Previous Page Table of Contents Next Page



In addition to the present low copra and coconut oil prices and an uncertain future many coconut plantations are experiencing declining production because of increasing senility with many plantations in excess of 50 and 60 years of age. Although higher yielding hybrids are being planted, the uptake of new technologies has been low in some countries because of poor farmer experience with hybrids. Monocropped coconuts are increasingly regarded as no longer an economic proposition and in some countries particularly in the South Pacific, copra and coconut products will not be a viable export unless the return to farm labour can be increased, and will be grown only for domestic uses.

It is suggested that lessons can be learned from other integrated systems such as the combination of Pinus radiata and pastoral agriculture in New Zealand, Australian agroforestry, silvopastoral systems in North America with slash pine (Pinus elliotti), loblolly pine (Pinus taeda), longleaf pine (Pinus palustris), Douglas fir (Pseudotsuga menziessi) and black walnut (Juglans nigra) and the integration of sheep and cattle with rubber (Hevea brasiliensis) and oil palm (Elaeis guianensis) particularly in Malaysia. In a number of cases predictive models have been developed to determine optimum combinations of crops, trees and livestock from the many feasible options which can be sustained from a given set of resources.

For the immediate future it is suggested developments will take place in three particular areas:

  • as the majority of coconut areas have been planted at traditional spacings there will be a need to continue screening forage species for shade tolerance and persistence, and especially low light situations of <50 percent light transmission. In addition to native species and those listed in Chapter 3, the grasses Stenotaphrum secundatum and Paspalum notatum cv. Competidor have demonstrated both shade tolerance and persistence in recent trials, with Paspalum wettsteinii, Paspalum malacophyllum and Brachiaria decumbens being less persistent under sustained grazing. Of the legumes, Arachis species have been impressive in trials with both A. pintoi and A. repens sufficiently competitive even with the more vigorous grasses to maintain a high legume content in frequently defoliated plots. Desmodium ovalifolium, D. intortum and D. heterophyllum have also performed well in on-going trials and Arachis glabrata and Aeschynomene americana cv. Glenn should be further investigated;

  • for tall coconut varieties traditional spacings have ranged from about 7 × 7 m to 9.1 × 9.1 m with densities from 120 to 200 trees ha-1. However, various sources have suggested that ‘the economics of spacing tree crops more widely to allow more light to reach the understorey and facilitate production of forage for livestock’ should be investigated. Several reports have indicated that the same tree density can be achieved by various planting systems with little effect on the mean coconut yield palm-1 and ha-1. Rectangular planting systems with wide between-row spacing (rather than square planting systems) are more suitable for intercropping. Recently tried hedgerow planting systems have used spacings of 12.2 × 4.6 m, 15 × 9 m, 9.1 × 5.5 m, 12 × 5 m, 18.3 × 4.6 m, 13.6 × 6.1 m, 14.5 × 6.1 m and 15 × 5 m. In Tonga, coconut yields from hedgerow plantings (15 × 5 m) were reported to be higher by 25 percent. If hedgerow planting systems are more widely adopted in future then with light transmission conditions in the inter-row areas being in excess of 80 or 90 percent throughout the life of the coconut trees then the need to identify shade tolerant forage species will be less of a priority and species which are used in open areas can be recommended. If new coconut layouts are adopted both for tall varieties and hybrids this will have implications not only for pasture species but also for stocking rates, grazing systems and the management and economics of the whole integrated system;

  • with overwhelming evidence that returns from monocrop coconuts can be increased substantially by intercropping and livestock production, there is likely to be increased development of coconut based multicropping systems where various management options are modelled before hand to maximize returns to the grower. This information will be available to growers so that they can select their preferred option on the basis of models where the latest prices and best estimates can be utilized to suggest future trends and likely returns.

This Chapter examines the trends in copra price, coconut plantation age, the likely influence of coconut varietal improvement, lessons which can be learned from other integrated systems and finally examines three key areas where further research is required.

