DEVELOPMENT OF THE FIJI PINE COMMISSION An account of the emergence of a major new industry

in a developing country. A. G. D. WHYTE*

ABSTRACT

This paper sketches the evolution of the Fiji Pine Com­mission, then provides a brief history of New Zealand's involve-ment in bilateral aid to the Fiji Fine Commission and how the School of Forestry at Canterbury became involved in the project Various contributions from the School are described, including brief outlines of the inventory and yield forecasting system, man­agement modelling, training of staff in implementing new tech­niques and research projects in foxtailing, spiral grain, sawn outturn and scheduling harvests. Recent involvement of School personnel in prefeasibility and feasibility studies on proposals to ultilise the expanding resource is also outlined. The paper con­cludes with a discussion of (a) the benefits and drawbacks of the School of Forestry contributions, (b) the problems encountered in getting techniques implemented, and (c) the efforts to com­plement other aid projects.

INTRODUCTION

The main purpose of this paper is to outline an unusual form of assistance to a forestry project in a developing country through contributions from New Zealand students. Aid for the Fiji Pine Scheme has contained a significant component of resident assist­ance from undergraduate and postgraduate students studying at the School of Forestry, University of Canterbury, Christchurch. This association has existed continuously since 1975 and may be of interest to both donors and recipients of aid, because it appears to have worked successfully. The intention here, there­fore, is to outline the framework within which this university contribution has been provided, to indicate briefly what it has comprised and to give possible explanations for its apparent success.

•School of Forestry, University of Canterbury, Christchurch.

412

FIJI PINE COMMISSION 413

HISTORICAL EVOLUTION OF THE FIJI PINE COMMISSION (FPC)

Plantings of Caribbean pine, Pinus caribaea Mor. var. hondurensis Barr, et Golf., and slash pine, Pinus elliottii Engel, in the 1950s and '60s form the oldest crops in the Fiji Pine Com­mission's present resource in the north-west of Viti Levu. These early plantings were carried out on highly degraded, fire-induced grasslands in the rain shadow zone of the island. A UNDP/FAO report on developing forestry and forest industries in Fiji (Forestal, 1972) led to a quickening in afforestation on com­munally owned lands which were virtually unused and uninhabit­ed. The programme was funded by loans from the Common­wealth Development Corporation and tlie Fiji government, then later with substantial aid from New Zealand, and, to a much lesser extent, from the U.K., Australia and the U.S. Peace Corps.

The Fiji Pine Scheme was set up in 1972 as a separate govern­ment department responsible to the Minister of Agriculture, Fisheries and Forests. Out of it arose the Fiji Pine Commission, established by Act No. 5 of 1976 to facilitate and develop an industry based on the growing, harvesting, processing and market­ing of species, primarily pines, in Fiji. The Commission is a statutory body which has entered into a partnership with the Fijian landowners to develop their land for industrial wood pro­duction and to encourage them to participate in the industry. The owners receive rentals on lhe leased land and royalties on the felled timber. The Commission is governed by a Board of ten members appointed by the Minister of Forests to represent the collective views of the Fiji government, the landowners, the providers of loan moneys, and the FPC management.

By mid-1982, the area established in pine plantations by, and under the management of, FPC amounted to 37 000 ha, consist­ing almost entirely of Caribbean pine, of which 4 500 ha are on Vanua Levu. About 90% of the area is under ten years of age, but, as a useful mix of sawlogs, poles and posts can be obtained from 15-year-old crops, the advent of very substantial quantities of utilisable wood is imminent. The present target is to establish a net stocked area of 60 000 ha which would give rise to a sustained annual yield of around 1 million m3.

In 1980, FPC called for proposals to utilise all or part of these burgeoning wood resources. Out of the four major bidders, Bri­tish Petroleum South West Pacific was invited by the FPC Board to carry out a detailed feasibility study on joint-venture processing

414 N.Z. JOURNAL OF FORESTRY

options for about one-third of the resource. The final report by the consultants is now to be submitted to BP and FPC in early 1983. Negotiations will still be proceeding for some time after that on how best to implement the recommendations of that report.

