Tài liệu Rubber plantation performance in the Northeast and East of Thailand in relation to environmental conditions docx

Lately, the rubber plantation area hasbeen decreasing in Malaysia, but in Thailand the trend has been reverse and plantations havestarted to spread to new areas in the East and Northeast

Rubber plantation performance in the Northeast and East of Thailand

in relation to environmental conditions

Laura Rantala

A thesis submitted for an M.Sc degree in Forest Ecology

Department of Forest Ecology/

Viikki Tropical Resources Institute (VITRI)

University of Helsinki

Finland

2006

This M.Sc thesis was done under the framework of a project “Improving the productivity ofrubber smallholdings through rubber agroforestry systems in Indonesia and Thailand” Theproject is being financed by the Common Fund for Commodities (CFC) It is coordinated bythe World Agroforestry Centre (ICRAF), and research partners include the Indonesian RubberResearch Institute, Kasetsart University (KU) and Prince of Songkhla University in Thailand,and the University of Helsinki (UH) I received funding from the UH for travel expenses toThailand and for participation in a bilateral exchange programme between the universities ofKasetsart and Helsinki

My initial knowledge of rubber cultivation and the tropical environment was limited to saythe least I am grateful to everyone involved in this work for the time they have generouslygiven for guiding me through the various stages of this work Firstly I wish to express mygratitude to my supervisor, Professor, Dr Olavi Luukkanen (UH), Director of the ViikkiTropical Resources Institute (VITRI), for making my participation in this project possible I

am grateful for his supervision, valuable comments and interest in my work During my fieldwork in Thailand, I received much academic as well as practical help from AssociateProfessor, Dr Suree Bhumibhamon and Dr Damrong Pipatwattanakul (KU) Without theirsupport my work in Thailand would not have been possible I am indebted to Dr VesaKaarakka (UH) for his help during various stages of my work and especially for thoughtfulcomments on my manuscript

In Thailand, I had the privilege to receive help from many people I want to mention the staffmembers of the Office of the Rubber Replanting Aid Fund in Bangkok, Nong Khai andBuriram, who kindly assisted me in finding suitable sites for field study I am grateful to Mr.Arak Chantuma and Mrs Pisamai Chantuma from Chachoengsao Rubber Research Centre forproviding me with the necessary facilities and assistance with the arrangements for my fieldwork I want to thank Mr and Mrs Chorruk, Mr and Mrs Choochit and Mrs SompongPuksa in Ban Kruen, Buriram, Mrs Boonhouse Nanoy, Mr Prasittiporn Sankarn and Mr andMrs Arlapol in Pak Khat, Nong Khai and Mrs Pa Noom Thurtong in Lad Krating forinformation, hospitality and for letting me conduct field inventories in their rubberplantations My field work would have not been possible without the help of Mr PrinKalasee, Mr Jakrapong Puakla, Ms Waranuch Chansuri, Ms Supanee Nakplang and Ms.Pantaree Kongsat I want to thank Mr Chakrit Na Takuathung for helping me in findingliterature from Thailand once I had already returned to Finland Finally I want to thank allthose who helped me and were very friendly to me making my short stay in KU and inThailand an unforgettable one

I want to thank Professor, Dr Jouko Laasasenaho and Timo Melkas for helping me withcalculating wood volume estimates for trees, and Riika Kilpikari for helping me withstatistics Thanks are also due to Dr Mohamed El Fadl for help in data search and comments

as well as to other VITRI staff and students for their comments Last but not least I want tothank my family and friends for their support

Dublin, November 2006

Laura Rantala

This study was financed by the Common Fund for Commodities, an intergovernmental

financial institution established within the framework of the United Nations, headquartered in Amsterdam, the Netherlands.

1 INTRODUCTION 5

1.1 Background of the study 5

1.2 Scope and objective of the study 10

2 LITERATURE REVIEW 11

2.1 Botany and distribution of Hevea brasiliensis 11

2.1.1 Distribution of Hevea brasiliensis in Thailand 12

2.2 Climatic requirements of the rubber tree 14

2.3 Soil requirements of the rubber tree 17

2.4 Rubber cultivation in Southeast Asia 18

2.4.1 General characteristics 18

2.4.2 Agroforestry practices 19

2.4.3 Environmental considerations 21

2.5 Uses of Hevea brasiliensis 22

3 MATERIAL AND METHODS FOR FIELD STUDY 23

3.1 Material 23

3.1.1 Field work and study areas 23

3.1.2 Plantation inventory 27

3.1.3 Interviews and field observations 28

3.1.4 Climatic conditions and soil types 28

3.2 Methods 31

3.2.1 Estimation of wood volume and biomass 31

3.2.2 Mann-Whitney's U-test 33

4 RESULTS 34

4.1 Plantation performance 34

4.1.1 Height and crown structure 34

4.1.2 Wood volume and biomass 37

4.2 Farming systems 44

4.2.1 General characteristics 44

4.2.2 Agroforestry practices and land use history 45

5 DISCUSSION 46

5.1 Variation in wood production potential between clones and study areas 46

5.2 Agroforestry practices in northeastern Thailand 49

5.3 Wood production potential in the Northeast and East compared to the South 50

5.4 Critical assessment of the study 54

5.4.1 Aims achieved 54

5.4.2 Limitations of the study 55

6 CONCLUSIONS AND RECOMMENDATIONS 57

REFERENCES 59

LIST OF ABBREVIATIONS

BPM 2 4 Bank Pertanian Malaysia's rubber clone number 24

BB19 10-year old RRIM 600 stand in Buriram, 14°38'50 N, 103°12'72 E

BR10 16-year old RRIM 600 stand in Buriram, 14°38'56 N, 103°12'79 E

BR16 16-year old RRIM 600 stand in Buriram, 14°38'56 N, 103°12'79 E

BR03 3-year old RRIM 600 stand in Buriram, 14°38'65 N, 103°13'47 E

CB16 16-year old BPM 24 stand in Chachoengsao, 13°5' N, 101°5' E

CB08 8-year old BPM 24 stand in Chachoengsao, 13°5' N, 101°5' E

CR16 16-year old RRIM 600 stand in Chachoengsao, 13°5' N, 101°5' E

CR06 6-year old RRIM 600 stand in Chachoengsao, 13°5' N, 101°5' E

CR03 3-year old RRIM 600 stand in Chachoengsao, 13°59'41 N, 101°43'81 E

CRRC Chachoengsao Rubber Research Center (of the Rubber Research Institute of Thailand)DBH Tree diameter at breast height (1.3 m)

DOA Department of Agriculture of Thailand

FAO Food and Agriculture Organization of the United Nations

LDD Land Development Department of Thailand

NB16 16-year old BPM 24 stand in Nong Khai, 18°37'11 N, 103°35'59 E

NB07 7-year old BPM 24 stand in Nong Khai, 18°36'09 N, 103°35'68 E

NR16 16-year old RRIM 600 stand in Nong Khai, 18°37'36 N, 103°35'60 E

NR08 8-year old RRIM 600 stand in Nong Khai, 18°36'09 N, 103°35'68 E

NR03 3-year old BPM 24 stand in Nong Khai, 18°37'07 N, 103°35'15 E

ORRA The Office of the Rubber Replanting Aid Fund

RFD Royal Forest Department of Thailand

RIS Rubber Information System developed by the Department of Agriculture of ThailandRRIM 600 Rubber Research Institute Malaysia's rubber clone number 600