10.1 Trends in copra (and coconut oil) price

In the last 15 years coconut products have suffered from a low international market value and competition from copra substitutes (Parawan and Ovalu, 1987). In fact, although there have been volatile price fluctuations (see de Silva, 1988) also there has been a long term decline in world copra prices since the 1950s (see Figure 226). This has not been limited to coconut, and other commodities showing international market price declines include cocoa, coffee, rubber, sugar and rice (FAO, 1993). Some declines have shown an average annual rate of 3%. Pehaut (1988) refers to the 25 years between 1960–1985 as “a fundamental chapter in the history of the world's oil crops economy”. Also with the development of crushing operations in producing countries, there has been a sharp drop in the copra trade and an increase in coconut oil trading (Daviron, 1994). In 1990 copra prices were so low that many producers simply left coconuts uncollected in the plantations (see Figure 227), as labour and other costs were higher than the anticipated return from the copra.

Figure 226

Figure 226. - Annual average price for copra: illustrating the structural decline in world copra price over time (Anon., 1993).

The problem facing the Tongan coconut producer was described (FAO, 1987b) as one of “declining returns per unit area and per man day … which has been brought about by the interplay of widely fluctuating international market prices and low yields of coconut palms. Fluctuating international market prices result in fluctuating domestic prices to the producer, despite the application of measures such as a stabilization fund that is designed to insulate the farmer from international market forces. Given the projected low international market prices for the next decade or more, it is not realistic to anticipate a substantial, long-term and sustained increase in domestic prices”. A more recent report (Anon., 1990c) also mentions processing problems, high wage rates and better returns from other crops, particularly root crops, as causes of the decline in the export of coconut products, the ‘traditional’ export of Tonga.

Opio (1988) commenting on the fact that “most farm households have either totally or partially neglected their coconut plantations” indicated that “coconut is no longer considered to be a remunerative cash crop by most farmers in Western Samoa”. “At the current prices, coconut production is not viable .... the return to farmers' effort is estimated at WS$ 0.56 man hour-1 which is less than the current mimimum wage of WS$ 0.65” (Opio, 1989). In many Pacific Islands (and other countries) rising wage levels and the low returns from copra (and coconut oil) mean that coconut farming is becoming less attractive (Dalla Rosa, 1993)! Foale and Ashburner (1994) suggest that “the attraction of coconut as a financial investment has disappeared almost completely, but its place in the household, local and regional economies remains secure and indeed grows in proportion to the rise in population of local communities”. The general situation of the coconut industries in Papua New Guinea, Solomon Islands, Vanuatu, Fiji and Tonga has been reviewed in Foale and Lynch (1994). They noted that “declining exports of copra and coconut oil were reported .... in some cases copra production has ceased and nuts surplus to domestic requirements are used solely for livestock”.

In Malaysia, the coconut industry has been described as stagnant, largely because of high cost and low productivity (Anon., 1992a). Cost of production in 1986 was US$ 304.22 ha-1 compared with US$ 237.76 in Philippines, US$ 217.84 in Indonesia, US$ 194.05 in India, US$ 125.55 in Papua New Guinea and US$ 75.87 ha-1 in Sri Lanka.

Figure 227

Figure 227. - Uncollected coconuts in Tonga in 1990, due to low copra prices.

Some idea of the fluctuating nature of copra prices, as well as the low price in 1990 and early 1991, is given by the domestic price of copra paid by the Vanuatu Commodities Marketing Board for hot air dried and smoked copra for selected months from 1982 to 1992 (Table 242). Monthly prices for coconut oil and copra 1986 to 1993 (US$ MT-1) as reported by APCC (Anon., 1990a; Taufikkurahman, 1991, 1994) are shown in Tables 243 and 244.