About 400 new permanent jobs arc likely to bo generated by 1985 if this development goes ahead. About one-half of these jobs are in harvesting, the remainder in mill operations, all of whioh are to be made as labour-intensive as is practicable, given the log categories and wood products that have to be produced. This number is in addition to the 120 permanent staff and 250 hourly workers that FPC already employs all year round, and to the 2 OOO or so contractors involved in nursery work, planting, releasing and other silvicultural practices who are engaged for shorter periods at a time.

Rentals, royalties and employment provide monetary benefits to the indigenous population, but there are also other benefits. On the social side, the expanding, extensive FPC road network (already over 600 km) has provided access to and from fifteen remote villages and numerous farms, good housing has been provided for FPC employees, the multi-racial composition of the FPC staff (Fijian, Indian, Chinese and European ethnic groups) has undoubtedly promoted better racial harmony, and the multi­plier effect has been of benefit to the infra-structural servicing groups in and around Lautoka and Nadi.

NEW ZEALAND/FIJI BILATERAL AID

In 1972 the Ministry of Foreign Affairs in New Zealand com­missioned a report by two senior Forest Service personnel, J. W. Levy and R. Usmar, on the viability of the Pine Scheme as it was then called. These two foresters recommended that aid from New Zealand would be a worthy project. Accordingly, aid has been given since that time in the form of funds to finance cost of establishment operations, donations of plant and equipment, and provision of special expertise in fields such as roading, vehicle repairs and maintenance, surveying, general forestry, silvi­cultural and wood-quality research, cattle grazing, mensuration, planning systems, utilisation and administration. Financial assist­ance to the Pine Commission since 1972 has exceeded NZ$8 million.

Three crucial benefits from this particular bilateral aid pro­gramme are worth emphasising:

FIJI PINE COMMISSION 415

(1) About half the annual cash grant is used to employ villagers in establishing new plantations which, together with rentals and royalties, represents welcome inputs of money away from urban centres

(2) Technology and systems of work are deliberately made labour intensive, so that as many rural people as possible can be gainfully employed

(3) There has been a firm, continued commitment to training local staff in all aspects of the work involved in modern industrial forestry.

Partly on account of these features, and because of other positive aspects of the whole project, the aid programme has been hailed a success.

Herein lies a problem, however, in that pressures are mounting daily to ensure the continued success of the scheme, particularly now that the technology and investments needed to utilise the whole wood resource will involve a massive increase in resources. The Fijian landowners were quick to appreciate the benefits of, and to learn the techniques needed for afforestation, but the next phase of harvesting and manufacturing will be much harder to assimilate within the framework of past and present Fijian culture, life-style and experience.

SCHOOL OF FORESTRY INVOLVEMENT WITH FPC

In 1974 the now General Manager of the Fiji Pine Commission, E. W. Gregor, sought assistance, through the Ministry of Foreign Affairs, from the Canterbury School of Forestry to refine manage­ment modelling and forward planning for the young resource. This represented a shift in emphasis from purely afforestation problems to a wider perspective of management of the maturing and expanding resource. The task was greatly simplified by the timeliness of this important decision: if it had been delayed any longer, a great deal of ground would have had to be covered in a much shorter time, which, from experience obtained in other projects, can lead to poor, hurried planning. Early in 1975, then, the writer visited Fiji, recommended the implementation of an integrated inventory system for monitoring early establishment, assessing development of the maturing resource and forecasting the outturn, in quality as well as quantity, of the final crops. It was also suggested that, if the system were to be adopted, young New Zealand students could be used to train the local staff and

416 N.Z. JOURNAL OF FORESTRY

also provide a research back-up in the field of mensuration and planning models.

