More severely, during the dry season, the Imperata grass and other weedy species increase fire

Murdoch University 2011 7

Chapter 2: Literature review frequency risk (Pickford et al., 1992), which is the most dangerous factor affecting many reforestation programmes. In a reforestation programme in Northern Thailand, fire prevention alone consumed more than 60% of the project funds (Svasti, 2000). Domestic animals can also be problematic. Farmers in many countries in the tropics allow their domestic cattle to roam freely in degraded forests (FORRU, 2006). These animals browse on and trample the planted seedlings and cause soil compaction.

The social aspect in reforestation programmes concerns mainly the security of land tenure (Friday et al., 1999), local involvement and mitigating the causes of degradation. Firstly, it is important to have security over the proposed lands at the first step of a reforestation project regardless of who carries out the work (e.g., companies, non- governmental organizations, government agencies, communities or individual farmers). Reforestation areas with unclear land tenure can easily stimulate conflicts between local communities and other stakeholders (Siregar et al., 2007). Secondly, having secured land tenure, the next step is to seek involvement from local communities. Economic benefit is the driving force in motivating local communities to become involved in reforestation programmes (FORRU, 2006). These benefits include job opportunities, access to non-timber forest products, income generated from timber harvesting at the end of the planting cycle and the right to use part of the plantations for farming activities such as agroforestry (Nawir and Gumartini, 2002). Motivation of local people and cooperation with governmental bodies are equally as important as using the best methods in reforestation (FORRU, 2006).Thirdly, the causes of forest degradation such as grazing, wood collection and fire have to be properly dealt with before attempting a reforestation programme. Only when these factors are controlled can reforestation efforts succeed (Schmidt, 2008).

In addition to the above three constraints, reforestation of degraded areas is usually expensive, especially on sites that are difficult to reach (Poopathy et al., 2005). Reforestation programmes generally involve the production and planting of nursery- raised seedlings which require substantial labor and capital inputs (Hardwick et al., 2000). Moreover, the long-term nature of forest plantations in generating financial return makes most reforestation programmes unattractive to funding agencies compared to other land use options (OTA, 1983; Poopathy et al., 2005). However, direct and indirect costs and benefits of reforestation projects are sometimes not incorporated into

Murdoch University 2011 8

Chapter 2: Literature review the project’s plan, therefore it fails to convince funding agencies (OTA, 1983). The planning for a reforestation programme is full of complex decision making procedures. It requires the reforestation managers to undertake systematic analysis of the ecological and socio-economic conditions of the proposed planting sites, and to use appropriate reforestation technologies accordingly.

2.3 Reforestation by direct seeding

Direct seeding has been widely practiced in restoration of degraded lands in Australia (National Academy of Sciences, 1981; Schmidt, 2000; Piggott et al., 2002). It can be undertaken in many ways such as broadcasting (seeds are broadcasted by hand or by machine), drilling (using a seed drill), pitting and seeding (by hand or machine) and imprinting (by a large heavy drum) (Bainbridge, 2007). Drilling or aerial sowing is an ideal method for revegetation of degraded lands in Australia (Thrall et al., 2001) and elsewhere where the areas requiring revegetation are quite large and/or are difficult to access. Direct seeding has been used or tested in a variety of site conditions including revegetation of mining sites (Rokich et al., 2002; Turner et al., 2006), desert and dryland (Bainbridge, 2007), abandoned agricultural lands (Engel and Parrotta, 2001; Woods and Elliott, 2004), in agroforestry systems (Mandal and Nielsen, 2004) and for restoration of rainforest species (Doust et al., 2006).

Success of direct seeding is influenced by a wide range of factors, including time of seeding and management after seeding (Bainbridge, 2007). Unfavorable site conditions, such as a prolonged dry season, grass infestation, strong wind and animal predation are among further constraints leading to the failure of direct seeding (Bainbridge, 2007). Weed invasion is known to reduce the success of direct seeding over planting of nursery-raised seedlings (Schmidt, 2008). Therefore, promotion of survival and early growth of direct-seeding is a challenge. Direct seeding can be an efficient method where cheap seeds are available or for reforestation of degraded sites with difficult access (Schmidt, 2008). In his review, Schmidt (2008) concludes that direct seeding is a viable method in reforestation under conditions where seed technology is applied to enhance seed germination and seedling survival, and competition from other plants or seed predation is controlled. Woods and Elliott (2004) recommend direct seeding in areas within community forests where local communities lack resources to producing

Murdoch University 2011 9

Chapter 2: Literature review seedlings. Three reforestation trials will be established in the field (Chapters 3, 5 and 7) where direct seeding will be employed. Results from these trials will be discussed in the context of the specific trial environmental conditions.

2.4 Research priorities in reforestation

The needs of research in reforestation or restoration of degraded tropical forests in Southeast Asia have been reviewed in the past (Blakesley et al., 2002; Hardwick et al., 2004). The following discussion highlights those issues of relevance to this study.

2.4.1 Species selection

Reforestation with multiple species to meet a variety of services and products, including local community requirements (Chapter 4, Section 4.5.1), is one of the main research topics needing immediate attention (Appanah, 2002). This type of reforestation must involve planting or seeding a number of species with the ability to capture the reforestation sites quickly, yet provide forest products and service within a short period. However, a key question is “Which species are suitable for such purposes?” In tropical Asia, a concept of Framework Species, originated from northern Queensland (Goosem and Tucker, 1995), is being adapted by the Forest Restoration Research Unit (FORRU) in northern Thailand to test species suitability for rehabilitation of degraded upland areas. The method involves the random planting of 20 to 30 tree species to re-capture the site and attract seed-dispersing wildlife (FORRU, 2006). It is suggested that this plantation will stimulate biodiversity recovery within four to five years and become a self-sustaining forest. A tree considered as a framework species must possess the following ecological characteristics (FORRU, 2006): 1) ease of propagation in the nursery; 2) high survival and rapid growth rate under harsh conditions; 3) broad crown structure to shade out weeds; and 4) early production of flowers, fruits and other wildlife-attracting resources. These characteristics are tested both under nursery and field conditions. So far, FORRU have evaluated 400 species out of ca. 1100 species in the northern region of Thailand (FORRU, 2006). However, under the influence of site conditions, a species that performs well on one site may fail on another. Therefore, selection of framework species has to be carried out for each ecological region. Selection of species for the studies in Chapter 3 through to Chapter 7 was based on their ecological, socio-economic and conservation values. Species characteristics and the

Murdoch University 2011 10

Chapter 2: Literature review reasons for their selection are given in the introduction of the chapter where a species is firstly used and in Appendix IV.

2.4.2 Seed procurement and storage

The availability of quality seeds is one of the principal barriers affecting the productivity of forest plantations (Varmola and Carle, 2002). Growth performance of a tree species is greatly influenced by genetic quality inherited from the parent trees (Schmidt, 2000). To maximize or maintain the genetic diversity of a forest plantation, seeds should be collected from at least 15 unrelated parent plants of a good seed source (e.g., Schmidt, 2000). However, a number of tropical trees produce only a small amount of seed every year, and others have a intermittent seed production with intervals of several years (Schmidt, 2000). More significantly, in fragmented forest landscapes or degraded areas, finding good quality parent plants for seed collection is very difficult. In these situations, storage of seeds to ensure seed availability every year becomes an essential part of reforestation programmes. Many tropical trees produce recalcitrant seeds, e.g., the Dipterocarpaceae, and long-term storage of this type of seed remains a challenge (Blakesley et al., 2002). To overcome the shortage of seed, research into improving methods for production of quality planting stock from stem cuttings (Florentine and Westbrooke, 2004) or from artificial seeds (Section 2.5.6) should be promoted. Procurement of seed is one of the preliminary tasks in any research experiment of the subsequent Chapters. Further, detailed discussion on issues related to seed procurement of two high-value timber species in Cambodia, A. xylocarpa and D. cochinchinensis, are given in Chapters 4 and 8.

2.4.3 Direct seeding

Research on direct seeding is considered a priority by many researchers (e.g., Elliott, 2000; Florentine and Westbrooke, 2004). Previous studies show that the suitability of species for direct seeding varies significantly between species tested (e.g., Engel and Parrotta, 2001; Woods and Elliott, 2004; Doust et al., 2006; Doust et al., 2008). Only a fraction of the species that had been tested in these studies are considered suitable for direct seeding. In addition, Doust et al. (2006) found that species with large seeds produced higher survival rates than those with smaller seeds. Seedlings from large seeds rely heavily on the nutrient content in the cotyledons whereas smaller seeded species are

Murdoch University 2011 11

Chapter 2: Literature review more dependent on soil nutrients during early growth. Thus, the larger seeded species have more time to adapt to nutrient impoverished sites and can built up root biomass before tapping soil nutrient reserves (Milberg and Lamont, 1997).

Evaluating the field performance of a range of species under different management regimes should be a continuing process as performance of a species varies between sites over time. This includes determining the efficacy of seed pretreatments (Woods and Elliott, 2004), improvement of microsite conditions of the seeding spots (Doust et al., 2006), and testing of seeding time (Turner et al., 2006; Doust et al., 2008). Generally, the aim of research on direct seeding is to find methods for promoting early survival as the newly emerged seedlings have to compete with other plants, predominantly invasive species, for survival.

Research in relation to inoculation of N2-fixing bacteria with legumes is a continuing process, although the practice has been adopted for over 100 years (Stephens and Rask, 2000). However, research on the use of beneficial microorganisms in reforestation by direct seeding is in its infancy. In recent years, soil microorganisms have been used to promote survival of direct seeding (e.g., Mandal and Nielsen, 2004; Thrall et al., 2005). Thrall et al. (2005) reported their field research in Australia in which acacia seeds

(Acacia acinacea Lindl., Acacia dealbata Link., Acacia mearnsii De Wild., Acacia paradoxa DC., Acacia pycnantha Benth. and Acacia rubida A. Cunn.) were pelleted with selected strains of N2-fixing bacteria before direct seeding. The results showed that inoculated plants grew 10 to 58% faster than the uninoculated control. This study indicates that N2-fixing bacteria can play a significant role in enhancing reforestation by direct seeding when effective strains are used. Mandal and Nielsen (2004) used crushed root nodules, Rhizobium-colonized soil and standard bacterial inoculants to inoculate during direct seeding of Calliandra species, and found that bacteria-colonized soil (7 000 cells g-1 soil) gave superior nodulation at 17 weeks after seeding. Mycorrhizal fungi, a group of beneficial organisms, have not been explored fully in reforestation with direct seeding. Previous studies with nursery-raised seedlings have shown improved growth and survival of inoculated plants over non-inoculated control plants (Section 2.5.2). Selection and/or production of appropriate strains of rhizobia (Chapter 6) or mycorrhizal fungi for inoculating seeds are in need of further investigation. This thesis will explore the benefits of utilization of soil microorganisms in reforestation by

Murdoch University 2011 12

Chapter 2: Literature review direct seeding with regards to Australian acacia and eucalypt species (Chapter 3) in addition to a number of key high-value timber species of Southeast Asia (Chapters 5-7).

2.4.4 Social issues

Motivation of local communities to involve themselves in reforestation projects and sustainable collection of forest products from rehabilitated forests should also be given high priority (Elliott, 2000). Participation of local communities is an important element in ensuring the success of any reforestation project. There are many ways to motivate local communities including salary payments for being employed in the project, being given the right to collect a range of non-timber forest products, the opportunity to develop reforested areas as ecotourism sites, an awareness about environmental degradation and protection of spiritual forests (FORRU, 2006). The social aspect varies with site according to the political, economic and cultural context of a community.

In tropical regions, quite often lands that are available for reforestation are of degraded forests which are generally the places where all members of local communities have unrestricted access for grazing, collection of fuelwood and other non-timber forest products. In this situation, without proper management intervention, reforestation programmes are unlikely to be successful. Research aimed at mitigating the impact of forest plantations on livelihoods of local communities and securing forest plantations into the future is very important. Social issues are very important in reforestation programmes. However, apart from some mention in Chapter4, this area lies outside the core business of this thesis and is not discussed further.

2.5 Some strategies to enhance reforestation success

2.5.1 Application of fertilizer and water retention polymers

Application of fertilizer to promote the early growth of planted seedlings has been practiced in reforestation experiments in tropical conditions (e.g., Nichols et al., 2001; Carpenter et al., 2004; Urgiles et al., 2009) as well as in arid and semiarid environments (e.g., Al-Humaid and Moftah, 2007) for many years. Furthermore, application of fertilizer during plantation establishment for some high-value timber species, such as Chukrasia tabularis A. Juss. (Kalinganire and Pinyopusarerk, 2000) and Hopea odorata

Murdoch University 2011 13

Chapter 2: Literature review

(Noor et al., 2002), is well established. Fertilizer may improve tree growth in the first two years (Nichols et al., 2001). Although, inorganic fertilizer is effective in promoting early growth of seedlings, appropriate fertilizer use is required to avoid harmful impacts observed in some agricultural systems. For example, in Northwest China, Guo et al. (2010) found that long-term application of N fertilizer (120 kg N fertilizer ha−1 year−1 for 17 years) in winter wheat (Triticum aestivum L.) cropping lead to accumulation of nitrate in the soil (to a depth of 300 cm), posing an environmental risk on groundwater. In Chapters 3, 5 and 7, fertilizer is used as a sole treatment or in combination with microorganisms or a water retention polymer. Its effect on growth and survival of nursery-raised seedlings or direct seeding in field trials will be discussed in each chapter. Other forms of soil conditioners such as mulches, composts, and manures have frequently been shown to increase soil fertility (Lal, 2004; Chang et al., 2007; Guo et al., 2010). However, their application in reforestation programmes can be constrained due to limited availability in large quantities. They are not considered further in this thesis.

In addition to fertilizer, maintaining soil moisture availability through application of water retention polymers has been tested in reforestation experiments, especially in arid and semi-arid regions (Hüttermann et al., 1999; Pausas et al., 2004; Al-Humaid and Moftah, 2007). Water retention polymers increase the water-holding capacity (prolonging soil moisture availability) of sandy soil, increasing plant survival and growth and may reduce surface evaporation (Hüttermann et al., 1999; Al-Humaid and Moftah, 2007). In continental Southeast Asia, the long dry season (Huke, 1982) and the irregular pattern of rainfall, may have a negative effect on newly planted seedlings. Therefore, maintaining soil moisture availability to sustain plant growth in the first growing season is important. Application of water retention polymers could be an option for prolonging soil moisture into the dry season. Significant effects of polymers on plant growth and survival were reported by Hüttermann et al. (1999) and Al-Humaid and Moftah (2007). To what extent this polymer addition is sufficient for improving survival and growth of planted seedlings in a given reforestation site is still an area of ongoing research. The effect of water retention polymers on growth and survival of direct seeded seedlings in reforestation experiments in the continental Southeast Asia is discussed further in Chapters 5, 7 and 8.

Murdoch University 2011 14

Chapter 2: Literature review

2.5.2 The potential benefits of microorganisms in reforestation

The use of mycorrhizal fungi in forestry started when there was a failure in introducing exotic pine plantations in many countries of the world, and since then there has been a growing interest in the potential benefits from inoculation of mycorrhizal fungi in reforestation programmes (Jackson and Mason, 1984). Mycorrhizal fungi have been recognized as a contributor to improve biological, chemical, and physical soil quality (Cardoso and Kuyper, 2006). Plants that form symbiotic associations with mycorrhizal fungi absorb more water and nutrients, mainly phosphorus, zinc, copper, potassium, nitrogen and calcium from the soil than non-mycorrhizal plants (Dahm, 2006; Saggin- Júnior and da Silva, 2006). Apart from impacts on the survival and growth rate of inoculated plants, mycorrhizal fungi may improve soil microbial communities and their functions (Duponnois et al., 2005). Many tree species cannot be successfully established on degraded sites without the presence of appropriate fungal partners (Kendrick, 2000; Lakhanpal, 2000). Results from recent studies have highlighted the potential roles of mycorrhizal fungi in improving growth of seedlings. For example, Chen et al. (2006b) explored the growth and mycorrhization of Eucalyptus urophylla S. T. Blake inoculated with spores of Scleroderma in a nursery in south China. Mycorrhizal inoculation improved total plant biomass up to 40% in a soil-based rooting medium and more than 400% in a potting mix (vermiculite, peat and river sand, 2:1:2, v:v:v). In an investigation of the effect of phosphorus fertilizer and ectomycorrhizal fungi on biomass production of an E. urophylla plantation in south China, Xu et al. (2002) inoculated E. urophylla seedlings in the nursery with mycelial slurries of three ectomycorrhizal fungi (Pisolithus albus (Cooke & Massee) Priest, Hebeloma westraliense Boug.,Tomm. & Mal. and Laccaria lateritia Malençon). Dependent on the isolate, fungi had both positive and negative effects on tree growth in the first three years, and after four and a half years, the positive effect was reduced. This highlights the importance of matching fungi with site type as well as the host.

Another beneficial plant-soil interaction is the symbiosis between a host-plant with N2- fixing bacteria. Members of the Leguminosae form symbiotic associations with N2- fixing bacteria which are economically and environmentally important in the agricultural sector (Freire and de Sá, 2006). In addition to association with bacteria, many N2-fixing trees form a symbiotic association with mycorrhizae, and both symbionts generate

Murdoch University 2011 15

Chapter 2: Literature review synergistic benefits to the host (Bainbridge, 2007). Although nitrogen is abundant in the atmosphere, only a very limited amount of nitrogen is available for plant use in many soils (Gentili and Jumpponen, 2006; Raman and Selvaraj, 2006). Moreover, on degraded sites, many legumes depend on effective nodulation and nitrogen fixation to ensure successful establishment, and thus seed inoculation with effective strains of rhizobia is very helpful (Bainbridge, 2007). Seed inoculation is an efficient and easy method of introducing rhizobia to the rhizosphere. Inoculation of rhizobia to legumes can be performed by coating on seeds or applying directly to the nursery or field soils in the forms of liquids or granules (Deaker et al., 2004).

Dual inoculation of N2-fixing bacteria and mycorrhizal fungi with a legume host may be of great success in enhancing plant growth if the right combination of organisms can be integrated (Gentili and Jumpponen, 2006), and this approach has increasingly captured the interests of researchers (Raman and Selvaraj, 2006). For example, in a nursery experiment in China, Cao et al. (1994) inoculated seedlings of Leucaena leucocephala

(Lam.) De Wit with AM fungi and N2-fixing bacteria. They discovered that dual inoculation was superior over single inoculation, and substantially improved shoot dry weight by 144%. Likewise, Zhong et al. (1994) demonstrated a great improvement in the growth of two species of Casuarina spp. after dual inoculation with Pisolithus and Frankia. However, information about the right combination of Rhizobium with mycorrhizal fungi for a host plant is very limited (Raman and Selvaraj, 2006). Appropriate microorganisms could help outplanted seedlings to establish and overcome harsh conditions of reforestation sites. While rhizobial inocula can be easily mass produced, mass production of mycorrhizal inocula, especially AM fungi, with the present technology, is difficult due to its plant-dependence characteristic. However, mycorrhizal fungi can be delivered to seedlings at the nursery level.

2.5.3 Principles and techniques for selection of microorganisms for reforestation

2.5.3.1 Methods for screening effective mycorrhizal fungi

Many species of indigenous mycorrhizal fungi are freely available in most natural soils with varying capacity for colonizing plant roots and promoting growth (Dodd and Thomson, 1994). However, host fungus specificity exists for many mycorrhizal fungi (e.g., Lesueur et al., 2001, Wu et al., 2002 and Cavallazzi et al., 2007). This implies the

Murdoch University 2011 16

Chapter 2: Literature review need to select appropriate fungal isolates for a particular plant species before embarking on inoculation of mycorrhizal fungi for seed or seedlings. Screening for effective isolates of AM and ECM fungi can comprise a series of experiments including: 1) collection of soil samples; 2) assessment of mycorrhizal potential (bioassay); 3) trapping culture; 4) mono-species culture; and 5) screening for the most effective isolates in a glasshouse or in the field. The step by step methodologies of the screening process are summarized below.

Collection of soil samples: Soil samples are generally collected from the rhizosphere of the studied plants (e.g., Lesueur et al., 2001, Cavallazzi et al., 2007) or from the sites of interest (e.g., Wu et al., 2002). Lesueur et al. (2001) collected soil samples from 0 to 20 cm depth, within 50 cm from the base of Calliandra calothyrsus Meissn. and with one sub-sample from each of the four directions. If samples have to be collected from many sites, sampling within a site is needed and the samples of each site can then be bulked.

Bioassay: The status of mycorrhizal fungi in a site for a specific plant can be evaluated by field observation (for ECM) (e.g., Chen et al., 2007) and collection of soil samples for bioassay experiments (for ECM and AM) (e.g., Bouamri et al., 2006, Chen et al., 2007) or by examining mycorrhizal infection on the roots of seedlings of the species of interest collected from natural habitats (Tawaraya et al., 2003). The last method can only determine infective species; it may not reflect the abundance and richness of mycorrhizal fungi in the area. Also, field observations on fungal sporocarps often underestimate the number of ECM fungal species due to the fact that not all fungi may fruit in any one year.

In bioassay experiments, soil samples collected from different habitats can be used to evaluate the inoculum potential of mycorrhizal fungi in soil (Brundrett et al., 1996). Results from the bioassay experiment indicate whether the studied site has sufficient inoculum of a compatible fungal species (Brundrett et al., 1996). Based on this finding, a decision on the need to introduce effective mycorrhizal fungi to the site can then be made (Brundrett et al., 1996; Chen et al., 2007). Cavallazzi et al. (2007) performed a bioassay experiment by homogenizing soil samples and then mixing each sample with sterilized sand (1:1, v:v), and the mixture was placed into pots. One pre-germinated corn seed was added to each pot. After 40 days, corn roots were assessed for AM infection.

Murdoch University 2011 17

Chapter 2: Literature review

Soil samples having the greatest mycorrhizal infection were selected for assessing species diversity in trap culture and isolating fungi in pure culture.

Sometimes, soil samples are collected intact using a soil corer. Bait plants are then planted in the intact soil to measure mycorrhizal inoculum potential of the soil (Brundrett et al., 1996). The core soil bioassay is superior to the method involving a serial dilution of soil samples with a sterilized substrate because it measures the mycorrhizal infection from intact hyphal networks as well as spores (Brundrett et al., 1996).

Trap culture: Trap cultures and bioassays are set up in a similar way, but their objectives and end products are different. Trap cultures are often used to obtain fresh and healthy AM spores for identification and determining species richness and diversity of soil samples (Lesueur et al., 2001), and to obtain spore inoculum to establish pure cultures (INVAM, http://invam.caf.wvu.edu/methods/cultures/trapcultures.htm). In the bioassay approach, the inoculum potential of a soil sample, including viable and infective propagules such as colonized roots, hyphal fragments and spores, is the main concern. Spores alone, however, may not be a good indicator of inoculum potential (Abbott and Robson, 1982). For example, spores of many Glomus species are inferior to hyphae or colonized roots in terms of root infection (INVAM, http://invam.caf. wvu.edu/otherinfo/articles/propagules.htm). In some ecosystems, such as arid areas, spores represent only a fraction of the infective propagules (Requena et al., 1996).

Bouamri et al. (2006) set up a trap culture by mixing each soil sample with sterilized sand (1: 1, v:v), whereas Lesueur et al. (2001) placing the soil samples as a layer in the cultured pot on top of a substrate comprising loam/sand/granules of compressed clay– Terragreen (1:1:1, v:v:v), and covering it with sterile substrate. Pots were then sown with seed of a suitable species such as sorghum or corn. The pots were maintained in a glasshouse for four to six months before soil cores were taken for assessment of spore populations (Lesueur et al., 2001; Bouamri et al., 2006). Sometimes, the trap culture is repeated for two consecutive cycles as repeat trapping may discover new AM species that were dormant during the first cycle (Lesueur et al., 2001). Bouamri et al. (2006) reported that two species of the genus Scutellospora were discovered during the second trapping event.

Murdoch University 2011 18

Chapter 2: Literature review

A variety of agricultural plants have been used as bait plants in bioassay and trap culture experiments for AM fungi. These include sorghum (Lesueur et al., 2001; Wu et al., 2002; Bouamri et al., 2006; Cavallazzi et al., 2007), corn (Cavallazzi et al., 2007) and clover (Brundrett and Abbott, 1995; Brundrett et al., 1996). These bait plants are selected based on two criteria: rapid root growth and ease in clearing roots to reveal internal structures (Brundrett et al., 1996). Sometimes, the plants to be inoculated in the screening experiment are also used (Lesueur et al., 2001). In Australia, Eucalyptus species are often used to assay for ECM fungi (Brundrett et al., 1996).

Pure cultures: In the case of AM fungi, pure cultures can be produced from viable spores freshly harvested from trap cultures by placing spores (with the same morphological characteristics) in contact with the root system of a host plant in pots filled with a sterilized potting mix, such as a mixture of sand and red-yellow podsol (2:1, v:v) (Cavallazzi et al., 2007). In the latter study, pots were maintained in a glasshouse with strict hygiene to prevent contamination by other fungal spores, and checked for spore purity after four months (Cavallazzi et al., 2007). Pure cultures were repeated in another cycle using larger pots and a mixture of soil inoculum from each culture with fresh sterilized substrate of sand and red-yellow podosol (1:1, v:v), and the pots were reseeded with the same hosts. After another four months, the individual soil inoculum was chopped, well mixed, and then stored at 4oC for studying the efficiency of AM isolates (Cavallazzi et al., 2007).

Even though an attempt is made from the beginning to select only the spores with similar morphological characteristics, it may not be possible to obtain pure cultures in the first cycle. Therefore, pot cultures are repeated many cycles until pure isolates are obtained (Calvente et al., 2004). To avoid this repetition, a pure pot culture is sometimes started by placing one healthy and fresh spore on the germinating root of a host plant grown in a sterile potting mix (Jansa et al., 2002). The plants are then grown in a glasshouse for a few months before harvesting the spores. Pure cultures from single spores, however, frequently encounter spore dormancy problems and thus can result in limited success (Gemma and Koske, 1988). INVAM (http://invam.caf.wvu.edu/otherinfo /articles/propagules.htm) reported that the single spore culture of many Glomus species produces a success rate of only 10 to 50%.

Murdoch University 2011 19

Chapter 2: Literature review

In contrast to AM, many ECM fungi can easily be isolated from the field, especially from fungal fruiting bodies (Dodd and Thomson, 1994). Isolation can be made using vegetative tissue of fruiting bodies, spores and mycorrhizal roots (Brundrett et al., 1996). Isolation from fruiting bodies is preferred because it is easy for many fungal genera and the species can be readily identified (Dodd and Thomson, 1994). Mycelia of many ECM fungi can be grown on a nutrient-enriched agar medium without the presence of a host plant (Dodd and Thomson, 1994; Brundrett et al., 1996). In addition, ECM isolates can be procured from existing culture collections, sub-cultured, and passaged through a host in vitro to regain vigour (Brundrett et al., 1996).

Screening of isolates: Methods of screening for effective mycorrhizal fungi were comprehensively discussed by Dodd and Thomson (1994). Many publications address selection of suitable strains for the host species of interest (e.g., Burgess et al., 1993; Thomson et al., 1994; Lesueur et al., 2001; Wu et al., 2002; Cavallazzi et al., 2007).

The efficiency of AM isolates for a particular tree/plant can be investigated by inoculating spores or soil containing spores and root fragments to the host plant. As an example, Gai et al. (2006) screened 14 isolates of AM fungi (five species) in a pot experiment in a glasshouse using autoclaved field soil and sand (3:1, w:w) and sweet potato (Ipomoea batatas (L.) Lam.) as the host plant. Each pot was inoculated with 200 spores of each isolate at the time of transplanting the host. After four months, results revealed that only two isolates, Glomus etunicatum W.N. Becker & Gerd. and Glomus mosseae (Nicolson & Gerdemann) Gerd. & Trappe, were effective for sweet potato.

Due to the influence of biotic and abiotic factors, the best isolates selected from a pot culture in the glasshouse may not maintain their effectiveness under field conditions. Therefore, screening experiments in the field are needed before any conclusion on the effectiveness of a species can be made. This was shown by the work of Wu et al. (2002), as out of five effective strains, selected in the glasshouse, only two strains were proved effective after testing in the field. Competition from other indigenous AM fungi for nutrients was one of the factors that altered this result (Wu et al., 2002).

The assessment of compatibility of ECM fungi with host plants can be conducted in- vitro using synthesis in sterile culture or in vivo in a glasshouse or nursery using

Murdoch University 2011 20

Chapter 2: Literature review pasteurized potting media (Brundrett et al., 1996). In an in vitro synthesis experiment, host-fungus compatibility can be assessed by placing the host plant’s roots in contact with mycelium of ECM fungi under favourable conditions. This method is frequently used because it is effective and facilitates the rapid screening of ECM fungi (Brundrett et al., 1996). However, it has some disadvantages as some fungi which are usually well- matched with a host plant in natural conditions cannot be produced in sterile culture or may even reduce their compatibility (Brundrett et al., 1996). Therefore, experiments using pasteurized soil or a potting medium are needed. As an example, Burgess et al. (1993) screened 16 ECM isolates in a glasshouse using steam-sterilized sand placed in non- draining plastic pots. Seeds of Eucalyptus globulus Labill. and Eucalyptus diversicolor F. Muell. were surface sterilized and germinated aseptically. Seedlings were inoculated with one of the fungal isolates, and then four inoculated seedlings were transplanted into each pot. The pots were maintained about three months until the harvest of seedlings for mycorrhizal assessment (Burgess et al., 1993). The result from this experiment is limited to the steam-sterilized sand used as rooting media; therefore, it may not be applicable in general nursery practices in which a variety of potting media are used.

Chen et al. (2006b) inoculated E. urophylla seedlings with spores of Scleroderma spp. in a nursery using four non-pasteurized field soils and a potting mix (vermiculite, peat and river sand: 2:1:2, v:v:v) as rooting media. The inoculated Scleroderma spp. were not only able to form ectomycorrhizas in non-pasteurized media, but they outperformed the indigenous species in all rooting media (with the highest mycorrhizal colonization found in the potting mix). Similarly, Brundrett et al. (2005) conducted large-scale ECM inoculation on Eucalyptus seedlings in forest tree nurseries in Australia and China using 90 isolates of 23 fungal genera, mainly obtained from forests or eucalypt plantations in Australia. In general, the success rate of fungal inoculation was 38%. This rate, however, varied between genera with agaricoid genera (Descolea and Laccaria) and hypogeous fungi being superior over the sequestrate genera (Pisolithus and Scleroderma). Again, these findings indicate the need to screen for effective ECM fungi in a non-aseptic environment, nursery or field sites, in order to select the right fungi before embarking on a large-scale inoculation programme. Inoculation of ECM fungi could be efficiently carried out in commercial nurseries with different management practices (Brundrett et al., 2005).

Murdoch University 2011 21

Chapter 2: Literature review

2.5.3.2 Methods for screening effective rhizobial bacteria for legumes

A number of rhizobial bacteria may infect many legume plants and vice-versa (Frioni et al., 1998; Yates et al., 2004; Menna et al., 2006). However, growth performance of the host plant is often improved by a specific strain(s) of rhizobia (Frioni et al., 1998). In addition, due to the variation of local edaphic and climatic conditions such as soil fertility, acidity and altitude, the native rhizobia may formulate a specific adaptation capability in their response to local site conditions (Melchor-Marrroquín et al., 1999). Therefore, selection of effective strains for a legume is important for a reforestation site. In principle, methods for screening of effective rhizobia are the same as for screening of mycorrhizal fungi (Section 2.5.3.1), and details can be found in the Handbook for Rhizobia by Somasegaran and Hoben (1994). The following section describes the general procedures and is based mainly on the research work of Lesueur et al. (2001).

Collection of root nodule samples: Samples of root nodules are normally collected from a wide range of geographical regions where the host plants grow naturally or from forest plantations where the legumes have been introduced successfully. Lesueur et al. (2001) collected 120 healthy nodules from 8 to 12 Calliandra calothyrsus trees in a population. The nodule samples from each population were bulked, placed in paper bags, and kept in a room with a temperature of 15 to 20°C before using in trap culture.

Trap culture: Trapping experiments in the glasshouse can be used to help isolate rhizobia. The following method for trapping N2-fixing bacteria is taken from Lesueur et al. (2001). Seedlings of the species of interest were pre-germinated and transplanted into sterile sand-filled plastic bags. Each bag received a dose of crushed nodule solution which was applied at the base of the plants. Treatments were separated by provenances and replicated five times. The pots were maintained for two months in a glasshouse, and then the same number of fresh nodules from each provenance was harvested for the isolation of rhizobia. Root nodules were surface sterilized by immersion in 95% ethanol for 30 s. Then, after cleaning, they were acidified in mercuric chloride (HgCl2, 0.1%) for 3 min. Each time after surface sterilization, nodules were cleaned several times with sterile distilled water. Nodules were crushed on a sterilized glass slide, and drops of crushed nodules were transferred onto yeast extracted based medium (YEM) agar plates. The isolates were kept in YEM at -80°C for later use.

Murdoch University 2011 22

Chapter 2: Literature review

Screening of rhizobia: Effective strain of N2-fixing bacteria for a specific legume in a specific soil condition has to be selected through screening experiments (Howieson et al., 1988; Lesueur et al., 2001). The techniques for selection of an effective strain of root nodule bacteria for a host plant(s) require a combination of glasshouse and field experiments (Howieson et al., 2000). An example of the work of Lesueur et al. (2001) describes the procedure: pre-germinated seedlings of Calliandra calothyrsus were transplanted into sand-filled plastic bags. The seedlings were then inoculated with strains obtained from the trapping culture. After maintaining in an axenic glasshouse for three months, seedlings were transplanted into a sterilized field soil, and were maintained for six months before harvesting. The most efficient strains were measured in terms of nodulation, growth and shoot nitrogen content (Lesueur et al., 2001). Screening in the glasshouse is useful for a large number of strains (Melchor-Marrroquín et al., 1999); however, the most promising strains may perform differently under local environmental conditions (Section 2.5.3.1). Methods for selection of effective N2-fixing bacteria for D. cochinchinensis are discussed in Chapter 6.

Nitrogen is often the nutrient that is most limiting for plant growth. Not surprisingly therefore, effective symbiotic associations between legumes and N2-fixing bacteria underpin the sustainability of agricultural (Menna et al., 2006) and industrial crop production. Therefore, selection and maintenance of effective strains, ability to form root nodules, efficiency in nitrogen fixing and resilience to soil toxicities, for important legumes are of great significance.

2.5.4 Seed coating/pelleting

Seed coating/pelleting refers to the covering of seed with a coating material like calcium alginate and lime. Other agents may be added including various germination- and early- survival promoting substances such as fertilizer and microorganisms (Schmidt, 2000; Turner et al., 2006; Schmidt, 2008). The practice is not economically feasible when applied in the nursery where fertilizer and microorganisms can easily be added to the nursery container. However, the coating technique is of importance for reforestation by direct seeding if coating enhances seed germination and survival (Schmidt, 2000). The methods for seed coating are relatively simple and have been comprehensively

Murdoch University 2011 23

Chapter 2: Literature review described (Brockwell,1962, Pijnenborg et al., 1991; Thrall et al., 2001; Bainbridge, 2007). In recent years, there has been interest in using seeds coated with various plant growth promoting substances to improve reforestation by direct seeding (e.g., Thrall et al., 2005; Turner et al., 2006). Results indicate that seed coating has the potential to improving reforestation.

2.5.4.1 Seed coating material and method

Brockwell (1962) classified material used for seed coating into three components: 1) adhesive such as gum arabic or methyl ethyl cellulose; 2) powder/carrier such as dolomite, organic nutrient, lime or charcoal; and 3) inoculant such as Rhizobium. Seed coating is performed by wetting seeds with adhesive and then rolling in a carrier for coating. Peat-based Rhizobium inoculum is then added to the pelleted seeds externally or by mixing with the pellet (Brockwell, 1962). In a similar way, Pijnenborg et al. (1991) first mixed an adhesive, methyl cellulose, and peat-based Rhizobium inoculum with seed (Medicago sativa L.) followed by CaCO3 as a protective layer. Adding

CaCO3 to the outer layer is very beneficial as it protects the microorganisms and the seeds by preserving moisture when sowing in a dry condition and reduces soil acidity when the seeds are sown in a highly acidic soil (Somasegaran and Hoben, 1994; Schmidt, 2000).

Among the carriers that have been used, peat is widely accepted as the best inoculant carrier for legumes (Somasegaran and Hoben, 1994; Thrall et al., 2001; Temprano et al., 2002; Kubota et al., 2008). Peat has been used as inoculant carrier for legumes for almost a century (Brockwell and Bottomley, 1995). However, since peat is not available everywhere, efforts have been made to search for alternative carriers with similar quality in enhancing the survival of inoculants while coated on seeds. For example, Davidson and Reuszer (1978), tested 12 coating materials for their effectiveness in promoting the survival of Rhizobium japonicum (Kirchner) Buchanan on soybean seed, under controlled temperature and humidity. They concluded that a mixture of charcoal with an adhesive was the most suitable carrier. Chao and Allexander (1984) found that rhizobia had a higher survival on seeds coated with mineral soil (Lima silt loam) than peat. In a study on survival of rhizobia on different inoculant formulations, Temprano et al. (2002) found that perlite can be a substitute for peat in retaining the survival of

Murdoch University 2011 24

Chapter 2: Literature review rhizobia during a six-month storage period. In the private sector, searching for the best carrier has been ongoing for over 30 years. Asano (1996) patented a seed coating material comprising 70 to 95 parts by weight of a clay mineral and 5 to 30 parts by weight of a hydrophobic compound. Dannelly (1981) used a water insoluble microgel to give protection to seed and as a carrier for fertilizers, herbicides and pesticides. This coating material do not dissolve immediately when placed in contact with water, but the outer layer swells and falls away, layer by layer, until the coating is removed. These examples have indicated that there are a range of materials that can be potential carriers for seed coating. In Cambodia, peat is a scarce resource, but other similar materials such as dusted coconut husk, charcoal as well as materials originated from soil such as termite mounds and clay soil are available locally.

In addition to the type of carrier, the characteristics of the carrier play an influential role on seed coating success. Carriers with large particles may not stick to the seed or become fragmented during handling. Temprano et al. (2002) and Brockwell (1962) suggested that a good carrier should pass through a 200 μm mesh sieve. The total weight of carrier on seed should constitute about 50 to 90% of the seed weight

(Brockwell, 1962; Asano, 1996). A carrier should have a pH within a range from 6.5 to 7.5 (Strijdom and van Rensburg, 1981). The advantages and disadvantages of seed coating for use in reforestation programmes are highlighted in Table 2.1.

Along with the carrier, the adhesive is also a subject of research (Temprano et al., 2002; Bardin and Huang, 2003). A range of materials have been tested such as polyvinyl alcohol, methyl cellulose, skim milk, sugar, vegetable and paraffin oils. However, until the present the best and widely used adhesive is gum arabic since it is most effective in enhancing the survival of rhizobia on seeds (Schmidt, 2000; Temprano et al., 2002; Deaker et al., 2004). In order to ensure high survival rate of microorganisms, coated seeds are recommended to be sown as soon as possible (e.g., Thrall et al., 2001). However, Bardin and Huang (2003) reported that the number of bacteria, like Erwinia rhapontici (Millard) Burkholder on coated seeds of sugar beet, remained constant during the first two months of storage at 5ºC. Methods for coating seeds of D. cochinchinensis with N2-fixing bacteria are described in Chapter 6. The effects of seed coating as a means to deliver N2-fixing bacteria to seeds are further discussed in Chapters 6 and 7.

Murdoch University 2011 25

Chapter 2: Literature review

Table 2.1 Advantages and disadvantages of seed coating technology in reforestation programmes

Description Reference

Advantages of seed coating 1. Coating produces a uniform seed size which facilitates handling, Schmidt, 2000 e.g., machine sowing, especially for small seeds. 2. Coating protects seeds from a variety of hazards and it is very Schmidt, 2000 useful for reforestation programmes by direct seeding since seeds used in this practice require some form of protection. 3. Various germination- and early-growth promoting Schmidt, 2000 substances/organisms such as fertilizer, spores of mycorrhizal fungi and rhizobia can be introduced to seeds during the process of seed coating. 4. Coated seeds can be stored for some time before sowing. Brockwell, 1962; Schmidt, 2000; Bardin and Huang, 2003 5. The planting season is extended because seeds can be planted Turner et al., 2006 earlier in the growing season. 6. Seed coating allow the handling of fertilizer and beneficial Thrall et al., 2005; Turner et microorganisms on the seed coat which are readily available at al., 2006 the early stage of seedling establishment. 7. Seed coating reduces post-dispersal seed predation. Johns and Greenup, 1976

Disadvantages of seed coating 1. The coated seeds exclude further pretreatment of seed. Schmidt, 2000 2. The process of seed coating has to be undertaken shortly before Schmidt, 2000 sowing because the coated seeds cannot be stored for a long. 3. Coatingtime. of seeds may delay or reduce germination of some Schmidt, 2000, 2008; Turner species and therefore reduce survival. et al., 2006; Bainbridge, 2007 4. Only a small amount of seed coating materials can be added to Schmidt, 2008 seeds (depending on seed size). 5. Seed coating incurs additional cost to the reforestation Turner et al., 2006 programme.

2.5.5 Biochar―an emerging technology for stimulating plant growth

Biochar is a fine-grained charcoal high in organic carbon produced from the slow pyrolysis of biomass (Chan et al., 2007). Under natural conditions, biochar is accumulated in soils in the form of incomplete combustion of biomass resulted from

Murdoch University 2011 26

Chapter 2: Literature review forest fires (Wardle et al., 1998; Carvalho et al., 2006; Kurth et al., 2006). Charcoal is a very stable material in natural environments, it can stay in the soil for decades or centuries (Carvalho et al., 2006). The application of biochar to soil is considered as a new approach for long-term reduction of carbon dioxide from the atmosphere (Lehmann et al., 2006). In recent years, the potential of fire-deposited charcoal in improving both chemical and biological properties of forest soils has been increasingly recognized (Kurth et al., 2006).

2.5.5.1 Effects of biochar on plant growth

As a soil amendment, the addition of biochar to soil helps to improve soil chemical and physical quality by reducing soil acidity (Glaser et al., 2002; Chan et al., 2007; Rondon et al., 2007; Steiner et al., 2007), improving water holding capacity (Glaser et al., 2002), and improving nutrient retention and nutrient availability, all of which can lead to increased crop yield and productivity (Glaser et al., 2002; Chan et al., 2007; Rondon et al., 2007; Steiner et al., 2007). Biochar is considered as a soil conditioner rather than a fertilizer (Steiner et al., 2007). In a pot experiment, Chan et al. (2007) discovered that adding biochar alone to soil did not increase yield of radish; however, it promoted efficient use of N fertilizer in plants when biochar was coupled with N fertilizer. In this experiment, a negative effect of biochar (at one rate) on plant growth was observed, but the reason(s) behind this remains unclear. In a field experiment with rice (Oryza sativa L.) and sorghum (Sorghum bicolor L.), Steiner et al. (2007) found that application of charcoal sustained soil fertility over four harvesting seasons; and they recommended that a mixture of charcoal and chicken manure might produce a best quality soil.

Based on a literature survey, Glaser et al. (2002) assumes that charcoal enhances nutrient availability for plants through two mechanisms: 1) nutrients are captured in the large number of continuous fine pores; and 2) the cation-exchange capacity on the edges of the charcoal is enhanced as a result of slow biological oxidation. Regarding water retention, Glaser et al. (2002) concluded that only sandy soil had higher available moisture after application of biochar, no change was observed in loamy soil, and clay soil even showed decreased soil moisture with increasing charcoal application.

Murdoch University 2011 27

Chapter 2: Literature review

2.5.5.2 Effects of biochar on microorganisms

The significance of charcoal in stimulating the development of AM fungi is relatively well-known in Japan (Nishio, 1996). Many studies have established that biochar stimulates beneficial microorganisms and thus promotes plant growth. Ishii and Kadoya (1994) investigated the effects of several kinds of charcoal on growth performance of citrus and development of AM. Total plant biomass was improved in response to charcoal application, and all charcoal treatments stimulated higher intensity of AM infection as compared to the control. However, applications of biochar do not always benefit mycorrhizal fungi. For example, Rondon et al. (2007) showed that nitrogen fixation by common bean (Phaseolus vulgaris L.) was significantly improved by moderate rates of biochar application, but there was no beneficial effect on root colonization by AM fungi.

The role of charcoal in stimulating soil microorganisms has been attributed to its physico-chemical properties (Saito and Marumoto, 2002).Warnock et al. (2007) believed that biochar affects the quantity and/or activity of mycorrhizas in the soil in four ways: 1) it alters nutrient availability that affects both plants and mycorrhizal fungi; 2) it brings about positive and/or negative effects to other soil microorganisms; 3) it alters plant-mycorrhizal fungal signaling processes or detoxifies allelochemicals; and 4) it provides a refuge from mycorrhizal grazers. Charcoal has a large number of pores which serve as a safe microhabitat for the spores that acquire their nutrients through mycelia extended from roots. However, little is known about which mechanism is the most important for a given site condition (Warnock et al., 2007).

Charcoal is a tool for stimulating plant growth (e.g., Steiner et al., 2007) while at the same time increasing carbon sequestration in soils (Lehmann et al., 2006). However, the present understanding of biochar as a soil amendment highlights the need for further investigation. Warnock et al. (2007) proposed a number of research priorities, one of which is to look at the negative effects of charcoal. Negative or neutral effects of charcoal application may possibly be under-reported. The environmental conditions and biochar properties that lead to such effects, as discussed above, have to be identified. Some of the recommendations from Glaser et al. (2002) include the need to understand chemical and physical properties of charcoal surfaces, and the need to evaluate the agronomic effectiveness and the economic viability of charcoal as a soil improver. Although biochar

Murdoch University 2011 28

Chapter 2: Literature review may be beneficial for reforestation programmes, it is not the main interest of this study. However, site preparation using fire can result in scattered patches of charcoal (Chapter 7), and the effect of this charcoal on seedling survival is discussed in that Chapter.

2.5.6 Artificial seeds

Artificial seed technology is a practice of producing tissue culture to obtain somatic embryos and encapsulate them in a matrix, such as an alginate hydrogel, enclosed in an artificial seed coat (Saiprasad, 2001). Production of artificial seeds from vegetative parts, such as shoots or nodal shoot segments (Sarkar and Naik, 1998; Danso and Ford- Lloyd, 2003; Chand and Singh, 2004), is easier than production from somatic embryos. The production of artificial seeds is useful for species that have difficulties in propagation by seed such as Pogonatherum paniceum (Lam.) Hack. that produces low nutrient seeds which cannot support seedling growth (Wang et al., 2007) or Saintpaulia ionantha Wendl. that produces non-viable seeds (Daud et al., 2008). In addition, the technology also facilitates the exchange of plant material and conservation of germplasm of endangered species or species with high economic value using appropriate storage techniques (Danso and Ford-Lloyd, 2003; Rai et al., 2008). In forestry, artificial seed technology could be a useful tool for production of seeds with storage difficulty such as dipterocarps or species with a long-term mast fruiting. This technology, however, has not been fully developed. With the current technologies, artificial seed technology presents advantages and disadvantage as summarized in Table 2.2. The discussion on artificial seeds in this literature review highlights a new way of producing seedlings of plants where the number of viable seeds is low or where recalcitrant seeds are not suitable for long-term storage. However, artificial seed is not a main concern of this thesis and will not be discussed further.

2.6 Concluding remarks

Reforestation is one of the key methods to counter the process of deforestation. Reforestation of degraded areas is a difficult task and usually expensive, as it is associated with many challenges: ecological, economic and social. Direct seeding is a low budget method of reforestation and its application is especially suitable for multi-

Murdoch University 2011 29

Chapter 2: Literature review

Table 2.2 Advantages and disadvantages of artificial seeds in reforestation programmes

Description Reference Advantages of artificial seeds 1. Seeds can be stored at 4oC for up to six months. Saiprasad, 2001 2. Seed is free of pathogens since it is produced from tissue culture. Saiprasad, 2001 3. Some useful additions such as microorganisms can be included into the Saiprasad, 2001 4. Superiorencapsulation and genetic matrix uniformfor improving seeds canvigor be of produced the somatic in large embryos. quantit ies at Maruyama et al., 1997; low costs. Panaia, 2006 5. Reduction of time in a tree breeding program since desirable gene Maruyama et al., 1997; sequences can be inserted into somatic cells. Saiprasad, 2001 6. In economic terms, the production of artificial seeds is 10 times more Panaia, 2006 efficient than tissue culture. 7. Artificial seed is a promising technology for production of transgenic Saiprasad, 2001 plants, seedless plants, polyploids with elite traits and plants with problems in seed production/propagation. 8. Implications of this technology are: 1) guaranteed and quality supply of Panaia, 2006 planting material for those species that are in high demand; and 2) reduced loss of biodiversity and increased efficiency of reforestation programmes.

Disadvantages of artificial seeds

1. Production of artificial seeds requires intensive labor, and the production Brischia et al., 2002 cost is high. 2. At present, there is a lack of synchrony of somatic embryos for the Saiprasad, 2001 production of synthetic seeds. 3. The production of artificial seed requires the development of specific Panaia, 2006 protocols for each species of interest. 4. With present technology, germination and growth of plantlets is often not Panaia, 2006 as vigorous as that of normal seed. This phenomenon is recommended as a priority research in the future.

species restoration for environmental and biodiversity conservation purposes. However, low survival rates indicate that the technology needs to be further developed before it can play a significant role in reforestation programmes. It is expected that seed technologies such as seed coating could lead to new opportunities for the improvement of direct seeding. In addition, if suitable beneficial microorganisms could be easily applied through seed coating, significant improvement in seedling survival should be achievable.

Murdoch University 2011 30

Chapter 3: Reforestation of a degraded eucalyptus woodland

CHAPTER 3

REFORESTATION OF A DEGRADED MEDITERRANEAN-TYPE EUCALYPTUS GOMPHOCEPHALA A. DC. WOODLAND IN WESTERN AUSTRALIA

a b

b

(a), Seedlings of Eucalyptus gomphocephala after three months established by direct seeding in a woodland adjacent to Yalgorup National Park, Western Australia. (b), A Scleroderma sp. fruiting body at the trial site.

Murdoch University 2011 31

Chapter 3: Reforestation of a degraded eucalyptus woodland

3.1 Introduction

In the southwest of Western Australia (WA), forests and woodlands of Eucalyptus gomphocephala A. DC. (tuart) have been reduced from more than 111 609 ha during pre-European settlement (Hopkins et al., 1996) to only 30 317 ha (Government of Western Australia, 2003). In addition, the species has declined in health and vigor in recent decades. The decline is severe in Yalgorup National Park and surrounding regions (Government of Western Australia, 2003). Research to diagnose the cause of this decline has been undertaken (Barber and Hardy, 2006; Cai et al., 2010), and one management option in reversing the decline of eucalypt woodlands is to restore the structure and function of the remnant forests through forest restoration (Yates and Hobbs, 1997a).