Although there has been an increase since the lows of 1990, and although World Bank commodity price projections (Anon., 1993) indicate that the copra price will continue to increase over the next few years (see Table 245), the long term outlook is uncertain. With limited production gains expected for coconut oil versus a large supply increase for palm kernel oil, some substitutions between the lauric oils will occur (coconut oil's share of lauric oil production has dropped from about 85% in the early 1970s to 75% in the mid-1980s. Over the next decade its share is likely to drop below 70%. Its share of the total export market for vegetable oils has also declined steadily, from 6% in 1986 to 5% in 1990). As copra production remains primarily a smallholder operation (95% of production comes from small-scale units of 0.2–0.4 ha in size - Daviron, 1994) and there has been little replanting in recent years then in countries like the Philippines (the major copra/coconut oil exporter) there is likely to be a decline in the production capacity of trees in the next 10 to 15 years. Also, domestic consumption of coconut is increasing in producing countries so that around 70% of all coconut production is now consumed domestically and only 30% is available for export (although the percentage exported is higher in South Pacific countries). Realizing that rehabilitation of the coconut industry is necessary to maintain the continuity of supply of coconut oil, the Philippines has taken a US$ 121 million loan from the World Bank (Persley, 1992) to undertake a major replanting programme. Although market studies by the World Bank and the Philippines Government have concluded that future demand for coconut products is good, both for industrial uses and as an edible oil, this may depend upon the Philippines (as the dominant exporter) ensuring that end-user confidence is maintained in the continued stability of its supplies. Persley (1992) suggests that “the share for coconut oil of both the non-edible and edible vegetable oil markets is easily sustainable, with some room for modest expansion, unless end-users become convinced that they cannot rely on a stable supply from the Philippines and other smaller exporters. In that event, the buyers will make long-term investment decisions in favour of substitute products, especially for industrial purposes. Deterioration of this industry would have important macro-economic consequences for the Philippines” and particularly for the millions of smallholders dependent on the industry. Hone (1994) indicates that “it is widely recognized that coconut oil and copra meal face difficulties in international markets” and reviews some of the problems while also examining possible opportunities for the production of aseptic coconut cream, skim milk and water. More recently Shrimpton (1995) described on-going research aimed at replacing diesel and other fuels with coconut oil in the Pacific Islands.

Table 242. - Domestic price paid by Vanuatu Commodities Marketing Board (VCMB) in Vatu (120 Vatu = US$1, April 1993) for hot air dried and smoke dried copra (Anon., 1993)

Date Price ChangeHot AirSmoke
April, 198218,00018,000
November, 198215,00013,000
August, 198320,00016,000
October, 198323,00019,000
December, 198325,00021,000
February, 198430,00026,000
August, 198436,00031,000
December, 198441,00036,000
September, 198535,00030,000
March, 198625,00020,000
May, 198618,00013,000
July, 198723,00018,000
October, 198730,00025,000
September, 198925,00020,000
August, 199020,00015,000
January, 199115,00010,000
January, 199222,00017,000
May, 199225,00020,000

In many countries in the South Pacific copra and other coconut products will not be a viable export unless the return to farm labour can be dramatically increased (Anon., 1990c). On the other hand, coconuts will continue to be grown as an intrinsic part of the multi-storey cropping system and used for a wide variety of food and other domestic uses. Thus in its traditional role as the “tree of life” in Polynesian society, coconut has an assured future.

Table 243. - Monthly prices of copra, 1986–1993 (US$ per MT, Philippines/Indonesia, Cif Europe)

Annual Average197309398348231286380295

Source: Anon. (1990a); Taufikkurahman (1991, 1994).

Table 244. - Monthly prices of coconut oil, 1986–1993 (US$ per MT, Cif Rotterdam)

Annual Average297442565517337433578450

Source: Anon. (1990a); Taufikkurahman (1991, 1994).

In coconut producing areas there has been a general realization that monocropped coconuts are no longer an economic proposition either at plantation or smallholder level. As indicated in Section 1.4 various intercrops and intergrazing schemes have been introduced and from depending solely on monocrop copra production many coconut farmers now depend on intercropping, multistorey or multiple cropping systems (Baconawa et al., 1985). In Indonesia (Mendra et al., 1991) plantations which used to produce only coconuts now produce coconuts, vanilla and forage for livestock (from leucaena used to support the vanilla). Dwiwarni et al. (1987) noted that various multi-storeyed coconut cropping systems gave highest income ha-1 while coconut in monoculture gave the lowest returns. In Malaysia, Papua New Guinea and Solomon Islands, cocoa and coconut is a popular combination. Other intercrops include coffee, citrus, cloves, banana, papaya, pineapple, kava, watermelon and various vegetables. In Tonga (FAO, 1987b) “some success has been achieved in the intercropping programme, notably in the area of banana and vanilla production. Greater intensification of the programme to include livestock production is an imperative”. In Philippines the more common intercrops include rice, corn, peanut, cassava, sweet potato, taro, pineapple, fruit trees, cacao, coffee, black pepper, vegetables and pasture for grazing by livestock (Aguilar and Benard, 1991). Foale and Ashburner (1994) suggest that the future for the coconut lies in its integration into multi-culture cropping systems.