RRIT Rubber Research Institute of Thailand

1 INTRODUCTION

1.1 Background of the study

The rubber tree, Hevea brasiliensis (Muell.) Arg., is a major crop for smallholders in Thailand

and an important commercial crop everywhere in Southeast Asia It is grown for latexproduction, while rubber wood is considered as a secondary product Therefore rubber isregarded as an agricultural crop However, recent improvements in wood technology have led

to rubber tree becoming increasingly important as a source of wood products (Evans andTurnbull 2004) Rubber wood has enjoyed an environmentally friendly reputation as a rawmaterial, because it is a by-product of latex production, and when grown in renewableplantations, it can substitute timber from natural forests

The natural range of Hevea, of the family Euphorbiaceae, covers the Amazon river basin and parts of the nearby uplands Within the genus, Hevea brasiliensis (also known as para rubber)

is one of the most widely distributed species It grows in an area South of the Amazon river,extending towards the west in Peru and the south to Bolivia and Brazil (Wycherley 1992)

The rubber tree has always been known for its latex, which was used by the ancientcivilizations of Central and South America The commercial and large-scale exploitation ofthe tree did not begin until in the last quarter of the 19th century With the arrival of cars,discovery of the pneumatic tyre and following increase in rubber prices, the produced amount

of plantation-originated rubber was soon larger than that of wild rubber At the same time,there were strong geo-political pressures to move the rubber production away from SouthAmerica (Jones and Allen 1992) While searching for a cash crop for its eastern colonies, theBritish identified rubber as a potential crop for planting in Southeast Asia (Hong 1999)

Rubber was first introduced in Asia in 1876, when seeds were first shipped from theAmazonas to the United Kingdom and further to Ceylon and planted there In the followingyear, rubber trees were planted in Singapore and Malaya (Hong 1999) Although rubber wasfirst an estate crop, local individual farmers soon adopted the crop and so they were drawninto the world commercial economy (Courtenay 1979) Nowadays rubber is cultivatedworldwide in most parts of the lowland humid tropics, but the production is heavily

concentrated into Asia, where over 90 % of the world’s natural rubber is being produced.

Rubber seeds were first brought to Thailand from Malaya in 1900 and planted in Trangprovince in southern Thailand (RFD 2000) Estate agriculture was for political reasonsdiscouraged in Thailand, unlike in Malaya, in the beginning of the 20th century Rubbergrowing became important as a smallholder crop, when local farmers responded to theimproved rubber prices in mid-1920s and planted rubber in southern Thailand (Courtenay1979) Favourable climatic conditions, free land areas and easy railway access enabled theadoption of rubber growing in the South (Pendleton 1962) Small areas were plantedelsewhere, mainly in Chantaburi province, where rubber seeds and seedlings from Malayawere first taken in 1908 Later the cultivation extended to some other eastern provinces (RFD2000)

Peninsular Malaysia has been the world's most important rubber cultivation area, and the

present wealth of this area was largely based on production of natural rubber (Collins et al.

1991) In the year 2005, Indonesia, Thailand and Malaysia produced 33 %, 23 % and 13 % ofthe world’s natural rubber, respectively (FAO 2006) Lately, the rubber plantation area hasbeen decreasing in Malaysia, but in Thailand the trend has been reverse and plantations havestarted to spread to new areas in the East and Northeast of Thailand 1 This area has been

referred to as non-traditional for rubber cultivation (Chantuma et al 2005) Today Thailand

has the second largest area of rubber plantations in the world following Indonesia, is theworld's largest producer of natural rubber (FAO 2006) and also the world leader in rubberwood production and export (LDD 2005a)

The rubber plantation area in Thailand is much larger than the area of forest plantations in thecountry According to FAO (2005), the total area of rubber plantations in Thailand was

1 680 000 ha in 2005 According to the statistics of the Rubber Research Institute of Thailand(RRIT 1996 cited in RFD 2000), the rubber plantation area was larger already in the year

2000, when it was recorded as 1 959 000 ha In comparison, the area of forest plantations in

Thailand in the year 2000 was 355 000 hectares The area of natural forest in the same yearwas 16 486 500 hectares (RFD 2001)

1 In this study, areas of Thailand are referred to as South, Central, East, Northeast and North A map of Thailand and names of provinces in these areas is in Appendix 1.

Rubber has been referred to as a woody agricultural crop (FAO 2005) together with the oilpalm and coconut In Thailand, the rubber plantation area is larger than the plantation area ofthese two crops In the year 2005, the plantation areas of rubber, oil palm and coconut were

1 680 000 ha, 315 000 ha and 343 000 ha, respectively (FAO 2006) The plantation areas ofboth oil palm and rubber have been growing Oil palm is cultivated in the South of Thailand,which is also the traditional area for rubber cultivation Competition for land area from othercrop species has been identified as one factor driving the establishment of rubber in newareas

In Thailand the smallholder rubber is intensively supported by the Royal Thai Government, informs of technology and production inputs such as seedlings, land preparation and fertilizer(Joshi 2005) In recent years the Thai Government has been promoting rubber planting also innew areas In the year 2004, the goal was to extend the planted area, with a target of onemillion rai (160 000 hectares) extension within two years from 2004 to 2006 (RRIT 2005).The establishment of new rubber plantations has been promoted especially in the North andNortheast of Thailand The estimated extension of rubber cultivation area is 400 000 hectares

by the year 2010 (RRIT 2005)

In contrast to Malaysia, where rubber is mainly grown on large estates, in Thailand 90 % ofrubber is grown in family-owned smallholdings 2 less than eight hectares in size, the averagearea of a plantation being only two hectares (Pratummintra 2005) Rubber yields per hectare

in Thailand are the highest of the three leading rubber-producing countries This is due togovernmental support for smallholder rubber cultivation, and especially to the use ofimproved planting material Of the three leading rubber producers, the yield per hectare islowest in Indonesia, where rubber has traditionally been grown in “jungle rubber“agroforestry systems In these systems, the low yields have been reported to result from a low

level of maintenance and use of non-improved planting material (Wibawa et al 2005).

Therefore, improving the productivity of rubber agroforestry has much potential especially in

2 In this study, the term smallholding is used to refer to family-owned small rubber plantations The Department

of Agriculture (DOA) of Thailand has classified smallholdings, medium-sized holdings and estates as those where rubber area is less than 8 hectares, 8-40 hectares and more than 40 hectares, respectively (Pratummintra 2005) According to Courtenay (1979), the smallholding is usually family-owned, managed by the family head and worked by family labour The plantation in turn is frequently owned by a company or a government enterprise, and usually professionally managed (Courtenay 1979) In this study, the term plantation is, however,

Indonesia In Thailand’s case, a potential for increased production could lie in theestablishment of rubber in new areas Therefore research on the performance of rubber inthese new areas is needed.

Rubber grows best in a climate similar to that in its area of origin in the Amazonas, where therainfall is heavy and there is no dry season (Rao and Vijayakumar 1992) In northeasternThailand, the annual rainfall is less than optimal for rubber and the dry season lasts forapproximately six months In this climate, smaller wood volumes per hectare have been

reported in comparison with plantations in the traditional cultivation area (Chantuma et al.