Two key colonizing species of the declining E. gomphocephala woodland, E. gomphocephala and Acacia saligna (Labill.) H. L. Wendl., were selected for this study because they represent two types of symbiotic relationships, i.e. non-N2-fixing and N2- fixing; ectomycorrhizal and endomycorrhizal woody plants, respectively. They are also important species in the revegetation of tuart woodlands on the Swan Coastal Plain. Tuart is an endemic species that occurs naturally on sandy soils of the Swan Coastal Plain of southwestern Australia (Rippey and Rowland, 2004). Tuart grows up to 42 m tall and the wood has been used for fuel, flooring blocks, framing, particle board, fibreboard and small posts, but is no longer harvested from the wild. The species is considered an important candidate for planting in adverse sites such as highly calcareous soils and well-drained slightly saline sites (Turnbull and Booth, 2002). Although slow growing, it has performed well in plantations in semi-arid regions of North Africa, and it is a plantation tree species in other Mediterranean countries such as Cyprus, Jordan, Israel and Syria (Turnbull and Booth, 2002). Acacia saligna occurs throughout coastal regions of south-western Australia on a range of soil types (Rippey and Rowland, 2004). It tolerates saline soil and is planted for soil stabilization, tannin, fodder, bush food, gum and fuelwood production (Brockwell et al., 2005). Approximately, 300 000 ha of A. saligna plantations have been established worldwide, mostly in North Africa, West Asia and parts of South America (Midgley and Turnbull, 2003). In arid zones of North Africa, A. saligna is considered as a multi-purpose tree species and is widely planted for fuelwood and fodder for animals or in agroforestry systems (El Nasr et al., 1996;

Murdoch University 2011 32

Chapter 3: Reforestation of a degraded eucalyptus woodland

Odenyo et al., 1997; Droppelmann and Berliner, 2000).

In the literature (Chapter 2), beneficial organisms (mycorrhizal fungi and N2-fixing bacteria) have shown a potential role in reforestation success. Selected strains of microorganisms have helped seedlings to establish and overcome harsh conditions of reforestation sites. Direct seeding is an alternative to planting nursery-raised seedlings, where multi-species plantations are to be established for biodiversity and environmental concerns. Direct seeding of E. gomphocephala and A. saligna has been tested or used in reforestation programs in WA and in some other parts of the world (e.g., Michaelides, 1979; Piggott et al., 2002; Ruthrof et al., 2003; Maslin and McDonald, 2004) with varying degrees of success. However, the value of microorganisms in promoting the survival and growth of seedlings from direct seeding of these two species has not yet been investigated.

The aim of this Chapter is to investigate the value of symbiotic bacteria and fungi in enhancing the early growth and survival of two key colonizing species, E. gomphocephala and A. saligna, in a disturbed and declining tuart woodland. Consideration of the possibilities of promoting early growth and survival of E. gomphocephala and A. saligna led to the following hypotheses: 1) That beneficial organisms and inorganic fertilizer enhance survival and early growth from direct seeding and from nursery-raised seedlings of these two species; and 2) That direct seeding and planting of nursery-raised seedlings are equally successful methods in reforestation of a harsh mediterranean-type environment. To test these hypotheses, two separate experiments were established, with direct seeding and nursery-raised seedlings, at one site adjacent to Yalgorup National Park, Western Australia.

3.2 Materials and methods

3.2.1 Site conditions and preparation

The experiments were set up on a private property adjacent to Yalgorup National Park (YNP), 80 km south of Perth, Western Australia. This location has a latitude of 32o 42' 18" S, longitude of 115o 38' 21'' E. The region experiences a mediterranean-type climate with hot, dry summers and cool, wet winters. Rainfall starts in April and finishes in October (Figure 3.1). The total annual rainfall in 2008 was 638 mm. The monthly

Murdoch University 2011 33

Chapter 3: Reforestation of a degraded eucalyptus woodland maximum and minimum rainfall in 2008 was 162 and 0 mm in July and January, respectively. The maximum temperature was 38oC in December 2007 or January 2008 and the minimum temperature was 5ºC in July (BOM, 2009). The area is located on the Swan Coastal Plain (Government of Western Australia, 2003). The soil is classified in the Spearwood Dune System, which is characterized by leached sand at the surface and pale yellow to reddish brown subsoils (McArthur and Bettenay, 1974). Limestone outcrops are common in the trial site. Soil samples (0 to 10 cm) were randomly collected across the blocks and bulked for physical and chemical analysis by a commercial laboratory, CSBP Limited (Cumming Smith British Petroleum) (Table 3.1).

Rainfall Mean temperature Minimum temperature Maximum temperature 200 40

160 C) 30 º 120 20 80 (mm) Rainfall

10 Temperature ( 40

0 0

08

07 07 08 08 08

07 08

07 08

07 07

08

07 08

-

- - - - -

- -

- -

- -

- -

-

Jul Jul

Jan

Jun Jun

Oct

Sep Feb

Apr

Dec

Mar

Aug Aug Nov May Month/year

Figure 3.1 Monthly rainfall and temperature (mean, absolute maximum and absolute minimum) during the period of the field trial. Climatic data were from Mandurah (station 009977), 32o 52' S and 115o 71' E, 10 km to the North of the trial site (Bureau of Meteorology, 2008).

The trial site contained scattered large tuart trees, but no native understorey remained and no natural regeneration was evident. The area had been used for cattle grazing and cattle yards, and there was no fertilizer use immediately prior to trial establishment (Pers. comm., Sarich’s property manager, 2008). Throughout the plot, the invasive species load was quite high, including species such as onion weed (Trachyandra sp.), lupins (Lupinus ssp.) and various grasses. Weeds were sprayed with glyphosate herbicide (Roundup® Ready To Use 10 ml of glyphosate in 1L of water) one week prior to trial establishment. The trial area was fenced with wire mesh to a height of 1.5 m to

Murdoch University 2011 34

Chapter 3: Reforestation of a degraded eucalyptus woodland exclude kangaroos and vehicles.

Table 3.1 Soil chemical and physical properties of the two reforestation experiments on private property, adjacent to Yalgorup National Park, Western Australia (depth 0-10 cm)

Soil property Value Soil property Value Clay (%) 3.96 ± 0.43 Extractable B (mg/kg)9 0.53 ± 0.03 Silt (%) 2.25 ± 0.62 Extractable S (mg/kg)10 5.58 ± 0.67 Sand (%) 93.80 ± 0.61 Extractable Fe (mg/kg)11 258.04 ± 1 11 pH CaCl2 6.40 ± 0.18 DTPA Cu (mg/kg) 14.040.23 ± 0.02 2 11 pH H2O 7.03 ± 0.11 DTPA Zn (mg/kg) 0.28 ± 0.02 Organic Carbon (%)3 1.59 ± 0.15 DTPA Mn (mg/kg)11 1.99 ± 0.45 Nitrate N (mg/kg)4 6.00 ± 0.41 DTPA Fe (mg/kg)11 14.46 ± 1.31 Ammonium N (mg/kg)4 4.50 ± 0.29 Exchangeable Ca (meq/100g)12 6.67 ± 0.54 Total N (%)5 0.14 ± 0.01 Exchangeable Mg (meq/100g)12 0.49 ± 0.05 Extractable K (mg/kg)6 35.75 ± 4.53 Exchangeable Na (meq/100g)12 0.09 ± 0.00 Extractable P (mg/kg)6 13.00 ± 1.47 Exchangeable K (meq/100g) 12 0.12 ± 0.00 Available P (mg/kg)7 6.03 ± 0.62 Exchangeable Al (meq/100g)13 0 ± 0.00 8 14 Total P (mg/kg) 301.00 ± 15.58 Soluble Cl2 (mg/kg) 19.50 ± 1.94 Electrical conductivity (dS/m)15 0.05 ± 0.00

The values are mean ± S.E. (n = 4). Unless otherwise stated, soil chemical properties were determined using methods described in Rayment and Higginson (1992) and their codes are indicated in the parentheses. 1 (4B2). 2 (4A1). 3 Walkley-Black (6A1). 4 2 M KCl (7C2). 5 Kjeldahl steam distillation 6 7 8 9 10 (7A1). Colwell (9B1). Olsen (9C2). Pooled/inhouse method. Hot 0.01 M CaCl2 extraction (12C1). 11 12 13 0.25 M KCl 40 (Blair et al., 1991). (12A1). 0.1 M NH4Cl at pH = 7 (15E1). 1 M KCl (15G1). 14(5A1). 15 1:5 soil:water extraction (3A1).

3.2.2 Procurement of materials

3.2.2.1 Seeds and seedlings

Seeds of E. gomphocephala and A. saligna were collected in the region in 2002. Seeds of A. saligna with no insect attack or mechanical damage were selected by hand, based on visual appearance. Prior to direct seeding, acacia seeds were pretreated by immersion in boiled water (100ºC) for 1 min followed by soaking in tap water overnight (Gunn, 2001). Seedlings of E. gomphocephala were raised in seedling trays (PL81F manufactured by Lännen Plant Systems, Finland) from December 2006 to early June 2007. Each had internal dimension of 41 × 41 × 73 mm. The potting mix was composed of 92% composted pine bark and 8% peat by volume. At the time of planting, the seedlings

Murdoch University 2011 35

Chapter 3: Reforestation of a degraded eucalyptus woodland were six-months old with an average height of 25 cm. Seedlings of A. saligna were raised in 56 cell seed trays (Kwikpot trays, manufactured by Garden City Plastic, Australia). The size of a cell was 35 × 35 × 70 mm. The rooting medium, comprising peat and perlite (2:1, v:v), was steam-pasteurized twice at 65ºC for 2 h. Seedlings were raised in a glasshouse for three months from March to early June 2007. At the time of planting, the average height of A. saligna was 15 cm. Seedlings of the two species received standard forest nursery fertigation.

3.2.2.2 Fertilizer

The field fertilizer treatment was application of commercial tablets (TyphoonTM controlled-release tablets, Langley Australia Pty Ltd, Perth). Each 10-g tablet contained: 20% N as ammonium and urea; 4.4% P as water soluble, citrate soluble and citrate insoluble; 8.2% K as sulphate; 4% Ca as phosphate; 6% S as sulphate; 0.2% Mg as oxide; 0.03% Cu as sulphate; 0.5% Zn as oxide; 0.33% Fe as sulphate; 0.16% Mn as sulphate; 0.01% Mo as molybdate; and 0.01% B as tetraborate.

3.2.2.3 Mycorrhizal fungi

Two sources of mycorrhizal fungi were used. The first, ectomycorrhizal fungi (Table 3.2, Experiment 1: ECM fungi) in the form of a spore slurry, consisting of Scleroderma cepa Pers., Pisolithus marmoratus (Berk.) E. Fisch., Laccaria laterita Malençon and Amanita eucalypti O.K. Mill. with roughly equal numbers of spore per taxon. The fungi were collected from natural habitats of tuart forests in the YNP region in 2006, and they had been stored at 4°C prior to use. The second fungal source was a commercial mycorrhizal inoculum (Table 3.2, Experiment 2: commercial mycorrhizas) provided by Zadco For Quality Gro Pty Ltd (Sydney) (Figure 3.2a). Sachets of clay-based carrier, contained a total of 21 000 propagules of Glomus intraradices Schenck & Smith, Glomus mosseae (Nicolson & Gerdemann) Gerd. & Trappe, Laccaria bicolor (Maire) P. D. Orton, Pisolithus tinctorius (Pers.) Coker & Couch, Scleroderma cepa, Scleroderma aff. geastrum and Scleroderma citrinum Pers.

Murdoch University 2011 36

Chapter 3: Reforestation of a degraded eucalyptus woodland

a b

Figure 3.2 (a), Commercial mycorrhizal inoculum and (b), application of inoculum below the seedling in the planting hole.

3.2.3 Experimental design and treatment

Two separate field experiments were established. The first investigated the effect of ectomycorrhizal (ECM) fungal spores and inorganic fertilizer on direct seeding and nursery-raised seedlings of E. gomphocephala. This was a 2 × 3 factorial experiment, consisting of two plant materials (seed and seedling) and three treatments (Table 3.2) in a randomized complete block (RCB) design. Therefore, each block contained six plots, three of which were randomly allocated to seed and the other three to seedling. Each plot was then randomly assigned one treatment. The second examined the effects of N2- fixing bacteria and mycorrhizal fungi on direct seeding and planted seedlings of A. saligna. This was a 2 × 8 factorial experiment consisting of two plant materials (seed and seedling) and eight treatments (Table 3.2) in a RCB design. Therefore, in a block there were 16 plots, eight of which were randomly allocated to seed and the other eight to seedlings. Each plot was randomly assigned one treatment. Each experiment contained three blocks at the same site. For both experiments, each plot contained 25 planting holes or seeding spots, five rows of five holes each. Spacing between adjacent holes in the same row and that between adjacent rows was 1 m.

Murdoch University 2011 37

Chapter 3: Reforestation of a degraded eucalyptus woodland

Table 3.2 Details of treatments and method of treatment application in the field trials in Western Australia

Treatment Method of treatment application

Experiment 1: Eucalyptus gomphocephala ECM fungi Each seedling/seeding spot received 10 ml of spore slurry (9 × 106 spores) buried in contact with roots on two sides (at 5 cm from each seedling). The same number of spores was applied ca. 0.5 cm beneath the seeds in the direct seeding plots. ECM fungi + fertilizer In addition to ECM fungi, one 10-g tablet was buried to a depth of 10 cm within 30 cm from each seedling at three months after trial establishment. Control No fertilizer or ECM inoculum

Experiment 2: Acacia saligna T1: root nodule Crushed root nodule solutions were made from root nodules (two days old) collected from young A. saligna trees in the Murdoch University’s nature reserve. The solution was applied to one-month old seedlings in the nursery or beneath the seeds at the time of seeding (ca. 1 nodule seedling-1 or seeding spot-1). T2: commercial mycorrhizasl A sachet of commercial inoculum was placed at the bottom of the hole fungi in contact with the root ball of each seedling or at ca. 0.5 cm below the seeds at the time of seeding (Figure 3.2b). T3: fertilizer Fertilizer was applied as in Experiment 1. T4 Root nodule + commercial mycorrhizal fungi. T5 Root nodule + fertilizer. T6 Root nodule + commercial mycorrhizal fungi + fertilizer. T7 Autoclaved commercial mycorrhizal fungi (at 121oC for 20 min). T8 No treatment applied.

3.2.4 Trial establishment

The trial was established in June 2007. Seedlings were planted using a pottiputki implement (one seedling in each planting hole). A trowel was used to make seeding spots, ca. 4 × 4 × 1 cm in size. Seeds ( three for A. saligna or three to five for E. gomphocephala) were placed in each seeding spot and covered with soil to a depth of ca. 0.2 - 1 cm. Direct seeding spots were thinned to 1 plant at three months, and weeds were carefully removed within a 30 cm radius around the seedlings for all experiments.

Murdoch University 2011 38

Chapter 3: Reforestation of a degraded eucalyptus woodland

3.2.5 Data collection and analysis

For both experiments, seedling survival was assessed at three, six and 13 months, representing the critical stages in the first year of growth: middle of the first wet season (August), early summer (November) and middle of the second wet season (July). The first growing season is considered to be, the critical period for plant survival (Savill et al., 1997; Castro et al., 2004). Seedling survival was calculated based on the number of planting or direct seeding holes with presence of a seedling at the time of assessment as percentage of the total planting/direct seeding holes per plot (n = 25). Seedling height at 13 and 40 months was measured between the soil level and the highest shoot tip. Percentages of survival rates were arc-sin square root transformed (Gomez and Gomez, 1984), and data on height growth in Experiment 2 were Napierian logarithmic transformed prior to statistical analysis to fit the normal distribution. The descriptive data presented in the tables are untransformed.

At 13 months, leaves were harvested (Experiment 2) for assessment of nutrient concentration. In each plot, 20 young-fully-expanded leaves (or phyllodes) were collected from 10 randomly selected trees, except in some direct seeding plots where the number of trees was less than 10. At 13 months, there were too few E. gomphocephala trees remaining in the direct seeding plots (Table 3.3) and therefore leaf nutrient assessment of this species was not undertaken. The samples were dried in an oven at 70ºC for 48 h to constant weight before sending to a commercial laboratory, CSBP Limited, for inorganic nutrient analysis by inductively coupled plasma atomic emission spectroscopy (ICP-AES) following microwave acid digestion (Zarcinas et al., 1987).

Data were analysed using SPSS Version 15.0. Univariate Analysis of Variance was applied to test the effects of treatments, plant material and their interaction on survival and height growth. The Duncan’s new Multiple Range Test was used to detect differences among means at p ≤ 0.05. Multivariate analysis of variance (MANOVA) was used to test the effects of inoculum and plant material on leaf nutrient concentration of A. saligna. Wilk’s Lambda multivariate test was initially used, followed by univariate tests for each nutrient element.

Murdoch University 2011 39

Chapter 3: Reforestation of a degraded eucalyptus woodland

3.3 Results

3.3.1 Experiment 1: Eucalyptus gomphocephala

3.3.1.1 Seedling survival

There was no effect of treatment on seedling survival (Table 3.3) at each of the three assessment times. There was a significant effect of plant material at each of the three assessments. The interaction between the two factors at each of the three assessments was not significant, indicating that the difference between plant materials was not significantly affected by the treatments used. The effects of treatment and plant material on survival rate did not change until 40 months (Appendix III).

Table 3.3 The effects of treatment and plant material on survival and growth of Eucalyptus gomphocephala established by direct seeding and nursery-raised seedlings

Survival three Survival six Survival 13 Height growth months (%) months (%) months (%) 13 months (cm) Treatment ECM fungi 88.00 ± 6.45 60.00 ± 16.69 46.00 ± 17.37 57.14 ± 12.19 ECM fungi + fertilizer 97.33 ± 1.69 58.67 ± 17.05 44.67 ± 18.25 54.59 ± 15.59 Control 84.67 ± 5.00 48.00 ± 16.36 42.67 ± 14.63 49.09 ± 10.01

Plant material Seed 82.67 ± 4.52 b 19.56 ± 5.14 b 7.56 ± 2.05 b 33.12 ± 4.54 b Seedling 97.33 ± 1.49 a 91.56 ± 2.35 a 81.33 ± 3.13 a 74.09 ± 9.07 a

Analysis of variance

Source d.f. F Sig. F Sig. F Sig. F Sig. Block 2 0.822 0.467 0.183 0.836 0.826 0.465 8.142 0.008 Treatment (Tr.) 2 2.421 0.139 2.273 0.154 0.221 0.805 0.395 0.683 Plant material (Pl.) 1 9.891 0.010 106.295 0.000 241.777 0.000 29.431 0.000 Tr. × Pl. 2 0.941 0.422 0.132 0.878 1.775 0.219 0.885 0.443 Error 10

Values are replicated block means (n = 3) ± S.E. The significant ANOVA probabilities are marked in bold. In a column, means followed by the same letters are not significantly different at p ≤ 0.05.

At three months, the survival rates of both planting materials was very high, 97.3% for nursery-raised seedlings and 83% for direct seeding. Over time there was an increase in mortality of both planting materials. More than 90% of the direct-seeded seedlings recorded in the first assessment died (Figure 3.3b) by the time of the assessment at 13

Murdoch University 2011 40

Chapter 3: Reforestation of a degraded eucalyptus woodland months, however, only 16% of the nursery-raised seedlings died over the same period. The significant decrease in the survival rate from direct seeding was observed at six months (November assessment), one month into the summer drought. At 13 months, the nursery-raised seedlings maintained their survival rate at 81.3%, ca. 11 fold higher than that of direct seeding, being 7.6%.

a b c c ) )

c d c c ) )

Figure 3.3 Seedlings of Eucalyptus gomphocephala established from direct seeding at (a) three months and (b) dead seedlings at the same period. Plants established by (c) seed and (d) seedling at 13 months. Bars in (a) and (b) represent 5 cm; poles in (c) and (d) painted with 20 cm-colour bands.

3.3.1.2 Seedling growth

At 13 months, there was no effect of treatment on height growth (Table 3.3). However, there was a block effect on height growth. Plants in Block 1 grew faster than those in Block 2 and Block 3 by 82 and 69%, respectively (data not presented). However, limited soil sampling showed that soil nutrients in the three blocks did not differ substantially (Appendix II). Plant material had a significant effect on height growth with the nursery-raised seedlings reaching 74.1 cm, more than double that of plants

Murdoch University 2011 41

Chapter 3: Reforestation of a degraded eucalyptus woodland from direct seeding (33.1 cm) (Figure 3.3c, d). The interaction of treatment and plant material was not significant, indicating that the difference in plant material was not significantly affected by the application of inoculum. The effects of treatment and plant material on height remained at 40 months (Appendix III).

3.3.2 Experiment 2: Acacia saligna

3.3.2.1 Seedling survival

The survival of A. saligna was significantly affected by treatment (Table 3.4) at each of the three assessments. There was a significant effect of plant material at each of the three assessments. There was no significant interaction between treatment and plant

Table 3.4 The effects of treatment and plant material on survival and growth of Acacia saligna established by direct seeding and nursery-raised seedlings

Survival three Survival six Survival 13 Height growth months (%) months (%) months (%) 13 months (cm)

Treatment T1: root nodule 80.67 ± 6.40 abcd 76.00 ± 8.00 ab 60.00 ± 12.65 ab 71.67 ± 12.71 abc T2: mycorrhizal fungi 71.33 ± 7.04 d 59.33 ± 11.43 c 50.67 ± 12.97 b 52.13 ± 11.66 c T3: fertilizer 92.67 ± 3.17 a 84.00 ± 5.93 a 78.67 ± 7.06 a 89.28 ± 14.39 a

T4: T1 + T2 84.00 ± 3.27 abcd 78.67 ± 4.22 ab 67.33 ± 6.96 ab 76.83 ± 19.12 abc T5: T1 + T3 92.00 ± 2.31 ab 85.33 ± 4.70 a 77.33 ± 7.77 a 76.21 ± 10.81 ab T6: T1 + T2 + T3 73.33 ± 5.90 cd 68.67 ± 6.81 bc 60.67 ± 7.96 b 84.24 ± 11.53 a

T7: autoclaved T2 74.67 ± 11.67 bcd 64.00 ± 13.06 bc 56.00 ± 15.49 b 73.18 ± 18.08 abc T8: control 86.00 ± 4.59 abc 70.00 ± 10.42 abc 55.33 ± 11.57 b 59.41 ± 12.53 bc Plant material Seed 73.33 ± 3.56 b 57.50 ± 3.85 b 42.50 ± 4.20 b 48.45 ± 3.73 b Seedling 90.33 ± 1.73 a 89.00 ± 1.60 a 84.00 ± 1.80 a 97.29 ± 5.72 a Analysis of variance

Source d.f. F Sig. F Sig. F Sig. F Sig. Block 2 0.302 0.742 0.245 0.784 0.038 0.963 1.416 0.258 Treatment (Tr.) 7 3.232 0.011 3.440 0.008 3.685 0.005 2.740 0.025 Plant material (Pl.) 1 30.201 0.000 84.522 0.000 120.57 0.000 69.146 0.000 Tr. × Pl. 7 1.955 0.095 1.873 0.110 2.132 0.070 1.073 0.405 Error 30

Values are replicated block means (n = 3) ± S.E. The significant ANOVA probabilities are marked in bold. In a column, means followed by the same letters are not significantly different at p ≤ 0.05.

Murdoch University 2011 42

Chapter 3: Reforestation of a degraded eucalyptus woodland material, indicating that the treatment effect did not differ with plant material tested and that plant material effects were not significantly affected by the application of treatments.

At all of the assessments, the inorganic fertilizer resulted in improved seedling survival over the control, and at 13 months this improvement was 42%. None of the sole microorganism treatments or their combination improved survival over the control. Moreover, the commercial mycorrhizal fungi significantly reduced seedling survival (17%) at three months compared with the control. At three months, the survival rates of both planting materials were high, being 90.3 % for nursery-raised seedling and 73.3% for direct seeding. Over time there was an increase in mortality of both planting materials. More than 40% of the direct-seeded seedlings recorded in the first assessment had died by the time of the assessment at 13 months. However, only 7% of the nursery- raised seedlings died over the same period. At 13 months, the overall survival rate was 84%, approximately double that of plants from direct seeding (42.5%). These trends continued at 40 months (Appendix III).

3.3.2.2 Seedling growth

At 13 months, there was a significant effect of treatment on height growth (Table 3.4). However, only inorganic fertilizer, and its combination with root nodule and mycorrhizal fungi (T6) resulted in a significant improvement, by 50 and 42%, respectively, over the control. Height growth of A. saligna was significantly affected by plant material. The nursery-raised seedling reached 97.3 cm, double that of direct seeding, being 48.5 cm (Figure 3.4b, c). The interaction of treatment and planting material was not significant, indicating that the difference in plant material was not significantly affected by the application of treatments. At 40 months, there was no longer an effect of treatment; however, plant material maintained its effectiveness on height growth (Appendix III).

3.3.2.3 Phillode nutrient concentration

Multivariate test (Wilks' Lambda) indicated that there was a significant effect of plant material on the phyllode nutrient concentration [F (14, 17) = 4.593, p < 0.01], but there was no effect of treatment [F (98, 116.438) = 1.253, p = 0.121] and an interaction

Murdoch University 2011 43

Chapter 3: Reforestation of a degraded eucalyptus woodland between treatment and planting material [F (98, 116.438) = 0.837, p = 0.817]. Univariate analysis (Table 3.5) indicated that five of the 14 nutrient elements (N, P, Ca, Cl and B) were significantly different between direct seeding and nursery-raised seedlings. Phyllodes of the nursery-raised seedlings contained higher concentrations of N, P and B whereas phyllodes from direct seeding had higher concentrations of Ca and Cl.

Table 3.5 Phyllode nutrient concentrations of Acacia saligna in direct seeding and nursery-raised seedlings as effects of planting materials after 13 months

Element From direct seeding From nursery-raised seedling F and p value (d.f. 1, 30) N (%) 2.15 ± 0.04 2.30 ± 0.04 F = 6.954, p = 0.013 P (%) 0.12 ± 0.00 0.13 ± 0.00 F = 7.375, p = 0.011 K (%) 0.65 ± 0.04 0.75 ± 0.03 F = 3.029, p = 0.092 S (%) 0.26 ± 0.01 0.29 ± 0.01 F = 3.749, p = 0.062 Na (%) 0.17 ± 0.01 0.17 ± 0.00 F = 1.102, p = 0.302 Ca (%) 1.70 ± 0.05 1.54 ± 0.04 F = 5.172, p = 0.030 Mg (%) 0.17 ± 0.00 0.16 ± 0.00 F = 2.753, p = 0.108 Cl (%) 1.25 ± 0.04 1.11 ± 0.04 F = 7.236, p = 0.012 Cu (mg/kg) 2.89 ± 0.40 2.72 ± 0.14 F = 0.204, p = 0.655 Zn (mg/kg) 12.38 ± 0.80 11.84 ± 0.56 F = 0.464, p = 0.501 Mn (mg/kg) 21.42 ± 1.03 19.96 ± 1.00 F = 0.888, p = 0.354 Fe (mg/kg) 30.53 ± 0.99 33.90 ± 1.41 F = 3.631, p = 0.066

NO3 (mg/kg) 40.42 ± 0.20 41.08 ± 0.29 F = 3.187, p = 0.084 B (mg/kg) 12.23 ± 0.50 14.96 ± 0.94 F = 7.216, p = 0.012

Values are replicated block means (n = 3) ± S.E. The significant ANOVA probabilities are marked in bold.

a b c

c c c ) ) )

Figure 3.4 Seedlings of Acacia saligna established from direct seeding at (a) three months and (b) 13 months. (c) Planted seedlings after 13 months. A bar in (a) represents 5 cm; poles in (b) and (c) painted with 20 cm colour bands.

Murdoch University 2011 44

Chapter 3: Reforestation of a degraded eucalyptus woodland

3.4 Discussion

3.4.1 Effects of treatments on survival and growth

In Experiment 2, A. saligna responded strongly to the sole fertilizer application in survival and height growth. At 13 months, the combination of fertilizer with crushed root nodule (T5) improved the survival rate by 40% and the combination of fertilizer with root nodules and fungi (T6) improved height growth by 42%. The fertilizer effect disappeared at 40 months (Appendix III). This could be because plants without the starter fertilizer application eventually were able to access soil nutrient stores with root exploration and growth caught up. Alternatively, other factors become limiting for the faster growing trees and the size difference between the two diminished over time. Diminished effect of fertilizer with time was reported by Taylor et al. (2006) from their three-year investigation of the influences of seedling browse protection and fertilizer application on growth of Nuttall oak (Quercus nuttallii Palm.) in west Alabama. In reforestation trials, fertilizer has been used solely or in combination with other survival/growth promoting materials such as water retention polymers (e.g., Al-Humaid and Moftah, 2003). However, application of fertilizer in field trials does not always result in significant improvement in plant growth and considerably higher amounts are often used than that applied in this trial. As an example, Vincent and Davies (2003) buried 100 g of NPK fertilizer (12:12:12) around planted seedlings of Dryobalanops aromatica Gaertn. f. and Shorea parvifolia Dyer at two sites in Sarawak, Malaysia. The fertilizer was applied one week after planting and then at six month intervals. After 22 months they found that the effect of fertilizer on plant growth (height and diameter) was site specific. The effect of fertilizer on seedling survival and growth is further discussed in Chapter 7.

These trials have shown that none of the inoculant treatments (single or combined) substantially improved survival or growth. There are numerous studies showing positive effects of inoculation with mycorrhizal fungi on seedlings in nurseries or in field trials (Bâ et al., 2002; Chen et al., 2006a; Chen et al., 2006b; Duponnois et al., 2007), but there are also reports of neutral and negative effects of mycorrhizal fungi on trees species (Xu et al., 2002; Appleton et al., 2003; Teste et al., 2004; Schwartz et al., 2006; Repáč, 2007). This is because the effectiveness of mycorrhizal fungi for a host plant depends on compatibility between the host-fungus partners, degree of mycorrhizal

Murdoch University 2011 45

Chapter 3: Reforestation of a degraded eucalyptus woodland dependency of the host, the effectiveness of mycorrhizal fungi under the inoculated site conditions, and the competition of indigenous mycorrhizal fungi (Ortega et al., 2004).

Acacia saligna forms symbiotic associations with AM fungi (Reddell and Warren, 1987), and a previous study (Benbrahim and Ismaili, 2002) has demonstrated that one of the species used, Glomus mosseae, is effective for this host. However, it is not known if it formed effective associations in this experiment due to the mixed inoculum that was used. It is possible that the commercial inoculum might have been ineffective as no quality control information was available on the batch used. For E. gomphocephala, the four ECM fungi were collected from an E. gomphocephala woodland in the trial region in 2006, and were identified by a mycologist at Murdoch University. Although there is no published report on fungus-host compatibility, it is likely that the four fungal species form symbiotic associations with E. gomphocephala as they are common species in these woodlands. There is a strong possibility that there was competition between the inoculated fungi with existing inoculum load for ECM fungi (e.g., Requena et al., 2001). For example, Lu et al. (1999) showed that ECM fungal diversity in E. globulus plantations increased by one to three species per year from establishment on ex-pasture sites presumably from movement of spores in air from remnant bush nearby. It was not possible for this study to identify which of these explanations is the major contributing factor, because mycorrhizal formation was not accessed in the seedlings. The need for inoculation with mycorrhizal fungi in the field is further discussed in Chapter 8.

As with mycorrhizal fungi, there was no effect of adding N2-fixing bacteria on survival and height growth of A. saligna. Competition from the N2-fixing bacteria at the trial site could be the cause as excavations of sample seedlings after 13 months of growth showed that nodulation occurred in trees in both inoculated and uninoculated plots of the two planting materials (data not presented). Acacia saligna is a promiscuous tree and associates with a diverse range of rhizobia (Marsudi et al., 1999; Yates et al., 2004). The experimental site at Yalgorup was adjacent to native woodland that contains numerous indigenous members of the Mimosaceae and Fabaceae. In addition, the disturbed land contained scattered exotic annual legumes, such as lupins (Lupinus ssp.).

The need for inoculation with N2-fixing bacteria is further discussed in Chapter 8.

Murdoch University 2011 46

Chapter 3: Reforestation of a degraded eucalyptus woodland

3.4.2 Seedling survival and growth―differences between plant material

At 13 months, the survival rate of direct seeding was 42.5% for A. saligna and 7.5% for E. gomphocephala, whereas survival of the nursery-raised seedlings for both species was more than 80%. There was greater loss of plants in the direct-seeded E. gomphocephala than in the A. saligna plots as survival rate at three months was 83% and 73%, respectively. Previous studies have shown that survival is often low with direct seeding. As an example, Doust et al. (2008) investigated the survival and growth performance of direct seeding of Acacia celsa Tindale, Alphitonia petriei C.T White & Braid, Castanospermum australe A. Cunn. & Fraser ex Hook., Cryptocarya oblata F. M. Bailey and Flindersia brayleyana F. Muell. on degraded sites in the wet tropical region of northeast Queensland, Australia, with annual rainfall from 2 643 to 3 393 mm.

The survival rates after two years were predominantly < 20% when the seed were sown early in the wet season (December). A low survival rate of E. gomphocephala established by direct seeding was previously reported by Ruthrof et al. (2003) from a study on the ability of Eucalyptus cladocalyx F. Muell., E. gomphocephala and Eucalyptus marginata Donn ex Smith to recruit seedlings in a long-unburnt area of King Park, Perth, with similar environmental condition as the trial site in this study. The number of germinants of the three eucalypts increased during the early part of the wet season (June and July), but all the newly established seedlings died towards summer. Therefore, the low survival rate of E. gomphocephala in this experiment was not an exception. This study showed that nursery-raised seedling had higher survival rates as compared to direct seeding for both tree species. For A. saligna, both survival rate and growth of the nursery-raised seedling at 13 months doubled that of direct seeding. Similarly, the superiority of the nursery-raised seedling of E. gomphocephala over the direct seeding was more than double in height and 11 fold in survival rate. These findings parallel field experiments undertaken by Löf et al. (2004) on sites with predominantly sandy soils of former farmlands in Denmark and southern Sweden. In this study, after four years, planted seedlings of four European broadleaved tree species (Fagus sylvatica L., Quercus robur L., Prunus avium L. and Crataegus monogyna Jacq.) survived better than plants from direct seeding.

The significant differences in survival between the two planting materials may be related to the need to develop extensive roots prior to the onset of the summer drought.

Murdoch University 2011 47

Chapter 3: Reforestation of a degraded eucalyptus woodland

The capacity of leaf area to drive belowground biomass early in recruitment (Kramer, 1969) is obviously a key factor. The survival rate of direct seeding plants decreased drastically following the increase in temperature coupled with cessation of rainfall in November (Tables 3.3 and 3.4). By late summer, soil in the region has no available water within about 1 m of the surface (McArthur and Bettenay, 1974). Presumably, larger plants were able to survive the summer because they possessed deep, branched root systems, which could access stored water in deeper profiles. This was simply a consequence of the older growth stage of the nursery-raised seedling with a developed root system able to make a rapid growth after outplanting. In addition, the nursery- raised seedlings of A. saligna must have had benefits from the symbiotic association with N2-fixing bacteria by starting nodulation and N2 fixation sooner than the germinants from direct seeding as the former had been inoculated with crushed nodules in the nursery. Considering the high survival rate of direct seeding of both species at the first assessment, it is worth following up these assumptions by amelioration of soil conditions to enhance soil water and nutrient availability for direct seeded seedlings through, for example, application of a water retention polymer in combination with fertilizer (Chapters 5 and 7).

3.4.3 Survival from direct seeding―comparison of the two species

In direct seeding, A. saligna had a much greater survival as compared to E. gomphocephala. This difference needs to be considered given that the two species were chosen to represent two guilds: legumes and non-legumes. The trial area is characterized by nutrient deficiency (Table 3.1). Soil in the study area contained high NO3–N, but very low organic carbon compared to soils under natural forest in Yalgorup National Park (Cai et al., 2010). However, the level of soil N in the study site was considerably lower than in agricultural soils in Australia (Hazelton and Murphy, 2007). The low level of soil mineral N might not be a serious problem for A. saligna as it had ability to form root nodules. Partial excavations of sample seedlings revealed that nodulation occurred in all plots (data not presented), hence there were compatible Bradyrhizobium in the soil. However, the effectiveness of these nodules in fixing N is unknown. Ecologically, A. saligna is a highly drought-resistant species (Nativ et al., 1999) and a larger part of its biomass is allocated to root growth (Witkowski, 1991). Having all these biological and ecological characteristics, the young Acacia seedlings should adapt well to nutrient-poor sites. In

Murdoch University 2011 48

Chapter 3: Reforestation of a degraded eucalyptus woodland contrast, as a non- N2-fixing species, the newly emerged Eucalypt seedling suffered from the limited soil N. This limitation restricts early root formation, development and penetration into a deeper soil (Price, 2006), such poor rooting may have led to the direct-seeded seedlings starting to die at three months (Figure 3.3b).

In addition to the soil nutrient status, the low survival rate of direct seeded E. gomphocephala may possibly be exacerbated by water repellency of the soil at the trial site. Water repellency is common in sandy soils in Australia (Harper and Gilkes, 1994). This soil characteristic resulted from waxy organic compounds coating soil particles (Franco et al., 1995; Roper, 2006). It occurs in soils with coarse texture with low clay content, and more severely in sandy soils with clay < 5% (Harper and Gilkes, 1994). Water repellent soil inhibits the penetration of water into the soil, and thus discourages seed germination and plant growth (DeBano, 1981).

The differences in survival between the two species may be explained in relation to the seed size. The effect of seed size on the early performance of seedlings was explained by Milberg and Lamont (1997) in an investigation into the effect of removal of cotyledons from the newly emerged seedlings of four woody species (Eucalyptus todtiana, F. Muell., Eucalyptus loxophleba Benth., Hakea psilorrhyncha R. M. Barker and Hakea circumalata Meisn.). The experiment was conducted in a glasshouse using two types of soils, sand and loam. Results showed that the nutrient content in cotyledons is an influential factor affecting the early survival of seedlings. After emergence, the early growth of larger-seeded seedlings relied more heavily on nutrient content in the cotyledons than from the soil as compared to smaller-seeded species that relied greatly on soil nutrients. The authors suggested that this characteristic enables the larger seeded species to adapt to nutrient impoverished sites as they had built up physical strength before tapping nutrients from the soil.

The low survival rate of E. gomphocephala after 13 months does not generally mean that direct seeding of this species is not applicable. In this trial, seedling survival at three months was very high, i.e. 83%. As discussed above, the significant decrease in survival after the first assessment was the result of unfavourable site conditions. Site selection should be further studied for direct seedling of this species. Elsewhere in Western Australia, ashbeds, which can be created during hot fires, have been reported to

Murdoch University 2011 49

Chapter 3: Reforestation of a degraded eucalyptus woodland promote natural regeneration and enhance growth and survival of E. gomphocephala seedlings (Ruthrof et al., 2003; Archibald et al., 2006; Barber and Hardy, 2006; Ruthrof and Close, 2006). Ashbeds are rich in soil mineral nutrients, mainly P and K (Humphreys and Lambert, 1965; Bauhus et al., 1993; Chambers and Attiwill, 1994). These nutrient-rich growing media can enhance early growth of seedlings and has been linked with increased rooting depth and profusely rooted plants, characteristics which are vital for surviving summer drought (Archibald et al., 2006). Therefore, further research on direct seeding of E. gomphocephala in its natural habitat may need to investigate the effect of ashbeds, by applying or creating on site, in conjunction with supplements of mineral nutrients and water retention polymers.

3.5 Concluding remarks

Mycorrhizal fungi and N2-fixing bacteria did not enhance the survival and growth of the two key colonizing tree species in declined woodland with a harsh mediterranean-type environment. However, inorganic fertilizer enhanced survival and early growth of A. saligna. The small fertilizer tablet used are typically of commercial produce marketed in Australia to boost early growth of woody planting stock, either seedlings or rooted cuttings. Planting of nursery-raised seedlings resulted in better survival and growth than direct seeding of the two species evaluated during the 13 months. This study indicates that direct seeding of E. gomphocephala is not promising. Site constrains that reduce seeding success may possibly be overcome by application of water retention polymer with inorganic fertilizer (Chapters 5 and 7). By contrast, direct seeding of the woody legume A. saligna was effective.

Murdoch University 2011 50

Chapter 4: Conservation of high-value timber species in Cambodia

CHAPTER 4

CONSERVATION AND UTILIZATION OF THREATENED HARDWOOD SPECIES THROUGH REFORESTATION―AN EXAMPLE OF AFZELIA XYLOCARPA (KRUZ.) CRAIB AND DALBERGIA COCHINCHINENSIS PIERRE IN CAMBODIA

a b

(a), A seven-year old plantation of Afzelia xylocarpa (arrows) inter-planted with Acacia mangium on a former Imperata grassland in Siem Reap province, Cambodia (2009). (b), Suckers of Dalbergia cochinchinensis (arrows) in a secondary forest in Siem Reap province (b, photo courtesy of Cambodia Tree Seed Project, 2004).

Murdoch University 2011 51

Chapter 4: Conservation of high-value timber species in Cambodia

4.1 Introduction

The importance of reforestation in Southeast Asia was briefly reviewed in Chapter 1. In this chapter, the use of threatened rosewood species in Cambodia is considered as a case study. Globally, more than 8 000 of an estimated total of 100 000 tree species are being threatened to varying degrees, and almost a thousand species are critically endangered (Oldfield et al., 1998). The number of threatened tree species is increasing (Oldfield, 2008). Often the most highly valued timber trees are the most threatened species in their native habitats (Theilade et al., 2005). Examples include Afzelia xylocarpa (Kruz.) Craib and Dalbergia spp. of Southeast Asia (Gillett and Sinovas, 2008). Afzelia xylocarpa and D. cochinchinensis Pierre are listed in the IUCN Red List as endangered (EN A1 cd) and vulnerable (VU A1 cd), respectively (IUCN, 2009). Neither A. xylocarpa nor D. cochinchinensis are CITES listed, although D. cochinchinensis has been listed as meeting CITES Appendix 2 criteria (UNEP- WCMC, 2007 ). At the national level, both species are listed as top priority reflecting their highly threatened status and need for immediate conservation measures (FA and CTSP, 2003b; FLD et al., 2006). Currently, trade in many of the high-value timber species is mainly based on the management of natural forests (Quesada and Stoner, 2004). Consequently, they are much sought after and have become rare.

Efforts have been made to conserve high-value timber species in situ or ex situ (e.g., Newton et al., 2003). In Southeast Asia, there are 2 656 protected areas covering 759 788 km2 or 16.39% of the land area (Chape et al., 2003). However, only 12% of the world’s threatened tree species are found in protected areas (IUCN, 1999). Moreover, protected areas are not always safe locations for protecting high value timber species since forest encroachment and illegal logging occur in some conservation areas (Newton et al., 2003; Kometter et al., 2004; WWF, 2004). A recent study has shown that protected areas in South and Southeast Asia have experienced high rates of deforestation during the last 20 years compared to other regions of the world (DeFries et al., 2005). Currently, 31 plant species in Cambodia are listed in the IUCN Red List as critically endangered (9), endangered (13) and vulnerable (9) (IUCN, 2009) although 21% of its total land area has been designated for conservation purposes (ICEM, 2003). Thus, in situ conservation of tropical forest genetic resources may not be effective if deforestation continues (Koskela and Amaral, 2002).

Increased use of threatened high-value timber species in reforestation programmes is

Murdoch University 2011 52

Chapter 4: Conservation of high-value timber species in Cambodia recognized as an option to protect them from extinction (Lillesø et al., 2002; Tamrakar, 2003). This not only results in the multiplication and distribution of threatened germplasm, and mitigates genetic erosion, but also minimizes pressure on their protected natural habitats (Lillesø et al., 2002). Worldwide, only 8% of threatened tree species are cultivated (IUCN, 1999). Planting is limited because of problems associated with slow growth (Newton et al., 2003) and limited silvicultural knowledge on plantation establishment and management (FAO, 2005; Devall and Smith, 2007). However, conservation through domestication and tree breeding programmes has been undertaken for a small number of high-value timber species of Southeast Asia, such as Chukrasia tabularis A. Juss. and Chukrasia velutina (M. Roemer) C. DC., in northern Australia (Pinyopusarerk and Kalinganire, 2003; Gunn et al., 2006), China, Lao PDR, Malaysia, Sri Lanka, Thailand and Vietnam (Pinyopusarerk and Kalinganire, 2003). The objective of this Chapter is to examine the current status of A. xylocarpa and D. cochinchinensis in their natural habitats, as well as trade and conservation measures applied in Cambodia. Opportunities for promoting reforestation that address conservation, economic development and poverty reduction are explored. Recommendations are made for improving growth and survival under plantation conditions.

4.2 Commercial status of Afzelia xylocarpa and Dalbergia cochinchinensis

4.2.1 Commercial attributes of the species

Worldwide, the genus Dalbergia is composed of about 100 species distributed throughout the tropical and subtropical regions (Soerianegara and Lemmens, 1994). However, Rosketko and Westley (1994) estimated that the number of species lies between 100 and 300, with the majority being trees, and the remainder shrubs or lianas. The majority of Dalbergia spp. (72) are threatened (Oldfield et al., 1998). There are 70 Dalbergia spp. in Asia; at least seven of them are high-value timber trees (Soerianegara and Lemmens, 1994). In Asia, a few species of Dalbergia are used in tree planting programmes including Dalbergia cochinchinensis (FA, 2007), Dalbergia latifolia Roxb. and Dalbergia sissoo Roxb. ex DC. (Soerianegara and Lemmens, 1994). The former is planted in Indonesia and D. sissoo is widely planted in South Asia (Dhakal et al., 2005a). Pakistan alone has more than 100 000 ha of irrigated plantations of D. sissoo (Anon., 1979). The genus Afzelia includes about 15 species, 10 of which are native to Africa, and Afzelia xylocarpa which is the only species native to mainland Southeast Asia

Murdoch University 2011 53

Chapter 4: Conservation of high-value timber species in Cambodia

(Soerianegara and Lemmens, 1994). The characteristics of A. xylocarpa and D. cochinchinensis are summarized in Table 4.1.

4.2.2 Utilization and trade

In Cambodia, timber species are classified into four categories based on wood quality and uses: luxury class and classes 1-3 (MAFF, 1986). The luxury class comprises 14 species, and harvesting of these species is subject to special permission from the Ministry of Agriculture, Forestry and Fisheries (MAFF, 1986). Two of the high value timber species from the luxury group, Afzelia xylocarpa (local name Beng) and Dalbergia cochinchinensis (local name Krangoung), have been promoted in reforestation programmes by the Forestry Administration (FA, 2007). These two species are known in the international trade as rosewoods (Schafer, 1957).

These two rosewoods are well recognized as high-value timber species in Indochina (Anon., 1979; MAFF, 1986; Soerianegara and Lemmens, 1994; Tri and Khanh, 1996; Gillett and Sinovas, 2008). Unfortunately, there is scant information on timber production and revenue. From 1930-1932, during the French colonial period, Cambodia produced more than 46 000 m3 of luxury woods (Cleary, 2005). From 1985-1988, Thailand had the capacity to produce about 30 000 m3 annually of A. xylocarpa for domestic uses (Soerianegara and Lemmens, 1994). A price survey carried out in 2008 with six furniture manufacturers and retailers in Phnom Penh, revealed that sawn wood (usually in the form of rough sawn timber planks or rough cant saw logs) of D. cochinchinensis was much more expensive than A. xylocarpa (Table 4.2). Sawn wood, harvested from natural forests, brought to the city is normally cut to a length of 1.5 to 2 m (Figure 4.1d), and the price varies greatly with width or diameter, which ranges from 15 to > 30 cm for D. cochinchinensis and 30 to > 80 cm for A. xylocarpa. Dalbergia cochinchinensis from Cambodia is reported to be sold in Victoria, Canada at about US$ 20 000 m-3 for “special select grade” and US$ 14 000 m-3 for “grade A”; the woods being used in musical instrument manufacturing and other fine woodworking, turning or carving (Carmichael, 2008). The VietNamNet Bridge (2007) reported that D. cochinchinensis was sold by the kilogram in Vietnam: the Chinese traders bought the wood at US$ 80 kg-1 (about US$ 76 000 m-3) while the Vietnamese traders paid US$ 43 to 62 kg-1 (about US$ 40 000 to 59 000 m-3). However, this price might be for the finished products.

Murdoch University 2011 54

Chapter 4: Conservation of high-value timber species in Cambodia

Table 4. 1 Species characteristics, habitat requirement and uses for Afzelia xylocarpa and Dalbergia cochinchinensis

Characteristic Description Reference Biological Large deciduous trees with a diameter up to 1 m for D. Keating and cochinchinensis and 2 m for A. xylocarpa and 30 m height. Bolza, 1982; Vũ, Slow-growing with long life, A. xylocarpa can live more than 1996; Jøker, 250 years. Seed number: 150 and 35 000 seeds kg-1, 2000a, 2000b; respectively. Seeds are orthodox which allows storage at room Baker et al., 2005 temperature for several years.

Ecological Natural distribution from Myanmar through Thailand, Lao Soerianegara and PDR, and Cambodia to southern Vietnam, mostly on well- Lemmens, 1994; drained sites in mixed deciduous lowland forests or dry Vũ, 1996; Jøker, evergreen forests. They are light demanding, and drought 2000a, 2000b; FA tolerant. Both species are deciduous in the dry season. The seed and CTSP, 2003b; collection season is November-December. Dalbergia Stibig et al., cochinchinensis seeds are dispersed by wind and A. xylocarpa 2004; Kaewkrom by wild animals. Both species have ability to regenerate et al., 2005; naturally. FORRU, 2006; Koonkhunthod et al., 2007

Habitat Occur at an altitude up to 900 m with average annual rainfall Soerianegara and requirement from 1 200-1 650 mm and a dry season of five to six months. Lemmens, 1994; They occur on a variety of soils, but A. xylocarpa prefers well- Dy Phon, 2000; drained clayey, loamy or lateritic soils while D. Jøker, 2000a, cochinchinensis favours deep sandy soil. In Cambodia, the 2000b; FA and species occur in many provinces with forest cover, especially CTSP, 2003b those in the north and northeast of the country.

Use The species are famous for their superb quality woods, which Keating and are red- or flame-coloured with prominent dark brown or black Bolza, 1982; streaks, hard and durable. They are predominantly used in the Soerianegara and luxury furniture industry (Figure 4.1a - c), musical instruments, Lemmens, 1994; sporting equipment, fine-art handicrafts, interior decorations, Vũ, 1996; Dy doors, windows and stairs. Bark of A. xylocarpa is used for Phon, 2000; tanning and veterinary medicine for curing some diseases of Prachaiyo, 2000; domestic animals. Stem of D. cochinchinensis is reported to be Jøker, 2000a, boiled for curing syphilis and anti-tumour and blood stasis. The 2000b; Palasuwan woods can be an excellence source for fuelwood as they have et al., 2005; high calorific values of 4 716 and 5 112 cal.g-1 for A. xylocarpa Delang, 2007; and D. cochinchinensis, respectively. FA, 2007

Murdoch University 2011 55

Chapter 4: Conservation of high-value timber species in Cambodia

a b

c d

Figure 4.1 (a) - (c), Some examples of furniture/souvenir products made from Afzelia xylocarpa and Dalbergia cochinchinensis on display in shops in Cambodia. (a), A souvenir shop in Phnom Penh where ca. 90% of the products are made of A. xylocarpa (photo courtesy of Moy Rotha, 2006). (b), A bed made of A. xylocarpa in a furniture shop in Pailin province (photo courtesy of So Than, 2006). (c), Chairs made of A. xylocarpa (left) and D. cochinchinensis (right) in a furniture shop in Phnom Penh (2009). (d), A pile of rough sawn wood of D. cochinchinensis in a furniture shop (photo courtesy of So Than, 2004).