In 1975, the first three students from Canterbury took up the challenge of first familiarising themselves with the suggested in­ventory procedures, then passing on thecc skills to local Fijian staff. The major benefit of this approach was that the local staff were far less reticent about questioning and discussing technical points than they would have been (and still are) with more senior training personnel. They identified ihcmcclves as of the same peer group as the students and, from the ensuing dialogue that became an integral part of the training procc:s, a much better appreciation of the techniques was obtained. It may well have been due to this commitment to comprehensive training that the inventory system has been run for several years now entirely by local Fijian staff.

Three students for each of the last eight years have now spent their long vacations working with the Fiji Pine Commission. In addition to training local staff in new techniques or refinements of existing ones, each student collects research data while in Fiji, then analyses these data during the following academic year at Canterbury and presents a dissertation, thesis or report on his or her findings. These findings are integrated within the existing inventory, forecasting or planning systems and form part of the training projects for the following year's students to undertake. The continuity of effort and the length of time (i.e., three months) for each student assignment have apparently contribu'ed greatly to the high rate of routine implementation of the research find­ings. This is no mean achievement in an applied discipline such as plantation forestry, where there is an ever-widening gap worldwide between knowledge possessed by researchers r.nd the extent to which this knowledge is accepted and routinely practised by forest managers.

TECHNICAL CONTENT OF SCHOOL OF FORESTRY CONTRIBUTIONS

A full list of School of Forestry research studies rela'ed to the Fiji Pine Commission is given in the Bibliography. Only a brief outline of the various stages of development and implementation of the research is given in this section. The extent of the contri­butions to forest industrial planning and management, however, is depicted in Fig. 1.

FIJI PINE COMMISSION 417

Routine-. Invent­ories of New and Maturing Plantat­ions.

Yield Forecasting Model for Maturing Crops

I

Models

Management Records of 3 Areas, Volumes and Operations

Forecasts of Potential Prod­uction up to 10 years ahead.

Long and Short-term Planning 9!/^

Detailed Forecasts of Outturn ' 1 year ahead for each logging Unit

[Wood quality, Utilization, 6

Marketing and Log Allocation Models

Rolling annual Harvest­ing, Industrial and Financial Plans for next Ten Years

Assessment of Actual"7" Outturn and Waste on Cutovers, Monitoring and Revising Yield

I Forecasting Model

FIG. 1: Schematic outline of inventory, modelling and planning systems for the FPC.

Boxes 1 and 2 refer to inventory and yield forecasting systems which are more fully documented in an earlier article (Whyte, 1978). They represent integrated systems for (1) monitoring the success of establishment and seedling survival, (2) collecting stand information on the present quantities of resources, and (3 fore­casting up to ten years ahead the potential volume assortments that could be produced. The yield forecasting model (sec also Broad, 1978) uses actual and predicted stand diameter statistics (mean and minimum diameler and variance of all diameters) together with age and stocking to forecast future numbers of stems by one centimetre dbhob classes using the Weibull distri­bution. Heights of trees are predicted from dbhob and age. Potential assortment volumes by one centimetre dbhob classes are estimated from regional, compatible volume and taper equations. Merchantability specifications and empirical cull fac­tors are then included in the forecasts of potential production for up to ten years ahead (Box 4). Use is also made of the manage­ment record system referred to in Box 3 (an early version of which was suggested by Bennison (1976) but which was developed in its present form by Fickes and Dunn (1980).

