European Tropical Forest Research Network

Organisations - Institutions - Programmes

• How forest plantations affect the soil
• Dung or forest biomass as fuel - even plantation trees with a bad reputation help conserve soil fertility
• Mixed species plantations of eucalyptus and acacia: growth, nutrition and soil changes
• Influence of poplar short-rotation coppice on some soil properties
• The effect of exotic tree plantations in northern Thailand on soil properties
• The carbon cycling in the eucalypt plantations in Congo
• The impact of tropical industrial tree plantations on site nutrient status, site productivity and economic profitability
• The long term memory of soils - how Amazonian dark earths reflect past land use
• Mangroves in the red river estuary in Vietnam: a sink or source of nutrients?

HOW FOREST PLANTATIONS AFFECT THE SOIL
An introduction to the theme and why we need more research on it
By Gerhard Glatzel

Even the casual hiker notices changes in soil properties when he enters a forest from grassland. Usually he sees leaf litter, fallen from forest trees on the soil surface and the soil is likely to be softer to the impact of his boots. In a forest not much foliage is consumed by herbivores and more or less integer leaves are shed from the canopy, when their functional lifespan is over. These dead or dying leaves are depleted of nutrients due to re-translocation into the tree and are consumed on the forest floor by a well structured community of organisms, which are capable of utilizing the tree litter. Fallen branches or dead trees harbour many highly specialised consumers as well as their parasites and predators. The shady and humid interior of forests allows soil organisms to act at the soil surface much more often than in the harsh climate outside the forest. The humus layer and topsoil in forest is therefore generally more porous and open to infiltration of water. Less evident but by no means unimportant are the effects of tree roots. Frequently more than 50 percent of the photosynthetic carbon gain in the canopy is used for root processes, sustaining not only root growth but a diverse microbial community deep within the soil. Tree roots of large trees bear a remarkable static and dynamic load. When trees sway in the wind, the lower parts of the soil profile are compacted, while the upper layer is lifted and loosened. The border between compaction and loosening depends very much on the root architecture of the tree species in the forest. On slopes tree roots bind the soil and prevent erosion. Another feature of forest soils is arborturbation, the turnover of soil by uprooting of trees. It creates inhomogeneity, another important feature of forest soils in view of regeneration ecology and biodiversity.

For all these reasons forest soil science is a well recognised, special field and it is sufficient to refer to textbooks on forest soils to answer the question of how forests affect the soil in general. In the context of forest plantations however, there are gaps in knowledge on various scales.

As a consequence of carbon trading, ecosystem restoration and increased demand for specific and uniform wood and fibre tree plantations will most likely increase significantly world wide during the coming decades. In order to understand and quantify the effect of forest plantations, not only the aboveground biomass has to be taken into account, but also how these plantations affect the soil. With regard to carbon sequestration, there is a lack of information on the dynamics of carbon incorporation in the soil as well as carbon saturation and potential carbon release from soil, if the plantation is destroyed or if the climate changes. Recent work shows, that soils have a long carbon memory. The assumption that forest soils store more carbon than steppe or scrub soils has been challenged. The correlation of carbon storage potential of forest soils with basic soil properties such as clay content, pH or soil hydrology is by far not as well worked out as in arable land. A much broader spectrum of soil properties of soils potentially suitable for tree plantations adds to the problem. Tree species used in forest plantations are physiologically much more diverse than agricultural crops. Rooting depth and root distribution in the soil profile determine carbon input, in particular to deeper soil layers, soil water relations and mineral nutrition. Leaves of evergreen trees are often highly protected by hard to degrade and toxic substances to ensure long functioning in the canopy. Deposited to the forest floor they resist quick decomposition and become substrate for unique decomposition food webs and humification processes. Herbaceous understorey has to be taken into account too. It is controlled by light availability and the structure and chemical composition of the litter deposited from the canopy. In terms of its direct contribution to carbon sequestration, the understoreuzy may not be very significant. Indirectly it may have a large effect because of its role in nutrient cycling and soil biology.

In the planning of forest plantations many factors other than carbon sequestration in plant biomass and soil have to be considered. Soil erosion is not necessarily minimised in forest plantations. Seasonally dense canopies of deciduous trees may suppress herbaceous forest floor vegetation, making the topsoil susceptible to erosion in the dry season, in particular if litter has been reduced by prescribed burning or by collection for fuel by the local population. Water repellent litter layers or soil sealing by waxy substances may increase runoff. Mechanical site preparation, chemical weed control, nutrient exports from harvesting and stool renewal in coppiced plantations may significantly impact on soil properties. Many plantation tree species are capable of outgrowing the supply of mineral nutrients provided by the soil. Deficiency symptoms, retarded growth, susceptibility to abiotic and biotic stress and even dieback may surprise some years after planting, if the soil is unknown and has not been tested.

Land use and conservation deserve special attention, in particular in subtropical and tropical countries. Degraded land or badlands targeted for carbon plantation may be so classified from a forester's perspective or that of a company in the carbon trade business. From a villager's viewpoint the same land may be pasture and source of various plant or animal resources for household use. Establishing and protecting plantations, usually on the more suitable sites, will inevitably increase pressure on the remaining land if food and fuel are scarce. As plantations offer little in terms of biodiversity, the over all situation is likely to deteriorate, if forest remnants are more heavily exploited. In new plantations biomass accumulation on the forest floor may increase fire danger both from natural causes and from herders in need of pasture. Prescribed burning may be used to reduce the risk of damaging fires, but it requires skills on the local level and acceptance by the educated public because a role of fire in carbon sequestration is difficult to explain. The role of plantations in landscape hydrology has many facets. Well established forests are usually considered beneficial because of better infiltration of rain water and more even discharge. From the perspective of local users dried up wells as a consequence of increased water use by plantations of fast growing tree species may be a catastrophe, despite all benefits further downstream. Where high water tables and salinity are a problem, plantations have been used to control interflow and seepage to effectively lower water tables at valley bottoms.

Unfortunately a very simple concept of the benefits of forest plantations has been followed in the carbon trade argumentation and it is to be feared that in reality forest plantations will not always produce the desired effects. In view of the local, regional and global importance it would be important to establish a global database on the effects of forest plantations on soils, landscapes and communities and to establish a scientific network to identify and address gaps in knowledge. The following contributions provide insight into some ongoing research in this field.

Something to read:

• Calder IR 2002. Forests and hydrological services: Reconciling public and science perceptions. Land Use and Water Resources Research 2.2:1-12
• Currie WS and Nadelhoffer KJ. 2002. The imprint of land-use history: Patterns of carbon and nitrogen in downed woody debris at the Harvard forest. Ecosystems 5:446-460
• Koerner W, Dupouey JL, Dambrine E and Benoît M 1997. Influence of past land use on the vegetation and soils of present day forests in the Vosges mountains, France. J of Ecology 85:351-358
• Little D, Farrell E and Collins J 1997. Land-use legacies and soil development in semi-natural ecosystems in the marginal uplands of Ireland. Catena 30:83-93
• Noble AD and Randall PJ 1998. How trees affect soils. RIRDC Publication No 98/16, 129 pages, worth to read

Gerhard Glatzel
Institute of Forest Ecology, UNI BOKU Vienna, Peter Jordan-Strasse 82, A-1190 Vienna, Austria
glatzel@woek.boku.ac.at

Top of page

DUNG OR FOREST BIOMASS AS FUEL - EVEN PLANTATION TREES WITH A BAD REPUTATION HELP CONSERVE SOIL FERTILITY
By Zerfu Hailu

Introduction
Ethiopian highlands suffer from a severe fuel wood shortage. Despite of a long history of Eucalyptus plantations on the Ethiopian highlands, the potential of the forest resources to supply fuel-wood on a sustainable yield basis is 12.5 million m3.year-1, vastly insufficient to satisfy the calculated demand of fuel-wood of 45 million m3.year-1 (Forestry Action Program, '94),. The projected demand for 2014 is about 88.9 million m3.year-1 and the projected supply on a sustainable yield basis will be only 8.84 million m3.year-1. This means that the demand will be ten times the sustainable supply.

The Ethiopian energy assessment report by United Nations Development Program (UNDP) and World Bank (1995), indicates that in the early 1990s 93% of the total national energy consumption was supplied by biomass fuel sources. Cow dung is widely used and its share is particularly high in the Ethiopian Highlands, increasing with increasing elevation. Above 2400 m the share of cow dung of the total biomass fuel consumption, is in the range of 41-65% (Mesfin 1991).