Table 4.2 Estimated price of sawn wood in Cambodia and the region in 2008

Species Estimated price Location Reference (US$ m-3) Afzelia xylocarpa 1 000-1 500 Cambodia Thea So, unpublished data 2008 715a Thailand Soerianegara and Lemmens, 1994 Dalbergia cochinchinensis 1 900-3 900 Cambodia Gillett and Sinovas, 2008 a Price in 1988.

4.3 Conservation of the species

4.3.1 Threats and conservation

The rapid economic growth in the last two decades, coupled with the increase in population, has resulted in increased demand of forested lands for development and for forest products. Illegal cutting, over exploitation and habitat loss, through conversion of

Murdoch University 2011 56

Chapter 4: Conservation of high-value timber species in Cambodia natural forests to other land uses, are the main threats causing deforestation and forest degradation in Cambodia (Strange et al., 2007; Gillett and Sinovas, 2008).

Illegal trade in the two species has been reported in many parts of Cambodia (Global Witness, 2004; FA, 2008; Gillett and Sinovas, 2008) even though harvesting is prohibited by law (MAFF, 1986). Illegal activities have been driven by high demand and high timber prices in the domestic and international markets (Gillett and Sinovas, 2008). A price survey revealed that the price of sawn timber of the two species in the capital city has increased about 40% from 2002 till 2008, and it is expected that it will keep growing in the years to come, as demand remains strong. Currently, there is a lack of information on the scale of illegal harvests and the remaining tree populations in natural habitats in Cambodia. Therefore, it is not possible to estimate the extent of impact caused by this illegal activity. A recent survey in the lowland forests of Stoeung Treng province, northern Cambodia, reported that log poaching has led to local extinction in some parts of their natural habitats (Francke et al., 2007).

Two approaches to forest genetic conservation, in situ and ex situ conservation, are being implemented for a range of priority tree species in Cambodia (FLD et al., 2006). This work comprises the five most threatened priority tree species A. xylocarpa, D. cochinchinensis, Dalbergia oliveri Gamble ex Prain, Pterocarpus macrocarpus Kurz and Fagraea fragrans Roxb. (FLD et al., 2006). In situ conservation plays a leading role in conservation of forest genetic resources in Cambodia, as 21% of the total land area has been put under various forms of protection, namely national parks, wildlife sanctuaries, protected forests, protected landscapes, and multiple use areas (ICEM, 2003). However, little is known about the population size or status of the two rosewood species in these protected areas. In addition to the protected areas, a network of 691 ha of natural forests has been established for conservation of genetic resources of the national priority tree species and as a source of planting materials for tree planting programmes (FLD et al., 2006). Of this area, 107 ha were designated for conservation (genebanks) of the two rosewoods (FLD et al., 2006). Finally, a one-hectare ex situ conservation stand of each species has been established in Cambodia. In neighbouring countries, Thailand, Lao PDR and Vietnam, the two rosewoods are of top priority, and in situ or ex situ conservation measures have been reported (FAO et al., 2004; Luoma-aho et al., 2004). Ex

Murdoch University 2011 57

Chapter 4: Conservation of high-value timber species in Cambodia situ conservation stands of A. xylocarpa and D. cochinchinensis covering 40 and 56 ha, respectively, exist in Thailand (FORGENMAP, 2002).

4.3.2 Tree domestication

Domestication is the process of bringing a wild plant under managed cultivation by refining its genetics to maximize human benefit (Thomson et al., 2001). Domestication of A. xylocarpa and D. cochinchinensis has emerged in Cambodia at the time of rapid decline of their natural populations and high demand for their timber. The concept of conservation of threatened high-value timber species through use stemmed from the National Forest Gene Conservation Strategy, which was developed in 2003 (FA and CTSP, 2003b). The idea is that increased planting of threatened high-value timber species in reforestation programmes will reduce pressure on their natural populations, and at the same time provide income from high value timber and thereby contribute towards economic development and social welfare (Kjær et al., 2001). The strategy encourages using planting materials (mainly from seed) of the priority tree species (mainly high-value timber species) in tree planting activities. To date, work on tree domestication carried out and completed by the Cambodia Tree Seed Project (CTSP) in collaboration with the Forestry Administration (FA) and its partners (FA and CTSP, 2003b; FLD et al., 2006; CTSP home page: www.treeseedfa.org) include: 1. Ecological zonation. Ten ecological zones that exhibit uniform ecological conditions have been defined. Afzelia xylocarpa was found in seven and D. cochinchinensis in six zones. The zonation serves as a practical guideline for movement of planting materials. It supports domestication by reducing the risk of economic loss due to planting of mal-adapted trees. Limited transfer is also likely to support conservation effort by sustaining patterns of geographic genetic differentiation including those generated by local adaptation (FAO et al., 2004). 2. Tree improvement. Two and three seed sources were established for A. xylocarpa and D. cochinchinensis, respectively (Figure 4.2). Good phenotypic trees were selected and marked in each seed source for seed collection (Table 4.3). In addition, seed production areas/ex situ conservation stands of more than one ha each were established in the Kbal Chhay Watershed Protected area in 2004. The two species were also included in a species screening trial established in the same year. The

Murdoch University 2011 58

Chapter 4: Conservation of high-value timber species in Cambodia

N

LEGEND Seed sources of Dalbergia cochinchinensis Seed sources of Afzelia xylocarpa Provincial town

Forest cover map 2006 Evergreen forest Deciduous forest Other forest Non forested area

Figure 4.2 Seed sources of Afzelia xylocarpa and Dalbergia cochinchinensis in Cambodia.

activities allow selection for improved timber quality, but also good health and vigour. The latter supports continued adaption to local conditions through natural selection, an important part of genetic conservation (Eriksson et al., 1993). 3. Seed procurement and nursery operations. These are now well defined. For both species, seeds are collected by cutting short branches bearing pods in December or pods are taken from the ground (A. xylocarpa). Pods are dried in full or partial sunlight for a few days to facilitate seed extraction. Seed of A. xylocarpa are pretreated by scarifying the seed coat followed by soaking in water overnight. Seedlings emerge after 7 to 10 days with high germination percentage, ca. 90%. For D. cochinchinensis, seeds are pre-germinated by soaking in water overnight. Seeds germinate at about seven days with germination percentage ca. 70 to 80%. They are ready for transplanting into containers at about two weeks. Provision of seed sources (germplasm availability) and guidelines on seed handling and nursery techniques are pre-requisites for integrated use and conservation, as they prevent technical and practical obstacles in the implementation of large-scale plantings.

Murdoch University 2011 59

Chapter 4: Conservation of high-value timber species in Cambodia

Table 4.3 Identified seed sources of Afzelia xylocarpa and Dalbergia cochinchinensis in natural forests in Cambodia

No. Species Location Area Number of Reference (province) (ha) mother trees 1 Afzelia xylocarpa Rottanak Kiri 18 marked27 FLD et al., 2006 2 Afzelia xylocarpa Rottanak Kiri 20 26 FLD et al., 2006

3 Dalbergia cochinchinensis Koh Kong 19 26 FLD et al., 2006

4 Dalbergia cochinchinensis Pursat 96 45 FLD et al., 2006

5 Dalbergia cochinchinensis Siem Reap 50 121 FLD et al., 2006

4.4 Plantation establishment and community restoration

4.4.1 Suitability for reforestation

Dalbergia cochinchinensis has long been assumed to be a slow-growing species when planted. However, under favourable site conditions it can grow quite fast (Eastman and Danborg, 1995). It is estimated that over a 50-year rotation, the volume production could reach 400 m3 ha-1 (FAO et al., 2004). Being native species, they are adapted to a variety of local climates and soil conditions. As a host for N2-fixing bacteria (Brewbaker, 1990) (Chapters 5 to 7), D. cochinchinensis can help to improve soil fertility and to increase host survival under marginal conditions typical of reforestation sites. The two species are easily propagated using a variety of planting materials, such as seeds, nursery-raised seedlings, and cuttings (Jøker, 2000a, 2000b). Dalbergia has a strong ability to regenerate from stumps after the tree has been cut and produces suckers from roots that remain attached to the stump. A field trial on direct seeding in Thailand (Chapter 5) has shown promising results provided seedlings are kept free from weeds for 14 months.

4.4.2 Plantations in Cambodia and the region

In 2002 and 2004, FA planted 550 ha of A. xylocarpa and D. cochinchinensis in three provinces, Mundul Kiri, Preah Sihanouk and Siem Reap (FA, 2007). To the best of my knowledge, this is the first large-scale plantation of the two species ever established in Cambodia and one of the few in the region (Table 4.4). The two rosewoods have also been used in small-scale plantations established by a number of public institutions, non-

Murdoch University 2011 60

Chapter 4: Conservation of high-value timber species in Cambodia governmental organizations and communities throughout the country and through the national tree planting ceremony (Arbor Day) (FA, 2007) (Figure 4.3c). The total area of these small-scale plantations, as well as the success of the plantings, are not known.

A few small-scale plantations of A. xylocarpa, mixed with Dalbergia spp. and Tectona grandis L. f., exist in Thailand (Soerianegara and Lemmens, 1994). In Lao PDR, there are reports that A. xylocarpa has been planted by various stakeholders, such as farmers, the public and private sectors in many parts of country (Saignaphet, 1995; Sati, 1995; Soumphonphakdy, 1995). In Savannakhet (Lao PDR), A. xylocarpa was the second most popular species after Eucalyptus in terms of area planted (Saignaphet, 1995). Even though plantings of A. xylocarpa and D. cochinchinensis are known within the region there has been no systematic effort to evaluate provenances or determine survival and growth rates for the two species.

Table 4.4 Total area under plantations of Afzelia xylocarpa and Dalbergia cochinchinensis in Cambodia and the region

Country Area (ha) Reference Afzelia xylocarpa Dalbergia cochinchinensis Cambodia 450 100 FA, 2007 Lao PDR 135 200 Phongoudome and Mounlamai, 2004 Vietnam 2 467 n. a. Nghia, 2004 n.a. Data not available.

4.4.3 Plantation technologies: a case study in the Kbal Chhay Watershed area

The Kbal Chhay Watershed Protected area (Figure 4.2) covers an area about 6 300 ha, located in Preah Sihanouk province. Detailed information on location, soil and climate is presented in Chapter 7, Section 7.2.1. Until the late 1990s, about 40% of the area was covered by luxuriant Imperata grass (McDonald, 2003). Establishment of mixed species plantations on denuded lands for watershed protection is one of the objectives of the long-term management plan of the area. In 2001, reforestation of denuded areas commenced with Acacia species (usually Acacia mangium A. Cunn. ex Benth. and Acacia auriculiformis A. Cunn. ex Benth.). In 2004, a total area of 250 ha of mixed species plantations of A. xylocarpa, D. cochinchinensis, and two other indigenous species, Dipterocarpus alatus and Hopea odorata, was established under the Acacia

Murdoch University 2011 61

Chapter 4: Conservation of high-value timber species in Cambodia canopies with a density of 500 seedlings ha-1 (FA, 2007). The indigenous species were planted between rows of Acacia at the second planting season without opening gaps.

4.4.3.1 Use of nurse trees

Planting indigenous trees under the canopy of a fast-growing species such as Acacia is widely practiced not only in Kbal Chhay, but also in other parts of Cambodia (FA, 2007). This is because many reforestation sites are generally infested by weeds and grasses, in which many slow-growing indigenous species are not able to compete for survival in the early stage of development. In addition, the grasses are a fire hazard. Acacia plantations are usually established with a density ranging from 1 666 to 2 083 stems ha-1 (FA, 2007). At canopy closure, the grasses are eliminated, and usually by the second or third planting season, the stands are thinned to provide gaps for underplanting with selected indigenous species (FA and FLD, 2007). Once the indigenous species has been introduced, monitoring of the Acacia canopy becomes important. However, this has not been taken into consideration yet in Kbal Chhay and in many other reforestation sites.

For shade intolerant species, like the two rosewoods, early removal of the canopy appears to be essential. A systematic thinning (25% stem removal) of Acacia plantations is recommended after the second planting season before underplanting with indigenous species (FA and FLD, 2007). Likewise, the canopy of nurse trees should be thinned in subsequent years. In Kbal Chhay, thinning of Acacia was delayed until age five years, and many D. cochinchinensis suffered from excessive competition and lack of sunlight. Today, many trees are stunted while others grow vertically towards the canopy layer resulting in slender stems, which are susceptible to wind break after thinning the Acacia (Figure 4.3a). Partial removal of the Acacia canopy, by lines or gaps, is needed in order for the underplanted trees to receive sufficient sunlight for their full development. In Indonesia, the growth of Anisoptera marginata Korth., an indigenous species, was doubled by early opening of the A. mangium canopy compared to that in the unopened stands (Otsamo, 1998b).

Apart from Acacia, other pioneer indigenous species have been used as a means to eliminate grass in the region. In Vietnam, for example, a method known as “accelerated pioneer-climax series” was designed to rapidly rehabilitate grassland, produce firewood

Murdoch University 2011 62

Chapter 4: Conservation of high-value timber species in Cambodia and small timber, and improve site conditions for the climax species (So, 2000). Several versions of this method have been applied. In the Imperata grasslands, pioneer species are planted at high stocking (up to 3 300 trees ha-1) and the trees are then thinned at year four to provide space for underplanting the climax species. The pioneer trees are thinned again in year eight and remaining trees removed in year 12. Sometimes, a mixed plantation is established simultaneously by alternating pioneer rows (such as Indigofera teysmanii Miq., Gliricidia sepium (Jacq.) Kunth ex Walp.) with rows of intermediate (Xylia dolabriformis Benth., D. cochinchinensis) or climax (Anisoptera cochinchinensis Pierre, D. alatus, H. odorata) species. The pioneers are thinned every two years, while the intermediate species are thinned over six years.

Other than acacias, that have been widely used to reclaim grasslands in Cambodia, there are a few species that are claimed to be as effective as acacias, such as Gmelina arborea Roxb. ex Sm. (Otsamo et al., 1997; Friday et al., 1999), Vitex pubescens Vahl., and G. cepium (Friday et al., 1999). These species grow fast, possess broad/dense canopies that suppress weeds, have thick fire-resistant bark, and sprout after fire (Friday et al., 1999). Other pioneer indigenous species such as Rhodomyrtus tomentosa (Aiton) Hassk. (local name Puoch), Knema corticosa Lour. (Sma Krabey), and Garcinia ferrea Pierre. (Prush) were suggested by McDonald (2003) as a substitute to acacias where biodiversity conservation is the main objective of reforestation. These species, however, have not yet been tested. A trial evaluating the effectiveness of a range of nurse species is required to broaden the range of potential nurse trees in Cambodia.

4.4.3.2 Plantation in open areas

While planting indigenous timber species under Acacia canopies is commonly adopted in Imperata grasslands, establishment of plantations in open areas is also practiced where weeds, especially grasses, are less dominant or where a small-scale plantation can be established and intensive weed control is part of the management plan. However, even with intensive weeding, avoiding fire damage has proved difficult in areas heavily infested with grasses. As an example, in the dry season of 2004, the second year after planting, a seed production area of A. xylocarpa was completely burnt by a wild fire. However, about 90% of seedlings coppiced from their bases at the onset of the wet season of the same year (FA and CTSP, 2005). To minimize the fire risk in the research

Murdoch University 2011 63

Chapter 4: Conservation of high-value timber species in Cambodia

a b

c d e

Figure 4.3 (a), (b) and (c), Plantations of Dalbergia cochinchinensis. (a), Dalbergia cochinchinensis planted under the canopy of Acacia auriculiformis without thinning since planting in 2004 in Kbal Chhay Watershed Protected area; (b), Dalbergia plot in a species selection trial established in an open area in the same location and year as (a). Note the Imperata grass is still present; (c), A small- scale plantation of D. cochinchinensis (seven years old) in an open area in Siem Reap province; (d), Enrichment planting of Afzelia xylocarpa (eight years old) in a degraded forest in Siem Reap province. Note the trees in (c) and (d) drop all their leaves in the dry season (January 2009) and; (e), A mother tree of D. cochinchinensis in an identified seed source in natural forest, Siem Reap province (e, photo courtesy of Cambodia Tree Seed Project).

Murdoch University 2011 64

Chapter 4: Conservation of high-value timber species in Cambodia plots and promote early growth, application of fertilizer (30 g of NPK 15:15:15 seedling-1 year-1) in combination with intensive weeding during the first four years was practiced until the trees became taller than the grass and were strong enough to withstand fire (Pers. comm., Moy Rotha, 2008). Application of inorganic fertilizer is generally recommended for some high-value timber species, for example 100 g of NPK 15:15:15 was given to each Chukrasia tabularis seedling in the first year (Kalinganire and Pinyopusarerk, 2000). Testing for optimum rates of inorganic fertilizer to boost the early growth of the two rosewoods in different soil types in Cambodia should be encouraged. Early growth of the two species from plantations in Cambodia as well as the region is shown in Table 4.5.

Table 4.5 Average early growth of Afzelia xylocarpa and Dalbergia cochinchinensis under plantation conditions in Cambodia and the region

Species Planting Age DBH Height Province, country Reference method (year) (cm) (m) Afzelia n.a. 3 n.a. 1.4 n.a. Soerianegara and xylocarpa Lemmens, 1994 En1 3 1.12 0.7 Borikhamsai, Lao PDR Lee, 2005 En 8 6.8 7.7 Siem Reap, Cambodia Thea So, unpublished data Dalbergia En 3 0.92 0.9 Borikhamsai, Lao PDR Lee, 2005 cochinchinensis Mo 5 6.7 5.7 Preah Sihanouk, Thea So, Cambodia unpublished data Mo 7 11.2 8.5 Siem Reap, Cambodia Thea So, unpublished data Mo 12 10 15.4 Sakaerat, Thailand Kamo et al., 2002 Mo 38 29 21.8 Dong Nai, Vietnam Nghia, 2000

1 En: enrichment planting in degraded forests. Mo: mono-species plantation in open area. 2 Root collar diameter. n.a: data not available.

A species selection trial including 19 indigenous (such as D. alatus, H. odorata, P. macrocarpus, and Tarrietia javanica) tree species was conducted in Kabal Chhay Watershed Protected area. A statistically sound comparison of growth and survival of A. xylocarpa and D. cochinchinensis is wanting. However, visual assessment of these two parameters in the species selection trial reveals that D. cochinchinensis performs better than the other 19 indigenous species during the first five year period. The trial shows

Murdoch University 2011 65

Chapter 4: Conservation of high-value timber species in Cambodia that A. xylocarpa and D. cochinchinensis do not need to be underplanted to Acacia unless the Imperata grass is of high risk. Results from a three-year old trial established in a logged-over forest in Lao PDR support this assumption as seedlings of both species planted in open sun (gap 14 × 14 m) showed much higher survival and growth rates than those in semi-shaded lines (2 m wide) (Lee, 2005).

Early plantation trials for A. xylocarpa and D. cochinchinensis pointed to problems with weed control, fire hazard and monitoring of nurse trees. It is hoped that the established plantations/trials will provide future results on appropriate management, growth increment, and costs necessary for determining the profitability of such plantings.

4.4.4 Silvicultural treatment of natural (degraded) forests

Assisted natural regeneration (ANR) is practiced in degraded forests where there are enough naturally established seedlings. The Forest Restoration Research Unit (FORRU) (2006) estimated that in a site with a density of seedlings and live stumps of mixed species over 1 562 ha-1, ANR alone can be a successful technique in rehabilitation of degraded forests provided forest fire and domestic animals are controlled. Some techniques for enhancing survival and growth of the naturally regenerated seedlings include removal of weeds close to the seedlings, removing overshadowing trees/branches, and supporting bent seedlings with sticks. Application of fertilizer could be practiced to stimulate early growth. Assisted natural regeneration has not been widely adopted in Cambodia, but prevention of forest fire is commonly practiced in many community forests.

Where there is a shortage of natural regeneration of A. xylocarpa and D. cochinchinensis, enrichment planting can be implemented in gaps or lines of degraded and/or deciduous forests. This technique has been tested with great success by local Forestry Administrations in Siem Reap province where A. xylocarpa seedlings were planted in lines in a degraded forest of 30 ha (Pers. comm., Suo Hai, 2008) (Figure 4.3d). ANR and enrichment planting may be most appropriate in community forests, the majority of which are located in degraded forests (IFSR, 2004) and where forest restoration to diversify forest products is the main management objective.

Murdoch University 2011 66

Chapter 4: Conservation of high-value timber species in Cambodia

4.4.5 Relationships between tree growth, tree form, site and stand management

The form of the tree (length of branch-free bole and bole straightness) is important for commercial utilisation. However, D. cochinchinensis is reputed to have poor stem form (Jøker, 2000b) (Figure 4.3a - c). Crookedness of stems has been reported in D. sissoo and D. latifolia (Soerianegara and Lemmens, 1994). Improvement of stem form of a timber tree species is possible, for example, Gunn et al. (2006) reported improved form of Chukrasia through provenance selection.

Knowledge on site requirement and genotype-environment interactions of A. xylocarpa and D. cochinchinensis at different site conditions is lacking. Consequently, planting of these species for economic purposes has not been widely practiced although the high timber price is well recognized. The long-term rotation, 50 years for D. cochinchinensis (FAO et al., 2004), for example, renders the species less attractive than their exotic counterparts. Many tree planters need to generate revenue within a short period of time. Thus, acacias and eucalypts are preferred, although the timber price and per hectare revenue of a high-value tree exceeds the revenue from fast-growing eucalypts or acacias (FAO et al., 2004). This is a common problem for many indigenous tree species (Tsai and Faridah-Hanum, 1992). A solution may be the reduction of the minimum log diameter at harvest, by shortening the rotation of the timber harvest to, for example, 30 years or less for D. cochinchinensis. Reduction in rotation length of high-value hardwood species such as teak has gained in popularity in Thailand and Malaysia, where a commercial harvest is anticipated within 15 to 20 years (Lee, 1999; Kijkar, 2001). Market surveys are needed in order to estimate marketability and prices for smaller sized timber of A. xylocarpa and D. cochinchinensis. Reduction in rotation length needs to take into account the higher proportion of much lower value sapwood compared to high-value heartwood that will be produced. It will be necessary to establish the pattern of heartwood production as affected by genetics, silviculture and rotation length. Proposed frameworks for plantation establishment in Cambodia are outlined in Table 4.6.

Murdoch University 2011 67

Chapter 4: Conservation of high-value timber species in Cambodia

Table 4.6 Proposed methods for reforestation using Afzelia xylocarpa and Dalbergia cochinchinensis

Method Description Potential stakeholder Reforestation In areas highly infested with weeds and grasses such as Imperata, first sState, of denuded reclaim the land with pioneer tree species, then underplant rosewoods private lands after about two to three years when the grasses are eliminated by investors, opening gaps of acacia canopy to ca. 25%. More gap opening is communities, required as the rosewoods grow, probably every two years. Where conservation weeds and Imperata grass are less competitive, the rosewoods do not organizations require nurse trees. ANR Assisted natural regeneration (ANR) can be practiced in a degraded State, forest where there are enough naturally established seedlings. communities Enrichment This method can be applied in gaps of logged-over forests and gaps or State, planting lines of degraded and/or deciduous forests. Line planting in logged-over communities, evergreen forests may not be appropriate since there remain a high conservation density of tall trees which could easily out-compete the underplanted organizations trees in a short period. Agroforestry For local livelihood development or biodiversity conservation, mixed State, on large species plantations should be adopted. Many species of Dalbergia are communities plantations well known as agroforestry species such as D. sissoo. Diversification and private of species with different rotations to maximize the harvests during the investors rotation of rosewoods should be of highest consideration. For example, mixture by row(s) with fast- or medium-growing species would produce additional harvests of fuelwood, poles/pillars or wood chips before harvesting the rosewoods. At canopy closure, some understorey shade- tolerant plants such as wild yams and medicinal plants can be introduced. Shade/nurse These trees can be planted to provide shade for agricultural crops like Farmers, trees coffee. D. cochinchinensis is more suitable for this purpose as it has communities, narrow and sparse crowns. Although A. xylocarpa casts heavy shade private due to their large canopies at maturity, they can be pruned to adjust the investors shade. Mature plantations can be used as nurse/support trees for rattan production which can be inter-planted between the tree rows. Living Afzelia xylocarpa and D. cochinchinensis can be planted as one or more Farmers, fences rows along the borders of home gardens, pagoda compounds or school monks yards. Paddy field Trees can be planted on mounds, bunds of paddy fields and banks of Farmers, trees ponds or canals. A legume tree grown on the bund of a rice field can communities improve soil fertility, but it may decrease rice production in areas under the shade close to the tree base (Sae-Lee et al., 1992); regular pruning is required at the mature stage.

Murdoch University 2011 68

Chapter 4: Conservation of high-value timber species in Cambodia

4.5 Challenges and opportunities

4.5.1 Importance of reforestation to the wellbeing of local communities

In Cambodia, local livelihoods are traditionally dependent on the forests as a source for subsistence and income (IFSR, 2004; Top et al., 2009). For communities living in or adjacent to the forests, their dependence on forests for livelihood is even higher. For them, forests are the only source of income, providing food, medicine and construction materials (Kim et al., 2008). Thus, there is a need to consider access to a variety of forest products by communities living adjacent to reforestation sites. Maintaining mixed species plantations during the rotation of rosewoods could secure different products useful to communities from the early stage of establishment until the end of the rotation (Tsai and Faridah-Hanum, 1992). Hence, pioneer and/or intermediate species should be selected based on the socio-economic value of their products to local communities in addition to standard plantation criteria. Afzelia xylocarpa and D. cochinchinensis are suited to many forms of tree planting that could benefit local communities (Table 4.6). Promotion of species beneficial to local livelihoods and economies will encourage participation in the plantation enterprise by local communities. If trials are promising, farmers may desire to grow D. cochinchinensis and A. xylocarpa on a small scale. If so, the best routes through which to move improved germplasm to farmers and large-scale growers should be determined in order to improve uptake and social benefit.

4.5.2 Seed supply strategy

The availability of genetically improved seeds is one of the principal barriers affecting the productivity of forest plantations (Varmola and Carle, 2002). There is a demand for seeds of A. xylocarpa and D. cochinchinensis, but it is difficult to estimate the amount required every year due to the lack of a long-term plan for tree planting programmes (FA and CTSP, 2003a). A survey on seed price in 2008 showed that seed of D. cochinchinensis collected from a seed source in Siem Reap was sold at US$150 kg-1 and the local FA managing the seed source had sold about 3 kg of seed (Thea So, unpublished data). Currently, the supply of genetically improved seed remains a challenge for Cambodia in promoting the planting of high-value timber species. This is exacerbated by the limited availability of seed sources and the weakness of the seed

Murdoch University 2011 69

Chapter 4: Conservation of high-value timber species in Cambodia sector. The identified seed sources in natural forests do not represent all ecological zones. In addition, the limited number of marked mother trees may not be able to produce sufficient seed to meet the demand of reforestation programmes. In order to respond to the immediate need of seeds, a seed supply strategy is proposed as follows. Firstly, encourage the use of seed from identified sources as many seed users are not aware of their existence. Secondly, there is a possibility that seed sources exist in natural forests across the country that have not been identified by FA. Seeds can be collected from these sources. However, finding a minimum of 15 mother trees, as required for reforestation practices (e.g., Schmidt, 2000), may be unrealistic. A minimal requirement in seed collection technique has to be followed, such as selecting mother trees with a reasonable phenotype and bulking the seeds from a number of un-related trees (Kjær et al., 2006). Thirdly, movement of seeds between zones with similar ecological conditions, as suggested by the Forest Gene Conservation Strategy, can be practiced (FA and CTSP, 2003b). Plantations of A. xylocarpa and D. cochinchinensis in Kbal Chhay have shown that the two species have adapted well in an ecological zone different from their natural habitats in Rottanak Kiri and Siem Reap province, respectively. Fourthly, procurement of seeds from neighboring countries such as Lao PDR and Thailand is an option. Sourcing of teak seeds from neighboring countries has long been practiced. Care should be taken to select the sources with similar ecological conditions as the intended planting sites.

Conversion of healthy stands in existing young plantations (Table 4.4) into seed production areas, by selective culling of inferior trees, is an option in securing seed supply in the near future. However, due to the lack of data on the number of mother trees used to establish existing stands, there is a risk of using inbred seeds pollinated by related trees. To reduce this risk, seeds should be collected from different sources in the same ecological zone and bulked in equal proportions (Dhakal et al., 2005b). For the long-term strategy, genetically improved seed sources are needed.

Promotion of a functional seed supply system should be undertaken at local and central FA offices. This could be achieved by supporting the local FA offices and communities managing the seed sources through the provision of on-the-job training on seed collection, handling, storage, and distribution. At the central level, a tree seed office is being proposed to meet the challenges in seed supply and tree improvement, but it is

Murdoch University 2011 70

Chapter 4: Conservation of high-value timber species in Cambodia anticipated that full function will take some years to be effective. Temporarily, a staff member with collective experience in the tree seed sector could be assigned to coordinate or assist seed users in procurement of seeds from various sources.

4.5.3 Ecological and biological characteristics―research needs

An overall limiting factor in promoting plantations of indigenous species is the lack of understanding of their ecological and biological features related to survival and growth characteristics under plantation conditions (Tsai and Faridah-Hanum, 1992). This gap has to be filled by promoting research and/or observations on the ecological and biological characteristics of A. xylocarpa and D. cochinchinensis. In an economic sense, it would be useful to document favourable conditions for plantings. Ecological surveys of plantations and natural habitats could provide information on required environmental conditions of the trees and their responses as a result of genotype and environment interactions. Permanent sample plots, established in plantations and wild stands, are needed to facilitate this understanding.

In addition, priority should be given to research aimed at improvement of stem form of A. xylocarpa and D. cochinchinensis as it is a desirable quality for timber production. Stem quality of a timber tree has long been recognized as influenced by silvicultural practices, such as site and site preparation, spacing, thinning, fertilizing and artificial pruning (Ehrenberg, 1970), as well as genes inherited from the mother tree (e.g., Vidakovic and Ahsan 1970). Field trials exploring suitable spacing, thinning, and shading regimes of nurse trees should be established in a number of ecological zones with major tree planting programmes. Based on results of such studies, guidelines can be prepared in the future on suitable site conditions and silvicultural practices.

4.5.4 Development of a tree improvement programme

In view of: a) the early promise of D. cochinchinensis and A. xylocarpa exhibited at Kbal Chhay, b) the desirability of conserving the germplasm present in existing seed sources, and c) the unavailability of genetically improved seed, it is recommended to initiate a tree domestication programme for A. xylocarpa and D. cochinchinensis. A provisional strategy to guide the process of tree improvement encompasses two options based on resources and local capacity.

Murdoch University 2011 71

Chapter 4: Conservation of high-value timber species in Cambodia

The first option for consideration is the Breeding Seed Orchard (BSO), which is a combination of the production of improved seeds and progeny testing (Dhakal et al., 2005b; Kjær et al., 2006). It is a low input, fast and simple method in tree domestication (Dhakal et al., 2005b; Kjær et al., 2006). The method for establishment of the first generation BSO is summarized below (Dhakal et al., 2005b; Kjær et al., 2006). A second generation of BSO can be established using seeds from the first generation. A substantial gain in the second generation is expected through selection. 1. Identify good phenotypic trees from local ecological zones and collect seed from not less than 30 to 40 trees. 2. Produce seedlings and plant them in the defined BSO. Seedlings can be mixed, which results in a Bulked BSO or kept by family in plots, which results in a Family BSO. 3. Based on desirable phenotypic characteristics, select the outstanding individuals or families according to the type of BSO, and then undertake thinning of the inferior individuals or families. In the Family BSO, only one tree is left in a plot. Seed collection is undertaken after the post-thinning flowering. Alternatively, a domestication strategy starting with provenance trials would allow the inclusion of a wide range of populations into the tests. Widely distributed species, like the rosewoods, are expected to demonstrate considerable variation in phenotypic characteristics which is the base population for improvement through selection and breeding. A summary of this method follows (based on Pinyopusarerk and Kalinganire, 2003): 1. Identify good phenotypic trees, organize provenance seed collection from wild stands or plantations from ecological zones (provenances) where the species is known to occur. Then establish provenance trials to assess the phenotypic differences leading to selection of outstanding populations and individuals. At this stage, comparison of provenances from Cambodia and those from the region would be useful. 2. Select and mark the superior provenances and individuals within a provenance, based on fast-growth and stem straightness, for grafting to establish a clone bank. This clone bank provides materials for establishment of clonal seed orchards (CSO) and clone trials. 3. Results from the clone tests are used to undertake thinning of the CSO by removing the poor-performed clones. Collect the seed crop from the post-thinning flowering for use. 4. Undertake provenance-progeny trials using seed from the superior trees selected in

Murdoch University 2011 72

Chapter 4: Conservation of high-value timber species in Cambodia

step 2 by infusion with new families of best ecotype when they are available. Then convert provenance trials into seedling seed orchards, by heavily culling inferior trees. Collect the seed crop from the post-thinning flowering for use.

The strategy recommended is far from complete and should be reviewed as further information is generated from on-going research. There are logistic and economic constraints tied to implementation of the suggested elements, but it is probably worth the effort considering the high value of the two species. Many natural forests, where D. cochinchinensis and A. xylocarpa occur, continue to be lost through deforestation. Considering the scarcity of seed sources and availability of good phenotypic seed trees left in the forests, a specific programme aimed at strengthening in situ genetic resource conservation of remaining populations of the species is recommended as a complement to the suggested domestication strategy.

Efficient domestication requires willing national and international collaboration so that essential resources can be brought to the task: the genetic resources of natural populations, the capacity to establish and manage field trials, and facilities for coordination (Pinyopusarerk and Kalinganire, 2003). In 2010, a national forest programme (NFP) for Cambodia was endorsed after prolonged consultation between the government, international aid agencies, NGOs, and other stakeholders (GoC, 2010). The programme identifies management and conservation of forest genetic resources including establishment of seed sources and development of multi-purpose tree plantations as one of six priority areas within the forestry sector. Thus, the government of Cambodia acknowledges that urgent measures are required to conserve and sustainable use the remaining genetic resources of economically important species. The NFP provides an excellent framework for future collaborative activity within this field.

4.6 Concluding remarks

The opportunities and challenges in promoting reforestation in Cambodia using A. xylocarpa and D. cochinchinensis have been discussed. The suitability of A. xylocarpa and D. cochinchinensis in reforestation programmes combined with attractive timber prices creates an ideal opportunity to promote their establishment in multipurpose plantations for genetic conservation and commercial outcomes. However, silvicultural

Murdoch University 2011 73

Chapter 4: Conservation of high-value timber species in Cambodia practices need to be tested and improved. Further research is needed on the effect of silvicultural practices on productivity and socio-economic benefits to local communities. The harvest of forest products to satisfy local needs for fuelwood, construction materials, and other non-timber forest products during the rotation of rosewoods should be maximized. This would enable plantations to contribute to economic development and poverty reduction, and encourage a wider range of stakeholders to participate in tree planting programmes.

Greater use of A. xylocarpa and D. cochinchinensis in reforestation may help reduce pressure on the resource in the remaining natural forests. Yet, programmes for tree improvement and increased supply of genetically superior seeds are wanting. The proposed domestication strategy for D. cochinchinensis and A. xylocarpa could serve as a model for the future conservation and sustainable use of Cambodia’s high-value timber species. The National Forest Programme of Cambodia provides an excellent framework for future collaborative activity within this field.

Murdoch University 2011 74

Chapter 5: Improving reforestation success of high-value and key forest tree species in Thailand

CHAPTER 5

IMPROVING REFORESTATION SUCCESS OF HIGH-VALUE AND KEY FOREST TREE SPECIES BY DIRECT SEEDING: FIELD TRIALS IN THAILAND

a b

Young seedlings of Xylia xylocarpa (a) and Afzelia xylocarpa (b) two weeks after direct seeding in Sakeaw province, Thailand.

Murdoch University 2011 75

Chapter 5: Improving reforestation success of high-value and key forest tree species in Thailand

5.1 Introduction

In Southeast Asia, many high-value timber species from the legume family, such as Afzelia xylocarpa (Kruz.) Craib and Dalbergia cochinchinensis Pierre (Chapter 4), are slow-growing (Soerianegara and Lemmens, 1994; Vũ, 1996), threatened or endangered (IUCN, 2009) in their natural habitats, and are placed on top of the priority species list of many countries (Cambodia, Lao PDR, Thailand and Vietnam) (FA and CTSP, 2003b; Nghia, 2004; Phongoudome and Mounlamai, 2004; Sumantakul, 2004). The listing indicates the significant role of the species in economic development and conservation value as well as the need for promoting them in reforestation programmes (Luoma-aho et al., 2004). However, apart from a few small-scale plantations managed for conservation purposes (Nghia, 2004; Phongoudome and Mounlamai, 2004; FA, 2007), these species have not been widely used in tree plantings. The slow-growing habit at the early stage of development (Soerianegara and Lemmens, 1994; Lee, 2005) is the main constraint for widespread planting of these species, as the young seedlings are vulnerable to early competition with weeds and damage from forest fire at many planting sites. One of the practical solutions in reforestation using slow-growing species is to plant them under the canopy of fast-growing nurse trees, mainly Acacia mangium and A. auriculiformis, when the grasses have been eliminated (Otsamo, 1998b; McNamara et al., 2006). Application of fertilizer is widely practiced to promote the early growth of newly planted seedlings to help them overcome the harsh conditions of some planting sites (Kalinganire and Pinyopusarerk 2000; Noor et al., 2002). A previous reforestation trial (Chapter 3) has shown that a legume tree species performed better than a non-legume in harsh conditions of the mediterranean-type environment. It is generally known that the majority of legume trees form symbiotic associations with

N2-fixing bacteria and mycorrhizal fungi. However, knowledge on the status of N2- fixing capacity or mycorrhizal fungal association or dependency of the high-value timber species of continental Southeast Asia is limited. In addition, the effects of microorganisms on the early growth, establishment and survival from direct seeding of high-value timber legumes have not been tested.

In Chapter 3 (Section 3.4), soil nutrient and moisture availability were most likely to be critical factors affecting survival of direct seeded Eucalyptus gomphocephala during the first year in a mediterranean-type environment. Soil solution, a mixture of soil water

Murdoch University 2011 76

Chapter 5: Improving reforestation success of high-value and key forest tree species in Thailand and dissolved nutrients, is a key medium for supporting plant and soil biota growth (Bardgett, 2005). In bare ground, such as agricultural land, the soil moisture content in the surface horizon is significantly lower than those under forest cover (Uhl et al., 1988). Therefore, maintaining soil moisture availability is essential to prevent plant mortality and to sustain growth. Traditionally, soil moisture can be maintained by application of organic matter including covering the soil surface with mulch (Russell, 1973). These methods, however, are time and resource consuming. Conserving soil water through application of water retention polymers has been discussed in Chapter 2, Section 2.5.1. It is likely that water retention polymers not only retain soil water during intermittent dry periods in the wet season, but extend soil water availability into the dry season. Such products are worth testing in reforestation by direct seeding as newly emerged seedlings are susceptible to drought. It is anticipated that sustaining water availability during the first growing season enhances growth of the newly emerged seedlings so they are better able to withstand the ensuing dry period.

The aim of this Chapter is to explore the possibility to enhance the early growth and survival of four high-value timber species in Southeast Asia, A. xylocarpa, D. cochinchinensis, Sindora cochinchinensis Baill. and Xylia xylocarpa (Roxb.) Taubert, by direct seeding using low-cost and readily applicable techniques that can be applied by a wide range of stakeholders. Two fast growing exotic species, A. mangium and Eucalyptus camaldulensis Dehnh., were also included in the study as reference points for the indigenous species. Consideration of the growth characteristic of the high-value timber trees and possibilities for promoting early growth and survival led to the following hypotheses: 1) That beneficial microorganisms, inorganic fertilizer and a water retention polymer enhance survival and early growth of direct seeded plantations of high-value timber species; and 2) That the growth performance of the tree species differ in reforestation by direct seeding on a former agricultural land. To test these hypotheses, two separate experiments were established at one site in Sakeaw province, Thailand, adjacent to the Thai-Cambodian border. This site was chosen based on three factors as follows: 1) It is the natural habitat of all indigenous species used in this study; 2) The climatic conditions represent a large region of continental Southeast Asia; and 3) Ease of access and logistical arrangement.

Murdoch University 2011 77

Chapter 5: Improving reforestation success of high-value and key forest tree species in Thailand

5.2 Materials and methods

5.2.1 Study site

The study site is located within the forest plantation concession of the Forest Industry Organization (FIO) of Thailand. FIO is a state-own forest enterprise under the Ministry of Natural Resources and Environment. It was established in 1956, and has a main role to operate commercial forest plantations in the whole of Thailand (http://www.fio.co.th/). The site was an old agricultural land that had been managed for crop production, six- year rotation of E. camaldulensis agroforestry, during the past 30 years. It represents an easy site for reforestation considering the site conditions: location, flatness, medium weed infestation, and the absence of seed predators. It is easily accessible from Cambodia as it is close to the Thai-Cambodian border. The area is administratively located in Wang Nam Yen District, Sakaew Province, around 200 km east of Bangkok (Figure 5.1). It is located on a flat area with an altitude of 80 m a.s.l. This location has a latitude of 13o 33' 54" N, and a longitude of 102o 10' 20'' E. Climate in the study area is governed by the monsoon and is characterized by two distinct seasons, the wet season, occurring from May to October, and the dry season, occurring from November to April. The eastern region of Thailand in 2008 received an accumulative rainfall of 1 939 mm (Figure 5.2). The mean annual temperature is 27ºC, the hottest month is April with a mean temperature of 29.8ºC (Thai Meteorological Department, 2008). The trial area was previously covered with a mixed deciduous forest dominated by Pterocarpus macrocarpus, A. xylocarpa, X. xylocarpa and D. cochinchinensis. Some of these species are still present in remnant vegetation nearby. The natural forest was cleared in the late 1970s. Since then, the area has been continuously planted with Eucalyptus camaldulensis, on a six years rotation, up to the establishment of the experimental plots (Pers. comm., Chaiya Junsawang, 2008).

Soil of the study site is sandy loam derived from a red-yellow podzol (Moormann and Rojanasoonthon, 1967). The top soil, which varies from 5 to 20 cm, is grey in colour. Gravels and stones constitute around 20 to 30%, and 40 to 50% of the soil composition in the top soil and sub-soil, respectively (Figure 5.3). From each experimental area, two soil samples were randomly collected and they were equally bulked before sending to the Laboratory of Forest Soils of the Faculty of Forestry, Kasetsart University, Thailand, for physical and chemical analyses (Table 5.1).

Murdoch University 2011 78

Chapter 5: Improving reforestation success of high-value and key forest tree species in Thailand

Legend:

National road Provincial road Gravel paved road

1 km N

Figure 5.1 Location of the research plot inside the plantation concession of the Forest Industry Organization (FIO) in Wang Nam Yen District, Sakaew Province, Thailand.

500 2008 Mean 1971 to 2000

400

300 (mm)

200

Rainfall 100

0 Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec

Figure 5.2 Monthly rainfall of the eastern region of Thailand in 2008 compared to the mean for 1971-2000 (Thai Meteorological Department, 2008).

Murdoch University 2011 79

Chapter 5: Improving reforestation success of high-value and key forest tree species in Thailand

Table 5.1 Physical and chemical properties of the soils in the experimental trial, Sakeaw province, Thailand

Experiment 1 Experiment 2 0-20 (cm) 21-50 (cm) 0-20 (cm) 21-50 (cm) Sand (%) 65 78 66 71 Silt (%) 19 9 18 16 Clay (%) 16 13 16 13

pH (H2O ratio 1:1) 6.04 5.54 5.32 5.9 Total C (%)1 1.73 0.58 1.16 0.45 Total N (%)1 0.16 0.07 0.11 0.05 P (mg/kg)2 8.91 2.03 4.17 1.8 K (mg/kg)3 174 63.94 70.3 28.16 CEC (cmol/kg) 4 10.24 4.99 8.25 5.56 OM (%)5 3.41 0.93 2.08 0.58 Ca (mg/kg)3 1 663.6 854.8 1 337.6 1 07 Mg (mg/kg)3 177.36 72.66 99.16 50.4

1 2 3 Total C and N: combustion method (Dumas method). Bray II. Extracted with NH4OAc pH 7.0 and 4 5 analyzed with an Atomic Absorption Spectrophotometer. 1 M NH4OAc pH 7.0. Walkley-Black acid digestion. The values are from single measurements.

a b

Figure 5.3 Soil profile (a) and land preparation before direct seeding (b).

5.2.2 Procurement of materials

5.2.2.1 Tree species selection and seed procurement

Four indigenous tree species were selected for this study based on criteria shown in Table 5.2. In addition, two fast-growing exotic plantation species, A. mangium and E.

Murdoch University 2011 80

Chapter 5: Improving reforestation success of high-value and key forest tree species in Thailand camaldulensis, were included in this study as reference points. Detailed information on A. xylocarpa and D. cochinchinensis were discussed in Chapter 4, and Sindora cochinchinensis and X. xylocarpa are described in Appendix IV. Seeds were obtained from a number of sources (Table 5.2). Seeds of uniform size and no signs of physical damage or insect attack were selected by visual examination (except for E. camaldulensis). The seeds were then pretreated based on seed characteristics before use. Only D. cochinchinensis was available for a seed germination test before using in the field (Appendix VII, Test1).

Table 5.2 Seed supplier, method of seed pretreatment and species selection criteria

Species Supplier Year Pre-germination treatment1 Criteria for selection2

Acacia mangium ATSC3, 2006 Immersed in boiled water for  Seed availability Australia 1 min  Socio-economic value

 N2-fixing capability Afzelia xylocarpa RFD4, 2003 Seed coat was scarified at the  Conservation value Thailand distal end opposite to the  Seed availability hilum using a file and then  Socio-economic value soaked in tap water overnight Dalbergia RFD, 2006 Soaked in tap water overnight  Conservation values cochinchinensis Thailand  Seed availability  Socio-economic values

 N2-fixing capability  Intermediate successional Eucalyptus ATSC, 2006 No pre-treatment  Seedstage availability camaldulensis Australia  Socio-economic value Sindora RFD, 2003 The same method as applied  Seed availability cochinchinensis Thailand to A. xylocarpa  Early successional stage  Socio-economic value  Conservation value Xylia xylocarpa FIO5, 2008 Soaked in tap water for 15  Seed availability

Thailand min.  N2-fixing capability  Socio-economic value  Conservation value  Early successional stage

1 After Gunn (2001), CTSP (2005) and Chapter 4 (Section 4.3.2). 2 Order of the list does not necessary reflect priority. 3 Australian Tree Seed Center, Canberra. 4 Royal Forest Department, Thailand. 5 Forest Industry Organization, Thailand.

Murdoch University 2011 81

Chapter 5: Improving reforestation success of high-value and key forest tree species in Thailand

5.2.2.2 Root nodule bacteria

Root nodules of A. mangium, D. cochinchinensis and X. xylocarpa were collected from mature trees planted within the FIO’s plantations or native remnant vegetation within a 2 km radius of the planting site. After collection, nodules were washed with tap water and then air-dried prior to storing over silica gel in clear plastic vials. One night prior to trial establishment, nodules were crushed into tap water to make a crushed nodule solution. Nodules were used because rhizobium had not yet been isolated and identified, and thus pure inoculum was unavailable (Chapter 6) for these species at the time the experiments were established.

5.2.2.3 Mycorrhizal fungi

Two types of mycorrhizal fungi, AM and ECM, were used in the experiments. Mycorrhizal inocula were procured from four sources in Thailand and in Western Australia (Table 5.3). All AM inoculum was thoroughly mixed by hand, giving a mixed inoculum with ca. 24 spores g-1 soil. The ECM inoculum was made up in the form of a

Table 5.3 Mycorrhizal inoculum used in the experiments in Sakeaw province, Thailand

No. Fungal species Amount of No. of spores Source/producer soil (kg) g-1 soil AM inoculum1 1 Scutellospora sp. 2 10 Department of Biology, Chiang Mai Glomus etunicatum University, Thailand W.N. Becker & Gerd.

2 Glomus sp. 2 30 Department of Soil Science, Kasetsart University, Thailand 3 n.a. 10 25 - 28 Myco Star, commercial inoculum produced by Amputpong Co. Ltd., Thailand 4 Glomus sp. and 5 25 Rhizobium Research Center, Kasetsart Acaulospora sp. University, Thailand

ECM inoculum2 1 Pisolithus alba Field collection in Perth, WA 2 Hebeloma sp. Field collection in Perth, WA

1 All four AM inocula were mixed, with equal numbers of spores from each inoculums, into one batch, giving a total of ca. 24 spores g-1 soil. 2 The ECM inoculum was made up in the form of spore slurry contained 99% of Pisolithus alba and 1% of Hebeloma sp.

Murdoch University 2011 82

Chapter 5: Improving reforestation success of high-value and key forest tree species in Thailand

spore slurry consisting of Pisolithus alba nom. ined. (99%) and Hebeloma sp. (1%). The fungi were collected from eucalypt bushland at Murdoch University in late April 2008. The ECM fungal taxa have been used previously in inoculation trials in south China (Brundrett et al., 2005) and P. alba is known to occur in eucalypt plantations in Thailand (Pers. comm., B. Dell, 2008).

5.2.2.4 Fertilizer

® Chemical fertilizer (14:14:14 N:P2O5:K2O, Osmocote Controlled Release Fertilizer, manufactured by Sotus International Co. Ltd., 77 Maungthongtanee Changwattana Road. Pakreat district, Nontaburi province, Thailand 11120) was purchased from a local market in Wang Nam Yen district town. Micronutrients, manufactured by Sahai Kaset Co. Ltd. (Farmers’ Friend), Thailand, were applied separately.

5.2.2.5 Water retention polymer

The water retention polymer, Water$ave (manufactured by Polymer Innovations Pty Ltd, Singleton, NSW, Australia), a super absorbent polymer, was used. The gel absorbs and retains up to 500 times its own weight in water which is slowly released to the root system when needed by the plant. According to the manufacturer, Water$ave improves soil porosity and aeration. When water is used by the plant, the polymer returns to its dry form, and it remains active for three to five years with a minor loss in efficiency over time (Polymer innovations, 2005).

5.2.3 Experimental design and treatment

Experiment 1: By direct seeding, this experiment investigated the effect of microorganisms with or without inorganic fertilizer on survival and growth of five tree species, Acacia mangium, Afzelia xylocarpa, D. cochinchinensis, S. cochinchinensis and X. xylocarpa. The layout was a split-plot design involving five tree species as sub-plot treatments in a 2 × 6 (or 12 plots) factorial experiment with two levels of inorganic fertilizer and six levels of microorganism treatments (Table 5.4) as main plots. Therefore, within a replicated block there were 12 main plots, arranged in a randomized complete block design (RCBD), six of which were randomly allocated to “with fertilizer” and the

Murdoch University 2011 83

Chapter 5: Improving reforestation success of high-value and key forest tree species in Thailand other six were allocated to “without fertilizer”. Each main plot was then subdivided into 10 planting rows to which five tree species were randomly allocated to two adjacent rows (Figure 5.5a). The spacing between two adjacent seeding spots within a row or between two adjacent rows was 1 m. Between adjacent plots, there were buffer strips 3 m wide. There were no trees in the buffer areas. This design resulted in 20 seeding spots per species and a total of 100 seeding spots per plot. There were three blocks in the same site.