2005) So far, comparative studies on the effect of climatic conditions to wood volume perhectare and to individual volumes of trees in relation to plantation age have not been done Inorder to contribute to improving the productivity of rubber cultivation in Thailand, this kind

of information is needed

It has been presented that unfavourable environmental conditions would more drastically

affect the latex yield than the timber production of rubber (Grist et al 1998) In areas where

rubber cultivation is less favored by environmental conditions, improved farming systemssuch as agroforestry could be an option for increasing the economical profitability as well asenvironmental and social benefits of rubber cultivation

Rubber plantations are usually established using vegetatively propagated and often improvedplanting material Clones perform differently in response to stress from external factors such

as drought (Rao and Vijayakumar 1992) The performance and wood production potential ofdifferent clones in the non-traditional cultivation area (North and Northeast) in Thailand hasnot yet been studied The results from such studies would be useful in determining whichclones would be best suited for marginal planting areas

Although latex is still the main product of rubber cultivation, wood selling can increase thetotal productivity and enable reaching a maximum productivity of the rubber plantationearlier This is possible because wood selling can shorten the latex tapping period, after whichtrees can be either felled or used for further tapping depending on the current prices of latex

and wood (Arshad et al 1997; Clément-Demange 2004).

The wood production potential of rubber at a given site depends mainly on clone, planting

density and tapping practices In the case of clones, their architecture, most importantly thebranching pattern, is a critical characteristic Breeding of more suitable clones could lead tobetter rubber wood productivity and increased income in the long term, but meanwhile clonalrecommendations can already be given (Clément-Demange 2004) The RRIT has alreadygrouped rubber clones into three classes according to their latex, timber and joint productionpotential Clonal recommendations for the non-traditional area in Thailand could be veryuseful in order to determine which clones can be best adapted to a marginal cultivationenvironment.

Plantation forestry and estate crops are controversial issues due to their reported negativesocial and environmental impacts Indeed, rubber plantation establishment has had somedirect negative environmental consequences in Thailand in the past The logging ban of allforests, which was declared in Thailand in 1989, was adopted following environmental

degradation caused by logging and rubber plantation development on forest land (Collins et

al 1991) After the ban, Thailand's timber has had to be taken from forest and rubber

plantations This has been one of the main factors driving the increasing utilisation of rubberwood for industrial purposes

Rubber has been and still is an important commercial crop in Thailand and Southeast Asia InThailand’s case, income from rubber cultivation is especially important for rubbersmallholders According to RRIT (2005), there are over one million rubber smallholders inthe country The demand for natural rubber has been predicted to rise from 8.4 million tonnes

in the year 2004 to 11.9 million tonnes in the year 2010 (Joshi 2005) As the demand forrubber wood products remains high as well, it is important to ensure a sustainable andsufficient future supply of rubber products while improving the productivity of farmingsystems in order to contribute to ensuring good income for rubber smallholders in Thailand

This report studied the performance and wood production potential of two rubber clones innortheastern Thailand The study was conducted under the framework of a Common Fund forCommodities (CFC)- funded project “Improving the Productivity of Rubber Smallholdingsthrough Rubber Agroforestry Systems” This project was coordinated by the WorldAgroforestry Centre (ICRAF), and partners included the Indonesian Rubber ResearchInstitute, Prince of Songkhla University and Kasetsart University in Thailand, and theUniversity of Helsinki This study was also a joint undertaking in the long series of academic

collaboration between the universities of Kasetsart and Helsinki.

1.2 Scope and objective of the study

The present study was carried out in Thailand in order to investigate the performance andwood production potential of two rubber clones, namely RRIM 600 and BPM 24, in threeareas under different climatic conditions in northeastern and eastern Thailand The woodproduction potential was assessed through estimating the wood volume of individual trees andplantations per hectare As this study focused on the forestry-related uses of rubber, latexyields were not measured However when assessing the general profitability of rubber, thelatex yield component is currently the most significant factor in determining the viability ofrubber cultivation

The general objective of this study was to investigate, using literature review and field datacollection, the wood production potential of two rubber clones in northeastern and easternThailand in relation to environmental conditions and to study the characteristics of rubberfarming systems in northeastern Thailand

The specific objectives of this study were:

1) To investigate the wood production potential (wood volume and clear bole volume asrelated to plantation age) of rubber clones in relation to geographical area and climaticconditions

2) To compare the wood production potential of rubber clones in different geographical areas.3) To preliminarily investigate the effects of site characteristics, especially the previous land-use history, on the performance of rubber

4) To preliminarily identify and study components of agroforestry systems used at rubberplantations

2 LITERATURE REVIEW

2.1 Botany and distribution of Hevea brasiliensis

Hevea brasiliensis is a tropical, deciduous tree, which grows 25-30 meters tall in its natural

distribution area Most of the planted trees are smaller, because they have been bred for theproduction of latex without taking much into account their wood production potential (Hong1999) The bole of the rubber tree is usually straight but quickly tapered, and heavy branching

is common The branching pattern is very variable, and the leading stem can be dominant orsoon divided into several heavy branches The tree is easily damaged by strong winds

(Lemmens et al 1995) Clonal variation in wind-resistance has been observed, depending on types of branching (Cilas et al 2004) Rubber tree matures at the age of seven to ten years,

after which latex tapping can be started When aiming at economic latex production, the lifecycle of a rubber plantation is 30-35 years, after which replanting is necessary

The current world-wide distribution of rubber plantations is presented in Figure 1 Apart fromIndonesia, Thailand and Malaysia, also India, Vietnam, China, Nigeria, Liberia, Sri Lankaand Brazil, in descending order, have large areas (over 100 000 ha) of rubber plantations(FAO 2006) In Table 1, the development in planted area and production of natural rubber inthe three leading rubber-producing countries is compared

Rubber plantation area, million ha and percentage of world total in 2004

1400; 17 %

1740; 21 %

649; 8 % 154; 2 % 13; 0 %

1676; 20 %

2675; 32 %

Indonesia Thailand Malaysia Rest of Asia Africa South America Others

Figure 1 Rubber plantation area in the world in thousand hectares, and percentage of the totalplanted area in the world in the year 2004 FAO 2006

Table 1 Rubber plantation area in 1000 hectares and the average production of natural

rubber in kilograms per hectare per year (kg-1ha -1a-1) between years 1985-2005 in Indonesia,Malaysia and Thailand (FAO 2006)

Area Prod.

1990 Area Prod.

1995 Area Prod.

2000 Area Prod.

2005 Area Prod.

2.1.1 Distribution of Hevea brasiliensis in Thailand

In 1996, the fourth survey on Thailand’s rubber plantation area was carried out by the RRITusing Landsat satellite images According to this survey, the total plantation area was 1 959

285 ha, of which 45 420 ha (2.3 %) were in the Northeast and North of Thailand The easternprovinces including Chachoengsao accounted for 12.3 % of the plantation area (RRIT 1996cited in RFD 2000) According to Chantuma (2005), presently 5 % of the plantations are innortheastern and 10 % in eastern Thailand The Thai Government has targeted enlarging thearea of rubber plantation by 48 000 hectares in the North and 112 000 ha in the Northeast of

Thailand (Chantuma et al 2005).

