Previous PageTable Of ContentsNext Page

Integrating the assessment of non-wood forest products into the forest inventory of a large area: experiences from Nepal

Christoph Kleinn,1 Risto Laamanen2 and Samar Bahadur Malla3

1Abteilung für Forstliche Biometrie, Universität Freiburg, Werderring 6, D-79085 Freiburg, Germany
2 Forest Resources Information System Project, Kathmandu, Nepal
3 Retired, former: Department of Plant Resources, Kathmandu, Nepal


This paper describes an attempt to integrate the assessment of some non-wood forest products (NWFPs) in the framework of a standard forest inventory. The study is restricted mainly to medicinal and aromatic plants, i.e., herbal NWFPs, a group that has economic importance in the hill regions of Nepal. Several specific difficulties are discussed. They arise from the fact that a regular forest inventory is not specifically designed for a survey of herbaceous plants. A pilot inventory carried out in two districts in Nepal is described. The experiences are reported and discussed.


Non-wood forest products (NWFPs) are attracting more and more attention from researchers worldwide. In many studies the importance of these products and their sustainable management is stressed. Their contribution to the subsistence of local populations and to the macroeconomic development differs very much from region to region. The promotion of NWFPs can and should complement the objectives of rural development and appropriate forest management (Hammet 1993), as they are sources of alternative employment and income generation (Sharma 1995). In short, the full economic valuation of endangered tropical forest ecosystems is not possible without consideration of the value of non-timber forest products (Hall & Bawa 1993). Sustainable reliance on NWFPs also creates the need to maintain and conserve biodiversity (Sharma 1995). Thus, management of NWFPs cannot be seen separately from general forest management, which, unlike forest plantation, affects vegetation and biodiversity in general.

It is frequently emphasized that detailed and systematic data about this natural resource is lacking (Malla et al. 1995). Sharma (1995) points out the lack of an ecological data base. Existing information is extremely scant regarding the status of the resource base, the probable impact of harvesting and collecting practices, etc., and area-specific sustainable harvesting. This, of course, is a situation that makes management of any resource difficult.

So it appears justified to think about the assessment of NWFPs in a general way. Forest inventories are evolving from a pure timber orientation towards multiresource (Lund & Wigton 1994). They should take into consideration other products and services and sometimes extend even to areas not falling under the definition of forest. Inclusion of NWFPs in land-use inventories therefore seems a straightforward idea.

However, there is no comprehensive publication yet that systematically addresses the specific problems in the context of the assessment of NWFPs. This paper makes a contribution towards such a publication by presenting practical experiences from Nepal.

Classification of NWFPs for inventory

The term non-wood forest products (NWFP) is used here in a more restrictive sense than that proposed by Chandrasekharan (1995), as for this study the service functions of forests are disregarded.

Many techniques have been developed to assess wood volume, its quality and its spatial distribution in traditional forest inventories. It is a minor issue to integrate the assessment of non-wood forest products, which show some relationship with known tree dimensions, into a forest inventory. Other products unrelated to the trees pose more severe problems, but they can be classified into three groups according to feasibility of estimating them in forest inventory (table 1).

Table 1. Summary of three groups of NWFP according to feasibility criteria of forest inventory

NWFP Group Group description Examples Comments
1 Non-wood
tree parts
Tree fruits, leaves, twigs (fodder, ornamental) Can be related to tree dimensions
2 Products from
´tree like´ plants
Bamboo, rattan Relatively easy measurable dimensions
3 Herbs and other
Medicinal and aromatic herbs Some specific properties to be taken into consideration when incorporating into standard forest inventories

Note that products like fuelwood, charcoal and small-dimension wood materials for carving are excluded in the definition of NWFPs. Yet they are included when the term non-timber forest products is used.