Experiment 2: By direct seeding, this experiment investigated the effect of inorganic fertilizer and water retention polymer on survival and growth of six species, the five species used in Experiment 1 and E. camaldulensis. The experimental trial was set up using a split-plot design involving six tree species in the sub-plots and four levels of treatments in the main plots (Table 5.4). Therefore, within a replicated block there were four main plots, arranged in RCBD (Figure 5.4). The arrangement of subplots within a main plot and the spacing within and between rows were the same as in Experiment 1, except that the main plot had 12 planting rows (Figure 5.5b). This design resulted in 20 direct seeding holes per subplot or 120 holes per main plot. Similar buffer strips, 3 m wide, were used between adjacent plots as Experiment 1; however, the size of buffer was not consistent due to site condition. There were three replicated blocks in the same site.

5.2.4 Site preparation and trial establishment

The trial site was prepared in mid April 2008. A small bulldozer was used to push the stumps of eucalypts, remaining from the clear cut, and the area was then disc-ploughed twice to 20 cm depth. Remaining debris was removed by hand (Figure 5.2b). The trial was established in mid May 2008 after the soil had received some monsoonal rains. Each seeding spot was marked with a 0.5 m stick to facilitate seeding, monitoring and maintenance. Three legume seeds were placed in each seeding spot [ca. 10 × 10 × 2-5 cm (depth, depending on seed size)] and covered with a soil layer to the depth of the seed. The use of three seeds is discussed in Section 5.4.1. It was difficult to differentiate the small seeds of E. camaldulensis from chaff; therefore, the number of seeds per seeding spot was not consistent although effort was made to target at least 3 seeds spot-1. Large seeding holes, 20 × 20 × 20 cm, were made to accommodate the 1-L volume of water retention polymer. Treatments and methods of treatment applications are shown in Table 5.4.

Murdoch University 2011 84

Chapter 5: Improving reforestation success of high-value and key forest tree species in Thailand

Table 5.4 Descriptions of the treatments used and methods of application in Experiments 1 and 2 undertaken in Sakeaw province, Thailand

Treatment Method of treatment application

Experiment 1 Microorganism Root nodule1 Crushed root nodule solution was applied at 2 cm below the seeds at the time of seeding (ca. 1 nodule seeding spot-1). AM Mixed AM inoculum (ca. 170 spores seeding spot-1) was applied at 2 cm below the seeds at the time of seeding. AM + ECM The same amount of mixed AM inoculum + 10 ml of ECM spore solution (ca. 3 × 108 spores seeding spot-1) were placed at 2 cm below the seeds at the time of seeding. Root nodule + AM As above. Root nodule + AM + ECM As above. Control No inoculum.

Fertilizer With fertilizer The set of six microorganism treatments above received 5 g of inorganic fertilizer (see text) + micronutrients per seeding spot applied at 10 cm from the seeding spot to a depth of 10 cm at the time of seeding. The application rate of micronutrients was 50 ml solution or ca. 0.675, 0.225, 0.3, 0.3, 0.113, 0.113, 0.038 and 0.008 mg of MgO, S, Fe, Mn, Cu, B, Zn and Mo, respectively (3 g of micronutrients dissolved in 20 L of water, as directed by the label). Without fertilizer A nother set of six microorganism treatments received no inorganic fertilizer. Experiment 2 Fertilizer The same amount of fertilizer used in Experiment 1 was applied the same way and at the same time. Polymer A solution of 2.5 g of water retention polymer dissolved in 1L of water was placed 20 cm below the seeds. Fertilizer + polymer As above. Control No fertilizer or polymer.

1 There were no root nodules available for A. xylocarpa and S. cochinchinensis, but direct seeding of these species was undertaken in all plots, and they received mycorrhizal treatments as the other species. However, these two species were excluded from statistical analysis.

Murdoch University 2011 85

Chapter 5: Improving reforestation success of high-value and key forest tree species in Thailand

Experiment 1 1.1.1 1.1.2 1.1.3 F2T5 F1T1 F1T2

1.1.4 1.1.5 1.1.6 1.1.7 1.1.8 1.1.9 To FIO Office, 1.5 km F2T1 F1T5 F2T4 F1T4 F2T6 F1T3

1.2.1 1.2.2 1.2.3 1.1.10 1.1.11 1.1.12 F2T2 F1T4 F2T6 F1T2 F2T3 F1T6

1.2.4 1.2.5 1.2.6 1.2.7 1.2.8 1.2.9 F2T3 F1T6 F1T2 F1T1 F1T3 F2T5 N N 1.2.10 1.2.11 1.2.12

F1T5 F2T4 F2T1

1.3.1 1.3.2 1.3.3 1.3.4 1.3.5 1.3.6 F2T3 F1T2 F2T5 F1T4 F2T1 F1T3

1.3.7 1.3.8 1.3.9 1.3.10 1.3.11 1.3.12 F1T1 F2T4 F1T6 F2T6 F1T5 F2T2

carpusmacrocarpus

o

Pter 2.2.3 2.2.1 Experiment 2 T1 T2

2.2.4 2.3.1 2.3.2 2.3.3 2.3.4

T4 T4 T1 T3 T2

Old plantation of of plantation Old

2.1.1 2.1.2 2.1.3 2.1.4 2.2.2 T3 T4 T1 T2 T3

Figure 5.4 Sketch map of the experimental plots in the field inside the FIO plantation in Sakeaw province. Plots marked with the same colour are in the same replicated block. Plot number: 1.1.1, denotes Experiment 1, Block 1, plot number 1. Treatments in Experiment 1: Microorganism: T1, root nodule; T2, AM; T3, AM + ECM; T4, root nodule + AM; T5, root nodule + AM + ECM; T6, control. Fertilizer: F1, with fertilizer; F2, without fertilizer. Experiment 2: T1, fertilizer; T2, polymer; T3, fertilizer + polymer; T4, control.

Murdoch University 2011 86

Chapter 5: Improving reforestation success of high-value and key forest tree species in Thailand

Two weeks after trial establishment, seeding spots without germinants were refilled with three seeds in both experiments. Seeding spots that contained one or more germinants were not reseeded. The refilling rates were ≤ 5% for A. mangium, A. xylocarpa, S. cochinchinensis, X. xylocarpa and D. cochinchinensis and ca. 60% for E. camaldulensis. No repeat treatments were applied to the refilling spots. At two months after trial establishment, thinning was undertaken where the seeding spots contained more than one seedling. It was observed that heavy rains had washed away some of the seeds, particularly E. camaldulensis, therefore reducing seedling emergence.

a b

Figure 5.5 Schematic presentation of the arrangement of the seeding spots in a plot in Experiment 1 (a) and Experiment 2 (b). In a plot, circles marked with different colour patterns represent different tree species. Species in Experiment 1: Acacia mangium; Afzelia xylocarpa; Dalbergia cochinchinensis; Sindora cochinchinensis and Xylia xylocarpa. Species in Experiment 2: all species in Experiment 1 plus Eucalyptus camaldulensis. Spacing in both Experiments: 1m between two adjacent rows or between two adjacent spots in a row. Broken lines represent plot boundaries. Plot size: 10 × 10 m in Experiment 1 and 10 × 12 m in Experiment 2.

5.2.5 Maintenance of the field trial

The surrounding areas of the trial plots were grazing grounds for herds of cattle from a nearby village. Therefore, fencing, using barbwire, was erected to a height of 1.5 m to protect the trial plots from animal and human interference. After two months, inferior seedlings with signs of unhealthy, slow growth, or attacked by insects were cut out at the ground level leaving one healthy seedling per seeding hole.

Although the trial area was ploughed before establishment, weeds could not be eliminated even during the first month after ploughing, and it became the main

Murdoch University 2011 87

Chapter 5: Improving reforestation success of high-value and key forest tree species in Thailand competitors for space, water and soil nutrients with the newly-emerged seedlings. Weeds included grasses, vines, bush and pioneer tree species of the secondary forest formation, in total about 20 species. Those that could be identified were Eupatorium odoratum L., Mimosa pudica L., Sida acuta Brum. f., Phyllanthus urinaria L., Nephelium hypoleucum Kurz, Imperata cylindrica, Cassia occidentalis L., Dioscorea oryzetorum Prain & Burkill, Dioscorea brevipetiolata Prain & Burkill and Dioscorea hispida Dennst.

Weeding in plots was carried out two months after seeding using hand tools. Weeding was then carried out every two months until the seedlings were six-months old (December 2008). A final weeding was carried out in the middle of the second rainy season, August 2009. The buffer zones between plots were treated once with herbicide at three months using Glyphosate isopropylammonium mixed with a surfactant (Thai brand name Mixer) with a respective rate of 300 ml and 200 ml in 20 L of water, as directed by the label.

5.2.6 Data collection and analysis

Survival rate, height and diameter (at 2 cm above ground level) growth were assessed at 20 months after trial establishment. Data analyses were carried out using SPSS statistical package Version 15.0. For all tests, the probability p ≤ 0.05 was used. For both experiments, Linear (Mixed Models) analysis was used (on advice of the Statistical Consulting Group, http://scg.maths.uwa.edu.au/). For post hoc analyses, a multiple comparison test, Sidak, was used to detect differences among means. ANOVA with Type III sums of squares was used, as this adjusts the treatment effects for any differences between blocks. The survival rate was calculated based on the number of direct seeding holes with presence of a seedling at the time of assessment as percentage of the total direct seeding holes per plot (n = 20). Percentages of survival rate are replicated block means across three blocks. They were transformed to arc-sin square root transformation (Gomez and Gomez, 1984) prior to statistical analysis. Height and diameter data in both experiments were Napierian logarithmic transformed before analysis to fit the normal distribution. The descriptive statistics presented in graphs and tables are untransformed data.

Murdoch University 2011 88

Chapter 5: Improving reforestation success of high-value and key forest tree species in Thailand

5.3 Results

5.3.1 Experiment 1

5.3.1.1 Seedling survival

After 20 months, microorganism had no significant effect on the survival rate (Table 5.5). Fertilizer had a significant effect on the survival rate. Application of inorganic fertilizer lowered the survival rate by 2%. The interaction of microorganism and fertilizer was not significant, suggesting that the fertilizer difference was not affected by microorganism.

Seedling survival rates differed significantly among tree species. However, the difference between the highest, X. xylocarpa, and the lowest, D. cochinchinensis, was minor, being 7%. In general, the three species exhibited remarkably high survival rates (> 92%). The interaction of microorganism and tree species was significant, indicating the differential response of tree species to microorganism (Figure 5.6a). Results from Sidak comparisons indicated that the significant interaction of microorganism and tree species was due primarily to the significant response of D. cochinchinensis to the co- inoculation of AM + ECM and root nodule + AM. The survival rate differences of these two treatments from the control were 15 and 17%, respectively. Acacia mangium and X. xylocarpa did not respond to the microorganism (Figure 5.6a). The interaction of fertilizer and tree species was not significant, indicating that the differences in seedling survival among tree species were not affected by fertilizer. There was no three-way interaction, microorganism × fertilizer × tree species, on survival rate.

5.3.1.2 Height and diameter growth

Height was not affected by microorganism, fertilizer, and there was no interaction between the two main factors. However, there was a significant difference in height among tree species. The interaction of microorganism and tree species or fertilizer and tree species was not significant, indicating that the differences in height among tree species were not affected by microorganism or fertilizer. There was no three-way interaction, microorganism × fertilizer × tree species, on height.

As for height, diameter was also not affected by microorganism and fertilizer. There was no interaction between the two main factors. There was a significant difference Murdoch University 2011 89

Chapter 5: Improving reforestation success of high-value and key forest tree species in Thailand

Table 5.5 Survival rate, height and diameter growth at 20 months in Experiment 1 as effects of microorganism, fertilizer and tree species in a field trial in Sakeaw province, Thailand

Survival rate (%) Height (m) Diameter (cm) Microorganism Root nodule 97.50 ± 0.93 2.59 ± 0.56 3.61 ± 0.56 AM 97.22 ± 0.92 2.61 ± 0.49 3.64 ± 0.48 AM + ECM 99.44 ± 0.56 2.59 ± 0.52 3.58 ± 0.48 Root nodule + AM 96.39 ± 1.39 2.33 ± 0.50 3.26 ± 0.52 Root nodule + AM + ECM 93.06 ± 3.16 2.43 ± 0.55 3.34 ± 0.55 Control 96.67 ± 1.76 2.40 ± 0.50 3.31 ± 0.48

Fertilizer With fertilizer 95.65 ± 1.17 b 2.57 ± 0.30 3.52 ± 0.29 Without fertilizer 97.78 ± 0.74 a 2.42 ± 0.29 3.40 ± 0.29

Tree species Acacia mangium 98.33 ± 0.66 a 5.43 ± 0.14 a 6.26 ± 0.18 a Dalbergia cochinchinensis 92.50 ± 1.78 b 1.13 ± 0.07 b 1.81 ± 0.09 c Xylia xylocarpa 99.31 ± 0.29 a 0.91 ± 0.05 c 2.30 ± 0.07 b Analysis of variance Source d.f. F Sig. F Sig. F Sig. Main-plot analysis Block 2 1.471 0.251 6.213 0.007 3.653 0.043 Microorganism (Mic.) 5 2.299 0.080 1.251 0.320 1.339 0.285 Fertilizer 1 5.285 0.031 1.082 0.309 0.696 0.413 Microorganism × fertilizer 5 2.190 0.092 1.682 0.181 1.433 0.252 Error 22 Sub-plot analysis Tree species 2 21.332 0.000 746.489 0.000 560.274 0.000 Microorganism × tree species 10 2.713 0.010 1.967 0.059 2.537 0.015 Fertilizer × tree species 2 0.317 0.730 0.161 0.852 1.773 0.181 Mic. × fertilizer × tree species 10 1.386 0.215 0.965 0.486 0.710 0.711 Error 48

Values are replicated block means (n = 3) ± S.E. The significant ANOVA probabilities are marked in bold. In a column, means followed by the same letters are not significantly different at p ≤ 0.05.

in diameter among tree species. Diameter was significantly affected by the interaction of microorganisms and tree species, indicating the differential response of tree species to microorganism. Results from Sidak comparisons indicated that the significant interaction of microorganism and tree species was due primarily to the significant

Murdoch University 2011 90

Chapter 5: Improving reforestation success of high-value and key forest tree species in Thailand response of D. cochinchinensis to the co-inoculation of AM + ECM. The inoculated tree had 43% larger diameters than the control trees. Acacia mangium and X. xylocarpa showed no response to inoculation (Figure 5.6b). The interaction of fertilizer and tree species was not significant, indicating that the differences in diameter growth among tree species were not affected by the fertilizer used. There was no three-way interaction, microorganism × fertilizer × tree species, on diameter growth.

a 120 a a a a a a a a a a a bc abc ab a a a 100 bc c

80 60 40 20

rate Survival (%) 0

b T1: nodule T4: nodule + AM 8 a a a a T2: AM T5: nodule + AM+ ECM 7 a a 6 T3: AM + ECM T6: control

(cm)

5 4 a Diameter 3 a ab a a a a abc abc a 2 c bc 1 0 Acacia mangium Dalbergia cochinchinensis Xylia xylocarpaxylocaroa

Figure 5.6 Effect of microorganism across different tree species on survival rate and diameter growth at 20 months in Experiment 1 in Sakeaw province, Thailand. In the same species, columns followed by the same letters are not significantly different at p ≤ 0.05.

5.3.2 Experiment 2

5.3.2.1 Seedling survival

After 20 months, treatment did not promote seedling survival (Table 5.6). There was a significant difference among tree species, A. mangium had the highest survival rate and E. camaldulensis had the lowest rate. The interaction of treatment and tree species was not significant, indicating that the survival differences among tree species were not

Murdoch University 2011 91

Chapter 5: Improving reforestation success of high-value and key forest tree species in Thailand affected by the treatments used.

Table 5.6 Survival rate, height and diameter growth at 20 months in Experiment 2 as effects of treatments and tree species in a field trial in Sakeaw province, Thailand

Survival rate (%) Height (m) Diameter (cm) Treatment Fertilizer 90.28 ± 5.37 2.14 ± 0.49 ab 3.06 ± 0.48 ab Polymer 91.94 ± 3.69 2.29 ± 0.49 ab 3.11 ± 0.48 ab Fertilizer + polymer 92.50 ± 3.51 2.40 ± 0.48 a 3.30 ± 0.50 a Control 84.72 ± 5.92 1.72 ± 0.40 b 2.40 ± 0.38 b Tree species Acacia mangium 99.58 ± 0.42 a 5.19 ± 0.16 a 6.10 ± 0.21 a Afzelia xylocarpa 91.67 ± 3.39 a 0.48 ± 0.03 d 1.17 ± 0.06 e Dalbergia cochinchinensis 97.50 ± 0.97 a 1.10 ± 0.10 b 1.74 ± 0.12 d Eucalyptus camaldulensis 64.58 ± 8.84 b 4.25 ± 0.33 a 4.68 ± 0.43 b Sindora cochinchinensis 97.92 ± 1.44 a 0.86 ± 0.10 c 1.75 ± 0.13 d Xylia xylocarpa 87.92 ± 6.47 a 0.95 ± 0.09 bc 2.38 ± 0.11 c

Analysis of variance Source d.f. F Sig. F Sig. F Sig. Main-plot analysis Block 2 0.008 0.992 4.959 0.054 4.183 0.073 Treatment 3 0.346 0.794 4.907 0.047 5.845 0.033 Error 6 Sub-plot analysis Tree species 5 9.672 0.000 271.031 0.000 178.773 0.000 Treatment × tree species 15 0.945 0.526 1.833 0.064 1.922 0.050 Error 40

Values are replicated block means (n = 3) ± S.E. The significant ANOVA probabilities are marked in bold. In a column, means followed by the same letters are not significantly different at p ≤ 0.05.

5.3.2.2 Height and diameter growth

Treatment had a significant effect on height (Table 5.6). The combination of fertilizer and water retention polymer resulted in a 40% increase in height over the control. There was a significant difference in height among tree species. Among the high-value indigenous trees, growth of A. xylocarpa was inferior to the other three species (Figure 5.8d) whereas D. cochinchinensis was the best (Figure 5.8e). The interaction of treatment and tree species was not significant, indicating that the height differences

Murdoch University 2011 92

Chapter 5: Improving reforestation success of high-value and key forest tree species in Thailand among tree species were independent of treatments used.

Treatment had a significant effect on diameter growth. The combination of fertilizer and polymer increased diameter by 38% compared to the control. There was a significant difference in diameter among tree species. The diameter growth of A. xylocarpa was inferior to that of D. cochinchinensis and X. xylocarpa (Figure 5.8d, e and h). The treatment × tree species interaction was just significant, indicating that there existed different responses of tree species to the treatments used. Results from Sidak comparisons indicated that the significant interaction of treatment and tree species was due to the response of E. camaldulensis to all treatments (Figure 5.7). Fertilizer, polymer and the combination of fertilizer and polymer increased diameter by 74, 86 and 95%, respectively, compared to the control. Other species did not respond to the treatments used.

T1: fertilizer T2: polymerT1 T2 T3: T3fertilizer T4 + polymer T4: control

8 a a a 7 a a a 6 a 5

4 b a 3 a a a a a a a a a a a 2 a a a a

Basal diameter diameter Basal(cm)

1

0

A. xylocarpa E. camaldulensis X. xylocarpa

A. mangium D.cochinchinensis S. cochinchinensis m

Figure 5.7 Effect of treatment on diameter growth across tree species at 20 months in Experiment 2 in Sakeaw province, Thailand. For the same tree species, columns followed by the same letters are not significantly different at p ≤ 0.05.

Murdoch University 2011 93

Chapter 5: Improving reforestation success of high-value and key forest tree species in Thailand

a b

c d e

f g h

Figure 5.8 (a) and (b), Overview of the trial plots established by direct seeding in Sakeaw province. (a) From left to right Acacia mangium, Xylia xylocarpa, Dalbergia cochinchinensis, Sindora cochinchinensis and Eucalyptus camaldulensis. (b) Xylia xylocarpa, left, and Acacia mangium, right. (c) - (h), Close-up views of the tree species. (c) Acacia mangium, (d) Afzelia xylocarpa, (e) Dalbergia cochinchinensis, (f) Eucalyptus camaldulensis, (g) Sindora cochinchinensis and (h) Xylia xylocarpa. Photos were taken in the early dry season, late December 2009, when the trees were 20 months old. Pole was painted with 20 cm colour bands.

Murdoch University 2011 94

Chapter 5: Improving reforestation success of high-value and key forest tree species in Thailand

5.4 Discussion

5.4.1 Can direct seeding of high-value timber species be an option for reforestation of former agricultural lands in tropical regions?

This research indicated that, with good site preparation and when weeds are intensively managed, direct seeding of high-value timber species is possible for reforestation of former agricultural lands of mainland Southeast Asia. In both experiments, the survival rate at 20 months for the indigenous species was very high, being more than 87%. However, care should be taken when claiming that the survival rate of indigenous species was high. In this study, three seeds were initially placed into a seeding spot. This was intentionally undertaken because: 1) The germination test of D. cochinchinensis (Appendix VII, Test 1) and a previous report on X. xylocarpa (Phongoudome, no date) suggested that germination rates of these two species in the field would be < 70%; 2) It was anticipated that a number of seeds or newly-emerged seedlings would be destroyed by predators (e.g., Woods and Elliott, 2004; Doust et al., 2008; Sovu et al., 2010), and some seeds would be washed away from seeding spots by rain flush; and 3) To ensure maximum survival rates which would help validate the interpretation of treatment (microorganism, fertilizer and polymer) effects. However, many seeding spots (no record was undertaken) required thinning after two months leaving one seedling per spot. If one seed was initially placed in a seeding spot and refilling was not undertaken, the final survival rate would have been significantly lower than the present result.

The high survival rates obtained in this study could possibly be due in part to the method of seeding, post seeding maintenance, species characteristics and environmental conditions (early rainfall, fencing, low seed/seedling predation). Seeds were subjected to pretreatment before direct seeding, except E. camaldulensis. Pretreatment of seeds to accelerate seed germination improves the efficiency of direct seeding as individual seed is ready to grow (Schmidt, 2008). In this study, seeds were buried to the seed thickness, therefore they were not exposed to desiccation before seed germination. In testing direct seeding techniques of 18 species in tropical northeast Queensland, Doust et al. (2006) discovered that burying seeds resulted in higher survival rates than broadcasting, sowing in furrows or sowing on mounds. All of the four indigenous tree species originate from deciduous forests (Vũ, 1996; Dy Phon, 2000) and thus can be presumed to have adapted

Murdoch University 2011 95

Chapter 5: Improving reforestation success of high-value and key forest tree species in Thailand to the harsh conditions of those forest lands. These species regenerate readily in natural habitats when optimal conditions (moisture, light and temperature) are met (e.g., Kaewkrom et al., 2005; Koonkhunthod et al., 2007). In addition, the trial was established at the beginning of the wet season, in the second half of May, after the soil had received some rain. Thus, the newly emerged seedlings had enough time to develop a rooting system in the subsoil in order to access stored water during the dry season. The trial site had been disc-ploughed twice before seeding, and the plots were kept free of weeds until the middle of the second rainy season (August 2009). The results may have been significantly different if weeding was not undertaken. Two non-weeded plots adjacent to Experiment 2, located inside the trial site, were fully covered with luxuriant weeds to a height of > 1m at eight months. At year two, none of the four indigenous and two fast- growing species, A. mangium and E. camaldulensis, had emerged from the weeds although germination was evident shortly after seeding (data not presented). Weed competition and the requirement of frequent weeding is a great disadvantage of direct seeding (Schmidt, 2000). Some researchers treated weeding frequency as a set of treatments, not only in direct seeding experiments (e.g., Willoughby and Jinks, 2009), but also in planting of nursery-raised seedlings (e.g., Günter et al., 2009). This is considered more in Chapter 7.

In a research trial established on a degraded land with an eroded soil surface, in Kampong Speu province, Cambodia, the survival rate of A. xylocarpa was 30% at three months (CTSP, 2005). However, on checking the blank seeding holes, it was apparent that the seeds were planted too deep, ca. 10 cm, thus reducing emergence. The environmental conditions of that trial site are quite similar to the Sakaew site with an annual rainfall of 1 500 mm. Another direct seeding trial undertaken by Tunjai (2005) in Chiang Mai province, Northern Thailand, resulted in an survival rate for A. xylocarpa of 52% and height growth of 29.3 cm after one year when weeding was undertaken every two months. On the very fertile, red volcanic soil derived from latosols (Crocker, 1963), in Kampong Cham province, Cambodia, height growth of the indigenous species from an eighteen-month-old agroforestry trial was more than double that of the Sakeaw trials (Table 5.7). This trial was established by planting of nursery-raised seedlings aged about eight months.

Murdoch University 2011 96

Chapter 5: Improving reforestation success of high-value and key forest tree species in Thailand

Table 5.7 Survival and growth of nursery-raised seedlings of four indigenous species planted in Kampong Cham province, Cambodia, at 18 months old (Uon Samol, unpublished data)

Species Survival (%) Height (m) Diameter (cm) Afzelia xylocarpa 89.69 1.45 ± 0.08 1.76 ± 1.06 Dalbergia cochinchinensis 82.47 3.28 ± 0.07 4.36 ± 0.74 Sindora cochinchinensis 95.88 2.18 ± 0.06 3.90 ± 0.79 Xylia xylocarpa 94.85 2.49 ± 0.05 4.77 ± 1.01

It is widely accepted that direct seeding significantly reduces the establishment cost (Bullard et al., 1992; Engel and Parrotta, 2001; Cole et al., 2011). Establishment cost may vary significantly between species and reforestation sites; therefore, a cost effective analysis is needed if a species with expensive seed is to be used in direct seeding. Restoration by direct seeding has not been widely practiced in Southeast Asia. The availability of seeds of the target species could be a limiting factor for this practice. Currently, finding sufficient seeds of high-value timber trees, like A. xylocarpa and D. cochinchinensis, is a challenge for reforestation practitioners in Cambodia because the country has limited number of seed sources and the tree seed sector is still weak (Chapter 4, Section 4.5.2). In addition to scarcity, seed of D. cochinchinensis is expensive, being US$ 200 kg-1 in Cambodia in 2009 (Pers. comm., Uon Samol, 2009).

Traditionally, many slow-growing indigenous species need to be raised and maintained in a nursery for about one year before out planting in the field (FORRU, 2006). With a survival rate more than 82% at the second year, the pertinent question is: are resources for nursery activities needed for these species? This study has produced an encouraging result for local communities and biodiversity conservation practitioners to restore community forests (Chapter 8) and conservation areas using a low-cost and simple technique―direct seeding.

5.4.2 Do microorganisms enhance reforestation by direct seeding?

The findings of Experiment 1 indicated that D. cochinchinensis responded to inoculation with microorganisms both in increased survival and diameter growth. Dual inoculation of

AM either with ECM or N2-fixing bacteria significantly improved the survival rate. However, only one microorganism treatment, the dual inoculation of AM with ECM, improved diameter growth. Bacteria or AM alone did not promote survival or diameter

Murdoch University 2011 97

Chapter 5: Improving reforestation success of high-value and key forest tree species in Thailand growth, and triple inoculation with AM, ECM and bacteria yielded the same result. The current knowledge of symbiotic organisms in D. cochinchinensis is limited. What we know so far is D. cochinchinensis forms symbiotic associations with ECM fungi (Chalermpong and Bunthaveekun, 1982), but the compatible species are unknown. The genus Dalbergia is known for its symbiotic association with AM fungi (Khan, 2001; Mridha and Dhar, 2007; Bargali, 2011). Bisht et al. (2009) reported a significant growth improvement in Dalbergia sissoo under glasshouse conditions when the species was inoculated with AM fungi (Gigaspora albida Schenck & Smith, Glomus intraradices and Acaulospora scrobiculata Trappe.), and the synergistic effect occurred when the fungi were co-inoculated with bacteria (Rhizobium leguminosarum bv. viciae). The type of rhizobia that are compatible with D. cochinchinensis is discussed elsewhere (Chapter 6).

Acacia mangium and X. xylocarpa did not respond to inoculation with bacteria or fungi. This could be due to the inoculants being ineffective, or competition from more effective soil-borne symbionts. Xylia xylocarpa is reported to form symbiotic associations with AM fungi (Lakshman et al., 2001; Youpensuk et al., 2004), ECM fungi (Chalermpong and Bunthaveekun, 1982) and slow-growing Bradyrhizobium spp. (Manassila et al., 2007). However, the compatibility of the inoculants used is unknown as they had not been tested with X. xylocarpa and the other two species before using in the present study. Another possibility is that this species may have low dependency on mycorrhizal fungi. Acacia mangium is reported to be compatible with a number of AM fungi (Dart et al., 1991; Ghosh and Verma, 2006; Aggangan et al., 2010) including Glomus etunicatum, the only fungus known to the species level in the inoculum used.

Some Pisolithus species are compatible with A. mangium (Duponnois and Bâ, 1999; Duponnois et al., 2002). Acacia mangium forms symbiotic association with fast- growing Rhizobium and slow-growing Bradyrhizobium (Dart et al., 1991; Galiana et al., 1998). Checking for nodulation or mycorrhizal infection in the field plots would help clarify assumptions about competition from indigenous microorganisms. That this was not undertaken is recognised as a weakness of the current study. It was not able to be investigated due to the very limited access to the site, distance from the base laboratory in Australia and strict quarantine measures for entry of soil and organisms into Australia. Persistence of inoculants in the field should be checked using molecular tools (e.g., Print et al., 2003) and compared to the inocula being trialed. It can be explored in the future in a trial established in Cambodia (Chapter 7) as the Bradyrhizobia have now

Murdoch University 2011 98

Chapter 5: Improving reforestation success of high-value and key forest tree species in Thailand been characterized (Chapter 6).

As inoculation with soil microorganisms was not effective for some species, the pertinent question is raised: is it worth inoculating microorganisms in reforestation sites? The need for inoculation of microorganisms in reforestation is further discussed in Chapter 8. These findings suggest the importance of selecting appropriate symbionts (Chapters 6) prior to using them in the field (Chapter 7).

5.4.3 Do inorganic fertilizer and water-retention polymers enhance reforestation by direct seeding?

Chemical fertilizer had no significant effect on plant growth in Experiment 1. There was a block effect on height (Table 5.5) with height in Block 1 being 14 and 15% greater than Blocks 2 and 3, respectively (data not presented). These effects possibly indicate a gradient from soil fertility at the lower slope (Block 1), to lower fertility at the upper slope (Block 3) (Figure 5.4). Additional soil analyses should have been undertaken to ascertain the cause of the block effect, but due to limited resources and logistic arrangements this work was not committed. Other possible contributing factors, for which data were not available, include soil depth, sub soil chemistry, hydrology and soil biota. Soil at the trial site is characterized by a high percentage of sand and low cation- exchange capacity (CEC) (Hazelton and Murphy, 2007). Soil with a high percentage of sand has a limited capacity to buffer soil nutrients and retain moisture. Compared to other reforestation studied sites in the tropics, such as an Imperata-dominated grassland in South Kalimantan, Indonesia (Otsamo et al., 1997) (the same depth, 0-20 cm), soil in Sakeaw trial had higher available P (almost 3 fold), comparable N, but a considerably low level of CEC.

In Experiment 2, treatment did not improve the survival. The combination of fertilizer with a polymer resulted in a significant height advantage (40%) over trees in the control plots. The effect of treatments on diameter growth of the six tree species was selective, as only E. camaldulensis responded positively to the treatments. Fertilizer and water retention polymer each improved diameter growth of E. camaldulensis by 74 and 86% over the control, respectively, and the improvement reached 95% when fertilizer was combined with the polymer. This trial has shown that E. camaldulensis is more

Murdoch University 2011 99

Chapter 5: Improving reforestation success of high-value and key forest tree species in Thailand responsive to treatment application than the other five large-seeded species. This finding is in accordance with a report from a glasshouse experiment that small-seeded seedlings from eight species of the genus Eucalyptus responded markedly to nutrient addition compared to the large-seeded species from the genus Banksia in nutrient-poor soil (Milberg et al., 1998). Seedlings of E. camaldulensis, with limited nutrient content in the cotyledons, are likely to require access to soil nutrients immediately after emergence. In contrast, seedlings from larger seeds, containing larger amounts of N, P

(Milberg et al., 1998) and K in the seed, use nutrient reserves from their cotyledons after seedling emergence and take longer to be fully dependent on soil nutrients. The transition period over which the newly emerged seedlings change their dependency from cotyledon reserves to external resources varies between species and growing environments (Kitajima, 2002). As an example, a newly-emerged legume, Hymenaea courbaril L., mainly used its cotyledon resources for about two months before fully exploiting external resources (Tiné et al., 2000).

Duangpatra and Attanandana (1992) reported that, in sandy soil of the northeast of Thailand, application of 70 g of a polymer (commercially known as ACRYHOPE) in each planting hole, together with 120 g of rock phosphate, increased height growth of cashew nut (Anacardium occidentale L.), green mango (Mangiferra indica L. cv. Keo- Sawei) and para-rubber (Hevea brasiliensis Muell, Arg) planted in the dry season (January to May 1988) with irrigation as an experimental factor. The polymer significantly enhanced survival of cashew nut and mango, but not rubber. In Saudi Arabia, Al-Humaid and Moftah (2003) conducted a drought stress experiment in the field using six-month old seedlings of Conocarpus erectus L. planted in 30-L plastic containers filled with 40 kg sandy soil mixed with five rates of a polymer (Stockosorb K400 from Germany), at the rate of 0, 40, 80, 160 or 240 g per container. Each container received 10 g NPK 20:20:20 fertilizer with micronutrients. The containers were buried such that the surface of the pot was at ground level. They reported that containers with 240 g polymer extended seedling survival for two months after the last watering (or three times longer compared to the control) and increased shoot growth, during the water stress period, by 300% over the control. Elsewhere, water retention polymers have been used and significant effects on plant growth and reduction of water usage have been reported (e.g., Hayat and Ali, 2004; Bhat et al., 2009).

Murdoch University 2011 100

Chapter 5: Improving reforestation success of high-value and key forest tree species in Thailand

The trend in Table 5.6 and Figure 5.7 suggest that increasing the amount of the polymer and fertilizer could bring about significant improvement in growth of indigenous species and is worth testing in the future. A much larger amount of inorganic fertilizer is generally recommended for planting of high-value timber species, for example 100 g of NPK 15:15:15 was given to each Chukrasia tabularis seedling in the first year (Kalinganire and Pinyopusarerk, 2000) (Chapter 4). However, the amount of fertilizer and its frequency of application is site dependent given the great variability in soils and the extent to which they have become degraded. In soil, roots depend on water, aeration and nutrients for their growth (Kramer, 1969), and these requirements can be provided by application of water retention polymer in combination with fertilizer. However, further study is needed to explore the optimum amount of polymer and fertilizer that could enhance direct seeding of indigenous species. Unlike microorganisms, water retention polymers and fertilizers are applicable for any tree species and are readily available. Polymers could be useful for overcoming weed competition in reforestation as the planting season can be extended into the late dry season (possibly April in Cambodia and Thailand) where weed competition is almost nonexistent. Seedlings then have additional time to harden before competition with weeds occurs. However, an optimum amount of water retention polymer that could hold enough water for seedling growth during the extended planting period has to be explored. Polymers are potentially disadvantageous for newly planted seedlings if there are continuous wet days as they can promote waterlogged conditions near seedling roots.

5.5 Concluding remarks

This study showed that beneficial microorganisms enhanced survival and diameter growth of D. cochinchinensis. However, the effects of organisms were selective as only the co-inoculation of AM with ECM or AM with N2-fixing bacteria substantially improved survival. The co-inoculation of AM with ECM significantly improved diameter growth. Although N2-fixing bacteria or AM alone did not promote survival or diameter growth of D. cochinchinensis, these findings indicate the potential of microorganism in improving survival and growth of a species with high conservation and economic value (Chapter 4). Therefore, inoculation of D. cochinchinensis with effective strains of N2-fixing bacteria (Chapters 6 and 7) or mycorrhizal fungi should be evaluated more widely in reforestation and plantation sites in the region.

Murdoch University 2011 101

Chapter 5: Improving reforestation success of high-value and key forest tree species in Thailand

Application of inorganic fertilizer in combination with a water retention polymer promoted height growth of all species tested. It would be interesting to test additional rates of fertilizer and polymer to determine if higher amounts could further enhance growth of indigenous species across a range of sites. This study indicated that the four high-value timber species, A. xylocarpa, D. cochinchinensis, S. cochinchinensis and X. xylocarpa are promising species for reforestation by direct seeding of former agricultural lands provided good site preparation and intensive weeding are part of the reforestation management plan. However, this finding is limited to a single trial site that had been managed for agricultural crop production in the last 30 years, and furthermore, the land was well prepared and the seedlings were kept free of weeds for 14 months. The findings may be quite different on more difficult sites, such as those covered by Imperata grassland, or with degraded soils or where site preparation and weeding are kept to a minimum (Chapter 7).

Murdoch University 2011 102

Chapter 6: Selection of N-fixing bacteria for Dalbergia cochinchinensis

CHAPTER 6

SELECTION OF EFFECTIVE N2-FIXING BACTERIA FOR DALBERGIA COCHINCHINENSIS PIERRE

Emergence of a seedling of Dalbergia cochinchinensis from a seed coated with N2-fixing bacteria in the glasshouse experiment.

Murdoch University 2011 103

Chapter 6: Selection of N-fixing bacteria for Dalbergia cochinchinensis

6.1 Introduction

In Chapter 4, the economic potential of planting Dalbergia cochinchinensis Pierre and the promotion of species for use in reforestation programmes for economic development and poverty reduction were discussed. However, improving the early growth of the species is a challenge for reforestation managers. As a legume, the species forms symbiotic associations with N2-fixing bacteria. Inoculation with these organisms may improve early growth of seedlings in the field if effective bacteria are identified and the planting sites have low populations of effective bacteria (Chapters 2 and 8). However, an effective N2-fixing strain has not been reported for this species. Therefore, the objectives of this study are as follows: 1) To find strains of effective N2-fixing bacteria for D. cochinchinensis under glasshouse conditions; and 2) To identify the taxa that formed nodules with D. cochinchinensis. Effective strains selected in this glasshouse experiment will be also used for inoculation of D. cochinchinensis in a field trial in Cambodia (Chapter 7). In addition, a seed coating technique (Chapter 2, Section 2.5.4) was evaluated as an inoculation method for D. cochinchinensis.

6.2 Materials and methods

6.2.1 Environmental conditions of the study site

The study was undertaken in a glasshouse at Murdoch University (South Street campus), which is about 15 km south from the centre of Perth, Western Australia. Murdoch University is on the Swan Costal Plain with a latitude of 32° 4' 1" S and a longitude of 115° 50' 3" E. Rainfall starts in April and finishes in October, and the total annual rainfall in 2010 was 495.8 mm (the lowest since 1972). The minimum monthly mean temperature is 16.5oC in July and the maximum monthly mean temperature is 34.9oC in February (BOM, 2011).

6.2.2 Seed procurement and germination

Seeds of Dalbergia cochinchinensis were supplied by the Royal Forest Department of Thailand. Seeds of uniform size and with no signs of physical damage were selected by visual examination for germination or seed coating. Seeds of D. cochinchinensis were pretreated by soaking in tap water overnight. The pre-treated seeds were surface

Murdoch University 2011 104

Chapter 6: Selection of N-fixing bacteria for Dalbergia cochinchinensis

sterilized by immersing them in 95% (v:v) ethanol for 30 s and then immersing in sodium hypochlorite (3%, v:v) for 4 min followed by washing in five changes of sterile distilled water. Twenty five seeds were evenly spread on sterile distilled water agar (0.9%, w:v) plates (90 mm diameter Petri dishes) and allowed to adhere to the agar for approximately 30 min so that the seeds will not fall from the agar when the plates were inverted in the incubation room. The inversion position of the plates is important as it allow the roots of seedlings to grow out of the agar. Germination plates were sealed with laboratory film (Parafilm®, Pechiney Plastic Packaging, Menasha, WI, USA) and completely wrapped in aluminum foil to exclude light. The germination plates were incubated at 28ºC. Seed germination was checked after four days. After seven days, when the majority of seeds had germinated with average radicle length of 5 mm, they were transferred into prepared pots (Howieson et al., 1995).

6.2.3 Isolation of N2-fixing bacteria and production of broth culture

Root nodules of D. cochinchinensis were collected from two-year old D. cochinchinensis trees planted in Kbal Chhay Watershed Protected Area, Cambodia (Chapter 4). The nodules were collected in early January 2010 and were kept in sterile plastic containers on top of cotton wool over silica gel until isolation in March 2010. Bacteria were isolated following the method of Somasegaran and Hoben (1994). The desiccated nodules of D. cochinchinensis were rehydrated by placing them into small beakers with sterile water, and the beakers were placed in a refrigerator overnight to allow the nodules to imbibe the water. The intact, undamaged nodules were immersed in 95% ethanol for 30 s followed by soaking in a 3% (v:v) solution of sodium hypochlorite for 4 min. The nodules were rinsed in five changes of sterile distilled water. Each surface sterilized nodule was crushed with a sterile wooden applicator stick in a large drop of sterile water in a sterile Petri dish; and the nodule suspension was streaked on a TY- agar [1 L of DI water, 12 g of agar (BBLTM Agar, grade A, Dickinson and Company, Sparks, MD, USA), 5 g of peptone (BactoTM Tryptone, Dickinson and Company, Sparks, MD, USA), 3 g of yeast extract (BBLTM Yeast Extract, Dickinson and

Company, Sparks, MD, USA) and 0.87 g of CaCl2.2H2O] plate using a sterile wooden stick. The plates were wrapped with laboratory film and incubated under temperature of 28ºC. The plates were checked for bacteria growth daily from the third day. Repeat isolations from the first plates were made until pure cultures were obtained. Gram

Murdoch University 2011 105

Chapter 6: Selection of N-fixing bacteria for Dalbergia cochinchinensis

staining was undertaken to check for the purity of the rhizobial bacteria. Three isolates were selected and coded as DS1, DS2 and DR2 (Table 6.2). Broth culture was produced in TY-medium (Somasegaran and Hoben, 1994) and shaken at 200 rpm at 28ºC for four days. Before inoculation to the seeds, the number of viable bacteria was determined (Table 6.2) (Somasegaran and Hoben, 1994).

6.2.4 Seed coating technique and counting of bacteria on coated seeds

Three materials were used for seed coating (Table 6.1). The method of seed coating was modified from Somasegaran and Hoben (1994). For each isolate, 100 seeds were surface sterilized (Section 6.2.1) and air-dried in a laminar flow hood. Sterile seeds were placed in a sterile container and were sprayed liberally with 2 mL of gum arabic solution (30%, w:v). The solution was prepared by heating 100 mL of distilled water to near boiling and adding in small lots of granular gum arabic (1 to 2 g) while continuously stirring the mixture until a total of 30 g was added. The container was then gently shaken until the seeds were uniformly coated. Gamma irradiated peat was added onto the sticky seeds and the container was shaken gently until the seeds were uniformly coated with peat. The content in the container was sprayed liberally with a broth culture of bacteria. Peat was applied as needed if the content was too damp. Finally, CaCO3 was added and the container was shaken until the seeds were uniformly coated with CaCO3 (Figure 6.1). The coated seeds were air-dried in a laminar flow hood for 1 h and then kept in a sterile container under temperature of 4ºC until the next day when used. The quantity of substrate taken up by a seed was ca. 160% of the seed weight and the pH (water) of the substrate was 7.1.

Table 6.1 Materials used for coating seeds of Dalbergia cochinchinensis

Material Function Supplier Gamma irradiated peat Carrier of bacteria BeckerUnderwood Australia (www.beckerunderwood.com)

Calcium carbonate (CaCO3) powder Coating material Unimin Australia Limited (26 Tomlinson Road, Welshpool, WA 6106) Gum arabic, in the form of graininess Adhesive Astarte Earthwise (St Marys South, NSW of normal household sugar 2760 Australia, www.earthwise.net.au)

Murdoch University 2011 106

Chapter 6: Selection of N-fixing bacteria for Dalbergia cochinchinensis

Gamma irradiated peat + broth Coating material: CaCO3 culture of N2-fixing bacteria Gum arabic

Seed

Figure 6.1 Schematic presentation of a cross section of the seed coating technique for Dalbergia cochinchinensis showing layers of substrates.

Samples of 10 coated seeds from each treatment were used for determining the number of bacterial cells. Each sample was divided into two subsamples of five seeds each. Each subsample was transferred into a 1.5 mL microcentrifuge tube containing 1 mL of saline (NaCl). The tubes were shaken in a moto-vortex mixer for 5 min to remove the inoculum off the seeds. A serial dilution from 10-1 to 10-6 was made for each subsample, and then 20 μL of each dilution was plated on TY-agar plates using the spread plate method (modified from Somasegaran and Hoben, 1994). The plates were incubated under temperature of 28°C, and the number of viable rhizobial colonies was counted after 7 to10 days (Table 6.2).

Table 6.2 Detailed treatments used in the glasshouse experiment for Dalbergia cochinchinensis

Treatment code Description of treatment Number of Bradyrhizobium cells T1 Seed coating with DS1a 14 000 seed-1 T2 Seed coating with DS2 3 000 seed-1 T3 Seed coating with DR2 4 000 seed-1 T4 Broth culture of DS1 21 × 107 seedling-1 T5 Broth culture of DS2 18 × 107 seedling-1 T6 Broth culture of DR2 19 × 107 seedling-1 T7 N+ control - T8 N- control - a DS1, DS2 and DR2 are codes given to isolates from Dalbergia cochinchinensis.

6.2.5 Potting, maintenance and harvest of plants

Thirty two black plastic pots with a size of 14 × 16 cm (base diameter × height) were used. Pots and other potting materials such as tubes and caps were sterilized by soaking in a 10% (v:v) solution of sodium hypochlorite for four days and then rinsed in several changes of boiling water. The pot was filled with a pasteurized (65ºC for 3 h) mix of river

Murdoch University 2011 107

Chapter 6: Selection of N-fixing bacteria for Dalbergia cochinchinensis

and yellow sand (1:1, v:v). The filled pots were steam-pasteurized again at 65ºC for 3 h. Three drainage holes were made at the bottom of the pots. The pH (water) of the potting mix, after pasteurizing, was 5.85. Nitrogen in the potting mix sand was flushed with boiling distilled water (Howieson et al., 1995).

A sterile watering tube (15 × 150 mm) was inserted in the middle of the pot to approximately two-thirds the soil depth and was covered with a sterile cap. Three holes were made in the soil to a depth of 0.5 cm using a sterile wooden stick. For treatments T1, T2 and T3, six coated seeds were sown in each pot, two seeds per hole, and they were then covered with the potting mix to the seed thickness. For treatments, T4, T5 and T6, one pre-germinated seed (with 5-10 mm radical root) was transferred into each hole using sterilized forceps. One milliliter of four-day old broth culture of bacteria was inoculated to the seed and the hole was filled with the potting mix. Treatments T7 and T8 were conducted the same way as T4, except that there was no inoculation with broth culture. When cotyledons were evident above the soil surface, seedlings in the seed coating pots were thinned to three per pot using sterile scissors to cut the seedlings at the soil surface, and the potting mix was covered liberally with sterile perlite (autoclaved at 121ºC for 20 min) to reduce the risk of contamination. Pots were randomly placed on a bench in a glasshouse (Figure 6.2) from late April to July 2010 (16 weeks) with average daily temperature ranging from 23 to 30ºC and minimum temperature at night of ca. 15ºC. All seedlings were watered three times a week, through the tube with 40 mL of sterile deionized water. After three weeks, all treatments were given weekly 40 mL of sterile nutrient solution consisting of macro- and micro- elements, as in Table 6.3. A working solution, the combination of all macro- and micronutrients,

Table 6.3 Chemicals used in the production of nutrient solutions (modified from Howieson et al., 1995) for the glasshouse experiment with Dalbergia cochinchinensis

Macronutrient Concentration (mg L-1) Micronutrient Concentration (µg L-1)

MgSO4.7H2O 306.9 H3BO3 92.6

KH2PO4 67.9 Na2MoO4.2H2O 3.6

K2SO4 437.6 ZnSO4.7H2O 107.6

FeEDTA 62.4 MnSO4.H2O 8.4

CaSO4.2H2O 204 CoSO4.7H2O 28.1

CuSO4.5H2O 25

Murdoch University 2011 108

Chapter 6: Selection of N-fixing bacteria for Dalbergia cochinchinensis

except CaSO4.2H2O, was prepared and then autoclaved at 121ºC for 20 min. The calcium solution (CaSO4.2H2O)was autoclaved separately at 121ºC for 20 min. The two solutions were then combined. Additionally, the nitrogen plus (N+) pots received 40 mL of potassium nitrate solution (5 g KNO3 in 1 000 mL of deionized water) per week.

Figure 6.2 Arrangement of pots of Dalbergia cochinchinensis in the glasshouse experiment.

6.2.6 Experimental design, data collection and analysis

A completely randomized design (CRD) experiment consisting of eight treatments (Table 6.2) with four replicated pots was established. Data on the number of root nodules and plant dry weight were collected at 16 weeks. Roots and shoots were oven dried at 60ºC for 48 h to constant weight before weighing. Data analyses were carried out using SPSS statistical package Version 15.0. For all tests, the probability p ≤ 0.05 was used. One-way ANOVA was used, and then significant differences between treatments were determined using Duncan’s Multiple Range Test. Data on numbers of root nodule were square-root transformed (Gomez and Gomez, 1984) to fit the normal distribution requirement before applying statistical analysis. The descriptive statistics presented in the table are for untransformed data.

6.2.7 DNA amplification, isolation and sequencing

6.2.7.1 Polymerase chain reaction-fingerprinting

Nine isolates from the DR2 inoculated bacteria (five from the seed coating treatment: T3.6.3, T3.6.1, T3.2, T3.10.3 and T3.7.2 and four from the broth culture treatment: T6.5.1, T6.5.2, T6.1.1 and T6.4) and the reference strains, DR2, DS1 and DS2, were prepared for polymerase chain reaction (PCR) amplification. The bacterial DNA were amplified using RPO1 primer (5'-AAT TTT CAA GCG TCG TGC CA-3') (Richardson

Murdoch University 2011 109

Chapter 6: Selection of N-fixing bacteria for Dalbergia cochinchinensis

et al., 1995). The PCR amplification were carried out in 20 µL reaction volumes containing 1 µL of bacterial cell adjusted to optical density (OD) 6 at 600 nm (Spectrophotometer Hitachi U-1100), 4 µL of 5 × PCR polymerisation buffer (Fisher Biotech Australia, www.fisherbiotech.com), 11.1 µL of double distilled water, 2.4 µL of

25 nM MgCl2, 1 µL of 50 nM RPO1 primer, and 0.5 µL Taq DNA polymerase (Fisher Biotech Australia, www.fisherbiotech.com). The amplifications were performed in an iCycler thermal cycler (Bio-Rad Laboratories, Hercules, CA, USA), which were programmed as follows: 94ºC for 5 min, followed by 5 cycles at 94ºC for 30 s, 50ºC for 20 s and 72ºC for 90 s. These were followed by 30 cycles at 94ºC for 30 s, 50ºC for 20 s and 72ºC for 90 s with final extension at 72ºC for 5 min.