In terms of latex production, suitable rubber growing areas can be found also in the traditional cultivation area in northeastern and northern Thailand The Department ofAgriculture of Thailand has created a rubber information system (RIS), where climatic andsoil profile data are stored in a regional geographic information system (GIS) database Amodel for maximum latex production potential that was validated by using existing latex yielddata from the eastern provinces was used to evaluate and map the production potential in theNorth and Northeast of Thailand.

non-Three rubber yield classes were determined In class L1 the production potential is over 2500

kg per hectare per year (kg-1ha-1a-1) According to the RIS, this class was not found in theNorth and Northeast, only in the South of Thailand The second best class, L2, where theproduction potential was estimated at 1500-2500 kg-1ha-1a-1 was found in an area of about 320

000 hectares in the Northeast and 160 000 hectares in the North of Thailand The third class,L3, where production is lower than 1500 kg-1ha-1a-1 and trees can not yet be exploited afterseven years from plantation establishment, was not regarded as a suitable area (Pratummintra2005)

Figure 2 The area of rubber plantations in Thailand in the year 2000 according to RFD 2000,and the share of the total area in different regions in 2005 (Chantuma 2005).

2.2 Climatic requirements of the rubber tree

The rubber tree is native to the evergreen tropical rainforests usually occurring within the 5°latitude of the equator The climate of this region is characterized by heavy rainfall and nodistinct dry season According to Rao and Vijayakumar (1992), the optimal climatic

conditions for the genus Hevea are:

A rainfall of 2000 mm or more, evenly distributed throughout the year with no severedry season and with 125-150 annual rainy days,

A maximum temperature of about 29-34 °C, minimum of about 20 °C and a monthlymean of 25-28 °C,

High atmospheric humidity of about 80 % with moderate wind, and

Bright sunshine for about 2000 hours in a year, at the rate of six hours a day in allmonths

In traditional rubber growing areas, the total rainfall ranges between 2000-4000 mm,distributed over 140-220 days, without more than one to four dry months (Rao andVijayakumar 1992) Rubber can successfully be cultivated under these kinds of humid

lowland tropical conditions, roughly between 15°N and 10°S (Lemmens et al 1995).

Cultivation of the tree has however expanded away from the equator to latitudes as far North

as 29°N in India, Myanmar and China, and down to 23°S in Brazil In Thailand, rubber hastraditionally been cultivated on the Malay Peninsula from 6-12°N and in areas with anaverage rainfall of around 2000 mm per year (Watson 1989) Cultivation in the East andNortheast of Thailand (up to 18°N) has mainly started during the last two decades

It is justified to make a distinction between the conditions that permit the survival of rubberand those that assure best growth and yield (Compagnon 1987) and a cultivation which iseconomically viable A general lower limit of annual rainfall for the economically viablecultivation of rubber can not be easily given, since environmental factors other than climatealso affect the survival of the tree (Compagnon 1987) A well-distributed annual rainfall of

1500 mm has sometimes been considered as a lower limit for commercial production

(Lemmens et al 1995) However, the requirement depends on the distribution of rain

throughout the year, length of dry season and soil water retention capacity In favorable soils,rubber could tolerate a dry season of four to five months, during which less than 100 mm ofrain is received and within this period, two to three months with rainfall less than 50 mm(Compagnon 1987)

Plants encountering high temperature in the absence of rainfall are driven to higher rate oftranspiration which in turn leads to moisture stress Effects of rainfall and temperature on thephotosynthetic rate (Sangsing 2004) and further the growth performance (Jiang 1988) and the

latex yield (Jiang 1988; Rao et al 1990; Rao et al 1996; Raj et al 2005) of rubber trees have

been derived In general, moisture stress has resulted in decreasing latex yields as well as

decreasing total production of dry matter According to Grist et al (1998), the growth and

latex yield of a tree are affected in different ways by soil moisture Moisture stress has moredramatic effects on the latex yield than on tree growth, as turgor pressure in latex vesselsinside the trunk of the tree is required to facilitate the latex flow

Clonal differences in photosynthetic rates (Nataraja and Jacob 1998; Sangsing 2004) and

tolerance to moisture stress (Rao et al 1990; Chandrashekar et al 1998; Raj et al 2005) have

been observed Priyadarshan et al (2005) studied the yield potential of several rubber clones

in marginal environments suffering from severe winds, low temperatures and highevaporation in northeastern India Clone RRIM 600 (Rubber Research Institute Malaysia,clone number 600) appeared to be able to adapt well to various conditions, and produced

moderate yield in all marginal environments mentioned (Priyadarshan et al 2005).

Chantuma et al (2005) studied the wood production potential of clone RRIM 600 in the

non-traditional rubber cultivation area of northeastern Thailand In Nong Khai province, thesurvival percentage in a 15-year old plantation was 90 and the wood volume was 138 m3ha-1

In Chachoengsao province, at a plantation aged 19, the survival was 79 % and wood volume

188 m3ha-1 Authors compared these results with figures from the traditional cultivation area

in Phuket and Surat Thani in southern Thailand, where plantations were 25 years old Survivalwas 78 % and 83 % and wood volume 256 and 300 m3ha-1, respectively (Chantuma et al.

2005) Wood volume was assessed based on tree girth According to this study it seemed thatrubber wood productivity in the non-traditional area could be almost comparable to that in theSouth of Thailand However, it would be interesting to include several plantations inconsideration, also in the drought area of the Northeast, as well as to compare theperformance of different clones It seems that the growth performance could be restricted inthe drought area, where trees encounter water stress especially during the hot and dry season

The optimum day temperature for rubber is 26-28 °C Night-time temperature drops to 10 °C

in Laos and Cambodia have not caused problems, but preferably the minimum temperatureshould not drop below 14-15 °C (Compagnon 1987) During periods of low temperature,slowing down of growth has been observed in China and in Northeast India In China, whererubber-growing areas lie between 18° and 24°N, the growth rate has been reported to slowdown drastically during the winter (Rao and Vijayakumar 1992) Cold damage, including thedeath of shoots and a decreasing latex flow, has occurred when trees encounter hot and coldconditions within one day and night temperature fall quickly to less than 5 °C and daytemperature rising to 15-20 °C (Watson 1989) Apart from latex flow and growth rate, coldconditions have been reported to affect the survival during wintering and outbreak orsuppression of diseases (Jiang 1988) Different clones appear to vary greatly in their coldresistance (Watson 1989)

Rubber trees shed their leaves annually, but the timing and intensity of leaf-shedding depends

on climatic condition and varies between clones (Lemmens et al 1995) In eastern and

northeastern Thailand rubber trees shed their leaves in December, and start to grow newleaves in January and February Trees in the South drop their leaves approximately twomonths later and start to produce new leaves in March and April (RFD 2000)

2.3 Soil requirements of the rubber tree

Rubber can grow on many soils, the best options being well drained (Lemmens et al 1995)

clayey and deep clay soils (Growing multipurpose… 1994), but it can withstand physicalconditions ranging from stiff clay with poor drainage to well drained sandy loam Soil waterretention capacity, depth and soil moisture are important factors determining the suitability of

a growing site Ground covering plants can help improving the soil physical properties

