Version classiqueVersion mobile

L’énergie dans le développement de la Nouvelle-Calédonie

Yves Le Bars
Elsa Faugère
Philippe Menanteau
et al.

Emissions and the reduction of greenhouse gas emissions in New Caledonia

Variations in coal stores in the vegetation of New Caledonia

Texte intégral

1The vegetation and afforestation of agricultural land currently not forested presents potential to increase carbon stores which, under certain conditions and within certain restrictions, could offset some of the emissions which result from the use of fossil fuels and deforestation.

2Aspects associated with the use of land, changes to land and to forests are classed as a specific “sector of activity” under the climate change convention and the Kyoto protocol.


The government’s and the Provinces’ respective competences with respect to forestation issues

3New Caledonia has a large area of public forest which extends across the best part of 11 900 km2 and which is made up of the “private domain” of the public communities. This domain of communities represented (in 2001) 64% of the overall surface area of New Caledonia, with the balance made up 19% in private ownership and 17% customary land (inalienable, nontransferable, incommutable and non-seizable). Management of the domain, which belongs to New Caledonia, is granted to the government (article 127 of the Organic Act) which stipulates that the domain must be managed “in compliance with existing regulations, and in particular provincial regulations in the domains for which the Provinces are responsible, and in the general interest”.

4When the system of Provinces was introduced (1989), the forest service serving Grande Terre was split out into a “forestry service” in North Province (integrated into the Department of Economic Development and the Environment, DDEE), and a “forest and timber” division within the South Province plant production and forest division (integrated with the Department of Rural Development, DDR). The government did not maintain a forestry service in so much as virtually the entire area of public forest remained its property. The Provinces have been invested with regulatory competences, giving them the power to enact standards which apply across the entire Provincial territory. Their designated agents have the power to record offences committed on the private domain of New Caledonia. In practice, there are so few such agents that there is hardly any real surveillance.

5Current reforestation activities are carried out by the Provinces, generally on New Caledonian land which then involves transfer of the ground concerned. More than 7 000 ha of formerly reforested plots of land are located in the domain of New Caledonia, most of them in the North Province, but with some also in the South Province.

6Areas of natural forest are worked by private companies using a system of Temporary Logging Permits (PTE), issued by the competent provincial department. This involves 0.46% of the domain of New Caledonia. Smaller scale logging, carried out by individuals, requires a timber chopping permit issued by the provincial departments. That fact that few requests have been submitted indicates that, in practice, activities are not heavily regulated.

7More than a quarter of the private domain of New Caledonia, or 86% of the surface area across ultrabasic land already suffers under the strain of mining concessions. This raises the question as to how decisions are taken, and in particular the respective roles of the public communities and those who work the land when it come to future choices regarding land clearance.

The use of rural spaces: the extensive livestock breeding which occupies the majority of the land in use is in decline

8Topography and soil quality restrictions form a heavy burden on agriculture. The best land has been given over to the cultivation of trees and market gardening. In 2002, when the previous agricultural census took place, the usable agricultural area (UAA) accounted for 13.5% of the country’s entire surface area, while the maintained agricultural area (MAA) represented 57% of the UAA, which leaves a difference of almost 110 000 ha of land that is not utilised or is poorly maintained, mostly made up of unproductive pasture land. Livestock breeding, which is more extensive (0.4 LU/ha on average), occupies 90% of the UAA, although this sector is in decline. This could lead to an increase in meat imports, shifting the associated methane emissions to another country. It should be noted that the UAA changed from 12% in 1991 to 13.5% in 2002 of the country’s total surface area (or 20 000 ha according to the agricultural census); the MAA maintained fell by more than 15%.

9The price of private land seems to be getting more exorbitant (1 to 3 million PF/ha) when starting new agricultural activities, even far from Greater Nouméa, and neither is there any mechanism to main agricultural land prices, with land often sold as part of a succession plan and given the absence of an agricultural pension plan. The land reorganisation body, the Group Operations for Land Management (Ogaf) continues, to be managed by state departments (the State Department for Agriculture, Forestry and the Environment, Dafe); in New Caledonia, these are small-scale operations which have never involved planting a final forest or biomass production. In 2002, the total workable surface area, not including the UAA, was more than 17 000 ha of non-productive fallow and almost 18 000 ha of “natural forest”; there was just 632 ha of land that had been reforested, most of it no doubt completed prior to 1990. So areas of savanna or degraded forest could no doubt in theory be forested. This could result in increased carbon stores in forests and in agricultural land. The issue of land use thus merits clarification by the local research and technical support institutes, such as the IAC. Should land be forested or should local agricultural production be increased? The response to this complex question will require a long-term study, new knowledge to be acquired and previous knowledge such as in the areas of agronomy and forestry to be resurrected. It would be dangerous to claim to be able to respond solely on the basis of existing information within this evaluation.

Forest areas and resources

10The definition used for implementation of the Kyoto protocol includes three simultaneous criteria: the minimum surface area of the formation classified (between 0.05 and 1 ha), the minimum coverage of adult lignous vegetation (between 10% or 30% of the surface area) and the minimal potential height of the plants at maturity (between 2 m to 5 m). This last reference ensures that all young lignous plants subsequently encouraged to grow can be classified as forest.

  • 1 Melaleuca quiquenerva (myrtle).

11In New Caledonia, unfortunately, the Territory’s only comprehensive aerial cover dates back to 1954! Taking this as the basis, the general forest inventory activities carried out in 1975 produced a map of plant formations on Grande Terre and some of the islands which distinguishes between the following formations: dense evergreen forest (23%), “ Niaoulis1“ formations (14%), other forest formations (1%), scrubland (25%), thicket (8%), savanna (22%). These percentages relate to Grande Terre (they change only marginally if the islands are also taken into account) and exclude agricultural and inhabited areas. At the time, agricultural areas and inhabited areas accounted for 5% and 2% of the surface area respectively. Note that in 2002, a maintained agricultural area (MAA) of around 7.5% was achieved which can probably be compared with the 5% referred to at the time.