Research Rationale
Expansion of forest plantations using fast growing, browsing resistant species, such as Eucalyptus, would help to alleviate the problem. Plantations of Eucalyptus species, however, are considered potentially harmful to the environment, because of excessive water use, soil nutrient depletion and adverse effects on biodiversity. As nutrient depletion and land degradation due to dung collection and its use as fuel is extremely harmful because of its adverse effects on food security, a scientific study was conducted to compare nutrient drain from Eucalyptus globulus fuel wood plantations and cow dung collection.

Materials and Methods

Tree biomass sampling
Tree biomass sampling was done in December 1999 and January 2000 at two different sites. Sample trees were felled at a stump height of 10 cm from the ground with a bow saw as practiced by the farming communities in the research areas. After felling and removal of branches total tree height up to the tip of the leading shoot was measured. Discs with a width of 3 cm were taken from the very bottom of the stem and at every 1 m interval along the length until the diameter over bark dropped below 2.5 cm.

Cow dung
Livestock dung was collected from the villages near the research sites. Five dung cake samples were taken from five different households. The cakes were collected from the dung-drying yard, as it was ready to be used for fuel or to be marketed. The dung cakes were broken down and mixed and out of the mixture a sub-sample of 100 gram was drawn for oven dry weight determination and nutrient analysis.

Sample preparation and chemical analysis of biomass samples
All biomass samples were oven-dried at 80 0C to constant weight and dry weight was determined. Samples were homogenized for further chemical analysis by milling (mesh = 0.5 mm). Chemical parameters were determined by standard procedures used in the analytical laboratory of the Institute of Forest Ecology at Vienna University of Agricultural Sciences

Biomass energy balance
Newcombe (1989) used a calorific value of 14.3 MJ.kg-1 for wood and 13.8 MJ.kg-1 for cow dung in his report of economic justification for rural afforestation in Ethiopia. This means that 0.952 kg of wood can substitute 1.0 kg of cow dung. Based on this calorific value equivalence the nutrient depletion by E. globulus fuel wood plantations could be compared to nutrient losses caused by the current use of cow dung as biomass fuel.

Results
Table 1 shows biomass and nutrient content of two Eucalyptus plantations in the Ethiopian Amhara Region as compared to equivalent amounts of cow dung (based on caloric equivalents). At Teda Ager research site the calculated stem wood biomass in a 11.5 years old E. globulus plantation was 46.3 t ha-1. This stem wood biomass can substitute 48.6 t of cow dung from its present use as a fuel with equivalent calorific value. At Weldeab Ager research site, calculated stem wood biomass in 14.5 years old E. globulus plantation was 176.5 t.ha-1, which can substitute 185.4 t of cow dung at equivalent calorific value.

Table 1. E.Globulus stem wood and equivalent livestock dung biomass in t and macronutrients in kg. ( Source Zerfu 2002)

Discussion
At Teda Ager research site, nitrogen in stem wood biomass was 32.9 kg while caloric equivalent amounts of dung contained 858 kg, which is more than 25 times as much. This means that substitution of dung as biomass fuel by Eucalyptus wood depletes the land far less. For one kg of nitrogen exported by burning Eucalyptus wood 26 kg of nitrogen in dung could be saved and returned to the fields as organic fertilizer. At the average conditions of the investigated sites (age and stocking), a hectare of E. globulus plantation represents more than 800 kg of net nitrogen recycling potential in the farming system in the form of organic fertilizer by replacement of dung from its present use as fuel resource. The recycling potential is even higher in Weldeab Ager (Table 1 Pg 7). For phosphorus the amount in the stem wood biomass was 11.9 kg at Teda Ager research site, and that of dung was 168.5 kg. This means that 1 kg of phosphorus removed with stem wood harvest for biomass fuel allows for the return of more than 14 kg of phosphorus from cow dung not used as fuel. The trend is similar with the other macronutrients (Table 1Pg 7).

The importance of soil organic matter for sustainability lies predominantly in those circumstances where management based on fossil-fuel sources is either impossible or undesirable, which is the case in many tropical farming systems (Swift & Woomer 1993). Although application of mineral fertilizer increases yields in arable farming, mineral fertilizer alone cannot sustain crop yields in the long run. Addition of organic matter, such as cow dung, in agricultural fields plays an important role in nutrient cycling, erosion control and the maintenance of favourable soil chemical and physical properties. However, with the present Ethiopian situation, it is unthinkable to bring cow dung into the farming system without having a substitute for its use as domestic fuel. At present no other species, be it indigenous or exotic, is available, which can substitute Eucalyptus in narrowing the ever-widening gap between demand and supply of fuel biomass. This is not only because of its rapid growth, but also its deep rooting and tolerance of drought as well as its resistance to extreme browsing and ease of propagation. Therefore only a step-by-step strategy will be able to alleviate the present biomass fuel crisis in the Ethiopian Highlands and to allow for consideration of other important issues such as biodiversity conservation and rebuilding of close to nature forests in the future.

Conclusion
The role of E. globulus plantations in Ethiopia's situation is far reaching when it is evaluated in view of its potential contribution to the farming system through substitution of dung. Therefore, E. globulus plantations are recommended and most likely the only realistic solution to the crisis, despite many critical views on Eucalyptus plantations. Under appropriate management practices dung is a renewable and sustainable organic fertilizer and it should be targeted as a key resource because of its positive effects on the physical and chemical properties of the soil, and consequently on the farming system and food security in the Ethiopian Highlands.

References

• Ethiopian Forestry action plan 1994, The challenge for development, volume 2, Ministry of Natural Resources development and Environmental Protection, Addis Ababa, Ethiopia.
• Mesfin Wolde-Mariam, 1991. Suffering Under God's Environment: A vertical study of the predicament of peasants in North-Central Ethiopia. Institute of Geography, University of Berne, Switzerland, pp 220.
• Newcombe, K. 1989. An economic justification for rural afforestation: The case of Ethiopia. In:Schramm, G. and Warford, J.J. 1989. Environmental management and economic development, International bank, Johns Hopkins University press, Baltimore, U.S.A.
• Swift, M.J. and Woomer, P., 1993. Organic matter and sustainability of agricultural systems: Definition and measurements. PP 3-17. In: Mulongoy, K. and Merckx, R. (eds), Proceeding of an international symposium held between 4-6 November, Leuven, Belgium.
• Zerfu, H., 2002. Ecological impact evaluation of Eucalyptus plantations in comparison with agricultural and grazing land-use types in the Highlands of Ethiopia. Ph.D. dissertation, Institute of Forest Ecology, Vienna University of Agricultural Sciences, Vienna.

Zerfu Hailu
Environmental Protection Land, Administration & Use Authority, P.O.Box 145, Bahir Dar, Ethiopia
Tel +251 8 206810, Fax +251 8 202275

Top of page

MIXED SPECIES PLANTATIONS OF EUCALYPTUS AND ACACIA: GROWTH, NUTRITION AND SOIL CHANGES
by Partap Khanna, Juergen Bauhus, Wilawan Wichiennopparat and Peter Snowdon

Introduction
Eucalypts and acacias are preferred plantation species in the tropics and sub-tropics because of their fast growth which is expected to meet the extensive demands of wood for construction, poles, pulp and fuel. Almost all the industrial plantations are monocultures, and questions are being raised about the sustainability of their growth and their effects on the site. Repeated harvesting of eucalypt plantations on short rotations may deplete site nutrients because of export of nutrients in the harvested wood and other tree components, and further losses may occur during inter-rotation site management practices e.g., slash burn, site preparation. Nitrogen (N) losses are likely to be very important for future growth. It is therefore appropriate to explore new systems of plantation management in which N may be added via fixation. As an attractive alternative to monoculture plantations, Mixed-species plantations can include an N-fixing species with more valuable tree species (Parrotta 1999) to improve both plant productivity and soil nitrogen dynamics (Khanna 1998), to increase biomass production (Binkley 1992, Montagnini et al. 1995, Binkley 1997) and to increase soil carbon sequestration (Kaye et al. 2000). In addition, a range of additional potential advantages may occur. For example, structural and biological diversity is increased while susceptibility to pest and disease outbreaks may be reduced. The faster-growing species in the mixture provide improved microclimate for establishment and growth of companion species (i.e., serve as a ‘nurse crop'), and improve stem form. From a commercial viewpoint, mixed-species plantations have the advantage of producing more than one product, giving their managers a better chance of adapting successfully when market demands change.

About 10 years ago we established three experiments (two in Thailand and one in Australia) where Eucalptus (E) and Acacia (A) were mixed in five proportions (100% E, 75%E+25%A, E50+A50; E25+A75; 100A) and were grown under two planting densities. In Thailand the trees in denser planting were thinned to remove Acacia at 28 months; and were harvested at 98 months. Preliminary data on the growth and biomass of trees and nutrient contents were published by Wichiennopparat et al. (1998) and the final data will be presented at the ‘International Conference on Eucalypt Productivity' at Hobart, Australia (http://www.cdesign.com.au/Eucprod/). Some data on the growth and nutrition of trees were presented by Khanna (1997, 1998) and Bauhus et al. (2000).