(Krishnakumar and Potty 1992) An optimal soil pH value for rubber is at 5-6 (Lemmens et

al 1995) The performance of the tree can be restricted where there is rocky surface, heavy

drainage or soil pH values above 6.5 or below 4 (Krishnakumar and Potty 1992)

In Thailand, rubber trees can be grown in many areas that are unsuitable for other commonlycultivated cash crops Rubber requires a modest level of soil nutrients when compared tocoffee, tea, coconut and oil palm Some fertilizer is however advantageous and can be needed

to replace nutrients lost (RFD 2000)

The Land Development Department of Thailand (LDD) has carried out research in easternThailand in order to identify soil types suitable for rubber planting in the East According tothe study, soil properties essential for rubber are soil depth of at least one meter and moderatefertility Shallow soil, heavy stone layer at or above 50 centimeters from soil surface and lowlevel of fertility were regarded as unsuitable conditions for rubber cultivation Suitable soilseries were found to cover 9 200 hectares in eastern Thailand (LDD 2005a)

2.4 Rubber cultivation in Southeast Asia

2.4.1 General characteristics

Because rubber has traditionally been classified as an agricultural crop, rubber plantations areconsidered as agricultural land and not as forest plantation However, the rubber tree is themost widely planted tree species in Southeast Asia (FAO 2005) The characteristics of rubberfarming systems vary within Southeast Asia In the beginning of the 20th century, estateplanting was encouraged in Malaya, while in Thailand and the Netherlands Indies rubberbecame an important crop for smallholders (Courtenay 1979) Still at present in PeninsularMalaysia rubber is grown on smallholdings and estate plantations, the latter beingcharacteristic to Malaysia while the smallholder rubber is dominant in Thailand Theplantations are for the most part 'monoculture', i.e consisting of a single crop In Indonesiathe practice is different- rubber is mainly cultivated in extensive and often complex3agroforestry systems, referred to as jungle rubber In these systems rubber is the main cropcultivated, but it is grown together with timber species, fruit trees, rattan or medicinal plants(Wibawa 2005)

Incentives for improving the productivity of rubber cultivation can sometimes be limited InIndonesia, where the productivity of natural rubber per hectare is low, yield could beimproved by increasing the number of trees per hectare, and by planting better yielding rubbervarieties However, expected land scarcity caused by outside land claims provides incentivesfor securing future land rights by forest clearing and rubber planting, and not so much forintensification of existing farming systems (Angelsen 1995)

Neither in Thailand is the land tenure secure in all cases Private land ownership is recognizedstep by step, from registration of land use to full ownership The registration of landoccupancy is at present the only form of land security for millions of people, and although

3 The complex rubber agroforestry system includes a variety of plants, trees as well as treelets (banana, cocoa, coffee), lianas and herbs which are all associated The structure and functioning of these systems has been reported to be close to that of a natural forest A simple agroforestry system in turn consists of a smaller number

of plants, usually no more than five tree species and annual species (paddy or upland rice, maize, vegetables, herbs) or treelets (Gouyon 2003).

these people are commonly regarded as owners of the land, a formal ownership is still missing(Luukkanen 2001).

Government agencies supporting rubber planting in Thailand are the Rubber ResearchInstitute of Thailand (RRIT) and the Office of the Rubber Replanting Aid Fund (ORRAF).The RRIT works under the Department of Agriculture (Ministry of Agriculture andCooperatives), and its responsibilities include rubber development plans, research, technologytransfer and control of natural rubber production, trade, exports and imports (RRIT 2005).ORRAF is also attached to the Ministry of Agriculture and Cooperatives and it is a non-profitenterprise carrying out governmental policies ORRAF's objective is to work with rubberfarmers on rubber production, processing and marketing through providing improved varieties

of rubber seedlings, aiding in the establishment of both new plantations and replantings andproviding technology and guidance (Chaninthornsongkhla 2005)

In Thailand, rubber seedlings are usually produced by bud grafting on rootstock in nurseries.Rubber seeds from high-yielding parents are first grown from four to eight months, untilstems reach a desired diameter at about 10 cm above ground, after which a grafting from adesired clone is attached Budwood clones are grown in specific bud-root gardens The RRIThas developed a certifying system in order to take care of the quality of planting materialproduced at nurseries

2.4.2 Agroforestry practices

Diversification of income through introducing food crops, timber trees or livestock in rubberfarming systems is a common practice in Southeast Asia In Thailand, simple agroforestrypractices such as intercropping and integration of fruit trees have been adopted atsmallholdings in order to diversify sources of income These practices have however not yetbeen formally recommended nor well documented (Joshi 2005) The RRIT has carried outresearch on various intercropping systems, and according to these studies, intercrops thatcould successfully be grown with rubber in Thailand are banana, papaya, pineapple and

upland rice (RRIT 2005) Cherdchom et al (2002) reported four main integrated rubber

farming systems in the South of Thailand emerging during the financial crisis in the late

1990's The major systems included 1) Rubber intercrop farming, 2) Rubber-rice farming, 3)Rubber-fruit tree farming, and 4) Rubber-livestock farming.

According to Joshi (2005), diversification of income sources through rubber agroforestrysystems could become more crucial in the non-traditional cultivation area, where rainfall islow and other conditions less favorable for rubber, than in the South of Thailand The LDDhas already recommended planting of food crops with rubber in eastern Thailand Fruit treessuch as durian, mangosteen and rambutan were also recommended in order to diversifysources of income (LDD 2005a)

When rubber trees are planted in widely used "standard" plantation pattern of 3 m x 7 m or

8 m, intercropping is generally possible only during the first years of rotation, before rubber

canopies close and do not allow the growth of light-demanding crops A study by Rodrigo et

al (2005) in Malaysia investigated the possibility to improve the productivity of rubber

agroforestry by altering planting patterns Considering overall performance of long-termintercropping, a double rubber row system with intercrops was identified as the best option

(Rodrigo et al 2005) Wibawa et al (2005) have also received encouraging results in

long-term intercropping using a rubber spacing of 6 m x 2 m x 14 m

Another study by Rodrigo et al (2004) demonstrates that apart from its overall economic

benefits, agroforestry can be beneficial to the growth of rubber trees Intensive intercropping

of young rubber with banana may result in an increase in growth and yield of rubber trees,and to a reduction in the length of the unproductive immature phase of rubber Intercroppinghad a positive effect on the growth of rubber throughout the six years of the study, with theresult that trees grown with intercrop were ready for tapping four months earlier than those

growing on their own (Rodrigo et al 2004).

In Malaysia rubber has generally been planted as monocrop, but to increase productivity,some farmers cultivate short term crops such as vegetables, corn, pineapple, groundnut andbanana between rubber rows during the first two and a half to three years of rotation Animproved intercropping system has been developed in order to sustain the productivity ofintercropping over a longer period of time In this system rubber is planted in one, double ortriple rows and the interhedges are planted with forest or fruit trees

To assess the financial viability of rubber plantation with integrated forest trees, aneconomical analysis was carried out comparing rubber agroforestry systems with integratedtimber trees to traditional monoculture plantations in terms of income in both smallholdingsand large estates For the smallholdings, projected income from integrated timber species

seemed attractive Hedge planting with rubber and teak (Tectona grandis) or sentang (Azadirachta excelsa) was identified an option for consideration Sentang or teak could

provide a bonus income at harvest while latex collection provides continuous supply of cash

before harvesting (Arshad et al 1997).