12The Atlas of New Caledonia (éditions IRD, 1981) includes a map – currently being updated – of the main types of vegetation. The DTSI (a territory department) whose mission is to centralise and integrate all map information, has just completed a map due for publication which shows land coverage and the main types of vegetation. It distinguishes between the following types of vegetation (the percentages are as a percentage of the entire territory): dense forest on sedimentary volcanic ground or on ultrabasic ground (22% and 9%, respectively), dense para-forest scrubland (8%), thick vegetation on ultrabasic ground (4%), lingo-herbaceous scrubland (9%), bush vegetation on sedimentary volcanic ground (14%), savanna (22%).

13The different methodology and system of naming used make comparison with the mapping completed in 1975 impossible. Indeed, the continuums between dense forest and degraded forest, between open forest and savanna, and between scrubland and different heights of thicket make any attempt to find some coherence intrinsically difficult, and would require a unique, robust and repeatable method of mapping to be developed and implemented in order to monitor vegetation, forested areas and the biomass multiple times.


The principal features of the physical environment

14Grande Terre is divided along its entire length by a central chain of mountains (average altitude 1 100 m) which end in the north with Mont Panié at 1 628 m and in the south with Mont Humboldt. Secondary mountain chains spread out along perpendicular axes running more or less east to west. The western slopes stretch out into wide coastal plains while the eastern slopes are sheer, and cut into by deep valleys. The south of New Caledonia is a vast plateau of ultrabasic rocks (average altitude 250 m). We must emphasise that in the majority of New Caledonia’s national forest areas, steep slopes are a powerful constraint to gathering timber, and more so to harvesting biomass where the value per unit of weight is even lower than that of timber. A distinction should thus be made between land where there is some hope of harvesting biomass, and land where it will be possible only to increase carbon stores provided these can then be protected, particularly against fire.

The factors which determine variations in natural vegetation

15Three factors explain the main natural variations in the vegetation of New Caledonia, and of Grande Terre in particular: the divergence between the west coast (arid) and the east coast (humid), the variations in altitude and differences in geological substrate. Variations in precipitations arise both from the contrast between the east and west and the gradient of the altitude. In more elevated areas, some of the precipitation manifests itself as mist.

16The impact of the geology is apparent in the major contrast between ultabasic rocks (also known as ultramafic or “peridotite”) and acidic rocks of a sedimentary volcanic nature. The ultrabasic rocks occupy the southern part of Grande Terre, forming the “grand massif du sud” which culminates with Mont Humboldt. These rocks are found in specific masses in the central region and along the western coast of Grande Terre. The soils derived from these rocks – ferralitic ferritic – are exceptionally low in nutriments (Ca, P) and, conversely, exceptionally rich in certain metals (chrome, manganese, nickel). This is where the soils derive their unusual name of “mine land” from. The Loyalty Islands, which are relatively flat, have a characteristic substrate (coral limestone, madreporic plateau) and a karst relief.

17These natural factors which affect the variety of vegetation have been interacting for at least 4 000 years coupled with the impact – direct or indirect – of human activities which changed and increased when the Europeans arrived. Among the indirect anthropic effect, we can count the introduction of several plant and animal species. Of these, there is a surplus population of rusa deer which means they form a systemic constraint on vegetation. In the same train of thought, we should also mention feral livestock, and the high number of pigs.

18Fires, most of which are anthropic in origin, are no doubt the source of human impact and their effects are among the most widespread in the area.

Biodiversity exceptional in its originality

19New Caledonia is known throughout the word as one of the “hot spots” of biodiversity, a title which expresses both the richness and originality (endemism) of the biological forms, but also the pressures and risks that weight upon the ecosystems concerned. The vulnerability of the species is increased in an island environment, and the planet’s small islands are thought to shelter around a sixth of the world’s threatened plant species. This exceptional biodiversity, a heritage from the island’s geological history, requires that special attention be paid to how these forest vegetation assets are managed. This concern has resulted in concrete terms in reserves being established, particularly during the course of the past two decades. However, this also requires “good practice” to be applied to the management of the entire region, without which these reserves could prove ineffective. It should be emphasised at this point that variations in carbon stores in the “nature reserves” do not have to be recorded under the Kyoto protocol.

  • 2 Synonyms: Albizzia falcataria, Falcataria moluccanna.

20Among the hereditary duties associated with New Caledonia’s diversity come vigilance with regard to the risk of a proliferation of species introduces into natural or semi-natural environments as defined by the concept of “biological invasion”. This duty may sometimes come into conflict with forestation aims designed to facilitate the economically viable production of biomass or to increase carbon stores. In fact, some species whose capacity to increase their numbers very rapidly is recognised in the islands of Asia and in the Pacific have been classified there as alien species that are outside their region of origin according to the International Union for Conservation of Nature (IUCN) database. These species include Paraserienthes falcataria2 and Pinus Caribaea, although this latter has been important in the afforestation of woodland in New Caledonia.

The principal types of vegetation

Dense, humid, evergreen forests

21The slopes of the central mountain chain are still covered with large swathes of dense, humid, evergreen forest found primarily on acidic rock (volcanic sedimentary) which accounts for around 300 000 ha, and on ultra-basic rocks which makes up 100 000 ha. Despite the variations due to altitude, this forest does have certain constants such as its closed canopy at an average height of 20 m made up primarily of evergreen leaf species which give rise to a more limited number of species (araucarias, kaoris, Houp). Another constant is the abundance of palms, pandanus and ferns. The specific richness and level of endemism are high in all areas, but particularly marked on the ultrabasic substrate and at altitude.