Growth and management
Mixing Eucalyptus species with N-fixing trees generally increases the total production of wood, and in some cases even the total production of Eucalyptus wood. For example, mixtures of Eucalyptus globulus and Acacia mearnsii produce 15% more eucalyptus biomass than pure stands of eucalypts in addition to the substantial growth of acacias (Fig.1). This resulted from the additional N that was made available to eucalypts when grown in mixtures with acacia and also from better utilisation of soil resources. Investigation by Bauhus et al. (2000) of fine roots of mixed-species plantation (at 6.5 years of age) suggested that increased productivity (most evident in the 50:50 acacia-eucalypt mixtures) was the result of stratification in the fine-root systems of the two species. A key challenge in managing mixed-species forests is establishing the desired competitive balance between species, so that each species achieve their potential by making optimum use of their resource niche. This challenge of balancing species may not be economically feasible in some temperate forests, but harvesting of small-diameter materials to remove improper species balance would have substantial fuelwood value in many tropical areas.

Fig. 1. Stem basal area of average trees of Eucalyptus globulus (E) and Acacia mearnsii (A) at 5.5 years of age in monoculture (100E and 100A) and 50E+50A mixed stands.

Effects on nutrient supply and soils
Inputs of carbon to the soil may be greater when N-fixing species are incorporated in the plantation mixture (Paul et al. 2002), which is likely to occur because increased productivity will cause greater organic matter input to the soil by high turnover rates of plant litter (above-ground residues and below-ground roots). In both temperate and tropical forests, the presence of N fixation trees has been shown to increase soil pools (Kaye et al. 2000) of both nitrogen and carbon, and to have variable effects on soil P (reviewed by Binkley and Giardina 1997). Studies have shown that the nitrogen status in mixed-species plantations that include an N-fixing species can be enhanced quickly e.g., one year after plantation establishment (Khanna, 1998). Bauhus et al. (2000) did not observe any change in C in the 0-30 cm of soil from mixed species when compared with the soil from pure stands. However a more differentiated soil sampling in smaller depth increments at 10 years after plantation establishment showed that soil organic carbon was highest in the 50:50 mixtures of eucalypts and acacias (Fig 2), and that the amount of soil carbon was related to above-ground productivity, but not to the percentage of acacia in mixtures. Nitrogen fertilisation of the pure eucalypt stands, however, did not increase soil organic carbon when compared to the unfertilised stands (shown as Efer in Fig 2) and that the soil from the surrounding native forest had higher C levels than in either fertilized or unfertilized eucalypt monocultures.

Fig 2. Organic carbon in 0-30 cm depth of soils from 10-yr old plantations of Eucalyptus globulus (E) and Acacia mearnsii (A) mixtures (25-100) and eucalypt monoculture receiving inorganic fertilization (Efer), and from the surrounding native eucalypt forest (NFst). Treatments carrying the same letter were not significantly different at (P<0.05). (Source: Pares 2002)

Future
It has been demonstrated that the productivity and vitality of mixed plantations can be higher than that of mono-specific plantations of either species used in the mixture. Mixing the species is expected to increase the amount and quality of the harvested wood by improving the form of the tree. Mixed species may return a higher amount and better quality of leaf and root litter resulting in a higher storage of organic carbon in the soil. However, very little is known about the environmental and economic benefits and the appropriate design and management of mixed plantations. Adequate control of competition may be essential to best manage the mixed species plantations. We suggest that the mixed species will provide :

• Increased stand productivity, better stand health, improved site fertility, enhanced soil ecological values, and improved carbon sequestration.
• Higher economic returns and better stability of plantations.
• Additional research efforts and social awareness to develop suitable management models for mixed species plantations.

We see an essential and significant role of international organisation such as CIFOR to facilitate the development and acceptance of such alternative plantation systems.

Acknowledgements
This project received financial support from ACIAR and the following institutions and many of their research and technical staff and students were involved:
CSIRO Forestry and Forest Products, P.O. Box E4008, Canberra-Kingston 2606, Australia.,
Silvicultural Research Division, Royal Forest Department, 61 Paholyothin Rd., Chatuchak, Bangkok, 10900 Thailand.
School of Resources, Environment and Society, The Australian National University, ACT 2002, Australia.

References

• Binkley, D. 1992. Mixtures of nitrogen-fixing and non-nitrogen-fixing tree species. In: Cannell, M. G. R., D. C. Malcolm, and P. A. Robertson (ed.) The ecology of mixed species stands of trees. London, Blackwell Scientific. pp. 99-123.
• Binkley, D. 1997. Bioassays of the influence of Eucalyptus saligna and Albizia falcataria on soil nutrient supply and limitation. For. Ecol. Manage. 91, 229-234.
• Binkley, D., C. Giardina and M. A. Bashkin. 2000. Soil phosphorous pools and supply under the influence of Eucalyptus saligna and nitrogen-fixing Albizia facaltaria. For. Ecol. Manage. 128, 241-247.
• Bauhus, J., P.K. Khanna and N. Menden. 2000. Aboveground and belowground interactions in mixed plantations of Eucalyptus globulus and Acacia mearnsii. Can. J. For. Res. 30, 1886-1894.
• Kaye, J. P., S. C. Resh, M. W. Kaye and R. A. Chimner. 2000. Nutrient and carbon dynamics in a replacement series of Eucalyptus and Albizia trees. Ecol. 81, 3267-3273.
• Khanna, P.K. 1997. Comparison of growth and nutrition of young monocultures and mixed stands of Eucalyptus globulus and Acacia mearnsii. For. Ecol. Manage. 94, 105-113.
• Khanna, P.K. 1998. Nutrient cycling under mixed tree systems in south-east Asia. Agrofor. Systems 38, 99-120.
• Montagnini, F., E. Gonzalez, C. Porras and R. Rheingans. 1995. Mixed and pure forest plantations in the humid neotropics: a comparison of early growth, pest damage and establishment costs. Commonw. For. Rev. 74, 306-314.
• Pares, A. 2002. Soil organic carbon sequestration in mixed and monospecific plantations of Eucalyptus globulus ssp. pseudoglobulus and Acacia mearnsii. Thesis B. Sci. (Hon), Forestry Program, School of Resources, Environment and Society, The Australian National University.
• Parrotta, J. A. 1999. Productivity, nutrient cycling, and succession in single- and mixed-species plantations of Casuarina equisetifolia, Eucalyptus robusta, and Leucaena leucocephala in Puerto Rico. For. Ecol. Manage.124, 45-77.
• Paul, K. I., P. J. Polglase, J. G. Nyakuengama and P. K. Khanna. 2002. Change in soil carbon following afforestation. For. Ecol. Manage. 168, 241-257
• Wichiennopparat, W., P.K. Khanna and P. Snowdon. 1998. Contribution of acacia to the growth and nutrient status of eucalypts in mixed-species stands at Ratchaburi, Thailand. In: Turnbull, J.W. Crompton, H.R. and K. Pinyopusarerk (eds) Recent Developments in Acacia Planting. Proc. Intern. Workshop Hanoi, Vietnam 27-30 Oct. 1997, pp281-287.

For further information contact;
Dr P.K. Khanna,
Institute of Soil Science and Forest Nutrition, Buesgenweg 2, 34077, Goettingen, Germany
Email : pkhanna@gwdg.de

Top of page

INFLUENCE OF POPLAR SHORT-ROTATION COPPICE ON SOME SOIL PROPERTIES
By A. Berthelot

Introduction
Fast growing plantations are likely to provide a huge quantity of raw material quickly and at a relatively low cost. Hybrid poplars, in french conditions and suitable soils, offer a high productivity. Thanks to the use of clonal varieties it is possible to obtain material with homogeneous and well-known properties. AFOCEL planted many plots of short-rotation coppice (SRC) of poplar in France. The stand density varies from 2,000 to 3,000 stems ha-1 and the rotation length is fixed at 7 or 8 years. The objective is to produce 10 to 12 dry tons ha-1 year-1 of total biomass, with a main proportion (75%) of pulpwood. Between 1991 and 1995, the first plots of industrial size were harvested and the effects of several felling systems were studied (Sutter et al., 1995). A European project also aimed to study the impact of debarking on the harvest cost and soil fertility (Toval et al., 1995). This paper, based on soil analyses carried out at planting and at the time of harvest, presents the evolutions of the most common properties of the soil.