In Indonesia, over 70 % of the total rubber area is jungle rubber agroforestry A jungle rubbercultivation system is usually established after slash-and-burn of secondary forest or old rubberarea Complex rubber agroforests have been observed to preserve many functions of a naturalforest and therefore they could provide many environmental services: maintainingbiodiversity, retaining soil water captivation capacity and sequestering carbon from the

atmosphere (Joshi et al 2002) However, complex agroforests are competing for land with

more intensive land use options When incentives for retaining the traditional agroforestrysystems are not available, farmers often choose land use forms that provide fewerenvironmental services Efficient compensation such as a reward practice could help preserveand promote complex agroforestry systems and the environmental services they provide

(Joshi et al 2002).

The production of latex in jungle rubber agfororestry is very low- only about a third of that inintensive monocultures Improved rubber agroforestry systems have been succesfullydeveloped, studied and promoted in Indonesia in order to improve the productivity of rubbercultivation According to Xavier (2004), promising results on integrating plantation treespecies grown for timber in rubber agroforests have been observed in Indonesia

2.4.3 Environmental considerations

Most of the original forest cover in Southeast Asia has been cleared for agriculture, includingrubber cultivation In recent times the expansion of rubber growing into primary forest hasbeen most common in Indonesia, as a result of population growth, insecurity of land rights,

land scarcity and rising rubber prices (Angelsen 1995) Obviously, intensive rubbercultivation can not be comparable to natural forest in terms of biodiversity, and rubbercultivation should therefore not extend to areas covered with natural forest In the case ofjungle rubber, as pointed out before, the complex agroforest could, however, perform manyecological functions, and when comparing rubber cultivation with other land use alternatives,the change from traditional shifting rice cultivation to smallholder rubber has been reported tohave various positive ecological effects in Indonesia (Angelsen 1995).

According to Balsiger et al (2000), the role of rubber tree as a carbon sink has often been

under-estimated Apparently due to its high leaf area index and the extra energy the treerequires to produce latex, it acts as an effective carbon sink

Intensive rubber growing areas can become vulnerable to soil nutrient loss and erosion thatresult from ground preparation and clear-cutting Growing rubber together with agriculturalcrops could be the best way to decrease these environmental impacts On steep slopes,terracing has been recommended to prevent erosion (Royal Forest Department 2000) TheLand Development Department (LDD 2005a) has recommended planting of vetiver grass inhilly areas for erosion control While latex harvesting is practiced, fertilizer may be required

to replace nutrients lost (RFD 2000)

2.5 Uses of Hevea brasiliensis

The most important product of Hevea brasiliensis is the latex produced in the bark of the tree

and made into natural rubber Rubber wood is generally considered as a by-product, and itscommercial value was almost non-existent until about 25 years ago The wood was mainlyused as fuelwood and for charcoal making The large supply and easy availability of rubberwood were not attractive enough to the wood processing industries in the past Lately, thedecreasing area and availability of natural forests for logging, increasing labour costs andother factors have favoured the emergence of rubber wood as a raw material for mechanicalwood industry, especially for the manufacture of furniture and wood-based panel (Hong1999)

Rubber wood can be a substitute for many species, including meranti (Shorea spp.), teak, oak

and pine (Balsiger et al 2000) The timber is moderately durable and light creamy in colour,

which makes it attractive and popular among consumers Rubber wood is also useful inmechanical and chemical pulping processes to produce paper with fair quality However,some problems remain as special attention needs to be given to remove latex residues fromthe pulp (Yussof 1999)

Thailand has a large rubber wood industry, and its products include furniture, particle board,parquet board and construction boles (RFD 2000) The annual export value of Thailand'sfurniture industry is more than 300 million US dollars (FAO 2005) Yet the rubber woodindustry in Thailand still faces some constraints and challenges within resource management

as well as industries, product and market development Although the resource base is large,the quality of raw material is restricted According to Anonymous (2000), the main problemsconcerning resource management and utilisation were inefficiency of rubber wood rawmaterial management due to insufficient promotion and development of high-yieldingcombined latex and timber clones, unfavourable infrastructure, and difficulties in loggingespecially during rainy season, and restricting regulations for logging

3 MATERIAL AND METHODS FOR FIELD STUDY

3.1 Material

3.1.1 Field work and study areas

Field work was carried out in northeastern and eastern Thailand between August andNovember of 2005 The field work was conducted together and in collaboration with projectpartners from CFC- funded project, “Improving the Productivity of Rubber Smallholdingsthrough Rubber Agroforestry Systems” Project partners involved in field work were studentsand staff from Kasetsart University, Bangkok

Thailand is situated in the tropical zone between latitudes 6-20° North (N) and longitudes 105° East (E) The climate is characterized by moderate rainfall and a hot dry summer The

98-country has a monsoon climate: Northeast monsoon from December to February (the dryseason), hot weather and variable winds of March, April and May, Southwest monsoon fromMay to October (the rainy season), and retreating monsoon period of October and November(Pendleton 1962) Maximum day temperatures in Thailand change relatively little during theyear In upper Thailand, the maximum temperature sometimes exceeds 40°C (Koteswaram1974).

The amount and timing of rain is much more important to nature and agriculture in Thailandthan is temperature Rainfall can be unpredictable, and the amount of rain can vary much fromplace to place and from year to year Most of Thailand receives the majority of rain during theSouthwest monsoon Generally, the quantity of rainfall decreases with increasing distancefrom the sea, but the amount of rainfall and the length of rainy season vary much depending

on area and altitude The greatest quantities of rain (4200 mm annually on average) arereceived on the West coast of the peninsula The peninsula in general is characterized byample and relatively evenly distributed rainfall On the other hand, the extreme Southeastcoast is very similar to the West coast of the peninsula (Pendleton 1962)

The driest regions of Thailand are found in the Northeast, on the Khorat platform, whichsuffers from lack of water in the dry season In the lower part of Khorat the average annualrainfall is only 1050 mm On the other hand, in the far Northeast, along the Mekong River,over 2030 mm is received annually Variation in the average amount of rainfall is thereforenotable in the Northeast During the Northeast monsoon, winds can be relatively cold inKhorat and thus also daily temperature variations are greater than those in the central valleyand in more maritime areas (Pendleton 1962).