The first group of NWFPs comprises products that come from trees. In this group products like fruits and nuts, bark, leaves, twigs for fodder or ornamental purposes, leaf litter and dry foliage, natural gum and resin are included. The occurrence of these products is dependent on the trees. So to estimate their presence, quantity and possibly quality of the produce, it is necessary to establish the relationships between the products and characteristics of the trees.

To assess the potential of fruit or nut production, for example, one needs to know the tree species and the conditions for fruiting. This is species specific; the variables determining season and intensity of fruiting might include the weather, tree age, tree dimension (stem and crown), phytosociological position in the stand, and the level of competition. In addition, the size of the fruit crop can be greater than what is available for harvest. Normally only part of the fruit is accessible to the harvesters. A similar problem occurs in timber-oriented inventories. Only part of the growing stock of the forest is for harvest. Some areas are excluded, e.g., for conservation reasons or because of problems of accessibility. Because of size or species, only a portion of the timber is utilizable or merchantable. While these proportions can be determined relatively easily for timber, it is more difficult to do so for the NWFPs of group 1.

The second group comprises products like bamboo, rattan, palm trunks, palm leaves and palm syrups. These products are different from those in the first group as they do not originate from dicotyledonous trees. With respect to a field survey, however, they exhibit characteristics similar to those of trees and can be relatively easily assessed. Bamboo, rattan and palm species are easily detectable and have clearly measurable dimensions (length and diameter), although there are some practical problems in carrying out accurate measurements.

The third group comprises non-woody plants parts, usually of the forest undergrowth. It comprises medicinal and aromatic herbs, roots and rhizomes, and ornamental plants, mushrooms. Much less experience exists in forest inventory with respect to the systematic assessment of these herbal NWFPs. These plants exhibit some characteristics quite different from those of the first two groups. Field assessment is a much more difficult task for the following reasons:

· Seasonality: Many of the usable herbs can be identified only in a particular period of the year. For the rest of the year they are not detectable as they are only in the form of seed or underground organs. If it is assumed that an inventory goes for direct identification of the herbs, field assessment is restricted to the short period of the year when they are growing. Therefore, to cover a large area, many field teams have to be employed to finish the survey in the short time available. This contrasts with a regular forest inventory where a longer part of the year is available for field measurements.

This issue of seasonality also means that permanent sample plots have to be remeasured at about the same time of the year on each occasion. This restriction poses some difficulties for organizing non-wood forest product assessments. The situation is complicated further by the fact that not all the NWFP species have the same seasonality and that the growing period is also a function of the climatic conditions of the year of the inventory.

· For some of the NWFPs in group 3 the harvest is destructive, with the entire plant or vital parts of it being removed (roots, tubers, rhizome). In this case, no traces of the former existence of these plants are left, and an inventory field team visiting such a site would classify it as one in which this species does not grow. Depending on the completeness of the harvest and the species occurrence in the vicinity, recolonialization may take some time.

In plant science, site conditions are generally assessed together with the occurrence. Each plant has a set of site demands, but whether plants grow on suitable sites depends on other factors (history or just pure chance). Thus, it is useful to assess not only the plants themselves but also the site. In this way, it is possible to draw conclusions from the site conditions about the potential occurrence of the plants. This, of course, requires detailed knowledge of the NWFP species or species groups. In the ideal case, from an inventory point of view, one would be able to correlate characteristics that can be discerned in aerial photos with habitats of the plants of interest. These are, for example, macro characteristics of site (like elevation, aspect, slope, geology) in a particular forest type that are easily detected. Others are species composition, density, age and stand structure.

· A great variety of herb species are usable. To be able to identify them, botanical knowledge and experience are necessary, which normally forest inventory field teams do not have. As in tropical forest inventories, where tree species are identified by local tree spotters, it is possible to employ local collectors for the field work. To guarantee consistency, it would be preferable to work with a permanent group of botanists. These, however, are highly specialized people, often too expensive to be employed throughout the entire fieldwork period.

Appropriate training for the field teams is required in any case. The species to be considered must be limited to the most important ones (Malla et al. 1995).