Materials and methods
AFOCEL has soil analyses for 9 sites at the time of planting and at the time of harvest (age 7 or 8). We also have the analyses for 5 plots, each year after felling for 2 methods of mechanical felling: debarked and pulplogs which had been debarked and pulplogs with bark. The analyses were carried out by agricultural laboratories, according to the methods usually used. Only the results concerning the upper layers are presented here. For the whole of the plots we profit from an estimate of the total above-ground biomass, calculated by AFOCEL mass tables (Bouvet and Berthelot, 1994). We also know the biomass remaining on the soil after felling (crowns and branches) as well as the quantity and the nutrient content of the bark remaining on the soil for the debarked treatment (Ranger et al., 1988 ; Berthelot et al., 2000). The previous crop was primarily made up of meadows (6), but also of arable lands (2) and a poplar plantation. Two plots are located on slightly acidic soils, all the others are located on alkaline soils.

Results
The results of the analyses at the plantation and at felling appear in table 1. The pH of the upper layer remains stable except for the two plots located in a slightly acidic geological context (Ecorches and Gavrus). In these two cases we note a light acidification due to the humification of the organic matter falling down on the soil each year (fig.1).

In half of the plots we observe a light enrichment of organic matter, even on old meadows. Curiously it is not the case on the two old arable lands. The particular case of the plot of Tourtenay must be mentioned: it is an old drained peaty swamp. The organic matter rate is very high is dropping regularly due to mineralisation (fig.2).

The nitrogen content in the upper layer varies in the same way as the organic matter. The rate of P2O5 does not change, and it is impossible to observe a clear tendency between the planting and harvest. On the other hand it appears clearly that the two plots planted on arable lands are definitely richer in P2O5 than the plots planted on meadows, reflecting the durable influence of the old fertilizations (fig.3). For K2O, we observe more significant variations but, there still, without particular tendency.

Table 2 shows the results for 5 plots, according to 2 methods of harvest (debarked and with bark). After harvest, an analysis was carried out every year for both methods. Only the most recent data are provided (1 to 3 years after harvest, according to the site). The variations are very weak and no difference is visible between both felling methods. The most visible tendencies once again relate to the pH of the upper layer (fig.4). We observe a slight increase in the pH probably due to the setting in light of the soil and its action on the organic matter. This phenomenon is more pronounced in the plot located in an acidic context (Ecorches).

Table 1 : soil analyses results of upper layer for 9 plots, at the time of the planting (initial) and at the time of harvest (final)

* P₂O₅  : Joret-Hebert méthod for alkaline soils and Dyer method for acidic soils (P = P₂O₅  / 2.29)
* K2O : Acetate d'ammonium method (K = K₂O / 1.205)

Table 2 : soil analyses results of upper layer for 5 plots, at the time of the plantation, at the time of harvest and 1 and 3 years after harvesting, with or without bark

* P₂O₅ : Joret-Hebert méthod for alkaline soils and Dyer method for acidic soils (P = P₂O₅ / 2.29)
* K2O : Acetate d'ammonium method (K = K2O / 1.205)

Conclusion
All these analyses confirm that only long term evolutions are easily noticeable. The variations observed here result from the change of land use: short-rotation coppice planted on former agricultural lands. On the other hand, the nature of the forestry interventions and their possible impact on the soil require heavy studies, standardized protocols and a long period of observation. To assess the impact of some practices over a period of only a few years appears quite hazardous compared to the precision of the analyses.

AFOCEL has an important network of trials, on many species: Pinus, Pseudotsuga, Picea, Populus, etc. (Gastine et al., submitted). We already perform soil analyses at planting for several tens of plots. It would be useful to carry out new analyses 5, 10 or 15 years after planting in order to characterize the evolution of soil.

References

• Berthelot A., Ranger J., 2000. Nutrient uptake and immobilization in a short-rotation coppice stand of hybrid poplars in north-west France. Forest Ecology and Management 128 167-179.
• Bouvet A., Berthelot A., 1994. Taillis à courtes rotations de peuplier - Tarifs masses et volumes. Annales de recherches sylvicoles AFOCEL, p 219-234.
• Gastine F., Bouvet A., Deleuze C., Monchaux P., 2002. Le réseau d'essais AFOCEL fête ses 40 ans ! Submitted to Rev. For. Fr.
• Ranger J., Nys C., Barnéoud C., 1986. Production et exportation d'éléments nutritifs de taillis de peuplier à courte rotation. Annales de recherches sylvicoles AFOCEL, p 183-225.
• Sutter B., Reuling. D., Berthelot A., Bonduelle P., 1995. La récolte du TCR de peuplier. Fiche informations-forêt AFOCEL, fasc. 508, p 141-152.
• Toval G., Soria F., Mansilla P., Bonduelle P., Sutter B., Petit H., Berthelot A., 1995. Mechanized logging and on-site dedebarking of short rotation coppice (eucalyptus and poplar): impact on silviculture, environment and economy. Eclair project n°0160 contract n° AGRE-CT91-0057 : Final technical report, 45 p.

For further information please contact:
A. Berthelot,
AFOCEL Nord-Est, route de Bonnencontre, 21170 Charrey-sur-Saône, France
Email: alain.berthelot@afocel.fr

Top of page

THE EFFECT OF EXOTIC TREE PLANTATIONS IN NORTHERN THAILAND ON SOIL PROPERTIES
By Roongreang Poolsiri

Introduction
The forest area in Thailand is decreasing steadily due to of the demand of land for agriculture, pasture and living. Most of the deforestation occurs in the highland on soils which are extremely susceptible to soil erosion. The destroyed or threatened forests are important for the water supply to local and downstream communities as well as for both plant and animal biodiversity. For the past three decades, the Thai government has attempted to restore degraded highland areas through reforestation carried out by the Royal Forest Department. Exotic tree species were preferred because they are fast growing and provide cover rapidly and thus protect the soil from ongoing erosion. More recently concerns were aired that exotic plantations use more water and nutrients for their growth. In order to obtain data, several exotic tree plantations were studied with regard to their nutrient uptake, storage and return to forest soils.

Litter fall represents a major biological pathway for element transfer from vegetation to soils. Knowledge of seasonality of nutrient and mass return through litter fall to the forest floor is important for plantation management. The objective of this study was to study soil properties and rooting as well as to quantify the litter fall and its nutrient status of four exotic tree species, i.e. Acacia confusa, Liquidambar formosana, Cinnamomum camphora and Cunninghamia lanceolata. These species were chosen because they are common and allow the comparison of evergreen vs, deciduous species including a nitrogen fixing legume and a conifer.

As the chemical analyses of the soil samples are not yet finished only the litter fall data are presented here.

Materials and Methods

Study site
This study was carried out at The Royal Project, Doi Angkhang, Chiangmai Province, which is located in the northern part of Thailand. It lies at a latitude of 19º 52¢ N and longitude of 99º 02¢ E. Doi Angkhang is situated on limestone and shale mountains laid down in a north south direction. On shale and sandstone the land is undulating. Limestone outcrops have been weathered by bicarbonate solving, producing a typical Karst topography (Phupharung, 1979). The study area has an altitude of 1,400 m msl. The experimental plots, 20 x 25 m in size, were selected in plantations of four exotic tree species, i.e. Acacia confusa, Cinnamomum camphora, Liquidambar formosana and Cunninghamia lanceolata which were 28 years old, fully stocked and growing well without visible nutrient deficiencies or growth defects.

Litter fall
Litter fall samples were collected during one year starting in March 2000. In each plantation there were 5 litter traps set up randomly in the plot. The litter trap area was 1x1 m2. Every month the litter was sorted and divided into 5 fractions for each plantation, i.e., leaves, branches, barks, flowers and fruit, and other fractions. All the samples were taken for oven dry weight determination and nutrient analysis.

Sample preparation and chemical analysis of litter fall samples
All the litter fall samples were oven-dried at 80 ºC to constant weight and oven-dried mass was determined. Samples were homogenized for further chemical analysis by milling (mesh=0.5). Chemical parameters were determined by standard procedures used in the analytical laboratory of the Institute of Forest Ecology at University of Agricultural Sciences, Vienna.

Results and discussions
The mean and fractioned (leaves, branches, fruits, and others) mass of the annual litter fall for A. confusa, L. formosana, C. camphora and C. lanceolata plantations are given in Table 1. In this study litterfall mass was calculated in g.m-2. The litterfall mass of the C. lanceolata plantation was uncommonly low due to tree pruning in the years before study. In order to allow a rough estimate of nutrient fluxes in the C. lanceolata plantation actual litter fall was adjusted to equivalent amounts of litterfall in the other plantations. The actual litter fall mass is given in the parentheses in the row below the assumed litter fall mass of the C. lanceolata plantation in the Table 1.