Table 245. - World Bank Commodity Price Projections, constant 1990 (Anon., 1993)

CommoditiesUS$ per tonne
Palm kernel219234243227

10.2 Coconut plantation age

One of the factors identified as contributing to the low income in the coconut production sector is the declining productivity of coconut trees (Arancon, 1988) due to senility (others identified by Aguilar and Benard, 1991 include the declining and unstable prices of coconut products, non-adoption of recommended coconut practices, underutilization of coconut farms due to tenure problems, ineffective or weak support services and poor credit facilities).

According to Anon. (1995) 'by the mid-eighties the economic competitiveness of coconut production became more and more doubtful … This has had dramatic consequences for research and development plans … In several major production centres, the vital long term replanting efforts for the regeneration of the productive stands of palms lost much official backing and public support. As a result, many countries now have an aging, unproductive stand of palms which are unsuitable for reviving the sector. Lack of drive in the replanting effort leads to lack of confidence in the processing sector and as a consequence to delays in the renewal of infrastructure. This in turn leads to uncompetitive systems and stagnating output".

The useful bearing life of tall varieties of coconut is usually said to be around 60 years. In Samoa, Opio (1987a) suggested that the yields of mature Local Talls are not economically viable beyond 40 years unless intercropped with other cash crops. Average yields in many producing countries are close to 0.5 tonnes ha-1 compared with figures of more than 4 tonnes ha-1 mentioned for hybrids (Opio, 1988; Persley, 1992), although the latter production can only be realized with a good fertilizer programme.

Of major concern is that in the Philippines, THE major exporter of copra and coconut oil, with about 70% of the coconut products traded in the world market according to Carlos (1993), in 1988 some 30% of the coconuts were more than 60 years of age (Anon., 1988b). Of equal concern is that another 39% of existing palms were between 31 and 60 years of age and only 13% were between 0 and 10 years of age (Arancon, 1988). Magat (1993) suggests that 15% of palms are senile and that unless measures are taken there will be a 2% decline per annum in coconut production. A major replanting programme was initiated with a US$ 121 million World Bank loan in 1990 with the aim of rehabilitating the industry. In a number of other countries the picture is not much different:

Figure 228

Figure 228. - Senile palms aged in excess of 70 years, Mulifanua, Western Samoa.

10.3 Coconut varietal improvement

Traditionally the majority of coconuts planted were various local tall varieties, sometimes poorly selected and with limited genetic potential. While improved selection procedures in tall populations can give limited improvements in yield potential it is now well established that only by using hybrids (and providing appropriate nutrient inputs) are coconut yields likely to be significantly increased (Manciot and Sivan, 1991). Coconut breeding over the past 30 years has demonstrated that hybrids between the ages of 10 and 20 years are capable of yielding up to 6.5 tonnes ha-1 year-1 of copra, under favourable environmental conditions and good management (Persley, 1992), with average production probably between 2.5 and 4.0 tonnes ha-1 year-1. This contrasts with the world average yield of 0.5 tonnes ha-1 year-1 and good tall yields of between 1.0 and 2.0 tonnes ha-1 year -1. The subjects of coconut varieties, selection and breeding have been reviewed by Ohler (1984). Work on coconut hybrids has been undertaken in a number of countries: Fiji, India, Ivory Coast, Jamaica, Sri Lanka and Vanuatu. Useful references include: Bourdeix et al., 1992; Chomchalow, 1993; Daniel et al., 1991; de Nucé de Lamothe, 1990; de Nucé de Lamothe et al., 1991; de Taffin and Sangare, 1989; de Taffin et al., 1991; Dootson et al., 1988; Duhamel, 1993; Foale, 1993; Le Saint and de Nucé de Lamothe, 1987; Manciot and Sivan, 1991; Nair, 1993; Nair et al., 1993; N'cho et al., 1988; Persley, 1992 and Sangare et al., 1988. De Nucé de Lamothe (1990) has summarized some of the hybrid work noting that in the Ivory Coast production of hybrids exceeds 5 tonnes of copra ha-1 year-1 under good conditions and even the first hybrids planted 25 years ago still produce 3.4 tonnes ha-1 year-1 compared with 1.6 tonnes ha-1 year for West African Talls under similar conditions. An indication of the early differences in copra yield performance in Western Samoa between hybrid (see Figure 229) and local tall coconut varieties receiving similar inputs is shown in Table 246 (Aveau, 1993) and a comparison of yield streams for hybrids and local Talls for Fiji, Vanuatu and W. Samoa is given in Table 247.