· Herbal plants are not as striking as trees. Trees can hardly be overlooked in a field plot. This does not hold for all herbal plants, and situations can occur in which some herbal plants are quite easily missed. To avoid negatively biased results, it is desirable to introduce `detection probabilities', as used in wildlife surveys (e.g., Thompson 1992). Several factors affect the chance to locate a herb, and they are likely to differ between species. Other important factors are the density of the understorey, the weather conditions, and the knowledge and experience of the team.

Detectability and ease of identification are problems that can reduce data quality. This is a major issue for the inventory of herbal NWFPs.

· Experience shows that control measurements are indispensable for achieving a satisfactory level of data quality. However, seasonality of herbal NWFP puts limits on the time available for control measurements, and the problem of detectability complicates the interpretation of differences between original and control measurements.

· The scale and the spatial distribution in the occurrence of herbal plant populations show a pattern completely different from that of trees. Field plots, as used for forest inventories, are not necessarily appropriate for the survey of herbal plants. Demarcation of the plot boundaries in the field by cords or tapes is very helpful to guarantee that a correct decision is made as to whether an individual is in or out of the plot area.

· Quantitative assessments of trees are relatively easy to make. With herbal plants this is not so. Even simple counting can be difficult if the plants occur in dense clusters. Biomass or weight of the entire plant or of parts of it cannot be measured directly. For some plants, the parts of interest are underground and not even visible (roots, tubers, rhizomes). Estimation then requires relations between simply measurable attributes and the target attribute. These relations have to be developed first. Again, because of the great variety of herbal NWFPs, a single generalized method of quantification may not be applicable to all species.

· Many of the NWFPs in group 3 do not occur only inside forests. They might also grow in shrubland or in woodlands that do not fall into the vegetation or land-use class of forest, or even of grasslands. This ecological amplitude must be known for the species under assessment, so that it is clear that a forest inventory is not capable of delivering the entire picture. Trees outside forest boundaries are also similarly ignored in inventories.

The above problems of seasonality, the great variety of species, and the detectability problems are a hindrance but not an overwhelming constraint to carrying out inventories. They do, however, pose additional organizational and reliability problems.

The simple classification of NWFPs shows that the main problems of NWFP assessment are with group 3. Their assessment extends more into the field of general vegetation research and quantitative techniques to sample vegetation. Experiences from Nepal with herbal NWFPs are presented here as a detailed example of making NWFP inventory.

Some experiences from Nepal

General situation

Nepal is one of the principal countries of the world producing non-wood forest products (see Sah & Dutta, this volume). Non-wood forest products, especially medicinal and aromatic plants, are important, particularly in the hill and mountain region. About 650 plants are utilized and commercialized in one form or another. Currently about 100 of these have a high commercial value and are traded. Medicinal and aromatic plants, as well as the NWFP-based industries form one of the six sector programmes in the Nepal Master Plan for the Forestry Sector (MPFS) 1989. The royalties and the levies paid when exporting NWFPs provide a considerable income for the government.

Harvesting NWFPs has grown from occasional collection by shepherds and mountain people to a large business involving full-time collectors, middlemen and dealers on several levels. These activities started to threaten the existence of some of the species, because some collectors leave no regenerative material on the site (Malla et al. 1995). Although nobody knows the actual situation in the field, the figures in the export statistics show that the trade with some of the products is decreasing dramatically. Assuming that the market demand has remained unchanged, this can be interpreted as a sign of the fact that these plants have been overexploited.

Management and monitoring of the harvest activities has to be based upon quantitative and spatial information about the resource. As a considerable portion of the NWFP comes from forests, it seems reasonable not to separate the NWFP issues from the other forest management issues. Their forest planning and management should be carried out within the framework of a general management plan. So it is natural that the assessment of NWFPs should be considered an important part of any forest inventory planning. But research is needed on methods for providing good quality data on NWFP.