The PCR products were examined on 1.5% agarose (Bio-Rad Laboratories, Hercules, CA, USA) gels in 1 × TAE buffer (w:v) [diluted from 50 × TAE buffer in 1 L consisting of 242 g Tris, 57.1 mL glacial acetic acid (GAA), and 18.612 g EDTA, the pH of buffer adjusted to 8.0 by GAA] and visualised using 10 × SYBR® Safe DNA gel stain (invitrogenTM, http://www.invitrogen.com). The PCR products were electrophoresed at 75 V for 90 min. Images were captured using the Quantity One software package (Bio-Rad Laboratories, Hercules, CA, USA). The step by step methodology of PCR amplification is given in Appendix V.

6.2.7.2 Partial 16S rRNA sequencing

Two isolates, T3.2 and T6.5.1, representing seed coating and broth culture inoculation, respectively, showing the same band signal as the reference strain, DR2, were selected for partial 16S rRNA gene amplification and sequenced to identify their identity according to the procedure described by Yanagi and Yamasato (1993). The partial 16S rRNA gene was amplified in both directions using the forward primer 20F (5'-AGT TTG ATC CTG GCT CA-3') and the reverse primer 1540R (5'-AAG GAG GTG ATC CAG CCG CA-3') at a concentration of 50 nM. Detailed methods of gene amplification, DNA extraction and purification, DNA sequencing reaction and post-reaction purification are given in Appendix V.

Partial 16S rRNA sequencing of PCR products ST1, ST2 and ST3 (forward direction of DR2, T3.2 and T6.5.1, respectively) and ST5, ST6 and ST7 (reverse direction of DR2, T3.2 and T6.5.1, respectively) was undertaken by the Western Australian State

Murdoch University 2011 110

Chapter 6: Selection of N-fixing bacteria for Dalbergia cochinchinensis

Agricultural Biotechnology Centre (SABC) using an AB Applied Biosystems (HITACHI) 3730xl DNA Analyser. Partial 16S rRNA sequences were assembled and checked with DNA Baser v2.91.5. Comparison of the partial 16S rRNA nucleotide gene sequences (with the homologue nucleotide sequences in the GenBank data base was performed using BLASTN 2.2.24+ (National Center for Biotechnology Information).

6.3 Results

6.3.1 Effective bacteria for Dalbergia cochinchinensis

Analysis of variance indicated that there was a significant effect of isolate on the number of root nodules (Table 6.4), root dry weight, shoot dry weight and total dry weight. Among the three isolates tested, DR2 was effective, and both methods of delivering the bacteria (seed coating and broth culture) were equally effective in all parameters tested. Seed coating or broth culture from DR2 improved total dry weight by ca. 150% over the N- control (Figures 6.3 and 6.4), but they were not as effective as the N+ control (only 60%). The other two isolates, DS1 and DS2, produced some root nodules with both broth culture and seed coating, but these bacteria were not effective. Root nodules were mainly attached

Table 6.4 Number of root nodules of Dalbergia cochinchinensis at 16 weeks as effects of inoculation treatment

Treatment Number of nodules per plant T1: Seed coating with DS1 11.25 ± 2.31 b T2: Seed coating with DS2 13.75 ± 2.14 b T3: Seed coating with DR2 25.00 ± 2.15 a T4: Broth culture of DS1 8.17 ± 2.56 b T5: Broth culture of DS2 9.92 ± 2.41 b T6: Broth culture of DR2 23.92 ± 3.47 a T7: N+ control 0.00 ± 0.00 c T8: N- control 0.00 ± 0.00 c Analysis of variance Source d.f. F Sig. Treatment 7 1.834 0.000 Error 24

Values are means (n = 4) ± S.E. Means followed by the same letter are not significantly different according to Duncan’s Multiple Range Test at p ≤ 0.05. The significant ANOVA probability is marked in bold.

Murdoch University 2011 111

Chapter 6: Selection of N-fixing bacteria for Dalbergia cochinchinensis

to the tap roots (Figure 6.5a - d). The majority of nodules were small (< 2 mm length), and the shape was generally globose. The outside colour was yellow and the internal colour was pink.

T1: seed coating with DS1 T5: broth culture of DS2 Root dry weight Shoot dry weight T2: seed coating with DS2 T6: broth culture of DR2 Total dry weight 300 T3: seed coating with DR2 T7: N+ control A 250 T4: broth culture of DS1 T8: N- control B B 200 A B B 150 Dry weight (g) B B a C C C 100 b C b C c C C c c C 50 c C c C 0 T1 T2 T3 T4 T5 T6 T7 T8

Figure 6.3 Root dry weight [F (7, 24) = 1.159, p < 0.001], shoot dry weight [F (7, 24) = 2.350, p < 0.001] and total dry weight [F (7, 24) = 1.802, p < 0.001] per plant of Dalbergia cochinchinensis showing effects of N2-fixing bacteria after 16 weeks in the glasshouse experiment. In the same parameter, columns followed by the same letters (lower, upper or upper in bold) are not significantly different at p ≤ 0.05.

Figure 6.4 Effect of N2-fixing bacteria on growth of Dalbergia cochinchinensis in the glasshouse experiment at 16 weeks. Comparisons of treatments T1-T6 to the un-inoculated controls, N- (left) and N+ (right).

Murdoch University 2011 112

Chapter 6: Selection of N-fixing bacteria for Dalbergia cochinchinensis

b c a d

Figure 6.5 Root nodules of Dalbergia cochinchinensis from the glasshouse experiment at 16 weeks. (a), Overview of root nodules from the isolate DR2. (b) - (d), Close view of root nodules from plants in (a). A bar in (a) represents 10 mm and bars in (b) - (d) represent 2 mm.

6.3.2 Molecular characterisation and identification of isolates

The electrophoretic patterns of PCR products showed that bacteria from DR2 inoculation (seed coating or broth culture) shared the same pattern as those in the reference isolate DR2 (lanes 1 to 10) (Figure 6.6) suggesting that they are the same bacteria. Partial 16S rRNA sequencing of the reference isolate DR2 and isolates T3.2 and T6.5.1, representing bacteria from seed coating with and broth culture of DR2, respectively, revealed that the strain is close to Bradyrhizobium elkanii Kuykendall et al., 1993 (Table 6.5).

6.4 Discussion

The aim of this study was to identify effective N2-fixing bacterial strains for D. cochinchinensis. This study indicated that the isolate DR2 was effective for D. cochinchinensis because nodules formed from it were healthy in appearance, pink inside, and plant growth was promoted. In the literature, data concerning N2-fixing bacteria or effective strain for D. cochinchinensis are lacking. Although the tree species has high economic potential (Chapter 4, Section 4.2.2), no effort has been made to

Murdoch University 2011 113

Chapter 6: Selection of N-fixing bacteria for Dalbergia cochinchinensis

L 1 2 3 4 5 6 7 8 9 10 11 12 L 10 000

3 000 2 500 2 000 1 500

1 000

750

500

250

Figure 6.6 Electrophoretic patterns of PCR products amplified with primer RPO1 of bacteria that nodulated Dalbergia cochinchinensis. Lane L, 1kb DNA ladder weight marker from 250 to 10 000 bp; lanes 1 to 5 bacteria from seed coating treatment with DR2 (strains T3.6.3, T3.6.1, T3.2, T3.10.3 and T3.7.2, respectively); lanes 6 to 9 bacteria from broth culture treatment of DR2 (strains T6.5.1, T6.5.2, T6.1.1 and T6.4, respectively); lane 10, reference isolate DR2; and lanes 11 and 12, isolates DS1 and DS2, respectively. Note: bacteria from lanes 3 and 6 were selected for partial 16S rRNA sequencing.

Table 6.5 Partial 16S rRNA sequences of the reference isolate DR2 and two isolates from DR2 inoculation with Dalbergia cochinchinensis, T3.2 (seed coating) and T6.5.1 (broth culture) compared with Gene Bank

Isolate Strain No. of Direction Gene Bank data base Homology Accession bases (%) No. DR2 ST1 509 Forward Bradyrhizobium elkanii strain 100 HQ171486.1 PRNB-22 16S rRNA gene ST5 522 Reverse Bradyrhizobium elkanii strain 100 HQ171487.1 PRNB-23 16S rRNA gene T3.2 ST2 364 Forward Bradyrhizobium elkanii strain 100 HQ171486.1 PRNB-22 16S rRNA gene ST6 653 Reverse Bradyrhizobium elkanii strain 100 GU433451.1 CCBAU 51010 16S rRNA gene T6.5.1 ST3 388 Forward Bradyrhizobium elkanii strain 100 EF394150.1 CCBAU 53142 16S rRNA gene ST7 811 Reverse Bradyrhizobium elkanii strain 100 GU433457.1 CCBAU 23090 16S rRNA gene

Murdoch University 2011 114

Chapter 6: Selection of N-fixing bacteria for Dalbergia cochinchinensis

promote seedling growth using its symbiotic association with N2-fixing bacteria.

Rasolomampianina et al. (2005) characterized 68 N2-fixing bacteria associated with eight Dalbergia species endemic to Madagascar. They found that Dalbergia formed root nodules with a diversity of N2-fixing bacteria including Bradyrhizobium, Mesorhizobium, Rhizobium, Azorhizobium, Phyllobacterium, Burkholderia and Ralstonia in its natural habitats. However, 70% of the bacterial strains isolated belonged to the genus Bradyrhizobium, and the remaining 30% was spread across the other 6 genera. Bradyrhizobium elkanii nodulates a variety of hosts, ranging from annual plants to shrubs and trees (Table 6.6). The species occurs in soils under natural forests and eucalypt plantations in Brazil (Pereira et al., 2008). Its geographical range, however, is not limited to tropical and sub-tropical regions (Table 6.6), as it is reported from European Mediterranean ecosystems (Cardinale et al., 2008; Rodríguez-Echeverría, 2010).

Table 6. 6 Some hosts nodulated by Bradyrhizobium elkani, based on Gene Bank and literature

Host plant Life form Region/country Gene Bank accession no. or reference Pongamia pinnata (L.) Pierre Tree Tropical and HQ171486.1, HQ171487.1 sub-tropical Glycine max (L.) Merr. Annual herb Tropical and EF394150.1 sub-tropical Arachis hypogaea L. and Lablab Annual herbs Southern China GU433451.1, GU433457.1 purpureus (L.) Sweet Pterocarpus indicus Willd. Tree Tropical region Lok et al., 2006; Manassila et al., 2007 Acacia auriculiformis A. Cunn. ex Trees Tropical region Manassila et al., 2007 Benth., A. mangium A. Cunn. ex Benth. Milletia leucantha Kurz., Xylia xylocarpa (Roxb.) Taubert Cytisus grandiflorus (Brot.) DC., Shrubs and Mediterranean Rodríguez-Echeverría, Cytisus scoparius (L.) Link, tree ecosystems of 2010 Acacia longifolia (Andrews) Willd. Portugal

In this glasshouse study, only one strain of three tested was effective. Identification of nodule crushes from this effective strain, based on RPO1 fingerprinting confirmed

Murdoch University 2011 115

Chapter 6: Selection of N-fixing bacteria for Dalbergia cochinchinensis

identity of the inoculant organism in the nodules, and thus nodulation was not due to contamination. Partial 16S rRNA sequencing revealed that all the rhizobia isolated were close to Bradyrhizobium elkanii. Considering the significant 150% increase in plant total dry weight, it is worth evaluating the effectiveness of DR2 in the field (Chapter 7). The perceived inferiority of DR2 compared to the N+ control may be because the inoculated seedlings were not given starter inorganic N and thus plants are likely to have been deficient in N before root nodules formed and began fixing N. Nodulation takes a number of weeks, for example, 40 days for the herbaceous legume Glycine wightii (Wight & Arn.) Verdc. (Whiteman, 1972).

This glasshouse experiment also showed that delivery of bacteria to seeds by seed coating is as effective as inoculation with broth culture despite the fact that the number of bacteria coated on the seed was less than 1% of the cells provided in broth culture. The coated seeds were sown the day after coating and it is not known how long the inoculum can survive. Forestier et al. (2001) reported from a glasshouse experiment that inoculation with a liquid culture of Rhizobium (1 mL) to the roots of one-week old seedlings of Leucaena leucocephala did not result in any difference in nodulation and growth from coating of the same bacteria with pre-germinated seeds. However, they did not report the number of cells used. In this glasshouse experiment, the total weight of carrier on the seed constituted about 160% of the seed weight. This was significantly higher than the rates used in previous studies, namely 50 to 90% of seed weight (Brockwell, 1962; Asano, 1996). However, the high dose deliberately ensures a high number of bacterial cells per seed. The pH of the carrier in this study (pH = 7.1) was in the range suggested by Strijdom and van Rensburg (1981).

Delivery of bacteria to seeds through seed coating/pelleting is gaining increased attention in reforestation studies by direct seeding (e.g., Homchan et al., 1989; Thrall et al., 2001; Thrall et al., 2005). However, commercial seed coating is yet to be widely adopted. One of the limiting factors in handling of bacteria on seed is the requirement to sow seed as soon as possible in order to ensure a high survival rate of the bacteria (e.g., Thrall et al., 2001), otherwise the coated seed should be stored at 4ºC to maintain survival of the bacteria (Temprano et al., 2002). Previous research has shown that non- refrigerated storage is not suitable for the survival of rhizobia in peat inoculants

Murdoch University 2011 116

Chapter 6: Selection of N-fixing bacteria for Dalbergia cochinchinensis

(Temprano et al., 2002). In addition, bacteria in non-sterile peat do not maintain their viability as well as those in sterile peat (Temprano et al., 2002). Avoiding exposure of coated seeds or inocula to high temperature or dry conditions during storage, transportation and planting may not always be possible, especially in tropical conditions. The requirement for refrigerated storage presents a major difficulty in promoting seed coating as a means to deliver bacteria to seeds in warm climates. Searching for alternative carriers to peat is ongoing (Chapter 2, Section 2.5.4.1), and a carrier that can maintain survival of bacteria in hot tropical climates is needed. Chao and Allexander (1984) reported that only 15% of cells of Rhizobium meliloti Dangeard and R. phaseoli Dangeard inoculated into mineral soil-based inoculants (Lima silt loam, pH (water) = 7.2), were viable at 37°C after 27 days. However, if the inoculum was coated on seeds, the survival rate of rhizobia might be lower than this considering the rapid loss of moisture from the small substrate that can be coated on seed. However, coated seeds can be stored in moisture proof containers. Soil-based inoculants, such as termite mound soils which have a high percentage of clay (Semhi et al., 2008) and are available locally in tropical regions, deserve investigation.

6.5 Conclusions

This glasshouse experiment showed that one strain, close to Bradyrhizobium elkanii, was effective for D. cochinchinensis. Furthermore, delivery of bacteria to seeds through seed coating was as effective as inoculation with broth culture under glasshouse conditions. The effectiveness of this isolate under field conditions in a tropical environment will be investigated in Chapter 7.

Murdoch University 2011 117

Chapter 7: Improving reforestation success of high-value tree species in Cambodia

CHAPTER 7

IMPROVING REFORESTATION SUCCESS BY DIRECT SEEDING OF HIGH- VALUE TIMBER SPECIES ON DEGRADED FOREST LAND: FIELD TRIALS IN CAMBODIA

A bole of Sindora cochinchinensis (arrow) in a community forest in Kampong Cham province (photo courtesy of Cambodia Tree Seed Project, 2006).

Murdoch University 2011 118

Chapter 7: Improving reforestation success of high-value tree species in Cambodia

7.1 Introduction

In Chapter 6, an effective strain of N2-fixing bacteria, DR2, belonging to the species Bradyrhizobium elkanii, was identified for Dalbergia cochinchinensis Pierre under glasshouse conditions. Two questions arise: does this strain of bacteria maintain its effectiveness under field conditions and can it be delivered with seed to enhance restoration in Cambodia? This Chapter therefore investigates this strain in a field trial established in the Kbal Chhay Watershed Protected area of Preah Sihanouk province, Cambodia.

Inoculation of N2-fixing bacteria in reforestation, as a treatment to increase the success of direct seeding, has not been widely practiced. However, a few studies indicate that inoculation of seeds with effective strains of Rhizobia prior to direct seeding can result in significant improvement in early plant growth (Homchan et al., 1989; Thrall et al., 2005). Peat has been widely used as a Rhizobium carrier (Chapter 2). However, peat- based inoculants do not tolerate high temperature (Chao and Allexander, 1984), especially when used in a coating on seeds (Davidson and Reuszer, 1978). Thus, this substrate may not be suitable for handling bacteria in tropical environments. Therefore, alternative carriers have been proposed, for example, mineral soil to enhance survival of Rhizobium (Chao and Allexander, 1984). In reforestation by direct seeding, two characteristics of a bacterial carrier are needed: effectiveness in the promotion of bacterial survival and convenience in handling in the field, especially when the bacteria are to be delivered separately from the seeds. Therefore, the present study also investigates the effectiveness of methods of handling N2-fixing bacteria on and off seed. The hypothesis for this study is: survival and growth of direct seeding of D. cochinchinensis is enhanced by inoculation with effective N2-fixing bacteria.

The second part of this study investigates the effect of inorganic fertilizer with and without a water retention polymer. This is a follow-up of Experiment 2, Chapter 5, in a field with contrasting environmental conditions from those in Thailand (Chapter 5). These trials were conducted at a degraded forest site with high rainfall and high intensity of weed competition. Weed control was included as an experimental factor. The hypotheses for this experiment are: 1) Growth performance is different for each high-value species in reforestation by direct seeding; 2) Growth and survival of indigenous species from direct seeding are enhanced by application of inorganic

Murdoch University 2011 119

Chapter 7: Improving reforestation success of high-value tree species in Cambodia fertilizer with or without a water retention polymer; and 3) Weeds may not be a serious problem for direct seeding of indigenous species provided that weeding is carried out once at three months after trial establishment.

7.2 Materials and methods

7.2.1 Study site

The study site was located in Kbal Chhay Watershed Protected area of Preah Sihanouk province, Cambodia about 200 km southwest of the capital city, Phnom Penh (Figure 7.1). The area has an altitude of 110 m a.s.l., a latitude of 10o 38' 7'' N and a longitude of 103o 35' 23'' E (point at the northeast corner). Kbal Chhay, covering an area of 6 202 ha, is protected by sub-decree No. 76 dated 2 November 1997. The trial site was part of the reforestation area (100 ha) undertaken by the Forestry Administration in 2010 where seedlings of Afzelia xylocarpa (Kruz.) Craib, D. cochinchinensis, Dipterocarpus alatus and Hopea odorata were planted in open area (Pers. comm., Head of the Kbal Chhay Watershed Protected Area, 2010).

The climate in the study area is governed by the monsoon and is characterized by two distinct seasons, the wet season, occurring from May to October and the dry season, occurring from December to April. The nearest meteorological station in Preah Sihanouk city, 10 km to the southeast, receives an annual rainfall of 3 256 mm (average from 1983 to 2009) (Figure 7.2). The mean annual temperature is 27.5ºC (average from 2000 to 2009), the hottest month is April with a mean maximum temperature of 34.3ºC and the coldest month is January, with a mean minimum temperature of 21.1ºC. The rainfall in 2010 was 2 150 mm (until the end of the wet season, November) much less than the average from 1983 to 2009 (Hydro-meteorological Department, 2010).

The area was formerly covered by wet-evergreen forest, but it was cleared during the 1990s as a result of illegal logging followed by land speculation. At the time of trial establishment, the area was covered with a mixture of grasses, shrubs and secondary forest regrowth, including pioneer and climax species,

Murdoch University 2011 120

Chapter 7: Improving reforestation success of high-value tree species in Cambodia

N Kbal Chhay Watershed Protected Area

Legend:

National road

Other road Kbal Chhay boundary

1km

Figure 7.1 Location of the trial site inside the Kbal Chhay Watershed Protected Area, Preah Sihanouk province, Cambodia.

Rainfall 2010 (end of November) Rainfall 1983 to 2009 Mean temperature 2000 to 2009 Absolute minimum temperature 2000 to 2009 Absolute maximum temperature 2000 to 2009

800 40

C)

°

(

600 30 400 20 200 10 Rainfall (mm) 0 0 Temperature Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Month

Figure 7.2 Monthly rainfall and temperature (mean, absolute maximum and absolute minimum) of Preah Sihanouk province, Cambodia (Hydro-meteorological Department, 2010).

Murdoch University 2011 121

Chapter 7: Improving reforestation success of high-value tree species in Cambodia with an average height of about 5 to 10 m. Woody plants were scattered among dense-tall grass, Saccharum spontaneum L., up to 3 m in height, and Imperata sp. grass. The main woody species were Bauhinia bassacensis Pierre ex Gagnep, Cochlospermum religiosum (L.) Alston, Irvingia malayana Oliv. ex Benn., Lithocarpus elegans (Blume) Hatus ex Soepadmo, Madhuca cochinchinensis (Pierre ex Dubard) H. J. Lam, Memecylon laevigatum Blume, Miliusa velutina Hook f. & Thomson, Mitrella mesnyi (Pierre) Bân, Prismatomeris tetrandra (Roxb.) Schum and Willughbeia edulis Roxb.

The soil of the study area is characterized by loamy sand derived from red-yellow podzols (Crocker, 1963). Around 10% of the area is covered with lateritic stone outcrops. The top soil, which is ca. 20 cm deep, is dark brown (Figure 7.6c). Two soil samples were collected randomly and bulked, and subsamples from the bulk sent to the National Agriculture Laboratory in Phnom Penh for physical and chemical analyses (Table 7.1).

Table 7.1 Physical and chemical properties of the soils in the experimental site in Preah Sihanouk

province, Cambodia

Soil property Value 0-20 cm 21-80 cm Sand (%) 78 70 Silt (%) 9 7 Clay (%) 13 23

pH (H2O ratio 1:25) 4.7 4.6 1 Total C (%o) 19.35 6.35 2 Total N (%o) 1.75 0.70 3 Total P (%o) 0.213 0.157 Available P (mg/kg)4 37 33 CEC (cmol/kg) 5 9.5 12 OM (%)6 3.2 1 Ca (cmol/kg)7 1.80 1.20 Mg (cmol/kg) 7 1.00 0.80 Na (cmol/kg) 7 0.43 0.43 K (cmol/kg) 7 0.38 0.06

The values are from single measurement. Methods for chemicals analysis: 1 Black; 2 Kjeldalh; 3Nitric 4 5 digest; (Olsen) Extracted with 0.5 M NaHCO3 + 0.5 M FNH4, pH = 8.5; Extracted with 1 M 6 7 CaCl2.2H2O, pH = 7; Walkley-Black acid digestion; Extracted with 1 M Ammonium acetate, pH = 7.

Murdoch University 2011 122

Chapter 7: Improving reforestation success of high-value tree species in Cambodia

7.2.2 Procurement of materials

7.2.2.1 Seeds

Four high-value timber tree species, Afzelia xylocarpa, Dalbergia cochinchinensis, Sindora cochinchinensis and Xylia xylocarpa, indigenous to Cambodia and Southeast Asia, were selected for the study. Reasons for selection of these species outlined in Chapter 5. Seeds were obtained from a number of sources (Table 7.2).

Table 7.2 Sources of seeds used in the direct seeding experiments in Preah Sihanouk province, Cambodia

Species Year Seed source Use in this study Afzelia xylocarpa 2006 Pursat province, Cambodia Experiment 2 Dalbergia cochinchinensis 2007 Royal Forest Department, Thailand Experiment 1 2009 Siem Reap province, Cambodia Experiment 2 Sindora cochinchinensis 2006 Pursat province, Cambodia Experiment 2 Xylia xylocarpa 2010 Pursat province, Cambodia Experiment 2

7.2.2.2 N2-fixing bacteria

Fresh cultures of three Bradyrhizobium isolates, DR2, DS1 and DS2 (see Chapter 6), were stored in 20% glycerol at -20ºC from March to April 2010. The bacteria were re- cultured in late April for preparation of seed coating and peat-based inoculum. All the three isolates used in the glasshouse (Chapter 6) were used in this study as the experiment had not been finished and the best strain had not been determined when the trial was established. It was expected, therefore, that using a multi-strain inoculum would be better than a single-strain (Thrall et al., 2001).

Three methods of delivering bacteria to seeds were tested: seed coating, clay-based inoculum and peat-based inoculum. The detailed method for seed coating presented in Chapter 6 (Section 6.2.3) was followed except that broth culture of the three isolates was mixed with equal cell counts during the seed coating. The quantity of substrate taken up by a seed and pH of the substrate was not determined. The number of cells per seed was determined (Chapter 6, Section 6.2.3) one day after seed coating (Table 7.3). The coated seeds (Figure 7.3b) were stored under temperature of 4ºC until transported to the field.

Murdoch University 2011 123

Chapter 7: Improving reforestation success of high-value tree species in Cambodia

Table 7.3 Number of bacteria per seed and per gram of sterile peat

Isolate Number of cells of Bradyrhizobium Seed coating with mixed culture of isolates DR2, DS1 and DS2 8 × 103 seed-1 Peat-based inoculum DR2 5 × 106 g-1 peat DS1 4 × 106 g-1 peat DS2 5 × 106 g-1 peat

a b

c d

Figure 7.3 (a), Seeds of Dalbergia cochinchinensis before coating. (b), Seeds of D. cochinchinensis coated with N2-fixing bacteria using gamma-irradiated peat as carrier and then pelleted with

CaCO3. (c), Peat-based N2-fixing bacteria, gamma irradiated peat bag inoculated with bacterial broth culture. (d), Clay-based inoculum (or clay balls).

For peat-based inoculum, fresh broth culture (four days) of the three isolates was separately inoculated into gamma-irradiated peat bags in early May 2010. Sterilized distilled water was added as necessary to increase the final moisture of the peat based inoculum to 45% (Temprano et al., 2002). The inoculated bags were incubated under temperature of 28ºC for seven days and then transferred to a cold room (4ºC) before using in the field. Prior to storing, the number of the bacterial colonies (Table 7.3) was

Murdoch University 2011 124

Chapter 7: Improving reforestation success of high-value tree species in Cambodia checked using a spread plate method (Somasegaran and Hoben, 1994). The peat based- inoculum from the three isolates was mixed, in equal proportions by weight, one day before using in the field. The clay-based inoculum (or clay balls) was prepared in the field, one day prior to use, by mixing 1 g of peat-based inoculum with 9 g of fine termite mound clay (the clay was available at the trial site). The mixture was then moistened and balled by hand (Figure 7.3d).

7.2.2.3 Fertilizer and water retention polymer

Two types of chemical fertilizer were used: NPK compound fertilizer (commercially known as AGROPHATE OX-BRAND, 15:15:15 N:P2O5:K2O + 2.5% CaO, 0.4% MgO and 0.02% B) and Urea (46% N). Both were manufactured by Thai Central Chemical Public Co., Ltd. (http://www. tcccthai.com) and were available at the local market. The same water retention polymer as in the trial in Thailand (Chapter 5) was used in this study.

7.2.3 Experimental design and treatments

Experiment 1: This was a 2 × 4 factorial experiment in a randomized complete block design (RCBD) involving two levels of N fertilizer and four options of delivering of N2- fixing bacteria to seeds of D. cochinchinensis (Table 7.4). Therefore, in a block there were eight plots, four of which were allocated randomly to nil N fertilizer, the other four to a plus N fertilizer (Figure 7.4). Each plot in a level of N fertilizer was then assigned with one option of N2-fixing bacteria treatment. Each plot contained 25 planting holes, five rows of five holes each, and spacing between adjacent holes in the same row was 1.5 m and that between adjacent rows was 3 m. Spacing between plots in Blocks 1 and 2 was 3 to 5 m whereas that in Block 3 was 1.5 to 3 m. There were no trees in the buffer areas. There were three replicated blocks in the same site.

Experiment 2: The layout was a split-plot design involving four tree species as sub-plot treatments in a 2 × 3 (or six plots) factorial experiment with two levels of weeding and three levels of fertilizer treatments, as main plots (Table 7.4). Therefore, within a block there were six main plots, arranged in an RCBD, three of which were assigned to “weeding once” and the other three were assigned to “weeding every three months” (Figure 7.4). Each main plot in each level of weeding was then assigned to a fertilizer treatment. Each main plot was subdivided into four sub-plots, each accommodated a

Murdoch University 2011 125

Chapter 7: Improving reforestation success of high-value tree species in Cambodia randomly-arranged tree species. Each sub-plot contained 25 seeding spots. Spacing between adjacent spots in the same row and that between adjacent rows were the same as in Experiment 1. The buffer areas between adjacent blocks were 5-m wide and those between adjacent plots were 3-m wide (Figure 7.4). There were no trees in the buffer areas. There were three replicated blocks in the same site.

Table 7.4 Descriptions of treatments and methods of application in Experiments 1 and 2 undertaken in Kbal Chay Watershed Area, Preah Sihanouk province, Cambodia

Treatment Method of treatment application Experiment 1

N2-fixing bacteria Seed coating Three coated seeds each with a mix of DR2, DS1 and DS2 were buried to ca. 0.5 cm at each seeding spot.

Peat-based N2-fixing bacteria One gram of peat-based inoculum containing an equal mix of DR2, DS1 and DS2 was applied immediately below the seeds.

Clay-based N2-fixing bacteria A 10 g ball of clay-based inoculum containing one gram of the peat- based inoculum described above was placed immediately below the seeds.

Control No N2-fixing bacteria. N fertilizer treatment

With N fertilizer N fertilizer (30 g of urea) was buried to a depth of 10 cm at two sides, each ca. 20 cm, of the seeding spot. Without N fertilizer No N fertilizer application.

Experiment 2 Fertilizer treatment

Fertilizer Inorganic fertilizer 15:15:15 (N:P2O5:K2O) (30 g) was applied at two opposite locations to the depth of 10 cm at about 20 cm from the seeding spot. Fertilizer + polymer A solution of 2.5 g of water retention crystal dissolved in a litre of water was placed 20 cm below the seeds, and the same amount of fertilizer was applied as above. Control No fertilizer or polymer. Weeding treatment Weeding once Weeding was carried out only one time at three months after trial establishment. All weeds in a row were removed using hand hoes (Figure 7.5). Weeding every three months Weeding was carried out every three months in the first year.

Murdoch University 2011 126

Chapter 7: Improving reforestation success of high-value tree species in Cambodia

T1W2-A T1W2-X T3W1-D T3W1-S 4 1

T1W2-S T1W2-D T3W1-X T3W1-A 3 m T2W2-X T2W2-A T1W1-S T1W1-D 5 2 T2W2-D T2W2-S T1W1-A T1W1-X

T2W1-A T2W1-S T 3W2-X T3W2-A 6 3

T2W1-X T2W1-D T3W2-S T3W2-D Exp eriment 2, Block 1 5 m T1W1-D T 1W1-S T2W2-X T2W2-D 4 1

T1W1-A T1W1-X T2W2-S T2W2-A

5.T4N1 1.T1N2 T3W2-X T3W2-A T2W1-X T2W1-A 5 2

6.T2N2 2.T2N1 T3W2-S T3W2-D T2W1-D T2W1-S

7.T4N2 3.T1N1 T3W1-A T3W1-S T1W2-A T1W2-D 6 3 8.T3N1 4.T3N2 T3W1-D T3W1-X T1W2-S T1W2-X

Exp eriment 1, Block 3 Exp eriment 2, Block 2

T3W2-D T3W2-S T1W1-S T1W1-A 4 1 T3W2-X T3W2-A T1W1-D T1W1-X

T2W1-X T2W1-D T1W2-D T1W2-X

5 2 T2W1-S T2W1-A T1W2-A T1W2-A

T3W1-D T3W1-S T2W2-X T2W2-D 6 3 T3W1-A T3W1-X T2W2-A T2W2-S Experiment 2, Block 3 6.T1N2 3.T2N1 1.T4N2

7.T3N1 2.T1N1

8.T2N2 5.T44.T3N12 1.T1N1

Experiment 1, Block 1 7.T4N2 5.T4N1 2.T3N2

8.T2N1 6.T1N2 3.T3N1

Experiment 1, Block 2 4.T2N2

Figure 7.4 Sketch map of the trial in Preah Sihanouk province, Cambodia, showing the arrangement of experiments, blocks and plots/subplots. Experiment 1: N2-fixing bacteria: T1, seed coating; T2, peat-based N2-fixing bacteria; T3, clay-based N2-fixing bacteria; and T4, control. N fertilizer: N1, with N fertilizer; N2, without N fertilizer. In each plot, the prefix numbers 1 to 8 denote the plot number. Experiment 2: Tree species: A, Afzelia xylocarpa; D, Dalbergia cochinchinensis, S, Sindora cochinchinensis; X, Xylia xylocarpa. Fertilizer treatment: T1, fertilizer; T2, fertilizer + polymer; and T3, control. Weeding treatment: W1, weeding once; and W2, weeding every three months. In each plot, numbers 1 to 6 (in the circle) denote the plot number. Dashed lines denote boundaries of the trial, along which a row of X. xylocarpa was planted.

Murdoch University 2011 127

Chapter 7: Improving reforestation success of high-value tree species in Cambodia

1.5 m

1.5 m 1.5 m 3 m

Figure 7.5 Arrangement of direct seeding spots in each plot of Experiment 2. Circles denote seeding spots. Circles with the same colour symbolize a subplot (or a tree species). The green bands are areas where weeding was not undertaken, and the white bands denote hand-weeded areas. Note: A plot in Experiment 1 has a size equal to a sub-plot in Experiment 2.

7.2.4 Site preparation and trial establishment

The land was prepared in mid May 2010 by clearing vegetation and burning debris on site. About half of the area (Experiment 1, Block 3 and Experiment 2, Blocks 1 and 2) was then disc-ploughed once using a tractor with rubber tyres. The other half of the area (Experiment 1, Blocks 1 and 2 and Experiment 2, Blocks 3) was not be able to be ploughed due to the presence of tree stumps (Figure 7.6a), however, hand weeding, using hoes, was undertaken in this part of the trial site in 1.5 m bands (Figure 7.5). The areas between the hand-weeding bands and those between blocks and plots were sprayed, about 10 days before trial establishment, with herbicide, Glyphosate isopropylammonium mixing with a surfactant (commercially known as Mixer) with a rate of 300 ml and 200 mL (respectively) in 20 L of water as directed by the label. Direct seeding was undertaken in the first week of June 2010 after the soil had received some monsoon rains. All seeds were pretreated as required (Chapter 5, Section 5.2.2) before seeding. Direct seeding spots were made by loosening the soil ca. 20 × 20 × 5 cm (width × width × depth) using a hand hoe. Deeper holes (20 cm depth) were made to accommodate the polymer (Table 7.4). In each spot, three seeds were buried to the depth of their thickness. Each seeding spot was marked with a stick of about 1 m to facilitate locating the spot during monitoring and maintenance.

Murdoch University 2011 128

Chapter 7: Improving reforestation success of high-value tree species in Cambodia

a b

c d

e

Figure 7.6 (a), Site preparation by clearing the land and burning the debris on site. (b), Half of the trial site was disc-ploughed once. (c), Soil profile. (d), Water-retention polymer in a seeding hole. (e), Direct seeding of Dalbergia cochinchinensis. Photos were taken in the first week of June 2010.

/ 7.2.5 Maintenance of the field trial

The trial site was not fenced since it is located inside a protected area where domestic animal or human disturbance is not a problem. Two weeding operations (Figure 7.5) were undertaken, at the end of August and end of November 2010, corresponding to three and six months after trial establishment. Care was taken to avoid disturbance to seedlings and the treatments when weeding close to the seedlings. All buffer areas between blocks and plots and the trial site boundaries (3 m width) were hand weeded at six

Murdoch University 2011 129

Chapter 7: Improving reforestation success of high-value tree species in Cambodia months using hoes to minimize fire risk. After two months, inferior seedlings with signs of unhealthy, slow growth, or attacked by insects were cut out at the ground level leaving one healthy seedling per seeding hole.

7.2.6 Data collection and analysis

Survival rate and height were assessed at six months after trial establishment. Data analyses were carried out using SPSS statistical package Version 15.0. For all tests, the probability p ≤ 0.05 was used. For both experiments, Linear (Mixed Models) analysis was used. For post hoc analyses, a multiple comparison test, Sidak, was used to detect differences among means.

Survival rate was calculated based on the number of direct seeding spots with presence of a seedling at the time of assessment as percentage of the total direct seeding spots per plot (Experiment 1) or sub-plot (Experiment 2) (n = 25). Percentages of survival are means across three blocks. They were arc-sin square root transformed (Gomez and Gomez, 1984) prior to statistical analysis to fit the normal distribution requirement. The descriptive statistics presented in tables are untransformed data. During the first six months after trial establishment, it was not possible to include the effect of weeding in the analysis of Experiment 2 (Table 7.4) as the data were collected at the time of the second weeding. Therefore, only data from “weeding every three months” were used for statistical analysis according to a split-plot design with three fertilizer treatments as the main plots and four tree species as the sub-plots. The trial will be continued beyond the period of this PhD and the effect of weeding can then be assessed.

7.3 Results

7.3.1 Experiment 1

After six months, seedling survival rate was not affected by N2-fixing bacteria, N fertilizer or the interaction of N2-fixing bacteria and N fertilizer (Table 7.5). There was a significant block effect on survival rate with survival rate in Block 1 being significantly higher than those in Blocks 2 and 3, by 27% and 65%, respectively (data not presented). Soil fertility could play a role in this outcome due to the presence of ashbeds (discussed in Chapter 3) and charcoal (Chapter 2) in Blocks 1 and 2 left as a result of site

Murdoch University 2011 130

Chapter 7: Improving reforestation success of high-value tree species in Cambodia preparation in which heaps of debris were burned on site (Appendix VIII). Height growth was not affected by N2-fixing bacteria, N fertilizer or the interaction of N2-fixing bacteria and N fertilizer.

Table 7.5 Survival rate and height of Dalbergia cochinchinensis at six months in Experiment 1 as effects of N2-fixing bacteria and N fertilizer in a field trial in Preah Sihanouk province, Cambodia

Survival (%) Height (cm)

N2-fixing bacteria Seed coating 51.33 ± 5.21 15.32 ± 1.77

Peat-based N2-fixing bacteria 50.00 ± 10.57 14.18 ± 1.50

Clay-based N2-fixing bacteria 59.33 ± 9.77 14.05 ± 2.01 Control 63.33 ± 10.90 14.86 ± 2.51

N fertilizer With N fertilizer 54.00 ± 6.83 15.06 ± 1.23

Without N fertilizer 58.00 ± 6.20 14.14 ± 1.44

Analysis of variance

Source d.f. F Sig. F Sig. Block 2 10.243 0.002 3.289 0.067

N2-fixing bacteria 3 0.938 0.449 0.123 0.945 N fertilizer 1 0.320 0.580 0.295 0.596

N2-fixing bacteria × N fertilizer 3 0.305 0.821 2.025 0.157 Error 14

Values are replicated block means (n = 3) ± S.E. The significant ANOVA probabilities are marked in bold.

7.3.2 Experiment 2

After six months, seedling survival rate was significantly affected by treatment with the combination of fertilizer and polymer improving the survival rate by 6% (Table 7.6). There was a significant effect of tree species. Xylia xylocarpa had the highest survival rate and D. cochinchinensis had the lowest rate. The interaction of treatment and tree species was not significant indicating that the survival rate differences among tree species were not affected by the treatments used.

Murdoch University 2011 131

Chapter 7: Improving reforestation success of high-value tree species in Cambodia

Height was not affected by treatment. However, there was a significant effect of tree species on height with A. xylocarpa being the tallest and X. xylocarpa the shortest (Figure 7.7). The interaction of treatment and tree species was not significant, indicating that the height differences among tree species were not affected by the treatments used.

Table 7.6 Survival rate and height growth of high-value timber trees at six months in Experiment 2 as effects of treatment and tree species in a field trial in Preah Sihanouk province, Cambodia

Survival (%) Height (cm) Treatment Fertilizer 85.00 ± 6.11 b 25.40 ± 1.79 Fertilizer + polymer 91.67 ± 4.25 a 23.21 ± 2.01 Control 86.00 ± 6.31 b 22.58 ± 2.49

Tree species Afzelia xylocarpa 96.00 ± 1.49 b 31.36 ± 1.19 a Dalbergia cochinchinensis 57.33 ± 4.52 c 17.04 ± 1.23 c Sindora cochinchinensis 97.33 ± 1.76 ab 28.98 ± 0.69 b Xylia xylocarpa 99.56 ± 0.44 a 17.54 ± 0.96 c

Analysis of variance Source d.f. F Sig. F Sig. Main-plot analysis Block 2 3.577 0.129 0.943 0.462 Treatment 2 8.022 0.040 1.574 0.313 Error 4 Sub-plot analysis Tree species 3 80.634 0.000 93.571 0.000 Treatment × tree species 6 2.535 0.059 1.970 0.124 Error 18

Values are replicated block means (n = 3) ± S.E. The significant ANOVA probabilities are marked in bold. In a column, means followed by the same letters are not significantly different at p ≤ 0.05.

Murdoch University 2011 132

Chapter 7: Improving reforestation success of high-value tree species in Cambodia

a b

c d

Figure 7.7 Direct-seeded seedlings in the field trial in Kbal Chhay Watershed Protected Area (Preah Sihanouk province, Cambodia). Six-month old seedlings of (a), Afzelia xylocarpa; (b), Dalbergia cochinchinensis; (c), Sindora cochinchinensis and (d), Xylia xylocarpa. Plants in (a), (b) and (c) are from the “weeding every three months” plots, and the photos were taken immediately after the second weeding. (d), A seedling of X. xylocarpa in the “weeding once” plot; note the competition from weeds and annual plants. The background of (c) shows a “weeding once” plot. Pole was painted with 20 cm bands. Photos were taken at the end of November 2010.

Murdoch University 2011 133

Chapter 7: Improving reforestation success of high-value tree species in Cambodia

7.4 Discussion

7.4.1 Seedling survival and growth in Experiment 1: effects of N2-fixing bacteria

The hypothesis for this study, that survival and growth of direct seeding of D. cochinchinensis is enhanced by inoculation with effective N2-fixing bacteria, was not established. After six months, there was no benefit of inoculation on plant growth. There are a number of possible explanations for this. Firstly, the inoculum may not have survived under the storage conditions in Cambodia. It took about one month before the inoculants were used in the field. All materials, coated seeds and peat-based inoculum, were stored at 4ºC until transportation to Phnom Penh. There, they were kept in a refrigerator until transportation to the field. Over all, the inocula were exposed to air temperature for about seven days before being used. It is possible that the bacteria, especially those coated on seeds, might not have survived or their number might have been reduced during transportation as when they were exposed to ambient temperature (e.g., Chao and Allexandre, 1984) and dryness (Russell, 1973). The number of bacteria on the coated seeds, 8 000 cells seed-1 or 24 000 cells seeding spot-1, was in the range used by previous researchers (e.g., Thrall et al., 2005). However, this number was determined one day after seed coating. The actual number of live bacteria immediately before seeding might have been significantly lower than this. The clay-based inoculum was prepared one day before usage and bacteria experienced dryness as the clay balls were dried at room temperature overnight to facilitate handling. While it is generally known that termite mound soils have a high clay content (Jouquet et al., 2004), the physical and chemical property of the clay-based inoculant was not determined.

Secondly, there may have been effective N2-fixing bacteria in the field soils. The B. elkanii used in the trial treatments originated from a young plantation of D. cochinchinensis, located about 3 km from the trial site, where inoculation has never been practiced. Dalbergia cochinchinensis is not in its natural range in this area and it was first introduced to the Kbal Chhay Watershed Protected area in 2004 (Chapter 4). A recent survey of tree species in Kbal Chhay by a botanist (McDonald, 2003) did not confirm the presence of D. cochinchinensis in the area. Therefore, it is likely that B. elkanii is associated with several hosts indigenous to the Kbal Chhay area. On checking for root nodules from samples of plants in Block 1, two seeding spots plot-1 (data not presented), it became clear that nodules were present in both inoculated and

Murdoch University 2011 134

Chapter 7: Improving reforestation success of high-value tree species in Cambodia uninoculated plots. Root nodules of D. cochinchinensis were also discovered in the uninoculated Experiment 2 (Figure 7.8). The occurrence of nodules in the uninoculated plants suggests that there was sufficient indigenous bacterial population in the soil, and these bacteria may be as effective as those of the inoculant organisms (Date, 1991). Therefore, it is worth investigating the bacteria that nodulated the control plots of Experiment 1 and seedlings in Experiment 2. Molecular analysis (Chapter 6) of root nodules collected from these sources would confirm if the inoculated B. elkanii are indigenous to the trial site. If the bacteria are not B. elkanii, indicating that D. cochinchinensis forms symbiotic association with other bacteria, then their effectiveness is worth investigating considering the non significant difference between the inoculated and uninoculated plots. In a field study on the influence of the population size of indigenous rhizobia on performance of introduced bacteria on annual and perennial legumes in Hawaii, Thies et al. (1991) reported that the effectiveness of the inoculated bacteria was inversely related to the number of native bacteria. If there were some effective strains, only 50 cells g-1 of soil can eliminate the effectiveness of any inoculated bacteria. In a review, Brockwell and Bottomley (1995) conclude that in a site with few or no naturally-occurring rhizobia, introduction of new strains is normally effective, and inoculation is useless where there are large rhizobial populations. It was not possible for this study to determine the number of bacteria in the soil before establishment of the trial, and this is recognised as a weakness of this study. It was not able to be investigated due to the limited access to the site and distance from the base laboratory in Australia. The need for inoculation of N2-fixing bacteria is further discussed in the general discussion (Chapter 8).

A third possibility for the ineffectiveness of the inoculated bacteria could be the soil acidity, pH 4.7 (Table 7.1). It is common for most legume plants to have reduced nodulation efficiency in acidic soils with pH < 5 (Bordeleau and Prévost, 1994; Zahran, 1999; Price, 2006). Soil acidity is usually associated with the presence of adverse concentrations of elements, such as Al and Mn toxicity, coupled with nutrient deficiencies, especially P and Mo (Bordeleau and Prévost, 1994). Previous studies showed that soil acidity limits bacterial survival and persistence in soils and decreases nodulation (Graham et al., 1982; Ibekwe et al., 1997). In a review, Bordeleau and Prévost (1994) stated that nodule number decreases with a decrease in soil Ca coupled with an increase in Al. Griffiths and McCormick (1984) studied the effects of soil

Murdoch University 2011 135

Chapter 7: Improving reforestation success of high-value tree species in Cambodia acidity on the capacity of Frankia to nodulate Alnus glutinosa (L.) Gaertn., grown in mine soils that had been limed to various pH values. They concluded that soil pH was an important factor affecting nodulation of A. glutinosa. The highest levels of nodulation were observed in the soils with pH 5.5 to 7.2, and nodulation was decreased in soils with pH < 5.5. Different rhizobia express their symbiotic efficiency differently in acidic sites. In a comparison of 12 strains of Bradyrhizobium for their effectiveness with groundnut (Arachis hypogaea L.) in acidic soils with pH ranging from 5.0 to 6.5, van Rossum et al. (1994) found that some strains were ineffective while others performed well. Liming is generally recommended to raise the pH of acid soils (Bordeleau and Prévost, 1994; Zahran, 1999; Price, 2006) and can increase plant productivity (e.g., Buerkert et al., 1990). In a study on effect of liming on nodulation and yield of common bean across four field sites with soil pH ranging between 4.6 and -1 5.0, Buerkert et al. (1990) reported that liming, 2 000 kg of Ca(OH)2 ha , increased soil pH by 0.4 to 1.3 units, yield by 313% and nodule dry weight by 110%. In the future, the ability of Bradyrhizobium elkanii isolates to survive and nodulate under acid soil conditions should be evaluated prior to their use in acid soils.

a b c

Figure 7. 8 Examples of root nodules from direct-seeded seedlings of Dalbergia cochinchinensis in the field trial in Preah Sihanouk province, Cambodia. Root nodules from a bacterial inoculated (a) and uninoculated (b) treatment of Experiment 1 (the arrow points to one of the nodules). (c), Root nodules from the uninoculated Experiment 2.

The fourth possibility is that the site was not constrained by N for early tree growth and thus the trees were not dependent on nodulation for growth and survival. The lack of a fertilizer response supports this view (see below).

Murdoch University 2011 136

Chapter 7: Improving reforestation success of high-value tree species in Cambodia

7.4.2 Seedling survival and growth in Experiments 1 and 2: effect of fertilizer and polymer

In Experiment 1, plants did not respond to N fertilizer application. There were no signs of nitrogen deficiency in the leaves of young seedlings from the “without N fertilizer” treatment (Figure 7.9) suggesting that the level of soil nitrogen was sufficient for D. cochinchinensis at age they were assessed. Discussion on plant response to fertilizer application is presented below for Experiment 2. In general, the survival rate of D. cochinchinensis was moderate, namely 60%. Although this rate is comparable with previous research results in tropical regions using different methods of seeding (e.g., Doust et al., 2008; Sovu et al., 2010), it was much lower than that obtained in the trial in Sakeaw (Chapter 5). This may have been a consequence of reduced germination as germination capacity was only 50% (test 2, Appendix VII).

The first hypothesis for Experiment 2, that growth performance is different for each high-value species, was established as results showed that there were significant differences in growth and survival rate among tree species tested. Except for D. cochinchinensis, survival rates of the other three species were high and were comparable to those achieved the trial in Thailand (Chapter 5). The survival rate of D. cochinchinensis in this experiment was also low, less than 60%. This could be explained in relation to seed quality as many seeds were broken, due to improper seed extraction techniques: the source of Dalbergia seed used in Experiment 2 was different from that used in Experiment 1 (Table 7.2). However, a seed germination test was not undertaken before using the seed. Although an attempt was made to select seeds that were sound in appearance, it was not possible to detect broken seeds until they were soaked and swollen in water during the pretreatment.

The second hypothesis for Experiment 2, that growth and survival of high-value indigenous species from direct seeding are enhanced by application of inorganic fertilizer with or without a water retention polymer, was established for the survival as the combination of fertilizer with polymer resulted in the highest survival rate, but there

Murdoch University 2011 137

Chapter 7: Improving reforestation success of high-value tree species in Cambodia

a e

b f

c g

d h

Figure 7.9 Leaves of Dalbergia cochinchinensis from Experiment 1, field trial in Preah Sihanouk province, in which N2-fixing bacteria and N fertilizer were used as treatments. (a) - (d), Leaves from plots treated with N fertilizer and (e) - (h), those from plots without N fertilizer treatment. (a) and

(e), seed coating; (b) and (f), peat based N2-fixing bacteria and (c) and (g), clay-based N2-fixing bacteria. (d) and (h), no bacteria. Bars represent 1 cm.