Litter deposition
The mean mass of the annual litter fall ranged from 575 g.m-2 for C. camphora plantation to 694 g.m-2 for the L. formosana plantation (Table 1). There were no significant differences among species. However the mean weight of annual litter of the C. camphora plantation was significantly lower than other non conifer plantations. The leaves fraction ranged from 337 g.m-2 for the C. lanceolata plantation to 535 g.m-2 for the L. formosana plantation (Table 1). The leaves fraction of A. confusa accounted for about 66% of the mean annual litter fall, while for the other plantations, that fraction accounted for between 75% and 77% of the mean annual litter fall.

The branches fraction ranged from 112 g.m-2 for the C. camphora plantation to 166 g.m-2 for the C. lanceolata plantation. Fruits fraction is composed of flower and fruit parts. These ranged from 3 g.m-2 for L. formosana plantations to 55 g.m-2 for C. lanceolata plantations. In this fraction the mean masses of L. formosana could not be analyzed because there was only one month's data for a year in this study. In the other parts fraction of L. formosana and C. camphora plantations there is no significant difference in values.

Table 1. Mean masses of annual litter flux and masses of litter fall fractioned in to leaves, branches, fruits, and other under 4 exotic tree plantations at Doi Angkhang, Chiangmai Province

¹ Means in the same column followed by the same letter are not significantly difference at 0.05 level of significance (LSD).
Values in parentheses are mean actual litter fall masses of the C. lanceolata plantation in the study year.

Table 2. Annual litter fall flux N, P, K, Ca and Mg (g.m^-2) of the fractions of the total elemental flux in each partitioned litter fraction under 3 exotic tree plantations at Doi Angkhang, Chiangmai Province

Nutrient flux in the litter fractions
Table 2 shows the nutrient flux in the litter fractions for 3 of the exotic tree plantations. For this study the litter fractions of C. lanceolata were not used for calculation and comparison as they were not representative because of the disturbance due to pruning. The nutrient flux was calculated in g.m-2 and is presented in Table 2.

The nutrient flux generally followed the pattern leaves > branches > fruits > others (Table 2). The nutrient flux of N, P, K, Ca and Mg of leaves varied according to stand. Acacia confusa had the highest nutrient flux of N compared with L. formosana and C. camphora. Liquidambar formosana had highest values for P, K, Ca and Mg. The nutrient flux from branches follows the same pattern described above, with a few exceptions. Branches of A. confusa were considerably higher in N and Mg, while C. camphora was higher in P and K. Branches of L. formosana had intermediate values for N, P and K but were higher in Ca. Fruits litter had consistently lower N, P, K, Ca and Mg nutrient fluxes than either leaves or branches litter (Table 2). Fruits litter from A. confusa had higher for all nutrient fluxes than the fruits fraction from other species. Nitrogen, P, K, Ca and Mg nutrient fluxes in the others fraction of L. formosana and C. camphora were always higher than in the fruits fraction.

Discussions and conclusions
The litter fall study provides an estimate of the magnitude of nutrients returned to the forest soil each year. It has to be seen in relation to other land use practices where very little is returned because of biomass harvesting or where nutrients are lost due to soil erosion. The contribution of different biomass fractions to annual litter contributions varies among the tree species. Typical for forests, the component with the highest share in terms of mass on an annual basis was leaves (Caldentey et al., 2001, Vogt et al., 1986) followed by branches, others and fruits fractions. The monthly litter fall depends mostly on factors responsible for leaf senescence and abscission. In areas with no prevalent or strongly seasonal water limitations, temporal variations of leaf litter on the forest floor are the combined result of rate of fall and decomposition of falling material, and the diverse responses of species to different environment cues (Cuevas and Lugo, 1998). In this study, the plantations have the same climate and environment, and there was no significant difference between the litter fall for each tree species; even C. camphora is an evergreen-broadleaf tree species.

As expected, the litter fall in the forest plantations has released the nutrients back to the forest soils. We could not calculate the nutrient releases from the C. lanceolata plantation because of the sample disturbance due to pruning. In accordance with theory the legume tree species A. confusa has the highest nitrogen levels in all fractions of the litter fall because it is capable of nitrogen fixation from the air.

These preliminary results show that between 60 and 100 kg of nitrogen are returned annually with a litter fall of between 5.5 to 7 tons of biomass. The ongoing soil analyses will show how this is reflected in soil chemistry. The beneficial effects of litter decomposition on topsoil porosity as compared to degraded open or agricultural land is obvious even to the casual observer.

References

• Caldentey, J., M. Ibarra and J. Hernández. 2001. Litter fluxes and decompostion in Nothofagus pumilio stands in the region of Magallanes, Chile. For. Ecol. Manage. 148 : 145-157.
• Cuevas, E. and A. E. Lugo. 1998. Dynamic of organic matter and nutrient return from litterfall in stands of ten tropical tree plantation species. For. Ecol. Manage. 112 : 263-279.
• Phupharung, B. 1979. Soil classification and land potential determination at Doi Angkhang, Chiangmai Province. M.Sc. Thesis, Kasetsart Univ., Bangkok.
• Vogt, K., C. Grier and D. Vogt. 1986. Production turnover and nutrient dynamic of above- and below-ground detritus of world forest. Adv. Ecol. Res. 15 : 303-377.

For further information please contact:
Roongreang Poolsiri
Department of Silviculture, Faculty of Forestry, Kasetsart University, Bangkok, Thailand. 10900
Email: fforrrp@ku.ac.th

Top of page

THE CARBON CYCLING IN THE EUCALYPT PLANTATIONS IN CONGO
By Jean-Pierre Bouillet, Yann Nouvellon, Jean de Dieu Nzila and Olivier Hamel

Since 1978, 42,000 ha of clonal Eucalyptus plantations have been established in the savannas around Pointe-Noire (Congo), on sandy soils with low nutrient content. Two studies focusing on carbon cycling have been conducted. The first one dealt with the effects of organic matter management on plantation productivity, and the second one dealt with CO2 fluxes and carbon sequestration within stands.

Organic matter management
Six treatments simulating different harvesting intensities, were designed to leave six levels of organic matter or aboveground biomass on site: 1) Residue removed (R): all aboveground organic residues (slash, litter,…) were removed from the plot, 2) Whole tree harvest(WTH): all aboveground components of the commercial trees were removed, 3) Stemwood and bark harvested (SBH): only the commercial-sized boles and associated bark were removed, 4) Stemwood harvested (SH): only debarked, commercial-sized boles were removed, 5) Organic residues were burned, following stemwood harvest, as in treatment SH, and 6) Double slash retained (DS): commercial trees were harvested as in the SBH treatment ; the residues from the WTH plots were added to the residues from this treatment.

One year after replanting, eucalypts in treatment R exhibited a lower total biomass compared to other treatments (5.9 t ha-1 vs a mean of 7.2 t ha-1 for the remaining treatments). The lowest values of nutrient content aboveground were observed in treatment R. Under the most favourable treatment (usually DS), plots accumulated 53, 63, 45, 145, and 92% more N, P, K, Ca and Mg, respectively, compared to treatment R. At 24 months, treatment R was significantly less productive than other treatments. The differences between the most and least productive treatments (DS and R) were 1.7 m in height, 7.4 cm in circumference, and 8.3 m3 ha-1 year-1 in total volume. Using the model proposed by Olson (1963), the coefficient of decomposition was found to be about 0.10, irrespective of treatment. It was estimated that a 50% loss in mass occurred within 6-8 months after clearcutting. The amount of nutrients released during slash decomposition varied considerably among treatments. Maximum values were reached in treatment DS, 20 months after the initial harvest, with 329 kg N ha-1, 41 kg P ha-1, 99 kg K ha-1, 73 kg Ca ha-1 and 52 kg Mg ha-1. This rapid release of nutrients leads to great risks of nutrient leaching, especially if there is a long delay between clearfelling and planting.

It was then shown that the productivity of the eucalypt plantations is largely dependent on conservative management of organic matter and nutrients. From an operational point of view, it is recommended: (i) to debark stems in the field; (ii) to retain stem tops on the site; (iii) to avoid slash burning; (iv) to reduce the delay between stand harvesting and crop planting.

CO2 fluxes and carbon sequestration
Since October 2000, CO2 and water fluxes have been measured within a young eucalypt stand. The main objective is to derive the net carbon ecosystem exchange (NEE = carbon sequestration) from continuous eddy flux measurements, and to compare these estimates to those obtained from:
1) Measurements of Net Primary Productivity (NPP = carbon stock increment in the aerial and below-ground tree compartments + litter production from the aboveground tree compartment + litter production from the fine root turnover) and soil heterotrophic respiration (Rh = soil CO2 efflux resulting from soil organic matter and litter decomposition), with NEE = NPP – Rh, and
2) Measurements of carbon stocks (soil and biomass) and their variations over a chronosequence.