Only a few studies could be found in which surveys of NWFPs had been considered. One had been carried out by the Department of Plant Research in the mountain area of Jumla. It was an evaluation of official statistics, and it is not clear how accurately the true situation is reflected. Malla (1991) gathered data on the type of NWFPs, their existence in different districts and the quantity said to have been collected. This study was based mainly on interviews with collectors and traders and on an evaluation of official statistics made to keep track of royalties and export levies.

However, it is generally accepted that an evaluation of official statistics and interviews with collectors and dealers cannot serve to obtain an objective picture of the NWFP situation, neither quantitatively nor with respect to their spatial distribution. Often the statistics are collated from information of the dealers, in regions other than their collection area (Edwards & Bowen 1993). This distorts the picture of their natural occurrence. At best it may help to get a general idea about the importance of the NWFP trade and about the relative market share of different products. Obviously, NWFPs are also subject to illegal collecting and smuggling, and such products would not appear in any official statistics.

Several studies stress the necessity to conduct objective surveys as the basis for management activities. In Edwards & Bowen (1993), it is stressed that inventories should seek to clarify the ecological requirements of species, area of land meeting these requirements and abundance of different species.

Case study

The Forest Resources Information System (FRIS) Project, funded by FINNIDA and carried out by the Finnish Forest and Park Service in cooperation with the Forestry Department in Kathmandu, has the objective of building up and establishing a comprehensive forest information system for Nepal. Within this project, a National Forest Inventory was designed and pilot studies were carried out in several districts, as it was felt that NWFPs form an important part of the forest resource in Nepal that requires inventorying. This was then put into practice in a pilot inventory in the two districts of Dhading and Nuwakot (Kleinn 1994, Laamanen et al. 1994).

A cluster layout is depicted in figure 1. To assess tree characteristics, two circular plots were used in one cluster (Radius 18 m each, plot area 1017.9 m2). They were 100 m or 200 m apart from each other, depending on terrain conditions. NWFPs were assessed in a strip connecting the two circular plots within a cluster; the area of the NWFP assessment strip was 1000 m2, with side lengths 200 x 5 m or 100 x 10 m. Altogether 20 clusters (40 plots) were laid out during the pilot inventory exercise.

W3735e08.GIF (5572 bytes)

Figure 1. Field plot design of pilot inventory, combining classical forest inventory and NWFP assessment.
NWFP assessment on a 1000-m2 strip (200 x 5 m or 100 x 10 m) connecting the two circular forest inventory plots.

The strip, demarcated by rope in the field, was made narrow to facilitate plant counting. Transects are standard field sample plot shapes employed in quantitative vegetation research, but the area of 1000 m2 is regarded too small in vegetation research to sample distribution or abundance. It was used here for reasons of feasibility of field work.

A team of NWFP specialists, consisting of a botanist, an assistant botanist, a field assistant and a local helper, accompanied every forest inventory field team. As soon as the centre point of the first circular plot was determined, the NWFP team started with the measurements on the strip while the forest inventory team worked on the circular plot. This procedure worked smoothly and without problems. To arrive at quantitative estimates of the products, the useful parts of a sample of five specimens of each species were weighed. The average weight was then used to estimate the total weight per unit area.

The pilot inventory had the objective of learning more about the sampling procedure and its feasibility rather than producing numbers about the occurrence of NWFPs; however, some general results (following Laamanen et al. 1994) can be presented:

· It was observed that the occurrence of some important NWFPs was well correlated with the surrounding forest type. For example, Woodfordia fruticosa, Asparagus racemosus and Osyris wightiana occurred in Schima-Castanopsis forests, in pure Shorea robusta (Sal) and mixtures of them. This supports the idea that indirect measures could be developed for assessing NWFPs. If one could succeed in correlating their occurrence with geographic attributes (physiographic zone, altitude, geology, soil, etc.) or stand characteristics (forest type, species composition, crown cover, mean dimensions of the trees) the sampling procedure would be considerably facilitated. Aerial photographs could even be usefully used.