418 N.Z. JOURNAL OF FORESTRY

Box 5 refers to pre-harvesting inventories carried out on crops which are candidates for felling in the following year. For each logging setting, tree size and defect data are assessed on a sample of about 400 trees. The information collected in this kind of inventory includes measurements of the position and extent of crook, sweep and any other lack of stem straightness, and also a sub-sample of dendrometer readings of upper stem diameters to allow checks to be made of the representativeness of the cur­rent volume and taper equations. The size and defect data are then analysed (see Box 6) using dynamic programming to identify the optimum bucking pattern for single stems assuming uncon­strained demands for any one log type, and then the DP results from all settings are pooled to obtain through linear programming the best allocation of bucking strategies that satisfies market de­mands (see Eng, 1982). Remeasurement of matched plots allows checks to be made of the forecasts of stand diameter statistics, while log-scaling and cut-over waste assessments, when added together, are used to calibrate the reliability of the yield fore­casting model (Box 7). From the results procured in Box 8, which identities the determination of the next year's production, man­agement modelling of the whole resource allows revision of both the short- and longer-term forecasts of output (Boxes 9 and 10) using a combination of generalised mathematical programming models (e.g., Whyte and Baird, 1982) and a simulation model (Garcia, 1981).

The important characteristics of the overall system are: (a) The coherent structure of the individual components, (b) the stand-table type of forecasting of future crop statistics, (c) the use of compatible taper and volume equations, (d) the use of inventory both to derive the yield forecasts and monitor the reliability of each component of the yield forecasting and bucking models, and (c) the recognition of the need to assess defect and its effect on production and utilisation. The bucking model recognises the need for something more than just the "best" strategy for each stand on its own, and so allocates a year's production on the basis of realistic market constraints.

Extensive use of the bucking model has been made by the consultants engaged by British Petroleum in helping them to estimate proportions of various end-use categories and in guid­ing them in the design of the handling requirements for various utilisation plants. Box 6 also contains reference to other research studies that have currently been of relevance in evaluating pos­sible utilisation options — e.g., the extent of spiral grain and its

FIJI PINE COMMISSION 419

effect on quality of poles and sawn outturn; the frequency and implications of foxtailing; sawn outturn of various end-use cate­gories in relation to size of logs and trees; analysis of past timber sales; and harvest scheduling models.

Much of the earlier data-processing was aimed at programmable and non-programmable calculators after initial development on large computers at the University. This self-sufficiency, although extremely useful in encouraging the participation of local staff, soon became a limiting factor that FPC staff wished to circum­vent. FPC staff then started to make use of larger computers in Lautoka and Suva, until the level of contracting became so high that a small micro-computer at FPC itself was justified. Within 12 months of the initial purchase, a second such machine has had to be ordered to cope with the increased demands. Training in the early local calculator systems and, latterly, appreciation courses in computer data-processing have paved the way for routine implementation of FPCs own computers by local, but non-computer specialist staff.

DISCUSSION

Tlie design of the inventory and yield forecasting system for FPC was tailored to suit the specific needs in 1975 of the organ­isation. Although those needs have changed as the emphasis has shifted from purely afforestation considerations and chip sales, towards large-scale industrial manufacture of a range of wood products, the original system is still basically the same; there has simply been the addition of special components which reflect various sophistication factors in the management of resources. The stability of the procedures is important where data are scarce and consistency of information is therefore of great benefit. The early start to designing a system proved invaluable, since the time was then available to put systems in place before there was a compelling need to have data and results in hand. The General Manager's foresight, therefore, sets a good example for others in his position to take note of and, moreover, to follow.

The inventory and yield forecasting systems were made de­liberately labour-intensive to employ as many rural people as possible both in field-work and in data-processing. The sound educational background of the Fijian staff together with the gradual build-up in sophistication of techniques has allowed local staff to run the routine work most competently and attain a

420 N.Z. JOURNAL OF FORESTRY

standard of management servicing that is at least on a par with forestry firms of that size in both developed and developing countries throughout the Pacific. The present level need not be stretched any higher, but merely consolidated by increasing the number of staff trained to the same depth.

The role played by forestry students in promoting useful dialogue between instructor and trainee should not be lightly dismissed. There are examples in several parts of the world of specialists' recommendations that lie idle simply because there has been insufficient interaction among specialists, trainers and trainees. It is generally recognised in UNDP projects, for example, that what is often missing in implementing recommended pro­cedures is the long-term physical presence of people to explain why things should be done in certain ways (A. J. Leslie, formerly Director of Forest Industries Division of FAO, pers. comm.). The forestry students have provided that type of support to a degree, and involvement in this way may well be the prime reason for the apparent success of this experimental approach.