Figure 3 Map of Thailand and study areas (district, province) (Map: Wikipedia, modifiedwww-document)

The first aim of this study was to investigate the performance of rubber in relation to climaticconditions in the non-traditional cultivation area Therefore it was necessary to conduct thestudy in areas with different amounts of rainfall The Northeast of Thailand has beenpromoted for rubber plantation establishment, and, on the other hand, considerable variation

in the amount of rainfall exists within the northeastern region enabling the identification ofsuitable study sites for comparison

Ban Kruen district in the province of Buriram in the lower part of the Khorat plain(approximately 14°N, 103°E), Pak Khat district in Nong Khai province by the Mekong River

in upper Khorat (approx 18°N, 103°E) and Lad Krating village in Chachoengsao province inthe central valley at the border of Khorat region (approx 13°N, 101°E) were selected as studysites The locations of study areas are marked on the map (Figure 3) Chachoengsao formallybelongs to the East of Thailand, but since it is situated at the border of the Khorat plateau, inthis study the Chachoengsao area also is referred to as Northeast, in order to make adistinction between the new cultivation area and the area of the South and East

The other aim of this study was to compare the performance of two or more clones Tooverview the situation of rubber cultivation in northeastern Thailand, the Office of RubberReplanting Fund (ORRAF) was contacted in Bangkok Local offices were also contacted inBuriram and Nong Khai With the help of ORRAF staff and local farmers, suitable studyareas were searched It soon became evident that finding several clones for this study wasmore difficult than previously expected The main clone planted at smallholdings in theNortheast of Thailand appeared to be RRIM 600 (Rubber Research Institute Malaysia, clonenumber 600), BPM 24 (Bank Pertanian Malaysia, clone number 24) occupying less land Forthis reason, the study was limited to clones RRIM 600 and BPM 24 Both of these cloneshave been classified as high-yielding latex clones by the RRIT (RFD 2000) According toSirianayu (2005), high-yielding timber clones or combined timber-latex clones have not yetbeen planted in northeastern Thailand

All plantations in Buriram and Nong Khai were smallholdings, and the size of individualplantations ranged from 0.6 to 3.2 ha The plantations in Chachoengsao, with the exception ofthe youngest plantation which was a smallholding, belonged to Chachoengsao RubberResearch Centre (CRRC), of the Rubber Research Institute of Thailand

The aim was to find plantations of both clones and of corresponding ages from all three studyareas, in order to make a comparison between clones and areas In Buriram, young BPM 24plantations could not be found Neither in Nong Khai nor in Chachoengsao could newlyestablished BPM 24 plantations be found The location of and other information on eachstudy site are presented in Table 2.

Table 2 Plantations included in this study *) Approximate coordinates

ha

No of trees/ha

101°5' E *)

RRIM600 6 0,6 440 2.5 m x 7 m, in blocks of

eight CR03 Chachoengsao 13°59'41 N

5 % of the total number of trees being measured As the number of trees grows larger, small

sample size leads to more reliable results than the same sample size at a smaller plantation(Kangas and Päivinen 2000) For this reason at plantations larger than 20 rai (3.2 ha) every

50th tree was chosen as a sample tree, resulting in 2 % of the total number of trees beingmeasured Altogether 302 trees were measured

The following was measured from sample trees: diameter at breast height (DBH), girth atbreast height, tree height, height of crown, length of branch-free stem (also referred to as clearbole) and the width of crown at two opposite points Diameter and girth were measured at theprecision of 0.1 cm using tallmeter Height attributes were measured at the precision of 0.5 musing Suunto hypsometer The length of branch-free stem was defined as the length of stemfrom the ground before any branches started Crown was defined to start where the foliagestarted Width of crown was measured from the ground using tape measure, at the precision of

10 cm The borders of each smallholding plantation were marked as way points using GarmineTrex Legend GPS navigator The area of each plantation was defined using GPS Theaccuracy of Garmin eTrex Legend is about 15 m

3.1.3 Interviews and field observations

The owners of each plantation were interviewed in Thai with the help of a translator Thebasic information of planted clone, area of plantation and age of the trees was obtained in thisway In addition, information on plantation history and agroforestry practises was collected.The list of questions asked is available in Appendix 2 The health of the trees, in case therewas any visible damage or illness, and signs of previous forest cover in plantation area werevisually observed in the field

3.1.4 Climatic conditions and soil types

Meteorological data were obtained from the Thai Meteorological Department (TMD).Rainfall data from the last 20 years included the amount of rainfall per month, amount ofrainy days per month and daily maximum Temperature data included monthly mean,minimum and maximum temperatures Weather stations where the data were collected wereNan Rong weather station, number 436401 in Buriram province, Chachoengsao weatherstation, number 423001 in Chachoengsao province and Nong Khai weather station, number

352201 in Nong Khai province For temperatures, data from Chachoengsao station were notavailable Data from Chon Buri weather station, number 459201 in Chon Buri province wereused instead Statistical data on the climatic conditions in study areas are presented in Table 3,Figures 4 and 5 and in Appendix 3.

Table 3 Climatic conditions in study areas, mean values in parentheses Data from the last

Dry period **)

Lad Krating,

Chachoengsao

790-1400 (1180) 51-118 (88) 28.6, 39.9, 13 6 months Ban Kruen, Buriram 840-1500 (1150) 85-142 (113) 27, 41.8, 8.5 6 months Pak Khat, Nong Khai 930-2250 (1570) 106-156 (128) 26.4, 42.8, 4.9 6 months

*) Lad Krating statistics from Chon Buri province, approximately 100 km to the South fromLad Krating

**) Number of months when rainfall is less than 100 mm, 20-year average

Figure 4 Mean annual rainfall in mm in study areas between the years 1984-2004

(TMD 2005)

1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 2004 600

Figure 5 Number of annual rainy days in study areas (TMD 2005).

Information on soil types was collected using soil maps, literature and the soil informationCD-ROM of Land Development Department of Thailand (LDD) Some of these documentswere translated from Thai into English In Buriram and Nong Khai, local “soil doctor”volunteers trained by LDD were interviewed for information on soil types, soil condition andproblems related to soil The LDD has trained soil doctors since 1995, and presently there are

55 000 soil doctors in villages in Thailand (LDD 2005b) They are in charge of coordinatingland development between the LDD and farmers in villages, transferring new technology totheir neighbours and participating in activities of the LDD (LDD 2005b)

Table 4 Soil conditions in study areas (Soilview 2.0, Soil map of Buriram province, Soil map

of Chachoengsao province, Characterization of established 2003, all published by the LDD;Pintha 2005)

vegetation

Land use Problems

Deep soil Dry

Dipterocarp forest

Sandy soil, water deficiency, low natural fertility, erosion Ban Kruen,

Buriram

Laterite

clay, laterite

Strongly acid to neutral

pH 4.5-7

Well drained

Shallow soil Dry open

Dipterocarp forest

Shallow soil, low natural fertility, erosion Pak Khat,

Very strongly acid to strongly acid

pH <4.5-5.5

Moderate Deep soil,

ground water depth <1.6 m

Dry Dipterocarp forest, Mixed deciduous forest

Low pH, nutrient loss

3.2 Methods

3.2.1 Estimation of wood volume and biomass

Bole volume (also referred to as wood volume, excludes bark, branches and leaves) and totaldry biomass were calculated for each tree using volume and biomass models developed for

Hevea brasiliensis by Räisänen (1997) in Mexico These models were created based on

material consisting of 19 trees for bole volume model and 9 trees for biomass model

Bole volume V b , dm 3 = 0.065789 * d 2.179986 * h 0.488780, excluding bark, branches and leaves

Total biomass B, kilograms = 0.066218 * d 2.131143 * hc 0.612696

In these models, d = DBH in centimetres, h = height in meters, and hc = height of crown in

meters

Another model, developed in Thailand by the RFD, was also used to estimate wood volumes.For creating RFD's volume model, 931 trees from the South and East of Thailand weremeasured

Volume V, m 3 = 0.0000461697 * x 2.0816, including bark and branches

(RFD cited in Urapeepatanapong 1989) (Equation 3)

In this model, x is girth at breast height (1.3 m) in centimeters.