· A similar need has been stressed by Malla et al. (1995) regarding a quantitative assessment of Rheum australe and Asparagus racemosus. They indicate the need to find out which of the NWFPs have similar demands on their habitat so that they can be classified by habitat. It will probably not be possible to do this for all of the 100 most important NWFPs of Nepal. In addition, it is equally important to find indicators to characterize the preferred habitat for each of the main groups of NWFPs. These indicators should be easily measurable.

· Seasonality is an important consideration. During the survey, many of the Asparagus tubers were found to be rotten already. The best time to survey these species would probably be after the rainy season, or in early autumn.

· Variation in number of NWFPs between the field plots was very high, indicating clustering of the individuals, which can be due to minor changes of site conditions. The 1000-m2 plot was therefore too small, particularly if a quantitative assessment is to be carried out. A sampling intensity of 5-10% of the target area is therefore proposed for a fair estimation. This is not, however, congruent with the basic idea of large-area inventories, which should normally have a much lower sampling intensity.

· In many plots NWFP assessment was hindered by the dense undergrowth of the understorey.

· Many of the medicinal plants have not been explored yet. A medicinal plant locally called `teet pindalu' and used for a variety of medicinal purposes was identified as Arundina graminifolia.

· On average it took 3 hours to complete one strip.

The conclusion of this study was that combining the forest inventory with a non-timber forest product assessment was technically feasible (Laamanen et al. 1994). However, for application over a large area the budget required would become a major constraint.

Discussion and conclusions

There is a need to reflect upon the usefulness and justification of NWFP within the framework of large area forest inventories. First it is important to be clear about the objectives and particularly to identify who will benefit from the information collected.

Large-area forest inventories are to provide information on the status and condition of the forests. Sampling intensity is usually small, while the units reported on are large. The results are generally used to support regional planning and policy formulation being evaluated and used by government departments. Often there is also an interest in such data from other organizations and projects dealing with forestry, land-use aspects, ecology, etc., on regional, national or international scales.

The benefits of improved information in regional planning or forest policy formulation cannot be easily quantified economically. Therefore, justifying the investment for an inventory of a large area is sometimes not easy, especially when NWFPs are included in the list of variables to be assessed. The questions to be answered must be clearly identified.

There are different schools of thought on the usefulness of NWFPs inventories. Godoy et al. (1993) recommend that the inventory (the stock quantity in the forest) and the flow (the quantity actually used by people) be distinguished. They further state that, for most purposes, the inventory is meaningless, being related neither to present nor to sustainable use. For both timber and non-wood forest products, only a part of the stock quantity is normally available for harvest and utilization. For economic evaluation, the proportion must clearly be known (explicitly for specific products and also for specific regions). However, there are many concepts of forest inventories and many ways to use the data collected. The authors of this paper therefore find the general argument of Godoy et al. (1993) not very useful.

A more specific criticism has pointed out by Edwards & Bowen (1993): who indicate that Macro level inventories of NWFPs are of limited value unless specific case studies are conducted to support the data. Within the overall objective of contributing to conservation by employing sustainable management of NWFPs, one has to realize that a large-area inventory is only one part of the process. Gunatilleke et al. (1993) have emphasized that socioeconomic, biological and silvicultural information is required for a reasonable multiple-use management of rain forests. Large-area forest inventories do deliver some pieces of needed information in this mosaic.

From large-area forest inventories, data on occurrence (geographical distribution) and quantity (number of plants, size, weight of usable parts, etc.) of selected NWFPs can be obtained. Not all usable plants can be included. A selection of the most important NWFPs, or groups of NWFPs, must always be made. From the experiences of our pilot study, it appears that at least 20 species can easily be managed without botanists, provided useful field manuals for identification of those plants are made available to well-trained field crews. Observations on additional species can of course be recorded as additional information. Further data collected refer to the surrounding vegetation type and to the topography that is possibly associated with the occurrence of the NWFP species.