Carbon stocks and soil respiration measurements were obtained over a chronosequence that includes 6 stands from 6 months up to 10 years. The eddy correlation measurements were obtained from the top of a tower erected within the 3 year-old stand of the chronosequence.

Soil CO2 efflux exhibited strong seasonal variations, reflecting the seasonal changes in soil water content. Maximum values were obtained during the wet season, while minimum values were obtained at the end of the dry season. Annual soil CO2 efflux was 11.8 t C ha-1 at eddy correlation site and 16.7 t C ha-1 at the 10 year-old stand. At each site, good relationships were obtained between volumetric water content of the surface soil and soil respiration. Rhizospheric and heterotrophic contributions to total soil CO2 efflux were estimated from comparison of soil CO2 efflux measured over trenched plots and over non-trenched plots: at the eddy correlation site, root respiration contributed to 25% of total annual CO2 efflux. First results obtained at other stands indicate that root contribution to total soil CO2 efflux increases with stand age, probably as a result of root biomass increase with stand age.

Similar to soil CO2 efflux, NEE measured by eddy correlation exhibited strong seasonal variations with lowest values (highest CO2 uptake) obtained during the wet season. At this time, minimum diurnal peaks of NEE were about -25 mmol m-2 s-1. By contrast, minimum peaks obtained at the end of the dry season were about -12 mmol m-2 s-1. Monthly NEE ranged from -85.3 g C m-2 up to 29.8 g C m-2. This positive value (net carbon emission) has been observed at the transition between the dry and the wet season, and resulted from a faster increase of ecosystem respiration than photosynthesis, after the first rains.

For the two-year period covered by eddy correlation measurements, mean annual NEE was -370 g C m-2 year-1. This corresponds to a net carbon uptake by the stands, but represents a small fraction of the gross primary production (-1990 g C m-2 year-1), due to high carbon loss by ecosystem respiration (1620 g C m-2 year-1). Aboveground respiration, root respiration, and heterotrophic respiration represented 27%, 19% and 54% of ecosystem respiration, respectively.

Over the same period NPP was -1203 g C m-2 year-1 (-676 g C m-2 year-1 for total tree biomass increment, and -527 g C m-2 year-1 for fine root turnover and litter production). Summing NPP and Rh provides another estimate of annual NEE (-330 g C m-2 year-1), slightly lower to the one obtained from eddy correlation measurements (-370 g C m-2 year-1). Carbon stocks measurements obtained at the other stands will further be used for estimating carbon sequestration over successive rotations.

For further information please contact :
Jean-Pierre Bouillet
Cirad-Forêt, Campus international de Baillarguet, TA 10C, 34 Montpellier Cedex 5 France
Tel: + (33) 04 67 59 38 66, Fax: + (33) 04 67 59 37 33
Email: jean-pierre.bouillet@cirad.fr

Top of page

THE IMPACT OF TROPICAL INDUSTRIAL TREE PLANTATIONS ON SITE NUTRIENT STATUS, SITE PRODUCTIVITY AND ECONOMIC PROFITABILITY
By Jens Mackensen and Horst Fölster

Introduction
The establishment of industrial plantations in the tropics is increasing. International efforts to credit plantations for carbon sequestration will further enhance this development. As the rate of carbon sequestration equals the stand productivity the main question is on how sustainable industrial plantation management can be - for both timber production and carbon sequestration? Industrial tree plantations are known for their nutrient mining impact. Nutrient loss is unavoidable but depends on the management intensity. For the generally weathered tropical soils already a modest soil nutrient loss can result in nutrient deficiency, which in turn has an impact on the stand productivity. The main objective of the research conducted was to assess management dependent nutrient losses and the economic impact of the required nutrient compensation, which is to be considered for economical carbon sequestration schemes.

Site and approach
In this assessment, we draw partly on global or regional data, partly on own data from the PT.IHM plantation concession in East-Kalimantan, Indonesia. Ali- and Acrisols are found on 80% of the site exhibiting low pH (pH(H2O): 4.5-4.8), high aluminum saturation (56-91%), an effective cation exchange capacity (ECEC) of 18-26 cmol+ kg-1 clay and a clay content of 20-42%. Eucalyptus deglupta and Acacia mangium are the dominant species. Investment calculation is based on mean annual increment (MAI) of 25m3 ha-1 for both species during a rotation length of 8 years resulting in an expected harvest volume of 200m3 ha-1.

By deriving relative values for nutrient losses due to stem harvest, leaching, erosion and slash burning and applying them to relevant site-specific parameters such as soil parameters, species, and management intensities, we derived the best estimate available for assessment of off-site losses under industrial plantation management. Nutrient input through precipitation and weathering was considered. The approach is considered a first approximation but no substitution for detailed and long-term site-specific nutrient losses. In awareness of the methodological uncertainties we applied a conservative estimate for management-dependent nutrient losses (Mackensen, 1998, 1999; Mackensen et al. 2001, in press).

Results
Average management induced nutrient losses within one rotation ranged from 10 to 50% of plant available soil nutrients. The calculated nutrient losses for the assumed harvest volume and according to management intensity and soil fertility ranged between 14-63% for N, 3-17% for P, 14-53% for K, 5-42% for Ca and 3-33% for Mg. Under high and medium-impact management scenarios nutrient losses due to leaching, burning and erosion were comparable or even higher than nutrient export caused by stem harvest. Slash burning contributed most to nutrient losses. Erosion caused significant losses of Ca , Mg and P. Ca-losses are especially high for Acacia stands, whereas K- and Mg-losses were higher in Eucalypt stands. Nutrient depletion will first occur on nutrient poor sites such as Ferral- and Arenosols or shallow Ali- and Acrisols at the study site, but is also expected to occur after 2-5 rotations on average plantation sites.

Compensation for these nutrient losses is essential to avoid distinctive depletion of soil nutrient storage. Without fertilizer application intensively managed tree plantations generally have a negative nutrient balance: within one rotation more nutrients are lost from the system than gained. Fertilizer compensation for nutrient losses incurred due to tree harvest increased standard plantation establishment cost by 18 to 33%. As a consequence, the internal rate of return dropped from 14% to 9-12%. Fertilization costs are species specific. Considering the fertilization costs, which can potentially make up for any carbon credit gain, strategies to reduce management-dependent nutrient losses are unavoidable. A low-impact management including alternatives to slash burning, soil-conserving harvesting techniques, and appropriate site selection are recommended.

Discussion
Plantations established on poor soils in the humid tropics incur significant nutrient losses due to harvesting and site management. Even though these losses can be reduced by low-impact management, they are still high compared to plant available soil nutrient stocks, accumulated losses under intensive short-rotation management will result in severe soil nutrient depletion and subsequently decrease stand productivity and carbon sequestration rates, if not compensated for. Compensation of nutrient losses will raise plantation establishment costs significantly, especially on marginal sites.

Our results point to the need for developing better nutrient management in industrial plantations. Nutrient losses and thus fertilization costs can be reduced by appropriate site management. The abandonment of slash burning is the single most important factor for potential nutrient savings. Between 10 to 50% of total nutrient losses could be avoided if alternatives to slash burning were adopted. Additional leaching and erosion losses triggered by slash burning will also be reduced.

References

• Mackensen J., Fölster H. & Klinge R. (2002). Assessment of management-dependent nutrient fluxes in tropical industrial tree plantations. Ambio, in press.
• Mackensen J., Fölster H. & Ruhiyat D. (2000). Cost-analysis for a sustainable management of fast growing tree plantations in East-Kalimantan, Indonesia. Forest Ecology and Management 131:239-253.
• Mackensen J. (1999) Nutrient management for industrial tree plantations (HTI) in Indonesia – Practical guideline towards integrated nutrient management. Tropical Forest Research, TÖB FTWF-17e, GTZ, Eschborn, pp. 96.
• Mackensen J. (1998) Untersuchung zur Nährstoffversorgung in schnellwachsenden Plantagensystemen in Ost-Kalimantan, Indonesien - ökologische und ökonomische Implikationen. Göttinger Beiträge zur Land- und Forstwirtschaft in den Tropen und Subtropen, 127:1-209.