22At an average altitude (300 to 1 000 metres), forests are found predominantly on the western slopes of the central mountain chain where they are better protected against wind and cyclones; most of them are located in the North Province. Above 1 000 m, the forest is characterised by the omnipresence of moss and lichens, a lower canopy (rarely exceeding a height of 10 m), a more marked presence of ferns and epiphytes, and by the existence of soils (mor or ranker) with a thick superficial organic horizon often greater than 20 cm. Forests at low altitude are a rarity today, and are restricted to the Blue River reserve where they are found on an ultrabasic substrate, to a few very dispersed forests in the valleys of the central mountain chain; all that remains of the “gum oak” forests, which once extended to the east coast at low altitude are a few limited trees.

23Precipitation ranges from 1 500 to 3 500 mm per annum at an average altitude, rising to 4 000 mm above 1 000 m, with some records (8-10 m) on the eastern slopes close to the ridges. These mountain forests thus play an important role in hydrology management in that they reduce the risks of erosion, in particular through their high soil absorption capacity; they increase the available water resource by trapping the humidity from mist.

24On ultramafic rocks, the dense, evergreen forest is characterised by specialist species, but the effect of the substrate is not necessarily seen in the physionomy of the vegetation in the absence of human disruption: the canopy is at a modest height, though reasonably dense. A large proportion of these forests has suffered as a result of mining coupled with fires, particularly around the summits along the west coast.

Low formations on ultrabasic rocks (“maquis miniers”)

25This is a low formation in its natural state with no degradation, and with specific orophilic features close to the summits. This ‘maquis’ scrubland is made up of a herbaceous stratum of Cyperales and a stratum of trees or shrubs sometimes exceeding 5 m in height plus a few araucarias and kaoris. Depending on the relative importance of each component, a distinction can be made between herbaceous, bush, or tree features. According to the 1975 survey, these formations occupy 450 000 ha.

The sclerophyllous forests

26This vegetation, also known as “arid forest”, probably once stretched across the entire west coast of Grande Terre at low and medium altitude (below 300 m) in areas with rainfall of less than 1 m/annum. These forests are made up of evergreen trees and bushes which have small leaves with cuticles (sclerophyllous) that have adapted to the dryness. The trees are small (15 m high and 40 cm in diameter at the biggest) and the forest floor may be made up of dense or less dense thicket with few herbaceous features. The richness of taxonomy does not match that of the dense, humid forest, but the arid forest’s flora is deeply unique, with several endemic species (10.4% of the country’s endemic species). In common with almost everywhere in the tropical world, the most arid vegetation formations have suffered most from human activities, fire and the pressure of cervidaes. It is estimated that natural arid forests, which could today occupy and area of 4 000 ha, take up only 1 to 2% of their original natural area. They fall within a specific protection programme.

Savanna and forms of degradation

27Savanna is found essentially along the eastern coast of Grande Terre in the driest climatic conditions; the savanna could undoubtedly be interpreted as a form of degradation of the sclerophyllous forest given the effects of logging, fires and grazing livestock and deer. Niaouli is, generally speaking, the main wooded area, although its local abundance is often described as being associated with the swampy part of the savanna. Another little documented aspect, although one that is important in understanding the system of fires, relates to the nature and composition of the herbaceous stratum. The 1975 survey recorded 230 000 ha of savanna-type formations “in Niaouli”, 370 000 ha of “savanna” and 220 000 ha of “thicket” (of which 36% on the islands).

Containing fires

28While deliberate burning is an ancient and modern practice throughout human society, it becomes particularly relevant in a tropical context where the climate has highly contrasting dry and rainy seasons. The very high quantities of biomass, in particular herbaceous, produced in the rainy season are far greater than those found in Mediterranean or temperate regions. Once it becomes dried out with the end of the rains, or if the rains become less frequent, herbaceous biomass or sub-lignous forms an ideal fuel. The reasons for starting fires are often the desire to “rid” the ground of biomass deemed to be a nuisance (obstacle, contains insects or snakes, biomass is dangerous if a fire is started accidentally). Cultivation and hunting are also often cited as justification. In any case, the areas actually burned sometimes largely exceed the initial intention when the fire was started. The ability to start fire is also open to malevolence such as in the case of disputes regarding land or other various conflicts. Proportional land ownership or rights of use, either implicit or contested, thus need to be defined. Such motivations and behaviour are not specific to New Caledonia.

29Areas burned were as much as 50 000 ha in some years, with an annual average of 20 000 ha, although there is no real annual balance of opinion. Attempts at mapping using Modis satellite images which show active fires rather than areas already burned, do not agree with the estimates issued by the civil protection agency. All estimates of surface area need to be relative to the plant formations affected because the consequences of a fire can be very different in ecological terms as well as in terms of the carbon released. A fire prevention research programme, in which forest managers and the WWF are participating, is underway starting with an analysis of why fires are started.

An initial attempt to estimate carbon stores in the vegetation of New Caledonia (according to Durrieu de Madron, 2009)

  • 3 In France, forest biomass achieves an average of around 69 t of carbon, or, approximately, 120 t of (...)

30The huge lack of specific information in New Caledonia has forced the author to make several approximations or extrapolations from data derived from the global literature. The results obtained should therefore be approached with caution and used only to provide some order of scale. Note that the values estimated for the dense forests are relatively high given that their canopies are relatively low (rarely above 25 m) which is a feature of the forests of New Caledonia. We could delve deeper into this question, but that would require a more specific study as well as data not currently available. In any case, stores of dry matter are certainly between 180 to 190 t/ha (for above ground biomass as an upper fork, corresponding to forests in a good state, usually not worked)3. These values do not take into account the areas used worked since the reference survey, which looked primarily forests with a relatively high areas of biomass.

31Of the forested areas described in the table, the dense forests and savanna of Niaoulis represent by far the largest areas at 59% and 36% respectively in the 1975 mapping activity. There is still huge uncertainty regarding the areas of savanna given their blurred definition and great variations in the density of their stratum of trees.