There have been problems, of course, in that some, but not all of these young instructors have a flair and the patience for the interactive teaching they must do. A few do not find it easy to adapt to a different environment from their own, while others shine more at the research side than teaching (and vice-versa, of course).

There is, too, a recurring problem of a young organisation such as FPC not being able to absorb short-term costs of research like slowing or halting normal production processes to reduce the possible effects of confounding factors during a special in­vestigation. Unless almost instantaneous savings can be guaran­teed in the long term, there is an obvious reluc'ancc on managers' parts to spend money just on the off-chance of deriving benefits. The most challenging problem, however, is the need to co­ordinate all aid projects of which the one examined here is but a single example at FPC. It is unrealistic to assume that the existing management alone can cope with co-ordinating re:earch, routine management and training aspects. There needs to be someone who has a foot in each camp and who has a long-term presence in the organisation to ensure that conflict and unneces­sary duplication are reduced to a minimum. FPC has appointed, therefore, a Special Projects Manager and also a Training Officer, both of whom have assumed responsibilities for such co-ordina­tion in the future.

FIJI PINE COMMISSION 421

ACKNOWLEDGEMENTS The co-operation and support of E. W. Gregor and the FPC

staff (particularly J. Dunn and J. Niumataiwalu), New Zealand aid personnel, and the students undertaking assignments at each stage in this ongoing project are gratefully acknowledged, as is the financial and administrative backing of the Ministry of Foreign Affairs without which the project could not have been undertaken.

B I B L I O G R A P H Y A N D R E F E R E N C E S Adams, P. M., 1978. Foxtailing in Pinus caribaea var. hondurensis at

Lololo Station, Fiji. Canterbury School of Forestry, B.For.Sc, dis-sertation; 70 pp.

Bennison, B. L., 1976. The development of a compartment register system. Canterbury School of Forestry, B.For.Sc, dissertation; 24 pp.

Broad, L. R., 1978. A revised yield forecasting model for Pinus caribaea in Fiji. Canterbury School of Forestry, M.For.Sc. Report: 56 pp.

Cromarty, P. L., 1981. Validation of FPC yield forecasting model. Canter­bury School of Forestry, B.For.Sc, dissertation; 211 pp.

de Kluyver, C. A.; Whyte, A. G. D.; Baird, F. T., et al., 1980. Forest harvest scheduling in Fiji: a comparative case study of linear pro­gramming, heuristic and decomposition techniques. N.Z. Oper. Res., 8 (1): 33-72.

Eng, G., 1982. A methodology for forest outturn assessment and optimal tree bucking. University of Canterbury M.Sc, thesis; 108 pp.

Eng, G.; Whyte, A. G. D., 1982. Optimal tree bucking Proc. Oper. Res. Soc. N.Z.: 91-8.

Evison, D. C, 1979. Pre-stratification for Fiji Pine Commission routine inventory. Canterbury School of Forestry, B.For.Sc, dissertation; 68 pp.

Fickes, J.; Dunn, J., 1980. FRRAP (Forest Reporting, Recording and Planning) Report Listing. FPC Report. 121 pp.

Forestal, 1972. The development of forests and forest industries in Fiji. Tech. Rep. FO/SF/FIJ2.

Garcia, O., 1981. IFS, an interactive forest simulator for long range planning. N.Z. Jl For. Sc, ll (1); 8-22.

Geiser, H., 1977. A yield forecast model for Pinus caribaea in Fiji. Canter­bury School of Forestry, M.For.Sc. report; 66 pp.

Geiser, H.; McCallum, W. J.; Twaddle, A. A., 1977. Tree volume equations over and under bark for Pinus caribaea and Pinus elliottii in Drasa/ Lololo, Nadarivatu and Nausori Highlands. FPC Report.