Jens Mackensen
United Nations Environment Programme, Division of Policy Development and Law, P.O.Box 30551, Nairobi, KENYA

Horst Fölster, Corresponding author
Institute for Forest Nutrition and Soil Science, University Göttingen, Büsgenweg 2, D-37077 Göttingen, GERMANY
Email: hfoelst@gwdg.de

Top of page

THE LONG TERM MEMORY OF SOILS - HOW AMAZONIAN DARK EARTHS REFLECT PAST LAND USE
By Bruno Glaser

Soils store valuable information on site and climate conditions under which they were formed (e.g. Zech et al., 2000). It has also been shown that land-use especially influences amounts and composition of soil organic matter (e.g. Srivastava and Singh, 1991; Guggenberger et al., 1994; Neufeldt, 1998; Glaser et al., 2000a). While classical research on the genesis of soils quantified the amounts of organic matter and nutrients, new analytical approaches based on sophisticated chemical biomarker analysis aim to identify their sources.

In Amazonia, nutrient poor Oxisols and Ultisols predominate. These soils can hardly be used for agriculture in a sustainable way. It is known that Oxisols can not be used for continuous cropping longer than for one or two years without application of high amounts of imported fertilizers such as NPK or super phosphate (Sanchez and Logan, 1992; Steiner, 1996). However, these fertilizers are not affordable for the poor smallholder farmers. Additionally, according to Tiessen et al. (1994) sustainable agriculture in the humid tropics is hardly possible even after intensive fertilization due to the low nutrient retention

Figure 1. Typical soil profile of a Terra Preta at the Hatahara site close to Iranduba near Manaus, Brazil. This is one of the most impressing Terra Preta sites comprising about 16 hectares which has been continuously excavated by Dr. Neves and his group (Neves, 2000; Neves et al., 2001; Petersen et al., 2001). Up to now, a digital topographic map of that site; about 20 m2 of excavations have been realised; and 19 samples were radiocarbon dated (Neves, 2000). This site also has an excellent preservation of organic matter including plant remains, human bones as well as terrestrial and aquatic faunal residues.

Capacity of many soils. Within this landscape of infertile soils anthropogenic dark earths known as Terra Preta (do Indio) occur in patches of up to several hundred hectares, covering in total about 10% of Amazonia (Mann, 2002). These soils are characterized by high amounts of stable and labile organic matter and nutrients which enable sustainable cropping. It is obvious that these dark earths are the product of intensive anthropogenic influence by pre-Columbian Indians (Figure 1 Pg 25).

Recent investigations have shown that the high amounts of stable soil organic matter in Terra Preta soils are mainly due to residues of incomplete combustion (pyrogenic carbon, black carbon) (Glaser et al., 2000b; Glaser et al., 2001a, b). It is assumed that pyrogenic carbon persists in this environment over centuries due to its chemical stability caused by the aromatic structure making the compound also resistant to microbial degradation. This assumption was emphasized by 14C ages of 1000 to 2000 years of this carbon type in Terra Preta soils (Glaser et al., 2001b). Slow oxidation during this time produced carboxylic groups on the edges of the aromatic backbone, which increased the nutrient retention capacity. It was concluded that pyrogenic carbon found in these anthropogenic soils not only acts as a significant carbon sink, but is also a key factor for maintaining the sustainable fertility of Terra Preta soils. Nevertheless, high amounts of pyrogenic carbon do not primarily contribute to higher nutrient contents. However, pyrogenic carbon plays an important role for nutrient retention and, thus, for reduction of nutrient leaching (Glaser et al., 2002; Lehmann et al., 2002). Other recalcitrant biomacromolecules such as lignin contributed only a minor part to soil organic matter (SOM).

Further investigations showed that organic matter is mainly stabilized via chemi-sorption to mineral surfaces whereas physical stabilization via entrapment into the interior of aggregates accounts for about 20% in Terra Preta soils compared to 10% in adjacent soils. Therefore, besides the occurrence of recalcitrant soil organic matter in pyrogenic forms, the stability of soil organic matter in Terra Preta can be partly explained by physical stabilization in aggregates. Additionally, higher amounts of soil organic matter in Terra Preta soils favor soil aggregation (Glaser, 1999).

The higher N contents were also explained by chemical recalcitrance as only 30% of total N in Terra Preta soils could be chemically identified, amino acid-N contributing 18-25%, amino sugar-N 4-7%, and inorganic N 1-2% (Glaser, 1999). It was speculated that the major part of the unknown N in Terra Preta soils consisted of heterocyclic N as it is known that charred residues contain such N forms. Finally, it is considered that the high fertility of anthropogenic dark earths results from a favorable conjunction of mineral and organic contributions, making these soils highly enriched in non-exchangeable forms of nutrients (Glaser, 1999).

Theoretically, only C and N can be produced in situ via photosynthesis and N fixation, respectively. All other elements such as P, K, Ca, and Mg must be incorporated from the surroundings for nutrient accumulation. In situ weathering can be excluded in Amazonia, at least on the heavily weathered Oxisols and Ultisols. Therefore, for the Terra Preta genesis, the following primary and secondary nutrient sources are possible:

1. Human and animal excrements (rich in P and N)
2. Waste including animal and fish bones (rich in P and Ca)
3. Ash residues of incomplete combustions (rich in Ca, Mg, K, P, and pyrogenic C)
4. Terrestrial plant biomass (e.g. green manure, compost)
5. Aquatic plant biomass (e.g. algae)

The aim of an ongoing research project is the small scale field sampling of archaeological remains and soil samples and the investigation of artifacts and "land-use" biomarkers in order to reconstruct the genesis of Terra Preta soils with special emphasis of the origin of soil nutrients. The occurrence of archaeological remains such as human and animal bones, fish bones and turtle backs helps us to identify major nutrient input paths especially of P. By means of lipid biomarker analysis which are especially stable in the environment, a differentiation between input of human and animal excrements as well as between terrestrial and aquatic biomass can be obtained.

If we look first at the potential to differentiate between human and animal excrements as nutrient sources for Terra Preta formation, sterols and bile acids have been proven to be especially useful. Preliminary results of sterol analysis of a Terra Preta show that human excrements do contribute to the nutrient richness of Terra Preta soils (Glaser, unpublished data). Such biomarkers are very stable in the environment, even under extreme environmental conditions (Evershed and Bethell, 1996; Simpson et al., 1998).

If we look further into the potential to differentiate between the input of terrestrial and aquatic biomass as nutrient sources in Terra Preta soils, the n-alkane pattern looks very promising. While cuticular waxes of terrestrial plants contain predominantly long chain n-alkanes (>C20), short chain n-alkanes (<C20) are typical for algae (e.g. Collister et al., 1994; Bourbonniere et al., 1997; Brincat et al., 2000; Filley et al., 2001; Hoefs et al., 2002). Preliminary results of the n-alkane distribution in a Terra Preta show the predominance of aquatic biomass. Thus if this result can be verified for a bigger sample collective the following conclusion with respect to the Terra Preta genesis could be envisioned: Terra Preta is formed via import of aquatic biomass from the alluvial area to Oxisols of the Terra Firme where nutrients are stored, recycled and thus accumulated by additional input of burning residues. Additional evidence for this hypothesis can be drawn from palynological analysis on Colombian Terra Preta sites (Mora et al., 1991: 50).

If we summarize the current knowledge obtained by applying sophisticated analytical chemistry to Amazonian dark earths, the following pattern of past land-use can be identified: Terra Preta is the product of intensive human occupation in the course of which tremendous amounts of burning residues (pyrogenic carbon), human excrements, food waste (animal bones, fish bones, turtle backs) and biomass (especially aquatic biomass from the alluvial area) is added to an initially infertile soil. However, much more research is needed to verify these results both on a smaller (within site variation) and large(from site to site) scale. The reconstruction of the former land-use history of a sustainable land-use system may help to produce such soils in the future, thus providing (i) economic development of poor countries, (ii) a reduction of the greenhouse gas effect by carbon sequestration, and (iii) a reduction of the risk of a further destruction of the Amazonian rain forest.

References

• Collister, J. W., Rieley, G., Stern, B., Eglinton, G., Fry, B. 1994. Compound-specific d13C analyses of leaf lipids from plants with differing carbon dioxide metabolism. Organic Geochemistry 21, 619-627.
• Evershed, R. P., Bethell, P. H. 1996. Application of multimolecular biomarker techniques to the identification of fecal material in archaeological soils and sediments.
• In ACS Symposium Series 625: Archaeological Chemistry: Organic, Inorganic and Biochemical Analysis. M. V. Orna, (Ed.), American Chemical Society, Archaeological Chemistry Symposium Series, Washington, DC pp. 157-172.
• Glaser, B. 1999. Eigenschaften und Stabilität des Humuskörpers der Indianerschwarzerden Amazoniens. Bayreuther Bodenkundliche Berichte 68, 196 pp.
• Glaser, B., Haumaier, L., Guggenberger, G., Zech, W. 2001. The Terra Preta phenomenon: a model for sustainable agriculture in the humid tropics. Naturwissenschaften 88, 37-41.
• Glaser, B., Lehmann, J., Zech, W. 2002. Ameliorating physical and chemical properties of highly weathered soils in the tropics with charcoal - a review. Biology and Fertility of Soils 35, 219-230.
• Mann, C. C. 2002. The real dirt on Rainforest fertility. Science 297, 922-923.
• Neves, E. G., Bartone, R. N., Petersen, J. B., Heckenberger, M. J. 2001. The timing of Terra Preta formation in the central Amazon: New data from three sites in the central Amazon., pp. 10.