32The signatories to the climate change Convention undertake to record emissions by source, i.e. whether they come from forests, and absorption by sink in order to contribute to the census of greenhouse gas emissions by country. This assumes the ability to be able to measure variations in stocks in an acceptable manner. This initial information is thus completely insufficient to determine such variations in New Caledonia’s forests with any precision. It is currently not even possible to provide a precise estimate of total current stores of carbon in New Caledonia’s forest ecosystems.

Table 6 – Order of scale - obtained by extrapolation and not through direct measurement - of the biomass and carbon stores in the main wooded formations of New Caledonia

Table 6 – Order of scale - obtained by extrapolation and not through direct measurement - of the biomass and carbon stores in the main wooded formations of New Caledonia

Note * But large variation in density of trees, thus biomass.
Note ** On the islands.
Note *** Less than 20 years in most cases.
Note **** Extrapolated directly from Wallis’ estimate.


Reforestation and new vegetation promotes an increase in stores

Reforestation of forests

33New Caledonia’s reforestation studies and programmes date back to the 1970 when tests were conducted with exotic species (eucalyptus, tropical pines), as well as indigenous species (kaoris, araucarias, gum oak); it proved possible to plant these native species on the ultrabasic substrate. Of the species introduced, Pinus caribaea (var. Hondurensis) and, to a lesser degree, Pinus elliottii have proven to be the most suitable, and have been the most used. Most of this “historical” reforestation (1970-1980) with pines took place in the North Province. Of the 4 600 ha completed, only 2 800 ha, primarily in the Tango sector, was in an acceptable state or sufficiently accessible to enable timber to be gathered. The conditions for gathering timber often proved to be more difficult that he planters estimated!

34Since provincialisation, the provincial departments have been responsible for monitoring plantations located mainly in New Caledonia’s land area as well as for new afforestation projects. Within the two Provinces, the key objective is to put forward rural activities and, if possible, provide products for use within the domestic market. New Caledonia is largely deficient in forestry products, in particular timber and logs. It is estimated that the percentage of needs met has dropped recently from 30% to 20%, principally due to the increase in demand. Imports come mainly from New Zealand (Pinus raddiata in particular).

  • In South Province, the DDR has adopted an ambitious policy, and estimates that 1 300 ha of forest land was reforested between 1989-1990. It would appear that the main thing with this reforestation is that it was carried out on public, non-agricultural land and is thus not included under article 3.3 of the protocol. Reforestation in the south of the Province using Pinus Caribaea var. Hondurensis has now ceased given the species’ tendency to spread in the maquis minier and increase the risk of fire there. Elsewhere, the DDR has led vegetation extraction programmes around certain of its plantations, in “areas of ecological interest”. At the same time, the DDR is ensuring that reforestation of “naturalised” Pinus on the Isle of Pines is managed. A total of 1 700 ha (of which 10-20 % araucariacae, relatively young) has been planted in South Province. The rotation period for pine is around 30 years, about a century for kaoris and 60 years for araucarias.

35Current and future reforestation is focused more on local species (kaoris, araucarias), despite lingering technical uncertainties. The system of supplying seeds to private nurseries, fully guaranteed by the IAC until now, was deemed unsatisfactory with the recruitment of an Inra (French National Institute for Agricultural Research) engineer seconded by the IAC to support the private nurseries. Some effect is expected from the new provisions of the investment code which expects to see up to 80% aid or private reforestation.

  • In North Province, reforestation is of the order of 15 ha par annum, although it may have achieved 30 ha per annum a few years ago. The total planted in forest areas since 1990 is of the order of 300 to 600 ha. In absolute terms, there seems to be no limit currently to suitable areas, certainly less so than in the South Province. However, setting up projects on customary land (and elsewhere) is a slow process which involves negotiation. Reforestation is concerned with five to six species which are subject to economic evaluation (timber in particular), including araucaria, faux-tamanou and gaiac. After a few years, areas that are sufficiently maintained and protected to provide viable vegetation – i.e. where the standing biomass substantially exceeds spontaneously degraded vegetation – again appear to account for less than half the afforested area (150 to 300 ha).

36Privately-owned land in New Caledonia with naturally favourable properties in terms of usage would seem to amount to less than 1 000 ha in the South Province. In the North Province, there could be more extensive areas suitable for reforestation, although Hygen et al. (2006) suggest even these do not exceed a few thousand hectares. (2006).

37The potential for reforestation of private land is today once more encouraged. However, as past experience has shown, the success of any planting depends on where fire breaks out, deer grazing, land issues, and, of course, when it comes down to promoting specific species to plant, the economic return.

38The true prospects for economically and socially acceptable forestation and reforestation thus seem restricted in the light of the information available at the time of this evaluation. From a technical point of view, it is difficult to envisage large-scale reforestation programmes other than with the benefit of new insights into potential forest production on the basis of long-term experiments in various situations which are representative of the conditions in New Caledonia, the option of automating harvesting or forestation etc.), and, finally, on the basis of the rural socio-economic situation in New Caledonia. These activities will need to be carried out locally by a team with a given track record in such areas. The technical knowledge validated locally indeed seemed to be lacking during this evaluation, despite earlier work undertaken, and renewed interest in the Provinces’ services. This fits with the questions formulated earlier regarding the allocation of land between agricultural production (livestock in particular) and forestry.

Revegetation of mining sites

39Prior to recent adoption of the mining code, which uses fiscal incentives, revegetation of the mining sites relied on the good will of industrialists. A specific study conducted by the DTSI estimates that a total of 20 000 ha of the land area was damaged by mining activities – there is reference to “orphan” sites where the current user does not (any longer?) have any responsibility for the damaged land. Currently, land rehabilitation efforts scarcely exceed 100 ha per annum across the whole of New Caledonia. This could soon reach 200 ha (it could take a century to revegetate these sites!). In any case, the small increase in non-alien species seriously restricts the increase in carbon stores than can be expected on this land. In addition, recourse to enrichment would risk encouraging alien species. In the best case scenario, it would be possible only to slowly restore carbon stores to where they were prior to mining activity. Even if this is successful – of which there is no guarantee and the land will remain fragile for a long time – reconstituting the biomass will be a slow process.