Murdoch University 2011 138

Chapter 7: Improving reforestation success of high-value tree species in Cambodia was no effect of treatment on growth. The significant response of survival to this treatment may be linked to the polymer as fertilizer alone was not effective. It was not expected to see an effect of polymer during the rainy season. In a pot study under glasshouse conditions to investigate the effect of polymer on soil water holding capacity with a five-day irrigation interval, Sivapalan (2001) reported that a significant amount of water was lost from the control pots compared to those in the polymer treated soils within 35 days after planting of soybean due to evaporation from the soil. However, this trend was reversed after 40 days when the plants started using excess water retained by the polymer. The amount of water retained by a sandy soil was significantly increased by 95% with addition of 0.07% polymer to the potting mix (w:w). It can be concluded that the polymer has a function to regulate water availability in the soil by absorbing water from the surrounding environment, including the top soil, and releasing it when there is an excess of water or plants require it. When the polymer was applied at the bottom of the seeding hole, it absorbs water from the soil surface and consequently reduces evaporation through limiting the movement of water from the sub-soil to the soil surface (Ouchi et al., 1990). In 2010, the Kbal Chhay area had an irregular pattern of rainfall with minimum of rain occurring in the first month of trial establishment (June) and the middle of the wet season (September) (Figure 7.2). This irregular pattern of rain may not have been a problem for the polymer-treated plants. Orzeszyna et al. (2006) and the manufacturer of geosynthetics (www.polyfabrics.com.au) patented a geocomposite consisting of a geotextile and superabsorbent synthetic polymers (SAP) enclosed inside. In a pot experiment, they applied geocomposite at the bottom of the pots, about 30 cm below grass seeds (name was not revealed); after six months, without precise quantitative analysis, they reported that the polymer promoted extensive root development of the grass compared to the control. In the Kbal Chhay trial, it is assumed that plants treated with polymer may have had more intensive root development resulting in better survival rates, and consequently leading to improved above ground growth at a later stage (Chapter 5). However, excavation of seedlings for studying root development was not undertaken, therefore, it is not sure if this assumption is valid.

Fertilizer application was not effective in promoting height growth although the amount of fertilizer applied in this study was significantly higher than those used in previous trials (Chapters 3 and 5) (though different rates of NPK were applied). One possibility is that fertilizer was applied too early, at the same time as direct seeding, when seedlings

Murdoch University 2011 139

Chapter 7: Improving reforestation success of high-value tree species in Cambodia were not ready to benefit from it. This was undertaken because frequent visits to the site were not possible due to the distance between the trial site and the based university. As discussed in Chapters 3 and 5, the newly emerged seedlings, especially those that are large seeded, like A. xylocarpa, S. cochinchinensis and X. xylocarpa, may rely heavily on nutrient reserves in their cotyledons in their early growth and might take about two months before starting to explore nutrient resources from the soil (e.g.,Tiné et al., 2000). During this period, some of the fertilizer could have been lost during heavy rain.

Soil acidity at the trial site (Section 7.4.1) could also have a role in the lack of response to fertilizer application. Soil acidity restricts the accessibility of seedlings to macro and micro nutrients, especially P, Ca, and Mo (Hazelton and Murphy, 2007). Acid soils can strongly adsorb much of the phosphate that is applied in NPK fertilizers. One of the consequence in highly acidic soil is the occurrence of Al toxicity which is often the most limiting factor for plant growth, especially where soil pH < 5 (Price, 2006). It is unclear how much soil chemistry may have been playing a role in restoration success at the site. However, as mentioned earlier, soil acidity did not block nodule development and survival at six months was good. In the future, soil physical and chemical properties should be determined before trial establishment in conjunction with bioassays for effective N2-fixing bacteria.

7.5 Concluding remarks

Inoculation of effective N2-fixing bacteria did not promote growth or survival of direct seeding of a high-value timber species in the field trial in Cambodia. It is likely that competition of indigenous bacteria plus soil fertility played a significant role in producing this outcome. Therefore, in future, reforestation sites should be surveyed to determine which sites should be targeted for inoculation (Chapter 8). Direct seeding of high-value timber species is possible in a former degraded forest where weed control was undertaken in the first six months. Application of combined fertilizer and polymer promoted survival rate of direct seeding of the high-value timber species. The effect of polymer is further discussed in Chapter 8.

Murdoch University 2011 140

Chapter 8: General discussion

CHAPTER 8

GENERAL DISCUSSION

A community forest in Kampong Speu province (central part of Cambodia), showing the need for forest restoration.

Murdoch University 2011 141

Chapter 8: General discussion

8.1 Major findings

This study has shown, based on survival and early growth, that direct seeding of four high-value timber species (Afzelia xylocarpa, Dalbergia cochinchinensis, Sindora cochinchinensis and Xylia xylocarpa) is suitable for reforestation programmes in Southeast Asia. As seeds of these high-value timber species are orthodox (seeds with hard coats suitable for long-time storage), the finding can also be applied to other species bearing orthodox seeds, such as Albizia lebbeck (L.) Benth., Dalbergia oliveri, Peltophorum dasyrachis (Miq.) Kurz and Cassia siamea L. The successful use of these species in direct seeding requires knowledge of seed pretreatment. Many legumes in the orthodox group undergo physical dormancy as they have hard and impermeable seed- coats which prevent imbibition (Schmidt, 2000). In order to enhance germination, seed dormancy should be broken by pretreatment before direct seeding. Different species need different methods of pretreatment (Gunn, 2001), and appropriate seed pretreatment ensures rapid and uniform germination (Jinks and Jones, 1996), a characteristic desirable for reforestation by direct seeding. The Australian Tree Seed Centre published a guide for seed pretreatment, mainly for species indigenous to Australia, however, some species indigenous to or used in continental Southeast Asia, e.g. Albizia lebbeck, Casuarina equisetifolia L., Leucaena leucocephala, Pterocarpus indicus, P. macrocarpus and Toona sureni (Blume) Merr., can be found in the list (Gunn, 2001). In addition, general rules on breaking seed dormancy can be found in Willan (1985) and Schmidt (2000). Without pretreatment, some seeds are able to stay in the top soil for up to four years without germination (Teketay and Granström, 1997). This is an advantage in natural regeneration as if all of one seedling cohort were to be lost, some seeds remain for recruitment in the future. Seed pretreatment is thus a prerequisite condition in any direct seeding programme using orthodox seeds.

This thesis has shown that sole inoculation of N2-fixing bacteria to legume trees produced mixed results. Under glasshouse conditions where fresh cultures of Bradyrhizobium elkanii were used under optimum soil conditions (suitable soil pH and absence of competing bacteria), a significant improvement in plant growth was achieved when inoculum was applied through seed coating or broth culture. By contrast, inoculation of seeds and seedlings with N2-fixing bacteria was not effective in one field experiment in a mediterranean-type ecosystem (Chapter 3) and two field experiments in

Murdoch University 2011 142

Chapter 8: General discussion

tropical regions (Chapters 5 and 7). In this study, a simple method in delivering bacteria to seeds/seedlings using crushed root nodules was initially tested (Chapters 3 and 5). This simple method was replaced by a more controlled approach in which bacteria were isolated and screened (Chapters 6 and 7) in conjunction with two options of delivering of bacteria to seeds: seed coating and bacterial inoculum. Competition with indigenous bacteria in the reforestation sites and soil acidity are factors that may have affected the outcomes of the field trials.

The obvious conclusion from these studies is that inoculation is not always needed. There is often a great diversity of natural rhizobia in the soils in terms of nodulation,

N2-fixation and tolerance to toxicity (Freire and de Sá, 2006). Inoculation of rhizobia is considered to be unnecessary when the indigenous population densities are high (Gentili and Jumpponen, 2006). In degraded sites, effective strains of N2-fixing bacteria may be absent or insufficient for the introduced host plants (OTA, 1983), and under these conditions inoculation of N2-fixing bacteria is beneficial. Catroux et al. (2001) suggest -1 introducing N2-fixing bacteria when the indigenous population is less than 100 cells g soil. A successful rhizobial inoculation was reported by Thrall et al. (2005) in reforestation by direct seeding using Acacia species in sites where the number of indigenous rhizobia was less than 65 cells g-1 soil. As discussed in Chapter 7, if a reforestation site contains some effective strains, only 50 cells g-1 of soil can eliminate the effectiveness of the inoculated bacteria (Thies et al., 1991). However, we can not rule out absolutely that either the site in Cambodia (Chapter 7) was not limited by N or that the indigenous nodulating bacteria were effective. The former requires assessment of the N cycle and more detailed study of plant N fertilizer responses in the presence of all other essential nutrients (e.g., Abdelgadir et al. 2010). The later could be explored in the future by assessment of N isotope ratios in the trees (e.g., Peoples et al., 1989;

Hardarson and Danso, 1993) as well as studies on the N2-fixing capacity of the indigenous rhizobia. There are many studies that have looked at the dependency of legumes on N fixation in tropical regions (e.g., Pule-Meulenberg and Dakota, 2009; de Freitas et al., 2010) reported from a study on assessment of symbiotic dependency of legumes in Botswana that 11 out of 18 of the studied legumes derived ca. ≥ 50% of their N from symbiotic fixation.

Murdoch University 2011 143

Chapter 8: General discussion

In developed countries, such as Australia, farmers are often recommended to inoculate all grain or pasture legume seeds at planting because the cost is not so high, ca. US$ 2 ha-1 (Herridge et al., 2002). Herridge (2002) estimated that the cost for N fertilizer application for legume crops in Vietnam accounts for US$ 25 to 30 million annually. If application of fertilizer N was replaced by inoculation with N2-fixing bacteria, the cost would only be US$ 1 million. In a review, Vessey (2004) suggests that, in agriculture, even where yield responses are not evident, inoculation with N2-fixing bacteria can still be beneficial as it increased N levels in the grain, which can fetch higher price in some markets, and increased N content in the plant which is important when legumes are used as animal fodder. In reforestation, there are plenty of poorer sites than those used in this study, such as the sandy site with little vegetation cover shown in Figure 8.1a. In this type of soil, a planted seedling might not grow well without an input of N. This type of site is one of the potential sites where inoculation with N2-fixing bacteria is likely to be important (Section 8.3.2). In Australia, the programme of legume inoculation resulting in a benefit:cost ratio of 17:1 or US$ 50 million annually with 1.5 million ha of farm land requiring inoculation annually (Herridge et al., 2002). In reforestation, where smaller areas are reforested annually, the inoculation cost might be significantly higher than those in agriculture, but the benefit might be great considering the single requirement of inoculation at the time of planting. However, the cost for inoculation in reforestation might be comparable to or even lower than legume field crops, if inoculation is undertaken in the nursery. Maybe for a legume tree, inoculation with rhizobia is like an insurance policy. Vessey (2004) suggests that inoculation of legume plants with N2-fixing bacteria is most important for long-lived plants as they produce large amounts of biomass over time, and consequently require large inputs of N.

In summary, although this study failed to show satisfactory results from the inoculation with N2-fixing bacteria in the field trials, it does not necessarily means that inoculation is unnecessary or unproductive; rather it is a warning that careful consideration is needed before deciding on inoculation. The need for inoculation is site specific; it is based on the history of land use of the site and local environmental conditions which include competition from indigenous bacteria, soil fertility and acidity. Therefore, there should be studies to determine the number of indigenous bacteria in the potential reforestation sites before deciding on an inoculation programme. Methods for determining the most- probable number of bacteria in soils are described in Somasegaran and Hoben (1994).

Murdoch University 2011 144

Chapter 8: General discussion

Sometimes, amelioration of soil conditions, such as increasing soil moisture or pH to the optimum level, is necessary to ensure the effectiveness of bacterial inoculation. Otherwise, effective bacteria for a particular site condition have to be discovered (Section 8.3.2).

8.2 Issues arising from the research

8.2.1 Transfer of knowledge and technology to Cambodia

As indicated in Chapter 1, previous studies have claimed that direct seeding is a low cost method for reforestation (Engel and Parrotta, 2001; Cole et al., 2011). While direct seeding has been employed in Cambodia (CTSP, 2005), delivery of soil organisms to seeds or seedlings during direct seeding or tree planting and application of a water retention polymer are new to the country. If some of the techniques, such as inoculation with N2-fixing bacteria and mycorrhizal fungi, are to be adopted in Cambodia, this will incur additional costs for the reforestation programme. The need for inoculation with soil microorganisms is not solely dependent on the site conditions (Section 8.1), but also the capacity to apply the technology, which include human resources and facilities and the adoption of the technology by stakeholders/users.

The Forestry Administration’s Forest Research Institute should be the right place for capital investment on N2-fixing bacteria and mycorrhizal fungi considering its impacts on the end-users, such as private tree planters, communities and nursery managers across the country. An ex-ante impact analysis by undertaking a rigorous cost-benefit projection for the new technology is the first step requirement (Bantilan and Johansen, 1995). Improved biomass gain from inoculation is the main rationale for the cost of inoculation of seeds and seedlings with N2-fixing bacteria and mycorrhizal fungi (Kuek, 1994). Based on the literature, evidence of growth improvement and economic productivity of plantations from adopting inoculation technology in plantation forestry (Chapter 2) can be used as a tool for convincing the policy makers and end users. Other rationales are the potentials of microorganism for sustainable plant production and soil conservation (Gianinazzi and Vosátka, 2004). However, it may be difficult to convince the end-users to adopt the technology based solely on the literature. Therefore, in the long run, field demonstration plots should be established under various ecological

Murdoch University 2011 145

Chapter 8: General discussion

conditions as they will generate accurate, site specific data on growth improvement and economic benefit from inoculation. These demonstration plots should run in parallel with continuous applied research on screening for effective bacteria and / or fungi for tree species used in reforestation programmes in Southeast Asia (Section 8.3.3).

8.2.2 Seed procurement and supply

The high establishment rates achievable by direct seeding of the high-value timber species in this study encourage the use of these species in reforestation, which not only contribute to economic development but also conservation of their genetic resources (Chapter 4). However, reforestation by direct seeding needs a large amount of seed (Weinland, 1998; Schmidt, 2008). This need presents a difficulty for promotion of direct seeding of some species, especially threatened species like D. cochinchinensis, as most mature trees have been removed illegally from their natural habitats (Chapter 4). There are a number of identified seed sources in natural forests across Cambodia where a total number of 20 priority species have been marked for seed collection (FLD et al., 2006). However, access to these sources is limited due to the lack of a seed distribution system (Chapter 4). In addition, a few hectares of seed production areas (SPA) of seven species were established recently in the Kbal Chhay Watershed Protected Area (FLD et al., 2006), but these sources will not be able to produce seeds in the near future. The majority of Dalbergia seeds (60 to 70%) used in the field trial in Cambodia (Chapter 7) were broken due to a poor seed extraction technique used by villagers, which resulted in low establishment rates. This highlights an issue in seed handling, including collection, processing and storage. It also reveals that poor quality seeds are being used in Cambodia (FA and CTSP, 2003a). In addition, it was not possible to find additional seeds for refilling the un-germinated holes. As discussed in Chapter 4, the problems in seed supply and distribution have to be improved in order that reforestation by direct seeding can become widely applicable.

Although this study showed that direct seeding of D. cochinchinensis is feasible, it may not be a good idea now to recommend it as an option in reforestation programmes considering the limited seed sources and high price of seed for this species in Cambodia. It is possible in the future that the current SPA and plantations will be able to produce seed. However, in the current situation, alternative production of planting

Murdoch University 2011 146

Chapter 8: General discussion

material for D. cochinchinensis, such as from cuttings should be considered. Planting material of another high-value timber species, Dalbergia melanoxylon, is mainly produced from cuttings as propagation by seeds is difficult due to poor seed viability (Amri et al., 2010). It is worth investing in propagation via cuttings for D. cochinchinensis as this will also open an opportunity to begin a tree improvement programme as the techniques enable selection of superior provenances (e.g., Amri et al., 2009) and clones for further propagation or for establishment of clonal seed orchards (Chapter 4).

8.2.3 Coping with weed problems in reforestation

One of the challenges for any reforestation programme is to cope with the risk caused by weeds/grasses, particularly Imperata grass (Chapters 4, 5 and 7). About 180 000 ha of the land surface area of Cambodia is covered with Imperata grass (Garrity et al., 1996). The presence of weeds and grasses on a reforestation site can cause two serious problems: fire in the dry season and competition for nutrients with and suppression of the newly planted or direct seeded seedlings.

Fire, which is sometimes started by local people, significantly reduces survival of planted seedlings (Otsamo et al., 1997; Hooper et al., 2002). Protection from forest fire can consume a significant part of a reforestation project budget (Chapter 2). In Cambodia, about 20 to 30% of the annual reforestation budget is allocated to fire prevention and suppression (Pers. comm., Uon Sam Ol, 2010). When fast-growing tree species are planted, fire prevention should continue for three years or so, until canopy closure. However, fire prevention can take much longer when slow-growing indigenous species are planted in open areas (Chapter 4). Common strategies in fire prevention are to disc-plough between tree rows at the onset of the dry season and to construct fire lines in and surrounding reforestation site. Sometime a lookout is also needed. Otsamo et al. (1995) reported, from a reforestation trial in an Imperata grassland in South Kalimantan, that disc-ploughing the whole area plus application of inorganic fertilizer resulted in improved growth of three fast-growing tree species, Acacia mangium, Gmelina arborea and Paraserianthes falcataria (L.) I. Nielsen.

Previous studies has shown that slow-growing indigenous species in Southeast Asia,

Murdoch University 2011 147

Chapter 8: General discussion

including Dalbergia latifolia, Cassia fistula L., Delonix regia (Hook.) Raf., Intsia bijuga (Colebr.) Kuntze, Lagerstroemia sp., Pinus merkusii, Pterocarpus indicus and P. macrocarpus, are not suitable for planting in open area of Imperata grassland (Otsamo et al., 1997). In contrast, fast-growing species, such as A. mangium, A. auriculiformis, A. leptocarpa, G. arborea, P. falcataria and Cassia siamea, have shown promise in reclaiming Imperata grassland (Otsamo et al., 1995). Management of research plots of indigenous species established on Imperata grassland in Kbal Chhay, Cambodia (Chapter 4) proved difficult, though it was possible, especially in coping with fire. The common practice in planting slow-growing indigenous species on a grassland is to interplant the species under the canopies of the fast-growing species (e.g., McNamara et al., 2006). An integrated strategy is needed in dealing with weed problems in reforestation.

8.2.4 Inoculation with mycorrhizal fungi

As discussed in Chapter 2, many studies have shown that inoculation with a relatively small number of effective mycorrhizal fungi improved growth of seedlings in nurseries or in field trials under particular conditions, for example in soils with limited phosphorus supply. In Chapter 3, none of the fungal treatments produced beneficial effects. In Chapter 5, mycorrhizal fungi improved establishment of D. cochinchinensis by ca. 20% and diameter growth by ca. 40%, but not the other two species, A. mangium and X. xylocarpa. Mycorrhizal fungi are ubiquitous in the topsoil of many ecosystems (e.g., Olsson et al., 1999; Singh et al., 2003; Repáč, 2007) though their number and diversity in different ecosystem may vary greatly depending on the level of soil disturbance. In addition, more than 80% of plant families form symbiotic associations with AM fungi (Gianinazzi-Pearson, 1996), and these fungi, in contrast to many ECM fungi, generally have limited host specificity (Helgason et al., 2002), but exceptions occur (Chapter 2). Therefore, careful considerations should be taken for the need for mycorrhizal inoculation and the chance of success (Schwartz et al., 2006). Inoculation with mycorrhizal fungi is necessary where indigenous population densities are extremely low or absent such as in degraded ecosystems where top soils have been removed through mining operations or are severely eroded (Jasper et al., 1987; Rao and Tak, 2002). Checking for AM fungal communities in the intended reforestation sites is necessary before initiating inoculation programmes as production of these fungi is

Murdoch University 2011 148

Chapter 8: General discussion

difficult and costly (Gentili and Jumpponen, 2006). Assessment of their population through bioassay experiments (Chapter 2) will indicate whether a reforestation site has sufficient inoculum of compatible fungal species (ECM and AM fungi) (Brundrett et al., 1996). Then, a decision can be made on the need to introduce effective mycorrhizal fungi to the reforestation site (Brundrett et al., 1996; Chen et al., 2007). Where inoculation is needed, local strains should be used (Schwartz et al., 2006) as they have adapted to the local biotic and abiotic conditions. This will minimize the spreading of harmful organisms that may accidentally contaminate the introduced inoculum

(Schwartz et al., 2006). In addition, transferring of mycorrhizal fungi between countries could lead to the spread of undesirable invasive fungi (Schwartz et al., 2006; Pringle et al., 2009; Vellinga et al., 2009). In a review, Litchman (2010) concludes that invasions by fungi and other microbes may significantly alter the structure of indigenous communities and affect ecosystem function. For example, an invasive mycorrhizal fungus in North America, Amanita phalloides (Vaill. ex Fr.) Link, is replacing native fungi in several ecosystems (Wolfe et al., 2010). This is of particular concern for biodiversity conservation areas. Currently, commercial inocula are available for mycorrhizal fungi; however, care should be taken when using the commercial products as there is no universal strain(s) that are compatible with large numbers of hosts and in a wide range of climatic conditions.

8.2.5 Enrichment planting/seeding of degraded forests

Although direct seeding can be employed as a means to established commercial tree plantations (Chapters 3, 5 and 7), its use may be most suitable where tree rows are not a requirement or where multi-species planting is desirable such as in watershed, protected and biodiversity conservation areas or community forests. In these situations, pre- germinated seeds can be buried in forest gaps where natural density is low (Figure 8.1a,b). In Cambodia, the aim of community forest management is to improve livelihoods of participating communities (Sunderlin, 2006) through access to a variety of forest products. However, the majority of community forests are severely degraded (Sunderlin, 2006). These forests are in need of restoration to increase productivity and improve livelihoods of local communities. Some community forests are too degraded to be able to recover by natural regeneration by seeds as there are limited or no mature trees that can produce seeds. Sometimes, degraded forests contain very few tree species,

Murdoch University 2011 149

Chapter 8: General discussion

especially dry dipterocarp forests (Figure 8.1a) dominated by D. tuberculatus, D. obtusifolius or Shorea obtusa. Local communities can use direct seeding of multiple tree species for enrichment of their community forests for achieving multiple forest products as the technique is relatively simple. It is expected that with some technical advice from the local Forestry Administration on seed pretreatment and seeding methods, local communities can restore their forests with low cost.

a b

Figure 8.1 Community forests in the central part of Cambodia, Kampong Chhnang province. (a), A severely degraded forest on a nutrient poor sandy soil in Kampong Chnang province, Cambodia. Note there are very few tree species and few mature trees left for natural regeneration. (b), Another community forest, with a signboard in the foreground, with lack of mature trees of commercially important species. Photos were taken in December 2009.

Sometimes, the condition of degraded forests is exacerbated by poor soil fertility and low soil moisture (Figure 8.1a), therefore, an integrated strategy needs to be developed for restoration or reforestation of these forests. Firstly, legume trees with multipurpose characteristic, such as X. xylocarpa, A. lebbeck, P. macrocarpus and C. siamea, should be used for reforestation as legumes are more tolerant to nutrient poor soil (Chapter 3) and they can improve soil fertility (Schroth et al., 2001). Seeds of these tree species are available in large quantities from natural forests of Pursat, Siem Reap, Preah Vihear and Rotanak Kiri provinces (FA and CTSP, 2003b). Secondly, soil moisture and nutrient availability may be promoted through application of water retention polymers in combination with organic or inorganic fertilizer (Section 8.3.1) during direct seeding or tree planting to ensure high survival and early growth.

8.3 Further research

Murdoch University 2011 150

Chapter 8: General discussion

This thesis has mostly focused on promoting access by young seedlings to soil nutrients and water in order that they can overcome the harsh conditions of reforestation sites in the first growing season. Opportunities for further research have been discussed in each of the experimental Chapters (Chapters 3, 5 - 7). However, it is worth reemphasizing the need for further research on three aspects of reforestation, promotion of soil moisture through application of water retention polymers, the use of N2-fixing bacteria for high-value trees and understanding the mycorrhizal dependency of reforestation species.

8.3.1 Utilization of water retention polymers for improving reforestation success

Application of a water retention polymer in combination with inorganic fertilizer promoted survival (Chapter 7) and height growth of all species tested (Chapter 5). The central and southern parts of Cambodia, such as Takeo, Kampong Speu and Kampong Chhnang provinces, could be suitable sites for testing polymers as the areas not only experience a long dry season and low annual rainfall, but also the topsoils have a high percentage of sand, generally ca. 90% (Hin et al., 2010) (Figure 8.1a). Water retention polymers do not contain essential plant nutrients, they are soil conditioners. Therefore, combining polymers with organic materials, such as compost or farm-yard manure and inorganic fertilizer, could provide an ideal substrate for optimizing early seedling growth. In previous studies, polymers were generally used in conjunction with fertilizer applications (Duangpatra and Attanandana, 1992; Al-Humaid and Moftah, 2007). Polymer materials are readily biodegradable in soil by ionic, fungal and microbial activities which convert them to non-harmful residues: water, carbon dioxide and organic matter (Barvenik, 1994; Stahl et al., 2000). More attention should be paid to research especially the following three aspects: 1. Testing application rates of polymer to determine an optimum amount that enhances growth of indigenous species across a range of sites with a long dry season and sandy soil texture. In addition, knowledge on the duration of the polymer effect in particular site conditions is also important as it would suggest the right timing of application. As proposed in Chapter 5, polymers can be used for extending the growing season, as plants can be planted/seeded late in the dry season, e.g. late April in Cambodia, where weed competition is limited. Thus, optimal rates of polymers that ensure plant growth before the arrival of the wet season have to be explored.

Murdoch University 2011 151

Chapter 8: General discussion

2. Determining the most suitable method of polymer application, such as whether to mix polymers with soil, or place them at the bottom of planting holes. A previous study showed that polymers can promote extensive root development (Orzeszyna et al., 2006), therefore, the depth of polymer placement in a planting hole could have significant effects on root development. Placing a polymer deep in the planting hole, e.g. 50 cm, may promote root development in the sub-soil layer and thus reduce root desiccation during the first-year’s dry period. It may not be appropriate to apply polymers to the soil surface as continuous or excessive application may form a membrane-like layer which limits water infiltration and increases surface runoff (Wu et al., 2010). 3. Testing of the effects of different polymers. There are many types of polymers (Gross and Kalra, 2002) and each of these polymers differ in their capacity to absorb and hold water and to respond to variation in soil temperature (Andry et al., 2009).

8.3.2 Screening effective N2-fixing bacteria

When there is a need for inoculation with effective bacteria in any reforestation site (Section 8.1), the next step is to screen for effective bacteria (Chapter 2) for the species to be planted. Difference species of Rhizobium and Bradyrhizobium vary in their tolerance to major environmental factors; therefore, screening for effective strains is needed for different sites (Keyser et al., 1993). Screening for effective bacteria should include strains from a wide range of ecological conditions as some tree species can be compatible with a range of bacteria. For example, Rasolomampianina et al. (2005) showed that the genus Dalbergia forms root nodules with a diversity of N2-fixing bacteria, mostly in Bradyrhizobium. The benefits of inoculation with effective N2-fixing bacteria were discussed in Chapter 2. In addition, biological nitrogen fixation is not only useful for the host plant, but also for the inter-planted or subsequent rotation plants, and is environmentally safe (Crews and Peoples, 2004). Inoculated bacteria can persist in the field for many years, and thus eliminate the need for reinoculation (Sanginga et al., 1994).

Selection for effective N2-fixing bacteria is needed for legume trees that are used in reforestation programmes, but priority should be given to species with high economic and conservation values, such as D. cochinchinensis and D. oliveri in Southeast Asia.

Murdoch University 2011 152

Chapter 8: General discussion

Sometimes, effective strains for a tree species have been discovered in other countries, and it is worth including these strains in screening experiments under local environmental conditions. The easy way of delivering bacteria to seedlings is by inoculation of seedlings in the nursery where the bacteria and the hosts can form symbiotic associations before outplanting to the field. When the bacteria are to be delivered to seeds/seedlings in the field, a range of carriers and techniques have to be investigated based on convenience in handling and effectiveness in promoting survival of bacteria. In Cambodia, various sites where inoculation with effective N2-fixing bacteria could be effective include mined sites, degraded lands where top soils has been removed by erosion, and nutrient-poor sandy soils of denuded land (Figure 8.1a). Sometimes, an integrated strategy may be needed to ensure the success of bacterial inoculation. This includes liming the topsoil to raise the pH to a level suitable for bacterial survival and expansion and improving soil moisture availability by application of water retention polymers.

8.3.3 Understanding the mycorrhizal dependency of reforestation species

While this study did not reveal any strong growth effect from mycorrhizal inoculation, sites such as Figure 8.1a were not used in any of the trials in Southeast Asia. There are no data on the extend of bare land in Cambodia, but sites with similar conditions as that of Figure 8.1a, which the Forestry Administration (FA, 2008b) classified as “other forests”, cover a total area of more than 1 million ha. In addition, as indicated in Section 8.2.3, about 180 000 ha of the land surface area of Cambodia is covered with Imperata grass (Garrity et al., 1996). These areas are likely to have low soil biodiversity and are likely to benefit from inoculation with mycorrhizal fungi if suitable inocula can be developed. As with N2-fixing bacteria, when inoculation with mycorrhizal fungi is considered important for a site, the next step is to select effective isolates for tree species to be planted. Local strains should be given priority as they have adapted to local environmental conditions of the planting site (Section 8.2.4). On the other hand, there are many gaps in knowledge as to the mycorrhizal status of many high-value timber species (particularly rosewoods), except the dipterocarps (e.g., Pampolina et al., 1994; See and Alexander, 1994; Ådjers et al., 1998; Yuwa-Amornpitak et al., 2006), and their dependency on mycorrhizal fungi. It is not clear whether succession of fungi takes place as these trees age. This basic knowledge must underpin future reforestation

Murdoch University 2011 153

Chapter 8: General discussion

efforts to use mycorrhizal fungi. Previous studies that surveyed mycorrhizal fungi in natural forests soils in Thailand revealed that most plants develop mycorrhizal associations (Yuwa-Amornpitak et al., 2006; Nandakwang et al., 2008), suggesting that forest soils are sources of fungal inoculum. Given the shortage of funds and researchers in Cambodia, simple techniques in mycorrhizal inoculation should be promoted. These include sowing seeds under the canopy of mother trees to allow seedlings to develop mycorrhizal association before collecting wildings for potting (Ådjers et al., 1998) and using soil inoculum collected from under the mother trees or forest soils as a component in the potting mix (Svasti, 2000). Sometimes, nursery managers retain trees in nursery beds for a source of mycorrhizal fungi and also for shade. However, it is often unknown how much mycorrhizal symbiosis develops in the roots of seedlings inoculated in these simple ways. As there is always a risk of the spread of pathogens with these methods, longer-term research should be undertaken with the aim of delivering cost-effective technologies for future reforestation efforts in Cambodia. Ideally, this can be facilitated through greater research collaboration within the region.

Murdoch University 2011 154

Appendix I

APPENDIX I

RELEVANT PUBLICATIONS

Refereed papers So, T., Thailade, I., Dell, B., 2010. Conservation and utilization of threatened hardwood species in reforestation―an example of Afzelia xylocarpa (Kruz.) Craib and Dalbergia cochinchinensis Pierre in Cambodia. Pacific Conservation Biology 16:101-116.

So, T., Ruthrof, K. X., Dell, B., 2011. Seed and seedling responses to inoculation with mycorrhizal fungi and root nodule bacteria: implications for restoration of degraded Mediterranean-type Tuart woodlands. Ecological Management and Restoration 12: 157-160.

Strange, N., Theilade, I., So, T., Sloth, A., Helles, F. 2007. Integration of species persistence, cost and conflicts: An evaluation of tree conservation strategies in Cambodia. Biological Conservation 137: 223- 236.

Conference presentation Thea, S., Ruthrof, X. K., Dell, B., 2009. Reforestation in Cambodia: the use of seeds. Abstract presented at SRI 2009 World Conference on Ecological Restoration. Making change in a changing world. 19th Conference of the Society for Ecological Restoration International (SRI). Perth, Western Australia, Australia. 23-27 August 2009.

Murdoch University 2011 155

Appendix II

APPENDIX II

SOIL CHEMICAL AND PHYSICAL PROPERTIES OF THE TRIAL SITE IN WESTERN AUSTRALIA

Table II.1 Soil chemical and physical properties of the trial site in Western Australia (depth 0-10 cm)

Soil property Experiment 1 Experiment 2 (Eucalyptus gomphocephala) (Acacia saligna) Block 1 Block 2 Block Clay (%) 3.92 5.05 33.91 2.96 Silt (%) 3.97 1.02 1.98 2 Sand (%) 92.11 93.93 94.11 .95.03 1 pH CaCl2 6.1 6.1 6.6 06.8 2 pH H2O 6.8 6.9 7.1 17.3 Organic Carbon (%)3 1.5 1.58 1.98 1.28 Nitrate N (mg/kg)4 6 5 6 7 Ammonium N (mg/kg)4 5 5 4 4 Total N (%)5 0.16 0.13 0.15 0.11 Extractable K (mg/kg)6 37 47 34 25 Extractable P (mg/kg)6 14 16 13 9

Available P (mg/kg)7 7.4 6.7 4.7 5.3

Total P (mg/kg)8 342 273 308 281 Extractable B (mg/kg)9 0.5 0.5 0.6 0.5 Extractable S (mg/kg)10 5.4 7.5 4.4 5 Extractable Fe (mg/kg)11 236.25 290.85 271.9 233.1 DTPA Cu (mg/kg)11 0.21 0.26 50.27 0.19 DTPA Zn (mg/kg)11 0.34 0.29 0.27 0.23 DTPA Mn (mg/kg)11 3.21 2.1 1.56 1.1 DTPA Fe (mg/kg)11 14.31 16.08 16.63 10.82 Exchangeable Ca (meq/100g)12 6.06 5.74 8.16 6.7 Exchangeable Mg (meq/100g)12 0.51 0.54 0.57 0.33 Exchangeable Na (meq/100g)12 0.08 0.09 0.11 0.07 Exchangeable K (meq/100g)12 0.12 0.13 0.12 0.1 Exchangeable Al (meq/100g)13 0 0 0 0 14 Soluble Cl2 (mg/kg) 19 18 25 16 Electrical Conductivity (dS/m)15 0.036 0.049 0.057 0.042

Values are single measurement. Unless otherwise stated, soil chemical properties were determined using methods described in Rayment and Higginson (1992) and their codes are indicated in the parentheses. 1 (4B2). 2 (4A1). 3 Walkley-Black (6A1). 4 2 M KCl (7C2). 5 Kjeldahl steam distillation (7A1). 6 Colwell 7 8 9 10 (9B1). Olsen (9C2). Pooled/inhouse method. Hot 0.01 M CaCl2 extraction (12C1). 0.25 M KCl 40 11 12 13 14 15 (Blair et al., 1991). (12A1). 0.1 M NH4Cl at pH = 7 (15E1). 1 M KCl (15G1). (5A1). 1:5 soil:water extraction (3A1).

Murdoch University 2011 156

Appendix III

APPENDIX III

EFFECTS OF TREATMENT AND PLANT MATERIAL ON SURVIVAL AND GROWTH OF EUCALYPTUS GOMPHOCEPHALA AND ACACIA SALIGNA AT 40 MONTHS

Table III.1 The effects of treatment and plant material on survival and growth of Eucalyptus gomphocephala established by direct seeding or using nursery-raised seedlings at 40 months

Treatment Survival (%) Height growth (m) Treatment ECM fungi 40.00 ± 15.73 1.54 ± 0.41 ECM fungi + fertilizer 39.33 ± 16.79 1.56 ± 0.37 Control 33.33 ± 13.25 1.49 ± 0.25

Plant material Seed 4.44 ± 1.94 b 1.03 ± 0.12 b Seedling 70.67 ± 4.27 a 1.81 ± 0.25 a

Analysis of variance

Source d.f. F Sig. F Sig. Block 2 0.399 0.681 8.000 0.020 Treatment 2 0.533 0.603 0.255 0.783 Plant material 1 107.246 0.000 8.060 0.030 Treatment × plant material 2 0.370 0.700 2.033 0.212 Error 10

Values are replicated block means (n = 3) ± S.E. The significant ANOVA probabilities are marked in bold. In a column, means followed by the same letters are not significantly different at p ≤ 0.05.

Murdoch University 2011 157

Appendix III

Table III.2 The effects of treatment and plant material on establishment and growth of Acacia saligna established by direct seeding or using nursery-raised seedlings at 40 months

Survival (%) Height growth (m)

Treatment T1: root nodule 60.00 ± 12.11 abc 2.53 ± 0.28 T2: mycorrhizas 50.67 ± 11.93 c 2.14 ± 0.37 T3: fertilizer 78.00 ± 6.77 a 3.01 ± 0.29 T4: T1 + T2 67.33 ± 8.31 abc 2.82 ± 0.37 T5: T1 + T3 76.00 ± 7.73 ab 2.89 ± 0.14 T6: T1 + T2 + T3 60.00 ± 7.28 bc 2.87 ± 0.22 T7: autoclaved T2 56.00 ± 15.28 c 2.55 ± 0.38 T8: control 54.00 ± 11.44 c 2.43 ± 0.19

Plant material Seed 42.50 ±4.08 b 2.21 ± 0.14 b Seedling 83.00 ±1.83 a 3.10 ± 0.09 a

Analysis of variance

Source d.f. F Sig. F Sig. Block 2 0.285 0.754 0.863 0.432 Treatment 7 3.443 0.008 1.972 0.093 Plant material 1 110.300 0.000 36.163 0.000 Treatment × plant material 7 1.738 0.138 1.624 0.167 Error 30

Values are replicated block means (n = 3) ± S.E. The significant ANOVA probabilities are marked in bold. In a column, means followed by the same letters are not significantly difference at p ≤ 0.05.

Murdoch University 2011 158

Appendix IV

APPENDIX IV

SPECIES INFORMATION OF SINDORA COCHINCHINENSIS

AND XYLIA XYLOCARPA

IV. 1 Sindora cochinchinensis

a b

c

Figure IV.1 (a), A Sindora cochinchinensis tree with fruits (photo courtesy of Cambodia Tree Seed Project. (b), Emergence of seedlings of S. cochinchinensis one week after direct seeding in Sakeaw province, Thailand. (c), Seeds of S. cochinchinensis at a street stand in Pursat province, Cambodia. These seeds have been roasted for using as tea or making desserts when the seed coat has been removed.

Sindora cochinchinensis is native to Cambodia, Laos, Vietnam, Thailand and the Malay peninsula (Dy Phon, 2000). It is generally found in dry dipterocarp forests and secondary forests (Vũ, 1996). Sindora cochinchinensis is a large evergreen tree, reaching a height up to 25 to 30 m and a diameter of 0.8 to 1 m when mature (Vũ, 1996). It is a light demanding species occurring in lowland and foothill forests. It grows best on deep sandy and loamy soils, but it can be found on poor and rocky soils. The trees flower in January to March and fruiting is evident in April to August (Vũ, 1996). The seeds are edible and use for making desserts (Figure IV.1c). They are sometimes roasted to make a tea-like drink (Dy Phon, 2000). Although the species is a member of the legume family, it apparently does not form symbiotic association

Murdoch University 2011 159

Appendix IV

with N2-fixing bacteria (Faria et al., 1987). The wood is resistant to the attack by termites, and is used for construction, furniture and fine art articles (Vũ, 1996). Sindora cochinchinensis is one of the priority species for conservation in Cambodia (FA and CTSP, 2003b), Thailand and Vietnam (Luoma-aho et al., 2004). However, the species has not been widely used in reforestations in these countries. The IUCN Red List classifies the species as Lower Risk Least Concern (IUCN, 2008).

IV.2 Xylia xylocarpa

a b

Figure IV. 2 (a), A mother tree of Xylia xylocarpa in an identified seed source in a mixed-deciduous forest in Siem Reap province, Cambodia (photo courtesy of Cambodia Tree Seed Project, 2004). The tree is being marked for seed collection. (b), Twenty-month old seedlings of X. xylocarpa (foreground) established by direct seeding in Sakeaw province, Thailand. Note a big variation in growth between seedlings. The seedling on the right reached a height of 2.4 m whereas that on the left achieved 1 m height (the colour bands on the pole represent 20 cm). The difference indicates a potential opportunity for growth improvement of the species. Acacia mangium plots are present in the background.

Xylia xylocarpa is distributed in Indochina countries (Dy Phon, 2000). It is a large deciduous tree that can reach a height of 20 to 25 m (Dy Phon, 2000) and a diameter of 40 to 60 cm (Vũ, 1996) (Figure IV.2a). It occurs in dense dipterocarp forests (Dy Phon, 2000) or semi-deciduous forests on deep sandy soils (Vũ, 1996). It is a lowland species, growing at an altitude of up to 850 m with mean annual rainfall of 1 200 to 1 700 mm (Phongoudome, no date). The tree is a light-demanding species and sheds its leaves early in the

Murdoch University 2011 160

Appendix IV

dry season. In mixed deciduous forests, it is found growing with other species such as Pterocarpus macrocarpus, Dipterocarpus alatus, Dipterocarpus obtusifolius and Lagerstroemia calyculata Kurz. The trees flower in March to June, and fruiting occurs in December (Vũ, 1996), and a kilogram of seed contains about 3 700 to 6 200 seeds. The germination rate varies from 40 to 77% (Phongoudome, no date).

The wood of X. xylocarpa is very hard, durable and resistant to termites and insects, and is used in pillars, construction, boat-building, railway sleepers and furniture (Vũ, 1996). The bark and fruit are used in traditional medicine against haemoptysies (Dy Phon, 2000). Xylia xylocarpa is one of the priority species of Indochina countries, and it has been conserved in situ or ex situ in all countries of the region (FA and CTSP, 2003b; Nghia, 2004; Phongoudome and Mounlamai, 2004; Sumantakul, 2004; FLD et al., 2006). Two in situ conservation stands of 121 ha were established in Cambodia in the early 2000s (FLD et al., 2006).

The species is suitable for rehabilitation of degraded lands left over from logged-over forest or shifting cultivation (So, 2000), and it is one of the priority species for plantation in Thailand (Pedersen, 2000). It has a strong natural regeneration ability, especially in open areas. A trial in Southern Cameroon showed that the species tolerates highly acidic and Al toxic soils (Kanmegne et al., 2000). Results from a field trial in Sabah, Malaysia, showed that the species can reach a mean total height of 23.2 m and mean DBH of 26.1 cm after nine years (Josue, 2004). Although it is a climax species, it has a presence in the early stage of forest succession (Kaewkrom et al., 2005; Koonkhunthod et al., 2007). This characteristic indicates the suitability of the species in planting, however, in Cambodia, the species has not yet been promoted as a plantation species.

Murdoch University 2011 161

Appendix V

APPENDIX V

PROTOCOLS FOR IDENTIFICATION OF ROOT NODULE BACTERIA OF DALBERGIA COCHINCHINENSIS

V.1 Checking for the persistence of the inoculated bacteria in the glasshouse experiment

V.1.1 Cell preparation

1. Cells from fresh plate culture (7 to 10 days) were transferred (3 to 4 loops) into 1.5 mL eppendorf tubes filled with 1 mL of 0.86 % saline (NaCl). 2. Eppendorf tubes were centrifuged at 10 000 rpm for 10 min. 3. The cells were washed by pouring the supernatant out of the eppendorf and then the eppendorf was refilled with 200 µl saline. Cells were mixed with the saline by pipetting up and down for a few times. 4. Fifty µL of cell solution was diluted into 950 µL saline (using new 1.5 mL eppendorf tubes). The tubes were shaken gently to ensure the solution was well mixed. 5. Optical density of the cells was measured and the data were recorded. 6. Cell solutions were adjusted to optical density to 6 (O.D. = 6) with saline solution. 7. One µL of cell solution was used for running PCR (Section V.1.2), and the remaining cells were stored under temperature of 4 oC (up to two weeks).

V.1.2 PCR Amplification

1. Small eppendorf tubes, 0.2 mL, were marked with an identification number (on the lid) corresponding to the isolate name. 2. Master mix (MM) was made up based on RPO1 primer reaction as shown in the table below:

No Materials 1 reaction (µL) 141 reactions (µL) 1 Buffer (5 ×) 4 56

2 MgCl2 2.4 33.6 3 Primer (RPO1) 1 14 4 Taq DNA 0.5 7

5 H2O (PCR) 11.1 155.4 6 Cells 1 - Total 20 266

1 There were 12 reactions to work with. Note: Materials no. 1- 4 were stored at -20ºC, and during the preparation of master mix they are transported in an ice box.

3. Each 0.2 mL-eppendorf tube was filled with 19 µL of master mix and 1 µL of cell solution (O.D. = 6), and the content was ready for running a PCR amplification. 4. The PCR run with programmes as follows:

Murdoch University 2011 162

Appendix V

Step 1 95ºC for 5 min Step 2 94ºC for 30 sec 50ºC for 20 sec 5 cycles 72ºC for 90 sec Step 3 94ºC for 30 sec 50ºC for 20 sec 35 cycles 72ºC for 90 sec Step 4 72ºC for 5 min 5. After PCR amplification, the PCR products were ready for running electrophoresis (next step). PCR products were stored under temperature of 4oC until they were detected through electrophoresis.

V.1.3 Electrophoresis

1. Solution of agarose gel, 1.5%, was prepared by adding 1.5 g of agarose powder (Bio-Rad Laboratories, Hercules, CA, USA) into 100 mL of 1 × TAE buffer in a 200 mL clean shott bottle. The bottle, with loosely covered lid, was microwaved for 1 min and the bottle was gently shaken to ensure the agarose was well mixed with the buffer. The bottle was microwaved for another min. 2. 8.5 µL of SYBR-Safe DNA gel stain (Invitrogen Molecular Probes, Eugene, OR, USA) were loaded into the buffer solution while it was still hot. The buffer solution was shaken gently and left for 5 min to cool down. 3. The agarose solution was poured into a gel chamber, with the comb in place, and the gel was left for 15 min to become solidified, and the comb was carefully removed. 4. The gel, together with the tray, was placed onto the electrophoresis apparatus with the wells facing the negative electrode. The gel was flood with buffer (1 × TAE buffer) to 2-3 mm over the gel. 5. Each PCR product eppendorf tube was loaded with 6 µL of loading dye (Blue/orange loading dye). 6. Each gel well was loaded with 25 µL of the PCR product. Wells at both sides of the PCR product were filled with 10 µL of ladder solution (Promega Corporation, Madison, WI, USA) (which had been mixed with 100 µL of loading dye). 7. The apparatus was run, after replacing the cover, at 75 V for 90 min (or until the yellow line was about 1 cm from the edge of the gel). 8. The gel was taken out of the bath and placed (without tray) on the floor of the UV light box, connected to a computer, for capturing the picture using Quantity One programme.

V.2 DNA amplification, extraction and sequencing

V.2.1 DNA amplification

1. Fresh cells were prepared (within seven days, all steps in V. 1.1 were repeated, if necessary).

Murdoch University 2011 163

Appendix V

2. The master mix was prepared as follows:

No Materials 1 reaction (µL) 121 reactions (µL) 1 Buffer (5×) 4 48

2 MgCl2 1.2 14.4 3 50 µM Primer (20F, Forwards) 1 12 4 50 µM Primer (1540R, Reverse) 1 12 5 Taq DNA 0.5 6

6 H2O (PCR) 11.3 135.6 7 Cells 1 -

1The volume of master mix was made up for 12 reactions.

3. A 0.2-mL-volume eppendorf tube, marked with a strain number, was filled with 19 µL of master mix and 1 µL of cell solution (O.D. = 6). 4. The content was amplified by running PCR with the following programmes: Step 1 95ºC for 5 min Step 2 95ºC for 30 sec 52ºC for 30 sec 30 cycles 72ºC for 60 sec Step 3 72ºC for 5 min 5. The PCR products were subjected to electrophoresis (Section V.1.3) or stored under temperature of 4oC until they were detected through electrophoresis. 6. A picture of the DNA fragment was taken (step 8, Section V.1.3) (Figure V.1), and the DNA product was excised from the agarose gel with a clean, sharp scalpel on a Mini- transilluminator (Bio-Rad Laboratories, Hercules, CA, USA). 7. The gel containing the DNA product was placed in 1.5 mL tubes and stored under temperature of 20oC below zero or directly subjected to DNA extraction (Section V.2.2).

L 1 2 3 4 5 6 7 8 9 L 10

1 500

Figure V. 1 Electrophoretic patterns of PCR products amplified with forward primer 20F and reverse primer 1540R. Lane L, 1kb DNA ladder weight marker from 250 to 10 000 bp; lanes 1 - 3 bacteria from reference isolate DR2; lanes 4 - 6, bacteria from the isolate T3.2 (seeds coated with DR2); lane 7 to 9, bacteria from the isolate T6.5.1 (inoculation with broth culture of DR2); and lanes 10, water.

V.2.2 DNA extraction (Clean up PCR product: after QIAquick PCR purification kit, 2006)

Murdoch University 2011 164

Appendix V

1. The gel containing DNA product was weighted in a 1.5 mL eppendorf tube. Three volumes of Buffer QG were added to 1 volume of gel. 2. The tubes were incubated at 50ºC for 10 min until the gel slice had completely dissolved. 3. The colour of the mixture was checked to make sure that it was yellow. (If the colour is orange or violet, 10 µL of 3 M sodium acetate, pH 5.0, have to be added to each tube). 4. One-gel volume of isopropanol was added to the sample and mixed. 5. A QIAquick spin column was placed in a provided 2 mL collection tube. The samples were transferred into the QIAquick column (to bind DNA), and were centrifuged at 13 000 rpm for 1 min. 6. The flow-through was discarded and the QIAquick column was placed back into the same collection tube (collection tubes are reused). 7. The QIAquick column was centrifuged at 13 000 rpm for 1 min after adding 0.5 mL of Buffer QG. The flow-through was discarded. 8. The QIAquick column was centrifuged at 13 000 rpm for 1 min after adding 0.75 mL of Buffer PE. The flow-through was discarded. 9. The QIAquick column was placed into a clean 1.5 mL eppendorf tube (without lid). 10. The QIAquick column was centrifuged at 13 000 rpm for 1 min to elute the DNA after adding 30 µL of double distilled water. The purified DNA was obtained, and it was transferred to a new tube (with lid) and stored under temperature of 20ºC below zero.

V.2.3 Determine concentration of DNA via gelelectrophoresis

1. Eight micro-liters of purified DNA were loaded into a 0.2 mL eppendorf tube followed by addingf 1.5 µL of loading dye (the rate was 1 volume of dye to 5 volumes of purified DNA). 2. The DNA concentration was determined by running the 1.5% gel (Section V.1.3) and capturing the picture (Figure V.2).

L 1 2 3 4 L

1 500

Figure V. 2 Concentration of DNA. Lane L, 1kb DNA ladder weight marker from 250 to 10 000 bp; lanes 1,2,3 and 4, DNA from reference isolate DR2, T3.2,T6.5.1 and water, respectively.