A full list of references can be obtained from the author at:
Institute of Soil Science and Soil Geography, University of Bayreuth, D-95440 Bayreuth, Germany
Email: bruno.glaser@uni-bayreuth.de

Top of page

MANGROVES IN THE RED RIVER ESTUARY IN VIETNAM: A SINK OR SOURCE OF NUTRIENTS?
By J.H.M. Wösten

Introduction
Mangroves are tidal forest ecosystems in sheltered saline to brackish environments. They act as a buffer between land and sea as they prevent erosion, reduce currents, attenuate waves and encourage sediment deposition and accretion (Augustinus, 1995). Moreover they provide food and shelter for birds, fish, shrimp and crab. At the same time they support commercial and recreational fisheries and deliver several direct and indirect services to the local population such as firewood.

Mangroves are threatened both by oceanic factors such as erosion caused by strong currents and high waves during typhoons, and by terrestrial factors such as intensification of agriculture and aquaculture. In the Red River Estuary communities of Sonneratia caeseolaris, Bruguiera gymnorhiza and Aegiceras corniculatum dominate.

As part of a larger study on sustainable management, this study focuses on understanding the cycling dynamics of carbon, nitrogen and phosphorus in the estuary. Details of this study are presented in Wösten et al. (2002). In studying the nutrient dynamics use is made of the LOICZ (Land Ocean Interactions in the Coastal Zone) – CABARET (Computer Assisted Budget Analysis for Research, Education, and Training) model (Gordon et al., 1996).

Study area and modelling approach
The estuary of the Red River is located about 110 km southeast of Hanoi, the capital of Vietnam, and consists of mudflats, mangrove swamps, salt marshes and sandy beaches. From the point where the Red River enters the sea, the area extends approximately 10 km to the north, 10 km to the south and 5 km seaward, covering a total area of 107 km2.

The LOICZ approach is a budgeting procedure describing the rate of material delivered to the system ("inputs"), the rate of material removed from the system ("outputs"), and the rate of change of material within the system ("internal sources or sinks"). In this approach four sequential budgets are established: (i) water budget, (ii) salt budget, (iii) budgets of P, and (iv) C:P and N:P ratios to calculate C and N sequestration. In this study, measured data for river discharge, precipitation, evaporation and groundwater flow are available. Based on these data, residual flow is calculated. To conserve salt in the estuary, the amount of salt leaving the estuary with residual flow is balanced by an amount of salt entering the estuary with mixing flow caused by winds, tides, or estuarine flow.

P loss is assumed to represent conversion of P to mangrove biomass and allows estimation of primary production (p) minus respiration (r). A positive value of the net metabolism (p – r) implies that production of the system is higher than respiration and therefore that C is captured in the system. With respect to the nitrogen budget, nitrogen fixation and denitrification are important. The expected amount of N uptake or release is related to the P uptake or release and to the N:P ratio of organic matter. Assuming that all P loss is sequestered in mangrove biomass ignores the uptake and release of nutrients by the abundant mineral particles in the estuary. Further research is needed to quantify this process of nutrient exchange with mineral particles.

Data Availability for the Red River Estuary
The budget model approach requires input data on water, salt and nutrients. These data were collected in Vietnam and are reported by Tri et al. (1999). Figure 1 shows that river discharge is by far the most important component in the water budget, while groundwater flow is negligible and precipitation and evaporation are also small. Large monthly differences exist in river discharge, with relatively high values in the wet months May-October and low values in the months November-April. Measured data for salinity, Dissolved Inorganic N (DIN) and Dissolved Inorganic P (DIP) concentrations are shown in Figure 2.

Water and nutrient balance
First, residual and mixing flows are calculated resulting in values of -108.106 m3 d-1 and 119.106 m3 d-1 respectively. The negative sign for residual flow indicates that there is a net water outflow from the estuary to the sea. Because residual flow carries salt out of the estuary, there is a compensating salt source in the form of a mixing flow which has a positive sign. Next components of the N and P budgets are calculated. Table 1 shows the calculated three major components of the material balance, daily import and daily export, and net balances. Daily import of N in the estuary is about 1090 kmol d-1 while daily export of N is 268 kmol d-1. This implies that mangroves in the estuary sequester the difference between import and export, yielding 822 kmol d-1 of N and thus act as a sink for N. Similar to N, Table 1 shows that daily import of P is about 389 kmol d-1 while daily export of P is 286 kmol d-1. This means that the mangroves in the estuary sequester the difference, yielding 103 kmol d-1 of P and thus also act as a sink for P.

Table 1: Calculated inorganic N and calculated inorganic P flows, daily import in the estuary system (FQ), daily export from the estuary system , and net balances (import-export) for the estuary system in kmol d-1.

Biomass production
With a calculated P uptake of 100 kmol d-1 which equals 3 100 kg P d-1 and a C:P ratio of 300 for mangroves (Tri et al., 1999), net carbon metabolism amounts to 30 Mmol d-1 which equals 360 000 kg C d-1 for the total area of 107 km2. Expressed per unit surface area this implies a carbon fixation of approximately 33 kg ha-1 d-1. Assuming that 40% of the dry matter is carbon, a maximum biomass growth rate of about 80 kg dry matter ha-1 d-1 is calculated. This calculated growth rate agrees well with rates measured by Gong and Ong (1990).

Biomass growth of a full grown mangrove stand in the Red River Estuary as measured by Tri et al. (1999) amounts to 31 kg dry matter ha-1d-1. Agreement between calculated and measured growth rates (80 versus 31 kg ha-1d-1 respectively) is considered to be acceptable. These values provide an independent validation mechanism and thus support the accuracy of the calculated N and P sequestration data.

Conclusions
Mangroves in the Red River Estuary function as a sink of nutrients. Independent data on mangrove growth rates support the calculated N and P sequestration data. As a follow up, it would be attractive to use the described budget approach to investigate the effects of realistic future scenarios on changes in river discharge and river nutrient concentrations. These changes could be caused by changes in the water retention of the Red River catchment or by predictions on the population growth of Hanoi. In such a scenario analysis it should be remembered that calculations are largely based on average hydrological and nutrient concentration data, whereas in reality these data vary considerably in time and space. Nevertheless, this study demonstrates that the described budget approach is capable of assessing the nutrient status of the Red River Estuary even if relatively few input data are available on hydrology and nutrient concentrations.

References

• Augustinus, P.G.E.F. 1995 Geomorphology and sedimentology of mangroves. In Geomorphology and sedimentology of estuaries (Perillo, G.M.E. ed). Elsevier Science, Amsterdam. pp. 333-357.
• Gong, W.K. & Ong, J.E. 1990 Plant biomass and nutrient flux in a managed mangrove forest in Malaysia. Estuarine, Coastal and Shelf Science 31, 519-530.
• Gordon, D.C., Boudreau, P.R., Mann, K.H., Ong, J.E., Silvert, W.L., Smith, S.V., Wattayakorn, G., Wulff, F. & Yanagi, T. 1996 LOICZ Biogeochemical modelling guidelines. LOICZ Reports and Studies 5, pp. 96, LOICZ, Texel, Netherlands.
• Tri, N.H., Ninh, N.H., Lien, T.V., Trinh, B., Chinh, N.T., Trong, T.D. & Secretario, F.T. 1999 Economic- Environmental modelling of coastal zones in the Red River delta. Final Report. Hanoi, Vietnam, 62 pp.
• Wösten, J.H.M., de Willigen, P., Tri, N.H., Lien, T.V., Smith, S.V. 2003 Nutrient dynamics in mangrove areas of the Red River Estuary in Vietnam. Estuarine, Coastal and Shelf Science (in press).

List of Figures
Figure 1. Measured monthly changes in river discharge (VQ), precipitation (VP), evaporation (VE) and groundwater (VG) in 106 m3 d-1.
Figure 2. Estimated monthly changes in concentrations of DIN and DIP in river water, and salinity in the estuary.

J.H.M. Wösten
Alterra Green World Research, P.O. Box 47, 6700 AA Wageningen, The Netherlands.
Email: j.h.m.wosten@alterra.wag-ur.nl

Top of page