Table 246. - Copra yield performance of a natural hybrid and typical local tall in kg per ha (Aveau, 1993)

Age in yearsHybridLocal Tall

Figure 229

Figure 229. - Hybrid coconuts, WSTEC, Western Samoa.

Table 247. - Coconut Yield Streams Per Hectare in the Pacific (in tons) (Opio, 1990b)

YearsFijiVanuatuW. Samoa
HybridsLocal TallsHybridsLocal TallsHybridsLocal Talls
  10     0     0     0     0     0     
  20     0     0     0     0     0     
  30     0     0     0     0     0     
  40.170     0.280     0.220     
15–19NA      1.60NA      1.40NA      1.53
20–24NA      1.38NA      1.29NA      1.27
25–29NA      1.58NA      1.55NA      1.60
30–34NA      1.43NA      1.47NA      1.55
35–40NA      1.21NA      1.20NA      1.16
Over 40NA      0.95NA      0.90NA      0.91

N.A. = Not available.

Source: 1. Fiji MPI (1985). Report of the Agricultural Commodity Committee, Suva, August 1985.
2. Calvez C., et al. (1985) Improvement of Coconut in Vanuatu and its importance to the Pacific Region, Oléagineux 40 (2), 447–486.
3. Efu S. (1986). Report on Coconut Development in Western Samoa. Coconut Seed Multiplication Project, Olomanu, Western Samoa.

However, there are problems and in a number of instances “the acceptance of hybrids by smallholders has been subject to a number of real or perceived constraints and poor performance has been reported under some local conditions. These constraints to hybrid production and use, which relate to lack of adaptation to specific environment and difference in nut size, need to be further investigated” (Persley, 1992). Although potentially earlier and heavier bearing, problems which have been mentioned include: rubbery copra, shorter bearing life requiring earlier replanting, smaller nuts with a larger handling component (and therefore costlier copra) and lower production. De Taffin et al. (1991) mention susceptibility to nut-fall caused by the fungus Phytophthora heveae; when the hybrid Maren was introduced to regions of Papua New Guinea which had the pests Scapanes and Rhyncophorous the hybrid suffered a much higher rate of fatal damage than the local tall coconut population (Foale, 1993); in Sri Lanka soil differences affected hybrid performance (Mahindapala, 1993) and Chomchalow (1993) indicates that in poor growing conditions (soil, weather and management) hybrids may produce rather poorly. In Philippines, Carlos (1993) indicates that “the foreign hybrids performed fairly well in three regions of Mindanao … they performed poorly in the Luzon and Visayan areas, which are exposed to occasional typhoons and uneven rainfall patterns … in many cases, the palms were cut down for producing small, few and difficult-to-husk nuts. Local hybrids performed much better.” In Tanzania, hybrids show far less tolerance to water deficit than the local tall population (Foale, 1994). In Vanuatu, the writer observed that the uptake of hybrid seednuts (produced at the IRHO Station) by smallholders and plantations was quite low due to poor experience of some growers with hybrids, the relatively high costs involved in replanting and the uncertain returns from copra. In Indonesia the use of hybrids involved an increase in labour associated with copra production and higher costs for inputs (Anon. 1988c). Foale (1994) suggests also that some of the high yield claims for hybrids should be treated with caution as these have often been based on the first few years of production and comparisons have been made between young hybrids and old talls. “The peak yield potential of hybrids is around 30% higher than for a tall population”.

However, in spite of various problems coconut replanting programmes are utilizing new hybrids and in some countries large areas have already been established. Thus in Indonesia, some 15% of the coconut area is under hybrids (Mahmud, 1993). A large number of dwarf × tall hybrids as well as improved high yielding tall varieties are being produced from seed gardens for distribution (Chomchalow, 1993; Mahindapala, 1993). If sufficiently large areas are established and the higher yield potential is achieved then the average copra yield ha-1 should rise and the returns per unit area and per man day may make copra production a more viable proposition. The key to higher and sustained yields is likely to be the availability and cost of fertilizer.

10.4 Identified problems in the pasture-cattle-coconut system

Although cattle are successfully grazed on both native and improved pasture under coconuts and this book has illustrated the various facets of the pasture-cattle-coconut system and the likely returns to the farmer, there are many areas requiring further investigation.