The causes of declining stores

Land clearing and urbanisation

40We have seen agricultural land use increase by 20 000 ha between 1991 and 2002, which is no doubt greater than the increase in dwelling and infrastructures for which we have no total figure. In both cases, land clearance was probably to the detriment of the savanna, or degraded rather than dense forests, although this cannot be confirmed.

Forest use

41In the natural forest, use has always been selective, with four species representing 80% of output (tamanou, houp, kaoris, “beech”). It is estimated that, since the Second World War, forest has covered 40 000 to 50 000 ha giving selective sampling of 15 to 20 m3/ha of timber (volume of trunks). The decrease in biomass stores per hectare of natural forest in recent decades could be of the order of 50 t of dry mater (25 t of carbon). However there are no real measures available.

42In the 1980s, 700 to 900 ha were used annually, giving a total output of around 14 000 m3 of timber. It could be concluded that, since assessment which enabled the value of standing biomass to be evaluated (cf. table 5), some 20 000 ha of forest have been worked and that, across this area, the reference values mentioned in relation to dense forest have decreased substantially by around 25%. However, these are approximations only.

43The system of Temporary Logging Permits (PTEs) has been unable to guarantee the sustainability of logging, and concerns have been expressed in regard to the resilience of the ecosystems in the Caledonian mountains. In fact, none of the sites worked over the past 30-40 years has been logged again. The postlogging assessment conducted by the Aoupinié pilot site (North Province) reached some rather pessimistic conclusions, despite the level of preparation and monitoring being higher than current practice. This study, conducted in 1996, confirms that at the average rate of logging for New Caledonia, there would be a decline in the health of the trees reserved, these trees would fail to increase significantly in volume after logging, there would be an invasion of heliophilous species (23% of the land area) which would block the regeneration of desirable species, and net signs of erosion on the paths (6% of the land area). This awareness, added to the socio-economic context (despite the fact that the work was subsidised) leads to an apparently inexorable decline in activity. In the South Province, moreover, logging of the natural forest will cease in 2012.

44The effect on variations in carbon stores of harvesting from the natural forest is thus becoming negligible. Reconstituting the biomass in zones worked previously could result in an increase in their stores.


45Changes to the El Niñño system in the Pacific region could also increase the frequency of dry periods and years, which could increase the risk of fire. Ongoing research (The National Research Agency (ANR) INC project) should deliver better understanding of the causes and risks of fire.

The expansion of mining areas

46In the ultrabasic geological zone, which is prized for its nickel, much of the surviving forest land has suffered under the strain of mining concessions (around 30% of the forest area in New Caledonia). Without more precise information about the pattern of land clearance under these concessions, we can assume that the 20 000 ha degraded by mining were degraded over the past 50 years, creating an annual pattern of around 400 ha.


47We assume, given its revenue per habitant, that New Caledonia is deemed to be an “industrialised” annex 1 country. The reasoning that follows assumes that New Caledonia will adjust to comply with the requirements of the Convention, and in particular of the Kyoto protocol.

48Article 4.1a of the Convention regarding inventories of greenhouse gas emissions states that all emissions must be registered by source and all removals by sinks. In the context of France, these variations in carbon stores in the forests of New Caledonian would be low. This is, however, not a “key” source where better understanding is a priority when we consider these emissions across the whole of “France plus the overseas territories (DOM), plus Territorial Overseas Communities (COM) and plus New Caledonia”. These should, however, be recorded more precisely if emissions from New Caledonia are to be considered separately.

Under the Kyoto protocol

49A distinction needs to be made here between variations in stores under article 3.3 (which become compulsory from the initial period of engagement), and variations under 3.4. Under article 3.4, France retained only the forest section, but application was optional and with variable geometry. Accounting is also specific.

Record of variations in stores in relation to article 3.3 of the protocol

50Article 3.3:

51“Net variations in greenhouse gas emissions by sources and absorption by sinks resulting from human activity associated with a change in land use and forestry and restricted to afforestation, reforestation and deforestation since 1990, which correspond with verifiable changes in carbon stores during each period of engagement are used by the annex I parties (by industrialised countries) to meet their commitments (the limitation and reduction of greenhouse gas emissions) stipulated under this article. Greenhouse gas emissions by sources and ‘absorption by sinks associated with these activities must be notified in a transparent and verifiable way and examined in accordance with articles 7 and 8.”

52All annex I countries which have made commitments under the Kyoto protocol are obliged to submit such a carbon balance sheet for the period between 2008-2012. Increases in forestation on non-forest land since 1990 which are of direct human origin may be counted.

  • At the opening of the balance sheet, as defined by article 3.3, the decrease in carbon stores between 2008 and 2012 as the result of urban expansion of the creation of infrastructures, land clearance for agricultural reasons and any expansion of mining areas must be noted.

  • The credits side of the balance sheet must show increases in carbon stores, but only between 2008 and 2012, increases in forestation on agricultural land since 1990, and in afforestation of mining land which was no longer deemed to be forest in 1990 (i.e. land that was cleared and not reforested prior to 1990). It should be noted that most reforestation completed since 1990 was undoubtedly on public, non-agricultural land, even if the vegetation was, probably, degraded. Such forestation or reforestation cannot thus be considered under article 3.3.

53The reference date for the official allocation of land is 1990 which poses a fundamental problem in New Caledonia given the lack of reference information regarding land allocation at that time. Detailed examination of the results of the agricultural census could be useful. It would also need to be demonstrated that whatever has become covered with vegetation is not the result of the land being abandoned, but of true forestation.

54Revegetation of mining land between 1990 and 2008 undoubtedly didn’t always succeed in vegetation that could be classified as forest. However this is of little importance if, over time, such vegetation meets the criteria of the FAO definition of forest. If, however, this increase in stock between 2008 and 2012 is too low to merit precise measurement, it can simply be ignored.