Data analysis will give a picture of the spatial distribution of the NWFPs. Because of the nature of large-area inventories, this is an estimate referring to larger geographical units, and comparison with other regional statistics will be informative. The data will give sound baseline information for identification of endangered species, particularly when remeasurements are available so that changes in the population can be observed.

On the research side, the data can be used to build relationships between auxiliary data and the occurrence of species. This will be useful during further inventories, but also as a general contribution to the knowledge of their biology. Such data, if easily retrievable from a database, will be most valuable for many kinds of studies, such as the adjustment of royalties or other means of fund raising. However, the availability of this data could lead to further exploitation of the resource.

It is difficult, to draw general conclusions from all of these considerations. The authors' opinion from this experience in Nepal is that the data collected in the large-area inventory was very worthwhile and can contribute to the development of a strategy for managing NWFPs. One of the outputs may be the protection of species that have been found to be endangered. This protection must be enacted by government, but its enforcement for the sustainable management of NWFPs can succeed only through community awareness and action.


Chandrasekharan C. 1995. Terminology, definition and classification of forest products other than wood. International Expert Consultation on Non-wood Forest Products. FAO Non-wood Forest Products No. 3. p 345-380.

Edwards D.M. & M.R. Bowen, eds. Focus on Jaributi. Proceedings of the Non-timber Forest Products Seminar, Kathmandu, 12 May 1993. FRSC Occasional Paper 2/93. 26 p.

Godoy R., Lubowski R. & Markandya A. 1993. A method for the economic valuation of non-timber tropical forest products. Economic Botany 47(3):220-233.

Gunatilleke I.A.U.N., Gunatilleke C.V.S. & Abeygunawardena P. 1993. Interdisciplinary research towards management on non-timber forest resources in lowland rain forest of Sri Lanka. Economic Botany 47(3):282-290.

Hall P. & Bawa K. 1993. Methods to assess the impact of extraction of non-timber tropical forest products on plant population. Economic Botany 47(3):234-247.

Hammet A.L. 1993. Non-timber forest products: profits or panacea? p 2-3. In: Focus on Jaributi, ed. D.M. Edwards & M.R. Bowen. Proceedings of the Non-timber Forest Products Seminar, Kathmandu, 12 May 1993. FRSC Occasional Paper 2/93.

Kleinn, C. 1994. Forest resources inventories in Nepal: status quo, needs, recommendations. FRIS Project Paper No. 1. Forestry Department, Kathmandu, Nepal. 80 p.

Laamanen R., Malla S.B., Battarai N.K. & Subedi M.N. 1994. Pilot forest inventory in Dhading and Nuwakot:assessment of minor forest products in Dhading. FRIS Project Paper No. 2. Forestry Department, Kathmandu, Nepal. 28 p + annexes.

Lund G. & Wigton W.G. 1994. A primer for designing multiple resource inventory (MRI) and monitoring programs. AIFM/IUFRO Conference on Multiple Resource Inventory and Monitoring of Tropical Rain Forests, 21-24 November 1994, Seremban, Malaysia.

Malla S.B. 1991. Medicinal plants: their utilisation and management and the biological diversity of the flora of Nepal. p 401-444. In: Background papers to the National Conservation Strategy for Nepal, vol. 2. World Congress Union, IUCN.

Malla S.B., Shakya P.R., Rajbhandari K.R., Battarai N.K. & Subedi M.N. 1995. Minor forests products of Nepal: general status and trade. FRIS Project Paper No. 4. Forestry Department, Kathmandu, Nepal. 27 p + annexes.

Sharma P. 1995. Non-wood forest products and integrated mountain development: observations from Nepal. International Expert Consultation on Non-wood Forest Products, FAO Non-wood Forest Products No. 3. p 157-166.

Thompson, S.K. 1992. Sampling. John Wiley & Sons. 343 p.

Previous PageTop Of PageNext Page