Shelton et al. (1987a) reviewed various aspects of the role of pastures in the plantations of Asia and the Pacific and identified problem areas and priority areas for further research:

  1. The identification of the most successful plant adaptive characteristics for shaded environments and the better understanding of the basis of grass/ legume competition under shade. This will lead to more efficient selection of improved species as well as to recommendations for the management of grasses and legumes under shade.

  2. The identification of more shade tolerant and resilient grasses is needed “Lack of persistence of grass species and invasion by unpalatable, broadleaf weeds are the main management difficulties in coconut plantations”.

  3. The influence of shade on soil nitrogen availability (and therefore plant growth and nitrogen content);

  4. The effect of shade on the nutritive value of grass species.

  5. The effects of pastures on plantation crop yields;

  6. In Malaysia, as the rearing of sheep on cover crops in young rubber plantations gains acceptance there is a need to “develop sheep grazing management systems which optimize animal production, weed control and stability of forage resources in a light environment which declines with planting age.”

  7. As the lower animal liveweight gains may be due to decreased forage production as well as reduced nutritive value of the grass species there is a need to investigate management practices aimed at increasing animal production, such as the introduction of legumes and concentrate supplementation.

An FAO report on integrating crops and livestock in West Africa (FAO, 1983d) identified other problems and areas where further information was required:

  1. The problem of plantation owner permission for herders to graze the plantations.

  2. The possibility of using plantations to grow fodder for sale.

  3. Methods of integration of food crops, tree crops and livestock.

  4. The use of intermediate and final by-products of tree crops as livestock feeds.

At the International Livestock-Tree Cropping Workshop held at MARDI in 1988 Reynolds identified a number of areas where research was required:

  1. The use of indigenous legumes, e.g., hetero (D. heterophyllum) which are possibly more persistent than many exotic legumes and are capable of fixing moderate amounts of N.

  2. The key problem of weed control (and improved management practices).

  3. The seasonality of forage production and the use of supplementary feed sources.

  4. Cut-and-carry stall/pen feeding systems based on various conventional and non-conventional feeds.

  5. Smallholder farming systems or integrated farming systems, i.e., small model units/farms of size 0.5, 1.0, 2.0 … × ha.

  6. Simple, robust, low input systems that are stable, low risk and sustainable.

  7. The economic, social and marketing aspects of intercropping.

  8. Different tree layouts based on findings in work with forest species, e.g. Hawke et al. (1984); Lewis et al. (1985); Percival and Knowles (1988).

  9. The influence of shade on milk production.

Recent findings in some of the areas mentioned above have been included in this book and research continues. Before discussing in detail some key areas where future developments are likely to take place it is considered appropriate at this stage to look briefly at other integrated systems that have had similar problems and perhaps to see what lessons can be learned.

10.5 Lessons from other integrated systems

As indicated throughout this book the growing of pastures and the integration of livestock is not a practice which is unique to coconut. Livestock have been grazed in forests for thousands of years (Nair, 1991; Payne, 1985; Rackham, 1986; Van Maydell, 1990) and in recent times livestock have been grazed under a range of tree crops including oil palm, rubber, clove, cashew, guava, kola, walnut, kapok, citrus, mango, other fruit-trees and in reafforestation projects. What can we learn from the experience of livestock integration with these other tree crops? What approaches and methodologies used elsewhere can be applied to the coconut-pasture-cattle system? Some information from recent experiences under rubber and oil palm has been incorporated into specific chapters but perhaps there is need to look more closely at other combinations which could have important lessons for the coconut-pasture-cattle combination. These include the combination of livestock and trees (silvopastoral agroforestry systems), as well as recent work on oilpalm and rubber.