V.2.4 DNA sequencing reaction

Murdoch University 2011 165

Appendix V

1. The cleaned DNAs were sequenced in two directions, forward (20F primer: 5'-AGT TTG ATC CTG GCT CA-3') and reverse (1450R primer: 5'-AAG GAG GTG ATC CAG CCG CA-3') (GeneWorks, Adelaide, Australia, http://www.geneworks.com.au/default.aspx?p=1). 2. Two master mixes were prepared, one for each primer (Forward and Reverse) for the ¼ sequencing reaction as follows:

No Materials 1 reaction (µL) 51 reactions (µL) 1 Double distilled water 1.8 9 2 Dye terminator mix 2 10 3 5 × Sequencing Buffer 1 5 4 3.2 p moles primer at 50 nM concentration 3.2 16 5 DNA template 2 - Total 10

1 Three reactions were to be sequenced.

3. The sequencing reaction was run with the following programmes: Step 1 Hold on at 96ºC for 2 min before loading the samples to heat cycler Step 2 96ºC for 10 sec 52ºC for 5 sec 25 cycles 60ºC for 4 min Step 3 14ºC for ∞ min (or final stage)

V.2.5 Post sequencing reaction clean up (after QIAquick PCR purification kit, 2006)

1. The sequencing reaction (10 µL) was transferred into a 0.5 mL eppendorf tube and the following added: 1 µL of EDTA (125 mM), 1 µL of sodium acetate pH 5.2 (3 M) and 25 µL of 100% ethanol. 2. The content was mixed by pipetting up and down and left under room temperature for 20 min. 3. The content was centrifuged at 14 000 rpm for 30 min. 4. Liquid in the tube was quickly removed using 200 µL pipette tips. The excess liquid was removed using Kimwipe tissue to dry the tube, but ensuring not to touch the pellet. 5. The pellet was rinsed by adding 125 µL of 70% ethanol and centrifuged (14 000 rpm) for 5 min. 6. The supernatant was removed as much as possible with a pipette tip, and then the tube was dried with a Kimwipe tissue. 7. The pellet was air-dried in the dark for 15 min. 8. Samples were submitted to the Western Australian State Agricultural Biotechnology Centre (SABC) for partial 16S rRNA nucleotide gene sequence analysis using AB Applied Biosystems (HITACHI) 3730xl DNA Analyser.

Murdoch University 2011 166

Appendix VI

APPENDIX VI

PARTIAL 16S rRNA NUCLEOTIDE GENE SEQUENCES

Partial 16S rRNA nucleotide gene sequence of strain ST1 (forward direction of strain DR2)

1 CGAGCGGGCATAGCAATATGTCAGCGGCAGACGGGTGAGTAACGCGTGGGAACGTACCTT 60 61 TTGGTTCGGAACAACTGAGGGAAACTTCAGCTAATACCGGATAAGCCCTTACGGGGAAAG 120 121 ATTTATCGCCGAAAGATCGGCCCGCGTCTGATTAGCTAGTTGGTGAGGTAATGGCTCACC 180 181 AAGGCGACGATCAGTAGCTGGTCTGAGAGGATGATCAGCCACATTGGGACTGAGACACGG 240 241 CCCAAACTCCTACGGGAGGCAGCAGTGGGGAATATTGGACAATGGGCGCAAGCCTGATCC 300 301 AGCCATGCCGCGTGAGTGATGAAGGCCCTAGGGTTGTAAAGCTCTTTTGTGCGGGAAGAT 360 361 AATGACGGTACCGCAAGAATAAGCCCCGGCTAACTTCGTGCCAGCAGCCGCGGTAATACG 420 421 AAGGGGGCTAGCGTTGCTCGGAATCACTGGGCGTAAAGGGTGCGTAGGCGGGTCTTTAAG 480 481 TCAGGGGTGAAATCCTGGAGCTCAACTCC 509

Partial 16S rRNA nucleotide gene sequence of strain ST5 (reverse direction of strain DR2)

1 ACAAGGCCCGGGAACGTATTCACCGTGGCGTGCTGATCCACGATTACTAGCGATTCCAAC 60 61 TTCATGGGCTCGAGTTGCAGAGCCCAATCCGAACTGAGACGGCTTTTTGAGATTTGCGAA 120 121 GGGTCGCCCCTTAGCATCCCATTGTCACCGCCATTGTAGCACGTGTGTAGCCCAGCCCGT 180 181 AAGGGCCATGAGGACTTGACGTCATCCCCACCTTCCTCGCGGCTTATCACCGGCAGTCTC 240 241 CTTAGAGTGCTCAACTAAATGGTAGCAACTAAGGACGGGGGTTGCGCTCGTTGCGGGACT 300 30 TAACCCAACATCTCACGACACGAGCTGACGACAGCCATGCAGCACCTGTCTCCGGTCCAG 360 361 CCGAACTGAAGAACTCCGTCTCTGGAGTCCGCGACCGGGATGTCAAGGGCTGGTAAGGTT 420 421 CTGCGCGTTGCGTCGAATTAAACCACATGCTCCACCGCTTGTGCGGGCCCCCGTCAATTC 480 481 CTTTGAGTTTTAATCTTGCGACCGTACTCCCCAGGCGGAATG 522

Partial 16S rRNA nucleotide gene sequence of strain ST2 (forward direction of strain T3.2)

1 CGAGCGGGCATAGCAATATGTCAGCGGCAGACGGGTGAGTAACGCGTGGGAACGTACCTT 60 61 TTGGTTCGGAACAACTGAGGGAAACTTCAGCTAATACCGGATAAGCCCTTACGGGGAAAG 120 121 ATTTATCGCCGAAAGATCGGCCCGCGTCTGATTAGCTAGTTGGTGAGGTAATGGCTCACC 180 181 AAGGCGACGATCAGTAGCTGGTCTGAGAGGATGATCAGCCACATTGGGACTGAGACACGG 240 241 CCCAAACTCCTACGGGAGGCAGCAGTGGGGAATATTGGACAATGGGCGCAAGCCTGATCC 300 301 AGCCATGCCGCGTGAGTGATGAAGGCCCTAGGGTTGTAAAGCTCTTTTGTGCGGGAAGAT 360 361 AATG 364

Partial 16S rRNA nucleotide gene sequence of strain ST6 (reverse direction of strain T3.2)

1 GGTGTGTACAAGGCCCGGGAACGTATTCACCGTGGCGTGCTGATCCACGATTACTAGCGA 60 61 TTCCAACTTCATGGGCTCGAGTTGCAGAGCCCAATCCGAACTGAGACGGCTTTTTGAGAT 120 121 TTGCGAAGGGTCGCCCCTTAGCATCCCATTGTCACCGCCATTGTAGCACGTGTGTAGCCC 180 181 AGCCCGTAAGGGCCATGAGGACTTGACGTCATCCCCACCTTCCTCGCGGCTTATCACCGG 240 241 CAGTCTCCTTAGAGTGCTCAACTAAATGGTAGCAACTAAGGACGGGGGTTGCGCTCGTTG 300

Murdoch University 2011 167

Appendix VI

301 CGGGACTTAACCCAACATCTCACGACACGAGCTGACGACAGCCATGCAGCACCTGTCTCC 360 361 GGTCCAGCCGAACTGAAGAACTCCGTCTCTGGAGTCCGCGACCGGGATGTCAAGGGCTGG 420 421 TAAGGTTCTGCGCGTTGCGTCGAATTAAACCACATGCTCCACCGCTTGTGCGGGCCCCCG 480 481 TCAATTCCTTTGAGTTTTAATCTTGCGACCGTACTCCCCAGGCGGAATGCTTAAAGCGTT 540 541 AGCTGCGCCACTAGTGAGTAAACCCACTAACGGCTGGCATTCATCGTTTACGGCGTGGAC 600 601 TACCAGGGTATCTAATCCTGTTTGCTCCCCACGCTTTCGTGCCTCAGCGTCAG 653

Partial 16S rRNA nucleotide gene sequence of strain ST3 (forward direction of strain T6.5.1)

1 TGCAGTCGAGCGGGCATAGCAATATGTCAGCGGCAGACGGGTGAGTAACGCGTGGGAACG 60 61 TACCTTTTGGTTCGGAACAACTGAGGGAAACTTCAGCTAATACCGGATAAGCCCTTACGG 120 121 GGAAAGATTTATCGCCGAAAGATCGGCCCGCGTCTGATTAGCTAGTTGGTGAGGTAATGG 180 181 CTCACCAAGGCGACGATCAGTAGCTGGTCTGAGAGGATGATCAGCCACATTGGGACTGAG 240 241 ACACGGCCCAAACTCCTACGGGAGGCAGCAGTGGGGAATATTGGACAATGGGCGCAAGCC 300 301 TGATCCAGCCATGCCGCGTGAGTGATGAAGGCCCTAGGGTTGTAAAGCTCTTTTGTGCGG 360 361 GAAGATAATGACGGTACCGCAAGAATAA 388

Partial 16S rRNA nucleotide gene sequence of strain ST8 (reverse direction of strain T6.5.1)

1 GCTGCCCCCTTTCGGTTAGCGCACCGTCTTCAGGTAAAACCAACTCCCATGGTGTGACGG 60 61 GCGGTGTGTACAAGGCCCGGGAACGTATTCACCGTGGCGTGCTGATCCACGATTACTAGC 120 121 GATTCCAACTTCATGGGCTCGAGTTGCAGAGCCCAATCCGAACTGAGACGGCTTTTTGAG 180 181 ATTTGCGAAGGGTCGCCCCTTAGCATCCCATTGTCACCGCCATTGTAGCACGTGTGTAGC 240 241 CCAGCCCGTAAGGGCCATGAGGACTTGACGTCATCCCCACCTTCCTCGCGGCTTATCACC 300 301 GGCAGTCTCCTTAGAGTGCTCAACTAAATGGTAGCAACTAAGGACGGGGGTTGCGCTCGT 360 361 TGCGGGACTTAACCCAACATCTCACGACACGAGCTGACGACAGCCATGCAGCACCTGTCT 420 421 CCGGTCCAGCCGAACTGAAGAACTCCGTCTCTGGAGTCCGCGACCGGGATGTCAAGGGCT 480 481 GGTAAGGTTCTGCGCGTTGCGTCGAATTAAACCACATGCTCCACCGCTTGTGCGGGCCCC 540 541 CGTCAATTCCTTTGAGTTTTAATCTTGCGACCGTACTCCCCAGGCGGAATGCTTAAAGCG 600 601 TTAGCTGCGCCACTAGTGAGTAAACCCACTAACGGCTGGCATTCATCGTTTACGGCGTGG 660 661 ACTACCAGGGTATCTAATCCTGTTTGCTCCCCACGCTTTCGTGCCTCAGCGTCAGTATCG 720 721 GGCCAGTGAGCCGCCTTCGCCACTGGTGTTCTTGCGAATATCTACGAATTTCACCTCTAC 780 781 ACTCGCAGTTCCACTCACCTCTCCCGAACTC 811

Murdoch University 2011 168

Appendix VII

APPENDIX VII

SEED GERMINATION OF DALBERGIA COCHINCHINENSIS

Two samples of Dalbergia cochinchinensis seed, supplied by the Royal Forest Department, Thailand collected in 2006 and 2007, were available for testing of their germination percentages. Germination tests were undertaken in March 2008 (test 1for the 2006 seedlot) and March 2010 (test 2 for the 2007 seedlot). Seeds of uniform size without physical damage were selected by visual examination. Seeds were pretreated by soaking in tap water overnight and then surface sterilized in bleach (Chapter 6). Twenty five seeds were evenly spread on sterile distilled water agar (0.9%: w:v) plates (90 mm diameter Petri dishes) and allowed to adhere to the agar for approximately 30 min. Plates were sealed with laboratory film (Parafilm®, Pechiney Plastic Packaging, Menasha, WI, USA) and then incubated at 28oC in the dark. Germination was checked after four days, and the last count was undertaken at 10 days (Table VII.1).

Table VII. 1 Germination percentages of two seedlots of Dalbergia cochinchinensis

Plate Germination percentage (%) Germination percentage (%) Test 1: 20 March 2008 Test 2: 13 March 2010

1 68 56 2 72 44 3 64 52 4 80 60 5 64 48 6 68 56 7 76 40 Mean ± S.E. 70.29 ± 2.29 50.86 ± 2.72

Murdoch University 2011 169

Appendix VIII

APPENDIX VIII

SKETCH MAP OF THE FIELD TRIAL IN CAMBODIA

Figure VIII. 1 Sketch map of the field trial in Kbal Chhay, Preah Sihanouk province, Cambodia, showing approximated locations of the charcoal remains which are marked with yellow patches. Details of the trial map is shown in Chapter 7, Section 7.2.3.

Murdoch University 2011 170

References

REFERENCES

Abbott, L. K. and Robson, A. D., 1982. The role of vesicular arbuscular mycorrhizal fungi in agriculture and the selection of fungi for inoculation. Australian Journal of Agricultural Research 33: 389-408.

Abdelgadir, E. M., Fadul, E. M., Fageer, E. A. and Ali, E. A., 2010. Response of wheat to nitrogen fertilizer at reclaimed high terrace salt-affected soils in Sudan. Journal of Agriculture and Social Sciences 6: 43-47.

Ådjers, G., Hadengganan, S., Kuusipalo, J., Otsamo, A. and Vesa, L., 1998. Production of planting stock from wildings of four Shorea species. New Forests 16: 185-197.

Aggangan, N. S., Moon, H. K. and Han, S. H., 2010. Growth response of Acacia mangium Willd. seedlings to arbuscular mycorrhizal fungi and four isolates of the ectomycorrhizal fungus Pisolithus tinctorius (Pers.) Coker and Couch. New Forests 39: 215-230.

Al-Humaid, A. I. and Moftah, A. E., 2007. Effects of hydrophilic polymer on the survival of buttonwood seedlings grown under drought stress. Journal of Plant Nutrition 30: 53-66.

Amri, E., Lyaruu, H. V. M., Nyomora, A. S. and Kanyeka, Z. L., 2009. Evaluation of provenances and rooting media for rooting ability of African Blackwood (Dalbergia melanoxylon Guill. & Perr.) stem cuttings. Research Journal of Agriculture and Biological Sciences 5: 524-532.

Amri, E., Lyaruu, H. V. M., Nyomora, A. S. and Kanyeka, Z. L., 2010. Vegetative propagation of African Blackwood (Dalbergia melanoxylon Guill. & Perr.): effects of age of donor plant, IBA treatment and cutting position on rooting ability of stem cuttings. New Forests 39: 183-194.

Andry, H., Yamamoto, T., Irie, T., Moritani, S., Inoue, M. and Fujiyama, H., 2009. Water retention, hydraulic conductivity of hydrophilic polymers in sandy soil as affected by temperature and water quality. Journal of Hydrology 373: 177-183.

Anon., 1979. Tropical legumes: resources for the future. National Academy of Sciences, Washington D.C.

Appanah, S., 2002. Introduction: restoration of degraded forests as opportunities for development. In Proceedings of the International Conference on Bringing back the forests: Policies and practices for degraded lands and forests, pp. 7-14 (Eds H. C. Sim, S. Appanah and P. B. Durst). Food and Agriculture Organization of the United Nations. Regional Office for Asia and the Pacific, Bangkok.

Appleton, B., Koci, J., French, S., Lestyan, M. and Harris, R., 2003. Mycorrhizal fungal inoculation of established street trees. Journal of Arboriculture 29: 107-110.

Archibald, R., Bowen, B., Hardy, G., McCaw, L. and Close, D., 2006. The role of fire in tuart decline at Yalgorup?Tuart decline research findings. Tuart Bulletin No.6. Perth.

Murdoch University 2011 171

References

Asano, H., 1996. Coated seeds. United States. Patent Number 5525131.

Bâ, A. M., Sanon, K. B. and Duponnois, R., 2002. Influence of ectomycorrhizal inoculation on Afzelia quanzensis Welw. seedlings in a nutrient-deficient soil. Forest Ecology and Management 161: 215-219.

Bainbridge, D. A., 2007. A guide for desert and dryland restoration. Island Press, Washington DC.

Baker, P. J., Bunyavejchewin, S., Oliver, C. D. and Ashton, P. S., 2005. Disturbance history and historical stand dynamics of a seasonal tropical forest in western Thailand Ecological Monographs 75: 317-343.

Bantilan, M. C. S. and Johansen, C., 1995. Research evaluation and impact analysis of biological nitrogen fixation. Plant and Soil 174: 279-286.

Barber, P. and Hardy, G., 2006. Research into the cause and management of tuart decline. Report of phase 1 activity (July 2003 – June 2006) and phase 2 objectives (March 2006 – Feb 2009). Tuart Health Research Group (THRG), Perth.

Bardgett, R. D., 2005. The biology of soil: a community and ecosystem approach. Oxford University Press, Oxford.

Bardin, S. D. and Huang, H. C., 2003. Efficacy of sticker for seed treatment with organic matter or microbial agents for the control of damping off of sugar beat. Plant Pathology Bulletin 12: 19-26.

Bargali, K., 2011. Screening of leguminous plants for VAM association and their role in restoration of degraded lands. Journal of American Science 7: 7-11.

Barvenik, F. W., 1994. Polyacrylamide characteristics related to soil applications. Soil Science 158: 235-243.

Bauhus, J., Khanna, P. K. and Raison, R. J., 1993. The effect of fire on carbon and nitrogen mineralization and nitrification in an Australian forest soil. Australian Journal of Soil Research 31: 621-639.

Benbrahim, K. F. and Ismaili, M., 2002. Interactions in the symbiosis of Acacia saligna with Glomus mosseae and rhizobium in a fumigated and unfumigated soil. Arid Land Research and Management 16: 365-376.

Bhat, N. R., Suleiman, M. K., Al-Menaie, H., Al-Ali, E. H., AL-Mulla, L., Christopher, A., Lekha, V. S., Ali, S. I. and George, P., 2009. Polyacrylamide polymer and salinity effects on water requirement of Conocarpus lancifolius and selected properties of sandy loam soil. European Journal of Scientific Research 25: 549-558.

Bisht, R., Chaturvedi, S., Srivastava, R., Sharma, A. K. and Johri, B. N., 2009. Effect of arbuscular mycorrhizal fungi, Pseudomonas fluorescens and Rhizobium leguminosarum on the growth and nutrient status of Dalbergia sissoo Roxb. Tropical Ecology 50: 231- 242.

Blair, G. J., Chinoim, N., Lefroy, R. D. B., Anderson, G. C. and Crocker, G. J., 1991. A

Murdoch University 2011 172

References soil sulfur test for pastures and crops. Analyst 109: 549-568.

Blakesley, D., Hardwick, K. and Elliott, S., 2002. Research needs for restoring tropical forests in Southeast Asia for wildlife conservation: framework species selection and seed propagation. New Forests 24: 165-174.

BOM, 2009. Bureau of Meteorology (BOM). Austraian Government. http://www.bom. gov.au/climate/averages/. Accessed August 2008.

Bordeleau, L. and Prévost, D., 1994. Nodulation and nitrogen fixation in extreme environments. Plant and Soil 161: 115-125.

Bouamri, R., Dalpé, Y., Serrhini, M. N. and Bennani, A., 2006. Arbuscular mycorrhizal fungi species associated with rhizosphere of Phoenix dactylifera L. in Morocco. African Journal of Biotechnology 5: 510-516.

Brewbaker, J. L., 1990. Nitrogen fixing trees. In Proceedings of the International Conference on fast growing trees and nitrogen fixing trees. Marburg 8 - 12 October 1989, pp. 253-262 (Eds D. Werner and P. Muller). Gustav Fischer Verlag, Stuttgart.

Brischia, R., Piccioni, E. and Standardi, A., 2002. Micropropagation and synthetic seed in M.26 apple rootstock (II): A new protocol for production of encapsulated differentiating propagules. Plant Cell, Tissue and Organ Culture 68: 137-141.

Brockwell, J., 1962. Studies on seed pelleting as an aid to legume seed inoculation. Australian Journal of Agricultural Research 13: 638-649.

Brockwell, J. and Bottomley, P. J., 1995. Recent advances in inoculant technology and prospects for the future. Soil Biology and Biochemistry 27: 683-697.

Brockwell, J., Searle, S. D., Jeavons, A. C. and Waayers, M., 2005. Nitrogen fixation in acacias: an untapped resource for sustainable plantations, farm forestry and land reclamation. Australian Centre for International Agricultural Research. Monograph No. 115.

Brundrett, M., Bougher, N., Dell, B., Grove, T. and Malajczuk, N., 1996. Working with mycorrhizas in forestry and agriculture. Australian Centre for International Agricultural Research Monograph 32, Canberra.

Brundrett, M., Nick Malajczuk, N., Mingqin, G., Daping, X., Snelling, S. and Dell, B., 2005. Nursery inoculation of Eucalyptus seedlings in Western Australia and Southern China using spores and mycelial inoculum of diverse ectomycorrhizal fungi from different climatic regions. Forest Ecology and Management 209: 193-205.

Brundrett, M. C. and Abbott, L. K., 1995. Mycorrhizal fungus propagules in the jarrah forest. II. Spatial variability in inoculum levels. New Phytologist 131: 461-469.

Bryson, C. T., Krutz, L. J., Ervin, G. N., Reddy, K. N. and Byrd Jr., J. D., 2010. Ecotype variability and edaphic characteristics for cogongrass (Imperata cylindrica) populations in Mississippi. Invasive Plant Science and Management 3: 199-207.

Buerkert, A., Cassman, K. G., de la Piedra, R. and Munns, D. N., 1990. Soil acidity and

Murdoch University 2011 173

References liming effects on stand, nodulation, and yield of common bean. Agronomy Journal 82: 749-754.

Bullard, S., Hodges, J. D., Johnson, R. L. and Straka, T. J., 1992. Economics of direct seeding and planting for establishing oak stands on old-field sites in the South. Southern Journal of Applied Forestry 16: 34-40.

Burgess, T. I., Malajczuk, N. and Grove, T. S., 1993. The ability of 16 ectomycorrhizal fungi to increase growth and phosphorus uptake of Eucalyptus globulus Labill. and E. diversicolor F. Muell. Plant and Soil 153: 155-164.

Cai, Y. F., Barber, P., Dell, B., O’Brien, P., Williams, N., Bowen, B. and Hardy, G., 2010. Soil bacterial functional diversity is associated with the decline of Eucalyptus gomphocephala. Forest Ecology and Management 260: 1047-1057.

Calvente, R., Cano, C., Ferrol, N., Azcón-Aguilar, C. and Barea, J. M., 2004. Analysing natural diversity of arbuscular mycorrhizal fungi in olive tree (Olea europaea L.) plantations and assessment of the effectiveness of native fungal isolates as inoculants for commercial cultivars of olive plantlets. Applied Soil Ecology 26: 11-19.

Cao, J. D., Tang, Y. Q., Qin, S. M. and Hou, Y. R., 1994. Dual inoculation of Mimosaceae seedlings with vesicular-arbuscular mycorrhizal fungi and rhizobium. In Proceeding of an International Symposium and Workshop on Mycorrhizas for Plantation Forestry in Asia, Kaiping, Guangdong Province, P.R. China, Vol. 62, pp. 119-121 (Eds M. Brundrett, B. Dell, N. Malajczuk and M. Q. Gong). Australian Centre for International Agricultural Research (ACIAR), Canberra.

Cardinale, M., Lanza, A., Bonnì, M. L., Marsala, S., Puglia, A. M. and Quatrini, P., 2008. Diversity of rhizobia nodulating wild shrubs of Sicily and some neighbouring islands. Archives of Microbiology 190: 461-470.

Cardoso, I. M. and Kuyper, T. W., 2006. Mycorrhizas and tropical soil fertility. Agriculture, Ecosystems and Environment 116: 72-84.

Carmichael, C., 2008. V. I. Exotic hardwoods. Victoria, Vancouver Island, BC, Canada. http://www.saers.com/~craig/wood/index.html. Accessed 15 December 2008.

Carpenter, F. L., Nichols, J. D., Pratta, R. T. and Young, K. C., 2004. Methods of facilitating reforestation of tropical degraded land with the native timber tree, Terminalia amazonia. Forest Ecology and Management 202: 281-291.

Carvalho, J. A., Neto, T. G. S., Veras, C. A. G., Alvarado, E. C., Sandberg, D. V. and Santos, J. C., 2006. Carbon sequestration through charcoal formation in Amazonian forest fires. Forest Ecology and Management 234S: S154.

Castro, J., Zamora, R., Hódar, J. A., Gómez, J. M. and Gómez-Aparicio, L., 2004. Benefits of using shrubs as nurse plants for reforestation in Mediterranean mountains: A 4-year study. Restoration Ecology 12: 352-358.

Catroux, G., Hartmann, A. and Revellin, C., 2001. Trends in rhizobial inoculant production and use. Plant and Soil 230: 21-30.

Murdoch University 2011 174

References

Cavallazzi, J. R. P., Filho, O. K., Stu¨rmer, S. L., Rygiewicz, P. T. and de Mendonça, M. M., 2007. Screening and selecting arbuscular mycorrhizal fungi for inoculating micropropagated apple rootstocks in acid soils. Plant Cell, Tissue and Organ Culture 90: 117-129.

Chalermpong, A. and Bunthaveekun, T., 1982. Surveying in ectomycorrhizae in dry evergreen forest of Thailand. Royal Forest Department, Minitry of Agriculture and Cooperatives, Bangkok.

Chambers, D. P. and Attiwill, P. M., 1994. The ash-bed effect in Eucalyptus regnans forest: Chemical, physical and microbiological changes in soil after heating or partial sterilisation. Australian Journal of Botany 42: 739-749.

Chan, K. Y., Van Zwieten, L., Meszaros, I., Downie, A. and Joseph, S., 2007. Agronomic values of greenwaste biochar as a soil amendment. Australian Journal of Soil Research 45: 629-634.

Chand, S. and Singh, A. K., 2004. Plant regeneration from encapsulated nodal segments of Dalbergia sissoo Roxb., a timber-yielding leguminous tree species. Journal of Plant Physiology 161: 237-243.

Chang, E. H., Chung, R. S. and Tsai, Y. H., 2007. Effect of different application rates of organic fertilizer on soil enzyme activity and microbial population. Soil Science and Plant Nutrition 53: 132-140.

Chao, W. L. and Allexander, M., 1984. Mineral soils as carriers for Rhizobium inoculants. Applied and Environmental Microbiology 47: 94-97.

Chape, S., Blyth, S., Fish, L., Fox, P. and Spalding, M. (Eds), 2003. 2003 United Nations list of protected areas. IUCN and UNEP-WCMC, Cambridge.

Chen, Y. L., Dell, B. and Malajczuk, N., 2006a. Effect of Scleroderma spore density and age on mycorrhiza formation and growth of containerized Eucalyptus globulus and E. urophylla seedlings. New Forests 31: 453-467.

Chen, Y. L., Kang, L. H. and Dell, B., 2006b. Inoculation of Eucalyptus urophylla with spores of Scleroderma in a nursery in south China: Comparison of field soil and potting mix. Forest Ecology and Management 222: 439-449.

Chen, Y. L., Liu, S. and Dell, B., 2007. Mycorrhizal status of Eucalyptus plantations in south China and implications for management. Mycorrhiza 17: 527-535.

Cleary, M., 2005. Managing the forest in colonial Indochina c.1900-1940. Modern Asian Studies 39: 257-283.

Cole, R. J., Holl, K. D., Keene, C. L. and Zahawi, R. A., 2011. Direct seeding of late- successional trees to restore tropical montane forest. Forest Ecology and Management 261: 1590-1597.

Cole, T. G., Yost, R. S., Kablan, R. and Olsen, T., 1996. Growth potential of twelve Acacia species on acid soils in Hawaii. Forest Ecology and Management 80: 175-186.

Murdoch University 2011 175

References

Crews, T. E. and Peoples, M. B., 2004. Legume versus fertilizer sources of nitrogen: ecological tradeoffs and human needs. Agriculture, Ecosystems and Environment 102: 279-297.

Crocker, C. D., 1963. General soil map. Scale 1: 1 000 000. Service Geographique des F. A. R. K., Phnom Penh.

CTSP, 2005. Direct seeding. Cambodia Tree Seed Project (CTSP). http://www. treeseedfa.org/doc/SeedTestingReport/DirectseedingEnglish.pdf. Accessed 15 April 2008.

Dahm, H., 2006. Role of mycorrhiza in forestry. In Handbook of Microbial Biofertilizers, pp. 241-270 (Ed M. K. Rai). Haworth Press, Binghamton, New York.

Daneshgar, P. and Jose, S., 2009. Imperata cylindrica, an alien invasive grass, maintains control over nitrogen availability in an establishing pine forest. Plant and Soil 320: 209- 218.

Dannelly, C. C., 1981. Seed coatings. United States. Patent Number 4249343.

Danso, K. E. and Ford-Lloyd, B. V., 2003. Encapsulation of nodal cuttings and shoot tips for storage and exchange of cassava germplasm. Plant Cell Reports 21: 718-725.

Dart, P., Umali-Garcia, M. and Almendras, A., 1991. Role of symbiotic associations in nutrition of tropical acacias. In Advances in tropical acacia research, pp. 13-19 (Ed J. W. Turnbull). Australian Centre for International Agricultural Research Proceedings No. 35. Canberra.

Date, R. A., 1991. Nodulation success and persistence of recommended inoculum strains for subtropical and tropical forage legumes in Northern Australia. Soil Biology and Biochemistry 23: 533-541.

Daud, N., Taha, R. M. and Hasbullah, N. A., 2008. Artificial seed production from encapsulated micro shoots of Saintpaulia ionantha Wendl. (African Violet). Journal of Applied Sciences 8: 4662-4667.

Davidson, F. and Reuszer, H. W., 1978. Persistence of Rhizobium japonicum on the soybean seed coat under controlled temperature and humidity. Applied and Environmental Microbiology 35: 94-96. de Freitas, A. D. S., de Souza, L. Q., de Sá Barretto Sampaio, E. V., Moura, P. M. and Meneze, R. S. C., 2010. Tree nitrogen fixation in a tropical dry vegetation in Northeast Brazil. In 19th World Congress of Soil Science, Soil Solutions for a Changing World. 1 - 6 August 2010, Brisbane, Australia, pp. 66-69 Published on DVD.

Deaker, R., Roughley, R. J. and Kennedy, I. R., 2004. Legume seed inoculation technology―a review. Soil Biology and Biochemistry 36: 1275-1288.

DeBano, L. F., 1981. Water repellent soils: a state-of-the-art. General Technical Report PSW-46. Pacific Southwest Forest and Range Experiment Station, California.

DeFries, R., Hansen, A., Newton, A. C. and Hansen, M. C., 2005. Increasing isolation

Murdoch University 2011 176

References of protected areas in tropical forests over the past twenty years. Ecological Applications 15: 19-26.

Delang, O. C., 2007. The role of medicinal plants in the provision of health care in Lao PDR. Journal of Medicinal Plants Research 1: 50-59.

Devall, M. S. and Smith, G. C., 2007. Forests in the balance: Linking tradition and technology in landscape mosaics Biological Conservation 137: 487-488

Dhakal, L. P., Jha, P. K. and Kjaer, E. D., 2005a. Mortality in Dalbergia sissoo (Roxb.) following heavy infection by Aristobia horridula (hope) bettles. Will genetic variation in susceptibility play a role in combating declining health? Forest Ecology and Management 218: 270-276.

Dhakal, L. P., Lillesø, J. P. B., Kjær, E. D., Jha, P. K. and Aryal, H. L., 2005b. Seed sources of agroforestry trees in a farmland context - a guide to tree seed source establishment in Nepal. Forest & Landscape Development and Environment Series 1- 2005, Hørsholm.

Dodd, J. C. and Thomson, B. D., 1994. The screening and selection of inoculant arbuscular-mycorrhizal and ectomycorrhizal fungi. Plant and Soil 159: 149-158.

Doronila, A. I. and Fox, J. E. D., 2000. Ecosystem development on a titanium dioxide residue pond after five years in Capel, Western Australia. International Journal of Surface Mining, Reclamation and Environment 14: 137-150.

Doust, S. J., Erskine, P. D. and Lamb, D., 2006. Direct seeding to restore rain forest species: Microsite effects on the early establishment and growth of rain forest tree seedlings on degraded land in the wet tropic in Australia. Forest Ecology and Management 234: 333-343.

Doust , S. J., Erskine, P. D. and Lamb, D., 2008. Restoring rainforest species by direct seeding: Tree seedling establishment and growth performance on degraded land in the wet tropics of Australia. Forest Ecology and Management 256: 1178-1188.

Droppelmann, K. J. and Berliner, P. R., 2000. Biometric relationships and growth of pruned and non-pruned Acacia saligna under runoff irrigation in northern Kenya. Forest Ecology and Management 126: 349-359.

Duangpatra, P. and Attanandana, T., 1992. Effects of high-water absorbing polymer on growth and drought endurance of cashew nut, green mango and para-rubber under tropical field conditions. Kasetsart Journal 26: 95-102.

Duponnois, R. and Bâ, A. M., 1999. Growth stimulation of Acacia mangium Willd by Pisolithus sp. in some Senegalese soils. Forest Ecology and Management 119: 209-215.

Duponnois, R., Founoune, H. and Lesueur, D., 2002. Influence of the controlled dual ectomycorrhizal and rhizobal symbiosis on the growth of Acacia mangium provenances, the indigenous symbiotic microflora and the structure of plant parasitic nematode communities. Geoderma 109: 85-102.

Duponnois, R., Founoune, H., Masse, D. and Pontanier, R., 2005. Inoculation of Acacia

Murdoch University 2011 177

References holoserices with ecthomycorrhizal fungi in a semi-arid site in Senegal: growth respond and influences of the mycorrhizal soil infectivity. Forest Ecology and Management 207: 351-362.

Duponnois, R., Plenchette, C., Prin, Y., Ducousso, M., Kisa, M., Bâ, A. M. and Galiana, A., 2007. Use of mycorrhizal inoculation to improve reafforestation process with Australian Acacia in Sahelian ecozones. Ecological Engineering 29: 105-112.

Dy Phon, P., 2000. Dictionary of plants used in Cambodia. Olympic Printing House, Phnom Penh.

Eastman, J. and Danborg, F. (Eds), 1995. National Workshop on Strengthening Re- afforestation Programmes in Lao PDR. Vientiane, 19-21 June 1995. FAO Regional Project STRAP. Field Document No. 4 GCP/RAS/142/JPN, Rome.

Ehrenberg, C., 1970. Breeding for stem quality. Unasylva 24: 23-31.

El Nasr, H. M. A., Kandil, H. M., El Kerdawy, A., Dawlat, Khamis, H. S. and El-Shaer, H. M., 1996. Value of processed saltbush and acacia shrubs as sheep fodders under the arid conditions of Egypt. Small Ruminant Research 24: 15-20.

Elliott, S., 2000. The Chiang Mai research agenda for the restoration of degraded forest lands for wildlife conservation in Southeast Asia. In Proceedings of the Workshop on Forest Restoration for Wildlife Conservation, January 30th - February 4th 2000, Chiang Mai, Thailand, pp. 383-411 (Eds S. Elliott, J. Kerby, D. Blakesley, K. Hardwick, K. Woods and V. Anusarnsunthorn). International Tropical Timber Organization and The Forest Restoration Research Unit, Chiang Mai.

Engel, V. L. and Parrotta, J. A., 2001. An evaluation of direct seeding for reforestation of degraded lands in central Sao Paulo state, Brazil. Forest Ecology and Management 152: 169-181.

Eriksson, G., Namkoong, G. and Roberds, J. H., 1993. Dynamic gene conservation for uncertain futures. Forest Ecology and Management 62: 15-37.

FA, 2007. Cambodia forestry statistics 2006. Forestry Administration (FA), Phnom Pehn.

FA, 2008a. Activities and results of forestry in Semester 1, 2008. Forest-Wildlife Magazine (in Khmer). Forestry Administration (FA), Phnom Penh.

FA, 2008b. Cambodia forest cover. Forest cover map change 2002-2006. Forestry Administration (FA), Phnom Penh.

FA and CTSP, 2003a. An assessment of tree seed demand, future requirement and constraints. Forestry Administration (FA) and Cambodia Tree Seed Project (CTSP), Phnom Penh.

FA and CTSP, 2003b. Forest gene conservation strategy. Part A: Conservation of forest genetic resources. Forestry Administration (FA) and Cambodia Tree Seed Project (CTSP), Phnom Penh.

Murdoch University 2011 178

References

FA and CTSP, 2005. Ex situ conservation in Kbal Chhay. Seed production areas and species and provenance trials. Forestry Administration (FA) and Cambodia Tree Seed Project (CTSP). Unpublished document.

FA and FLD, 2007. Guideline. Income generation model for rural people through participatory thinning of state owned plantation while improving plantation management and forest governance (joint plantation management). Forestry Administration (FA) and Forest & Landscape Denmark (FLD), Phnom Penh.

FAO, 2005. Helping forests take cover: on forest protection, increasing forest cover and future approaches to reforesting degraded tropical landscapes in Asia and the Pacific. RAP Publication No. 2005/13. Food and Agriculture Organization of the United Nations (FAO), Regional Office for Asia and the Pacific, Bangkok.

FAO, 2007. State of the world's forests 2007. Part I: Progress towards sustainable forest management. Food and Agriculture Organization of the United Nation, Rome.

FAO, 2010a. Global forest resources assessment 2010- key findings. Food and Agriculture Organization (FAO) of the United Nation, Rome.

FAO, 2010b. Global forest resources assessment 2010. Main report. Food and Agriculture Organisation of the United Nations. FAO Forestry Paper 163, Rome.

FAO, FLD and IPGRI, 2001. Forest genetic resources conservation and management. Vol. 2: In managed natural forests and protected areas (in situ). International Plant Genetic Resources Institute, Rome.

FAO, FLD and IPGRI, 2004. Forest genetic resources conservation and management. Vol. 1: Overview, concepts and some systematic approaches. International Plant Genetic Resources Institute, Rome.

Faria, S. M. d., Lima, H. C. d., Franco, A. A., Mucci, E. S. F. and Sprent, J. I., 1987. Nodulation of legume trees from South East Brazil. Plant and Soil 99: 347-356.

FLD, CTSP and FA, 2006. Conservation of valuable and endangered tree species in Cambodia 2001 - 2006―a case study. Forest & Landscape Development and Environment Series 3-2006. Danish Centre for Forest, Landscape and Planning, Aalborg.

Florentine, S. K. and Westbrooke, M. E., 2004. Restoration on abandoned tropical pasturelands—do we know enough? Journal for Nature Conservation 12: 85-94.

Forestier, S., Alvarado, G., Badjel, S. B. and Lesueur, D., 2001. Effect of Rhizobium inoculation methodologies on nodulation and growth of Leucaena leucocephala. World Journal of Microbiology and Biotechnology 17: 359-362.

FORGENMAP, 2002. Consultancy Report 20. Conservation Strategy for Forest Genetic Resources of Thailand. Forest Genetic Resources Conservation and Management Programme (FORGENMAP), the Royal Forest Department, Danced and DFSC.

FORRU, 2006. How to plant a forest: The principles and practice of restoring tropical forests. The Forest Restoration Research Unit (FORRU), Biology Department, Science

Murdoch University 2011 179

References

Faculty, Chiang Mai University, Chiang Mai.

Francke, M. S., Saveng, I., Theilade, I. and Schmidt, L., 2007. Deciduous trees of Prey Long. In A floral and faunal biodiversity assessment of Prey Long. Forest & Landscape Working Papers no. 25-2007, pp. 68-81 (Eds A. Olsson and D. Emmett). Forest & Landscape Denmark.

Franco, C. M. M., Tate, M. E. and Oades, J. M., 1995. Studies on non-wetting sands. I. The role of intrinsic particulate organic matter in the development of water-repellency in non-wetting sands. Australian Journal of Soil Research 33: 253-263.

Freire, J. R. J. and de Sá, E. L. S., 2006. Sustainable agriculture and the rhizobial/legume symbiosis. In Handbook of microbial biofertilizers, pp. 183-202 (Ed M. K. Rai). Haworth Press, Binghamton.

Friday, K. S., Drilling, M. E. and Garrity, D. P., 1999. Imperata grassland rehabilitation using agroforestry and assisted natural regeneration. International Center for Research in Agroforestry, Southeast Asian Regional Research Programme, Bogor.

Frioni, L., Malatés, D., Irigoyen, I. and Dodera, R., 1998. Promiscuity for nodulation and effectivity in N2-fixing legume tree Acacia caven in Uruguay. Applied Soil Ecology 7: 239-244.

Gai, J. P., Feng, G., Christie, P. and Li, X. L., 2006. Screening of arbuscular mycorrhizal fungi for symbiotic efficiency with sweet potato. Journal of Plant Nutrition 29: 1085-1094.

Galiana, A., Gnahoua, G. M., Chaumont, J., Lesueur, D., Prin, Y. and Mallet, B., 1998. Improvement of nitrogen fixation in Acacia mangium through inoculation with rhizobium. Agroforestry Systems 40: 297-307.

Garrity, D. P., Soekardi, M., van Noordwijk, M., de la Cruz, R., Pathak, P. S., Gunasena, H. P. M., van So, N., Huijun, G. and Majid, N. M., 1996. The Imperata grasslands of tropical Asia: area, distribution, and typology. Agroforestry Systems 36: 3- 29.

Gemma, J. N. and Koske, R. E., 1988. Seasonal variation in spore abundance and dormancy of Gigaspora gigantea and in mycorrhizal inoculum potential of dune soil. Mycologia 80: 211-216.

Gentili, F. and Jumpponen, A., 2006. Potential and possible uses of bacterial and fungal biofertilizers. In Handbook of microbial biofertilizers, pp. 1-28 (Ed M. K. Rai). Haworth Press, Binghamton.

Ghosh, S. and Verma, N. K., 2006. Growth and mycorrhizal dependency of Acacia mangium Willd. inoculated with three vesicular arbuscular mycorrhizal fungi in lateritic soil. New Forests 31: 75-81.

Gianinazzi-Pearson, V., 1996. Plant cell responses to arbuscular mycorrhizal fungi: Getting to the roots of the symbiosis. The Plant Cell 8: 1871-1883.

Gianinazzi, S. and Vosátka, M., 2004. Inoculum of arbuscular mycorrhizal fungi for

Murdoch University 2011 180

References production systems: science meets business. Canadian Journal of Botany 82: 1264- 1271.

Gillett, H. and Sinovas, P. (Eds), 2008. Strategies for the sustainable use and management of timber tree species subject to international trade: south east Asia. Dalbergia cochinchinensis. UNEP World Conservation Monitoring Centre, Cambridge.

Glaser, B., Lehmann, J. and Zech, W., 2002. Ameliorating physical and chemical properties of highly weathered soils in the tropics with charcoal―a review. Biology and Fertility of Soils 35: 219-230.

Global Witness, 2004. Laundering of illegal timber undermines forestry reform in Cambodia. http://www.globalwitness.org/media_library_detail.php/335/en/laundering _of_illegal_timber_undermines_forestry_r. Accessed 25 June 2009.

GoC, 2010. National Forest Programme. Final draft. Government of Cambodia (GoC), http://www.twgfe.org/nfp/Living_NFP_doc.php. Accessed 28 January 2010.

Gomez, K. A. and Gomez, A. A., 1984. Statistical procedures for agricultural research. John Wiley & Sons, New York.

Goosem, S. P. and Tucker, N. I. J., 1995. Repairing the rainforest—theory and practice of rain forest re-establishment in north Queensland's wet tropics. Wet Tropics Management Authority, Cairns.

Government of Western Australia, 2003. A Tuart Atlas: Extent, density and condition of tuart woodlands on the Swan Coastal Plain. pp. 38 Prepared by the Department of Conservation and Land Management for the Tuart Response Group.

Graham, P. H., Viteri, S. E., Mackie, F., Vargas, A. T. and Palacios, A., 1982. Variation in acid soil tolerance among strains of Rhizobium phaseoli. Field Crops Research 5: 121-128.

Griffiths, A. P. and McCormick, L. H., 1984. Effects of soil acidity on nodulation of Alnus glutinosa and viability of Frankia. Plant and Soil 79: 429-434.

Gross, R. A. and Kalra, B., 2002. Biodegradable polymers for the environment. Science 297: 803-807.

Gunn, B., 2001. Australian tree seed centre operations manual. CSIRO Forestry and Forest Products. Australian Tree Seed Centre, Canberra.

Gunn, B. V., Aken, K. and Pinyopusarerk, K., 2006. Provenance performance of Chukrasia in a five-year-old field trial in the Northern Territory, Australia. Australian Forestry 69: 122-127.

Günter, S., Gonzalez, P., Álvarez, G., Aguirre, N., Palomeque, X., Haubrich, F. and Weber, M., 2009. Determinants for successful reforestation of abandoned pastures in the Andes: Soil conditions and vegetation cover. Forest Ecology and Management 258: 81-91.

Guo, S., Wu, J., Dang, T., Liu, W., Li, Y., Wei, W. and Syers, J. K., 2010. Impacts of

Murdoch University 2011 181

References fertilizer practices on environmental risk of nitrate in semiarid farmlands in the Loess Plateau of China. Plant and Soil 330: 1-13.

Hardarson, G. and Danso, S. K. A., 1993. Methods for measuring biological nitrogen fixation in grain legumes. Plant and Soil 152: 19-23.

Hardwick, K., Healey, J. R., Elliott, S. and Blakesley, D., 2004. Research needs for restoring seasonal tropical forests in Thailand: accelerated natural regeneration. New Forests 27: 285-302.

Hardwick, K. A., Healey, J. R. and Blakesley, D., 2000. Research needs for the ecology of natural regeneration of seasonally dry tropical forests in Southeast Asia. In Proceedings of the Workshop on Forest Restoration for Wildlife Conservation, 30 January - 4 February 2000, Chiang Mai, Thailand, pp. 165-180 (Eds S. Elliott, J. Kerby, D. Blakesley, K. Hardwick, K. Woods and V. Anusarnsunthorn). International Tropical Timber Organization and The Forest Restoration Research Unit, Chiang Mai.

Harper, R. J. and Gilkes, R. J., 1994. Soil attributes related to water repellency and the utility of soil survey for predicting its occurrence. Australian Journal of Soil Research 32: 1109-1124.

Hayat, R. and Ali, S., 2004. Water absorption by synthetic polymer (Aquasorb) and its effect on soil properties and tomato yield. International Journal of Agriculture and Biology 6: 998-1002.

Hazelton, P. and Murphy, B., 2007. Interpreting soil test results: what do all the numbers mean? CSIRO Publishing, Collingwood.

Helgason, T., Merryweather, J. W., Denison, J., Wilson, P., Young, J. P. W. and Fitter, A. H., 2002. Selectivity and functional diversity in arbuscular mycorrhizas of co- occurring fungi and plants from a temperate deciduous woodland. Journal of Ecology 90: 371-384.

Herridge, D., 2002. Legume N and inoculants: Global and Vietnamese perspectives. In Inoculants and nitrogen fixation of legumes in Vietnam, pp. 7-9 (Ed D. Herridge). Australian Centre for International Agriculture Research No.109e.

Herridge, D., Gemell, G. and Hartley, E., 2002. Legume inoculants and quality control. In Inoculants and nitrogen fixation of legumes in Vietnam, pp. 105-115 (Ed D. Herridge). Australian Centre for International Agriculture Research No.109e.

Hin, S., Bell, R. W., Newsome, D. and Seng, V., 2010. Understanding variability in texture and acidity among sandy soils in Cambodia. In 19th World Congress of Soil Science, Soil Solutions for a Changing World, pp. 50-53 Published on DVD, Brisbane.

Homchan, J., Date, R. A. and Roughley, R. J., 1989. Responses to inoculation with root- nodule bacteria by Leucaena leucocephala in soils of N.E Thailand. Tropical Grasslands 23: 92-97.

Hooper, E., Condit, R. and Legendre, P., 2002. Responses of 20 native tree species to reforestation strategies for abandoned farmland in Panama Ecological Applications 12: 1626-1641.

Murdoch University 2011 182

References

Hopkins, A. J. M., Coker, J., Beeston, G. R., Bowan, P. and Harvey, J. M., 1996. Conservation status of vegetation types throughout Western Australia. National Reserves System Co-operative Program, Project No. N703, Final Report. Department of Conservation and Land Management, Perth.

Howieson, J. G., Ewing, l. M. A. and D'Antuono, M. F., 1988. Selection for acid tolerance in Rhizobium meliloti. Plant and Soil 105: 179-188.

Howieson, J. G., Loi, A. and Carr, S. J., 1995. Biserrula pelecinus L.―a legume pasture species with potential for acid, duplex soils which is nodulated by unique root-nodule bacteria. Australian Journal of Agricultural Research 46: 997-1009.

Howieson, J. G., Malden, J., Yates, R. J. and O’Hara, G. W., 2000. Techniques for the selection and development of elite inoculant strains of Rhizobium leguminosarum in southern Australia. Symbiosis 28: 33-48.

Huke, R. E., 1982. Agroclimatic and dry-season maps of South, Southeast, and East Asia. International Rice Research Institute (IRRI), Los Baños, Laguna.

Humphreys, F. R. and Lambert, M. J., 1965. An examination of a forest site which has exhibited the ash-bed effect. Australian Journal of Soil Research 3: 81-94.

Hüttermann, A., Zommorodi, M. and Reise, K., 1999. Addition of hydrogels to soil for prolonging the survival of Pinus halepensis seedlings subjected to drought. Soil and Tillage Research 50: 295-304.

Hydro-meteorological Department, 2010. Monthy rainfall and temperature of Preah Sihanouk province 1983 - 2009. Provincial Hydro-meteorological Department, Preah Sihanoukville.

Ibekwe, A. M., Angle, J. S., Chaney, R. L. and Van Berkum, P., 1997. Enumeration and nitrogen fixation potential of Rhizobium leguminosarum biovar trifolii grown in soil with varying pH values and heavy metal concentrations. Agriculture, Ecosystems and Environment 61: 103-111.

ICEM, 2003. Cambodia national report on protected areas and development. Review of protected areas and development in the lower Mekong River region. International Centre for Environmental Management (ICEM), Indooroopilly.

IFSR, 2004. Cambodia Independent Forest Sector Review 2004. Main report. Royal Danish Embassy Development Cooperation Section and Forestry Administration. Independent Forest Sector Review (IFSR). http://www.cambodia-forest-sector.net. Accessed 15 June 2007.

Ishii, T. and Kadoya, K., 1994. Effects of charcoal as a soil conditioner on citrus growth and vesicular-arbuscular mycorrhizal development. Journal of the Japanese Society for Horticultural Science 63: 529-535.

IUCN, 1999. Threatened trees. World Conservation 3-4: 12.

IUCN, 2008. 2008 IUCN Red List of Threatened Species. www.iucnredlist.org. Accessed 8 November 2008.

Murdoch University 2011 183

References

IUCN, 2009. IUCN Red list of threatened species. Version 2009.1. http://www.iucnre dlist.org. Accessed 12 September 2009.

Jackson, R. M. and Mason, P. A., 1984. Mycorrhiza. Edward Arnold, London.

Jansa, J., Mozafar, A., Anken, T., Ruh, R., Sanders, I. R. and Frossard, E., 2002. Diversity and structure of AMF communities as affected by tillage in a temperate soil. Mycorrhiza 12: 225-234.

Jasper, D. A., Robson, A. D. and Abbott, L. K., 1987. The effect of surface mining on the infectivity of vesicular-arbuscular mycorrhizal fungi. Australian Journal of Botany 35: 641-652.

Jinks, R. L. and Jones, S. K., 1996. The effect of seed pretreatment and sowing date on the nursery emergence of Sitka spruce (Picea sitchensis [Bong.] Carr.) seedlings. Forestry 69: 335-345.