55In the case of New Caledonia, a negative balance sheet is probably to be expected (net decrease in stores accounted under article 3.3), principally due to the effects of agricultural, mining and urban land clearance. Carbon stores on land areas to be cleared between 2008 and 2012 (or during the subsequent commitment periods after 2012) will therefore need to be measured.

Record of variations in stores in relation to article 3.4 of the protocol

56Article 3.4:

57“Prior to the first session of the conference of the parties (……)each party included in annex I shall provide, for consideration by the Subsidiary Body for Scientific and Technical Advice, data to establish its level of carbon stocks in 1990 and to enable an estimate to be made of its changes in carbon stocks in subsequent years. At its first session, or as soon as practicable thereafter, the Conference of the Parties (……) decide upon modalities, rules and guidelines as to how, and which, additional human-induced activities related to changes in greenhouse gas emissions by sources and removals by sinks in the agricultural soils and the land-use change and forestry categories shall be added to, or subtracted from, the assigned amounts for Parties included in Annex I, taking into account uncertainties, transparency in reporting, verifiability, the methodological work of the Intergovernmental Panel on Climate Change, the advice provided by the Subsidiary Body for Scientific and Technological Advice in accordance with Article 5 and the decisions of the Conference of the Parties. Such a decision shall apply in the second and subsequent commitment periods. A Party may choose to apply such a decision on these additional human-induced activities for its first commitment period, provided that these activities have taken place since 1990.”

58This article defines the optional provisions through to the end of the first period of commitment, i.e. through to the end of 2012, after which they become restrictive. Compared with the previous article, these provisions offer annex I countries, from 2008, the possibility of benefiting from additional carbon credits (note that this is with certain restrictions only and with specific accounting; for clarification, please read “Emissions and the reduction of greenhouse gas emissions in New Caledonia” and “New Caledonia’s international integration” on the CD ROM). The balance sheet needs to be beneficial to the country, which is not always the case. This is why Canada has not continued to manage its forests under article 3.4, because the average carbon stores in its forests risk falling between 2008 and 2012.

59Only a small proportion of this increase in forest (amounting to around 15%) can be accounted under this initial period of commitment. We still do not know how this proportion will change after 2012. This option allows possible compensation for a negative carbon balance under the accounting system in article 3.3.

60It is highly possible, although not certain, that the growth of existing forests in New Caledonia over the period concerned will be positive given the output from earlier planting (before 1990) and, most importantly, the gradual slowing down of forest use. It may be assumed that the 50 000 to 60 000 ha worked over the past 50-60 years (almost 20% of the dense forest land area) is part of a strategy of reconstitution and slow recapitalisation of the biomass. This increase could also be cancelled out by the degradation of the dense for-est through processes which are not part of formal forest use. Given the lack of precise information, this provides food for thought only.

61To go further, it would be necessary to determine the changes in stocks in existing forests (outside 3.3) using an inventory procedure which complies with the requirements of the climate change convention and the Kyoto protocol.

62Article 3.4 is also concerned with the reduction of greenhouse gas emissions resulting in particular from the management and growth of forests not considered under article 3.3, cultivation and revegetation. These are not discussed here, but rather under “New Caledonia’s international integration” on the CD ROM.

The influence of the perimeter considered on defining the measures required between now and 2012

63Taken against France as a whole, the changes in stocks in New Caledonia are relatively minor. Conversely, when considered in isolation from France as a whole, the consequences under Kyoto look rather different. The measures applied thus need to be more precise, particularly to avoid debits under article 3.3 with effect from 2008 and, undoubtedly, from 2012 under article 3.4 if its adoption becomes compulsory as negotiations so far would suggest.

64The situation in New Caledonia is, at first sight, apparently paradoxical: its obligations are those of an annex I country whereas its system of information regarding land occupancy and, more precisely, forest resources, is not very well developed. This can be largely explained by the absence of true interest from industry in the forest. This does not prevent forests on the mainland which are scarcely profitable (in the Mediterranean or in the mountains) from being recorded by the IFN in the same way as afforestations which are clearly more productive. However the additional costs this entails cannot be ignored.

Perspectives post 2012 under the Kyoto protocol

65This requires a recording strategy that is consistent with the principles defined by the IPCC in 2006. However, the information required depends on the perimeters under consideration. Given New Caledonia’s limited experience in reporting activities, the general thinking of the GOFC-GOLD (2008) work, while not a “turnkey” solution, could provide a source of inspiration. The expertise of the National Forest Inventory (IFN) could certainly be sollicited on a regular basis during the period in which protocols for measuring and monitoring are being established and launched. In order to meet the forest and carbon stores inventory and monitoring requirements efficiently, an approach at two levels is preferable; this would automatically bring together observation by teledetection with measures on the ground to verify the teledetection approach (ground truth) along with additional measures. It should, however, be highlighted that teledetection approaches are not sufficient under article 3.3: it is also necessary to be able to determine in which areas stores are located, such as whether afforestations are on agricultural land or not.


The importance of mapping on monitoring

66It is important to define a naming system for the different types of vegetation which can be applied consistently in map form in space and time, and which suits any methods of the observation which could be used recurrently. The initial period of commitment under the protocol will require information to be available for a five-year period, between 2008 and 2012 for example, either directly or by interpolation between the records created prior to 2008 and in 2012.

67This naming system needs to incorporate the fundamental dichotomy between “forest” and “non-forest” according to the definition which will be officially adopted by New Caledonia, and allow for reasonable subdivisions within these two classifications to qualify changes in land allocation and quantify their consequences in terms of carbon capture and storage.

68The subdivisions under the “forest” classification will need to include strata of sampling relevant to the measuring activities on the ground which are intended to estimate the level of biomass and of carbon. Finally, there will be a need to evaluate “degradation” in the sense of carbon stores declining on land still classified as “forest” by means of changes in the land area between the mapping sub-classifications under “forestry” before the information about any degradation is refined with information from the ground.