Hazelman, E. C, 1977. The involvement of the indigenous Fijians in the logging industry of the Fiji Pine Commission. Canterbury School of Forestry, M.For.Sc. report; 66 pp.

Katz, A , 1978. Root-pruning of potted nursery stock of P. caribaea var. hondurensis. Canterbury School of Forestry, B.For.Sc, dissertation; 34 pp.

Manley, B. R., 1976. Yield projection of Pinus caribaea at Drasa, Fiji, using the Weibull function. N.Z. For. Serv., For. Res. lnst. Mensura­tion Rep. No. 52; 30 pp. (unpubl.).

422 N.Z. JOURNAL OF FORESTRY

McEwen, S. E., 1976. A study of foxtail height, diameter and frequency in a three year old stand of Pinus caribaea var. hondurensis. Canter­bury School of Forestry, B.For.Sc, dissertation; 38 pp.

Morey, C. W., 1982. Relative demands and actual sales from Drasa sawmill and post and pole yard. Canterbury School of Forestry, B.For.Sc, dissertation; 93 pp.

Rolfe, M. JL., 1980. Log scaling at Drasa Post and Pole Yard. Canterbury School of Forestry, B.For.Sc, dissertation; 62 pp.

Sargento, J., 1978. Vegetative propagation of Pinus caribaea and Pinus ooocarpa by means of needle fascicles. Canterbury School of Forestry, M.For.Sc. thesis; 92 pp.

Sole, J., 1981. Stumpage Appraisal 1980. FPC Report 42 pp. Tan, C. K., 1982. Sawn outturn of Pinus caribaea (Fiji Pine) in Fiji.

Canterbury School of Forestry, B.For.Sc, dissertation; 96 pp. Whyte, A. G. D., 1978. Inventory and yield forecasting systems for

Caribbean pine in Fiji. Proc. For. Res. lnst. Symp. 20: 41-7. 1980. "Integrating wood quality with quantity information: an

example with Fiji Pine." IUFRO Div. 5 Conference, Oxford 1980; 17 pp.

Whyte, A. G. D.; Adams, P. M.; McEwen, S. E., 1981. Foxtailing of Pinus caribaea var. hondurensis in Fiji: frequency, distribution of occurrence and wood properties. For Ecol. & Management. 3: 237-43.

Whyte, A. G. D.; Baird, F. T., 1982. "Integrated modelling of the New Zealand forest industry." Reg. Sci. Association Conference. Canberra. 6 to 8 Dec; 21 pp.

Whyte, A. G. D.; Wiggins, P.C; Wong, T. W., 1980. "Spiral grain in Fijian grown Pinus caribaea var. hondurensis and its effects on sawn outturn." IUFRO Div. 5 Conference. Oxford 1980; 18 pp.

Wiggins, P. C, 1979. Spiral grain in Pinus caribaea in Fiji. Canterbury School of Forestry, B.For.Sc, dissertation; 167 pp.

Wong, T. W., 1980. Twist in timber sawn from Pinus caribaea, with a range of spiral grain angles, in Fiji. Canterbury School of Forestry, B.For.Sc, dissertation; 151 pp.

Wybourne, L. J., 1982. A review of spacing and thinning in Pinus caribaea var. hondurensis in Fiji. Canterbury School of Forestry, B.For.Sc. dissertation; 117 pp.

There have been formal contributions by two other members of the School of Forestry staff but pertaining to different aspects of forest management. For the sake of bibliographical completeness, they are shown below.

McKelvey, P. J., 1978. Legislative Provision for Some Fiji Pine Commission Operations. Unpubl, report; 7 pp.

—• 1978. Preliminary Thoughts on FPC and Multiple-use. Unpubl. report, 8 pp.

1978. Comments on FPC Research. Unpubl, report, 6 pp. O'Reilly, R. N., 1979. Environmental Guidelines for Logging and Roading

at FPC. Unpubl, report, 17 pp.