Johns, G. G. and Greenup, L. R., 1976. Pasture seed theft by ants in northern New South Wales. Australian Journal of Experimental Agriculture and Animal Husbandry 16: 249-256.

Jøker, D., 2000a. Afzelia xylocarpa (Kurz) Craib. Danida Forest Seed Centre. Seed Leaflet No. 6. Danida Forest Seed Centre, Humlebaek.

Jøker, D., 2000b. Dalbergia cochinchinensis Pierre. Danida Forest Seed Centre. Seed Leaflet No. 26. Danida Forest Seed Centre, Humlebaek.

Josue, J., 2004. Some wood properties of Xylia xylocarpa planted in Sabah. Sepilok Bulletin 1: 1-15.

Jouquet, P., Tessier, D. and Lepage, M., 2004. The soil structural stability of termite nests: role of clays in Macrotermes bellicosus (Isoptera, Macrotermitinae) mound soils. European Journal of Soil Biology 40: 23-29.

Kaewkrom, P., Gajaseni, J., Jordan, C. F. and Gajaseni, N., 2005. Floristic regeneration in five types of teak plantations in Thailand. Forest Ecology and Management 210: 351- 361.

Kalinganire, A. and Pinyopusarerk, K., 2000. Chukrasia: Biology, cultivation and utilisation. Australian Centre for International Agricultural Research Technical Reports No. 49, Canberra.

Kamo, K., Vacharangkura, T., Tiyanon, S., Viriyabuncha, C., Nimpila, S. and Doangsrisen, B., 2002. Plant species diversity in tropical planted forests and implication for restoration of forest ecosystems in Sakaerat, northeastern Thailand. Japan Agricultural Research Quarterly 36: 111-118.

Kanmegne, J., Bayomock, L. A., Duguma, B. and Ladipo, D. O., 2000. Screening of 18 agroforestry species for highly acid and aluminum toxic soils of the humid tropics. Agroforestry Systems 49: 31-39.

Keating, W. G. and Bolza, E., 1982. Characteristics, properties and uses of timbers.

Murdoch University 2011 184

References

Volume 1. South-east Asia, Northern Australia and the Pacific. Inkata Press, Melbourne.

Kendrick, B., 2000. The fifth kingdom. Third edition. Focus Publishing, Newburyport.

Keyser, H. H., Somasegaran, P. and Bohlool, B. B., 1993. Rhizobial ecology and technology. In Soil microbial ecology: applications in agricultural and environmental management, pp. 205-226 (Ed F. B. Metting). Marcel Dekker, New York.

Khan, A. G., 2001. Relationships between chromium biomagnification ratio, accumulation factor, and mycorrhizae in plants growing on tannery effluent-polluted soil. Environment International 26: 417-423.

Kijkar, S., 2001. Timber plantation development in Thailand. In Proceedings of the International Conference on Timber Plantation Development. Food and Agriculture Organization of the United Nations, Manila.

Kim, S., Sasaki, N. and Koike, M., 2008. Assessment of non-timber forest products in Phnom Kok community forest, Cambodia. Asia Europe Journal 6: 345-354.

Kitajima, K., 2002. Do shade-tolerant tropical tree seedlings depend longer on seed reserves? Functional growth analysis of three Bignoniaceae species. Functional Ecology 16: 433-444

Kjær, E. D., Dhakal, L. P., Lillesø, J. P. B. and Graudal, L., 2006. Application of low input breeding strategies for tree improvement in Nepal. In Proceedings of the IUFRO Division 2 Joint Conference on Low Input Breeding and Genetic Conservation of Forest Tree Species, 9-13 October 2006, pp. 103-109 (Ed K. I. Fikret). Antalya.

Kjær, E. D., Graudal, L. and Nathan, I., 2001. Ex situ conservation of genetic resources of commercial tropical tree species. In In situ and ex situ conservation of commercial tropical trees pp. 127-146 (Eds B. A. Theilges, S. D. Sastrapradja and A. Rimbawanto). Gadjah Mada University, Yogyakarta.

Kometter, R. F., Martinez, M., Blundell, A. G., Gullison, R. E., Steininger, M. K. and Rice, R. E., 2004. Impacts of unsustainable mahogany logging in Bolivia and Peru. Ecology and Society 9: 12.

Koonkhunthod, N., Sakurai, K. and Tanaka, S., 2007. Composition and diversity of woody regeneration in a 37-year-old teak (Tectona grandis L.) plantation in Northern Thailand. Forest Ecology and Management 247: 246-254.

Koskela, J. and Amaral, W. A. N., 2002. Conservation of tropical forest genetic resources: IPGRI’s efforts and experiences. In Proceedings of the Southeast Asian moving workshop on conservation, management and utilization of forest genetic resources, 25 February-10 March 2001, Thailand, pp. 191-206 (Eds J. Koskela, S. Appanah, A. P. Pedersen and M. D. Markopoulos). FORGENMAP/IPGRI/FORSPA/ DFSC/RFD, Bangkok.

Kramer, P. J., 1969. Plant and soil water relationship: A modern synthesis. McGraw- Hill Book Company, New York.

Murdoch University 2011 185

References

Kubota, A., Hoshiba, K. and Bordon, J., 2008. Effect of fertilizer-N application and seed coating with Rhizobial inoculants on soybean yield in Eastern Paraguay. Revista Brasileira de Ciência do Solo 32: 1627-1633.

Kuek, C., 1994. Issues concerning the production and use of inocula of ectomycorrhizal fungi in increasing the economic productivity of plantations. In Management of mycorrhizas in agriculture, horticultre and forestry, pp. 221-230 (Eds A. D. Robson, L. K. Abbott and N. Malajczuk). Kluwer Academic Publishers, Dordrecht.

Kurth, V. J., MacKenzie, M. D. and DeLuca, T. H., 2006. Estimating charcoal content in forest mineral soils. Geoderma 137: 135-139.

Lakhanpal, T. N., 2000. Ectomycorrhiza―an overview. In Mycorrhizal biology, pp. 101-118 (Eds K. G. Mukerji, B. P. Chamola and J. Singh). Kluwer Academic / Plenum Publishers, New York.

Lakshman, H. C., Rajanna, L., Inchal, R. F., Mulla, F. I. and Srinivasulu, Y., 2001. Survey of VA - mycorrhizae in agroforestry and its implications on forest trees. Tropical Ecology 42: 283-286.

Lal, R., 2004. The potential of carbon sequestration in soils of south Asia. ISCO 2004 - 13th International Soil Conservation Organisation Conference - Brisbane, July 2004. Conserving Soil and Water for Society: Sharing Solutions, Brisbane.

Lamb, D., 2002. Is it possible to reforest degraded tropical lands to achieve economic and also biodiversity benefits? In Bringing back the forests. Policies and practices for degraded lands and forests: Proceedings of an International Conference, Kuala Lumpur, Malaysia, 2002., pp. 17-25 (Eds H. C. Sim, S. Appanah and P. B. Durst). Food and Agriculture Organization of the United nations, Bangkok, Thailand.

Lee, D. K. (Ed), 2005. Restoration of degraded forest ecosystem in the south east Asian tropical region. Annual Report for Year 2005 - 2006. ASEAN-Korea Environmental Cooperation Project (AKECOP). ASEAN-Korea Environmental Cooperation Unit, Seoul.

Lee, S. S., 1999. Forest health in plantation forests in South-East Asia. Australasian Plant Pathology 28: 283-291.

Lehmann, J., Gaunt, J. and Rondon, M., 2006. Bio-char sequestration in terrestrial ecosystems―a review. Mitigation and Adaptation Strategies for Global Change 11: 403-427.

Lesueur, D., Ingleby, K., Odee, D., Chamberlain, J., Wilson, J., Manga, T. T., Sarrailh, J. M. and Pottinger, A., 2001. Improvement of forage production in Calliandra calothyrsus: methodology for the identification of an effective inoculum containing Rhizobium strains and arbuscular mycorrhizal isolates. Journal of Biotechnology 91: 269-282.

Lillesø, J. P. B., Dhakal, L. P., Kjaer, E. D., Nathan, I. and Shrestha, R., 2002. Conservation of trees through use by local people and decentralized seed distribution supported by a tree seed programme. In Proceedings of the Southeast Asian moving workshop on conservation, management and utilization of forest genetic resources, 25

Murdoch University 2011 186

References

February-10 March 2001, Thailand, pp. 241-253 (Eds J. Koskela, S. Appanah, A. P. Pedersen and M. D. Markopoulos). FORSPA Publication No. 31/2002. FORSPA and FAO, Bangkok.

Litchman, E., 2010. Invisible invaders: non-pathogenic invasive microbes in aquatic and terrestrial ecosystems. Ecology Letters 13: 1560-1572.

Löf, M., Thomsen, A. and Madsen, P., 2004. Sowing and transplanting of broadleaves (Fagus sylvatica L., Quercus robur L., Prunus avium L. and Crataegus monogyna Jacq.) for afforestation of farmland. Forest Ecology and Management 188: 113-123.

Lok, E. H., O’Hara, G. and Dell, B., 2006. Nodulation of the legume Pterocarpus indicus by diverse strains of rhizobia. Journal of Tropical Forest Science 18: 188-194.

Lu, X., Malajczuk, N., Brundrett, M. and Dell, B., 1999. Fruiting of putative ectomycorrhizal fungi under blue gum (Eucalyptus globulus) plantations of different ages in Western Australia. Mycorrhiza 8: 255-261.

Luoma-aho, T., Hong, L. T., Ramanatha Rao, V. and Sim, H. C. (Eds), 2004. APFORGEN priority species. Proceedings of the inception workshop on forest genetic resources conservation and management. Asia Pacific forest genetic Resources Programme (APFORGEN). Kepong, Malaysia, 15-18 July, 2003. IPGRI-APO, Serdang.

Macedo, M. O., Resende, A. S., Garcia, P. C., Boddey, R. M., Jantalia, C. P., Urquiaga, S., Campello, E. F. C. and Franco, A. A., 2008. Changes in soil C and N stocks and nutrient dynamics 13 years after recovery of degraded land using leguminous nitrogen- fixing trees. Forest Ecology and Management 255: 1516-1524.

MAFF, 1986. Tree species classification and diameter limits for harvest. Forest Decision 050. Ministry of Agriculture, Forestry and Fisheries (MAFF). (in Khmer), Phnom Penh.

MAFF, 2006. Master plan for national agricultural research. Ministry of Agriculture, Forestry and Fisheries (MAFF), Phnom Penh.

Manassila, M., Nuntagij, A., Kotepong, S., Boonkerd, N. and Teaumroong, N., 2007. Characterization and monitoring of selected rhizobial strains isolated from tree legumes in Thailand. African Journal of Biotechnology 6: 1393-1402.

Mandal, T. and Nielsen, N. E., 2004. An improved low-input method for establishing calliandra hedgerows on small-scale farms in western Kenya. Agroforestry Systems 60: 227-231.

Marsudi, N. D. S., Glenn, A. R. and Dilworth, M. J., 1999. Identification and characterization of fast- and slow-growing root nodule bacteria from South-Western Australian soils able to nodulate Acacia saligna. Soil Biology and Biochemistry 31: 1229-1238.

Maruyama, E., Kinoshita, I., Ishii, K., Shigenaga, H., Ohba, K. and Saito, A., 1997. Alginate-encapsulated technology for the propagation of the tropical forest trees: Cedrela odorata L., Guazuma crinita MART., and Jacaranda mimosaefolia D. DON.

Murdoch University 2011 187

References

Silvae Genetica 46: 17-23.

Maslin, B. R. and McDonald, M. W., 2004. AcaciaSearch: evaluation of Acacia as a woody crop option for southern Australia. Rural Industries Research and Development Corporation (RIRDC). RIRDC publication no. 03/017, Barton.

McArthur, W. M. and Bettenay, E., 1974. The development and distribution of the soils of the Swan Coastal Plan, Western Australia. CSIRO, Melbourne.

McDonald, J. A., 2003. Improving water retention capacities of the Kbal Chhay Watershed by reforestation reclamation initiatives. Support to Capacity Building of Sustainable Management of the Kbal Chhay Watershed Project. Danish International Development Agency, Phnom Penh.

McNamara, S., Tinh, D. V., Erskine, P. D., Lamb, D., Yates, D. and Brown, S., 2006. Rehabilitating degraded forest land in central Vietnam with mixed native species plantings. Forest Ecology and Management 233: 358-365.

Melchor-Marrroquín, J. I., Vargas-Hernández, J. J., Ferrera-Cerrato, R. and Krishnamurthy, L., 1999. Screening Rhizobium spp. strains associated with Gliricidia sepium along an altitudinal transect in Veracruz, Mexico. Agroforestry Systems 46: 25- 38.

Menna, P., Hungria, M., Barcellos, F. G., Bangele, E. V., Hess, P. N. and Martínez- Romero, E., 2006. Molecular phylogeny based on the 16S rRNA gene of elite rhizobial strains used in Brazilian commercial inoculants. Systematic and Applied Microbiology 29: 315-332.

Michaelides, E. D., 1979. Mini-monograph on Acacia cyanophylla.Technical Consultation on Fast-Growing Plantation Broadleaved Trees for Mediterranean and Temperate Zones. Lisbon, Portugal 16-20 October. FAO, Rome.

Midgley, S., Pinyopusarerk, K., Harwood, C. and Doran, J., 1996. Exotic plant species in Vietnam's economy - The contributions of Australian trees. Seminar on Environment and Development in Vietnam. Australian National University. http://coombs.anu. edu.au/~vern/env_dev/papers/pap04.html. Accessed 23 March 2008.

Midgley, S. J. and Turnbull, J. W., 2003. Domestication and use of Australian acacias: case studies of five important species. Australian Systematic Botany 16: 89-102.

Milberg, P. and Lamont, B. B., 1997. Seed/cotyledon size and nutrient content play a major role in early performance of species on nutrient-poor soils. New Phytologist 137: 665-672.

Milberg, P., Pérez-Fernández, M. A. and Lamont, B. B., 1998. Seedling growth response to added nutrients depends on seed size in three woody genera. Journal of Ecology 86: 624-632.

Miller, J. R. and Hobbs, R. J., 2007. Habitat restoration—Do we Know what we’re doing? Restoration Ecology 15: 382-390.

Moormann, F. R. and Rojanasoonthon, S., 1967. Thailand. General soil conditions.

Murdoch University 2011 188

References

Land Development Department, Kasetsart University, the Applied Scientific Research Corporation of Thailand. http://eusoils.jrc.it/esdb_archive/EuDASM/Asia/images/ maps/th2004_1so.jpg. Accessed 25 June 2008.

Mridha, M. A. U. and Dhar, P. P., 2007. Biodiversity of arbuscular mycorrhizal colonization and spore population in different agroforestry trees and crop species growing in Dinajpur, Bangladesh. Journal of Forestry Research 18: 91-96.

Nandakwang, P., Elliott, S., Youpensuk, S., Dell, B., Teaumroong, N. and Lumyong, S., 2008. Arbuscular mycorrhizal status of indigenous tree species used to restore seasonally dry tropical forest in Northern Thailand. Research Journal of Microbiology 3: 51-61.

National Academy of Sciences, 1981. Sowing forests from the air. National Academy Press, Washington D.C.

Nativ, R., Ephrath, J. E., Berliner, P. R. and Saranga, Y., 1999. Drought resistance and water use efficiency in Acacia saligna. Australian Journal of Botany 47: 577-586.

Nawir, A. A. and Gumartini, T., 2002. Company-community partnership outgrower schemes in forestry plantations in Indonesia: an alternative to conventional rehabilitation programmes. In Bringing back the forests. Policies and practices for degraded lands and forests, pp. 317-329 (Eds H. C. Sim, S. Appanah and P. B. Durst). Food and Agricultural Organization of the United Nations, Kuala Lumpur, Malaysia.

Newton, A., Oldfield, S., Fragoso, G., Mathew, P., Miles, L. and Edwards, M., 2003. Towards a global tree conservation atlas: mapping the status and distribution of the world’s threatened tree species. UNEP-WCMC/FFI.

Nghia, N. H., 2000. Use of native species in forest rehabilitation and conservation in Vietnam. In Proceedings of the Workshop on Forest Restoration for Wildlife Conservation, 30 January - 4 February 2000, Chiang Mai, Thailand, pp. 105-107 (Eds S. Elliott, J. Kerby, D. Blakesley, K. Hardwick, K. Woods and V. Anusarnsunthorn). International Tropical Timber Organization and The Forest Restoration Research Unit, Chiang Mai.

Nghia, N. H., 2004. Status of forest genetic resources conservation and management in Vietnam. In Forest genetic resources conservation and management. Proceedings of the Asia Pacific Forest Genetic Resources Programme (APFORGEN) Inception Workshop, Kepong, Malaysia, 15 - 18 July, 2003, pp. 290-301 (Eds T. Luoma-aho, L. T. Hong, V. Ramanatha Rao and H. C. Sim). IPGRI-APO, Serdang.

Nichols, J. D., Rosemeyer, M. E., Carpenter, F. L. and Kettler, J., 2001. Intercropping legume trees with native timber trees rapidly restores cover to eroded tropical pasture without fertilization. Forest Ecology and Management 152: 195-209.

Nishio, M., 1996. Microbial fertilizers in Japan. Food and Fertilizer Technology Center Extension Bulletin. Food and Fertilizer Technology Center, Taipei.

Noor, M. M., Aminah, H. and Zaki, A. M., 2002. Merawan Siput Jantan. In A manual for forest plantation establishment in Malaysia, pp. 189-197 (Ed B. Krishnapillay). Forest Research Institute Malaysia, Kuala Lumpur.

Murdoch University 2011 189

References

Odenyo, A. A., Osuji, P. O. and Karanfil, O., 1997. Effect of multipurpose tree (MPT) supplements on ruminal ciliate protozoa. Animal Feed Science Technology 67: 169-180.

Oldfield, S., 2008. Choices for tree conservation. Oryx 42: 159-160.

Oldfield, S., Lusty, C. and MacKinven, A., 1998. The world list of threatened trees. World Conservation Press, Cambridge.

Olsson, P. A., Thingstrup, I., Jakobsen, I. and Bååth, E., 1999. Estimation of the biomass of arbuscular mycorrhizal fungi in a linseed field. Soil Biology and Biochemistry 31: 1879-1887.

Ortega, U., Duñabeitia, M., Menendez, S., Gonzalez-Murua, C. and Majada, J., 2004. Effectiveness of mycorrhizal inoculation in the nursery on growth and water relations of Pinus radiata in different water regimes. Tree Physiology 24: 65-73.

Orzeszyna, H., Garlikowski, D. and Pawlowski, A., 2006. Using of geocomposite with superabsorbent synthetic polymers as water retention element in vegetative layers. International Agrophysics 20: 201-206.

OTA, 1983. Sustaining tropical forest resources: Reforestation of degraded lands (Background paper No.1). Office of Technology Assessment (OTA). U.S. Government Printing Office, Washington D.C.

Otsamo, A., Ådjers, G., Hadi, T. S., Kuusipalo, J., Tuomela, K. and Vuokko, R., 1995. Effect of site preparation and initial fertilization on the establishment and growth of four plantation tree species used in reforestation of Imperata cylindrica (L.) Beauv. dominated grasslands. Forest Ecology and Management 73: 271-277.

Otsamo, A., Ådjers, G., Hadi, T. S., Kuusipalo, J. and Vuokko, R., 1997. Evaluation of reforestation potential of 83 tree species planted on Imperata cylindrica dominated grassland. A case study from South Kalimantan, Indonesia. New Forests 14: 127-143.

Otsamo, R., 1998b. Removal of Acacia mangium overstorey increased growth of underplanted Anisoptera marginata (Dipterocarpaceae) on an Imperata cylindrica grassland site in South Kalimantan, Indonesia. New Forests 16: 71-80.

Ouchi, S., Nishikawa, A. and Kameda, E., 1990. Soil-improving effect of a super-water- absorbent-polymer II. Evaporation, leaching of salts and growth of vegetables. Japanese Journal of Soil Science and Plant Nutrition 61: 606-613.

Palasuwan, A., Soogarun, S., Lertlum, T., Pradniwat, P. and Wiwanitkit, V., 2005. Inhibition of heinz body induction in an invitro model and total antioxidant activity of medicinal Thai plants. Asian Pacific Journal of Cancer Prevention 6: 458-463.

Pampolina, N. M., de la Cruz, R. E. and Garcia, M. U., 1994. Ectomycorrhizal roots and fungi of Philippine Dipterocarps. In Proceedings of the International Symposium and Workshop on Mycorrhiza for Plantation Forestry in Asia, Kaiping, Guangdong province, P.R. China, Vol. 62, pp. 47-50 (Ed M. Brundett, Dell, B., Malajcuzk, N., and Gong Migquin). Australian Centre for International Agricultural Research (ACIAR), Canberra.

Murdoch University 2011 190

References

Panaia, M., 2006. Developing synthetic seeds for clonal propagation of Australian plants-Somatic embryogenesis as a precursor to synthetic seeds. Rural Industries Research and Development Corporation. Australian Goverment, Canberra.

Pausas, J. G., Bladé, C., Valdecantos, A., Seva, J. P., Fuentes, D., Alloza, J. A., Vilagrosa, A., Bautista, S., Cortina, J. and Vallejo, R., 2004. Pines and oaks in the restoration of Mediterranean landscapes of Spain: New perspectives for an old practice―a review. Plant Ecology 171: 209-220.

Pedersen, A. P., 2000. Tree species selection in Thailand: Various species for various purposes. In Proceedings of the Workshop on Forest Restoration for Wildlife Conservation, 30 January - 4 February 2000, Chiang Mai, Thailand, pp. 223-233 (Eds S. Elliott, J. Kerby, D. Blakesley, K. Hardwick, K. Woods and V. Anusarnsunthorn). International Tropical Timber Organization and The Forest Restoration Research Unit, Chiang Mai.

Peoples, M. B., Faizah, A. W., Rerkasem, B. and Herridge, D. F., 1989. Methods for evaluating nitrogen fixation by nodulated legumes in the field. Australian Centre for International Agricultural Research, Canberra.

Pereira, R. M., da Silveira, E. L., Carareto-Alves, L. M. and Lemos, E. G. M., 2008. Evaluation of possible rhizobacteria populations in soils under forest species. Revista Brasileira de Ciência do Solo 32: 1921-1927.

Phongoudome, C., no date. Species monograph no. 49. Xylia xylocarpa (Leguminosae, Mimosoideae,Bean or pea family) Mai Daeng. Lao Tree Seed Project (LTSP), Vientiane.

Phongoudome, C. and Mounlamai, K., 2004. Status of forest genetic resources conservation and management in Lao PDR. In Forest genetic resources conservation and management. Proceedings of the Asia Pacific Forest Genetic Resources Programme (APFORGEN) Inception Workshop, Kepong, Malaysia, 15 - 18 July, 2003, pp. 183-205 (Eds T. Luoma-aho, L. T. Hong, V. Ramanatha Rao and H. C. Sim). IPGRI-APO, Serdang.

Pickford, S., Suharti, M. and Wibowo, A., 1992. A note on fuelbeds and fire behavior in alang-alang (Imperata cylindrica). International Journal of Wildland Fire 2: 41-46.

Piggott, J. P., Brown, P. H. and Williams, M. R., 2002. Direct seeding trees on farmland in the Western Australian Wheatbelt. Resource Management Technical Report No. 146, Department of Agriculture Western Australia.

Pijnenborg, J. W. M., Lie, T. A. and Zehnder, A. J. B., 1991. Nodulation of lucerne (Medicago sativa L.) in an acid soil: Effects of inoculum size and lime-pelleting. Plant and Soil 131: 1-10.

Pinyopusarerk, K. and Kalinganire, A., 2003. Domestication of Chukrasia. Australian Centre for International Agricultural Research Monograph No 98. CSIRO Forestry and Forest Products, Canberra.

Polymer innovations, 2005. Water$ave. http://www.polymerinnovations. com.au/?pid =120&page=40. Accessed 15 April 2008.

Murdoch University 2011 191

References

Poopathy, V., Appanah, S. and Durst, P. B. (Eds), 2005. Helping forests take cover. RAP PUBLICATION 2005/13. Food and Agriculture Organization of the United Nations. Regional Centre for Asia and the Pacific, Bangkok.

Prachaiyo, B., 2000. Farmers and forests: a changing phase in Northeast Thailand. Southeast Asian Studies 38: 271-446.

Price, G. (Ed), 2006. Australian soil fertility manual. Fertilizer Industry Federation of Australia and CSIRO Publishing, Collingwood.

Prin, Y., Galiana, A., Le Roux, C., Méléard, B., Razafimaharo, V., Ducousso, M. and Chaix, G., 2003. Molecular tracing of Bradyrhizobium strains helps to correctly interpret Acacia mangium response to inoculation in a reforestation experiment in Madagascar. Biology and Fertility of Soils 37: 64-69.

Pringle, A., Bever, J. D., Gardes, M., Parrent, J. L., Rillig, M. C. and Klironomos, J. N., 2009. Mycorrhizal symbioses and plant invasions. Annual Review of Ecology, Evolution, and Systematics 40: 699-715.

Pule-Meulenberg, F. and Dakota, F. D., 2009. Assessing the symbiotic dependency of grain and tree legumes on N2 fixation for their N nutrition in five agro-ecological zones of Botswana. Symbiosis 48: 68-77.

Quesada, M. and Stoner, K. E., 2004. Threats to the conservation of tropical dry forest in Costa Rica. In Biodiversity conservation in Costa Rica: Learning the lessons in a seasonal dry forest, pp. 266-280 (Eds G. W. Frankie, A. Mata and S. B. Vinson). University of California Press, Berkeley.

Rai, M. K., Jaiswal, V. S. and Jaiswal, U., 2008. Encapsulation of shoot tips of guava (Psidium guajava L.) for short-term storage and germplasm exchange. Scientia Horticulturae 118: 33-38.

Raman, N. and Selvaraj, T., 2006. Tripartite relationship of Rhizobium, AMF, and host in growth promotion. In Handbook of Microbial Biofertilizers, pp. 51-88 (Ed M. K. Rai). Haworth Press, Binghamton.

Rao, A. V. and Tak, R., 2002. Growth of different tree species and their nutrient uptake in limestone mine spoil as influenced by arbuscular mycorrhizal (AM)-fungi in Indian arid zone. Journal of Arid Environments 51: 113-119.

Rasolomampianina, R., Bailly, X., Fetiarison, R., Rabevohitra, R., Béna, G., Ramaroson, L., Raherimandimby, M., Moulin, L., De Lajudie, P., Dreyfus, B. and Avarre, J. C., 2005. Nitrogen-fixing nodules from rose wood legume trees (Dalbergia spp.) endemic to Madagascar host seven different genera belonging to α - and β - Proteobacteria. Molecular Ecology 14: 4135-4146.

Reddell, P. and Warren, R., 1987. Inoculation of acacias with mycorrhizal fungi: potential benefits. In Australian acacias in developing countries. Australian Centre for International Agricultural Research proceedings no. 16, pp. 50-53 (Ed J. W. Turnbull). Australian Centre for International Agriculture Research, Canberra.

Repáč, I., 2007. Ectomycorrhiza formation and growth of Picea abies seedlings

Murdoch University 2011 192

References inoculated with alginate-bead fungal inoculum in peat and bark compost substrates. Forestry 80: 517-530.

Requena, N., Jeffries, P. and Barea, J. M., 1996. Assessment of natural mycorrhizal potential in a desertified semiarid ecosystem. Applied and Environmental Microbiology 62: 842-847.

Requena, N., Perez-Solis, E., Azcón-Aguilar, C., Jeffries, P. and Barea, J. M., 2001. Management of indigenous plant-microbe symbioses aids restoration of desertified ecosystems. Applied and Environmental Microbiology 67: 495-498.

Richardson, A. E., Viccars, L. A., Watson, J. M. and Gibson, A. H., 1995. Differentiation of Rhizobium strains using the polymerase chain reaction with random and directed primers. Soil Biology and Biochemistry 27: 515-524.

Rippey, E. and Rowland, B., 2004. Coastal plants. Perth and the south-west region. University of Western Australia Press, Perth.

Rodríguez-Echeverría, S., 2010. Rhizobial hitchhikers from Down Under: invasional meltdown in a plant–bacteria mutualism? Journal of Biogeography 37: 1611-1622.

Rokich, D. P., Dixon, K. W., Sivasithamparam, K. and Meney, K. A., 2002. Smoke, mulch, and seed broadcasting effects on woodland restoration in Western Australia. Restoration Ecology 10: 185-194.

Rondon, M. A., Lehmann, J., Ramírez, J. and Hurtado, M., 2007. Biological nitrogen fixation by common beans (Phaseolus vulgaris L.) increases with bio-char additions. Biology and Fertility of Soils 43: 699-708.

Roper, M. M., 2006. Potential for remediation of water repellent soils by inoculation with wax-degrading bacteria in south-western Australia. Biologia, Bratislava 61: S358- S362.

Rosketko, J. M. and Westley, S. B. (Eds), 1994. Dalbergia sissoo production and use: a field manual. Nitrogen Fixing Tree Association and Taiwan Forestry Research Institute, Morrilton.

Russell, E. W., 1973. Soil conditions and plant growth. Longman, London.

Ruthrof, K. and Close, D., 2006. Tuart regeneration and restoration. Government of Western Australia, Perth.

Ruthrof, K. X., Loneragan, W. A. and Yates, C. J., 2003. Comparative population dynamics of Eucalyptus cladocalyx in its native habitat and as an invasive species in an urban bushland in south-western Australia. Diversity and Distributions 9: 469-483.

Sae-Lee, S., Vityakon, P. and Prachaiyo, B., 1992. Effects of trees on paddy bund on soil fertility and rice growth in Northeast Thailand. Agroforestry Systems 18: 213-223.

Saggin-Júnior, O. J. and da Silva, E. M. R., 2006. Production of seedlings inoculated with arbuscular mycorrhizal fungi and their performance after outplanting. In Handbook of microbial biofertilizers, pp. 353-394 (Ed M. K. Rai). Haworth Press, Binghamton.

Murdoch University 2011 193

References

Saignaphet, S., 1995. Re-afforestation activities in Savannakhet province. In National Workshop on Strengthening Re-afforestation Programmes in Lao PDR, Vientiane, 19 - 21 June 1995, pp. 51-55 (Eds J. Eastman and F. Danborg). FAO Regional Project STRAP. Field Document No. 4 GCP/RAS/142/JPN, Rome.

Saiprasad, G. V. S., 2001. Artificial seeds and their applications. Resonance-Journal of Science Education 6: 39-47.

Saito, M. and Marumoto, T., 2002. Inoculation with arbuscular mycorrhizal fungi: the status quo in Japan and the future prospects. Plant and Soil 244: 273-279.

Sanginga, N., Danso, S. K. A., Mulongoy, K. and Ojeifo, A. A., 1994. Persistence and recovery of introduced Rhizobium ten years after inoculation on Leucaena leucocephala grown on an Alfisol in southwestern Nigeria. Plant and Soil 159: 199-204.

Sarkar, D. and Naik, P. S., 1998. Synseeds in potato: an investigation using nutrient- encapsulated in vitro nodal segments. Scientia Horticulturae 73: 179-184.

Sati, K., 1995. Re-afforestation activities in Bo Keo province. In National Workshop on Strengthening Re-afforestation Programmes in Lao PDR. Vientiane, 19 - 21 June 1995, pp. 92-94 (Eds J. Eastman and F. Danborg). FAO Regional Project STRAP. Field Document No. 4 GCP/RAS/142/JPN, Rome.

Savill, P., Evans, J., Auclair, D. and Falck, J., 1997. Plantation silviculture in Europe. Oxford University Press, Oxford.

Schafer, E. H., 1957. Rosewood, dragon's blood, and lac. Journal of the American Oriental Society 77: 129-136.

Schmidt, L., 2000. Guide to handling of tropical and subtropical forest seed. Danida Forest Seed Center, Humlebaek.

Schmidt, L., 2008. A review of direct sowing versus planting in tropical afforestation and land rehabilitation. Development and Environment Series 10-2008. Forest & Landscape Denmark, Aalborg.

Schroth, G., Salazar, E. and da Silva, J. P., 2001. Soil nitrogen mineralization under tree crops and a legume cover crop in multi-strata agroforestry in central Amazonia: spatial and temporal patterns. Experimental Agriculture 37: 253-267.

Schwartz, M. W., Hoeksema, J. D., Gehring, C. A., Johnson, N. C., Klironomos, J. N., Abbott, L. K. and Pringl, A., 2006. The promise and the potential consequences of the global transport of mycorrhizal fungal inoculum. Ecology Letters 9: 501-515.

See, L. S. and Alexander, I. J., 1994. The response of seedlings of two dipterocarp species to nutrient additions and ectomycorrhizal infection. Plant and Soil 163: 299- 306.

Semhi, K., Chaudhuri, S., Clauer, N. and Boeglin, J. L., 2008. Impact of termite activity on soil environment: A perspective from their soluble chemical components. International Journal of Environmental Science and Technology 5: 431-444.

Murdoch University 2011 194

References

Setiadi, Y., 2000. Mycorrhizal seedling production for enhancing rehabilitation of degraded forest in Indonesia. In Proceedings of the Workshop on Forest Restoration for Wildlife Conservation, 30 January - 4 February 2000, Chiang Mai, Thailand, pp. 235- 243 (Eds S. Elliott, J. Kerby, D. Blakesley, K. Hardwick, K. Woods and V. Anusarnsunthorn). International Tropical Timber Organization and The Forest Restoration Research Unit, Chiang Mai.

Singh, S. S., Tiwari, S. C. and Dkhar, M. S., 2003. Species diversity of vesicular- arbuscular mycorrhizal (VAM) fungi in jhum fallow and natural forest soils of Arunachal Pradesh, northeastern India. Tropical Ecology 44: 207-215.

Siregar, U. J., Rachmi, A., Massijaya, M. Y., Ishibashi, N. and Ando, K., 2007. Economic analysis of sengon (Paraserianthes falcataria) community forest plantation, a fast growing species in East Java, Indonesia. Forest Policy and Economics 9: 822-829.

Sivapalan, S., 2001. Effect of polymer on soil water holding capacity and plant water use efficiency. Proceedings 10th Australian Agronomy Conference. Hobart.

So, N. V., 2000. The potential of local tree species to accelerate natural forest succession on marginal grasslands in southern Vietnam. In Proceedings of the Workshop on Forest Restoration for Wildlife Conservation, 30 January - 4 February 2000, Chiang Mai, Thailand, pp. 135-148 (Eds S. Elliott, J. Kerby, D. Blakesley, K. Hardwick, K. Woods and V. Anusarnsunthorn). International Tropical Timber Organization and The Forest Restoration Research Unit, Chiang Mai.

Sodhi, N. S., Koh, L. P., Brook, B. W. and Ng, P. K. L., 2004. Southeast Asian biodiversity: an impending disaster. TRENDS in Ecology and Evolution 19: 654-660.

Soerianegara, I. and Lemmens, R. H. M. J. (Eds), 1994. Plant resources of South-East Asia no. 5 (1). Timber trees: Major commercial timbers. PROSEA, Bogor.

Somasegaran, P. and Hoben, H. J., 1994. Handbook for rhizobia. Springer-Verlag, New York.

Soumphonphakdy, B., 1995. Re-afforestation programmes in Vientiane province. In National Workshop on Strengthening Re-afforestation Programmes in Lao PDR, Vientiane, 19 - 21 June 1995, pp. 45-48 (Eds J. Eastman and F. Danborg). FAO Regional Project STRAP. Field Document No. 4 GCP/RAS/142/JPN, Rome.

Sovu, Savadogo, P., Tigabu, M. and Odén, P. C., 2010. Restoration of former grazing lands in the highlands of Laos using direct seeding of four native tree species: Seedling establishment and growth performance. Mountain Research and Development 30: 232- 243.

Stahl, J. D., Cameron, M. D., Haselbach, J. and Aust, S. D., 2000. Biodegradation of superabsorbent polymers in soil. Environmental Science and Pollution Research 7: 83- 88.

Steiner, C., Teixeira, W. G., Lehmann, J., Nehls, T., de Macêdo, J. L. V., Blum, W. E., H. and Zech, W., 2007. Long term effects of manure, charcoal and mineral fertilization on crop production and fertility on a highly weathered Central Amazonian upland soil. Plant and Soil 291: 275-290.

Murdoch University 2011 195

References

Stephens, J. H. G. and Rask, H. M., 2000. Inoculant production and formulation. Field Crops Research 65: 249-258.

Strange, N., Theilade, I., So, T., Sloth, A. and Helles, F., 2007. Integration of species persistence, costs and conflicts: An evaluation of tree conservation strategies in Cambodia. Biological Conservation 137: 223-236.

Strijdom, B. W. and van Rensburg, H. J., 1981. Effect of steam sterilization and gamma irradiation of peat on quality of Rhizobium inoculants. Applied and Environmental Microbiology 41: 1344-1347.

Sumantakul, V., 2004. Status of forest genetic resources conservation and management in Thailand. In Forest genetic resources conservation and management. Proceedings of the Asia Pacific Forest Genetic Resources Programme (APFORGEN) Inception Workshop, Kepong, Malaysia, 15 - 18 July, 2003, pp. 265-289 (Eds T. Luoma-aho, L. T. Hong, V. Ramanatha Rao and H. C. Sim). IPGRI-APO, Serdang.

Sunderlin, W. D., 2006. Poverty alleviation through community forestry in Cambodia, Laos, and Vietnam: An assessment of the potential. Forest Policy and Economics 8: 386-396.

Svasti, M. R. S., 2000. Rivers in jeopardy: A village community's response to the destruction of their upper watershed forests in the Mae Soi valley catchment, Northern Thailand. In Proceedings of the Workshop on Forest Restoration for Wildlife Conservation, 30 January - 4 February 2000, Chiang Mai, Thailand, pp. 123-134 (Eds S. Elliott, J. Kerby, D. Blakesley, K. Hardwick, K. Woods and V. Anusarnsunthorn). International Tropical Timber Organization and The Forest Restoration Research Unit, Chiang Mai.

Tamrakar, P. R., 2003. State of forest genetic resources conservation and management in Nepal. Forest genetic resources working papers, Working Paper FGR/69E. Forest Resources Development Service, Forest Resources Division. FAO, Rome.

Tawaraya, K., Takaya, Y., Turjaman, M., Tuah, S. J., Limin, S. H., Tamai, Y., Cha, J. Y., Wagatsuma, T. and Osaki, M., 2003. Arbuscular mycorrhizal colonization of tree species grown in peat swamp forests of Central Kalimantan, Indonesia. Forest Ecology and Management 182: 381-386.

Taylor, T. S., Loewenstein, E. F. and Chappelka, A. H., 2006. Effect of animal browse protection and fertilizer application on the establishment of planted Nuttall oak seedlings. New Forests 32: 133-143.

Teketay, D. and Granström, A., 1997. Germination ecology of forest species from the highlands of Ethiopia. Journal of Tropical Ecology 13: 805-831.

Temprano, F. J., Albareda, M., Camacho, M., Daza, A., Santamaria, C. and Rodriguez- Navarro, D. N., 2002. Survival of several Rhizobium/Bradyrhizobium strains on different inoculant formulations and inoculated seeds. International Microbiology 5: 81- 86.

Teste, F. P., Schmidt, M. G., Berch, S. M., Bulmer, C. and Egger, K. N., 2004. Effects of ectomycorrhizal inoculants on survival and growth of interior Douglas-fir seedlings

Murdoch University 2011 196

References on reforestation sites and partially rehabilitated landings. Canadian Journal of Forest Research 34: 2074-2088.

Thai Meteorological Department, 2008. Thailand annual weather summary, 2008. Thai Meteorological Department. http://www.tmd.go.th/en. Accessed 16 December 2009.

Theilade, I., Luoma-aho, T., Rimbawanto, A., Nguyen, H. N., Greijmans, M., Nashatul, Z. N. A., Sloth, A., So, T. and Burgess, S., 2005. An overview of the conservation status of potential plantation and restoration species in Southeast Asia. A paper presented at the Symposium on Tropical Rainforest Rehabilitation and Restoration - Existing Knowledge and Future Directions. Kota Kinabalu, 26-28 June, 2005.

Thies, J. E., Singleton, P. W. and Bohlool, B. B., 1991. Influence of the size of indigenous rhizobial populations on establishment and symbiotic performance of introduced rhizobia on field-grown legumes. Applied and Environmental Microbiology 57: 19-28.

Thomson, B. D., Grove, T. S., Malajczuk, N. and Hardy, G. E. S. J., 1994. The effectiveness of ectomycorrhizal fungi in increasing the growth of Eucalyptus globulus Labill. in relation to root colonization and hyphal development in soil. New Phytologist 126: 517-524.

Thomson, L., Midgley, S., Pinyopusarerk, K. and Kalinganire, A., 2001. Tree domestication: the Australian experience in partnerships with special reference to the Asia-Pacific region. In Southeast Asian Moving Workshop on Conservation, Management and Utilisation of Forest Genetic Resources. Bangkok (Thailand), 25 Feb - 10 Mar 2001, pp. 207-221 (Eds J. Koskela, S. Appanah, A. P. Pedersen and M. D. Markopoulos). FORSPA Publication (FAO), no. 31, Bangkok.

Thrall, P. H., Millsom, D. A., Jeavons, A. C., Waayers, M., Harvey, G. R., Bagnall, D. J. and Brockwell, J., 2005. Seed inoculation with effective root-nodule bacteria enhance revegetation success. Journal of Applied Ecology 42: 740-751.

Thrall, P. H., Murray, B. R., Watkin, E. L. J., Woods, M. J., Baker, K., Burdon, J. J. and Brockwell, J., 2001. Bacterial partnerships enhance the value of native legumes in rehabilitation of degraded agricultural lands. Ecological Management and Restoration 2: 233-235.

Tiné, M. A. S., Cortelazzo, A. L. and Buckeridge, M. S., 2000. Xyloglucan mobilisation in cotyledons of developing plantlets of Hymenaea courbaril L. (Leguminosae- Caesalpinoideae). Plant Science 154: 117-126.

Top, N., Mizoue, N., Ito, S., Kai, S., Nakao, T. and Ty, S., 2009. Effects of population density on forest structure and species richness and diversity of trees in Kampong Thom Province, Cambodia. Biodiversity and Conservation 18: 717-738.

Tri, M. V. and Khanh, T. C., 1996. Rules and regulations of the Government of Vietnam on collection and exportation of biological materials. Journal of Ethnopharmacology 51: 173-176.

Tsai, L. M. and Faridah-Hanum, I., 1992. Indigenous species plantations for Malaysia. In Proceedings of the National Seminar on Indigenous Species for Forest Plantations, 23

Murdoch University 2011 197

References

- 24 April 1992, Faculty of Forestry, Universiti Pertanian Malaysia, Serdang, pp. 1-7 (Eds A. S. Sajap, R. A. Kader, M. S. H. Othman, A. Mohamed, F. H. Ibrahim and M. H. Sahri). Faculty of Forestry, Universiti Pertanian Malaysia, Serdang.

Tunjai, P., 2005. Appropriate tree species and techniques for direct seeding for forest restoration in Chiang Mai and Lamphun Provinces. MSc thesis. Chiang Mai University, Chiang Mai.

Turnbull, J. W. and Booth, T. H., 2002. Eucalypts in cultivation: an overview. In Eucalyptus: The Genus Eucalyptus, Vol. 22, pp. 52-74 (Ed J. J. W. Coppen). Taylor & Francis, London.

Turner, S. R., Pearce, B., Rokich, D. P., Dunn, R. R., Merritt, D. J., Majer, J. D. and Dixon, K. W., 2006. Influence of polymer seed coatings, soil raking, and time of sowing on seedling performance in post-mining restoration. Restoration Ecology 14: 267-277.

Uhl, C., Buschbacher, R. and Serrão, E. A. S., 1988. Abandoned pastures in eastern Amazonia. I. Patterns of plant succession. Journal of Ecology 76: 663-681.

UNEP-WCMC, 2007. Tree Conservation Information Service. UNEP World Conservation Monitoring Centre, Cambridge.

Urgiles, N., Loján, P., Aguirre, N., Blaschke, H., Günter, S., Stimm, B. and Kottke, I., 2009. Application of mycorrhizal roots improves growth of tropical tree seedlings in the nursery: a step towards reforestation with native species in the Andes of Ecuad. New Forests 38: 229-239. van Rossum, D., Muyotcha, A., de Hoop, B. M., van Verseveld, H. W., Stouthamer, A. H. and Boogerd, F. C., 1994. Soil acidity in relation to groundnut-Bradyrhizobium symbiotic performance. Plant and Soil 163: 165-175.

Varmola, M. I. and Carle, J. B., 2002. The importance of hardwood plantations in the tropics and sub-tropics. International Forestry Review 4: 110-121.

Vellinga, E. C., Wolfe, B. E. and Pringle, A., 2009. Global patterns of ectomycorrhizal introductions. New Phytologist 181: 960-973.

Vessey, J. K., 2004. Benefits of inoculating legume crops with rhizobia in the northern Great Plains. Crop Management: Online, doi:10.1094/CM-2004-0301-1004-RV.

Vidakovic, M. and Ahsan, J., 1970. The inheritance of crooked bole in shisham (Dalbergia sissoo Roxb.). Silvae Genetica 19: 94-98.

VietNamNet Bridge, 2007. Rare trees threatened nationwide. http://english. vietnamnet.vn/social/2007/08/732584. Accessed 12 November 2008.

Vincent, A. and Davies, S. J., 2003. Effects of nutrient addition, mulching and planting- hole size on early performance of Dryobalanops aromatica and Shorea parvifolia planted in secondary forest in Sarawak, Malaysia. Forest Ecology and Management 180: 261-271.

Vũ, V. D. (Ed), 1996. Vietnam forest trees. Agricultural Publishing House, Hanoi.

Murdoch University 2011 198

References

Wade, M. R., Gurr, G. M. and Wratten, S. D., 2008. Ecological restoration of farmland: progress and prospects. Philosophical Transactions of the Royal Society B: Biological Sciences 363: 831-847.

Wang, W. G., Wang, S. H., Wu, X. A., Jin, X. Y. and Chen, F., 2007. High frequency plantlet regeneration from callus and artificial seed production of rock plant Pogonatherum paniceum (Lam.) Hack. (Poaceae). Scientia Horticulturae 113: 196-201.

Wardle, D. A., Zackrisson, O. and Nilsson, M. C., 1998. The charcoal effect in boreal forests: mechanisms and ecological consequences. Oecologia 115: 419-426.

Warnock, D. D., Lehmann, J., Kuyper, T. W. and Rillig, M. C., 2007. Mycorrhizal responses to biochar in soil―concepts and mechanisms. Plant and Soil 300: 9-20.

Weinland, G., 1998. Plantations. In A review of dipterocarps: Taxonomy, ecology and silviculture, pp. 151-185 (Eds S. Appanah and J. M. Turnbull). Center for International Forestry Research, Bogor.

Whiteman, P. G., 1972. The effects of inoculation and nitrogen application on seedling growth and nodulation of Glycine wightii and Phaseolus astropurpureus in the field. Tropical Grasslands 6: 11-16.

Willan, R. L., 1985. A guide to forest seed handling with special reference to the tropics. Food and Agriculture Organization of the United Nations, Rome.

Willoughby, I. and Jinks, R. L., 2009. The effect of duration of vegetation management on broadleaved woodland creation by direct seeding. Forestry 82: 343-359.

Witkowski, E. T. F., 1991. Growth and competition between seedlings of Protea repens (L.) L. and the alien invasive, Acacia saligna (Labill.) Wendl. in relation to nutrient availability. Functional Ecology 5: 101-110.

Wolfe, B. E., Richard, F., Cross, H. B. and Pringle, A., 2010. Distribution and abundance of the introduced ectomycorrhizal fungus Amanita phalloides in North America. New Phytologist 185: 803-816.

Woods, K. and Elliott, S., 2004. Direct seeding for forest restoration on abandoned agricultural land in Northern Thailand. Journal of Tropical Forest Science 16: 248-259.

Wu, S. F., Wu, P. T., Feng, H. and Bu, C. F., 2010. Influence of amendments on soil structure and soil loss under simulated rainfall China’s loess plateau. African Journal of Biotechnology 9: 6116-6121.

Wu, T., Hao, W., Lin, X. and Shi, Y., 2002. Screening of arbuscular mycorrhizal fungi for the revegetation of eroded red soils in subtropical China. Plant and Soil 239: 225- 235.

WWF, 2004. Are protected areas working? An analysis of forest protected areas by WWF. The World Wide Fund For Nature (WWF) International, Gland.

Xu, D., Dell, B., Malajczuk, N. and Gong, M., 2002. Effect of phosphorus application and ectomycorrhizal fungal inoculation on biomass production of Eucalyptus urophylla

Murdoch University 2011 199

References plantation in South China. Plant nutrition – Food Security and Sustainability of Agro- Ecosystems 92: 646-647.

Yanagi, M. and Yamasato, K., 1993. Phylogenetic analysis of the family Rhizobiaceae and related bacteria by sequencing of 16S rRNA gene using PCR and DNA sequencer. FEMS Microbiology Letters 107: 115-120.

Yates, C. J. and Hobbs, R. H., 1997a. Temperate eucalypt woodlands: a review of their status, processes threatening their persistence and techniques for restoration. Australian Journal of Botany 45: 949-973.

Yates, C. J. and Hobbs, R. H., 1997b. Woodland restoration in the Western Australian wheatbelt: a conceptual framework using a state and transition model. Restoration Ecology 5: 28-35.

Yates, R. J., Howieson, J. G., Nandasena, K. G. and O’Hara, G. W., 2004. Root-nodule bacteria from indigenous legumes in the north-west of Western Australia and their interaction with exotic legumes. Soil Biology and Biochemistry 36: 1319-1329.

Youpensuk, S., Lumyong, S., Dell, B. and Rerkasem, B., 2004. Arbuscular mycorrhizal fungi in the rhizosphere of Macaranga denticulata Muell. Arg., and their effect on the host plant. Agroforestry Systems 60: 239-246.

Yuwa-Amornpitak, T., Vichitsoonthonkul, T., Tanticharoen, M., Cheevadhanarak, S. and Ratchadawong, S., 2006. Diversity of ectomycorrhizal fungi on Dipterocarpaceae in Thailand. Journal of Biological Sciences 6: 1059-1064.

Zahran, H. H., 1999. Rhizobium-legume symbiosis and nitrogen fixation under severe conditions and in an arid climate. Microbiology and Molecular Biology Reviews 63: 968-989.

Zarcinas, B. A., Cartwright, B. and Spouncer, L. R., 1987. Nitric acid digestion and multi-element analysis of plant material by inductively coupled plasma spectrometry. Communications in Soil Science and Plant Analysis 18: 131-141.

Zhong, C. L., Gong, M. Q., Chen, Y. and Wang, F. Z., 1994. Inoculation of Casuarina with ectomycorrhizal fungi, vesicular-arbuscular mycorrhizal fungi and Frankia. In Proceeding of an International Symposium and Workshop on Mycorrhizas for Plantation Forestry in Asia, Kaiping, Guangdong Province, P.R. China, Vol. 62, pp. 122-126 (Eds M. Brundrett, B. Dell, N. Malajczuk and G. Mingquin). Australian Centre for International Agricultural Research (ACIAR), Canberra.

Murdoch University 2011 200