The role of spatial, aerial or satellite observation: interest in and the limitations of different techniques and various sources of information

69Given the objectives, there is a fundamental dichotomy between spatial information that is medium resolution, high resolution (pixels of the order of 10-20 m or more) and very high resolution (pixels less than 4-5 m in size). These values also apply to multi-spectral optical images, i.e. those that pro-vide information in several channels of the visible spectrum and in one to two channels of the infra-red spectrum.

70Modern aerial photography, for example, as supplied by IGN from the 1970s onwards, has fine enough pixels to deliver very high resolution postscanning, whether final interpretation is visual or semi-automatic. Information from near infra-red technology, which is important in defining the characteristics of different types of vegetation, is, however, not always there: the photographs provided by IGN to the National Research Institute (IFN) include this information, whereas the IGN coverage of the coast of Guyana does provide access. In panchromatic mode, very high resolution (VHR) satellite images today permit a quality of visual interpretation approaching that achieved with aerial photography.

71According to GOFC-GOLD, costs of 0.1 to 0.5 $/km2 and of 2 to 40 $/km2 have been put forward for HR and VHR coverage respectively (or aerial photographs). There is thus a noticeable difference in cost. However, it would seem that HR information can not meet the needs and conditions of New Caledonia considered in isolation within a restricted area (18 000 km2), and of its very varied vegetation which is down to natural and human reasons. It should be noted, however, that the coast of Guyana, which has a land area similar to that of New Caledonia, is covered fairly regularly by aerial photography (IGN). It is not unfeasible for the Overseas Departments (DOM) and Territorial Overseas Communities (COM) to obtain VHR data as part of ten-year map update strategy.

72Looking ahead, there are some methods of directly estimating biomass on the basis of radar teledetection data, in particular using the Japanese Alos satellite radar from the Japanese space agency and the Alos Kyoto & Carbon initiative. With the range of values for above ground biomass (AGB) observed not substantially exceeding 200 t/ha, there is a belief that problems of saturation with this type of signal could perhaps cripple its application in New Caledonia. However, a prefeasibility study would be needed and this would then require some work to calibrate the results with the biomass measured on the ground.

Recurrent measures on the ground and monitoring the growth of trees

73If well managed, spatial observation can facilitate changes in land allocation to be monitored, and diachronic mapping of sub-classes within the “forest” classification. It cannot, in itself, enable carbon stores in these different classes (flux) to be measured, nor provide precise values of their average above ground biomass (stocks). For this, ground inventories need to be prepared according to a systematic sampling plan for the area which is repeated over time (inventory cycle).

74The sampling points enable interpretations derived from teledetection documents to be verified on the ground and, most importantly, provide an inventory of the trees present in a reference area (small area), a botanical identification and measure of them using a simple variable such as trunk diameter. This is a traditional approach to forest inventories which could take inspiration, with some adjustment to new aims, from the protocols used in the past in New Caledonia (1975 inventory, local inventories between 1980-1990). Visits to the sampling sites on the ground can also provide the opportunity to gather additional information about the vegetation and the environment.

75More stringent measures will need to be considered over a sub-group of locations on the ground in order to establish a system of allometric relations which will ultimately enable the total above ground biomass of a tree to be estimated based on the measurement of its diameter. Such relations do not currently exist for New Caledonia, but can be developed either directly or indirectly. Since these measures are destructive, requiring trees to be felled and cut up, it may make sense to benefit from the opportunity provided by the few logging companies still operating, even if this would probably mean rebalancing the sample strategy to provide better coverage of the variety of forests in New Caledonia. In the absence of a simple and operational measuring technique a posteriori the growth of tropical trees, characterising their growth requires that reference areas (“permanent growth sites”) be monitored in an ongoing basis. These must be located such that they cover the main forest formation classifications, including plantings.

76All this work, which is needed to monitor changes in carbon stores in the forest vegetation, can be managed only if a “forest resources” inventory cell is established, which does not currently exist. This cell could be made up of three to four permanent members to carry out the field work, and one or two members working closely with the DTSI for the mapping-related activities. In addition to complying with the commitments made under the climate change convention, the results obtained on a regular basis by an inventory cell could help with the management of forest resources led by the Provinces’ technical departments. As is the case on the mainland, they could also make a useful contribution to characterising habitats and conserving the biotopes.


77The increase in carbon stores in constructions is not, currently, included. However, changes in stores may be recorded under the Convention. It is, however, difficult to determine these stores with sufficient precision.


The quantities of carbon at play in the land and forest areas can not compensate for what will be emitted by the metal industry

78While it is difficult, without precise data, for New Caledonia, to indicate the carbon balance as defined by the Kyoto protocol, it would nevertheless seem sensible to set out the challenges in relation to other sources of green-house gas emissions in the Territory.

79Let’s assume that the balance of emissions under article 3.3 is equal to zero for the period of commitment under consideration, between 2008 and 2012 for example, or between 2012 and 2020. This leads back to the thinking that declining carbon stores due to forest clearance would be compensated by increases in carbon stores in agricultural forests established since 1990. This assumption is no doubt too optimistic given the low level of afforestation on agricultural land, but we will persist with it for the moment.

What would the contribution under article 3.4 be if New Caledonia was considered separately by 2012?

80Let’s assume that the average annual increase in forest in New Caledonia, where the ground is relatively poor, could reach 1 t of carbon per ha (2 t of dry matter/ha or 4 m3 of “saleable” wood), which is probably optimistic, allowing 3.67 t of CO2 per ha to be captured over a given number of years. Applying the accounting rules under article 3.4, that would result in 0.55 t of CO2 per ha. With nickel production emitting around 20 t of CO2 per tonne, around 36 ha of forest would be needed to compensate for the emissions generated for one tonne of nickel. Producing 200 000 t of nickel would thus require 7.2 million ha which clearly hugely exceeds the surface area of New Caledonia. If we assume that the accounting rules will change in 2012 and that 50% rather than 15% of the carbon stores in forests can be included (which is not very probable since this would, in many countries, endanger activities associated with wood cutting and conversion), a surface area of around 11 ha of for-est per tonne of nickel, or 4.4 million ha in total, would still be needed.