This model gives the total volume of a tree, which can be further divided into sawn timber(usable volume), fuelwood and wood residues, and pole (Urapeepatanapong 1989)

Mean square error for the inventory results (sample y) was calculated using the followingformula:

Mean Square Error, Se = var y (Kangas and Päivinen 2000) (Equation 4)

Mean variance, var = ( 1- (n/N) * (S y 2 /n) )

where n = number of samples, N = total number of trees, Sy = distribution

(Kangas and Päivinen 2000) (Equation 5)

Mean square errors for results on wood volume and biomass using these volume modelsvaried from 1.8 % to 17.8 % Mean square errors are presented in Figure 6

Figure 6 Mean square error percentage for wood volume and biomass estimates for eachplantation Plantation abbreviations: BB19- Buriram, BPM 24, 19 years; BR16- Buriram, RRIM 600, 16 years; BR10- Buriram, RRIM 600, 10 years; NB16- Nong Khai, BPM24, 16 years; NR16- Nong Khai, RRIM 600, 16 years; NB07- Nong Khai, BPM 24, 7 years, NR08- Nong Khai, RRIM 600, 8 years; CB16- Chachoengsao, BPM 24, 16 years; CR16- Chachoengsao, RRIM 600, 16 years; CB08- Chachoengsao, BPM 24, years; CR06- Chachoengsao, RRIM 600, 6 years

In plantations NB16 and CB08, the distribution values within population grew large, because

a few trees included as sample trees were larger than average

In addition, the volume of clear bole was calculated This is the volume of the economicallyvaluable lower part of trunk, before branching begins For calculating this volume, trunkdiameter at the point where first branches start was needed These diameters were notmeasured in the field, and to attain values for these diameters, a taper curve was used A tapercurve for rubber trees growing in Mexico was created by Räisänen (1997) and it was found to

be relatively similar to that of silver birch (Betula pendula), created by Laasasenaho (1982).

BB19 1 ha BR16 1.3 ha BR10 0.8 ha NB16 0.6 ha NR16 2 ha NR08 0.6 ha NB07 1.3 ha CR16 0.8 ha CB16 4.8 ha CR06 0.6 ha CB08 6.4 ha 0

2 4 6 8 10

12

14

16

18

Mean square error for estimated wood

volume and biomass

Volume, m3/ha (Räisänen) Volume, m3/ha (RFD) Biomass, kg/ha (Räisänen)

Therefore the taper curve of silver birch was used in this study to get estimates for clear bolevolume of rubber trunks A polynomial form model for silver birch was developed byLaasasenaho (1982) Using this model, a taper curve could be calculated when tree height andDBH were known From this taper curve it was possible to get estimates for diameter atbranching point and further calculate the volume of the clear bole In order to get closerestimates of volume, a correction equation was used.

The number of trees per hectare for each plantation and further the wood volume per hectarewas calculated using total number of trees and the area of plantation However, someproblems were encountered Firstly, the exact area of all plantations could not be verified due

to technical problems Trees were not distributed evenly and planting densities varied: mostcommon planting pattern was 3 m x 7 m, leading to approximately 480 planted trees perhectare However, higher densities such as 3 m x 6 m and 2.5 m x 7 m were used at someplantations, and at one plantation the planting density was 3 m x 8 m In CRRC, trees wereplanted in blocks, and at one plantation the total number of trees per hectare was as low as

180 Highest number of trees per hectare was 730 To minimize the effects of estimatednumber of trees per hectare and planting pattern to the results, the wood volume per hectarewas calculated as though standard planting density of 3 m x 7 m had been used at allplantations and all trees had survived This creates over-estimation of wood volume, but at thesame time it facilitates the comparison of two clones and different areas

3.2.2 Mann-Whitney's U-test

In order to investigate variation between populations, i.e 1) the variation in wood volumebetween clones RRIM 600 and BPM 24 in each study site and 2) the variation in woodvolume within clones RRIM 600 and BPM 24 between study sites, a non-parametric Mann-Whitney's U-test was chosen, because the sample size in some plantations was small (lessthan 20 trees) and therefore a test based on the application of normal distribution risks giving

inaccurate results (Ranta et al 1989) Mann-Whitney's U-test was carried out in SPSS 14.0

for Windows software

4 RESULTS

4.1 Plantation performance

4.1.1 Height and crown structure

On average, RRIM 600 clones aged 16 were 17.4 meters tall while BPM 24 clones aged

16-19 were 15.3 meters tall Yet BPM 24 always had taller clear boles than RRIM 600 Onaverage, BPM 24 had a clear bole measuring 4.9 meters while RRIM 600 had 3.8 meters ofclear bole In plantations aged 16 or more, BPM 24 clones had a clear bole measuring 4.5meters while the clear bole of RRIM 600 trees was approximately 3.6 meters tall

Clone RRIM 600 grew taller than clone BPM 24 in all three areas There were no differences

in average height of RRIM 600 trees between Buriram (14°4' N, 103°2' E, average height 16.7m) and Chachoengsao (13°5'N, 101°5'E, average height 16.5 m) In Nong Khai (18°4'N,103°3'E) average height was taller, 19.1 meters In all three areas the height of RRIM 600trees varied to a similar extent but within different minimum and maximum figures Inascending order, height of 16 years old RRIM 600 trees varied between 10 and 19.5 meters inChachoengsao, 12.3-21.5 m in Buriram and 16-25 m in Nong Khai

BPM 24 clones aged 16 or older grew much taller in Nong Khai (average height being 18 m)compared to Buriram (13.9 m) and Chachoengsao (14.9 m) In Buriram, height of 19-year oldtrees varied between 10-19 meters, in Chachoengsao (trees aged 16) between 10-19.5 m and

in Nong Khai (trees aged 16) between 12-22 m

Figure 7 Average height and proportion of clear bole (in brown) in meters in plantations,excluding young (three-year old) plantations Plantation abbreviations: BB19- Buriram, BPM 24, 19 years;

BR16- Buriram, RRIM 600, 16 years; BR10- Buriram, RRIM 600, 10 years; NB16- Nong Khai, BPM24, 16 years;

NR16-Nong Khai, RRIM 600, 16 years; NB07- NR16-Nong Khai, BPM 24, 7 years, NR08- NR16-Nong Khai, RRIM 600, 8 years; Chachoengsao, BPM 24, 16 years; CR16- Chachoengsao, RRIM 600, 16 years; CB08- Chachoengsao, BPM 24, years;

CB16-CR06- Chachoengsao, RRIM 600, 6 years.

Average height of RRIM 600 trees as related to plantation age

Plantation age, years

Figure 8 Development of mean height in meters as related to plantation age in three RRIM

600 plantations in Nong Khai and Chachoengsao and in two plantations in Buriram

As seen in Figure 8, the growth development of clone RRIM 600 seemed to be faster in NongKhai than in Chachoengsao and Buriram Trees also seemed to reach a taller maximum height

in Nong Khai than in the other two study areas By the end of a rotation (approx 30 years),according to this estimate, an average height of rubber trees in Nong Khai could be about 25