10.5.1 Pinus radiata and livestock in New Zealand

Although located in a temperate area it is felt that the combination of Pinus radiata and pastoral agriculture in New Zealand is such a system with the advantage that it has been well documented (Anon., 1974a, 1975d; 1989a, Arthur-Warsop, 1985; Butcher, 1988; Cox et al., 1988; Fenton et al., 1972; Gadgil et al., 1986; Gillingham et al., 1976; Hawke, 1991; Hawke et al., 1993; Hawke and Wedderburn, 1994; Knowles, 1972, 1975, 1988, 1989, 1991; Knowles and Klomp, 1976; Knowles and Percival, 1983; Knowles and Tahan, 1979; Knowles and West, 1984, 1988; Knowles et al, 1973; Koehler et al., 1986; Marnane et al., 1982; Paton, 1988; Percival et al., 1984a, 1984b, 1984c, 1984d, 1988; Percival and Knowles, 1983, 1988; Sparking et al., 1989; Steel and Percival, 1984; Stewart, 1985; Tustin et al., 1979; Walker, 1984; West et al., 1981, 1987, 1988; Whiteside and Sutton, 1983).

McQueen (1978) indicated that “grazing of indigenous forest had been a common practice since herbivorous mammals were introduced to New Zealand 200 years ago”. As well as feeding the animals this was a means of clearing undergrowth for weed control and access to timber trees.

Silvopastoral agroforestry systems were first considered in New Zealand in 1979, as a result of developments in plantation forestry with ‘direct sawlog’ regimes for radiata pine (Fenton and Sutton, 1978). Under such regimes the stand is kept open by being thinned to a final crop of 200 to 250 stems ha-1. Grazing with sheep and cattle was considered an opportune way of utilizing the undergrowth and also of getting an early return, particularly where trees were planted on pasture (Fenton et al., 1972; Knowles, 1972). Workers at the Forest Research Institute (McQueen, 1978) realized that this somewhat negative control of undergrowth could be turned into a positive economic gain from livestock if grass could be grown satisfactorily among the trees (the same realization has led to the establishment of improved pastures under coconuts rather than just using livestock to control existing vegetation). Through research, and demonstration on a farm scale, the concept has developed into a viable land-use system which encompasses various options for both the farmer and forester. These include forest grazing, shelterbelts managed for timber production, plantations established on pasture and fenceline growing of timber trees (see Figure 230).

Over the last 30 years a considerable database of information has been built up on the following (under P. radiata):

(Some of the major findings have also been reviewed by Torres, 1983).

Figure 230

Figure 230. - Pinus radiata for timber planted on fence lines in New Zealand.

Control of pampas grass (Cortaderia sp.) by grazing cattle has proved to be particularly effective and Lotus uliginosis (syn. pedunculatus) cv. Maku has been shown to have outstanding characteristics for forest grazing. Treating pasture or strip spraying with a herbicide before planting sturdy, well conditioned seedlings gave increased tree growth and survival in agroforestry situations. Once trees reach 1 to 2 m in height they are relatively safe from sheep. Dairy cattle were found to cause unacceptable levels of damage and so are generally not grazed amongst young trees. Debarking by cattle and sheep is usually associated with overgrazing, but can occur unpredictably. An alternative to grazing during the tree establishment phase involves inter-row cropping for hay and silage, with trees being planted in rows 5 to 15 m apart. The use of protective fencing or repellents is uncommon, and livestock are usually excluded for the first 12 to 18 months as a simple means of avoiding damage. Torres (1983) suggests an alternative, to prevent early damage to trees by animals, would be to establish trees in combination with food crops which are substituted by pastures when trees are out of reach of animals. He recommends that cattle should not be introduced until trees are more than 4 m high. In early trials set up to measure the effects of the tree crop on the understorey, the major factor limiting livestock numbers under trees up to 8 years was often slash rather than canopy shading. With improved (P. radiata) genetic material now available, instead of the earlier ratios of 4:1 and 5:1, ratios of planted stocking/final crop stocking of 3:1 for seedlings and 2:1 for rooted cuttings are used resulting in a reduction in the amount of slash produced and a reduction in pasture covered by slash.

The recommendation on silvicultural management calls for planting pines in improved pasture at 2 × 5 m spacing (1,000 trees ha-1); thin at age 5 to 500 trees ha-1 and prune the remaining trees to 30% of height; prune trees to 40% tree height at age 8; thin to 200 trees at age 12 and prune remaining trees to 50% of height; and finally clearfell trees at age 25 (Knowles et al., 1973).