81In other words, even assuming such increases in carbon stores and very favourable methods of accounting, the forests of New Caledonia will be able to compensate for only a small proportion of the emissions caused by nickel production currently.

What would the contribution be if New Caledonia is considered as part of France by 2012?

82Since increases in carbon stores eligible under the article relating to forests are capped at an inclusive value which cannot be renegotiated (around 15% of the annual increases forecast in 2000 for the period 20082012, any additional increase in carbon stores in forests does not mean an additional carbon credit. Conversely, under article 3.3, debits would continue to be accounted.

Some recommendations regarding the relationship between forests and greenhouse gases in New Caledonia

Under the United Nations climate change convention

83Across France as a whole, including New Caledonia, it is France that pro-vides the information required to the United Nations.

84If inventories of greenhouse gases from New Caledonia had to be supplied for this territory only, priorities in terms of the measures to be stepped up as a priority could change and new, sometimes costly, measures could be imposed.

Under the Kyoto protocol

85For the moment, New Caledonia has no obligations under the Kyoto protcol. If New Caledonia was keen to adhere to the protocol via France, it would have to prepare the inventories imposed under article 3.3 and article 3.4, even if its contribution to changes in carbon stores on the greenhouse gas balance sheet remained minimal.

86Under article 3.3, it would no doubt be sufficient to record decreases in stores resulting from the deforestation of mining land, urbanism and increases in agricultural land. Increases in stocks due to the forestation of agricultural land could undoubtedly be left out. This would no doubt result in a balance sheet which would be a little less favourable than measuring everything, but it would certainly be easier and less expensive to put together.

87For article 3.4, it would be necessary to measure changes in carbon stores in forests remaining as forests, but given the fact that France has some margin (it does not use all the increases in stocks achieved in its forests) it would be sufficient to measure increases in stocks with a low sample rate. It would therefore be easy to demonstrate, even if the increase in carbon stores in New Caledonia was negligible, that this has no notable effect on France’s commitments, including New Caledonia. If stocks decrease, it would be necessary to be a little more precise, but primarily perhaps to determine rounded up values to properly demonstrate that the increase in stocks under 3.4 is still, in France, above value z (i.e. above the 15% increase in stocks by 2010 which was forecast in 2000). As long as the increase in stock remains above value z, it will change nothing on the final balance.

88Given that France has, since 2008, adopted the “forest” section of 3.4, it would undoubtedly be possible, unless the protocol is amended prior to 2012, to comply with articles 3.3 and 3.4 from 2012. It would be sufficient to simply recommend the measures which are a little more precise than currently be implemented with a view to gradually approaching the level of precision for measures as required by the United Nations.

89If New Caledonia wished to enter into commitments of a different nature, perhaps autonomously, as considered in one of the scenarios reviewed in “New Caledonia’s international integration”, it would undoubtedly need to adopt some more sophisticated means of measurement in order to evaluate its carbon stores, particularly in its forests.

90In conclusion, we must make the following recommendations:

  • conduct a prospective study into the use of land for agriculture and forest in a way which is theoretically possible and economically viable to ensure New Caledonia’s sustained development. This will involve identifying any potential increase in agricultural and forestry output for food and nonfood uses as well as the sites where carbon stores could be increased with a reasonable capacity to protect them. This question merits clarification by local industrialists, and could be drive, for example, by the IAC (Institut agronomique calédonien/Caledonian Agricultural Institute);

  • establish a forestry department to gather information and with the ability to issue orders for forest inventories to be conducted in accordance with the Convention’s and protocol’s requirements as well as the requirements of various other environmental conventions. This dedicated department would need to be given the right tools to conduct inventories by teledetection and field work;

  • our recommendations to ensure the management and sustainable development of forest resources and thus contribute in a modest way to improving New Caledonia’s greenhouse gas balance are:

  • to increase, if possible, harvestable timber planting activities at the end of the cycle: using local araucariacae, even if the reforested areas are not to be very extensive (a few thousand ha). To achieve this, an applied forest research cell will need to be rebuilt which will work closely with the Province departments; voluntary forest policies will also need to be drawn up.;

  • to reconstitute stocks in virtually degraded natural forests, if possible using regeneration methods that are low cost per hectare, for example direct sowing of pre-germinated seeds. This will involve the development of new technologies and thus research programmes fully focused on this objective;

  • to try to fight fires more effectively, in particular to protect dense for-est formations from the risks of degradation from the savanna or open formations. This is a tough task, particularly given the absence of any economic evaluation of the forests. Awareness thus needs to be developed of the populations that protect the forest since this is also part of the fight against cli-mate change.

91Implementing these recommendations will certainly not have a huge effect in changing the balance of CO2 emissions and storage, but it could have associated benefits in terms of land development and protecting the biodiversity of the water cycle, etc.

92Interest in biomass to produce energy is also highlighted under “New energy production and storage technology”.


1 Melaleuca quiquenerva (myrtle).

2 Synonyms: Albizzia falcataria, Falcataria moluccanna.

3 In France, forest biomass achieves an average of around 69 t of carbon, or, approximately, 120 t of dry matter. Reductions in stores of carbon in forests converted to cultures approach 85 t/ha, or around 312 t of CO2/ha.

Table des illustrations

Titre Table 6 – Order of scale - obtained by extrapolation and not through direct measurement - of the biomass and carbon stores in the main wooded formations of New Caledonia
Fichier image/jpeg, 199k

© IRD Éditions, 2010

Conditions d’utilisation :

Rechercher dans OpenEdition Search

Vous allez être redirigé vers OpenEdition Search