A major trial was established in 1973 at Tikitere, near Rotorua with P. radiata planted on to ryegrass/white clover pasture at initial stocking rates of 250, 500, 1,000 and 2,000 stems ha-1, which were thinned to final crop stockings of 50, 100, 200 and 400 stems ha-1 by 8 years (see Table 248 and Figure 231). Open pasture control plots were included. Data from Tikitere and elsewhere were used to develop a model relating pasture yields to tree size. The equation shows the relationship between the yield of understorey pasture (expressed as a percentage of open pasture yield) and two readily measured forest parameters, green crown length ha-1 (GCL) and mean green crown length (MCL).

Relative pasture yield = 24.8582 + 100.951e-0.00026138GCL - 16.3029 loge MCL

This is expressed graphically in Figure 232 and Table 249. With the exception of the 50 stems ha-1 regime, there was a clear pattern of decreasing pasture yields with increasing tree stocking and age. The effects on pasture production of different spatial arrangements of the 100 stems ha-1 treatment is discussed in section 10.6.2. Rye grass and white clover contents of pastures declined with stand age (and density).

Table 248. - Arrangement of Pinus radiata on the Tikitere trial (After Percival et al., 1984d)

Final crop tree density
(stems ha-1)
 Initial tree density
(stems ha-1)
Initial spacing (m)Average final spacing
Between rowsWithin rows
  50..25014.22.814 × 14
100..5007.12.810 × 10
  100*..5003.5/24.81.428 × 5
200..10007.11.47 × 7
400..20003.51.45 × 5

* Two rows 3.5 m apart, then 24.8 m to the next two rows.

Figure 231

Figure 231. - Pinus radiata at 50 stems ha-1 on the Tikitere trial in 1991.

Livestock numbers carried on understorey pasture are sensitive to both tree size (age) and density. An illustration of the effect of tree age and density on livestock (sheep) carrying capacity of pastures under P. radiata is shown in Figure 233. The more pines ha-1 and the older the pines, then the lower the carrying capacity. Also, the more pines ha-1 then the shorter the agricultural phase.

If it is assumed that livestock carrying capacity at the site was 10 breeding ewe equivalents ha-1 then over 28 years an unplanted hectare would yield 280 sheep grazing years. Table 250 shows how this figure would be reduced by the various tree stocking rates. Because the understorey grazing yields are concentrated in the first half of the rotation, discounting at different interest rates changes these proportions. This is shown in Table 251.

Table 249. - The effects of stocking and age of radiata pine on annual pasture dry matter production expressed as a percentage of that from open pasture (i.e. 100 = no trees) (Hawke, 1991).

Tree age
Final tree stocking (stems ha-1)SED
  3116  92  92  8010.3
  6  84  75  68  -  6.3
  8  92  89  72  53  5.3
10  -  -  -  12  3.7
13  85  50  18  -  5.2
15  -  32  -  -  4.4

Figure 232

Figure 232. - Pasture yield/tree growth relationship from a number of sites in the Central North Island, New Zealand (After Knowles and West, 1984).

Figure 233

Figure 233. - Effect of tree stocking and age on livestock carrying capacity (after Hawke et al., 1984).

Figure 234

Figure 234. - Pinus radiata at 100 stems ha-1 in a hedge row system (28 m × 5 m) at the Tikitere trial in 1991.

Table 250. - Predicted grazing availability over a tree rotation of 28 years (in sheep grazing years) (After Percival and Knowles, 1988)

 UnplantedFinal Crop Stocking (Stems ha-1)
Grazing years ha-128077       51       38       
Proportion of unplanted

Table 251. - Discounted grazing availability over a tree rotation of 28 years (as a proportion of unplanted land). (After Percival and Knowles, 1988)

Discount rate
UnplantedFinal Crop Stocking (Stems ha-1)

An interesting aspect of the Tikitere trial were the two 100 stems ha-1 treatments, one with an average final spacing of 10 × 10 m and the other a hedgerow system (see Figure 234) with an average final spacing of 28 × 5 m, which provided ample grazing areas between the twin rows of P. radiata. The twin row (hedgerow) treatment reduced the decline in pasture production compared with even spacing (see Table 252) although the trade off was a smaller tree diameter in the double rows! This planting system will be referred to again later in this Chapter (see Section 10.6.2).

Table 252. - The effects of spacial arrangement of radiata pine on pasture dry matter production expressed as a percentage of that from open pasture (i.e. 100 = no trees). (Hawke, 1991).

Tree ageFinal tree stocking (stems ha-1)
100 even spacing100 twin rows

Previous Page Top of Page Next Page